HELIOS NAD-Seq: A Next-Generation Capture and Sequencing Protocol for NAD-Capped RNAs with Superior Targeting and Processing

HELIOS NAD-Seq: A Next-Generation Capture and Sequencing Protocol for NAD-Capped RNAs with Superior Targeting and Processing | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article HELIOS NAD-Seq: A Next-Generation Capture and Sequencing Protocol for NAD-Capped RNAs with Superior Targeting and Processing Marvin Möhler, Tae Wha Seong, Frederik Weber, Héctor Gabriel Morales Filloy, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7650670/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 9 You are reading this latest preprint version Abstract Background Nicotinamide adenine dinucleotide (NAD) can function as a non-canonical initiating nucleotide, producing NAD-capped RNAs across all domains of life. This 5′-modification has been implicated in RNA stability and host–pathogen interactions. Existing identification methods, such as NAD captureSeq, require large RNA inputs, show limited specificity, and have low sample throughput, restricting their utility for low-input or large-scale studies. Results We developed HELIOS NAD-Seq ( H igh-efficiency E nzyme modification with L ow I nput O f S ample for NAD- RNA Seq uencing), a protocol that combines a pyridyl-based ADP-ribosyl cyclase substrate, 3-picolylamine biotin, with early sample barcoding to enable high-yield, one-step biotinylation of NAD-RNAs. HELIOS NAD-Seq reduces RNA input requirements by at least 50-fold compared to NAD captureSeq, increases specificity for NAD-caps, and allows simultaneous preparation of libraries from at least 16 sample groups in quadruplicate within four days. Applied to Escherichia coli , HELIOS NAD-Seq confirmed NAD-capping for 89.5% of previously identified transcripts from NAD captureSeq and detected 242 additional, predominantly low-abundance RNAs. Time-course profiling of NAD-capping across the bacterial growth curve identified a core set of metabolism-related transcripts showing low expression in stationary phase, sharp peaks in early exponential growth, and decreases in the plateau phase, suggesting a link between metabolic state and NAD-capping dynamics. Conclusions HELIOS NAD-Seq is a robust and sensitive high-throughput platform for NAD-RNA profiling enabling low-input and large-scale studies. Its application to E. coli demonstrates technical advantages over existing protocols and reveals dynamic NAD-capping patterns associated with metabolism. NAD-capped RNA non-canonical initiating nucleotide RNA modification RNA metabolism transcriptomics Escherichia coli high-throughput sequencing RNA capping NAD-Seq Figures Figure 1 Figure 2 Figure 3 Background The modification of RNA with nicotinamide adenine dinucleotide (NAD + ) at the 5' end resembles the eukaryotic 7-methylguanosine (m 7 G) cap. Unlike the m 7 G cap, NAD + and related metabolites such as NADH, FAD, dephospho-coenzyme A, and others can be incorporated into RNA as non-canonical initiating nucleotides (NCIN) during transcription primarily through cotranscriptional processes.( 2 – 8 ) Originally identified as the first prokaryotic RNA cap in 2009 by Liu and coworkers,( 9 ) the NAD-modification remains the most extensively studied non-canonical RNA cap( 10 , 11 ). Since the development of NAD captureSeq by Jäschke and colleagues in 2015,( 1 , 12 ) NAD-capped RNAs have been found across all domains of life—from bacteria( 1 , 13 – 15 ) and archaea( 15 ) to yeast( 16 , 17 ), plants( 18 – 20 ), and animals( 21 , 22 ), including human cells.( 23 ) The widespread occurrence of NAD-capping, coupled with the existence of specific enzymes (“erasers”)( 13 , 17 , 23 – 30 ) and both prokaryotic and eukaryotic RNA polymerases as “writers”( 1 , 2 , 13 , 23 , 31 , 32 ) supports the notion that NAD-capping is a conserved and functionally relevant biological phenomenon. Recent studies have linked NAD + and FAD capping to viral infection processes,( 30 , 33 , 34 ) while suggested functions in cells include RNA stability( 1 , 35 ), regulation of gene expression( 33 , 34 ), and involvement in processes like RNAylation—a covalent attachment of NAD-capped RNAs to proteins( 36 ). Despite this, our understanding of the physiological roles of NAD-capping remains limited, and further exploration is hindered by the limitations of current detection methods. Cap-specific NAD-RNA identification protocols play a central role in the discovery of NAD-RNAs with potentially new roles in a wide range of organisms. The recurring element of those capture protocols, which are based on NAD captureSeq,( 1 , 12 ) is the use of a primary alcohol substrate class to modify the NAD-cap structure in a key capture reaction catalyzed by the enzyme ADP-ribosyl cyclase (ADPRC) from Aplysia californica .( 37 , 38 ) This modification reaction is generally used to attach a tag molecule to the NAD-cap, which allows either the specific purification or the detection of NAD-RNAs from a total RNA sample, permitting the sequencing and identification of NAD-capped RNA species.( 1 , 18 , 21 , 22 , 30 , 39 – 41 ) As discussed in our recent review article, NAD-RNA capture protocols would benefit substantially from improvements in ADPRC reaction yield and reaction specificity, reductions in total RNA demand, and an increase in sample processing capacity.( 42 ) Currently, most methods advise to start from around 100 µg of total RNA per replicate (or an equivalent amount of e.g. polyadenylated RNA fraction),( 1 , 18 , 21 , 22 , 40 , 41 ) which creates a barrier to analyzing samples with low RNA yield or NAD-capping levels (often only around 0.1–0.2% of the transcriptome).( 11 ) Undesired reactivity and off-target effects, especially with abundant m 7 G-capped RNAs, affect the evaluation and reliability of the obtained sequencing data.( 42 ) Finally, only a low number of samples can be processed in parallel, which makes the analysis of sample screens (e.g. total RNA from different biological conditions or growth states) time-consuming and more error-prone.( 42 ) To overcome these challenges, we introduce HELIOS NAD-Seq ( H igh-efficiency E nzyme modification with L ow I nput O f S ample for NAD- RNA Seq uencing), a novel protocol that significantly improves yield, specificity, and throughput. Using a pyridyl-based ADPRC substrate, 3-picolylamine biotin (3PAB), HELIOS NAD-Seq enables one-step, highly efficient biotinylation of NAD-RNAs, requiring at least 50-fold less total RNA compared to previous methods. Its reaction conditions avoid common off-target modifications, such as labeling of m 7 G-capped RNAs, and do not need divalent cations or additional click-chemistry steps. This combination of high reactivity, specificity, and low input requirement makes HELIOS NAD-Seq suitable for diverse biological samples. Moreover, the protocol incorporates early barcoding and multiplexing of samples via 3'-ligation with barcoded adapters, allowing simultaneous processing of at least 16 sample groups, each in quadruplicate, within just four days. This design minimizes sample handling bias, enhances reproducibility, and greatly increases processing capacity. The sequencing data obtained for E. coli samples demonstrate that HELIOS NAD-Seq can recover most transcripts previously identified by NAD captureSeq, with the added ability to detect many additional low-abundance NAD-RNAs. Comparisons also reveal higher specificity and sensitivity, even with a 50-fold reduction in total RNA input. The high-throughput approach enables detailed analyses of NAD-capping dynamics during different growth phases, uncovering potential links between cellular metabolism and NAD-RNA prevalence, and highlighting the broad applicability of this method for expanding our understanding of NAD-capping across diverse organisms and conditions. Results ADPRC reaction with PAB derivatives NAD-capped RNA (Fig. 1 a) can be specifically modified through an ADPRC-catalyzed transglycosylation reaction targeting the NAD + -cap structure. This enzyme removes the nicotinamide base via nucleophilic attack at the ribose-1’-carbon, facilitating the subsequent attack of a suitable nucleophile to yield a retention of the ribose-substitution pattern( 43 ) (Fig. 1 b). Established protocols for NAD-RNA capture have utilized this reaction with primary alcohols as nucleophilic substrates for ADPRC.( 1 , 18 , 21 , 22 , 30 , 39 – 41 ) When performing the enzymatic transglycosylation with alkyne- and azide-containing primary alcohols (4-pentyn-1-ol and 3-azidopropan-1-ol, Fig. S1 a-b), an additional step is required to attach biotin or other tags via click chemistry. However, these chemical reactions present limitations, such as RNA degradation in copper-catalyzed azide-alkyne-cycloaddition (CuAAC) and specificity issues with reactive, internal alkynes in strain-promoted azide-alkyne cycloaddition (SPAAC). The recently introduced HEEB substrate ( N -[2-(2-hydroxyethoxy)ethyl]biotinamide; Fig. S1 c) allows one-step biotinylation of NAD-RNAs at a lower substrate concentration. ( 21 , 22 ) However, a side reactivity of the ADPRC-catalyzed reaction of primary alcohol substrates with m 7 G-capped RNAs has been reported,( 39 , 40 ) which could lead to false-positive identification of NAD-capped RNAs if incomplete m 7 G-cap depletion techniques are used.( 21 , 40 , 41 ) Due to the consistent modification of NAD-RNA using primary alcohols in the presence of ADPRC in all these protocols,( 1 , 18 , 21 , 22 , 30 , 39 – 41 ) there remains significant potential for innovation in this key capture step.( 42 ) To improve the efficiency of NAD-RNA modification, we introduce here a novel substrate class, based on N -nucleophilic compounds, which have shown high yields and rapid kinetics during ADPRC transglycosylation at the dinucleotide level (NAD + ).( 44 , 45 ) We identified picolylamine-biotin (PAB) conjugates, in particular 3-picolylamine biotin (3PAB, Fig. 1 c), as potent novel substrates for modifying both NAD + and NAD-RNA. The modular design features a nucleophilic pyridyl head that resembles nicotinamide, connected via a reverse amide and optionally a linker to a functional end (Fig. 1 c). For NAD-RNA identification protocols, this end group can be a high-affinity binder such as biotin, but other functionalities, like fluorophores, can be introduced by selecting different end groups. Synthesis of PAB conjugates involved NHS-activation of biotin derivatives followed by amide coupling with primary amines on 3- or 4-picolylamine (Fig. S2a-b). Besides biotin variants (3PAB, 4PAB), corresponding desthiobiotin conjugates (3PADB, 4PADB) were prepared similarly. In ADPRC transglycosylation assays, these PAB compounds enabled efficient, one-step biotinylation of NAD + -modified targets, with rapid kinetics and high yields (Fig. 1 d). With NAD + as substrate, the conversion to a biotinylated ADPR product (ADPR-3PAB) was completed within 30 minutes at low enzyme concentrations (0.025 mol%) and a threefold excess of PAB (Fig. 1 e). High-performance liquid chromatography (HPLC) and high-resolution mass spectrometry (HR-MS) confirmed the formation of the expected products (Fig. 1 e, Fig. S3). Notably, the reaction with PAB conjugates proceeded independently of Mg 2+ ions and was unaffected by EDTA, which chelates divalent cations and can mitigate RNA hydrolysis (Fig. S4a-c)( 46 – 48 ). NADH was inert in these reactions, indicating specificity for the oxidized state (Fig. S5). Kinetic comparisons revealed that all PAB derivatives could efficiently convert NAD + to the biotinylated ADPR within five minutes or less, with the reactivity ranking as 3PAB > 3PADB > 4PAB > 4PADB (Fig. 1 f, S7d). Quantitative analysis based on HPLC peak areas indicated near-complete conversion for all compounds within these times (Fig. 1 g–h, S8). The k cat / K m value calculated from the exponential fit curve was 7.59 ×10 5 M − 1 s − 1 which is within a reasonable range for ADPRC-catalyzed transglycosylation reactions.( 45 ) For RNA-level applications, we developed a robust assay (Fig. 1 j): a standard transglycosylation reaction using 3PAB and a 113 nt model NAD-modified RNA (about 48% NAD-modified according to acryloylaminophenyl boronic acid (APB)-PAGE analysis( 49 ) which retards NAD-capped RNA). The reaction was performed with excess E. coli total RNA, and after quenching and extraction, the RNA was incubated with streptavidin (Fig. S6). The biotinylated RNA exhibited reduced mobility in electrophoretic mobility shift assays (EMSAs), confirming successful labeling (Fig. 1 k). This method, termed StrepShift assay, was performed using agarose gels and complemented with northern blot analysis where needed. We observed that the efficiency of transglycosylation on RNA was lower than on dinucleotides, likely due to the larger size of oligonucleotides and potential RNA degradation. Maintaining RNA integrity is crucial for accurate NAD-RNA identification. Even with EDTA-containing buffers and reaction times of 30 minutes, higher enzyme concentrations (e.g., 380 nM) caused significant degradation of 113 nt model RNA and total E. coli RNA (Fig. S9). To mitigate this, we reduced the enzyme concentration 10-fold to 38 nM in the HELIOS NAD-Seq protocol, which prevented detectable RNA degradation (Fig. S9). In comparative assays, 3PAB consistently outperformed other pyridyl conjugates in labeling efficiency with the 113 nt model RNA and total E. coli RNA (Fig. 1 l). Reactions were conducted with reaction times from 0.5 to 30 minutes, followed by precipitation, streptavidin incubation, and the StrepShift assay (Fig. 1 k, S10). Due to weaker streptavidin binding, desthiobiotin derivatives (3PADB, 4PADB) produced less pronounced mobility shifts. The reaction with 3PAB approached completion within 30 minutes with a k cat / K m value of 6.17 ×10 5 M − 1 s − 1 , with slower kinetics observed for other conjugates (Fig. S11). The reactivity ranking was confirmed as 3PAB > 4PAB > 3PADB > 4PADB (Fig. S11d). Further, we synthesized biotin conjugates with extended polyethylene glycol linkers, N(PEG) 3 B and N(PEG) 11 B (Fig. S12a). These were prepared via similar NHS-activation methods. However, their efficiency in transglycosylation reactions decreased with longer linkers, yielding only 63% and 25% biotinylation relative to 3PAB, respectively (Fig. S13). Consequently, the smaller 3PAB remained the preferred substrate for efficient NAD-RNA biotinylation. The HELIOS NAD-Seq protocol The workflow of HELIOS NAD-Seq for Illumina sequencing is illustrated in Fig. 2 and hinges on several key concepts. First, total RNA is specifically biotinylated via ADPRC transglycosylation using our novel nucleophilic substrate 3PAB in a single reaction step. This is followed by 3'-ligation of a barcoded adapter, enabling early multiplexing of biological replicates and negative controls, each uniquely barcoded. Biotinylated RNA species are then captured on streptavidin magnetic beads, and all samples and controls are pooled for simultaneous processing. After reverse transcription with adapter-specific primers, the RNA is hydrolyzed with NaOH, and the resulting cDNA is recovered. The cDNA molecules then undergo a second adapter ligation, PCR amplification, and library purification via native PAGE. From up to 16 total RNA samples—each in quadruplicate with controls—the entire HELIOS NAD-Seq protocol can be completed in four days. Starting with total RNA extracted from an organism of choice, typically 4 µg per sample (split into 2 µg for positive and negative controls), internal RNA standards (Fig. S14) are added at approximately 0.25 fmol/µg RNA for normalization and quality control. For protocol optimization, we reduced the initial RNA input and enzyme concentration: ADPRC was used at 38 nM, a tenfold decrease from NAD captureSeq conditions, to minimize RNA degradation (Fig. S9). The nucleophilic substrate (3PAB) concentration was lowered by roughly 70-fold, compared to NAD captureSeq, to 15 mM, maintaining full conversion efficiency. This setup ensures rapid, specific labeling of NAD-capped RNAs, avoiding side reactivity with 5' modifications such as the m 7 G cap, as confirmed by our specificity assays (Fig. 2 b, Fig. S16a). To verify biotinylation efficiency, we used model RNA standards (IS-036) spiked into total RNA before the reaction, which were UV-crosslinked onto nylon membranes, incubated with HRP-streptavidin, and detected via luminol. Biotinylation of NAD-capped species occurred within 0.5 minutes, with no signal detected for non-NAD species even after 30 minutes (Fig. 2 b, S16b). For early sample multiplexing, a barcoded adapter is ligated to the 3'-end of RNA transcripts prior to streptavidin bead capture (Fig. 2 a). Inspired by the NAD captureSeq AD3 adapter,( 12 ) it contains a randomized 5’-region to reduce sequence bias, followed by a Truseq-compatible 3'-region blocked with a C3 spacer and a 4-base barcode (with a 2-base spacer)( 50 ) for multiplexing four biological replicates and controls. A set of 12 unique barcodes was designed using the BARCOSEL tool( 51 ), ensuring a Hamming distance of at least three to minimize sample misassignment (Fig. S17a). 5’-phosphorylated barcoded adapters (Fig. 2 c) were further modified by pre-adenylation using ImpA (Fig. S17b, PAGE analysis in Fig. S17c)( 12 ). After 3'-ligation and heat inactivation, samples and controls are pooled and subjected to streptavidin capture. In vitro, the binding strength of NAD-RNA to streptavidin was assessed via dot blot with radioactive probing: biotinylated RNA generated with various PAB derivatives demonstrated strong, specific interactions, with the weakest binding observed for desthiobiotin variants (Fig. S18). The captured RNA was washed with urea buffers, eluted with biotin, and analyzed by northern blotting using probes against the model RNAs and ribosomal RNAs as controls (Figs. S18–S21). The approach confirmed effective, specific binding that was robust even after multiple washes. Further downstream protocol steps (see Fig. 2 a) were adapted largely from the NAD captureSeq workflow( 1 , 12 ). Each of those steps has been optimized for the lower oligonucleotide input of HELIOS NAD-Seq and, where applicable, the use with streptavidin magnetic beads. After on-bead reverse transcription (RT) using the primer RTp2, the RNA part of the formed RNA-cDNA duplex is hydrolyzed by incubation with 150 mM NaOH solution( 12 ). The released cDNA is recovered and further treated via a C-tailing reaction using terminal transferase and second adapter ligation. After this, we used NEBNext universal forward and indexed i7 reverse primers for PCR amplification of the libraries, which were purified and size-selected by native PAGE. Gel extraction and recovery of the DNA yielded libraries ready for Illumina NGS sequencing. To evaluate the capabilities of our novel HELIOS NAD-Seq protocol, we sought to compare it to the performance of NAD captureSeq, which we regularly apply in our laboratories. For this purpose, we replicated the experiments on E. coli K12 JM109, which were highlighted in the original NAD captureSeq paper( 1 ). E. coli K12 JM109 were grown to the late exponential phase (OD 600 ≈ 2.0) and harvested (Fig. S22). Due to the low RNA input required for HELIOS NAD-Seq, as little as 1.0 mL of culture yielded sufficient total RNA for 10 protocol runs with 2 µg RNA each after purification and DNase digest (Fig. S23). HELIOS NAD-Seq was performed starting from 2 µg total RNA in biological triplicates (for control purposes with slightly elevated internal RNA standard amounts of 1 fmol) following the standard protocol and barcoding with adapters carrying bc01–03 ligated to samples treated with ADPRC and 3PAB, and adapters with bc04–06 ligated to controls treated with ADPRC and 3-picolylamine. Samples and negative controls were pooled before streptavidin magnetic bead capture. After native PAGE purification of the PCR-amplified and multiplexed libraries, a size selection in the region between 150 and 300 bp was performed to comply with the procedure of the original NAD captureSeq (Fig. S24). Due to the mild conditions of HELIOS NAD-Seq, for which only low degradation during the protocol application is expected, it was already possible at this stage to conclude that mainly shorter E. coli RNA transcripts were NAD-capped. Sequencing reads were then processed, aligned, quantified, and analyzed as described in Methods (“HELIOS NAD-Seq—Bioinformatic analysis”). The sequencing data revealed that 94.2% of over 132 million reads could be assigned to the six barcodes, with even distribution across biological replicates. Positive samples exhibited 35–43 million reads, while negative controls contained 5.5–6.1 million, reflecting the successful enrichment of NAD-capped RNAs. After removing reads from internal standards, the positive samples still had nearly 200 times more reads than the controls (6.2–7.5 million vs. 22–50 thousand). Enrichment analysis confirmed the high specificity of HELIOS NAD-Seq, showing that off-target effects, such as non-specific labeling of m 7 G-capped RNAs, were effectively eliminated (Tables S1–S2). From the dataset, 30,080,435 reads (26.6%) mapped to 3,557 out of 4,621 annotated gene features in E. coli K12 JM109 and the pUC19 plasmid( 1 ). Applying strict detection criteria (baseMean ≥ 50, log2-fold change ≥ 3, adjusted p-value ≤ 0.01), we identified 389 transcripts (associated with 344 unique gene features) as high-confidence NAD-capped RNAs (Table S3). These transcripts exhibited a conserved promoter motif (TANNNT) akin to the standard E. coli promoter, typically located at positions − 12 to − 7 upstream of the transcription start site (TSS), which is always an adenosine (+ 1A) in NAD transcripts (Fig. S25a–b). A manual analysis of TSS indicated that 207 transcripts (53.2%) matched known or suspected TSSs, most being protein-coding RNAs or regulatory small RNAs (Table S4). Many were full-length sRNAs, while most mRNAs appeared as truncated 5'-fragments. Comparison with previous NAD captureSeq datasets( 1 ) confirmed most of our top hits—75 of 80 genes from dataset 1, 42 of 44 from dataset 2, and 64 of 70 from dataset 3—were also identified as NAD-capped, with HELIOS NAD-Seq demonstrating significantly higher enrichment factors (Tables S5–S7)( 1 ). Moreover, HELIOS NAD-Seq identified an additional 251 gene features not detected previously, reflecting increased sensitivity. (Fig. 2 e). Some highly abundant NAD-capped RNAs (e.g., from yfjI, ykgS, ymdB, bfr, and yuaQ loci) were new findings. Overlap with the NAD tagSeq II data was markedly lower, likely due to differences in strain identity, growth conditions, and sample timing and is shown in Fig.S26. Overall, the results validate that HELIOS NAD-Seq, starting from substantially lower RNA inputs, offers superior specificity and sensitivity, facilitating accurate determination of NAD-capping at transcript-level resolution. The negative controls of the NAD-capped transcript copA had 0 reads while the non-capped transcript serX had almost no background signal (Fig. S27). To evaluate the specificity of HELIOS NAD-Seq relative to NAD captureSeq( 1 ), we compared the log₂ fold-changes of genes identified by both protocols. As shown in Fig. S28, HELIOS NAD-Seq yields substantially higher log₂ fold-changes, often several orders of magnitude greater, underscoring the high specificity of the ADPRC reaction with the novel 3PAB substrate in positive samples and the absence of biotinylation with 3PA in negative controls. Application of HELIOS NAD-Seq to identify dynamic changes in NAD-capped RNAs To demonstrate the advantages of HELIOS NAD-Seq over NAD captureSeq, we performed an experiment that had previously been challenging due to the large RNA input requirements, extensive hands-on time, and limited multiplexing capacity of NAD captureSeq. Total RNA was collected from E. coli at 30-minute intervals, spanning the stationary, exponential, and plateau phases of the growth curve (Fig. 3 a) resulting in 128 fractions to be processed. After ADPRC reaction and ligation of 8 different barcoded adapters, all 8 samples for each time point (n = 16) could be multiplexed, effectively reducing the number of samples to be processed by a factor of 8. Due to this high multiplexing potential, the entire protocol up to Illumina library preparation could be completed within 4 working days. After sequencing, NAD-capped transcripts were identified at each timepoint, by comparing positive samples to negative controls. A dramatic change in NAD-capping levels was observed across the timecourse, with potential NAD-capped gene species counts increasing more than 14-fold from 0.5 hours to 3.5 hours after inoculation reaching a maximum of 654 transcripts at the 3.5 hours mark, compared to 46 transcripts at the 0.5 hour mark. This peak coincides with the early exponential phase of E. coli growth (Fig. 3 a). Between 3.5 and 4 hours post inoculation, NAD-capping levels dropped sharply (Fig. 3 a), potentially marking the onset of overflow metabolism shown in Fig. 3 b, which is discussed in detail in the Discussion section. KEGG pathway analysis of NAD-capped transcripts at individual timepoints revealed that many genes at 3.5 hours were involved in metabolic pathways, including biosynthesis of secondary metabolites, biosynthesis of amino acids, carbon metabolism, carbon fixation by the Calvin cycle, biosynthesis of various plant secondary metabolites, and cyanoamino acid metabolism (Figure S29). The full list of NAD-capped transcripts detected in least at two timepoints, including gene name, category, maximum baseMean, maximum log₂ fold change, and lowest adjusted p-values, is provided in Table S8. Transcript start site (TSS) mapping analysis revealed a sharp enrichment at the + 1 position for 46.9% of the positive samples for reads that map to the E. coli genome and 67.2% for reads that map to RNAI of the pUC19 plasmid. Four representative TSS profiles are shown in Figure S30. Abundant alternative start sites identified in Table S4 may, in part, contribute to positive samples displaying additional peaks. Furthermore, motif analysis consistently identified a − 12 to − 7 promoter element (TANNNT) in the upstream regions of NAD-capped transcripts in 90.6% of positive samples. Four representative motif analysis plots are presented in Figure S31. To quantify changes in NAD-capping across timepoints, read counts were normalized for sequencing depth and further adjusted using a set of stably expressed reference tRNAs which will be referred to as the “reference set” (Fig. 3 c). These genes exhibited uniform expression across timepoints, as shown by three representative species that remained stable in contrast to dynamic transcripts such as RNAI (Figure S32a). The resulting normalized read counts allowed accurate comparison of NAD-capped transcript abundance across the growth curve of E.coli . To validate the normalization strategy, two internal standards (IS-036 and IS-060) were selected based on their consistent capture at the + 1 TSS position across all four positive samples and timepoints. Two representative TSS plots for each internal standard are shown in Figure S33. Application of the two-step normalization strategy led to a marked reduction in the coefficient of variation (%CV) of their read counts across all 16 timepoints. The reduction in %CV indicates that the two-step normalization strategy effectively compensates for differences in read depth and minor expression changes, resulting in uniform capture of the two internal standards across timepoints. For IS-036, the %CV decreased from 106.67% to 43.46%, and for IS060, from 86.83% to 26.56%, with the most substantial improvement observed after the second normalization step (Figure S32b). These results support the validity of the normalization approach for quantifying fluctuations in NAD-capped RNA abundance throughout the E. coli growth curve. For a detailed description of the normalization steps see Methods section: “HELIOS NAD-Seq – Normalization in time course experiment” . RNAI, the most abundant NAD-capped RNAI encoded on the pUC19 plasmid, was used to evaluate the effectiveness of the normalization strategy. The resulting normalized RNAI read counts from HELIOS NAD-Seq closely mirrored the NAD-RNA abundance profile observed via northern blot on APB-PAGE, with probes 5′-end–labeled using γ-[³²P]-ATP (Fig. 3 d, Figure S34). Both ppp-RNAI and NAD-RNAI were highest at 0.5 hours post inoculation and lowest at 3.5 hours post inoculation. Capping ratios exhibited a similar pattern despite smaller differences between the two timepoints, with a roughly 3-fold reduction, as compared to the 7-fold reduction of ppp-RNAI and 22-fold reduction of NAD-RNAI (Figure S35). Both approaches revealed high RNAI levels at the 0.5 hour mark, a dramatic decrease during the early and late exponential phases, and a subsequent increase during the plateau phase. RNAI displayed expression patterns that did not fit any of the expression patterns of NAD-capped gene species that map to the endogenous E.coli genome (Fig. 3 e,f). To identify NAD-capped RNAs with dynamic expression profiles across the growth curve, gene species that give rise to potential NAD-capped RNAs in at least 2 of the 16 timepoints were selected (n = 257). Following the two-step normalization, mean z-scores were calculated for each gene across timepoints and visualized in a heatmap, where z-score represents the normalized expression level of a transcript, scaled relative to their mean and standard deviation, allowing comparison of dynamic patterns independent of absolute abundance (Fig. 3 e). The resulting heatmap revealed three distinct expression clusters. Cluster 1, comprising of 122 out of a total of 257 genes, contained genes with high NAD-capping during the stationary and early exponential phases, followed by a steep decline in the late exponential phase and a subsequent rise in the plateau phase. Cluster 1 revealed enrichment in 4 metabolic pathways including biosynthesis of secondary metabolites, ether lipid metabolism, folate biosynthesis, and biosynthesis of cofactors (Figure S36a). Cluster 2, comprising 94 genes, showed a sharp peak in the early exponential growth phase of E. coli followed by a similar pattern as observed in cluster 1. Cluster 2 revealed enrichment of 3 metabolic pathways (Figure S36b). Cluster 3, comprising 41 genes, displayed a biphasic pattern, with peaks during both the exponential and plateau phases. But with its low number of potential NAD-capped transcripts it displayed only tRNA biosynthesis as the enriched pathway (Figure S36c). The strong temporal peak at the early exponential phase for all 3 clusters may potentially align with a period of maximal NAD availability driven by high cell division( 52 ).These cluster-specific trends are further summarized in Fig. 3 f, where the average z-scores of all the genes within the cluster across timepoints are plotted for each cluster. A complete list of gene names, functional categories, KEGG pathways, and the cluster in which they were detected is provided in Table S9. Discussion In our recent review, we established a matrix to evaluate NAD-RNA identification protocols based on yield, specificity, evaluability, and throughput( 42 ). An ideal next-generation method should detect NAD-RNAs with minimal degradation and off-target effects, even from low-input samples, while enabling robust and scalable analysis adaptable to complex experimental settings. Additionally, a parallelization of protocol steps is needed to increase the throughput of biological samples and to allow the transition to new and more complex use cases. HELIOS NAD-Seq addresses these criteria and demonstrates significant improvements over existing methods. It maintains the versatility of NAD captureSeq while advancing the 4 benchmark criteria( 42 ). HELIOS NAD-Seq detects NAD-RNAs from at least 50-fold less total RNA compared to NAD captureSeq and NAD tagSeq( 1 , 39 ) with reduced degradation achieved through a 10-fold lower ADPRC concentration. Specificity is enhanced, eliminating off-target modifications of 5′-caps such as m 7 G and enabling the identification of hundreds of additional low-abundance NAD-capped transcripts. Throughput is further increased by early barcoding, which provides an 8-fold higher multiplexing capacity and reduces hands-on time. Evaluability is improved in qualitative terms, reflected by a more stringent adjusted p-value threshold (0.01 versus 0.05 for NAD captureSeq), though quantitative assessment remains limited, as HELIOS NAD-Seq cannot yet determine the proportion of the total transcript pool that is NAD-capped. A key factor behind its performance is the switch from O -nucleophilic hydroxyl to N -nucleophilic pyridyl substrates—specifically, the highly reactive 3PAB—mimicking the natural substrate of Aplysia californica ADP-ribosyl cyclase. This enables rapid, high-yield biotinylation of NAD + and NAD-capped RNAs via ADPRC transglycosylation, with full labeling achieved in about three minutes—a significant improvement over the 30-minute reaction time previously required with 4-pentynol and 3-azido-1-propanol ( 1 , 18 , 39 , 40 ). Importantly, the absence of divalent cations (e.g., Mg²⁺), which can promote RNA hydrolysis, further preserves RNA integrity. The reactions are highly specific for NAD caps, with no off-target labeling of RNAs bearing m 7 G caps, as confirmed by specificity assays (Fig. 2 a, S16). The conditions also allow for a 10-fold reduction in enzyme concentration, minimizing degradation even with low RNA inputs. Despite widespread use of ADPRC in NAD-RNA detection protocols, characterizations have mostly focused on dinucleotide reactions with NAD + , leaving a knowledge gap regarding its activity on NAD-capped oligonucleotides( 12 , 40 ). To address this, we developed in vitro assays to rigorously analyze the kinetics and specificity of ADPRC transglycosylation on NAD-capped RNAs. These revealed that, using 3PAB, a 70-fold lower substrate concentration compared to prior methods suffices, with total RNA inputs reduced by over 50-fold relative to NAD captureSeq( 12 , 18 , 39 – 41 ). In earlier protocols, negative controls were often performed without ADPRC( 12 ). We can now confirm that the treatment of negative controls with ADPRC and 3-picolylamine in HELIOS NAD-Seq better mimics the conditions in the sample reactions. Our approach with early pooling and barcoded adapters minimizes bias and enables multiplexing of up 128 samples, for 16 biological conditions (including replicates and controls) in parallel—an eightfold increase in throughput. In HELIOS NAD-Seq, hydrophilic streptavidin magnetic beads are used for capture and purification, offering lower nonspecific binding and milder washing conditions than previous methods ( 12 , 50 ). Our investigations showed the optimal balance between biotinylation efficiency and binding strength with the 3PAB substrate, which may also facilitate future direct NAD-RNA sequencing via nanopores. Downstream processing was adapted for low input, streamlining library preparation to be completed within four working days. Validation in E. coli confirmed our method's superior sensitivity, recovering nearly all previously identified NAD-capped transcripts and uncovering hundreds of new ones, often originating from alternative transcription start sites. By exploiting its multiplexing capacity, we captured NAD-capping dynamics across 16 timepoints of the E. coli growth curve, revealing distinct expression clusters with sharp peaks during early exponential growth and declines in later phases. Functional enrichment of these clusters identified many NAD-capped species to be related to metabolic pathways, underscoring its potential role in coordinating transcriptional output with metabolic state. The temporal pattern of Clusters 1 and 2 may suggest a link between NAD-capping and the metabolic state of the cell. The dramatic peak in the number of potential NAD-capped species may reflect elevated NAD + and NADH availability during active cell division, consistent with previous studies using dynamic single-cell fluorescence measurements( 52 ). The sharp drop in NAD-RNA levels after the early exponential phase may coincide with the onset of overflow metabolism( 53 – 55 )—a well-characterized shift in E. coli from aerobic respiration to fermentation during rapid growth. This phenomenon, analogous to the Warburg effect in cancer cells, is thought to arise from multiple factors, including limitations in the respiratory capacity of the inner membrane( 54 ), redirection of proteomic resources away from energy-efficient pathways( 53 ), and redox signaling through the NADH/NAD⁺ ratio( 55 ). Importantly, aerobic respiration coupled with the TCA cycle produces approximately 8 NADH molecules per glucose, which are efficiently converted back to 8 NAD⁺ via the electron transport chain. In contrast, fermentation yields only 2 NAD⁺ per glucose, which may drastically reduce the total available NAD⁺ pool (Fig. 3 b). Thus, the metabolic switch to fermentation may lead to a decline in NAD⁺ availability, potentially reducing the rate of NAD-capping during RNA transcription. This mechanistic link may explain the pronounced decrease in NAD-RNA abundance between 3.5 hours and 4 hours post inoculation (Fig. 3 A). The temporal expression pattern of RNAI (Fig. 3 d) are aligned to observations of decreased plasmid concentration at fast growth rates( 56 ). Jasiecki et al. also showed that RNAI degradation rates are lower at higher growth rates( 57 ) which could potentially explain why NAD-capping rates are lower during the early to late exponential growth phases of the E.coli . Conclusion In summary, HELIOS NAD-Seq provides a highly sensitive and specific platform for NAD-RNA detection, enabling robust profiling from minimal RNA input with reduced degradation and negligible off-target effects. By integrating early barcoding and optimized chemistry, the method achieves high multiplexing capacity and improved throughput, while uncovering both known and novel NAD-capped transcripts. Application to E. coli revealed dynamic capping patterns linked to growth phase and metabolic state, suggesting a connection between NAD + availability, overflow metabolism, and RNA regulation. These advances establish HELIOS NAD-Seq as a versatile tool for investigating the biological roles of NAD-capping and its contribution to cellular physiology. Methods Stock preparation PAB conjugates. Stock solutions (300 mM) of 3PAB, 4PAB, 3PADB, 4PADB, N(PEG) 3 B, and N(PEG) 11 B were prepared from pure solid, dissolved in DMSO and stored as 100 µL aliquots at − 20°C. ADPRC transglycosylation reactions on the dinucleotide level . Standard ADPRC transglycosylation reactions were performed in the presence of 5 mM NAD + (or 5 mM NADH), 15 mM 3PAB (or other N -nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 1.3 µM ADPRC enzyme (380 nM for product identification reactions). Buffer tests were performed in the presence of 1 mM NAD + , 1.5 mM 4PAB, 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), either with 5 mM MgCl 2 or 5 mM EDTA), 5% DMSO, and 1.3 µM ADPRC enzyme. The reactions were started by adding a freshly prepared enzyme dilution in H 2 O (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 30 min. If necessary, the reaction mixtures were diluted to 200 µL with H 2 O, before they were filtered through 10 kDa Amicon centrifugal filters and analyzed by RP-HPLC (gradient: 1–50% ACN in 50 min). For the buffer tests, 3 kDa Amicon centrifugal filters were used to prepare for RP-HPLC analysis (gradient: 2–5% buffer B in 20 min, then 15–30% in 10 min, then 30–70% in 20 min, total run time of 50 min). Collected peak solutions were analyzed by HR-MS. Kinetic Analysis of ADPRC transglycosylation reactions on the dinucleotide level . ADPRC transglycosylation reactions were performed in the presence of 5 mM NAD + , 15 mM 3PAB (or other N -nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 1.3 µM ADPRC enzyme. The reaction was started by adding a freshly prepared enzyme dilution in H 2 O (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 30 min. After predefined reaction times (8 s, 16 s, 25 s, 35 s, 45 s, 60 s, 90 s, 120 s, 180 s, 300 s), 30 µL aliquots were removed from the reaction mixture and quenched by addition to a mixture of 170 µL H 2 O and 400 µL P/C/I solution, followed by vigorous mixing on a vortex mixer. A sample termed »0 s« was removed before starting the reaction. Extractions with P/C/I and diethyl ether were performed, followed by removal of ether traces in an Eppendorf concentrator, before the samples were filtered through Amicon centrifugal filters (≤ 10 kDa) and analyzed by HPLC (gradient: 1–50% ACN in 50 min) in separate runs for each time-point. The areas of NAD + and product peaks in the chromatogram for 260 nm absorption were calculated using the ChemStation analysis software. StrepShift assay . For the electrophoretic mobility shift assay (EMSA) of biotin-modified RNA (usually after ADPRC transglycosylation) with streptavidin (streptavidin gel-shift assay), precipitated RNA samples containing biotinylated RNA species were standardly incubated in the presence of streptavidin in aqueous solution. In general, 6 µL of RNA sample (~ 600 ng RNA, for the kinetic analysis, for example, a mixture of around 570 ng E. coli total RNA and 30 ng model RNA) were incubated with 3 µL streptavidin (1 µg/µL) at 37°C for 15 min. For each analysis, a positive control with known biotinylation level was treated in addition to the experiment samples. Next, 7.5 µL of the RNA-streptavidin incubation mixture were added to 1.5 µL of agarose gel loading buffer (6x) and gel electrophoresis performed on a 2% agarose gel. For the analysis of the ADPRC transglycosylation reaction, the contained RNA species were additionally visualized using the Northern blotting technique. Northern blot . After agarose gel electrophoresis, a capillary transfer setup was used for transfer of nucleic acids onto a nylon membrane. For this, a small pedestal with a sheet of Whatman chromatography paper that reached into a reservoir of 0.5x TBE buffer and five more gel-sized sheets of Whatman chromatography paper (soaked in 0.5x TBE buffer) on top was prepared. On top of this, the agarose gel was placed, followed by a gel-sized nylon membrane (soaked in 0.5x TBE buffer), while carefully avoiding or removing air bubbles between gel and membrane. This was topped off with five more gel-sized sheets of Whatman chromatography paper (soaked in 0.5x TBE buffer), a stack of dry paper towels and a suitable weight. With this setup, capillary transfer was performed overnight at room temperature. After this, the nylon membrane was dried at air and transferred RNA species were photo cross-linked to the membrane surface in a UV chamber. The membrane was then blocked with ROTI Hybri-Quick in a hybridization tube (42°C, 1 h, rotating) before the hybridization solution was exchanged and typically 5 µL of a radioactively labeled, complementary RNA probe was added, which was followed by overnight incubation (42°C, rotating). The membrane was rinsed, washed, air-dried, and then placed on a storage phosphor screen, which was used for readout after incubation for between 6 h and 3 days at room temperature. ADPRC transglycosylation reactions on the RNA level . Standard ADPRC transglycosylation reactions were performed in the presence of 113 nt model RNA (~ 1:1 mixture of NAD/pppRNAIII leader), which was applied in a total amount of 5 µg (pure model RNA) or 0.5 µg (mixed with 4.5 µg E.coli total RNA), 15 mM 3PAB (or other N -nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 38 nM ADPRC enzyme, in a total reaction volume of 300 µL. Reactions assessing the potential of N(PEG) x B derivatives were performed accordingly, but in the presence of 10 µg model RNA and 380 nM ADPRC enzyme, in a total reaction volume of 100 µL. The reactions were started by adding a freshly prepared enzyme dilution in H 2 O (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 30 min. The reactions were quenched by addition of 300 µL P/C/I (optionally with additional 200 µL H 2 O for lower reaction volumes), or aliquots of 30 µL were removed from the reaction mixture and added to a mixture of 270 µL H 2 O and 300 µL P/C/I. Following extractions with P/C/I and diethyl ether and the removal of ether traces in an Eppendorf concentrator, the RNA samples were recovered by isopropanol precipitation and the nucleic acid concentrations were determined by UV-Vis spectroscopy, before a StrepShift assay was performed. Kinetic Analysis of ADPRC transglycosylation reactions on the RNA level . ADPRC transglycosylation reactions were performed in the presence of 2.5 ng/µL of a 113 nt model RNA (RNAIII leader, 47.9% NAD-modified), 47.5 ng/µL E. coli total RNA, 15 mM 3PAB (or other N -nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 38 nM ADPRC enzyme, in a total reaction volume of 300 µL. The reaction was started by adding 150 µL of a freshly prepared enzyme dilution in H 2 O (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 45 min. After predefined reaction times (0.5 min, 1 min, 2 min, 3 min, 5 min, 10 min, 20 min, 30 min, 45 min), 30 µL aliquots were removed from the reaction mixture and quenched by addition to a mixture of 270 µL H 2 O and 300 µL P/C/I solution, followed by vigorous mixing on a vortex mixer. A sample termed »0 s« was removed before starting the reaction. Extractions with P/C/I (three times) and diethyl ether (twice) were performed, followed by removal of ether traces in an Eppendorf concentrator. 3 M NaOAc (pH 5.5) to reach a final concentration of 0.3 M and 1 µL glycogen (RNA-grade) were added, and the RNA samples were recovered by isopropanol precipitation. The nucleic acid concentration of the recovered samples was determined by UV-Vis spectroscopy, before a StrepShift assay was performed. The contained RNA species were transferred to nylon membranes and probed with a radioactively-labeled RNAIII leader probe using the Northern blotting technique. Signal intensities recorded for shifted and non-shifted RNAIII leader species at each time-point of the reaction were analyzed and plotted. ADPRC transglycosylation specificity tests on the RNA level . For the treatment of differently capped RNA species in an ADPRC transglycosylation setup, several IS-036 model RNA stocks with defined 5’-end structures and concentrations of 200 ng/µL were prepared: a) 5’-pppA, b) 1:1 5’-GpppA and 5’-pppA, c) 1:1 5’-m 7 GpppA and 5’-pppA, d) 1:1 5’-NppA (NAD) and 5’-pppA, e) 1:19 5’-NppA (NAD) and 5’-pppA. From these stocks, 50 µL solutions containing 5 µg IS-036 model RNA, 30 mM 3PAB, and 2x ADPRC buffer (with EDTA) were prepared, from which two samples à 5 µL were removed and diluted with 5 µL H 2 O each. One of those samples was stored as a negative control, the other was incubated at 37°C for 30 min without addition of enzyme (»0 s« sample). To the residual volume of 40 µL (for each of the RNA mixes), an equal volume of ADPRC enzyme dilution (76 nM) was added to start the reaction. After pre-defined reaction times (0.5 min, 1 min, 2 min, 3 min, 5 min, 10 min, 20 min, 30 min), 10 µL aliquots were removed from the reaction mixture and quenched by addition to a mixture of 290 µL H 2 O and 300 µL P/C/I solution, followed by vigorous mixing on a vortex mixer. Extractions with P/C/I (three times) and diethyl ether (twice) were performed, followed by removal of ether traces in an Eppendorf concentrator. 3 M NaOAc (pH 5.5) to reach a final concentration of 0.3 M was added, and the RNA samples were recovered by isopropanol precipitation (due to the low RNA amount of around 500 ng per sample, the vessels were stored at − 20°C for 3 days). The nucleic acid concentration of the recovered samples was determined by UV-Vis spectroscopy and fluorescence spectroscopy. Then, dot blot experiments were performed in technical quadruplicates, using 1 µL (~ 50 ng IS-036 model RNA) per sample and time-point, and probing with streptavidin-HRP conjugate. Capture and Elution of Biotinylated RNA using Streptavidin Magnetic Beads . The preparation of biotinylated RNA species for this experiment was performed by ADPRC transglycosylation of 20 µg of a 113 nt model RNA (~ 1:1 mixture of NAD/pppRNAIII leader) under standard reaction conditions in the presence of 15 mM 3PAB (or other N -nucleophilic substrate) and 380 nM ADPRC enzyme, in a total reaction volume of 200 µL. The reaction was quenched and extracted after 5 min of incubation at 37°C following the standard protocol. After RNA sample recovery, the biotinylation levels were determined using a StrepShift assay. For the capture on streptavidin magnetic beads, 10 ng of the partially biotinylated model RNAs were mixed with 390 ng E. coli total RNA. Hydrophilic streptavidin magnetic beads were prepared according to the manufacturer’s instructions. 25 µL of bead suspension each were transferred to PCR tubes in an 8-tube stripe. Using a magnetic stand, the bead storage solution was removed, and the beads were washed three times with 100 µL of 1x immobilization buffer, before they were each blocked by incubation at 20°C for 30 min with a solution of 100 µg/mL acetylated BSA in 1x immobilization buffer, followed by three further washes with 1x immobilization buffer. In general, upon the addition of a new solution to the beads, the suspension was mixed by pipetting up and down five times. Then, the prepared mixes of model RNA and total RNA, with a total volume of 25 µL each, were added to the beads and incubated at 20°C for 1 h. The supernatant solution, as well as the solutions of three washes with 100 µL streptavidin wash buffer (containing 8 M urea, or 1 M urea for some PADB-reacted samples) and one wash with 100 µL H 2 O, were collected. The elution of bound RNA was performed first by incubation of the beads with 25 µL buffered solution (1 mM EDTA, 10 mM Tris-HCl, pH 7.5, optionally containing 0.4 mM biotin) at different temperatures of (40°C, 60°C, or 80°C) for 10 min. Optionally, this was followed by an elution step with 25 µL buffered biotin solution at 80°C for 10 min in order to fully remove bound biotinylated RNA. The solutions of all elution steps were collected. The contained RNA was analyzed by dot blot experiments that, where applicable, were performed in technical triplicates, using 1 µL per collected sample (1.5 µL for elution samples in the elution condition optimization experiment), and probing with radioactively labeled RNA probes complementary to either RNAIII leader, or E. coli ribosomal RNAs (5S, 16S, 23S). Dot blot analysis . Typically, 1 µL aliquots of RNA-containing, aqueous solutions were spotted onto a nylon membrane. After drying at air, RNA species were photo-crosslinked to the membrane surface in a UV chamber. Probing with radioactively labeled, complementary RNA probes was performed following a Northern blotting technique. Probing with streptavidin-HRP conjugate was performed according to the manufacturer’s instructions, after blocking the nylon membrane with 3% BSA fraction V in 1x TBS-T. After removing the blocking solution, a labeling solution of 1x TBS-T containing 42 ng/mL streptavidin-HRP conjugate was added, followed by incubation for 30 min (room temperature, rotating). After rinsing, washing and air drying, the membrane was stained with SignalFire ECL Plus reagent according to the manufacturer’s instructions. After readout at a ChemiDoc Imager device, signal intensities were determined using the ImageLab software. HELIOS NAD-Seq – standard protocol . The protocol is standardly performed in a total of 4 working days. Day 1 : HELIOS NAD-Seq typically starts from total RNA samples in quadruplicates, which have been extracted from an organism of choice, treated by DNase I, and recovered by ethanol or isopropanol precipitation. For each replicate (numbered as 1, 2, 3, 4, et cetera ), a 20.5 µL total RNA mix containing 4.1 µg of total RNA and a desired amount of internal RNA standards (e.g. 1 fmol) is prepared. For the initial ADPRC transglycosylation with 3PAB, a diluted ADPRC stock is prepared, containing 5 mM HEPES (pH 7.0), 0.5 mM MgCl 2 , 5% glycerol, and 380 nM ADPRC enzyme (100 µL suffice for 5 replicate samples plus negative controls). Then, for each replicate, a 50 µL sample (S) mix containing 10 µL total RNA mix, 30 mM 3PAB, and reagents (2x ADPRC buffer with EDTA, 10% DMSO), as well as a 50 µL negative control (N) mix containing 10 µL total RNA mix, 30 mM 3PA, and reagents (2x ADPRC buffer with EDTA, 10% DMSO) are prepared. With the descriptors for positive samples (S) and negative controls (N), and the numbering of the replicates as 1, 2, 3, 4, et cetera , a set of mixes termed S1, S2, S3, S4, et cetera , and N1, N2, N3, N4, et cetera , is obtained. The vessels containing S and N mixes (50 µL), as well as an equal number of vessels containing 40 µL H 2 O, are pre-heated to 37°C. Then, 10 µL of the ADPRC dilution are mixed with a pre-heated water aliquot (40 µL), and the full 50 µL are added to the pre-heated reaction mix S1 and mixed by pipetting, before further incubation at 37°C. This procedure is repeated for reaction mix S2 after either 30 s or 1 min. With this, after either 3:30 min or 7:00 min, eight ADPRC transglycosylation reactions (for S1, S2, S3, S4, N1, N2, N3, and N4) have been started, all either 30 s or 1 min apart. After 20 min of incubation at 37°C, a mixture of 200 µL H2O and 300 µL P/C/I solution is added to quench the S1 reaction, followed directly by vigorous mixing using a vortex mixer. This is again repeated for the other ADPRC transglycosylation reaction mixes in the same sequence as performed for starting the reactions, with either 30 s or 1 min in between. Aqueous and organic phases are separated by centrifugation (13000 rpm, 2 min, room temperature) and the aqueous phase is treated by two more extractions with P/C/I (300 µL each), and, after transferring the aqueous phase to a 1.5 mL LoBind tube, followed by two extractions with diethyl ether (300 µL each). After discarding the organic phase after the last extraction and centrifugation step, residual traces of ether are removed in an Eppendorf concentrator (~ 3 min, room temperature). Around 270 µL of aqueous phase should typically remain for each replicate sample (S1, S2, S3, S4) and negative control (N1, N2, N3, N4), which are then each mixed with 30 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, with 1 volume (300 µL) of cold isopropanol (− 20°C). The ADPRC-reacted RNA samples are precipitated at − 20°C for 2 h, and recovered by centrifugation (14900 g, 90 min, 4°C). The formed pellets are washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before they each are dissolved in 10 µL H 2 O. For 3’-adapter ligation with barcoded adapters, 75 µL of a 2.4x ligation master mix (2.4x standard ligation buffer, 120 µg/mL acetylated BSA, 12 mM DTT, 24% DMSO, 2 U/µL RNaseOUT, 0.6 U/µL RNA ligase 1, 12 U/µL RNA ligase 2 (truncated, K227Q)) are prepared for the treatment of eight samples. For eight samples, e.g. AD3-id4-bc01 through -bc08 can be used. To 1 µL of each required adenylated 3’-adapter (~ 100 µM total concentration, and 50 µM pre-adenylated) are added 0.7 µL H 2 O and 8.5 µL of the 2.4x ligation master mix. From the resulting 2x adapter/ligation master mixes, 10 µL are added to the 10 µL of ADPRC-reacted RNA samples (e.g. bc01 for S1, bc02 for S2, bc03 for S3, bc04 for S4, bc05 for N1, bc06 for N2, bc07 for N3, and bc08 for N4), which are incubated at 4°C overnight. Day 2 : Hydrophilic streptavidin magnetic beads are removed from the storage at 4°C and allowed to incubate at room temperature for around 30–45 min, after which the container is carefully agitated to create a homogeneous suspension. 25 µL of magnetic beads per PCR tube are transferred using a 200 µL pipette tip that was cut off with a scalpel to increase the diameter of the opening. Then, 25 µL of 1x immobilization buffer is added to each PCR tube, the beads are suspended by pipetting up and down for five times, and the tubes are placed on a magnetic stand. After a pellet of beads has formed, the supernatant is removed carefully. The described procedure is performed for every addition and removal of a new solution, optionally with an additional incubation time after mixing and before pelleting on a magnetic stand. The beads are washed two additional times with 1x immobilization buffer (50 µL each), before they are incubated at 20°C for 5 min with 25 µL of a bead blocking solution containing 1x immobilization buffer, and 100 µg/mL acetylated BSA. After this incubation, the beads are washed two times with 1x immobilization buffer (50 µL each) before addition of the ADPRC-reacted, 3’-ligated RNA samples. After overnight incubation of the 3’-ligation reaction mixes (S1, S2, S3, S4, N1, N2, N3, N4) with the respective barcoded adapters, the reactions are heat-inactivated at 65°C for 15 min, an equal volume (20 µL) of 2x immobilization buffer is added to each vessel, and the different solutions are combined. For eight samples, a total volume of 320 µL multiplexed solution is obtained, which is mixed and divided equally on the PCR tubes containing BSA-blocked, washed streptavidin magnetic beads. It is recommended to not use more than 180 µL of RNA-containing solution per 25 µL of initially used bead suspension. After incubation at 20°C for 5 min to allow for the binding of biotinylated RNA species to the beads, the supernatant is removed, and 50 µL of 1x immobilization buffer is added. At this point, it is advised to combine the suspensions of two PCR tubes into one. By doing this, the RNA-containing solution of eight samples, which was split before, is unified again after the bead-binding step. The beads are washed once with 1x high salt wash buffer and once with 1x low salt wash buffer (100 µL each). The on-bead reverse transcription is performed by adding 20 µL of RT master mix (5 µM of RTp2 primer, 0.5 mM of each dNTP (dATP, dCTP, dGTP, dTTP), 1x SSIV buffer, 5 mM DTT, 2 U/µL RNaseOUT, and 10 U/µL SuperScript IV Reverse Transcriptase) to the magnetic beads, which, after suspending, is followed by incubation at 55°C for 10 min. After adding an equal volume (20 µL) of 2x immobilization buffer, the bead suspension is further incubated at 20°C for 5 min to allow for the re-binding of biotinylated RNA species in a hybrid with complementary cDNA. The beads are washed once with 1x immobilization buffer, once with 1x high salt wash buffer and once with 1x low salt wash buffer (100 µL each). To hydrolyze the RNA and release the cDNA, a 50 µL solution of 0.15 M NaOH is added, followed by incubation at 55°C for 15 min. After incubation, the supernatant is transferred to a 1.5 mL LoBind tube. The beads are washed once with 50 µL H 2 O and the supernatant is also transferred to the same 1.5 mL LoBind tube. The full volume of 100 µL is then mixed with 11 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, with 2.5 volumes (280 µL) of cold absolute ethanol (− 20°C). The multiplexed cDNA sample is precipitated at − 20°C for 2 h, and recovered by centrifugation (14900 g, 60 min, 4°C). The formed pellet is washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before it is dissolved in 10 µL H 2 O, and transferred to a PCR tube. The C-tailing of the cDNA is performed by addition of an equal volume (10 µL) of a TdT master mix (1 mM CTP, 2x TdT buffer, 1 U/µL TdT enzyme), followed by incubation at 37°C for 20 min. Then, the reaction is heat-inactivated at 70°C for 10 min. The 2nd adapter ligation reaction is performed by addition of an equal volume (20 µL) of a 2nd ligation master mix (10 µM sense cDNA anchor, 10 µM antisense cDNA anchor, 20 µM ATP, 2x T4 DNA ligase buffer, 3 Weiss U/µL T4 DNA ligase), followed by overnight incubation at 4°C. Day 3 : After overnight incubation of the 2nd ligation reaction mix, the volume of 40 µL is diluted to 100 µL with H 2 O, followed by heat-inactivation at 65°C for 10 min, before transfer to a 1.5 mL LoBind tube. The full volume of 100 µL is then mixed with 11 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, with 2.5 volumes (280 µL) of cold absolute ethanol (− 20°C). The multiplexed, fully ligated cDNA sample is precipitated at − 20°C for 2 h, and recovered by centrifugation (14900 g, 60 min, 4°C). The formed pellet is washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before it is dissolved in 20 µL H 2 O. The multiplexed, fully ligated cDNA is PCR-amplified in a 50 µL solution containing 2–8 µL cDNA sample, 2 µM NEBNext Universal Primer, 2 µM NEBNext Index Primer, and 20 µL of a Q5 master mix (0.5 mM of each dNTP (dATP, dCTP, dGTP, dTTP), 2.5x Q5 reaction buffer, 0.05 U/µL Q5 HotStart High-Fidelity DNA Polymerase), with a standard PCR amplification protocol: After initial activation (98°C, 40 s), a sequence of denaturation (98°C, 10 s), annealing (69°C, 20 s), and extension (72°C, 30 s) is performed 9 times, before a sequence of denaturation (98°C, 10 s), annealing (66°C, 20 s), and extension (72°C, 30 s) is performed up to 16 times, followed by a final extension (72°C, 10 min). After PCR amplification, the solution is transferred to a 1.5 mL LoBind tube. The full volume of 50 µL is then mixed with 6 µL NaOAc (3 M, pH 5.5), and, after short incubation at room temperature, with 2.5 volumes (140 µL) of cold absolute ethanol (− 20°C). The DNA sample is precipitated at − 20°C for 30–60 min, and recovered by centrifugation (14900 g, 60 min, 4°C). The formed pellet is washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before it is dissolved in 15 µL H 2 O. For native PAGE purification, an equal volume (15 µL) of 2x native PAGE loading buffer (with bromophenol blue dye) is added to the PCR-amplified DNA samples. This sample is loaded on a freshly prepared native PA gel (50 mL of gel solution are prepared from 16.7 mL Rotiphorese 30% sequencing gel concentrate (29:1), 5 mL TBE buffer (10x), and 28.3 mL H 2 O, and polymerization is set off using 500 µL APS (10%) and 50 µL TEMED), flanked on each side by 1 µL of GeneRuler UltraLow Range DNA ladder, and the native PAGE is run in 1x TBE buffer at 17 W for roughly 2.5 h (typically stopped slightly before the bromophenol blue dye runs out). The gel is removed, submerged in around 300 mL 1x TBE buffer containing 5 µL SYBR Gold, and incubated like this for 5 min. Then, the gel is shortly washed in water, placed between two sheets of transparent plastic tubing, and analyzed on a Typhoon FLA 9500 biomolecular imager (PMT voltage typically 500 V) or another suitable device. The image is printed out according to its actual size, placed under the plastic tubing to fit the gel, and the area between 150 bp and 300 bp (marked by two bands in the DNA ladder) is excised using a clean scalpel. The excised gel pieces are crushed using a suitable method, and mixed with 0.3 M NaOAc solution (pH 5.5) for overnight elution at 20°C (750 rpm shaking). Day 4 : The gel pieces are removed from the overnight elution mix using a 50 mL filter falcon tube and centrifugation (5000 g, 5 min, 20°C). Depending on the total volume, the flow-through is divided equally into 5 mL tubes, to each of which are added 10 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, 2.5 volumes of cold absolute ethanol (− 20°C). The multiplexed DNA libraries are precipitated at − 20°C for 1 h, and recovered by centrifugation (15000 g, 60 min, 4°C). The formed pellets are washed twice with 70% EtOH (150 µL each), followed by centrifugation (15000 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before they are typically dissolved in a total of 20–30 µL H 2 O, and combined in a single 1.5 mL LoBind tube. The size selected DNA libraries are analyzed using a Qubit spectrofluorometer and a suitable kit according to the manufacturer’s instructions, in order to determine the concentration (in ng/µL) of the stock solution, and using a Bioanalyzer device, a suitable kit and electrophoresis chip according to the manufacturer’s instructions, in order to determine the average size (in bp) of the multiplexed libraries. Both obtained values are further used to determine the stock concentration (in nM), for example with the Promega Biomath Calculator web tool. The library is then typically adjusted to a volume of 20 µL with a DNA concentration of 10 nM (depending on the specific requirements for NGS), and ready for analysis by next-generation sequencing at a sequencing facility of choice. HELIOS NAD-Seq – Bioinformatic analysis Raw paired-end reads were demultiplexed using Cutadapt (vX.X) with custom barcode sequences corresponding to samples bc01–bc08, using degenerate barcode patterns (e.g., TCAAGTNNNNNNG) for both Read 1 and Read 2 with up to eight random nucleotides (Figure S17a). Demultiplexing was performed by assigning reads to barcodes based on a 5’ overlap with a minimum overlap of 12 nt (-O 12), applying the -g and -A options to match barcode–adapter sequences on forward and reverse reads, respectively. To remove 5’-end leading Gn introduced during NAD capture adapter ligation, an in-house python script was used to trim G-rich patterns from the 5′ ends of forward reads via regular expressions (e.g., ^[GN]*), discarding reads shorter than 18 nt after trimming. Further quality and adapter trimming was performed with Trimmomatic (v0.39) in paired-end mode using the parameters ILLUMINACLIP:TruSeq3-PE.fa:1:30:15, SLIDINGWINDOW:4:20, and MINLEN:18, retaining both paired and unpaired reads. Processed reads were then subjected to a three-stage hierarchical alignment pipeline using Bowtie2 (v2.4.5): first, alignment to a custom spike-in RNA reference (spikeRnas) in end-to-endmode, retaining only unaligned reads via --un-conc and --un; second, alignment of spike-in-depleted reads to an E. coli tRNA + rRNA reference index in local mode, again retaining only unaligned reads; and third, alignment of remaining reads to the E. coli K-12 genome (NC_00913.3 including pUC19c plasmid) in local mode, generating SAM files for paired-end, R1 singleton, and R2 singleton reads. SAM files from the genome alignment step were then filtered with a custom python script to retain only alignments whose read sequence began with adenosine (A) at the 5’ end (the first base of the SEQ field in the SAM file). To quantify transcription from intergenic regions, a custom GTF annotation file was generated from the E. coli reference annotation using an in-house python script that identified genomic intervals between annotated genes on the same strand and chromosome, emitting each as a feature of type "intergenic". Read counts were obtained using featureCounts (v2.0.6, Subread package) with the -t intergenic option and the custom GTF file as the annotation reference; paired-end files were processed with the -p flag to ensure fragment-level counting, whereas single-end files were processed without it. Differential expression analysis between positive NAD-capture samples and matched negative controls was then performed separately for each timepoint using PyDESeq2 (v0.4.10) to identify transcripts significantly enriched in the positive fraction with an adjusted p value of below 0.01 and Log2 fold-change (sample/control) of above 3, which were considered NAD-capped candidates. To assess transcription start site (TSS) positions of NAD-capped RNAs, read start distributions were computed from SAM alignment files. For reads mapping to the E. coli genome and puc19C plasmid, a custom python script parsed gene intervals from the reference GTF and counted reads based on their position relative to the annotated TSS, stratified by strand. Reads were binned by 1-based distance from the TSS defined by the genome assembly ASM584v2 (up to 30 bp both upstream and downstream), and the resulting distributions were visualized as bar plots per barcode and timepoint. For internal standard spike-in controls, a separate script was used to calculate the read start position (TSS) per reference sequence without gene annotations. Positions were aggregated across all reads in each timepoint and barcode, and TSS distributions were plotted individually for each spike-in reference. To visualize nucleotide preferences around transcription start sites (TSS), sequence logos were generated using a custom Python script in combination with WebLogo (v3). Filtered reads were exported from SAM files to FASTA format, with each header containing the alignment count and genomic location. Sequences were grouped by chromosome and replicated according to their alignment counts to reflect read abundance. Each sequence was then trimmed or padded to a uniform window of 41 nucleotides, corresponding to positions − 20 to + 20 relative to the TSS. For each chromosome in each sample, a weighted FASTA file was created and used as input for WebLogo. HELIOS NAD-Seq – Normalization in time course experiment To quantify read coverage at the 3′ ends of genes, a custom GTF file was generated from the E. coli reference annotation (GCF_000005845.2_ASM584v2) by extracting gene features and defining variable 3′ windows based on gene length and biotype. For tRNA, rRNA, and genes shorter than 100 bp, the full gene body was retained. For genes 100–200 bp, a region starting 50 bp downstream of the TSS to the gene end was used, and for genes longer than 200 bp, the region began 100 bp downstream of the TSS. Coordinates were adjusted strand-specifically. This modified GTF was then used with featureCounts (subread v2.0.6) to quantify reads from aligned SAM files, with paired-end and single-end reads processed separately. Following read counting, PyDESeq2 (v0.4.10) was used to identify stably expressed genes across timepoints, selecting genes with log2 fold change between − 1 and + 1 and a baseMean ≥ 100 in at least 10 out of 16 timepoints. This filtering yielded 21 consistently expressed tRNA genes. Their counts were normalized by the number of assigned reads to the E. coli genome and used to construct 10 representative sampling sets by binning genes into expression-based strata and scaling mid- and low-expression genes to match the high-expression bin. These sampling sets were used to derive timepoint-specific normalization factors relative to timepoint 1, which were then applied to normalize NAD-capture read counts in each timepoint across all replicates. For downstream expression pattern analysis, genes identified as NAD-capped in at least 3 of the 16 timepoints were clustered using dynamic time warping (DTW). Log-transformed and z-scored expression values (aggregated by mean across four replicates) were input into TimeSeriesKMeans (tslearn, DTW metric, k = 3), and clustering results were visualized using heatmaps and average trajectory plots. Genes not detected in at least two timepoints were excluded prior to clustering. Declarations Ethics approval and consent to participate: Not applicable Consent for publication: Not applicable Availability of data and materials: Raw sequencing data have been deposited in the NCBI Sequence Read Archive (SRA), bioproject # PRJNA1327118, SRA # SRR35337667 - SRR35337683. Competing interests: The authors declare that they have no competing interests. Funding: The project has received funding from the European Research Council under the European Union’s Horizon 2020 research and innovation program (Grant 882789 RNACoenzyme) and from the German Research Council (DFG; Project 439669440, TRR319, Subproject A02). MM was supported by a doctoral scholarship of the German Academic Scholarship Foundation. Authors contributions: AJ, MM, and HGMF initially devised the project, with critical input from FP. MM, TWS, FW, HGMF, MW, ET, EM, AC conducted experiments, MM, TWS and YZ analyzed and visualized data, AJ procured funding and supervised the project, MM and TWS wrote the initial manuscript draft, all authors participated in manuscript editing. Acknowledgements: We thank the Brunner lab (BZH, Heidelberg University) for access to the qPCR machine, Heiko Rudy and Tobias Timmermann for mass spectrometry and NMR measurements, Daniel Wolf for assistance in synthesis, Drs. Christian Löcherer, Jens Frindert, Christian Schmitt, and Franziska Grün for providing materials or methods, and Drs. Martin Gärtner and Bastian Bühler (all IPMB, Heidelberg University) for valuable discussions. Author information: MM and TWS contributed equally to this work. References Cahová H, Winz M-L, Höfer K, Nübel G, Jäschke A. NAD captureSeq indicates NAD as a bacterial cap for a subset of regulatory RNAs. Nature. 2015;519(7543):374. Bird JG, Zhang Y, Tian Y, Panova N, Barvík I, Greene L, et al. The mechanism of RNA 5′ capping with NAD+, NADH and desphospho-CoA. Nature. 2016;535(7612):444. Malygin AG, Shemyakin MF. Adenosine, NAD and FAD can initiate template-dependent RNA synthesis catalyzed by Escherichia coli RNA polymerase. FEBS Letters. 1979;102(1):51-4. Huang F. Efficient incorporation of CoA, NAD and FAD into RNA by in vitro transcription. Nucleic Acids Res. 2003;31(3):e8. Möhler M, Höfer K, Jäschke A. Synthesis of 5′-Thiamine-Capped RNA. Molecules. 2020;25(23):5492. Depaix A, Grudzien-Nogalska E, Fedorczyk B, Kiledjian M, Jemielity J, Kowalska J. Preparation of RNAs with non-canonical 5′ ends using novel di- and trinucleotide reagents for co-transcriptional capping. Frontiers in Molecular Biosciences. 2022;9:854170. Weber F, Motzkus NA, Brandl L, Möhler M, Alempijevic A, Jäschke A. Identification and in vitro characterization of UDP-GlcNAc-RNA cap-modifying and decapping enzymes. Nucleic Acids Res. 2024;52(10):5438-5450. doi:10.1093/nar/gkae353. Mlynarska-Cieslak A, Depaix A, Grudzien-Nogalska E, Sikorski PJ, Warminski M, Kiledjian M, et al. Nicotinamide-Containing Di- and Trinucleotides as Chemical Tools for Studies of NAD-Capped RNAs. Organic Letters. 2018;20(23):7650-5. Chen YG, Kowtoniuk WE, Agarwal I, Shen Y, Liu DR. LC/MS analysis of cellular RNA reveals NAD-linked RNA. Nature chemical biology. 2009;5(12):879. Grudzien-Nogalska E, Bird JG, Nickels BE, Kiledjian M. “NAD-capQ” detection and quantitation of NAD caps. RNA. 2018;24(10):1418-25. Wang J, Alvin Chew BL, Lai Y, Dong H, Xu L, Balamkundu S, et al. Quantifying the RNA cap epitranscriptome reveals novel caps in cellular and viral RNA. Nucleic Acids Res. 2019;47(20):e130. Winz M-L, Cahová H, Nübel G, Frindert J, Höfer K, Jäschke A. Capture and sequencing of NAD-capped RNA sequences with NAD captureSeq. Nature protocols. 2017;12(1):122. Frindert J, Zhang Y, Nübel G, Kahloon M, Kolmar L, Hotz-Wagenblatt A, et al. Identification, biosynthesis, and decapping of NAD-capped RNAs in B. subtilis. Cell reports. 2018;24(7):1890-901. Morales-Filloy HG, Zhang Y, Nübel G, George SE, Korn N, Wolz C, et al. The 5′ NAD Cap of RNAIII Modulates Toxin Production in Staphylococcus aureus Isolates. Journal of Bacteriology. 2020;202(6):e00591-19. Ruiz-Larrabeiti O, Benoni R, Zemlianski V, Hanišáková N, Schwarz M, Brezovská B, et al. NAD+ capping of RNA in Archaea and Mycobacteria. bioRxiv. 2021:2021.12.14.472595. Walters RW, Matheny T, Mizoue LS, Rao BS, Muhlrad D, Parker R. Identification of NAD+ capped mRNAs in Saccharomyces cerevisiae. Proc Natl Acad Sci USA. 2017;114(3):480-5. Zhang Y, Kuster D, Schmidt T, Kirrmaier D, Nübel G, Ibberson D, et al. Extensive 5′-surveillance guards against non-canonical NAD-caps of nuclear mRNAs in yeast. Nat Commun. 2020;11(1):5508. Zhang H, Zhong H, Zhang S, Shao X, Ni M, Cai Z, et al. NAD tagSeq reveals that NAD+-capped RNAs are mostly produced from a large number of protein-coding genes in Arabidopsis. Proc Natl Acad Sci USA. 2019;116(24):12072-7. Wang Y, Li S, Zhao Y, You C, Le B, Gong Z, et al. NAD+-capped RNAs are widespread in the Arabidopsis transcriptome and can probably be translated. Proc Natl Acad Sci USA. 2019;116(24):12094-102. Dong H, Wang X, Tan C, Gao L, Cui J, Liu L, et al. NAD+-capped RNAs are widespread in rice (Oryza sativa) and spatiotemporally modulated during development. Science China Life Sciences. 2022;65(10):2121-4. Niu K, Zhang J, Ge S, Li D, Sun K, You Y, et al. ONE-seq: epitranscriptome and gene-specific profiling of NAD-capped RNA. Nucleic Acids Res. 2023;51(2):e12. Ge S, Wang X, Wang Y, Dong M, Li D, Niu K, et al. Hidden features of NAD-RNA epitranscriptome in Drosophila life cycle. iScience. 2024;27(1):108618. Jiao X, Doamekpor SK, Bird JG, Nickels BE, Tong L, Hart RP, et al. 5′ end nicotinamide adenine dinucleotide cap in human cells promotes RNA decay through DXO-mediated deNADding. Cell. 2017;168(6):1015-27. e10. Höfer K, Li S, Abele F, Frindert J, Schlotthauer J, Grawenhoff J, et al. Structure and function of the bacterial decapping enzyme NudC. Nature chemical biology. 2016;12(9):730. Grudzien-Nogalska E, Wu Y, Jiao X, Cui H, Mateyak MK, Hart RP, et al. Structural and mechanistic basis of mammalian Nudt12 RNA deNADding. Nature Chemical Biology. 2019;15(6):575-82. Sharma S, Grudzien-Nogalska E, Hamilton K, Jiao X, Yang J, Tong L, et al. Mammalian Nudix proteins cleave nucleotide metabolite caps on RNAs. Nucleic Acids Res. 2020;48(12):6788-98. Pan S, Li K-e, Huang W, Zhong H, Wu H, Wang Y, et al. Arabidopsis DXO1 possesses deNADding and exonuclease activities and its mutation affects defense-related and photosynthetic gene expression. Journal of Integrative Plant Biology. 2020;62(7):967-83. Abele F, Höfer K, Bernhard P, Grawenhoff J, Seidel M, Krause A, et al. A Novel NAD-RNA Decapping Pathway Discovered by Synthetic Light-Up NAD-RNAs. Biomolecules. 2020;10(4):513. Sharma S, Yang J, Grudzien-Nogalska E, Shivas J, Kwan KY, Kiledjian M. Xrn1 is a deNADding enzyme modulating mitochondrial NAD-capped RNA. Nat Commun. 2022;13(1):889. Wolfram-Schauerte M, Moskalchuk A, Pozhydaieva N, Ramírez Rojas AA, Schindler D, Kaiser S, et al. T4 phage RNA is NAD-capped and alters the NAD-cap epitranscriptome of Escherichia coli during infection through a phage-encoded decapping enzyme. bioRxiv. 2024. Julius C, Yuzenkova Y. Bacterial RNA polymerase caps RNA with various cofactors and cell wall precursors. Nucleic Acids Res. 2017;45(14):8282-90. Bird JG, Basu U, Kuster D, Ramachandran A, Grudzien-Nogalska E, Towheed A, et al. Highly efficient 5' capping of mitochondrial RNA with NAD+ and NADH by yeast and human mitochondrial RNA polymerase. eLife. 2018;7:e42179. Sherwood AV, Rivera-Rangel LR, Ryberg LA, Larsen HS, Anker KM, Costa R, et al. Hepatitis C virus RNA is 5'-capped with flavin adenine dinucleotide. Nature. 2023;619(7971):811-8. Benoni B, Benoni R, Trylcova J, Grab K, Pačes J, Weber J, et al. HIV-1 infection reduces NAD capping of host cell snRNA and snoRNA. bioRxiv. 2022:2022.11.10.515957. Wolfram-Schauerte M, Höfer K. NAD-capped RNAs – a redox cofactor meets RNA. Trends in Biochemical Sciences. 2023;48(2):142-55. Wolfram-Schauerte M, Pozhydaieva N, Grawenhoff J, Welp LM, Silbern I, Wulf A, et al. A viral ADP-ribosyltransferase attaches RNA chains to host proteins. Nature. 2023;620(7976):1054-62. Lee HC, Aarhus R. ADP-ribosyl cyclase: an enzyme that cyclizes NAD+ into a calcium-mobilizing metabolite. Cell Regulation. 1991;2(3):203-9. Hellmich MR, Strumwasser F. Purification and characterization of a molluscan egg-specific NADase, a second-messenger enzyme. Cell Regulation. 1991;2(3):193-202. Zhang H, Zhong H, Wang X, Zhang S, Shao X, Hu H, et al. Use of NAD tagSeq II to identify growth phase-dependent alterations in E. coli RNA NAD+ capping. Proc Natl Acad Sci USA. 2021;118(14):e2026183118. Hu H, Flynn N, Zhang H, You C, Hang R, Wang X, et al. SPAAC-NAD-seq, a sensitive and accurate method to profile NAD+-capped transcripts. Proc Natl Acad Sci USA. 2021;118(13):e2025595118. Sharma S, Yang J, Favate J, Shah P, Kiledjian M. NADcapPro and circNC: methods for accurate profiling of NAD and non-canonical RNA caps in eukaryotes. Commun Biol. 2023;6(1):406. Möhler M, Jäschke A. Future Perspectives for the Identification and Sequencing of Nicotinamide Adenine Dinucleotide-Capped RNAs. Accounts of Chemical Research. 2023;56(21):3000-9. Cohen MS, Chang P. Insights into the biogenesis, function, and regulation of ADP-ribosylation. Nat Chem Biol. 2018;14(3):236-243. doi:10.1038/nchembio.2568 Lee HC, Aarhus R. Structural Determinants of Nicotinic Acid Adenine Dinucleotide Phosphate Important for Its Calcium-mobilizing Activity. Journal of Biological Chemistry. 1997;272(33):20378-83. Preugschat F, Tomberlin GH, Porter DJ. The base exchange reaction of NAD+ glycohydrolase: identification of novel heterocyclic alternative substrates. Archives of biochemistry and biophysics. 2008;479(2):114-20. Bashkin JK, Jenkins LA. The Role of Metals in the Hydrolytic Cleavage of DNA and RNA. Comments on Inorganic Chemistry. 1994;16(1-2):77-93. Li Y, Breaker RR. Kinetics of RNA Degradation by Specific Base Catalysis of Transesterification Involving the 2‘-Hydroxyl Group. Journal of the American Chemical Society. 1999;121(23):5364-72. Waters RS, Bryden NA, Patterson KY, Veillon C, Anderson RA. EDTA chelation effects on urinary losses of cadmium, calcium, chromium, cobalt, copper, lead, magnesium, and zinc. Biol Trace Elem Res. 2001 Dec;83(3):207-21. doi:10.1385/BTER:83:3:207 Nubel G, Sorgenfrei FA, Jaschke A. Boronate affinity electrophoresis for the purification and analysis of cofactor-modified RNAs. Methods. 2017;117:14-20. Watkins CP, Zhang W, Wylder AC, Katanski CD, Pan T. A multiplex platform for small RNA sequencing elucidates multifaceted tRNA stress response and translational regulation. Nature Communications. 2022;13(1):2491. Somervuo P, Koskinen P, Mei P, Holm L, Auvinen P, Paulin L. BARCOSEL: a tool for selecting an optimal barcode set for high-throughput sequencing. BMC Bioinformatics. 2018;19(1):257. Zhang Z, Milias-Argeitis A, Heinemann M. Dynamic single-cell NAD(P)H measurement reveals oscillatory metabolism throughout the E. coli cell division cycle. Scientific Reports. 2018;8(1):2162. Basan M, Hui S, Okano H, Zhang Z, Shen Y, Williamson JR, et al. Overflow metabolism in Escherichia coli results from efficient proteome allocation. Nature. 2015;528(7580):99-104. Szenk M, Dill KA, de Graff AMR. Why Do Fast-Growing Bacteria Enter Overflow Metabolism? Testing the Membrane Real Estate Hypothesis. Cell Systems. 2017;5(2):95-104. Vemuri GN, Altman E, Sangurdekar DP, Khodursky AB, Eiteman MA. Overflow metabolism in Escherichia coli during steady-state growth: transcriptional regulation and effect of the redox ratio. Appl Environ Microbiol. 2006;72(5):3653-61. doi:10.1128/AEM.72.5.3653-3661.2006 Klumpp S. Growth-Rate Dependence Reveals Design Principles of Plasmid Copy Number Control. PLOS ONE. 2011;6(5):e20403. Jasiecki J, Wegrzyn G. Growth-rate dependent RNA polyadenylation in Escherichia coli . EMBO Rep. 2003;4(2):172–7. doi:10.1038/sj.embor.embor733 Additional Declarations No competing interests reported. Supplementary Files MoehlerHeliosGenomeBiologySI.pdf Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 09 Feb, 2026 Reviews received at journal 29 Jan, 2026 Reviewers agreed at journal 09 Jan, 2026 Reviews received at journal 22 Nov, 2025 Reviewers agreed at journal 07 Nov, 2025 Reviewers invited by journal 02 Oct, 2025 Editor assigned by journal 23 Sep, 2025 Submission checks completed at journal 19 Sep, 2025 First submitted to journal 18 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7650670","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":525910703,"identity":"b4b225df-a1d0-41ce-8d80-ddc74455c53e","order_by":0,"name":"Marvin Möhler","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Marvin","middleName":"","lastName":"Möhler","suffix":""},{"id":525910704,"identity":"ca3e4750-5263-4bd0-861a-217c74a07a99","order_by":1,"name":"Tae Wha Seong","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Tae","middleName":"Wha","lastName":"Seong","suffix":""},{"id":525910705,"identity":"d3f98df2-d93b-43d1-9e68-ef5fc066c715","order_by":2,"name":"Frederik Weber","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Frederik","middleName":"","lastName":"Weber","suffix":""},{"id":525910706,"identity":"2e167f59-8cce-4401-93c5-cbd77a00f88c","order_by":3,"name":"Héctor Gabriel Morales Filloy","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Héctor","middleName":"Gabriel Morales","lastName":"Filloy","suffix":""},{"id":525910707,"identity":"174202d9-851c-4338-acc3-7122bf512229","order_by":4,"name":"Marleen Wölke","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Marleen","middleName":"","lastName":"Wölke","suffix":""},{"id":525910708,"identity":"3615bb28-16ba-4e16-bb48-1f3cd4e2ab94","order_by":5,"name":"Elif Tosun","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Elif","middleName":"","lastName":"Tosun","suffix":""},{"id":525910709,"identity":"27b918b5-b397-4fbf-95b9-146a3b1ef198","order_by":6,"name":"Edrea Mendonca","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Edrea","middleName":"","lastName":"Mendonca","suffix":""},{"id":525910710,"identity":"d2ce58b6-4153-4973-9f7f-7390a9acb341","order_by":7,"name":"Andrijana Crenner","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Andrijana","middleName":"","lastName":"Crenner","suffix":""},{"id":525910711,"identity":"4735e3a5-6409-47e0-a67a-8fe327ffa5b5","order_by":8,"name":"Yaqing Zhang","email":"","orcid":"","institution":"Heidelberg University","correspondingAuthor":false,"prefix":"","firstName":"Yaqing","middleName":"","lastName":"Zhang","suffix":""},{"id":525910712,"identity":"9154bf34-3103-4295-9a6e-9f91f343f32b","order_by":9,"name":"Frank Preugschat","email":"","orcid":"","institution":"Neolaiabio Inc.","correspondingAuthor":false,"prefix":"","firstName":"Frank","middleName":"","lastName":"Preugschat","suffix":""},{"id":525910713,"identity":"5515446f-5713-4e28-ab47-7c041d9544d3","order_by":10,"name":"Andres Jäschke","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIie3OMQrCMBSA4ffoKu1qEOwVUgpdepmIUJfo7GamuChdew9BHCMdXHKAbAqCk2DByUGwLbo4mI4O+YckhHy8ALhcfxiKdmEQ1LtC4dVrV0JEV/J2DKhqDx2ItyzL6rHLhrGZ7hXuUvAPwjJglWVkrXmcmBlTqCdAtGUMCp54KOejreG0fMoSqGEWkt/u95osNgWnChtyPFlIwWGAkjPa/xDzW9TkkpC1zKJCXxsy6RFt+ViUj8/VQ47DYMnjCmU69A/KQsTXRc/yLYDQ+sLlcrlcLx63S5FAW7RxAAAAAElFTkSuQmCC","orcid":"","institution":"Heidelberg University","correspondingAuthor":true,"prefix":"","firstName":"Andres","middleName":"","lastName":"Jäschke","suffix":""}],"badges":[],"createdAt":"2025-09-18 14:23:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7650670/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7650670/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93576583,"identity":"021bddc1-018a-4c7c-8139-5a6d676ca4a3","added_by":"auto","created_at":"2025-10-15 09:36:38","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1583453,"visible":true,"origin":"","legend":"","description":"","filename":"MoehlerHeliosGenomeBiology.docx","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/54d1f625f7ff36fb2cbd3cce.docx"},{"id":93576256,"identity":"aaafa3e8-354a-4935-84b0-3a6308c769c3","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12005,"visible":true,"origin":"","legend":"","description":"","filename":"df3b2895ff894118bd6747f004cda0e2.json","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/69792f431c67905ee23e261a.json"},{"id":93576270,"identity":"22822801-1a55-4e1c-b3ad-2fedb4f2de10","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4953575,"visible":true,"origin":"","legend":"","description":"","filename":"MoehlerHeliosGenomeBiologySI.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/d1ab6bd5ad2bccc6eba91435.pdf"},{"id":93576260,"identity":"4bdb33e8-82f2-4542-ab14-4f424a6eb351","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":180965,"visible":true,"origin":"","legend":"","description":"","filename":"df3b2895ff894118bd6747f004cda0e21enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/5c34bb5d0794291d401eae5d.xml"},{"id":93577775,"identity":"5adea100-119f-484e-86ec-bc1ed8517e9e","added_by":"auto","created_at":"2025-10-15 09:44:38","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":525221,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/02af666bb550c913abb6e137.png"},{"id":93576586,"identity":"dedd65b4-a96c-45e2-8623-199db1a07848","added_by":"auto","created_at":"2025-10-15 09:36:38","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":473370,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/1b668c6b005bdd0588e5832d.png"},{"id":93576262,"identity":"1c4024fe-3340-4fb5-83c0-7a46c2574350","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":408970,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/15ebfe86d9f0bd57aac8bd6c.png"},{"id":93576268,"identity":"797c43b1-f78a-4788-90db-9c1cb9acf14a","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69586,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/ae6b5ea4dd0e3149ba733561.png"},{"id":93576259,"identity":"5111ec42-4e31-46c2-b228-60e5fc8ddf33","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56339,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/83de9a76f0a43b0a3a47be8b.png"},{"id":93576587,"identity":"afbd85f2-9767-4067-9a7d-edd3bfa6be85","added_by":"auto","created_at":"2025-10-15 09:36:38","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69785,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/50747e24f78475e35c723afe.png"},{"id":93576265,"identity":"bbd65a2d-e639-4443-9b04-a36ac2ca36cc","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":179011,"visible":true,"origin":"","legend":"","description":"","filename":"df3b2895ff894118bd6747f004cda0e21structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/97710966e5b400a74caa31b8.xml"},{"id":93576267,"identity":"d67d1318-e2ed-47ab-b493-3d2a48b6b2df","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"html","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":190450,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/87ddbab50d71ef66d9df2c2c.html"},{"id":93576584,"identity":"239d454f-7272-48a7-94fd-9a8396a971b9","added_by":"auto","created_at":"2025-10-15 09:36:38","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":138514,"visible":true,"origin":"","legend":"\u003cp\u003eNAD-cap structure of RNA and specific targeting by ADPRC transglycosylation with the novel 3-picolylamine biotin substrate. a)\u0026nbsp;Chemical structure of the nicotinamide adenine dinucleotide (NAD\u003csup\u003e+\u003c/sup\u003e) cap structure at the 5’-end of RNA via a 5’,5’-pyrophosphate bridge between a nicotinamide ribonucleotide and the 5’-adenosine of the oligonucleotide chain. b)\u0026nbsp;ADP-ribosyl cyclase (ADPRC)-catalyzed transglycosylation reaction replacing the nicotinamide moiety at the 1’-riboside position with a suitable nucleophile. c)\u0026nbsp;Structure and concept of the \u003cem\u003eN\u003c/em\u003e-nucleophilic ADPRC substrate 3-picolylamine biotin (3PAB). d)\u0026nbsp;ADPRC transglycosylation reaction of NAD\u003csup\u003e+\u003c/sup\u003e or NAD-capped RNA with 3PAB as a nucleophilic substrate. e)\u0026nbsp;HPLC analysis of the ADPRC-catalyzed conversion of NAD\u003csup\u003e+\u003c/sup\u003e to nicotinamide (NA) and ADP-ribosyl 3-picolylamine biotin conjugate (ADPR-3PAB, identified by HR-MS). The chromatograms show the analysis of the reaction mix in the presence (orange) and absence of enzyme (blue). f)\u0026nbsp;Simplified protocol for the kinetic analysis of ADPRC transglycosylation reactions of NAD\u003csup\u003e+\u003c/sup\u003e using 3PAB.\u0026nbsp;g) Analysis of aliquots withdrawn at different time-points for ADPRC-3PAB transglycosylations of NAD\u003csup\u003e+\u003c/sup\u003e by HPLC; magnified view of the ADPR-3PAB reaction product peak (see Fig. S3a for full chromatogram). h)\u0026nbsp;ADPR-3PAB formation data (derived from product area in HPLC chromatograms) for each time-point of reaction triplicates plotted against the reaction time with exponential fitting. j)\u0026nbsp;Simplified protocol for the kinetic analysis of ADPRC transglycosylation reactions of a model RNA (113 nt) using 3PAB. k) Analysis of aliquots withdrawn at different time-points for ADPRC-3PAB transglycosylations of a model RNA (113\u0026nbsp;nt) via agarose gel electrophoresis and northern blot. l) Model RNA (113\u0026nbsp;nt)-3PAB formation data (derived from northern blotting and phosphor imaging readout) for each time-point of reaction triplicates plotted against the reaction time with exponential fitting.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/ba0e05f41eb611a9955c6b46.jpg"},{"id":93576254,"identity":"c24aa59d-5f12-4a2b-8788-b2ede26829bb","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":103234,"visible":true,"origin":"","legend":"\u003cp\u003ea)\u0026nbsp;Scheme illustrating the workflow of HELIOS NAD-Seq for the identification of NAD-capped RNAs. By default, HELIOS NAD-Seq starts with the ADPRC-catalyzed 3PAB-modification of NAD-RNA contained in around 2\u0026nbsp;µg of total RNA, followed by early multiplexing of a full sample group (biological replicates and negative controls) after 3’-ligation with barcoded adapters, streptavidin magnetic bead purification of the biotinylated 3PAB-modified RNA species, on-bead reverse transcription, bead release of the complementary cDNA, second adapter ligation, PCR amplification, and library preparation including native PAGE purification and size selection, before the samples are ready for sequencing with the Illumina platform. b)\u0026nbsp;Dot blot fluorescence readout for the ADPRC specificity assay with 3PAB (see Fig.\u0026nbsp;S15) shows transglycosylation of NAD-capped model RNA (36\u0026nbsp;nt), while no reactivity towards common 5’-end structures, such as 5’-triphosphate (ppp) and the m\u003csup\u003e7\u003c/sup\u003eG-cap (m\u003csup\u003e7\u003c/sup\u003eGppp) is observed. c)\u0026nbsp;Design of barcoded adapters AD3-id4. A set of twelve sequencing-ready 4-letter barcodes was chosen for HELIOS NAD-Seq (see Fig.\u0026nbsp;S17a). Adapters are pre-adenylated (see Fig.\u0026nbsp;S17b) before use in 3’-ligation reactions. d)\u0026nbsp;Number of RNA transcripts identified from the sequencing data set of HELIOS NAD-Seq with \u003cem\u003eE.\u0026nbsp;coli\u003c/em\u003e K12 JM109 samples, which possessed a transcription start site (TSS) near or before the 5’-end of \u003cem\u003eE.\u0026nbsp;coli\u003c/em\u003e genes, and their assignment to a distinct group of gene features. e)\u0026nbsp;Comparison of sequencing hits from the NAD captureSeq data sets by \u003cem\u003eCahová et al. \u003c/em\u003eand HELIOS NAD-Seq with \u003cem\u003eE.\u0026nbsp;coli\u003c/em\u003e K12 JM109 samples, highlighting the superior sensitivity of the HELIOS NAD-Seq protocol.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/4c83a0a1f58722458f8e35ff.jpg"},{"id":93576255,"identity":"d9c72715-76fa-4421-b16e-5d7385c5c11c","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":166125,"visible":true,"origin":"","legend":"\u003cp\u003ea) NAD-capped transcript abundance across 16 \u003cem\u003eE. coli\u003c/em\u003e growth timepoints. Total RNA was collected in quadruplicates at 30-minute intervals and potential NAD-capped gene species were identified at each time point. NAD-capping peaked after 3 hours and 30 minutes which corresponds to the early exponential phase. A full list of NAD-capped transcript candidates that appear in at least 2 time points can be found in Table S8. b) Reduced NAD regeneration during fermentation (2 NAD⁺ per glucose vs. 8 in respiration) may underlie the decline in NAD-capped RNA. c) Two-step normalization workflow for NAD-capping analysis. Sequencing depth was first used to normalize total reads per timepoint. In the second step, expression levels were adjusted using 21 “reference set” genes to account for variability across samples. Reference gene stability is illustrated for three representative tRNAs in Figure S32a.\u003cstrong\u003e \u003c/strong\u003eValidation of RNAI quantification by northern blotting on APB-PAGE using 5′-end γ-[³²P]-ATP–labeled probes (Figure S34). d) Normalized HELIOS NAD-Seq read counts closely matched RNAI abundance profiles from northern blotting. e) Heatmap of NAD-capped RNAs detected in ≥2 of 16 timepoints after two-step normalization, showing mean z-scores across the growth curve. (C1-cluster1, C2-cluster2, C3-cluster3) Full gene lists with annotations and clusters are provided in Table S9. f) Average z-scores across timepoints for each NAD-capped RNA cluster, summarizing the trends observed in Figure 3e.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/5d4b733ea230601ced92d7f7.jpg"},{"id":93578020,"identity":"2d6b9a65-44b8-4c55-b0db-3d1417a8b224","added_by":"auto","created_at":"2025-10-15 09:52:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1350367,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/900148f9-d453-4e42-bca9-5df3d8f8e509.pdf"},{"id":93576264,"identity":"3d96e22f-bf19-466b-b4d9-e44400e90447","added_by":"auto","created_at":"2025-10-15 09:28:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":4953575,"visible":true,"origin":"","legend":"","description":"","filename":"MoehlerHeliosGenomeBiologySI.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7650670/v1/5e64996dc77a2d70553f9c84.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eHELIOS NAD-Seq: A Next-Generation Capture and Sequencing Protocol for NAD-Capped RNAs with Superior Targeting and Processing\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Background","content":"\u003cp\u003eThe modification of RNA with nicotinamide adenine dinucleotide (NAD\u003csup\u003e+\u003c/sup\u003e) at the 5' end resembles the eukaryotic 7-methylguanosine (m\u003csup\u003e7\u003c/sup\u003eG) cap. Unlike the m\u003csup\u003e7\u003c/sup\u003eG cap, NAD\u003csup\u003e+\u003c/sup\u003e and related metabolites such as NADH, FAD, dephospho-coenzyme A, and others can be incorporated into RNA as non-canonical initiating nucleotides (NCIN) during transcription primarily through cotranscriptional processes.(\u003cspan additionalcitationids=\"CR3 CR4 CR5 CR6 CR7\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eOriginally identified as the first prokaryotic RNA cap in 2009 by Liu and coworkers,(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e) the NAD-modification remains the most extensively studied non-canonical RNA cap(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Since the development of NAD captureSeq by J\u0026auml;schke and colleagues in 2015,(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e) NAD-capped RNAs have been found across all domains of life\u0026mdash;from bacteria(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) and archaea(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) to yeast(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e), plants(\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e), and animals(\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e), including human cells.(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e) The widespread occurrence of NAD-capping, coupled with the existence of specific enzymes (\u0026ldquo;erasers\u0026rdquo;)(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan additionalcitationids=\"CR24 CR25 CR26 CR27 CR28 CR29\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e) and both prokaryotic and eukaryotic RNA polymerases as \u0026ldquo;writers\u0026rdquo;(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e) supports the notion that NAD-capping is a conserved and functionally relevant biological phenomenon. Recent studies have linked NAD\u003csup\u003e+\u003c/sup\u003e and FAD capping to viral infection processes,(\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e) while suggested functions in cells include RNA stability(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e), regulation of gene expression(\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e), and involvement in processes like RNAylation\u0026mdash;a covalent attachment of NAD-capped RNAs to proteins(\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). Despite this, our understanding of the physiological roles of NAD-capping remains limited, and further exploration is hindered by the limitations of current detection methods.\u003c/p\u003e\u003cp\u003eCap-specific NAD-RNA identification protocols play a central role in the discovery of NAD-RNAs with potentially new roles in a wide range of organisms. The recurring element of those capture protocols, which are based on NAD captureSeq,(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e) is the use of a primary alcohol substrate class to modify the NAD-cap structure in a key capture reaction catalyzed by the enzyme ADP-ribosyl cyclase (ADPRC) from \u003cem\u003eAplysia californica\u003c/em\u003e.(\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e) This modification reaction is generally used to attach a tag molecule to the NAD-cap, which allows either the specific purification or the detection of NAD-RNAs from a total RNA sample, permitting the sequencing and identification of NAD-capped RNA species.(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan additionalcitationids=\"CR40\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eAs discussed in our recent review article, NAD-RNA capture protocols would benefit substantially from improvements in ADPRC reaction yield and reaction specificity, reductions in total RNA demand, and an increase in sample processing capacity.(\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e) Currently, most methods advise to start from around 100 \u0026micro;g of total RNA per replicate (or an equivalent amount of e.g. polyadenylated RNA fraction),(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e) which creates a barrier to analyzing samples with low RNA yield or NAD-capping levels (often only around 0.1\u0026ndash;0.2% of the transcriptome).(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e) Undesired reactivity and off-target effects, especially with abundant m\u003csup\u003e7\u003c/sup\u003eG-capped RNAs, affect the evaluation and reliability of the obtained sequencing data.(\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e) Finally, only a low number of samples can be processed in parallel, which makes the analysis of sample screens (e.g. total RNA from different biological conditions or growth states) time-consuming and more error-prone.(\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eTo overcome these challenges, we introduce HELIOS NAD-Seq (\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eH\u003c/span\u003eigh-efficiency \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eE\u003c/span\u003enzyme modification with \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eL\u003c/span\u003eow \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eI\u003c/span\u003enput \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eO\u003c/span\u003ef \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eS\u003c/span\u003eample for \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eNAD-\u003c/span\u003eRNA \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eSeq\u003c/span\u003euencing), a novel protocol that significantly improves yield, specificity, and throughput. Using a pyridyl-based ADPRC substrate, 3-picolylamine biotin (3PAB), HELIOS NAD-Seq enables one-step, highly efficient biotinylation of NAD-RNAs, requiring at least 50-fold less total RNA compared to previous methods. Its reaction conditions avoid common off-target modifications, such as labeling of m\u003csup\u003e7\u003c/sup\u003eG-capped RNAs, and do not need divalent cations or additional click-chemistry steps. This combination of high reactivity, specificity, and low input requirement makes HELIOS NAD-Seq suitable for diverse biological samples.\u003c/p\u003e\u003cp\u003eMoreover, the protocol incorporates early barcoding and multiplexing of samples via 3'-ligation with barcoded adapters, allowing simultaneous processing of at least 16 sample groups, each in quadruplicate, within just four days. This design minimizes sample handling bias, enhances reproducibility, and greatly increases processing capacity.\u003c/p\u003e\u003cp\u003eThe sequencing data obtained for \u003cem\u003eE. coli\u003c/em\u003e samples demonstrate that HELIOS NAD-Seq can recover most transcripts previously identified by NAD captureSeq, with the added ability to detect many additional low-abundance NAD-RNAs. Comparisons also reveal higher specificity and sensitivity, even with a 50-fold reduction in total RNA input. The high-throughput approach enables detailed analyses of NAD-capping dynamics during different growth phases, uncovering potential links between cellular metabolism and NAD-RNA prevalence, and highlighting the broad applicability of this method for expanding our understanding of NAD-capping across diverse organisms and conditions.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eADPRC reaction with PAB derivatives\u003c/h2\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eNAD-capped RNA (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea) can be specifically modified through an ADPRC-catalyzed transglycosylation reaction targeting the NAD\u003csup\u003e+\u003c/sup\u003e-cap structure. This enzyme removes the nicotinamide base via nucleophilic attack at the ribose-1\u0026rsquo;-carbon, facilitating the subsequent attack of a suitable nucleophile to yield a retention of the ribose-substitution pattern(\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). Established protocols for NAD-RNA capture have utilized this reaction with primary alcohols as nucleophilic substrates for ADPRC.(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan additionalcitationids=\"CR40\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e) When performing the enzymatic transglycosylation with alkyne- and azide-containing primary alcohols (4-pentyn-1-ol and 3-azidopropan-1-ol, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea-b), an additional step is required to attach biotin or other tags via click chemistry. However, these chemical reactions present limitations, such as RNA degradation in copper-catalyzed azide-alkyne-cycloaddition (CuAAC) and specificity issues with reactive, internal alkynes in strain-promoted azide-alkyne cycloaddition (SPAAC). The recently introduced HEEB substrate (\u003cem\u003eN\u003c/em\u003e-[2-(2-hydroxyethoxy)ethyl]biotinamide; Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ec) allows one-step biotinylation of NAD-RNAs at a lower substrate concentration. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e) However, a side reactivity of the ADPRC-catalyzed reaction of primary alcohol substrates with m\u003csup\u003e7\u003c/sup\u003eG-capped RNAs has been reported,(\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e) which could lead to false-positive identification of NAD-capped RNAs if incomplete m\u003csup\u003e7\u003c/sup\u003eG-cap depletion techniques are used.(\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e) Due to the consistent modification of NAD-RNA using primary alcohols in the presence of ADPRC in all these protocols,(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan additionalcitationids=\"CR40\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e) there remains significant potential for innovation in this key capture step.(\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eTo improve the efficiency of NAD-RNA modification, we introduce here a novel substrate class, based on \u003cem\u003eN\u003c/em\u003e-nucleophilic compounds, which have shown high yields and rapid kinetics during ADPRC transglycosylation at the dinucleotide level (NAD\u003csup\u003e+\u003c/sup\u003e).(\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e) We identified picolylamine-biotin (PAB) conjugates, in particular 3-picolylamine biotin (3PAB, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec), as potent novel substrates for modifying both NAD\u003csup\u003e+\u003c/sup\u003e and NAD-RNA. The modular design features a nucleophilic pyridyl head that resembles nicotinamide, connected via a reverse amide and optionally a linker to a functional end (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec). For NAD-RNA identification protocols, this end group can be a high-affinity binder such as biotin, but other functionalities, like fluorophores, can be introduced by selecting different end groups.\u003c/p\u003e\u003cp\u003eSynthesis of PAB conjugates involved NHS-activation of biotin derivatives followed by amide coupling with primary amines on 3- or 4-picolylamine (Fig. S2a-b). Besides biotin variants (3PAB, 4PAB), corresponding desthiobiotin conjugates (3PADB, 4PADB) were prepared similarly.\u003c/p\u003e\u003cp\u003eIn ADPRC transglycosylation assays, these PAB compounds enabled efficient, one-step biotinylation of NAD\u003csup\u003e+\u003c/sup\u003e-modified targets, with rapid kinetics and high yields (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ed). With NAD\u003csup\u003e+\u003c/sup\u003e as substrate, the conversion to a biotinylated ADPR product (ADPR-3PAB) was completed within 30 minutes at low enzyme concentrations (0.025 mol%) and a threefold excess of PAB (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee). High-performance liquid chromatography (HPLC) and high-resolution mass spectrometry (HR-MS) confirmed the formation of the expected products (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee, Fig. S3). Notably, the reaction with PAB conjugates proceeded independently of Mg\u003csup\u003e2+\u003c/sup\u003e ions and was unaffected by EDTA, which chelates divalent cations and can mitigate RNA hydrolysis (Fig. S4a-c)(\u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e). NADH was inert in these reactions, indicating specificity for the oxidized state (Fig. S5).\u003c/p\u003e\u003cp\u003eKinetic comparisons revealed that all PAB derivatives could efficiently convert NAD\u003csup\u003e+\u003c/sup\u003e to the biotinylated ADPR within five minutes or less, with the reactivity ranking as 3PAB\u0026thinsp;\u0026gt;\u0026thinsp;3PADB\u0026thinsp;\u0026gt;\u0026thinsp;4PAB\u0026thinsp;\u0026gt;\u0026thinsp;4PADB (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ef, S7d). Quantitative analysis based on HPLC peak areas indicated near-complete conversion for all compounds within these times (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eg\u0026ndash;h, S8). The k\u003csub\u003ecat\u003c/sub\u003e / K\u003csub\u003em\u003c/sub\u003e value calculated from the exponential fit curve was 7.59 \u0026times;10\u003csup\u003e5\u003c/sup\u003e M\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003es\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e which is within a reasonable range for ADPRC-catalyzed transglycosylation reactions.(\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eFor RNA-level applications, we developed a robust assay (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ej): a standard transglycosylation reaction using 3PAB and a 113 nt model NAD-modified RNA (about 48% NAD-modified according to acryloylaminophenyl boronic acid (APB)-PAGE analysis(\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e) which retards NAD-capped RNA). The reaction was performed with excess \u003cem\u003eE. coli\u003c/em\u003e total RNA, and after quenching and extraction, the RNA was incubated with streptavidin (Fig. S6). The biotinylated RNA exhibited reduced mobility in electrophoretic mobility shift assays (EMSAs), confirming successful labeling (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ek). This method, termed StrepShift assay, was performed using agarose gels and complemented with northern blot analysis where needed.\u003c/p\u003e\u003cp\u003eWe observed that the efficiency of transglycosylation on RNA was lower than on dinucleotides, likely due to the larger size of oligonucleotides and potential RNA degradation. Maintaining RNA integrity is crucial for accurate NAD-RNA identification. Even with EDTA-containing buffers and reaction times of 30 minutes, higher enzyme concentrations (e.g., 380 nM) caused significant degradation of 113 nt model RNA and total \u003cem\u003eE. coli\u003c/em\u003e RNA (Fig. S9). To mitigate this, we reduced the enzyme concentration 10-fold to 38 nM in the HELIOS NAD-Seq protocol, which prevented detectable RNA degradation (Fig. S9).\u003c/p\u003e\u003cp\u003eIn comparative assays, 3PAB consistently outperformed other pyridyl conjugates in labeling efficiency with the 113 nt model RNA and total \u003cem\u003eE. coli\u003c/em\u003e RNA (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003el). Reactions were conducted with reaction times from 0.5 to 30 minutes, followed by precipitation, streptavidin incubation, and the StrepShift assay (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ek, S10). Due to weaker streptavidin binding, desthiobiotin derivatives (3PADB, 4PADB) produced less pronounced mobility shifts. The reaction with 3PAB approached completion within 30 minutes with a k\u003csub\u003ecat\u003c/sub\u003e / K\u003csub\u003em\u003c/sub\u003e value of 6.17 \u0026times;10\u003csup\u003e5\u003c/sup\u003e M\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003es\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, with slower kinetics observed for other conjugates (Fig. S11). The reactivity ranking was confirmed as 3PAB\u0026thinsp;\u0026gt;\u0026thinsp;4PAB\u0026thinsp;\u0026gt;\u0026thinsp;3PADB\u0026thinsp;\u0026gt;\u0026thinsp;4PADB (Fig. S11d).\u003c/p\u003e\u003cp\u003eFurther, we synthesized biotin conjugates with extended polyethylene glycol linkers, N(PEG)\u003csub\u003e3\u003c/sub\u003eB and N(PEG)\u003csub\u003e11\u003c/sub\u003eB (Fig. S12a). These were prepared via similar NHS-activation methods. However, their efficiency in transglycosylation reactions decreased with longer linkers, yielding only 63% and 25% biotinylation relative to 3PAB, respectively (Fig. S13). Consequently, the smaller 3PAB remained the preferred substrate for efficient NAD-RNA biotinylation.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eThe HELIOS NAD-Seq protocol\u003c/h3\u003e\n\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe workflow of HELIOS NAD-Seq for Illumina sequencing is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and hinges on several key concepts. First, total RNA is specifically biotinylated via ADPRC transglycosylation using our novel nucleophilic substrate 3PAB in a single reaction step. This is followed by 3'-ligation of a barcoded adapter, enabling early multiplexing of biological replicates and negative controls, each uniquely barcoded. Biotinylated RNA species are then captured on streptavidin magnetic beads, and all samples and controls are pooled for simultaneous processing. After reverse transcription with adapter-specific primers, the RNA is hydrolyzed with NaOH, and the resulting cDNA is recovered. The cDNA molecules then undergo a second adapter ligation, PCR amplification, and library purification via native PAGE. From up to 16 total RNA samples\u0026mdash;each in quadruplicate with controls\u0026mdash;the entire HELIOS NAD-Seq protocol can be completed in four days.\u003c/p\u003e\u003cp\u003eStarting with total RNA extracted from an organism of choice, typically 4 \u0026micro;g per sample (split into 2 \u0026micro;g for positive and negative controls), internal RNA standards (Fig. S14) are added at approximately 0.25 fmol/\u0026micro;g RNA for normalization and quality control. For protocol optimization, we reduced the initial RNA input and enzyme concentration: ADPRC was used at 38 nM, a tenfold decrease from NAD captureSeq conditions, to minimize RNA degradation (Fig. S9). The nucleophilic substrate (3PAB) concentration was lowered by roughly 70-fold, compared to NAD captureSeq, to 15 mM, maintaining full conversion efficiency. This setup ensures rapid, specific labeling of NAD-capped RNAs, avoiding side reactivity with 5' modifications such as the m\u003csup\u003e7\u003c/sup\u003eG cap, as confirmed by our specificity assays (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb, Fig. S16a). To verify biotinylation efficiency, we used model RNA standards (IS-036) spiked into total RNA before the reaction, which were UV-crosslinked onto nylon membranes, incubated with HRP-streptavidin, and detected via luminol. Biotinylation of NAD-capped species occurred within 0.5 minutes, with no signal detected for non-NAD species even after 30 minutes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb, S16b).\u003c/p\u003e\u003cp\u003eFor early sample multiplexing, a barcoded adapter is ligated to the 3'-end of RNA transcripts prior to streptavidin bead capture (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). Inspired by the NAD captureSeq AD3 adapter,(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e) it contains a randomized 5\u0026rsquo;-region to reduce sequence bias, followed by a Truseq-compatible 3'-region blocked with a C3 spacer and a 4-base barcode (with a 2-base spacer)(\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e) for multiplexing four biological replicates and controls. A set of 12 unique barcodes was designed using the BARCOSEL tool(\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e), ensuring a Hamming distance of at least three to minimize sample misassignment (Fig. S17a). 5\u0026rsquo;-phosphorylated barcoded adapters (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec) were further modified by pre-adenylation using ImpA (Fig. S17b, PAGE analysis in Fig. S17c)(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAfter 3'-ligation and heat inactivation, samples and controls are pooled and subjected to streptavidin capture. In vitro, the binding strength of NAD-RNA to streptavidin was assessed via dot blot with radioactive probing: biotinylated RNA generated with various PAB derivatives demonstrated strong, specific interactions, with the weakest binding observed for desthiobiotin variants (Fig. S18). The captured RNA was washed with urea buffers, eluted with biotin, and analyzed by northern blotting using probes against the model RNAs and ribosomal RNAs as controls (Figs. S18\u0026ndash;S21). The approach confirmed effective, specific binding that was robust even after multiple washes.\u003c/p\u003e\u003cp\u003eFurther downstream protocol steps (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea) were adapted largely from the NAD captureSeq workflow(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Each of those steps has been optimized for the lower oligonucleotide input of HELIOS NAD-Seq and, where applicable, the use with streptavidin magnetic beads. After on-bead reverse transcription (RT) using the primer RTp2, the RNA part of the formed RNA-cDNA duplex is hydrolyzed by incubation with 150 mM NaOH solution(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). The released cDNA is recovered and further treated via a C-tailing reaction using terminal transferase and second adapter ligation. After this, we used NEBNext universal forward and indexed i7 reverse primers for PCR amplification of the libraries, which were purified and size-selected by native PAGE. Gel extraction and recovery of the DNA yielded libraries ready for Illumina NGS sequencing.\u003c/p\u003e\u003cp\u003eTo evaluate the capabilities of our novel HELIOS NAD-Seq protocol, we sought to compare it to the performance of NAD captureSeq, which we regularly apply in our laboratories. For this purpose, we replicated the experiments on \u003cem\u003eE. coli\u003c/em\u003e K12 JM109, which were highlighted in the original NAD captureSeq paper(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). \u003cem\u003eE. coli\u003c/em\u003e K12 JM109 were grown to the late exponential phase (OD\u003csub\u003e600\u003c/sub\u003e\u0026thinsp;\u0026asymp;\u0026thinsp;2.0) and harvested (Fig. S22). Due to the low RNA input required for HELIOS NAD-Seq, as little as 1.0 mL of culture yielded sufficient total RNA for 10 protocol runs with 2 \u0026micro;g RNA each after purification and DNase digest (Fig. S23).\u003c/p\u003e\u003cp\u003eHELIOS NAD-Seq was performed starting from 2 \u0026micro;g total RNA in biological triplicates (for control purposes with slightly elevated internal RNA standard amounts of 1 fmol) following the standard protocol and barcoding with adapters carrying bc01\u0026ndash;03 ligated to samples treated with ADPRC and 3PAB, and adapters with bc04\u0026ndash;06 ligated to controls treated with ADPRC and 3-picolylamine. Samples and negative controls were pooled before streptavidin magnetic bead capture. After native PAGE purification of the PCR-amplified and multiplexed libraries, a size selection in the region between 150 and 300 bp was performed to comply with the procedure of the original NAD captureSeq (Fig. S24). Due to the mild conditions of HELIOS NAD-Seq, for which only low degradation during the protocol application is expected, it was already possible at this stage to conclude that mainly shorter \u003cem\u003eE. coli\u003c/em\u003e RNA transcripts were NAD-capped. Sequencing reads were then processed, aligned, quantified, and analyzed as described in Methods (\u0026ldquo;HELIOS NAD-Seq\u0026mdash;Bioinformatic analysis\u0026rdquo;).\u003c/p\u003e\u003cp\u003eThe sequencing data revealed that 94.2% of over 132\u0026nbsp;million reads could be assigned to the six barcodes, with even distribution across biological replicates. Positive samples exhibited 35\u0026ndash;43\u0026nbsp;million reads, while negative controls contained 5.5\u0026ndash;6.1\u0026nbsp;million, reflecting the successful enrichment of NAD-capped RNAs. After removing reads from internal standards, the positive samples still had nearly 200 times more reads than the controls (6.2\u0026ndash;7.5\u0026nbsp;million vs. 22\u0026ndash;50 thousand). Enrichment analysis confirmed the high specificity of HELIOS NAD-Seq, showing that off-target effects, such as non-specific labeling of m\u003csup\u003e7\u003c/sup\u003eG-capped RNAs, were effectively eliminated (Tables S1\u0026ndash;S2).\u003c/p\u003e\u003cp\u003eFrom the dataset, 30,080,435 reads (26.6%) mapped to 3,557 out of 4,621 annotated gene features in \u003cem\u003eE. coli\u003c/em\u003e K12 JM109 and the pUC19 plasmid(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Applying strict detection criteria (baseMean\u0026thinsp;\u0026ge;\u0026thinsp;50, log2-fold change\u0026thinsp;\u0026ge;\u0026thinsp;3, adjusted p-value\u0026thinsp;\u0026le;\u0026thinsp;0.01), we identified 389 transcripts (associated with 344 unique gene features) as high-confidence NAD-capped RNAs (Table S3). These transcripts exhibited a conserved promoter motif (TANNNT) akin to the standard \u003cem\u003eE. coli\u003c/em\u003e promoter, typically located at positions \u0026minus;\u0026thinsp;12 to \u0026minus;\u0026thinsp;7 upstream of the transcription start site (TSS), which is always an adenosine (+\u0026thinsp;1A) in NAD transcripts (Fig. S25a\u0026ndash;b). A manual analysis of TSS indicated that 207 transcripts (53.2%) matched known or suspected TSSs, most being protein-coding RNAs or regulatory small RNAs (Table S4). Many were full-length sRNAs, while most mRNAs appeared as truncated 5'-fragments.\u003c/p\u003e\u003cp\u003eComparison with previous NAD captureSeq datasets(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) confirmed most of our top hits\u0026mdash;75 of 80 genes from dataset 1, 42 of 44 from dataset 2, and 64 of 70 from dataset 3\u0026mdash;were also identified as NAD-capped, with HELIOS NAD-Seq demonstrating significantly higher enrichment factors (Tables S5\u0026ndash;S7)(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Moreover, HELIOS NAD-Seq identified an additional 251 gene features not detected previously, reflecting increased sensitivity. (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee). Some highly abundant NAD-capped RNAs (e.g., from yfjI, ykgS, ymdB, bfr, and yuaQ loci) were new findings. Overlap with the NAD tagSeq II data was markedly lower, likely due to differences in strain identity, growth conditions, and sample timing and is shown in Fig.S26. Overall, the results validate that HELIOS NAD-Seq, starting from substantially lower RNA inputs, offers superior specificity and sensitivity, facilitating accurate determination of NAD-capping at transcript-level resolution. The negative controls of the NAD-capped transcript copA had 0 reads while the non-capped transcript serX had almost no background signal (Fig. S27). To evaluate the specificity of HELIOS NAD-Seq relative to NAD captureSeq(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e), we compared the log₂ fold-changes of genes identified by both protocols. As shown in Fig. S28, HELIOS NAD-Seq yields substantially higher log₂ fold-changes, often several orders of magnitude greater, underscoring the high specificity of the ADPRC reaction with the novel 3PAB substrate in positive samples and the absence of biotinylation with 3PA in negative controls.\u003c/p\u003e\n\u003ch3\u003eApplication of HELIOS NAD-Seq to identify dynamic changes in NAD-capped RNAs\u003c/h3\u003e\n\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo demonstrate the advantages of HELIOS NAD-Seq over NAD captureSeq, we performed an experiment that had previously been challenging due to the large RNA input requirements, extensive hands-on time, and limited multiplexing capacity of NAD captureSeq.\u0026nbsp;Total RNA was collected from \u003cem\u003eE. coli\u003c/em\u003e at 30-minute intervals, spanning the stationary, exponential, and plateau phases of the growth curve (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea) resulting in 128 fractions to be processed. After ADPRC reaction and ligation of 8 different barcoded adapters, all 8 samples for each time point (n\u0026thinsp;=\u0026thinsp;16) could be multiplexed, effectively reducing the number of samples to be processed by a factor of 8. Due to this high multiplexing potential, the entire protocol up to Illumina library preparation could be completed within 4 working days. After sequencing, NAD-capped transcripts were identified at each timepoint, by comparing positive samples to negative controls. A dramatic change in NAD-capping levels was observed across the timecourse, with potential NAD-capped gene species counts increasing more than 14-fold from 0.5 hours to 3.5 hours after inoculation reaching a maximum of 654 transcripts at the 3.5 hours mark, compared to 46 transcripts at the 0.5 hour mark. This peak coincides with the early exponential phase of \u003cem\u003eE. coli\u003c/em\u003e growth (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea).\u003c/p\u003e\u003cp\u003eBetween 3.5 and 4 hours post inoculation, NAD-capping levels dropped sharply (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea), potentially marking the onset of overflow metabolism shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb, which is discussed in detail in the Discussion section. KEGG pathway analysis of NAD-capped transcripts at individual timepoints revealed that many genes at 3.5 hours were involved in metabolic pathways, including biosynthesis of secondary metabolites, biosynthesis of amino acids, carbon metabolism, carbon fixation by the Calvin cycle, biosynthesis of various plant secondary metabolites, and cyanoamino acid metabolism (Figure S29). The full list of NAD-capped transcripts detected in least at two timepoints, including gene name, category, maximum baseMean, maximum log₂ fold change, and lowest adjusted p-values, is provided in Table S8.\u003c/p\u003e\u003cp\u003eTranscript start site (TSS) mapping analysis revealed a sharp enrichment at the +\u0026thinsp;1 position for 46.9% of the positive samples for reads that map to the \u003cem\u003eE. coli\u003c/em\u003e genome and 67.2% for reads that map to RNAI of the pUC19 plasmid. Four representative TSS profiles are shown in Figure S30. Abundant alternative start sites identified in Table S4 may, in part, contribute to positive samples displaying additional peaks. Furthermore, motif analysis consistently identified a \u0026minus;\u0026thinsp;12 to \u0026minus;\u0026thinsp;7 promoter element (TANNNT) in the upstream regions of NAD-capped transcripts in 90.6% of positive samples. Four representative motif analysis plots are presented in Figure S31.\u003c/p\u003e\u003cp\u003eTo quantify changes in NAD-capping across timepoints, read counts were normalized for sequencing depth and further adjusted using a set of stably expressed reference tRNAs which will be referred to as the \u0026ldquo;reference set\u0026rdquo; (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec). These genes exhibited uniform expression across timepoints, as shown by three representative species that remained stable in contrast to dynamic transcripts such as RNAI (Figure S32a). The resulting normalized read counts allowed accurate comparison of NAD-capped transcript abundance across the growth curve of \u003cem\u003eE.coli\u003c/em\u003e.\u003c/p\u003e\u003cp\u003eTo validate the normalization strategy, two internal standards (IS-036 and IS-060) were selected based on their consistent capture at the +\u0026thinsp;1 TSS position across all four positive samples and timepoints. Two representative TSS plots for each internal standard are shown in Figure S33. Application of the two-step normalization strategy led to a marked reduction in the coefficient of variation (%CV) of their read counts across all 16 timepoints. The reduction in %CV indicates that the two-step normalization strategy effectively compensates for differences in read depth and minor expression changes, resulting in uniform capture of the two internal standards across timepoints. For IS-036, the %CV decreased from 106.67% to 43.46%, and for IS060, from 86.83% to 26.56%, with the most substantial improvement observed after the second normalization step (Figure S32b). These results support the validity of the normalization approach for quantifying fluctuations in NAD-capped RNA abundance throughout the \u003cem\u003eE. coli\u003c/em\u003e growth curve. For a detailed description of the normalization steps see Methods section: \u003cem\u003e\u0026ldquo;HELIOS NAD-Seq \u0026ndash; Normalization in time course experiment\u0026rdquo;\u003c/em\u003e.\u003c/p\u003e\u003cp\u003eRNAI, the most abundant NAD-capped RNAI encoded on the pUC19 plasmid, was used to evaluate the effectiveness of the normalization strategy. The resulting normalized RNAI read counts from HELIOS NAD-Seq closely mirrored the NAD-RNA abundance profile observed via northern blot on APB-PAGE, with probes 5\u0026prime;-end\u0026ndash;labeled using γ-[\u0026sup3;\u0026sup2;P]-ATP (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed, Figure S34). Both ppp-RNAI and NAD-RNAI were highest at 0.5 hours post inoculation and lowest at 3.5 hours post inoculation. Capping ratios exhibited a similar pattern despite smaller differences between the two timepoints, with a roughly 3-fold reduction, as compared to the 7-fold reduction of ppp-RNAI and 22-fold reduction of NAD-RNAI (Figure S35). Both approaches revealed high RNAI levels at the 0.5 hour mark, a dramatic decrease during the early and late exponential phases, and a subsequent increase during the plateau phase. RNAI displayed expression patterns that did not fit any of the expression patterns of NAD-capped gene species that map to the endogenous \u003cem\u003eE.coli\u003c/em\u003e genome (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee,f).\u003c/p\u003e\u003cp\u003eTo identify NAD-capped RNAs with dynamic expression profiles across the growth curve, gene species that give rise to potential NAD-capped RNAs in at least 2 of the 16 timepoints were selected (n\u0026thinsp;=\u0026thinsp;257). Following the two-step normalization, mean z-scores were calculated for each gene across timepoints and visualized in a heatmap, where z-score represents the normalized expression level of a transcript, scaled relative to their mean and standard deviation, allowing comparison of dynamic patterns independent of absolute abundance (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee). The resulting heatmap revealed three distinct expression clusters. Cluster 1, comprising of 122 out of a total of 257 genes, contained genes with high NAD-capping during the stationary and early exponential phases, followed by a steep decline in the late exponential phase and a subsequent rise in the plateau phase. Cluster 1 revealed enrichment in 4 metabolic pathways including biosynthesis of secondary metabolites, ether lipid metabolism, folate biosynthesis, and biosynthesis of cofactors (Figure S36a). Cluster 2, comprising 94 genes, showed a sharp peak in the early exponential growth phase of \u003cem\u003eE.\u003c/em\u003e coli followed by a similar pattern as observed in cluster 1. Cluster 2 revealed enrichment of 3 metabolic pathways (Figure S36b). Cluster 3, comprising 41 genes, displayed a biphasic pattern, with peaks during both the exponential and plateau phases. But with its low number of potential NAD-capped transcripts it displayed only tRNA biosynthesis as the enriched pathway (Figure S36c). The strong temporal peak at the early exponential phase for all 3 clusters may potentially align with a period of maximal NAD availability driven by high cell division(\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e).These cluster-specific trends are further summarized in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef, where the average z-scores of all the genes within the cluster across timepoints are plotted for each cluster. A complete list of gene names, functional categories, KEGG pathways, and the cluster in which they were detected is provided in Table S9.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn our recent review, we established a matrix to evaluate NAD-RNA identification protocols based on yield, specificity, evaluability, and throughput(\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e). An ideal next-generation method should detect NAD-RNAs with minimal degradation and off-target effects, even from low-input samples, while enabling robust and scalable analysis adaptable to complex experimental settings. Additionally, a parallelization of protocol steps is needed to increase the throughput of biological samples and to allow the transition to new and more complex use cases.\u003c/p\u003e\u003cp\u003eHELIOS NAD-Seq addresses these criteria and demonstrates significant improvements over existing methods. It maintains the versatility of NAD captureSeq while advancing the 4 benchmark criteria(\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e). HELIOS NAD-Seq detects NAD-RNAs from at least 50-fold less total RNA compared to NAD captureSeq and NAD tagSeq(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e) with reduced degradation achieved through a 10-fold lower ADPRC concentration. Specificity is enhanced, eliminating off-target modifications of 5\u0026prime;-caps such as m\u003csup\u003e7\u003c/sup\u003eG and enabling the identification of hundreds of additional low-abundance NAD-capped transcripts. Throughput is further increased by early barcoding, which provides an 8-fold higher multiplexing capacity and reduces hands-on time. Evaluability is improved in qualitative terms, reflected by a more stringent adjusted p-value threshold (0.01 versus 0.05 for NAD captureSeq), though quantitative assessment remains limited, as HELIOS NAD-Seq cannot yet determine the proportion of the total transcript pool that is NAD-capped.\u003c/p\u003e\u003cp\u003eA key factor behind its performance is the switch from \u003cem\u003eO\u003c/em\u003e-nucleophilic hydroxyl to \u003cem\u003eN\u003c/em\u003e-nucleophilic pyridyl substrates\u0026mdash;specifically, the highly reactive 3PAB\u0026mdash;mimicking the natural substrate of \u003cem\u003eAplysia californica\u003c/em\u003e ADP-ribosyl cyclase. This enables rapid, high-yield biotinylation of NAD\u003csup\u003e+\u003c/sup\u003e and NAD-capped RNAs via ADPRC transglycosylation, with full labeling achieved in about three minutes\u0026mdash;a significant improvement over the 30-minute reaction time previously required with 4-pentynol and 3-azido-1-propanol (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). Importantly, the absence of divalent cations (e.g., Mg\u0026sup2;⁺), which can promote RNA hydrolysis, further preserves RNA integrity. The reactions are highly specific for NAD caps, with no off-target labeling of RNAs bearing m\u003csup\u003e7\u003c/sup\u003eG caps, as confirmed by specificity assays (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea, S16). The conditions also allow for a 10-fold reduction in enzyme concentration, minimizing degradation even with low RNA inputs.\u003c/p\u003e\u003cp\u003eDespite widespread use of ADPRC in NAD-RNA detection protocols, characterizations have mostly focused on dinucleotide reactions with NAD\u003csup\u003e+\u003c/sup\u003e, leaving a knowledge gap regarding its activity on NAD-capped oligonucleotides(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). To address this, we developed in vitro assays to rigorously analyze the kinetics and specificity of ADPRC transglycosylation on NAD-capped RNAs. These revealed that, using 3PAB, a 70-fold lower substrate concentration compared to prior methods suffices, with total RNA inputs reduced by over 50-fold relative to NAD captureSeq(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan additionalcitationids=\"CR40\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn earlier protocols, negative controls were often performed without ADPRC(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). We can now confirm that the treatment of negative controls with ADPRC and 3-picolylamine in HELIOS NAD-Seq better mimics the conditions in the sample reactions. Our approach with early pooling and barcoded adapters minimizes bias and enables multiplexing of up 128 samples, for 16 biological conditions (including replicates and controls) in parallel\u0026mdash;an eightfold increase in throughput.\u003c/p\u003e\u003cp\u003eIn HELIOS NAD-Seq, hydrophilic streptavidin magnetic beads are used for capture and purification, offering lower nonspecific binding and milder washing conditions than previous methods (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e). Our investigations showed the optimal balance between biotinylation efficiency and binding strength with the 3PAB substrate, which may also facilitate future direct NAD-RNA sequencing via nanopores.\u003c/p\u003e\u003cp\u003eDownstream processing was adapted for low input, streamlining library preparation to be completed within four working days. Validation in \u003cem\u003eE. coli\u003c/em\u003e confirmed our method's superior sensitivity, recovering nearly all previously identified NAD-capped transcripts and uncovering hundreds of new ones, often originating from alternative transcription start sites. By exploiting its multiplexing capacity, we captured NAD-capping dynamics across 16 timepoints of the \u003cem\u003eE. coli\u003c/em\u003e growth curve, revealing distinct expression clusters with sharp peaks during early exponential growth and declines in later phases. Functional enrichment of these clusters identified many NAD-capped species to be related to metabolic pathways, underscoring its potential role in coordinating transcriptional output with metabolic state.\u003c/p\u003e\u003cp\u003eThe temporal pattern of Clusters 1 and 2 may suggest a link between NAD-capping and the metabolic state of the cell. The dramatic peak in the number of potential NAD-capped species may reflect elevated NAD\u003csup\u003e+\u003c/sup\u003e and NADH availability during active cell division, consistent with previous studies using dynamic single-cell fluorescence measurements(\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). The sharp drop in NAD-RNA levels after the early exponential phase may coincide with the onset of overflow metabolism(\u003cspan additionalcitationids=\"CR54\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e)\u0026mdash;a well-characterized shift in \u003cem\u003eE. coli\u003c/em\u003e from aerobic respiration to fermentation during rapid growth. This phenomenon, analogous to the Warburg effect in cancer cells, is thought to arise from multiple factors, including limitations in the respiratory capacity of the inner membrane(\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e), redirection of proteomic resources away from energy-efficient pathways(\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e), and redox signaling through the NADH/NAD⁺ ratio(\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). Importantly, aerobic respiration coupled with the TCA cycle produces approximately 8 NADH molecules per glucose, which are efficiently converted back to 8 NAD⁺ via the electron transport chain. In contrast, fermentation yields only 2 NAD⁺ per glucose, which may drastically reduce the total available NAD⁺ pool (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). Thus, the metabolic switch to fermentation may lead to a decline in NAD⁺ availability, potentially reducing the rate of NAD-capping during RNA transcription. This mechanistic link may explain the pronounced decrease in NAD-RNA abundance between 3.5 hours and 4 hours post inoculation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). The temporal expression pattern of RNAI (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed) are aligned to observations of decreased plasmid concentration at fast growth rates(\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e). Jasiecki et al. also showed that RNAI degradation rates are lower at higher growth rates(\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e) which could potentially explain why NAD-capping rates are lower during the early to late exponential growth phases of the \u003cem\u003eE.coli\u003c/em\u003e.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, HELIOS NAD-Seq provides a highly sensitive and specific platform for NAD-RNA detection, enabling robust profiling from minimal RNA input with reduced degradation and negligible off-target effects. By integrating early barcoding and optimized chemistry, the method achieves high multiplexing capacity and improved throughput, while uncovering both known and novel NAD-capped transcripts. Application to \u003cem\u003eE. coli\u003c/em\u003e revealed dynamic capping patterns linked to growth phase and metabolic state, suggesting a connection between NAD\u003csup\u003e+\u003c/sup\u003e availability, overflow metabolism, and RNA regulation. These advances establish HELIOS NAD-Seq as a versatile tool for investigating the biological roles of NAD-capping and its contribution to cellular physiology.\u003c/p\u003e\n\n\n\n"},{"header":"Methods","content":"\u003cp\u003e\u003cb\u003eStock preparation PAB conjugates.\u003c/b\u003e Stock solutions (300 mM) of 3PAB, 4PAB, 3PADB, 4PADB, N(PEG)\u003csub\u003e3\u003c/sub\u003eB, and N(PEG)\u003csub\u003e11\u003c/sub\u003eB were prepared from pure solid, dissolved in DMSO and stored as 100 µL aliquots at − 20°C.\u003c/p\u003e\u003cp\u003e\u003cb\u003eADPRC transglycosylation reactions on the dinucleotide level\u003c/b\u003e. Standard ADPRC transglycosylation reactions were performed in the presence of 5 mM NAD\u003csup\u003e+\u003c/sup\u003e (or 5 mM NADH), 15 mM 3PAB (or other \u003cem\u003eN\u003c/em\u003e-nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 1.3 µM ADPRC enzyme (380 nM for product identification reactions). Buffer tests were performed in the presence of 1 mM NAD\u003csup\u003e+\u003c/sup\u003e, 1.5 mM 4PAB, 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), either with 5 mM MgCl\u003csub\u003e2\u003c/sub\u003e or 5 mM EDTA), 5% DMSO, and 1.3 µM ADPRC enzyme. The reactions were started by adding a freshly prepared enzyme dilution in H\u003csub\u003e2\u003c/sub\u003eO (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 30 min. If necessary, the reaction mixtures were diluted to 200 µL with H\u003csub\u003e2\u003c/sub\u003eO, before they were filtered through 10 kDa Amicon centrifugal filters and analyzed by RP-HPLC (gradient: 1–50% ACN in 50 min). For the buffer tests, 3 kDa Amicon centrifugal filters were used to prepare for RP-HPLC analysis (gradient: 2–5% buffer B in 20 min, then 15–30% in 10 min, then 30–70% in 20 min, total run time of 50 min). Collected peak solutions were analyzed by HR-MS.\u003c/p\u003e\u003cp\u003e\u003cb\u003eKinetic Analysis of ADPRC transglycosylation reactions on the dinucleotide level\u003c/b\u003e. ADPRC transglycosylation reactions were performed in the presence of 5 mM NAD\u003csup\u003e+\u003c/sup\u003e, 15 mM 3PAB (or other \u003cem\u003eN\u003c/em\u003e-nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 1.3 µM ADPRC enzyme. The reaction was started by adding a freshly prepared enzyme dilution in H\u003csub\u003e2\u003c/sub\u003eO (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 30 min. After predefined reaction times (8 s, 16 s, 25 s, 35 s, 45 s, 60 s, 90 s, 120 s, 180 s, 300 s), 30 µL aliquots were removed from the reaction mixture and quenched by addition to a mixture of 170 µL H\u003csub\u003e2\u003c/sub\u003eO and 400 µL P/C/I solution, followed by vigorous mixing on a vortex mixer. A sample termed »0 s« was removed before starting the reaction. Extractions with P/C/I and diethyl ether were performed, followed by removal of ether traces in an Eppendorf concentrator, before the samples were filtered through Amicon centrifugal filters (≤ 10 kDa) and analyzed by HPLC (gradient: 1–50% ACN in 50 min) in separate runs for each time-point. The areas of NAD\u003csup\u003e+\u003c/sup\u003e and product peaks in the chromatogram for 260 nm absorption were calculated using the ChemStation analysis software.\u003c/p\u003e\u003cp\u003e\u003cb\u003eStrepShift assay\u003c/b\u003e. For the electrophoretic mobility shift assay (EMSA) of biotin-modified RNA (usually after ADPRC transglycosylation) with streptavidin (streptavidin gel-shift assay), precipitated RNA samples containing biotinylated RNA species were standardly incubated in the presence of streptavidin in aqueous solution. In general, 6 µL of RNA sample (~ 600 ng RNA, for the kinetic analysis, for example, a mixture of around 570 ng \u003cem\u003eE. coli\u003c/em\u003e total RNA and 30 ng model RNA) were incubated with 3 µL streptavidin (1 µg/µL) at 37°C for 15 min. For each analysis, a positive control with known biotinylation level was treated in addition to the experiment samples. Next, 7.5 µL of the RNA-streptavidin incubation mixture were added to 1.5 µL of agarose gel loading buffer (6x) and gel electrophoresis performed on a 2% agarose gel. For the analysis of the ADPRC transglycosylation reaction, the contained RNA species were additionally visualized using the Northern blotting technique.\u003c/p\u003e\u003cp\u003e\u003cb\u003eNorthern blot\u003c/b\u003e. After agarose gel electrophoresis, a capillary transfer setup was used for transfer of nucleic acids onto a nylon membrane. For this, a small pedestal with a sheet of Whatman chromatography paper that reached into a reservoir of 0.5x TBE buffer and five more gel-sized sheets of Whatman chromatography paper (soaked in 0.5x TBE buffer) on top was prepared. On top of this, the agarose gel was placed, followed by a gel-sized nylon membrane (soaked in 0.5x TBE buffer), while carefully avoiding or removing air bubbles between gel and membrane. This was topped off with five more gel-sized sheets of Whatman chromatography paper (soaked in 0.5x TBE buffer), a stack of dry paper towels and a suitable weight. With this setup, capillary transfer was performed overnight at room temperature. After this, the nylon membrane was dried at air and transferred RNA species were photo cross-linked to the membrane surface in a UV chamber. The membrane was then blocked with ROTI Hybri-Quick in a hybridization tube (42°C, 1 h, rotating) before the hybridization solution was exchanged and typically 5 µL of a radioactively labeled, complementary RNA probe was added, which was followed by overnight incubation (42°C, rotating). The membrane was rinsed, washed, air-dried, and then placed on a storage phosphor screen, which was used for readout after incubation for between 6 h and 3 days at room temperature.\u003c/p\u003e\u003cp\u003e\u003cb\u003eADPRC transglycosylation reactions on the RNA level\u003c/b\u003e. Standard ADPRC transglycosylation reactions were performed in the presence of 113 nt model RNA (~ 1:1 mixture of NAD/pppRNAIII leader), which was applied in a total amount of 5 µg (pure model RNA) or 0.5 µg (mixed with 4.5 µg \u003cem\u003eE.coli\u003c/em\u003e total RNA), 15 mM 3PAB (or other \u003cem\u003eN\u003c/em\u003e-nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 38 nM ADPRC enzyme, in a total reaction volume of 300 µL. Reactions assessing the potential of N(PEG)\u003csub\u003ex\u003c/sub\u003eB derivatives were performed accordingly, but in the presence of 10 µg model RNA and 380 nM ADPRC enzyme, in a total reaction volume of 100 µL. The reactions were started by adding a freshly prepared enzyme dilution in H\u003csub\u003e2\u003c/sub\u003eO (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 30 min. The reactions were quenched by addition of 300 µL P/C/I (optionally with additional 200 µL H\u003csub\u003e2\u003c/sub\u003eO for lower reaction volumes), or aliquots of 30 µL were removed from the reaction mixture and added to a mixture of 270 µL H\u003csub\u003e2\u003c/sub\u003eO and 300 µL P/C/I. Following extractions with P/C/I and diethyl ether and the removal of ether traces in an Eppendorf concentrator, the RNA samples were recovered by isopropanol precipitation and the nucleic acid concentrations were determined by UV-Vis spectroscopy, before a StrepShift assay was performed.\u003c/p\u003e\u003cp\u003e\u003cb\u003eKinetic Analysis of ADPRC transglycosylation reactions on the RNA level\u003c/b\u003e. ADPRC transglycosylation reactions were performed in the presence of 2.5 ng/µL of a 113 nt model RNA (RNAIII leader, 47.9% NAD-modified), 47.5 ng/µL \u003cem\u003eE. coli\u003c/em\u003e total RNA, 15 mM 3PAB (or other \u003cem\u003eN\u003c/em\u003e-nucleophilic substrate), 1x ADPRC buffer (50 mM Na-HEPES (pH 7.0), 5 mM EDTA), 5% DMSO, and 38 nM ADPRC enzyme, in a total reaction volume of 300 µL. The reaction was started by adding 150 µL of a freshly prepared enzyme dilution in H\u003csub\u003e2\u003c/sub\u003eO (preheated to 37°C) to an equal volume of a 2x master mix of reactants and reagents (preheated to 37°C), followed by incubation at 37°C for 45 min. After predefined reaction times (0.5 min, 1 min, 2 min, 3 min, 5 min, 10 min, 20 min, 30 min, 45 min), 30 µL aliquots were removed from the reaction mixture and quenched by addition to a mixture of 270 µL H\u003csub\u003e2\u003c/sub\u003eO and 300 µL P/C/I solution, followed by vigorous mixing on a vortex mixer. A sample termed »0 s« was removed before starting the reaction. Extractions with P/C/I (three times) and diethyl ether (twice) were performed, followed by removal of ether traces in an Eppendorf concentrator. 3 M NaOAc (pH 5.5) to reach a final concentration of 0.3 M and 1 µL glycogen (RNA-grade) were added, and the RNA samples were recovered by isopropanol precipitation. The nucleic acid concentration of the recovered samples was determined by UV-Vis spectroscopy, before a StrepShift assay was performed. The contained RNA species were transferred to nylon membranes and probed with a radioactively-labeled RNAIII leader probe using the Northern blotting technique. Signal intensities recorded for shifted and non-shifted RNAIII leader species at each time-point of the reaction were analyzed and plotted.\u003c/p\u003e\u003cp\u003e\u003cb\u003eADPRC transglycosylation specificity tests on the RNA level\u003c/b\u003e. For the treatment of differently capped RNA species in an ADPRC transglycosylation setup, several IS-036 model RNA stocks with defined 5’-end structures and concentrations of 200 ng/µL were prepared: a) 5’-pppA, b) 1:1 5’-GpppA and 5’-pppA, c) 1:1 5’-m\u003csup\u003e7\u003c/sup\u003eGpppA and 5’-pppA, d) 1:1 5’-NppA (NAD) and 5’-pppA, e) 1:19 5’-NppA (NAD) and 5’-pppA. From these stocks, 50 µL solutions containing 5 µg IS-036 model RNA, 30 mM 3PAB, and 2x ADPRC buffer (with EDTA) were prepared, from which two samples à 5 µL were removed and diluted with 5 µL H\u003csub\u003e2\u003c/sub\u003eO each. One of those samples was stored as a negative control, the other was incubated at 37°C for 30 min without addition of enzyme (»0 s« sample). To the residual volume of 40 µL (for each of the RNA mixes), an equal volume of ADPRC enzyme dilution (76 nM) was added to start the reaction. After pre-defined reaction times (0.5 min, 1 min, 2 min, 3 min, 5 min, 10 min, 20 min, 30 min), 10 µL aliquots were removed from the reaction mixture and quenched by addition to a mixture of 290 µL H\u003csub\u003e2\u003c/sub\u003eO and 300 µL P/C/I solution, followed by vigorous mixing on a vortex mixer. Extractions with P/C/I (three times) and diethyl ether (twice) were performed, followed by removal of ether traces in an Eppendorf concentrator. 3 M NaOAc (pH 5.5) to reach a final concentration of 0.3 M was added, and the RNA samples were recovered by isopropanol precipitation (due to the low RNA amount of around 500 ng per sample, the vessels were stored at − 20°C for 3 days). The nucleic acid concentration of the recovered samples was determined by UV-Vis spectroscopy and fluorescence spectroscopy. Then, dot blot experiments were performed in technical quadruplicates, using 1 µL (~ 50 ng IS-036 model RNA) per sample and time-point, and probing with streptavidin-HRP conjugate.\u003c/p\u003e\u003cp\u003e\u003cb\u003eCapture and Elution of Biotinylated RNA using Streptavidin Magnetic Beads\u003c/b\u003e. The preparation of biotinylated RNA species for this experiment was performed by ADPRC transglycosylation of 20 µg of a 113 nt model RNA (~ 1:1 mixture of NAD/pppRNAIII leader) under standard reaction conditions in the presence of 15 mM 3PAB (or other \u003cem\u003eN\u003c/em\u003e-nucleophilic substrate) and 380 nM ADPRC enzyme, in a total reaction volume of 200 µL. The reaction was quenched and extracted after 5 min of incubation at 37°C following the standard protocol. After RNA sample recovery, the biotinylation levels were determined using a StrepShift assay. For the capture on streptavidin magnetic beads, 10 ng of the partially biotinylated model RNAs were mixed with 390 ng \u003cem\u003eE. coli\u003c/em\u003e total RNA. Hydrophilic streptavidin magnetic beads were prepared according to the manufacturer’s instructions. 25 µL of bead suspension each were transferred to PCR tubes in an 8-tube stripe. Using a magnetic stand, the bead storage solution was removed, and the beads were washed three times with 100 µL of 1x immobilization buffer, before they were each blocked by incubation at 20°C for 30 min with a solution of 100 µg/mL acetylated BSA in 1x immobilization buffer, followed by three further washes with 1x immobilization buffer. In general, upon the addition of a new solution to the beads, the suspension was mixed by pipetting up and down five times. Then, the prepared mixes of model RNA and total RNA, with a total volume of 25 µL each, were added to the beads and incubated at 20°C for 1 h. The supernatant solution, as well as the solutions of three washes with 100 µL streptavidin wash buffer (containing 8 M urea, or 1 M urea for some PADB-reacted samples) and one wash with 100 µL H\u003csub\u003e2\u003c/sub\u003eO, were collected. The elution of bound RNA was performed first by incubation of the beads with 25 µL buffered solution (1 mM EDTA, 10 mM Tris-HCl, pH 7.5, optionally containing 0.4 mM biotin) at different temperatures of (40°C, 60°C, or 80°C) for 10 min. Optionally, this was followed by an elution step with 25 µL buffered biotin solution at 80°C for 10 min in order to fully remove bound biotinylated RNA. The solutions of all elution steps were collected. The contained RNA was analyzed by dot blot experiments that, where applicable, were performed in technical triplicates, using 1 µL per collected sample (1.5 µL for elution samples in the elution condition optimization experiment), and probing with radioactively labeled RNA probes complementary to either RNAIII leader, or \u003cem\u003eE. coli\u003c/em\u003e ribosomal RNAs (5S, 16S, 23S).\u003c/p\u003e\u003cp\u003e\u003cb\u003eDot blot analysis\u003c/b\u003e. Typically, 1 µL aliquots of RNA-containing, aqueous solutions were spotted onto a nylon membrane. After drying at air, RNA species were photo-crosslinked to the membrane surface in a UV chamber. Probing with radioactively labeled, complementary RNA probes was performed following a Northern blotting technique. Probing with streptavidin-HRP conjugate was performed according to the manufacturer’s instructions, after blocking the nylon membrane with 3% BSA fraction V in 1x TBS-T. After removing the blocking solution, a labeling solution of 1x TBS-T containing 42 ng/mL streptavidin-HRP conjugate was added, followed by incubation for 30 min (room temperature, rotating). After rinsing, washing and air drying, the membrane was stained with SignalFire ECL Plus reagent according to the manufacturer’s instructions. After readout at a ChemiDoc Imager device, signal intensities were determined using the ImageLab software.\u003c/p\u003e\u003cp\u003e\u003cb\u003eHELIOS NAD-Seq – standard protocol\u003c/b\u003e. The protocol is standardly performed in a total of 4 working days. \u003cem\u003eDay 1\u003c/em\u003e: HELIOS NAD-Seq typically starts from total RNA samples in quadruplicates, which have been extracted from an organism of choice, treated by DNase I, and recovered by ethanol or isopropanol precipitation. For each replicate (numbered as 1, 2, 3, 4, \u003cem\u003eet cetera\u003c/em\u003e), a 20.5 µL total RNA mix containing 4.1 µg of total RNA and a desired amount of internal RNA standards (e.g. 1 fmol) is prepared. For the initial ADPRC transglycosylation with 3PAB, a diluted ADPRC stock is prepared, containing 5 mM HEPES (pH 7.0), 0.5 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 5% glycerol, and 380 nM ADPRC enzyme (100 µL suffice for 5 replicate samples plus negative controls). Then, for each replicate, a 50 µL sample (S) mix containing 10 µL total RNA mix, 30 mM 3PAB, and reagents (2x ADPRC buffer with EDTA, 10% DMSO), as well as a 50 µL negative control (N) mix containing 10 µL total RNA mix, 30 mM 3PA, and reagents (2x ADPRC buffer with EDTA, 10% DMSO) are prepared. With the descriptors for positive samples (S) and negative controls (N), and the numbering of the replicates as 1, 2, 3, 4, \u003cem\u003eet cetera\u003c/em\u003e, a set of mixes termed S1, S2, S3, S4, \u003cem\u003eet cetera\u003c/em\u003e, and N1, N2, N3, N4, \u003cem\u003eet cetera\u003c/em\u003e, is obtained. The vessels containing S and N mixes (50 µL), as well as an equal number of vessels containing 40 µL H\u003csub\u003e2\u003c/sub\u003eO, are pre-heated to 37°C. Then, 10 µL of the ADPRC dilution are mixed with a pre-heated water aliquot (40 µL), and the full 50 µL are added to the pre-heated reaction mix S1 and mixed by pipetting, before further incubation at 37°C. This procedure is repeated for reaction mix S2 after either 30 s or 1 min. With this, after either 3:30 min or 7:00 min, eight ADPRC transglycosylation reactions (for S1, S2, S3, S4, N1, N2, N3, and N4) have been started, all either 30 s or 1 min apart. After 20 min of incubation at 37°C, a mixture of 200 µL H2O and 300 µL P/C/I solution is added to quench the S1 reaction, followed directly by vigorous mixing using a vortex mixer. This is again repeated for the other ADPRC transglycosylation reaction mixes in the same sequence as performed for starting the reactions, with either 30 s or 1 min in between. Aqueous and organic phases are separated by centrifugation (13000 rpm, 2 min, room temperature) and the aqueous phase is treated by two more extractions with P/C/I (300 µL each), and, after transferring the aqueous phase to a 1.5 mL LoBind tube, followed by two extractions with diethyl ether (300 µL each). After discarding the organic phase after the last extraction and centrifugation step, residual traces of ether are removed in an Eppendorf concentrator (~ 3 min, room temperature). Around 270 µL of aqueous phase should typically remain for each replicate sample (S1, S2, S3, S4) and negative control (N1, N2, N3, N4), which are then each mixed with 30 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, with 1 volume (300 µL) of cold isopropanol (− 20°C). The ADPRC-reacted RNA samples are precipitated at − 20°C for 2 h, and recovered by centrifugation (14900 g, 90 min, 4°C). The formed pellets are washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before they each are dissolved in 10 µL H\u003csub\u003e2\u003c/sub\u003eO. For 3’-adapter ligation with barcoded adapters, 75 µL of a 2.4x ligation master mix (2.4x standard ligation buffer, 120 µg/mL acetylated BSA, 12 mM DTT, 24% DMSO, 2 U/µL RNaseOUT, 0.6 U/µL RNA ligase 1, 12 U/µL RNA ligase 2 (truncated, K227Q)) are prepared for the treatment of eight samples. For eight samples, e.g. AD3-id4-bc01 through -bc08 can be used. To 1 µL of each required adenylated 3’-adapter (~ 100 µM total concentration, and 50 µM pre-adenylated) are added 0.7 µL H\u003csub\u003e2\u003c/sub\u003eO and 8.5 µL of the 2.4x ligation master mix. From the resulting 2x adapter/ligation master mixes, 10 µL are added to the 10 µL of ADPRC-reacted RNA samples (e.g. bc01 for S1, bc02 for S2, bc03 for S3, bc04 for S4, bc05 for N1, bc06 for N2, bc07 for N3, and bc08 for N4), which are incubated at 4°C overnight. \u003cem\u003eDay 2\u003c/em\u003e: Hydrophilic streptavidin magnetic beads are removed from the storage at 4°C and allowed to incubate at room temperature for around 30–45 min, after which the container is carefully agitated to create a homogeneous suspension. 25 µL of magnetic beads per PCR tube are transferred using a 200 µL pipette tip that was cut off with a scalpel to increase the diameter of the opening. Then, 25 µL of 1x immobilization buffer is added to each PCR tube, the beads are suspended by pipetting up and down for five times, and the tubes are placed on a magnetic stand. After a pellet of beads has formed, the supernatant is removed carefully. The described procedure is performed for every addition and removal of a new solution, optionally with an additional incubation time after mixing and before pelleting on a magnetic stand. The beads are washed two additional times with 1x immobilization buffer (50 µL each), before they are incubated at 20°C for 5 min with 25 µL of a bead blocking solution containing 1x immobilization buffer, and 100 µg/mL acetylated BSA. After this incubation, the beads are washed two times with 1x immobilization buffer (50 µL each) before addition of the ADPRC-reacted, 3’-ligated RNA samples. After overnight incubation of the 3’-ligation reaction mixes (S1, S2, S3, S4, N1, N2, N3, N4) with the respective barcoded adapters, the reactions are heat-inactivated at 65°C for 15 min, an equal volume (20 µL) of 2x immobilization buffer is added to each vessel, and the different solutions are combined. For eight samples, a total volume of 320 µL multiplexed solution is obtained, which is mixed and divided equally on the PCR tubes containing BSA-blocked, washed streptavidin magnetic beads. It is recommended to not use more than 180 µL of RNA-containing solution per 25 µL of initially used bead suspension. After incubation at 20°C for 5 min to allow for the binding of biotinylated RNA species to the beads, the supernatant is removed, and 50 µL of 1x immobilization buffer is added. At this point, it is advised to combine the suspensions of two PCR tubes into one. By doing this, the RNA-containing solution of eight samples, which was split before, is unified again after the bead-binding step. The beads are washed once with 1x high salt wash buffer and once with 1x low salt wash buffer (100 µL each). The on-bead reverse transcription is performed by adding 20 µL of RT master mix (5 µM of RTp2 primer, 0.5 mM of each dNTP (dATP, dCTP, dGTP, dTTP), 1x SSIV buffer, 5 mM DTT, 2 U/µL RNaseOUT, and 10 U/µL SuperScript IV Reverse Transcriptase) to the magnetic beads, which, after suspending, is followed by incubation at 55°C for 10 min. After adding an equal volume (20 µL) of 2x immobilization buffer, the bead suspension is further incubated at 20°C for 5 min to allow for the re-binding of biotinylated RNA species in a hybrid with complementary cDNA. The beads are washed once with 1x immobilization buffer, once with 1x high salt wash buffer and once with 1x low salt wash buffer (100 µL each). To hydrolyze the RNA and release the cDNA, a 50 µL solution of 0.15 M NaOH is added, followed by incubation at 55°C for 15 min. After incubation, the supernatant is transferred to a 1.5 mL LoBind tube. The beads are washed once with 50 µL H\u003csub\u003e2\u003c/sub\u003eO and the supernatant is also transferred to the same 1.5 mL LoBind tube. The full volume of 100 µL is then mixed with 11 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, with 2.5 volumes (280 µL) of cold absolute ethanol (− 20°C). The multiplexed cDNA sample is precipitated at − 20°C for 2 h, and recovered by centrifugation (14900 g, 60 min, 4°C). The formed pellet is washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before it is dissolved in 10 µL H\u003csub\u003e2\u003c/sub\u003eO, and transferred to a PCR tube. The C-tailing of the cDNA is performed by addition of an equal volume (10 µL) of a TdT master mix (1 mM CTP, 2x TdT buffer, 1 U/µL TdT enzyme), followed by incubation at 37°C for 20 min. Then, the reaction is heat-inactivated at 70°C for 10 min. The 2nd adapter ligation reaction is performed by addition of an equal volume (20 µL) of a 2nd ligation master mix (10 µM sense cDNA anchor, 10 µM antisense cDNA anchor, 20 µM ATP, 2x T4 DNA ligase buffer, 3 Weiss U/µL T4 DNA ligase), followed by overnight incubation at 4°C. \u003cem\u003eDay 3\u003c/em\u003e: After overnight incubation of the 2nd ligation reaction mix, the volume of 40 µL is diluted to 100 µL with H\u003csub\u003e2\u003c/sub\u003eO, followed by heat-inactivation at 65°C for 10 min, before transfer to a 1.5 mL LoBind tube. The full volume of 100 µL is then mixed with 11 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, with 2.5 volumes (280 µL) of cold absolute ethanol (− 20°C). The multiplexed, fully ligated cDNA sample is precipitated at − 20°C for 2 h, and recovered by centrifugation (14900 g, 60 min, 4°C). The formed pellet is washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before it is dissolved in 20 µL H\u003csub\u003e2\u003c/sub\u003eO. The multiplexed, fully ligated cDNA is PCR-amplified in a 50 µL solution containing 2–8 µL cDNA sample, 2 µM NEBNext Universal Primer, 2 µM NEBNext Index Primer, and 20 µL of a Q5 master mix (0.5 mM of each dNTP (dATP, dCTP, dGTP, dTTP), 2.5x Q5 reaction buffer, 0.05 U/µL Q5 HotStart High-Fidelity DNA Polymerase), with a standard PCR amplification protocol: After initial activation (98°C, 40 s), a sequence of denaturation (98°C, 10 s), annealing (69°C, 20 s), and extension (72°C, 30 s) is performed 9 times, before a sequence of denaturation (98°C, 10 s), annealing (66°C, 20 s), and extension (72°C, 30 s) is performed up to 16 times, followed by a final extension (72°C, 10 min). After PCR amplification, the solution is transferred to a 1.5 mL LoBind tube. The full volume of 50 µL is then mixed with 6 µL NaOAc (3 M, pH 5.5), and, after short incubation at room temperature, with 2.5 volumes (140 µL) of cold absolute ethanol (− 20°C). The DNA sample is precipitated at − 20°C for 30–60 min, and recovered by centrifugation (14900 g, 60 min, 4°C). The formed pellet is washed twice with 70% EtOH (150 µL each), followed by centrifugation (14900 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before it is dissolved in 15 µL H\u003csub\u003e2\u003c/sub\u003eO. For native PAGE purification, an equal volume (15 µL) of 2x native PAGE loading buffer (with bromophenol blue dye) is added to the PCR-amplified DNA samples. This sample is loaded on a freshly prepared native PA gel (50 mL of gel solution are prepared from 16.7 mL Rotiphorese 30% sequencing gel concentrate (29:1), 5 mL TBE buffer (10x), and 28.3 mL H\u003csub\u003e2\u003c/sub\u003eO, and polymerization is set off using 500 µL APS (10%) and 50 µL TEMED), flanked on each side by 1 µL of GeneRuler UltraLow Range DNA ladder, and the native PAGE is run in 1x TBE buffer at 17 W for roughly 2.5 h (typically stopped slightly before the bromophenol blue dye runs out). The gel is removed, submerged in around 300 mL 1x TBE buffer containing 5 µL SYBR Gold, and incubated like this for 5 min. Then, the gel is shortly washed in water, placed between two sheets of transparent plastic tubing, and analyzed on a Typhoon FLA 9500 biomolecular imager (PMT voltage typically 500 V) or another suitable device. The image is printed out according to its actual size, placed under the plastic tubing to fit the gel, and the area between 150 bp and 300 bp (marked by two bands in the DNA ladder) is excised using a clean scalpel. The excised gel pieces are crushed using a suitable method, and mixed with 0.3 M NaOAc solution (pH 5.5) for overnight elution at 20°C (750 rpm shaking). \u003cem\u003eDay 4\u003c/em\u003e: The gel pieces are removed from the overnight elution mix using a 50 mL filter falcon tube and centrifugation (5000 g, 5 min, 20°C). Depending on the total volume, the flow-through is divided equally into 5 mL tubes, to each of which are added 10 µL NaOAc (3 M, pH 5.5), 1 µL of glycogen (RNA-grade), and, after short incubation at room temperature, 2.5 volumes of cold absolute ethanol (− 20°C). The multiplexed DNA libraries are precipitated at − 20°C for 1 h, and recovered by centrifugation (15000 g, 60 min, 4°C). The formed pellets are washed twice with 70% EtOH (150 µL each), followed by centrifugation (15000 g, 10 min, 4°C), and dried in an Eppendorf concentrator (~ 3 min, room temperature), before they are typically dissolved in a total of 20–30 µL H\u003csub\u003e2\u003c/sub\u003eO, and combined in a single 1.5 mL LoBind tube. The size selected DNA libraries are analyzed using a Qubit spectrofluorometer and a suitable kit according to the manufacturer’s instructions, in order to determine the concentration (in ng/µL) of the stock solution, and using a Bioanalyzer device, a suitable kit and electrophoresis chip according to the manufacturer’s instructions, in order to determine the average size (in bp) of the multiplexed libraries. Both obtained values are further used to determine the stock concentration (in nM), for example with the Promega Biomath Calculator web tool. The library is then typically adjusted to a volume of 20 µL with a DNA concentration of 10 nM (depending on the specific requirements for NGS), and ready for analysis by next-generation sequencing at a sequencing facility of choice.\u003c/p\u003e\u003ch3\u003eHELIOS NAD-Seq – Bioinformatic analysis\u003c/h3\u003e\u003cp\u003eRaw paired-end reads were demultiplexed using Cutadapt (vX.X) with custom barcode sequences corresponding to samples bc01–bc08, using degenerate barcode patterns (e.g., TCAAGTNNNNNNG) for both Read 1 and Read 2 with up to eight random nucleotides (Figure S17a). Demultiplexing was performed by assigning reads to barcodes based on a 5’ overlap with a minimum overlap of 12 nt (-O 12), applying the -g and -A options to match barcode–adapter sequences on forward and reverse reads, respectively. To remove 5’-end leading Gn introduced during NAD capture adapter ligation, an in-house python script was used to trim G-rich patterns from the 5′ ends of forward reads via regular expressions (e.g., ^[GN]*), discarding reads shorter than 18 nt after trimming. Further quality and adapter trimming was performed with Trimmomatic (v0.39) in paired-end mode using the parameters ILLUMINACLIP:TruSeq3-PE.fa:1:30:15, SLIDINGWINDOW:4:20, and MINLEN:18, retaining both paired and unpaired reads. Processed reads were then subjected to a three-stage hierarchical alignment pipeline using Bowtie2 (v2.4.5): first, alignment to a custom spike-in RNA reference (spikeRnas) in end-to-endmode, retaining only unaligned reads via --un-conc and --un; second, alignment of spike-in-depleted reads to an E. coli tRNA + rRNA reference index in local mode, again retaining only unaligned reads; and third, alignment of remaining reads to the E. coli K-12 genome (NC_00913.3 including pUC19c plasmid) in local mode, generating SAM files for paired-end, R1 singleton, and R2 singleton reads. SAM files from the genome alignment step were then filtered with a custom python script to retain only alignments whose read sequence began with adenosine (A) at the 5’ end (the first base of the SEQ field in the SAM file). To quantify transcription from intergenic regions, a custom GTF annotation file was generated from the E. coli reference annotation using an in-house python script that identified genomic intervals between annotated genes on the same strand and chromosome, emitting each as a feature of type \"intergenic\". Read counts were obtained using featureCounts (v2.0.6, Subread package) with the -t intergenic option and the custom GTF file as the annotation reference; paired-end files were processed with the -p flag to ensure fragment-level counting, whereas single-end files were processed without it. Differential expression analysis between positive NAD-capture samples and matched negative controls was then performed separately for each timepoint using PyDESeq2 (v0.4.10) to identify transcripts significantly enriched in the positive fraction with an adjusted p value of below 0.01 and Log2 fold-change (sample/control) of above 3, which were considered NAD-capped candidates. To assess transcription start site (TSS) positions of NAD-capped RNAs, read start distributions were computed from SAM alignment files. For reads mapping to the \u003cem\u003eE. coli\u003c/em\u003e genome and puc19C plasmid, a custom python script parsed gene intervals from the reference GTF and counted reads based on their position relative to the annotated TSS, stratified by strand. Reads were binned by 1-based distance from the TSS defined by the genome assembly ASM584v2 (up to 30 bp both upstream and downstream), and the resulting distributions were visualized as bar plots per barcode and timepoint. For internal standard spike-in controls, a separate script was used to calculate the read start position (TSS) per reference sequence without gene annotations. Positions were aggregated across all reads in each timepoint and barcode, and TSS distributions were plotted individually for each spike-in reference. To visualize nucleotide preferences around transcription start sites (TSS), sequence logos were generated using a custom Python script in combination with WebLogo (v3). Filtered reads were exported from SAM files to FASTA format, with each header containing the alignment count and genomic location. Sequences were grouped by chromosome and replicated according to their alignment counts to reflect read abundance. Each sequence was then trimmed or padded to a uniform window of 41 nucleotides, corresponding to positions − 20 to + 20 relative to the TSS. For each chromosome in each sample, a weighted FASTA file was created and used as input for WebLogo.\u003c/p\u003e\u003ch3\u003eHELIOS NAD-Seq – Normalization in time course experiment\u003c/h3\u003e\u003cp\u003eTo quantify read coverage at the 3′ ends of genes, a custom GTF file was generated from the \u003cem\u003eE. coli\u003c/em\u003e reference annotation (GCF_000005845.2_ASM584v2) by extracting gene features and defining variable 3′ windows based on gene length and biotype. For tRNA, rRNA, and genes shorter than 100 bp, the full gene body was retained. For genes 100–200 bp, a region starting 50 bp downstream of the TSS to the gene end was used, and for genes longer than 200 bp, the region began 100 bp downstream of the TSS. Coordinates were adjusted strand-specifically. This modified GTF was then used with featureCounts (subread v2.0.6) to quantify reads from aligned SAM files, with paired-end and single-end reads processed separately.\u003c/p\u003e\u003cp\u003eFollowing read counting, PyDESeq2 (v0.4.10) was used to identify stably expressed genes across timepoints, selecting genes with log2 fold change between − 1 and + 1 and a baseMean ≥ 100 in at least 10 out of 16 timepoints. This filtering yielded 21 consistently expressed tRNA genes. Their counts were normalized by the number of assigned reads to the \u003cem\u003eE. coli\u003c/em\u003e genome and used to construct 10 representative sampling sets by binning genes into expression-based strata and scaling mid- and low-expression genes to match the high-expression bin. These sampling sets were used to derive timepoint-specific normalization factors relative to timepoint 1, which were then applied to normalize NAD-capture read counts in each timepoint across all replicates.\u003c/p\u003e\u003cp\u003eFor downstream expression pattern analysis, genes identified as NAD-capped in at least 3 of the 16 timepoints were clustered using dynamic time warping (DTW). Log-transformed and z-scored expression values (aggregated by mean across four replicates) were input into TimeSeriesKMeans (tslearn, DTW metric, k = 3), and clustering results were visualized using heatmaps and average trajectory plots. Genes not detected in at least two timepoints were excluded prior to clustering.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics approval and consent to participate: Not applicable\u003c/p\u003e\n\u003cp\u003eConsent for publication: Not applicable\u003c/p\u003e\n\u003cp\u003eAvailability of data and materials: Raw sequencing data have been deposited in the NCBI Sequence Read Archive (SRA), bioproject # PRJNA1327118, SRA # SRR35337667 - SRR35337683.\u003c/p\u003e\n\u003cp\u003eCompeting interests: The authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003eFunding: The project has received funding from the European Research Council under the European Union’s Horizon 2020 research and innovation program (Grant 882789 RNACoenzyme) and from the German Research Council (DFG; Project 439669440, TRR319, Subproject A02). MM was supported by a doctoral scholarship of the German Academic Scholarship Foundation.\u003c/p\u003e\n\u003cp\u003eAuthors contributions: AJ, MM, and HGMF initially devised the project, with critical input from FP. MM, TWS, FW, HGMF, MW, ET, EM, AC conducted experiments, MM, TWS and YZ analyzed and visualized data, AJ procured funding and supervised the project, MM and TWS wrote the initial manuscript draft, all authors participated in manuscript editing.\u003c/p\u003e\n\u003cp\u003eAcknowledgements: We thank the Brunner lab (BZH, Heidelberg University) for access to the qPCR machine, Heiko Rudy and Tobias Timmermann for mass spectrometry and NMR measurements, Daniel Wolf for assistance in synthesis, Drs. Christian Löcherer, Jens Frindert, Christian Schmitt, and Franziska Grün for providing materials or methods, and Drs. Martin Gärtner and Bastian Bühler (all IPMB, Heidelberg University) for valuable discussions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAuthor information: MM and TWS contributed equally to this work.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eCahov\u0026aacute; H, Winz M-L, H\u0026ouml;fer K, N\u0026uuml;bel G, J\u0026auml;schke A. NAD captureSeq indicates NAD as a bacterial cap for a subset of regulatory RNAs. Nature. 2015;519(7543):374.\u003c/li\u003e\n\u003cli\u003eBird JG, Zhang Y, Tian Y, Panova N, Barv\u0026iacute;k I, Greene L, et al. The mechanism of RNA 5\u0026prime; capping with NAD+, NADH and desphospho-CoA. Nature. 2016;535(7612):444.\u003c/li\u003e\n\u003cli\u003eMalygin AG, Shemyakin MF. Adenosine, NAD and FAD can initiate template-dependent RNA synthesis catalyzed by Escherichia coli RNA polymerase. FEBS Letters. 1979;102(1):51-4.\u003c/li\u003e\n\u003cli\u003eHuang F. Efficient incorporation of CoA, NAD and FAD into RNA by in vitro transcription. Nucleic Acids Res. 2003;31(3):e8.\u003c/li\u003e\n\u003cli\u003eM\u0026ouml;hler M, H\u0026ouml;fer K, J\u0026auml;schke A. Synthesis of 5\u0026prime;-Thiamine-Capped RNA. Molecules. 2020;25(23):5492.\u003c/li\u003e\n\u003cli\u003eDepaix A, Grudzien-Nogalska E, Fedorczyk B, Kiledjian M, Jemielity J, Kowalska J. Preparation of RNAs with non-canonical 5\u0026prime; ends using novel di- and trinucleotide reagents for co-transcriptional capping. Frontiers in Molecular Biosciences. 2022;9:854170.\u003c/li\u003e\n\u003cli\u003eWeber F, Motzkus NA, Brandl L, M\u0026ouml;hler M, Alempijevic A, J\u0026auml;schke A. Identification and in vitro characterization of UDP-GlcNAc-RNA cap-modifying and decapping enzymes. Nucleic Acids Res. 2024;52(10):5438-5450. doi:10.1093/nar/gkae353.\u003c/li\u003e\n\u003cli\u003eMlynarska-Cieslak A, Depaix A, Grudzien-Nogalska E, Sikorski PJ, Warminski M, Kiledjian M, et al. Nicotinamide-Containing Di- and Trinucleotides as Chemical Tools for Studies of NAD-Capped RNAs. Organic Letters. 2018;20(23):7650-5.\u003c/li\u003e\n\u003cli\u003eChen YG, Kowtoniuk WE, Agarwal I, Shen Y, Liu DR. LC/MS analysis of cellular RNA reveals NAD-linked RNA. Nature chemical biology. 2009;5(12):879.\u003c/li\u003e\n\u003cli\u003eGrudzien-Nogalska E, Bird JG, Nickels BE, Kiledjian M. \u0026ldquo;NAD-capQ\u0026rdquo; detection and quantitation of NAD caps. RNA. 2018;24(10):1418-25.\u003c/li\u003e\n\u003cli\u003eWang J, Alvin Chew BL, Lai Y, Dong H, Xu L, Balamkundu S, et al. Quantifying the RNA cap epitranscriptome reveals novel caps in cellular and viral RNA. Nucleic Acids Res. 2019;47(20):e130.\u003c/li\u003e\n\u003cli\u003eWinz M-L, Cahov\u0026aacute; H, N\u0026uuml;bel G, Frindert J, H\u0026ouml;fer K, J\u0026auml;schke A. Capture and sequencing of NAD-capped RNA sequences with NAD captureSeq. Nature protocols. 2017;12(1):122.\u003c/li\u003e\n\u003cli\u003eFrindert J, Zhang Y, N\u0026uuml;bel G, Kahloon M, Kolmar L, Hotz-Wagenblatt A, et al. Identification, biosynthesis, and decapping of NAD-capped RNAs in B. subtilis. Cell reports. 2018;24(7):1890-901.\u003c/li\u003e\n\u003cli\u003eMorales-Filloy HG, Zhang Y, N\u0026uuml;bel G, George SE, Korn N, Wolz C, et al. The 5\u0026prime; NAD Cap of RNAIII Modulates Toxin Production in Staphylococcus aureus Isolates. Journal of Bacteriology. 2020;202(6):e00591-19.\u003c/li\u003e\n\u003cli\u003eRuiz-Larrabeiti O, Benoni R, Zemlianski V, Hani\u0026scaron;\u0026aacute;kov\u0026aacute; N, Schwarz M, Brezovsk\u0026aacute; B, et al. NAD+ capping of RNA in Archaea and Mycobacteria. bioRxiv. 2021:2021.12.14.472595.\u003c/li\u003e\n\u003cli\u003eWalters RW, Matheny T, Mizoue LS, Rao BS, Muhlrad D, Parker R. Identification of NAD+ capped mRNAs in Saccharomyces cerevisiae. Proc Natl Acad Sci USA. 2017;114(3):480-5.\u003c/li\u003e\n\u003cli\u003eZhang Y, Kuster D, Schmidt T, Kirrmaier D, N\u0026uuml;bel G, Ibberson D, et al. Extensive 5\u0026prime;-surveillance guards against non-canonical NAD-caps of nuclear mRNAs in yeast. Nat Commun. 2020;11(1):5508.\u003c/li\u003e\n\u003cli\u003eZhang H, Zhong H, Zhang S, Shao X, Ni M, Cai Z, et al. NAD tagSeq reveals that NAD+-capped RNAs are mostly produced from a large number of protein-coding genes in Arabidopsis. Proc Natl Acad Sci USA. 2019;116(24):12072-7.\u003c/li\u003e\n\u003cli\u003eWang Y, Li S, Zhao Y, You C, Le B, Gong Z, et al. NAD+-capped RNAs are widespread in the Arabidopsis transcriptome and can probably be translated. Proc Natl Acad Sci USA. 2019;116(24):12094-102.\u003c/li\u003e\n\u003cli\u003eDong H, Wang X, Tan C, Gao L, Cui J, Liu L, et al. NAD+-capped RNAs are widespread in rice (Oryza sativa) and spatiotemporally modulated during development. Science China Life Sciences. 2022;65(10):2121-4.\u003c/li\u003e\n\u003cli\u003eNiu K, Zhang J, Ge S, Li D, Sun K, You Y, et al. ONE-seq: epitranscriptome and gene-specific profiling of NAD-capped RNA. Nucleic Acids Res. 2023;51(2):e12.\u003c/li\u003e\n\u003cli\u003eGe S, Wang X, Wang Y, Dong M, Li D, Niu K, et al. Hidden features of NAD-RNA epitranscriptome in Drosophila life cycle. iScience. 2024;27(1):108618.\u003c/li\u003e\n\u003cli\u003eJiao X, Doamekpor SK, Bird JG, Nickels BE, Tong L, Hart RP, et al. 5\u0026prime; end nicotinamide adenine dinucleotide cap in human cells promotes RNA decay through DXO-mediated deNADding. Cell. 2017;168(6):1015-27. e10.\u003c/li\u003e\n\u003cli\u003eH\u0026ouml;fer K, Li S, Abele F, Frindert J, Schlotthauer J, Grawenhoff J, et al. Structure and function of the bacterial decapping enzyme NudC. Nature chemical biology. 2016;12(9):730.\u003c/li\u003e\n\u003cli\u003eGrudzien-Nogalska E, Wu Y, Jiao X, Cui H, Mateyak MK, Hart RP, et al. Structural and mechanistic basis of mammalian Nudt12 RNA deNADding. Nature Chemical Biology. 2019;15(6):575-82.\u003c/li\u003e\n\u003cli\u003eSharma S, Grudzien-Nogalska E, Hamilton K, Jiao X, Yang J, Tong L, et al. Mammalian Nudix proteins cleave nucleotide metabolite caps on RNAs. Nucleic Acids Res. 2020;48(12):6788-98.\u003c/li\u003e\n\u003cli\u003ePan S, Li K-e, Huang W, Zhong H, Wu H, Wang Y, et al. Arabidopsis DXO1 possesses deNADding and exonuclease activities and its mutation affects defense-related and photosynthetic gene expression. Journal of Integrative Plant Biology. 2020;62(7):967-83.\u003c/li\u003e\n\u003cli\u003eAbele F, H\u0026ouml;fer K, Bernhard P, Grawenhoff J, Seidel M, Krause A, et al. A Novel NAD-RNA Decapping Pathway Discovered by Synthetic Light-Up NAD-RNAs. Biomolecules. 2020;10(4):513.\u003c/li\u003e\n\u003cli\u003eSharma S, Yang J, Grudzien-Nogalska E, Shivas J, Kwan KY, Kiledjian M. Xrn1 is a deNADding enzyme modulating mitochondrial NAD-capped RNA. Nat Commun. 2022;13(1):889.\u003c/li\u003e\n\u003cli\u003eWolfram-Schauerte M, Moskalchuk A, Pozhydaieva N, Ram\u0026iacute;rez Rojas AA, Schindler D, Kaiser S, et al. T4 phage RNA is NAD-capped and alters the NAD-cap epitranscriptome of Escherichia coli during infection through a phage-encoded decapping enzyme. bioRxiv. 2024.\u003c/li\u003e\n\u003cli\u003eJulius C, Yuzenkova Y. Bacterial RNA polymerase caps RNA with various cofactors and cell wall precursors. Nucleic Acids Res. 2017;45(14):8282-90.\u003c/li\u003e\n\u003cli\u003eBird JG, Basu U, Kuster D, Ramachandran A, Grudzien-Nogalska E, Towheed A, et al. Highly efficient 5\u0026apos; capping of mitochondrial RNA with NAD+ and NADH by yeast and human mitochondrial RNA polymerase. eLife. 2018;7:e42179.\u003c/li\u003e\n\u003cli\u003eSherwood AV, Rivera-Rangel LR, Ryberg LA, Larsen HS, Anker KM, Costa R, et al. Hepatitis C virus RNA is 5\u0026apos;-capped with flavin adenine dinucleotide. Nature. 2023;619(7971):811-8.\u003c/li\u003e\n\u003cli\u003eBenoni B, Benoni R, Trylcova J, Grab K, Pačes J, Weber J, et al. HIV-1 infection reduces NAD capping of host cell snRNA and snoRNA. bioRxiv. 2022:2022.11.10.515957.\u003c/li\u003e\n\u003cli\u003eWolfram-Schauerte M, H\u0026ouml;fer K. NAD-capped RNAs \u0026ndash; a redox cofactor meets RNA. Trends in Biochemical Sciences. 2023;48(2):142-55.\u003c/li\u003e\n\u003cli\u003eWolfram-Schauerte M, Pozhydaieva N, Grawenhoff J, Welp LM, Silbern I, Wulf A, et al. A viral ADP-ribosyltransferase attaches RNA chains to host proteins. Nature. 2023;620(7976):1054-62.\u003c/li\u003e\n\u003cli\u003eLee HC, Aarhus R. ADP-ribosyl cyclase: an enzyme that cyclizes NAD+ into a calcium-mobilizing metabolite. Cell Regulation. 1991;2(3):203-9.\u003c/li\u003e\n\u003cli\u003eHellmich MR, Strumwasser F. Purification and characterization of a molluscan egg-specific NADase, a second-messenger enzyme. Cell Regulation. 1991;2(3):193-202.\u003c/li\u003e\n\u003cli\u003eZhang H, Zhong H, Wang X, Zhang S, Shao X, Hu H, et al. Use of NAD tagSeq II to identify growth phase-dependent alterations in E. coli RNA NAD+ capping. Proc Natl Acad Sci USA. 2021;118(14):e2026183118.\u003c/li\u003e\n\u003cli\u003eHu H, Flynn N, Zhang H, You C, Hang R, Wang X, et al. SPAAC-NAD-seq, a sensitive and accurate method to profile NAD+-capped transcripts. Proc Natl Acad Sci USA. 2021;118(13):e2025595118.\u003c/li\u003e\n\u003cli\u003eSharma S, Yang J, Favate J, Shah P, Kiledjian M. NADcapPro and circNC: methods for accurate profiling of NAD and non-canonical RNA caps in eukaryotes. Commun Biol. 2023;6(1):406.\u003c/li\u003e\n\u003cli\u003eM\u0026ouml;hler M, J\u0026auml;schke A. Future Perspectives for the Identification and Sequencing of Nicotinamide Adenine Dinucleotide-Capped RNAs. Accounts of Chemical Research. 2023;56(21):3000-9.\u003c/li\u003e\n\u003cli\u003eCohen MS, Chang P. Insights into the biogenesis, function, and regulation of ADP-ribosylation. Nat Chem Biol. 2018;14(3):236-243. doi:10.1038/nchembio.2568\u003c/li\u003e\n\u003cli\u003eLee HC, Aarhus R. Structural Determinants of Nicotinic Acid Adenine Dinucleotide Phosphate Important for Its Calcium-mobilizing Activity. Journal of Biological Chemistry. 1997;272(33):20378-83.\u003c/li\u003e\n\u003cli\u003ePreugschat F, Tomberlin GH, Porter DJ. The base exchange reaction of NAD+ glycohydrolase: identification of novel heterocyclic alternative substrates. Archives of biochemistry and biophysics. 2008;479(2):114-20.\u003c/li\u003e\n\u003cli\u003eBashkin JK, Jenkins LA. The Role of Metals in the Hydrolytic Cleavage of DNA and RNA. Comments on Inorganic Chemistry. 1994;16(1-2):77-93.\u003c/li\u003e\n\u003cli\u003eLi Y, Breaker RR. Kinetics of RNA Degradation by Specific Base Catalysis of Transesterification Involving the 2\u0026lsquo;-Hydroxyl Group. Journal of the American Chemical Society. 1999;121(23):5364-72.\u003c/li\u003e\n\u003cli\u003eWaters RS, Bryden NA, Patterson KY, Veillon C, Anderson RA. EDTA chelation effects on urinary losses of cadmium, calcium, chromium, cobalt, copper, lead, magnesium, and zinc. Biol Trace Elem Res. 2001 Dec;83(3):207-21. doi:10.1385/BTER:83:3:207\u003c/li\u003e\n\u003cli\u003eNubel G, Sorgenfrei FA, Jaschke A. Boronate affinity electrophoresis for the purification and analysis of cofactor-modified RNAs. Methods. 2017;117:14-20.\u003c/li\u003e\n\u003cli\u003eWatkins CP, Zhang W, Wylder AC, Katanski CD, Pan T. A multiplex platform for small RNA sequencing elucidates multifaceted tRNA stress response and translational regulation. Nature Communications. 2022;13(1):2491.\u003c/li\u003e\n\u003cli\u003eSomervuo P, Koskinen P, Mei P, Holm L, Auvinen P, Paulin L. BARCOSEL: a tool for selecting an optimal barcode set for high-throughput sequencing. BMC Bioinformatics. 2018;19(1):257.\u003c/li\u003e\n\u003cli\u003eZhang Z, Milias-Argeitis A, Heinemann M. Dynamic single-cell NAD(P)H measurement reveals oscillatory metabolism throughout the E. coli cell division cycle. Scientific Reports. 2018;8(1):2162.\u003c/li\u003e\n\u003cli\u003eBasan M, Hui S, Okano H, Zhang Z, Shen Y, Williamson JR, et al. Overflow metabolism in Escherichia coli results from efficient proteome allocation. Nature. 2015;528(7580):99-104.\u003c/li\u003e\n\u003cli\u003eSzenk M, Dill KA, de Graff AMR. Why Do Fast-Growing Bacteria Enter Overflow Metabolism? Testing the Membrane Real Estate Hypothesis. Cell Systems. 2017;5(2):95-104.\u003c/li\u003e\n\u003cli\u003eVemuri GN, Altman E, Sangurdekar DP, Khodursky AB, Eiteman MA. Overflow metabolism in \u003cem\u003eEscherichia coli\u003c/em\u003e during steady-state growth: transcriptional regulation and effect of the redox ratio. Appl Environ Microbiol. 2006;72(5):3653-61. doi:10.1128/AEM.72.5.3653-3661.2006\u003c/li\u003e\n\u003cli\u003eKlumpp S. Growth-Rate Dependence Reveals Design Principles of Plasmid Copy Number Control. PLOS ONE. 2011;6(5):e20403.\u003c/li\u003e\n\u003cli\u003eJasiecki J, Wegrzyn G. Growth-rate dependent RNA polyadenylation in \u003cem\u003eEscherichia coli\u003c/em\u003e. \u003cstrong\u003eEMBO Rep.\u003c/strong\u003e 2003;4(2):172\u0026ndash;7. doi:10.1038/sj.embor.embor733\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"genome-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gbio","sideBox":"Learn more about [Genome Biology](https://genomebiology.biomedcentral.com/)","snPcode":"13059","submissionUrl":"https://submission.springernature.com/new-submission/13059/3","title":"Genome Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"NAD-capped RNA, non-canonical initiating nucleotide, RNA modification, RNA metabolism, transcriptomics, Escherichia coli, high-throughput sequencing, RNA capping, NAD-Seq","lastPublishedDoi":"10.21203/rs.3.rs-7650670/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7650670/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cb\u003eBackground\u003c/b\u003e\u003c/p\u003e\u003cp\u003eNicotinamide adenine dinucleotide (NAD) can function as a non-canonical initiating nucleotide, producing NAD-capped RNAs across all domains of life. This 5\u0026prime;-modification has been implicated in RNA stability and host\u0026ndash;pathogen interactions. Existing identification methods, such as NAD captureSeq, require large RNA inputs, show limited specificity, and have low sample throughput, restricting their utility for low-input or large-scale studies.\u003c/p\u003e\u003cp\u003e\u003cb\u003eResults\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe developed HELIOS NAD-Seq (\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eH\u003c/span\u003eigh-efficiency \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eE\u003c/span\u003enzyme modification with \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eL\u003c/span\u003eow \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eI\u003c/span\u003enput \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eO\u003c/span\u003ef \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eS\u003c/span\u003eample for \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eNAD-\u003c/span\u003eRNA \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eSeq\u003c/span\u003euencing), a protocol that combines a pyridyl-based ADP-ribosyl cyclase substrate, 3-picolylamine biotin, with early sample barcoding to enable high-yield, one-step biotinylation of NAD-RNAs. HELIOS NAD-Seq reduces RNA input requirements by at least 50-fold compared to NAD captureSeq, increases specificity for NAD-caps, and allows simultaneous preparation of libraries from at least 16 sample groups in quadruplicate within four days. Applied to \u003cem\u003eEscherichia coli\u003c/em\u003e, HELIOS NAD-Seq confirmed NAD-capping for 89.5% of previously identified transcripts from NAD captureSeq and detected 242 additional, predominantly low-abundance RNAs. Time-course profiling of NAD-capping across the bacterial growth curve identified a core set of metabolism-related transcripts showing low expression in stationary phase, sharp peaks in early exponential growth, and decreases in the plateau phase, suggesting a link between metabolic state and NAD-capping dynamics.\u003c/p\u003e\u003cp\u003e\u003cb\u003eConclusions\u003c/b\u003e\u003c/p\u003e\u003cp\u003eHELIOS NAD-Seq is a robust and sensitive high-throughput platform for NAD-RNA profiling enabling low-input and large-scale studies. Its application to \u003cem\u003eE. coli\u003c/em\u003e demonstrates technical advantages over existing protocols and reveals dynamic NAD-capping patterns associated with metabolism.\u003c/p\u003e","manuscriptTitle":"HELIOS NAD-Seq: A Next-Generation Capture and Sequencing Protocol for NAD-Capped RNAs with Superior Targeting and Processing","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-15 09:28:33","doi":"10.21203/rs.3.rs-7650670/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-09T12:41:25+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-29T06:35:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"288056334796309478540080272742545728187","date":"2026-01-09T08:52:17+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-22T12:13:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"123153461366080092132421259486432922471","date":"2025-11-07T07:45:25+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-02T14:00:05+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-23T11:56:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-19T06:30:05+00:00","index":"","fulltext":""},{"type":"submitted","content":"Genome Biology","date":"2025-09-18T14:19:33+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"genome-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"gbio","sideBox":"Learn more about [Genome Biology](https://genomebiology.biomedcentral.com/)","snPcode":"13059","submissionUrl":"https://submission.springernature.com/new-submission/13059/3","title":"Genome Biology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"611ee392-463d-4e9d-9712-1d36e097dd54","owner":[],"postedDate":"October 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-02-09T12:54:06+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-15 09:28:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7650670","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7650670","identity":"rs-7650670","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}