Production of Archaeal Inositol Phospholipids in Engineered Saccharomyces cerevisiae

Production of Archaeal Inositol Phospholipids in Engineered Saccharomyces cerevisiae | 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 Article Production of Archaeal Inositol Phospholipids in Engineered Saccharomyces cerevisiae Tong Si, Jinze Li, Shizhe Zhang, Erpeng Guo, Zhilai Hong This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8072063/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Despite phospholipid backbone divergence between Archaea (ether-isoprenoid/glycerol-1-phosphate membranes) and Bacteria/Eukarya (ester-fatty acid/glycerol-3-phosphate membranes), known as the ‘lipid divide’, inositol headgroups are conserved across all domains. Inositol lipids from thermophilic archaea form exceptionally stable liposomes but are difficult to obtain at scale. Here, we engineered eukaryotic Saccharomyces cerevisiae for heterologous biosynthesis of unsaturated archaetidylinositol, revealing unexpected promiscuity of endogenous yeast enzymes toward archaeal substrates. Metabolic engineering achieved production levels reaching 6.5% of total cellular lipids, enabling structural, physiological, and biophysical characterization. Production triggered pleiotropic lipid metabolism changes and enhanced thermotolerance, with engineered yeast retaining 44% viability after 30 minutes at 50°C, a condition baseline strains cannot survive. Purified archaeal inositol lipid formed thermostable liposomes maintaining integrity across 25-95°C. This yeast platform will enable access to versatile archaetidylinositol derivatives with tailored bioactivity and material properties for both evolutionary studies and biotechnology applications. Biological sciences/Biochemistry/Lipids/Membrane lipids Biological sciences/Biophysics/Membrane structure and assembly Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Life on Earth is classified into three domains: Bacteria, Archaea, and Eukarya 1 , distinguished by their membrane lipid compositions. Bacterial and eukaryotic membranes contain fatty acids ester-linked to sn -glycerol-3-phosphate (G3P), while archaeal membranes possess isoprenoid chains ether-linked to sn -glycerol-1-phosphate (G1P) 2 . This 'lipid divide' emerged early in the evolution of life and represents a central evolutionary puzzle 3 . Despite this structural divergence of phospholipid backbones, certain polar headgroups including myo -inositol are shared across all three domains. In eukaryotes, phosphatidylinositol (PI) serves critical roles beyond membrane structure, functioning as a precursor for phosphoinositides that regulate membrane trafficking, cytoskeletal organization, and signal transduction 4, 5 . The biosynthetic pathways for inositol phospholipids (Figure 1), however, differ fundamentally between domains. In eukaryotes like yeast Saccharomyces cerevisiae , PI synthase (Pis1p) catalyzes condensation of CDP-diacylglycerol (CDP-DAG) with free myo -inositol 6 . In contrast, archaeal and bacterial pathways employ d- myo -inositol-3-phosphate (abbreviated as inositol-3-phosphate in this work; also called l- myo -inositol-1-phosphate) rather than free myo -inositol as the headgroup precursor 7, 8 . In archaea, archaetidylinositol phosphate synthase (AIPS) utilizes CDP-archaeol and inositol-3-phosphate to form archaetidylinositol phosphate, which is subsequently dephosphorylated to yield the final product (Figure 1) 9 . Despite different substrate preferences, phylogenetic analyses suggest that eukaryotic PI synthases and prokaryotic AIPS enzymes share common ancestry, but the evolutionary origin of inositol lipids, whether bacterial, archaeal, or predating domain divergence, remains actively debated 7 . Understanding enzyme substrate specificity and catalytic flexibility across domains may provide insights into this evolutionary puzzle. In thermophilic archaea such as Thermococcus and Pyrococcus , archaetidylinositol and its glycosylated derivatives constitute up to 98% of membrane phospholipids 10, 11 , making them the predominant polar lipids. The prevalence of these ether-linked inositol lipids contributes to membrane stability for survival at extreme temperatures, likely through enhanced hydrogen bonding networks and optimized membrane packing 12 . Archaeosomes, liposomes formulated from archaeal lipids, retain this remarkable stability against high temperature, low pH, oxidative stress, and enzymatic degradation 13 , making them attractive for drug delivery, vaccine adjuvants, and other biotechnological applications. For example, clinical studies have shown archaeosomes to be safe and effective adjuvants that stimulate both humoral and cell-mediated immune responses 14 . Despite these advantages, both the challenges in engineering and cultivating thermophilic archaea at industrial scale, and the difficulties in chemical synthesis, have restricted commercial development of archaeosome-based technologies. These limitations have motivated synthetic biology approaches to produce archaeal lipids in industrially tractable hosts. Previous efforts have demonstrated partial reconstitution of archaeal lipid biosynthesis in microbial chassis such as Escherichia coli 15-18 and S. cerevisiae 19 . Our group previously engineered yeast to produce CDP-archaeol through expression of archaeal glycerol-1-phosphate dehydrogenase (G1PDH), geranylgeranylglyceryl phosphate synthase (GGGPS), digeranylgeranylglyceryl phosphate synthase (DGGGPS), and CDP-archaeol synthase (CarS), along with enhanced GGPP supply (Figure 1). While this resulted in hybrid neutral lipids containing both archaeal and eukaryotic components 19 , polar lipid production remained minimal, limiting studies of membrane properties and biotechnological applications. In parallel, efforts in E. coli have achieved production of archaeal phospholipids with ethanolamine and glycerol polar heads 17, 20 . However, to our knowledge, robust production of archaetidylinositol has never been achieved in heterologous hosts. In this study, we sought to produce unsaturated di- O -phytanyl phosphatidylinositol (uDoPhPI) by introducing AIPS to our previously constructed CDP-archaeol-producing yeast. Using modular pathway assembly, we evaluated the contributions of individual archaeal enzymes, assessed strategies to enhance precursor supply, and examined enzyme promiscuity between yeast and archaeal lipid biosynthetic pathways. We characterized the physicochemical properties of purified uDoPhPI, including its ability to form stable liposomes, and evaluated the thermotolerance of both uDoPhPI-containing vesicles and engineered yeast strains. Results Engineering and Structural Characterization of uDoPhPI Production in S. cerevisiae In our previous work 19 , we engineered a yeast strain (designated 4U), which is derived from the Gty116 chassis with enhanced GGPP supply 21 , and expressed heterologous archaeal lipid biosynthetic genes from Methanosarcina acetivorans (MaG1PDH, MaGGGPS, MaDGGGPS) and Archaeoglobus fulgidus (AfCarS). This strain produced the CDP-archaeol intermediate and accumulated primarily neutral archaeal lipids (DGGGOH) and hybrid DGGGO-acyl ester (DGGGO-FA) species through metabolic competition for the DGGGP precursor (Figure 1). Based on tandem MS (MS/MS) fragmentation patterns, baseline uDoPhPI was detected despite lacking a dedicated AIPS, suggesting promiscuous activity of endogenous yeast PI synthase (Pis1p) toward CDP-archaeol. But production levels in the 4U strain were too low for purification and structural characterization 19 . To enhance uDoPhPI production, we considered two candidate genes (Figure 1): Aeropyrum pernix archaetidylinositol phosphate synthase (ApAIPS), whose activity has been confirmed in vitro to catalyze condensation of CDP-archaeol with inositol-3-phosphate 22 ; and yeast INO1 encoding ScIno1p, the rate-limiting enzyme for inositol-3-phosphate biosynthesis 23 . Using modular Golden Gate assembly (Extended Data Fig. 1), we created pathway variants with individual or combined overexpression of ApAIPS and ScIno1p. Semi-quantitative analysis by mass spectrometry (MS) using selected ion monitoring (m/z 896.6 → 335.0 [M+NH 4 ] + ) revealed that uDoPhPI production increased in the order: 4U < 4U+A < 4U+I < 4U+AI (Figure 2), identifying the latter as optimal for large-scale purification (Extended Data Fig. 2). We fermented 4 L of the 4U+AI strain under galactose induction, harvesting after 48 hours to obtain 24.0 g dry cell weight (DCW). Modified Bligh-Dyer extraction yielded 2.02 g total lipids, corresponding to 84.2 mg lipid/g DCW, consistent with previous reports 24 . Silica gel chromatography with chloroform-methanol-ammonium hydroxide gradient separated the crude extract (Extended Data Fig. 2a). MS ion monitoring traced uDoPhPI in the 70-80% chloroform fractions, which were pooled and lyophilized to yield 131.4 mg material, corresponding to 6.5% of total yeast lipid and a yield of 5.5 mg uDoPhPI/g DCW. High-resolution MS analysis (Extended Data Fig. 2b, c) confirmed the molecular formula C 49 H 83 O 11 P with an observed m/z value of 896.6010 [M+NH 4 ] + , matching the calculated mass of 896.6011 (Δ = -0.11 ppm). Additionally, the fragmentation patterns of uDoPhPI agree with data in the literature in both positive and negative mode MS/MS spectra (Extended Data Fig. 2d). The isolated uDoPhPI purity was determined to be >96.8% based on A 210 peak area (Extended Data Fig. 2e). NMR spectroscopy further confirmed the uDoPhPI structure (Extended Data Fig. 3 and Supplementary Fig. 1). The 1 H NMR spectrum (400 MHz, CDCl 3 /CD 3 OD 5:2) showed diagnostic signals for unsaturated geranylgeranyl chains: δ 5.33 (d, 2H), δ 5.11 (d, 6H), δ 1.90-2.15 (m, 24H), δ 1.67 (s, 12H) and δ 1.60 (s, 18H). The myo -inositol headgroup appeared as complex multiplets between δ 3.25-4.23, while glycerol signals overlapped in the δ 3.45-4.17 region. 13 C NMR confirmed the structure with characteristic quaternary carbons at δ 140.5-131.0, olefinic carbons at δ 124.2-120.3, glycerol backbone at δ 76.8, 69.4, 65.2, and inositol ring carbons between δ 76.4-71.2. These NMR data are consistent with values reported in the literature for similar compounds 25, 26 . The detailed 2D NMR analysis of uDoPhPI (Supplementary Fig. 1) clearly showed the presence of the myo -inositol headgroup in COSY spectrum, and the linkage of geranylgeranyl chains to the glycerol skeleton in HMBC correlations, further confirmed its structure. Along with HR-MS, these combined spectral data established the structure as unsaturated di- O -phytanyl phosphatidylinositol with C 20 isoprenoid chains ether-linked to a glycerol phosphate backbone 19, 25, 27 . Notably, the glycerol backbone chirality cannot be assigned using these methods. Metabolic Impact of uDoPhPI Production on Yeast Lipid Metabolism We characterized the impact of uDoPhPI overproduction on cellular lipid metabolism, focusing on endogenous phospholipids, archaeal neutral lipids, and hybrid lipid species. We optimized an LC-MS/MS method with enhanced chromatographic resolution and established multiple reaction monitoring (MRM) for quantitative analysis (Extended Data Fig. 4). For uDoPhPI, we used the purified material to generate calibration curves enabling absolute quantification (Extended Data Fig. 4a). Commercial C 34:1 PI standard was utilized to quantify yeast endogenous PI species (Extended Data Fig. 4b). For neutral lipids including yeast TAGs, archaeal DGGGOH, and hybrid DGGGO-acyl esters (Extended Data Fig. 4c, d, e), we performed relative quantification using MRM ion counts normalized to the 4U baseline strain (Extended Data Fig. 5). To further assess compositional shift trends within each lipid class, individual species were compared using class-specific ion counts (Figure 3), and caution is warranted in interpreting absolute abundances between molecular species due to potential differences in ionization efficiencies. Using these quantitative approaches, we first examined the inositol phospholipid pools (Figure 2). The 4U baseline produced 11.5±1.2 mg uDoPhPI/g DCW despite lacking ApAIPS, confirming endogenous enzyme promiscuity. Expression of ApAIPS in 4U+A resulted in a 2.58±0.19-fold increase in uDoPhPI levels, demonstrating functional archaeal enzyme activity in yeast. ScINO1 overexpression in 4U+I yielded a 3.06±0.34-fold increase, confirming inositol precursor limitation. The combined 4U+AI strain achieved the highest production at 3.95±0.54-fold over baseline. The yield value obtained here using small culture volumes (45.6±6.2 mg uDoPhPI/g DCW, Figure 2) was substantially higher than that obtained during large-scale purification (5.5 mg uDoPhPI/g DCW, Extended Data Fig. 2), likely reflecting differences in cultivation conditions, lipid extraction efficiencies, and chromatographic purification losses between the two methods. This enhanced uDoPhPI production came partially at the expense of endogenous PI (Figure 3a and Extended Data Fig. 5a). For the abundant PI species in yeast 28 , C 34:1 PI and C 32:1 PI, ScINO1 overexpression alone increased their levels by 1.27±0.11-fold and 1.25±0.24-fold, respectively, confirming inositol precursor limitation for native PI biosynthesis. However, ApAIPS expression either alone (0.98±0.05-fold and 1.14±0.04-fold) or combined with ScINO1 overexpression (0.82±0.12-fold and 1.07±0.24-fold) had modest effects on C 34:1 PI and C 32:1 PI relative to the 4U baseline. The comparison between 4U+AI and 4U+I revealed that enhanced inositol supply due to ScIno1p overexpression was partitioned between the native and archaeal PI pathways when ApAIPS was present. For C 32:0 PI, a minor PI molecule in yeast 28 , a distinct pattern was observed. All engineered strains (4U+A, 4U+I, and 4U+AI) showed substantial increases in C 32:0 PI (7.36±2.16-fold, 6.44±0.75-fold, and 7.91±1.06-fold, respectively) relative to 4U, but the impact on overall endogenous PI content remained limited due to the low basal abundance of C 32:0 PI. The total PI content across all strains (4.77-6.41 mg/g CDW, Figure 3a) remained within the range reported for yeast 28 . Notably, neutral and hybrid lipid accumulation increased across all engineered strains (Figure 3b, c and Extended Data Fig. 5b, c, d). DGGGOH levels rose 5.36±1.30-fold with ApAIPS, 7.09±2.15-fold with ScIno1p, and 5.95±1.19-fold with both modifications (Figure 3c and Extended Data Fig. 5c). For the hybrid DGGGO-acyl esters (Figure 3c and Extended Data Fig. 5d), the abundant DGGGO-C 16:1 and DGGGO-C 16:0 showed comparable enhancement in fold changes relative to DGGGOH, whereas the less abundant species showed more dramatic increases, similar to the case with the minor C 32:0 PI. TAGs also increased in all engineered strains (Figure 3b and Extended Data Fig. 5b), with C 45:1 TAG showing the most dramatic enhancement (8.90±0.73-fold) in 4U+A. Flow cytometric analysis of Nile Red-stained cells confirmed elevated neutral lipid accumulation (Extended Data Fig. 6), with mean fluorescence intensity increasing 2.57±0.51-fold in 4U+A, 2.92±0.60-fold in 4U+I, and 4.73±0.94-fold in 4U+AI relative to 4U. These pleiotropic effects suggest that manipulating inositol metabolism impacts broader lipid homeostasis beyond the targeted phospholipid pathways, consistent with previous findings on crosstalk between storage lipid metabolism and PI biosynthesis 29 . Dissecting Enzyme Promiscuity in the Inositol Lipid Pathway Having observed promiscuous activity of yeast Pis1p toward CDP-archaeol in the 4U strain, we investigated whether other endogenous enzymes exhibit similar cross-domain substrate flexibility. Using Golden Gate assembly for modular gene replacement (Extended Data Fig. 1), we substituted individual archaeal genes with Venus fluorescent protein in the 4U+AI background to assess the contribution of each enzyme and identify potential metabolic crosstalk between archaeal and eukaryotic pathways. We first examined MaG1PDH (Figure 2), which generates the stereochemically distinct glycerol-1-phosphate backbone characteristic of archaeal lipids. Replacement of MaG1PDH in the 4U-G+AI strain resulted in severely reduced uDoPhPI production (>77% decrease relative to 4U+AI) yet detectable levels persisted (0.89±0.03-fold relative to the 4U baseline). Based on our previous stereochemical analysis showing that yeast-produced DGGGOH contains exclusively the G1P backbone even without heterologous G1PDH expression 19 , we infer that the residual uDoPhPI also likely contains the archaeal G1P backbone. However, we note that direct stereochemical confirmation of uDoPhPI chirality would be needed to conclusively establish this inference, as our current NMR data do not allow determination of glycerol chirality. The endogenous G1P synthesis reveals a cryptic metabolic capability in S. cerevisiae that parallels observations in E. coli 17 , though the substantial reduction in uDoPhPI levels upon G1PDH deletion confirms that archaeal G1PDH remains essential for efficient production. Interestingly, removal of AfCarS in 4U-C+AI maintained robust uDoPhPI production at 2.32±0.18-fold relative to the 4U baseline (Figure 2), representing 58% of the complete pathway level in 4U+AI. This demonstrates that endogenous yeast CDP-diacylglycerol synthase Cds1p promiscuously accepts the archaeal DGGGP substrate to generate CDP-archaeol, paralleling the flexibility of Pis1p in accepting CDP-archaeol for PI synthesis. These differential impacts reveal distinct metabolic constraints: while G1P supply requires dedicated archaeal enzymes due to limited endogenous activity, both the CDP-activation and PI synthesis steps exhibit considerable enzymatic promiscuity. This suggests that the evolutionary lipid divide between domains may be maintained more by differential precursor availability than by absolute enzymatic specificity, with key transferases retaining ancestral flexibility for structurally diverse substrates. As our previous work 19 established that MaGGGPS and MaDGGGPS are essential for DGGGOH production in yeast with complete loss upon their deletion, their testing for uDoPhPI production was not performed in this study. Enhanced Thermotolerance in uDoPhPI-Producing Yeast Strains To investigate whether uDoPhPI production confers stress resistance similar to that observed in E. coli strains containing archaeal ether lipids 17, 30 , we subjected engineered yeast strains to heat shock at 50°C and monitored survival by colony-forming unit counts (Figure 4). The 4U+AI strain, which produces the highest uDoPhPI levels (Figure 2), exhibited markedly enhanced thermotolerance compared to controls. This strain retained 76.54±6.36% viability after 10 minutes at 50°C, while the parent Gty116 and 4U strains showed only 37.81±16.38% and 4.24±1.90% survival, respectively. The protective effect persisted throughout the time course, with 4U+AI maintaining 43.85±3.88% viability at 30 minutes and 5.75±1.63% at 60 minutes, when all other strains showed nearly complete loss of viability (Figure 4). Conversely, the 4U-G+AI strain with severely reduced uDoPhPI production (Figure 2) displayed the poorest thermotolerance (Figure 4), with only 1.92±1.92% survival at 10 minutes and complete loss of viability by 20 minutes. This inverse correlation between uDoPhPI levels and heat sensitivity indicates that archaeal phospholipid content directly influences membrane thermostability in yeast. Formation and Characterization of uDoPhPI liposomes Archaeal lipids form stable archaeosomes that have been explored as drug delivery vehicles and vaccine adjuvants. Previous studies of archaetidylinositol-based liposomes have relied on either chemically synthesized, saturated archaetidylinositol (di- O -phytanyl phosphatidylinositol, DoPhPI) 26 or natural lipid extracts from thermophilic archaea containing predominantly glycosylated forms (~90% glucosyl-DoPhPI, ~10% DoPhPI) 13 . Whether non-glycosylated, unsaturated DoPhPI produced through heterologous yeast biosynthesis could form stable vesicular structures, however, remained unknown. We therefore investigated the membrane-forming capacity and stability of our yeast-derived uDoPhPI. First, using the electroformation method, giant unilamellar vesicles (GUVs) were successfully prepared from uDoPhPI/DOPE-Atto647 (1000:1) with uniform spherical morphology and efficient calcein encapsulation, demonstrating intact bilayer formation (Figure 5a). For small unilamellar vesicle (SUV) characterization, purified uDoPhPI was hydrated in calcein solution and extruded through 100 nm polycarbonate membranes. Dynamic light scattering analysis (DLS) revealed a monodisperse population with mean diameter of 111.0±4.1 nm. DPPC liposomes prepared using the identical method showed a larger average diameter of 210.1±6.2 nm, potentially reflecting differences in membrane curvature stress between ester and ether lipids. During four-week storage at 4°C, uDoPhPI archaeosomes demonstrated colloidal stability with minimal size change (117.2±0.2 nm) (Figure 5b), while DPPC liposomes rapidly aggregated, preventing accurate DLS analysis after a few hours. Membrane integrity assessment showed DPPC liposomes underwent substantial calcein leakage, increasing from 13.5% to 53.6% by four weeks (Figure 5c). In contrast, uDoPhPI archaeosomes effectively retained their cargo with only marginal leakage increase from 8.3% to 13.4% (Figure 5c), demonstrating superior long-term stability. To assess SUV thermal stability, we monitored temperature-dependent calcein release in the range of 25-95°C (Figure 5d). DPPC liposomes exhibited minimal leakage below their phase transition (42°C) but showed rapid release reaching ~90% at 50°C. In contrast, uDoPhPI liposomes retained nearly complete calcein encapsulation across the entire temperature range with no detectable release even at 95°C (Figure 5d). Our results demonstrate that unsaturated DoPhPI, despite lacking the saturation and glycosylation typical of natural archaeal membranes, retains the vesicle-forming capacity and remarkable thermostability suitable for biotechnological applications. Discussion This study establishes robust production of uDoPhPI in engineered S. cerevisiae , offering advantages over extraction from extremophilic archaea or complex chemical synthesis. This was achieved through combined expression of archaeal AIPS and overexpression of yeast inositol-3-phosphate synthase in a previously engineered chassis overproducing the CDP-archaeol precursor. Accumulation of uDoPhPI (45.6±6.2 mg/g DCW) substantially exceeded endogenous PI levels (6.4±1.6 mg/g DCW) by >6-fold, suggesting that the distinct structures of archaeal lipids may escape feedback mechanisms regulating native PI homeostasis in yeast. Metabolic engineering also triggered pleiotropic effects beyond the targeted inositol lipid pathway. Enhanced neutral lipid accumulation may reflect either a sink mechanism for fatty acids displaced from phospholipid biosynthesis or altered metabolic regulation. The redistribution across neutral, hybrid, and phospholipid pools demonstrates cellular flexibility to accommodate fundamentally different membrane chemistries, providing a platform for investigating evolutionary transitions in membrane composition. Baseline uDoPhPI production in yeast strains lacking archaeal AIPS or CarS expression reveals remarkable promiscuity of yeast enzymes, likely reflecting ancestral substrate flexibility. Moreover, detection of endogenous G1P synthesis in yeast, evidenced by residual DGGGOH and uDoPhPI production without heterologous G1PDH, parallels observations in E. coli 17 and an expanding catalog of bacterial species harboring cryptic archaeal-like capabilities 31, 32 . This latent G1P biosynthesis capability in yeast, combined with promiscuous metabolism of archaeal intermediates, supports models in which LUCA possessed mixed membranes with subsequent domain-specific optimization 2 , rather than strict incompatibility driving the lipid divide. The direct correlation between cellular uDoPhPI content and heat resistance parallels both the prevalence of archaetidylinositol in hyperthermophilic archaea 10, 11 and the thermostability of archaetidylinositol-based vesicles 13, 26 . Formation of stable, uniform SUVs with remarkable colloidal stability positions uDoPhPI archaeosomes as promising vehicles for drug delivery and vaccine formulations 14, 33 . Additionally, the enhanced organismal robustness suggests potential applications in engineering industrial yeast strains for high-temperature bioprocesses 34, 35 . Future work may pursue three directions. First, systems metabolic engineering could be employed to approach the high yields (20-30% of total lipids) achieved in E. coli 17 . Second, introduction of lipid structural modifications, including isoprenoid chain saturation for enhanced oxidative stability 18, 36 and alternative archaeal polar headgroups, could enable designer archaeosomes with tailored properties. Third, mechanistic studies of uDoPhPI subcellular localization, dynamics, and interactions with membrane proteins would provide insights into the emergence of complex endomembrane systems in eukaryotes. Such studies are particularly relevant, as Asgard archaea, the closest archaeal relatives to eukaryotes, encode both AIPS homologs (Supplementary Fig. 2) and eukaryotic signature proteins that may interact with inositol lipids 37, 38 . Together, our platform for producing archaeal phospholipids in a genetically tractable eukaryote opens new avenues for studying membrane evolution, developing thermostable delivery vehicles, and engineering organisms with hybrid membrane properties optimized for extreme environments. Methods Recombinant DNA construction Plasmids constructed and used in this study are listed in Supplementary Table 1. Synthetic primers and genes were purchased from Genewiz (Suzhou, China) and listed in Supplementary Table 2. The ApAIPS gene ( Aeropyrum pernix , APE1526) was synthesized with codon optimization for expression in S. cerevisiae . The yeast ScINO1 gene (YJL153C) was amplified from S. cerevisiae genomic DNA. Restriction endonucleases, T4 DNA ligase, and Gibson Assembly kit were purchased from New England Biolabs (Ipswich, MA, USA). Phanta Max Master Mix was purchased from Vazyme Biotech Co., Ltd (Nanjing, China). QIAprep Spin Plasmid Mini-prep kits from Qiagen (Valencia, CA, USA) were employed to prepare plasmid DNA from E. coli . PCR and digestion products were purified by QIAquick PCR Purification and Gel Extraction kits (Qiagen). Genomic DNA isolation was performed using the YeaStar Genomic DNA kit (Zymo Research). Archaeal pathway plasmids were constructed using the Golden Gate Assembly MoClo-YTK kit (Addgene, #1000000061) 39 with modifications. The system employed a hierarchical two-level assembly strategy (Extended Data Fig. 1). Briefly, Level 1 plasmids (pYTK095 series) served as individual transcription units containing the inducible promoter P GAL1 , CDS, terminator T ENO1 , 6×His_3×Flag tags, and linking elements (ConL, ConR) with BsmB I restriction sites. Level 1 plasmids housing archaeal lipid biosynthesis genes (MaG1PDH, MaGGGPS, MaDGGGPS, and AfCarS) were constructed in our previous study 19 . For the present work, new Level 1 plasmids for gene overexpression (ApAIPS or ScINO1) or gene replacement (with Venus fluorescent protein gene) were generated by two-fragment Gibson assembly of (1) PCR products or synthetic genes with 20-30 bp flanking homology and (2) PCR-amplified pYTK095 backbone vectors. Level 2 plasmids (pYTK096 series) contained complete metabolic pathways composed of multiple transcription units, along with homologous recombination sequences for integration at the 106a locus in the S. cerevisiae genome. To assemble the Level 2 plasmids, six Level 1 plasmids, including pYTK095-MaG1PDH, pYTK095-MaGGGPS, pYTK095-MaDGGGPS, pYTK095-AfCarS, pYTK095-ApAIPS and pYTK095-ScINO1, were mixed with the Level 2 acceptor vector pYTK096-ccdB in BsmB I assembly reactions. For gene deletion variants, the targeted Level 1 plasmid was substituted with its Venus replacement counterpart. Golden Gate assembly reactions contained 0.02 pmol of each DNA component, 1000 U T4 DNA ligase, 0.5 μL of T4 ligase buffer, 5 U BsmB I restriction endonuclease, and water to a 5 μL final volume. The assembly protocol consisted of 45 cycles of 42°C for 1 min and 16°C for 1 min, followed by 65°C for 20 min. Assembly products were transformed into E. coli DH5α by standard chemical transformation. Transformants were selected at 37°C on Luria-Bertani (LB) agar plates supplemented with either 50 μg/mL kanamycin or 100 μg/mL ampicillin, followed by liquid cultivation for plasmid isolation. Yeast strain construction Yeast strains used in this study are listed in Supplementary Table 3. S. cerevisiae Gty116 was a generous gift from Professor Jay D. Keasling 21 . The baseline 4U strain 19 , constructed previously in the Gty116 chassis, contains M. acetivorans G1PDH (Ma3686), GGGPS (Ma3969), DGGGPS (Ma0961), and A. fulgidus CarS (AF1740). Sequence-verified Level 2 plasmids were linearized with Not I and used to transform Gty116 competent cells using the standard heat shock method 40 . The plasmid pCUT-KanMX encoding the CRISPR/Cas system was co-transformed with linearized donor DNA to induce a double-strand break at the 106a locus, thereby allowing efficient and scarless integration of the donor DNA. Transformants were selected on YPD agar medium supplemented with 800 μg/mL G418 (Beijing OKA Biotech, Beijing, China) and correct genomic integration was confirmed by diagnostic colony PCR. Production, purification, and structural analysis of uDoPhPI To prepare seed cultures, the highest-producing strain (4U+AI) was first cultivated in baffled shake flasks containing YPD medium (1% yeast extract, 2% peptone, 2% glucose) at 30°C and 200 rpm. For inducible production, cells were then inoculated at an initial OD 600 of 0.1 in 4 L of YPG medium (1% yeast extract, 2% peptone, 2% galactose) and cultured under the same conditions for 48 hours. For uDoPhPI purification (Extended Data Fig. 2), cells were harvested by centrifugation, yielding 24 g dry cell weight. The cell pellet was ground with a mortar and pestle in liquid nitrogen. The crude extract was mixed with 500 mL of methanol/chloroform (2:1, v/v) in a 2 L separating funnel. Phase separation was induced by sequential addition of 200 mL of chloroform and 200 mL of water. The lower organic phase was collected, and extraction was repeated 3-5 times on the remaining material. Pooled organic extracts were concentrated using a rotary evaporator. The concentrated extract was redissolved in 2 mL of n-hexane and 2 mL of chloroform and applied to a silica gel column (200-300 mesh). uDoPhPI was eluted using a chloroform-methanol-ammonium hydroxide gradient (75:25:0.4, v/v/v), with fractions monitored by mass spectrometry (m/z 896.6 [M+NH 4 ] + ). Fractions containing uDoPhPI were pooled and lyophilized, yielding 131.4 mg of purified material. Purity was assessed by HPLC using an Agilent 1290 HPLC instrument consisting of a G7104A flexible pump, a G7117B DAD detector, and a G7167B autosampler. For NMR analysis, purified uDoPhPI was dissolved in 500 µL of deuterated chloroform and 200 µL of deuterated methanol (70 mg/mL), then NMR data was acquired on a 400 MHz Bruker Avance III spectrometer. NMR data are reported as follows (Extended Data Fig. 3 and Supplementary Fig. 1): chemical shifts ( δ , in ppm), multiplicity (s = singlet, d = doublet, t = triplet), coupling constants ( J , in Hz), and integration, using TMS as the internal reference. Topspin ver. 4.5.0 (Bruker Biospin) was used for NMR spectral acquisition and data processing. High-resolution mass spectrometry (HR-MS) was performed using a timsTOF Flex mass spectrometer (Bruker Daltonics) to confirm the molecular formula. Mass spectra were acquired in electrospray ionization (ESI) positive ion mode with the following parameters: tims ramp time 100 ms, PASEF on, 100 ms, scan range m/z, 100–1350; 1/k 0 , 0.55–1.87 V.s/cm 2 ). Untargeted profiling data were acquired in ESI positive ion mode and negative ion modes. Yeast lipid extraction for quantification Yeast cultivation and inducible production were performed as described above. After 48 hours, ~2 mL aliquots corresponding to 40 OD 600 units (equivalent to 12 mg dry cell weight) were collected for total lipids extraction using a modified Bligh and Dyer method optimized for yeast phospholipids 41 . Cells were pelleted by centrifugation (10,000 × g, 5 min) and lysed by vigorous agitation with 0.1 mm glass beads (3 cycles of 2 min vortexing with 1 min cooling on ice) directly in the chloroform/methanol extraction solvent (1:2 v/v). The mixture was thoroughly vortexed for 1 min at room temperature. Phase separation was induced by adding chloroform and water (1:1 v/v). After centrifugation (2,000 × g, 5 min), the lower organic layer containing lipids was carefully collected. The upper aqueous layer and interface were re-extracted once using the same procedure, and the organic phases were pooled. The combined organic extracts were dried under a nitrogen stream and stored at -80°C until analysis. Before analysis, lipids were resuspended in 500 µL of methanol/isopropanol (1:1 v/v). Quantitative Lipid Analysis by LC-MS/MS Lipid extracts were analyzed using a triple quadrupole LC-MS/MS system equipped with an Agilent 1290 HPLC and an Agilent 6470B triple quadrupole mass spectrometer operated in ESI positive ion mode. Quantification was performed using multiple reaction monitoring (MRM) mode (Extended Data Fig. 4). Three different chromatographic methods were employed to achieve optimal separation of different lipid classes. For archaeal uDoPhPI and DGGGOH (Method 1), chromatographic separation was achieved using two C18 AR UPLC columns (2.1 × 150 mm, 2 µm; ACE) connected in series and maintained at 55 °C with a flow rate of 0.5 mL/min. A 2 µL sample volume was injected and eluted with the following gradient: 0 min, 72% B; 8 min, 100% B; 11 min, 72% B, followed by 8 min equilibration. Mobile phase A consisted of 50% acetonitrile and 50% water with 5 mM ammonium acetate, and mobile phase B consisted of 50% isopropanol and 50% acetonitrile supplemented with 0.1% aqueous solution of 4 mM ammonium formate. Samples were diluted 1000-fold for DGGGOH detection to avoid saturation. MRM transitions were m/z 896.6 → 335 (uDoPhPI, [M+NH 4 ] + ) and 654.4 → 273.3 (DGGGOH, [M+NH 4 ] + ) at fragmentor voltage 135 V and collision energy 20 V. For endogenous yeast phosphatidylinositol species (Method 2), a single C 18 AR UPLC column (2.1 × 150 mm, 2 µm; ACE) was used with the same chromatographic and MS conditions as Method 1. MRM transitions were m/z 854.6 → 577.5 (C 34:1 PI, [M+NH 4 ] + ), 828.5 → 551.5 (C 32:0 PI, [M+NH 4 ] + ), and 826.5 → 549.6 (C 32:1 PI, [M+NH 4 ] + ). For yeast TAGs and hybrid DGGGO-acyl esters (Method 3), these neutral lipids were separated on a single C18 AR UPLC column (2.1 × 150 mm, 2 µm; ACE) maintained at 55°C with a flow rate of 0.2 mL/min. A 2 µL sample was injected and eluted with the following gradient: 0 min, 0% B; 3.3 min, 0% B; 5 min, 24% B; 23 min, 65% B; 24 min, 90% B; 28 min, 90% B, followed by 5 min equilibration. Mobile phase A was methanol and mobile phase B was isopropanol, both containing formic acid and ammonium hydroxide (100:0.04:0.1, v/v/v). MRM transitions were monitored at fragmentor voltage 135 V and collision energy 10 V. For yeast TAGs, the three most abundant species were analyzed using the quantifier ion m/z 549.5 and corresponding [M+NH 4 ] + precursor ions for C 45:1 TAG (m/z 738.6), C 47:1 TAG (m/z 766.5), and C 53:2 TAG (m/z 848.6). For DGGGO-acyl esters, the quantifier ion of m/z 273.3 was selected for relative quantification of corresponding [M+NH 4 ] + precursor ions for DGGGO-C 8:0 (m/z 780.7), DGGGO-C 10:0 (m/z 808.6), DGGGO-C 12:0 (m/z 836.6), DGGGO-C 14:1 (m/z 862.6), DGGGO-C 16:1 (m/z 890.7), DGGGO-C 16:0 (m/z 892.7), DGGGO-C 18:1 (m/z 918.7), and DGGGO-C 18:0 (m/z 920.7). Flow cytometric quantification of cellular lipids Neutral lipid content was assessed by Nile Red staining and flow cytometry 42 . After 48-hour galactose induction as described above, cells were collected, washed twice with phosphate buffered saline (PBS, pH 7.4), and stained with 100 μM Nile Red (Thermo Fisher Scientific, Waltham, MA, USA) for 30 minutes at room temperature in the dark. Stained cells were washed, resuspended in PBS, and analyzed using a CytoFLEX S Flow Cytometer (Beckman Coulter, Indianapolis, IN, USA) with 488 nm excitation and 585 nm emission detection. For each sample, 10,000 gated events were collected, and mean fluorescence intensity was determined using CytExpert software to quantify relative neutral lipid levels across strains. Thermotolerance Assay The engineered yeast strains were cultured under galactose induction for 48 hours as described above. The final cultures were diluted to an OD 600 of 0.001 in YPD medium to ensure suitable cell density for colony-forming unit (CFU) quantification. Equal volumes (20 μL) of diluted cultures were exposed to 50°C for different durations (0, 10, 20, 30, and 60 min), while untreated samples maintained at 30°C served as controls. Following thermal treatment, samples were immediately plated on YPD agar medium and incubated at 30°C for 48 hours. Colonies were counted and survival rates were calculated as the percentage of viable cells relative to the untreated control at each time point. Data are presented as the mean ± standard error of the mean (SEM) from four independent biological replicates. Giant Unilamellar Vesicle (GUV) Preparation Commercial reference lipids, including DOPE-Atto647N (1,2-dioleoyl- sn -glycero-3-phosphoethanolamine labeled with Atto647N) and 1,2-dipalmitoyl- sn -glycero-3-phosphocholine (DPPC), were purchased from Avanti Polar Lipids (Alabaster, AL, USA), and stored in chloroform at -20°C until use. GUVs were prepared using the electroformation method 43 using a VesiclePrepPro device (Nanion Technologies GmbH, Munich, Germany). Purified uDoPhPI and DOPE-Atto647N were mixed at a 1000:1 molar ratio in chloroform. A 15 μL aliquot of the lipid mixture (1 mg/mL) was spread homogeneously on the conductive side of an indium tin oxide (ITO)-coated glass coverslide (Visiontek Systems Ltd, Chester, UK). The lipid-coated ITO slide was placed under vacuum for at least 15 min to ensure complete solvent evaporation. A rubber O-ring (18 mm inner diameter) was positioned on the lipid-coated surface and filled with 280 μL of 200 mM sucrose solution containing 20 μM calcein. A second ITO-coated slide was placed on top to seal the chamber. The assembled electroformation chamber was connected to VesiclePrepPro electrodes, and an alternating current field was applied using a multi-step protocol as described in Supplementary Table 4. Following electroformation, GUVs were harvested by gentle centrifugation and visualized immediately by Nikon AX confocal laser scanning microscopy using a CFI Plan Apochromat Lambda D 60× oil objective (Nikon, Tokyo, Japan) with excitation at 488 nm (calcein) and 647 nm (DOPE-Atto647N). Small Unilamellar Vesicle (SUV) Preparation Calcein-loaded SUVs were prepared by the thin-film hydration and extrusion method 44 . Purified uDoPhPI (5 mg) in chloroform was transferred to an amber glass vial and the solvent was evaporated under a stream of nitrogen, followed by drying under vacuum for 30 min to remove residual solvent. The dried lipid film was hydrated with 500 μL of 106 mM calcein solution, vortexed vigorously, and sonicated for 30 min in an ultrasonic bath at 45°C. The resulting multilamellar vesicle suspension was extruded 21 times through a polycarbonate membrane (100 nm pore size) using a mini-extruder (Avanti Polar Lipids) to generate SUVs. Unencapsulated calcein was removed by gel filtration on a Sephadex G-50 column (Sigma-Aldrich, St. Louis, MO, USA) pre-equilibrated with PBS (pH 7.0) without calcium and magnesium (Cytiva, Chicago, IL, USA). DPPC liposomes were prepared using the identical procedure. Particle size distributions and polydispersity indices were measured by dynamic light scattering (DLS) using a DynaPro NanoStar (Wyatt Technology, Santa Barbara, CA, USA) at 25°C. Temperature-dependent Calcein Release Assay Temperature-dependent calcein release from uDoPhPI and DPPC vesicles was assessed at pH 7.0 over the temperature range of 25-95°C. Liposome suspensions were diluted to 0.1 mg/mL lipid concentration in PBS. Samples were incubated at each test temperature for 5 min, and 200 μL aliquots were dispensed in triplicate into a Corning black 96-well microplate (Catalog#3916, Corning, NY, USA). Fluorescence intensity was measured using an Agilent BioTek Synergy H1 microplate reader (Winooski, Vermont, USA) with excitation and emission wavelengths of 488 nm and 528 nm, respectively. The percentage of calcein released was calculated as F / F max × 100%, where F represents the fluorescence intensity after incubation at the test temperature, and F max represents the maximal fluorescence obtained after complete vesicle lysis with 2 μL of 10% (v/v) Triton X-100 added at the test temperature. Bioinformatics Analysis For AIPS sequence similarity network (SSN) analysis, ApAIPS was used as a query to perform a protein BLAST search in the NCBI ClusteredNR database (limited to Asgard group, taxid:1935183) at a threshold of e-value<10 -5 . The top 100 BLAST hits were retrieved and pooled with 458 archaeal AIPS sequences from InterPro family IPR054868 (accessed on Oct 17, 2025). SSN analysis was performed using the online EFI-EST tool 45 (https://efi.igb.illinois.edu/efi-est/, FASTA mode) with an alignment score threshold of 30. The 70% identity representative node network comprising 313 nodes and 48,803 edges was downloaded and visualized using Cytoscape (Version: 3.10.3). Declarations Declaration of competing interests The authors declare no competing financial interest. Acknowledgements This work was financially supported by the National Key Research and Development Program of China (2021YFA0910800), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0480000), Guangdong S&T Program (2024B1111140001), and Shenzhen Science and Technology Program (RCJC20221008092810021). We thank the Shenzhen Synthetic Biology Infrastructure for instrument support and technical assistance. We thank Professor Jay D. Keasling for providing S. cerevisiae Gty116. Data availability Data are available from the corresponding authors upon reasonable request. References Woese, C.R., Kandler, O. & Wheelis, M.L. Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya. Proc Natl Acad Sci U S A 87 , 4576–4579 (1990). Lombard, J., Lopez-Garcia, P. & Moreira, D. The early evolution of lipid membranes and the three domains of life. Nat Rev Microbiol 10 , 507–515 (2012). Koga, Y. Early Evolution of Membrane Lipids: How did the Lipid Divide Occur? J Mol Evol 72 , 274–282 (2011). Posor, Y., Jang, W. & Haucke, V. Phosphoinositides as membrane organizers. Nat Rev Mol Cell Biol 23 , 797–816 (2022). Steinfeld, N., Giridharan, S.S.P., Kauffman, E.J. & Weisman, L.S. in Phosphoinositides: Methods and Protocols. (ed. R.J. Botelho) 1–17 (Springer US, New York, NY; 2021). Nikawa, J., Kodaki, T. & Yamashita, S. Primary structure and disruption of the phosphatidylinositol synthase gene of Saccharomyces cerevisiae . J Biol Chem 262 , 4876–4881 (1987). Esposti, M.D. Eukaryotes inherited inositol lipids from bacteria: implications for the models of eukaryogenesis. FEBS Lett 597 , 2484–2496 (2023). Morii, H., Ogawa, M., Fukuda, K., Taniguchi, H. & Koga, Y. A revised biosynthetic pathway for phosphatidylinositol in Mycobacteria. J Biochem 148 , 593–602 (2010). Morii, H., Kiyonari, S., Ishino, Y. & Koga, Y. A Novel Biosynthetic Pathway of Archaetidyl-myo-inositol via Archaetidyl-myo-inositol Phosphate from CDP-archaeol and d-Glucose 6-Phosphate in Methanoarchaeon Methanothermobacter thermautotrophicus Cells. J Biol Chem 284 , 30766–30774 (2009). Tourte, M. et al. Novel Intact Polar and Core Lipid Compositions in the Pyrococcus Model Species, P. furiosus and P. yayanosii , Reveal the Largest Lipid Diversity Amongst Thermococcales. Biomolecules 10 (2020). Morii, H. et al. A novel phosphoglycolipid archaetidyl(glucosyl)inositol with two sesterterpanyl chains from the aerobic hyperthermophilic archaeon Aeropyrum pernix K1. Biochim Biophys Acta 1436 , 426–436 (1999). Chong, P.L., Chang, A., Yu, A. & Mammedova, A. Vesicular and Planar Membranes of Archaea Lipids: Unusual Physical Properties and Biomedical Applications. Int J Mol Sci 23 (2022). Gmajner, D., Ota, A., Sentjurc, M. & Ulrih, N.P. Stability of diether C(25,25) liposomes from the hyperthermophilic archaeon Aeropyrum pernix K1. Chem Phys Lipids 164 , 236–245 (2011). McCluskie, M.J., Deschatelets, L. & Krishnan, L. Sulfated archaeal glycolipid archaeosomes as a safe and effective vaccine adjuvant for induction of cell-mediated immunity. Hum Vaccin Immunother 13 , 2772–2779 (2017). Jain, S., Caforio, A. & Driessen, A.J.M. Biosynthesis of archaeal membrane ether lipids. Frontiers in Microbiology Volume 5 - 2014 (2014). Isobe, K. et al. Geranylgeranyl reductase and ferredoxin from Methanosarcina acetivorans are required for the synthesis of fully reduced archaeal membrane lipid in Escherichia coli cells. J Bacteriol 196 , 417–423 (2014). Caforio, A. et al. Converting Escherichia coli into an archaebacterium with a hybrid heterochiral membrane. Proc Natl Acad Sci U S A 115 , 3704–3709 (2018). Chen, H. et al. Engineering archaeal membrane‐spanning lipid GDGT biosynthesis in bacteria: Implications for early life membrane transformations. mLife 4 , 193–204 (2025). Zhang, J. et al. Biosynthesis of Hybrid Neutral Lipids with Archaeal and Eukaryotic Characteristics in Engineered Saccharomyces cerevisiae . Angew Chem Int Ed Engl 62 , e202214344 (2023). Hoekzema, M., Jiang, J. & Driessen, A.J.M. Optimizing Archaeal Lipid Biosynthesis in Escherichia coli . ACS Synth Biol 13 , 2470–2479 (2024). Reider Apel, A. et al. A Cas9-based toolkit to program gene expression in Saccharomyces cerevisiae. Nucleic Acids Res 45 , 496–508 (2017). Morii, H., Ogawa, M., Fukuda, K. & Taniguchi, H. Ubiquitous distribution of phosphatidylinositol phosphate synthase and archaetidylinositol phosphate synthase in Bacteria and Archaea, which contain inositol phospholipid. Biochem Biophys Res Commun 443 , 86–90 (2014). Hirsch, J.P. & Henry, S.A. Expression of the Saccharomyces cerevisiae inositol-1-phosphate synthase (INO1) gene is regulated by factors that affect phospholipid synthesis. Mol Cell Biol 6 , 3320–3328 (1986). Slaughter, J.C. & Minabe, M. Fatty acid-containing lipids of the yeast Saccharomyces cerevisiae during post-fermentation decline in viability. J Sci Food Agric 65 , 497–501 (1994). Lamosa, P. et al. Occurrence of 1-glyceryl-1-myo-inosityl phosphate in hyperthermophiles. Appl Environ Microbiol 72 , 6169–6173 (2006). Ruiz, J. et al. Membrane plasticity induced by myo-inositol derived archaeal lipids: chemical synthesis and biophysical characterization. Phys Chem Chem Phys 25 , 16273–16287 (2023). Jensen, S.M., Brandl, M., Treusch, A.H. & Ejsing, C.S. Structural characterization of ether lipids from the archaeon Sulfolobus islandicus by high-resolution shotgun lipidomics. J Mass Spectrom 50 , 476–487 (2015). Xia, J.-M. & Yuan, Y.-J. Comparative Lipidomics of Four Strains of Saccharomyces cerevisiae Reveals Different Responses to Furfural, Phenol, and Acetic Acid. J Sci Food Agric 57 , 99–108 (2009). Gaspar, M.L., Hofbauer, H.F., Kohlwein, S.D. & Henry, S.A. Coordination of storage lipid synthesis and membrane biogenesis: evidence for cross-talk between triacylglycerol metabolism and phosphatidylinositol synthesis. J Biol Chem 286 , 1696–1708 (2011). Yoshida, R., Motoyama, K., Ito, T. & Hemmi, H. Effects of producing high levels of hyperthermophile-specific C25,C25-archaeal membrane lipids in Escherichia coli . Biochem Biophys Res Commun 729 , 150349 (2024). Palyzová, A., Guschina, I.A. & Řezanka, T. Chiral analysis of glycerol phosphates- can bacteria biosynthesize heterochiral phospholipid membranes? J Chromatogr A 1676 (2022). Palyzova, A. et al. Stereochemistry of phosphatidylglycerols from thermotolerant bacteria isolated thermal springs. J Chromatogr A 1739 , 465517 (2025). Santhosh, P.B. & Genova, J. Archaeosomes: New Generation of Liposomes Based on Archaeal Lipids for Drug Delivery and Biomedical Applications. ACS Omega 8 , 1–9 (2023). Chi, Z., Kohlwein, S.D. & Paltauf, F. Role of phosphatidylinositol (PI) in ethanol production and ethanol tolerance by a high ethanol producing yeast. J Ind Microbiol Biotechnol 22 , 58–63 (1999). Henderson, C.M., Zeno, W.F., Lerno, L.A., Longo, M.L. & Block, D.E. Fermentation temperature modulates phosphatidylethanolamine and phosphatidylinositol levels in the cell membrane of Saccharomyces cerevisiae. Appl Environ Microbiol 79 , 5345–5356 (2013). Rao, A. & Driessen, A.J.M. Unraveling the multiplicity of geranylgeranyl reductases in Archaea: potential roles in saturation of terpenoids. Extremophiles 28 , 14 (2024). Eme, L. et al. Inference and reconstruction of the heimdallarchaeial ancestry of eukaryotes. Nature 618 , 992–999 (2023). Liu, Y. et al. Expanded diversity of Asgard archaea and their relationships with eukaryotes. Nature 593 , 553–+ (2021). Lee, M.E., DeLoache, W.C., Cervantes, B. & Dueber, J.E. A Highly Characterized Yeast Toolkit for Modular, Multipart Assembly. ACS Synth Biol 4 , 975–986 (2015). Gietz, R.D. & Schiestl, R.H. Large-scale high-efficiency yeast transformation using the LiAc/SS carrier DNA/PEG method. Nat Protoc 2 , 38–41 (2007). Bligh, E.G. & Dyer, W.J. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37 , 911–917 (1959). Rostron, K.A. & Lawrence, C.L. in Histochemistry of Single Molecules. (eds. C. Pellicciari & M. Biggiogera) 219–229 (Springer New York, New York, NY; 2017). Dreher, Y., Jahnke, K., Bobkova, E., Spatz, J.P. & Göpfrich, K. Division and Regrowth of Phase‐Separated Giant Unilamellar Vesicles**. Angew Chem Int Ed 60 , 10661–10669 (2021). Dutta, S., Watson, B., Mattoo, S. & Rochet, J.-C. Calcein Release Assay to Measure Membrane Permeabilization by Recombinant Alpha-Synuclein. Bio-Protocol 10 , e3690 (2020). Zallot, R., Oberg, N. & Gerlt, J.A. The EFI Web Resource for Genomic Enzymology Tools: Leveraging Protein, Genome, and Metagenome Databases to Discover Novel Enzymes and Metabolic Pathways. Biochemistry 58 , 4169–4182 (2019). Additional Declarations There is NO Competing Interest. Supplementary Files aPISINatSynth.docx Supplementary Information ExtendedDataFigures.docx Cite Share Download PDF Status: Posted Version 1 posted 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-8072063","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":542536172,"identity":"b4204e8a-b6e8-4b11-8d00-3da62f46aaa9","order_by":0,"name":"Tong Si","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAy0lEQVRIiWNgGAWjYPACCTkGZmLV8kC1GAO1MDaQooUhEaicSC327L2HXzD8sUif3878/AFDjR0D/2wCOnl4zqVZMLZJ5G44zGbYwHAsmUHizgECWiRyzAwYG4BamBmAWtgOMBhIJBChheGPRLp8M/vHBoZ/xGkxfsDABlR2mMewgbGNGC1nzpgxJLZJGG44zFM4I7EvmUfiBgEt7O09xh8+/KmTl+8/vuHDh292cvwzCGgBAjaESxIQEYUXMH8gRtUoGAWjYBSMYAAA4CY5l/8TdcwAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0003-2985-9014","institution":"Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences","correspondingAuthor":true,"prefix":"","firstName":"Tong","middleName":"","lastName":"Si","suffix":""},{"id":542536173,"identity":"b92bde79-381f-4373-90b3-064e7e48762d","order_by":1,"name":"Jinze Li","email":"","orcid":"","institution":"Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Jinze","middleName":"","lastName":"Li","suffix":""},{"id":542536174,"identity":"88a5ab9c-7cad-4b99-ad9c-e3bc7902fefe","order_by":2,"name":"Shizhe Zhang","email":"","orcid":"","institution":"Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Shizhe","middleName":"","lastName":"Zhang","suffix":""},{"id":542536175,"identity":"d92588ef-f9a5-4ec9-957a-0b082cbb623f","order_by":3,"name":"Erpeng Guo","email":"","orcid":"","institution":"Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Erpeng","middleName":"","lastName":"Guo","suffix":""},{"id":542536176,"identity":"16fb356d-201b-4840-a1a5-9b0a765a4575","order_by":4,"name":"Zhilai Hong","email":"","orcid":"","institution":"Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Zhilai","middleName":"","lastName":"Hong","suffix":""}],"badges":[],"createdAt":"2025-11-10 02:50:52","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8072063/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8072063/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95611255,"identity":"34a8bc0e-6d11-4095-9224-649a2d704e97","added_by":"auto","created_at":"2025-11-11 08:05:42","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":245018,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBiosynthetic pathways of unsaturated di-\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eO\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e-phytanyl phosphatidylinositol (uDoPhPI) in engineered \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eS. cerevisiae\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e.\u003c/strong\u003e The schematic contrasts archaeal (blue shading) and eukaryotic (yellow shading) lipid species, including uDoPhPI3P: unsaturated di-\u003cem\u003eO\u003c/em\u003e-phytanyl phosphatidylinositol-phosphate; uDoPhPI: unsaturated di-\u003cem\u003eO\u003c/em\u003e-phytanyl phosphatidylinositol; DGGGOH: 2,3-di-\u003cem\u003eO-\u003c/em\u003egeranylgeranyl-\u003cem\u003esn\u003c/em\u003e-glycerol; PI: phosphatidylinositol; DAG: diacylglycerol; TAG: triacylglycerol. Heterologous archaeal genes include G1PDH: glycerol-1-phosphate dehydrogenase; GGGPS: geranylgeranylglyceryl phosphate synthase; DGGGPS: di-\u003cem\u003eO\u003c/em\u003e-geranylgeranylglycery-l-phosphate synthase; CarS: CDP-archaeol synthase; AIPS: archaetidylinositol phosphate synthase. Endogenous yeast genes include INO1: inositol-3-phosphate synthase; LPP: lipid phosphate phosphatase; CDS1: phosphatidate cytidylyltransferase; PIS1: CDP-diacylglycerol-inositol 3-phosphatidyltransferase.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/29d58a1fe9f524e6882a76d6.png"},{"id":95611254,"identity":"cff4f348-491d-4066-afcf-7c68434f50ba","added_by":"auto","created_at":"2025-11-11 08:05:42","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":256563,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProduction of uDoPhPI in engineered yeast strains\u003c/strong\u003e. Left panel: schematic representation of pathway components present in each strain. Strain genotypes 4U contains MaG1PDH, MaGGGPS, MaDGGGPS, and AfCarS; 4U+A adds ApAIPS; 4U+I adds ScINO1; 4U+AI adds both ApAIPS and ScINO1; 4U-G+AI lacks MaG1PDH (replaced with Venus) but contains ApAIPS and ScINO1; 4U-C+AI lacks AfCarS (replaced with Venus) but contains ApAIPS and ScINO1. Right panel: uDoPhPI abundance as measured by LC-MS/MS multiple reaction monitoring (MRM) normalized by cell biomass. Fold changes are relative to the baseline 4U strain (set as 100%). Data represent mean ± SEM from three independent biological replicates.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/e382285b436fe1d835f9afad.png"},{"id":95611258,"identity":"12129cf9-8427-4b90-bb0c-32bb8f5d4df4","added_by":"auto","created_at":"2025-11-11 08:05:42","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":279401,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImpact of uDoPhPI pathway engineering on cellular lipid metabolism. \u003c/strong\u003eRelative quantification was performed using ion counts by LC-MS/MS MRM. \u003cstrong\u003e(a)\u003c/strong\u003eYeast PI species. \u003cstrong\u003e(b)\u003c/strong\u003e Yeast TAG species. \u003cstrong\u003e(c)\u003c/strong\u003e Archaeal and hybrid lipids with DGGGOH backbone. All data represent mean ± SEM from three independent biological replicates. Strain genotypes are as defined in Figure 2. Normalized fold changes for each lipid molecule relative to the 4U baseline are presented in Extended Data Fig. 5.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/c28cd2b48e0e6b0be6bdc6e4.png"},{"id":95656424,"identity":"6c7bf5db-4a84-4269-86d7-185eb564737b","added_by":"auto","created_at":"2025-11-11 16:18:40","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":340769,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnhanced thermotolerance in uDoPhPI-producing yeast strains. \u003c/strong\u003eEngineered strains were cultured under galactose induction for 48 hours, diluted to OD₆₀₀ 0.001, and subjected to heat shock at 50°C for the indicated durations. \u003cstrong\u003e(a)\u003c/strong\u003e Survival curves. Percentage of viable cells (colony-forming units) relative to untreated controls maintained at 30°C. Data represent mean ± SEM from four independent biological replicates. \u003cstrong\u003e(b)\u003c/strong\u003e Representative images of agar plates showing colony formation. Strain genotypes are as defined in Figure 2. Gty116 is the chassis strain without introduction of archaeal pathways.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/5d4e3d36e1573c5e37080798.png"},{"id":95611257,"identity":"e529506c-ddd4-4f4e-a840-c80ff69193b9","added_by":"auto","created_at":"2025-11-11 08:05:42","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":713638,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCharacterization of uDoPhPI liposome formation and stability.\u003c/strong\u003e \u003cstrong\u003e(a)\u003c/strong\u003e Confocal microscopy images of calcein-loaded uDoPhPI GUVs prepared by electroformation. Top: individual vesicles; bottom: vesicle population. Green: encapsulated calcein; magenta: membrane marker DOPE-Atto647N. \u003cstrong\u003e(b) \u003c/strong\u003eMean diameters of SUV measured by DLS. \u003cstrong\u003e(c) \u003c/strong\u003eTime-dependent calcein release from SUVs during storage at 4°C. \u003cstrong\u003e(d) \u003c/strong\u003eTemperature-dependent calcein release from SUVs at pH 7.0. Data in b-d represent mean ± SEM from three independent preparations. Black square: DPPC; Red cycle: uDoPhPI; Red triangle in (\u003cstrong\u003eb\u003c/strong\u003e): uDoPhPI after 4-week storage at 4°C.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/1402378ebea89bb757ddf981.png"},{"id":96708049,"identity":"c567923d-d1fe-46b7-8afd-f4b61bb0a6b5","added_by":"auto","created_at":"2025-11-25 09:53:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2882099,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/0173d26d-16cb-4ab0-9a20-b41d6c6b48f0.pdf"},{"id":95611256,"identity":"3e03ccb6-62ef-444d-bfcc-38c2aa98ea6f","added_by":"auto","created_at":"2025-11-11 08:05:42","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2560103,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"aPISINatSynth.docx","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/25e16964010a2895629d0ea3.docx"},{"id":95611260,"identity":"9ff8e2bb-f1ba-4528-95bf-b2acf45c6097","added_by":"auto","created_at":"2025-11-11 08:05:42","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":7101855,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-8072063/v1/4c825bfa8b5ff9f0081133bc.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Production of Archaeal Inositol Phospholipids in Engineered Saccharomyces cerevisiae","fulltext":[{"header":"Introduction","content":"\u003cp\u003eLife on Earth is classified into three domains: Bacteria, Archaea, and Eukarya\u003csup\u003e1\u003c/sup\u003e, distinguished by their membrane lipid compositions. Bacterial and eukaryotic membranes contain fatty acids ester-linked to \u003cem\u003esn\u003c/em\u003e-glycerol-3-phosphate (G3P), while archaeal membranes possess isoprenoid chains ether-linked to \u003cem\u003esn\u003c/em\u003e-glycerol-1-phosphate (G1P)\u003csup\u003e2\u003c/sup\u003e. This \u0026apos;lipid divide\u0026apos; emerged early in the evolution of life and represents a central evolutionary puzzle\u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eDespite this structural divergence of phospholipid backbones, certain polar headgroups including \u003cem\u003emyo\u003c/em\u003e-inositol are shared across all three domains. In eukaryotes, phosphatidylinositol (PI) serves critical roles beyond membrane structure, functioning as a precursor for phosphoinositides that regulate membrane trafficking, cytoskeletal organization, and signal transduction\u003csup\u003e4, 5\u003c/sup\u003e. The biosynthetic pathways for inositol phospholipids (Figure 1), however, differ fundamentally between domains. In eukaryotes like yeast \u003cem\u003eSaccharomyces cerevisiae\u003c/em\u003e, PI synthase (Pis1p) catalyzes condensation of CDP-diacylglycerol (CDP-DAG) with free \u003cem\u003emyo\u003c/em\u003e-inositol\u003csup\u003e6\u003c/sup\u003e. In contrast, archaeal and bacterial pathways employ d-\u003cem\u003emyo\u003c/em\u003e-inositol-3-phosphate (abbreviated as inositol-3-phosphate in this work; also called l-\u003cem\u003emyo\u003c/em\u003e-inositol-1-phosphate) rather than free \u003cem\u003emyo\u003c/em\u003e-inositol as the headgroup precursor\u003csup\u003e7, 8\u003c/sup\u003e. In archaea, archaetidylinositol phosphate synthase (AIPS) utilizes CDP-archaeol and inositol-3-phosphate to form archaetidylinositol phosphate, which is subsequently dephosphorylated to yield the final product (Figure\u0026nbsp;1)\u003csup\u003e9\u003c/sup\u003e. Despite different substrate preferences, phylogenetic analyses suggest that eukaryotic PI synthases and prokaryotic AIPS enzymes share common ancestry, but the evolutionary origin of inositol lipids, whether bacterial, archaeal, or predating domain divergence, remains actively debated\u003csup\u003e7\u003c/sup\u003e. Understanding enzyme substrate specificity and catalytic flexibility across domains may provide insights into this evolutionary puzzle.\u003c/p\u003e\n\u003cp\u003eIn thermophilic archaea such as \u003cem\u003eThermococcus\u003c/em\u003e and \u003cem\u003ePyrococcus\u003c/em\u003e, archaetidylinositol and its glycosylated derivatives constitute up to 98% of membrane phospholipids\u003csup\u003e10, 11\u003c/sup\u003e, making them the predominant polar lipids. The prevalence of these ether-linked inositol lipids contributes to membrane stability for survival at extreme temperatures, likely through enhanced hydrogen bonding networks and optimized membrane packing\u003csup\u003e12\u003c/sup\u003e. Archaeosomes, liposomes formulated from archaeal lipids, retain this remarkable stability against high temperature, low pH, oxidative stress, and enzymatic degradation\u003csup\u003e13\u003c/sup\u003e, making them attractive for drug delivery, vaccine adjuvants, and other biotechnological applications. For example, clinical studies have shown archaeosomes to be safe and effective adjuvants that stimulate both humoral and cell-mediated immune responses\u003csup\u003e14\u003c/sup\u003e. Despite these advantages, both the challenges in engineering and cultivating thermophilic archaea at industrial scale, and the difficulties in chemical synthesis, have restricted commercial development of archaeosome-based technologies.\u003c/p\u003e\n\u003cp\u003eThese limitations have motivated synthetic biology approaches to produce archaeal lipids in industrially tractable hosts. Previous efforts have demonstrated partial reconstitution of archaeal lipid biosynthesis in microbial chassis such as \u003cem\u003eEscherichia coli\u003c/em\u003e\u003csup\u003e15-18\u003c/sup\u003e and \u003cem\u003eS. cerevisiae\u003c/em\u003e\u003csup\u003e19\u003c/sup\u003e. Our group previously engineered yeast to produce CDP-archaeol through expression of archaeal glycerol-1-phosphate dehydrogenase (G1PDH), geranylgeranylglyceryl phosphate synthase (GGGPS), digeranylgeranylglyceryl phosphate synthase (DGGGPS), and CDP-archaeol synthase (CarS), along with enhanced GGPP supply (Figure 1). While this resulted in hybrid neutral lipids containing both archaeal and eukaryotic components\u003csup\u003e19\u003c/sup\u003e, polar lipid production remained minimal, limiting studies of membrane properties and biotechnological applications. In parallel, efforts in \u003cem\u003eE. coli\u003c/em\u003e have achieved production of archaeal phospholipids with ethanolamine and glycerol polar heads\u003csup\u003e17, 20\u003c/sup\u003e. However, to our knowledge, robust production of archaetidylinositol has never been achieved in heterologous hosts.\u003c/p\u003e\n\u003cp\u003eIn this study, we sought to produce unsaturated di-\u003cem\u003eO\u003c/em\u003e-phytanyl phosphatidylinositol (uDoPhPI) by introducing AIPS to our previously constructed CDP-archaeol-producing yeast. Using modular pathway assembly, we evaluated the contributions of individual archaeal enzymes, assessed strategies to enhance precursor supply, and examined enzyme promiscuity between yeast and archaeal lipid biosynthetic pathways. We characterized the physicochemical properties of purified uDoPhPI, including its ability to form stable liposomes, and evaluated the thermotolerance of both uDoPhPI-containing vesicles and engineered yeast strains.\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003ch3\u003eEngineering and Structural Characterization of uDoPhPI Production in \u003cem\u003eS. cerevisiae\u003c/em\u003e\u003c/h3\u003e\n\u003cp\u003eIn our previous work\u003csup\u003e19\u003c/sup\u003e, we engineered a yeast strain (designated 4U), which is derived from the Gty116 chassis with enhanced GGPP supply\u003csup\u003e21\u003c/sup\u003e, and expressed heterologous archaeal lipid biosynthetic genes from \u003cem\u003eMethanosarcina acetivorans\u003c/em\u003e (MaG1PDH, MaGGGPS, MaDGGGPS) and \u003cem\u003eArchaeoglobus fulgidus\u003c/em\u003e (AfCarS). This strain produced the CDP-archaeol intermediate and accumulated primarily neutral archaeal lipids (DGGGOH) and hybrid DGGGO-acyl ester (DGGGO-FA) species through metabolic competition for the DGGGP precursor (Figure 1). Based on tandem MS (MS/MS) fragmentation patterns, baseline uDoPhPI was detected despite lacking a dedicated AIPS, suggesting promiscuous activity of endogenous yeast PI synthase (Pis1p) toward CDP-archaeol. But production levels in the 4U strain were too low for purification and structural characterization\u003csup\u003e19\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo enhance uDoPhPI production, we considered two candidate genes (Figure 1): \u003cem\u003eAeropyrum pernix\u003c/em\u003e archaetidylinositol phosphate synthase (ApAIPS), whose activity has been confirmed \u003cem\u003ein vitro\u0026nbsp;\u003c/em\u003eto catalyze condensation of CDP-archaeol with inositol-3-phosphate\u003csup\u003e22\u003c/sup\u003e;\u0026nbsp;and yeast \u003cem\u003eINO1\u003c/em\u003e encoding ScIno1p, the rate-limiting enzyme for inositol-3-phosphate biosynthesis\u003csup\u003e23\u003c/sup\u003e. Using modular Golden Gate assembly (Extended Data Fig. 1), we created pathway variants with individual or combined overexpression of ApAIPS and ScIno1p. Semi-quantitative analysis by mass spectrometry (MS) using selected ion monitoring (m/z 896.6\u0026nbsp;\u0026rarr;\u0026nbsp;335.0 [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e) revealed that uDoPhPI production increased in the order: 4U \u0026lt; 4U+A \u0026lt; 4U+I \u0026lt; 4U+AI\u0026nbsp;(Figure 2), identifying the latter as optimal for large-scale purification (Extended Data Fig. 2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe fermented 4 L of the 4U+AI strain under galactose induction, harvesting after 48 hours to obtain 24.0 g dry cell weight (DCW). Modified Bligh-Dyer extraction yielded 2.02 g total lipids, corresponding to 84.2 mg lipid/g DCW, consistent with previous reports\u003csup\u003e24\u003c/sup\u003e. Silica gel chromatography with chloroform-methanol-ammonium hydroxide\u0026nbsp;gradient separated the crude extract (Extended Data Fig. 2a). MS ion monitoring traced uDoPhPI in the 70-80% chloroform fractions, which were pooled and lyophilized to yield 131.4 mg material, corresponding to\u0026nbsp;6.5% of total yeast lipid and a yield of 5.5 mg uDoPhPI/g\u0026nbsp;DCW. High-resolution MS analysis (Extended Data Fig. 2b, c) confirmed the molecular formula C\u003csub\u003e49\u003c/sub\u003eH\u003csub\u003e83\u003c/sub\u003eO\u003csub\u003e11\u003c/sub\u003eP with an observed \u003cem\u003em/z\u0026nbsp;\u003c/em\u003evalue of 896.6010 [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e, matching the calculated mass of 896.6011 (\u0026Delta; = -0.11 ppm).\u0026nbsp;Additionally, the fragmentation patterns of uDoPhPI agree with data in the literature in both positive and negative mode MS/MS spectra (Extended Data Fig. 2d). The isolated uDoPhPI purity was determined to be \u0026gt;96.8% based on A\u003csub\u003e210\u003c/sub\u003e peak area (Extended Data Fig. 2e).\u003c/p\u003e\n\u003cp\u003eNMR spectroscopy further confirmed the uDoPhPI structure (Extended Data Fig. 3\u0026nbsp;and\u0026nbsp;Supplementary Fig. 1). The \u003csup\u003e1\u003c/sup\u003eH NMR spectrum (400 MHz, CDCl\u003csub\u003e3\u003c/sub\u003e/CD\u003csub\u003e3\u003c/sub\u003eOD 5:2) showed diagnostic signals for unsaturated geranylgeranyl chains: \u003cem\u003e\u0026delta;\u003c/em\u003e 5.33 (d, 2H), \u003cem\u003e\u0026delta;\u003c/em\u003e 5.11 (d, 6H), \u003cem\u003e\u0026delta;\u003c/em\u003e 1.90-2.15 (m, 24H), \u003cem\u003e\u0026delta;\u003c/em\u003e 1.67 (s, 12H) and \u003cem\u003e\u0026delta;\u003c/em\u003e 1.60 (s, 18H). The \u003cem\u003emyo\u003c/em\u003e-inositol headgroup appeared as complex multiplets between \u003cem\u003e\u0026delta;\u003c/em\u003e 3.25-4.23, while glycerol signals overlapped in the \u003cem\u003e\u0026delta;\u003c/em\u003e 3.45-4.17 region. \u003csup\u003e13\u003c/sup\u003eC NMR confirmed the structure with characteristic quaternary carbons at \u003cem\u003e\u0026delta;\u003c/em\u003e 140.5-131.0, olefinic carbons at \u003cem\u003e\u0026delta;\u003c/em\u003e124.2-120.3, glycerol backbone at \u003cem\u003e\u0026delta;\u003c/em\u003e 76.8, 69.4, 65.2, and inositol ring carbons between \u003cem\u003e\u0026delta;\u003c/em\u003e 76.4-71.2. These NMR data are consistent with values reported in the literature for similar compounds\u003csup\u003e25, 26\u003c/sup\u003e. The detailed 2D NMR analysis of uDoPhPI (Supplementary Fig. 1) clearly showed the presence of the \u003cem\u003emyo\u003c/em\u003e-inositol headgroup in COSY spectrum, and the linkage of geranylgeranyl chains to the glycerol skeleton in HMBC correlations, further confirmed its structure. Along with HR-MS, these combined spectral data established the structure as unsaturated di-\u003cem\u003eO\u003c/em\u003e-phytanyl phosphatidylinositol with C\u003csub\u003e20\u003c/sub\u003e isoprenoid chains ether-linked to a glycerol phosphate backbone\u0026nbsp;\u003csup\u003e19, 25, 27\u003c/sup\u003e. Notably, the glycerol backbone chirality cannot be assigned using these methods.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMetabolic Impact of uDoPhPI Production on Yeast Lipid Metabolism\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe characterized the impact of uDoPhPI overproduction on cellular lipid metabolism, focusing on endogenous phospholipids, archaeal neutral lipids, and hybrid lipid species. We optimized an LC-MS/MS method with enhanced chromatographic resolution and established multiple reaction monitoring (MRM) for quantitative analysis (Extended Data Fig. 4). For uDoPhPI, we used the purified material to generate calibration curves enabling absolute quantification (Extended Data Fig. 4a). Commercial C\u003csub\u003e34:1\u003c/sub\u003e PI standard was utilized to quantify yeast endogenous PI species (Extended Data Fig. 4b). For neutral lipids including yeast TAGs, archaeal DGGGOH, and hybrid DGGGO-acyl esters (Extended Data Fig. 4c, d, e), we performed relative quantification using MRM ion counts normalized to the 4U baseline strain (Extended Data Fig. 5). To further assess compositional shift trends within each lipid class, individual species were compared using class-specific ion counts (Figure 3), and caution is warranted in interpreting absolute abundances between molecular species due to potential differences in ionization efficiencies.\u003c/p\u003e\n\u003cp\u003eUsing these quantitative approaches, we first examined the inositol phospholipid pools (Figure 2). The 4U baseline produced 11.5\u0026plusmn;1.2 mg uDoPhPI/g DCW despite lacking ApAIPS, confirming endogenous enzyme promiscuity. Expression of ApAIPS in 4U+A resulted in a 2.58\u0026plusmn;0.19-fold increase in uDoPhPI levels, demonstrating functional archaeal enzyme activity in yeast. ScINO1 overexpression in 4U+I yielded a 3.06\u0026plusmn;0.34-fold increase, confirming inositol precursor limitation. The combined 4U+AI strain achieved the highest production at 3.95\u0026plusmn;0.54-fold over baseline. The yield value obtained here using small culture volumes (45.6\u0026plusmn;6.2 mg uDoPhPI/g DCW,\u0026nbsp;Figure 2) was substantially higher than that obtained during large-scale purification (5.5 mg uDoPhPI/g\u0026nbsp;DCW,\u0026nbsp;Extended Data Fig. 2), likely reflecting differences in cultivation conditions, lipid extraction efficiencies, and chromatographic purification losses between the two methods.\u003c/p\u003e\n\u003cp\u003eThis enhanced uDoPhPI production came partially at the expense of endogenous PI (Figure 3a\u0026nbsp;and\u0026nbsp;Extended Data Fig. 5a). For the abundant PI species in yeast\u003csup\u003e28\u003c/sup\u003e, C\u003csub\u003e34:1\u003c/sub\u003e PI and C\u003csub\u003e32:1\u003c/sub\u003e PI, ScINO1 overexpression alone increased their levels by 1.27\u0026plusmn;0.11-fold and 1.25\u0026plusmn;0.24-fold, respectively, confirming inositol precursor limitation for native PI biosynthesis. However, ApAIPS expression either alone (0.98\u0026plusmn;0.05-fold and 1.14\u0026plusmn;0.04-fold) or combined with ScINO1 overexpression (0.82\u0026plusmn;0.12-fold and 1.07\u0026plusmn;0.24-fold) had modest effects on C\u003csub\u003e34:1\u003c/sub\u003e PI and C\u003csub\u003e32:1\u003c/sub\u003e PI relative to the 4U baseline. The comparison between 4U+AI and 4U+I revealed that enhanced inositol supply due to ScIno1p overexpression was partitioned between the native and archaeal PI pathways when ApAIPS was present. For C\u003csub\u003e32:0\u003c/sub\u003e PI, a minor PI molecule in yeast\u003csup\u003e28\u003c/sup\u003e, a distinct pattern was observed. All engineered strains (4U+A, 4U+I, and 4U+AI) showed substantial increases in C\u003csub\u003e32:0\u003c/sub\u003e PI (7.36\u0026plusmn;2.16-fold, 6.44\u0026plusmn;0.75-fold, and 7.91\u0026plusmn;1.06-fold, respectively) relative to 4U, but the impact on overall endogenous PI content remained limited due to the low basal abundance of C\u003csub\u003e32:0\u0026nbsp;\u003c/sub\u003ePI. The total PI content across all strains (4.77-6.41 mg/g CDW,\u0026nbsp;Figure 3a) remained within the range reported for yeast\u003csup\u003e28\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eNotably, neutral and hybrid lipid accumulation increased across all engineered strains (Figure 3b, c\u0026nbsp;and\u0026nbsp;Extended Data Fig. 5b, c, d). DGGGOH levels rose 5.36\u0026plusmn;1.30-fold with ApAIPS, 7.09\u0026plusmn;2.15-fold with ScIno1p, and 5.95\u0026plusmn;1.19-fold with both modifications (Figure 3c\u0026nbsp;and\u0026nbsp;Extended Data Fig. 5c). For the hybrid DGGGO-acyl esters (Figure 3c\u0026nbsp;and\u0026nbsp;Extended Data Fig. 5d), the abundant DGGGO-C\u003csub\u003e16:1\u003c/sub\u003e and DGGGO-C\u003csub\u003e16:0\u003c/sub\u003e showed comparable enhancement in fold changes relative to DGGGOH, whereas the less abundant species showed more dramatic increases, similar to the case with the minor C\u003csub\u003e32:0\u0026nbsp;\u003c/sub\u003ePI. TAGs also increased in all engineered strains (Figure 3b\u0026nbsp;and\u0026nbsp;Extended Data Fig. 5b), with C\u003csub\u003e45:1\u0026nbsp;\u003c/sub\u003eTAG showing the most dramatic enhancement (8.90\u0026plusmn;0.73-fold) in 4U+A. Flow cytometric analysis of Nile Red-stained cells confirmed elevated neutral lipid accumulation (Extended Data Fig. 6), with mean fluorescence intensity increasing\u0026nbsp;2.57\u0026plusmn;0.51-fold in 4U+A,\u0026nbsp;2.92\u0026plusmn;0.60-fold in 4U+I, and 4.73\u0026plusmn;0.94-fold in 4U+AI relative to 4U. These pleiotropic effects suggest that manipulating inositol metabolism impacts broader lipid homeostasis beyond the targeted phospholipid pathways, consistent with previous findings on crosstalk between storage lipid metabolism and PI biosynthesis\u003csup\u003e29\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDissecting Enzyme Promiscuity in the Inositol Lipid Pathway\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHaving observed promiscuous activity of yeast Pis1p toward CDP-archaeol in the 4U strain, we investigated whether other endogenous enzymes exhibit similar cross-domain substrate flexibility. Using Golden Gate assembly for modular gene replacement (Extended Data Fig. 1), we substituted individual archaeal genes with Venus fluorescent protein in the 4U+AI background to assess the contribution of each enzyme and identify potential metabolic crosstalk between archaeal and eukaryotic pathways.\u003c/p\u003e\n\u003cp\u003eWe first examined MaG1PDH (Figure 2), which generates the stereochemically distinct glycerol-1-phosphate backbone characteristic of archaeal lipids. Replacement of MaG1PDH in the 4U-G+AI strain resulted in severely reduced uDoPhPI production (\u0026gt;77% decrease relative to 4U+AI) yet detectable levels persisted (0.89\u0026plusmn;0.03-fold relative to the 4U baseline). Based on our previous stereochemical analysis showing that yeast-produced DGGGOH contains exclusively the G1P backbone even without heterologous G1PDH expression\u003csup\u003e19\u003c/sup\u003e, we infer that the residual uDoPhPI also likely contains the archaeal G1P backbone. However, we note that direct stereochemical confirmation of uDoPhPI chirality would be needed to conclusively establish this inference, as our current NMR data do not allow determination of glycerol chirality. The endogenous G1P synthesis reveals a cryptic metabolic capability in \u003cem\u003eS. cerevisiae\u003c/em\u003e that parallels observations in \u003cem\u003eE. coli\u003c/em\u003e\u003csup\u003e17\u003c/sup\u003e, though the substantial reduction in uDoPhPI levels upon G1PDH deletion confirms that archaeal G1PDH remains essential for efficient production.\u003c/p\u003e\n\u003cp\u003eInterestingly, removal of AfCarS in 4U-C+AI maintained robust uDoPhPI production at 2.32\u0026plusmn;0.18-fold relative to the 4U baseline (Figure 2), representing 58% of the complete pathway level in 4U+AI. This demonstrates that endogenous yeast CDP-diacylglycerol synthase Cds1p promiscuously accepts the archaeal DGGGP substrate to generate CDP-archaeol, paralleling the flexibility of Pis1p in accepting CDP-archaeol for PI synthesis. These differential impacts reveal distinct metabolic constraints: while G1P supply requires dedicated archaeal enzymes due to limited endogenous activity, both the CDP-activation and PI synthesis steps exhibit considerable enzymatic promiscuity. This suggests that the evolutionary lipid divide between domains may be maintained more by differential precursor availability than by absolute enzymatic specificity, with key transferases retaining ancestral flexibility for structurally diverse substrates.\u003c/p\u003e\n\u003cp\u003eAs our previous work\u003csup\u003e19\u003c/sup\u003e established that MaGGGPS and MaDGGGPS are essential for DGGGOH production in yeast with complete loss upon their deletion, their testing for uDoPhPI production was not performed in this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEnhanced Thermotolerance in uDoPhPI-Producing Yeast Strains\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate whether uDoPhPI production confers stress resistance similar to that observed in \u003cem\u003eE. coli\u003c/em\u003e strains containing archaeal ether lipids\u003csup\u003e17, 30\u003c/sup\u003e, we subjected engineered yeast strains to heat shock at 50\u0026deg;C and monitored survival by colony-forming unit counts (Figure\u0026nbsp;4). The 4U+AI strain, which produces the highest uDoPhPI levels (Figure 2), exhibited markedly enhanced thermotolerance compared to controls. This strain retained 76.54\u0026plusmn;6.36% viability after 10 minutes at 50\u0026deg;C, while the parent Gty116 and 4U strains showed only 37.81\u0026plusmn;16.38% and 4.24\u0026plusmn;1.90% survival, respectively. The protective effect persisted throughout the time course, with 4U+AI maintaining 43.85\u0026plusmn;3.88% viability at 30 minutes and 5.75\u0026plusmn;1.63% at 60 minutes, when all other strains showed nearly complete loss of viability (Figure\u0026nbsp;4).\u003c/p\u003e\n\u003cp\u003eConversely, the 4U-G+AI strain with severely reduced uDoPhPI production (Figure 2) displayed the poorest thermotolerance (Figure 4), with only 1.92\u0026plusmn;1.92% survival at 10 minutes and complete loss of viability by 20 minutes. This inverse correlation between uDoPhPI levels and heat sensitivity indicates that archaeal phospholipid content directly influences membrane thermostability in yeast.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFormation and Characterization of uDoPhPI liposomes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eArchaeal lipids form stable archaeosomes that have been explored as drug delivery vehicles and vaccine adjuvants. Previous studies of archaetidylinositol-based liposomes have relied on either chemically synthesized, saturated archaetidylinositol (di-\u003cem\u003eO\u003c/em\u003e-phytanyl phosphatidylinositol, DoPhPI)\u003csup\u003e26\u003c/sup\u003e or natural lipid extracts from thermophilic archaea containing predominantly glycosylated forms (~90% glucosyl-DoPhPI, ~10% DoPhPI)\u003csup\u003e13\u003c/sup\u003e. Whether non-glycosylated, unsaturated DoPhPI produced through heterologous yeast biosynthesis could form stable vesicular structures, however, remained unknown.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe therefore investigated the membrane-forming capacity and stability of our yeast-derived uDoPhPI. First, using the electroformation method, giant unilamellar vesicles (GUVs) were successfully prepared from uDoPhPI/DOPE-Atto647 (1000:1) with uniform spherical morphology and efficient calcein encapsulation, demonstrating intact bilayer formation (Figure 5a).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor small unilamellar vesicle (SUV) characterization, purified uDoPhPI was hydrated in calcein solution and extruded through 100 nm polycarbonate membranes. Dynamic light scattering analysis (DLS) revealed a monodisperse population with mean diameter of 111.0\u0026plusmn;4.1 nm. DPPC liposomes prepared using the identical method showed a larger average diameter of 210.1\u0026plusmn;6.2 nm, potentially reflecting differences in membrane curvature stress between ester and ether lipids. During four-week storage at 4\u0026deg;C, uDoPhPI archaeosomes demonstrated colloidal stability with minimal size change (117.2\u0026plusmn;0.2 nm) (Figure 5b), while DPPC liposomes rapidly aggregated, preventing accurate DLS analysis after a few hours. Membrane integrity assessment showed DPPC liposomes underwent substantial calcein leakage, increasing from 13.5% to 53.6% by four weeks (Figure 5c). In contrast, uDoPhPI archaeosomes effectively retained their cargo with only marginal leakage increase from 8.3% to 13.4% (Figure 5c), demonstrating superior long-term stability.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo assess SUV thermal stability, we monitored temperature-dependent calcein release in the range of 25-95\u0026deg;C (Figure 5d). DPPC liposomes exhibited minimal leakage below their phase transition (42\u0026deg;C) but showed rapid release reaching ~90% at 50\u0026deg;C. In contrast, uDoPhPI liposomes retained nearly complete calcein encapsulation across the entire temperature range with no detectable release even at 95\u0026deg;C (Figure 5d). Our results demonstrate that unsaturated DoPhPI, despite lacking the saturation and glycosylation typical of natural archaeal membranes, retains the vesicle-forming capacity and remarkable thermostability suitable for biotechnological applications.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study establishes robust production of uDoPhPI in engineered \u003cem\u003eS. cerevisiae\u003c/em\u003e,\u0026nbsp;offering advantages over extraction from extremophilic archaea or complex chemical synthesis. This was achieved through combined expression of archaeal AIPS and overexpression of yeast inositol-3-phosphate synthase in a previously engineered chassis overproducing the CDP-archaeol precursor. Accumulation of uDoPhPI (45.6\u0026plusmn;6.2 mg/g DCW) substantially exceeded endogenous PI levels (6.4\u0026plusmn;1.6 mg/g DCW) by \u0026gt;6-fold, suggesting that the distinct structures of archaeal lipids may escape feedback mechanisms regulating native PI homeostasis in yeast. Metabolic engineering also triggered pleiotropic effects beyond the targeted inositol lipid pathway. Enhanced neutral lipid accumulation may reflect either a sink mechanism for fatty acids displaced from phospholipid biosynthesis or altered metabolic regulation. The redistribution across neutral, hybrid, and phospholipid pools demonstrates cellular flexibility to accommodate fundamentally different membrane chemistries, providing a platform for investigating evolutionary transitions in membrane composition.\u003c/p\u003e\n\u003cp\u003eBaseline uDoPhPI production in yeast strains lacking archaeal AIPS or CarS expression reveals remarkable promiscuity of yeast enzymes, likely reflecting ancestral substrate flexibility. Moreover, detection of endogenous G1P synthesis in yeast, evidenced by residual DGGGOH and uDoPhPI production without heterologous G1PDH, parallels observations in \u003cem\u003eE. coli\u003c/em\u003e\u003csup\u003e17\u003c/sup\u003e and an expanding catalog of bacterial species harboring cryptic archaeal-like capabilities\u003csup\u003e31, 32\u003c/sup\u003e. This latent G1P biosynthesis capability in yeast, combined with promiscuous metabolism of archaeal intermediates, supports models in which LUCA possessed mixed membranes with subsequent domain-specific optimization\u003csup\u003e2\u003c/sup\u003e, rather than strict incompatibility driving the lipid divide.\u003c/p\u003e\n\u003cp\u003eThe direct correlation between cellular uDoPhPI content and heat resistance parallels both the prevalence of archaetidylinositol in hyperthermophilic archaea\u003csup\u003e10, 11\u003c/sup\u003e and the thermostability of archaetidylinositol-based vesicles\u003csup\u003e13, 26\u003c/sup\u003e. Formation of stable, uniform SUVs with remarkable colloidal stability positions uDoPhPI archaeosomes as promising vehicles for drug delivery and vaccine formulations\u003csup\u003e14, 33\u003c/sup\u003e. Additionally, the enhanced organismal robustness suggests potential applications in engineering industrial yeast strains for high-temperature bioprocesses\u003csup\u003e34, 35\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFuture work may pursue three directions. First, systems metabolic engineering could be employed to approach the high yields\u003cem\u003e\u0026nbsp;\u003c/em\u003e(20-30% of total lipids) achieved in \u003cem\u003eE. coli\u003c/em\u003e\u003csup\u003e17\u003c/sup\u003e. Second, introduction of lipid structural modifications, including isoprenoid chain saturation for enhanced oxidative stability\u003csup\u003e18, 36\u003c/sup\u003e and alternative archaeal polar headgroups, could enable designer archaeosomes with tailored properties. Third, mechanistic studies of uDoPhPI subcellular localization, dynamics, and interactions with membrane proteins would provide insights into the emergence of complex endomembrane systems in eukaryotes. Such studies are particularly relevant, as Asgard archaea, the closest archaeal relatives to eukaryotes, encode both AIPS homologs (Supplementary Fig. 2) and eukaryotic signature proteins that may interact with inositol lipids\u003csup\u003e37, 38\u003c/sup\u003e. Together, our platform for producing archaeal phospholipids in a genetically tractable eukaryote opens new avenues for studying membrane evolution, developing thermostable delivery vehicles, and engineering organisms with hybrid membrane properties optimized for extreme environments.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eRecombinant DNA construction\u003c/p\u003e\n\u003cp\u003ePlasmids constructed and used in this study are listed in\u0026nbsp;Supplementary Table 1. Synthetic primers and genes were purchased from Genewiz (Suzhou, China) and listed in\u0026nbsp;Supplementary Table 2. The\u003cem\u003e\u0026nbsp;\u003c/em\u003eApAIPS gene (\u003cem\u003eAeropyrum pernix\u003c/em\u003e, APE1526) was synthesized with codon optimization for expression in \u003cem\u003eS. cerevisiae\u003c/em\u003e. The yeast ScINO1 gene (YJL153C)\u0026nbsp;was amplified from \u003cem\u003eS. cerevisiae\u003c/em\u003e genomic DNA. Restriction endonucleases, T4 DNA ligase, and Gibson Assembly kit were purchased from New England Biolabs (Ipswich, MA, USA). Phanta Max Master Mix was purchased from Vazyme Biotech Co., Ltd (Nanjing, China). QIAprep Spin Plasmid Mini-prep kits from Qiagen (Valencia, CA, USA) were employed to prepare plasmid DNA from \u003cem\u003eE. coli\u003c/em\u003e. PCR and digestion products were purified by QIAquick PCR Purification and Gel Extraction kits (Qiagen). Genomic DNA isolation was performed using the YeaStar Genomic DNA kit (Zymo Research).\u003c/p\u003e\n\u003cp\u003eArchaeal pathway plasmids were constructed using the Golden Gate Assembly MoClo-YTK kit (Addgene, #1000000061) \u003csup\u003e39\u003c/sup\u003e with modifications. The system employed a hierarchical two-level assembly strategy (Extended Data Fig. 1). Briefly, Level 1 plasmids (pYTK095 series) served as individual transcription units containing the inducible promoter P\u003csub\u003eGAL1\u003c/sub\u003e, CDS, terminator T\u003csub\u003eENO1\u003c/sub\u003e, 6×His_3×Flag tags, and linking elements (ConL, ConR) with \u003cem\u003eBsmB\u003c/em\u003eI restriction sites. Level 1 plasmids housing archaeal lipid biosynthesis genes (MaG1PDH, MaGGGPS, MaDGGGPS, and AfCarS) were constructed in our previous study \u003csup\u003e19\u003c/sup\u003e. For the present work, new Level 1 plasmids for gene overexpression (ApAIPS or ScINO1) or gene replacement (with Venus fluorescent protein gene) were generated by two-fragment Gibson assembly of (1) PCR products or synthetic genes with 20-30 bp flanking homology and (2) PCR-amplified pYTK095 backbone vectors.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLevel 2 plasmids (pYTK096 series) contained complete metabolic pathways composed of multiple transcription units, along with homologous recombination sequences for integration at the 106a locus in the \u003cem\u003eS. cerevisiae\u003c/em\u003e genome. To assemble the Level 2 plasmids, six Level 1 plasmids, including pYTK095-MaG1PDH, pYTK095-MaGGGPS, pYTK095-MaDGGGPS, pYTK095-AfCarS, pYTK095-ApAIPS and pYTK095-ScINO1, were mixed with the Level 2 acceptor vector pYTK096-ccdB in \u003cem\u003eBsmB\u003c/em\u003eI assembly reactions. For gene deletion variants, the targeted Level 1 plasmid was substituted with its Venus replacement counterpart. Golden Gate assembly reactions contained 0.02 pmol of each DNA component, 1000 U T4 DNA ligase, 0.5 μL of T4 ligase buffer, 5 U \u003cem\u003eBsmB\u003c/em\u003eI restriction endonuclease, and water to a 5 μL final volume. The assembly protocol consisted of 45 cycles of 42°C for 1 min and 16°C for 1 min, followed by 65°C for 20 min.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAssembly products were transformed into \u003cem\u003eE. coli\u0026nbsp;\u003c/em\u003eDH5α by standard chemical transformation. Transformants were selected at 37°C on Luria-Bertani (LB) agar plates supplemented with either 50 μg/mL kanamycin or 100 μg/mL ampicillin, followed by liquid cultivation for plasmid isolation.\u003c/p\u003e\n\u003cp\u003eYeast strain construction\u003c/p\u003e\n\u003cp\u003eYeast strains used in this study are listed in Supplementary Table 3.\u003cem\u003e\u0026nbsp;S. cerevisiae\u003c/em\u003e Gty116 was a generous gift from Professor Jay D. Keasling \u003csup\u003e21\u003c/sup\u003e. The baseline 4U strain \u003csup\u003e19\u003c/sup\u003e, constructed previously in the Gty116 chassis, contains \u003cem\u003eM. acetivorans\u003c/em\u003e G1PDH (Ma3686), GGGPS (Ma3969), DGGGPS (Ma0961), and \u003cem\u003eA. fulgidus\u003c/em\u003e CarS (AF1740). Sequence-verified Level 2 plasmids were linearized with \u003cem\u003eNot\u003c/em\u003eI and used to transform Gty116 competent cells using the standard heat shock method \u003csup\u003e40\u003c/sup\u003e. The plasmid pCUT-KanMX encoding the CRISPR/Cas system was co-transformed with linearized donor DNA to induce a double-strand break at the 106a locus, thereby allowing efficient and scarless integration of the donor DNA. Transformants were selected on YPD agar medium supplemented with 800 μg/mL G418 (Beijing OKA Biotech, Beijing, China) and correct genomic integration was confirmed by diagnostic colony PCR.\u003c/p\u003e\n\u003cp\u003eProduction, purification, and structural analysis of uDoPhPI\u003c/p\u003e\n\u003cp\u003eTo prepare seed cultures,\u0026nbsp;the highest-producing strain (4U+AI)\u0026nbsp;was first cultivated in baffled shake flasks containing YPD medium (1% yeast extract, 2% peptone, 2% glucose) at 30°C and 200 rpm. For inducible production, cells were then inoculated at an initial OD\u003csub\u003e600\u003c/sub\u003e of 0.1 in 4 L of YPG medium (1% yeast extract, 2% peptone, 2% galactose) and cultured under the same conditions for 48 hours.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor uDoPhPI purification (Extended Data Fig. 2), cells were harvested by centrifugation, yielding 24 g dry cell weight. The cell pellet was ground with a mortar and pestle in liquid nitrogen. The crude extract was mixed with 500 mL of methanol/chloroform (2:1, v/v) in a 2 L separating funnel. Phase separation was induced by sequential addition of 200 mL of chloroform and 200 mL of water. The lower organic phase was collected, and extraction was repeated 3-5 times on the remaining material. Pooled organic extracts were concentrated using a rotary evaporator. The concentrated extract was redissolved in 2 mL of n-hexane and 2 mL of chloroform and applied to a silica gel column (200-300 mesh). uDoPhPI was eluted using a chloroform-methanol-ammonium hydroxide gradient (75:25:0.4, v/v/v), with fractions monitored by mass spectrometry (m/z 896.6 [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e). Fractions containing uDoPhPI were pooled and lyophilized, yielding 131.4 mg of purified material. Purity was assessed by HPLC using an Agilent 1290 HPLC instrument consisting of a G7104A flexible pump, a G7117B DAD detector, and a G7167B autosampler.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor NMR analysis, purified uDoPhPI was dissolved in 500 µL of deuterated chloroform and 200 µL of deuterated methanol (70 mg/mL), then NMR data was acquired on a 400 MHz Bruker Avance III spectrometer. NMR data are reported as follows (Extended Data Fig. 3\u0026nbsp;and\u0026nbsp;Supplementary Fig. 1): chemical shifts (\u003cem\u003eδ\u003c/em\u003e, in ppm), multiplicity (s = singlet, d = doublet, t = triplet), coupling constants (\u003cem\u003eJ\u003c/em\u003e, in Hz), and integration, using TMS as the internal reference. Topspin ver. 4.5.0 (Bruker Biospin) was used for NMR spectral acquisition and data processing.\u003c/p\u003e\n\u003cp\u003eHigh-resolution mass spectrometry (HR-MS) was performed using a timsTOF Flex mass spectrometer (Bruker Daltonics) to confirm the molecular formula. Mass spectra were acquired in electrospray ionization (ESI) positive ion mode with the following parameters: tims ramp time 100 ms, PASEF on, 100 ms, scan range m/z, 100–1350; 1/k\u003csub\u003e0\u003c/sub\u003e, 0.55–1.87 V.s/cm\u003csup\u003e2\u003c/sup\u003e). Untargeted profiling data were acquired in ESI positive ion mode and negative ion modes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eYeast lipid extraction for quantification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYeast cultivation and inducible production were performed as described above. After 48 hours, ~2 mL aliquots corresponding to 40 OD\u003csub\u003e600\u003c/sub\u003e units (equivalent to 12 mg dry cell weight) were collected for total lipids extraction using a modified Bligh and Dyer method optimized for yeast phospholipids \u003csup\u003e41\u003c/sup\u003e. Cells were pelleted by centrifugation (10,000 × g, 5 min) and lysed by vigorous agitation with 0.1 mm glass beads (3 cycles of 2 min vortexing with 1 min cooling on ice) directly in the chloroform/methanol extraction solvent (1:2 v/v). The mixture was thoroughly vortexed for 1 min at room temperature. Phase separation was induced by adding chloroform and water (1:1 v/v). After centrifugation (2,000 × g, 5 min), the lower organic layer containing lipids was carefully collected. The upper aqueous layer and interface were re-extracted once using the same procedure, and the organic phases were pooled. The combined organic extracts were dried under a nitrogen stream and stored at -80°C until analysis. Before analysis, lipids were resuspended in 500 µL of methanol/isopropanol (1:1 v/v).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eQuantitative Lipid Analysis by LC-MS/MS\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLipid extracts were analyzed using a triple quadrupole LC-MS/MS system equipped with an Agilent 1290 HPLC and an Agilent 6470B triple quadrupole mass spectrometer operated in ESI positive ion mode. Quantification was performed using multiple reaction monitoring (MRM) mode (Extended Data Fig. 4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThree different chromatographic methods were employed to achieve optimal separation of different lipid classes. For archaeal uDoPhPI and DGGGOH (Method 1), chromatographic separation was achieved using two C18 AR UPLC columns (2.1 × 150 mm, 2 µm; ACE) connected in series and maintained at 55 °C with a flow rate of 0.5 mL/min. A 2 µL sample volume was injected and eluted with the following gradient: 0 min, 72% B; 8 min, 100% B; 11 min, 72% B, followed by 8 min equilibration. Mobile phase A consisted of 50% acetonitrile and 50% water with 5 mM ammonium acetate, and mobile phase B consisted of 50% isopropanol and 50% acetonitrile supplemented with 0.1% aqueous solution of 4 mM ammonium formate. Samples were diluted 1000-fold for DGGGOH detection to avoid saturation. MRM transitions were m/z 896.6 → 335 (uDoPhPI, [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e) and 654.4 → 273.3 (DGGGOH, [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e) at fragmentor voltage 135 V and collision energy 20 V.\u003c/p\u003e\n\u003cp\u003eFor endogenous yeast phosphatidylinositol species (Method 2), a single C\u003csub\u003e18\u003c/sub\u003e AR UPLC column (2.1 × 150 mm, 2 µm; ACE) was used with the same chromatographic and MS conditions as Method 1. MRM transitions were m/z 854.6 → 577.5 (C\u003csub\u003e34:1\u003c/sub\u003e PI, [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e), 828.5 → 551.5 (C\u003csub\u003e32:0\u003c/sub\u003e PI, [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e), and 826.5 → 549.6 (C\u003csub\u003e32:1\u003c/sub\u003e PI, [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e).\u003c/p\u003e\n\u003cp\u003eFor yeast TAGs and hybrid DGGGO-acyl esters (Method 3), these neutral lipids were separated on a single C18 AR UPLC column (2.1 × 150 mm, 2 µm; ACE) maintained at 55°C with a flow rate of 0.2 mL/min. A 2 µL sample was injected and eluted with the following gradient: 0 min, 0% B; 3.3 min, 0% B; 5 min, 24% B; 23 min, 65% B; 24 min, 90% B; 28 min, 90% B, followed by 5 min equilibration. Mobile phase A was methanol and mobile phase B was isopropanol, both containing formic acid and ammonium hydroxide (100:0.04:0.1, v/v/v). MRM transitions were monitored at fragmentor voltage 135 V and collision energy 10 V. For yeast TAGs, the three most abundant species were analyzed using the quantifier ion m/z 549.5 and corresponding [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e precursor ions for C\u003csub\u003e45:1\u003c/sub\u003e TAG (m/z 738.6), C\u003csub\u003e47:1\u003c/sub\u003e TAG (m/z 766.5), and C\u003csub\u003e53:2\u003c/sub\u003e TAG (m/z 848.6). For DGGGO-acyl esters, the quantifier ion of m/z 273.3 was selected for relative quantification of corresponding [M+NH\u003csub\u003e4\u003c/sub\u003e]\u003csup\u003e+\u003c/sup\u003e precursor ions for DGGGO-C\u003csub\u003e8:0\u003c/sub\u003e (m/z 780.7), DGGGO-C\u003csub\u003e10:0\u003c/sub\u003e (m/z 808.6), DGGGO-C\u003csub\u003e12:0\u0026nbsp;\u003c/sub\u003e(m/z 836.6), DGGGO-C\u003csub\u003e14:1\u0026nbsp;\u003c/sub\u003e(m/z 862.6), DGGGO-C\u003csub\u003e16:1\u003c/sub\u003e (m/z 890.7), DGGGO-C\u003csub\u003e16:0\u0026nbsp;\u003c/sub\u003e(m/z 892.7), DGGGO-C\u003csub\u003e18:1\u003c/sub\u003e (m/z 918.7), and DGGGO-C\u003csub\u003e18:0\u003c/sub\u003e (m/z 920.7).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFlow cytometric quantification of cellular lipids\u003c/p\u003e\n\u003cp\u003eNeutral lipid content was assessed by Nile Red staining and flow cytometry \u003csup\u003e42\u003c/sup\u003e. After 48-hour galactose induction as described above, cells were collected, washed twice with phosphate buffered saline (PBS, pH 7.4), and stained with 100 μM Nile Red (Thermo Fisher Scientific, Waltham, MA, USA) for 30 minutes at room temperature in the dark. Stained cells were washed, resuspended in PBS, and analyzed using a CytoFLEX S Flow Cytometer (Beckman Coulter, Indianapolis, IN, USA) with 488 nm excitation and 585 nm emission detection. For each sample, 10,000 gated events were collected, and mean fluorescence intensity was determined using CytExpert software to quantify relative neutral lipid levels across strains.\u003c/p\u003e\n\u003cp\u003eThermotolerance Assay\u003c/p\u003e\n\u003cp\u003eThe engineered yeast strains were cultured under galactose induction for 48 hours as described above. The final cultures were diluted to an OD\u003csub\u003e600\u003c/sub\u003e of 0.001 in YPD medium to ensure suitable cell density for colony-forming unit (CFU) quantification. Equal volumes (20 μL) of diluted cultures were exposed to 50°C for different durations (0, 10, 20, 30, and 60 min), while untreated samples maintained at 30°C served as controls. Following thermal treatment, samples were immediately plated on YPD agar medium and incubated at 30°C for 48 hours. Colonies were counted and survival rates were calculated as the percentage of viable cells relative to the untreated control at each time point. Data are presented as the mean ± standard error of the mean (SEM) from four independent biological replicates.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGiant Unilamellar Vesicle (GUV) Preparation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCommercial reference lipids, including DOPE-Atto647N (1,2-dioleoyl-\u003cem\u003esn\u003c/em\u003e-glycero-3-phosphoethanolamine labeled with Atto647N) and 1,2-dipalmitoyl-\u003cem\u003esn\u003c/em\u003e-glycero-3-phosphocholine (DPPC), were purchased from Avanti Polar Lipids (Alabaster, AL, USA), and stored in chloroform at -20°C until use.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGUVs were prepared using the electroformation method \u003csup\u003e43\u003c/sup\u003e using a VesiclePrepPro device (Nanion Technologies GmbH, Munich, Germany). Purified uDoPhPI and DOPE-Atto647N were mixed at a 1000:1 molar ratio in chloroform. A 15 μL aliquot of the lipid mixture (1 mg/mL) was spread homogeneously on the conductive side of an indium tin oxide (ITO)-coated glass coverslide (Visiontek Systems Ltd, Chester, UK). The lipid-coated ITO slide was placed under vacuum for at least 15 min to ensure complete solvent evaporation. A rubber O-ring (18 mm inner diameter) was positioned on the lipid-coated surface and filled with 280 μL of 200 mM sucrose solution containing 20 μM calcein. A second ITO-coated slide was placed on top to seal the chamber. The assembled electroformation chamber was connected to VesiclePrepPro electrodes, and an alternating current field was applied using a multi-step protocol as described in\u0026nbsp;Supplementary Table 4. Following electroformation, GUVs were harvested by gentle centrifugation and visualized immediately by Nikon AX confocal laser scanning microscopy using a CFI Plan Apochromat Lambda D 60× oil objective (Nikon, Tokyo, Japan) with excitation at 488 nm (calcein) and 647 nm (DOPE-Atto647N).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSmall Unilamellar Vesicle (SUV) Preparation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCalcein-loaded SUVs were prepared by the thin-film hydration and extrusion method \u003csup\u003e44\u003c/sup\u003e. Purified uDoPhPI (5 mg) in chloroform was transferred to an amber glass vial and the solvent was evaporated under a stream of nitrogen, followed by drying under vacuum for 30 min to remove residual solvent. The dried lipid film was hydrated with 500 μL of 106 mM calcein solution, vortexed vigorously, and sonicated for 30 min in an ultrasonic bath at 45°C. The resulting multilamellar vesicle suspension was extruded 21 times through a polycarbonate membrane (100 nm pore size) using a mini-extruder (Avanti Polar Lipids) to generate SUVs. Unencapsulated calcein was removed by gel filtration on a Sephadex G-50 column (Sigma-Aldrich, St. Louis, MO, USA) pre-equilibrated with PBS (pH 7.0) without calcium and magnesium\u0026nbsp;(Cytiva, Chicago, IL, USA). DPPC liposomes were prepared using the identical procedure. Particle size distributions and polydispersity indices were measured by dynamic light scattering (DLS) using a DynaPro NanoStar (Wyatt Technology, Santa Barbara, CA, USA) at 25°C.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTemperature-dependent Calcein Release Assay\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTemperature-dependent calcein release from uDoPhPI and DPPC vesicles was assessed at pH 7.0 over the temperature range of 25-95°C. Liposome suspensions were diluted to 0.1 mg/mL lipid concentration in PBS. Samples were incubated at each test temperature for 5 min, and 200 μL aliquots were dispensed in triplicate into a Corning black 96-well microplate (Catalog#3916, Corning, NY, USA). Fluorescence intensity was measured using an Agilent BioTek Synergy H1 microplate reader (Winooski, Vermont, USA) with excitation and emission wavelengths of 488 nm and 528 nm, respectively. The percentage of calcein released was calculated as F / F\u003csub\u003emax\u003c/sub\u003e × 100%, where F represents the fluorescence intensity after incubation at the test temperature, and F\u003csub\u003emax\u003c/sub\u003e represents the maximal fluorescence obtained after complete vesicle lysis with 2 μL of 10% (v/v) Triton X-100 added at the test temperature.\u003c/p\u003e\n\u003cp\u003eBioinformatics Analysis\u003c/p\u003e\n\u003cp\u003eFor AIPS sequence similarity network (SSN) analysis, ApAIPS was used as a query to perform a protein BLAST search in the NCBI ClusteredNR database (limited to Asgard group, taxid:1935183) at a threshold of e-value\u0026lt;10\u003csup\u003e-5\u003c/sup\u003e. The top 100 BLAST hits were retrieved and pooled with 458 archaeal AIPS sequences from InterPro family IPR054868 (accessed on Oct 17, 2025). SSN analysis was performed using the online EFI-EST tool\u003csup\u003e45\u003c/sup\u003e (https://efi.igb.illinois.edu/efi-est/, FASTA mode) with an alignment score threshold of 30. The 70% identity representative node network comprising 313 nodes and 48,803 edges was downloaded and visualized using Cytoscape (Version: 3.10.3).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eDeclaration of competing interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing financial interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was financially supported by the National Key Research and Development Program of China (2021YFA0910800), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0480000), Guangdong S\u0026amp;T Program (2024B1111140001), and Shenzhen Science and Technology Program (RCJC20221008092810021).\u0026nbsp;We thank the Shenzhen Synthetic Biology Infrastructure for instrument support and technical assistance. We thank Professor Jay D. Keasling for providing \u003cem\u003eS. cerevisiae\u003c/em\u003e Gty116.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData are available from the corresponding authors upon reasonable request.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eWoese, C.R., Kandler, O. \u0026amp; Wheelis, M.L. Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e, 4576\u0026ndash;4579 (1990).\u003c/li\u003e\n\u003cli\u003eLombard, J., Lopez-Garcia, P. \u0026amp; Moreira, D. The early evolution of lipid membranes and the three domains of life. \u003cem\u003eNat Rev Microbiol\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 507\u0026ndash;515 (2012).\u003c/li\u003e\n\u003cli\u003eKoga, Y. Early Evolution of Membrane Lipids: How did the Lipid Divide Occur? \u003cem\u003eJ Mol Evol\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e, 274\u0026ndash;282 (2011).\u003c/li\u003e\n\u003cli\u003ePosor, Y., Jang, W. \u0026amp; Haucke, V. Phosphoinositides as membrane organizers. \u003cem\u003eNat Rev Mol Cell Biol\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 797\u0026ndash;816 (2022).\u003c/li\u003e\n\u003cli\u003eSteinfeld, N., Giridharan, S.S.P., Kauffman, E.J. \u0026amp; Weisman, L.S. in Phosphoinositides: Methods and Protocols. (ed. R.J. Botelho) 1\u0026ndash;17 (Springer US, New York, NY; 2021).\u003c/li\u003e\n\u003cli\u003eNikawa, J., Kodaki, T. \u0026amp; Yamashita, S. Primary structure and disruption of the phosphatidylinositol synthase gene of \u003cem\u003eSaccharomyces cerevisiae\u003c/em\u003e. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e262\u003c/strong\u003e, 4876\u0026ndash;4881 (1987).\u003c/li\u003e\n\u003cli\u003eEsposti, M.D. Eukaryotes inherited inositol lipids from bacteria: implications for the models of eukaryogenesis. \u003cem\u003eFEBS Lett\u003c/em\u003e \u003cstrong\u003e597\u003c/strong\u003e, 2484\u0026ndash;2496 (2023).\u003c/li\u003e\n\u003cli\u003eMorii, H., Ogawa, M., Fukuda, K., Taniguchi, H. \u0026amp; Koga, Y. A revised biosynthetic pathway for phosphatidylinositol in Mycobacteria. \u003cem\u003eJ Biochem\u003c/em\u003e \u003cstrong\u003e148\u003c/strong\u003e, 593\u0026ndash;602 (2010).\u003c/li\u003e\n\u003cli\u003eMorii, H., Kiyonari, S., Ishino, Y. \u0026amp; Koga, Y. A Novel Biosynthetic Pathway of Archaetidyl-myo-inositol via Archaetidyl-myo-inositol Phosphate from CDP-archaeol and d-Glucose 6-Phosphate in Methanoarchaeon Methanothermobacter thermautotrophicus Cells. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e284\u003c/strong\u003e, 30766\u0026ndash;30774 (2009).\u003c/li\u003e\n\u003cli\u003eTourte, M. et al. Novel Intact Polar and Core Lipid Compositions in the \u003cem\u003ePyrococcus\u003c/em\u003e Model Species, \u003cem\u003eP. furiosus\u003c/em\u003e and \u003cem\u003eP. yayanosii\u003c/em\u003e, Reveal the Largest Lipid Diversity Amongst Thermococcales. \u003cem\u003eBiomolecules\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e (2020).\u003c/li\u003e\n\u003cli\u003eMorii, H. et al. A novel phosphoglycolipid archaetidyl(glucosyl)inositol with two sesterterpanyl chains from the aerobic hyperthermophilic archaeon \u003cem\u003eAeropyrum pernix\u003c/em\u003e K1. \u003cem\u003eBiochim Biophys Acta\u003c/em\u003e \u003cstrong\u003e1436\u003c/strong\u003e, 426\u0026ndash;436 (1999).\u003c/li\u003e\n\u003cli\u003eChong, P.L., Chang, A., Yu, A. \u0026amp; Mammedova, A. Vesicular and Planar Membranes of Archaea Lipids: Unusual Physical Properties and Biomedical Applications. \u003cem\u003eInt J Mol Sci\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e (2022).\u003c/li\u003e\n\u003cli\u003eGmajner, D., Ota, A., Sentjurc, M. \u0026amp; Ulrih, N.P. Stability of diether C(25,25) liposomes from the hyperthermophilic archaeon Aeropyrum pernix K1. \u003cem\u003eChem Phys Lipids\u003c/em\u003e \u003cstrong\u003e164\u003c/strong\u003e, 236\u0026ndash;245 (2011).\u003c/li\u003e\n\u003cli\u003eMcCluskie, M.J., Deschatelets, L. \u0026amp; Krishnan, L. Sulfated archaeal glycolipid archaeosomes as a safe and effective vaccine adjuvant for induction of cell-mediated immunity. \u003cem\u003eHum Vaccin Immunother\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2772\u0026ndash;2779 (2017).\u003c/li\u003e\n\u003cli\u003eJain, S., Caforio, A. \u0026amp; Driessen, A.J.M. Biosynthesis of archaeal membrane ether lipids. \u003cem\u003eFrontiers in Microbiology\u003c/em\u003e \u003cstrong\u003eVolume 5 - 2014\u003c/strong\u003e (2014).\u003c/li\u003e\n\u003cli\u003eIsobe, K. et al. Geranylgeranyl reductase and ferredoxin from Methanosarcina acetivorans are required for the synthesis of fully reduced archaeal membrane lipid in Escherichia coli cells. \u003cem\u003eJ Bacteriol\u003c/em\u003e \u003cstrong\u003e196\u003c/strong\u003e, 417\u0026ndash;423 (2014).\u003c/li\u003e\n\u003cli\u003eCaforio, A. et al. Converting \u003cem\u003eEscherichia coli\u003c/em\u003e into an archaebacterium with a hybrid heterochiral membrane. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e115\u003c/strong\u003e, 3704\u0026ndash;3709 (2018).\u003c/li\u003e\n\u003cli\u003eChen, H. et al. Engineering archaeal membrane‐spanning lipid GDGT biosynthesis in bacteria: Implications for early life membrane transformations. \u003cem\u003emLife\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 193\u0026ndash;204 (2025).\u003c/li\u003e\n\u003cli\u003eZhang, J. et al. Biosynthesis of Hybrid Neutral Lipids with Archaeal and Eukaryotic Characteristics in Engineered \u003cem\u003eSaccharomyces cerevisiae\u003c/em\u003e. \u003cem\u003eAngew Chem Int Ed Engl\u003c/em\u003e \u003cstrong\u003e62\u003c/strong\u003e, e202214344 (2023).\u003c/li\u003e\n\u003cli\u003eHoekzema, M., Jiang, J. \u0026amp; Driessen, A.J.M. Optimizing Archaeal Lipid Biosynthesis in \u003cem\u003eEscherichia coli\u003c/em\u003e. \u003cem\u003eACS Synth Biol\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2470\u0026ndash;2479 (2024).\u003c/li\u003e\n\u003cli\u003eReider Apel, A. et al. A Cas9-based toolkit to program gene expression in Saccharomyces cerevisiae. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 496\u0026ndash;508 (2017).\u003c/li\u003e\n\u003cli\u003eMorii, H., Ogawa, M., Fukuda, K. \u0026amp; Taniguchi, H. Ubiquitous distribution of phosphatidylinositol phosphate synthase and archaetidylinositol phosphate synthase in Bacteria and Archaea, which contain inositol phospholipid. \u003cem\u003eBiochem Biophys Res Commun\u003c/em\u003e \u003cstrong\u003e443\u003c/strong\u003e, 86\u0026ndash;90 (2014).\u003c/li\u003e\n\u003cli\u003eHirsch, J.P. \u0026amp; Henry, S.A. Expression of the Saccharomyces cerevisiae inositol-1-phosphate synthase (INO1) gene is regulated by factors that affect phospholipid synthesis. \u003cem\u003eMol Cell Biol\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 3320\u0026ndash;3328 (1986).\u003c/li\u003e\n\u003cli\u003eSlaughter, J.C. \u0026amp; Minabe, M. Fatty acid-containing lipids of the yeast Saccharomyces cerevisiae during post-fermentation decline in viability. \u003cem\u003eJ Sci Food Agric\u003c/em\u003e \u003cstrong\u003e65\u003c/strong\u003e, 497\u0026ndash;501 (1994).\u003c/li\u003e\n\u003cli\u003eLamosa, P. et al. Occurrence of 1-glyceryl-1-myo-inosityl phosphate in hyperthermophiles. \u003cem\u003eAppl Environ Microbiol\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e, 6169\u0026ndash;6173 (2006).\u003c/li\u003e\n\u003cli\u003eRuiz, J. et al. Membrane plasticity induced by myo-inositol derived archaeal lipids: chemical synthesis and biophysical characterization. \u003cem\u003ePhys Chem Chem Phys\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 16273\u0026ndash;16287 (2023).\u003c/li\u003e\n\u003cli\u003eJensen, S.M., Brandl, M., Treusch, A.H. \u0026amp; Ejsing, C.S. Structural characterization of ether lipids from the archaeon \u003cem\u003eSulfolobus islandicus\u003c/em\u003e by high-resolution shotgun lipidomics. \u003cem\u003eJ Mass Spectrom\u003c/em\u003e \u003cstrong\u003e50\u003c/strong\u003e, 476\u0026ndash;487 (2015).\u003c/li\u003e\n\u003cli\u003eXia, J.-M. \u0026amp; Yuan, Y.-J. Comparative Lipidomics of Four Strains of \u003cem\u003eSaccharomyces cerevisiae\u003c/em\u003e Reveals Different Responses to Furfural, Phenol, and Acetic Acid. \u003cem\u003eJ Sci Food Agric\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 99\u0026ndash;108 (2009).\u003c/li\u003e\n\u003cli\u003eGaspar, M.L., Hofbauer, H.F., Kohlwein, S.D. \u0026amp; Henry, S.A. Coordination of storage lipid synthesis and membrane biogenesis: evidence for cross-talk between triacylglycerol metabolism and phosphatidylinositol synthesis. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e286\u003c/strong\u003e, 1696\u0026ndash;1708 (2011).\u003c/li\u003e\n\u003cli\u003eYoshida, R., Motoyama, K., Ito, T. \u0026amp; Hemmi, H. Effects of producing high levels of hyperthermophile-specific C25,C25-archaeal membrane lipids in \u003cem\u003eEscherichia coli\u003c/em\u003e. \u003cem\u003eBiochem Biophys Res Commun\u003c/em\u003e \u003cstrong\u003e729\u003c/strong\u003e, 150349 (2024).\u003c/li\u003e\n\u003cli\u003ePalyzov\u0026aacute;, A., Guschina, I.A. \u0026amp; Řezanka, T. Chiral analysis of glycerol phosphates- can bacteria biosynthesize heterochiral phospholipid membranes? \u003cem\u003eJ Chromatogr A\u003c/em\u003e \u003cstrong\u003e1676\u003c/strong\u003e (2022).\u003c/li\u003e\n\u003cli\u003ePalyzova, A. et al. Stereochemistry of phosphatidylglycerols from thermotolerant bacteria isolated thermal springs. \u003cem\u003eJ Chromatogr A\u003c/em\u003e \u003cstrong\u003e1739\u003c/strong\u003e, 465517 (2025).\u003c/li\u003e\n\u003cli\u003eSanthosh, P.B. \u0026amp; Genova, J. Archaeosomes: New Generation of Liposomes Based on Archaeal Lipids for Drug Delivery and Biomedical Applications. \u003cem\u003eACS Omega\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1\u0026ndash;9 (2023).\u003c/li\u003e\n\u003cli\u003eChi, Z., Kohlwein, S.D. \u0026amp; Paltauf, F. Role of phosphatidylinositol (PI) in ethanol production and ethanol tolerance by a high ethanol producing yeast. \u003cem\u003eJ Ind Microbiol Biotechnol\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 58\u0026ndash;63 (1999).\u003c/li\u003e\n\u003cli\u003eHenderson, C.M., Zeno, W.F., Lerno, L.A., Longo, M.L. \u0026amp; Block, D.E. Fermentation temperature modulates phosphatidylethanolamine and phosphatidylinositol levels in the cell membrane of Saccharomyces cerevisiae. \u003cem\u003eAppl Environ Microbiol\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 5345\u0026ndash;5356 (2013).\u003c/li\u003e\n\u003cli\u003eRao, A. \u0026amp; Driessen, A.J.M. Unraveling the multiplicity of geranylgeranyl reductases in Archaea: potential roles in saturation of terpenoids. \u003cem\u003eExtremophiles\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 14 (2024).\u003c/li\u003e\n\u003cli\u003eEme, L. et al. Inference and reconstruction of the heimdallarchaeial ancestry of eukaryotes. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e618\u003c/strong\u003e, 992\u0026ndash;999 (2023).\u003c/li\u003e\n\u003cli\u003eLiu, Y. et al. Expanded diversity of Asgard archaea and their relationships with eukaryotes. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e593\u003c/strong\u003e, 553\u0026ndash;+ (2021).\u003c/li\u003e\n\u003cli\u003eLee, M.E., DeLoache, W.C., Cervantes, B. \u0026amp; Dueber, J.E. A Highly Characterized Yeast Toolkit for Modular, Multipart Assembly. \u003cem\u003eACS Synth Biol\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 975\u0026ndash;986 (2015).\u003c/li\u003e\n\u003cli\u003eGietz, R.D. \u0026amp; Schiestl, R.H. Large-scale high-efficiency yeast transformation using the LiAc/SS carrier DNA/PEG method. \u003cem\u003eNat Protoc\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 38\u0026ndash;41 (2007).\u003c/li\u003e\n\u003cli\u003eBligh, E.G. \u0026amp; Dyer, W.J. A rapid method of total lipid extraction and purification. \u003cem\u003eCan J Biochem Physiol\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 911\u0026ndash;917 (1959).\u003c/li\u003e\n\u003cli\u003eRostron, K.A. \u0026amp; Lawrence, C.L. in Histochemistry of Single Molecules. (eds. C. Pellicciari \u0026amp; M. Biggiogera) 219\u0026ndash;229 (Springer New York, New York, NY; 2017).\u003c/li\u003e\n\u003cli\u003eDreher, Y., Jahnke, K., Bobkova, E., Spatz, J.P. \u0026amp; G\u0026ouml;pfrich, K. Division and Regrowth of Phase‐Separated Giant Unilamellar Vesicles**. \u003cem\u003eAngew Chem Int Ed\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 10661\u0026ndash;10669 (2021).\u003c/li\u003e\n\u003cli\u003eDutta, S., Watson, B., Mattoo, S. \u0026amp; Rochet, J.-C. Calcein Release Assay to Measure Membrane Permeabilization by Recombinant Alpha-Synuclein. \u003cem\u003eBio-Protocol\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e3690 (2020).\u003c/li\u003e\n\u003cli\u003eZallot, R., Oberg, N. \u0026amp; Gerlt, J.A. The EFI Web Resource for Genomic Enzymology Tools: Leveraging Protein, Genome, and Metagenome Databases to Discover Novel Enzymes and Metabolic Pathways. \u003cem\u003eBiochemistry\u003c/em\u003e \u003cstrong\u003e58\u003c/strong\u003e, 4169\u0026ndash;4182 (2019).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8072063/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8072063/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Despite phospholipid backbone divergence between Archaea (ether-isoprenoid/glycerol-1-phosphate membranes) and Bacteria/Eukarya (ester-fatty acid/glycerol-3-phosphate membranes), known as the ‘lipid divide’, inositol headgroups are conserved across all domains. Inositol lipids from thermophilic archaea form exceptionally stable liposomes but are difficult to obtain at scale. Here, we engineered eukaryotic Saccharomyces cerevisiae for heterologous biosynthesis of unsaturated archaetidylinositol, revealing unexpected promiscuity of endogenous yeast enzymes toward archaeal substrates. Metabolic engineering achieved production levels reaching 6.5% of total cellular lipids, enabling structural, physiological, and biophysical characterization. Production triggered pleiotropic lipid metabolism changes and enhanced thermotolerance, with engineered yeast retaining 44% viability after 30 minutes at 50°C, a condition baseline strains cannot survive. Purified archaeal inositol lipid formed thermostable liposomes maintaining integrity across 25-95°C. This yeast platform will enable access to versatile archaetidylinositol derivatives with tailored bioactivity and material properties for both evolutionary studies and biotechnology applications.","manuscriptTitle":"Production of Archaeal Inositol Phospholipids in Engineered Saccharomyces cerevisiae","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-11 08:05:37","doi":"10.21203/rs.3.rs-8072063/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"65f847a2-00b4-45a2-b552-43f48fa39e80","owner":[],"postedDate":"November 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":57713138,"name":"Biological sciences/Biochemistry/Lipids/Membrane lipids"},{"id":57713139,"name":"Biological sciences/Biophysics/Membrane structure and assembly"}],"tags":[],"updatedAt":"2025-11-21T17:07:36+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-11 08:05:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8072063","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8072063","identity":"rs-8072063","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}