Resin-embedded FISH Enables High-resolution Spatial Analysis of Gene Expression in Developing Rice Seeds

Resin-embedded FISH Enables High-resolution Spatial Analysis of Gene Expression in Developing Rice Seeds | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Resin-embedded FISH Enables High-resolution Spatial Analysis of Gene Expression in Developing Rice Seeds Ting-Ting Yang, Yi-Ping Ho, Swee-Suak Ko This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9466150/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background High-resolution spatial analysis of gene expression is essential for understanding developmental regulation in plant reproductive tissues. However, conventional paraffin embedding and chromogenic in situ hybridization (ISH) are often distorted by poor tissue preservation, high background, and limited multiplexing capacity, particularly in starch-rich cereal tissues. Results In this study, we established and evaluated an optimized workflow that combined acrylic resin embedding with fluorescence in situ hybridization (FISH) for precise morphological, histochemical, and molecular analyses of rice ( Oryza sativa ) reproductive tissues. Resin-embedded sections preserved cellular architecture during prolonged ISH procedures and enabled high-resolution visualization of developing seed structures. Histochemical staining revealed dynamic redistribution of carbohydrates and proteins from maternal tissues to the endosperm during seed maturation. Both chromogenic ISH and FISH reliably detected the spatial expression patterns of key seed development genes, including Rice Prolamin Box Binding Factor ( RPBF ) and Granule-bound starch synthase I ( GBSS1 ). Notably, FISH provided superior signal clarity and reduced background compared with chromogenic ISH, particularly in reactive oxygen species–rich tissues such as rice anthers. Furthermore, dual-color FISH enabled simultaneous detection of RPBF and GBSSI transcripts, revealing coordinated co-expression within the developing endosperm at cellular resolution. Conclusion This work demonstrates that resin embedding combined with FISH is a robust and versatile platform for spatial gene expression analysis in cereal reproductive tissues and offers significant advantages for developmental and functional genomics studies. Endosperm differentiation Fluorescence in situ hybridization (FISH) Resin embedding Rice seed development Spatial gene expression Dual-color FISH Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Background RNA in situ hybridization (RNA ISH) is a robust molecular technique that enables the detection and localization of specific RNA transcripts directly within fixed cells and tissue sections. Unlike RT-qPCR, which reports only bulk or averaged transcript levels, RNA ISH preserves the tissue and cellular architecture, enabling spatial visualization of transcripts at cellular and subcellular resolution. This capability is particularly valuable for studying heterogeneous tissues, developmental patterns, and cell-type-specific gene regulation (Levsky and Singer 2003 ; Miya et al. 2024 ). RNA ISH exhibits high specificity, as it relies on sequence-specific probes that can discriminate among closely related transcripts or splice variants (Femino et al. 1998 ). It also offers high sensitivity, enabling the detection of low-abundance transcripts. Single-molecule RNA ISH techniques enable visualization of individual RNA molecules as discrete punctate signals, facilitating quantitative analysis at the single-cell level (Raj et al. 2008 ; Zenklusen and Singer 2010 ). Dual-color RNA ISH further extends this capability by simultaneously detecting two gene transcripts within the same tissue section, allowing direct comparison of co-expression or cell-type-specific gene expression while reducing section-to-section variation (Levsky and Singer 2003 ). Additionally, RNA ISH can be combined with immunohistochemistry or immunofluorescence to visualize RNA and protein targets in the same tissue section, enabling integrated analysis of transcriptional and protein expression patterns (Femino et al. 1998 ). Overall, RNA ISH offers higher spatial resolution, morphological context, and specificity compared with RT-qPCR, making it a powerful tool for dissecting gene expression patterns in complex tissues. Despite its advantages, RNA ISH has several limitations. One major constraint is that RNA ISH protocols require careful optimization of tissue preparation, probe hybridization, and signal amplification, often involving specialized reagents and imaging systems (Choi et al. 2010 ). In addition, RNA ISH is generally low-throughput compared with next-generation sequencing–based methods such as RNA sequencing or spatial transcriptomics. Traditional RNA ISH assays typically target a limited number of genes per experiment, making them less suitable for unbiased transcriptome-wide analysis (Ståhl et al. 2016 ). Although multiplex RNA ISH approaches have expanded detection capacity, they remain constrained by probe design complexity and imaging limitations. Quantitative analysis in RNA ISH can also be challenging. Signal intensity and transcript counts may be influenced by probe efficiency, tissue permeability, and imaging parameters, which can affect reproducibility across experiments and laboratories (Choi et al. 2010 ). High-quality tissue sections are essential for accurate spatial gene expression profiling. However, certain plant tissues, such as rice seeds that are rich in starchy endosperm, are particularly susceptible to structural damage during RNA in situ hybridization. Prolonged hybridization and washing procedures lasting 2–3 days often lead to paraffin tissue swelling or separation of the endosperm from surrounding tissues. The high starch content and relatively loose cellular organization of the endosperm reduce its resistance to mechanical and osmotic stress caused by repeated buffer exchanges, elevated hybridization temperatures, and protease treatments. Together, these factors compromise tissue integrity, adversely affecting probe penetration, signal uniformity, and overall experimental reproducibility. Resin (plastic) embedding allows semithin sections (~ 1 µm) that provide superior preservation of morphology, cell walls, and intracellular structures compared with paraffin wax sections, which are generally thicker and less resolving. Using this approach, C 4 photosynthetic maize tissues exhibited clear starch accumulation within chloroplasts of bundle sheath cells (Edwards et al. 2001 ). The enhanced structural stability provided by resin embedding is particularly advantageous for fragile, starch-rich tissues, such as endosperm, and offers more consistent section quality and improved morphological interpretation compared with conventional paraffin-based methods. Rice endosperm development is characterized by the coordinated accumulation of starch and storage proteins, a process tightly regulated at the transcriptional level. Among the key transcription factors involved, Rice Prolamin Box Binding Factor (RPBF) plays a central role in controlling seed storage protein gene expression. Previous studies have demonstrated that RPBF is a key transcription factor involved in endosperm development. Loss of RPBF function results in chalky endosperm, reduced seed size, and defective grain filling (Kawakatsu et al. 2009 ). RPBF binds to the conserved prolamin box present in the promoters of many endosperm-expressed storage protein genes (Wu et al. 2023 ; Yamamoto et al. 2006 ). In contrast, genes involved in starch biosynthesis, such as Granule-Bound Starch Synthase I ( GBSSI/Wx1 ), are regulated by a distinct but overlapping transcriptional network and do not contain canonical P-box motifs in their promoters. Although RPBF does not directly regulate GBSSI , both storage protein and starch biosynthetic pathways are developmentally synchronized, suggesting that RPBF may indirectly influence carbon–nitrogen allocation during endosperm maturation (Ning et al. 2023 ). GBSSI is specifically localized to starch granules in endosperm amyloplasts and is responsible for the synthesis of amylose, a key determinant of grain cooking and eating quality. Natural variation or mutation in Wx1 alters amylose content, leading to changes in grain texture, including waxy or low-amylose phenotypes, and can influence endosperm structure and grain filling efficiency (Wang et al. 2015 ). Despite advances in seed biology, high-resolution spatial analysis of gene expression in rice reproductive tissues remains limited by poor tissue preservation and low signal clarity during in situ hybridization. In particular, the coordinated expression of key genes, such as RPBF and GBSSI , during seed development remains poorly understood at the cellular level. This study aimed to establish an optimized workflow combining resin embedding with fluorescence in situ hybridization (FISH) to improve tissue integrity and detection sensitivity. Using this approach, we sought to characterize morphological features, histochemical changes, and spatial gene expression patterns, providing new insights into the regulation of rice seed development. Results Paraffin-embedding Tissue Reduced Image Quality after ISH Classic paraffin-embedded sections maintained good image quality following short-term staining, such as hematoxylin staining for 50 seconds, which clearly visualized starch distribution and tissue morphology of caryopsis coat (CC), dorsal vascular bundle (DVB), nucellar epidermis (NE), aleurone layer (AL), and starchy endosperm (SE) (Fig. 1A) . However, when subjected to the multi-step in situ hybridization (ISH) procedure lasting approximately three days, the paraffin sections exhibited significant deterioration of tissue integrity. The prolonged exposure to high temperature, enzymatic digestion, and repeated washing steps caused tissue deformation and partial detachment from the slides (Fig. 1B) . As a result, the morphological structure became distorted, and the resolution of cellular features was markedly reduced, hindering accurate localization and interpretation of gene expression signals. Clearly, conventional paraffin embedding is not suitable for high-quality ISH analysis of rice seed tissues. Embedding and Sectioning of LR White–embedded Rice Seeds To overcome the tissue deformation and section instability associated with paraffin-based RNA in situ hybridization, we established a resin-based ISH workflow optimized for rice seeds. Because rice seeds are substantially larger than tissues typically processed by resin embedding, several modifications to standard embedding and sectioning procedures were required. To accommodate seed size, 5-mL microcentrifuge tube caps were used as embedding molds. Following London Resin (LR) White infiltration, samples were oriented in the molds and covered with a glass coverslip to exclude oxygen during polymerization, a critical factor for efficient LR White curing. This modification enabled consistent and uniform polymerization and generated mechanically stable resin blocks. After polymerization, the LR White blocks exhibited substantial hardness, making them difficult to trim using conventional razor blades. To overcome this limitation, the resin blocks were mounted onto a trimming chuck using hot-melt adhesive. The resin block trimming was performed using an ultrasonic knife, which significantly improved trimming efficiency and precision while minimizing mechanical stress and potential damage to the embedded seed tissues. The workflow is summarized in Fig. 2 , and a detailed trimming procedure is presented in the accompanying video ( Supplemental File 1 ). Overall, this optimized embedding and sectioning strategy enabled the reproducible preparation of high-quality LR White sections of rice seeds, with improved structural integrity and section stability compared to paraffin-based methods. Resin-embedding Improves Anatomical Resolution in Rice Seed Sections Developing seeds of the rice cultivar TNG67 were collected at representative developmental stages: S2 (6 days after pollination (DAP)), S3 (9 DAP), and S4 (16 DAP). Samples were fixed, embedded in LR-white resin, transversely sectioned at a thickness of 2 μm, and stained with TBO for microscopic examination. Resin-embedded whole seeds exhibited well-preserved tissue integrity, capturing both the rigid structure of the husk and the loosely organized texture of the starchy endosperm ( Fig. S1 ). At the early developmental stages (S2), seed cross-sections revealed a prominently thickened caryopsis coat derived from maternal tissues, which enclosed a relatively small and incompletely developed endosperm. The cellular organization of the endosperm at this stage appeared less compact, and the aleurone layer (AL) was not yet clearly differentiated. As seed development progressed to the S3 stage, one AL and grain filling were observed. At the S4 stage, a marked reduction in the thickness of the caryopsis coat was observed. In contrast, the endosperm exhibited substantial expansion, occupying a larger proportion of the seed volume, accompanied by clear differentiation and thickening of the aleurone layer to two layers. The high-resolution TBO-stained resin sections allowed precise visualization of tissue boundaries and cellular architecture, providing a robust morphological framework for subsequent analyses of gene expression and storage reserve accumulation during rice seed maturation ( Fig. 3 ). Histochemical Staining of Resin-embedded Tissues Reveals Dynamic Nutrient Accumulation During Seed Development To investigate the spatial and temporal patterns of nutrient accumulation during rice seed development, transverse sections of resin-embedded seeds collected at stages S2 to S4 were subjected to histochemical staining using periodic acid–Schiff (PAS), iodine–potassium iodide, and coomassie brilliant blue (CBB). PAS staining, which detects polysaccharides, revealed distinct developmental stage–dependent distribution patterns. At the early S2 stage, strong PAS signals were predominantly detected in the maternal caryopsis coat (CC), while the filial endosperm showed relatively weak staining. By the S3 stage, PAS signals were observed at comparable levels in both the CC and endosperm, indicating a shift in polysaccharide distribution. At the S4 stage, PAS staining was largely absent from the CC and became strongly enriched in the endosperm, suggesting a developmental shift of carbohydrate accumulation from maternal tissues to filial tissues during seed maturation (Fig. 4A) . Consistent with the PAS staining results, iodine staining revealed similar patterns of starch accumulation across developmental stages. At S2, starch signals were weak and mainly associated with maternal tissues. As development progressed, starch accumulation increased markedly within the endosperm. Notably, iodine staining in the starchy endosperm shifted from red at S2 to deep blue at the later stages, indicating progressive changes in starch content and composition, with increasing amylose accumulation as endosperm cells matured ( Fig. 4B) . CBB staining was used to visualize protein accumulation within developing seeds. At the early stages, protein signals were relatively weak; however, by the S4 stage, strong CBB staining was observed, indicating substantial storage protein accumulation. At this stage, protein signals were detected in both the CC and endosperm compartments, with pronounced enrichment in the endosperm, consistent with active storage protein deposition during seed maturation ( Fig. 4C) Together, these histochemical analyses demonstrated coordinated and stage-specific redistribution of carbohydrates and proteins during rice seed development, supporting the transition from maternal tissue support to endosperm-centered nutrient storage as seeds mature. Fluorescence In Situ Hybridization (FISH) Reduces ROS-related Background Conventional chromogenic in situ hybridization (ISH) commonly uses nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) as a substrate for signal detection; however, this reaction is highly sensitive to reactive oxygen species (ROS). In rice anthers, elevated ROS levels are intrinsically associated with tapetal programmed cell death (PCD), a well-characterized and essential developmental process during microsporogenesis (Hu et al. 2011; Yi et al. 2016). As a result, chromogenic ISH frequently produces strong nonspecific background signals, including in sense-probe negative controls, which can obscure authentic hybridization patterns ( Fig. 5A, B ). Consistent with these observations, NBT staining revealed prominent superoxide accumulation in anther tissues, particularly in tapetal cells, meiocytes, and vascular bundles ( Fig. 5C ). A section stained with K-citrate buffer alone, used as a negative control, showed no detectable signal ( Fig. 5D ). In contrast, FISH largely eliminated ROS-associated background, enabling clearer and more specific detection of gene expression in rice anther tissues ( Fig. 5E, F ). In rice seed tissues, no detectable autofluorescence was observed in the Cy3 or Cy5 channels, indicating minimal background interference for probes labeled with these fluorophores. In contrast, strong autofluorescence was detected in the DAPI channel, which may interfere with the interpretation of FISH signals ( Fig. S2 ). Resin-embedded Tissues Enable Precise Spatial and Temporal Analysis of Gene Expression To investigate the spatial expression patterns of key seed development markers, we examined the seed-specific genes Rice Prolamin Box Binding Factor ( RPBF ) and Granule-bound Starch Synthase I ( GBSS1 ) using both chromogenic in situ hybridization (ISH) and fluorescence ISH (FISH) on resin-embedded rice seed tissue sections ( Fig. 6, 7 ). These two ISH technologies produced similar expression patterns. Notably, the high-quality preservation of tissue morphology in resin sections allowed clear visualization of gene expression at cellular resolution. Chromogenic in situ hybridization showed negligible ISH background with sense probes of RPBF and GBSSI ( Fig. S3 ). RPBF expression was strongly detected in maternal carpel tissues during early seed development (stages S1 and S2, asterisks), but expression declined markedly as seeds matured, with only minimal signal observed in the aleurone layer at later stages, suggesting temporal regulation of RPBF during early seed development ( Fig. 6 upper panel). GBSS1 , a key enzyme in amylose synthesis, showed a strong ISH signal in the maternal tissue but a weak signal in filial tissue at S1, the caryopsis coat of S2, and signals intensifying in the aleurone layer and endosperm at S4 developmental stages, indicating active starch biosynthesis during endosperm filling ( Fig. 6, bottom panel). Fluorescence In Situ Hybridization (FISH) Reveals Clear and Reproducible Gene Expression Patterns Resin-embedded tissue sections identical to those used for chromogenic ISH were employed to examine RPBF and GBSSI expression by fluorescence in situ hybridization (FISH). FISH produced distinct, well-defined hybridization signals with low background, allowing clear visualization of gene expression patterns (Fig. 7) . Importantly, the spatial distribution of ISH signals detected by FISH was consistent with that obtained using chromogenic ISH, demonstrating high reproducibility between the two approaches. These results confirm that FISH reliably detects gene expression in resin-embedded rice tissues while providing improved signal clarity and imaging resolution. Dual-color FISH Reveals Co-expression Patterns of RPBF and GBSSI To investigate the spatial co-expression patterns of two genes within the same tissue section, dual-color fluorescence in situ hybridization (DISH) was performed using gene-specific probes targeting RPBF and GBSSI . The dual-color FISH analysis enabled simultaneous visualization of the transcripts and provided high spatial resolution of their expression domains. The results showed that RPBF and GBSSI transcripts were detected in overlapping regions of the developing rice endosperm, indicating co-expression within the same cell types. In cells exhibiting overlapping signals, the two transcripts displayed distinct yet coordinated spatial distributions. Notably, the DISH signals were clearly distinguishable with minimal background interference, demonstrating the robustness of this approach for resolving gene co-expression patterns in complex tissues. These observations suggest that RPBF and GBSSI are co-expressed in the same developmental context, supporting a potential regulatory relationship during endosperm development. Discussion A major challenge in studying rice reproductive tissues is preserving the structural integrity of seeds during multi-stage preparation protocols such as in situ hybridization (ISH). Rice grains are relatively large and anatomically complex, containing both rigid maternal tissues and highly hydrated filial tissues rich in starch and storage proteins. These features make them particularly susceptible to deformation during conventional paraffin embedding and sectioning. In this study, we addressed these limitations by optimizing an acrylic resin–based embedding workflow using LR White, which has been widely applied in high-resolution histological studies of plant systems (Newman and Hobot 1999 ). Several critical modifications were introduced to adapt standard protocols to better suit rice seeds. First, customized embedding molds were employed to accommodate whole developing grains and maintain proper orientation. Second, oxygen exclusion during polymerization was implemented to ensure consistent curing of LR White resin, as oxygen is known to inhibit free-radical polymerization and compromise block quality. Third, ultrasonic trimming proved efficient for sectioning hardened resin blocks. Compared with conventional razor blades, ultrasonic knives significantly improved trimming precision while minimizing mechanical stress and preventing tissue fracture. These technical refinements collectively enabled the reproducible preparation of high-quality sections with minimal distortion (Fig. 2 , 3 ). The resulting resin-embedded sections exhibited excellent preservation of cellular architecture across developmental stages. Toluidine blue staining clearly resolved key anatomical features, including the husk, caryopsis coat, nucellar epidermis, aleurone layer, and starchy endosperm ( Fig. S1 , Fig. 3 ). The ability to simultaneously preserve both rigid and soft tissues highlights the superior infiltration and mechanical support provided by acrylic resins, particularly in plant tissues with dense cell walls and high carbohydrate content (Newman and Hobot 1999 ). This level of structural preservation is essential for accurate interpretation of tissue-specific processes during seed development. In addition to morphological analyses, resin sections proved highly compatible with histochemical staining. Periodic acid–Schiff (PAS) and iodine staining revealed dynamic changes in carbohydrate distribution during seed maturation (Fig. 4 ). Specifically, polysaccharides were initially enriched in maternal tissues such as the caryopsis coat and subsequently redistributed to the endosperm, where starch accumulation increased progressively. These observations are consistent with established models of cereal grain filling, in which assimilates are transported from maternal tissues to the developing endosperm (Olsen 2004 ; Sabelli and Larkins 2009 ). Similarly, the accumulation of storage proteins at later developmental stages reflects the transition toward endosperm maturation and reserve deposition. Together, these results demonstrate that resin embedding supports integrated structural and biochemical analyses, providing a comprehensive framework for studying seed development. Precise spatial analysis of gene expression is critical for understanding the regulatory networks governing seed development. However, conventional paraffin-based ISH methods often fail to maintain tissue integrity during prolonged hybridization and washing steps, particularly in starch-rich tissues such as rice endosperm. In this study, we demonstrated that resin-embedding combined with fluorescence in situ hybridization (FISH) provides a robust and reproducible platform for cellular-resolution gene expression analysis. Paraffin sections showed substantial deformation, distortion, and detachment from slides following ISH procedures (Fig. 1 ), consistent with previous reports highlighting the fragility of paraffin-embedded plant tissues during multi-step protocols (Jackson 1991 ). These artifacts significantly compromise the accuracy of signal localization and limit the interpretation of gene expression patterns. In contrast, resin-embedded sections maintained structural integrity throughout the ISH workflow, enabling precise mapping of transcripts within defined cellular contexts. An additional advantage of the resin-based approach is improved signal clarity in FISH analyses. Background fluorescence was observed in the DAPI channel, likely due to intrinsic autofluorescence from cell wall components or storage materials in seeds; however, signals in longer-wavelength channels such as Cy3 and Cy5 were clear and reproducible ( Fig. S2) . This observation is consistent with previous studies showing that plant tissues often exhibit strong autofluorescence in the blue spectrum, necessitating the use of red or far-red fluorophores for optimal signal detection (Duncan et al. 2016 ; Ruzin 2024 ). Therefore, careful selection of fluorophores is essential to minimize background and maximize sensitivity in plant FISH experiments. A key finding of this study is the reduced susceptibility of FISH to reactive oxygen species (ROS)-related background compared with chromogenic ISH. In rice anthers, high ROS levels are associated with tapetal programmed cell death, a critical developmental process (Hu et al. 2011 ; Yi et al. 2016 ). Under these conditions, chromogenic detection using NBT/BCIP substrates produced strong nonspecific signals that overlapped with endogenous ROS distribution, even in negative controls (Fig. 5 ). This limitation reflects the inherent sensitivity of enzymatic colorimetric reactions to oxidative environments (Jackson, 1991 ). In contrast, FISH largely eliminated such background interference, enabling more specific and reliable detection of gene expression. These results underscore the advantages of fluorescence-based detection, particularly for tissues undergoing oxidative stress or active metabolic changes. Both chromogenic ISH and FISH performed on resin sections yielded consistent spatial and temporal expression patterns for key seed developmental genes. RPBF transcripts were predominantly detected in maternal tissues during early stages and declined as seeds matured, whereas GBSSI expression increased progressively in the aleurone layer and starchy endosperm (Figs. 6 , 7 ) . These patterns are consistent with the known roles of RPBF as a transcriptional regulator of storage protein genes and GBSSI as a key enzyme in amylose biosynthesis (Sano 1984 ; Kawakatsu et al. 2009 ). The ability to recapitulate these established expression dynamics validates the reliability and sensitivity of the resin-based ISH platform. Importantly, superior preservation of tissue morphology enabled precise localization of transcripts at cellular resolution. Distinct cell layers and tissue compartments could be clearly resolved, allowing accurate assignment of gene expression domains. This level of detail is difficult to achieve with paraffin-based methods, particularly in cereal seeds where structural distortion is common. Thus, resin embedding represents a critical methodological advance for spatial transcript analysis in plant tissues. Multiplex FISH enhances analytical power by enabling simultaneous detection of multiple transcripts within the same tissue. Dual-color FISH revealed overlapping expression of RPBF and GBSSI in developing endosperm cells (Fig. 8 ), indicating coordinated gene activity during seed development. RPBF expression was prominent in maternal carpel tissues at the early stages (S1–S2) but declined during maturation, suggesting a role in early endosperm differentiation (Yamamoto et al. 2006 ). In contrast, GBSSI expression increased progressively, with strong accumulation in the aleurone layer at later stages (Fig. 6 , 7 ), consistent with its role in amylose biosynthesis during grain filling (Liu et al. 2014 ). Notably, co-expression of both genes was observed in the aleurone layer at 3 DAP (Fig. 8 F), suggesting a potential interaction during early grain development. Although this overlap may reflect coordinated regulation or complementary functions in starch metabolism, its biological significance remains unclear and warrants further functional and regulatory studies. The ability to visualize such interactions at cellular resolution demonstrates the value of resin-based multiplex FISH for dissecting complex gene regulatory networks in rice seed development. Beyond targeted gene analysis, the methods developed in this study have important implications for validating high-throughput transcriptomic datasets. RNA sequencing (RNA-seq) has revolutionized plant biology by providing comprehensive gene expression profiles across developmental stages and conditions. However, RNA-seq lacks spatial resolution, making it difficult to assign gene expression to specific cell types or tissues. Integrating transcriptomic data with spatial techniques such as ISH and FISH is therefore essential for linking gene expression patterns to developmental processes (Wu et al. 2020 ). The resin-based platform established here provides a reliable and versatile tool for such validation. Enabling high-resolution localization of transcripts within intact tissue contexts allows researchers to confirm and refine gene expression patterns identified through RNA-seq. Moreover, the compatibility of resin sections with multiplex FISH opens up new possibilities for analyzing gene co-expression networks and cell-type–specific regulatory mechanisms. This approach is particularly valuable for complex tissues such as developing seeds, where multiple cell types interact dynamically during development. Conclusion In summary, this study establishes resin embedding combined with FISH as a powerful and versatile approach for spatial transcript analysis in rice reproductive tissues. By overcoming the structural limitations of paraffin embedding and reducing the background interference associated with chromogenic detection, this method enables high-resolution, reliable visualization of gene expression. The ability to perform multiplex analyses further enhances its utility for investigating coordinated gene regulation. Collectively, these advances provide a robust framework for studying the molecular mechanisms underlying rice seed development and offer broad applications for functional genomics in cereal crops. Materials and Methods Plant Materials Rice ( Oryza sativa L.) cultivar TNG67 plants were grown in soil under controlled conditions in the genetically modified organism (GMO) greenhouse at the Biotechnology Center, Academia Sinica (Tainan, Taiwan). Flowers were tagged at anthesis, and developing seeds were collected at 6 days after pollination (DAP; stage S2), 6 DAP (S3), and 16 DAP (S4). Harvested seeds were immediately fixed and processed for histological analysis, in situ hybridization, and fluorescence in situ hybridization as described below. Paraffin Tissue Sectioning Paraffin sectioning was performed as previously described (Chen et al., 2016). Briefly, developing seeds were fixed, dehydrated through a graded ethanol series, cleared, and embedded in paraffin. Sections of appropriate thickness were cut using a rotary microtome and mounted onto SuperFrost PLUS microscope slides (Cat. S7441, Matsunami, Japan). LR White Resin Embedding and Sectioning of Rice Seeds Rice ( Oryza sativa ) cv. TNG67 seeds were harvested at defined developmental stages ( i.e. , 3, 6, 9, and 16 days after pollination, DAP). Samples were fixed in freshly prepared 4% (w/v) paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.2, vacuum infiltrated for 20–30 min, followed by incubation at 4°C for 12–16 h. After fixation, tissues were washed three times (10 min each) in 0.1 M PB at 4°C. [ Note: For RNA-based applications ( i.e. , ISH or FISH), all solutions and tools were prepared using DEPC-treated water and RNase-free conditions.] Fixed tissues were dehydrated through a graded ethanol series as follows: 30% ethanol, 30 min; 50% ethanol, 30 min; 70% ethanol, 30 min; 85% ethanol, 30 min; 95% ethanol, 30 min; 100% ethanol, three changes, 30 min each. [Complete dehydration was critical to ensure efficient resin infiltration.] Then, dehydrated samples were infiltrated with LR White resin (AGR1281; Agar Scientific, Wetzlar, Germany) using a stepwise ethanol–resin series at 4°C with gentle agitation: Ethanol:LR White (2:1, v/v), 2 h; Ethanol:LR White (1:1, v/v), 2 h; Ethanol:LR White (1:2, v/v), 2 h; and 100% LR White, overnight. The resin was replaced with fresh 100% LR White and incubated for an additional 6–8 h to ensure complete infiltration. The sample was transferred into a tube cap of 5 mL Eppendorf tube, filled with fresh LR White resin, and oriented appropriately for sectioning. Polymerization was carried out at 60°C for 24 h in a dry oven. Polymerized blocks were allowed to cool to room temperature before trimming. In this study, resin blocks were trimmed with a Sonic Saber knife (Phrozen, Taiwan) and sectioned on a rotary microtome (Leica RM2255, Leica, Buffalo Grove, IL, USA) equipped with a diamond knife. Tissues were sectioned to a thickness of 2 µm. Sections were collected onto SuperFrost PLUS microscope slides (Cat. S7441, Matsunami, Japan). Slides were dried on a 42°C slide warmer for 1–2 h or overnight at room temperature to ensure firm adhesion. Histological Staining Histological staining was performed using standard procedures. Hematoxylin (ScyTek Laboratories, USA) staining was used to examine general tissue morphology. Toluidine blue O (TBO; Sigma-Aldrich, USA) staining was used to visualize cellular structure. 1% iodine–potassium iodide (I₂/KI) staining was used to detect starch accumulation, and Coomassie Brilliant Blue (Bio-Rad, USA) staining was performed to visualize storage proteins. All staining procedures were conducted according to the manufacturers’ instructions or established protocols. Stained or hybridized sections were examined using a bright-field microscope (Zeiss, Germany) for comparative analyses across developmental stages. In Situ hybridization (ISH) probe synthesis Rice genes RPBF (Os02g0252400), GBSSI (Os06g0133000), and RbohF (Os08g0453700) were amplified and cloned for probe synthesis. Gene-specific primer sets were as follows: RPBF , 5′-AGCGGCGGCGCATTATCCC-3′ and 5′-AAGTAGCGCGGCTGCGCCATG-3′; GBSSI , 5′-TCTCCGAGAGGTTCAGGTCATCC-3′ and 5′-CCTGGCGATGCCGGAGATG-3′, and RbohF , 5′-GTCCACCAAGGTTGCTTGTAGA-3′ and 5′-CGTAGTTCCTGAAATCCTGCG-3′. The amplified gene-specific coding sequences were ligated into the pGEM-T easy vector (Promega, USA) following the manufacturer’s protocol. Digoxigenin (DIG)-labeled sense and antisense RNA probes were synthesized using the SP6/T7 DIG RNA Labeling Kit (cat. no. 11175025910, Roche) as previously described (Chen et al. 2016). Chromogenic In Situ hybridization The chromogenic in situ hybridization (ISH) was performed according to the protocol described by Chen et al. (2016) with minor modifications. Hybridization temperatures were optimized for each probe and set at 50°C for RPBF , 55°C for GBSSI , and 60°C for RbohF . After hybridization and washing, DIG-labeled probes were detected using nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) substrate solution to visualize hybridization signals. RNA Fluorescence In Situ Hybridization RNA Fluorescence in situ hybridization ( RNA FISH) was performed according to previously described protocols (Ko et al. 2021) with minor modifications. Briefly, DIG-labeled RNA probes were detected using anti-DIG-POD Fab fragments (cat. no. 11207733910, Roche), followed by tyramide signal amplification (TSA) using the TSA Plus Cyanine 5 (Cy5) detection kit (cat. no. NEL745001KT, PerkinElmer). Hybridization temperatures were at 50°C for RPBF , 55°C for GBSSI , and 60°C for RbohF , respectively. Fluorescent signals were examined under a Zeiss LSM-980 confocal microscope, using Cy3 or Cy5 filters at excitation and emission wavelengths of 561 nm/597 nm or 639 nm/740 nm, respectively. Dual-color FISH The dual-color FISH (DISH) technique was used to show the co-expression of RPBF and GBSSI transcripts as described previously (Ko et al. 2021). A DIG RNA labeling kit (SP6/T7) (cat. no. 11175025910, Roche) and Fluorescein RNA Labeling Mix (cat. no. 11685619910, Roche) were used to synthesize probes according to the manufacturer’s instructions. Probes of RPBF and GBSSI were simultaneously hybridized with tissue slices overnight at 53°C. Then, anti-fluorescein isothiocyanate (FITC)-POD was conjugated with GBSSI probe overnight at room temperature, and tissues were reacted with TSA working solution containing TSA Plus Cyanine 3 (Cy3) (cat. no. NEL744001KT, PerkinElmer) (at 1:100 dilution) at room temperature for 5 min, and then washed twice. Afterward, tissues were treated with 3% hydrogen peroxide at room temperature for 10 min to inactivate the first peroxidase of anti-FITC-POD. After that, anti-DIG-POD was conjugated with the RPBF probe at 37 °C for 2 h. Tissues were then stained with TSA working solution containing Cy5 amplification reagent (at 1:300 dilution) at room temperature for 5 min, and washed twice. Fluorescence signals were observed under a Zeiss LSM-980 confocal microscope, using Cy3 or Cy5 filters at excitation and emission wavelengths of 561 nm/597 nm or 639 nm/740 nm, respectively. Abbreviations AL: aleurone layer; CBB: coomassie brilliant blue; CC: caryopsis coat; DAF: days after flowering; En: endosperm; FISH: fluorescent in situ hybridization; GBSS1 : Granule-bound Starch Synthase I ; ISH: in situ hybridization; NE, nucellar epidermis; PAS: periodic acid–Schiff; RbohF : Respiratory Burst Oxidase Homolog F ; RPBF : Rice Prolamin Box Binding Factor ; TBO: toluidine blue O Declarations Competing interests The authors declare that they have no competing interests. Funding This work was supported by the Academia Sinica Biotechnology Center in Southern Taiwan. Author Contribution S.-S. Ko conceived and designed the research. T.-T. Yang and Y.-P. Ho performed the in situ hybridization and tissue sectioning. S.-S. Ko wrote the manuscript. All authors read and approved the final manuscript. Acknowledgement We are grateful to Dr. Wann-Neng Jane for guidance with resin sectioning. We thank the DNA Sequencing Core Facility of the Institute of Biomedical Sciences, Academia Sinica for providing DNA sequencing services. We also thank Ms. Miranda Loney for assistance with English editing. References Chen TK, Yang HT, Fang SC, Lien YC, Yang TT, Ko SS (2016) Hybrid-Cut: An improved sectioning method for recalcitrant plant tissue samples. J Vis Exp 117:e54754. doi:10.3791/54754 Choi HM, Chang JY, Trinh LA, Padilla JE, Fraser SE, Pierce NA (2010) Programmable in situ amplification for multiplexed imaging of mRNA expression. Nature biotechnology 28 (11):1208–1212 Duncan S, Olsson TS, Hartley M, Dean C, Rosa S (2016) A method for detecting single mRNA molecules in Arabidopsis thaliana. Plant methods 12 (1):13 Edwards GE, Franceschi VR, Ku MS, Voznesenskaya EV, Pyankov VI, Andreo CS (2001) Compartmentation of photosynthesis in cells and tissues of C4 plants. Journal of Experimental Botany 52 (356):577–590 Femino AM, Fay FS, Fogarty K, Singer RH (1998) Visualization of single RNA transcripts in situ. Science 280 (5363):585–590 Hu L, Liang W, Yin C, Cui X, Zong J, Wang X, Hu J, Zhang D (2011) Rice MADS3 regulates ROS homeostasis during late anther development. Plant Cell 23 (2):515–533. doi:10.1105/tpc.110.074369 Jackson D (1991) In situ hybridisation in plants. In: Molecular Plant Pathology: A Practical Approach. Oxford University Press, Oxford, pp 163–174 Kawakatsu T, Yamamoto MP, Touno SM, Yasuda H, Takaiwa F (2009) Compensation and interaction between RISBZ1 and RPBF during grain filling in rice. Plant J 59 (6):908–920. doi:10.1111/j.1365-313X.2009.03925.x Ko SS, Li MJ, Ho YC, Yu CP, Yang TT, Lin YJ, Hsing HC, Chen TK, Jhong CM, Li WH, Sun-Ben Ku M (2021) Rice transcription factor GAMYB modulates bHLH142 and is homeostatically regulated by TDR during anther tapetal and pollen development. J Exp Bot 72 (13):4888–4903. doi:10.1093/jxb/erab190 Levsky JM, Singer RH (2003) Fluorescence in situ hybridization: past, present and future. J Cell Sci 116 (Pt 14):2833–2838. doi:10.1242/jcs.00633 Liu D, Wang W, Cai X (2014) Modulation of amylose content by structure-based modification of OsGBSS1 activity in rice (Oryza sativa L.). Plant Biotechnol J 12 (9):1297–1307. doi:10.1111/pbi.12228 Miya M, Hibara K-i, Itoh J-i (2024) Spatial Analysis of gene expression by In situ hybridization. In: Rice: Methods and Protocols. Springer, pp 49–59 Newman GR, Hobot JA (1999) Resins for combined light and electron microscopy: a half century of development. The Histochemical Journal 31 (8):495–505 Ning L, Wang Y, Shi X, Zhou L, Ge M, Liang S, Wu Y, Zhang T, Zhao H (2023) Nitrogen-dependent binding of the transcription factor PBF1 contributes to the balance of protein and carbohydrate storage in maize endosperm. The Plant Cell 35 (1):409–434 Olsen O-A (2004) Nuclear endosperm development in cereals and Arabidopsis thaliana. The Plant Cell 16 (suppl_1):S214–S227 Raj A, van den Bogaard P, Rifkin SA, van Oudenaarden A, Tyagi S (2008) Imaging individual mRNA molecules using multiple singly labeled probes. Nat Methods 5 (10):877–879. doi:10.1038/nmeth.1253 Ruzin SE (2024) Techniques in light microscopy. Oxford University Press, Sabelli PA, Larkins BA (2009) The development of endosperm in grasses. Plant Physiol 149 (1):14–26. doi:10.1104/pp.108.129437 Sano Y (1984) Differential regulation of waxy gene expression in rice endosperm. Theoretical and applied genetics 68 (5):467–473 Ståhl PL, Salmén F, Vickovic S, Lundmark A, Navarro JF, Magnusson J, Giacomello S, Asp M, Westholm JO, Huss M (2016) Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science 353 (6294):78–82 Wang K, Hasjim J, Wu AC, Li E, Henry RJ, Gilbert RG (2015) Roles of GBSSI and SSIIa in determining amylose fine structure. Carbohydrate polymers 127:264–274 Wu MW, Liu J, Bai X, Chen WQ, Ren Y, Liu JL, Chen MM, Zhao H, Yao X, Zhang JD (2023) Transcription factors NAC20 and NAC26 interact with RPBF to activate albumin accumulations in rice endosperm. Plant Biotechnology Journal 21 (5):890 Wu TY, Müller M, Gruissem W, Bhullar NK (2020) Genome Wide Analysis of the Transcriptional Profiles in Different Regions of the Developing Rice Grains. Rice (N Y) 13 (1):62. doi:10.1186/s12284-020-00421-4 Yamamoto MP, Onodera Y, Touno SM, Takaiwa F (2006) Synergism between RPBF Dof and RISBZ1 bZIP activators in the regulation of rice seed expression genes. Plant physiology 141 (4):1694–1707 Yi J, Moon S, Lee YS, Zhu L, Liang W, Zhang D, Jung KH, An G (2016) Defective Tapetum Cell Death 1 (DTC1) regulates ROS levels by binding to metallothionein during tapetum degeneration. Plant Physiol 170 (3):1611–1623. doi:10.1104/pp.15.01561 Zenklusen D, Singer RH (2010) Analyzing mRNA expression using single mRNA resolution fluorescent in situ hybridization. Methods in enzymology 470:641–659. doi:10.1016/S0076-6879(10)70026-4 Additional Declarations No competing interests reported. Supplementary Files VideoTrimResinblock.mp4 Additional file 1. Video showing the trimming of a resin block using a Sonic Saber knife. SUPPFIG.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 12 May, 2026 Reviewers agreed at journal 01 May, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers invited by journal 29 Apr, 2026 Editor assigned by journal 20 Apr, 2026 Submission checks completed at journal 20 Apr, 2026 First submitted to journal 19 Apr, 2026 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-9466150","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":633179286,"identity":"4f9b7603-e2b1-4d81-8a98-b7495ca25e5d","order_by":0,"name":"Ting-Ting Yang","email":"","orcid":"","institution":"Academia Sinica Biotechnology Center in Southern Taiwan","correspondingAuthor":false,"prefix":"","firstName":"Ting-Ting","middleName":"","lastName":"Yang","suffix":""},{"id":633179289,"identity":"701cf1ac-733b-4bf3-9609-b34bdc345d2f","order_by":1,"name":"Yi-Ping Ho","email":"","orcid":"","institution":"Academia Sinica Biotechnology Center in Southern Taiwan","correspondingAuthor":false,"prefix":"","firstName":"Yi-Ping","middleName":"","lastName":"Ho","suffix":""},{"id":633179293,"identity":"92670111-0f3b-4d9c-b5b8-b37d2aa55390","order_by":2,"name":"Swee-Suak Ko","email":"data:image/png;base64,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","orcid":"","institution":"Academia Sinica Biotechnology Center in Southern Taiwan","correspondingAuthor":true,"prefix":"","firstName":"Swee-Suak","middleName":"","lastName":"Ko","suffix":""}],"badges":[],"createdAt":"2026-04-20 03:23:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9466150/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9466150/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108831999,"identity":"6d1cee87-5590-4c75-93b8-7c8aa7addc09","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1337570,"visible":true,"origin":"","legend":"\u003cp\u003eConventional paraffin embedding results in substantial loss of tissue image quality following\u003cem\u003e in situ\u003c/em\u003e hybridization. (\u003cstrong\u003eA\u003c/strong\u003e) Paraffin section before \u003cem\u003ein situ \u003c/em\u003ehybridization (ISH). (\u003cstrong\u003eB\u003c/strong\u003e) Loss of seed morphological integrity after a 3-day ISH procedure. TNG67 seeds at developmental stage S3 were embedded in paraffin and transversely sectioned to a thickness of 10 µm. Scale\u003cstrong\u003e \u003c/strong\u003ebars = 50 µm.\u003c/p\u003e","description":"","filename":"FIG1.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/3998de90dbac0ccc4e98b7e0.png"},{"id":108832001,"identity":"a72171b4-d15c-49c3-ad73-07711d4f70fb","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1585720,"visible":true,"origin":"","legend":"\u003cp\u003eA schematic overview of the workflow for embedding, trimming, and sectioning rice seeds in LR White resin. (\u003cstrong\u003eA\u003c/strong\u003e) Rice seed samples were embedded in 5-mL tube caps, covered with a cover glass, and (\u003cstrong\u003eB\u003c/strong\u003e) polymerized at 60°C for 24 h under oxygen-free conditions; and then (\u003cstrong\u003eC\u003c/strong\u003e) the resin block was trimmed to a trapezoidal shape. (\u003cstrong\u003eD\u003c/strong\u003e) Top view of a trimmed tissue block. (E) Finally, samples were sectioned at 2 µm thickness using a rotary microtome.\u003c/p\u003e","description":"","filename":"FIG2.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/d0d80e2e5406a7bf17a0bbfb.png"},{"id":108832003,"identity":"fa1608b5-1c31-4b2c-a224-0072eaa1853e","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":6806478,"visible":true,"origin":"","legend":"\u003cp\u003eAnatomical features of resin-embedded rice seed tissues at different developmental stages. Rice seeds were collected at 6 days after pollination (DAP; S2), 9 DAP (S3), and 16 DAP (S4), followed by fixation, LR White resin embedding, and sectioning into 2-µm-thick cross sections. (\u003cstrong\u003eA\u003c/strong\u003e) Seed morphology of TNG67 rice at the S3 stage. (\u003cstrong\u003eB\u003c/strong\u003e) Magnified view of the red boxed region in (A). (\u003cstrong\u003eC\u003c/strong\u003e) Developing seeds of TNG67 rice at S2 to S4. Major tissue structures are indicated, including the aleurone layer (AL), caryopsis coat (CC), dorsal vascular bundle (DVB), nucellar projection (NP), and nucellar epidermis (NE). Scale bars = 50 µm.\u003c/p\u003e","description":"","filename":"FIG3.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/dfb0ff76fc6edcda554beb59.png"},{"id":108832004,"identity":"70261957-24de-4093-b9d8-b9ab990bba60","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":6316708,"visible":true,"origin":"","legend":"\u003cp\u003eHistochemical detection of nutrient components in resin-embedded seed tissues. (\u003cstrong\u003eA\u003c/strong\u003e) Polysaccharides visualized by periodic acid–Schiff (PAS) staining. (\u003cstrong\u003eB\u003c/strong\u003e) Starch detected using 1% iodine–potassium iodide (I₂–KI) solution. (\u003cstrong\u003eC\u003c/strong\u003e) Proteins stained with Coomassie Brilliant Blue. Scale bars = 50 µm.\u003c/p\u003e","description":"","filename":"FIG4.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/1863c7561ec7005958aa447c.png"},{"id":108832007,"identity":"a745087e-de5d-4e35-9256-d8c947d6bf38","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":4938982,"visible":true,"origin":"","legend":"\u003cp\u003eFluorescence \u003cem\u003ein situ\u003c/em\u003e hybridization (FISH) eliminates background associated with conventional \u003cem\u003ein situ \u003c/em\u003ehybridization (ISH) in rice anthers. Chromogenic ISH detection of \u003cem\u003eRbohF\u003c/em\u003e transcripts in rice anthers. (\u003cstrong\u003eA\u003c/strong\u003e) Antisense probe of \u003cem\u003eRbohF\u003c/em\u003e. (\u003cstrong\u003eB\u003c/strong\u003e) The \u003cem\u003eRbohF\u003c/em\u003e sense probe was used as a negative control. (\u003cstrong\u003eC\u003c/strong\u003e) Section stained superoxide with 0.5 mM nitro blue tetrazolium (NBT). (\u003cstrong\u003eD\u003c/strong\u003e) Section stained with K-citrate buffer only as a negative control. (E, F) FISH detection of \u003cem\u003eRbohF\u003c/em\u003e transcripts. (\u003cstrong\u003eE\u003c/strong\u003e) Antisense probe of \u003cem\u003eRbohF\u003c/em\u003e labeled with Cy5 showing specific hybridization signals with minimal background. (\u003cstrong\u003eF\u003c/strong\u003e) Corresponding bright-field image. Arrows indicate signals in tapetal cells, asterisks indicate signals in young microspores, and arrowheads indicate signals in vascular bundles. Scale bars = 20 µm.\u003c/p\u003e","description":"","filename":"FIG5.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/48e65548dbec9e6a19895e8c.png"},{"id":108832008,"identity":"c345dd7b-d053-421b-bc06-caf5747c62e8","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":3066295,"visible":true,"origin":"","legend":"\u003cp\u003eChromogenic \u003cem\u003ein situ \u003c/em\u003ehybridization reveals spatial expression patterns of \u003cem\u003eRPBF\u003c/em\u003eand \u003cem\u003eGBSSI\u003c/em\u003e in developing rice seeds. Developing seeds of TNG67 rice at S1 (3 DAP) to S4 stage (16 DAP) were analyzed by chromogenic \u003cem\u003ein situ \u003c/em\u003ehybridization (ISH) using gene-specific antisense probes. Arrows indicate hybridization signals in the aleurone layer (AL); arrowheads indicate signals in the dorsal vascular bundle (DVB); asterisks denote signals in maternal tissues, including the nucellar epidermis (NE) and nucellar projection (NP). Scale bars = 20 µm.\u003c/p\u003e","description":"","filename":"FIG6.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/4aef9269120b918a9e5d0023.png"},{"id":108977602,"identity":"21e11de5-a1e6-41ff-a4c5-b0275cf72298","added_by":"auto","created_at":"2026-05-11 11:32:17","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":2915508,"visible":true,"origin":"","legend":"\u003cp\u003eFluorescence \u003cem\u003ein situ \u003c/em\u003ehybridization (FISH) showing spatial and temporal gene expression patterns of \u003cem\u003eRice Prolamin Box Binding Factor \u003c/em\u003e(\u003cem\u003eRPBF)\u003c/em\u003eand \u003cem\u003eGranule-Bound Starch Synthase I \u003c/em\u003e(\u003cem\u003eGBSSI\u003c/em\u003e) in developing rice seeds. Arrows indicate hybridization signals in the aleurone layer; arrowheads indicate signals in the dorsal vascular bundle (DVB); asterisks indicate signals in maternal tissues, including the nucellar epidermis (NE) and nucellar projection (NP); arrowheads indicate signals in the aleurone layer. Cy3 channels were detected at 561 nm /597 nm, Cy5 channels were detected at 639 nm/740 nm. Scale bars = 20 µm.\u003c/p\u003e","description":"","filename":"FIG7.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/431b540d904772374e15860c.png"},{"id":108832006,"identity":"9e17a286-0b15-4fdd-8667-6b3c752a79d3","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":5514031,"visible":true,"origin":"","legend":"\u003cp\u003eDual-color fluorescence \u003cem\u003ein situ\u003c/em\u003e hybridization (FISH) reveals spatial and temporal co-expression of \u003cem\u003eRice Prolamin Box Binding Factor \u003c/em\u003e(\u003cem\u003eRPBF\u003c/em\u003e) and \u003cem\u003eGranule-Bound Starch Synthase I \u003c/em\u003e(\u003cem\u003eGBSSI\u003c/em\u003e) during rice seed development. Developing rice seed sections hybridized with \u003cem\u003eRPBF \u003c/em\u003e(green) and \u003cem\u003eGBSSI \u003c/em\u003e(magenta) probes at successive developmental stages: (\u003cstrong\u003eA\u003c/strong\u003e) 3 days after pollination (DAP); (\u003cstrong\u003eB\u003c/strong\u003e) 6 DAP (stage S2); (\u003cstrong\u003eC\u003c/strong\u003e) 9 DAP (stage S3); and (\u003cstrong\u003eD\u003c/strong\u003e) 16 DAP (stage S4). (\u003cstrong\u003eE\u003c/strong\u003e) Higher-magnification view of the maternal tissues at 3 DAP shown in (A, maternal). (\u003cstrong\u003eF\u003c/strong\u003e) Higher-magnification view of the aleurone layer in tissues at 3 DAP shown in (A, filial). The white signal indicates overlap of\u003cem\u003e RPBF \u003c/em\u003eand \u003cem\u003eGBSSI\u003c/em\u003e expression. Scale bars = 20 µm.\u003c/p\u003e","description":"","filename":"FIG8.png","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/d3f7bd2a0c852504cc92953b.png"},{"id":109249353,"identity":"3de49fa9-e685-4dc7-b2ee-00d47a310c02","added_by":"auto","created_at":"2026-05-14 08:49:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":36816163,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/843ab4c5-c107-4627-9085-09b5149f28e7.pdf"},{"id":108832000,"identity":"402866b6-afe5-433a-8604-46044c13acb7","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":8230555,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAdditional file 1. \u003c/strong\u003eVideo showing the trimming of a resin block using a Sonic Saber knife.\u003c/p\u003e","description":"","filename":"VideoTrimResinblock.mp4","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/769552bbf161cadf69c65e29.mp4"},{"id":108832002,"identity":"8d599079-f4ff-4796-baa7-eadfb5b1530b","added_by":"auto","created_at":"2026-05-08 20:11:33","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":782294,"visible":true,"origin":"","legend":"","description":"","filename":"SUPPFIG.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9466150/v1/3156b59d9e9880ef3a76bdc2.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Resin-embedded FISH Enables High-resolution Spatial Analysis of Gene Expression in Developing Rice Seeds","fulltext":[{"header":"Background","content":"\u003cp\u003eRNA \u003cem\u003ein situ\u003c/em\u003e hybridization (RNA ISH) is a robust molecular technique that enables the detection and localization of specific RNA transcripts directly within fixed cells and tissue sections. Unlike RT-qPCR, which reports only bulk or averaged transcript levels, RNA ISH preserves the tissue and cellular architecture, enabling spatial visualization of transcripts at cellular and subcellular resolution. This capability is particularly valuable for studying heterogeneous tissues, developmental patterns, and cell-type-specific gene regulation (Levsky and Singer \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Miya et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eRNA ISH exhibits high specificity, as it relies on sequence-specific probes that can discriminate among closely related transcripts or splice variants (Femino et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). It also offers high sensitivity, enabling the detection of low-abundance transcripts. Single-molecule RNA ISH techniques enable visualization of individual RNA molecules as discrete punctate signals, facilitating quantitative analysis at the single-cell level (Raj et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Zenklusen and Singer \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Dual-color RNA ISH further extends this capability by simultaneously detecting two gene transcripts within the same tissue section, allowing direct comparison of co-expression or cell-type-specific gene expression while reducing section-to-section variation (Levsky and Singer \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Additionally, RNA ISH can be combined with immunohistochemistry or immunofluorescence to visualize RNA and protein targets in the same tissue section, enabling integrated analysis of transcriptional and protein expression patterns (Femino et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). Overall, RNA ISH offers higher spatial resolution, morphological context, and specificity compared with RT-qPCR, making it a powerful tool for dissecting gene expression patterns in complex tissues.\u003c/p\u003e \u003cp\u003eDespite its advantages, RNA ISH has several limitations. One major constraint is that RNA ISH protocols require careful optimization of tissue preparation, probe hybridization, and signal amplification, often involving specialized reagents and imaging systems (Choi et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). In addition, RNA ISH is generally low-throughput compared with next-generation sequencing\u0026ndash;based methods such as RNA sequencing or spatial transcriptomics. Traditional RNA ISH assays typically target a limited number of genes per experiment, making them less suitable for unbiased transcriptome-wide analysis (St\u0026aring;hl et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Although multiplex RNA ISH approaches have expanded detection capacity, they remain constrained by probe design complexity and imaging limitations. Quantitative analysis in RNA ISH can also be challenging. Signal intensity and transcript counts may be influenced by probe efficiency, tissue permeability, and imaging parameters, which can affect reproducibility across experiments and laboratories (Choi et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHigh-quality tissue sections are essential for accurate spatial gene expression profiling. However, certain plant tissues, such as rice seeds that are rich in starchy endosperm, are particularly susceptible to structural damage during RNA \u003cem\u003ein situ\u003c/em\u003e hybridization. Prolonged hybridization and washing procedures lasting 2\u0026ndash;3 days often lead to paraffin tissue swelling or separation of the endosperm from surrounding tissues. The high starch content and relatively loose cellular organization of the endosperm reduce its resistance to mechanical and osmotic stress caused by repeated buffer exchanges, elevated hybridization temperatures, and protease treatments. Together, these factors compromise tissue integrity, adversely affecting probe penetration, signal uniformity, and overall experimental reproducibility.\u003c/p\u003e \u003cp\u003eResin (plastic) embedding allows semithin sections (~\u0026thinsp;1 \u0026micro;m) that provide superior preservation of morphology, cell walls, and intracellular structures compared with paraffin wax sections, which are generally thicker and less resolving. Using this approach, C\u003csub\u003e4\u003c/sub\u003e photosynthetic maize tissues exhibited clear starch accumulation within chloroplasts of bundle sheath cells (Edwards et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). The enhanced structural stability provided by resin embedding is particularly advantageous for fragile, starch-rich tissues, such as endosperm, and offers more consistent section quality and improved morphological interpretation compared with conventional paraffin-based methods.\u003c/p\u003e \u003cp\u003eRice endosperm development is characterized by the coordinated accumulation of starch and storage proteins, a process tightly regulated at the transcriptional level. Among the key transcription factors involved, Rice Prolamin Box Binding Factor (RPBF) plays a central role in controlling seed storage protein gene expression. Previous studies have demonstrated that RPBF is a key transcription factor involved in endosperm development. Loss of RPBF function results in chalky endosperm, reduced seed size, and defective grain filling (Kawakatsu et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). RPBF binds to the conserved prolamin box present in the promoters of many endosperm-expressed storage protein genes (Wu et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Yamamoto et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). In contrast, genes involved in starch biosynthesis, such as \u003cem\u003eGranule-Bound Starch Synthase I\u003c/em\u003e (\u003cem\u003eGBSSI/Wx1\u003c/em\u003e), are regulated by a distinct but overlapping transcriptional network and do not contain canonical P-box motifs in their promoters. Although RPBF does not directly regulate \u003cem\u003eGBSSI\u003c/em\u003e, both storage protein and starch biosynthetic pathways are developmentally synchronized, suggesting that RPBF may indirectly influence carbon\u0026ndash;nitrogen allocation during endosperm maturation (Ning et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). GBSSI is specifically localized to starch granules in endosperm amyloplasts and is responsible for the synthesis of amylose, a key determinant of grain cooking and eating quality. Natural variation or mutation in \u003cem\u003eWx1\u003c/em\u003e alters amylose content, leading to changes in grain texture, including waxy or low-amylose phenotypes, and can influence endosperm structure and grain filling efficiency (Wang et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDespite advances in seed biology, high-resolution spatial analysis of gene expression in rice reproductive tissues remains limited by poor tissue preservation and low signal clarity during \u003cem\u003ein situ\u003c/em\u003e hybridization. In particular, the coordinated expression of key genes, such as \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e, during seed development remains poorly understood at the cellular level. This study aimed to establish an optimized workflow combining resin embedding with fluorescence \u003cem\u003ein situ\u003c/em\u003e hybridization (FISH) to improve tissue integrity and detection sensitivity. Using this approach, we sought to characterize morphological features, histochemical changes, and spatial gene expression patterns, providing new insights into the regulation of rice seed development.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eParaffin-embedding Tissue Reduced Image Quality after ISH\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eClassic paraffin-embedded sections maintained good image quality following short-term staining, such as hematoxylin staining for 50 seconds, which clearly visualized starch distribution and tissue morphology of caryopsis coat (CC), dorsal vascular bundle (DVB), nucellar epidermis (NE), aleurone layer (AL), and starchy endosperm (SE) \u003cstrong\u003e(Fig. 1A)\u003c/strong\u003e. However, when subjected to the multi-step \u003cem\u003ein situ\u003c/em\u003e hybridization (ISH) procedure lasting approximately three days, the paraffin sections exhibited significant deterioration of tissue integrity. The prolonged exposure to high temperature, enzymatic digestion, and repeated washing steps caused tissue deformation and partial detachment from the slides \u003cstrong\u003e(Fig. 1B)\u003c/strong\u003e. As a result, the morphological structure became distorted, and the resolution of cellular features was markedly reduced, hindering accurate localization and interpretation of gene expression signals. Clearly, conventional paraffin embedding is not suitable for high-quality ISH analysis of rice seed tissues.\u0026nbsp;\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eEmbedding and Sectioning of LR White\u0026ndash;embedded Rice Seeds\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo overcome the tissue deformation and section instability associated with paraffin-based RNA \u003cem\u003ein situ\u003c/em\u003e hybridization, we established a resin-based ISH workflow optimized for rice seeds. Because rice seeds are substantially larger than tissues typically processed by resin embedding, several modifications to standard embedding and sectioning procedures were required. To accommodate seed size, 5-mL microcentrifuge tube caps were used as embedding molds. Following London Resin (LR) White infiltration, samples were oriented in the molds and covered with a glass coverslip to exclude oxygen during polymerization, a critical factor for efficient LR White curing. This modification enabled consistent and uniform polymerization and generated mechanically stable resin blocks. After polymerization, the LR White blocks exhibited substantial hardness, making them difficult to trim using conventional razor blades. To overcome this limitation, the resin blocks were mounted onto a trimming chuck using hot-melt adhesive. The resin block trimming was performed using an ultrasonic knife, which significantly improved trimming efficiency and precision while minimizing mechanical stress and potential damage to the embedded seed tissues. The workflow is summarized in \u003cstrong\u003eFig. 2\u003c/strong\u003e, and a detailed trimming procedure is presented in the accompanying video (\u003cstrong\u003eSupplemental File 1\u003c/strong\u003e). Overall, this optimized embedding and sectioning strategy enabled the reproducible preparation of high-quality LR White sections of rice seeds, with improved structural integrity and section stability compared to paraffin-based methods.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResin-embedding Improves Anatomical Resolution in Rice Seed Sections\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDeveloping seeds of the rice cultivar TNG67 were collected at representative developmental stages: S2 (6 days after pollination (DAP)), S3 (9 DAP), and S4 (16 DAP). Samples were fixed, embedded in LR-white resin, transversely sectioned at a thickness of 2 \u0026mu;m, and stained with TBO for microscopic examination. Resin-embedded whole seeds exhibited well-preserved tissue integrity, capturing both the rigid structure of the husk and the loosely organized texture of the starchy endosperm (\u003cstrong\u003eFig. S1\u003c/strong\u003e). At the early developmental stages (S2), seed cross-sections revealed a prominently thickened caryopsis coat derived from maternal tissues, which enclosed a relatively small and incompletely developed endosperm. The cellular organization of the endosperm at this stage appeared less compact, and the aleurone layer (AL) was not yet clearly differentiated. As seed development progressed to the S3 stage, one AL and grain filling were observed. At the S4 stage, a marked reduction in the thickness of the caryopsis coat was observed. In contrast, the endosperm exhibited substantial expansion, occupying a larger proportion of the seed volume, accompanied by clear differentiation and thickening of the aleurone layer to two layers. The high-resolution TBO-stained resin sections allowed precise visualization of tissue boundaries and cellular architecture, providing a robust morphological framework for subsequent analyses of gene expression and storage reserve accumulation during rice seed maturation (\u003cstrong\u003eFig. 3\u003c/strong\u003e).\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eHistochemical Staining of Resin-embedded Tissues Reveals Dynamic Nutrient Accumulation During Seed Development\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate the spatial and temporal patterns of nutrient accumulation during rice seed development, transverse sections of resin-embedded seeds collected at stages S2 to S4 were subjected to histochemical staining using periodic acid\u0026ndash;Schiff (PAS), iodine\u0026ndash;potassium iodide, and coomassie brilliant blue (CBB). PAS staining, which detects polysaccharides, revealed distinct developmental stage\u0026ndash;dependent distribution patterns. At the early S2 stage, strong PAS signals were predominantly detected in the maternal caryopsis coat (CC), while the filial endosperm showed relatively weak staining. By the S3 stage, PAS signals were observed at comparable levels in both the CC and endosperm, indicating a shift in polysaccharide distribution. At the S4 stage, PAS staining was largely absent from the CC and became strongly enriched in the endosperm, suggesting a developmental shift of carbohydrate accumulation from maternal tissues to filial tissues during seed maturation \u003cstrong\u003e(Fig. 4A)\u003c/strong\u003e. Consistent with the PAS staining results, iodine staining revealed similar patterns of starch accumulation across developmental stages. At S2, starch signals were weak and mainly associated with maternal tissues.\u0026nbsp;As development progressed, starch accumulation increased markedly within the endosperm. Notably, iodine staining in the starchy endosperm shifted from red at S2 to deep blue at the later stages, indicating progressive changes in starch content and composition, with increasing amylose accumulation as endosperm cells matured (\u003cstrong\u003eFig. 4B)\u003c/strong\u003e. CBB staining was used to visualize protein accumulation within developing seeds. At the early stages, protein signals were relatively weak; however, by the S4 stage, strong CBB staining was observed, indicating substantial storage protein accumulation. At this stage, protein signals were detected in both the CC and endosperm compartments, with pronounced enrichment in the endosperm, consistent with active storage protein deposition during seed maturation (\u003cstrong\u003eFig. 4C)\u003c/strong\u003e Together, these histochemical analyses demonstrated coordinated and stage-specific redistribution of carbohydrates and proteins during rice seed development, supporting the transition from\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003ematernal tissue support to endosperm-centered nutrient storage as seeds mature. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFluorescence\u0026nbsp;\u003cem\u003eIn Situ\u0026nbsp;\u003c/em\u003eHybridization (FISH) Reduces ROS-related Background\u003cbr\u003e\u003c/strong\u003eConventional chromogenic \u003cem\u003ein situ\u003c/em\u003e hybridization (ISH) commonly uses nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) as a substrate for signal detection; however, this reaction is highly sensitive to reactive oxygen species (ROS). In rice anthers, elevated ROS levels are intrinsically associated with tapetal programmed cell death (PCD), a well-characterized and essential developmental process during microsporogenesis (Hu et al. 2011; Yi et al. 2016). As a result, chromogenic ISH frequently produces strong nonspecific background signals, including in sense-probe negative controls, which can obscure authentic hybridization patterns (\u003cstrong\u003eFig. 5A, B\u003c/strong\u003e). Consistent with these observations, NBT staining revealed prominent superoxide accumulation in anther tissues, particularly in tapetal cells, meiocytes, and vascular bundles (\u003cstrong\u003eFig. 5C\u003c/strong\u003e). A section stained with K-citrate buffer alone, used as a negative control, showed no detectable signal (\u003cstrong\u003eFig. 5D\u003c/strong\u003e). In contrast, FISH largely eliminated ROS-associated background, enabling clearer and more specific detection of gene expression in rice anther tissues (\u003cstrong\u003eFig. 5E, F\u003c/strong\u003e). In rice seed tissues, no detectable autofluorescence was observed in the Cy3 or Cy5 channels, indicating minimal background interference for probes labeled with these fluorophores. In contrast, strong autofluorescence was detected in the DAPI channel, which may interfere with the interpretation of FISH signals (\u003cstrong\u003eFig. S2\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eResin-embedded Tissues Enable Precise Spatial and Temporal Analysis of Gene Expression\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate the spatial expression patterns of key seed development markers, we examined the seed-specific genes \u003cem\u003eRice Prolamin Box Binding Factor\u003c/em\u003e (\u003cem\u003eRPBF\u003c/em\u003e) and \u003cem\u003eGranule-bound Starch Synthase I\u003c/em\u003e (\u003cem\u003eGBSS1\u003c/em\u003e) using both chromogenic\u003cem\u003e\u0026nbsp;in situ\u003c/em\u003e hybridization (ISH) and fluorescence ISH (FISH) on resin-embedded rice seed tissue sections (\u003cstrong\u003eFig. 6, 7\u003c/strong\u003e). These two ISH technologies produced similar expression patterns. Notably, the high-quality preservation of tissue morphology in resin sections allowed clear visualization of gene expression at cellular resolution. Chromogenic \u003cem\u003ein situ\u0026nbsp;\u003c/em\u003ehybridization showed negligible \u003cem\u003eISH\u003c/em\u003e background with sense probes of \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e (\u003cstrong\u003eFig. S3\u003c/strong\u003e). \u003cem\u003eRPBF\u003c/em\u003e expression was strongly detected in maternal carpel tissues during early seed development (stages S1 and S2, asterisks), but expression declined markedly as seeds matured, with only minimal signal observed in the aleurone layer at later stages, suggesting temporal regulation of \u003cem\u003eRPBF\u003c/em\u003e during early seed development (\u003cstrong\u003eFig. 6\u0026nbsp;\u003c/strong\u003eupper panel). \u003cem\u003eGBSS1\u003c/em\u003e, a key enzyme in amylose synthesis, showed a strong ISH signal in the maternal tissue but a weak signal in filial tissue at S1, the caryopsis coat of S2, and signals intensifying in the aleurone layer and endosperm at S4 developmental stages, indicating active starch biosynthesis during endosperm filling (\u003cstrong\u003eFig. 6,\u0026nbsp;\u003c/strong\u003ebottom panel). \u0026nbsp;\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eFluorescence \u003cem\u003eIn Situ\u003c/em\u003e Hybridization (FISH) Reveals Clear and Reproducible Gene Expression Patterns\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eResin-embedded tissue sections identical to those used for chromogenic ISH were employed to examine \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e expression by fluorescence \u003cem\u003ein situ\u0026nbsp;\u003c/em\u003ehybridization (FISH). FISH produced distinct, well-defined hybridization signals with low background, allowing clear visualization of gene expression patterns \u003cstrong\u003e(Fig. 7)\u003c/strong\u003e. Importantly, the spatial distribution of ISH signals detected by FISH was consistent with that obtained using chromogenic ISH, demonstrating high reproducibility between the two approaches. These results confirm that FISH reliably detects gene expression in resin-embedded rice tissues while providing improved signal clarity and imaging resolution. \u0026nbsp;\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eDual-color FISH Reveals Co-expression Patterns of \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate the spatial co-expression patterns of two genes within the same tissue section, dual-color fluorescence \u003cem\u003ein situ\u003c/em\u003e hybridization (DISH) was performed using gene-specific probes targeting \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e. The dual-color FISH analysis enabled simultaneous visualization of the transcripts and provided high spatial resolution of their expression domains. The results showed that \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e transcripts were detected in overlapping regions of the developing rice endosperm, indicating co-expression within the same cell types. In cells exhibiting overlapping signals, the two transcripts displayed distinct yet coordinated spatial distributions. Notably, the DISH signals were clearly distinguishable with minimal background interference, demonstrating the robustness of this approach for resolving gene co-expression patterns in complex tissues. These observations suggest that \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e are co-expressed in the same developmental context, supporting a potential regulatory relationship during endosperm development.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eA major challenge in studying rice reproductive tissues is preserving the structural integrity of seeds during multi-stage preparation protocols such as \u003cem\u003ein situ\u003c/em\u003e hybridization (ISH). Rice grains are relatively large and anatomically complex, containing both rigid maternal tissues and highly hydrated filial tissues rich in starch and storage proteins. These features make them particularly susceptible to deformation during conventional paraffin embedding and sectioning. In this study, we addressed these limitations by optimizing an acrylic resin\u0026ndash;based embedding workflow using LR White, which has been widely applied in high-resolution histological studies of plant systems (Newman and Hobot \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1999\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSeveral critical modifications were introduced to adapt standard protocols to better suit rice seeds. First, customized embedding molds were employed to accommodate whole developing grains and maintain proper orientation. Second, oxygen exclusion during polymerization was implemented to ensure consistent curing of LR White resin, as oxygen is known to inhibit free-radical polymerization and compromise block quality. Third, ultrasonic trimming proved efficient for sectioning hardened resin blocks. Compared with conventional razor blades, ultrasonic knives significantly improved trimming precision while minimizing mechanical stress and preventing tissue fracture. These technical refinements collectively enabled the reproducible preparation of high-quality sections with minimal distortion (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe resulting resin-embedded sections exhibited excellent preservation of cellular architecture across developmental stages. Toluidine blue staining clearly resolved key anatomical features, including the husk, caryopsis coat, nucellar epidermis, aleurone layer, and starchy endosperm (\u003cb\u003eFig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The ability to simultaneously preserve both rigid and soft tissues highlights the superior infiltration and mechanical support provided by acrylic resins, particularly in plant tissues with dense cell walls and high carbohydrate content (Newman and Hobot \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). This level of structural preservation is essential for accurate interpretation of tissue-specific processes during seed development.\u003c/p\u003e \u003cp\u003eIn addition to morphological analyses, resin sections proved highly compatible with histochemical staining. Periodic acid\u0026ndash;Schiff (PAS) and iodine staining revealed dynamic changes in carbohydrate distribution during seed maturation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Specifically, polysaccharides were initially enriched in maternal tissues such as the caryopsis coat and subsequently redistributed to the endosperm, where starch accumulation increased progressively. These observations are consistent with established models of cereal grain filling, in which assimilates are transported from maternal tissues to the developing endosperm (Olsen \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Sabelli and Larkins \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Similarly, the accumulation of storage proteins at later developmental stages reflects the transition toward endosperm maturation and reserve deposition. Together, these results demonstrate that resin embedding supports integrated structural and biochemical analyses, providing a comprehensive framework for studying seed development.\u003c/p\u003e \u003cp\u003ePrecise spatial analysis of gene expression is critical for understanding the regulatory networks governing seed development. However, conventional paraffin-based ISH methods often fail to maintain tissue integrity during prolonged hybridization and washing steps, particularly in starch-rich tissues such as rice endosperm. In this study, we demonstrated that resin-embedding combined with fluorescence \u003cem\u003ein situ\u003c/em\u003e hybridization (FISH) provides a robust and reproducible platform for cellular-resolution gene expression analysis.\u003c/p\u003e \u003cp\u003eParaffin sections showed substantial deformation, distortion, and detachment from slides following ISH procedures (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), consistent with previous reports highlighting the fragility of paraffin-embedded plant tissues during multi-step protocols (Jackson \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1991\u003c/span\u003e). These artifacts significantly compromise the accuracy of signal localization and limit the interpretation of gene expression patterns. In contrast, resin-embedded sections maintained structural integrity throughout the ISH workflow, enabling precise mapping of transcripts within defined cellular contexts.\u003c/p\u003e \u003cp\u003eAn additional advantage of the resin-based approach is improved signal clarity in FISH analyses. Background fluorescence was observed in the DAPI channel, likely due to intrinsic autofluorescence from cell wall components or storage materials in seeds; however, signals in longer-wavelength channels such as Cy3 and Cy5 were clear and reproducible (\u003cb\u003eFig. S2)\u003c/b\u003e. This observation is consistent with previous studies showing that plant tissues often exhibit strong autofluorescence in the blue spectrum, necessitating the use of red or far-red fluorophores for optimal signal detection (Duncan et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Ruzin \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Therefore, careful selection of fluorophores is essential to minimize background and maximize sensitivity in plant FISH experiments.\u003c/p\u003e \u003cp\u003eA key finding of this study is the reduced susceptibility of FISH to reactive oxygen species (ROS)-related background compared with chromogenic ISH. In rice anthers, high ROS levels are associated with tapetal programmed cell death, a critical developmental process (Hu et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Yi et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Under these conditions, chromogenic detection using NBT/BCIP substrates produced strong nonspecific signals that overlapped with endogenous ROS distribution, even in negative controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e5\u003c/span\u003e). This limitation reflects the inherent sensitivity of enzymatic colorimetric reactions to oxidative environments (Jackson, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1991\u003c/span\u003e). In contrast, FISH largely eliminated such background interference, enabling more specific and reliable detection of gene expression. These results underscore the advantages of fluorescence-based detection, particularly for tissues undergoing oxidative stress or active metabolic changes.\u003c/p\u003e \u003cp\u003eBoth chromogenic ISH and FISH performed on resin sections yielded consistent spatial and temporal expression patterns for key seed developmental genes. \u003cem\u003eRPBF\u003c/em\u003e transcripts were predominantly detected in maternal tissues during early stages and declined as seeds matured, whereas \u003cem\u003eGBSSI\u003c/em\u003e expression increased progressively in the aleurone layer and starchy endosperm (Figs.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e6\u003c/span\u003e, \u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e7\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e. These patterns are consistent with the known roles of \u003cem\u003eRPBF\u003c/em\u003e as a transcriptional regulator of storage protein genes and \u003cem\u003eGBSSI\u003c/em\u003e as a key enzyme in amylose biosynthesis (Sano \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1984\u003c/span\u003e; Kawakatsu et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). The ability to recapitulate these established expression dynamics validates the reliability and sensitivity of the resin-based ISH platform.\u003c/p\u003e \u003cp\u003eImportantly, superior preservation of tissue morphology enabled precise localization of transcripts at cellular resolution. Distinct cell layers and tissue compartments could be clearly resolved, allowing accurate assignment of gene expression domains. This level of detail is difficult to achieve with paraffin-based methods, particularly in cereal seeds where structural distortion is common. Thus, resin embedding represents a critical methodological advance for spatial transcript analysis in plant tissues.\u003c/p\u003e \u003cp\u003eMultiplex FISH enhances analytical power by enabling simultaneous detection of multiple transcripts within the same tissue. Dual-color FISH revealed overlapping expression of \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e in developing endosperm cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e8\u003c/span\u003e), indicating coordinated gene activity during seed development. \u003cem\u003eRPBF\u003c/em\u003e expression was prominent in maternal carpel tissues at the early stages (S1\u0026ndash;S2) but declined during maturation, suggesting a role in early endosperm differentiation (Yamamoto et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn contrast, \u003cem\u003eGBSSI\u003c/em\u003e expression increased progressively, with strong accumulation in the aleurone layer at later stages (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e6\u003c/span\u003e, \u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e7\u003c/span\u003e), consistent with its role in amylose biosynthesis during grain filling (Liu et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Notably, co-expression of both genes was observed in the aleurone layer at 3 DAP (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e8\u003c/span\u003eF), suggesting a potential interaction during early grain development. Although this overlap may reflect coordinated regulation or complementary functions in starch metabolism, its biological significance remains unclear and warrants further functional and regulatory studies. The ability to visualize such interactions at cellular resolution demonstrates the value of resin-based multiplex FISH for dissecting complex gene regulatory networks in rice seed development.\u003c/p\u003e \u003cp\u003eBeyond targeted gene analysis, the methods developed in this study have important implications for validating high-throughput transcriptomic datasets. RNA sequencing (RNA-seq) has revolutionized plant biology by providing comprehensive gene expression profiles across developmental stages and conditions. However, RNA-seq lacks spatial resolution, making it difficult to assign gene expression to specific cell types or tissues. Integrating transcriptomic data with spatial techniques such as ISH and FISH is therefore essential for linking gene expression patterns to developmental processes (Wu et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe resin-based platform established here provides a reliable and versatile tool for such validation. Enabling high-resolution localization of transcripts within intact tissue contexts allows researchers to confirm and refine gene expression patterns identified through RNA-seq.\u0026nbsp;Moreover, the compatibility of resin sections with multiplex FISH opens up new possibilities for analyzing gene co-expression networks and cell-type\u0026ndash;specific regulatory mechanisms. This approach is particularly valuable for complex tissues such as developing seeds, where multiple cell types interact dynamically during development.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, this study establishes resin embedding combined with FISH as a powerful and versatile approach for spatial transcript analysis in rice reproductive tissues. By overcoming the structural limitations of paraffin embedding and reducing the background interference associated with chromogenic detection, this method enables high-resolution, reliable visualization of gene expression. The ability to perform multiplex analyses further enhances its utility for investigating coordinated gene regulation. Collectively, these advances provide a robust framework for studying the molecular mechanisms underlying rice seed development and offer broad applications for functional genomics in cereal crops.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e\u003cstrong\u003ePlant Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRice (\u003cem\u003eOryza sativa\u003c/em\u003e L.) cultivar TNG67 plants were grown in soil under controlled conditions in the genetically modified organism (GMO) greenhouse at the Biotechnology Center, Academia Sinica (Tainan, Taiwan). Flowers were tagged at anthesis, and developing seeds were collected at 6 days after pollination (DAP; stage S2), 6 DAP (S3), and 16 DAP (S4). Harvested seeds were immediately fixed and processed for histological analysis, \u003cem\u003ein situ\u003c/em\u003e hybridization, and fluorescence \u003cem\u003ein situ\u003c/em\u003e hybridization as described below.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eParaffin Tissue Sectioning\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eParaffin sectioning was performed as previously described (Chen et al., 2016). Briefly, developing seeds were fixed, dehydrated through a graded ethanol series, cleared, and embedded in paraffin. Sections of appropriate thickness were cut using a rotary microtome and mounted onto SuperFrost PLUS microscope slides (Cat. S7441, Matsunami, Japan). \u0026nbsp;\u003cbr\u003e\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLR White Resin Embedding and Sectioning of Rice Seeds\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRice (\u003cem\u003eOryza sativa\u003c/em\u003e) cv. TNG67 seeds were harvested at defined developmental stages (\u003cem\u003ei.e.\u003c/em\u003e, 3, 6, 9, and 16 days after pollination, DAP). Samples were fixed in freshly prepared 4% (w/v) paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.2, vacuum infiltrated for 20\u0026ndash;30 min, followed by incubation at 4\u0026deg;C for 12\u0026ndash;16 h. After fixation, tissues were washed three times (10 min each) in 0.1 M PB at 4\u0026deg;C. [\u003cem\u003eNote:\u003c/em\u003e For RNA-based applications (\u003cem\u003ei.e.\u003c/em\u003e, ISH or FISH), all solutions and tools were prepared using DEPC-treated water and RNase-free conditions.] Fixed tissues were dehydrated through a graded ethanol series as follows: 30% ethanol, 30 min; 50% ethanol, 30 min; 70% ethanol, 30 min; 85% ethanol, 30 min; 95% ethanol, 30 min; 100% ethanol, three changes, 30 min each. [Complete dehydration was critical to ensure efficient resin infiltration.] Then, dehydrated samples were infiltrated with LR White resin (AGR1281;\u0026nbsp;Agar\u0026nbsp;Scientific, Wetzlar, Germany) using a stepwise ethanol\u0026ndash;resin series at 4\u0026deg;C with gentle agitation: Ethanol:LR White (2:1, v/v), 2 h; Ethanol:LR White (1:1, v/v), 2 h; Ethanol:LR White (1:2, v/v), 2 h; and 100% LR White, overnight. The resin was replaced with fresh 100% LR White and incubated for an additional 6\u0026ndash;8 h to ensure complete infiltration. The sample was transferred into a tube cap of 5 mL Eppendorf tube, filled with fresh LR White resin, and oriented appropriately for sectioning. Polymerization was carried out at 60\u0026deg;C for 24 h in a dry oven. Polymerized blocks were allowed to cool to room temperature before trimming.\u003c/p\u003e\n\u003cp\u003eIn this study, resin blocks were trimmed with a Sonic Saber knife (Phrozen, Taiwan) and sectioned on a rotary microtome (Leica RM2255, Leica, Buffalo Grove, IL, USA) equipped with a diamond knife. Tissues were sectioned to\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003ea\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003ethickness of 2 \u0026micro;m.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSections were collected onto SuperFrost PLUS microscope slides (Cat. S7441, Matsunami, Japan). Slides were dried on a 42\u0026deg;C slide warmer for 1\u0026ndash;2 h or overnight at room temperature to ensure firm adhesion. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHistological Staining\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHistological staining was performed using standard procedures. Hematoxylin (ScyTek Laboratories, USA) staining was used to examine general tissue morphology. Toluidine blue O (TBO; Sigma-Aldrich, USA) staining was used to visualize cellular structure. 1% iodine\u0026ndash;potassium iodide (I₂/KI) staining was used to detect starch accumulation, and Coomassie Brilliant Blue (Bio-Rad, USA) staining was performed to visualize storage proteins. All staining procedures were conducted according to the manufacturers\u0026rsquo; instructions or established protocols. Stained or hybridized sections were examined using a bright-field microscope (Zeiss, Germany) for comparative analyses across developmental stages.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eIn Situ\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;hybridization (ISH) probe synthesis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRice genes \u003cem\u003eRPBF\u003c/em\u003e (Os02g0252400), \u003cem\u003eGBSSI\u003c/em\u003e (Os06g0133000), and \u003cem\u003eRbohF\u003c/em\u003e (Os08g0453700) were amplified and cloned for probe synthesis. Gene-specific primer sets were as follows: \u003cem\u003eRPBF\u003c/em\u003e, 5\u0026prime;-AGCGGCGGCGCATTATCCC-3\u0026prime; and 5\u0026prime;-AAGTAGCGCGGCTGCGCCATG-3\u0026prime;; \u003cem\u003eGBSSI\u003c/em\u003e, 5\u0026prime;-TCTCCGAGAGGTTCAGGTCATCC-3\u0026prime; and 5\u0026prime;-CCTGGCGATGCCGGAGATG-3\u0026prime;, and \u003cem\u003eRbohF\u003c/em\u003e, 5\u0026prime;-GTCCACCAAGGTTGCTTGTAGA-3\u0026prime; and 5\u0026prime;-CGTAGTTCCTGAAATCCTGCG-3\u0026prime;. The amplified gene-specific coding sequences were ligated into the pGEM-T easy vector (Promega, USA) following the manufacturer\u0026rsquo;s protocol. Digoxigenin (DIG)-labeled sense and antisense RNA probes were synthesized using the SP6/T7 DIG RNA Labeling Kit (cat. no. 11175025910, Roche) as previously described (Chen et al. 2016).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eChromogenic\u003cem\u003e\u0026nbsp;In Situ\u0026nbsp;\u003c/em\u003ehybridization\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe chromogenic \u003cem\u003ein situ\u0026nbsp;\u003c/em\u003ehybridization (ISH) was performed according to the protocol described by Chen et al. (2016) with minor modifications. Hybridization temperatures were optimized for each probe and set at 50\u0026deg;C for \u003cem\u003eRPBF\u003c/em\u003e, 55\u0026deg;C for \u003cem\u003eGBSSI\u003c/em\u003e, and 60\u0026deg;C for \u003cem\u003eRbohF\u003c/em\u003e. After hybridization and washing, DIG-labeled probes were detected using nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) substrate solution to visualize hybridization signals.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRNA Fluorescence \u003cem\u003eIn Situ\u0026nbsp;\u003c/em\u003eHybridization\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRNA Fluorescence \u003cem\u003ein situ\u0026nbsp;\u003c/em\u003ehybridization\u003cstrong\u003e\u0026nbsp;(\u003c/strong\u003eRNA FISH) was performed according to previously described protocols (Ko et al. 2021) with minor modifications. Briefly, DIG-labeled RNA probes were detected using anti-DIG-POD Fab fragments (cat. no. 11207733910, Roche), followed by tyramide signal amplification (TSA) using the TSA Plus Cyanine 5 (Cy5) detection kit (cat. no. NEL745001KT, PerkinElmer). Hybridization temperatures were at 50\u0026deg;C for \u003cem\u003eRPBF\u003c/em\u003e, 55\u0026deg;C for \u003cem\u003eGBSSI\u003c/em\u003e, and 60\u0026deg;C for \u003cem\u003eRbohF\u003c/em\u003e, respectively. Fluorescent signals were examined under a Zeiss LSM-980 confocal microscope, using Cy3 or Cy5 filters at excitation and emission wavelengths of 561 nm/597 nm or 639 nm/740 nm, respectively.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDual-color FISH \u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe dual-color FISH (DISH) technique was used to show the co-expression of \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e transcripts as described previously (Ko et al. 2021). A DIG RNA labeling kit (SP6/T7) (cat. no. 11175025910, Roche) and Fluorescein RNA Labeling Mix (cat. no. 11685619910, Roche) were used to synthesize probes according to the manufacturer\u0026rsquo;s instructions. Probes of \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e were simultaneously hybridized with tissue slices overnight at 53\u0026deg;C. Then, anti-fluorescein isothiocyanate (FITC)-POD was conjugated with \u003cem\u003eGBSSI\u003c/em\u003e probe overnight at room temperature, and tissues were reacted with TSA working solution containing TSA Plus Cyanine 3 (Cy3) (cat. no. NEL744001KT, PerkinElmer) (at 1:100 dilution) at room temperature for 5 min, and then washed twice. Afterward, tissues were treated with 3% hydrogen peroxide at room temperature for 10 min to inactivate the first peroxidase of anti-FITC-POD. After that, anti-DIG-POD was conjugated with the \u003cem\u003eRPBF\u003c/em\u003e probe at 37 \u0026deg;C for 2 h. Tissues were then stained with TSA working solution containing Cy5 amplification reagent (at 1:300 dilution) at room temperature for 5 min, and washed twice. Fluorescence signals were observed under a Zeiss LSM-980 confocal microscope, using Cy3 or Cy5 filters at excitation and emission wavelengths of 561 nm/597 nm or 639 nm/740 nm, respectively.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAL: aleurone layer; CBB: coomassie brilliant blue; CC: caryopsis coat; DAF: days after flowering; En: endosperm; FISH: fluorescent \u003cem\u003ein situ\u003c/em\u003e hybridization; \u003cem\u003eGBSS1\u003c/em\u003e: \u003cem\u003eGranule-bound Starch Synthase I\u003c/em\u003e ; ISH: \u003cem\u003ein situ\u003c/em\u003e hybridization; NE, nucellar epidermis; PAS: periodic acid\u0026ndash;Schiff; \u003cem\u003eRbohF\u003c/em\u003e: \u003cem\u003eRespiratory Burst Oxidase Homolog F\u003c/em\u003e; \u003cem\u003eRPBF\u003c/em\u003e: \u003cem\u003eRice Prolamin Box Binding Factor\u003c/em\u003e; TBO: toluidine blue O\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis work was supported by the Academia Sinica Biotechnology Center in Southern Taiwan.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eS.-S. Ko conceived and designed the research. T.-T. Yang and Y.-P. Ho performed the in situ hybridization and tissue sectioning. S.-S. Ko wrote the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe are grateful to Dr. Wann-Neng Jane for guidance with resin sectioning. We thank the DNA Sequencing Core Facility of the Institute of Biomedical Sciences, Academia Sinica for providing DNA sequencing services. We also thank Ms. Miranda Loney for assistance with English editing.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eChen TK, Yang HT, Fang SC, Lien YC, Yang TT, Ko SS (2016) Hybrid-Cut: An improved sectioning method for recalcitrant plant tissue samples. J Vis Exp 117:e54754. doi:10.3791/54754\u003c/li\u003e\n\u003cli\u003eChoi HM, Chang JY, Trinh LA, Padilla JE, Fraser SE, Pierce NA (2010) Programmable in situ amplification for multiplexed imaging of mRNA expression. Nature biotechnology 28 (11):1208\u0026ndash;1212\u003c/li\u003e\n\u003cli\u003eDuncan S, Olsson TS, Hartley M, Dean C, Rosa S (2016) A method for detecting single mRNA molecules in Arabidopsis thaliana. Plant methods 12 (1):13\u003c/li\u003e\n\u003cli\u003eEdwards GE, Franceschi VR, Ku MS, Voznesenskaya EV, Pyankov VI, Andreo CS (2001) Compartmentation of photosynthesis in cells and tissues of C4 plants. Journal of Experimental Botany 52 (356):577\u0026ndash;590\u003c/li\u003e\n\u003cli\u003eFemino AM, Fay FS, Fogarty K, Singer RH (1998) Visualization of single RNA transcripts in situ. Science 280 (5363):585\u0026ndash;590\u003c/li\u003e\n\u003cli\u003eHu L, Liang W, Yin C, Cui X, Zong J, Wang X, Hu J, Zhang D (2011) Rice MADS3 regulates ROS homeostasis during late anther development. Plant Cell 23 (2):515\u0026ndash;533. doi:10.1105/tpc.110.074369\u003c/li\u003e\n\u003cli\u003eJackson D (1991) In situ hybridisation in plants. In: Molecular Plant Pathology: A Practical Approach. Oxford University Press, Oxford, pp 163\u0026ndash;174\u003c/li\u003e\n\u003cli\u003eKawakatsu T, Yamamoto MP, Touno SM, Yasuda H, Takaiwa F (2009) Compensation and interaction between RISBZ1 and RPBF during grain filling in rice. Plant J 59 (6):908\u0026ndash;920. doi:10.1111/j.1365-313X.2009.03925.x\u003c/li\u003e\n\u003cli\u003eKo SS, Li MJ, Ho YC, Yu CP, Yang TT, Lin YJ, Hsing HC, Chen TK, Jhong CM, Li WH, Sun-Ben Ku M (2021) Rice transcription factor GAMYB modulates bHLH142 and is homeostatically regulated by TDR during anther tapetal and pollen development. J Exp Bot 72 (13):4888\u0026ndash;4903. doi:10.1093/jxb/erab190\u003c/li\u003e\n\u003cli\u003eLevsky JM, Singer RH (2003) Fluorescence in situ hybridization: past, present and future. J Cell Sci 116 (Pt 14):2833\u0026ndash;2838. doi:10.1242/jcs.00633\u003c/li\u003e\n\u003cli\u003eLiu D, Wang W, Cai X (2014) Modulation of amylose content by structure-based modification of OsGBSS1 activity in rice (Oryza sativa L.). Plant Biotechnol J 12 (9):1297\u0026ndash;1307. doi:10.1111/pbi.12228\u003c/li\u003e\n\u003cli\u003eMiya M, Hibara K-i, Itoh J-i (2024) Spatial Analysis of gene expression by In situ hybridization. In: Rice: Methods and Protocols. Springer, pp 49\u0026ndash;59\u003c/li\u003e\n\u003cli\u003eNewman GR, Hobot JA (1999) Resins for combined light and electron microscopy: a half century of development. The Histochemical Journal 31 (8):495\u0026ndash;505\u003c/li\u003e\n\u003cli\u003eNing L, Wang Y, Shi X, Zhou L, Ge M, Liang S, Wu Y, Zhang T, Zhao H (2023) Nitrogen-dependent binding of the transcription factor PBF1 contributes to the balance of protein and carbohydrate storage in maize endosperm. The Plant Cell 35 (1):409\u0026ndash;434\u003c/li\u003e\n\u003cli\u003eOlsen O-A (2004) Nuclear endosperm development in cereals and Arabidopsis thaliana. The Plant Cell 16 (suppl_1):S214\u0026ndash;S227\u003c/li\u003e\n\u003cli\u003eRaj A, van den Bogaard P, Rifkin SA, van Oudenaarden A, Tyagi S (2008) Imaging individual mRNA molecules using multiple singly labeled probes. Nat Methods 5 (10):877\u0026ndash;879. doi:10.1038/nmeth.1253\u003c/li\u003e\n\u003cli\u003eRuzin SE (2024) Techniques in light microscopy. Oxford University Press, \u003c/li\u003e\n\u003cli\u003eSabelli PA, Larkins BA (2009) The development of endosperm in grasses. Plant Physiol 149 (1):14\u0026ndash;26. doi:10.1104/pp.108.129437\u003c/li\u003e\n\u003cli\u003eSano Y (1984) Differential regulation of waxy gene expression in rice endosperm. Theoretical and applied genetics 68 (5):467\u0026ndash;473\u003c/li\u003e\n\u003cli\u003eSt\u0026aring;hl PL, Salm\u0026eacute;n F, Vickovic S, Lundmark A, Navarro JF, Magnusson J, Giacomello S, Asp M, Westholm JO, Huss M (2016) Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science 353 (6294):78\u0026ndash;82\u003c/li\u003e\n\u003cli\u003eWang K, Hasjim J, Wu AC, Li E, Henry RJ, Gilbert RG (2015) Roles of GBSSI and SSIIa in determining amylose fine structure. Carbohydrate polymers 127:264\u0026ndash;274\u003c/li\u003e\n\u003cli\u003eWu MW, Liu J, Bai X, Chen WQ, Ren Y, Liu JL, Chen MM, Zhao H, Yao X, Zhang JD (2023) Transcription factors NAC20 and NAC26 interact with RPBF to activate albumin accumulations in rice endosperm. Plant Biotechnology Journal 21 (5):890\u003c/li\u003e\n\u003cli\u003eWu TY, M\u0026uuml;ller M, Gruissem W, Bhullar NK (2020) Genome Wide Analysis of the Transcriptional Profiles in Different Regions of the Developing Rice Grains. Rice (N Y) 13 (1):62. doi:10.1186/s12284-020-00421-4\u003c/li\u003e\n\u003cli\u003eYamamoto MP, Onodera Y, Touno SM, Takaiwa F (2006) Synergism between RPBF Dof and RISBZ1 bZIP activators in the regulation of rice seed expression genes. Plant physiology 141 (4):1694\u0026ndash;1707\u003c/li\u003e\n\u003cli\u003eYi J, Moon S, Lee YS, Zhu L, Liang W, Zhang D, Jung KH, An G (2016) Defective Tapetum Cell Death 1 (DTC1) regulates ROS levels by binding to metallothionein during tapetum degeneration. Plant Physiol 170 (3):1611\u0026ndash;1623. doi:10.1104/pp.15.01561\u003c/li\u003e\n\u003cli\u003eZenklusen D, Singer RH (2010) Analyzing mRNA expression using single mRNA resolution fluorescent in situ hybridization. Methods in enzymology 470:641\u0026ndash;659. doi:10.1016/S0076-6879(10)70026-4\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"rice","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"rice","sideBox":"Learn more about [Rice](http://thericejournal.springeropen.com)","snPcode":"12284","submissionUrl":"https://submission.nature.com/new-submission/12284/3","title":"Rice","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Endosperm differentiation, Fluorescence in situ hybridization (FISH), Resin embedding, Rice seed development, Spatial gene expression, Dual-color FISH","lastPublishedDoi":"10.21203/rs.3.rs-9466150/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9466150/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eHigh-resolution spatial analysis of gene expression is essential for understanding developmental regulation in plant reproductive tissues. However, conventional paraffin embedding and chromogenic \u003cem\u003ein situ\u003c/em\u003e hybridization (ISH) are often distorted by poor tissue preservation, high background, and limited multiplexing capacity, particularly in starch-rich cereal tissues.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eIn this study, we established and evaluated an optimized workflow that combined acrylic resin embedding with fluorescence \u003cem\u003ein situ\u003c/em\u003e hybridization (FISH) for precise morphological, histochemical, and molecular analyses of rice (\u003cem\u003eOryza sativa\u003c/em\u003e) reproductive tissues. Resin-embedded sections preserved cellular architecture during prolonged ISH procedures and enabled high-resolution visualization of developing seed structures. Histochemical staining revealed dynamic redistribution of carbohydrates and proteins from maternal tissues to the endosperm during seed maturation. Both chromogenic ISH and FISH reliably detected the spatial expression patterns of key seed development genes, including \u003cem\u003eRice Prolamin Box Binding Factor\u003c/em\u003e (\u003cem\u003eRPBF\u003c/em\u003e) and \u003cem\u003eGranule-bound starch synthase I\u003c/em\u003e (\u003cem\u003eGBSS1\u003c/em\u003e). Notably, FISH provided superior signal clarity and reduced background compared with chromogenic ISH, particularly in reactive oxygen species\u0026ndash;rich tissues such as rice anthers. Furthermore, dual-color FISH enabled simultaneous detection of \u003cem\u003eRPBF\u003c/em\u003e and \u003cem\u003eGBSSI\u003c/em\u003e transcripts, revealing coordinated co-expression within the developing endosperm at cellular resolution.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThis work demonstrates that resin embedding combined with FISH is a robust and versatile platform for spatial gene expression analysis in cereal reproductive tissues and offers significant advantages for developmental and functional genomics studies.\u003c/p\u003e","manuscriptTitle":"Resin-embedded FISH Enables High-resolution Spatial Analysis of Gene Expression in Developing Rice Seeds","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-08 20:11:27","doi":"10.21203/rs.3.rs-9466150/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-13T01:26:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"266306074074038956809665917864589596955","date":"2026-05-02T03:08:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"333674166801385375799986493490310742654","date":"2026-04-30T03:12:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-30T03:03:53+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-20T13:15:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-20T13:15:22+00:00","index":"","fulltext":""},{"type":"submitted","content":"Rice","date":"2026-04-20T03:10:05+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"rice","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"rice","sideBox":"Learn more about [Rice](http://thericejournal.springeropen.com)","snPcode":"12284","submissionUrl":"https://submission.nature.com/new-submission/12284/3","title":"Rice","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6aaeb6ac-743b-4153-9714-3445990f833f","owner":[],"postedDate":"May 8th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-13T01:26:37+00:00","index":10,"fulltext":""},{"type":"reviewerAgreed","content":"266306074074038956809665917864589596955","date":"2026-05-02T03:08:33+00:00","index":9,"fulltext":""},{"type":"reviewerAgreed","content":"333674166801385375799986493490310742654","date":"2026-04-30T03:12:11+00:00","index":8,"fulltext":""},{"type":"reviewersInvited","content":"2","date":"2026-04-30T03:03:53+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-08T20:11:27+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-08 20:11:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9466150","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9466150","identity":"rs-9466150","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}