{"paper_id":"1f7059eb-5170-4ee8-aff9-7dc7ce26ce43","body_text":"A facile method for fluorescent visualization of newly synthesized fibrous collagen by \ncapturing the allysine aldehyde groups serving as cross-link precursors  \n \nJunpei Kuroda1,2,3*, Kazunori K. Fujii 4, Sugiko Futaki 5, Azumi Hirata 5, Yuki Taga6, Takaki \nKoide4,7. \n \n1 Laboratory of Morphogenesis, JT Biohistory Research Hall, Takatsuki, Osaka, Japan. \n2 Graduate School of Science, Osaka University, Toyonaka, Osaka,Japan. \n3 Graduate School of Frontier Bioscience, Osaka University, Suita, Osaka,Japan. \n4 Department of Chemistry and Biochemistry, Waseda University, Shinjuku-ku, Tokyo, Japan \n5 Department of Anatomy and Cell Biology, Osaka Medical and Pharmaceutical University, \nTakatsuki, Osaka, Japan. \n6 Nippi Research Institute of Biomatrix, Toride, Ibaraki, Japan \n7 Waseda Research Institute for Science and Engineering, Waseda University, Shinjuku-ku, \nTokyo, Japan. \n \n*Corresponding author.  \nEmail: junpei.kuroda@brh.co.jp   \n \n \nABSTRACT  \n \nThe fibrous structures of collagen provide physical strength and stability to tissues and \norgans. Abnormalities in their orientation, growth, and remodeling cause morphogenetic \ndefects and serious diseases including fibrosis, so it is important to clarify how collagen \nfibers are correctly oriented and grown within tissues. However, this mechanism remains \nelusive, as few methods have been available to fluorescently stain collagen fibers with a \nsimple protocol and to observe their structure in three dimensions. Here we present a facile \nmethod that enables fluorescent staining of collagen fibers in vertebrate tissues. In our \nmethod using DAF-FM, known as a NO detection probe, premature collagen fibers can be \nvisualized via covalent binding to the allysine residues serving as precursors of cross-linking \nstructures of collagen. In addition, we showed that the labeling method  using two \nfluorescent probes with different colors, DAF-FM and DAR-4M, allows for pulse-chase \nobservation of newly synthesized collagen fibers. Our method will be a breakthrough \ntechnique in future collagen studies. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nINTRODUCION \n \nThe optimal function and physical properties of tissues are highly dependent on the \norientation pattern of the supramolecular structure of extracellular matrix (ECM) and its \ndensity (1, 2). Collagen is known as a core ECM protein that forms fibers, and it not only \nprovides physical strength to the tissues but also serves as a scaffold for cells to maintain \ntissue function and homeostasis ( 1–6). In the morphogenesis of various tissues such as \nbone, tendon, skin, and cornea, regularly assembled fibrous collagen contributes to normal \nshaping as a building material for supporting tissues ( 2, 5). During the process of bone \nformation, abnormalities of collagen fibers in the orientation and in the density due to errors \nin formation of suprastructure lead to a bone disease, osteogenesis imperfecta ( 7, 8). In \naddition, abnormal production of collagen fibers and disruption of their remodeling \nprocesses trigger fibrosis in various tissues (9, 10). In order to clarify the mechanisms of the \npathogenesis of these diseases, it is necessary to understand the dynamics of collagen \nfibers distributed in tissues. Therefore, it is essential to develop techniques to easily \nvisualize collagen architecture in living tissues, as well as a new imaging tool to track the \ndynamics of collagen fiber growth. \n  To visualize and analyze collagen fibers, several techniques have been developed. \nAmong these, fluorescent staining and labeling are powerful tools to obtain information \nabout the shape and spatial distribution of collagen fibers within tissues. Immunofluorescent \nstaining is a common method to observe the distribution of these fibers in various studies \n(11–13). Despite recent advancements in tissue transparency technologies that facilitate \ndeep tissue immunofluorescent staining, achieving high-clarity images with minimal \nbackground noise and nonspecific signals remains a significant challenge. SHG imaging \nwith multiphoton microscopy is a noninvasive technique to visualize collagen fibers in \nvarious tissues without fixation or staining ( 14, 15). However, this method is not suitable for \nobserving at early morphogenesis stages such as embryogenesis, as it is difficult to detect \nsufficient second-harmonic generation (SHG) signal in immature fine fibers. Alternatively, \nrecent studies have achieved in vivo fluorescence imaging of fibrillar coll agen using GFP \nlabeling (16–19). This technique is effective for live imaging, similar to SHG imaging, and it \nalso allows the visualization of fine fibers that are hard to see with SHG imaging. Despite \nthese advantages, this observation system requires a great deal of effort to introduce gene \nexpression constructs and is only applicable to visualization of fibrillar collagen composed of \ntype I collagen. In addition to this, there is concern that in many cases, labeling of collagen \nmolecules with fluorescent proteins such as GFP may interfere with normal folding of \ncollagen molecules and normal fibrogenesis of collagen. Recently, it has been reported that \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nthe networks of collagen fibers in various tissues are clearly stained with a synthetic colorant, \nFast Green FCF ( 20). Although this method with low molecular weight chemicals allows \nvisualization of collagen fibers in very deep regions of tissues, specific staining is prevented \nunder hydrophilic conditions, and furthermore, it requires many steps to prepare the sample \nfor staining. Furthermore, new visualization techniques targeting the precursors of \ncross-linking structures of collagen have recently been reported ( 21–25), but there is still no \nrapid and easy method for fluorescent staining of collagen fibers in various vertebrate \ntissues. In addition, the development of pulse-chase analysis, which allows for the \nfluorescent labeling of collagen fibers in living tissues and the subsequent tracking of their \ngrowth process, could be a breakthrough imaging tool to understand collagen dynamics \nduring tissue growth. \n  Here, we provide a novel fluorescence imaging method for collagen fibers using \ndiaminofluorescein-FM (DAF-FM), which is widely used as a detection probe for nitric oxide \n(NO) produced by cells, and its analogs ( 26–28). We focused on the previous studies that \nreported DAF-FM DA (diacetate) fluorescently stains the notochord, cartilage, bone and \nactinotrichia, which are structures of collagen fibers oriented at the tip of zebrafish fin \n(29–33). Recently, Ohashi et al. reported that this probe fluorescently labeled collagen fibers \nin the axolotl skin ( 34). Based on the results of these studies, we hypothesized that this \nprobe could interact with collagen fibers and fluorescently stain them, apart from NO. In this \nstudy, we found that DAF-FM fluorescently labels the collagen fibers by a rapid and simple \nmethod while the mouse fibroblasts are alive in vitro . We also showed that pulse-chase \nanalysis of collagen fibers formed in cell culture condition can be performed by labeling \nthem with DAF-FM and another probe, diaminorhodamine-4M (DAR-4M), at different time \npoints. Using the collagen fibers produced by culture cells, our mass spectrometric analysis \nsuccessfully revealed that DAF-FM reacted to the aldehyde groups of allysine serving as the \nprecursors of interchain cross-link.  Furthermore, we found that collagen fibers, which are \noriented deep within mouse tissues, can be clearly and three-dimensionally stained using \nDAF-FM DA.  In addition, we also found that  in vivo  imaging of the collagen fibers using \nDAF-FM DA can be performed in other vertebrates including zebrafish and axolotl.  Finally, \nwe demonstrated that the pulse-chase observation of collagen fibers with these two \nfluorescent probes can be applied to in vivo imaging of tissues such as zebrafish bones. Our \nmethod is an innovative imaging technique that targets the cross-linking precursors of \ncollagen fibers, and its rapid and easy method will become a standard tool in future collagen \nresearch. \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \n \nRESULTS \n \nDAF-FM and DAR-4M enable fluorescent labeling of collagen fibers formed by \ncultured cells \n \nPrevious studies have reported that DAF-FM DA, a known NO detection probe, is \neffective for fluorescent staining of collagen fiber-rich tissues such as notochord, cartilage, \nbone, and skin ( 29–34).\n Based on the results of these studies, we hypothesized that this \nprobe could fluorescently stain collagen fibers independently of NO. In this research, we first \ndecided to test this possibility by culturing mouse fibroblasts, which have a high activity for \nthe production of collagen fibers. In this experiment, we used DAF-FM, a membrane \nimpermeable version of DAF-FM DA. Before starting the culture experiment, we coated \nculture dishes with purified collagen to check if these substrates would be fluorescently \nstained by DAF-FM. We then confirmed that DAF-FM reacts little with purified collagen used \nas culture substrates (fig. S1). Therefore, we seeded mouse embryonic fibroblasts (MEFs) \non gelatin-coated culture dishes and stained them with DAF-FM while the cells were alive \nafter culture (Fig. 1A). As a result, we found that the network of fiber structures formed by \nMEFs was fluorescently labeled by DAF-FM (Fig. 1B). To determine whether these fibrous \nstructures were collagen fibers, we conducted double staining with DAF-FM and BindCOL, a \ncyclic peptide that hybridizes to denatured portions of collagen ( 35). Importantly, fibers \nlabeled by DAF-FM were also simultaneously labeled by Bin dCOL after denaturation by \nheat treatment, indicating that DAF-FM fluorescently stains collagen fibers formed by the \ncultured fibroblasts (Fig. 1B). We also found that when co-stained with DAF-FM and its \nderivative, DAR-4M, the two probes fluorescently label the same fiber structures (fig. S2). \nWe also performed double staining with DAR-4M using an antibody to type I collagen (Col1) \nand found that the DAR-4M staining merged very well with the Col1 antibody staining (fig. \nS2). These results indicate that DAF-FM and DAR-4M can fluorescently label collagen fibers \nproduced by fibroblasts under culture conditions using a simple method.  \nWe next asked whether the fluorescence of collagen fibers observed after DAF-FM \nstaining was due to the action of NO produced by the cells. Therefore, to clarify the \ninvolvement of NO, we treated MEFs under the culture conditions to remove NO and \nobserved the fluorescence of fibers after DAF-FM staining. As a result, no significant \ndifference was observed in the intensity of the fluorescent signal from the fibers under NO \nremoval conditions compared to control (fig. S3), indicating that NO is not involved in the \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nfluorescence emission of collagen fibers by DAF-FM.   \nWe also investigated the effect of DAF-FM staining on the cell activities. We treated MEFs \nin the culture condition with DAF-FM and then performed a BrdU incorporation assay to \nassess the activity of cell proliferation (fig. S4). As a result, there was no significant \ndifference in BrdU incorporation between cells treated with DAF-FM and untreated cells (fig. \nS4). In addition, to determine the effect of DAF-FM treatment on cell survival, we checked \nwhether apoptosis was enhanced after DAF-FM treatment. We used nuclear staining \nreagents to detect apoptosis and f ound no significant change in cell death between cells \ntreated with DAF-FM and untreated cells (fig. S4). Based on these results, we conclude that \nDAF-FM does not adversely affect cell activities, at least at low concentrations of treatment, \nand that this probe allows the visualization of collagen fibers produced by living cells \nindependently of NO. \n \n \nFig. 1. A rapid and simple method using DAF-FM for visualization of the collagen \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nfibers formed by culture cells. (A) Schematic diagram of DAF-FM staining for the collagen \nfibers formed by MEFs. Fully proliferated MEFs after 1 to 2 weeks of culture were incubated \nin 10μ M DAF-FM solution for 1hr. After the incubation, DAF-FM solution was replaced with \nfresh medium and the collagen fibers visualized by DAF-FM were observed. ( B) \nRepresentative fluorescent images of the collagen fibers labeled by DAF-FM (green) at \nculture day 10. Actin cytoskeleton was stained with Phalloidin (magenta), and nucleus was \nstained with Hoechst (blue). ( C) Representative fluorescent images of the collagen fibers \nco-stained with DAF-FM (green) and BindCOL (blue). Scale bar = 50 \nμ m.  \n \n \n \n \nfig. S1. Reactivity of DAF-FM with purified collagen. (A) Schematic diagram of DAF-FM \nstaining for the purified collagen substrates. ( B) Representative confocal images of the \npurified collagen after DAF-FM staining. Scale bar = 50 μ m. \n \n \n \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \nfig. S2. Fluorescent staining of the collagen fibers formed by culture cells using \nDAR-4M. (A) Schematic diagram of DAR-4M staining for the collagen fibers formed by \nMEFs. (B) Upper panels show representative fluorescent images of the collagen fibers \nvisualized with DAF-FM (green) and DAR-4M (magenta) at culture day 10. Both probes \nvisualized the same fibers. Lower panels show representative fluorescent images of the \ncollagen fibers visualized with DAR-4M (magenta) and anti-Col1 antibody staining at culture \nday 10. The fluorescent signals of DAR-4M and anti-Col1 antibody staining were merged \nwell on the same fibers. Scale bar = 50 \nμ m. \n \n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \n \n \n \n \nfig. S3. NO-independent fluorescent visualization of collagen fibers by DAF-FM. (A) \nSchematic diagram of DAF-FM staining under NO removal conditions for the collagen fibers \nformed by MEFs.  (B) Representative fluorescent images of the collagen fibers visualized \nwith DAF-FM (green) at culture day 10. Clear fluorescent signals of collagen fibers stained \nwith DAF-FM were detected under C-PTIO (NO removal) and L-NAME (NOS inhibition) \ntreatment conditions, similar to the control. The fluorescence intensity plots for each \ncondition are shown in the right panel. Scale bar = 50 \nμ m. \n \n \n \n \n \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \nfig. S4. No negative effects on cell activities with DAF-FM treatment. (A) Experimental \nworkflow to investigate the effect of DAF-FM treatment on cell division. ( B) Representative \nimages of anti-BrdU antibody staining. Cultured MEFs were incubated with DMSO (control) \nor DAF-FM solution. All cell nuclei were stained with Hoechst (blue), and cell nuclei \nincorporating BrdU were stained with anti-BrdU antibody (magenta). Collagen fibers were \nstained with DAF-FM (green). ( C) Number of BrdU-positive cells was counted under each \ncondition. There was no significant difference in the percentage of BrdU-positive cells \nbetween the conditions. ( D) Experimental workflow to investigate the effect of DAF-FM \ntreatment on cell death. (E) Representative images of Nuclear Blue staining. Cultured MEFs \nwere incubated with DMSO (control) or DAF-FM solution. All cell nuclei were stained with \nSyto 82 (orange), and cell nuclei of apoptotic cells were stained with Nuclear Blue (blue). \nCollagen fibers were stained with DAF-FM (green). ( F) Number of apoptotic cells was \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\ncounted under each condition. There was no significant difference in the percentage of \napoptotic cells between the conditions. Scale bar = 50 μ m. \n \nFluorescent labeling of collagen fibers with DAF-FM via interaction with cross-linking \nintermediate structures \n \nTo clarify the affinity and specificity of DAF-FM for collagen fibers, we attempted to identify \nthe mechanism by which this fluorescent probe labels collagen fibers. DAF-FM is widely \nused as a probe that reacts with NO to emit fluorescence ( 36), but the result in fig. S3 \nindicates that NO is not involved in the fluorescent labeling of collagen fibers by DAF-FM. \nHere we note that DAF-FM has recently been reported to emit green fluorescence by \nbinding to aldehydes as well as NO ( 37). In addition, it is known that \nε -amino groups of \nlysine residues in the N- and C-terminal non-triple helical domains (telopeptides) of collagen \nα -chain are modified to aldehyde groups by lysyl oxidase (LOX), and collagen molecules \nundergo intra- and intermolecular cross-linking via these aldehyde groups, resulting in the \npolymerization and growth of collagen fibers ( 38–40). Based on these facts, DAF-FM was \nsuggested to fluorescently label collagen fibers by interacting with aldehydes, which are \nintermediate structures in collagen cross-linking. Therefore, we examined the fluorescence \nafter DAF-FM staining under conditions that inhibit the formation of collagen cross-linking. \nWe treated MEFs in culture condition with \n/i4 -aminopropionitrile (BAPN), known as an \ninhibitor of LOX ( 41), in this experiment (Fig. 2A). As a result, inhibition of collagen \ncross-linking formation by treatment with BAPN suppressed fluorescent labeling of fibers by \nDAF-FM (Fig. 2B). On the other hand, when MEFs were cultured under normal conditions \nand then stained with DAF-FM in the presence of BAPN, clear fluorescence of collagen \nfibers was observed, similar to the control (fig. S5A). These results suggest that the \nfluorescent labeling of collagen fibers by DAF-FM is closely related to the modification of \ncollagen by LOX.  \nWe next tried to determine the specific target sites of collagen cross-linking structure that \nDAF-FM interacts with. First, proteins were extracted from MEF culture dishes treated with \nDAF-FM and SDS-PAGE was performed under reducing condition using DTT to examine \nwhether DAF-FM fluorescence was retained (Fig. 2C). As a result, when MEFs were \ncultured in the absence of BAPN and stained with DAF-FM, specific bands were detected in \nthe lysate-derived sample at the positions of molecular weights predicted to be \nα 1(/i4 ) and \nα 2(/i4 ) chains (Fig. 2D). On the other hand, these specific bands were not detected in the \nlysate-derived sample when the same experiment was performed in the presence of BAPN \n(Fig. 2D, fig. S5B). We also tested whether the same bands could be detected by Bin dCOL \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nafter the same experiment. In the lysate-derived sample, specific bands labeled with \nBindCOL were detected at the positions of molecular weights predicted to be α 1(/i4 ) and \nα 2(/i4 ) (Fig. 2D, fig. S5B). These experimental results indicate that under culture conditions \nin the presence of BAPN, production of collagen fibers itself occurs normally, but the \ninhibition of cross-linking suppresses fluorescence caused by DAF-FM. Furthermore, the \nfluorescence caused by DAF-FM in the two specific bands did not disappear after pepsin \ntreatment, indicating that DAF-FM specifically reacts with collagen molecules that are \nprotease-resistant due to their triple helical structure (fig. S5C). In addition, our results also \nsuggest that DAF-FM fluoresce through covalent binding to specific structures that occur \nduring the cross-linking formation process. \nWe further attempted to identify the binding site of DAF-FM by LC-MS analysis (Fig. 2C). \nIn this experiment, we analyzed three collagenase/pepsin-digested peptides derived from \ntelopeptide domains of type I collagen (\nα 1-N telopeptide, α 1-C telopeptide and α 2-N \ntelopeptide) which contain lysine or hydroxylysine participating in cross-link formation. The \ntelopeptidyl lysine residues are not hydroxylated in skin ( 42), while they are largely \nconverted to hydroxylysine in bone ( 43). Thus, we analyzed lysine-containing telopeptides \nfor the MEF samples. MS/MS sequence analysis confirmed that DAF-FM binds to the \naldehyde groups of lysine residues within the α 2-N telopeptide in the DAF-FM-labeled \nsample (Fig. 2E and F). Extracted ion chromatograms showed that a strong peak of this \ntelopeptide labeled with DAF-FM was detected only in the DAF-FM-labeled sample (Fig. \n2G). Extracted ion chromatogram peaks corresponding to the theoretical molecular weight \nof \nα 1-N and α 1-C telopeptides attached with DAF-FM were only slightly detected in the \nDAF-FM-labeled sample, and MS/MS sequence confirmation could not be performed (data \nnot shown). The mass shift observed in the MS/MS spectra suggests that DAF-FM reacted \nwith aldehyde groups to form fluorophores through the same mechanism as previously \nreported (37). The predicted fluorophore is generated by the reaction of DAF-FM with the \naldehyde group of the allysine residue (Fig. 2H). Based on these results, we conclude that \nDAF-FM fluorescently labels collagen fibers by covalently binding to the aldehyde groups of \nallysine (and possibly hydroxylysine) residues in the telopeptide, intermediate structures in \ncollagen cross-linking.  \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \n \nFig. 2. Mechanism of DAF-FM fluorescence targeting collagen cross-linking. (A) \nSchematic diagram of DAF-FM staining for the collagen fibers formed by MEFs under the \nculture condition with BAPN. ( B) Representative fluorescent images of the collagen fibers \nlabeled by DAF-FM (green) under the culture condition without BAPN and with BAPN. Scale \nbar = 50 \nμ m. (C) Experimental workflow for sample preparation for SDS-PAGE and LC-MS \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nanalysis. ( D) SDS-PAGE analysis of DAF-FM-labeled collagen. DAF-FM fluorescence of \nproteins derived from the lysate and medium at each condition was examined (upper panel), \nand BindCOL staining of the same protein samples at each condition was performed (lower \npanel). ( E) MS/MS spectra of α 2-N telopeptide containing lysine ( m/z 696.8085, z = 2) \nderived from control (non-labeled) samples. ( F) MS/MS spectra of α 2-N telopeptide \ncontaining DAF-FM-labeled allysine ( m/z 892.3239, z = 2) derived from DAF-FM-labeled \nsamples. ( G) Monoisotopic extracted ion chromatograms of the DAF-FM-labeled α 2-N \ntelopeptide (m/z 892.3352 ± 0.02, z = 2) for the control (blue) and DAF-FM-labeled sample \n(red). (H) Proposed mechanism of collagen fiber visualization with DAF-FM.  \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \nfig. S5. DAF-FM fluorescence suppressed by inhibition of collagen cross-linking \nformation (A) Representative fluorescent images of the collagen fibers labeled by DAF-FM \n(green) under the various culture conditions. DAF-FM fluorescence of the collagen fibers \nproduced by MEFs is not suppressed when BAPN is present only during DAF-FM staining, \nbut it is significantly suppressed when BAPN is present during MEF culture. Scale bar = 50 \nμ m. ( B) SDS-PAGE analysis of DAF-FM-labeled collagen with or without DTT. DAF-FM \nfluorescence of proteins produced by MEFs in the presence or absence of BAPN was \nexamined (upper panel), and BindCOL staining of the same protein samples was performed \n(lower panel). ( C) DAF-FM fluorescence of pepsin-digested or undigested proteins \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nproduced by MEFs was examined (upper panel), and CBB staining of the same protein \nsamples was performed (lower panel).  \n \n \nFluorescent visualization of collagen fibers with DAF-FM DA in cartilage and \nnotochord of mouse embryos \n \nUsing mouse culture cells, we successfully demonstrated that DAF-FM fluorescently labels \ncollagen fibers through a mechanism distinct from that of NO detection. We then tried to \nDAF-FM staining using mouse tissues to determine if this probe is also useful for fluorescent \nimaging of collagen fibers  in vivo . In previous studies, DAF-FM DA, which exhibits plasma \nmembrane permeability ( 27), was used for whole-body fluorescence staining of zebrafish \n(29–32). Therefore, we assumed that DAF-FM DA has high tissue penetration and decided \nto use this probe for fluorescent staining of mouse tissues. We first examined whether two \ncollagen-rich tissues, cartilage and notochord, could be fluorescently stained with DAF-FM \nDA in mouse embryos. Unfixed embryos at embryonic day (E) 14.5 were incubated \novernight in a diluted DAF-FM DA solution (Fig. 3A and B). After incubation, tails of embryos \nwere cleared to examine the three-dimensional orientation of collagen fibers in the cartilage \nand notochord. Confocal microscopy imaging revealed that strong green fluorescent signals \nwere detected in the cartilage primordium and notochord (Fig. 3C-C”). To determine whether \nthese DAF-FM DA signals were derived from collagen fibers, the sections of DAF-FM \nDA-labeled embryo were incubated with an anti-type II collagen (Col2) antibody. Reactivity \nof type II collagen, the major extracellular matrix component of the cartilage, was detected in \nthe area where DAF-FM DA fluorescence was observed (Fig. 3D), indicating that DAF-FM \nDA also fluorescently labels developing collagen fibers in mouse embryos. \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \nFig. 3. DAF-FM DA enables clear fluorescent visualization of the collagen fibers in \ncartilage and notochord of mouse embryos with a simple method.  (A) Schematic \ndiagram of DAF-FM DA staining for the collagen fibers in mouse embryos. (B) The \nfluorescent image of the E14.5 embryo after DAF-FM DA staining. (C) The 3D fluorescent \nimage of the embryonic tail stained with DAF-FM DA. Slice images of the area within the \nwhite dotted box are shown in (C’ and C’’). ca, cartilage; nc, notochord. (D) The fluorescent \nimages of antibody staining of the cryo-section samples of the tail. The tissue sections \nlabeled with DAF-FM DA (green) were stained with anti-Col2 antibody (magenta) and \nHoechst (blue). Scale bar = 2 mm (B) and 50 \nμ m (C to D).  \n \n \n \nFluorescent visualization of collagen fibers with DAF-FM DA in tendons and cartilage \nof postnatal mouse \n \nWe next investigated whether DAF-FM DA is useful in fluorescent visualization of collagen \nfibers in postnatal mice. To simultaneously observe cartilage and other collagen-rich tissues \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nsuch as tendons, the tails of new-born mice were stained with DAF-FM DA. After incubation \nin DAF-FM DA solution, tail samples were cleared with transparency reagents. As a result, \nstrong DAF-FM DA signals were observed in the tendon and in fibrocartilage of the \nintervertebral disc deep inside the tail (Fig. 4A). To simultaneously observe the distribution of \ncollagen fibers and surrounding cells, nuclear staining was performed in the tail samples \nlabelled with DAF-FM DA (Fig. 4B). Each XY cross-sectional image scanned by confocal \nmicroscopy displayed clearly labeled fiber structures in the tendon and intervertebral \nfibrocartilage (Fig. 4C). We observed that the nuclei were positioned in the gaps between \nfibers (Fig. 4C). We further confirmed that the fluorescence of DAF-FM DA overlapped well \nwith the SHG signal in tissue sections of intervertebral discs (Fig. 4D). The orientation \npattern of the fibers labelled with each fluorescence are approximately consistent (Fig. 4D). \nThe signal intensity of SHG was not uniform and differed for each fiber, while the DAF-FM \nDA signal was detected uniformly throughout the fiber (Fig. 4D). In conclusion, we \ndemonstrated that DAF-FM DA can be used to observe collagen fibers in postnatal mice \ntissues with high resolution enough to identify individual fibers by an easy method. \n \n \n \n \n \n \n \n \n \n \n \n \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \nFig. 4. Fluorescent visualization of the collagen fibers in tendon and cartilage of \npostnatal mouse using DAF-FM DA. ( A) The confocal fluorescent image with depth \ncolor-coded MIP of the tendon and cartilage in postnatal mouse tail stained with DAF-FM \nDA. (B) The 3D reconstructed confocal image in the area within the white dotted box of (A). \nThe collagen fibers were stained with DAF-FM DA (green), and all nuclei were stained with \nSyto 82 (orange). ( C) XY sectional views at an intervertebral region. te, tendon; fc, \nfibrocartilage. White dotted lines indicate the fibrocartilage orientation. ( D) The fluorescent \nimages of cryo-section samples at the tail intervertebral disc region. Orientation and \nfluorescent intensity of the collagen fibers visualized with DAF-FM DA (green) and SHG \n(blue) were plotted, respectively. Scale bar = 100 \nμ m (A to C) and 40 μ m (D). \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nFluorescent visualization of collagen fibers in aquatic vertebrates using DAF-FM DA \n \nTo further demonstrate the utility of collagen fiber imaging using DAF-FM DA, we \nattempted DAF-FM DA staining in tissues of other vertebrate animals. Recently, Kuroda et al \nreported that DAF-FM DA fluorescently labels actinotrichia oriented in the tip of zebrafish \nfins (32). More recently, Ohashi et al reported that this probe fluorescently labels the \ncollagen fiber network that develops three-dimensionally in the dermis of the axolotl ( 34). \nWe tested whether this probe labels collagen fibers in other collagen-rich tissues by staining \nthe whole bodies of these animals. In this experiment, living juvenile animals of zebrafish \nand axolotl were incubated in DAF-FM DA solution (fig. S6A and B). Firstly, we confirmed \nthat DAF-FM DA fluorescently visualized the lattice-like collagen fiber structures that \ndevelop in the dermis of juvenile zebrafish and axolotl, similar to the recent report (fig. S6C \nand D). Additionally, we found that tendons at the myoseptum junction were visualized with \nstrong fluorescence (fig. S6C). And also, clear fluorescence was observed in the tendons \nthat develop in the joint regions of each fin bone in zebrafish (fig. S6C). Furthermore, \ntendons, as well as ligaments, in the digits of axolotl were visualized by DAF-FM DA with \ndistinct signals (fig. S6F). These results demonstrate that DAF-FM DA is applicable for \nwhole-body staining of collagen fibers in the tissues of living aquatic vertebrates. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \nfig. S6. Fluorescent visualization of collagen fibers  in zebrafish and axolotl  using \nDAF-FM DA. ( A, B) Schematic diagram of DAF-FM DA staining for the collagen fibers in \nzebrafish and axolotl. Living juvenile zebrafish and axolotl were incubated overnight in 10μ M \nDAF-FM solution. After the incubation, DAF-FM DA fluorescence in the animal tissues were \nimaged by a confocal microscopy. ( C, D ) Representative fluorescent image with depth \ncolor-coded MIP of the body skin labeled by DAF-FM DA in zebrafish and axolotl, \nrespectively. The magnified MIPs of the areas within the white boxes are shown in the right \npanels of each image. White dotted lines indicate the orientation of collagen fibers. ( E) \nRepresentative fluorescent images of the tendon labeled by DAF-FM (green) in the \nzebrafish fin bones. Fin bones were stained with Alizarin Red (magenta). (F) Representative \nfluorescent images of the tendon and ligament labeled by DAF-FM DA (green) in the axolotl \nforelimb digits. te, tendon; li, ligament. Scale bar = 50 \nμ m (B, C and F) and 200 μ m (E). \n \n \nPulse-chase observation using two-color fluorescent probes to analyze the growth \ndynamics of collagen fiber in vitro and in vivo \n \nWe finally tested pulse-chase observation, which tracks the growth process of collagen \nfibers, using two fluorescent probes with different colors, DAF-FM and DAR-4M. For this \nexperiment, it is essential that the fluorescence of the labeled fibers remains stable over an \nextended period without fading. Therefore, we fluorescently labeled collagen fibers with \nDAF-FM only once during MEF culture and then examined whether the fluorescence faded \nby temporal observations in the same area (fig. S7A). As a result, the fluorescence of \ncollagen fibers labeled with DAF-FM hardly faded during the culture process after staining \n(fig. S7B). We then attempted the experiment to color-code collagen fibers during their \ngrowth by using a combination of the two fluorescent probes at different times during the \nculture process (Fig. 5A). As a result, we successfully distinguished between old fibers, \nformed four days ago, and newly formed fibers by performing DAF-FM and DAR-4M staining \nat different time points (Fig. 5B). Our result demonstrated the growth of a network of \ncollagen fibers through the process of thickening by the deposition of new collagen around \nold fibers (Fig. 5B’  asterisk) and the formation of new fibers near old fibers (Fig. 5B’  \narrowheads).  \nWe further attempted to perform the pulse-chase observation of collagen fibers in living \nanimal tissues. Here, we focused on the development of vertebrae in zebrafish. We first \nevaluated the reactivity of the two probes to the collagen fibers distributed in the notochord \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nat the 7 days post fertilization (dpf) stage, before the start of vertebral calcification (fig. S8A). \nAt the 7 dpf stage, strong fluorescence was detected in the notochord after DAF-FM DA \nstaining, but only weak fluorescence was detected after DAR-4M AM (acetoxymethyl ester) \nstaining (fig. S8A). On the other hand, at the stage of advanced calcification around the \nnotochord associated with vertebral bone formation, strong fluorescence was detected in \nthe intervertebral regions and in the neural and hemal spines after staining with both probes \n(fig. S8B). However, at this stage, no clear fluorescence was observed in the centrum \nregions with either probe (fig. S8B). We found that the fluorescence signals in the \nintervertebral regions and spines emitted with DAF-FM DA were similarly detected with \ncollagen hybridizing peptide (CHP),\n a peptide that binds to the denatured collagen chains \n(44), staining after denaturation by heat treatment (fig. S8C). To clarify the dynamics in the \ndistribution and production of collagen fibers during early osteogenesis in the vertebral \nregion, we first incubated zebrafish larvae at the 7 dpf stage in DAF-FM DA solution and \nlabeled the collagen fibers distributed at the notochord with green fluorescence (Fig. 5C). \nAfter staining with DAF-FM DA, the fish were bred in a circulating system tank for two weeks. \nSubsequently, the fish were incubated in DAR-4M AM solution (Fig. 5C). As a result, \ncollagen fibers distributed at the notochord, initially labeled with DAF-FM DA, exhibited a \nremarkable change in their distribution pattern during 2 weeks of growth (Fig. 5D). The \nfluorescence of collagen fibers uniformly labeled at the notochord before bone formation \nemitted a strong signal specifically in the intervertebral disc region during osteogenesis (Fig. \n5D). This fact indicates that collagen fibers uniformly distributed around the notochord at \nlarval stage are reorganized during bone formation and reused as structurally specific \ncomponents of the intervertebral disc. On the other hand, collagen fibers labeled with \nDAR-4M AM showed more extensive fluorescence in the intervertebral region compared to \nthose with DAF-FM DA (Fig. 5D). Furthermore, no obvious signal was observed in the \ncentrum region, whereas strong fluorescence labeled with DAR-4M AM was observed in the \nneural and hemal spines (Fig. 5D). This result suggests that in the early stages of vertebra \nformation, collagen production is more active in the spines and intervertebral regions than in \nthe vertebral bodies. In summary, we successfully tracked the changes of collagen \ndistribution pattern and visualized the active areas of collagen production in living zebrafish \nby applying the two probes at different times during the osteogenesis process. Hence, we \ndemonstrated that DAF-FM DA and DAR-4M AM are useful for pulse-chase observation of \ncollagen fibers during the growth of living tissues. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \n \nFig. 5. Pulse-chase observation using two different fluorescent probes to understand \ngrowth manner of collagen fibers in vitro and in vivo. (A) Schematic diagram illustrating \nthe pulse-chase observation of the collagen fibers formed by MEFs, using DAF-FM and \nDAR-4M. MEF cultured for 6 days were first stained with DAF-FM. After DAF-FM staining, \nthe staining solution was replaced with fresh medium and MEFs were incubated for 4 days. \nSubsequently, DAR-4M staining was performed, and the fluorescence of collagen fibers was \nobserved. (B) Representative fluorescent images of the collagen fibers labeled by DAF-FM \n(green, labeled 4 days ago) and DAR-4M (magenta, newly labeled). Magnified images in the \nwhite box area and cross section images at the position of the white dotted line are shown in \nthe right panels. Asterisks indicate the fibers that have thickened due to the additional \ngrowth of older fibers, and arrowheads indicate newly formed fibers between old fibers. ( C) \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nSchematic diagram illustrating the pulse-chase observation of the collagen fibers during the \nformation of vertebral bones in zebrafish, using DAF-FM DA and DAR-4M AM. Living \nzebrafish larvae at 7 dpf were first stained with DAF-FM DA. After the staining, the fish were \nreturned to fresh tank water and bred for two weeks. Subsequently, DAR-4M AM staining \nwas performed, and the fluorescence of the collagen fibers around the vertebrae was \nobserved. (D) Representative fluorescent images of the collagen fibers around the vertebrae \nlabeled by DAF-FM DA (green, labeled 14 days ago) and DAR-4M AM (magenta, newly \nlabeled). Cross section images at the position of the white dotted lines in each fluorescent \nimage are shown in the lower panels. Asterisks indicate the intervertebral regions. \nArrowheads and arrows indicate neural spines and hemal spines, respectively. Scale bar = \n10 \nμ m (B), 5 μ m (B’) and 50 μ m (D). \n \n \n \nfig. S7. Fluorescence of the collagen fibers labeled with DAF-FM hardly fade after \nwashout. (A) Schematic diagram of the temporal observation after DAF-FM staining for the \ncollagen fibers formed by MEFs. ( B) Representative fluorescent images of the collagen \nfibers labeled with DAF-FM at day 0, day 2 and day4 after labeling. Fluorescent mean \nintensity values of the DAF-FM at each time point are shown in the right panel. Scale bar = \n100 \nμ m. \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \nfig. S8. Reactivity of DAF-FM DA and DAR-4M AM with zebrafish notochord and \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nvertebral bones. ( A) Representative confocal images of the collagen fibers in the \nnotochord at 7 dpf simultaneously labeled with DAF-FM DA and DAR-4M AM. Each image \nwith fluorescence intensity color-coded MIPs is shown in the lower panels. ( B) \nRepresentative confocal images of the collagen fibers in the intervertebral regions and \nspines at 20 dpf simultaneously labeled with DAF-FM DA and DAR-4M AM. Each image with \nfluorescence intensity color-coded MIPs is shown in the lower panels. ( C) Representative \nconfocal images of the collagen fibers in the intervertebral regions and spines at 24 dpf \nco-labeled with DAF-FM DA and CHP-Cy3. Cross section images at the position of the white \ndotted lines in each fluorescent image are shown in the lower panels. nt, notochord; ns, \nneural spine; hs, hemal spine. Open arrowheads indicate the myoseptum tendon, arrows \nindicate unknown nonspecific signals detected in DAR-4M AM labeled samples, and \nasterisks indicate the intervertebral discs. Scale bar = 50 \nμ m. \n \n \nDISCUSSION  \n \nIn general, collagen fibers are distributed deep within animal tissues and form a \nthree-dimensional network, making it difficult to stain them using whole mount tissues.  \nFurthermore, due to the complexity of collagen structures, it is difficult to directly label \ncollagen with fluorescent molecules, and many issues remain for imaging of collagen fibers \nusing current technology. To overcome these issues, here we used cultured cells and \nvertebrate tissues to establish a method for visualizing collagen fibers with small-molecule \nfluorescent probes. One of the advantages of our imaging method is that it requires no \nspecial treatment prior to staining and the protocol is very simple. Even very thick tissues, \nsuch as the fibrocartilage in the intervertebral disc (Fig4), which is more than 300 \nμ m thick, \ncan be fluorescently stained clearly enough to distinguish the shape of individual fibers by \nsimply immersing them overnight in a 10 \nμ M DAF-FM DA dilute solution. In our method, \nobservation of collagen fibers distributed deep within thick tissues such as cartilage and \ntendons in mice requires immersing the samples in a clearing reagent overnight after \nstaining. However, DAF-FM emits via covalent bonding with allysine residues, so the \nfluorescence does not fade after transparency, making it a powerful imaging tool for \nthree-dimensional visualization of fiber networks inside thick tissues. In addition, our \ntechnology has several other powerful advantages for the analysis of collagen dynamics. \nFirst, treatment with DAF-FM solutions at low concentrations of 5-10 \nμ M does not exhibit \ncytotoxicity. Second, DAF-FM does not stain purified collagen or old collagen fibers that lack \naldehyde groups. Instead, it binds to aldehyde groups in allysine residues before \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\ncross-linking occurs after secretion, enabling the staining of various fibrous collagens. \nBecause DAF-FM covalently binds to the precursor of collagen cross-linking, the \nfluorescence of collagen fibers labeled with DAF-FM remains stable and hardly fade or \ndiffuse after washout. And third, by utilizing the above-mentioned reaction specificity and \napplying two fluorescent probes with different colors at separate time points, it is possible to \ndistinguish between old and newly formed collagen fibers that develop during their growth. \nAlthough several collagens labeled with fluorescent proteins have been developed for \nvisualization ( 16–19, 45, 46), it is questionable whether such modified collagen with \nrelatively large globular domains can fold and form fibrils normally. Additionally, this \ntechnique allows us to observe the distribution and morphology of collagen fibers only within \na specific time frame and does not provide information on how individual fibers grow. On the \nother hand, our fluorescent tag specifically labels normally secreted collagen, and the tag is \nsmall enough compared to the fluorescent proteins to minimize interference with the fibril \nformation process. Combining DAF-FM and DAR-4M for pulse-chase observation is \ncurrently the only technique to overcome this problem, and we expect that it will be a \nbreakthrough in future collagen research. \nWe revealed that DAF-FM and DAR-4M, known as detection probes for NO, label \ncollagen fibers fluorescently in a manner independent of NO (fig. S3). Collagen fibers \nundergo a characteristic growth process in which three \nα -chains combine to form a single \nunit and these trimers polymerize through cross-link formation ( 5, 40). We successfully \nidentified that DAF-FM covalently reacts to the aldehyde group of allysine residue, \nprecursors of collagen cross-linking, located in the telopeptide domain of type I collagen by \nLC-MS analysis (Fig. 2E-G). Several recent studies focusing on LOX-mediated cross-link \nformation have proposed novel methods for detecting collagen fibers ( 21–25). In these \nmethods, probes are immobilized by hydrazone or oxime formation targeting aldehyde \ngroups of allysine residues for specific detection of collagen. While these chemical bond \nformations are in principle reversible reactions, DAF-FM reacts with aldehyde groups \nthrough an annulation reaction (Fig. 2H), which is thought to prevent elimination reactions \neffectively and result in long lifetime of labeling. In fact, fluorescence from DAF-FM-labelled \ncollagen hardly faded for several days in vitro (fig. S7) and remained detectable for at least \nseveral days to weeks both in vitro  and in vivo (Fig. 5). Furthermore, the fluorescence of \nDAF-FM-labelled collagen could be observed even after SDS-PAGE (fig. S5C) and transfer \nto membranes (Fig. 2D, fig. S5B). Other advantages of using DAF-FM and DAR-4M include \nthat the probes are non-peptidic, low-molecular-weight compounds with high tissue \npermeability and that the two amine substituents on their benzene ring react with the targets \nto emit fluorescence, making them easy to use and minimizing interference from unreacted \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nprobes during fluorescence observation.  \nA very recent report has also shown that overnight treatment with low concentrations of \nDAF-FM DA does not adversely affect the growth of zebrafish tissues ( 32). However, if the \naldehyde groups of lysine residues in the telopeptides of collagen molecules are masked by \nDAF-FM and its derivatives by more strong treatment, such as longer-term incubation with \nhigher concentrations of the probes, the cross-link formation of collagen fibers might be \ninhibited. In order to perform in vivo  pulse-chase analysis of the collagen fibers in various \nanimal tissues in the future, it is necessary to verify the suitable experimental conditions that \ndo not inhibit cross-link formation.  \nSeveral types of collagen molecules, including types I, II, III, V and XI, are known to form \ncollagen fibers (4, 5, 47). All of these fibers polymerize and mature through a LOX-mediated \ncross-linking process (4, 5). In principle, DAF-FM and DAR-4M could label all these types of \ncollagen fibers, and it is not possible to distinguish which type of collagen fiber is labeled \nwith these probes. For the observation of tissues that simultaneously contain these types of \ncollagen fibers, it may be necessary to use specific antibodies or other markers in \ncombination to clarify which type of collagen fiber is stained. We found in this study that the \nreactivity of DAF-FM DA and DAR-4M AM, for the visualization of collagen fibers, varied \ndepending on the developmental stage of animals (fig. S8). We speculate that DAR-4M AM \nalso fluorescently labels collagen fibers via the same mechanism of action as DAF-FM DA, \nbut the cause for the low reactivity of DAR-4M AM at some developmental stages is \ncurrently unknown. In addition, we evaluated the reactivity of the probes only in a limited \nnumber of animal tissues in the current study, and future validation of the reactivity of \nDAR-4M, in particular, is needed using tissues from various animal species. We also found \nthat DAF-FM DA and DAR-4M AM do not uniformly label all collagen fibers in some tissues \n(Fig. 5, figS8). Considering the mechanism of action of these probes, it is possible that the \ncollagen fibers at specific regions of the tissues with high LOX activity or rapid turnover \ncould preferentially react with the probes and emit strong fluorescence, resulting in \nvariations in labeling by the probes. In other words, this property can be utilized in the future \nas a tool for measuring LOX activity or collagen turnover rate in some tissues.\n \nPrevious techniques have been developed as a strategy to target collagen molecules \nthemselves for the visualization of collagen fibers, but our method targets the intermediate \nstructure of cross-linking, enabling clear fluorescent labeling of collagen fibers \nthree-dimensionally oriented in animal tissues. In this study, we demonstrated that imaging \nwith DAF-FM/DAR-4M is a powerful tool for the pulse-chase observation of collagen fibers in \nin vitro system using mammalian cells and in in vivo system using zebrafish. In the future, \nwe expect to apply our method to the analysis of collagen fiber dynamics during mammalian \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\ntissue growth. Furthermore, our method could contribute to understanding the mechanisms \nof severe diseases such as fibrosis, which is caused by disruptions of the regulation of \ncollagen fiber production and remodeling. \n \n \nMATERIALS AND METHODS \n \nAnimal maintenance and tissue preparation \nPreparation of mouse tissues was carried out at Osaka Medical and Pharmaceutical \nUniversity. All mouse experiments were approved by the Institutional Review Board of \nOsaka Medical and Pharmaceutical University and performed in accordance with the Guide \nfor the Animal Care and Use of Laboratory Animals of Osaka Medical and Pharmaceutical \nUniversity. ICR mice were purchased from Japan SLC, Inc. (Shizuoka, Japan). Mice were \neuthanized under deep anesthesia using isoflurane inhalation for adults and hypothermia for \npups and embryos. Tissues were harvested in phosphate buffered saline (PBS) and utilized \nfor whole-mount staining or fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (pH \n7.4) for 2 days at 4°C.  Zebrafish were maintained under the standard laboratory conditions \nand treated as previously described ( 48). AB strains were used as wild type zebrafish. \nAlbino axolotls were raised and maintained under the conditions with 14 h of light/10\n/i4 h of \ndark cycles at 20 /i4 °C and fed commercial granular solid food 2 times a day. Zebrafish and \naxolotls were anesthetized with tricaine (MS-222) at optimal concentrations according to \neach body size. All experiments using these animals were approved by the animal care and \nuse at Osaka University. \n \nMicroscopy and image analysis \nThe fluorescent images were obtained using confocal microscopes: LSM 780 (Carl Zeiss), \nSTELLARIS8 (Leica), and FV1000 (Olympus), as well as a two-photon microscope, A1R \nMP+/Ti2-E (Nikon). 890 nm laser excitation and a 440 nm SP emission filter was used for \nSHG imaging of collagen fibers. ZEN (Carl Zeiss), LAS X (Leica), FV10-ASW (Olympus), \nand Fiji were used as image software for z projections. 3D image analysis was processed \nusing Imaris 10.1.1 (Oxford Instruments). The fluorescence intensity values and orientation \nangles of collagen fibers were measured using FIJI. \n \nDAF-FM/DAR-4M staining for the visualization of collagen fibers formed by cultured \ncells \nDAF-FM (Goryo Chemical, SK1003-01) and DAR-4M (Goryo Chemical, SK1005-01) were \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nused for staining of the collagen fibers produced by mouse embryonic fibroblast (MEF) \n(Reprocell, RCHEFC003). The cells were cultured in Dulbecco’s Modified Eagle’s Medium \n(DMEM, FUJIFILM Wako Pure Chemical Corporation) supplemented with 10% fetal bovine \nserum (FBS), 200 \nμ M L-ascorbic acid phosphate magnesium salt n-hydrate (FUJIFILM \nWako Pure Chemical Corporation), 100 U/mL penicillin, and 100 µg/mL streptomycin \n(Sigma-Aldrich) at 37°C in a 5% CO 2 atmosphere. 35 mm glass bottom dishes coated with \n0.1% gelatin solution (Nacalai Tesque, 19895-75) were used for the MEF culture experiment. \nFor the staining of collagen fibers deposited around the cultured cells, DAF-FM and \nDAR-4M were diluted with culture medium (DMEM with 10 % FBS) and adjusted to \nconcentrations of 6.9 \nμ M and 9.8 μ M, respectively. Before staining with the DAF-FM or \nDAR-4M solutions, the cells were washed three times with DMEM without L-ascorbic acid \nphosphate magnesium salt n-hydrate. The cells were then incubated with the diluted \nDAF-FM or DAR-4M solutions for 1 hour at 37°C. After incubation, the samples stained with \nthe probes were washed with fresh medium and observed using a confocal microscope. For \nthe pulse-chase observation of collagen fibers formed in vitro culture condition, the cells \nwere first incubated with DAF-FM solution, then after 4 days of culture with fresh medium, \nthey were next incubated with DAR-4M solution. For the analysis using a confocal \nmicroscope, the fluorescent signals of the collagen fibers stained with DAF-FM and DAR-4M \nwere detected with 488 nm and 561 nm lasers, respectively. \n \nDAF-FM staining of the purified collagen \nDAF-FM (Goryo Chemical, SK1003-01) was used for staining of the purified collagen \nsubstrates. Gelatin (Nacalai Tesque, 19895-75), collagen type I (Nitta Gelatin, Cellmatrix \nType I-P) and collagen type \n/i4  (AteloCell, CL-22) were prepared on 35 mm glass bottom \ndishes as substrates for DAF-FM staining. Gelatin and collagen type /i4  were adjusted to \nconcentrations of 0.1% and 0.03 mg/ml, respectively, and coated on glass. Collagen type I \nwas adjusted to a concentration of 1.8 mg/ml and neutralized by 1N NaOH to gel on glass. \nEach substrate was incubated with the DAF-FM solution diluted with DMEM with 10 % FBS, \nadjusted to a concentration of 6.9 \nμ M, for 1 hour at 37°C. After the incubation, the samples \nstained with DAF-FM were washed with fresh medium and observed using a confocal \nmicroscope. \n \n \nDAF-FM DA/DAR-4M AM staining for the visualization of collagen fibers in vertebrate \ntissues \nDAF-FM DA (Goryo Chemical, SK1004-01) and DAR-4M AM (Goryo Chemical, SK1006-01) \nwere used for whole-mount staining to visualize vertebrate animal tissues. For the staining \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nof whole-mount mouse tissues, DAF-FM DA was diluted with PBS, adjusted to \nconcentrations of 5 μ M, and used for staining. Mouse embryos or dissected tissues, \nincluding P0 - P1 skins and P14 tails, were incubated in the staining solution under dark \nconditions and treated for 12 hours at room temperature. After the staining, they were \nwashed with PBS and fixed with 4% paraformaldehyde (PFA) in PBS O/N at 4°C. After the \nfixation, they were treated with RapiClear 1.52 (SJL, RC152001) O/N at room temperature \nfor optical tissue clearing and observed using a confocal or two-photon microscope. For the \nstaining of whole-mount juvenile of axolotls, DAF-FM DA and DAR-4M AM were diluted with \nbreeding water, adjusted to concentrations of 5 \nμ M and 10 μ M, respectively, and used for \nstaining. Living juvenile of axolotls were bathed in the staining solution under dark \nconditions and treated for 12 hours at 20\n/i4 °C. After the staining, they were anesthetized with \ntricaine (MS-222) at an optimal concentration and fixed with 4% PFA in PBS O/N at 4°C. \nAfter the fixation, their skins were dissected and observed using a confocal microscope. \nAxolotl forelimbs were observed using a two-photon microscope after subsequent \ntransparency treatment with RapiClear 1.49 (SJL, RC149001). For the staining of \nwhole-mount zebrafish larvae, DAF-FM DA and DAR-4M AM were diluted with breeding \nwater, adjusted to concentrations of 5 \nμ M and 10 μ M, respectively, and used for staining. \nLiving zebrafish larvae were bathed in DAF-FM DA solution under dark conditions and \ntreated for 12 hours at room temperature. After the staining, they were anesthetized with \nMS-222 at an optimal concentration and fixed with 4% PFA in PBS O/N at 4°C. After the \nfixation, their tissues including skins, tendons, and bones were observed with a confocal \nmicroscope. For the pulse-chase observation of vertebral collagen fibers, zebrafish larvae at \n7dpf were first stained with DAF-FM DA, then after 2 weeks of breeding in a circulating tank, \nthey were next stained with DAR-4M AM. For the observation using a confocal microscope, \nthe fluorescent signals of the collagen fibers stained with DAF-FM DA and DAR-4M AM \nwere detected with 488 nm and 561 nm lasers, respectively. \n \nDetection of DAF-FM modification of lysine at the telopeptide domains of type I \ncollagen \nThe cell/matrix layers were sequentially digested with bacterial collagenase and pepsin to \nanalyze DAF-FM modification of telopeptidyl lysine in type I collagen as previously reported \n(49). In brief, the samples were heated at 80°C for 30 minutes, and digestion with 0.01 \nmg/mL of recombinant collagenase from Grimontia hollisae (Nippi, Tokyo, Japan) (50) was \nperformed in 100 mM Tris-HCl/5 mM CaCl\n2 (pH 7.5) at 37°C for 16 hours. After addition of \nacetic acid (final 0.5 M), digestion with 0.01 mg/mL of pepsin was further performed at 37°C \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nfor 16 hours. The peptide solutions were subjected to LC-MS analysis on a maXis II \nquadrupole time-of-flight mass spectrometer (Bruker Daltonics, Bremen, Germany) coupled \nto a Shimadzu Prominence UFLC-XR system (Shimadzu, Kyoto, Japan) using an Ascentis \nExpress C18 HPLC column (5 µm particle size, L × I.D. 150 mm × 2.1 mm; Supelco, \nBellefonte, PA, USA) ( 49). Peaks of peptides containing lysine or allysine labeled with \nDAF-FM (+392.061 Da) were detected in extracted ion chromatograms. \n \nAntibody staining of collagen  \nFor the antibody staining of collagen type I,  the cultured MEFs were fixed with 4% PFA in \nPBS and blocked with 1% bovine serum albumin (BSA)/PBS. After blocking, they were \nincubated with anti-mouse collagen type I rabbit polyclonal antibody (Rockland Inc, \n600-401-103-0.1, 1:100 dilution) solution in 1% BSA/PBS O/N at 4°C. Next day, they were \nwashed with PBS, and incubated with goat anti-rabbit IgG (H+L) antibody, FITC conjugate \n(Invitrogen, 65-6111, 1:200 dilution) solution in 1% BSA/PBS for 2 hours at room \ntemperature. For the antibody staining of collagen type \n/i4 , the mouse E14.5 embryos were \nfixed with 4% PFA in PBS. After fixation, their tails were dissected and sequentially \nimmersed in 10%, 20%, and 30% sucrose solutions in PBS and in a 1:1 solution of 30% \nsucrose and Tissue-Tek O.C.T Compound (Sakura Finetek Japan, 4583). Subsequently, \nthey were embedded in O.C.T Compound and frozen using dry ice. The tissues were then \nsectioned at a thickness of 10 µm using a cryomicrotome CM1850 (Leica). The cryo-section \nsamples were blocked with 2% bovine serum albumin (BSA)/PBS. After blocking, they were \nincubated with anti-chick collagen type \n/i4  mouse monoclonal antibody (DSHB, /i4 -/i4 6B3, \n1:100 dilution) solution in 2% BSA/PBS O/N at 4°C. Next day, they were washed with PBS \nand incubated with goat anti-mouse IgG (H+L) antibody, Alexa 594 conjugate (Invitrogen, \nA11020, 1:200 dilution) solution in 1% BSA/PBS for 1 hour at room temperature. \nImmunofluorescence images were obtained using a confocal microscope. \n \nStaining with denatured collagen-binding peptide \nFor staining of the collagen fibers deposited around cultured cells, the cells were treated \nwith phosphate-buffered saline (PBS) heated at 95°C for 1 minute to heat-denature \nextracellular collagen. They were then fixed with a 4% PFA and blocked with 2% BSA/PBS. \nAfter blocking, they were incubated with 5 µg/mL of BindCOL, biotin-conjugated (Funakoshi, \nFDV-0035) in 1% BSA/PBS or 3 µg/mL of fluorescein-conjugated soCMP6-7(Glu)2 in 1% \nBSA/PBS O/N at 4°C and washed with PBS (35). The cell samples incubated with BindCOL \nsolution were stained with streptavidin, Alexa 647 conjugate (Invitrogen, S32357, 1:200 \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\ndilution) solution in 1% BSA/PBS for 1 hour at room temperature and washed with PBS after \nstaining. Finally, the fluorescence signals of denatured collagen fibers were imaged using a \nconfocal microscope. For staining of the collagen fibers distributed in zebrafish tissues, the \ntissues were fixed with a 4% PFA. After fixation, they were heat-treated in a thermo bath at \n80°C for 10 minutes to denature extracellular collagen. They were then blocked with 2% \nBSA/PBS and incubated with 5 µM of CHP-Cy3 in 2% BSA/PBS O/N at 4°C, and then \nwashed with PBS. Finally, the fluorescence signals of denatured collagen fibers were \nimaged using a confocal microscope. \n \nStaining of collagen deposited around BAPN-treated cultured cells \nA previously established MEF clones were used in this staining (51). The cells were cultured \nin DMEM with 10% FBS and 200 \nμ M L-ascorbic acid phosphate magnesium salt n-hydrate \nat 37°C in a 5% CO2 atmosphere. The medium was replaced with HFDM-1(+) (Cell Science \n& Technology Institute Inc.) containing 100 U/mL penicillin and 100 µg/mL streptomycin after \nthe cells had reached confluence in 35 mm glass-bottom dishes. Confluent MEFs were \nincubated in HFDM-1(+) with or without the addition of 500 µM 3-aminopropionitrile \nfumarate (BAPN, Sigma-Aldrich) for 2 days in 35 mm glass-bottom dishes. Subsequently, \nthe cells were washed with PBS and incubated in a medium containing DAF-FM/DMSO \n(final concentration of 6.9 µM DAF-FM, and 0.1% DMSO) or DMSO (final concentration of \n0.1%) at a dilution of 1/1000 for an additional hour. In cases where BAPN was added, it was \nintroduced to a concentration of 500 µM. After washing the cells with PBS, they were fixed in \na 4% paraformaldehyde phosphate buffer solution for 10 minutes. The cells were washed \nwith PBS and observed using a confocal laser microscope. \n \nStaining of cellular actin and nucleus \nFor staining the actin cytoskeleton and cell nuclei of the cultured MEFs, the cells were fixed \nwith 4% PFA. After fixation, they were washed with PBS and incubated with a solution of \nPhalloidin-iFluor 594 conjugated (AAT Bioquest; 1:300 dilution) and Hoechst (Dojindo; 1:500 \ndilution) in PBS for 2 hours at room temperature. For staining of cell nuclei in frozen sections \nof mouse tail tissues, the samples were fixed with 4% PFA. After fixation, they were washed \nwith PBS and incubated with a solution of Hoechst (Dojindo; 1:500 dilution) in PBS O/N at \n4°C.\n For staining of cell nuclei in P14 mouse tail and P1 mouse skin, the tissue samples \nwere fixed with 4% PFA. After fixation, they were washed with PBS and incubated with a \nsolution of Syto 82 (Invitrogen; 1:1000 dilution) in PBS O/N at 4°C. Each sample was\n \nwashed in PBS after staining, and nuclear fluorescence was captured using confocal \nmicroscopy.  \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n \nBrdU incorporation assay \nMEFs were incubated in culture medium for 6 days and then incubated in DAF-FM solution \nat a concentration of 6.9 μ M for 1 hour. After staining with DAF-FM, cells were washed with \nfresh medium and treated for 12 hours in medium containing BrdU (Abcam, ab142567) \nadjusted to a concentration of 10 \nμ M. Cells were fixed with cold 70% EtOH for 5 minutes at \nroom temperature, and then treated with 0.1N HCl solution containing 0.1% Triton for 30 \nminutes at 37°C to increase permeability of the cell nuclei. After several washes in PBS, cell \nsamples were blocked for 1 hour with a 1% BSA/PBS solution for antibody staining. Mouse \nmonoclonal anti-BrdU antibody (Molecular Probe, A21300, 1:200 dilution) was used as the \nprimary antibody, and cell samples were incubated O/N at 4°C in the antibody solution \ncontaining 1% BSA/PBS. The next day, they were washed with PBS and incubated with \ngoat anti-mouse IgG antibody, Alexa 594 conjugated (Invitrogen, A11020, 1:200 dilution) \nsolution with 1% BSA/PBS for 2 hours at room temperature. Cell nuclei were simultaneously \nstained with Hoechst (Dojindo; 1:500 dilution) and the percentage of BrdU-positive nuclei \nwas compared between cells without DAF-FM staining (control) and cells with DAF-FM \nstaining.\n  \n \nCell viability assay \nMEFs were incubated in culture medium for 7 days and then incubated with DAF-FM \nsolution at a concentration of 6.9 μ M for 1 hour.  After staining with DAF-FM, cells were \nwashed in fresh medium and incubated in Syto 82 (Invitrogen, S11363, 1:1000 dilution) \nsolution at 37°C for 30 minutes to label the nuclei of living cells. Cells were then washed in \nfresh medium and incubated in Nuclear Blue (AAT Bioquest, Live or Dead Cell Viability \nAssay Kit, 22788, 1:200 dilution) solution for 30 minutes at 37°C to label the nuclei of dead \ncells. After nuclear staining, cells were washed with PBS, and the percentage of Nuclear \nBlue-positive nuclei was compared between cells without DAF-FM staining (control) and \ncells with DAF-FM staining. \n \nDrug treatment for NO removal and NOS inhibition \nMEFs were cultured for 10 days in the culture medium as described above. They were then \ntreated under the following three different conditions: DMSO (0.1% in DMEM) as control, \n2-(4-Carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (C-PTIO, Dojindo) (500 \nμ M in DMEM) known as a NO remover ( 52), and N G-nitro-L-arginine methyl ester \nhydrochloride (L-NAME, Dojindo) (500 μ M in DMEM) known as an inhibitor of NOS ( 53). In \nthe control and L-NAME treatment experiment, after 24 hours of treatment, MEFs were \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nincubated for 1 hour at 37°C in DAF-FM solution (6.9 μ M) diluted with DMEM containing the \ncorresponding drug (0.1% DMSO and 500 μ M L-NAME). In C-PTIO treatment experiment, \nafter 30 minutes of the treatment, MEFs were incubated in DAF-FM solution (6.9 μ M) diluted \nwith DMEM containing C-PTIO (500 μ M) for 1 hour at 37°C. After DAF-FM staining, \nfluorescent images of the collagen fibers produced by MEFs were imaged using a confocal \nmicroscope and the fluorescence intensity values were measured by FIJI image analysis \nsoftware.   \n \nWestern blot analysis of collagen in cell layers \nAs in the experiment of staining collagen deposited around BAPN-treated cultured cells, \nconfluent MEFs in 35 mm dishes were incubated in HFDM-1(+) with or without BAPN for 2 \ndays and then treated with DAF-FM/DMSO or DMSO for 1 hour. Following a wash with PBS, \nthe cell layers were dissolved in SDS-PAGE sample buffer (50 mM Tris-HCl [pH 6.7], 10% \nglycerol, and 2% SDS) and heated at 95°C for 5 minutes. The protein concentration of these \nSDS samples was determined using Pierce™ BCA protein assay kit (Thermo Fisher \nScientific, Waltham, MA, USA). SDS-PAGE was conducted on a 5% polyacrylamide gel with \n91 mM 1,4-dithiothreitol (DTT)-reduced or non-reduced samples, and proteins on the gel \nwere transferred to nitrocellulose membranes. Fluorescent bands were visualized with a \nCCD imager LAS-3000 (Fujifilm, Tokyo, Japan). Subsequently, the membranes were \nblocked with 5% skim milk/Tris-buffered saline (TBS; 50 mM Tris-HCl pH 7.4, 150 mM NaCl) \nand washed with TBS. They were treated with 1 µg/mL of biotin-conjugated Bind COL in 2% \nskim milk/TBS to detect collagen polypeptides (35). The membranes were washed with TBS, \ntreated with streptavidin-HRP (Thermo Fisher Scientific, 1:5000 dilution) in 2% skim \nmilk/TBS, and washed with TBS containing 0.1% Tween-20. Collagen bands were detected \nwith a CCD imager LAS-3000 using Pierce™ ECL western blotting substrate kit (Thermo \nFisher Scientific). \n \nSDS-PAGE analysis of cell layers treated with pepsin \nConfluent MEFs in 35 mm dishes were incubated in HFDM-1(+) containing DAF-FM/DMSO \n(final concentration 6.9 µM DAF-FM, 0.1% DMSO) or DMSO (final concentration 0.1% \nDMSO) for 2 days. The cell layers were washed with PBS and those collected with cell \nscrapers were treated with 0.1 M HCl, with or without 100 µg/mL pepsin (Sigma-Aldrich), at \n4°C for 16 hours. After neutralization with NaOH, these samples were mixed with 5 × SDS \nsample buffer and heated at 95°C for 5 minutes. Proteins in 91 mM DTT-reduced or \nnon-reduced samples were separated by SDS-PAGE on an 8% polyacrylamide gel, and \nfluorescent bands were visualized with a CCD imager LAS-3000. Subsequently, protein \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nbands were visualized by Coomassie Brilliant Blue R-250 staining. \n \nStatistical analysis \nStatistical analyses were performed using GraphPad Prism version 9.5.1 (731). All data are \npresented as mean ± SD. An unpaired two-tailed Student’s t-test was used to assess the \nstatistical significance of the differences between the means of two independent groups. \n \nANOVA followed by Tukey's multiple comparison test was conducted to evaluate the \nstatistical significance of differences among the three groups. \n \n \nAcknowledgments: We are grateful to members of the Kondo laboratory (laboratory of \npattern formation) at Osaka University and members of the Kuroda laboratory (laboratory of \nmorphogenesis) at JT Biohistory Research Hall. We would like to thank Dr. Ritsuko Suyama \nand Dr. Ritsuko Morita for managing the confocal microscopes at the Gr aduate School of \nFrontier Biosciences, Osaka University. We acknowledge the Leica imaging lab and the \nNikon imaging center at Osaka University for their support with obtaining the fluorescence \nimaging data. \n \nAuthor Contributions: J.K. and T.K. designed research; J.K. wrote the paper; S.F. and A.H. \nprovided mouse samples; J.K., K.F. and Y .T. conducted experiments; J.K. and K.F. \nexamined the conditions for staining of the collagen fibers produced by culture cells; J.K., \nS.F. and A.H. examined the conditions for staining of the collagen fibers in mouse tissues; \nJ.K., K.F. and Y .T. analyzed data; K. F., S. F., A.H., Y .T. and T.K. provided critical comments \nfor improving the manuscript. \n \nCompeting interests: The authors declare no competing interests  \n \nSupplemental materials: This article contains supplemental materials.  \n \nFunding: This research was funded by Japan Society for the Promotion of Science (JSPS) \nKAKENHI, Grant Number 21K06200 and Japan Science and T echnology Agency (JST) \nFOREST Program, Grant Number JPMJFR224P . \n \n \nREFERENCES AND NOTES \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n1.  F. W. Keeley, R. P. Mecham, Eds., Evolution of Extracellular Matrix  (Springer Berlin \nHeidelberg, Berlin, Heidelberg, 2013; \nhttps://link.springer.com/10.1007/978-3-642-36002-2)Biology of Extracellular Matrix. \n2.  C. Frantz, K. M. Stewart, V. M. Weaver, The extracellular matrix at a glance. Journal of \nCell Science 123, 4195–4200 (2010). \n3.  K. E. Kadler, C. Baldock, J. Bella, R. P . Boot-Handford, Collagens at a glance. Journal \nof Cell Science 120, 1955–1958 (2007). \n4.  S. Ricard-Blum, The Collagen Family. Cold Spring Harbor Perspectives in Biology  3, \na004978–a004978 (2011). \n5.  J. K. Mouw, G. Ou, V. M. Weaver, Extracellular matrix assembly: a multiscale \ndeconstruction. Nat Rev Mol Cell Biol 15, 771–785 (2014). \n6.  A. L. Fidler, S. P. Boudko, A. Rokas, B. G. Hudson, The triple helix of collagens – an \nancient protein structure that enabled animal multicellularity and tissue evolution. \nJournal of Cell Science 131, jcs203950 (2018). \n7.  H. Kuivaniemi, G. Tromp, D. J. Prockop, Mutations in fibrillar collagens (types I, II, III, \nand XI), fibril-associated collagen (type IX), and network-forming collagen (type X) \ncause a spectrum of diseases of bone, cartilage, and blood vessels. Hum. Mutat.  9, \n300–315 (1997). \n8.  J. Etich, L. Leßmeier, M. Rehberg, H. Sill, F. Zaucke, C. Netzer, O. Semler, \nOsteogenesis imperfecta-pathophysiology and therapeutic options. Mol Cell Pediatr  7, \n9 (2020). \n9.  T. A. Wynn, T. R. Ramalingam, Mechanisms of fibrosis: therapeutic translation for \nfibrotic disease. Nat Med 18, 1028–1040 (2012). \n10.  L. J. Dooling, K. Saini, A. A. Anla\nş , D. E. Discher, Tissue mechanics coevolves with \nfibrillar matrisomes in healthy and fibrotic tissues. Matrix Biol 111, 153–188 (2022). \n11.  R. J. Goodyear, X. Lu, M. R. Deans, G. P . Richardson, A tectorin-based matrix and \nplanar-cell-polarity genes are required for normal collagen-fibril orientation in the \ndeveloping tectorial membrane. Development, dev.151696 (2017). \n12.  R. Seidel, M. Blumer, E.-J. Pechriggl, K. Lyons, B. K. Hall, P. Fratzl, J. C. Weaver, M. N. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nDean, Calcified cartilage or bone? Collagens in the tessellated endoskeletons of \ncartilaginous fish (sharks and rays). Journal of Structural Biology 200, 54–71 (2017). \n13.  A. E. Mayorca-Guiliani, O. Willacy, C. D. Madsen, M. Rafaeva, S. Elisabeth Heumüller, \nF. Bock, G. Sengle, M. Koch, T. Imhof, F. Zaucke, R. Wagener, T. Sasaki, J. T. Erler, R. \nReuten, Decellularization and antibody staining of mouse tissues to map native \nextracellular matrix structures in 3D. Nat Protoc 14, 3395–3425 (2019). \n14.  P . J. Campagnola, A. C. Millard, M. Terasaki, P. E. Hoppe, C. J. Malone, W. A. Mohler, \nThree-Dimensional High-Resolution Second-Harmonic Generation Imaging of \nEndogenous Structural Proteins in Biological Tissues. Biophysical Journal  82, \n493–508 (2002). \n15.  X. Chen, O. Nadiarynkh, S. Plotnikov, P . J. Campagnola, Second harmonic generation \nmicroscopy for quantitative analysis of collagen fibrillar structure. Nat Protoc  7, \n654–669 (2012). \n16.  J. L. Morris, S. J. Cross, Y . Lu, K. E. Kadler, Y . Lu, S. L. Dallas, P . Martin, Live imaging \nof collagen deposition during skin development and repair in a collagen I – GFP fusion \ntransgenic zebrafish line. Developmental Biology 441, 4–11 (2018). \n17.  L. A. Shiflett, L. M. Ti ede-Lewis, Y . Xie, Y . Lu, E. C. Ray, S. L. Dallas, Collagen \nDynamics During the Process of Osteocyte Embedding and Mineralization. Front. Cell \nDev. Biol. 7, 178 (2019). \n18.  R. Kashimoto, S. Furukawa, S. Yamamoto, Y . Kamei, J. Sakamoto, S. Nonaka, T. M. \nWatanabe, T. Sakamoto, H. Sakamoto, A. Satoh, Lattice-patterned collagen fibers and \ntheir dynamics in axolotl skin regeneration. iScience 25, 104524 (2022). \n19.  H. Hino, S. Kondo, J. Kuroda, In vivo imaging of bone collagen dynamics in zebrafish. \nBone Rep 20, 101748 (2024). \n20.  G. Timin, M. C. Milinkovitch, High-resolution confocal and light-sheet imaging of \ncollagen 3D network architecture in very large samples. iScience 26, 106452 (2023). \n21.  M. R. Aronoff, P . Hiebert, N. B. Hentzen, S. Werner, H. Wennemers, Imaging and \ntargeting LOX-mediated tissue remodeling with a reactive collagen peptide. Nat Chem \nBiol 17, 865–871 (2021). \n22.  H. Wang, A. Poe, L. Pak, K. Nandakumar, S. Jandu, J. Steppan, R. Löser, L. \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nSanthanam, An in situ activity assay for lysyl oxidases. Commun Biol  4, 840 (2021). \n23.  H. Ma, I. Y . Zhou, Y . I. Chen, N. J. Rotile, I. Ay, E. A. Akam, H. Wang, R. S. Knipe, L. P . \nHariri, C. Zhang, M. Drummond, P. Pantazopoulos, B. F. Moon, A. T. Boice, S. E. \nZygmont, J. Weigand-Whittier, M. Sojoodi, R. A. Gonzalez-Villalobos, M. K. Hansen, K. \nK. Tanabe, P . Caravan, Tailored Chemical Reactivity Probes for Systemic Imaging of \nAldehydes in Fibroproliferative Diseases. J Am Chem Soc 145, 20825–20836 (2023). \n24.  P . Hiebert, G. Antoniazzi, M. Aronoff, S. Werner, H. We nnemers, A lysyl \noxidase-responsive collagen peptide illuminates collagen remodeling in wound healing. \nMatrix Biol 128, 11–20 (2024). \n25.  E. A. Akam-Baxter, D. Bergemann, S. J. Ridley, S. To, B. Andrea, B. Moon, H. Ma, Y . \nZhou, A. Aguirre, P. Caravan, J. M. Gonzalez-Rosa, D. E. Sosnovik, Dynamics of \ncollagen oxidation and cross linking in regenerating and irreversibly infarcted \nmyocardium. Nat Commun 15, 4648 (2024). \n26.  H. Kojima, N. Nakatsubo, K. Kikuchi, S. Kawahara, Y . Kirino, H. Nagoshi, Y . Hirata, T. \nNagano, Detection and Imaging of Nitric Oxide with Novel Fluorescent Indicators: \nDiaminofluoresceins. Anal. Chem. 70, 2446–2453 (1998). \n27.  H. Kojima, Y . Urano, K. Kikuchi, T. Higuchi, Y . Hirata, T. Nagano, Fluorescent \nIndicators for Imaging Nitric Oxide Production. Angew. Chem. Int. Ed.  38, 3209–3212 \n(1999). \n28.  H. Kojima, M. Hirotani, N. Nakatsubo, K. Kikuchi, Y . Urano, T. Higuchi, Y . Hirata, T. \nNagano, Bioimaging of Nitric Oxide with Fluorescent Indicators Based on the \nRhodamine Chromophore. Anal. Chem. 73, 1967–1973 (2001). \n29.  S. Lepiller, V. Laurens, A. Bouchot, P . Herbomel, E. Solary, J. Chluba, Imaging of nitric \noxide in a living vertebrate using a diaminofluorescein probe. Free Radical Biology and \nMedicine 43, 619–627 (2007). \n30.  J. Renn, B. Pruvot, M. Muller, Detection of nitric oxide by diaminofluorescein visualizes \nthe skeleton in living zebrafish. J. Appl. Ichthyol. 30, 701–706 (2014). \n31.  A. Huysseune, U. G. Larsen, D. Larionova, C. L. Matthiesen, S. V. Petersen, M. Muller, \nP . E. Witten, Bone Formation in Zebrafish: The Significance of DAF-FM DA Staining for \nNitric Oxide Detection. Biomolecules 13, 1780 (2023). \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n32.  J. Kuroda, H. Hino, S. Kondo, Dynamics of actinotrichia, fibrous collagen structures in \nzebrafish fin tissues, unveiled by novel fluorescent probes. PNAS Nexus  3, pgae266 \n(2024). \n33.  K. Miyamoto, J. Kuroda, S. Kamimura, Y . Sasano, G. Abe, S. Ansai, N. Funayama, M. \nUesaka, K. Tamura, Actinotrichia-independent developmental mechanisms of spiny \nrays facilitate the morphological diversification of Acanthomorpha fish fins. bioRxiv \n[Preprint] (2025). https://doi.org/10.1101/2025.03.01.640274. \n34.  A. Ohashi, H. Sakamoto, J. Kuroda, Y . Kondo, Y . Kamei, S. Nonaka, S. Furukawa, S. \nYamamoto, A. Satoh, Keratinocyte-driven dermal collagen formation in the axolotl skin. \nNat Commun 16, 1757 (2025). \n35.  K. K. Takita, K. K. Fujii, K. Ishii, T. Koide, Structural optimization of cyclic peptides that \nefficiently detect denatured collagen. Org Biomol Chem  17, 7380–7387 (2019). \n36.  H. Li, A. Wan, Fluorescent probes for real-time measurement of nitric oxide in living \ncells. Analyst 140, 7129–7141 (2015). \n37.  R. Nasuno, Y . Yoshikawa, H. Takagi, Acetaldehyde reacts with a fluorescent nitric \noxide probe harboring an o-phenylenediamine structure that interferes with fluorometry. \nFree Radical Biology and Medicine 187, 29–37 (2022). \n38.  R. Rucker, T. Kosonen, M. Clegg, A. Mitchell, B. Rucker, J. Uriu-Hare, C. Keen, Copper, \nlysyl oxidase, and extracellular matrix protein cross-linking. The American Journal of \nClinical Nutrition 67, 996S-1002S (1998). \n39.  S. D. Vallet, S. Ricard-Blum, Lysyl oxidases: from enzyme activity to extracellular \nmatrix cross-links. Essays in Biochemistry 63, 349–364 (2019). \n40.  M. D. Shoulders, R. T. Raines, Collagen structure and stab ility. Annu Rev Biochem 78, \n929–958 (2009). \n41.  S. S. Tang, P. C. Trackman, H. M. Kagan, Reaction of aortic lysyl oxidase with \nbeta-aminopropionitrile. J Biol Chem 258, 4331–4338 (1983). \n42.  M. Terajima, Y . Taga, W. A. Cabral, Y . Liu, M. Nagasawa, N. Sumida, Y . Kayashima, P . \nChandrasekaran, L. Han, N. Maeda, I. Perdivara, S. Hattori, J. C. Marini, M. Yamauchi, \nCyclophilin B control of lysine post-translational modifications of skin type I collagen. \nPLOS Genetics 15, e1008196 (2019). \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n43.  T. Saito, M. Terajima, Y . Taga, F. Hayashi, S. Oshima, A. Kasamatsu, Y . Okubo, C. Ito, \nK. Toshimori, M. Sunohara, H. Tanzawa, K. Uzawa, M. Yamauchi, Decrease of lysyl \nhydroxylase 2 activity causes abnormal collagen molecular phenotypes, defective \nmineralization and compromised mechanical properties of bone. Bone 154, 116242 \n(2022). \n44.  J. Hwang, Y . Huang, T. J. Burwell, N. C. Peterson, J. Connor, S. J. Weiss, S. M. Yu, Y . \nLi, In Situ Imaging of Tissue Remodeling with Collagen Hybridizing Peptides. ACS \nNano 11, 9825–9835 (2017). \n45.  Y . Lu, S. A. Kamel ‐ El Sayed, K. Wang, L. M. Tiede ‐ Lewis, M. A. Grillo, P. A. Veno, V. \nDusevich, C. L. Phillips, L. F. Bonewald, S. L. Dallas, Live Imaging of Type I Collagen \nAssembly Dynamics in Osteoblasts Stably Expressing GFP and mCherry ‐ Tagged \nCollagen Constructs. J of Bone & Mineral Res 33, 1166–1182 (2018). \n46.  T. Tanaka, K. Moriya, M. Tsunenaga, T. Yanagawa, H. Morita, T. Minowa, Y .-I. Tagawa, \nN. Hanagata, Y . Inagaki, T. Ikoma, Visualized procollagen I α 1 demonstrates the \nintracellular processing of propeptides. Life Sci Alliance 5, e202101060 (2022). \n47.  M. Van Der Rest, R. Garrone, Collagen family of proteins. The FASEB Journal  5, \n2814–2823 (1991). \n48.  ZFIN Publication: Westerfield, 1995. https://zfin.org/ZDB- PUB-970327-24. \n49.  M. Terajima, Y . Taga, W. A. Cabral, M. Nagasawa, N. Sumida, S. Hattori, J. C. Marini, \nM. Yamauchi, Cyclophilin B Deficiency Causes Abnormal Dentin Collagen Matrix. J \nProteome Res 16, 2914–2923 (2017). \n50.  N. Teramura, K. Tanaka, K. Iijima, O. Hayashida, K. Suzuki, S. Hattori, S. Irie, Cloning \nof a novel collagenase gene from the gram-negative bacterium Grimontia (Vibrio) \nhollisae 1706B and its efficient expression in Brevibacillus choshinensis. J Bacteriol \n193, 3049–3056 (2011). \n51.  N. Nagai, M. Hosokawa, S. Itohara, E. Adachi, T. Matsushita, N. Hosokawa, K. Nagata, \nEmbryonic lethality of molecular chaperone hsp47 knockout mice is associated with \ndefects in collagen biosynthesis. J Cell Biol 150, 1499–1506 (2000). \n52.  T. Akaike, M. Yoshida, Y . Miyamoto, K. Sato, M. Kohno, K. Sasamoto, K. Miyazaki, S. \nUeda, H. Maeda, Antagonistic action of imidazolineoxyl N-oxides against \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\nendothelium-derived relaxing factor/.NO through a radical reaction. Biochemistry  32, \n827–832 (1993). \n53.  S. Kagiyama, T. Tsuchihashi, I. Abe, M. Fujishima, Cardiovascular effects of nitric \noxide in the rostral ventrolateral medulla of rats. Brain Res 757, 155–158 (1997). \n \n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted June 24, 2025. ; https://doi.org/10.1101/2025.06.19.660320doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}