Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Apr 10.
Published in final edited form as: J Biomech. 2025 Sep 10;192:112956. doi: 10.1016/j.jbiomech.2025.112956

Knockout of collagen V in development, but not in skeletal maturity, leads to long-term deficits in supraspinatus tendon mechanics in mice

Jeremy D Eekhoff 1, Mitchell J Hallman 1, Michael S DiStefano 1, Rebecca L Betts 1, Courtney A Nuss 1, Stephanie N Weiss 1, Andrew F Kuntz 1, Louis J Soslowsky 1,*
PMCID: PMC13063014  NIHMSID: NIHMS2162669  PMID: 40945377

Abstract

Collagen V is a key matrix protein involved in fibril nucleation and lateral fibril growth during extracellular matrix assembly. Genetic mouse models have been used to investigate the role of collagen V in tendon, which showed deficient mechanical properties and aberrant fibril structure in the absence of collagen V. However, the lasting effects of collagen V deficiency later into adulthood remain unknown, as well as the role of collagen V in maintaining a mature matrix. This study therefore investigated the long-term effects of collagen V reduction on tendon as well as its role in mature tendon matrix in adulthood. Tendon-targeted conditional Col5a1 knockout, which excises Col5a1 alleles early in development, had long-term impact on tendon structure and function in 300-day old mice. Gene expression was altered with differential expression of primarily matrix and matrix remodeling genes. Regional changes in cellular shape and density were consistent with typical behavior in tendinopathy. Fibril diameters were increased due to dysregulated lateral growth. Deficits in mechanical properties indicate a weaker tendon matrix after knockout, although deformation patterns of collagen fibrils were not affected. In contrast, inducing collagen V knockdown in a mature tendon matrix at 120-days old did not cause substantial changes in any of the above-mentioned properties in 300-day old mice. In conclusion, these findings highlight the important function of collagen V in matrix assembly that has lasting effects into later ages, even though collagen V has little role in homeostatic maintenance of a mature tendon matrix.

Keywords: Tendon, Collagen V, Biomechanics, Development, Homeostasis

1. Introduction

Proper tendon function requires the coordinated action of the extracellular matrix at multiple length scales. As mechanical forces are transferred between muscle and bone through the tendon, the tendon matrix responds through the interplay of various mechanisms, such as sliding, realignment, uncrimping, and deformation, within and between collagen fibrils and fibers (Miller et al., 2012; Silva Barreto et al., 2023; Szczesny and Elliott, 2014). Therefore, the hierarchical structure of the tendon matrix is essential for its capacity to transfer load within the musculoskeletal system.

This structure is composed primarily of collagen I, which is chiefly responsible for bearing the tensile load sustained by tendon. However, the assembly of this structure relies on other components of the extracellular matrix that play an important role despite low abundance relative to collagen I. For example, short leucine-rich proteoglycans decorin and biglycan are required for the formation and maintenance of a functional tendon matrix, despite not directly contributing to its mechanical integrity (Fessel and Snedeker, 2009; Gordon et al., 2015; Robinson et al., 2017; Beach et al., 2022). Additionally, other collagen types, including collagens V, VI, XI, XII, and XIV have also been shown to be important for tendon matrix assembly with impacts on both matrix structure and mechanical properties (Ansorge et al., 2009; Izu et al., 2021, 2011; Wenstrup et al., 2011).

Collagen V, a regulatory collagen that forms heterotypic fibrils with collagen I (Mak et al., 2016), is notable for its association with classical Ehlers-Danlos Syndrome (cEDS) in human patients, which is caused by mutations in COL5A1 or COL5A2 (Bowen et al., 2017; Shirley et al., 2012). One of the characteristic features of cEDS is joint hypermobility, which indicates poor function of tendons and ligaments spanning the joint and can lead to joint pain, dislocations, and sprains. Col5a1 haploinsufficient mice have been used as a model for cEDS, demonstrating altered matrix assembly and impaired healing in tendon (Wenstrup et al., 2011; Johnston et al., 2017; Machol et al., 2022; Carlson et al., 2024). Conditional tendon-targeted knockout of Col5a1 led to more dramatic deficits in tendon mechanics and abnormal collagen structure, indicating a dose-dependent response to reduction of collagen V (Connizzo et al., 2015; Sun et al., 2015). This effect was dependent on location within the supraspinatus tendon, which exhibits regional variation due to its complex multiaxial loading environment (Connizzo et al., 2015, 2016a; DiStefano et al., 2025a). In contrast, inducible knockdown of Col5a1 during tendon healing did not appreciably impair healing outcomes (Leiphart et al., 2022).

Despite this substantial body of work on how collagen V influences tendon development and healing, it remains unknown how collagen V impacts maintenance of a mature tendon matrix. Additionally, because these prior studies were conducted at a younger age, the long-term impact of collagen V deficiency later into adulthood is not yet understood. Therefore, the objective of this study was to determine (1) the role of collagen V in maintaining functional tendon matrix in adulthood and (2) the long-term impact of collagen V deficiency during and beyond tendon development. We hypothesized that loss of collagen V at early development would lead to alterations in matrix structure and reduced mechanical integrity in 300-day old mice, while loss of collagen V during adulthood would have similar, but more modest, effects on tendon properties.

2. Methods

2.1. Animals

Tendon-targeted conditional ScxCre;Col5a1f/+ (T-Het) and ScxCre; Col5a1f/f (T-Null) mice, inducible ROSA26-CreERT2;Col5a1f/+ (I-Het) and ROSA26-CreERT2;Col5a1f/f (I-Null) mice, and Cre-negative (WT) controls were used in this IACUC approved study. All mice were male to reduce variability and maintain statistical power. Inducible models (I-Het and I-Null) received three doses of tamoxifen (100 mg/kg) on consecutive days at skeletal maturity (120 days, or approximately 17 weeks, old) to induce excision of floxed Col5a1 alleles. All mice were sacrificed at 300 days, or approximately 43 weeks, old, which approximately corresponds to middle-age in humans (Flurkey et al., 2007).

2.2. Gene expression

Supraspinatus tendon samples (n = 4/group) used for gene expression were harvested immediately after sacrifice, flash frozen in liquid nitrogen, and stored at −80 °C (DiStefano et al., 2025b). Samples were later thawed in RNAlater-ICE and the insertion region (0–1 mm from humeral insertion) and midsubstance region (1–2 mm from humeral insertion) were separated and mechanically homogenized in TRIzol. RNA was isolated, converted to cDNA, and pre-amplified using standard protocols. Genes related to tendon health and pathology were pre-selected for analysis, including categories of collagens, non-collagenous matrix, matrix remodeling, cell–cell and cell-ECM proteins, and cell and inflammatory markers. Expression of these genes was measured using RT-qPCR reactions within a 96×96 microfluidic array. For each sample, ΔCT values were calculated as difference in cycle thresholds between the average of housekeeper genes Abl1, Rn18s, and Rps17, and the experimental gene fold change (FC) between genotypes was calculated as 2ΔCT-ΔCT. Genes were considered differently regulated between genotypes where |log2(FC)|>1 and p < 0.05.

2.3. Histology

Shoulders were freshly harvested after sacrifice and fixed in paraformaldehyde, decalcified with ethylenediaminetetraacetic acid, processed, and embedded in paraffin (n = 8–11/group). After embedding, samples were sectioned in the coronal plane at 7 um thickness and stained with hematoxylin and eosin (H&E). Images of the supraspinatus tendon and surrounding tissues were acquired at 20X magnification. Regions of interest (100×200 μm) within the insertion and midsubstance regions of the tendon were manually selected for quantitative analysis. Cells were segmented using automated color and intensity thresholds, and cell density and median nuclear aspect ratio were calculated for each region of interest using a custom MATLAB script.

2.4. Transmission electron microscopy

Samples designated for fibril morphology (n = 6–10/group) analysis using transmission electron microscopy (TEM) were fixed in Karnovsky’s fixative following sacrifice, then post-fixed and stained with 1 % osmium tetroxide and dehydrated in ethanol (Dunkman et al., 2013). Insertion and midsubstance regions of the supraspinatus tendon were isolated and embedded in Epon resin. Ultrathin (60–80 nm) sections were collected and stained with UranyLess and 1 % phosphotungstic acid. Transmission electron images were acquired at 60,000X magnification and collagen fibril diameters were measured from up to 10 images per sample (DiStefano et al., 2025a).

2.5. Mechanical testing

Mechanical testing was performed on samples stored at −20 °C (n = 10/group). Supraspinatus tendon-humerus complexes were harvested and finely dissected to remove non-tendinous soft tissue. Verhoeff’s stain was used to apply lines on the tendon at 0, 1, 2, and 2.5 mm from the humeral insertion to demarcate the insertion and midsubstance regions and set the gauge length for testing. A custom laser displacement device was used to measure tendon cross-sectional area, which was used to calculate engineering stress (Peltz et al., 2009). The free end of the tendon was secured at the 2.5 mm stain line between pieces of sandpaper using cyanoacrylate glue, and the humerus was potted in a custom 3D-printed pot using polymethyl methacrylate. Prepared samples were then tested using a universal test frame (Instron 5848) while hydration and temperature were maintained by testing in a phosphate buffered saline bath at 37 °C.

A 0.025 N preload was applied to the tendon to remove slack prior to starting the test protocol. Preconditioning was performed between 0.5 and 1.5 % grip strain for 10 cycles at 0.25 Hz. After a 1-minute rest, a suite of viscoelastic testing was performed. Stress relaxation was performed consecutively at 3, 5, and 7 % grip strain, with 10 min holds at each strain level to allow for the tendon to reach equilibrium. At the end of each hold, sinusoidal frequency sweeps were performed using 10 cycles at an amplitude of 0.125 % grip strain and frequencies of 0.1, 1, 5, and 10 Hz. Following the final frequency sweep, the tendon was returned to its initial length and allowed to rest for 1 min, then ramped to failure at a strain rate of 0.1 % per second to determine quasistatic properties. Images were captured during the ramp to failure to optically measure regional Lagrangian strain of the tissue. Percent relaxation, dynamic modulus, and phase shift were calculated from the viscoelastic testing, and regional moduli and maximum stress were determined using the ramp to failure data (Supplemental Fig. S1). Only material properties are reported because of differences in body weight between mouse lines that could confound comparison of structural properties (Supplemental Fig. S2).

2.6. Atomic force microscopy

Samples were prepared as described above for mechanical testing and loaded into the universal test frame (n = 9–10/group). Preconditioning was performed between 0 and 1 % grip strain for 10 cycles at 1 Hz. Following a 1-minute rest, samples were ramped to either 0, 3, or 6 % grip strain and immediately flash frozen using rapid freeze spray, then quickly removed from the test fixture and embedded in optimal cutting temperature compound while keeping the tendon frozen to maintain the applied strain. Longitudinal sections were collected at 20 μm thickness and fixed in 10 % neutral buffered formalin.

Collagen fibrils within the sections were visualized with an atomic force microscope. Images were captured using tapping mode in air with a VTESPA-300 cantilever probe. Five 2×2 μm images at 512×512 pixel resolution were captured within each region for each sample, spanning across multiple tissue sections to capture the heterogeneity across the full width of the tissue. Fibril d-periods were measured using Fourier transform analysis. Regional average d-period length, local variance (average variance in d-period length within individual images) and global variance (variance in d-period length across entire sample) were calculated (DiStefano et al., 2025a).

2.7. Statistics

All experimentation and analysis was performed blinded. Data were visually inspected and tested using Shapiro-Wilk tests to determine deviations from normality that would impact the results of parametric testing. Outliers, defined a priori as data less than the first quartile minus 2.2 times the interquartile range (IQR) or the third quartile plus 2.2 times the IQR, were identified and removed. The effect of genotype on the expression of Col5a1, all histologic data, mechanics data, and atomic force microscopy was analyzed using analysis of variance. Where appropriate, post-hoc t-tests with Bonferroni correction were used to compare differences between specific genotypes. Volcano plots to compare gene expression profiles between genotypes were generated using multiple t-tests. Fibril diameter distributions were analyzed using Kolmogorov-Smirnov tests. Comparisons were not made between conditional and inducible models due to the use of different Cre drivers.

3. Results

3.1. Col5a1 expression

Analysis of gene expression (Fig. 1) confirmed knockout of Col5a1 in the insertion and midsubstance regions of the supraspinatus tendon in T-Null samples. Similarly, tamoxifen-induced knockdown decreased Col5a1 expression in both regions of I-Null tendons.

Fig. 1.

Fig. 1.

Open in a new tab

Expression of Col5a1 was successfully reduced in the insertion and midsubstance regions of T-Null and I-Null tendons. Data shown as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001.

3.2. Conditional Col5a1 knockout changes gene expression profile

Gene expression experienced changes with knockout of collagen V (Fig. 2; Supplemental Table S1). The most substantial changes were found in T-Null tendons, where out of 92 genes, there were 33 differentially regulated genes (DEGs) in the insertion and 17 DEGs in the midsubstance relative to WT, and 46 DEGs in the insertion and 17 DEGs in the midsubstance relative to T-Het. Changes in gene expression in T-Het tendons were more modest, with 13 DEGs in the insertion and 9 DEGs in the midsubstance relative to WT tendons. Inducible models had few changes in expression, with no more than 3 DEGs for any specific comparisons between WT, I-Het, and I-Null genotypes in either region.

Fig. 2.

Fig. 2.

Open in a new tab

Volcano plots show differences in gene expression between genotypes in the insertion and midsubstance regions. There were several DEGs in T-Het and T-Null tendons compared to WT, while changes in expression of I-Het and I-Null tendons were minimal in both regions.

3.3. Cell morphology and density exhibit regional changes in T-null tendons

Quantitative analysis of H&E images (Fig. 3A) revealed region-specific differences in cell morphology and density in T-Null tendons, where there was a subtle decrease in nuclear aspect ratio in the insertion (Fig. 3B) and an increase in cell density in the midsubstance compared to WT and T-Het tendons (Fig. 3C). In contrast, induction of Col5a1 knockdown at maturity did not cause histological differences in I-Het or I-Null samples.

Fig. 3.

Fig. 3.

Open in a new tab

Histological analysis of cell shape and density. (A) Representative H&E images from each genotype and region used for analysis. (B) Nuclear aspect ratio was reduced in the insertion of T-Null tendons, while (C) cellular density was increased in the midsubstance of T-Null tendons. Data shown as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001.

3.4. Fibril diameters are increased with conditional Col5a1 knockout

Fibril diameter distributions were determined using transmission electron microscopy (Fig. 4A). All comparisons between genotypes were statistically significant (p < 0.001). In T-Null tendons, distributions trended toward larger diameter fibrils in both the insertion and midsubstance regions, with a qualitatively larger effect in the midsubstance (Fig. 4B). Further, there were abnormally shaped fibrils in both regions of T-Null samples, deviating from the typical circular cross-section. There was a similar, though more modest, shift towards larger diameter fibrils in both regions of I-Null tendons. Differences in other genotypes were subtle and may not be functionally meaningful.

Fig. 4.

Fig. 4.

Open in a new tab

Nanoscale analysis of fibril size and morphology. (A) Representative TEM images from each genotype and region. Some fibrils in T-Null tendons exhibited irregular cross-sectional shape. (B) Fibril diameter distributions shifted toward larger fibril diameters in both regions of T-Null tendons, with a qualitatively larger shift in the midsubstance. There was a similar, although more subtle, shift toward larger diameter fibrils in I-Null tendons.

3.5. Tendon mechanics are altered with conditional reduction of collagen V

Quasistatic mechanics of the supraspinatus tendon was affected by reduction of collagen V in the conditional knockout model. Mechanical properties of T-Null tendons exhibited reductions in the modulus of both insertion and midsubstance regions and a decrease in max stress (Fig. 5). However, T-Het tendons did not differ from WT in any measured property. Inducible knockdown of collagen V after development did not affect quasistatic material properties.

Fig. 5.

Fig. 5.

Open in a new tab

Analysis of quasistatic mechanical properties. (A) Insertion modulus, (B) midsubstance modulus, and (C) max stress were reduced in T-Null tendons, while other genotypes were not significantly different from WT. Data shown as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001.

Viscoelastic properties were also affected in T-Null tendons (Fig. 6). Percent relaxation was increased at 7 % strain (Fig. 6A). Dynamic modulus was decreased at all frequencies at 7 % strain, and at 0.1 Hz at 5 % strain (Fig. 6B). Changes in phase shift were more widespread, with tangent delta increased at all frequencies and strain levels with the exception of 10 Hz at 3 % strain (Fig. 6C). In contrast, T-Het, I-Het, and I-Null viscoelastic properties were not different than WT.

Fig. 6.

Fig. 6.

Open in a new tab

Viscoelastic properties were affected by genotype. (A) Percent relaxation was increased at 7 % strain in T-Null tendons. (B) Dynamic modulus was decreased in T-Null tendons at 7 % at all frequencies, and at 5 % strain at 0.1 Hz. (C) Tangent delta was increased at nearly all strain levels and frequencies in T-Null tendons, with the exception of 3 % strain at 10 Hz. Data shown as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001.

3.6. Collagen V does not influence nanoscale fibril deformation patterns

Regional collagen fibril deformation under applied loading was quantified using atomic force microscopy measurements of fibril d-period. Average fibril d-period, which represents strain within collagen fibrils, did not vary by genotype at any strain level (Fig. 7A). Similarly, neither local (Fig. 7B) nor global (Fig. 7C) d-period variance, which represent heterogeneity in fibril deformation, were affected.

Fig. 7.

Fig. 7.

Open in a new tab

Regional fibril deformation patterns were analyzed using atomic force microscopy. (A) average fibril d-period, (B) local fibril d-period variance, and (C) global fibril d-period variance were all unaffected by genotype in both insertion and midsubstance regions at all strain levels. Data shown as mean ± standard deviation.

4. Discussion

While previous research has largely focused on elucidating the roles of collagen V in development, healing, and maturity in early adulthood, the present work investigated the longer-term effects of collagen V deficiency later in life. As expected, deficits in tendon properties that occur from early developmental knockout of collagen V continue to manifest with more advanced age. However, inducible knockdown of collagen V after the formation of a mature tendon matrix did not produce substantial changes in the tendon.

Analysis of regional gene expression indicated considerable changes in the expression profiles with conditional knockout of collagen V. Many of the DEGs were related to matrix remodeling or matrix structure, suggesting an attempt to compensate for the loss of collagen V. Upregulation of Mmp13, which is associated with both tendon rupture and overuse, was consistently among the largest changes in expression in both regions of T-Null tendons (Sun et al., 2008; Del Buono et al., 2012; Nilsson et al., 2024). Matrix metalloproteinases Mmp2 and Mmp14 were similarly upregulated, although with lower fold changes. However, Adamts1 was downregulated, demonstrating more nuanced changes in matrix remodeling than a global increase in all proteinases. Expression of Eln and Fbn2, both structural components of elastic fibers, was increased in both insertion and midsubstance regions, suggesting potential changes to tendon matrix composition that could impact mechanical function (Eekhoff et al., 2017, 2021, 2023; Ishizaki et al., 2024). Interestingly, Col5a2 was also upregulated in T-Null samples, which may indicate an attempt to shift to other collagen V isoforms with the loss of Col5a1 expression (Mak et al., 2016). Other changes to matrix-associated genes were region-specific. For example, Col2a1 saw increased expression in the midsubstance, while Col3a1 and Col6a2 were upregulated in the insertion. Some non-collagenous matrix genes, such as Lum and Postn, were also upregulated only in the insertion region. Increased expression of Lox, Loxl1, and Loxl2 in the insertion correspond with a previously reported increase in cross-linking with collagen V knockout in tendon (Connizzo et al., 2016a; Eekhoff et al., 2018). Taken together, these changes in gene expression indicate a sweeping attempt to restore a functional tendon matrix. Moreover, other genes related to cell–cell and cell-ECM communication, such as Cdh11 and Cd44, and other cell signaling processes, such as Igf1, Spp1, and Vegfb, were also differentially expressed with conditional knockout of collagen V, highlighting the far-reaching effects on tendon activity. Inducible knockdown, however, did not produce appreciable changes in gene expression. Only a small number of genes were differentially expressed, which were inconsistent between regions and all had relatively low fold changes.

Conditional knockout of collagen V also led to change in cell shape and density. In the insertion region, T-Null tendons had more rounded cells which typically suggests a stronger fibrocartilaginous phenotype (Archambault et al., 2007; Xu and Murrell, 2008). However, gene markers such as Sox9 and Col2a1 were not upregulated in the insertion, so the change in cell shape may be due to other factors, such as a shift in cell phenotype in response to the softer tendon matrix. In the midsubstance region, cell density was increased with conditional Col5a1 knockout. Hypercellularity is commonly associated with tendinopathy, so this increased midsubstance density likely reflects degenerative changes within T-Null tendons (Planckaert et al., 2025). Again, these changes contrast with the inducible knockdown model, which did not exhibit differences from WT controls.

In addition to initial fibril nucleation, one of the primary roles of collagen V is regulating lateral growth of collagen fibrils (Mak et al., 2016). In this study, this function of collagen V was assessed by measuring the distribution of fibril diameters regionally within the supraspinatus tendon. In both regions of the tendon, diameter distributions were shifted towards larger diameter fibrils in T-Null samples, and some fibrils demonstrated an irregular shaped cross-section. This shift is consistent with the previously found role for collagen V in regulating lateral growth and previous experiments with collagen V knockout in tendon, where absence of collagen V during fibrillogenesis leads to unrestricted lateral expansion and larger fibrils (Connizzo et al., 2016a; Sun et al., 2015). While there are conflicting reports regarding the mechanical consequences of this shift towards increased fibril diameter (Connizzo et al., 2016b; Derwin and Soslowsky, 1999; Goh et al., 2012; Lavagnino et al., 2005; Rigozzi et al., 2010), these data indicate dysregulation of matrix assembly that deviates from healthy behavior in the WT controls. Changes in the fibril diameter distribution of the inducible genotypes were more subtle, although a slight shift towards larger diameters was still observed in I-Null tendons. In addition to the implications of this change in fibril diameter in the context of tendon pathology, another potential application is in tissue engineering where replicating the fibril diameter of native tendon has remained a longstanding challenge (Tam and Baar, 2025). Temporal modulation of collagen V may help regulate fibril diameters in engineered tendon constructs.

Deficits in mechanical properties with conditional collagen V knockout are consistent with changes in gene expression and structure that suggest detrimental changes to the tendon matrix. Both quasistatic and viscoelastic properties demonstrate that T-Null tendons are composed of a weaker matrix with less resistance to deformation and failure as well as more viscous behavior. Percent relaxation and dynamic modulus were only significantly different than controls at higher strains, suggesting that the early response to loading and associated mechanisms, such as uncrimping, are less affected by collagen V knockout, while fibrils exhibit weaker behavior at higher strains when they are more fully engaged. This is further corroborated by data from atomic force microscopy, which did not find differences in fibril deformation behavior up to 6 % strain. The dysregulation of fibril nucleation and lateral growth in the absence of collagen V therefore does not affect the quarter-staggered arrangement collagen triple helices within fibrils and how it deforms under load. In contrast with the T-Null tendons, inducible knockdown of collagen V did not affect mechanical properties of the supraspinatus tendon.

The observed deficits in tendon structure–function in the T-Null mice are consistent with studies investigating the same model at younger ages (Connizzo et al., 2015; Sun et al., 2015; Connizzo et al., 2016a,c). The aberrant tendon matrix that forms in the absence of collagen V is not naturally repaired after the matrix is developed. While tendon does employ intrinsic repair mechanisms for incurred microdamage (Stauber et al., 2020), these mechanisms were insufficient to make up for the dysregulated matrix assembly in T-Null tendons. Despite this major role for collagen V in tendon development with lasting impact into later ages, results from the inducible knockdown model show that collagen V has little role in maintaining a mature tendon matrix after it has been formed. This finding contrasts with other matrix proteins, such as biglycan, which are necessary to maintain properties of mature tendon (Beach et al., 2022; Darrieutort-Laffite et al., 2023). While it is often considered to be a relatively passive tissue, coordinated regulation of molecular and mechanical cues are necessary for homeostatic maintenance of tendon after development (Sharir and Zelzer, 2011; Grinstein et al., 2019; Wang et al., 2021; Montagna et al., 2022). The differences in this study between the phenotypes of T-Null and I-Null mice demonstrate that distinct mechanisms are involved in tendon development and homeostasis and, consequently, the necessity of distinguishing between them.

There are some limitations of the current work that could provide exciting areas for future research. There may be sex-dependent roles for collagen V that would require additional work in females to discover. Evaluation of gene expression was restricted to the pre-selected genes for the microfluidic array, and more comprehensive analysis of the tendon transcriptome using RNASeq could reveal changes that were not included in this study. In addition, gene expression changes were determined using a relatively small sample size, which may have limited the power to detect more subtle changes. Still, we were still able to find significant changes in the conditional knockout model that corresponded to differences in tendon structure and function. Further investigation could determine how these changes in gene expression translate into potential differences in protein composition, which were not evaluated here. In particular, changes in collagen V protein content were not determined in this study, although previous work demonstrated a reduction in collagen V protein in the T-Null model confirming the validity of the Col5a1f/f mouse (Sun et al., 2015). Although the overall quantity of collagen V in tendon is low, the reduction in Col5a1 expression between 150 and 300 days old in supraspinatus tendon indicates that there is ongoing production of collagen V into adulthood that would be knocked down in the inducible model (DiStefano et al., 2025a). Moreover, other factors may have influenced tendon properties in this study. Differences in body weight are largely accounted for by focusing on material properties that are not dependent on tendon size, yet changes in weight could still indirectly affect tendon function. Gait abnormalities in T-null mice (Sun et al., 2015) may have similarly affected tendon properties through differences in limb use or physical activity. Finally, different tendons often exhibit different behavior due to their specific functional requirements (Batson et al., 2003; Quigley et al., 2018; Hefferan et al., 2025), and other tendons may respond differently to the loss of Col5a1 expression.

This study demonstrates the lasting effects of developmental collagen V knockout on tendon health. Despite this major role for collagen V in tendon development with continued impact into later ages, results from the inducible knockdown model show that collagen V has little role in maintaining a mature tendon matrix after it has been formed. Consequently, the role for collagen V in tendon is primarily limited to developmental stages with only minor impacts in maturity. These findings continue to elucidate the functions of minor matrix molecules, and collagen V in particular, in different stages and additionally provide important information highlighting the need for early intervention strategies for cEDS patients.

Supplementary Material

1
2

Funding

This study was supported by the NIH/NIAMS (R01AR070750) and the Penn Center for Musculoskeletal Disorders (P30AR069619).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jbiomech.2025.112956.

Footnotes

CRediT authorship contribution statement

Jeremy D. Eekhoff: Writing – original draft, Supervision, Methodology, Investigation. Mitchell J. Hallman: Writing – review & editing, Investigation. Michael S. DiStefano: Writing – review & editing, Methodology, Investigation. Rebecca L. Betts: Writing – review & editing, Investigation. Courtney A. Nuss: Writing – review & editing, Investigation. Stephanie N. Weiss: Writing – review & editing, Resources. Andrew F. Kuntz: Writing – review & editing, Supervision. Louis J. Soslowsky: Writing – review & editing, Software, Project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Ansorge HL, Meng X, Zhang G, Veit G, Sun M, Klement JF, Beason DP, Soslowsky LJ, Koch M, Birk DE, 2009. Type XIV collagen regulates fibrillogenesis. J. Biol. Chem 284, 8427–8438. 10.1074/jbc.M805582200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Archambault JM, Jelinsky SA, Lake SP, Hill AA, Glaser DL, Soslowsky LJ, 2007. Rat supraspinatus tendon expresses cartilage markers with overuse. J. Orthop. Res 25, 617–624. 10.1002/jor.20347. [DOI] [PubMed] [Google Scholar]
  3. Batson EL, Paramour RJ, Smith TJ, Birch HL, Patterson-Kane JC, Goodship AE, 2003. Are the material properties and matrix composition of equine flexor and extensor tendons determined by their functions? Equine Vet. J 35, 314–318. 10.2746/042516403776148327. [DOI] [PubMed] [Google Scholar]
  4. Beach ZM, Bonilla KA, Dekhne MS, Sun M, Adams TH, Adams SM, Weiss SN, Rodriguez AB, Shetye SS, Birk DE, Soslowsky LJ, 2022. Biglycan has a major role in maintenance of mature tendon mechanics. J. Orthop. Res 40, 2546–2556. 10.1002/jor.25299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bowen JM, Sobey GJ, Burrows NP, Colombi M, Lavallee ME, Malfait F, Francomano CA, 2017. Ehlers–Danlos syndrome, classical type. Am. J. Med. Genet. C Semin. Med. Genet 175, 27–39. 10.1002/ajmg.c.31548. [DOI] [PubMed] [Google Scholar]
  6. Carlson JA, Shetye SS, Sun M, Weiss SN, Birk DE, Soslowsky LJ, 2024. Collagen V haploinsufficiency in female murine patellar tendons results in altered matrix engagement and cellular density, demonstrating decreased healing. J. Orthop. Res 42, 950–960. 10.1002/jor.25740. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Connizzo BK, Adams SM, Adams TH, Birk DE, Soslowsky LJ, 2016a. Collagen V expression is crucial in regional development of the supraspinatus tendon. J. Orthop. Res 34, 2154–2161. 10.1002/jor.23246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Connizzo BK, Adams SM, Adams TH, Jawad AF, Birk DE, Soslowsky LJ, 2016b. Multiscale regression modeling in mouse supraspinatus tendons reveals that dynamic processes act as mediators in structure–function relationships. J. Biomech 49, 1649–1657. 10.1016/j.jbiomech.2016.03.053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Connizzo BK, Freedman BR, Fried JH, Sun M, Birk DE, Soslowsky LJ, 2015. Regulatory role of collagen V in establishing mechanical properties of tendons and ligaments is tissue dependent. J. Orthop. Res 33, 882–888. 10.1002/jor.22893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Connizzo BK, Han L, Birk DE, Soslowsky LJ, 2016c. Collagen V-heterozygous and -null supraspinatus tendons exhibit altered dynamic mechanical behaviour at multiple hierarchical scales. Interface Focus 6, 20150043. 10.1098/rsfs.2015.0043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Darrieutort-Laffite C, Beach ZM, Weiss SN, Eekhoff JD, Soslowsky LJ, 2023. Knockdown of biglycan reveals an important role in maintenance of structural and mechanical properties during tendon aging. J. Orthop. Res 41, 2287–2294. 10.1002/jor.25536. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Del Buono A, Oliva F, Longo UG, Rodeo SA, Orchard J, Denaro V, Maffulli N, 2012. Metalloproteases and rotator cuff disease. J. Shoulder Elbow Surg 21, 200–208. 10.1016/j.jse.2011.10.020. [DOI] [PubMed] [Google Scholar]
  13. Derwin KA, Soslowsky LJ, 1999. A quantitative investigation of structure-function relationships in a tendon fascicle model. J. Biomech. Eng 121, 598–604. 10.1115/1.2800859. [DOI] [PubMed] [Google Scholar]
  14. DiStefano MS, Eekhoff JD, Weiss SN, Nuss CA, Betts RL, Kuntz AF, Soslowsky LJ, 2025a. Murine supraspinatus tendons demonstrate aging-related changes in multiscale mechanics, structure, and gene expression. J. Orthop. Res 10.1002/jor.70013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. DiStefano MS, Weiss SN, Nuss CA, Betts RL, Han B, Kuntz AF, Soslowsky LJ, 2025b. Mature murine supraspinatus tendons demonstrate regional differences in multiscale structure, function and gene expression. PLoS One 20, e0318809. 10.1371/journal.pone.0318809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dunkman AA, Buckley MR, Mienaltowski MJ, Adams SM, Thomas SJ, Satchell L, Kumar A, Pathmanathan L, Beason DP, Iozzo RV, Birk DE, Soslowsky LJ, 2013. Decorin expression is important for age-related changes in tendon structure and mechanical properties. Matrix Biol. 32, 3–13. 10.1016/j.matbio.2012.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Eekhoff JD, Abraham JA, Schott HR, Solon LF, Ulloa GE, Zellers JA, Cannon PC, Lake SP, 2023. Fascicular elastin within tendon contributes to the magnitude and modulus gradient of the elastic stress response across tendon type and species. Acta Biomater. 163, 91–105. 10.1016/j.actbio.2022.03.025. [DOI] [PubMed] [Google Scholar]
  18. Eekhoff JD, Fang F, Kahan LG, Espinosa G, Cocciolone AJ, Wagenseil JE, Mecham RP, Lake SP, 2017. Functionally distinct tendons from elastin haploinsufficient mice exhibit mild stiffening and tendon-specific structural alteration. J. Biomech. Eng 139, 111003. 10.1115/1.4037932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Eekhoff JD, Fang F, Lake SP, 2018. Multiscale mechanical effects of native collagen cross-linking in tendon. Connect. Tissue Res 59, 410–422. 10.1080/03008207.2018.1449837. [DOI] [PubMed] [Google Scholar]
  20. Eekhoff JD, Steenbock H, Berke IM, Brinckmann J, Yanagisawa H, Wagenseil JE, Lake SP, 2021. Dysregulated assembly of elastic fibers in fibulin-5 knockout mice results in a tendon-specific increase in elastic modulus. J. Mech. Behav. Biomed. Mater 113, 104134. 10.1016/j.jmbbm.2020.104134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Fessel G, Snedeker JG, 2009. Evidence against proteoglycan mediated collagen fibril load transmission and dynamic viscoelasticity in tendon. Matrix Biol. 28, 503–510. 10.1016/j.matbio.2009.08.002. [DOI] [PubMed] [Google Scholar]
  22. Flurkey K, Mcurrer J, Harrison D, 2007. Mouse Models in Aging Research, in: The Mouse in Biomedical Research. Elsevier, pp. 637–672. DOI: 10.1016/B978-012369454-6/50074-1. [DOI] [Google Scholar]
  23. Goh KL, Holmes DF, Lu Y, Purslow PP, Kadler KE, Bechet D, Wess TJ, 2012. Bimodal collagen fibril diameter distributions direct age-related variations in tendon resilience and resistance to rupture. J. Appl. Physiol 113, 878–888. 10.1152/japplphysiol.00258.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Gordon JA, Freedman BR, Zuskov A, Iozzo RV, Birk DE, Soslowsky LJ, 2015. Achilles tendons from decorin- and biglycan-null mouse models have inferior mechanical and structural properties predicted by an image-based empirical damage model. J. Biomech 48, 2110–2115. 10.1016/j.jbiomech.2015.02.058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Grinstein M, Dingwall HL, O’Connor LD, Zou K, Capellini TD, Galloway JL, 2019. A distinct transition from cell growth to physiological homeostasis in the tendon. eLife 8. DOI: 10.7554/elife.48689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hefferan SA, Blaker CL, Ashton DM, Little CB, Clarke EC, 2025. Structural variations of tendons: a systematic search and narrative review of histological differences between tendons, tendon regions, sex, and age. J. Orthop. Res 43, 994–1011. 10.1002/jor.26060. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ishizaki Y, Wang J, Kim J, Matsumoto T, Maeda E, 2024. Contributions of collagen and elastin to elastic behaviours of tendon fascicle. Acta Biomater. 176, 334–343. 10.1016/j.actbio.2024.01.014. [DOI] [PubMed] [Google Scholar]
  28. Izu Y, Adams SM, Connizzo BK, Beason DP, Soslowsky LJ, Koch M, Birk DE, 2021. Collagen XII mediated cellular and extracellular mechanisms regulate establishment of tendon structure and function. Matrix Biol. 95, 52–67. 10.1016/j.matbio.2020.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Izu Y, Ansorge HL, Zhang G, Soslowsky LJ, Bonaldo P, Chu M-L, Birk DE, 2011. Dysfunctional tendon collagen fibrillogenesis in collagen VI null mice. Matrix Biol. 30, 53–61. 10.1016/j.matbio.2010.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Johnston JM, Connizzo BK, Shetye SS, Robinson KA, Huegel J, Rodriguez AB, Sun M, Adams SM, Birk DE, Soslowsky LJ, 2017. Collagen V haploinsufficiency in a murine model of classic Ehlers–Danlos syndrome is associated with deficient structural and mechanical healing in tendons. J. Orthop. Res 35, 2707–2715. 10.1002/jor.23571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lavagnino M, Arnoczky SP, Frank K, Tian T, 2005. Collagen fibril diameter distribution does not reflect changes in the mechanical properties of in vitro stress-deprived tendons. J. Biomech 38, 69–75. 10.1016/j.jbiomech.2004.03.035. [DOI] [PubMed] [Google Scholar]
  32. Leiphart RJ, Weiss SN, DiStefano MS, Mavridis AA, Adams SA, Dyment NA, Soslowsky LJ, 2022. Collagen V deficiency during murine tendon healing results in distinct healing outcomes based on knockdown severity. J. Biomech 144, 111315. 10.1016/j.jbiomech.2022.111315. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Machol K, Polak U, Weisz-Hubshman M, Song I-W, Chen S, Jiang M-M, Chen-Evenson Y, Weis MAE, Keene DR, Eyre DR, Lee BH, 2022. Molecular alterations due to Col5a1 haploinsufficiency in a mouse model of classic Ehlers-Danlos syndrome. Hum. Mol. Genet 31, 1325–1335. 10.1093/hmg/ddab323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mak KM, Png CYM, Lee DJ, 2016. Type V collagen in health, disease, and fibrosis. Anat. Rec 299, 613–629. 10.1002/ar.23330. [DOI] [PubMed] [Google Scholar]
  35. Miller KS, Connizzo BK, Feeney E, Soslowsky LJ, 2012. Characterizing local collagen fiber re-alignment and crimp behavior throughout mechanical testing in a mature mouse supraspinatus tendon model. J. Biomech 45, 2061–2065. 10.1016/j.jbiomech.2012.06.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Montagna C, Svensson RB, Bayer ML, Rizza S, Maiani E, Yeung C-Y-C, Filomeni G, Kjær M, 2022. Autophagy guards tendon homeostasis. Cell Death Dis. 13. 10.1038/s41419-022-04824-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Nilsson N, Alim MDA, Dietrich-Zagonel F, Concaro S, Brorsson A, Nilsson Helander K, Eliasson P, 2024. The delayed presentation of achilles tendon ruptures is associated with marked alterations in the gene expression of COL1A1, MMPs, TIMPs, and IL-6. Am. J. Sports Med 52, 164–173. 10.1177/03635465231212669. [DOI] [PubMed] [Google Scholar]
  38. Peltz CD, Perry SM, Getz CL, Soslowsky LJ, 2009. Mechanical properties of the long-head of the biceps tendon are altered in the presence of rotator cuff tears in a rat model. J. Orthop. Res 27, 416–420. 10.1002/jor.20770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Planckaert G, Burssens A, Stappers F, Coudenys J, Demolder S, Kaya I, Van Der Linden M, Gonçalves A, Lemeire K, Pavie B, Van Ovost E, Burssens P, Vanhaecke A, Van Dorpe J, Pringels L, Wezenbeek E, Snedeker J, De Bock K, Bonar F, Cook J, Victor J, Elewaut D, Gracey E, 2025. AI-driven histologic analysis of human Achilles tendinopathy provides a roadmap to unravel pathogenesis. Ann. Rheum. Dis 84, 866–876. 10.1016/j.ard.2025.01.027. [DOI] [PubMed] [Google Scholar]
  40. Quigley AS, Bancelin S, Deska-Gauthier D, Légaré F, Kreplak L, Veres SP, 2018. In tendons, differing physiological requirements lead to functionally distinct nanostructures. Sci. Rep 8, 4409. 10.1038/s41598-018-22741-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Rigozzi S, Müller R, Snedeker JG, 2010. Collagen fibril morphology and mechanical properties of the Achilles tendon in two inbred mouse strains. J. Anat 216, 724–731. 10.1111/j.1469-7580.2010.01225.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Robinson KA, Sun M, Barnum CE, Weiss SN, Huegel J, Shetye SS, Lin L, Saez D, Adams SM, Iozzo RV, Soslowsky LJ, Birk DE, 2017. Decorin and biglycan are necessary for maintaining collagen fibril structure, fiber realignment, and mechanical properties of mature tendons. Matrix Biol. 64, 81–93. 10.1016/j.matbio.2017.08.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sharir A, Zelzer E, 2011. Tendon homeostasis: the right pull. Curr. Biol 21, R472–R474. 10.1016/j.cub.2011.05.025. [DOI] [PubMed] [Google Scholar]
  44. Shirley ED, DeMaio M, Bodurtha J, 2012. Ehlers-Danlos syndrome in orthopaedics: etiology, diagnosis, and treatment implications. Sports Health Multidiscip. Approach 4, 394–403. 10.1177/1941738112452385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Silva Barreto I, Pierantoni M, Hammerman M, Törnquist E, Le Cann S, Diaz A, Engqvist J, Liebi M, Eliasson P, Isaksson H, 2023. Nanoscale characterization of collagen structural responses to in situ loading in rat Achilles tendons. Matrix Biol. 115, 32–47. 10.1016/j.matbio.2022.11.006. [DOI] [PubMed] [Google Scholar]
  46. Stauber T, Blache U, Snedeker JG, 2020. Tendon tissue microdamage and the limits of intrinsic repair. Matrix Biol. 85–86, 68–79. 10.1016/j.matbio.2019.07.008. [DOI] [PubMed] [Google Scholar]
  47. Sun HB, Li Y, Fung DT, Majeska RJ, Schaffler MB, Flatow EL, 2008. Coordinate regulation of IL-1β and MMP-13 in rat tendons following subrupture fatigue damage. Clin. Orthop 466, 1555–1561. 10.1007/s11999-008-0278-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sun M, Connizzo BK, Adams SM, Freedman BR, Wenstrup RJ, Soslowsky LJ, Birk DE, 2015. Targeted deletion of collagen V in tendons and ligaments results in a classic Ehlers-Danlos syndrome joint phenotype. Am. J. Pathol 185, 1436–1447. 10.1016/j.ajpath.2015.01.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Szczesny SE, Elliott DM, 2014. Interfibrillar shear stress is the loading mechanism of collagen fibrils in tendon. Acta Biomater. 10, 2582–2590. 10.1016/j.actbio.2014.01.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Tam KT, Baar K, 2025. Using load to improve tendon/ligament tissue engineering and develop novel treatments for tendinopathy. Matrix Biol. 135, 39–54. 10.1016/j.matbio.2024.12.001. [DOI] [PubMed] [Google Scholar]
  51. Wang T, Wagner A, Gehwolf R, Yan W, Passini FS, Thien C, Weissenbacher N, Lin Z, Lehner C, Teng H, Wittner C, Zheng Q, Dai J, Ni M, Wang A, Papadimitriou J, Leys T, Tuan RS, Senck S, Snedeker JG, Tempfer H, Jiang Q, Zheng MH, Traweger A, 2021. Load-induced regulation of tendon homeostasis by SPARC, a genetic predisposition factor for tendon and ligament injuries. Sci. Transl. Med 13. 10.1126/scitranslmed.abe5738. [DOI] [PubMed] [Google Scholar]
  52. Wenstrup RJ, Smith SM, Florer JB, Zhang G, Beason DP, Seegmiller RE, Soslowsky LJ, Birk DE, 2011. Regulation of Collagen fibril nucleation and initial fibril assembly involves coordinate interactions with collagens V and XI in developing tendon. J. Biol. Chem 286, 20455–20465. 10.1074/jbc.M111.223693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Xu Y, Murrell GAC, 2008. The basic science of tendinopathy. Clin. Orthop 466, 1528–1538. 10.1007/s11999-008-0286-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2162669-supplement-1.docx (527.5KB, docx)
2
NIHMS2162669-supplement-2.xlsx (44.3KB, xlsx)

RESOURCES