Early delivery of human umbilical cord mesenchymal stem cells improves healing in a rat model of Achilles tendinopathy

Tendinopathy is a common musculoskeletal disorder that arises due to overuse, injury and excessive use of certain medications, such as corticosteroids and fluoroquinolones [1,2]. Clinical manifestations of tendinopathy include pain and swelling around the tendon, often accompanied by functional limitations that significantly impact daily activities [1]. The Achilles tendon, being the largest tendon in the human body, plays a crucial role in functions such as standing and walking [3]. Achilles tendinopathy has a lifetime prevalence of 23.9% in athletes and 5.9% in the general population [4]. Unfortunately, the healing ability of tendon tissue is limited due to the absence of adequate cells and blood vessels [5]. Traditional conservative therapies, including nonsteroidal anti-inflammatory drugs, glucocorticoids, extracorporeal shockwave therapy and eccentric exercise, can alleviate symptoms but fall short of repairing the injured tendon [6–8].

Mesenchymal stem cells (MSCs) derived from human umbilical cords exhibit strong cloning and proliferation abilities, along with multiple differentiation and immune regulation capabilities, and possess an immune-exempt status [9,10]. Numerous clinical studies have corroborated their safety [11]. Furthermore, the use of these stem cells encounters less controversy, given recognition of the human umbilical cord as medical waste. These advantages render hUC-MSCs suitable for diseases resulting from chronic inflammation and associated tissue damage [12].

While some studies suggest that MSC transplantation alone may not effectively repair Achilles tendon injury [13,14], others report contrasting results [15–17]. These discrepancies may be attributed to variations in the timing of stem cell delivery. Positive effects were consistently observed in studies where injections were administered during the early stages of tendinopathy, while injections given during advanced stages demonstrated limited efficacy. Consequently, the experiments in this study were designed to explore the efficacy of MSC intervention at different stages and its potential therapeutic mechanism.

Methods

Achilles tendinopathy model

In this study, 39 male Sprague–Dawley rats weighing between 250 and 300 g were used. All the animals were sourced from the Animal Laboratory Center of Sun Yat-sen University and housed under controlled conditions, with a 12-h light–dark cycle, a laboratory temperature maintained at 25 ± 1°C, and a relative humidity range of 50–70%. The study received approval from the Animal Ethics Committee of Sun Yat-sen University (SYSU-IACACC-2022-000200). Treatment of the rats adhered to pertinent guidelines and regulations, ensuring ethical standards were maintained throughout the study. Furthermore, the research strictly complied with ARRIVE guidelines, promoting transparency and rigor in reporting.

Two rats were accommodated per cage; a 1-week acclimatization period was allowed, during which all rats had unrestricted access to food and water. Of the 39 rats, 32 underwent the chemical injury method [18,19] to establish Achilles tendinopathy models, while the remaining seven served as normal controls. The Achilles tendinopathy models were established by inducing anesthesia with 2% sodium pentobarbital and carefully inserting a 4.5-UI needle into the Achilles tendon, positioned 3 mm above the calcaneal and advanced 5 mm along the long axis. Subsequently, 50 μl of collagenase I solution (10 mg/ml; Sigma-Aldrich, Shanghai, China) was injected, which was based on previous relevant studies [20,21] and our pre-experiments. The needle was retracted slowly during the injection process, and this procedure was performed on both sides of the Achilles tendons. A successful model was established after 1 week and 4 weeks of one injection (Figure 1). To ensure the robustness of the modeling results, two rats were utilized for each time point (Supplementary File 1.1). All the models completed strictly according to this protocol were considered successful.

MSC delivery & sampling

Following the successful establishment of the tendinopathy model, 28 rats were randomly allocated into four groups using a random-number table: MSC-1 (MSC transplantation 1 week after collagenase I injection), control-1 (saline solution injection 1 week after collagenase I injection), MSC-2 (MSC transplantation 4 weeks after collagenase I injection) and control-2 (saline solution injection 4 weeks after collagenase I injection). Fifth-generation hUC-MSCs sourced from the SALIAI Stem Cells Company (G05003) were employed, with 5 × 10⁵ stem cells dissolved in 50 μl of normal saline injected into each Achilles tendon using 4.5 UI needles for the MSC groups. The control group received a 50-μl injection of normal saline. Stem cells or saline were delivered into and around bilateral Achilles tendons twice, with a 1-week interval between administrations. To mitigate subjective bias, the operator performing the injections was blinded to the injection contents. Sampling was conducted 4 weeks after the conclusion of the second treatment (Figure 1). During sampling, the right Achilles tendons of seven rats in each group underwent a biomechanics assay. Four left Achilles tendons from each group were dissected in the middle, with one-half allocated for histological analysis and the other half for gene expression analysis. Another three left Achilles tendons from each group were designated for transmission electron microscopy. It is noteworthy that all assessors involved in subsequent analyses were blinded to the group affiliations of the samples or sections.

Hematoxylin & eosin staining

Upon sampling, the Achilles tendons were promptly immersed in a dedicated tissue-specific fixative (Servicebio, Wuhan, China) for 48 h. Subsequently, the tendons underwent dehydration using sucrose solution and paraffin embedding, and were sliced into 10-μm sections for staining with hematoxylin and eosin reagent. Histological assessment of the tendon involved the evaluation of six parameters [22,23]: collagen fiber structure, collagen fiber arrangement, nuclear roundness, inflammation, vascular proliferation and cell density. The cumulative histological score, ranging from 0 to 18, serves as an indicator of the tendon's condition, with lower scores indicative of a closer resemblance to a normal Achilles tendon. Three expert assessors, blinded to the group assignments, conducted the assessments. This comprehensive histological scoring system provides a multifaceted understanding of the tendon's histopathological characteristics.

RT-qPCR

For RT-qPCR analysis, Achilles tendon samples were meticulously minced and ground with grinding beads in a frozen mill. Total RNA extraction was carried out using Trizol (Invitrogen, Shanghai, China). Subsequently, first-Strand cDNA was synthesized using the SuperScript III First-Strand Synthesis Kit (Invitrogen) following the manufacturer's protocol. The resultant cDNA served as the template for subsequent RT-qPCR, employing iQ SYBR Green Supermix (Bio-Rad, Guangzhou, China). Primers (Table 1) were synthesized by Shanghai Sangon Biotechnology Co., Ltd (Shanghai, China). The relative expression levels of PCR products were determined using the 2^-ΔΔCT method, with calculations carried out relative to the Control-1 group.

Immunohistochemical examinations

The 10-μm sections underwent deparaffinization and were subsequently treated with 10% hydrogen peroxide. Antigen retrieval was performed by heating the sections in a 0.01-mol/l sodium citrate solution at multiple intervals. Following this, blocking with a blocking solution ensued for 1 h. Subsequently, the sections were incubated overnight at 4°C with primary antibodies targeting collagen I, collagen III, MMP-3 and TIMP-3 (Supplementary File 1.2). After thorough washing to remove the primary antibodies, the cells were stained using the DAB chromogenic kit (Gene Tech, Beijing, China), and the nuclei were counterstained with hematoxylin. Hematoxylin-stained nuclei appeared blue, while positive DAB expression manifested as brown. Each type of antibody underwent staining simultaneously in one batch, and all samples were imaged with identical microscope settings at the same time for consistency. The mean optical density of positive expression in four randomly selected fields at 200× magnification was analyzed using ImageJ 1.53c. The optical density of the whole staining area and the integrated density of positive staining were measured. The mean optical density (integrated density/whole staining area) was then calculated as the outcome.

Transmission electron microscope assay

Fresh tissue underwent fixation using a transmission electron microscope-specific fixative (Servicebio), followed by dehydration, embedding and sectioning into 60-nm ultrathin sections. These sections were stained with 2% uranium acetate-saturated alcohol solution for 8 min in the absence of light. Subsequently, they were rinsed three times in 70% ethanol and then three times in ultrapure water. A second staining step involved immersion in 2.6% lead citrate for 8 min, followed by another three rinses in ultrapure water. After drying with filter paper, the cuprum grids were carefully placed on a grid board and air-dried overnight at room temperature. The stainied sections were then examined and photographed using a transmission electron microscope (HITACHI, Tokyo, Japan). ImageJ 1.53c was employed for the quantitative analysis of collagen density, diameter and area in the cross-section of the Achilles tendon. Three randomly selected images were analyzed for each sample, with approximately 40 fibrils measured per image.

Biomechanical testing

The Achilles tendon was detached from the calcaneus and immersed in saline. Both ends of the tendon were secured onto a jig, with gauze evenly wrapped to prevent slippage and breakage at the clamps. Tensile testing was performed using a Zhiqu ZQ990 tensile machine. The Achilles tendon was stretched at a controlled speed of 1 mm/s until fracture occurred, with a preload of 2N applied. The stress–strain curve derived from the testing provided essential parameters such as the maximum force and stiffness (calculated as the maximum force divided by the deformation at maximum force).

Statistical analysis

Statistical analysis of the data was conducted using R language (version 4.3.2). GraphPad Prism (version 9.4.1) was employed for the visualization of statistics charts. Data were presented as mean ± standard deviation, and qualitative data were expressed as numbers and percentages. The nlme package in R was utilized for the implementation of the linear mixed effects model analysis, treating individuals as random effects and groups as fixed effects. Subsequently, the emmeans package was used for post-pairwise comparisons among MSC and control groups, with values adjusted based on the false discovery rate method. A significance level of p < 0.05 was considered statistically significant.

Results

Histological performance improvement due to MSC delivery

As depicted in Figure 2, the normal Achilles tendon exhibited a characteristic parallel collagen arrangement with elongated fusiform nuclei interspersed among the fibrils. In the control groups, collagen arrangement became disordered, accompanied by inflammatory cell infiltration and the emergence of calcified regions, leading to an increased number of tendon cells. Following the injection of hUC-MSCs, a notable improvement in fiber arrangement was observed compared with the control groups. The cell count was reduced and a remarkable absence of calcium salt deposition was noted in the MSC groups in contrast to the control groups. Semiquantitative scores for the MSC groups were significantly lower than those for the control groups (both p < 0.01), with the MSC-1 group yielding lower scores than the MSC-2 group (p < 0.05).

mRNA level changes induced by MSC delivery

Figure 3 illustrates the PCR results, revealing distinctive expression patterns. In both MSC-1 and MSC-2 groups, the expressions of TNF-α and IL-1β were significantly lower than those in the control-1 and control-2 groups (p < 0.01, p < 0.05 for MSC-1 and both p < 0.05 for MSC-2). Notably, the expression in the MSC-1 group was slightly lower than that in the MSC-2 group, although this difference did not reach significance. This collective data suggest a substantial reduction in inflammation following the administration of stem cells, both in the early and advanced stages of Achilles tendinopathy. Examining MMP3 expression, the MSC-1 group exhibited significantly lower levels compared with the control-1 (p < 0.05) and MSC-2 (p < 0.05) groups. In addition, the MSC-2 group demonstrated markedly lower expression than the control-2 group (p < 0.01). In contrast, TIMP3 expression in the MSC-1 group was significantly higher than that in the control-1 (p < 0.01) and MSC-2 (p < 0.05) groups. The MSC-2 group also exhibited higher expression compared with the control-2 group (p < 0.05).

Promotion of normal collagen matrix synthesis by MSC delivery

As depicted in Figure 4, collagen I, being the primary component of normal tendon tissue, exhibited higher expression in both the MSC-1 (p < 0.01) and MSC-2 (p < 0.01) groups compared with the control-1 and control-2 groups, respectively. Remarkably, the MSC-1 group demonstrated significantly greater expression than the MSC-2 group (p < 0.05). Conversely, collagen III, associated with Achilles tendon fibrosis, exhibited reduced expression in both MSC-1 (p < 0.01) and MSC-2 (p < 0.05) groups compared with their respective control groups. It is noteworthy that exercising the collagen antibody faced challenges binding to mature collagen fibrils within the matrix, thereby enabling mainly staining for newly synthesized collagen, but the effect of staining for part mature collagen should be considered. MMP-3 expression followed a similar trend, being lower in the MSC groups (both p < 0.01) compared with their respective control groups. In contrast, TIMP-3 expression showed an opposing trend with higher expressions in the MSC-1 (p < 0.01) and MSC-2 groups (p < 0.05) compared with their respective control groups. Notably, the MSC-1 group exhibited higher expression compared with the MSC-2 group, but the difference was not statistically significant (p > 0.05).

Effect of MSC delivery on tendon fibrils thickness

Collagen I, characterized by larger fibers than collagen III, was examined in a transverse section of the Achilles tendon using transmission electron microscopy, as illustrated in Figure 5. In normal and MSC tissues, most fibrils exhibited a larger size, while control groups generally featured smaller fibrils. The results demonstrated that both the MSC-1 (both p < 0.01) and MSC-2 (both p < 0.01) groups displayed significantly larger diameters and a larger average cross-sectional area of collagen fibrils compared with their respective control groups. Moreover, the MSC-1 group exhibited larger diameters and a larger average cross-sectional area of collagen fibrils than the MSC-2 group (both p < 0.05). The density of fibrils in both the MSC-1 (p < 0.01) and MSC-2 (p < 0.01) groups were lower than that of their respective control groups, with no significant difference observed between the MSC groups.

Enchancement of tendon strength by MSC delivery

All tendon ruptures occurred between the two clips. As illustrated in Figure 6, the maximum force and stiffness observed in the MSC and control groups were notably lower than those in the normal group. However, the MSC-1 group exhibited a significantly higher maximum force than the control-1 group (p < 0.05), and the MSC-2 group demonstrated a slightly higher maximum force than the control-2 group (p > 0.05). The difference in maximum force between MSC-1 and MSC-2 groups was not significant (p > 0.05). Furthermore, the stiffness of the MSC-1 group was significantly higher than the both the control-1 (p < 0.01) and MSC-2 (p < 0.05) groups.

Discussion

Tendon injuries generally include acute rupture and chronic tendinopathy (or tendinosis) [24]. Regarding this study, we established collagenase-induced tendinopathy models through chemical injury. Tendinopathy arises from the failure of cumulative microinjuries to heal, primarily due to the scarcity of cells and blood vessels [5,25]. The injection of collagenase into the Achilles tendon triggers an inflammatory response, characterized by an acute phase within 3–7 days (early) followed by a proliferative phase lasting approximately 3 weeks, with subsequent remodeling occurring at 4–8 weeks after injury (advanced) [24,26]. The timing of treatment in this study was strategically chosen based on this process [19]. Previous research has demonstrated that the application of MSCs can enhance tendon healing through paracrine effects, influencing the associated inflammatory response and matrix remodeling [27]. Our study revealed that the administration of hUC-MSCs, both in the early and advanced stages of tendinopathy, ameliorated the pathological features of the Achilles tendon. Following early intervention, the Achilles tendon exhibited elevated expression of type I collagen, larger collagen fibrils and improved biomechanical properties.

During the early stages of tendon injury, there is a decrease in type I collagen and an increase in type III collagen, leading to scar tissue formation, eventually replaced by type I collagen in the advanced stage [27,28]. Prolonged absence of quick repairs can result in an abnormal collagen I/III ratio, potentially compromising tendon strength, causing mechanical microdamage and inducing degeneration [29]. Studies have suggested that stem cell therapy can stimulate tendon cells, accelerate scar tissue formation, restore the collagen ratio, normalize tendon structure early and facilitate the recovery of tendon strength, thus preventing mechanical microinjury and further degenerative changes [16,30]. In this study, the synthesis of type I collagen increased, and the proportion of type III collagen decreased in the stem cell group, with a more pronounced effect observed with early delivery of hUC-MSCs.

Our findings demonstrate that treatment with hUC-MSCs effectively reduced the expression of inflammatory factors IL-1β and TNF-α in injured Achilles tendons, with early MSC delivery exhibiting a potentially greater efficacy in mitigating inflammation. Following tendon injury, M1 macrophages predominantly assume a proinflammatory role by releasing cytokines such as IL-1β, TNF-α and IL-6. This response is followed by the accumulation of M2 macrophages, which play an anti-inflammatory role by releasing cytokines such as IL-10 and TGF-β [28]. Prior studies have indicated that MSCs possess the ability to regulate the inflammatory response by inhibiting the pro-M1 phenotype and promoting the polarization of macrophages to the M2 phenotype [31,32]. Moreover, matrix synthesis and degradation in healthy tendons are tightly regulated by MMP and TIMP. Overexpression of MMP-3 can stimulate collagen matrix degradation, whereas TIMP-3 inhibits the activity of several MMPs [33,34]. Our results indicate that the MSC group exhibited lower MMP-3 expression and higher TIMP-3 expression compared with the control group. This trend was more prominent in the MSC-1 group than in the MSC-2 group. Therefore, it may be inferred that early intervention not only reduces inflammation but also influences the expression of MMP and TIMP, ultimately promoting the synthesis of type I collagen.

Injured tendons harbor progenitor cells with robust chondrogenic potential, rendering them susceptible to heterotopic ossification in an inflammatory milieu [35]. Histological assessments revealed degenerative changes in tendons of the control group. The transplantation of MSCs demonstrated a noteworthy reduction in inflammation and a potential decrease in calcification deposition in the MSC groups during tendon repair.

The microenvironment plays a pivotal role in inflammation and matrix synthesis during various stages of tendinopathy. While numerous studies have investigated MSC delivery for treating tendinopathy [36], our research introduces a novel aspect by examining the impact of MSC delivery time, representing a significant innovation. Our study provides a preliminarily exploration of the therapeutic effects of MSCs in both early and advanced tendinopathy stages, shedding light on their impact on inflammation and collagen synthesis. In future investigations, we intend to delve into the interactive effects between the microenvironment at different stages of tendinopathy and MSCs, paving the way to identify potential strategies for further enhancing the efficacy of MSCs.

Despite the valuable insights gained from this study, certain limitations should be acknowledged. First, while the collagenase model can serve as a reference, it is not a perfect simulation of human Achilles tendinopathy. Second, the study design lacked a sufficient number of time points, limiting the assessment of both early and prolonged performances of hUC-MSCs in treating Achilles tendinopathy. An early time point to explore inflammation and a prolonged time point to assess therapeutic effect should be added in the future work. Finally, young collagenase-induced tendinopathy rats may not fully mirror the overuse-induced tendinopathy seen in humans. Although tendon degeneration is a common manifestation in both cases, including features such as cell and vascular hyperplasia, disordered collagen fibers and heterotopic ossification [16,37], caution should be exercised in extrapolating the results of this research to human diseases.

Conclusion

The administration of hUC-MSCs proves to be a promising intervention for ameliorating the pathological features of collagenase-induced Achilles tendinopathy in rats. This therapeutic approach effectively enhances collagen matrix synthesis and remodeling, mitigates local inflammation and improves the biomechanical properties of the Achilles tendon. Importantly, our findings underscore the greater efficacy hUC-MSC interventional therapy in the early stage of Achilles tendinopathy compared with the advanced stage.