A distinct transition from cell growth to physiological homeostasis in the tendon

Abstract

Changes in cell proliferation define transitions from tissue growth to physiological homeostasis. In tendons, a highly organized extracellular matrix undergoes significant postnatal expansion to drive growth, but once formed, it appears to undergo little turnover. However, tendon cell activity during growth and homeostatic maintenance is less well defined. Using complementary methods of genetic H2B-GFP pulse-chase labeling and BrdU incorporation in mice, we show significant postnatal tendon cell proliferation, correlating with longitudinal Achilles tendon growth. Around day 21, there is a transition in cell turnover with a significant decline in proliferation. After this time, we find low amounts of homeostatic tendon cell proliferation from 3 to 20 months. These results demonstrate that tendons harbor significant postnatal mitotic activity, and limited, but detectable activity in adult and aged stages. It also points towards the possibility that the adult tendon harbors resident tendon progenitor populations, which would have important therapeutic implications.

Keywords: developmental biology; homeostasis; mouse; musculoskeletal; regenerative medicine; stem cells; tendon; tissue growth.

Figure 1.. H2B-GFP expression is induced upon the addition of Dox to timed pregnant females from E10 to P0 (dark green); Dox is removed for the chase period of 0–2 years and H2B-GFP protein is diluted (light green) in proportion to cell division.

(A) At birth (P0), longitudinal sections of Achilles (B–B’’) tendons (n = 3 mice) show extensive H2B-GFP⁺ (green B, B'; white B’’) labeling of Hoechst⁺ nuclei (blue, B, B’). SHG is shown in white (B, C). Histogram showing that more than 95% of the cells are H2B-GFP⁺ at P0 (D). After 680 days, Achilles (C–C’’) tendons (n = 3 mice) have qualitatively fewer H2B-GFP⁺ (green C, C'; white C’’) labeled Hoechst⁺ nuclei (blue C, C’). Histogram showing only 20% of the cells are H2B-GFP⁺ and the H2B-GFP intensity has decreased at 645 days (E). For the histograms, a representative is shown; tendons from n > 3 mice were examined independently. Scale Bars, 50 µm; Vertical red lines (H, O) indicate the control GFP beads for standardizing intensity and gates.

Figure 1—figure supplement 1.. Flow cytometry analysis of H2B-GFP⁺ tendon cells showing the exclusion of CD31⁺ and CD45⁺ cells; CD31^-/CD45^- cells (80–90%) were used for our H2B-GFP analysis.

(A) A representative flow cytometry analysis of H2B-GFP expression in a H2B-GFP heterozygous mouse that never received Dox, indicating a relatively small amount (<1%) of transgene expression occurs in the absence of Dox induction (B). n = 3 mice per experiment.

Figure 2.. Analysis of H2B-GFP⁺ cells from postnatal to aged stages.

The percentage of H2B-GFP⁺ tendon cells decreases from P0 to P645, with a rapid decline of H2B-GFP⁺ tendon cells by 14 days post birth and a more graduate decrease in H2B-GFP⁺ cells from P14 to P645 (A). Analysis of H2B-GFP⁺ tendon cells shows logarithmic decay (log2) in the percentage of H2B-GFP⁺ cells (B). H2B-GFP⁺ tendon cells decrease in their intensity of GFP from P0 to P645 (C; stages examined are labeled and shown with different shades of green; histogram is a representative from n = 4 mice). Using the change in H2B-GFP intensity, a model was used to calculate the daily cell proliferation rates (D), which shows significantly higher proliferation rates from P0 to P21, and lower proliferation rates at P80 and P600. Tendons from n > 3 mice were examined per each stage shown on the graph. All error bars represent standard deviation.

Figure 3.. Analysis of tendon cell proliferation using short and long pulses of BrdU/EdU.

Examination of BrdU incorporation 24 hr after BrdU injection reveals higher BrdU incorporation rates at P0 through P14 compared with P21 and later stages (A). Representative flow cytometry of 24 hr BrdU labeled Scx-Cre;TdTom⁺ tendon cells shows 30.3% BrdU⁺/TdTom⁺ cells at P8 and 6.65% BrdU⁺/TdTom⁺ cells at P28 (B). 24 hr EdU labeling identifies a significant number of proliferating cells at P2 (C, A’, B’), but no EdU⁺ cells were observed at P60 (C, C’,D’). Administration of BrdU for over 90 days after P30 shows low but detectable levels of BrdU incorporation from 120 to 400 days (D). Example flow cytometry shows 2.18% BrdU⁺/TdTom⁺ cells after 3 months of BrdU incorporation (1–4 months, (E). Quantification of BrdU⁺/TdTom⁺ cells in tendon sections (F) shows a similar percentage of incorporation as the flow cytometry analysis and significantly more BrdU incorporating nuclei in outer Achilles tendon regions compared with inner regions at either 3 and 6 months. For all experiments, n = 3 mice were examined per stage. Error bars are standard deviation. Scale bars = 100 µm.

Figure 3—figure supplement 1.. No significant differences in the percentage of BrdU⁺ cells were observed between Scx-Cre;TdTom⁺ cells and Scx-GFP⁺ cells after 120 days of BrdU incorporation.

n = 3 mice per experiment. Error bars are standard deviation.

Figure 4.. Expression of tendon and matrix related genes changes during the transition in cell division rate.

RT-qPCR of selected markers of proliferation (Mki67; blue), tendon transcription factors (TFs, Scx and Mkx; green), and extracellular matrix (Col1a2, Col3a1, and Fmod; gray). Relative expression was calculated using the ∆C_T method using Gapdh as the reference gene. For all genes assayed, significant differences between all six time points were found via ANOVA (p<0.05). Stars indicate significant differences based on Tukey’s HSD compared to P0 only (*p<0.05; **p<0.01; ***p<0.001). See Supplementary file 1 for ANOVA statistics and full report of post hoc pairwise comparisons. n = 3 biological replicates per time point. Boxplot edges represent the interquartile range (IQR) and the middle line represents the median. Whiskers represent 1.5 x IQR.

Figure 5.. Tendon growth corresponds to changes in tendon cell density and matrix organization.

Two photon microscopy images of Achilles tendons showing second harmonic generation signal (purple, A-D’; white, A’’’–D’’’), Hoechst⁺ nuclei (white, A–D’), and Scx-GFP⁺ cells (green, A–D, A’’–D’’). Images show changes in cell density, cell shape, and collagen organization at P0 (A–A’’’), P7 (B–B’’’), P14 (C–C’’’), and P28 (D–D’’’). Using optical sections from 2-photon images, we found that tendon cell density significantly decreases at postnatal stages (E, each point represents cell counts from an optical section; n = 3 mice analyzed per stage). The Achilles tendon undergoes significant longitudinal growth from P0 to P30; no significant increases in Achilles length is detected after P30 (F, each point represents measurements from one mouse Achilles; n > 3 mice). Error bars are standard deviation. Scale bars = 100 µm.

Figure 6.. Our model of tendon growth shows significant proliferation prior to P30 that coincides with expansion and maturation of the tendon matrix.

By P30, tendon cells transition from relatively high to very low rates of proliferation, signifying a transition from growth to physiological tissue maintenance. Others have shown that matrix maturation and expansion continue after P30, but these changes do not involve proliferation-driven cell growth or longitudinal growth of the tendon.