Mesenchymal stem (stromal) cells are known to regenerate defects of bone, cartilage, and tendon. In contrast to hematopoietic stem cells, there has never been a consensus on how to unambiguously define MSCs. To remedy this, the International Society of Cellular Therapy proposed in a consensus conference a set of criteria describing human MSCs.1 According to this definition, human MSCs grow adherently to plastic and express the cell surface proteins CD73, CD90, and CD 105 but lack expression of CD 14, CD34, and CD45. Furthermore, MSCs are able to differentiate in vitro into at least 3 mesenchymal lineages, such as osteoblasts, adipocytes, and chondrocytes.1 In mice and humans, such MSCs have been isolated from several types of tissues, including BM, periosteum, fetal tissues, and AT.2 In the study reported here, we therefore set out to investigate the expression of the aforementioned characteristic surface antigens on equine AT-derived MSCs and their in vitro differentiation potential in more detail.
Equine MSCs have been isolated from BM,3,4 AT,5-7 PB,8,9 and umbilical cord blood.610 The AT above the dorsal gluteal muscle allows simple access for biopsy specimen collection, making such AT-derived MSCs a convenient source of cells for therapeutic applications. For example, successful resolution of articular defects after application of equine BM-derived MSCs has been reported,11 but the applied MSCs were not characterized in detail. Such detail may be important, however, because bleeding was provoked during the preparation of the cartilage defect.11 Therefore, it could not be established whether the cartilage repair was attributable to MSCs entering the defect directly from BM with the bleeding or MSCs being expanded in vitro and transplanted into the defect. This principle of cartilage repair by iatrogenic bleeding and infiltration of stem cells without transplantation of additional cells has been an established therapy for decades.12 But there have been few randomized, double-blinded studies13 in which various treatments, including cellular treatment, have been evaluated in large animals.
Several studies have been carried out to evaluate various biological properties of equine MSCs, including their long-term expansion, differentiation potential, chondrogenesis in hydrogel culture, and expression of stem cell markers. According to the associated reports, equine AT-derived MSCs express Oct-4, TRA1-60, and TRA1-816 and have a weaker general differentiation potential and markedly slower cell division than equine BM-derived MSCs.3,7 However, the expansion of MSCs for clinical purposes may be limited by cellular senescence after expansion of cells in vitro, including replicative senescence associated with the shortening of telomeres with time.14
In addition, induction of adipogenesis in equine MSCs is rather difficult.6 The addition of rabbit serum to the adipogenic induction medium yields adipocyte-like phenotypes but no maturation of adipocytes.3 Thus, equine MSCs seem to be characterized to some extent functionally, but less thoroughly, with respect to their phenotype and the expression of typical cell surface antigens. We therefore also set out to evaluate the proliferative behavior, telomere length, immunophenotype, and differentiation capacity of AT-derived MSCs in more detail in accordance with the recently defined criteria.1
Materials and Methods
Isolation and expansion of MSCs—The MSCs were isolated from the AT of 6 live healthy adult horses as described elsewhere.7 Briefly, tissue aspirates were minced and digested with 0.075% collagenase IIa at 37°C for 45 to 60 minutes. Then, AT-derived MSCs were collected by centrifugation (600 × g for 5 minutes at 20°C). The cells were resuspended in 10 mL of basal medium (Dulbecco modified Eagle medium with a low glucose concentration and without glutamine, plus 25mM HEPES),b 10% fetal bovine serum,c 2mM glutamine,d 100 U of penicillin-streptavidine/mL, and 500 ng of partricine/mL. The cell suspension was filtered through a series of 100-, 70-, and 40-μm filters,f centrifuged (600 × g for 5 minutes), and resuspended in 10 mL of basal medium. Cells were subsequently counted and plated onto flasksg at a cell density of approximately 20,000 cells/cm2.
Equine BM- and PB-derived MSCs were isolated in accordance with published protocols3,9 and characterized as described previously. For cultivation of PB-derived MSCs, basal medium was enriched with 20% fetal bovine serum.9
Cell proliferation—Cells were detached enzymaticallyh washed with PBS solution, and counted with a hemocytometer at each in vitro culture passage, each of which is identified here as P followed by the passage number (eg, PI is the first passage). Viability was evaluated by means of trypan blue dye exclusion. Numbers of viable MSCs were converted to duplication rates (ie, population duplications per day) by use of the following equation:
Duplication rate = (log[Nx/Nx+1]/log[2]/T
in which Nx is the cell number at passage x, Nx+1 is the cell number at passage x+1, and T is the passage culture duration in days. For further expansion, cells were plated in new flasks at a density of 4,000 cells/cm2.
Telomere length assay—Telomere lengths of the AT-derived MSCs were measured with a fluorescein isothiocyanate-labeled, probe-based peptide nucleic acid kiti in accordance with the manufacturer's protocol. The fluorescence intensities of the AT-derived MSCs were calculated relative to those of control cell line T-cell lymphoma 1301. The 1301 cells have a defined telomere length of 30 kbp. Consequently, the telomere lengths of AT-derived MSCs can be computed as absolute values. Measurements were performed by use of the FCM equipment^ and data were analyzed with associated software.k
Differentiation of equine MSCs—Differentiation of AT-derived MSCs was induced at the second and fourth passages (P2 and P4) to generate osteogenic, adipogenic, and chondrogenic lineages. For osteogenic and adipogenic differentiation, the cells were plated onto 6-well plates1 in basal medium at a starting density of 5,000 cells/cm2. After 24 hours of incubation, the basal medium was replaced with the respective differentiation medium. The osteogenic medium was basal medium supplemented with 100μM ascorbic acid,a 10mM β-glycerophosphate,a and 1μM dexamethasone.a The adipogenic medium contained 4.5 g of D-glucose7L, 100μM indomethacine,m 10 μg of insulina/mL, 0.5mM 3-isobutyl-1-methylxanthine,a 1μM dexamethasone,a and 5% rabbit serum.n Chondrogenic differentiation was induced in micromass culture. To do so, 400,000 cells were resuspended in 20 μL of chondrogenic medium without TGF-β3 (basal medium plus 4.5 g of D-glucose/L, 350μ L-proline, and 100nM dexamethasone) and were plated onto a 96-well plate.1 After 3 hours, 200 μL of chondrogenic medium with TGF-β3a (10 ng/mL) was added to the early condensed cell pellets. Differentiation media were changed 3 times a week. Adipogenesis was terminated after 7 days, osteogenesis after 3 weeks, and chondrogenesis after 4 weeks.
The outcome of differentiation was monitored by cytochemical staining and by analysis of marker gene expression by qRT-PCR assay. Adipocyte-specific in-tracellular lipid vesicles were made visible with oil red staining. The calcium apatite crystals produced by osteoblasts were stained with alizarin red and the von Kossa staining protocol. To detect the cartilage-specific enrichment of proteoglycans, chondrogenic pellets were embedded in tissue cassette,0 frozen in liquid nitrogen, cut into 7-μm slices with a cryotome, and fixed on glass plates with methanol before staining with Alcian blue.
FCM—Human MSCs were isolated from femoral bone, expanded as described elsewhere,15 and used to titrate the anti-human antibodies used in this study prior to staining of equine MSCs. Use of human MSCs was approved by the local ethics committee.
Equine AT-derived MSCs were isolated and expanded as described earlier, and after harvesting 0.5 × 106 cells/sample to 1 × 106 cells/sample, the MSCs were washed with PBS solution and preincubated with immune globulinp (10%, diluted 1:20) to block nonspecific binding of the antibodies to equine cells. The MSCs were incubated with anti-human monoclonal antibodies against CD14,q CD34,r CD45,S CD73,t CD90,U CD105,V and CD164W in the conditions recommended by the antibody suppliers. After the MSCs were washed with PBS solution, cells were resuspended in 200 μL of FCM buffer (PBS solution containing 2% fetal calf serum, 2mM EDTA, and 0.01% sodium azide) and antibody binding was measured via FCM. Negative com-pensation particlesx and unstained equine MSCs were used as negative control samples, and antibody-stained human MSCs were used as positive control samples. The MSCs were gated according to forward and side scatter, and data sets for antibody staining were acquired in the automatic compensation mode of the flow cytometerj The FCM data were analyzed and processed with the aid of FCM analysis softwarey as described.16 The difference between the geometric mean value of the fluorescence intensities of unstained samples and that of stained samples was calculated as the MFI shift.
qRT-PCR assay—To confirm the antigen expression patterns of AT-derived MSCs detected by FCM and to extend the small spectrum of surface markers testable by antibodies, a qRT-PCR-based analysisz was performed on AT-derived MSCs from 3 horses. To this end, mRNA was isolated to generate cDNA as described elsewhere.17 Primers were constructed to allow the specific amplification of cDNA encoding the equine CD14, CD34, CD45, CD73, CD90, CD105, CD140b, and CD164 transcripts (Appendix). The amounts of these transcripts were normalized to the housekeeping gene GAPDH and to serial dilutions of a recombinant DNA standard in each PCR assay.9,17 Furthermore, the effect of in vitro differentiation was assessed by qRT-PCR-mediated measurement of the induction of lineage-specific marker genes. For this purpose, equine mRNA-specific oligonucleotide primers were designed for the transcripts of genes for Col1, Col2, Col10, aggrecan, osteopontin, LPL, and PPARγ2. Transcript amounts of the latter genes were quantified relative to GAPDH and to β-actin.18 All PCR data were processed by use of a standard software programaa and are reported as means of normalized transcript amounts ± SD. The integrity and size of the PCR products were monitored by melting point analysis and gel electrophoresis as described elsewhere.17 Immunophenotypes of BM-derived MSCs from 2 horses and PB-derived MSCs from 1 horse were analyzed by qRT-PCR assay as well.
Statistical analysis—A 2-sided modified Student t test was performed to compare the normalized qRT-PCR data between individual experimental groups (eg, AT- vs BM-derived MSCs). Values of P < 0.05 were considered significant. Results are reported as mean ± SD.
Results
Expansion of AT-derived MSCs—The AT-derived MSCs grew adherently to plastic, had the characteristic spindle-shaped phenotype in P2 of in vitro culture, and grew in typical patterns (Figure 1). At P4, this appearance was not substantially altered, although the cells appeared to become more trapezoidal. At P5 or P6, 4 of 6 AT-derived MSC preparations ceased to proliferate and finally detached from the bottom of the flask. Only the cells of 2 horses could be cultivated to P8. For the most part, these cells had the typical phenotype. However, some cells had flattened cell bodies and long filopodial membrane extensions. These cells were poly-nucleated and contained several large vesicles.
Graphic display of AT-derived MSC duplication rates in culture over time revealed an irregular pattern of cellular proliferation (Figure 2). Initially, the mean ± SD duplication rate decreased from P1 (0.25 ± 0.14 duplicates/d) to P2 (0.20 ± 0.14 duplicates/d). Later, the duplication rate increased, peaking at P4 with 0.42 ± 0.20 duplicates/d. At P5, proliferation rates declined to 0.31 ± 0.06 duplicates/d. At that point, proliferation of AT-derived MSCs from 4 of the 6 horses ceased, but cells of the other 2 donors could be propagated up to P8, at which point their duplication rates increased to 0.44 ± 0.23 duplicates/d. However, at P9, these cells ended spontaneous proliferation and detached as well.
Telomere length of AT-derived MSCs—To investigate differences between the cells that stopped proliferating after 5 passages and those cells that continued to proliferate to P8, telomere lengths were determined in AT-derived MSCs at P3 (n = 3 horses) and at P4 (3). At P3, telomere length was 6.8 ± 1.7 kbp, shortening to 6.6 ±2.1 kbp at P4. This represented a reduction of approximately 200 bp/passage or 50 bp/cell division.
The AT-derived MSCs of 1 horse proliferated up to P8. For this horse as well, telomere length was tracked. The telomeres spanned 5.6 kbp at P3 and were shorter in P5 (3.7 kbp) but grew again to 6.4 kbp at P8. As such, the spontaneous activation of telomerase activity in AT-derived MSCs appeared to extend the duration of proliferation in vitro; however, stable clones did not appear to be generated. Because the cells detached at P9, there was no evidence of spontaneous transformation of equine AT-derived MSCs.
Photomicrographs of cultured equine AT-derived MSCs at P2 (A), P4 (B), and P8 (C). Cells grew adherently on cell culture dishes. A—Cells at P2 have a fibroblastoid phenotype and grow in parallel bundles. Bar = 250 μm. B—Cells at P4 develop a more flattened cell body. Bar = 250 μm. C—Cellular flattening is more prominent in P8. Subsets of polynucleated cells with aberrant cell phenotypes were evident at this stage, featuring highly enlarged cell bodies with long filopodial projections and large vesicles. Bar = 100 μm
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of cultured equine AT-derived MSCs at P2 (A), P4 (B), and P8 (C). Cells grew adherently on cell culture dishes. A—Cells at P2 have a fibroblastoid phenotype and grow in parallel bundles. Bar = 250 μm. B—Cells at P4 develop a more flattened cell body. Bar = 250 μm. C—Cellular flattening is more prominent in P8. Subsets of polynucleated cells with aberrant cell phenotypes were evident at this stage, featuring highly enlarged cell bodies with long filopodial projections and large vesicles. Bar = 100 μm
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of cultured equine AT-derived MSCs at P2 (A), P4 (B), and P8 (C). Cells grew adherently on cell culture dishes. A—Cells at P2 have a fibroblastoid phenotype and grow in parallel bundles. Bar = 250 μm. B—Cells at P4 develop a more flattened cell body. Bar = 250 μm. C—Cellular flattening is more prominent in P8. Subsets of polynucleated cells with aberrant cell phenotypes were evident at this stage, featuring highly enlarged cell bodies with long filopodial projections and large vesicles. Bar = 100 μm
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Mean ± SD rate of duplication per day of cultured AT-derived MSCs at various passages. Rates rose until P4 and then decreased (n = 6 horses; black squares). In cells of 2 of the 6 equine donors, a renewed increase was evident in duplication rates from P6 to P8 until the cells detached at P9 (gray circles)
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Mean ± SD rate of duplication per day of cultured AT-derived MSCs at various passages. Rates rose until P4 and then decreased (n = 6 horses; black squares). In cells of 2 of the 6 equine donors, a renewed increase was evident in duplication rates from P6 to P8 until the cells detached at P9 (gray circles)
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Mean ± SD rate of duplication per day of cultured AT-derived MSCs at various passages. Rates rose until P4 and then decreased (n = 6 horses; black squares). In cells of 2 of the 6 equine donors, a renewed increase was evident in duplication rates from P6 to P8 until the cells detached at P9 (gray circles)
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
FCM and qRT-PCR assay—Only a few of the antihuman monoclonal antibodies reacted specifically with the equine AT-derived MSCs (data not shown). Strong shifts in MFI were detected with antibodies against the MSC-associated CD90 (MFI, 7.629) and CD105 (MFI, 985) molecules and with antibody against CD14 (MFI, 1.042; Figure 3). Little to no signal was detected with antibodies against the hematopoietic markers CD34 (MFI, 0.1) and CD45 (MFI, 19.1) or against CD73 (MFI, 9.2) and CD164 (MFI, 1.1). In P4 of in vitro culture, the AT-derived MSCs had the same phenotype (data not shown).
Results of the qRT-PCR analysis of AT-, BM-, and PB-derived MSCs were summarized (8Figure 4; Table 1). The BM-derived MSCs had slightly different values from those of AT-derived MSCs, but in PB-derived MSCs, expression of the monocytic marker antigen CD14 (10%) and of the endothelial and hematopoietic marker antigen CD34 (21%), as calculated relative to GAPDH mRNA expression, was slightly elevated in comparison with respective values for AT- and BMderived MSCs. Thus, the anti-CD90 and anti-CD105 antibodies appeared to detect equine MSCs, whereas the anti-CD73 and anti-CD164 reagents failed to bind equine MSCs.
Mean ± SD percentage transcript amounts for target genes for various cell surface antigens relative to that for housekeeping gene GAPDH In AT-derived MSCs from 3 horses, BM-derived MSCs from 2 horses, and PB-derived MSCs from 1 horse.
| Marker, by type | AT-derived MSCs | BM-derived MSCs | PB-derived MSCs |
|---|---|---|---|
| Inclusion | |||
| CD73 | 87 ± 37 | 58 ± 29 | 34 |
| CD90 | 108 ± 11 | 86 ± 99 | 65 |
| CD105 | 20 ± 9 | 8 ± 7 | 59 |
| CD140b | 28 ± 9 | 14 ± 4 | 4 |
| CD164 | 58 ± 30 | 41 ± 23 | 42 |
| Exclusion | |||
| CD14 | 7.2 ± 3.8 | 4 ± 07 | 10 |
| CD34 | 0.4 ± 0.3 | 0.0 ± 0.0 | 21 |
| CD45 | 0.0 ± 0.0 | 0.6 ± 0.8 | 0.7 |
Differentiation of AT-derived MSCs—After 3 weeks of in vitro differentiation, cultured cells had contracted cell structures that were not observed in control cultures without induction medium. These structures were more abundant in cells osteogenically differentiated at P2 than in cells differentiated at P4 (Figure 5). Von Kossa and alizarin red staining of osteoblasts derived from AT-derived MSCs at P2 was more intense than that in P4-derived cells. However, both were obviously positive for von Kossa or alizarin red staining in comparison to untreated control cells.
A strong but nonsignificant induction of transcripts encoding osteopontin (15.6 ± 17.3-fold), Col1 (6.4 ± 7.1- fold), and b-actin (8.2 ± 13.6-fold) was recorded after osteogenic differentiation of AT-derived MSCs at P2 (Figure 5). Parallel to the results from cytochemical staining, induction of mRNA encoding osteopontin (4.41 ± 3.8-fold), Col1 (0.47 ± 0.22-fold), and b-actin (1.12 ± 0.54-fold) was lower in cells differentiated from MSCs at P4. Expressions of the marker genes by cells after osteogenic differen-tiation from MSCs in P4 were 71.7%, 97.9%, and 86.5% of the expression observed in cells differentiated from MSCs in P2. Thus, the osteogenic differentiation capacity of equine AT-derived MSCs appeared higher in early-passage cells than in later-passage cells.
Results of FCM for MSCs cultured from the AT of 3 horses, indicating the immunophenotype with respect to CD14 (A), CD34 (B), CD45 (C), CD73 (D), CD90 (E), CD105 (F), and CD164 (G) at P2. Results for immunostained MSCs are displayed as gray histograms, and results for unstained MSCs are represented by dashed lines. Expression of CD14, CD90, and CD105 was evident. A weak signal was detected for CD45, but there was no signal for CD34, CD73, and CD164
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Results of FCM for MSCs cultured from the AT of 3 horses, indicating the immunophenotype with respect to CD14 (A), CD34 (B), CD45 (C), CD73 (D), CD90 (E), CD105 (F), and CD164 (G) at P2. Results for immunostained MSCs are displayed as gray histograms, and results for unstained MSCs are represented by dashed lines. Expression of CD14, CD90, and CD105 was evident. A weak signal was detected for CD45, but there was no signal for CD34, CD73, and CD164
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Results of FCM for MSCs cultured from the AT of 3 horses, indicating the immunophenotype with respect to CD14 (A), CD34 (B), CD45 (C), CD73 (D), CD90 (E), CD105 (F), and CD164 (G) at P2. Results for immunostained MSCs are displayed as gray histograms, and results for unstained MSCs are represented by dashed lines. Expression of CD14, CD90, and CD105 was evident. A weak signal was detected for CD45, but there was no signal for CD34, CD73, and CD164
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Relative expression (normalized to amount of GAPDH mRNA) of various antigens in AT-derived MSCs from 3 horses (black bars), BM-derived MSCs from 2 horses (light gray bars), and PB-derived MSCs from 1 horse (dark gray bars). Transcripts encoding the MSC-typical cell surface markers CD73, CD90, and CD105 as well as the cell surface markers CD14, CD140b, and CD164 were detected in all 3 cell types. The PB-derived MSCs expressed significantly (P ≤ 0.05) larger amounts of CD105 and CD140b than did AT-derived MSCs. Transcripts of CD45 were not found. In PB-derived MSCs, significant (P < 0.001) expression of CD34 was detected, compared with expression in AT- and BM-derived MSCs
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Relative expression (normalized to amount of GAPDH mRNA) of various antigens in AT-derived MSCs from 3 horses (black bars), BM-derived MSCs from 2 horses (light gray bars), and PB-derived MSCs from 1 horse (dark gray bars). Transcripts encoding the MSC-typical cell surface markers CD73, CD90, and CD105 as well as the cell surface markers CD14, CD140b, and CD164 were detected in all 3 cell types. The PB-derived MSCs expressed significantly (P ≤ 0.05) larger amounts of CD105 and CD140b than did AT-derived MSCs. Transcripts of CD45 were not found. In PB-derived MSCs, significant (P < 0.001) expression of CD34 was detected, compared with expression in AT- and BM-derived MSCs
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Relative expression (normalized to amount of GAPDH mRNA) of various antigens in AT-derived MSCs from 3 horses (black bars), BM-derived MSCs from 2 horses (light gray bars), and PB-derived MSCs from 1 horse (dark gray bars). Transcripts encoding the MSC-typical cell surface markers CD73, CD90, and CD105 as well as the cell surface markers CD14, CD140b, and CD164 were detected in all 3 cell types. The PB-derived MSCs expressed significantly (P ≤ 0.05) larger amounts of CD105 and CD140b than did AT-derived MSCs. Transcripts of CD45 were not found. In PB-derived MSCs, significant (P < 0.001) expression of CD34 was detected, compared with expression in AT- and BM-derived MSCs
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
The chondrogenic differentiation of AT-derived MSCs of 4 horses from P2 resulted in micromass pellets with dim Alcian blue staining overall (Figure 6). Staining was more intense at the periphery of cell pellets, whereas the centers had rather faint staining. Micromass pellets generated from AT-derived MSCs from 2 horses at P4 had even weaker overall staining.
Chondrogenic differentiation of AT-derived MSCs from P2 yielded a robust and, for some markers, significant induction of gene expression. Transcripts encoding Col2 were significantly (3,544 ± 1,942-fold; P = 0.035) greater in number than with undifferentiated control cells. The expression of aggrecan (3,956 ± 2,695-fold; P = 0.060) and Col10 (1,391 ± 1,536-fold; P = 0.135) was higher, whereas expression of Col1 was significantly (0.26 ± 0.32-fold; P = 0.018) lower than in the control cells. Chondrogenically differentiated AT-derived MSCs in P4 had lower and nonsignificant induction values: for Col2, only a 1,758 ± 961-fold (P = 0.234) induction was detected, which represented 49% of P2 induction. Aggrecan had a 321 ± 251-fold (P = 0.322) induction, compared with that in control cells. Overall, this represented 8% of value obtained after differentiation of MSCs in P2. Expression of Col10 was reduced (0.15 ± 0.14-fold; P = 0.074) relative to that in undifferentiated control cells. This represented 0.01% of P2 value. The gene expression of Col1 was somewhat higher in P4-derived chon-drocytes than in P2-derived chondrocytes (0.46 ± 0.63-fold increase; P = 0.444; 177% of the P2 value). But the expression of Col1 in chondrocytes remained lower than the expression detected in MSCs prior to differentiation. As observed for osteogenic differentiation, the chondrogenic potential of equine AT-derived MSCs decreased with the duration of in vitro culture. Most notably, induction of expression of Col10, a marker for chondrocyte hypertrophy, appeared to be abolished in chondrocytes generated from P4 AT-derived MSCs.
Adipogenic differentiation was conducted in AT-derived MSCs from 4 horses at P2 and from 2 horses at P4, and the cells were maintained for 7 days in the adipogenic induction medium. Because the cells detached after day 8 of differentiation in the first set of investigations, we harvested the cells on day 7 of adipogenic differentiation in all experiments thereafter.
In comparison to controls, oil red staining of P2-derived adipocytes revealed cells with small lipid vesicles and a flattened phenotype (Figure 7). Even more cells containing such lipid vesicles were detected in adipogenically differentiated cells from MSCs in P4, but no fusion to large vesicles was observed before the cells detached. In P2-derived adipocytes, the expression of PPARγ2 was increased 2.52 ± 1.02- fold (P = 0.059) and the LPL expression was induced 1.97 ± 1.53-fold (P = 0.292) higher than that in control cells. In P4-derived cells, a significant (P = 0.03) PPARγ2 induction was observed (6.08 ± 1.76-fold higher than in control cells), whereas LPL expression was not different in P4-derived adipocytes versus control cells (0.96 ± 0.86-fold). The difference in expression induction of marker genes between P4- and P2-derived adipocytes was 300% for PPARγ2 (P = 0.041) and 58.3% for LPL (P = 0.362).
Photomicrographs of control (A and D), P2 (B and E), and P4 (C and F) cultures of equine AT-derived MSCs stained with alizarin red (A, B, and C) or the von Kossa method (D, E, and F) after induction of osteogenic differentiation. Cells were expanded without stimulation of osteogenic differentiation to P2 and P4, then osteogenic differentiation was induced. In contrast to staining in undifferentiated control cells, von Kossa and alizarin red staining was intense in cells at P2 and weaker in cells differentiated at P4. Bar = 250 μm in all panels but C, in which bar = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of control (A and D), P2 (B and E), and P4 (C and F) cultures of equine AT-derived MSCs stained with alizarin red (A, B, and C) or the von Kossa method (D, E, and F) after induction of osteogenic differentiation. Cells were expanded without stimulation of osteogenic differentiation to P2 and P4, then osteogenic differentiation was induced. In contrast to staining in undifferentiated control cells, von Kossa and alizarin red staining was intense in cells at P2 and weaker in cells differentiated at P4. Bar = 250 μm in all panels but C, in which bar = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of control (A and D), P2 (B and E), and P4 (C and F) cultures of equine AT-derived MSCs stained with alizarin red (A, B, and C) or the von Kossa method (D, E, and F) after induction of osteogenic differentiation. Cells were expanded without stimulation of osteogenic differentiation to P2 and P4, then osteogenic differentiation was induced. In contrast to staining in undifferentiated control cells, von Kossa and alizarin red staining was intense in cells at P2 and weaker in cells differentiated at P4. Bar = 250 μm in all panels but C, in which bar = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of equine AT-derived MSCs from P2 (A) and P4 (B) cultures after induction of chondrogenic differentiation. Cells were expanded without stimulation of differentiation to P2 and P4, at which point chondrogenic differentiation was induced. Intensity of staining was visibly greater in cells differentiated from AT-derived MSCs at P2 than in cells differentiated at P4. Alcian blue stain; bar = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of equine AT-derived MSCs from P2 (A) and P4 (B) cultures after induction of chondrogenic differentiation. Cells were expanded without stimulation of differentiation to P2 and P4, at which point chondrogenic differentiation was induced. Intensity of staining was visibly greater in cells differentiated from AT-derived MSCs at P2 than in cells differentiated at P4. Alcian blue stain; bar = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of equine AT-derived MSCs from P2 (A) and P4 (B) cultures after induction of chondrogenic differentiation. Cells were expanded without stimulation of differentiation to P2 and P4, at which point chondrogenic differentiation was induced. Intensity of staining was visibly greater in cells differentiated from AT-derived MSCs at P2 than in cells differentiated at P4. Alcian blue stain; bar = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of equine AT-derived MSCs from control (A), P2 (B), and P4 (C) cultures after induction of adipogenic differentiation. Cells were expanded without stimulation of differentiation to P2 and P4, at which point adipogenic differentiation was induced. Lipid-containing vesicles were not observed in undifferentiated AT-derived MSCs after staining (A), but a greater number of stained vesicles were detected in cells differentiated from P4 AT-derived MSCs (C) than in cells from P2 AT-derived MSCs (B). Oil red stain; bars = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of equine AT-derived MSCs from control (A), P2 (B), and P4 (C) cultures after induction of adipogenic differentiation. Cells were expanded without stimulation of differentiation to P2 and P4, at which point adipogenic differentiation was induced. Lipid-containing vesicles were not observed in undifferentiated AT-derived MSCs after staining (A), but a greater number of stained vesicles were detected in cells differentiated from P4 AT-derived MSCs (C) than in cells from P2 AT-derived MSCs (B). Oil red stain; bars = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Photomicrographs of equine AT-derived MSCs from control (A), P2 (B), and P4 (C) cultures after induction of adipogenic differentiation. Cells were expanded without stimulation of differentiation to P2 and P4, at which point adipogenic differentiation was induced. Lipid-containing vesicles were not observed in undifferentiated AT-derived MSCs after staining (A), but a greater number of stained vesicles were detected in cells differentiated from P4 AT-derived MSCs (C) than in cells from P2 AT-derived MSCs (B). Oil red stain; bars = 100 μm.
Citation: American Journal of Veterinary Research 71, 10; 10.2460/ajvr.71.10.1228
Discussion
Studies7,19 have revealed the bipotentiality of equine AT-derived MSCs to differentiate into osteogenic and adipogenic lineages. The present study had the goal of further analyzing the immunophenotype of equine AT-derived MSCs, their differentiation potential, and their expansion capacity.
In our study, the morphology of AT-derived MSCs changed during in vitro culture. In early passages, the AT-derived MSCs had the typical long, narrow, spindle-shaped appearance. This phenotype changed to a more flattened and elongated cell body at later passages. The cells ceased to proliferate and detached from plastic flasks at P5 to P6. The investigated duplication rates of 0.20 to 0.42 duplicates/d corresponded to a doubling time of 5 to 2.4 days, which is comparable to recently reported values.7 At the same time, this growth pattern of MSCs suggested that sufficient numbers of cells must be harvested in the first place to be able to generate the amount of cells required for transplantation within 5 passages of in vitro culture.
Only cells from 2 of the 6 horses could be cultivated until P9, but then they grew with accelerated duplication rates. This particular behavior of AT-derived MSCs was evident in other studies3,7,14 as well. We confirmed these findings and thereby corroborated assumptions of processes related to but not identical to transformations taking place in MSCs in vitro. Furthermore, we found that telomere length did not shrink between P6 and P8 in such AT-derived MSCs, whereas usually a reduction in telomere length of approximately 200 bp/passage occurs in MSCs. In high-passage cultures, we detected morphologically altered cells with multiple nuclei and numerous vacuoles (Figure 1). Because monocytes, macrophages, and osteoclasts have a highly restricted proliferation capacity in vitro, the morphologically altered cells observed in P8 were most likely not cells of these types. Furthermore, in contrast to MSC cultures derived from the BM, contamination of the AT-derived MSC cultures with multinucleated osteoclasts can be excluded. One could hypothesize that this in vitro observation may be caused by the expansion of rapidly cycling cells or by activation of transformation-associated factors such as proto-oncogenes attributable to the in vitro culture conditions. Spontaneous transformation with induction of proto-oncogenes in long-term cultures of MSCs has been reported.20 However, the MSCs investigated in the present study detached at P9, which is inconsistent with transformation or immortalization of the cells. Detailed analyses of telomerase reverse transcriptase activity, regulation of embryonic factors and proto-oncogenes, or mutations of apoptosis-regulating genes such as P53 may help to elucidate cause and development of this behavior of equine AT-derived MSCs in vitro.
The mean telomere length of equine AT-derived MSCs at P3 is 6.8 kbp, which is approximately 75% of the telomere length reported for human BM-derived MSCs.21 One reason for this rather short telomere length may be the shorter life span of horses22 or the original tissue used to harvest the cells (BM- vs AT-derived MSCs). For example, the endosteal stem cell niche contains early hematopoietic stem cells, whereas later stage hematopoietic stem cells are enriched in the vascular stem cell niche.23 Comparably, later- stage MSCs might be enriched in ATs, as this represents a vascular stem cell niche, whereas the BM might harbor MSCs in early stages of differentiation. Finally, technical differences in determining telomere length (FCM vs PCR assay) and interindividual variations in the samples may also account for the different telomere lengths observed in this and other studies.21
Equine AT-derived MSCs are highly prone to osteogenic differentiation. We detected an elevated expression of osteopontin after differentiation of AT-derived MSCs and intense von Kossa and alizarin red staining. The cell clusters generated during differentiation represent centers of production of calcium apatite and thus centers of osteogenic differentiation.24 The AT-derived MSCs also had good chondrogenic differentiation potential. We detected high expression of Col2 and aggrecan in differentiated cells derived from P2 MSCs but a rather low expression of Col1, compared with expression in undifferentiated MSCs. The expression of aggrecan in P2-derived cells could be confirmed by Alcian blue staining, which stains proteoglycans. However, the Alcian blue technique is not a quantitative method and a variety of factors influence the color intensity. We therefore interpreted these results only together with the analyses of transcripts encoding marker collagenous genes. With this strategy, we detected an induction of Col10, a marker for hypertrophic chondrocytes, in P2- derived chondrocytes. Its expression reflects a terminal chondrogenic differentiation of the cells. In contrast to the high expression in P2-derived cells, the expression of the collagenous marker genes for Col2 and aggrecan is lower in P4-derived chondrocytes. This rather low expression of Col10 in cells differentiated from P4 MSCs may be explained by a reduction in chondrogenic differentiation potential with time in in vitro culture.25 Accordingly, the gene expression of Col1 was slightly increased in cells generated by chondrogenic differentiation of MSCs from P4, compared with expression in the P2-derived chondrocytes.
In contrast to the differentiation of MSCs into the other 2 lineages, adipogenic differentiation capacity of AT-derived MSCs improved with higher passage numbers, as was apparent in the higher induction of the major adipogenic regulator PPARγ2 and more intense oil red staining. Again, we could confirm the necessity of rabbit serum supplementation for sufficient induction of adipogenesis in equine MSCs. We harvested and stained the cells on day 7 after adipogenic induction as described.7 After this point, the cells started to detach. A terminal differentiation into mature adipocytes with large, fused lipid vesicles could not be detected. This finding of hampered adipogenesis in equine AT-derived MSCs was interesting because such cells are derived from AT and are believed to regenerate this tissue. Thus, we concluded that adipogenic differentiation was initiated sufficiently by our stimuli, but some factors promoting further differentiation into mature adipocytes may have been lacking in our in vitro culture systems.
Immunophenotype results for equine MSCs in the present study are the first to suggest that equine MSCs fulfill the minimal criteria for human MSCs.1 Most antibodies applied were specific to human MSCs but did not cross-react with the corresponding epitopes on equine MSCs. This included the antibody against CD73. But the anti-human monoclonal antibodies against CD14, CD90, and CD105 stained equine MSCs specifically. It should be noticed that CD14 is not a typical antigen on MSCs, at least in mice and humans.1 Because few horse-specific antibodies are available, the expression of some of the important MSC surface markers was determined by qRT-PCR analysis. The positive staining of equine MSCs by antibodies against CD90 and CD105 as determined by FCM was confirmed by the results of that molecular analysis. The expression of mRNA encoding the MSC marker antigen CD73 was detected in equine AT-, BM-, and PB-derived MSCs.
Interestingly, CD14 was expressed in all equine MSC types and could be detected by 2 independent methods: FCM and transcript analysis. The CD14 antigen is known to be expressed on monocytes and macrophages. Therefore, the CD14 signal, particularly in the transcription analysis, might have been caused by contaminating monocytes. However, expression of CD14 has been detected on adipose-derived cells ex vivo but appears to wane with consecutive passages.26 Furthermore, contaminating macrophages would yield a small number of CD14-positive cells in the total population but not a bright staining of the total population in an FCM histogram. In addition, macrophages would yield a higher normalized CD14 mRNA signal in the qRT-PCR amplifications. Yet, we detected the CD14 antigen on MSCs independently of their passage (ie, P2 and P4) and we found CD14 transcripts in bulk cultures and, at comparable expression levels, even in cells after sorting of the CD90-positive fraction of the AT-derived MSCs (data not shown). Contaminating monocytes or macrophages do not proliferate in vitro at rates comparable to those of MSC in an MSC-optimized medium. Consequently, the cells expressing CD14 as detected by FCM and qRT-PCR assay would fade away with time of in vitro expansion. We therefore concluded that equine MSCs, in contrast to human MSCs, express the LPS receptor CD14. Perhaps the expression of CD14 makes these cells more sensitive to gram-negative bacteria. Therefore, the utmost care must be taken to avoid adverse effects on MSCs during expansion of the cells for clinical applications. Interestingly, equine PB-derived MSCs showed the highest expression of CD14 and also expressed CD34. This molecule is normally expressed on hematopoietic stem cells and on endothelial cells. But recent studies27,28 described a CD14-positive subset of cells expressing CD34 at low levels. They were derived from peripheral human blood and were referred to as monocyte-derived multipotent cells. Accordingly, the equine PB-derived MSCs could be similar to these multipotent cells.
Human MSCs lack expression of CD45,29 and we could not detect CD45-encoding mRNA in equine MSCs in our study. The slight shift generated by the anti-CD45 antibody on equine MSCs therefore appeared to be a nonspecific staining. Contamination of AT with hematopoietic cells expressing CD45 could be excluded because such reactivity was observed in AT cells ex vivo but not in cells after 2 passages in another study.26
The analysis of telomere length and the investigation of gene expression yielded results with a certain degree of variability. Cellular differences among the cells used might account for this. Furthermore, samples from only 6 horses were available. Therefore, statistical tests for some experimental results did not reach significance.
In the study reported here, equine MSCs expressed the lineage markers reported for human MSCs, specifically CD90 and CD105 as detected by FCM and qRT-PCR assay, as well as CD73, CD140b, and CD164 as determined by qRT-PCR assay. Equine AT-derived MSCs had differentiation capacities to generate osteoblasts or chondrocytes in vitro and therefore represent a suitable cellular source for regenerative treatments of bone or cartilage defects, particularly when expanded for only a few passages. In contrast to human MSCs, equine AT-derived MSCs express the LPS receptor CD14, and equine PB-derived MSC also express CD34, an antigen found in humans or mice on hematopoietic stem cells and on endothelial cells. We conclude that equine AT-derived MSCs have a phenotype comparable to the phenotype reported for MSCs and therefore substantially fulfill the minimal criteria established by the International Society of Cellular Therapy for MSCs.1 The expression and possible function of CD14 on equine MSCs should be investigated in future experiments.
Abbreviations
| AT | Adipose tissue |
| BM | Bone marrow |
| Col1 | Type I collagen |
| Col2 | Type II collagen |
| Col10 | Type X collagen |
| FCM | Flow cytometry |
| GAPDH | Glyceraldehyde-3-phospate-dehydrogenase |
| LPL | Lipoprotein lipase |
| MFI | Mean fluorescence intensity |
| MSC | Mesenchymal stem (stromal) cells |
| PB | Peripherally obtained blood |
| PPARγ2 | Peroxisome proliferator-activated receptor γ2 |
| qRT | Quantitative reverse transcription |
Sigma-Aldrich, Taufkirchen, Germany.
PAA Laboratories GmbH, Pasching, Austria.
GIBCO, Invitrogen, Carlsbad, Calif.
Lonza, Basel, Switzerland.
Biochrom, Berlin, Germany.
BD Biosciences, Heidelberg, Germany.
T75 flask, BD Falcon, Franklin Lakes, NJ.
Accutase, PAA Laboratories GmbH, Pasching, Austria.
Telomere PNA Kit, Dako, Glostrup, Denmark.
LSRII Flow Cytometer, BD Biosciences, Heidelberg, Germany.
BD fluorescence-activated cell sorting DIVA, BD Biosciences, Heidelberg, Germany.
Greiner, Frickenhausen, Germany.
Calbiochem, Darmstadt, Germany.
Invitrogen, Darmstadt, Germany.
Tissue Tek, Sakura Finetek, Zoeterwoude, The Netherlands.
Gamunex, Talecris Biotherapeutics Inc, Research Triangle Park, NC.
Clone 134620, R&D Systems, Minneapolis, Minn.
Clone 43A1, Santa Cruz Biotechnology, Santa Cruz, Calif.
Clone 2D1, R&D Systems, Minneapolis, Minn.
Clone AD2, BD Biosciences, Heidelberg, Germany.
Clone Thy1-A1, R&D Systems, Minneapolis, Minn.
Clone SN6, AdB Serotec, Martinsried, Germany.
lone 67D2, Biolegend, San Diego, Calif.
CompBeads, BD Biosciences, Heidelberg, Germany.
FlowJo, TreeStar Inc, Ashland, Ore.
LightCycler, Roche, Mannheim, Germany.
Microsoft Office Excel 2007, Microsoft Corp, Redmond, Wash.
References
- 1↑
Dominici MLe Blanc KMueller I, et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 2006;8:315–317.
- 2↑
Chamberlain GFox JAshton B et al. Concise review: mesenchymal stem cells: their phenotype, differentiation capacity, immunological features, and potential for homing. Stem Cells 2007;25:2739–2749.
- 3↑
Vidal MAKilroy GEJohnson JR et al. Cell growth characteristics and differentiation frequency of adherent equine bone marrow- derived mesenchymal stromal cells: adipogenic and osteogenic capacity. Vet Surg 2006;35:601–610.
- 4↑
Worster AANixon AJBrower-Toland BD et al. Effect of transforming growth factor β1 on chondrogenic differentiation of cultured equine mesenchymal stem cells. Am J Vet Res 2000;61:1003–1010.
- 5
Kisiday JDKopesky PWEvans CH et al. Evaluation of adult equine bone marrow- and adipose-derived progenitor cell chondrogenesis in hydrogel cultures. J Orthop Res 2008;26:322–331.
- 6↑
Reed SAJohnson SE. Equine umbilical cord blood contains a population of stem cells that express Oct4 and differentiate into mesodermal and endodermal cell types. J Cell Physiol 2008;215:329–336.
- 7↑
Vidal MAKilroy GELopez MJ, et al. Characterization of equine adipose tissue-derived stromal cells: adipogenic and osteogenic capacity and comparison with bone marrow-derived mesenchymal stromal cells. Vet Surg 2007;36:613–622.
- 8↑
Giovannini SBrehm WMainil-Varlet P et al. Multilineage differentiation potential of equine blood-derived fibroblast-like cells. Differentiation 2008;76:118–129.
- 9↑
Koerner JNesic DRomero JD et al. Equine peripheral bloodderived progenitors in comparison to bone marrow-derived mesenchymal stem cells. Stem Cells 2006;24:1613–1619.
- 10↑
Koch TGHeerkens TThomsen PD et al. Isolation of mesenchymal stem cells from equine umbilical cord blood. BMC Biotechnol 2007;7:26.
- 11↑
Wilke MMNydam DVNixon AJ. Enhanced early chondrogenesis in articular defects following arthroscopic mesenchymal stem cell implantation in an equine model. J Orthop Res 2007;25:913–925.
- 13↑
Guthrie M. Win place, or cell. Can a company harness stem cells to treat injured horses? The Scientist 2008;22:75.
- 14↑
Siddappa RLicht Rvan Blitterswijk C et al. Donor variation and loss of multipotency during in vitro expansion of human mesenchymal stem cells for bone tissue engineering. J Orthop Res 2007;25:1029–1041.
- 15↑
Felka TSchäfer RSchewe B et al. Hypoxia reduced the inhibitory effect of IL-1β on chondrogenic differentiation of FCS-free expanded MSC. Osteoarthritis Cartilage 2009;17:1368–1376.
- 16↑
Herzenberg LATung JMoore WA et al. Interpreting flow cytometry data: a guide for the perplexed. Nat Immunol 2006;7:681–685.
- 17↑
Alexander DJudex MMeyringer R et al. Transcription factor Egr-1 activates collagen expression in immortalized fibroblasts or fibrosarcoma cells. Biol Chem 2002;383:1845–1853.
- 18↑
Arnhold SJGoletz IKlein H et al. Isolation and characterization of bone marrow-derived equine mesenchymal stem cells. Am J Vet Res 2007;68:1095–1105.
- 19↑
Vidal MARobinson SOLopez MJ et al. Comparison of chondrogenic potential in equine mesenchymal stromal cells derived from adipose tissue and bone marrow. Vet Surg 2008;37:713–724.
- 20↑
Li HFan XKovi RC et al. Spontaneous expression of embryonic factors and p53 point mutations in aged mesenchymal stem cells: a model of age-related tumorigenesis in mice. Cancer Res 2007;67:10889–10898.
- 21↑
Bonab MMAlimoghaddam KTalebian F et al. Aging of mesenchymal stem cell in vitro. BMC Cell Biol 2006;7:14.
- 22↑
Budiansky S. The nature of horses: exploring equine evolution intelligence and behaviour. New York: The Free Press 1997.
- 23↑
Orkin SHZon LI. Hematopoiesis: an evolving paradigm for stem cell biology. Cell 2008;132:631–644.
- 24↑
McDuffee LAAnderson GIWright GM et al. In vitro heterogeneity of osteogenic cell populations at various equine skeletal sites. Can J Vet Res 2006;70:277–284.
- 25↑
Muraglia ACancedda RQuarto R. Clonal mesenchymal progenitors from human bone marrow differentiate in vitro according to a hierarchical model. J Cell Sci 2000;113:1161–1166.
- 26↑
McIntosh KZvonic SGarrett S, et al. The immunogenicity of human adipose-derived cells: temporal changes in vitro. Stem Cells 2006;24:1246–1253.
- 27↑
Romagnani PAnnunziato FLiotta F et al. CD14+CD34low cells with stem cell phenotypic and functional features are the major source of circulating endothelial progenitors. Circ Res 2005;97:314–322.
- 28↑
Seta NKuwana M. Human circulating monocytes as multipotential progenitors. Keio J Med 2007;56:41–47.
- 29↑
Bühring HJBattula VLTreml S, et al. Novel markers for the prospective isolation of human MSC. Ann N Y Acad Sci 2007;1106:262–271.
Appendix
Primers constructed to allow the specific amplification of cDNA of various proteins In equine AT-derived MSCs
| Protein | RefSeq ID* of genomic and transcript sequence | Sequence of primer (5′ to 3′) |
|---|---|---|
| Aggrecan | NC_009144.1 | TGCACAGACCCCGCCAGCTA |
| XM_001499504 | GTCTCTAAACTCAGTCCACG | |
| CD14 | NC_009157.2 | GACCTATAGACCATGGTGCG |
| AF200416 | AGTTTCTAAAAGGCGCCTCC | |
| CD34 | NC_009148.1 | CTCCAGCTGTGAGGACTTTA |
| XM_001491596 | AAGTTCTGGATCCCCATCCT | |
| CD45 | XR_036036 | CTCCTCATTCACTG CAGAGA |
| EF576851 | GGTACTGCTCAAATGTGGGA | |
| CD73 | NC_009153.1 | AACGGCACCATTACCTTGGA |
| XM_001500115 | GGGAGGATCAGCTTATACAC | |
| CD90 | NC_009150.1 | TCTCCTGCTGACAGTCTTGC |
| XM_001503225.1 | GGACCTTGATGTTGTACTTGC | |
| CD105 | NC_009168.1 | TTCTGGGCCACTGGTGAATA |
| XM_001500078 | TGCAATGCAGACTCGAGATG | |
| CD140b | NC_009157.1 | CGACGAGATCTACGAGATCA |
| XM_001501493 | GGTGTTGACTTCATTCAGGG | |
| CD164 | NC_009153.1 | GTATGAAACCTGTGAAAGCCG |
| XM_001502103 | GTTGTACCTGATGTAGTAGC | |
| LPL | NW_001809456.1 | CACCACTCCCTGAAGTTTCCA |
| XM_001489577.1 | CCACAGGTAGTATTGCACCA | |
| Osteopontin | NC_009146.1 | CGCAGATCTGAAGACCAGTA |
| XM_001496152 | GGAATGCTCACTGGTCTCAT | |
| PPARγ2 | NC_009159.1 | TTATTCTCAGTGGAGACCGC |
| XM_001492411 | CACCCACTCCTACAGGAAAT | |
| Col1 | NC_009147.1 | CAGCACTCTTACACCTGTTG |
| XM_001492939 | GGCAAATGCTCTGCACACTT | |
| Col2 | NC_009149 | GCTTCCACTTCAGCTATGGA |
| U62528 | TGTTTCGTGCAGCCATCCTT | |
| Col10 | NC_009153.1 | CCAAGAGGTGCCCCTGGAAT |
| XM_001504101 | GTTGCCTGTTATACACAATC |
RefSeq ID refers to the access numbers of the gene sequences in the National Center for Biotechnology Information (NCBI).
