Fig 1.
Effective bolus delivery of hCol1 and experimental timeline.
Hazelnut cream was used as a vehicle to deliver daily bolus doses of hCol1 to mice such that a delivery of a 150mg mixture provided a daily bolus dose of either 3.8mg (LD) or 38mg (HD) hCol1 (Control = hazelnut cream alone). Experimental mixtures were placed on autoclavable ceramic tiles (A) and presented to individually house mice (B) at the same time daily. After 5–7 days of training with vehicle alone, mice consumed the full amount presented within 2 minutes. Panel (C) depicts the experimental timeline. Mice were presented hazelnut cream daily in the bolus feeding regimen for a 1 week training period (blue line), and then Control, LD and HD daily supplements were initiated and continued for the remainder of the experiment (green line). After 4 weeks of supplementation, MLI (right knee) and Sham (left knee) surgery was performed (t = 0), followed by tissue harvests at 3 weeks and 12 weeks post-surgery. (D) To confirm successful delivery of hCol1, serum hProline levels were quantified via ELISA. Serum samples collected 1 week before (-1) and 2 weeks after surgery were harvested 3 hours after the mice consumed supplements (left graph). Serum samples collected 3 and 12 weeks after surgery were harvested 1 hour after consumption of the supplements (right graph). Symbols (○) represent the hProline level in the serum of individual mice. Bars represent the average hProline level for each experimental group (± SEM, N = 6). Significant differences between groups were identified via two-way ANOVA with a Tukey’s multiple comparisons post-test (*p<0.05, **p<0.01 compared to Control).
Table 1.
qRTPCR primer sequences.
Fig 2.
Cartilage loss following MLI in hCol1-fed mice is reduced.
Panel (A) presents an array of representative 40x Toluidine Blue/Fast Green stained sagittal sections from the medial compartment of sham and MLI joints 12 weeks post-injury under various treatment conditions (control = vehicle, LD = 3.8mg hCol1/day, HD = 38mg hCol1/day). Joint structures are labeled (F = femur, M = meniscus, T = tibia) and the black scale bar depicts 100μm. Total tibial cartilage area was determined in these representative sections as well as a series of similarly stained serial sections from all experimental joints at both 3 weeks (B) and 12 weeks (C) post-MLI using an automated approach (Visiopharm System). Symbols (○) represent the average tibial cartilage area of 3 sections/joint. Bars represent the average tibial cartilage area for each experimental group (± SEM, N = 6). Significant differences between experimental groups were identified via one-way ANOVA with a Tukey’s multiple comparisons post-test (*p<0.05, **p<0.01 compared to Control Sham; xp<0.05, xxp<0.01 compared to Control MLI).
Fig 3.
hCol1 is chondroprotective in the early stages of murine PTOA.
Panel (A) presents an array of representative 40x Safranin O/Fast Green stained sagittal sections (40x) from the medial compartment of sham and MLI joints 3 weeks post-injury under various treatment conditions (control = vehicle, LD = 3.8mg hCol1/day, HD = 38mg hCol1/day). Joint structures are labeled (F = femur, M = meniscus, T = tibia) and the black box denotes the area shown in the zoomed images, where the tidemarks are denoted with a yellow dashed line. Black scale bars depict 100μm. Cartilage architecture was evaluated using the Osteomeasure System to determine the tibial uncalcified cartilage area (B), tibial calcified cartilage area (C), the number of chondrocytes in the tibial uncalcified cartilage (D), and the number (E) and percentage (F) of Safranin-O positive (SafO+) chondrocytes in the tibial uncalcified cartilage. OARSI scoring of the sections analyzed by histomorphometry was also performed (G). For histomorphometry and cell counting, symbols (○) represent the average measurement made from 3 sections/joint. For OARSI Scoring, symbols (○) represent the average score for each joint based on scoring of 3 sections/joint by four observers. Bars in all graphs represent the average for each experimental group (± SEM, N = 6). Significant differences between experimental groups in the histomorphometry data (B-F) were identified via one-way ANOVA with a Tukey’s multiple comparisons post-test (*p<0.05, **p<0.01, ***p<0.001 compared to Control Sham; xp<0.05, xxp<0.01 compared to Control MLI). Significant differences between experimental groups in the OARSI data (G) were identified via a Kruskal-Wallis Test with a Dunn’s multiple comparisons post-test (*p<0.05, compared to Control Sham).
Fig 4.
hCol1 protects against cartilage loss in mid to late stage murine PTOA.
Panel (A) presents an array of representative 40x Safranin O/Fast Green stained sagittal sections from the medial compartment of sham and MLI joints 12 weeks post-injury under various treatment conditions (control = vehicle, LD = 3.8mg hCol1/day, HD = 38mg hCol1/day). Joint structures are labeled (F = femur, M = meniscus, T = tibia) and the tidemarks are denoted with a yellow dashed line in the zoomed images. Black scale bars depict 100μm. Cartilage architecture was evaluated using the Osteomeasure System to determine the tibial uncalcified cartilage area (B), the tibial calcified cartilage (C), the number of chondrocytes in the tibial uncalcified cartilage (D), and the number (E) and percentage (F) of Safranin-O positive (SafO+) chondrocytes in the tibial uncalcified cartilage. OARSI scoring of the sections analyzed by histomorphometry was also performed (G). For histomorphometry and cell counting, symbols (○) represent the average measurement made from 3 sections/joint. For OARSI Scoring, symbols (○) represent the average score for each joint based on scoring of 3 sections/joint by four observers. Bars in all graphs represent the average for each experimental group (± SEM, N = 6). Significant differences between experimental groups in the histomorphometry data (B-F) were identified via one-way ANOVA with a Tukey’s multiple comparisons post-test (*p<0.05, **p<0.01, ***p<0.001 compared to Control Sham; xp<0.05, xxp<0.01 compared to Control MLI, N = 6). Significant differences between experimental groups in the OARSI data (G) were identified via a Kruskal-Wallis Test with a Dunn’s multiple comparisons post-test (*p<0.05, **p<0.01 compared to Control Sham).
Fig 5.
hCol1 reduces MMP13 levels in articular cartilage of mice following MLI.
3 weeks post-injury (Sham or MLI), knee joints were harvested from mice and hypertrophic chondrocytes were analyzed by immunohistochemistry of MMP13 and ColX. Representative sagittal sections depict (A) MMP13 and (B) ColX stained chondrocytes (brown) with cell nuclei counterstained with hematoxylin (blue). Yellow dasted lines highlight the tide mark, separating calcified cartilage from uncalcified cartilage. Joint structures are labeled (F = femur, M = meniscus), and the black scale bar depicts 100μm.
Fig 6.
Chondrocyte apoptosis post-MLI is reduced in hCol1-supplemented mice.
Joints were harvested from mice 3 weeks post-injury (Sham or MLI) and apoptotic cells were identified via TUNEL staining. Representative 100x sagittal sections (right column of panels) show the overall scope of cellular apoptosis (green), with all cell nuclei DAPI labeled (blue). The yellow dashed lines depict the articular cartilage surface and the red dashed lines outline the anterior and posterior horns of the meniscus. The region demarcated with the white box is magnified in the right column (zoom). White scale bars depict 100μm.
Fig 7.
Synovial hyperplasia is reduced in mice supplemented with hCol1.
(A) Tissue sections stained with Safranin O/Fast Green were used to examine the synovium. Representative 40x sagittal sections from Sham and MLI joints of mice supplemented with Control (vehicle, hazelnut cream), LD hCol1 or HD hCol1 that were harvested at 3 and 12 weeks post-injury are depicted. Joint structures are labeled (F = femur, M = meniscus, T = tibia), and synovial membranes are demarcated with black arrows. The red line highlights the thickness of hyperplastic synovium in the Control MLI section and the black scale bar depicts 100μm. A synovial scoring method was also employed to quantify synovial hyperplasia at both 3 weeks (B) and 12 weeks (C) post-injury. Symbols (○) represent the average Synovial Score for each joint based on scoring of 3 sections/joint by four observers. Bars represent the average Synovial Score per experimental group (± standard deviation, N = 5–8 joints). Significant differences between experimental groups were identified via a Kruskal-Wallis Test with a Dunn’s multiple comparisons post-test (*p<0.05, **p<0.01, ***p<0.001 compared to Control Sham).
Fig 8.
Post-injury upregulation of TNF in the synovium is reduced in mice supplemented with hCol1.
(A) Representative TNF immunostained sagittal sections (100x) from Sham and MLI joints of mice supplemented with Control (vehicle, hazelnut cream), LD hCol1 or HD hCol1 that were harvested at 3 and 12 weeks post-injury are shown. Joint structures are labeled (F = femur, M = meniscus, T = tibia), synovial membranes are demarcated with red arrows, and brown staining of the tissue indicates intensity and location of TNF expression. The black scale bar depicts 100μm. mRNA was purified from synovial tissue collected from a separate cohort of similarly-treated mice at 3 weeks (B) and 12 weeks (C) post-injury. qRTPCR was performed to quantify Tnf expression level. Symbols (○) represent the Tnf level in each synovial sample and bars represent the average Tnf level for each experimental group (± SEM, N = 6). Significant differences between groups were identified via one-way ANOVA with a Tukey’s multiple comparisons post-test (*p<0.05 compared to Control Sham, xxp<0.01 compared to Control MLI).