Calorie restriction with regular chow, but not a high-fat diet, delays onset of spontaneous osteoarthritis in the Hartley guinea pig model

Abstract

Background: Obesity is a leading risk factor for osteoarthritis (OA). In contrast, calorie restriction (CR) may lessen OA due to improved systemic inflammatory status and reduced weight-bearing. The aim of this study was to determine how CR with regular chow versus a high-fat diet (HFD) alters OA progression using the Hartley guinea pig model of disease.

Methods: Twenty-four male guinea pigs were allocated to four groups at 2 months of age: (1) ad libitum regular chow (obese), (2) CR regular chow (lean), (3) ad libitum HFD, and (4) CR HFD. Animals in both HFD groups ate identical amounts and were combined into one HFD group for analyses. At 5 months, hind limbs were harvested for microcomputed tomography (microCT) and histopathologic evaluation of knee OA. Total body, gonad fat, and infrapatellar fat pad (IFP) masses were recorded. IFPs were collected for gene expression analysis. Immunohistochemistry for monocyte chemoattractant protein-1 (MCP-1) was performed on intact joints. Serum was utilized for protein C3 measurement. All data were compared using ordinary one-way ANOVA analyses with Tukey's post-hoc tests.

Results: Body mass in the lean and HFD groups were similar and lower than the obese group. Despite this, gonad fat pads in the HFD group were comparable to the obese group. MicroCT and histologic OA scores were similar in obese and HFD groups; both scores were significantly lower in the lean group. Obese and HFD groups displayed increased gene expression of pro-inflammatory and catabolic mediators in IFPs relative to lean animals. Consistent with this, immunohistochemistry for MCP-1 in knee joints demonstrated strong positive staining in obese and HFD groups but was minimally detected in lean animals. Serum protein C3 levels were also statistically higher.

Conclusions: This study demonstrated that CR with a regular chow diet lessened knee OA in the Hartley guinea pig and was associated with decreased local and systemic inflammation compared to obese animals. HFD animals, although under CR conditions, had OA scores and inflammatory markers similar to obese animals. Thus, diet composition, and not solely body weight, may be a key factor in development of OA.

Keywords: Calorie restriction; Hartley guinea pig; High-fat diet; Obesity; Osteoarthritis.

Fig. 1

Total body weight (a) and tibial length (b) in obese, lean, and HFD groups. Black line represents mean value. Open red triangles designate animals receiving the restricted HFD; closed triangles define animals on the ad libitum HFD. ****P < 0.0001

Fig. 2

Weight of gonad fat (a) and IFP (b) in obese, lean, and HFD groups. Black line represents mean value. Open red triangles designate animals receiving the restricted HFD; closed triangles define animals on the ad libitum HFD. ****P < 0.0001

Fig. 3

MicroCT and OARSI histologic scores in obese, lean, and HFD groups. Black line represents mean value. *P < 0.05, **P < 0.01

Fig. 4

a Dorsal and b sagittal reconstructions from a calorie-restricted animal with no OA lesions. MicroCT OA score of 0. c Dorsal reconstruction from an obese animal. There is sclerosis and small enthesiophytes on the medial femoral condyle (red arrows) and mild sclerosis of the central tibial plateau. d Sagittal reconstruction from same animal. Mild sclerosis of the cranial patella and caudal tibial condyle is present (blue arrows). MicroCT OA score of 6. e Dorsal reconstruction of an animal on the HFD. On the medial tibial condyle, there is a small osteophyte (green arrow). f HFD sagittal reconstruction. There is a small osteophyte on the proximal patella (green arrow), as well as mild sclerosis on the cranial patella (blue arrow). MicroCT OA score of 4

Fig. 5

a T blue photomicrograph of the medial compartment from a calorie-restricted animal. The articular surface is smooth with only mild superficial proteoglycan loss. b, c T blue photomicrographs from obese animals. The articular surfaces are fibrillated, there is proteoglycan loss in the superficial and middle zones, and there is cell clustering. d–f T blue photomicrographs from HFD-fed animals. There is regional to diffuse proteoglycan loss in the superficial and middle zones, mild articular surface irregularity, occasional cell clustering, and focal cell loss within the superficial zone

Fig. 6

Total platelet counts (a), protein C3 levels (b), serum cholesterol (c), and serum BUN (d) in obese, lean, and HFD groups. Black line represents mean values. Open red triangles designate animals receiving the restricted HFD; closed triangles define animals on the ad libitum HFD. *P < 0.05, **P < 0.01, ****P < 0.0001

Fig. 7

Representative images of immunostaining for MCP-1 on the medial tibial plateau for the ad libitum regular chow group (a), the HFD group (b), and calorie-restricted group (c). × 400

Fig. 8

The mean percentage of positive cells (a) and mean integrated intensity (b) of MCP-1 immunostaining in articular cartilage from both femoral condyles and tibial plateaus in obese, lean, and HFD groups. Black line represents mean values. Open red triangles designate animals receiving the restricted HFD; closed triangles define animals on the ad libitum HFD. **P < 0.01, ***P < 0.001, ****P < 0.0001