New insights into the role of matrix metalloproteinase 3 (MMP3) in bone

Abstract

The Matrix Metalloproteinases are important regulators of bone metabolism and can influence bone mass and bone remodeling. We investigate the role of Matrix Metalloproteinase 3 (MMP3) on bone in mice, by using Mmp3 knockout (Mmp3 KO) in the context of estrogen deficiency, and in human, by analyzing the association of promoter polymorphism with bone mineral density in postmenopausal women and with MMP3 expression. We presented evidence in this paper that Mmp3 KO significantly increases trabecular bone mass and trabecular number and does not affect cortical bone thickness. We also found that Mmp3 KO protects from the deleterious effects of ovariectomy on bone mineral density in mice by preventing deterioration of bone microarchitecture. The effect of Mmp3 KO does not involve bone formation parameters but instead acts by inhibition of bone resorption, leading to a reduced bone loss associated to ovariectomy. By studying a human cohort, we found that a polymorphism located in the promoter of the human MMP3 gene is associated with bone mineral density in postmenopausal women and found that MMP3 rs632478 promoter variants are associated with change in promoter activity in transfection experiments. In conclusion MMP3, although weakly expressed in bone cells, could be one of the important regulators of sex hormone action in bone and whose activity could be targeted for therapeutic applications such as in Osteoporosis.

Keywords: SNP; bone mineral density; bone resorption; estrogen deficiency; matrix metalloproteinase 3; osteoporosis.

FIGURE 1

Bone mineral density (BMD) analysis of femur and vertebrae from wild‐type and Mmp3 KO mice after ovariectomy and sham‐surgery. (A) BMD was measured by DEXA before (A) and 30 days after ovariectomy or sham‐surgery (B). Change in BMD is expressed in % of baseline (C). BMD is expressed in g/cm². All values are expressed as mean ± SD. Significance was determined using ANOVA and post hoc Tukey's tests. (**, p < 0.005), n = 7–9 animals per experimental groups

FIGURE 2

MicroCT and histology analyses of mouse femur in 4‐month‐old wild‐type and Mmp3 KO mice after ovariectomy or sham‐surgery. Two‐dimensional (2D) microCT (A) and histological (Toluidine blue staining) (B) pictures are representative images of the distal femur (coronal sections). (C) Microarchitecture parameters were measured by histomorphometry on 2D histological sections (left panel) and by 3D microCT (right panel). Bone volume per tissue volume (BV/TV, %), trabecular thickness (Tb.Th, μm), trabecular number (Tb.N, 1/μm), and trabecular spacing (Tb.Sp, μm) are presented as mean ± SD. Significance was determined using ANOVA and post hoc Tukey's tests. (*p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.0001), n = 5–7 animals per experimental group for histological analysis and n = 8–10 for microCT analysis

FIGURE 3

Bone resorption parameters measured after ovariectomy or sham‐surgery in wild‐type and Mmp3 KO mice. (A) Urinary deoxypyridinoline cross‐links level (DPD) normalized by the amount of urinary creatinine (DPD/creatinine) was measured before (3‐month‐old [3‐mo] mice) and 30 days after surgery (4‐month‐old [4‐mo] mice) in the Sham groups (n = 6–8 animals per experimental groups). (B) DPD/creatinine was also measured in all groups before sacrifice at 4 months of age. Osteoclast parameters were determined by histomorphometry after tartrate‐resistant acid phosphatase staining of femur coronal sections. Osteoclast surfaces per bone surface (Oc.S/BS, C) and osteoclast number per tissue volume (N.Oc/TV, D) were quantified (n = 3–4 animals per experimental groups). All values are expressed as mean ± SD. Significance was determined using the nonparametric Kruskal–Wallis test and the Dunn posttest (^## p < 0.005) or using ANOVA and post hoc Tukey's tests. (*p < 0.05; **p < 0.005; ***p < 0.0005)

FIGURE 4

Effect of human MMP3 promoter polymorphism rs632478 on BMD and MMP3 expression. Women BMD was expressed as Z‐score at femoral neck (A) and at L2‐L4 lumbar spine (B), and compared between the different rs632478 genotypes. Z‐scores at femoral neck (n = 487; G/G n = 124, G/T n = 261, T/T n = 102) and at lumbar spine (n = 485; G/G n = 123, G/T n = 260, T/T n = 102) were measured in a cohort of French women with osteoporosis. Values are expressed as mean ± SD. (C) Relative expression of MMP3 mRNA was measured in cortical bone of human tibial plateau and was presented by class of genotype (G/G n = 8; G/T n = 16; T/T n = 7; n = 31) (C). Values are expressed as mean ± SD (*P < 0.05). Significance was determined using ANOVA and post hoc Tukey's tests. Variance p values are indicated on each panel

FIGURE 5

Effect of SNP variation on the promoter activity of the human MMP3 gene. A, Scheme representing the SNP variations with frequency over 1% found in the 2.0 kb region of the promoter located 5′ of the Human MMP3 gene. (A) Various combinations of haplotypes were determined. SNP rs679620 (Lys45Glu) located in the MMP3 gene coding sequence is also indicated. Percentage of the MAF was estimated for each SNP in a subpopulation of the 1000genome project with European ancestry. (B) Fragments of 2.0 and 1.5 kb of the MMP3 promoter region, both containing the rs632478 G/T variant were cloned in front of the luciferase reporter cDNA sequence in a pGL4.10 reporter plasmid. The plasmids were co‐transfected into MC3T3‐E1 cells along with a construct expressing constitutively the Renilla reporter gene as internal control. Promoter activity is expressed as the ratio of luciferase activity over Renilla activity. All values are expressed as mean ± SD and reported to the expression of the rs632478 C promoter variant. Significance was determined using unpaired t‐ test. (**p < 0.005; ***p < 0.0005) (n = 4 independent transfection experiments performed in triplicate)