Regulation of Bone Mass in Mice by the Lipoxygenase Gene Alox15

See allHide authors and affiliations

Science  09 Jan 2004:
Vol. 303, Issue 5655, pp. 229-232
DOI: 10.1126/science.1090985


The development of osteoporosis involves the interaction of multiple environmental and genetic factors. Through combined genetic and genomic approaches, we identified the lipoxygenase gene Alox15 as a negative regulator of peak bone mineral density in mice. Crossbreeding experiments with Alox15 knockout mice confirmed that 12/15-lipoxygenase plays a role in skeletal development. Pharmacologic inhibitors of this enzyme improved bone density and strength in two rodent models of osteoporosis. These results suggest that drugs targeting the 12/15-lipoxygenase pathway merit investigation as a therapy for osteoporosis.

Osteoporosis is one of the most common bone and mineral disorders in all aging communities. It is characterized by low bone mass (and thus, low bone strength), which results in fractures from relatively minor trauma. Although life-style and environmental factors play key roles in the development of osteoporosis, there is now clear evidence that genetic factors are also of great importance (1). Bone mineral density (BMD) achieved in early adulthood (peak bone mass) is a major predictor of osteoporotic fracture risk. Genetic segregation analyses in inbred mouse strains (2) have identified linkage between peak BMD and several chromosomal regions (or quantitative trait loci, QTLs), but the identities of the underlying genes remain unknown. Recent studies suggest that regulatory variation is important in a variety of complex traits (3). Quantitative gene expression studies can identify genetic variation affecting transcription within genes contributing to differences in complex traits. This is particularly useful for analysis of traits for which a priori gene candidates do not exist.

To identify genes that might regulate BMD, we investigated a region on mouse chromosome 11 that strongly influences peak BMD (4). We generated a DGA/2 (D2) background congenic mouse with an 82-megabase (Mb) region of chromosome 11 replaced by the corresponding region of the C57BL/6 (B6) genome. The congenic mice with the B6 chromosome 11 region had increased peak BMD (whole body and femoral) and improved measures of femoral shaft strength (failure load and stiffness) relative to heterozygous or D2 littermates (Fig. 1, A to D). Linkage analysis of the congenic B6D2F2 population narrowed the BMD QTL to a 31-Mb region between 54.7 and 85.4 Mb on chromosome 11 (Fig. 1E) (5).

Fig. 1.

Analysis of chromosome 11 of F2 congenic mice (n = 8 to 10) bearing B6, D2, or heterozygous (Hz) alleles for the chromosome 11 introgressed region. Values are expressed as means ± SEM. (A) Peak whole-body BMD. Analysis of variance (ANOVA) indicated a significant effect of genotype on BMD (*P < 0.0001). Methods are given in (5). (B) Femoral BMD. *P < 0.005 by ANOVA. Isolated femora were tested to failure in response to three-point bending. *P < 0.01 by ANOVA. (C) Femoral shaft failure load (ultimate force expressed as newtons). (D) Femoral shaft stiffness (expressed as newtons per millimeter of displacement). (E) To refine the chromosome 11 BMD QTL, additional microsatellite markers within the introgressed region were selected for genotyping, and the congenic F2 intercross data set was analyzed with the Map Manager QT software program (28). The peak logarithm of the likelihood for linkage (or LOD score) mapped to a 31-Mb chromosomal region between D11Mit86 and D11Mit355, at 54.7 and 85.4 Mb distal from the centromere of chromosome 11, respectively. For comparison, the chromosomal location of Alox15 at 71 Mb is indicated by an arrow.

We next analyzed gene expression in B6 and D2 mice to identify the genetic locus within this region. Microarray analysis of kidney tissue indicated that Alox15, located in the middle of the identified QTL interval at 71 Mb, was the only differentially expressed gene within this chromosomal region. Alox15 expression in the D2 kidney was nearly 20 times that observed in B6 kidney. Quantitative reverse transcription polymerase chain reaction (RT-PCR) analysis using Alox15-specific oligonucleotide primers confirmed expression differences in both whole kidney (Fig. 2A) and osteoblastic cell cultures (Fig. 2B). Analysis of genomic DNA identified 15 polymorphisms in the Alox15 gene that distinguished the D2 and B6 strains (fig. S1).

Fig. 2.

Expression of Alox15 mRNA in kidney and osteoblastic cultures. (A) Expression of Alox15 mRNA in whole kidney. Total RNA was isolated from excised kidneys of individual 8-week-old B6 and D2 mice (5). RT-PCR analyses were performed using primers specific for 12/15-LO and β-actin. Expression of Alox15 mRNA in each sample was corrected for β-actin levels and expressed relative to the corresponding B6 value. Each data point was generated by at least three replicates, and data were analyzed with the paired t test (*P < 0.01). (B) Total RNA was isolated from primary neonatal calvarial osteoblast cultures of B6 and D2 mice (5). RT-PCR analyses were performed using primers specific for Alox15, CD36, osteocalcin (OC), and β-actin. Expression of specific mRNA species in each sample was corrected for β-actin levels and expressed relative to the corresponding B6 value. Each data point was generated by at least three replicates, and data were analyzed with the paired t test (*P < 0.01).

The Alox15 gene encodes 12/15-lipoxygenase (12/15-LO), an enzyme that converts arachidonic and linoleic acids into endogenous ligands for the peroxisome proliferator–activated receptor–γ (PPARγ) (68). Activation of this pathway in marrow-derived mesenchymal progenitors stimulates adipogenesis and inhibits osteoblastogenesis (9, 10). The fatty acid translocase, CD36, is positively regulated by 12/15-LO products acting through their interaction with PPARγ (6). Expression of CD36 mRNA was markedly up-regulated in neonatal calvarial cell cultures isolated from D2 mice compared with those from B6 mice, whereas osteocalcin (a marker of differentiated osteoblastic function) was substantially reduced (Fig. 2B), which indicates the biological impact of increased Alox15 expression. Moreover, transient overexpression of 12/15-LO in murine bone marrow stromal cell cultures restricted osteoblast differentiation as evidenced by reductions in alkaline phosphatase activity and secretion of osteocalcin (fig. S2). On the basis of these in vitro observations, we hypothesize that genetically determined, constitutively high 12/15-LO expression limits peak bone mass attainment by suppressing osteogenesis through activation of PPARγ-dependent pathways.

To explore the role of 12/15-LO in skeletal development in vivo, we first examined the skeletal phenotype of 12/15-LO knockout (12/15-LOKO) mice (11). As the targeted mutation (Alox15tm1Fun) is maintained on a B6 genetic background, we compared the skeletal phenotype of 12/15-LOKO mice to that of age-matched B6 progenitors. Although body weight and whole-body BMD were similar for the two strains, femoral BMD and biomechanical indices of femoral shaft strength were increased in the 12/15-LOKO mice (Fig. 3A). In companion studies designed to limit the possible effects of background strain, we examined the effect of crossbreeding D2 and 12/15-LOKO mice on bone mass acquisition. Heterozygous offspring from this pairing were interbred, and F2 mice with 0, 1, or 2 copies of the 12/15-LO D2 allele were obtained. The 12/15-LOKO-D2 F2 population was composed of approximately equal numbers of male (n = 141) and female (n = 151) mice, and the Alox15 genotype frequency conformed to Mendelian segregation expectations with 70 (24%) homozygous knockout (no 12/15-LO allele) mice, 153 (52%) heterozygous (single 12/15-LO allele from D2) mice, and 69 (24%) homozygous D2 (both 12/15-LO alleles from D2) mice. 12/15-LO–deficient F2 mice exhibited significantly higher whole-body BMD than that of mice homozygous for the 12/15-LO D2 allele (Fig. 3B). Furthermore, 12/15-LO–deficient F2 mice exhibited improved femoral structural competence as evidenced by increased failure load and stiffness (Fig. 3B). Thus, reduced expression of Alox15 rescued mice from the low–bone mass phenotype associated with the 12/15-LO D2 allele.

Fig. 3.

Effect of genetic manipulations of 12/15-LO on skeletal development. (A) 12/15-LO–knockout (12/15-LOKO) mice (n = 25) and B6 (wild-type, n = 28) mice used in these experiments were bred under identical conditions. Femoral BMD and biomechanical measures of femoral shaft strength were determined as described (5). Values are expressed as means ± SEM. *P < 0.01 and †P < 0.05 by paired t test. (B) Genetically heterogeneous F2 mice (n = 15/group) with 0 or 2 copies of the 12/15-LO D2 allele were generated from parental 12/15-LOKO and D2 mice. Whole-body BMD and biomechanical measures of femoral shaft strength were determined as described (5). Values are expressed as means ± SEM. *P < 0.01 and †P < 0.05 by paired t test.

Finally, we examined the skeletal effects of pharmacological inhibitors of 15-lipoxygenase in two rodent models of osteoporosis. Constitutive expression of an interleukin 4 (IL-4) transgene in B6 mice (Tg-IL-4) results in reduced peak bone mass and biomechanical strength (12). As IL-4 is known to upregulate Alox15 expression in a number of tissues, we hypothesized that increased Alox15 expression may contribute to the defective skeletal phenotype in this model. At the time of weaning, transgenic Tg-IL-4 mice were pair-fed for 12 weeks with either control rodent feed or feed supplemented with the specific 12/15-LO inhibitor, PD146176 (13, 14). Treatment with PD146176 resulted in increased whole-body BMD, femoral BMD, and femoral shaft failure load (Fig. 4). To determine the impact of 12/15-LO inhibition on bone loss, we tested another 12/15-LO inhibitor, RO4508159 [[[5-(5,6-difluoro-1H-indol-2-yl)-2-methoxyphenyl]amino]sulfonyl]-carbamic acid-isobutyl ester in ovariectomized rats, a standard osteopenia assay. Two weeks after ovariectomy, RO4508159 was administered daily by oral gavage for 11 weeks. Control groups (both rats that had sham operations and controls that were ovariectomized) received vehicle only. Treatment with RO4508159 attenuated ovariectomy-induced bone loss (Fig. 4B) at both the spine (45%, P < 0.01) and proximal femur (44%, P < 0.05). These results indicate that pharmacological inhibition of 12/15-LO in vivo can improve bone mass and strength during skeletal development, as well as offset the bone loss that accompanies estrogen deficiency.

Fig. 4.

Skeletal effect of pharmacological inhibition of 12/15-LO activity. (A) Transgenic mice overexpressing interleukin 4(Tg-IL-4) were pair-fed with either control rodent feed or feed supplemented with a specific 12/15-LO inhibitor (PD146176; 300 μg/g of feed) for 12 weeks (n = 20 per group). Whole-body BMD and biomechanical measures of femoral shaft strength were determined as described previously (5). Values are expressed as means ± SEM. *P < 0.01 by paired t test. (B) Effect of pharmacological inhibition of 12/15-LO activity on ovariectomy-induced bone loss. Three-month-old rats were ovariectomized and administered 1 mg/kg per day RO4508159 by daily oral gavage starting 2 weeks after ovariectomy and continuing for 11 weeks. Control groups, both rats that were not ovariectomized (but had sham operations) and ovariectomized rats, received vehicle only (n = 10 per group). The BMD values for the spine and right hip were determined on anesthetized animals (5). Rates of BMD loss for each group were compared to the rate for sham-operated rats at the appropriate time point and expressed as mean ± SEM. *P < 0.05 and †P < 0.01 by paired t test.

Our findings demonstrate that genetic variation within the 12/15-LO locus contributes to naturally occurring variation in peak bone mass and identify a novel pathway that regulates skeletal development. However, it remains possible that other genes within this 31-Mb region on chromosome 11 also contribute to the observed variation in peak BMD.

Lipoxygenases are nonheme, iron-containing enzymes that catalyze the oxygenation of certain polyunsaturated fatty acids, such as lipids and lipoproteins. 15-Lipoxygenase has been implicated in the pathogenesis of several diseases, including atherosclerosis (15), asthma (16), cancer (17), and glomerulonephritis (18). The biological functions of murine or human 15-LO have not yet been determined with certainty. Nevertheless, there is accumulating evidence to suggest a potential mechanism by which overexpression of 12/15-LO could exert a negative effect on skeletal development. Pluripotent marrow stromal cells can differentiate into one of several mature forms including adipocytes and osteoblasts, a process regulated by both protein and lipid factors. In many instances, lipid regulation of differentiation is mediated through PPAR-dependent signaling pathways. Linoleate is the most abundant fatty acid in low density lipoprotein (LDL) and is thought to be the largest reservoir of 12/15-LO substrate. Oxidized LDLs serve as PPARγ ligands (19) and have been shown to activate CD36 expression (20). Furthermore, oxidized lipids inhibit osteoblastic differentiation from preosteoblasts in vitro (21, 22) and bone formation in vivo (23). In addition, 5-lipoxygenase metabolites of arachidonic acid inhibit bone formation in vitro (24) and 5-LO–deficient mice exhibit increased cortical thickness (25); however, no BMD QTL has been identified on chromosome 6 where Alox5 resides.

The identification of Alox15 as a susceptibility gene for peak BMD in mice may have relevance to human osteoporosis. An autosomal genome screen for spinal BMD in 17 extended pedigrees found linkage to a chromosomal region (17p13.1) containing the genes encoding human 12-LO and 15-LO (26). In addition, an association between a single-nucleotide polymorphism of PPARγ and BMD was identified in postmenopausal women (27). Further studies in both animal models and human populations will be required to gain a deeper understanding of the role the 12/15-LO pathway plays in processes leading to peak bone mass attainment. If 12/15-LO is confirmed to contribute to human osteoporosis risk, inhibitors of the enzyme may merit investigation as a treatment for osteoporosis. Such inhibitors have already been developed for other indications (14).

Supporting Online Material

Materials and Methods

Figs. S1 and S2

References and Notes

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article