PLCL1 rs7595412 variation is not associated with hip bone size variation in postmenopausal Danish women

Osteoporosis is as a skeletal disorder characterized by compromised bone strength predisposing a person to an increased risk of fracture [1]. Bone mineral density (BMD) is commonly used to diagnose osteoporosis and to predict individual fracture risk [2‐4]. Recent genome-wide studies based on hip BMD identified novel susceptibility genes for osteoporosis [5, 6]. Besides BMD, a growing body of evidence suggested that bone size (BS) per se also plays an important role in determining bone strength and fracture risk [7‐9]. Women not only lose bone density after menopause but also have an increase in skeletal size as a result of periosteal apposition [10]. Previous studies showed that women with hip fractures have larger hip BS [11‐20]. Bone must indeed be flexible since it must be able to absorb energy by deforming, to shorten and widen when compressed, and to lengthen and narrow in tension without cracking [21]. Previous studies have demonstrated that BS variation is under strong genetic control, with heritability greater than 50% [9, 22‐25]. Moreover, segregation analyses have suggested that at least one major gene for BS variation exists in Caucasians [26‐28] and in Chinese [29]. Candidate genes have been reported to influence hip BS [30‐32]. However, these studies have presented conflicting data, due in part to small sample size and differences in the genetic background of control and case subjects [33]. Recently, the first genome-wide association study to search for novel genes underlying hip BS variation was conducted [34]. Four SNPs associated with hip BS in the PLCL1 locus had or approached genome-wide significance level in women. PLCL1 encodes an inositol 1,4,5-trisphosphate (IP3) binding protein that can inhibit IP3 mediated calcium signaling [35], an important pathway that regulates the response of bone cells to mechanical signals [36, 37]. The most significant SNP, rs7595412, located in intron 3, achieved a p value of 3.72 × 10^-7. Carriers of the major A allele have around 5 cm² or 17% larger hip BS than non-carriers. In the present study, we analyzed the effect of rs7595412 on hip BS and osteoporosis traits in 1,191 postmenopausal Danish women.

The clinical characteristics of the 1,191 postmenopausal Danish women are presented in Table 1. We genotyped all of them for the rs7595412 SNP and identified 943 A/A, 229 A/G and 19 G/G subjects (Table 2). The genotypic distribution was in Hardy-Weinberg equilibrium (p = 0.24). The PLCL1 polymorphism was associated neither with hip BS nor with spine BS variations (Table 2). Furthermore, a BMD test measuring the mineral density (such as calcium) in the distal part of the arm, femur neck, total femur, total spine was performed (Table 2). No association between the rs7595412 SNP and BMD levels was found. Serum osteocalcin and crosslaps concentrations were also measured in the studied individuals (Table 2). Women carrying at least one copy of the A allele had lower levels of osteocalcin as compared with those homozygous for the G allele (A/A + A/G: 29.18 [19.50-43.65] ng/ml vs G/G: 35.57 [21.47-58.93] ng/ml, p = 0.03) whereas no effect on Crosslaps was detected. History of vertebral and hip fractures was also recorded (Table 2). Women homozygous for the A allele were more affected by such fractures than those carrying at least one copy of the G allele (A/A: 18.8% vs A/G + G/G: 14.9%, p = 0.04) whereas no effect on hip fractures was observed.

Due to its incidence and clinical consequences, osteoporosis followed by vertebral, hip, and forearm fractures represents an outstanding problem of nowadays' health care. Because of its high mortality rate hip fractures are of special interest. The proportion of fractures caused by postmenopausal osteoporosis increases with age. Costs of examinations and treatment of women with postmenopausal osteoporosis and fractures are also increasing and represent a significant amount all over the world. Several risk factors are known in the pathogenesis of osteoporosis, first of all the lack of sufficient calcium and vitamin D intake, age, circumstances known to predispose falling, but also genetic factors. Osteodensitometry by DXA is among the most important method to evaluate osteoporosis, since decrease in BMD, defined as the ratio of the bone mineral content to BS, strongly correlates with fracture incidence. Hip BS, as such, was also found to be a valuable marker for hip fracture [11‐20]. Recently, a genome-wide association study identified the rs7595412 SNP (minor allele frequency = 11.7%), located in the PLCL1 locus, as strongly associated with hip BS variation in 50-year-old subjects [34].

In the present study of postmenopausal Danish women who were 20 years older, we did not observe such association (minor allele frequency = 11.2%). Women over 65 years of age are at particular risk to develop osteoporosis which may partly explain this lack of association. The age-related changes in bone size after the menopause caused by endocortical resorption and periosteal bone apposition have been shown to occur especially in postmenopausal women with increased bone loss [10] but this point is still under debate [41]. The PLCL1 genetic variation was also nominally associated with lower levels of serum osteocalcin. Markers of bone resorption, like Crosslaps, are usually elevated in postmenopausal women with osteoporosis as compared with normal postmenopausal women, but the markers of bone formation, like osteocalcin, are much less elevated and may indeed be decreased [42, 43]. This pattern of changes in bone turnover suggests that an extent of imbalance of bone resorption and bone formation occurs in osteoporosis [44]. Serum levels of osteocalcin were previously found to be either lower [45, 46], similar [47, 48] or even slightly elevated [49, 50] in patients with postmenopausal osteoporosis than in the control subjects. In their study, Liu and colleagues did not find any effect of PLCL1 SNPs on hip and spine BMD [34]. We confirmed the lack of effect on spine BMD and found no association with femur BMD. However, Liu and colleagues found marginally significant association of the PLCL1 SNPs with spine BS [34]. We did not observe any effect on spine BS but our results suggest an effect on the backbone since the proportion of vertebral fractures was slightly higher in postmenopausal women homozygous for the rs7595412 A allele than in the other subjects. At menopause, bone turnover increases, leading to poorer bone quality and if the increased bone resorption is not balanced with bone formation, the risk of fracture increases too [51]. Given the low proportion of hip fractures in our study samples, we could not correctly assess the impact of the PLCL1 rs7595412 polymorphism on this phenotype.