The authors state that they have no conflicts of interest.
Genetic analyses with mouse congenic strains for distal Chr1 have identified three closely linked QTLs regulating femoral vBMD, mid-diaphyseal cortical thickness, and trabecular microstructure in a sex-dependent fashion. The homologous relationship between distal mouse Chr 1 and human 1q21–24 offers the possibility of finding common regulatory genes for cortical and trabecular bone.
Introduction: The distal third of mouse chromosome 1 (Chr 1) has been shown to carry a major quantitative trait locus (QTL) for BMD from several inbred mouse strain crosses. Genetic and functional analyses are essential to identify genes and cellular mechanisms for acquisition of peak bone mass.
Materials and Methods: Nested congenic sublines of mice were developed with a C57BL/6J (B6) background carrying <1- to 9-Mbp-sized segments donated from C3H/HeJ (C3H). Isolated femurs from 16-wk-old female and male mice were measured by pQCT and μCT40 for volumetric (v)BMD, mid-diaphyseal cortical thickness, and distal trabecular phenotypes. Static and dynamic histomorphologic data were obtained on selected females and males at 16 wk.
Results and Conclusions: We found that the original BMD QTL, Bmd5, mapped to distal Chr 1 consists of three QTLs with different effects on vBMD and trabecular bone in both sexes. Compared with B6 controls, femoral vBMD, BMD, and cortical thickness (p < 0.0001) were significantly increased in congenic subline females, but not in males, carrying C3H alleles at QTL-1. Both females and males carrying C3H alleles at QTL-1 showed marked increases in BV/TV by μCT compared with B6 mice (p < 0.0001). Females increased BV/TV by increasing trabecular thickness, whereas males increased trabecular number. In addition, the μCT40 data showed two unique QTLs for male trabecular bone, QTL-2 and QTL-3, which may interact to regulate trabecular thickness and number. These QTLs are closely linked with and proximal to QTL-1. The histomorphometric data revealed sex-specific differences in cellular and bone formation parameters. Mice and humans share genetic homology between distal mouse Chr 1 and human Chr 1q20–24 that is associated with adult human skeletal regulation. Sex- and compartment-specific regulatory QTLs in the mouse suggest the need to partition human data by sex to improve accuracy of mapping and genetic loci identification.
Peak bone mass in women and men is generally accepted as being achieved by the end of the second or beginning of the third decade of life.(1–3) Genetic studies have shown that BMD is strongly influenced by heritable factors in humans.(4,5) Estimates of heritability for BMD in humans have ranged from 40% to 90%, depending on the model system.(6) In general, twin studies have indicated that the proportion of variance of BMD accounted for by genetic factors approaches 85–90%.(7) Studies with siblings have yielded somewhat lower estimates of heritability of ∼50–60% that are complicated by issues of environment and site specificity.(8) Nevertheless, locating and identifying those genes that regulate BMD in normal populations is essential to a better understanding of bone biology and for therapeutic inroads to bone disorders, such as osteoporosis.
Genome-wide studies in human populations have been performed with adult women and men to search for genetic association of chromosomal markers with DXA-based BMD of wrists, lumbar spine, and multiple sites in the hip. Population studies have included white American, white European, black, Mexican American, and Mainland Chinese. The outcomes have yielded a rich source of significant and suggestive linkages to 14 autosomes and chromosome (Chr) X,(8–17) even though no single gene accounts for a major part of the trait variation.(18,19) In particular, the frequency of findings with respect to human Chr 1q and bone are provocative. For example (1) absorptive hypercalciuria coupled with decreased spinal aBMD was mapped to 1q23–24, with soluble adenylate cyclase as a strong candidate gene(20,21); (2) quantitative trait loci (QTLs) for lumbar spine aBMD have been reported for 1q21–23,(12) at chromosomal locations of 250(22) and 270 cM(17); and (3) a male-specific QTL for spine aBMD mapped to 1q.(23) Without doubt, BMD is a very complex genetic trait with many genes participating in its regulation.
We and others, using recombinant inbred(24,25) and F2 hybrid progenies,(26–28) have mapped QTLs for whole body aBMD by DXA or femoral vBMD by pQCT to Chr 1 of the mouse. Testing of large segment congenic strains derived from introgressing a major portion of distal Chr 1 into the C57BL/6J (B6) background confirmed the presence of femoral volumetric (v)BMD QTLs regulated by genetic alleles from C3H/HeJ (C3H) or from Castaneus/EiJ (CAST) strains of mice.(29,30) In this report, the focus is on the distal region of mouse Chr 1 that has substantial homology with human Chr 1q21–24 and has a QTL region with the strongest association for vBMD regulation found in crosses of B6 and C3H inbred mice. Data are presented from congenic sublines that fine map the regulation of normal femoral cortical and trabecular bone parameters to a small region of distal Chr 1 with strong sex-dependent effects.
MATERIALS AND METHODS
Husbandry: The inbred mouse strains used for the studies reported herein were obtained from our research colonies at The Jackson Laboratory, Bar Harbor, ME, USA. The B6 and B6.C3H Chr 1 congenic sublines of mice were produced by pair matings, with progeny weaned at 22–25 days of age and housed in groups of two to five of the same sex in polycarbonate cages (324 cm2) with sterilized white pine shavings. Colony environmental conditions included 14:10-h light:dark cycles, with free access to acidified water (pH 2.5 with HCl to retard bacterial growth) that contained 0.4 mg/ml of vitamin K (menadione Na bisulfite), and irradiated NIH31 diet containing 6% fat, 19% protein, Ca:P of 1.15:0.85, and vitamin and mineral fortification (Purina Mills International, Brentwood, MO, USA). All procedures involving mice were reviewed and approved by the Institutional Animal Care and Use Committee of The Jackson Laboratory.
Congenic sublines and genotyping: We have previously reported eight mouse sublines derived from the original B6.C3H-1T congenic strain, several of which showed increased femoral vBMD in females.(30) The genetic segments carried in two of these B6.C3H sublines (hereafter shorthand designations: 1–5, 1–3, etc.) are shown in Fig. 1. The 1–5 congenic strain was previously shown to carry C3H alleles in the Chr 1 region bounded by the markers D1Mit445 and D1Mit403 (genetic interval, 70.0–101.2 cM) and designated Bmp5. Additional genotyping showed that the genetic interval in 1–5 was modestly larger (D1Mit218@67.0 to D1Mit511@109.6 cM), representing physical map positions of 129.86 and 193.86 Mbp, respectively, obtained from NCBI Ensembl Build 36, release 42. To resolve the genetic location of the Bmd5 QTL present in subline 1–5, but not 1–3, new congenic sublines were developed by mating B6.C3H-1–5 N10 generation mice with B6 progenitor strain mice, followed by intercrossing their N11F1 offspring to obtain segregating N11F2 progeny. Genotyping these progeny identified smaller C3H segments resulting from meiotic recombination within the 67.0- to 109.6-cM segment. Mice carrying unique segments were backcrossed to B6 mice. The resultant N12F1 offspring were genotyped, and pairs of mice carrying the desired small segments were intercrossed for N12F2 progeny used to establish new sublines (designated 1–11, 1–20, 1–19, 1–16, 1–12, 1–18, 1–17, 1–14, and 1–15) that are homozygous for different overlapping segments of the C3H genome on distal Chr 1.
Mice were genotyped by preparing genomic DNA from digestion of 1-mm tail tips in 0.5 ml 50 mM NaOH for 10 min at 95°C, and pH was adjusted to 8.0 with 1 M Tris-HCl. Genotyping of individual mouse DNA was accomplished by PCR using oligonucleotide primer pairs (Mit markers, www-genome.wi.mit.edu/cgi-bin/mouse/index) from several sources (Research Genetics, Birmingham, AL, USA; Invitrogen, Carlsbad, CA, USA; IDT, Coralville, IA, USA; Qiagen, Valencia, CA, USA). These primer pairs amplify simple CA repeated sequences of anonymous genomic DNA that are of different sizes and, through gel electrophoresis, can uniquely discriminate between B6 and C3H genomes. Details of standard PCR reaction conditions have been described previously.(31) PCR products from B6, C3H, and F1 hybrids were used as electrophoretic standards in every gel to identify the genotypes of mice (i.e., b6/b6, b6/c3, c3h/c3h). Five markers representing flanking sequences for CA repeats were designed with the aid of Frodo.wi. mit.edu/cgi-bin/primer3/ and MacVector. Custom primer pairs were obtained from Invitrogen and tested against B6 and C3H genomic DNA to confirm polymorphisms (listed in Table 1). In addition, five SNPs within the 170- to 179-Mbp region (NCBI Ensembl Build 36, Release 42) assayed for the sublines are presented in this report. Assays for SNPs were performed at KBiosciences, Herts, UK.
Table Table 1.. Sequences of Custom-Designed Primer Pairs to Discriminate C3H From B6 Alleles on Distal Chr 1
Necropsy for sample collection: At 16 wk of age, when mice have acquired their adult femoral mass,(32) females and males from the B6 progenitors and the newly developed distal Chr 1 sublines were necropsied, whole body weights were recorded, and tissue samples were collected. Skeletal preparations (lumbar vertebral columns, pelvis, and attached hind limbs) were placed in 95% EtOH for a period of not less than 2 wk. Lumbar vertebral columns, tibias, and femurs were dissected free of remaining muscle and connective tissue and placed in 95% EtOH for storage until subsequent analyses were conducted.
pQCT for vBMD bone densitometry
vBMD was measured on the left femur from groups (n = 20–43) of female and male B6 and congenic subline mice. Isolated femur lengths were measured with digital calipers (Stoelting, Wood Dale, IL, USA), and femurs were measured for density using the SA Plus densitometer (Orthometrics, White Plains, NY, USA). Calibration of the SA Plus instrument was accomplished with a manufacturer supplied phantom and with hydroxyapatite standards of known density (50–1000 mg/mm3) with cylindrical dimensions (2.4 mm diameter × 24 mm length) that approximate mouse femurs. Accuracy of linear measures was checked with defined thickness aluminum foils. The bone scans were analyzed with threshold of 710 and 570 mg/cm3, using Orthometrics software version 5.50 yielding cortical bone areas that were consistent with histomorphometrically derived periosteal values. Mineral content was determined with thresholds of 220 and 400 mg/cm3, selected so that mineral from most partial voxels (0.07 mm) would be included in the analysis. Precision of the SA Plus for repeated measurement of a single femur was found to be within 1.2–1.4%. Isolated femurs were scanned at seven locations at 2-mm intervals, beginning 0.8 mm from the distal ends of the epiphyseal condyles. Because of variation in femur lengths, the femoral head could not be scanned at the same location for each bone and thus was not included in final data. Total vBMD values were calculated by dividing the total mineral content by the total bone volume (mg/mm3). Cortical thickness was obtained at the midshaft scan.
μCT40 for distal trabecular bone
Trabecular bone components, as measured by μCT40 are heritable phenotypes(33) distinct from vBMD as measured by pQCT. Femurs from groups (n = 20–28) of female and male B6 and congenic subline mice were scanned using μCT (μCT40; Scanco Medical AG, Bassersdorf, Switzerland) to evaluate trabecular bone volume fraction and microarchitecture in the metaphyseal region of the distal femur. In addition, cortical thickness data were obtained at the midshaft. The μCT40 unit is calibrated weekly with a phantom standard provided by Scanco before beginning bone scans. The femurs were scanned at low resolution, energy level of 45 KeV, and intensity of 177 μA. The distal trabecular scan started ∼0.6 mm proximal to the growth plate and extended proximally 1.5 mm. Approximately 150 cross-sectional slices were made at 12-μm interval at the distal end beginning at the edge of the growth plate and extending in a proximal direction, and 100 contiguous slices were selected for analysis. These were contoured inside the endosteal edge of the cortical shell to obtain the total volume (TV) of the space, followed by analysis of the trabecular bone (BV) with the Scanco software version 5.0. The scans for midshaft cortical thickness were obtained by 18 slices at the exact midpoint of the femur. These slices were contoured by user-defined thresholds for cortical bone and iterated across slices using the Scanco software.
Groups of B6 and 1–11 congenic females and males were dual-labeled for bone histomorphometric analyses. Mice were injected with 20 mg/kg calcein intraperitoneally at 15 wk, and 50 mg/kg demeclocycline 7 days later. Mice were necropsied 48 h after the demeclocycline injection. Femurs were dissected free of muscle tissue and fixed in 70% EtOH. The femur samples were embedded undecalcified in methyl methacrylate. Longitudinal sections, 5 μm thick, were cut on a Micron microtome (Micron; Richard-Allan Scientific, Kalmazoo, MI, USA) and stained with 0.1% toluidine blue, pH 6.4. Static parameters of bone formation and resorption were measured in a defined area between 725 and 1270 μm from the growth plate, using an Osteomeasure morphometry system (Osteometrics, Atlanta, GA, USA). For dynamic histomorphometry, mineralizing surface per bone surface and mineral apposition rate were measured in unstained sections under UV light as described.(34) Bone formation rate was calculated. The terminology and units used are those recommended by the Histomorphometry Nomenclature Committee of the American Society for Bone and Mineral Research.(35)
Statistical evaluation was undertaken using JMP version 6.0 software (SAS, Cary, NC, USA). For the pQCT and μCT40 data, any body size differences between strains were accounted for by using body weight and femur length as covariates when they contributed significantly to an overall analysis of covariance (ANCOVA) model.(36) Data are expressed as adjusted least squares mean ± SE in all figures. Differences between means for B6 and each congenic subline were tested by Student's t-test, with significance declared when p ≤ 0.05 was observed.
Functional correlates for vBMD in subline B6.C3H-1–11 in distal Chr 1
Figure 1 presents the physical location of two sublines derived from the original 1T and a new subline, 1–11, derived from 1–5. The B6 progenitor alleles are shown as black squares and C3H alleles as open squares. Sublines are listed across the top of the figure, whereas a diagrammatic representation of Chr 1 on the left side indicates the C3H allelic region present in the original 1T congenic strain. The DNA markers on the map are ordered by physical map distance (nucleotide number in Mbp) from the centromeric end (www.ensembl.org/Mus_musculus). Subline 1–5 carried the vBMD QTL region, designated Bmd5,(30) and therefore was used to produce additional sublines for biologic and genetic analyses. We found that the 1–11 subline carried a similar phenotype as 1–4, but within a smaller genetic segment of ∼9 Mbp. The vBMD, total femoral mineral, and mid-diaphyseal cortical thickness from female and male femurs are shown in Fig. 2. The increased vBMD seen in 1–5 and in 1–11 subline females is reflected in the increased mineral and thickness parameters. Surprisingly, the 1–5 and 1–11 males showed decreased vBMD, which was reflected in significantly reduced total mineral, with a tendency for both sublines to have reduced cortical thickness.
To gain insight about functional changes in long bones resulting from introgression of C3H alleles in the distal region of Chr 1, we examined static and dynamic histomorphometric measures obtained from femurs of female and male congenic 1–11 mice at 16 wk of age. Table 2 shows that congenic 1–11 females had decreased osteoid surface, increased mineralizing surface, and increased bone formation. In contrast, the congenic 1–11 males showed none of the differences observed in the females. Males did show decreased trabecular thickness, increased number of osteoblasts, and increased mineral apposition rate, but the bone formation rate was not significantly different from B6 male controls. Thus, the C3H alleles in 1–11 mice produced different effects in females and males.
Table Table 2.. Histomorphometric Date From Distal Femurs of B6 and 1–11 Subline Females and Males at 16 wk of Age
Fine mapping of the vBMD QTL in B6.C3H-1–11 subline
Genetic analyses were pursued by production of nested congenic sublines derived from subline 1–11 as described above. Seventeen DNA markers that defined the proximal and distal ends of the C3H genomic segments in each subline are presented on the right side of the haplotype map in Fig. 3. The progenitor alleles, subline designations, and DNA markers are presented in Fig. 1.
The genetic region presented in Fig. 3 encompasses ∼4.4% of Chr 1 genomic sequence near the distal end. The flanking markers for each congenic subline establish approximate genetic boundaries of the chromosomal segment carrying C3H alleles. Three QTL are identified on Fig. 3 as follows: (1) QTL-1 is flanked by genetic markers, rs30595455 and rs6197487 (0.14 Mbp); (2) QTL-2 is flanked by D1Mit111 and rs3710340 (2.93 Mbp); and (3) QTL-3 is flanked by rs3710340 and D1Kls6–1 (1.27 Mbp). These QTL are addressed below.
Femoral density and thickness in congenic sublines for distal Chr 1
Figures 4A and 4B present the femoral vBMD data for progenitor B6 and congenic subline females and males. The sublines are partitioned into two groups of haplotypes (1–11, 1–20, 1–19, 1–16, 1–12 and 1–18, 1–17, 1–14, 1–15) for analyses, according to the presence or absence of D1Mit355. Female vBMD increased in five congenic sublines (1–11, 1–20, 1–19, 1–16, 1–12) compared with B6 controls. These five sublines carry the C3H alleles at the genetic marker, D1Mit355, which defines the QTL-1 region (Fig. 3). Furthermore, sublines 1–19 and 1–12 narrow the QTL-1 candidate region to the interval between markers rs30595455 and rs6197487. In contrast, the vBMD in subline males carrying the C3H alleles at QTL-1 showed decreased vBMD, with four sublines (1–11, 1–19, 1–16, 1–12) significantly lower than B6 male controls. When females and males from the remaining four sublines(1–18, 1–17, 1–14, 1–15) carrying B6 alleles at QTL-1 were analyzed, none showed consistent changes in vBMD.
Figures 4C and 4D present total femoral mineral measured by pQCT for females and males from the nine congenic sublines. The patterns of mineral content closely match the total vBMD patterns observed in both sexes. Thus, changes in mineral in distal Chr 1 are key to the observed femoral vBMD in Figs. 4A and 4B for both females and males.
Figures 4E and 4F present μCT40 data for the femoral mid-diaphyseal cortical thicknesses from all sublines. The data show that all females carrying C3H alleles for QTL-1 have increased cortical thickness. As might be expected, the males from four of the five sublines carrying C3H alleles for QTL-1 have decreased cortical thickness. Mid-diaphyseal cortical thickness data for females and males from the four sublines (1–18, 1–17, 1–14, 1–15) were not statistically different from that of the B6 controls. These μCT40 data are consistent with data on the sublines generated by pQCT.
Trabecular bone in congenic sublines for distal Chr 1
μCT40 was used to phenotype the distal femoral trabecular compartment in the congenic sublines. Figure 5A shows that females from the five sublines carrying C3H alleles at QTL-1 had increased BV/TV compared with B6 controls (similar to pQCT vBMD), whereas Fig. 5B shows that males from sublines 1–11, 1–20, and 1–16 had increased BV/TV, whereas the remaining two sublines were the same as B6 control males. The females from sublines 1–18, 1–17, 1–14, and 1–15 did not show consistent, interpretable changes in BV/TV. On the other hand, males from three of these four sublines (1–17, 1–14, 1–15) had increased BV/TV, whereas subline 1–18 increased but did not reach significance (p = 0.195). These changes suggest a second region of Chr 1 that has a significant impact on male BV/TV, designated QTL-2.
Figures 5C and 5D summarize the female and male trabecular thickness data. Female trabecular thickness was increased in all sublines (1–11, 1–20, 1–19, 1–16, 1–12) carrying C3H alleles for QTL-1. In contrast, males of these sublines did not differ from B6 controls. Females from sublines 1–18, 1–17, 1–14, and 1–15 carrying C3H alleles for QTL-2 showed no differences from B6 controls in trabecular thickness. In contrast, males from 1–18 and 1–17, but not 1–14 or 1–15, had increased trabecular thickness compared with B6 controls. These changes thickness indicate the presence of additional genetic regulation in males, designated QTL-3.
Figures 5E and 5F show female and male data for distal femoral trabecular number. Females from four of five sublines carrying C3H alleles for QTL-1 did not show significant increases in trabecular number. Females carrying C3H alleles for QTL-2 (1–18, 1–17, 1–14, 1–15) did not show consistent changes for trabecular number. Males from four of five sublines (carrying C3H alleles at QTL-1) showed increased trabecular numbers. Males from sublines 1–14 and 1–15 showed increased trabecular numbers, whereas males from sublines 1–18 and 1–17 showed no such increase.
The responses of subline males carrying C3H alleles for the QTL-2 region differ. Sublines 1–18 and 1–17 increased trabecular thickness, whereas 1–14 and 1–15 increased trabecular numbers. Both responses lead to increases in BV/TV.
Our original study for genetic determinants of femoral vBMD in B6C3F2 female progeny yielded 10 chromosomes carrying a density-related QTL. Chr 1 had the strongest association with vBMD, accounted for the largest portion of vBMD variance, and the interval map was suggestive of more than one QTL.(26) Additional studies with different inbred strain crosses by our group and others have also shown QTLs on Chr 1 for vBMD and for aBMD.(24,25,27,28,37,38) The linkage of Chr 1 with regulation of bone properties has been confirmed by B6.C3H congenic strains with B6 as the genetic background carrying large chromosomal segments from C3H, as well as from Castaneus (B6.CAST-1).(30) The B6.CAST-1 congenic sublines revealed the location of active CAST alleles to a region of 172–185 Mbp on the distal region of Chr 1.(29)
In this report, the region of distal Chr 1, subline 1–11, extending from 170–179 Mbp, showed an increase in female vBMD of 3.1%. Although an increase of 8–9% in female vBMD was observed for Chr 1 at N6F2 generation mice,(30) it is likely that the smaller ∼9-Mbp region represented in subline 1–11 does not carry all of the Chr 1 QTL contributing to the difference in vBMD between B6 and C3H progenitor females.(39) It is also possible that interactions between C3H alleles at Chr 1 QTL(s) and C3H alleles at bone QTL on other chromosomes have disappeared as continued backcrossing made the congenic background more B6-like. Finally, the genetic differences described in this report are subtle and represent alleles contributing to regulation of normal bone density and microstructure phenotypes. Thus, one would not expect to observe profound phenotypic changes often introduced by gene knockout or overexpression studies. However, the strong influence of C3H alleles in this region of distal Chr 1 have a measurable impact on vBMD such that positional cloning strategies with nested congenic strains are feasible.
Histomorphometric indices of 1–11 female and male bone showed sex differences of a subtle nature. Data in Table 2 showed that, for 9 of 11 measures, the 1–11 mice either strongly tended toward or were significantly different from B6 controls. Although the direction of change in eight of nine variables (increase or decrease) was opposite for females and males, both sexes showed increased bone formation when C3H alleles were present. Despite the absence of differences in either osteoblast or osteoclast numbers between B6 and 1–11 females, there were increases in mineralizing surface and bone formation that would yield more trabecular bone. Increased trabecular bone in 1–11 females was observed in the μCT40 data presented in Fig. 5. Males showed increases in osteoblast numbers and mineral apposition rate that would predict an increase in the trabecular bone in 1–11 males, as seen in Fig. 5. These data would suggest that the genetic regulation in 1–11 region impacts differential activity of bone cell populations rather than cellular proliferation.
Inspection of pQCT and μCT40 data for femurs from the nine congenic sublines suggest three distinct bone regulatory QTLs dependent on sex. The net effect of C3H alleles at QTL-1 across the five sublines (1–11, 1–20, 1–19, 1–16, 1–12) was to increase female vBMD and to increase trabecular thickness, resulting in increased BV/TV. On the other hand, the net effect of C3H alleles at QTL-1 for males of these sublines was to decrease vBMD, decrease trabecular thickness, and increase trabecular number, resulting in a variable change for BV/TV. Collectively, these observations showed that QTL-1 affects cortical and trabecular bone in a sex-specific manner. We speculate that the males have higher trabecular BV/TV because of a genetic determinant that enhances trabecular bone at the expense of cortical bone. Periosteal and endosteal bone histomorphometry data would be needed to evaluate this postulate.
A second genomic segment, designated QTL-2, was revealed by trabecular data from distal femurs for sublines 1–18, 1–17, 1–14, and 1–15. These sublines do not carry C3H alleles for QTL-1, and thus QTL-1 is not affecting the bone phenotypes in these four sublines. The females of these four sublines did not show consistent changes from B6 in trabecular bone parameters. Males from three of four sublines (1–17, 1–14, 1–15) increased BV/TV, whereas subline 1–18 increased BV/TV, but not significantly, possibly a sampling issue. Therefore, we chose to conservatively locate QTL-2 to address two possibilities: (1) QTL-2 includes subline 1–18 and lies between D1Kls12–1 and D1Kls11–2 or (2) QTL-2 does not include 1–18 and lies between D1Mit111 and D1Kls11–2.
A third genomic region, designated QTL-3, was revealed by differential changes in the trabecular bone compartment of males from sublines 1–18, 1–17, 1–14, and 1–15. Males of sublines 1–18 and 1–17 carrying C3H alleles for QTL-3 showed increased trabecular thickness but no increase in trabecular number. Males of sublines 1–14 and 1–15 carrying B6 alleles at QTL-3 showed no increase in trabecular thickness but did increase trabecular number. We speculate that QTL-3 acts in a dominant fashion to QTL-2 in sublines 1–18 and 1–17, blocking an increase in trabecular number. A less likely alternative is that the increase in trabecular number in sublines 1–14 and 1–15 result from additional genetic regulatory sequences proximal to D1Kls27–1. Thus, QTL-3, which has no effect in females, is male specific in its affect on trabecular thickness.
Both μCT40 and histomorphometry data showed that increased trabecular bone can be achieved by different mechanisms. For example, the μCT40 data showed the 1–11 females obtained their increased trabecular bone by increased thickness and number. The 1–11 males achieved their modest increase by increasing trabecular number despite a decrease in trabecular thickness. The histomorphometric data suggest that the normal complement of osteoblasts in 1–11 females is more effective in bone formation. In the absence of C3H alleles at QTL-1, we did not observe phenotypic differences in females by either pQCT or μCT40. Thus, we speculated that the histomorphometric data in 1–11 females is the consequence of QTL-1. On the other hand, 1–11 males showed effects of all three QTL; thus, the cellular actions seen in the histomorphometric analyses cannot be ascribed to any one of the three QTL.
Candidate genes for vBMD in the distal region of mouse Chr 1 are of interest in light of the homologous relationship with human 1q21–24 wherein evidence of human BMD regulation has been reported.(12,17,20–22,40) The smallest QTL-1 region (175.20–175.34 Mbp) carries one known and one predicted genes currently listed in the NCBI Ensembl database. The known gene, Gm1313, is listed in the Genomics Institute, Novartis Research Foundation (GNF) expression gene database, http://symatlas.gnf.org/SymAtlas/, derived from pooled B6 male and female tissues. GM1313 is highly expressed in bone marrow, but marginally expressed in bone without marrow. The second gene, EG240921, is described as “novel” based on prediction from genomic sequence; no expression information was listed in the Novartis database. These two genes are part of the P200 complex,(41) consisting of up to 12 genes that are described as interferon inducible/activated genes. The meiotic recombination between markers D1Mit355 and rs6197487 partitioned this complex such that most of these P200 genes have been eliminated as candidates. One of these eliminated genes, Ifi204, was initially attractive as a candidate based on reported regulation of osteoblast differentiation through interaction with CBFA1 protein.(42) Whether either of the two candidate genes exerts a role in congenic osteoblasts remains to be determined.
Consideration of possible candidate genes in QTL-2 and QTL-3 remains challenging. The QTL-2 region is 2.93 Mbp in size and is currently reported to contain 71 genes, many of which have significant changes in expression in bone, bone marrow, or both tissues, as reported in the GNF database. The QTL-3 region is 1.27 Mbp in size and is currently reported to contain 31 genes, some with variable gene expression changes in bone, bone marrow, or both according to the GNF database. Because both of these QTL seem to be male specific, comparison of male bones from sublines 1–18, 1–17, 1–14, and 1–15 with B6 should yield differences in either gene expression or protein related to QTL-2 and QTL-3, whereas similar comparison between females should not show differences. Thus, use of sex-specific expression may aid the task of gene identification.
The importance of sex-related differences, where phenotypic expression may be opposite in effect, affect different compartments, or present in one but not both sexes is fully evident in the bone literature. For example, in mice, different QTLs for females and males have been reported for mouse femoral cross-sectional area,(43) aBMD,(44) and vBMD,(45) whereas sex differences have been reported in ERα and β responses(46) and PTH treatment.(47) In humans, sex-specific QTLs for men and women have been reported for phenotypes of hip and spine aBMD.(12,23,48)
The results in this report have several important caveats. First, sex has a profound effect on cortical and trabecular phenotypes affected by expression of the three QTLs in distal Chr 1. An obvious hypothesis is that this could be an example of gonadal steroid regulation of cortical and trabecular QTL gene expression. Second, our congenic sublines provide evidence for two candidate genes in QTL-1; however, they have not yielded definitive evidence for candidate genes for either QTL-2 or QTL-3. More sublines are necessary to discriminate the precise location and phenotypic effects of both QTL-2 and QTL-3. Nevertheless, the nested congenic subline approach reduced the original F2 genomic region from 150 Mbp with a thousand or more genes to three candidate regions of 0.14–2.93 Mbp with smaller numbers of genes. Third, as progress is made to fully annotate the genomic sequences of the mouse, additional genes, perhaps noncoding RNA, will be discovered in one or more QTL regions that will be the candidate genes being sought. In fact, knocking in C3H alleles of this region onto B6 may define new genetic elements. Fourth, the histomorphometry data provided some insight to cellular effects attributable to QTL-1 on distal Chr 1 in females but not males.
Histomorphometric data from subline 1–12 carrying only QTL-1 could provide better insight to the cellular and dynamic effects of this genetic regulation on bone of both sexes.
In summary, congenic strains with small donor chromosomal segments can decompose a QTL consisting of a number of closely linked genes that function together, yielding the final bone phenotype. In this report, nested congenic sublines and phenotyping by both pQCT and μCT40 have revealed three closely linked QTLs that regulate bone parameters encompassing a region of ∼9 Mb containing ∼145 genes. The QTL-1 region identified by pQCT encompasses 0.14 Mb, the QTL-2 region identified by μCT40 encompasses a region as large as 2.93 Mb, and the QTL-3 region identified by μCT40 includes 1.27 Mb. Comparisons of congenic females and males carrying the same genetic complement showed strong sex effects on regulation of specific bone compartments. The congenic data complement the meta analysis for human spine and femoral neck BMD QTLs showing that loci regulate different bone sites in a sex-specific manner,(49) suggesting that these strains provide an experimental tool for identifying homologous bone genes and their functions.
The authors thank Drs AM Dorward and LD Shultz for critical review of the manuscript during its development and Deeana Smith for performing bone histomorphometry. This research was supported by NIH Grants AR-43618 and DAMD17–96-1–6306 (WGB); EY015073 (LRD); AR45433 (CJR); and DK42424 and DK45227 (EC).