Serum alkaline phosphatase activity is regulated by a chromosomal region containing the alkaline phosphatase 2 gene (Akp2) in C57BL/6J and DBA/2J mice

Jennifer E. Foreman1,2, David A. Blizard1, Glenn Gerhard5, Holly A. Mack1,3, Dean H. Lang1,4, Kathryn L. Van Nimwegen1,3, George P. Vogler1,3, Joseph T. Stout1, Zakariya K. Shihabi6, James W. Griffith7, Joan M. Lakoski8, Gerald E. McClearn1,3 and David J. Vandenbergh1,2,3

1 Center for Developmental and Health Genetics
2 Intercollege Program in Genetics
3 Department of Biobehavioral Health
4 Department of Kinesiology, The Pennsylvania State University, University Park, Pennsylvania
5 Weis Center for Research, Geisinger Clinic, Danville, Pennsylvania
6 Department of Pathology, Wake Forest University Medical Center, Winston-Salem, North Carolina
7 Department of Comparative Medicine, College of Medicine, The Pennsylvania State University, Hershey, Pennsylvania
8 Department of Pharmacology, University of Pittsburgh, Pittsburgh, Pennsylvania


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Quantitative trait locus (QTL) analyses were conducted to identify chromosomal regions that contribute to variability in serum alkaline phosphatase (AP) enzyme activity in mice derived from the C57BL/6J (B6) and DBA/2J (D2) inbred strains. Serum AP was measured in 400 B6D2 F2 mice at 5 mo and 400 B6D2 F2 mice at 15 mo of age that were genotyped at 96 microsatellite markers, and in 19 BXD recombinant inbred (RI) strains at 5 mo of age. A QTL on the distal end of chromosome 4 was present in all sex- and age-specific analyses with a peak logarithm of odds (LOD) score of 20.36 at 58.51 cM. The Akp2 gene, which encodes the major serum AP isozyme, falls within this QTL region at 70.2 cM where the LOD score reached 13.2 (LOD significance level set at 4.3). Serum AP activity was directly related to the number of D2 alleles of a single nucleotide polymorphism in the 5'-flanking region of the Akp2 gene, although no strain-related differences in hepatic expression of Akp2 RNA were found. A variety of sequence polymorphisms in this chromosomal region could be responsible for the differences in serum AP activity; the Akp2 gene, however, with several known amino acid substitutions between protein sequences of the B6 and D2 strains, is a leading candidate.

blood; aging; liver enzymes; bone enzymes; quantitative trait locus


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
ALKALINE PHOSPHATASE (AP) is a family of enzymes found in essentially all tissues that catalyze the hydrolysis of inorganic or organic phosphate esters with a broad specificity for ligand, including pyrophosphate. Detectable amounts of the enzyme are found in serum, derived mainly from hepatocytes and osteoclasts (37), and AP is a frequently assayed enzyme used in blood analysis for medical tests as an indicator of disease status (20). An elevated level of serum AP may be an indication of abnormal liver or bone function (14, 17). Normal values of serum AP activity also vary with age and gender, with an increase over time of up to 20% in men and up to 37% in women (4, 18).

In humans, there are four AP isozymes encoded by four homologous genes: placental, intestinal, germ cell, and tissue nonspecific AP (TNAP) (21). The importance of the TNAP gene in humans has been shown by the autosomal recessive disease hypophosphatasia, which has a range of phenotypes from stillbirth to loss of teeth in old age (21). These phenotypes are caused by abnormalities in bone mineralization associated with polymorphisms in the TNAP gene (21).

Five AP genes are present in mice, Akp1Akp5, in addition to one pseudogene. The Akp1, -3, -4, and -5 genes are found on chromosome (Chr) 1 at 100.6, 51.7, 71.6, and 54.0 cM, respectively. The Akp2 gene encodes the liver/bone/kidney isozyme and maps to mouse Chr 4 at 70.2 cM (1, 30, 36). The mouse Akp2 gene is also referred to as TNAP (21), because it is homologous to the TNAP gene in humans and maps to a region on mouse Chr 4 that is syntenic with human Chr 1 (28, 35). Furthermore, the Akp2 gene in mice has been shown to be functionally homologous to the TNAP gene in humans with symptoms of hypophosphatasia in Akp2-deficient mice. Most Akp2 –/– knockout mice show no detectable serum AP activity (21) while 30% of the knockouts show low levels of serum activity, which was attributed to the intestinal AP isozyme. Heterozygous mice had 50% lower AP enzyme activity in serum than wild-type mice (21), consistent with what would be expected if the Akp2 locus were the primary source of serum AP activity.

Polymorphisms in the Akp2 alleles found in C57BL/6J (B6) and DBA/2J (D2) mouse strains were originally detected by cellulose acetate electrophoresis. The AP isozyme from bone migrated as a broad band after neuraminidase treatment, with the B6 form migrating more rapidly than the D2 form (36). Several online single nucleotide polymorphism (SNP) databases, including the National Center for Biotechnology Information (NCBI; http://www.ncbi.nlm.nih.gov) and Roche (http://mousesnp.roche.com/cgi-bin/msnp.pl), list multiple SNPs at the Akp2 locus of B6 and D2 strains.

To define further the genetic basis of variability in serum AP activity, we conducted quantitative trait locus (QTL) analyses to identify chromosomal regions associated with serum AP levels in F2 and BXD recombinant inbred (RI) mice at 150 days of age (adult) and F2 mice at 450 days of age (middle age). A major QTL on Chr 4 encompassing the Akp2 locus was associated with variation in serum AP levels in B6D2 mice.


    METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Sample
As part of an age-related study, a B6D2 intercross was produced to provide 400 F2 animals (equally divided by sex) at 150 and 450 days of age. The progenitor strain mice (C57BL/6J x DBA/2J) were purchased from the Jackson Laboratory (Bar Harbor, ME) and maintained in a colony at The Pennsylvania State University. The colony was well established when F1s for this study were generated. Reciprocal F1 mating pairs were created to balance any maternal effects and balance for the origin of the Y chromosome (15). The objective was to examine the F2 cross and RI strains derived from two genetically different strains of mice on various age-related phenotypes. The animals in the current study are referred to as being 150 and 450 days of age, the central age of three blood samples for each age group. Other publications using the same sets of animals but only using measurements taken at death refer to the animals by their age at death, 200 and 500 days. Details on the mating scheme of these animals and other particulars in the study are reported by Lionikas et al. (15).

Measurement of Serum AP
Triplicate measures of the AP activity levels were taken for the F2 mice, and a single measure was taken for the RI mice. Blood was obtained by tail nick for the 150-day-old mice at ~120, 148, and 175 days of age ± 2 wk. Similarly, for the 450-day-old mice, blood was drawn at 424, 452, and 480 days of age ± 2 wk (times 1, 2, and 3, respectively, for each age group) in the F2 mice. The three time points were separated by 50 days to allow the mice to recover from the previous blood draw. The progenitor and RI strains had blood taken at only one time point at 183 days of age ± 2 wk. All procedures with mice followed the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council, National Academy of Science, 1996) and were approved by The Pennsylvania State University Institutional Animal Care and Use Committee.

After serum was collected by standard procedures, the samples were frozen and sent to the Clinical Blood Chemistry Laboratory at Wake Forest Medical Center for analysis on an ADVIA 1650 analyzer (Bayer, Terrytown, NY). A standard battery of blood tests was performed on the samples including an assessment of AP levels. The enzyme was measured kinetically based on hydrolysis of p-nitrophenyl phosphate at 410 nm using 2-methyl-1-propanol at pH 10.3 as a buffer. A problem in transportation of the samples from time 1 in the 150-day-old F2 mice and 450-day-old RI mice resulted in the inability to measure AP activity accurately; therefore, the F2 150-day data from time point 1 were not included in these analyses, and the 450 RI data were lost.

Genotyping
Tail DNA extraction and purification were carried out as previously described (32). Genotyping was conducted using a set of 96 markers from the MIT genomic database (http://www-genome.wi.mit.edu). Primers for the genotyping were purchased from Research Genetics, (Huntsville, AL). A table describing the primers for each marker, the fluorescent dye tag on the forward primer, and the pools of primers prepared for multiplex PCR amplification are available on the web (http://www.cdhg.psu.edu/GeneticMarkerAnalysis/) and are described in more detail in an earlier publication (32).

AP Genotyping
The Celera mouse database, which contains sequence information on B6 and D2 strains, was searched for SNPs at the Akp2 locus. Short sequences encompassing these SNPs were then searched for restriction enzyme recognition sequences. One SNP residing 679 bp upstream of the transcription start site generated a PstI site in D2 mice but not in the B6 strain. This PstI restriction fragment length polymorphism (RFLP) was then used to distinguish between the B6 and D2 Akp2 alleles and genotype the 150-day-old F2 population. This same polymorphism was used to map the gene to Chr 4 (29). Genomic DNA was amplified in a 10-µl final volume consisting of forward (5'-AGA GGA TAG TGT CTG CGG CTT AT-3') and reverse (5'-GCC CTT AAC TAA GTC AGT AGG AAG C-3') primers (Integrated DNA Technology, Coralville, IA) at 2 µM, 2.5 mM MgCl2, 10 mM deoxyribonucleotide triphosphates (dNTPs), 10% Tween 20, and 4 mM Spermidine, using either AmpliTaq Gold (Applied Biosystems) or JumpStart Taq (Sigma-Aldrich). After an initial denaturation step at 95°C for 2 min (AmpliTaq) or 94°C for 1 min (JumpStart), 35 cycles were carried out at 95°C for 45 s, 59°C for 45 s, and 72°C for 60 s, followed by a final extension step of 72°C for 10 min.

A portion of the PCR product (5 µl) was digested in a 15-µl volume with 5 units of PstI restriction enzyme and the appropriate buffer provided by the supplier (Gibco BRL) for 2 h at 37°C. The entire digest was electrophoresed for 90 min at 150 V on a 5% polyacrylamide gel. The gel was stained with ethidium bromide and visualized under ultraviolet (UV) light. The B6 allele (no PstI site) was 236 bp in size, and the D2 allele consisted of two fragments with 167 and 69 bp.

mRNA Quantification
Akp2 gene expression was analyzed using mRNA isolated from liver tissue. Messenger RNA was quantified by RT-PCR on mRNA isolated from the progenitor strains and verified using Northern blot analysis of F2 animals. Four mice of each strain and sex, 16 total, were used for the RT-PCR verification. DNA, RNA, and protein were extracted using Trizol (Invitrogen Life Technologies, Carlsbad, CA). The mRNA quality was tested before use in RT-PCR using the Agilent Bioanalyzer. A Taqman gene expression assay kit for the mouse Akp2 gene was used for the RT-PCR and run on an ABI 7300. The liver samples collected from the F2 150-day-old mice were analyzed for both sexes and all three possible genotypes at the Akp2 site. Northern blots were prepared following standard protocols (2, 31). The gels were loaded with 5 µg of RNA from each mouse, and the filters were cross-linked by UV light while still moist.

A DNA probe to detect Akp2 mRNA was synthesized by PCR for a part of the last exon of the Akp2 gene to minimize possible cross-hybridization to other Akp genes. Primers were chosen using Primer3 (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) to generate an 841-bp fragment. The primers, forward (5'-AACAACTACCAGGCCCAGTC-3') and reverse (5'-TTCCAAACAGGACAGCCACT-3'), were used in a PCR amplification of mouse genomic DNA using the same conditions as for the genotyping PCR, except that MgCl2 was at 1.5 mM, with 33 cycles at 95°C for 45 s, 64°C for 45 s, and 72°C for 90 s. The product was purified by gel electrophoresis on a 5% acrylamide gel. To ensure that the correct DNA had been amplified, a portion of the product was digested in separate reactions with the restriction enzymes BsmFI, BglII, and AccI. All fragments detected by gel electrophoresis were of the predicted sizes based on analysis using a restriction enzyme site mapper (http://www.restrictionmapper.org/).

Radioactively labeled probe DNA was prepared with the random primer labeling kit Prime-It II (Stratagene, Cedar Creek, TX). The probe was added to the Northern blot following the manufacturer’s protocol for UltraHyb (Ambion, Austin, TX), washed, and exposed to film (XLS1; Kodak, Rochester, NY). Densitometry of the resulting autoradiograms was carried out on an Amersham Biosciences PDSI. The images were then quantified using ImageQuant software (Amersham Biosciences, Piscataway, NJ).

Sequencing
A 252-bp genomic fragment of the mouse Akp2 gene corresponding to base pair positions 12,145 to 12,396 of Akp2 sequence data contained in the NCBI nucleotide database (accession ID: AF285233) was amplified from genomic DNA obtained from DBA/2J liver tissue using forward primer 5'-gccatgtccacaggtctctt-3' and reverse primer 5'-agcatcttcgcatccactct-3. The genomic fragment was amplified with the following thermal cycling conditions: 5 min at 94°C, followed by 39 cycles of denaturation at 94°C for 30 s, annealing at 60°C for 1 min, and elongation at 72°C for 1 min. A single band was obtained and excised following agarose gel electrophoresis for DNA isolation (Qiagen). The isolated DNA fragment was then subject to automated sequencing (Genewiz).

Data Analyses
The enzyme activity data differed statistically from a normal distribution, and the F2 150-day data had two outlying points. Points that were four standard deviations above the mean were excluded. The data were adjusted to a normal distribution using a square-root transformation. Initial QTL analyses were performed on both the transformed data and the untransformed data. No appreciable differences in QTL detection were present between the two data sets; therefore, only the results for the untransformed data are reported. QTL analysis was conducted on each sex separately and with the sexes combined. For combined analyses at each age, adjustment for the sex differential (males lower) was performed by adding the difference between the male and female means to each male value. Thus the means of both sexes were centered on the same value without changing the variance. Strain means were used for the RI analyses.

Interval and Composite Interval Mapping
Interval mapping (IM) and composite interval mapping (CIM) were used to detect QTL regions for serum AP using QTL Cartographer 2.0 (33). IM estimates QTL positions using two flanking markers incorporating recombination distances between the two markers. This method removes the confounding effect that occurs between recombination distance and effect size in single-marker mapping, allowing a more precise estimate of effect size and location. CIM was used to distinguish between a single QTL of large effect size and two closely linked QTLs of smaller effect size. CIM takes into account other significant chromosomal areas found during the analysis, using multiple regression to separate the effects of linked QTLs (38, 39). This type of analysis additionally increases the power and precision to detect a QTL because the test statistic is independent of QTL effects at other regions, and it can include unlinked markers to remove genetic variance attributed to QTLs linked to these markers.

The analyses were performed with the individual AP time point measures, as well as the average AP values for all time points within an age group. All three measures gave similar results, but the mean results gave marginally higher logarithm of odds (LOD) scores due to a reduction in the within-individual measurement error. Only the results utilizing the averages of the AP measures are reported. For the F2 mice, a QTL was considered suggestive if the LOD score ranged from 2.8 to 4.2 and significant with a LOD score of 4.3 or higher. In the RI mice, a QTL was reported as significant with a LOD score of 3.3 or higher. A QTL was considered confirmed if it appeared in both F2 age groups as suggestive or significant, or the QTL was at least at the suggestive level in one F2 age group and also appeared in the RI group with an LOD score >1.5 (12).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Enzyme Activity
AP activity levels in the progenitor strains were measured at 150 days and were higher than the levels seen in the 150-day F2. The progenitor strains had an AP activity at 150 days of age that was numerically equivalent to those seen in the F2 mice at 450 days of age (Table 1). The 150-day RI strain means ranged from 41 to 154 U/l in females and from 35 to 125 U/l in males (data not shown). Analysis of serum chemistry indicated that males and females for both F2 age groups and the two parental strains differ in serum AP activity. For the progenitor strains, significant sex (F = 38.66, P value < 0.001) and strain effects (F = 6.03, P value = 0.018) were determined using one-way ANOVA. D2 strain females had the highest mean serum AP activity and B6 strain males the lowest. The F2 generation showed a similar sex effect in both the 150-day (mean of samples 2 and 3; F = 73.39, P value < 0.001) and 450-day (mean of 3 samples; F = 81.29, P value < 0.001) age groups.


View this table:
[in this window]
[in a new window]
 
Table 1. Male and female mean serum AP activity for progenitor strains, F2 and RI population

 
QTL Mapping
QTL analysis of the relationship between serum AP activity and genotypes in the F2 for both F2 age groups and 150-day-old RI mice revealed a highly significant locus on Chr 4, one significant QTL on Chr 3, two suggestive QTL on Chr 3, a suggestive QTL on Chr 12, and a suggestive QTL on Chr 18 (Fig. 1 and Table 2). The significant QTLs are designated as Salpa3 and -4 for Serum AP Activity. Salpa1 and -2 were previously identified by Srivastava et al. (26). Salpa3, the highly significant QTL detected by IM on Chr 4, had a peak LOD score of 20.36 at 58.51 cM in the F2 150-day combined group and a peak LOD score of 20.99 at 65.91 cM in the F2 450-day combined group. Salpa3 was also present in all other analyses, except in the male RI 150-day group (Table 2), and is further refined in the CIM analyses for both age groups (peak LOD 19.33 at 58.51 cM for the F2 150-day combined group and peak LOD 21.73 at 65.91 cM for the F2 450-day combined group; see Fig. 2A). Salpa3 was present in both male and female data in the F2 150-day group, with IM LOD scores of 10.06 at 56.51 cM and 11.0 at 60.51 cM. The Salpa3 region accounted for ~20–34% (R2 estimates across the multiple QTL analyses) of the genetic variance in serum AP activity.



View larger version (37K):
[in this window]
[in a new window]
 
Fig. 1. Composite interval mapping (CIM) results for alkaline phosphatase (AP) levels from F2 150- and 450-day-old mice. Combined male and female data were used for both age groups. Solid line is the AP mean for the 150-day-old mice, and dashed line is the AP mean for the 450-day-old mice. Horizontal line at 4.3 represents the cutoff for significance.

 

View this table:
[in this window]
[in a new window]
 
Table 2. QTL table

 


View larger version (16K):
[in this window]
[in a new window]
 
Fig. 2. Detailed representation of quantitative trait locus (QTL) on chromosome (Chr) 4 (Salpa3). Marker positions are shown on as carrot marks; they are, from centromere to telomere, D4mit104, D4mit214, D4mit255, D4mit204, D4Akp2, and D4mit190. The D4Akp2 marker was genotyped after the first QTL analysis (A), and its effect is shown (B). A: CIM analysis of AP mean for the F2 150-day-old mice (solid line) and AP mean for the F2 450-day-old mice (dashed line). B: interval mapping (IM; solid line) and CIM (dashed line) analyses of QTL on Chr 4 including genotype information on Akp2 as a marker for the F2 150-day-old mice.

 
Salpa4, the significant QTL found on Chr 3, was identified in the male F2 450-day combined CIM analysis at 55.21 cM (LOD 6.05). Salpa4 accounted for ~9.43% of the variance and approached the suggestive significance level in the F2 450-day male IM analysis (LOD 2.76 at 53.21 cM; Table 2). One of the suggestive QTLs was identified proximal to Salpa4 in the F2 150-day IM and CIM male analyses at 25.81 cM (LOD 3.16) and 15.81 cM (LOD 3.01). The other suggestive QTL on Chr 3 was identified in the F2 450-day male CIM analysis at 73.61 cM (LOD 3.81; Fig. 1 and Table 2).

Candidate Gene Analyses
Salpa3.
The Salpa3 region on Chr 4 was found to encompass the Akp2 gene (http://www.informatics.jax.org/). After generation of genotypes at the Akp2 Pstl polymorphism, it was possible to perform further analyses. First, the Akp2 genotypes were included in a new QTL analysis. Salpa3 did not shift closer to the Akp2 site, but the shoulder of the QTL that extends over Akp2 increased (Fig. 2B). Second, ANOVA was utilized to test whether AP activity differs by genotype at the Akp2 site for the F2 150-day age group. D2 homozygous mice had significantly higher mean serum AP activity levels than the B6 homozygotes, with the heterozygotes falling in between (D2, 43.90 ± 1.90; B6, 27.38 ± 1.18; and D2/B6, 33.16 ± 0.95 U/l; F ratio = 32.5712, P value < 0.0001; Fig. 3). A Tukey’s pairwise comparison determined that the heterozygotes (D2/B6) were significantly different from both homozygotes (D2s and B6s), indicating that the Akp2 gene follows an additive model of inheritance for the F2 150-day age group (data not shown).



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 3. Mean AP levels for males (gray) and females (black) grouped by Akp2 genotype. AP levels indicate a sexual dimorphism. Average values for the combined sexes by genotype are as follows: B6, 27.38 ± 1.18; heterozygous, 33.21 ± 0.95; and D2, 44.06 ± 1.92 U/l.

 
A difference between the sexes was also present. Females for each genotype at the Akp2 locus had significantly higher serum AP activity levels than males with corresponding genotypes (t = 6.17, B6; t = 8.354, heterozygote; and t = 4.795, D2; P value < 0.001; Fig. 3). The highest male levels were similar to the lowest female levels. All the analyses were performed on both sexes separately as well as a combined group corrected for sex differences. The highly significant QTL on Chr 4 was identified in the separate male and female analyses as well as the combined analysis. All three analyses account for similar variance (R2).

Additionally, statistical associations using the F2 150- and 450-day combined sexes data were evaluated with information from the markers on either side of the Akp2 gene using ANOVA. This model tests whether assigning the mice to specific groups (i.e., based on genotypic information) is more informative than randomly assigning the mice to arbitrary groups. The data used in this analysis were grouped by the total number of D2 alleles at the two flanking markers (0–4), without regard to the marker source of the D2 allele, and showed a highly significant difference between the mean AP level for the mice assigned to these groups (F ratio = 15.177, P value < 0.0001; Fig. 4). There appears to be an additive effect, and the marker information is predictive of serum AP levels, meaning the more D2 alleles, the higher the serum AP activity levels.



View larger version (18K):
[in this window]
[in a new window]
 
Fig. 4. Histogram of AP means compared with the no. of D2 alleles for markers D4Mit204 and D4Mit190 that flank the Akp2 gene. Vertical axis is mean AP serum level of groups, and horizontal axis indicates the no. of D2 alleles for each age group. Mean AP levels (U/l) for each group are shown at bottom.

 
Interestingly, there were different relationships between serum AP activity levels and allele status for the markers flanking the Akp2 gene when analyzed separately. The D2 alleles on the centromeric side of the Akp2 gene (D4Mit204) had a greater effect than those telomeric (D4Mit190; data not shown). This result was unexpected, because the markers are roughly equidistant from the Akp2 gene. Due to this observation and the fact that the major LOD peak for Salpa3 did not shift when the Akp2 genotypes were added to the QTL analysis (Fig. 2B), a stepwise linear regression analysis was performed to determine whether genotypes at the Akp2 site added predictive value to the existing model (D4Mit204 and D4Mit190). In the analysis, the mice homozygous for D2 at one marker were given a value of one to determine effects of the increasing allele; all other genotypes were set to zero for the first variable. For the second variable, the heterozygous genotype was given a value of one to determine dominant effects. The regression was performed using SAS statistical software, and the genotypes at the Akp2 locus were added as the only predictor. Neither of the flanking markers, D4Mit204 or D4Mit190, added any statistically significant predictive value to the model (P value = 0.092 for D4Mit204 and 0.940 for D4Mit190).

Sequencing of Akp2 mRNA confirmed a polymorphism between the two strains that causes an amino acid substitution (data not shown). The B6 allele encodes a leucine at amino acid 324, while the D2 allele encodes a proline. A diagram of the Akp2 genomic structure highlighting the regions analyzed is shown in Fig. 5A. The mouse Akp2 sequence was also compared with the human sequence, which corresponds to the D2 allele at position 324 with the proline conserved (Fig. 5B).



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 5. A: schematic of the Akp2 gene. The Pst1 site (1) in the D2 allele is altered by a polymorphism in the B6 allele, CT[C/G]CAG, and was used to genotype the mice. The intact PCR product is 236 bp, and the cut splits it into 167- and 69-bp fragments. In exon X, aa324, there is a leucine (L)-to-proline (P) amino acid substitution (2) between the B6 and D2 alleles, respectively. The final exon, XII, was used to create the radioactive probe (3) of 842 bp. Italics indicate the PCR primers. B: protein alignment of mouse and human amino acid sequence including the L-to-P substitution. D2 and human alleles both have a P in this position.

 
Salpa4.
Salpa4 (Chr 3), identified in the F2 male 450-day IM and CIM analyses, had a 15.6-cM region of interest between 45.2 and 60.8 cM. This QTL accounted for between 3.43 and 9.43% of the variance, with the male-only analysis having the largest variance (Table 2). This region contains 80 genes identified by the Mouse Genome Informatics (MGI) web site (http://www.informatics.jax.org/). The suggestive QTL identified in the F2 male 150-day IM and CIM analyses on Chr 3 overlapped with part of the Salpa4 region between 27.8 and 43.5 cM. This overlapping region of 15.7 cM includes 65 genes listed in MGI database, and candidate genes within this QTL region are listed in Table 3.


View this table:
[in this window]
[in a new window]
 
Table 3. Candidate gene analysis

 
Suggestive QTL
The suggestive QTL identified in the F2 450-day male CIM analysis on Chr 3 had a 16.0-cM region of interest (1-LOD drop-off from peak) between 67.6 and 83.6 cM and accounted for between 4.64 and 9.67% of the variance. This region contained 26 genes identified by the MGI database. The suggestive QTL on Chr 12 identified in the F2 450-day female CIM analysis had a 21.0-cM region of interest between 21.0 and 42.0 cM, containing 70 genes listed in the MGI database, and accounted for 7.02% of the variance. A suggestive QTL on Chr 18 was identified in the F2 450-day combined and female CIM analyses, with a peak LOD score of 3.8 at 47.0 cM. The QTL region was between 39.0 and 55.0 cM where 27 genes were identified in the MGI database. Refer to Table 3 for a list of candidate genes for these three suggestive QTLs.

mRNA Quantification
RNA from liver was examined by RT-PCR and Northern blots hybridized to a cDNA probe generated from the Akp2 gene. The RNA was extracted from liver because tissue samples from these animals were available from necropsy. In adult humans, it has been shown that the liver isozyme accounts for roughly one-half of the total AP activity in the serum (10, 36), although a similar analysis has not been reported for the mouse. An ANOVA of the RT-PCR results demonstrated that there were no significant strain (F = 0.02, P value = 0.901) or sex (F = 0.30, P value = 0.749) differences in the 18S (ribosomal RNA) control; therefore, analysis of the untransformed gene count was performed (Fig. 6A). The gene count ANOVA indicated no significant strain (F = 0.32, P value = 0.580) or sex (F = 0.05, P value = 0.951) differences. Analysis of the Northern blots and resulting autoradiographs showed that there was only one mRNA species detectable and confirmed the RT-PCR results (Fig. 6B).



View larger version (73K):
[in this window]
[in a new window]
 
Fig. 6. Quantification of mRNA expression levels. A: histogram of RT-PCR results. B: photograph of an ethidium bromide-stained filter that shows RNA loading levels (left) and audioradiogram of a P32 Akp2 probe-hybridized filter (right). Lanes: 1, BB female; 2, BD male; 3, BB male; 4, BD male.

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
This study of a B6D2 intercross at 150 and 450 days of age, supplemented by data from 150-day-old BXD RIs, revealed several significant QTLs. The major QTL, Salpa3, which accounted for up to 30% of variation in serum AP activity, contained the Akp2 gene. An SNP in the Akp2 gene promoter was highly associated with serum AP activity, although RT-PCR indicated no significant differences in steady-state RNA expression levels of the progenitor alleles (Fig. 6A). This result indicates that the D2 allele was not expressed at greater levels than the B6 allele, and, therefore, it is unlikely that expression differences account for the increased enzyme activity levels associated with the D2 allele. Northern blot analysis also indicated steady-state RNA expression levels among the F2 genotypes and a single gene transcript; therefore, an alternative splicing event is also unlikely to be responsible (Fig. 6B).

The possibility exists that transcriptional differences in AP are present in bone, the isoform of which is responsible for roughly one-half of the serum activity in postadolescent animals (10). Although transcriptional differences in regulation of Akp2 may occur in tissues other than liver, this is most likely not the case for the bone isoform because it shares a promoter with the liver isoform (27). Therefore, transcriptional differences of the Akp2 gene between D2 and B6 strains are not likely to be responsible for Salpa3. Because the promoter SNP is closely linked to other SNPs in the Akp2 gene, other mechanisms could be responsible for differences in serum AP activity.

Several SNPs are present that distinguish B6 and D2 mice and result in amino acid substitutions. One important potentially functional amino acid substitution in the Akp2 gene is a leucine (L) in the B6 strain to a proline (P) in the D2 strain substitution at position 324 (L324P; GenBank sequence GI-40787774). Sequencing of this site confirmed the polymorphism indicated in the databases. A basic local alignment search tool (BLAST) comparison of the mouse Akp2 gene and human TNAP gene shows 58% sequence identity overall, and shows the area of the polymorphism L324P to be in a region that is highly conserved between the two sequences, with a proline appearing in the human sequence (http://www.ncbi.nlm.nih.gov/BLAST/).

The crystal structure of the human placental AP isozyme (13) provides a model to predict the potential effect of the L324P mutation in the liver/bone/kidney isozyme. The mouse polymorphic site aligns with the human P303 position. This amino acid appears to be in a hydrogen-bonded turn that comes between an alpha helix and an extended beta strand in the enzyme’s calcium-binding region (13). The conversion of the proline (human and D2 allele) to a leucine (B6 allele) would most likely disrupt the turn. A disruption of this nature would result in conformational changes that could cause a decrease in the enzyme activity. Lower activity levels are observed in the B6 strain, which is consistent with a less active enzyme given that similar mRNA levels were shown. The proline in this position is also conserved in three other human AP genes (13) as well as BALB/C Akp2 allele (GI-11692622). (For candidate gene analysis information on other QTLs, see Table 3.)

Additional experiments are necessary to determine the precise cause of the enzyme activity differences in the two mouse strains. When the Akp2 genotype information is included in the QTL analysis, the peak over this site is not redefined, as would be expected if the activity were due to a single gene effect. In the stepwise regression analysis, addition of the Akp2 genotypes into the model indicated that they were the most useful for prediction of enzyme activity. These results indicate that it is likely Salpa3 is, in fact, two tightly linked QTLs with another effect gene linked to Akp2. Generating a congenic mouse by breeding a D2 or B6 Akp2 gene onto the opposing strain background would positively identify Akp2 as a candidate gene for the quantitative trait (QT gene) and potentially clarify any epigenetic effects that may be occurring. To identify directly the polymorphism that accounts for the QTL effect, the PstI, P324L, and other polymorphisms around Akp2 would require analysis in separate lines of mice.

A recent publication mapped QTLs for serum AP activity to Chr 2, -6, and -14, but not Chr 4, in the MRL/MpJ and SJL/J mouse strains (26). Our analysis also reported regions of interest on Chr 2 and -14, but they did not reach the suggestive level. Srivastava et al. (26) did not report 1-LOD intervals, but the LOD intervals reported here for the QTLs on these chromosomes (Table 2) do not include the region of Srivastava’s peak centimorgan positions (26). Only female mice were used by Srivastava et al., and, interestingly, the QTLs on Chr 2 and -14 in our analyses were present in three female-specific and one combined analysis but were not present in any of the male-specific analyses. The fact that Srivastava et al. did not identify the highly significant peak on Chr 4 emphasizes the need for studies in multiple strains to broaden the scope of our knowledge of allelic variation (6a). It is possible that the variation between B6 and D2 mice at the Akp2 locus does not exist between MRL/MpJ and SJL/J strains.

AP serum levels, as well as being an indication of disease, increase normally with age. The gene(s) responsible for this increase could be used as a biomarker for aging, and genetic information about such genes might lead to a better understanding of the biological processes of aging. Chronological age has been shown to be an inadequate predictor of health state, quality of later life, and longevity. Thus it is important to pursue alternative indicators of age (16). Our results confirm that AP enzyme activity increases with age, and that the QTL investigated on Chr 4 is responsible for a significant part of the variation seen in activity of this enzyme. These findings support the hypothesis that the Akp2 gene is a useful biomarker of aging.


    GRANTS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
This work was supported by National Institute on Aging (NIA) Research Grant P01-AG-14731 and National Research Service Award Training Grant T32-AG-000276 (also from NIA).


    ACKNOWLEDGMENTS
 
We thank Kim Seymour, Susan Lingenfelter, and David Bienusthe for collection of the animal data.


    FOOTNOTES
 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: D. J. Vandenbergh, Center for Developmental and Health Genetics, The Pennsylvania State Univ., 101 Amy Gardner House, Univ. Park, PA 16802-2317 (e-mail: djv4{at}psu.edu).


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Bahary N, Zorich G, Pachter JE, Leibel RL, and Friedman JM. Molecular genetic linkage maps of mouse chromosomes 4 and 6. Genomics 11: 33–47, 1991.[CrossRef][ISI][Medline]
  2. Beckers T, Schmidt P, and Hilgard P. Highly sensitive northern hybridization of rare mRNA using a positively charged nylon membrane. Biotechniques 16: 1074–1078, 1994.[ISI][Medline]
  3. Bhattacharyya M, Ray S, Bhattacharya S, and Chakrabarti A. Chaperone activity and prodan binding at the self-associating domain of erythroid spectrin. J Biol Chem 279: 55080–55088, 2004.[Abstract/Free Full Text]
  4. Brigden ML and Heathcote JC. Problems in interpreting laboratory tests. What do unexpected results mean? Postgrad Med 107: 145–146, 151–142, 155–148 passim, 2000.[CrossRef][Medline]
  5. Clancy BM, Johnson JD, Lambert AJ, Rezvankhah S, Wong A, Resmini C, Feldman JL, Leppanen S, and Pittman DD. A gene expression profile for endochondral bone formation: oligonucleotide microarrays establish novel connections between known genes and BMP-2-induced bone formation in mouse quadriceps. Bone 33: 46–63, 2003.[CrossRef][ISI][Medline]
  6. Clement-Jones M, Schiller S, Rao E, Blaschke RJ, Zuniga A, Zeller R, Robson SC, Binder G, Glass I, Strachan T, Lindsay S, and Rappold GA. The short stature homeobox gene SHOX is involved in skeletal abnormalities in Turner syndrome. Hum Mol Genet 9: 695–702, 2000.[Abstract/Free Full Text]
  7. Complex Trait Consortium. The Collaborative Cross, a community resource for the genetic analysis of complex traits. Nat Genet 36: 1133–1137, 2004.[CrossRef][ISI][Medline]
  8. Davies R, Schuurman A, Barker CR, Clothier B, Chernova T, Higginson FM, Judah DJ, Dinsdale D, Edwards RE, Greaves P, Gant TW, and Smith AG. Hepatic gene expression in protoporphyic Fech mice is associated with cholestatic injury but not a marked depletion of the heme regulatory pool. Am J Pathol 166: 1041–1053, 2005.[Abstract/Free Full Text]
  9. Hahm K, Sum EY, Fujiwara Y, Lindeman GJ, Visvader JE, and Orkin SH. Defective neural tube closure and anteroposterior patterning in mice lacking the LIM protein LMO4 or its interacting partner Deaf-1. Mol Cell Biol 24: 2074–2082, 2004.[Abstract/Free Full Text]
  10. Hentunen TA, Choi SJ, Boyce BF, Dallas MR, Dallas SL, Shen-Ong GL, and Roodman GD. A murine model of inflammatory bone disease. Bone 26: 183–188, 2000.[CrossRef][ISI][Medline]
  11. Hitz J, Daigle G, Petitclerc C, Schiele F, and Siest G. Automated quantification of bone and liver alkaline phosphatase isoenzymes of human serum. Clin Chim Acta 107: 203–210, 1980.[CrossRef][ISI][Medline]
  12. Jodar Gimeno E, Munoz-Torres M, Escobar-Jimenez F, Quesada Charneco M, Luna del Castillo JD, and Olea N. Identification of metabolic bone disease in patients with endogenous hyperthyroidism: role of biological markers of bone turnover. Calcif Tissue Int 61: 370–376, 1997.[CrossRef][ISI][Medline]
  13. Lander E and Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 11: 241–247, 1995.[CrossRef][ISI][Medline]
  14. Le Du MH, Stigbrand T, Taussig MJ, Menez A, and Stura EA. Crystal structure of alkaline phosphatase from human placenta at 1.8 A resolution. Implication for a substrate specificity. J Biol Chem 276: 9158–9165, 2001.[Abstract/Free Full Text]
  15. Limdi JK and Hyde GM. Evaluation of abnormal liver function tests. Postgrad Med J 79: 307–312, 2003.[Abstract/Free Full Text]
  16. Lionikas A, Blizard DA, Vandenbergh DJ, Glover MG, Stout JT, Vogler GP, McClearn GE, and Larsson L. Genetic architecture of fast- and slow-twitch skeletal muscle weight in 200-day-old mice of the C57BL/6J and DBA/2J lineage. Physiol Genomics 16: 141–152, 2003.[Abstract/Free Full Text]
  17. McClearn GE. Biomarkers of age and aging. Exp Gerontol 32: 87–94, 1997.[CrossRef][ISI][Medline]
  18. McComb RB, Bowers GN, and Posen S. Alkaline Phosphatase. New York: Plenum, 1979.
  19. McPherson K, Healy MJ, Flynn FV, Piper KA, and Garcia-Webb P. The effect of age, sex and other factors on blood chemistry in health. Clin Chim Acta 84: 373–397, 1978.[CrossRef][ISI][Medline]
  20. Munns CF, Glass IA, Flanagan S, Hayes M, Williams B, Berry M, Vickers D, O’Rourke P, Rao E, Rappold GA, Hyland VJ, and Batch JA. Familial growth and skeletal features associated with SHOX haploinsufficiency. J Pediatr Endocrinol Metab 16: 987–996, 2003.[ISI][Medline]
  21. Murphy JE and Kantrowitz ER. Why are mammalian alkaline phosphatases much more active than bacterial alkaline phosphatases? Mol Microbiol 12: 351–357, 1994.[ISI][Medline]
  22. Narisawa S, Frohlander N, and Millan JL. Inactivation of two mouse alkaline phosphatase genes and establishment of a model of infantile hypophosphatasia. Dev Dyn 208: 432–446, 1997.[CrossRef][ISI][Medline]
  23. Rahman S, Oberdorf A, Montecino M, Tanhauser SM, Lian JB, Stein GS, Laipis PJ, and Stein JL. Multiple copies of the bone-specific osteocalcin gene in mouse and rat. Endocrinology 133: 3050–3053, 1993.[Abstract]
  24. Raymond MH, Schutte BC, Torner JC, Burns TL, and Willing MC. Osteocalcin: genetic and physical mapping of the human gene BGLAP and its potential role in postmenopausal osteoporosis. Genomics 60: 210–217, 1999.[CrossRef][ISI][Medline]
  25. Sabuncu T, Aksoy N, Arikan E, Ugur B, Tasan E, and Hatemi H. Early changes in parameters of bone and mineral metabolism during therapy for hyper- and hypothyroidism. Endocr Res 27: 203–213, 2001.[CrossRef][ISI][Medline]
  26. Srivastava AK, Masinde G, Yu H, Baylink DJ, and Mohan S. Mapping quantitative trait loci that influence blood levels of alkaline phosphatase in MRL/MpJ and SJL/J mice. Bone 35: 1086–1094, 2004.[CrossRef][ISI][Medline]
  27. Studer M, Terao M, Gianni M, and Garattini E. Characterization of a second promoter for the mouse liver/bone/kidney-type alkaline phosphatase gene: cell and tissue specific expression. Biochem Biophys Res Commun 179: 1352–1360, 1991.[CrossRef][ISI][Medline]
  28. Swallow DM, Povey S, Parkar M, Andrews PW, Harris H, Pym B, and Goodfellow P. Mapping of the gene coding for the human liver/bone/kidney isozyme of alkaline phosphatase to chromosome 1. Ann Hum Genet 50: 229–235, 1986.[ISI][Medline]
  29. Terao M and Mintz B. Cloning and characterization of a cDNA coding for mouse placental alkaline phosphatase. Proc Natl Acad Sci USA 84: 7051–7055, 1987.[Abstract/Free Full Text]
  30. Terao M, Pravtcheva D, Ruddle FH, and Mintz B. Mapping of gene encoding mouse placental alkaline phosphatase to chromosome 4. Somat Cell Mol Genet 14: 211–215, 1988.[CrossRef][ISI][Medline]
  31. Tsang SS, Yin X, Guzzo-Arkuran C, Jones VS, and Davison AJ. Loss of resolution in gel electrophoresis of RNA: a problem associated with the presence of formaldehyde gradients. Biotechniques 14: 380–381, 1993.[ISI][Medline]
  32. Vandenbergh DJ, Heron K, Peterson R, Shpargel KB, Woodroffe A, Blizard DA, McClearn GE, and Vogler GP. Simple tests to detect errors in high-throughput genotype data in the molecular laboratory. J Biomol Tech 14: 9–16, 2003.[Abstract/Free Full Text]
  33. Wang S, Basten CJ, and Zeng ZB. Windows QTL Cartographer 2.0. Raleigh, NC: Dept Statistics, NC State Univ, 2001–2004.
  34. Weintraub BD and Szkudlinski MW. Development and in vitro characterization of human recombinant thyrotropin. Thyroid 9: 447–450, 1999.[ISI][Medline]
  35. Weiss MJ, Spielman RS, and Harris H. A high-frequency RFLP at the human liver/bone/kidney-type alkaline phosphatase locus. Nucleic Acids Res 15: 860, 1987.[ISI][Medline]
  36. Wilcox FH and Taylor BA. Genetics of the Akp-2 locus for alkaline phosphatase of liver, kidney, bone, and placenta in the mouse. Linkage with the Ahd-1 locus on chromosome 4. J Hered 72: 387–390, 1981.[ISI][Medline]
  37. Wiwanitkit V. High serum alkaline phosphatase levels, a study in 181 Thai adult hospitalized patients. BMC Fam Pract 2: 2, 2001.[CrossRef][Medline]
  38. Zeng ZB. Precision mapping of quantitative trait loci. Genetics 136: 1457–1468, 1994.[Abstract/Free Full Text]
  39. Zeng ZB. Theoretical basis for separation of multiple linked gene effects in mapping quantitative trait loci. Proc Natl Acad Sci USA 90: 10972–10976, 1993.[Abstract/Free Full Text]




This Article
Abstract
Full Text (PDF)
All Versions of this Article:
23/3/295    most recent
00062.2005v1
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Foreman, J. E.
Articles by Vandenbergh, D. J.
PubMed
PubMed Citation
Articles by Foreman, J. E.
Articles by Vandenbergh, D. J.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2005 by the American Physiological Society.