Physiol. Genomics AJP: Regulatory, Integrative and Comparative Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Physiol. Genomics 35: 296-304, 2008. First published September 16, 2008; doi:10.1152/physiolgenomics.90260.2008
1094-8341/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
35/3/296    most recent
90260.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Korstanje, R.
Right arrow Articles by Millonig, J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Korstanje, R.
Right arrow Articles by Millonig, J. H.
Received 30 May 2008; accepted in final form 28 August 2008.
Physiological Genomics 35:296-304 (2008)
1094-8341/08 $8.00 © 2008 American Physiological Society

Quantitative trait loci affecting phenotypic variation in the vacuolated lens mouse mutant, a multigenic mouse model of neural tube defects

Ron Korstanje 1,*, Jigar Desai 3,*, Gloria Lazar 3, Benjamin King 1, Jarod Rollins 1, Melissa Spurr 1, Jamie Joseph 3, Sindhuja Kadambi 3, Yang Li 2, Allison Cherry 1, Paul G. Matteson 3, Beverly Paigen 1 and James H. Millonig 3,4,5

1 Jackson Laboratory, Bar Harbor, Maine
2 Groningen Bioinformatics Centre, Groningen Biomolecular Sciences and Biotechnology Institute, Haren, The Netherlands
3 Center for Advanced Biotechnology and Medicine, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey
4 Department of Neuroscience and Cell Biology, UMDNJ-Robert Wood Johnson Medical School, Piscataway, New Jersey
5 Department of Genetics, Rutgers University, Piscataway, New Jersey


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
The vacuolated lens (vl) mouse mutant arose spontaneously on the C3H/HeSn background and exhibits neural tube defects (NTDs), congenital cataract, and occasionally a white belly spot. We previously reported that 1) the vl phenotypes are due to a mutation in an orphan G protein-coupled receptor (GPCR), Gpr161; 2) the penetrance of the vl NTD and cataract phenotypes are affected by genetic background, allowing three unlinked quantitative trait loci (QTL) to be mapped (modifiers of vacuolated lens, Modvl1-3); and 3) phenotype-based bioinformatics followed by genetic and molecular analysis identified a lens-specific transcription factor that contributes to the cataract-modifying effect of Modvl3. We now extend this analysis in three ways. First, using the Gpr161 mutation to unequivocally identify mutant adults and embryos, we determined that ~50% of vl/vl NTD-affected embryos die during development. Second, the MOLF/Ei genetic background suppresses this embryonic lethality but increases the incidence of the adult belly spot phenotype. Additional QTL analysis was performed, and two novel modifiers were mapped [Modvl4, logarithm of odds ratio (LOD) 4.4; Modvl5, LOD 5.0]. Third, phenotype-based bioinformatics identified candidate genes for these modifiers including two GPCRs that cause NTD or skin/pigmentation defects (Modvl4: Frizzled homolog 6; Modvl5: Melanocortin 5 receptor). Because GPCRs form oligomeric complexes, these genes were resequenced and nonsynonymous coding variants were identified. Bioinformatics and protein modeling suggest that these variants may be functional. Our studies further establish vl as a multigenic mouse model for NTDs and identify additional QTL that interact with Gpr161 to regulate neurulation.

quantitative trait locus analysis; Gpr161; Frizzled homolog 6; Melanocortin 5 receptor


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
NEURAL TUBE DEFECTS (NTDs) are the second-most common human congenital disorder, with an incidence of ~1/1,000 in American Caucasian live births. Increased risk for NTDs is due to both genetic susceptibility and nongenetic factors such as the environment (13). Although >250 mouse NTD mutants have been identified, few of them model the multigenic basis of the human disorder (12, 23).

The vacuolated lens (vl) mutation was identified initially by the congenital cataract phenotype observed in postnatal mice. A small percentage of these mutants also display a white belly spot (14). Two vl-associated NTD phenotypes have been reported. One group of embryos displayed lumbar-sacral spina bifida, while in the other subset the neural tube closed but an attenuated dorsal midline and a dilated dorsal ventricle were observed (44–47). These later phenotypes are remarkably similar to a closed human NTD called embryonic hydromyelia (22).

Neurulation requires bending of the neural plate medially and dorso-laterally, elevation of the neural folds, and then finally apposition and fusion of the neural folds on the dorsal midline (11). Previous histological, ultrastructural electron microscopy (EM), and in vitro culture studies indicate that the vl mutation affects the last step of neurulation, apposition, and fusion of the neural folds (11, 20, 44, 46–48). These studies led to the conclusion that the vl spina bifida phenotype is likely due to a complete failure of neural fold fusion, while the vl closed NTD phenotype results from abnormal or incomplete apposition and fusion.

We previously reported (28) that these vl phenotypes are due to a mutation in an orphan G protein-coupled receptor (GPCR) called Gpr161. Approximately 360 human nonsensory GPCRs exist, with the ligand identified for ~200 receptors and the remaining 160 receptors called orphan GPCRs because their endogenous ligands are not known (16). Our previous in situ analysis demonstrated that Gpr161 is expressed in the lateral neural folds during neural tube closure and at all stages of lens development, consistent with the vl NTD and cataract phenotypes. The vl mutation is an 8-bp deletion, which causes a frameshift and premature stop codon truncating ~70% of the cytoplasmic COOH-terminal tail (143/203 amino acids) (28). We have demonstrated that mutant protein is expressed on the plasma membrane but the COOH-terminal tail truncation results in reduced receptor-mediated endocytosis. It is well established that attenuation of GPCR signaling is accomplished through receptor-mediated endocytosis, which is commonly initiated through COOH-terminal tail phosphorylation (21, 25, 34, 35, 42). These previous studies identified Gpr161 as a new receptor-ligand mediated pathway necessary for neural tube closure and lens development.

To map the vl locus and clone Gpr161, the mutation was crossed onto three different genetic backgrounds (C57BL/6J, CAST/Ei, and MOLF/Ei). We previously reported (28) that the penetrance of the vl NTD and cataract phenotypes was significantly affected in these crosses, enabling us to map three quantitative trait loci (QTL) (modifiers of vacuolated lens, Modvl1-3). In these B6 and CAST crosses, ~50% of adult F2 vl/vl mice display an obvious lumbar-sacral lesion and hindlimb paralysis (28), phenocopying important aspects of human spina bifida that have not been replicated in other mouse mutants (11, 23, 41). The modifying effects of these backgrounds on the vl NTD phenotype were mapped to chromosomes 5 (Modvl1C57BL/6J) and 1 (Modvl2CAST/Ei). Modvl3 affects the incidence of the cataract phenotype in the MOLF/Ei cross. Previous phenotype-based bioinformatics analysis identified Foxe3, a winged helix forkhead transcription factor, as a candidate for Modvl3. Foxe3 is an important regulator of lens development, and mutations in the gene result in cataract in both mice and humans (1, 29, 33, 38). Our resequencing of Foxe3 identified a nonsynonymous coding variant between C3H and MOLF. Additional bioinformatic, genetic, and molecular data demonstrated that this coding variant is functional and likely contributes to the Modvl3 cataract phenotype (28, 29).

We now extend these previous results by 1) identifying additional modifying effects for vl NTD-associated phenotypes (embryonic lethality, adult belly spot) on the three mixed backgrounds (B6/C3H, CAST/C3H, and MOLF/C3H), 2) performing further QTL analysis and mapping two novel modifiers (Modvl4 and 5), and 3) employing bioinformatics to identify candidate genes responsible for these modifying effects. These studies further establish vl as a mouse model for studying the multigenic basis of NTDs.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Animals and Crosses
Cryopreserved vacuolated lens mice were rederived from the Jackson Laboratory. To determine the penetrance of the different adult and embryonic phenotypes, +/vl x +/vl matings were performed and when necessary pregnant females were killed and litters were genotyped and phenotyped. To map the vl locus and to generate F2 vl/vl mice for QTL analysis, vl/vl C3H mice were crossed to three different inbred strains (C57BL/6J, CAST/EiJ, and MOLF/EiJ), which were obtained from the Jackson Laboratory. For the C57BL/6J and CAST/EiJ crosses wild-type females were mated to C3H/HeSnJ-vl/vl males and for the MOLF/EiJ cross wild-type males were mated to C3H/HeSnJ-vl/vl females to produce F1 progeny. The F1 mice were then intercrossed to produce F2 progeny. Because the vl mutation arose on the C3H background, vl/vl were identified by the C3H/C3H genotype of simple sequence-length polymorphism (SSLP) markers flanking the vl locus (B6: D1Mit143, D1Mit15; CAST and MOLF: D1Mit506, D1Mit15) (31). For the B6 and CAST crosses, 86 and 94 F2 vl/vl mice were typed for genome scan markers, respectively, and then analyzed for QTL. For the MOLF cross, 92 F2 vl/vl mice were typed for genome scan markers and analyzed for QTL. To refine the identified QTL intervals, 35 F2 C3H/MOLF mice were added and flanking SSLP markers to selected QTL regions were genotyped.

Mice were housed in a climate-controlled facility with a 14:10-h light-dark cycle and free access to food and water throughout the experiment. After weaning, mice were maintained on a chow diet (Old Guilford 234A, Guilford, CT). All experiments were approved by the Animal Care and Use Committees of the Jackson Laboratory and UMDNJ-Robert Wood Johnson Medical School.

DNA Isolation and Genotyping
Tail DNA was isolated and genotyped as described previously (28). The B6, CAST, and MOLF crosses used 75, 60, and 80 SSLP markers distributed across the genome, respectively. To genotype for the vl mutation, standard PCR conditions at 55°C annealing temperature were used with primers that flank the 8-bp deletion (F: 5'CCGTTCTACCAATGCCAACTTTG3'; R: 5'GTGAGGGGTTTTCAGGGTTTTTAC3'; 168-bp amplicon +/+, 160-bp amplicon vl/vl).

Statistical Analysis
We performed genomewide scans for QTL by using the method of Sen and Churchill (39), which is similar to the interval mapping procedure of Lander and Botstein (27) but uses a different imputation algorithm. First, we carried out one-dimensional genome scans on a single-QTL basis to detect QTL with main effects. Logarithm of odds ratio (LOD) scores were computed at 2-cM intervals across the genome, and significance was determined by permutation testing (9). Significant QTL meet or exceed the 95% genomewide thresholds, respectively. Simultaneous genome scans for all pairs of markers were then implemented to detect epistatic interactions. The search strategy has been described previously (39, 40). Briefly, the genome scan searches through all pairs of loci by fitting a two-way ANOVA model with an interaction item. A LOD score contrasting the full model to a null model (with no genomic effects) is computed, and genomewide significance is established by permutation analysis. A secondary test for the significance of the interaction term is computed only for those pairs that pass the genomewide screening. A stringent nominal significance level (0.001) is used for the interaction test, and only those locus pairs passing both tests are deemed to be interacting. The software package used in this study, R/QTL version 0.97-21, is available at http://www.biostat.jhsph.edu/~kbroman/qtl/ (4).

To test whether loci segregated with the lethality in F2 progeny, we analyzed all markers by {chi}2-test (single-marker search) and contingency table testing (marker pair search). A {chi}2-analysis of the data was performed to determine whether the segregation of alleles differed from the expected for normal Mendelian segregation. {chi}2-Tests were performed on each marker separately except for markers on distal chromosome 1 flanking the vl locus. The obtained P values determined whether segregation distortion was observed, which would suggest the existence of protective loci. To investigate whether possible interacting marker pairs demonstrated distorted segregation, contingency table testing was applied to the genotyping data. Each possible pair of markers was tested, except for neighboring linked marker pairs.

Bioinformatic and Resequencing Analysis
To identify candidate genes that may contribute to the modifying effects of the mapped QTL, we searched www.informatics.jax.org/phenotypes.shtml for genes within the 95% confidence intervals (CI) that caused similar phenotypes. Validated single nucleotide polymorphisms (SNPs) between C3H and MOLF for candidate genes were then identified by mining dbSNP (www.ncbi.nlm.nih.gov/genome/guide/mouse/). Noncoding SNPs were mapped relative to potential cis-regulatory sequences (ESPERR Regulatory Potential; genome.ucsc.edu), and potential effects on transcription factor binding were investigated by using Transcription Element Search Software (TESS; www.cbil.upenn.edu/tess). To identify and validate additional coding polymorphisms in Fzd6 and Mc5r, C57BL/6J, C3H/HeSnJ, and MOLF/EiJ genomic DNA was obtained from the Jackson Laboratories (http://www.jax.org/dnares/index.html), and the coding regions plus splicing sites were amplified with custom-designed primers (Primer3; frodo.wi.mit.edu/). Direct sequencing was performed on the PCR products with Big Dye Terminator Cycle Sequencing Chemistry and the ABI 3700 Sequence Detection System (Applied Biosystems). Results were analyzed with Sequencher software (version 4.8). The potential functional effects of nonsynonymous coding SNPs were evaluated with PolyPhen (genetics.bwh.harvard.edu/pph/), which also generated protein alignments between orthologs and paralogs. NetPhos (www.cbs.dtu.dk/services/NetPhos/) at www.expasy.org/ determined the potential effect of coding SNPs on protein phosphorylation. Protein secondary structure analysis was performed with McVector v9.0 (Robson-Garnier algorithm).

Protein Modeling
Protein sequences for Mc5r in mouse, rat, human, cow, chicken, frog, zebrafish, and pufferfish were obtained from Ensembl release 48. These protein sequences were aligned with CLUSTALW2 (http://www.ebi.ac.uk/Tools/clustalw2/), and the alignment was viewed in JalView version 2.3 (http://www.jalview.org) (10). A molecular model of three-dimensional protein structure for mouse MC5R (SWISS-PROT protein sequence P41149) was obtained from the ModBase resource (http://modbase.compbio.ucsf.edu/modbase) that used the X-ray crystallographic structure of bovine rhodopsin (PDB code 1L9H) as a template (32). RasMol version 2.7.2.1 (http://www.rasmol.org) was used to visualize the structure (36). The potential functional effects of Mc5r rs8256628 were evaluated by PANTHER (http://www.pantherdb.org/tools/csnpScoreForm.jsp) and SIFT (http://blocks.fhcrc.org/sift/SIFT.html). The effect of the Tyr121 substitution on Mc5r ModBase structure was visualized by SwissPDBViewer version 3.7 (http://www.expasy.org/spdbv) (18).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Penetrance of vl Phenotypes
C3H isogenic background.
Before the positional cloning of the vl locus, published experiments were limited to a small number of matings in which adult mice with cataracts were either mated to each other or mated to unaffected littermates. Mutant adults were identified previously by their cataract phenotype, which also occasionally displayed a belly spot. For the published vl embryonic analysis, pregnant females were killed and embryos with an obvious NTD defect were examined experimentally (44, 46, 47). Because vl mutants were identified by their phenotype and not by the causative mutation, the penetrance of the cataract, belly spot, and NTD phenotypes has never been determined precisely. In addition, since open NTD phenotypes in mice typically result in embryonic lethality (11, 20, 23), some percentage of mutant embryos may die during gestation, but this has never been investigated.

To determine the penetrance of the vl NTD phenotypes and whether the vl mutant displays embryonic lethality on an isogenic C3H background, +/vl intercrosses were performed. Over 100 adult progeny were generated and genotyped for the vl mutation. Vl homozygotes were observed ~50% less than their +/+ littermates (+/+: 26/109, +/vl: 69/109, vl/vl: 14/109), indicating that half of the vl homozygotes die before weaning (Table 1). When embryonic day (E)9.5–E12.5 litters were killed and genotyped, all of the vl/vl progeny displayed neural tube phenotypes: 41% of vl/vl embryos displayed lumbar-sacral spina bifida (23/56), while all of the remaining vl/vl embryos displayed a posterior spinal cord phenotype similar to human embryonic hydromyelia (Table 1). For litters killed between E14.5 and E18.5 (n = 64 embryos), no spina bifida-affected vl/vl embryos were observed and an increase in resorptions was detected. These data indicate that the vl mutation results in embryonic lethality on the C3H background, which has not been previously recognized. This lethality is likely due to the embryonic spina bifida phenotype, since other open mouse NTD mutants typically die during embryogenesis or at birth (20, 23).


View this table:
[in this window]
[in a new window]

  		Table 1. Penetrances of vl phenotypes on C3H background

 
The penetrance of the adult cataract and belly spot phenotype was also determined. Through our intercrosses to maintain the vl colony over the last 6 years, >400 vl/vl adult mice have been generated and 100% display congenital cataracts, with <5% also exhibiting a small white belly spot (Table 1). The percentages of these vl phenotypes on the isogenic C3H background were then used to determine the effect of the three modifying backgrounds (MOLF/Ei, CAST/Ei, B6) in our previous crosses.

MOLF/Ei cross.
We previously reported (28) that the MOLF background significantly reduced the penetrance of the cataract phenotype (Table 2). Our results now indicate that the MOLF background has additional effects on the penetrance of the vl-associated embryonic lethality and adult belly spot phenotype (Tables 2–4).


View this table:
[in this window]
[in a new window]

  		Table 2. Summary of vl-modifying effects

 
If the MOLF background completely rescues the vl-associated lethality, we would expect 25% of the F2 C3H/MOLF progeny to genotype as vl/vl. In our cross 22.5% of the F2 progeny were vl homozygotes, which is not statistically different from the expected 25% ({chi}2; P = 0.09) (Table 3). These data indicate that the MOLF background can rescue the vl embryonic lethality phenotype (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 3. Modifying effect of genetic background on vl-associated lethality

 
The surviving F2 C3H/MOLF vl/vl mice were of two phenotypic classes: 1) 49.6% of the F2 vl/vl mice used for our QTL analysis (63/127) displayed no obvious postnatal phenotype and were indistinguishable from their +/+ littermates; these data demonstrate that the MOLF background is able to rescue all obvious vl postnatal phenotypes in a subset of vl/vl. 2) The remaining C3H/MOLF vl/vl F2 progeny (50.4%, 64/127) were phenotypically abnormal, exhibiting congenital cataract and/or a white belly spot. Although we previously noted (28) that the penetrance of the cataracts was reduced considerably (Table 2), we could now determine that that the MOLF background significantly exacerbated the incidence of the belly spot phenotype (40.9%, 52/127) (Tables 2 and 4).


View this table:
[in this window]
[in a new window]

  		Table 4. Modifying effect of genetic background on vl belly spot phenotype

 
Together these results indicate that ~50% of the C3H/MOLF vl/vl embryos are rescued for the embryonic lethality, which is likely due to the spina bifida. As adults these rescued vl/vl F2 C3H/MOLF mice are either phenotypically normal or display a white belly spot. Interestingly, a belly spot phenotype is often due to defects in early central nervous system (CNS) development, suggesting that the MOLF modifiers may rescue the spina bifida defect but subsequent spinal cord development is abnormal.

B6 and CAST/Ei crosses.
We reported previously (28) that the B6 and CAST backgrounds result in a significant proportion of vl/vl F2 adult mice displaying an open lumbar-sacral lesion and hindlimb paralysis (Table 2). Our results now indicate two additional effects of the B6 and CAST backgrounds on the vl embryonic lethality and belly spot phenotypes (Table 2).

If the vl lethality was fully penetrant, we would expect ~50% of the embryos to die. Thus only ~12.5% of the adult F2 progeny would genotype as vl/vl. The B6 and CAST backgrounds increased the incidence of the vl-associated lethality since the number of F2 vl/vl mice generated from these crosses was less than the expected 12.5% ({chi}2; B6: P = 0.015, CAST: P < 0.001) (Table 3). These data indicate that the B6 and CAST backgrounds have modifiers that exacerbate the vl embryonic lethality phenotype (Table 2).

For the belly spot phenotype, the B6 and CAST backgrounds had different effects. The B6 background increased the penetrance for the belly spot phenotype. The CAST background had no effect (Table 4). Thus the B6, CAST, and MOLF backgrounds differentially affect the penetrance of the various vl-associated phenotypes (Table 2).

QTL Analysis
Given the modifiability of the belly spot and lethality phenotypes by MOLF, CAST, and B6 backgrounds, we decided to use these traits to map the genomic location of additional modifying loci. Additional QTL analysis was performed on F2 vl/vl progeny from the three different intercrosses to detect both additive and interactive significant QTL (P < 0.05) in our three crosses (Table 5). No interacting QTL were identified, so only single QTL are presented.


View this table:
[in this window]
[in a new window]

  		Table 5. Summary of significant modifier loci for vl mutation

 
Both the MOLF and B6 backgrounds increased the penetrance of the belly spot phenotype. To map possible modifier loci, F2 B6/C3H and MOLF/C3H vl/vl progeny with a belly spot were given a numerical value of 1, while an absence of a belly spot was scored as 0. QTL analysis was then performed for these two crosses, and three QTL were identified. For the B6 cross, a QTL was mapped to chromosome 5 with a peak at 44 cM linked to D5Mit309 (LOD 3.7). The B6 alleles segregated with the presence of a belly spot in a recessive fashion (Table 5). Interestingly, the same QTL was mapped previously when the adult spina bifida phenotype was used as a trait (Modvl1, D5Mit309, 44 cM, LOD 3.7) but with the B6 allele segregating in a dominant fashion (28). These data suggest that the same locus influences both phenotypes. Modvl1 accounts for 18.7% of the F2 belly spot phenotypic variance.

When the belly spot phenotype was used as a trait in the MOLF cross, two additional QTL (Modvl4 and Modvl5) were identified. Modvl4 maps to chromosome 15, with a peak at 15 cM linked to D15Mit252 (LOD 4.4). Modvl5 maps to chromosome 18, with a peak at 41 cM linked to D18Mit50 (LOD 5.0). Modvl4 and Modvl5 account for 12.7% and 16.8% of the F2 phenotypic variance. For both Modvl4 and 5, the C3H allele segregates with the belly spot phenotype in an additive fashion, while the MOLF alleles for these QTL are inherited with phenotypically normal F2 vl/vl progeny (Table 5, Fig. 1).


Figure 1
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 1. MOLF/Ei belly spot quantitative trait locus (QTL) analysis. A: genomewide scan of the (C3H/HeSn-vl/vl x MOLF/Ei) F2 intercross for belly spot. Suggestive (P = 0.10) and significant (P = 0.05) thresholds are indicated. LOD, logarithm of odds ratio. B: LOD score plot of chromosome 15 with Modvl4. C: allele effect plot of the peak marker (D15Mit252). D: LOD score plot of chromosome 18 with Modvl5. E: allele effect plot of the peak marker (D18Mit50).

 
We also investigated whether any QTL could be identified for the increase in vl-associated lethality on the C3H/B6 and C3H/CAST backgrounds (Table 3); {chi}2 and contingency table testing were used to determine whether any marker alleles were underrepresented in F2 vl/vl mice, but no significant QTL were identified (data not shown).

Bioinformatic Analysis
We were previously successful (28) in using bioinformatics and resequencing to identify a functional SNP in Foxe3 that contributes to the cataract-modifying effects of Modvl3. A similar approach was then employed for Modvl1, 2, 4, and 5. The 95% CIs for these QTL were searched (www.informatics.jax.org) for genes that result in NTDs or skin/pigmentation defects when mutated. Fourteen genes were identified: three for Modvl1, four for Modvl2, three for Modvl4, and four for Modvl5 (Table 6).


View this table:
[in this window]
[in a new window]

  		Table 6. Candidate genes for vl modifiers

 
To prioritize candidates for further analysis, GPCRs were examined. It is well established that GPCRs form oligomeric complexes with each other (3, 30), so we hypothesized that functional polymorphisms affecting the activity or expression of different GPCRs could potentially bypass the mutant effects of the Gpr161 COOH-terminal tail truncation. The above phenotype-based bioinformatic analysis identified two GPCRs that cause NTD and skin/pigmentation defects: Frizzled homolog 6 (Fzd6) for Modvl4 and Melanocortin 5 receptor (Mc5r) for Modvl5. Our analysis identified no GPCRs for Modvl1 or 2. Both Fzd6 and Mc5r are expressed during early CNS development (genome.ucsc.edu). The Fzd6 knockout results in craniorachischisis, a open neural tube along the entire anterior-posterior axis, while both the Fzd6 and Mc5r mutants exhibit a hair or skin phenotype (7, 19, 43). These observations focused our resequencing and bioinformatic analysis on Fzd6 and Mc5r.

Fzd6.
To identify possible functional polymorphisms, available dbSNP data were first mined for validated polymorphisms between C3H and MOLF. None of the validated SNPs between these strains is in protein coding sequence, but four of them mapped in or near (<100 bp) predicted cis-regulatory sequences (ESPERR Regulatory Potential; genome.ucsc.edu) (24, 26). All of these SNPs are situated in binding sites for transcription factors expressed during neural tube closure or in the skin and may affect binding (Table 7).


View this table:
[in this window]
[in a new window]

  		Table 7. Bioinformatic analysis for Fzd6 and Mc5r noncoding SNPs

 
We then resequenced the Fzd6 protein-coding region from C3H and MOLF to either identify novel polymorphisms or determine whether previously annotated SNPs were polymorphic between these two strains. This analysis validated polymorphic differences for five SNPs and identified one new polymorphism (Table 8).


View this table:
[in this window]
[in a new window]

  		Table 8. Resequencing analysis for Fzd6 and Mc5r protein-coding polymorphisms

 
Two of these polymorphisms (rs50388052, rs48579948) result in a nonsynonymous coding changes, and one 3-bp deletion causes a deletion of an amino acid. All of these amino acids are situated in the COOH-terminal tail of Fzd6. We then investigated whether any of these polymorphisms are predicted to be functional. The bioinformatic analysis did not identify rs48579948 or the deleted amino acid as potentially functional, but significant results were obtained for rs50388052. This SNP replaces a polar SerC3H residue at position 618 with a nonpolar ProMOLF residue (S618P). Interestingly, the ProMOLF allele is conserved among multiple vertebrate species in Fzd6 and a related protein, Fzd3, while the Ser allele has only been reported for C3H and B6 mouse strains (Fig. 2; data not shown). The S618P change is predicted to be damaging by PolyPhen [position-specific independent counts (PSIC) score difference = 1.748] and to affect the secondary structure of the COOH-terminal tail (Fig. 2). S618P also removes a potential Ser phosphorylation site in the COOH-terminal tail (data not shown), which is often used to initiate receptor-mediated endocytosis (16). These data are consistent with the S618P alteration possibly affecting Fzd6 activity.


Figure 2
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 2. Fzd6C3H and MOLF allelic differences. A: the TC3H to CMOLF polymorphism for rs50388052 (*) is shown with flanking sequence. B: the TC3H to CMOLF substitution replaces a serine (SC3H) with a proline (PMOLF) at amino acid 618 in the COOH-terminal tail terminus of the G protein-coupled receptor (GPCR). This amino acid change plus flanking amino acids (613–623 in mouse) is shown for C3H, MOLF, and other vertebrate species. The P618 allele (boxed) is evolutionarily conserved. C: the L618 to P618 change is predicted to alter Fzd6 secondary structure, with the P618 allele truncating a β-sheet and a turn but extending an {alpha}-helix. Predictions for the Robson-Garnier algorithm are shown for a 20-amino acid region.

 
Mc5r.
Data mining and resequencing identified or validated five noncoding polymorphisms in potential cis-regulatory sequences and six protein-coding SNPs. Three of the noncoding SNPs are predicted to affect transcription factor binding (Table 7). Four of the coding SNPs are nonsynonymous (Table 8), but only rs8256628 is predicted to be potentially deleterious (PSIC score difference = 1.876). Rs8256628 is situated at the start of the second transmembrane domain and substitutes a polar TyrC3H residue at position 121 for a hydrophobic PheMOLF residue (Y121F). Interestingly, CLUSTALw2 protein sequence alignments indicate that the PheMOLF allele is conserved among a number of vertebrate Mc5r orthologs (pufferfish, zebrafish, frog, chicken, rat, cow, human) and all McR paralogs (Mc1r-4r) (Fig. 3; data not shown). The Tyr121 allele is only observed in a subset of Mus musculus strains (C3H/HeSn, A/J, DBA/2J, BALB/cJ, NOD/LtJ) (www.ncbi.nlm.nih.gov/SNP), suggesting that this SNP may be functional.


Figure 3
View larger version (47K):
[in this window]
[in a new window]

  		Fig. 3. Mcr5C3H and MOLF allelic differences. A: the AC3H to TMOLF polymorphism for rs8286628 (*) is shown with flanking sequence. B: the AC3H to TMOLF substitution replaces a tyrosine (YC3H) with a phenylalanine (FMOLF) at amino acid 121 in the second transmembrane domain. This amino acid change plus flanking amino acids (117–126 in mouse) is shown for C3H, MOLF, and other vertebrate species. The F121 allele (boxed) is evolutionarily conserved. C: detailed view of F121 (yellow) and neighboring side chains from the modeled structure for MC5R. The second {alpha}-helix is shown in red and the third {alpha}-helix in blue. The S179, L180, and I183 residues from the third {alpha}-helix are shown in white and are within 3.8 of the side chain of F121.

 
Protein modeling was then performed to investigate the potential effects of this substitution. Mc5r is part of the class A rhodopsin-like GPCR gene family as classified by the GPCR database (http://www.gpcr.org). A molecular model of three-dimensional protein structure for mouse Mc5r was obtained from the ModBase resource, which used the X-ray crystallographic structure of bovine rhodopsin as a template (32). Phe121 is predicted to be in close apposition with three residues (S179, L180, and I183) in the third transmembrane {alpha}-helix (Fig. 3). Previous reports have established that the third {alpha}-helix of bovine rhodopsin undergoes a large conformational change on ligand binding and activation (8). This possible functional effect of the Y121F substitution was examined with PANTHER cSNP and SIFT tools. PANTHER cSNP reports a subPSEC (substitution position-specific evolutionary conservation) score of –6.18, with a score of 0 being neutral and greater negative values predictive of more deleterious substitutions. The SIFT analysis also indicates that the Y121F substitution may be functional. The effect of the Tyr substitution was then evaluated with protein modeling by mutating F121 to a tyrosine. Each of the nine rotamers, which are frequently observed side chain conformations compiled from crystallographic structures, was tested for favorable interactions with neighboring amino acids. Three of the rotamers had neutral interactions, while the other six were unfavorable, consistent with our bioinformatic results that the Y121F substitution may affect Mc5r function. These analyses have identified possible functional SNPs in both Fzd6 for Modvl4 and Mc5r for Modvl5 that may contribute to the modifying effects of these QTL and will be the basis of future analysis.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Our C3H phenotypic analysis indicates that the vl NTD and cataract phenotypes are fully penetrant but <5% of the vl/vl exhibit a belly spot as adults. These phenotypes are inherited in a recessive manner, with no vl heterozygotes exhibiting any of the phenotypes. The vl mutation also results in lethality in ~50% of vl/vl. Interestingly, a nearly equal percentage of E9–E12 vl/vl embryos display spina bifida, consistent with this NTD phenotype causing lethality. This possibility is supported by three other observations: 1) Spina bifida-affected C3Hvl/vl embryos are not observed at later developmental ages, which is also concomitant with increased resorptions. 2) Other mouse NTD mutants universally do not survive past P0 (20, 23), and three adult C3Hvl/vl mice do not exhibit an obvious lumbar-sacral lesion, indicating that either the spina bifida-affected embryos die or the defect repairs itself later in development. All remaining C3Hvl/vl embryos display a closed NTD phenotype. These vl/vl mutants are likely to survive embryogenesis but go on to display congenital cataract with 100% penetrance postnatally.

Adult C3Hvl/vl mice can also display a white belly spot at a low frequency. This phenotype is indicative of a melanocyte defect. Melanocytes are derived from neural crest cells, which are specified from the extreme lateral neural plate cells during neurulation (5, 37). These lateral neural plate cells are also important for neural fold apposition and fusion, which is affected by the vl mutation. EM studies have determined that cellular protrusions normally extend from these neural plate cells, which then interdigitate on contact during neural fold fusion (17). Interestingly, Gpr161 is specifically expressed in lateral neural plate cells during neurulation, and in vl the cellular protrusions have an abnormal ultrastructural morphology (48). Thus in vl/vl embryos with a closed NTD phenotype, aberrant neural fold apposition and fusion may also cause a neural crest/melanocyte defect that then results in the postnatal belly spot phenotype.

Our MOLF cross increases the frequency of the belly spot phenotype but also rescues the vl-associated lethality. This result suggests that MOLF modifiers can bypass the spina bifida phenotype but on some occasions subsequent spinal cord development is abnormal, resulting in the belly spot/neural crest phenotype. Our B6 QTL analysis also supports this conclusion. The B6 background increases both the frequency of the belly spot phenotype and the number of postnatal vl/vl mice exhibiting lumbar-sacral spina bifida and hindlimb paralysis. When the belly spot and spina bifida were used as traits, the same QTL was mapped to chromosome 5 (Modvl1), but interestingly the opposite allele effects were observed (B6 recessive for belly spot, C3H dominant for spina bifida). These results suggest that the underlying locus influences both phenotypes and further support the hypothesis that the vl closed NTD and belly spot phenotypes are causally related.

When the belly spot was used as a trait in our MOLF cross, Modvl4 and Modvl5 were identified. Approximately 50% of adult vl/vlC3H/MOLF mice are also phenotypically indistinguishable from +/+ littermates, indicating that the MOLF background is able to completely rescue all obvious postnatal vl-associated phenotypes. For both Modvl4 and 5, the MOLF/Ei alleles segregate with phenotypically normal mice, suggesting that these modifiers may be sufficient to suppress the lethality, neurulation, and neural crest defects. Our initial characterization of Modvl5 congenic mice supports this possibility (Lazar G, Desai J, Matteson PG, unpublished observations).

Unlike the MOLF cross, the B6/C3H and CAST/C3H backgrounds increased the incidence of the vl-associated lethality. These results suggest that the B6 and CAST backgrounds have modifiers that exacerbate the incidence or severity of the embryonic spina bifida phenotype. However, it is also possible that these modifiers cause lethality through a different mechanism. We attempted to map the position of these modifiers by identifying B6 and CAST alleles that were underrepresented in the F2 vl/vl mice. No significant findings were observed, suggesting that these loci could not be mapped because of their heterogeneity or low penetrance.

The B6 and CAST backgrounds both modify the adult spina bifida and embryonic lethality phenotypes in a similar manner, while these traits are affected differently by the MOLF cross. Both the B6 and CAST crosses resulted in adult F2 vl/vl mice with spina bifida, which was never observed on the C3H/MOLF background. The B6 and CAST backgrounds also increased the embryonic lethality, while the MOLF background rescued this phenotype. One factor that may play a role in these results is the directionality of the crosses (+/+B6 or CAST x vl/vlC3H vs. vl/vlC3H x +/+MOLF), suggesting that unrecognized sex differences may contribute to some of the vl-modifying effects.

We have shown that the vl mutation is due to a frameshift mutation in the orphan GPCR Gpr161, which truncates a significant portion of the COOH-terminal tail. The mutation does not affect the developmental expression of the Gpr161 gene or targeting of the mutant receptor to the plasma membrane. Instead, the truncation decreases steady-state protein levels but also results in increased cell surface expression due to reduced receptor-mediated endocytosis (28). Our data are in agreement with numerous published reports demonstrating a role of the COOH-terminal tail in receptor-mediated endocytosis (2, 6, 16, 25, 34, 35, 42). In addition, the COOH-terminal tail functions as a scaffold for other proteins [GPCR interacting proteins (GIPs)], which then positively or negatively regulate receptor activity (2, 15). These findings are consistent with the vl mutation causing a complex Gpr161 signaling phenotype, which could vary between cell types depending on the expression of GIPs or other proteins in the Gpr161 pathway. This could explain why all three backgrounds can modify the various vl phenotypes and why different QTL have been mapped for the cataract, spina bifida, and belly spot phenotypes.

Our previous analysis (28) indicated that Foxe3, a forkhead transcription factor important for lens development, is a contributor to the Modvl3 cataract-modifying effect. We have now used a similar approach to identify biologically relevant candidate genes for the other four QTL. Resequencing and bioinformatic analysis were focused on two GPCRs, Fzd6 and Mc5r, that may bypass the mutant effects of the vl mutation. One nonsynonymous coding SNP in each gene (Fzd6-rs50388052; Mc5r-rs8256628) is of particular interest. For both SNPs the MOLF allele is evolutionarily conserved among different paralogs and orthologs, and the substitution is predicted to alter protein activity and structure. The Fzd6 SerC3H allele for rs50388052 is expected to affect the secondary structure of the COOH-terminal tail as well as provide an additional site for phosphorylation, both of which may affect the binding of GIPs and GPCR activity. Mc5r protein modeling supports the idea that rs8256628 could be functional with a majority of the polar TyrC3H rotamers being unfavorable. Molecular and biochemical methods will be used to investigate the potential functionality of these SNPs. Other candidate genes identified by our bioinformatics will be analyzed similarly. Together with ongoing congenic analysis these experiments will hopefully identify the genes and polymorphisms that contribute to the modifying effect of these QTL.

In summary, our QTL analysis has demonstrated that numerous loci from three different backgrounds affect the penetrance and expressivity of the vl NTD phenotypes. The identification of these modifier genes will aid in understanding how the Gpr161 pathway regulates neural fold apposition and fusion. The multigenic inheritance of the vl NTD phenotypes on these mixed backgrounds will also serve as a mouse model to study the genetic and developmental basis of this common human disorder.


   GRANTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by research grants from The New Jersey Commission on Spinal Cord Research (02-3016-SCR-S-0, 04-2901-SCR-E-0) to J. H. Millonig.


   ACKNOWLEDGMENTS
 
We thank Theolyn Gilley and Taslima Rahman for technical support.


   FOOTNOTES
 
Address for reprint requests and other correspondence: J. H. Millonig, Center for Advanced Biotechnology and Medicine, 679 Hoes Lane, Piscataway, NJ 08854-5638 (e-mail: millonig{at}cabm.rutgers.edu).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

* R. Korstanje and J. Desai contributed equally to this work. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Blixt A, Mahlapuu M, Aitola M, Pelto-Huikko M, Enerback S, Carlsson P. A forkhead gene, FoxE3, is essential for lens epithelial proliferation and closure of the lens vesicle. Genes Dev 14: 245–254, 2000.[Abstract/Free Full Text]
  2. Bockaert J, Fagni L, Dumuis A, Marin P. GPCR interacting proteins (GIP). Pharmacol Ther 103: 203–221, 2004.[CrossRef][Web of Science][Medline]
  3. Breitwieser GE. G protein-coupled receptor oligomerization: implications for G protein activation and cell signaling. Circ Res 94: 17–27, 2004.[Abstract/Free Full Text]
  4. Broman KW, Wu H, Sen S, Churchill GA. R/qtl: QTL mapping in experimental crosses. Bioinformatics 19: 889–890, 2003.[Abstract/Free Full Text]
  5. Bronner-Fraser M, Fraser SE. Cell lineage analysis reveals multipotency of some avian neural crest cells. Nature 335: 161–164, 1988.[CrossRef][Web of Science][Medline]
  6. Cao TT, Deacon HW, Reczek D, Bretscher A, von Zastrow M. A kinase-regulated PDZ-domain interaction controls endocytic sorting of the beta2-adrenergic receptor. Nature 401: 286–290, 1999.[CrossRef][Web of Science][Medline]
  7. Chen W, Kelly MA, Opitz-Araya X, Thomas RE, Low MJ, Cone RD. Exocrine gland dysfunction in MC5-R-deficient mice: evidence for coordinated regulation of exocrine gland function by melanocortin peptides. Cell 91: 789–798, 1997.[CrossRef][Web of Science][Medline]
  8. Choi G, Landin J, Galan JF, Birge RR, Albert AD, Yeagle PL. Structural studies of metarhodopsin II, the activated form of the G-protein coupled receptor, rhodopsin. Biochemistry 41: 7318–7324, 2002.[CrossRef][Web of Science][Medline]
  9. Churchill GA, Doerge RW. Empirical threshold values for quantitative trait mapping. Genetics 138: 963–971, 1994.[Abstract]
  10. Clamp M, Cuff J, Searle SM, Barton GJ. The Jalview Java alignment editor. Bioinformatics 20: 426–427, 2004.[Abstract/Free Full Text]
  11. Copp AJ, Greene ND, Murdoch JN. The genetic basis of mammalian neurulation. Nat Rev Genet 4: 784–793, 2003.[CrossRef][Web of Science][Medline]
  12. Davidson CE, Li Q, Churchill GA, Osborne LR, McDermid HE. Modifier locus for exencephaly in Cecr2 mutant mice is syntenic to the 10q25.3 region associated with neural tube defects in humans. Physiol Genomics 31: 244–251, 2007.[Abstract/Free Full Text]
  13. Detrait ER, George TM, Etchevers HC, Gilbert JR, Vekemans M, Speer MC. Human neural tube defects: developmental biology, epidemiology, and genetics. Neurotoxicol Teratol 27: 515–524, 2005.[CrossRef][Web of Science][Medline]
  14. Dickie M. Vacuolated lens. Mouse News Lett 36: 39–40, 1967.
  15. El Far O, Betz H. G-protein-coupled receptors for neurotransmitter amino acids: C-terminal tails, crowded signalosomes. Biochem J 365: 329–336, 2002.[Web of Science][Medline]
  16. Gainetdinov RR, Premont RT, Bohn LM, Lefkowitz RJ, Caron MG. Desensitization of G protein-coupled receptors and neuronal functions. Annu Rev Neurosci 27: 107–144, 2004.[CrossRef][Web of Science][Medline]
  17. Geelen JA, Langman J. Ultrastructural observations on closure of the neural tube in the mouse. Anat Embryol (Berl) 156: 73–88, 1979.[CrossRef][Medline]
  18. Guex N, Peitsch MC. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18: 2714–2723, 1997.[CrossRef][Web of Science][Medline]
  19. Guo N, Hawkins C, Nathans J. Frizzled6 controls hair patterning in mice. Proc Natl Acad Sci USA 101: 9277–9281, 2004.[Abstract/Free Full Text]
  20. Harris MJ, Juriloff DM. Mini-review: toward understanding mechanisms of genetic neural tube defects in mice. Teratology 60: 292–305, 1999.[CrossRef][Web of Science][Medline]
  21. Heydorn A, Sondergaard BP, Ersboll B, Holst B, Nielsen FC, Haft CR, Whistler J, Schwartz TW. A library of 7TM receptor C-terminal tails. Interactions with the proposed post-endocytic sorting proteins ERM-binding phosphoprotein 50 (EBP50), N-ethylmaleimide-sensitive factor (NSF), sorting nexin 1 (SNX1), and G protein-coupled receptor-associated sorting protein (GASP). J Biol Chem 279: 54291–54303, 2004.[Abstract/Free Full Text]
  22. Ikenouchi J, Uwabe C, Nakatsu T, Hirose M, Shiota K. Embryonic hydromyelia: cystic dilatation of the lumbosacral neural tube in human embryos. Acta Neuropathol (Berl) 103: 248–254, 2002.[CrossRef][Medline]
  23. Juriloff DM, Harris MJ. Mouse models for neural tube closure defects. Hum Mol Genet 9: 993–1000, 2000.[Abstract/Free Full Text]
  24. King DC, Taylor J, Elnitski L, Chiaromonte F, Miller W, Hardison RC. Evaluation of regulatory potential and conservation scores for detecting cis-regulatory modules in aligned mammalian genome sequences. Genome Res 15: 1051–1060, 2005.[Abstract/Free Full Text]
  25. Koch T, Schulz S, Schroder H, Wolf R, Raulf E, Hollt V. Carboxyl-terminal splicing of the rat mu opioid receptor modulates agonist-mediated internalization and receptor resensitization. J Biol Chem 273: 13652–13657, 1998.[Abstract/Free Full Text]
  26. Kolbe D, Taylor J, Elnitski L, Eswara P, Li J, Miller W, Hardison R, Chiaromonte F. Regulatory potential scores from genome-wide three-way alignments of human, mouse, and rat. Genome Res 14: 700–707, 2004.[Abstract/Free Full Text]
  27. Lander ES, Botstein D. Mapping mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121: 185–199, 1989.[Abstract/Free Full Text]
  28. Matteson PG, Desai J, Korstanje R, Lazar G, Borsuk TE, Rollins J, Kadambi S, Joseph J, Rahman T, Wink J, Benayed R, Paigen B, Millonig JH. The orphan G protein-coupled receptor, Gpr161, encodes the vacuolated lens locus and controls neurulation and lens development. Proc Natl Acad Sci USA 105: 2088–2093, 2008.[Abstract/Free Full Text]
  29. Medina-Martinez O, Jamrich M. Foxe view of lens development and disease. Development 134: 1455–1463, 2007.[Abstract/Free Full Text]
  30. Milligan G. G protein-coupled receptor dimerization: function and ligand pharmacology. Mol Pharmacol 66: 1–7, 2004.[Abstract/Free Full Text]
  31. Mu J, Gilley T, Turner R, Paigen B. High-resolution genetic map of mouse chromosome 1 between D1Mit227 and D1Mit15 by use of an intercross of C57BL/6J x C3HeB/FeJ vl/vl. Mamm Genome 7: 770, 1996.[CrossRef][Web of Science][Medline]
  32. Okada T, Fujiyoshi Y, Silow M, Navarro J, Landau EM, Shichida Y. Functional role of internal water molecules in rhodopsin revealed by X-ray crystallography. Proc Natl Acad Sci USA 99: 5982–5987, 2002.[Abstract/Free Full Text]
  33. Ormestad M, Blixt A, Churchill A, Martinsson T, Enerback S, Carlsson P. Foxe3 haploinsufficiency in mice: a model for Peters' anomaly. Invest Ophthalmol Vis Sci 43: 1350–1357, 2002.[Abstract/Free Full Text]
  34. Prossnitz ER. Novel roles for arrestins in the post-endocytic trafficking of G protein-coupled receptors. Life Sci 75: 893–899, 2004.[CrossRef][Web of Science][Medline]
  35. Roth A, Kreienkamp HJ, Nehring RB, Roosterman D, Meyerhof W, Richter D. Endocytosis of the rat somatostatin receptors: subtype discrimination, ligand specificity, and delineation of carboxy-terminal positive and negative sequence motifs. DNA Cell Biol 16: 111–119, 1997.[Web of Science][Medline]
  36. Sayle RA, Milner-White EJ. RASMOL: biomolecular graphics for all. Trends Biochem Sci 20: 374, 1995.[CrossRef][Web of Science][Medline]
  37. Selleck MA, Bronner-Fraser M. Origins of the avian neural crest: the role of neural plate-epidermal interactions. Development 121: 525–538, 1995.[Abstract]
  38. Semina EV, Brownell I, Mintz-Hittner HA, Murray JC, Jamrich M. Mutations in the human forkhead transcription factor FOXE3 associated with anterior segment ocular dysgenesis and cataracts. Hum Mol Genet 10: 231–236, 2001.[Abstract/Free Full Text]
  39. Sen S, Churchill GA. A statistical framework for quantitative trait mapping. Genetics 159: 371–387, 2001.[Abstract/Free Full Text]
  40. Sugiyama F, Churchill GA, Higgins DC, Johns C, Makaritsis KP, Gavras H, Paigen B. Concordance of murine quantitative trait loci for salt-induced hypertension with rat and human loci. Genomics 71: 70–77, 2001.[CrossRef][Web of Science][Medline]
  41. Volpe J. Neurology of the Newborn (3rd ed.). Philadelphia, PA: Saunders, 1995, p. 5–21.
  42. von Zastrow M, Kobilka BK. Antagonist-dependent and -independent steps in the mechanism of adrenergic receptor internalization. J Biol Chem 269: 18448–18452, 1994.[Abstract/Free Full Text]
  43. Wang Y, Guo N, Nathans J. The role of Frizzled3 and Frizzled6 in neural tube closure and in the planar polarity of inner-ear sensory hair cells. J Neurosci 26: 2147–2156, 2006.[Abstract/Free Full Text]
  44. Wilson DB, Wyatt DP. Aberrant convergence of the neural folds in the mouse mutant vl. Teratology 45: 105–112, 1992.[CrossRef][Web of Science][Medline]
  45. Wilson DB, Wyatt DP. Closure of the posterior neuropore in the vl mutant mouse. Anat Embryol (Berl) 178: 559–563, 1988.[CrossRef][Medline]
  46. Wilson DB, Wyatt DP. In vitro expression of neural tube pathology in the vl mutant mouse. J Neuropathol Exp Neurol 52: 253–259, 1993.[Web of Science][Medline]
  47. Wilson DB, Wyatt DP. Pathogenesis of neural dysraphism in the mouse mutant vacuolated lens (vl). J Neuropathol Exp Neurol 45: 43–55, 1986.[Web of Science][Medline]
  48. Wilson DB, Wyatt DP. Ultrastructural defects in the apical neural folds in mutant embryos with spina bifida. Acta Neuropathol (Berl) 79: 94–100, 1989.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
35/3/296    most recent
90260.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Korstanje, R.
Right arrow Articles by Millonig, J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Korstanje, R.
Right arrow Articles by Millonig, J. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.