The squiggle tail (squig) mutation in mice is associated with a deletion in the mesenchyme homeobox 1 (Meox1) gene.

OBJECTIVE: We have taken a positional approach to assign the spontaneous squiggle tail (squig) mutation in mice to a specific gene defect.
RESULTS: A large panel of backcross mice was produced and characterized to map squig to high genetic resolution on mouse Chromosome (Chr) 11. Two overlapping candidate genes that co-localized with squig (Meox1, for mesenchyme homeobox 1; and Gm11551, which encodes a lncRNA located entirely within the first intron of Meox1) were fully sequenced to discover any squig-specific defects. This analysis revealed a 3195 bp deletion that includes all of Meox1, Exon 1 but does not disrupt Gm11551. We recommend that the squig mutation be renamed Meox1squig, and suggest that this variant may offer an appropriate animal model for Klippel-Feil syndrome 2 (KFS2) in humans.

Introduction

The recessive squiggle tail mutation (abbreviated squig) arose spontaneously in the BALB/cJ inbred mouse strain at The Jackson Laboratory (Bar Harbor, ME, USA) in 2013, and has been maintained on a segregating, coisogenic background since that time. Mice homozygous for squig display a shortened and very curly tail and are frequently smaller than their non-mutant littermates (see Additional file 1: Figure S1). In 2016, Karst et al. [1] mapped squig to mouse Chromosome (Chr) 11, based on the analysis of single nucleotide polymorphisms (SNPs) among a small set of F2 homozygotes, but the genetic resolution achieved was not sufficient to suggest any causative gene.

As a basis for assigning the squiggle tail phenotype to a specific genetic cause—which would facilitate the further analysis of this interesting variant and help to identify an orthologous human disorder—here we have fine-mapped squig with respect to various microsatellite and SNP markers on mouse Chr 11. This analysis identified a small set of co-localizing candidate genes, and we now suggest that one of these, Meox1 (for mesenchyme homeobox 1), harbors the squig defect.

Main text

Methods

Mice

Mice from the standard inbred strains C57BL/6 J (JAX stock #000664) and BALB/cJ (JAX stock #000651), and co-isogenic BALB/cJ-squig/GrsrJ mice (JAX stock #026620) were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). The squig mutation was maintained at CCSU by crossing heterozygotes with mutant homozygotes. Mutants were reliably identified (with at least 99.3% penetrance in our colony) by their short and curly tails that are apparent from birth (see Additional file 1: Figure S1). At the end of the study, mice were killed by cervical dislocation or by use of CO2 gas added to a chamber (typically their home cage) using a compressed gas cylinder fitted with a flow meter adjusted to displace only 30–70% of the chamber volume per minute (consistent with the recommendations of the AVMA Guidelines for the Euthanasia of Animals, 2020 Edition). Only the P.I. (TRK) who was trained at The Jackson Laboratory (Bar Harbor, ME) and has over 30 years of experience, performed euthanasia.

DNA isolation and marker typing

Genomic DNA was isolated from 2 mm tail-tip biopsies taken from two- to three-week-old mice using NucleoSpin® Tissue kits (Macherey–Nagel, Düren, Germany; distributed by Clontech Laboratories, Inc., Mountain View, CA, USA), as directed. DNA samples from standard inbred and mutant strains that we do not routinely maintain in our colony were purchased from The Jackson Laboratory’s Mouse DNA Resource.

The polymerase chain reaction (PCR) was performed in 13 µl reactions using the Titanium® PCR kit from Clontech Laboratories, as directed. Oligonucleotide primers for PCR were designed and synthesized by Integrated DNA Technologies, Inc. (Coralville, IA, USA), based on sequence information available online [2, 3]. To score PCR product sizes for dimorphic microsatellite markers, reactions plus 3 µl loading buffer were electrophoresed through 3.5% NuSieve® agarose (Lonza, Rockland, ME, USA) gels. Gels were stained with ethidium bromide and photographed under ultraviolet light. In addition to eight standard microsatellite markers [4] on Chr 11, eight DNA markers based on single nucleotide polymorphisms previously reported to differ between strains BALB/cJ and C57BL/6 J [2, 3] were scored. These markers (herein designated SNP#) are described in detail in Additional file 2: Table S1 and Additional file 3: Table S2.

Sequence analysis

For DNA sequence analysis, about 1.5 µg of individual PCR amplimers were purified and concentrated into a 30 µl volume using NucleoSpin® PCR Clean-up kits, and then shipped to the Keck Foundation Resource Laboratory at Yale University (New Haven, CT, USA) for primer-extension analysis.

Meox1squig

To rapidly determine genotypes at the squig locus (especially among phenotypically wild type mice) we used a standard PCR assay that employed three primers: a single forward primer (F1, 5’-GTTACCAGGAGGTGCTCAAA-3’) that annealed 5’ to the Meox1 deletion and two reverse primers—one that annealed within the Meox1 deletion (R1, 5’-GTGAAATGTGAGAGAGGAGAGG-3') and one that annealed 3’ to the deletion, within Gm11551, Exon 2 (R2, 5’-CCAGATCCCAGCAATCAAGATA-3'). Primers F1 and R1 direct the amplification of a 268 bp product specific to wild type BALB/cJ templates; the F1, R2 primer pair direct the amplification of a 456 bp product specific to squig templates.

Results

To genetically map the squig mutation, F1 heterozygotes (made by crossing BALB/c-squig/squig mice with standard C57BL/6 J mice) were crossed with squig/squig mutants, producing 1008 backcross (N2) offspring that segregated for alternative alleles of squig and numerous molecular markers. Guided by Karst’s previous mapping efforts [1], DNAs isolated from this N2 panel were typed for eight, PCR-scorable microsatellite (dinucleotide repeat) markers [4] known to map to distal Chr 11. Additional file 4: Figure S2 shows the string of markers transmitted by the F1 parent to each of these 1008 N2 progeny. This haplotype analysis suggested that the squig gene must be located within the 3.2 cM region between D11Mit59 and D11Mit360 (Fig. 1A). This genomic interval includes Rpl27 (for ribosomal protein L27), and, because defects in the related genes Rpl24 and Rpl38 have been shown to cause tail abnormalities in mice [5, 6] respectively, we investigated Rpl27 as the potential basis of squig. However, DNA sequence analysis of all exons of Rpl27 in squig mutants (data not shown) revealed no defects compared to wild type.

Fig. 1

Genetic and physical maps of the squig region on distal mouse Chr 11. A Low-resolution genetic map of squig and eight microsatellite markers, based on the backcross panel described in Additional file 1 (Figure S2). The number of crossovers (out of 1008 meioses) located in each marker-defined interval is shown. As described in Figure S2, the squig mutation must be located within the 3.17 cM interval between markers D11Mit59 and D9Mit360. B The D11Mit59 to D11Mit360 region is expanded to show the relative positions of 8 SNP markers used to type the 32 panel members recombinant in that 3.24 Mb interval. The number of crossovers located in each marker/SNP-defined interval is shown. The squig mutation must lie between SNP10 and SNP14, and was never separated from SNP6 nor SNP13. C The 0.6 Mb span between SNP10 and SNP14 includes 18 protein coding genes (shown as colored boxes). At least 5 of these (Arl4d, Meox1, Sost, Dusp3 and Cd300lg; shown as orange or green boxes) are known to be expressed in the axial skeleton [2]; only Meox1 (green box) is known to affect tail morphology when disrupted in mice [8, 9]. D The Meox1 gene is expanded to show the 3 exons it comprises (with a 2 kb scale bar). Green boxes represent coding regions; white boxes indicate the 5’ and 3’ untranslated regions. Meox1, Intron 1–2 includes a two-exon lncRNA (Gm11551, shown as blue boxes) that is transcribed from the complementary strand

DNA samples from the 32 mice identified as having a crossover between D11Mit59 and D11Mit360 were typed next for eight SNPs known to lie in that 3.24 Mb interval. These eight SNP markers are described in detail in Additional file 2: Table S1 and Additional file 3: Table S2 and are designated herein as SNP#. This analysis located six crossovers that fell centromeric to squig (between SNP10 and SNP6), and one crossover that fell distal to squig (between SNP13 and SNP14) (see Fig. 1B), thus restricting the location of squig between SNP10 and SNP14 (and very near SNP6 and SNP13, which were not meiotically separated from each other or from squig).

The 0.6 Mb span from SNP10 to SNP14 includes 18 expressed genes (Fig. 1C) (which, incidentally, do not include Rpl27). While four of these genes (Arl4d, Sost, Dusp3 and Cd300lg) have been associated with abnormal bone morphology or mineralization [2] the homeodomain-containing transcription factor gene, Meox1 (for mesenchyme homeobox 1), became our primary gene candidate due to its well-established role in axial skeleton formation [7-10]. DNA sequence analysis of all Meox1 exons in wild type BALB/cJ and BALB/c-squig/squig genomic DNA revealed a 3195 bp deletion that extends from the Meox1 promoter region to include all of Exon 1 and part of Intron 1–2 (Fig. 2). The deletion does not extend as far as predicted gene Gm11551 [12] which encodes a lncRNA that is entirely contained within Intron1-2 of Meox1 and is transcribed from the complementary DNA strand (see Fig. 1D).

Fig. 2

3195 bp including Meox1, Exon 1 are deleted in the squiggle tail variant. Bases shown in lower-case green are from the Meox1 promoter region, uppercase black letters are from Meox1, Intron 1–2. The upper boxed region represents Meox1, Exon 1; the 5’UTR is shown in white, the coding region is shown in green. Exon 2 of the lncRNA-encoding gene known as Gm11551, which lies entirely within Intron 1–2 of Meox1, is represented by a blue box (also see Fig. 1D). The number of base pairs that compose each segment is shown. A three-base direct repeat (5’-GAG-3’) found at the deletion breakpoint, a typical characteristic of spontaneous deletions in mammals [11], is underlined

Next, we used a “3-primer” PCR test (see Additional file 5: Figure S3A) to rapidly screen for the presence of the squig-associated deletion of Meox1, Exon 1 among 28 standard mouse strains but found no Meox1, Exon 1 defect in any of them (see Additional file 5: Figures S3B and Additional file 6: S4), suggesting that the 3195 bp deletion is specific to the squig mutation.

Discussion

Because Meox1 and squig map to the same small region on Chr 11, because squig mutants display a severe, specific defect in Meox1, and because engineered mutations in Meox1 produce similar recessive vertebral anomalies [including hemivertebrae, tail kinks and craniovertebral fusions, see [8, 9]] we suggest that squig is a spontaneous mutant allele of Meox1 and recommend that its official designation be changed to Meox1.

At least four independent defects in the human MEOX1 orthologue [13-15] have been associated with Klippel-Feil syndrome-2 (KFS2), an autosomal recessive condition characterized by a short neck, low occipital hairline and reduced bilateral neck movements resulting from the fusion of cervical vertebrae [16]. Especially because the previously-described mouse variants [8, 9] are no longer extant, we suggest that the Meox1 variant described herein (available from The Jackson Laboratory as JAX stock #026620) could provide a highly relevant animal model for this inherited human disorder.

Limitations

Rigorous proof for this gene assignment would require rescue of the mutant phenotype by the single addition of a wild type Meox1 allele, or complementation testing with an extant, engineered Meox1 mutation, for example. These formal tests were not performed here. The location of the antisense lncRNA Gm11551 within the first intron of mouse Meox1 may suggest a cis regulatory relationship [17] that warrants further investigation, although it is notable that the human MEOX1 orthologue does not harbor a similar lncRNA gene. While the Meox1 deletion does not disrupt the Gm11551 coding sequence, we did not verify the normal expression or processing of Gm11551 RNA.

Additional file 1: Figure S1. A wild type heterozygote (left) and a mutant squig/squig mouse (right) at 10 days of age.

Additional file 2: Table S1. Description of SNP markers referred to in the Girard et al. (2022) text.

Additional file 3: Table S2. Location of SNP markers referred to in the Girard et al. (2022) text.

Additional file 4: Figure S2. Segregation of alleles of squig and eight microsatellite markers among 1008 intraspecific backcross progeny.

Additional file 5: Figure S3. A deletion including Meox1, Exon 1 appears to be specific to the squig mutation.

Additional file 6: Figure S4. Original, uncropped photographs that were used to produce Figure S3B (as required by the editor).