Figure 1.
Deafness and cochlear pathology of rda and rda2J mutant mice.
A. ABR thresholds (dB SPL) of rda/rda and rda 2J/rda2J mutant mice and non-mutant controls tested at 33–48 days of age. ABRs for 8, 16, and 32 kHz stimulus frequencies were not detected in any of the rda/rda (N = 9) and rda2J/rda2J (N = 8) mutant mice tested, even with the maximum stimulus presentation of 100 dB SPL. Heterozygous +/rda (N = 8) and +/rda2J (N = 4) mice and homozygous +/+ B6 mice (N = 17) were combined as controls because their ABR thresholds did not significantly differ from one another. Error bars represent standard deviations of the threshold means. B, C. Cross sections through the basal turn of the cochlea from a +/rda heterozygous control mouse (B) and a littermate rda/rda mutant mouse (C) examined at 4 months of age. Note the complete degeneration of the organ of Corti (oc) and decreased density of spiral ganglion cells (sgc) in the rda/rda cochlea. Scale bars represent 100 microns. Cochlear cross sections of rda2J mutant mice (not shown) exhibited this same pathology.
Figure 2.
Cochlear hair cell abnormalities in rda and rda2J mutant mice.
Examination of the apical surfaces of cochlear hair cells by scanning electron microscopy showed normal bundle morphology in both inner hair cells (IHC) and outer hair cells (OHC) of rda/rda mutant mice at P0 (A). At P7 the stereocilia adjacent to the kinocilium (marked by arrow in H) had degenerated in some of the OHC bundles (marked by asterisks in B, D) of rda/rda mice, compared with the fully intact bundles of wildtype (+/+) controls (C). IHC bundles of rda/rda mice retained a normal appearance at P10 (D, K), but by P15 (F, L) all IHCs of mutant mice exhibited stereocilia elongations and fusions as compared with the normal IHC bundle morphology of age-matched controls (E). IHCs at P35 illustrate the striking degree of stereocilia elongations and fusions seen in rda/rda mutants (N) compared with age-matched controls (M). In contrast to IHCs, the OHC stereocilia of rda/rda mice did not elongate or fuse, but they did degenerate over time. By P15, nearly all OHCs had lost stereocilia at the bundle peak and a few (marked by asterisks) had lost the entire bundle (F, I), as compared with the intact OHC bundles of control mice (J). The IHC and OHC bundle abnormalities of rda2J/rda2J mutant mice (G) were very similar to those of rda/rda mice (F). Scale bars: A–G, 10 microns; H–N, 5 microns. All hair cells shown in the figure are from the middle region of the cochlea, but similar bundle abnormalities were seen in other regions (not shown), consistent with the absence of an ABR at all test frequencies (Fig. 1A).
Figure 3.
Elmod1 gene structure and expression.
A. The mouse Elmod1 gene spans 63.84 kb and is comprised of 11 exons transcribed on the reverse strand of the NCBI genomic DNA reference sequence (Chr 9: 53,823,108 to 53,759,267 bp, Build 37). B. The 2605 nt reference mRNA sequence (NM_177769) encodes a 326 amino acid protein (NP_808437). Amino acids 132–305 (encoded by exons 6–11) comprise a conserved ELMO/CED-12 domain (pfam04727, IPR006816) characteristic of the ELMO protein family. Diagrams A and B were downloaded from the Ensembl website (http://www.ensembl.org/). C. Elmod1 gene expression was examined by northern blot analysis. Commercially prepared blots from mouse embryos and adult tissues (MTN blots, Clontech, Palo Alto, CA) were hybridized with a mouse Elmod1 cDNA probe corresponding to nucleotides 426–904 of the NM_177769 reference cDNA sequence. Blots contained purified Poly A+ RNA from 7 day (lane 1), 11 day (lane 2), 15 day (lane 3) and 17 day (lane 4) embryos and from heart (H), brain (B), spleen (S), lung (Lu), liver (Li), skeletal muscle (M), kidney (K), and testis (T) of adult mice. A single transcript of about 2,600 nucleotides was detected, most highly expressed in adult brain.
Figure 4.
Molecular analysis of the Elmod1rda mutation.
Exons 1–5 of Elmod1 are deleted in rda mutant chromosomes. A. Southern blots of genomic DNA from mice with +/+ (lanes marked 1), +/rda (lanes marked 2) and rda/rda (lanes marked 3) genotypes, digested with BamHI (Bam), EcoRI (Eco), PstI (Pst), and PvuII (Pvu) restriction enzymes. The blots shown were hybridized with Elmod1 cDNA probes corresponding to exons 3–7 (nucleotides 426–904 of NM_177769, blot on left) and exons 3–4 (nucleotides of NM_177769,blot on right). Elmod1 mRNA is not expressed in rda/rda mutant mice. B. Northern blots of total RNA extracted from brains of adult C57BL/6J controls (+/+) and rda/rda mutants was hybridized first with an Elmod1 cDNA probe (nucleotides 426–904 of NM_177769, top figure) and subsequently with a beta actin control probe (Actb, bottom) to evaluate RNA loading concentrations. C. PCR with primer pairs spanning the genomic region around Elmod1 was used to define the extent of the rda deletion. Primers immediately flanking the mutation amplified a 963 bp product in mutant (rda) but not wildtype (+) DNA, and sequencing of this product identified a 757 bp insertion. A three-primer assay was developed to distinguish homozygous wildtype mice (+), rda/+ heterozygotes (het), and rda/rda homozygous mutants (rda). D. The molecular structure of the rda mutation was precisely defined as a 137,786 bp deletion with a 757 bp inverted insertion at the site of the deletion. Exons 1–5 of Elmod1 and the entire Tnfaip8l3 gene are included in the deletion.
Figure 5.
Molecular analysis of the Elmod1rda-2J mutation.
A. Southern blots of genomic DNA from mice with +/+ (lanes marked 1), +/rda2J (lanes marked 2) and rda2J/rda2J (lanes marked 3) genotypes, digested with EcoRI (Eco), PstI (Pst), PvuII (Pvu), and MspI (Msp) restriction enzymes. The blot was hybridized with an Elmod1 cDNA probe corresponding to exons 7–11. Although the same quantity of DNA was loaded in each lane, the intensity of Elmod1-hybridizing bands is greater in in EcoRI- and PvuII-digested DNA samples from rda2J/rda2J mice than samples from +/+ mice and exhibit additional bands (indicated by arrows) in PstI and MspI digested DNA. B. A northern blot of total RNA extracted from brains of adult C57BL/6J control (lane 1), rda2J/+ heterozygotes (lanes 2 and 3), rda2J/rda2J homozygotes (lanes 4 and 5), and rda/rda homozygote (lane 6; negative control) was hybridized with an Elmod1 cDNA probe. Wildtype Elmod1 transcript (∼2600 nt) was not detected in RNA from rda2J/rda2J mice; however, a 3200 nt mutant transcript (about 600 nucleotides larger than wildtype) was abundantly expressed. C. PCR products from rda2J–derived cDNAs indicate an intragenic duplication. cDNAs from mice with +/+ (lanes marked 1), +/rda2J (lanes marked 2) and rda2J/rda2J (lanes marked 3) genotypes were used as PCR templates in combination with primers specific to Elmod1 exons. Expected wildtype PCR products are marked by asterisks, and unexpected large PCR products unique to rda2J samples are indicated by arrows. 500 bp (lane marked L1) and 100 bp (lane marked L2) ladders were used to estimate PCR product sizes, and predicted sizes are given in Table S2B. D. A western blot of protein extracts from adult brains of rda/rda (lane 1, negative control), rda2J/rda2J (lane 2), and +/+ B6 mice (lane 3), shows that the predicted 38 kDa wildtype ELMOD1 protein is absent from both rda and rda2J mutant mice, but a larger 62 kDa mutant protein can be seen in rda2J mutants (indicated by arrow in lane 2), corresponding to the predicted duplication of 202 amino acids. The polyclonal ELMOD1 antibody cross-reacted with other unknown proteins (non-specific bands, NS). E. The large PCR products obtained from rda2J–derived cDNA (indicated by arrows in C) and multiple combinations of exon-specific primers were used to determine the DNA sequence and structure of the mutant Elmod1 transcript, and this analysis indicated a duplication of exons 3–8, as shown in the diagram of the presumed mutant gene structure.
Figure 6.
In situ expression of Elmod1 mRNA in the inner ear.
Expression was detected in inner hair cells (IHC) and outer hair cells (OHCs) of the cochlea (A) and vestibular hair cells (VHCs) of the crista ampullaris (B) in inner ears from wildtype (+/+) mice at postnatal day 7 (P7). Mutant rda/rda mice served as negative controls for probe specificity, as seen by the lack of detectable Elmod1 expression in cochlear hair cells of P7 rda/rda mice (C). Elmod1 expression was not detected above background staining in inner ears of P2 wildtype mice (D) or in P15 mice (not shown). The general location of hair cells (HCs) is indicated in the P2 cochlea (D), because inner and outer hair cells are not clearly distinguishable in cryosections at this age. For reference, the Reissner's membrane (RM) and stria vascularis (SV) of the cochleae are indicated with arrows. All panels are at the same magnification; the scale bar represents 50 micrometers.