Double homozygous waltzer and Ames waltzer mice provide no evidence of retinal degeneration

Breeding to generate Cdh23^v-6J and Pcdh15^av-Jfb double mutant mice

Animal procedures were conducted in accordance with the NIDCD/NIH Animal Care and Use Committee, protocol 1126–03. The Pcdh15^av-Jfb mutant allele arose spontaneously in offspring segregating a gastrin-releasing peptide receptor (Grpr) null allele (B6.129-Grpr^tm1Jfb) at the fifth backcross generation [45] and was maintained afterward by intercrossing animals. The Cdh23^v-6J allele arose on the B10.A-H2^h4 (4R)/SgDvEg congenic strain at The Jackson Laboratory (Bar Harbor, ME). Cdh23^v-6J mice were kindly provided by Konrad Noben-Trauth (NIDCD/NIH), which were then backcross-intercrossed with C57BL/6J mice for six generations.

Pcdh15 is located 10 cM from Cdh23 on mouse chromosome 10 [17]. To produce a chromosome with Pcdh15^av-Jfb in cis to Cdh23^v-6J, we crossed heterozygous Pcdh15^av-Jfb females with homozygous Cdh23^v-6J males. Double heterozygous F₁ progeny (Cdh23^v-6J +/+ Pcdh15^av-Jfb) were intercrossed. Mice homozygous for Cdh23^v-6J and heterozygous for Pcdh15^av-Jfb (Cdh23^v-6J +/Cdh23^v-6J Pcdh15^av-Jfb), which carry a chromosome with a crossover between Cdh23^v-6J and Pcdh15^av-Jfb, were identified. These mice were then crossed with C57BL/6J to obtain double cis-heterozygotes (+ +/Cdh23^v-6J Pcdh15^av-Jfb). Intercrossing double cis-heterozygotes produced mice that were homozygous for both Cdh23^v-6J and Pcdh15^av-Jfb. All of the genotypic combinations had a C57BL/6J genetic background.

Genotyping

Genomic DNA was isolated from tail biopsies (DNeasy Blood and Tissue Kit; Qiagen, Valencia, CA). The Pcdh15^av-Jfb allele was genotyped by PCR amplification of exon 17 of Pcdh15 (GenBank AF281899) using the forward primer 5′-GAT CTT CGC ATC CAA TGC AG-3′ and reverse primer 5′-AGC ATT TCG TTC TGG GTG AAT A-3′. Amplimers were sequence verified using Big DyeTerminator (ABI version 3.1; Applied Biosystems, Foster City, CA), size separated on a 3730xl DNA Analyzer (Applied Biosystems), and data analyzed using Lasergene (DNASTAR Inc.,

Madison, WI). The Cdh23^v-6J mutation was genotyped by PCR amplification of exon 9 of Cdh23 (GenBank AY026062) by using the forward primer 5′-AAA GGG CGC AGT AAT CTG TG-3′ and reverse primer 5′-TCC ACA CCT TCC AAG TAG GG-3′. Amplimers were sequence verified as we have described.

Scanning electron microscopy

Scanning electron microscopy (SEM) examinations were performed on the organ of Corti. Mice were euthanized with CO₂, decapitated, and their cochleae were removed and fixed in a solution containing 2.5% glutaraldehyde in 0.1 M cacodylate buffer supplemented with 2 mM CaCl₂. To expedite fixation of the organ of Corti, we perfused the cochlea with the fixative via the holes at the apex and at the round window. This was followed incubation of cochlea in the same fixative for 1.5 h at room temperature. The organ of Corti was dissected from the modiolus in Hank's balanced salt solution (Invitrogen, Carlsbad, CA) and dehydrated with an acetone series. The specimens were critical point dried from liquid CO₂ (Bal-Tec CPD030, Leica Microsystems, Wetzlar, Germany), sputter-coated (Balzers MED010, Balzers, Liechtensein) with platinum (5 nm, controlled by STM-100/MF film thickness monitor, Sycon Instruments, East Syracuse, NY), and observed with a field-emission scanning electron microscope (S-4500, Hitachi Technologies, Schaumburg, IL).

Histological evaluation

Histology was performed on 32 different tissues from P120 male mice obtained from two wild-type (+ +/+ +), two cis-heterozygotes (+ +/Cdh23^v-6J Pcdh15^av-Jfb), and two double homozygotes (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb). Tissues examined included brain, thymus, spleen, pancreas, lymph nodes, liver, kidneys, adrenal gland, salivary glands, Harderian gland, trachea, thyroid, esophagus, aorta, lung, testes, epididymis, urinary bladder, ovaries, uterus, oviducts, cervix, prostate, seminal vesicles, preputial gland, heart, tongue, skeletal muscle, sciatic nerve, eyes, stomach, small intestine, cecum, colon, rectum, skin, sternum, vertebrae, femur, and spinal cord. Tissues were fixed in buffered aqueous zinc formalin, embedded in paraffin, sectioned, and stained with hematoxylin and eosin (HE) then examined with bright-field microscopy. Tissue samples containing bone were decalcified in a solution of formic acid and sodium citrate before sectioning.

To evaluate histological differences between wild-type, heterozygous, and double homozygous mutant mice retinas, we measured the outer nuclear layer (ONL) width at 200 µm intervals on midretinal sections beginning 200 µm from the center of the retina, close to the optic nerve head. We measured ONL width from the retinas of two wild-type, two heterozygous (+ +/Cdh23^v-6J Pcdh15^av-Jfb), and two 4-month-old double homozygous (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb) mutant mice. Eight measurements on each side of the central retina were made using digital photomicrographs taken with a 40X objective.

Serum chemical and hematological analyses

Sera were obtained from six P120 male mice. There were two wild-types, two cis-heterozygotes (+ +/Cdh23^v-6J Pcdh15^av-Jfb), and two double homozygotes (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb), which were analyzed by the Department of Laboratory Medicine at the NIH for albumin, total bilirubin, total protein, alkaline phosphatase, ALT, AST, LDH, BUN, amylase, calcium, lipase, cholesterol, triglycerides, glucose, inorganic phosphorus, creatine kinase, gamma-glutamyl transpeptidase, and uric acid measurements. Hematological parameters measured were hemoglobin, white blood cell count, hematocrit, red blood cell count, mean corpuscular volume (MCV), and mean corpuscular hemoglobin concentration (MCHC). We used single-factor ANOVA to compare body to brain weight, spleen to brain weight, thymus to brain weight, heart to brain weight, liver to brain weight, and kidney to brain weight. No significant differences were found between mutant and controls (data not shown).

Electroretinograms

Electroretinograms (ERGs) were recorded from one 5-month-old and one 9-month-old cis-heterozygote (+ +/Cdh23^v-6J Pcdh15^av-Jfb), a 3-month-old and a 5-month-old homozygote for Cdh23^v-6J that were also heterozygous for Pcdh15^av-Jfb (Cdh23^v-6J +/Cdh23^v-6J Pcdh15^av-Jfb), and a 3-month-old, two 5-month-old and one 9-month-old double-homozygotes (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb). Mice were dark-adapted for 12 h before they were anesthetized by an intraperitoneal injection of 80 mg/kg ketamine and 4 mg/kg xylazine. The pupils were dilated with topical 0.5% tropicamide and 0.5% phenylephrine HCl, and body temperature of the mice was maintained near 38 °C with a heating pad. ERGs were recorded simultaneously from both eyes. This was done by placing gold wire loops on the cornea with a drop of methylcellulose after administration of 1% proparacaine topical anesthetic. Gold wires were also put on the sclera at the limbus as the differential electrodes and a ground wire was attached to the left paw.

Rod pathway function was evaluated by eliciting scotopic ERG responses in the dark-adapted state using single xenon photostrobe flashes (PS33 Photopic Stimulator; Grass Instrument Co., West Warwick, RI) delivered in a Ganzfield light integrating sphere with inter-stimulus intervals of 3 to 60 s depending on stimulus intensity. Stimulus intensities in the range of −6.9 to +0.6 log cd-s/m² were obtained using neutral density filters (Wratten, Eastman Kodak; Rochester, NY). Responses were amplified 5,000X at 0.1–1,000 Hz frequency band (3 dB cut points) using a 60 Hz line filter (CP511 AC amplifier, Grass Instrument Co).

Cone pathway function was evaluated by eliciting photopic responses in a light-adapted state on a rod-suppressing white background of 34 cd/m² using single flashes at 2 s inter-stimulus intervals. For each intensity, 20 responses were averaged. A-waves were measured from the pre-stimulus baseline to the initial trough. B-waves were measured either from the baseline or from the a-wave trough when present. Implicit times were measured from flash onset to the a-wave maximum and the b-wave maximum.

PCR analyses of inner ear and retinal cDNAs

Poly(A)⁺ RNA, isolated from P1-P5 inner ear and eye tissues of 50 C57BL/6J mice using Poly(A)Pure^TM mRNA reagents (Ambion, Austin, TX), and cDNA was prepared using an oligo dT primer and PowerScript^TM reverse transcriptase (Clontech, Palo Alto, CA). Cdh23 (GenBank AY026062) splice variants were amplified using a forward primer (5′-ATC ACA CGG AAG GTG AAT ATC CAA G-3′) in exon 5 and a reverse primer (5′-CAG TCA AGG TGA AAC GCT GGA TAA G-3′) in exon 15. cDNAs were PCR amplified using LA-Taq (Takara Mirus, Madison, WI) by denaturing at 95 °C for 2 min, 30 cycles at 95 °C for 45 s, 60 °C for 30 s and 68 °C for 3 min followed by a 15 min extension at 72 °C. PCR products were cloned and fully sequenced.

Physical and biochemical evaluation

Mice heterozygous for a mutant allele of Cdh23 and Pcdh15 (+ +/Cdh23^v-6J Pcdh15^av-Jfb) are physically indistinguishable from their wild-type or single heterozygous littermates. By comparison, double homozygous mice (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb) are hyperactive, exhibit head tossing and circling behavior and, as expected, are deaf. This phenotype was expected since homozygosity for either mutant allele alone (Cdh23^v-6J or Pcdh15^av-Jfb) results in congenital deafness and vestibular dysfunctions [17,18]. However, there was no additional pathophysiology in the double homozygous mice. With the exception of the inner ear, histological examinations of 32 tissues, a chemistry panel, and hematology tests of serum derived from two wild type, two double cis-heterozygotes (+ +/Cdh23^v-6J Pcdh15^av-Jfb) and two double homozygous (Cdh23^v-6J Pcdh15^av-Jfb/ Cdh23^v-6J Pcdh15^av-Jfb) males revealed no pathological or biochemical abnormalities (data not shown).

Phenotype of the eye

To explore the effect of the mutations on retinal function, we compared ERGs from age-matched mutant mice since increasing age is correlated directly with a decline in b-wave amplitude [46]. Figure 1 shows waveforms from a group of four 5-month-old mice, that included one cis-heterozygote (+ +/Cdh23^v-6J Pcdh15^av-Jfb), one Cdh23^v-6J homozygote that was also a heterozygote for Pcdh15^av-Jfb (Cdh23^v-6J+/Cdh23^v-6J Pcdh15^av-Jfb), and two double homozygotes (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb). These waveforms showed comparable shape, growth of amplitude with intensity, and implicit time for both rod and cone pathway-mediated responses. The double homozygous line (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb) and the double heterozygous line had similar amplitudes suggesting that there was no effect of the additional mutant alleles on retinal responses. The same outcome was observed for 3-month-old and 9-month-old double homozygous mice (data not shown).

We next examined retinal morphology of 5-month-old wild-type (n=2), double heterozygous (+ +/Cdh23^v-6J Pcdh15^av-Jfb; n=2) and double homozygous (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb; n=2) mice. A portion of representative sections taken from the same region of a retina of each of these genotypes is shown in Figure 2. The width of the ONL was measured at 16 points in sections of the mouse retina from wild-type (mean 52.57 µm; SD 0.76), double heterozygotes (mean 54.21 µm; SD 0.03), and double homozygotes for Cdh23^v-6J and Pcdh15^av-Jfb mutant alleles (mean 52.88 µm; SD 0.76; Figure 3). We found no statistical difference (one-way ANOVA, ad hoc Bonferroni, SPSS 16 for Windows, SPSS, Chicago, IL) between the three genotypes.

Hair cell stereocilia bundles

The most prominent histopathological feature observed in the organ of Corti of neonatal (P2) double homozygous mice (Cdh23^v-6J Pcdh15^av-Jfb/Cdh23^v-6J Pcdh15^av-Jfb) is the disruption of the normally compact inner ear hair cell stereocilia bundle morphology (Figure 4). Generally, the hair cell phenotype of mice homozygous for Cdh23^v-6J (Figure 4B) is somewhat less severe than mice homozygous for the Pcdh15^av-Jfb allele (Figure 4C). Hair cell phenotypes of Cdh23^v-6J and Pcdh15^av-Jfb single homozygotes were not modified by the presence of one additional mutant allele of Pcdh15 or Cdh23, respectively, as evaluated by SEM (Figure 4). However, the abnormal hair cell phenotypic features were additive in the double homozygotes (Figure 4D). The apical surface lacked microvilli, the stereocilia bundles were displaced, and the kinocilia were misplaced. In addition, the staircase architecture was also disrupted (Figure 4D), but some links between stereocilia were still present (Figure 4D, inset).

Splice variant of Cdh23

The Cdh23^v-6J allele is a G>T transversion (c.904G>T), which introduces a stop codon (p.E302X) in exon 9 [18]. To determine if there were inframe splice variants of Cdh23 that did not include exon 9, we used a forward primer in exon 5 and reverse primer in exon 13 and PCR amplified inner ear and retina cDNA of wild-type and Cdh23^v-6J mice. In addition to the reported transcript [18], there is an alternative splice variant of Cdh23 that does not include exons 7 to 9 (Figure 5; GenBank EU681829 and EU681830). We previously reported multiple shorter isoforms from mouse inner ear and retina that do not include exons 1–39 of Cdh23 (Figure 5; isoforms b, c) [29]. Thus the Ames waltzer Cdh23^v-6J allele is not deficient for all isoforms of Cdh23 but null only for isoforms that include exon 9.