*P. E. Schauwecker, Department of Cell and Neurobiology, University of Southern California Keck School of Medicine, 1333 San Pablo Street, BMT 403, Los Angeles, CA 90089-9112, USA. E-mail: email@example.com
Inbred strains of mice differ in their susceptibility to excitotoxin-induced cell death, but the genetic basis of individual variation in differential susceptibility is unknown. Previously, we identified a highly significant quantitative trait locus (QTL) on chromosome 18 that influenced susceptibility to kainic acid-induced cell death (Sicd1). Comparison of susceptibility to seizure-induced cell death between reciprocal congenic lines for Sicd1 and parental background mice indicates that genes influencing this trait were captured in both strains. Two positional gene candidates, Galr1 and Mbp, map to 55 cM, where the Sicd1 QTL had been previously mapped. Thus, this study was undertaken to determine if Galr1 and/or Mbp could be considered as candidate genes. Genomic sequence comparison of these two functional candidate genes from the C57BL/6J (resistant at Sicd1) and the FVB/NJ (susceptible at Sicd1) strains showed no single-nucleotide polymorphisms. However, expression studies confirmed that Galr1 shows significant differential expression in the congenic and parental inbred strains. Galr1 expression was downregulated in the hippocampus of C57BL/6J mice and FVB.B6-Sicd1 congenic mice when compared with FVB/NJ or B6.FVB-Sicd1 congenic mice. A survey of Galr1 expression among other inbred strains showed a significant effect such that ‘susceptible’ strains showed a reduction in Galr1 expression as compared with ‘resistant’ strains. In contrast, no differences in Mbp expression were observed. In summary, these results suggest that differential expression of Galr1 may contribute to the differences in susceptibility to seizure-induced cell death between cell death-resistant and cell death-susceptible strains.
Linkage studies using (C57BL/6J X FVB/NJ)N2 mice mapped three susceptibility loci, with the most significant locus, named seizure-induced cell death 1 (Sicd1), to the distal region of mouse chromosome 18 (Schauwecker et al. 2004). To confirm genetic linkage on distal chromosome 18, we created the congenic strain, FVB.B6-Sicd1, in which the relevant donor segment from the resistant C57BL/6J strain was placed on the susceptible FVB/NJ background. The presence of C57BL/6 chromosome 18 alleles on an FVB genetic background conferred protection against seizure-induced cell death, as compared with FVB/NJ parental controls (Lorenzana et al. 2007; Schauwecker et al. 2004). These results suggested that the causal gene(s) influencing susceptibility to seizure-induced cell death may reside in the Sicd1 locus.
As we verified that the Sicd1 locus significantly reduced susceptibility to kainic acid (KA)-induced excitotoxic cell death in the congenic FVB.B6-Sicd1 strain (Lorenzana et al. 2007; Schauwecker et al. 2004), we took a candidate gene approach to identify the causal susceptibility gene(s) responsible for the Sicd1 effect. The genetic variation underlying this quantitative trait locus (QTL) could consist of polymorphisms in either the coding region, thus altering the amino acid sequence of the translated protein, or the regulatory region, affecting expression of a gene. Among the many candidate genes present in the 12-Mb Sicd1 region, galanin receptor 1 (Galr1) and myelin basic protein (Mbp) are the most compelling. Both of these genes approximate the peak position of the Sicd1 QTL at 55 cM (Schauwecker et al. 2004) and have relevance to modulation of neuronal excitability and seizure threshold (Donarum et al. 2006; Jacoby et al. 2002; Mathis et al. 2000; Mazarati et al. 2000, 2006; McColl et al. 2006; Zini et al. 1993b).
In this paper, we describe a systematic gene identification strategy in which we conducted comparative genomic sequencing, expression analyses and comparative cDNA sequencing of two putative candidate genes, Galr1 and Mbp. To examine the possibility that variation in one of these genes could be involved in the susceptibility to seizure-induced excitotoxic cell death, we assessed the expression of mRNA for Galr1 and Mbp in the hippocampus of our two strains and between the FVB.B6-Sicd1 and the B6.FVB-Sicd1 congenic strains and their respective parental background strains, FVB/NJ or C57BL/6J. We also conducted a strain survey to examine the association between the Sicd1 genotype and the susceptibility to seizure-induced cell death among inbred strains of mice.
Materials and methods
Mice and generation of the congenic FVB.B6-Sicd1 and B6.FVB-Sicd1 strains
Male C57BL/6J, FVB/NJ, BALB/cJ, DBA/2J, SJL/J and 129T2/SvEmsJ mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA) at age 6–8 weeks. The congenic FVB.B6-Sicd1 and B6.FVB-Sicd1 strains were generated at the Zilkha Neurogenetic Institute at the University of Southern California Keck School of Medicine, as previously described (Schauwecker et al. 2004). All mice were maintained on a 12-h light/dark schedule with food and water available ad libitum. All experiments were approved by the University of Southern California Institutional Animal Care and Use Committee, in accordance with The National Institutes of Health Guide for the Care and Use of Laboratory Animals.
Quantitative polymerase chain reaction
For RNA isolation, mice were anesthetized with Avertin and killed by decapitation. Whole brains were removed by dissection and both hippocampi were rapidly dissected out on ice. Both combined hippocampi from a mouse whole brain (∼50 mg) were homogenized and total RNA was extracted using the RiboPure kit™ (Ambion, Austin, TX, USA) following manufacturer’s protocols. The quantity and quality of RNA were estimated with spectrophotometric analysis (OD260/280) and RNA was stored at −20°C. Reactions were set up for two-step reverse transcriptase polymerase chain reaction (RT-PCR).
First-strand cDNA was synthesized from total hippocampal RNA with random decamers using Reverse-IT RTase Blend™ (Abgene, Rochester, NY, USA). Relative basal hippocampal transcript levels were measured using the SYBR Green real-time quantitative PCR method following manufacturers’ instructions (Applied Biosystems, Foster City, CA, USA; Roche Applied Sciences, Indianopolis, IN, USA). Primer sequences for Galr1, Mbp and β-actin were designed using primer 3 (Rozen & Skaletsky 2000; summarized in Table 1) based on the murine Galr1, Mbp and β-actin sequences obtained from the Ensembl genome browser (release 36). All primers were purchased from Sigma-Genosys (The Woodlands, TX, USA) as high-performance liquid chromatography purified oligos and were tested to determine whether they produced a single band on an agarose gel. At least two independent replicate experiments were performed for each gene, and all samples were run in triplicate to minimize intra-assay variation. Primers were further evaluated by melting curve analysis of the PCR product to check amplification of a single major peak at a temperature compatible with the amplicon size and base composition.
Table 1. Quantitative real-time PCR primers of Galr1, Mbp and β-actin
Size in bp
For very discrete expression changes, we used the relative standard curve method, which gives highly accurate quantitative results and requires the least amount of validation (Applied Biosystems). Standard curves were prepared for the target (Galr1 or Mbp), and the endogenous control, β-actin, and the quantity of the target was normalized to the quantity of β-actin. β-actin was used as an internal reference gene because it had smaller variability among individual mouse strains than other housekeeping genes, such as glyceraldehyde 3-phosphate dehydrogenase or 18s rRNA. The same pool of standard cDNA was used for generation of standard curves throughout the study to ensure the accuracy of real-time PCR results.
The ratio of target gene mRNA expression for each individual strain as compared with C57BL/6J mice was determined according to the standard curve method. This method determines the changes in the target gene expression relative to changes in the internal standard (β-actin) and corrects for differences in the efficiency of amplification for the primer pairs for each gene. Statistical significance was determined using a one-way analysis of variance (anova) and intergroup differences were analyzed by a Student Newman–Keuls post hoc analysis using the statistical software package, SigmaStat 3.0 (Jandel Scientific, San Rafael, CA, USA). Data are presented as the mean ± SEM, and were considered significant at P < 0.05.
Galr1 and Mbp genomic sequencing
The Galr1 gene has three coding exons spanning 13.3 kb of genomic DNA and gives rise to a 348-amino-acid protein. The Mbp gene has seven coding exons spanning 110.5 kb of genomic DNA and gives rise to a 250-amino-acid protein. Galr1 and Mbp are arranged in a divergent direction separated by 68 kb of genomic sequence on mouse chromosome 18qE, 55 cM. The genomic DNA sequence of the promoter, 5′-untranslated region (UTR), all coding exons, 3′-UTR and splice sites on Galr1 and Mbp was determined in a total of two C57BL/6J and six FVB/NJ mice by PCR amplification of genomic DNA. Briefly, a small piece (∼1 cm) of the tip of the tail of each of the C57BL/6J and FVB/NJ mice was cut off using sharp scissors. Mouse tail genomic DNA was extracted and purified using the DNeasy kit™ (Qiagen, Valencia, CA, USA) following manufacturers’ protocols.
Sequencing primers for Galr1 and Mbp were designed using primer 3 (Rozen & Skaletsky 2000; Tables 2 and 3) based on the murine Galr1 and Mbp sequences obtained from the Ensembl genome browser (release 36). All primers were purchased from Sigma-Genosys) as unpurified desalted oligos. Mouse tail genomic DNA was amplified with gene-specific primers by standard 50-μl PCR reactions (Dieffenbach et al. 1993). Amplification conditions included 35 cycles of denaturation at 94°C for 30 seconds, annealing at the optimal temperature of each primer pair for 30 seconds and extension at 72°C for 50 seconds with initial denaturation at 94°C for 3 min and final extension at 72°C for 3 min. The resulting PCR products were resolved on 2% ethidium bromide-stained agarose gels and the purity of amplified PCR fragments was verified as a single sharp band. Bands of appropriate size were excised with a razor blade, purified and extracted using QIAEX II Gel Purification kits™ (Qiagen) according to manufacturers’ instructions.
Table 2. Sequencing primers of Galr1 genomic DNA
Genomic DNA region
Size in bp
Table 3. Sequencing primers of the Mbp genomic DNA
Genomic DNA region
Size in bp
Both Galr1 strands of each amplicon were sequenced using the corresponding forward and reverse PCR primers and 806 bp of the Galr1 promoter region was sequenced using a TOPO TA cloning kit for sequencing (Invitrogen) following manufacturers’ instructions. The 806-bp amplicon was gel purified and subcloned into pCR 4-TOPO (Invitrogen) before sequencing. DNA sequencing was performed by the Norris Cancer Facility at University of Southern California (USC) on an Applied Biosystems 3730 DNA analyzer by the Big Dye Terminator Cycle Sequencing kit, version 3 (Perkin-Elmer Biosystems, Foster City, CA, USA). Sequencing results were manually checked to eliminate any possible miss-calls and then analyzed using T-Coffee version 2.00 (Notredame et al. 2000) to align DNA sequences between the C57BL/6J and the FVB/NJ mouse strains.
Hippocampal Galr1 cDNA sequencing
The cDNA prepared from hippocampal RNA of 6- to 8-week-old male C57BL/6J and FVB/NJ mice was used for mutation screening. The primer pair amplifying the 1904-bp full-length Galr1 transcript was designed using primer 3 (Rozen & Skaletsky 2000; summarized in Table 4) based on the murine Galr1 sequence obtained from the Ensembl genome browser (release 36). Standard 50-μl PCR reactions (Dieffenbach et al. 1993) were performed on 10 μl of first-strand hippocampal cDNA synthesis reaction using the primer pair amplifying the 1904-bp full-length hippocampal Galr1 transcript. Amplification conditions included 40 cycles of denaturation at 94°C for 45 seconds, annealing at 57°C for 45 seconds and extension at 68°C for 2 min 30 seconds, with initial denaturation at 94°C for 3 min and final extension at 68°C for 15 min.
Table 4. Reverse transcriptase primer pair of full-length hippocampal Galr1 cDNA
The purity of amplified PCR fragments was verified as a single sharp band on an ethidium bromide-stained agarose gel. Bands of 1904 bp were excised with a razor blade, purified and extracted using the QIAEX II™ gel purification kit (Qiagen) according to manufacturers’ instructions. The resulting 1904-bp amplicon was subject to direct sequencing using primers summarized in Table 5 to detect any potential aberrant splicing and/or any potential amino acid polymorphism(s). Complementary DNA sequencing was performed on an ABI 3730 DNA analyzer by cycle sequencing using 3′-fluorescent-labeled dideoxynucleotides (dye terminator chemistry). All chromatogram data were checked manually to eliminate any possible miss-calls and sequencing results were analyzed using T-Coffee (Notredame et al. 2000) to align DNA sequences between the two mouse strains.
Table 5. Classification of seizure parameters in inbred strains of mice after systemic administration of KA
Seizure parameters, percentage of mice
Rearing and falling
Duration of seizures (h)
While KA induced a similar level of stage irrespective of mouse strain (F = 0.719; P = 0.621), a significant strain-dependent difference in the duration of severe seizures was observed (*P < 0.05).
Adult male C57BL/6J, FVB/NJ, BALB/cJ, DBA/2J, SJL/J and 129T2/SvEmsJ, homozygous FVB.B6-Sicd1 and homozygous B6.FVB-Sicd1 congenic mice (6–8 weeks old) were used in these studies. Kainic acid (Diagnostic Chemical, Ltd, Charlottetown, PEI, Canada) was dissolved in isotonic saline (pH 7.3) and administered s.c. to adult male mice. Preliminary dose response studies had defined seizure thresholds and showed consistent seizures among all six inbred mouse strains and both congenic strains, with a mortality of less than 25% with a dose of 25 mg/kg, s.c. (Schauwecker & Steward 1997). Kainic acid solutions were prepared fresh on the day of each experiment. Following KA injections, mice were monitored every 15 min for 4 h to determine seizure parameters: the onset, the duration and percentage of mice achieving each seizure stage according to the Racine classification of seizure stages (Racine 1972).
In order to evaluate the severity of KA-induced excitotoxic brain damage in different strains of mice, brains from each strain of mice were processed for light microscopic histopathologic evaluation according to previously published methods (Schauwecker et al. 2004). Briefly, 7 days after seizure induction by KA, mice were anesthetized with Avertin and transcardially perfused with 4% paraformaldehyde in 0.1 m phosphate buffer (pH 7.4). Brains were removed and post-fixed in 30% sucrose for at least 12–18 h for cryoprotection. Horizontal (35 μm) sections were cut on a sliding microtome and immersed in phosphate buffer (pH 7.4); free floating until histological processing was started. Every sixth section (∼210 μm) was processed for cresyl violet staining to assess cell loss and morphology. An alternate series of sections were stained with a modified Gallyas silver stain, which stains degenerating fibers, synaptic terminals and cell bodies (Nadler & Evenson 1983 as modified in Schauwecker 2003) and examined for the appearance of degenerative debris. An additional series of sections were stained with Fluoro-Jade B, a fluorescent marker for dying neurons, according to the method outlined previously (Schmued & Hopkins 2000).
Morphological assessment of neuronal damage
The number of degenerating neurons in both the right and the left hippocampus from every sixth section (240-μm separation distance) in four brain regions (CA3, CA1, dentate hilus and dentate gyrus), which represented various levels of the hippocampus, was visually estimated and a histological damage score was assigned on a 0–3 grading scale according to the following criteria: grade 0, absence of pyknotic cells; grade 1.0, mild (<25% of cells pyknotic); grade 2.0, moderate (<50% of hippocampal neurons pyknotic) and grade 3.0, extensive (>50% of cells pyknotic) according to a previously defined scale (Fujikawa 1995, 1996; Fujikawa et al. 1994; Schauwecker et al. 2004). All grading was performed blindly by an observer who was naïve to the strain.
There was no obvious difference in neuronal damage between hemispheres, so values for right and left hemispheres were averaged for each mouse. For the hippocampus, scores from sections were averaged and used for calculating group values. As histological damage scores were normally distributed, we were able to use standard parametric methods of data analysis. Thus, to determine whether differences in histological scores existed among the groups of mice, results were assessed statistically by one-way anova with the computer program SigmaStat version 3.00 (Jandel Scientific) and intergroup differences were analyzed by Student Newman–Keuls post hoc test.
Neuronal loss quantification
We counted cells in defined areas of CA1, CA3, the dentate hilus and the dentate gyrus in a blinded manner using unbiased stereological methods on cresyl violet-stained sections as described (Schauwecker & Steward 1997; Schauwecker et al. 2000). The Nissl-stained neurons in area CA3, area CA1, the dentate hilus and the dentate gyrus were counted in both the right and the left hippocampus and counting was initiated within the ventral hippocampus at the first point where hippocampal subfields could be easily identified. This level corresponded to horizontal section 54, based on the atlas of Sidman et al. (1971). Hippocampal subfields were based on Franklin and Paxinos (1997) classification and discrimination between the CA3 and the dentate hilus region was based on morphological features and locations of the cells (Sousa et al. 1998; West et al. 1991). Specifically, for dentate hilar cell counts, the hilus was operationally defined as the region bordered by the supra- and infrapyramidal granule cell layers and excluding the densely packed pyramidal neurons of area CA3.
Neuron counts were made in all subfields and the numbers for each side were averaged into single values for each animal. Surviving cells were counted only if they were contained within the pyramidal cell layer, dentate hilus or dentate gyrus, possessed a visible nucleus and characteristic neuronal morphology and had a cell body larger than 10 μm. Six square counting frames (200 × 200 μm) were randomly placed in the pyramidal layer of fields CA1 and CA3 or in the dentate hilus or dentate gyrus in 4–5 regularly spaced horizontal sections from each animal. Neuronal nuclei were evaluated at three different focal planes and only those in the focal plane were counted with a × 40 objective and considered as a counting unit. Stereological analysis was performed with the aid of Image-Pro Plus software version 4.00 (Media Cybernetics, Inc., Silver Spring, MD, USA) in combination with a SPOT digital camera (Diagnostic Instruments, Inc., Sterling Heights, MI, USA) and a motorized Z-stage (Optiscan, Prior Scientific, Fairfax, VA, USA). Final cell counts are expressed as the percentage of cells as compared with intact mice. Results were assessed statistically by one-way anova using the computer program, SigmaStat version 3.00 (Jandel Scientific), and intergroup differences were analyzed by the Student Newman–Keuls post hoc test. Data were considered significant at P < 0.05.
Phenotypic analysis of susceptibility to seizure-induced cell death in Sicd reciprocal congenic strains
In order to further confirm the chromosome 18 QTL, termed Sicd1, reciprocal congenic lines were constructed for the Sicd1 interval (D18Mit141–D18Mit25). For FVB.B6-Sicd1 congenic mice, we introgressed the interval containing the putative resistant B6 alleles onto the susceptible FVB background. In contrast, for the B6.FVB-Sicd1 congenic mice, we introgressed the interval containing the putative susceptibility FVB alleles onto the resistant B6 background. Phenotypic analysis of the extent of seizure-induced cell death in both strains confirmed that a gene(s), located between the D18Mit141 and the telomere on chromosome 18, plays an intrinsic role in susceptibility to seizure-induced cell death. This conclusion is supported by data showing the dramatic reduction in susceptibility to seizure-induced cell death with the FVB.B6-Sicd1 strain relative to the FVB background strain (Fig. 1; Lorenzana et al. 2007), and the significant increase in susceptibility to seizure-induced cell death within the B6.FVB-Sicd1 strain relative to the B6 background strain (Fig. 1).
Prioritization of candidate genes
As we were able to verify that the Sicd1 QTL significantly affected susceptibility to seizure-induced cell death in our congenic FVB.B6-Sicd1 strain (Schauwecker et al. 2004), we sought to identify candidate genes within the Sicd1 interval (D18Mit141–D18Mit25) that might explain this effect. The peak LOD (D18Mit186–D18Mit4) of Sicd1 covers about 13 Mb (the physical interval 72–85 Mb). However, the region defined by a one LOD drop (the 99% confidence interval for the true QTL location) is about 5 Mb (the physical interval 79–84 Mb) (Schauwecker et al. 2004). Analysis of the Sicd1 gene content indicates that this reduced 5-Mb region contains approximately 12 genes that show homology with the human genome (The Build 36 assembly by National Center for Biotechnology Information (NCBI); Fig. 2). Among the 12 genes within this region, we identified two potential Sicd1-encoded candidate genes, Galr1 and Mbp, based on the following criteria: (1) physically residing in the Sicd1 locus, (2) detected as expressed in the hippocampus as shown in the Allen Brain Atlas (http://www.brain-map.org/aba/mouse/brain/Galr1.html; http://www.brain-map.org/aba/mouse/brain/Mbp.html), (3) human homology and (4) correlated with hippocampal excitability and/or hyperexcitability.
Galr1 mRNA is differentially expressed in congenic and background strains
We hypothesized that strain-specific genetic variation, through its effects on gene expression, in one or more genes between C57BL/6J (B6) and FVB/NJ (FVB) is responsible for the Sicd1 effect. To investigate further whether variation in Galr1 expression determines susceptibility to KA-induced excitotoxic cell death, we measured the basal hippocampal transcript abundances of Galr1 to determine if variation in Galr1 expression is associated with susceptibility to KA-induced excitotoxic cell death. We used quantitative real-time PCR (qRT-PCR) because Galr1 is expressed at a relatively low level in the hippocampus. We found a significant difference in basal levels of Galr1 transcripts that were ∼40% less in a cell death-resistant B6 than in a cell death-susceptible FVB hippocampi (Fig. 3a).
Similarly, in homozygous FVB.B6-Sicd1 congenic and parental background FVB mice, quantitative PCR verified the strain-specific pattern of Galr1 gene expression with respect to the B6 congenic segment. Specifically, Galr1 mRNA expression was downregulated (∼45%) in congenic hippocampus as compared with FVB (Fig. 3b). In contrast, quantitative PCR verified that Galr1 mRNA expression was significantly increased in the hippocampus of B6.FVB-Sicd1 congenic mice as compared with C57BL/6J mice. A clear relationship was found: congenic FVB.B6-Sicd1 with the B6 allele at Galr1 showed significantly lower levels of hippocampal Galr1 expression than control FVB littermates, whereas congenic B6.FVB-Sicd1 with the FVB allele at Galr1 showed significantly higher levels of hippocampal Galr1 expression than control B6 littermates.
Inbred strains are differentially susceptible to KA-induced excitotoxic cell death
We previously found that certain inbred mouse strains show marked differences in susceptibility to KA-induced excitotoxic cell death (Schauwecker & Steward 1997). In the present study, we assessed two additional strains, SJL/J and DBA/2J for susceptibility to KA-induced excitotoxic cell death. Preliminary dose response studies defined seizure thresholds and showed consistent seizures among all six inbred mouse strains, with a mortality of less than 25% when a dose of 25 mg/kg was administered s.c. (Schauwecker & Steward 1997). Systemic administration of KA induced a characteristic sequential behavioral response: stage 1, immobility; stage 2, forelimb and/or tail extension, rigid posture; stage 3, repetitive movements, head bobbing; stage 4, rearing and falling; stage 5, continuous rearing and falling; stage 6, severe tonic-clonic seizures. Table 5 summarizes the percentage of animals achieving each seizure stage (1–5) and the duration of severe seizures for all six strains. Severe seizures (stages 4 and 5) lasted on average 1.6 h and no significant strain-dependent effects were found with regard to qualitative differences in seizure intensity or the percentage of animals achieving stage 5 seizures among the representative strains. However, a significant strain-dependent difference in the duration of stage 5 seizures was observed in DBA/2J (cell death-susceptible) and in SJL/J (cell death-resistant) strains as compared with the other strains.
We assessed KA-induced excitotoxic neuronal death in the six strains of mice by light microscopic histopathologic evaluation. Systemic administration of KA induced the degeneration and reduction of hippocampal neurons in area CA3, the dentate hilus and sporadic neuronal reduction in area CA1 in susceptible strains (Fig. 3c–e; FVB/N, DBA/2J and 129T2/SvEmsJ). Consistent with previous studies (Ben-Ari et al. 1984; Nadler & Cuthbertson 1980; Nadler et al. 1980; Sperk et al. 1983), neurons within the dentate granule cell layer and area CA2 were spared. In contrast, three representative resistant strains (C57BL/6J, BALB/cJ and SJL/J) showed no detectable reduction of neurons within the hippocampus proper and no indication was noted of damage to neuronal nuclei in any hippocampal region or in the septum, amygdala, pyriform cortex, neocortex or thalamic nuclei in cresyl violet-stained sections (Fig. 4). In addition, sections from susceptible mice that were processed for the Gallyas silver stain for degeneration displayed intense argyrophilic deposits within the stratum oriens and stratum pyramidale of the CA3 subfield and within the dentate hilus. In contrast, those strains resistant to excitotoxic cell death displayed no detectable evidence of degenerative debris or reduction of neurons in any of the hippocampal subfields after KA administration.
To confirm these results, we performed quantitative analyses of hippocampal neuron numbers. The susceptible strains (FVB/NJ, DBA/2J and 129T2/SvEmsJ) showed a reduction on average of 43% of dentate hilar neurons, 70% of CA3 pyramidal neurons and 38% of CA1 pyramidal neurons 7 days after KA administration as compared with wild-type control injected with the same amount of saline (F(1,5) = 66.413; P < 0.001; Fig. 5). These results indicated that KA induces differential vulnerability of neurons in the hippocampus in a strain-dependent manner although KA-induced seizures are qualitatively comparable.
Haplotype analysis of hippocampal Galr1 mRNA expression
To confirm the hypothesis that differences in susceptibility to seizure-induced cell death are associated with variation in the expression of Galr1, we conducted haplotype analyses on a total of six inbred mouse strains, divergent for the phenotype of seizure-induced cell death to determine whether variation in Galr1 expression observed among FVB and B6 strains was conserved as a whole in other inbred mouse strains. Quite clearly, we found that the representative cell death-resistant strains (C57BL/6J, BALB/cJ and SJL/J) showed significantly lower hippocampal Galr1 expression levels than did representative cell death-susceptible strains (FVB/NJ, DBA/2J and 129T2/SvEmsJ) (Fig. 6). Thus, we confirmed and extended the results observed among FVB and B6 strains.
Sequencing and identification of single-nucleotide polymorphisms in Galr1
As the Galr1 locus may be associated with susceptibility to KA-induced excitotoxic cell death in mice, we searched for underlying sequence variation of the Galr1 gene between B6 and FVB strains. To look for variation that could predict a functional consequence, the coding region of Galr1 was sequenced from the B6 and FVB inbred strains. We found no sequence differences in the coding region of Galr1. Similarly, no identified sequence alterations were found between the FVB and the B6 inbred strains within any of the 5′- or 3′-regulatory regions, splice sites or promoter region (Fig. 7).
Hippocampal Mbp mRNA expression and sequence analyses
Mbp is widely distributed in a number of tissues, but available data show some of the highest expression in brain (Mathisen et al. 1993). Using quantitative PCR, we found that Mbp was not differentially regulated in the hippocampus of either FVB or B6 or homozygous FVB.B6-Sicd1 congenic mice (Fig. 8a,b). Although we found no significant strain-dependent difference in basal expression for Mbp, genetic variation can affect the function of the gene product, and we searched for sequence variation of the Mbp gene between B6 and FVB strains as well. We manually sequenced the promoter, 5′-UTR, all seven coding exons, 3′-UTR, and splice sites of Mbp on B6 and FVB genomes. Similar to our results for Galr1, we found no sequence variation in the promoter, 5′-UTR, all coding exons, 3′-UTR and splice sites for Mbp (Fig. 9).
The cDNA sequence analysis of Galr1
Any critical DNA sequence change(s) underlying Sicd1 could be a result of changes in the protein-coding region that change the encoded protein or of changes in the regulatory region of the gene that influence expression of the gene. Here, we sequenced the protein-coding region of Galr1 using the B6 and FVB progenitor strains to detect any potential amino acid polymorphism(s) caused by nuclear post-transcriptional RNA editing (Fig. 10). We PCR amplified the whole 1904-bp amplicon of full-length Galr1 cDNA and subsequently subjected it to sequencing in C57BL/6J and FVB/NJ strains. No variation in cDNA sequence was observed between the two inbred strains examined.
Phenotyping studies of a new congenic strain developed in our laboratory have confirmed the significant genetic influence of a chromosome 18 QTL: Sicd1 for susceptibility to seizure-induced cell death (Schauwecker et al. 2004). To validate the Sicd1 effect and to move towards identification of the quantitative trait gene(s) for Sicd1, we generated reciprocal congenic lines of mice. These strains were differentially susceptible to seizure-induced cell death because they had received a chromosomal donor region from either FVB/NJ that contained the susceptible Sicd1 alleles (B6.FVB-Sicd1) or C57BL/6J that contained the resistant Sicd1 alleles (FVB.B6-Sicd1). These results give additional support for Sicd1 containing a gene (or a set of closely linked genes) contributing to seizure-induced cell death.
Because these congenic strains retained the phenotype of the original linked QTL, we hypothesized that candidate genes responsible for conferring susceptibility to seizure-induced excitotoxic cell death were present within the identified 12-Mb Sicd1 interval on chromosome 18. The mechanism(s) underlying phenotypic strain differences in susceptibility to seizure-induced cell death could involve genes affecting hippocampal excitability, hyperexcitability and glutamate release. Therefore, genes encoding proteins involved in these processes are likely candidates for modifying susceptibility to KA-induced excitotoxic cell death. Two such genes in the reduced Sicd1 interval satisfied this functional criteria, Galr1 and Mbp, and were subjected to candidate gene analysis. Combining the gene expression data generated in the congenic FVB.B6-Sicd1 strain with the association data of inbred strains of mice segregating for the trait of susceptibility to KA-induced excitotoxic cell death, we were able to identify Galr1 as a putative causal susceptibility gene for KA-induced excitotoxicity. In contrast, we did not find any evidence to support Mbp as a causal susceptibility gene for KA-induced excitotoxicity.
It was hypothesized that the gene(s) producing the difference in response to kainate-induced cell death between B6 and FVB mice would be differentially expressed in the hippocampus of untreated control mice. We were especially interested in the Galr1 and Mbp candidate genes because of their published associations with modifying hippocampal excitability and modifying hyperexcitability (Ben-Ari 1990; Donarum et al. 2006; Mazarati et al. 2000; Noveroske et al. 2005). To confirm the identity of any putative candidate genes, expression studies were performed using qRT-PCR, and any gene expression differences observed would presumably be the result of changes in allele frequencies because of selection rather than a response to kainate treatment. We found that B6 mice showed reduced levels of hippocampal Galr1 mRNA as compared with FVB mice. Second, the congenic FVB.B6-Sicd1 strain carrying the B6-derived resistant allele at the Sicd1 region, where Galr1 is located, on the FVB genetic background showed significantly decreased basal hippocampal Galr1 expression compared with the susceptible FVB control littermates. In contrast, the congenic B6.FVB-Sicd1 strain carrying the FVB-derived susceptible allele at the Sicd1 region on the B6 genetic background showed significantly enhanced basal hippocampal Galr1 expression compared with the resistant B6 control littermates.
We speculate that the downregulation of Galr1 observed in B6 and FVB.B6-Sicd1 mice supports its role as a modulator of neuronal excitability in the hippocampus (Ben-Ari 1990; Zini et al. 1993a,b). This speculation is supported by two lines of evidence suggesting that the neuropeptide galanin is a powerful regulator of seizure activity and neuronal excitability. Previous studies have showed that acute administration of galanin receptor agonists can effectively attenuate convulsive activity induced by the pro-convulsant picrotoxin (Mazarati & Lu 2005) and can delay kindling epileptogenesis (Mazarati et al. 2006). Second, loss-of-function experiments have showed enhanced susceptibility to excitotoxin-induced neuronal injury in Galr1 knockout mice (Mazarati et al. 2004), and a recent study by McColl et al. (2006) show Galr1 knockout mice exhibit spontaneous partial seizures with impaired synaptic inhibition in the hippocampus. In contrast, gain-of-function experiments have showed that galanin overexpression is known to decrease hippocampal neuronal injury resulting from limbic seizures (Haberman et al. 2003; Mazarati et al. 2000) presumably through Galr1 receptor modulation (Mazarati & Lu 2005). Furthermore, the fact that our reciprocal congenic strains were differentially susceptible to seizure-induced cell death would suggest that differences in the level of Galr1 expression alone may be sufficient to confer enhanced susceptibility to KA-induced neuronal death on a genetically resistant background, as was the case for the B6.FVB-Sicd1 congenic strain. Thus, taken together, our findings support the observation that Galr1 can regulate neuronal excitability in the hippocampus.
We predicted that the gene, or regulatory region of the gene, producing the difference in response to kainate between B6 and FVB mice would be polymorphic between these two strains. Thus, we documented the sequence variation between these strains for the putative candidate genes, Galr1 and Mbp. Sequence analysis showed no sequence variation in the promoter, 5′-UTR, coding exons, 3′-UTR or splice sites for Galr1 or Mbp between B6 and FVB strains. While sequence analysis of Galr1 detected no polymorphisms in the regions analyzed, it is important to note that SNPs in noncoding sequences may also affect the levels and forms of mRNA transcripts. However, based on a survey of SNPs identified in the Perlegan mouse database (http://mouse.perlegen.com/mouse/) in conjunction with the Center for Rodent genetics (http://www.niehs.nih.gov/crg/) and for intronic regions of Galr1 and Mbp, there is no evidence that an SNP in a noncoding region exists between FVB and B6. Second, while sequence analysis of Galr1 detected no polymorphisms, we cannot rule out Galr1 as a putative QTL gene. For example, Galr1 could be differentially regulated by an upstream mechanism. The downregulation of Galr1 expression in the hippocampus of B6 and FVB.B6-Sicd1 mice indicates a positive association between Galr1 expression and reduced susceptibility to excitatory amino acid-induced cell death. Whether or not changes in Galr1 activity are involved in producing the phenotype of reduced susceptibility remains to be determined. The mechanism by which Galr1 could influence susceptibility in this mouse model is not yet known, but could result from changes in the regulatory region of the gene, as suggested by the observed strain differences in expression.
While we did not find any evidence to support Mbp as a causal susceptibility gene for KA-induced excitotoxicity, there is still a possibility that other gene(s) within the Sicd1 region are involved as well. Of the 12 known genes in the region defined by the one LOD drop (79–84 Mb), five of these have been published on. For example, a nucleotide substitution in the gene encoding Ctdp1 results in an autosomal recessive disorder called congenital cataracts facial dysmorphism neuropathy (Varon et al. 2003). Nfatc1 is considered as the master transcription factor for osteoclasts – the bone resorbing cells that play a key role both in the normal bone remodeling and in the skeletal osteopenia of arthritis, osteoporosis, periodontal disease and certain malignancies (Sato et al. 2007; Zhao et al. 2007). Sall3 is a member of the Spalt gene family that encodes putative transcription factors. These Spalt homologues are widely expressed in neural and mesodermal tissues during early embryogenesis. Sall3 is required for the development of nerves that are derived from the hindbrain and for the formation of adjacent branchial arch derivatives (Parrish et al. 2004) and has also been showed to be an epigenetic hotspot of aberrant DNA methylation (Ohgane et al. 2004). Setbp1 is a protein that binds to the acute undifferentiated leukemia-associated protein, SET. SET is thought to play a key role in leukemogenesis by its nuclear localization and protein–protein interactions. While the function of Setbp1 remains unknown, it has been proposed to play a key role in the mechanism of SET-related leukemogenesis and tumorigenesis (Minakuchi et al. 2001). Lastly, Pard6g (Par6) is a key member of a multicomponent polarity complex that controls a variety of cellular processes, such as asymmetric cell division, establishment of epithelial apico-basal polarity and polarized cell migration (Bose & Wrana 2006; Yoshimura et al. 2006). While none of these genes, at present, is thought to play a relevant role in modulating neuronal excitability on its own, we cannot exclude that some of these genes, which play important roles in cellular-signaling pathways might be involved subsequently at some level. However, at present, we can only provide evidence that Galr1 is at least one of the causal susceptibility genes for KA-induced excitotoxic cell death in mice.
In summary, our results identify Galr1 as a promising candidate gene, but additional work is necessary to establish with certainty that Galr1 is a seizure-induced cell death susceptibility gene. Until then, it must be kept in mind that another gene within the QTL interval in linkage disequilibrium with Galr1 may ultimately be shown to contribute all or part of the QTL effect. Importantly, our congenic model and prospective sublines will be used to narrow and refine the differential locus on chromosome 18, as well as to examine gene interactions and subphenotypes in the control of excitatory amino acid-induced cell death susceptibility, with the ultimate goal of identifying the set of genes responsible for this complex trait. Validation of Galr1 as causal for the trait of susceptibility to seizure-induced cell death will involve the construction of animals that are genetically altered with respect to Galr1 activity followed by screens for variations in the trait of susceptibility to seizure-induced cell death.
This work was supported by NIH NS38696 to Paula Elyse Schauwecker. We thank Dr Wayne Frankel and Dr Thomas McCown for thoughtful comments on the manuscript and helpful discussions.