|
|
Proc Natl Acad Sci U S A. 2000 September 26; 97(20): 11038–11043. | PMCID: PMC27144 |
Copyright © 2000, The National Academy of Sciences Neurobiology Regional and strain-specific gene expression mapping in the adult
mouse brain Rickard Sandberg,*† Rie Yasuda,†‡ Daniel G. Pankratz,* Todd A. Carter,* Jo A. Del Rio,§ Lisa Wodicka,§ Mark Mayford,‡ David J. Lockhart,§ and Carrolee Barlow*¶ *The Salk Institute for Biological Studies, The Laboratory of Genetics, 10010 North Torrey Pines Road, La Jolla, CA 92037; ‡Department of Neurosciences, University of California, San Diego, La Jolla, CA 92093-0691; and §Genomics Institute of the Novartis Research Foundation, 3115 Merryfield Row, San Diego, CA 92121 Received June 6, 2000. To determine the genetic causes and molecular mechanisms
responsible for neurobehavioral differences in mice, we used highly
parallel gene expression profiling to detect genes that are
differentially expressed between the 129SvEv and C57BL/6 mouse
strains at baseline and in response to seizure. In addition, we
identified genes that are differentially expressed in specific brain
regions. We found that approximately 1% of expressed genes are
differentially expressed between strains in at least one region of the
brain and that the gene expression response to seizure is significantly
different between the two inbred strains. The results lead to the
identification of differences in gene expression that may account for
distinct phenotypes in inbred strains and the unique functions of
specific brain regions. Keywords: seizure, C57BL, 6, 129SvEv, oligonucleotide
array, amygdala Neurobehavioral studies have
advanced substantially through the use of mouse genetics. Many studies
have shown that inbred mouse strains exhibit significant variation in
several central nervous system (CNS) phenotypes. For example, despite
similar seizure susceptibility, inbred strains exhibit large
differences in neuronal cell death after seizures ( 1) vary greatly in
their behavioral response to drugs of addiction, and show marked
differences in behavioral testing (for a review see ref. 2). With the
advent of highly parallel gene expression studies using DNA arrays
( 3– 5), it is now possible to ask the questions: what is the
interacting set of genes that account for the differences between
inbred mouse strains and which genes are responsible for the unique
structures and functions of specific brain regions? We have applied
gene expression profiling of multiple brain regions in two commonly
used inbred strains that differ in their neurobehaviorial phenotypes,
the 129SvEv and C57BL/6 strains (for a review see ref. 6 and for
revised nomenclature of 129 strains see ref. 7). We determined the number and pattern of genes that are differentially
expressed in multiple brain regions in these strains of mice and in
response to seizure. Animal Use and Tissue Collection. All animal procedures were performed according to protocols approved by
The Salk Institute for Biological Studies Animal Care and Use
Committee. Male C57BL/6 and 129SvEv mice were purchased from
Taconic Farms at an age of 7 weeks and housed individually for 1 week
before death. Two samples were prepared and analyzed from different
mice for each strain. For animals used in the seizure analysis,
pentylenetetrazol solution was administered s.c.
at a dose of 50 mg/kg. All animals had a similar seizure
response as assessed by using standard criteria ( 8). Animals were
killed 60 min after seizure. Dissections were done between 14.00–17.00
h on wet ice covered with parafilm. Cortical dissections included the
entire cortex except olfactory bulbs. The midbrain consists of the
brain dissected free of cortex, pons and medulla. Cortex, cerebellum,
midbrain, and hippocampus were prepared in duplicate from two different
mice of each strain. To obtain sufficient tissue from amygdala and
entorhinal cortex, the microdissected regions of seven animals were
pooled. Dissected tissue was directly frozen on dry ice and stored at
−80°C. Mouse embryonic fibroblasts were prepared for each strain
according to standard protocols from six embryos at day 13.5 ( 9). RNA Preparation/Northern Blot Analysis. Tissues were placed into TRIzol (GIBCO/BRL) (added to the
frozen tissues at approximately 1 ml per 100 mg tissue) and homogenized
(Polytron, Kinematica, Lucerne, Switzerland) at maximum speed for
90–120 sec. Subsequent steps were done according to the
manufacturer's instructions. Labeling of samples, hybridization, and
scanning were performed as described ( 4). Northern blot analysis was
performed by using 20 μg of total RNA, and probes were derived from
random priming of 500–700 base pair fragments derived from expressed
sequence tags available from I.M.A.G.E. consortium. Blots were scanned
with a PhosphorImager (Molecular Dynamics). Gene Expression Analysis. Two different arrays (GeneChip, Affymetrix, Santa Clara, CA) were used
that together represent 13,069 probe sets corresponding to more than
10,000 genes and expressed sequence tags (Mu11KsubA and Mu11KsubB).
Data analysis was performed by using genechip version 3.1
(Affymetrix) and nfueggo 2.1 c
(Lockhart and Lockhart, San Diego). We used the genechip
software global scaling algorithm to compare all 24 samples (48 total
arrays, 24 SubA and 24 SubB arrays). We scaled all samples to a target
intensity of 200. A target intensity of 200 has been shown to
correspond to ≈3–5 transcripts per cell ( 4). All strain variation analyses were done by comparing
C57BL/6 to 129SvEv. To generate data for Fig.
and Table
1, all C57BL/6 samples were
compared with all 129SvEv samples (24 experiments, six regions per
strain prepared and performed in duplicate used to generate 12
comparison files). The criteria used to detect differences in gene
expression were a fold change of 1.8 or greater and a difference call
(as described in the genechip software) of increase,
marginal increase, decrease, or marginal decrease and a signal change
greater than 50 in 8/12 comparisons. For data in Table 3, which
is published as supplemental data on the PNAS web site, www.pnas.org,
the duplicate C57BL/6 samples were compared with those from the
corresponding duplicate 129SvEv brain region and the same criteria were
used (for 3/4 comparisons). To determine expression differences
in response to seizure the same criteria (for 4/4 comparisons)
were used.
| Table 1Strain-specific variation across all brain
regions |
Brain Region Gene Identification. To identify genes with region-restricted expression patterns, genes
were classified as present in a region if the gene had a call of
present in at least three of four samples. Similarly, to classify genes
as clearly not detected, we used a call of absent in four of four brain
samples (absent or expression at levels below the threshold of
detection). The signal from one brain region was compared with all
other brain regions and genes with significant differences were
included (P < 0.05 by using a Student's t
test). These data were used to generate Venn diagrams representing
overlapping and nonoverlapping gene expression patterns (see Fig. ). To detect region-specific variation (both restriction and enrichment),
the standard criteria above were used, with the additional criterion
that the gene must be scored as present in at least 80% of the
comparisons (e.g., in comparison of amygdala to cerebellum, midbrain,
hippocampus, entorhinal cortex, 14 of the total 16 samples had to be in
agreement). Genes were classified as ( i)
restricted/highly enriched if they were called absent in all
other regions, ( ii) enriched if detected in all other
regions but with higher levels in the region in question,
( iii) decreased if detected in all other regions but lower
in the region in question, and ( iv) not detected if scored
as absent in all four samples but present in another region. Note that
the number of genes in Table 2 is less
than the number represented in the Venn diagrams in the restricted and
absent categories because stricter criteria were used. However, all
genes identified in Table 2 also were identified in the analysis used
to generate the Venn diagrams.
| Table 2Genes with restricted expression or not detectable in
specific brain regions |
Gene Expression Differences Between C57BL/6 and 129SvEv
Mice. Gene expression profiles were measured for multiple brain regions
in two different mouse strains (C57BL/6 and 129SvEv). The
regions studied were cortex, hippocampus, amygdala, entorhinal cortex,
midbrain, and cerebellum plus passage 1 mouse embryonic fibroblasts
(MEFs). In total, 24 samples from six brain regions and four samples
from MEFs were analyzed. Duplicate samples were prepared from different
animals from each region for each strain. Of the 13,069 probe sets
analyzed, 7,169 (55%) gave a hybridization signal consistent with a
call of present (refs. 3 and 4 and Materials and Methods) in
at least one brain region. This finding indicates that at least 55% of
the genes covered on the murine arrays are detected in one or more
areas of the adult male mouse brain. To estimate experimental reproducibility within a strain and
brain region, the number of genes that scored as differentially
expressed in comparisons of all duplicate brain samples from the same
strain was determined. On average, only two genes of 13,069 (0.017%)
met the criteria as differentially expressed in replicate measurements
(see Materials and Methods). To determine which genes were
differentially expressed between C57BL/6 and 129SvEv mice, all
C57BL/6 brain samples were compared with all corresponding
129SvEv samples by using similar conservative criteria. Twenty-four
genes were identified that were differentially expressed in all six
brain regions of C57BL/6 compared with 129SvEv (see Fig. and
Table 1). We found that the mRNA for the murine leukemia virus gene
(derived from the endogenous retrovirus isolated from a
C57BL/6-derived cDNA library) was detected only in
C57BL/6 ( 10), as expected. In addition, the mRNA abundance for
the Gas5 gene was lower in the 129SvEv strain. It is known
that the Gas5 gene in 129 strains harbors mutations that
alter mRNA stability ( 11): this stability difference likely accounts
for the ≈2-fold decrease in steady-state mRNA abundance in 129SvEv
compared with C57BL/6. These findings validate the approach for
identifying biologically relevant gene expression differences. We next determined which genes were differentially expressed in
specific brain regions between the two strains of mice (see
Materials and Methods for analysis criteria). A total of 73
genes were differentially expressed in at least one brain region
between the two strains (73 of the 7,169 expressed genes or ≈1.0% of
genes expressed in the adult male mouse brain). Twenty-four of these
73 genes were already identified and described above. The
remaining 49 are listed at the web site
( ftp://ftp.gnf.org/pub/ papers/brainstrain/)
and Table 3. A similar comparison by using MEFs showed that 115 genes
were differentially expressed between the strains (0.88% of all
measured genes or 1.2% of the genes expressed in MEFs). In general,
genes differentially expressed between the strains in one brain region
showed either a consistent trend in all other regions or were not
detected in other regions in either strain (see Table 3). Only two of
the 73 genes showed a pattern that was different in different regions.
The level of glutathione peroxidase mRNA was lower by approximately
9-fold in the midbrain of C57BL/6 compared with the 129SvEv
midbrain. By contrast, in the cerebellum the level of glutathione
peroxidase was higher by a factor of more than 1.5-fold in
C57BL/6. The mRNA abundance for the other gene of unknown
function (Table 3) was lower by approximately 8-fold in the entorhinal
cortex of C57BL/6 compared with 129SvEv. In contrast, the mRNA
was more abundant by more than 1.5-fold in the cerebellum of
C57BL/6. This finding suggests that the majority of genes
identified as differentially expressed in one brain region between the
two strains did not meet the strict criteria used in Table 1, but did
show a similar trend in all other brain regions. Northern Blot and Reverse Transcription–PCR Analysis. To test the accuracy in detecting differentially expressed genes and to
determine whether differences were unique to one or many strains,
Northern blot analysis was performed on total RNA from hippocampus and
cerebellum from C57BL/6, 129SvEv, first-generation offspring of
C57BL/6 and 129SvEv, and BALB/c strains. The analysis was
performed for genes shown in Table 1 that scored as absent in one
strain but present in the other. These included CAP,
PAM, spi2, and the gene similar to
ste20-like kinase. We found that spi2 was
detectable in the hippocampus of the 129SvEv mouse brain but was absent
in the C57BL/6 hippocampus consistent with the array results
(see Fig. ). However, this gene is
expressed in the hippocampus of all other strains tested, indicating
that the lack of expression in C57BL/6 is unique to this strain.
Northern blot analysis was not sufficiently sensitive to detect mRNA
for CAP, PAM, or the gene highly similar to
ste-20. Therefore, semiquantitative reverse
transcription–PCR was done by using RNA from C57BL/6 and
129SvEv. Consistent with the array results, CAP and PAM were
more highly expressed in 129SvEv, and ste20-like kinase was
more abundant in C57BL/6 (data not shown).
Gene Expression Differences in Response to Seizure. The cellular response to neurotoxic insults varies between inbred
strains of mice ( 1), and the C57BL/6 strain is resistant to
seizure-induced hippocampal cell death. To test whether gene expression
analysis could detect differences that correlate with known phenotypic
variability in CNS response, the expression profiles of hippocampus and
cerebellum in the two strains 1 h after seizure induction using
pentylenetetrazol were determined. As shown, the C57BL/6 mice
had a significantly greater overall transcriptional response to seizure
induction (Fig. a). This was
evident mainly as an increase in the number of genes induced in the
hippocampus (49 in C57BL/6 compared with 12 in 129SvEv,
P < 0.001). The transcriptional response of several
known immediate-early genes was compared between the two strains (Fig.
b), including members of the fos and
jun family, serum and glucocorticoid-regulated kinase
( sgk), growth factor inducible immediate early gene
(3CH134), cox-2, and the transcription factors
KROX20 and zif/268. All but one of these
genes, sgk, showed a similar level of postseizure induction
in the two strains (Fig. b). Therefore, the immediate-early
response to seizure is similar between the two strains whereas the
overall transcriptional response is blunted in 129SvEv. It will be
important to determine which of these genes account for the
differential response in these strains (for the complete list of genes
see
ftp://ftp.gnf.org/pub/papers/brainstrain/).
To test whether genes differentially regulated at baseline might
contribute to variation in seizure response, the level of induction of
the genes that were differentially expressed between the two strains
(for hippocampus) was assessed. Of the 32 genes differentially
expressed in the hippocampus (24 from Table 1 and eight from Table 3),
only seven showed a greater than 1.8-fold change in response to seizure
in at least one strain ( CAP, GIRK3,
MEF-2C, and PDNP2, and novels AA114725, AA048853,
and AA035993) and the response between the two strains was different
for CAP, GIRK3, PDNP2, and two novel
genes AA035992 and AA114725 ( P < 0.05, Student's
t test, Fig. c). Most of these genes, although
different at baseline and differentially responsive, showed a similar
directional response (Fig. c). CAP was the only
gene repressed in 129SvEv but induced in C57BL/6. This pattern
of expression suggests that changes in response to seizure for specific
genes may be similar between the two strains, but that baseline
differences dictate what type of transcriptional response is required. Brain Region-Specific Differences in Gene Expression. Finally, we identified genes that were uniquely expressed or highly
enriched in one brain region. To determine the likelihood of error
caused by dissection inconsistency, we compared four independently
obtained samples from the same brain region. No genes met the criteria
for differential gene expression, indicating that mouse-to-mouse
differences and dissections did not contribute significant variability
in the array measurements. We next compared the expression profiles of
cortex, cerebellum, and midbrain within the same strain and found that,
on average, a relatively small number of genes (70/13,069 or
0.54%) showed clear differences (see Materials and Methods
for analysis criteria). In contrast, 13.6% (1,780/13,069) of
the monitored genes were differentially expressed between brain and
fibroblasts, even though the two very different types of cell
populations express a similar overall number of genes. A further analysis was used to identify genes expressed uniquely in
particular brain regions (see Materials and Methods and Fig.
). An additional analysis was used to
detect region-specific variation (both restriction and enrichment) (see
Table 2 and
ftp://ftp.gnf.org/pub/papers/brainstrain/),
which indicates that the cerebellum appears to be the most unique
region of those tested. Twenty-three genes were expressed in the
cerebellum that were not detected in other regions (Fig. a)
and another 28 were not expressed in cerebellum but were present in
other brain regions (Fig. a). Importantly, genes such as
PCP-2, a known cerebellar-specific gene, and NMDA
NR2C, a known cerebellar-specific
N-methyl- d-aspartate receptor subunit,
were identified as being specifically expressed in the cerebellum,
providing further validation of the approach. In contrast to the
cerebellum, the structures of the medial temporal lobe (hippocampus,
amygdala, and entorhinal cortex) showed extremely similar expression
profiles. Only eight genes were unique to one of the three regions
(Fig. b). Of the seven genes present in hippocampus but not
amygdala or entorhinal cortex, six also were expressed outside of the
medial temporal lobe (Fig. b). There was only one gene
uniquely expressed in the amygdala and none in the entorhinal cortex.
This finding suggests that forebrain structures, despite some
functional differences, are highly similar at the molecular level.
Finally, the midbrain was interesting in that, although 10 genes were
uniquely expressed, no genes were exclusively absent.
The level of consistency between our expression data and published
results was considerable. As shown in Table 2, 14 genes were highly
enriched or restricted to the cerebellum. Of the known genes, the
regional expression patterns were entirely consistent with published
findings for 10 of 11 genes. Only MB-IRK2 was inconsistent
in that we were unable to detect mRNA for IRK2 in any region
except the cerebellum, whereas published reports found expression in
the cortex and hippocampus, with higher levels in the cerebellum ( 12).
The greater than 90% concordance with published results suggests that
the gene expression patterns are being accurately measured in the
highly parallel array-based experiments. We have generated a catalogue of brain region-specific gene
expression differences that might contribute to the unique
neurobehavioral phenotypes of these commonly used strains of mice. We
determined which genes are consistently differentially expressed
between these strains and also found that the two strains differ
markedly in their transcriptional response to seizure. Finally, we used
these data to determine brain region-specific differences in gene
expression. Our findings suggest that gene expression profiling of
inbred strains may be a useful tool for dissecting the molecular
mechanisms of behavioral variation. Candidate Gene Analysis. Although these data are correlative, candidate genes were
identified for further study to determine their role in mediating
strain-specific phenotypes. Virtually all of the known genes observed
to be differentially expressed have previously defined roles in the
CNS. It is interesting to speculate that the resistance to some forms
of neurotoxic insults in C57BL/6 ( 1) is caused by the
combination of a decrease in the expression of genes involved in
mediating neuronal damage ( GluR1) ( 13) and
( spi2/eb4) ( 14, 15) and an increase in the expression
of a gene that augments the cellular response to stress
( ste-20) ( 16). Gene Expression Profiling as a Method to Augment Quantitative Trait
Loci Analysis. Several of the differentially expressed genes are encoded in
chromosomal regions thought to harbor genes important for strain
differences in CNS phenotypes ( 8, 17– 19). Although quantitative trait
loci analysis is powerful for mapping susceptibility loci to chromosome
intervals, many genes reside in these large intervals, and extensive
additional work is required to identify the specific gene or genes
involved. Our findings suggest that an expression-based strategy is
useful in identifying candidate genes responsible for quantitative
traits. For example, GIRK3 (more highly expressed in
129SvEv) is located on chromosome 1 in a region that has been shown to
contain one or more of the genes that contribute to strain differences
for free running period and locomotor activity ( 20), aspects of fear
conditioned response (cued and contextual) ( 21, 22), open field
emotionality ( 23), and acute pentobarbital-induced seizures ( 24). This
gene plays a role in maintaining resting potential and in controlling
excitability of the cell ( 25) and should be considered a candidate for
involvement in modulating multiple CNS phenotypes. PAM (more
abundant in 129SvEv) is a key bifunctional enzyme in the activation of
neuropeptides ( 26). The gene encodes two different enzymes,
peptidylglycine α-hydroxylating monooxygenase and
peptidyl-α-hydroxyglycine α-amidating lyase. These enzymes function
sequentially in a two-step pathway of peptide amidation. This gene maps
to chromosome 1 at 57.5 cM, and an ethanol-induced loss of righting
reflex locus has been mapped to chromosome 1 between 43 and 59 cM ( 27).
Interestingly, changes in several neural peptides, such as neurotensin,
have been linked to ethanol sensitivity, providing a potential link
between PAM and modifications of peptides involved in mediating ethanol
responses ( 28). Another two genes differentially expressed between the
strains, I2RF5 and a G-protein subunit, are located on
distal mouse chromosome 4. This region of chromosome 4 has been linked
to quantitative trait loci for alcohol drinking preference, saccharin
and sucrose preference ( 29– 32), and methyl β-carboline-3-carboxylate
seizure susceptibility ( 33). These genes are good candidates for
further study and suggests that gene expression profiling may be a
useful and more rapid approach for identifying or establishing the role
of a set of genes involved in a particular complex trait. Strain Variation in Seizure Response. The seizure experiments were designed to determine which genes
might contribute to strain-specific responses. Several of the genes
identified as different between the two strains were genes whose
expression was differentially altered in response to seizure. In
addition, the marked difference in transcriptional response between
these strains suggests that changes in expression may account for
strain variation in the cellular consequences of seizure response. We
also note that very few genes were repressed (nine genes, see
ftp://ftp.gnf.org/pub/papers/brainstrain/)
1 h after seizure induction. It will be interesting to study the
function of the genes repressed in response to seizure at early as well
as later time points and to further define their role in seizure
response. Gene Expression Profiling in Nonisogenic Strains. It is important to consider the implications of these results for
studies that use nonisogenic strains of mice. Many laboratories have
generated mice with targeted mutations in genes (knockouts) or
overexpressed genes (transgenics) and reported novel behavioral
phenotypes. The neurobehavioral phenotype of a particular mouse results
not only from the specific alteration induced by a targeted mutation,
but also from the effects of modifiers, which may differ significantly
based on genetic background ( 34). We have estimated that the 129SvEv
gene expression profile is significantly different (≈1% of expressed
genes) from that of other strains commonly used in transgenic
experiments, such as C57BL/6. Use of nonisogenic mouse strains
is therefore likely to produce situations where differences may be
identified, but it will be difficult to know with certainty whether the
differences are caused by the specific perturbation or are heavily
influenced by other differences caused by variation in genetic
background among nonisogenic littermates. By considering the results
presented here, it should be possible to exclude genes that differ
because of genetic background alone and to identify modifiers that
modulate phenotypes in different genetic backgrounds. Brain Region Gene Expression Profiling. The high concordance between our data and published results
indicates that array-based gene expression profiling can be used to
determine which genes are expressed and where. However, we found that
it was important to use stringent analysis criteria coupled with
statistical tests to ensure that the expression profiles are
interpreted appropriately. As it becomes possible to use this
technology for nuclei or even small cell populations in the CNS,
higher-resolution, region-specific, and cell-type specific information
will be gained. Studies of the regulatory elements for the uniquely
expressed genes may be useful in identifying promoters that could be
used to drive expression in specific cell types or tissues in animal
models. By making the complete data sets available on the web, we
encourage others to investigate the data to uncover more candidates for
further study. This study demonstrates the feasibility and utility of brain region
expression profiling and lays the foundation for asking system-level
questions. The expression results serve as a framework to begin to
understand the factors responsible for the variation in phenotypes
involving behavior, drug sensitivity, and neurotoxic-induced cell
death. There is no doubt that advances in gene targeting technology,
robust behavioral analysis, and global gene expression measurements
will provide new avenues for studying the brain and further our ability
to understand the interplay between the genes that give rise to complex
behaviors and unique brain functions. We thank P. G. Schultz for inspiration and support, Daniel Lockhart
for the development of the nfueggo software, J. Hogenesch,
R. Vega, and members of the Lockhart laboratory for assistance with
gene expression profiling, J. A. Greenhall for assistance with PCR, K.
G. Xanthopoulos, L. Tarantino, and A. Wynshaw-Boris for reviewing the
manuscript, and members of the Barlow laboratory for helpful
discussions. This work was supported in part by funds to C.B. from the
Frederick B. Rentschler Developmental Chair and the Esther A. and
Joseph Klingenstein Fund. | | CNS | central nervous system | | | MEF | mouse embryonic
fibroblast |
1. Schauwecker P E, Steward O. Proc Natl Acad Sci USA. 1997;94:4103–4108. [PubMed] 2. Crawley J N, Belknap J K, Collins A, Crabbe J C, Frankel W, Henderson N, Hitzemann R J, Maxson S C, Miner L L, Silva A J, et al. Psychopharmacology. 1997;132:107–124. [PubMed] 3. Lockhart D J, Dong H, Byrne M C, Follettie M T, Gallo M V, Chee M S, Mittmann M, Wang C, Kobayashi M, Horton H, Brown E L. Nat Biotechnol. 1996;14:1675–1680. [PubMed] 4. Wodicka L, Dong H, Mittmann M, Ho M H, Lockhart D J. Nat Biotechnol. 1997;15:1359–1367. [PubMed] 5. Lipshutz R J, Fodor S P, Gingeras T R, Lockhart D J. Nat Genet. 1999;21:20–24. [PubMed] 6. Simpson E M, Linder C C, Sargent E E, Davisson M T, Mobraaten L E, Sharp J J. Nat Genet. 1997;16:19–27. [PubMed] 7. Threadgill D W, Yee D, Matin A, Nadeau J H, Magnuson T. Mamm Genome. 1997;8:390–393. [PubMed] 8. Ferraro T N, Golden G T, Smith G G, St. Jean P, Schork N J, Mulholland N, Ballas C, Schill J, Buono R J, Berrettini W H. J Neurosci. 1999;19:6733–6739. [PubMed] 9. Hogan B, Beddington R, Costantini F, Lacy E. Manipulating the Mouse Embryo: A Laboratory Manual. Plainview, NY: Cold Spring Harbor Lab. Press; 1994. 10. Kubo Y, Nakagawa Y, Kakimi K, Matsui H, Higo K, Wang L, Kobayashi H, Hirama T, Ishimoto A. J Gen Virol. 1994;75:881–888. [PubMed] 11. Muller A J, Chatterjee S, Teresky A, Levine A J. Mamm Genome. 1998;9:773–774. [PubMed] 12. Karschin C, Dissmann E, Stuhmer W, Karschin A. J Neurosci. 1996;16:3559–3570. [PubMed] 13. Pellegrini-Giampietro D E, Peruginelli F, Meli E, Cozzi A, Albani-Torregrossa S, Pellicciari R, Moroni F. Neuropharmacology. 1999;38:1607–1619. [PubMed] 14. Inglis J D, Lee M, Davidson D R, Hill R E. Gene. 1991;106:213–220. [PubMed] 15. Chiang C S, Hong J H, Stalder A, Sun J R, Withers H R, McBride W H. Int J Radiat Biol. 1997;72:45–53. [PubMed] 16. Pombo C M, Tsujita T, Kyriakis J M, Bonventre J V, Force T. J Biol Chem. 1997;272:29372–29379. [PubMed] 17. Ferraro T N, Golden G T, Smith G G, Schork N J, St. Jean P, Ballas C, Choi H, Berrettini W H. Mamm Genome. 1997;8:200–208. [PubMed] 18. Demarest K, McCaughran J, Jr, Mahjubi E, Cipp L, Hitzemann R. J Neurosci. 1999;19:549–561. [PubMed] 19. Crabbe J C, Phillips T J, Buck K J, Cunningham C L, Belknap J K. Trends Neurosci. 1999;22:173–179. [PubMed] 20. Mayeda A R, Hofstetter J R. Behav Genet. 1999;29:171–176. [PubMed] 21. Caldarone B, Saavedra C, Tartaglia K, Wehner J M, Dudek B C, Flaherty L. Nat Genet. 1997;17:335–337. [PubMed] 22. Wehner J M, Radcliffe R A, Rosmann S T, Christensen S C, Rasmussen D L, Fulker D W, Wiles M. Nat Genet. 1997;17:331–334. [PubMed] 23. Flint J, Corley R, DeFries J C, Fulker D W, Gray J A, Miller S, Collins A C. Science. 1995;269:1432–1435. [PubMed] 24. Buck K, Metten P, Belknap J, Crabbe J. Mamm Genome. 1999;10:431–437. [PubMed] 25. Kubo Y, Baldwin T, Jan Y, Jan L. Nature (London). 1993;362:127–133. [PubMed] 26. Ouafik L H, Stoffers D A, Campbell T A, Johnson R C, Bloomquist B T, Mains R E, Eipper B A. Mol Endocrinol. 1992;6:1571–1584. [PubMed] 27. Markel P D, Bennett B, Beeson M, Gordon L, Johnson T E. Genome Res. 1997;7:92–99. [PubMed] 28. Duncan C C, Erwin V G. Alcohol. 1992;9:23–29. [PubMed] 29. Bachmanov A A, Tordoff M G, Beauchamp G K. Alcohol Clin Exp Res. 1996;20:201–206. [PubMed] 30. Bachmanov A A, Reed D R, Ninomiya Y, Inoue M, Tordoff M G, Price R A, Beauchamp G K. Mamm Genome. 1997;8:545–548. [PubMed] 31. Tarantino L M, McClearn G E, Rodriguez L A, Plomin R. Alcohol Clin Exp Res. 1998;22:1099–1105. [PubMed] 32. Blizard D A, Kotlus B, Frank M E. Chem Senses. 1999;24:373–385. [PubMed] 33. Martin B, Clement Y, Venault P, Chapouthier G. J Hered. 1995;86:274–279. [PubMed] 34. Gerlai R. Trends Neurosci. 1996;19:177–181. [PubMed]
|
See commentary "
