Gene expression profiling using microarrays is a powerful technology widely used to study regulatory networks. Profiling of mRNA levels in mutant organisms has the potential to identify genes regulated by the mutated protein.
Using tissues from multiple lines of knockout mice we have examined genome-wide changes in gene expression. We report that a significant proportion of changed genes were found near the targeted gene.
Citation: Valor LM, Grant SGN (2007) Clustered Gene Expression Changes Flank Targeted Gene Loci in Knockout Mice. PLoS ONE 2(12): e1303. https://doi.org/10.1371/journal.pone.0001303
Academic Editor: Abraham Sonenshein, Tufts University, United States of America
Received: July 5, 2007; Accepted: November 15, 2007; Published: December 12, 2007
Copyright: © 2007 Valor, Grant. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was funded by the Wellcome Trust. They supported the research and played no specific role in the details of the research.
Competing interests: The authors have declared that no competing interests exist.
Microarray technology has been widely assimilated by gene expression researchers and its success is largely due to its power of assaying thousands of transcripts in a genome wide manner. One of the most popular experimental designs has been to determine the gene expression changes that occur in knockout mice where it is correlated with phenotypic changes. The usual objective is to obtain insights into the role of the mutated protein by identifying particular genes under its control, and retrieve candidate genes that may explain the altered phenotype in the mutant animal.
The creation of knockout mice involves the modification of the genome of mouse embryonic stem (ES) cells through a variety of processes including homologous recombination or other mutagenic procedures . The majority of gene targeting experiments have been performed in ES cells derived from the 129 strain of mice. Following germ-line transmission of the mutant allele, the founder mice are often backcrossed onto other strains, creating hybrid animals. The relative contribution of the genetic background of the ES cell and the strain to which it was bred has been widely studied .
We have studied the effects of knockout (KO) mutations on the mechanisms of learning and memory , . Patterns of neural activity induced by learning or artificial stimuli lead to activation of signal transduction pathways, which ultimately lead to the cellular changes underlying learning and memory. The two major cellular changes involve strength of synaptic transmission (synaptic plasticity) and gene expression (activity-dependent gene expression). It is clear from studies in the sensory-motor synapses of Aplysia californica and the hippocampus of mice and rats that neurotransmitter receptors initiate the signalling events that lead to both synaptic plasticity and activity-dependent gene expression , , , , , , . L-glutamate is the dominant excitatory neurotransmitter in the central nervous system of vertebrates and glutamate receptors are of special relevance to synaptic plasticity. The ionotropic N-methyl-D-aspartate (NMDA) and metabotropic (mGLuR) subtypes of glutamate receptor are commonly involved in experimental plasticity-driven transcriptional changes. Both the NMDA and mGluR receptors are connected by a series of adaptor proteins ,  and can be physically isolated from mouse brain in ∼2000-kDa multiprotein complexes called NRC or MASC (NRC, NMDA Receptor Complex; MASC, MAGUK Associated Signaling Complex) , , , . NRCs are organized into receptor, adaptor, signalling, cytoskeletal and novel proteins, and includes cell adhesion molecules, kinases, phosphatases, GTPase-activating proteins and Ras and MAPK pathway components. NRC/MASC form a subset of the post-synaptic density which is thought to be comprised of a set of complexes. Such a complex offers an attractive model for the organisation and coordination signalling networks , , . It was hypothesised that the function of the NRC is to orchestrate the signalling pathways downstream of receptor activation and coordinate both synaptic plasticity and activity-dependent gene expression amongst other changes.
We were specifically interested in the previously characterized KO mice for PSD-95 and SAP102 , . Both proteins are from the MAGUK (Membrane Associated GUanylate Kinase) family, which members share a common structure of three PDZ (PSD-95/Dlg/ZO-1), an SH3 (Src-homology-3) and guanylate kinase (GK) domains and together enable protein-protein interactions. The postsynaptic density 95 (PSD-95) protein is a homolog of the Drosophila suppressor gene, dlg  and acts as a scaffold to organize postsynaptic proteins into signalling complexes  or anchoring complexes for glutamate receptors and cytoskeletal proteins, as reviewed elsewhere . Mice lacking PSD-95 or SAP102 were found to have enhanced forms of NMDA receptor-dependent long-term potentiation , , ,  and exhibited impairments in spatial learning (and other forms of behavioural plasticity) , , . Our original aim was to identify those transcripts that were controlled by PSD-95 and SAP102 and therefore try to track their dependent signalling pathways. Surprisingly, and because of the unbiased nature of microarray analyses, we found that the most significantly changing genes were clustered in the neighbourhood of the KO gene. We observed that this phenomenon was frequent in KO studies and, therefore, special care should be taken in the interpretation of this kind of experiments.
To determine the gene expression changes in the hippocampus of mice lacking SAP102 or PSD-95, total RNA was isolated from the hippocampus of SAP102−/Y and PSD-95−/− mutant mice and littermate wild-types. RNA was hybridized to MG-430 2.0 arrays (Affymetrix) that contained 45,037 probe sets excluding controls. The data was normalized and filtered (see Materials and Methods). We first examined the expression data from SAP102 mutants and found that 65% (13/20) of the significantly changed probe sets were encoded by chromosome X, which also encodes SAP102. This unexpected finding suggests that either SAP102 preferentially regulates genes on the X-chromosome or that there is some cis-acting effect of the mutation.
We focused on the SAP102 regulated genes on the X-chromosome and found they were clustered around the SAP102/dlgh3 gene (Figure 1). These genes spanned a region of ∼98 Mb (Figure 1A, Table 1). Within this cluster was the hprt (hypoxanthine guanine phosphoribosyl transferase) gene, a selection marker extensively used in ES cells to generate transgenic mice , , , . In our case, the ES cell clone used to generate our knockout mice was E14TG2a HM-1, which harbours an hprt allele containing a 55 kb-deletion in its 5′-flanking region . Using a PCR protocol specifically designed to genotype this deletion , we found it was correlated with the reduction of HPRT expression (data not shown). The hprt and SAP102/dlgh3 loci are separated by ∼48 Mb and since the hprt locus of 129S5 mice does not contain the deletion, these data suggested that the observed changes in hprt gene expression in SAP102 mutant mice was the result of a linked mutation originated in the ES cells. Following gene targeting of SAP102/dlgh3 gene in E14TG2a ES cells (129P2 background), the chimera was crossed to 129S5 and subsequently backcrossed. This provided an important clue that genetic background of the ES cells might be contributing to the gene expression profiling.
White histogram, results from 129S5 strain animals; black histogram, results from C57BL/6j after selection of hprt deletion (see text). B, probe sets from PSD-95−/− along chromosome 11, where the position of the mutated gene (dlgh4) is indicated. White histogram, hippocampus mRNA; black histogram; cortex mRNA.
To further explore this hypothesis, we examined the degree of linkage between both loci. Using the MGI database (www.informatics.jax.org) we estimated the percentage of independent segregation of hprt and SAP102/dlg3 to be 27% (27 cM). We tested this using a cross between SAP102/dlg3 heterozygous females and 129S5 (wild type) males, which produced 37 litters, and subsequent genotyping. Only 27% of the heterozygous females had two wild-type alleles of hprt and 29% of E17.5 male embryos showed segregation of hprt and SAP102/dlg3 mutations. These data are in keeping with the conclusion that the entire portion between both loci originated in the original E14TG2a ES cells.
These observations suggest that further backcrossing and marker selection would result in a reduction of the E14TG2a DNA. To explore this possibility we analyzed microarray data from SAP102−/Y hippocampi that contained the wild-type allele of hprt mutation. In this experiment we used SAP102−/Y mutant mice that had been backcrossed 3 generations onto the C57BL/6j background. In this case, only six probe sets were changed from the X chromosome (∼29% of the significant probe sets; 6/21) and spanned a shorter region of ∼48 Mb (Figure 1A, Table 1). This clustering was significantly different from a random distribution along the X chromosome (P<0.05, χ2 = 26.6 d.f. = 16). Four of the five probe sets were common to the previous experiment (Table 1) and the change was similar in direction and magnitude, thus reproducing the effect of the ES cell-derived sequences and suggesting changes were independent of the host backcrossing genetic background. Since we used C57BL/6j animals for the new analysis, we could take advantage of the availability of primers that distinguished C57- and 129-derived strains genotypes. We focused on Atp7a, which expression was changed in the microarray experiment and is found ∼5 Mb from SAP102/dlg3 and tested if it was of 129 or C57BL/6 origin. Using a published genotyping protocol (in which the amplicon covers exon 8 of Atp7a ), we found that the all samples that showed changes in Atp7a expression were of 129 origin (Figure 2). These data further support the conclusion that the changes in gene expression in the mutant mice involving the chromosome X can be explained by genetic background effects.
Two bands were observed for Atp7a reactions, one of 694 bp corresponding to 129-derived sequences and another of 637 bp corresponding to C57Bl6-derived sequences (upper panel). The appearance of the 694 bp-band coincided with the presence of the SAP102 mutation (-/y) (lower panel). Wild-type 129S5 (129) and C57BL/6j (BL6) were used as positive controls.
Similar effects were observed for other mutations involving autosomes. We took advantage of the results obtained in another knockout for a MAGUK protein, PSD-95, which is encoded on chromosome 11. Using an identical microarray platform to that used for SAP102 mutants, we found that ∼47% (8/17) of the significantly changed probes in hippocampus were encoded on chromosome 11. As shown in Figure 1B there was significant clustering of probe sets around the dlgh4 locus and spanned a region of ∼5 Mb (P<0.001, χ2 = 33.7 d.f. = 11). This shorter length of the ES cell-derived region in PSD-95 mutants may be due to the larger number of backcrosses (>10 in PSD-95 onto a C57BL/6 background) compared to the SAP102 mutants (∼48 Mb after 3 backcrossed on C57BL/6). These data are consistent with the view that the number of backcrosses helps to shorten the length of ES cell-derived sequences as well as marker selection (hprt in the SAP102 mutants) and should be considered in microarray profiling experiments in knock-out animals. The above data is based on studies of hippocampus transcript levels and we asked if similar results would be observed in other regions of the nervous system. Using cortex RNA from PSD-95 mutants we found that 4 of the 6 probe sets that were significant in the hippocampus were also significant in the cortex of the same animals (Table 2), thus reproducing the effects of ES-cell derived sequences in gene expression. For further details on the statistical analysis of both SAP102 and PSD-95 mutants, see Tables 3 and 4.
We also validated the microarray experiments performed in SAP102 and PSD-95 mutants by qRT-PCR. As shown in Figure 3 the changes in gene expression detected with microarrays were confirmed using qRT-PCR for SAP102, HPRT, Fundc2, and BB315069 (3′ to Fundc2) in the SAP102 mutants and PSD-95, Fbxo39 and BG092359 (3′ to Fbxo39) in the PSD-95 mutants.
Hippocampus mRNA was extracted and level of mRNA detected using microarrays or qRT-PCR. A, SAP102 mutant mice; results for SAP102, Hprt, Fundc2, BB315069 (3′ to Fundc2) are shown. B. PSD-95 mutant mice; results for PSD-95, Fbxo39 and BG092359 (3′ to Fbxo39). y-axis: fold change of mutant mRNA levels compared to control; mean±s.e.m. P, qRT-PCR data; M, microarray data. M1, M2… refers to different probe sets for the same gene. Note different scale of y-axes.
From the above experiments on brain mRNA from SAP102 and PSD-95 mutants it seems likely that these effects of genes linked to the mutant allele might be generally important in other knockout lines and other tissues. To address this issue we analyzed recent data deposited in ArrayExpress (www.ebi.ac.uk/arrayexpress/) that employed the same format of arrays as used in our study. The purpose was not to exhaustively compile the existent data but to provide further examples. Data from four mutant genes were analyzed: Foxa3 , Nebulin (neb) , Ercc1  and Fibulin-4 (efemp2)  (Figure 4), using RNA derived from testes, skeletal muscle, liver and aorta respectively.
y-axis: number of changed genes, x-axis: chromosomal position. The arrow indicates the location of the mutated genes in each case.
In all four knockouts we found changes in gene expression in genes near to the targeted locus; however the potential boundary between ES cell and host DNA was not clear in all mutants (Table 3, Figure 4). The number of changed genes was plotted for the length of the relevant chromosome and the position of the targeted gene indicated. This allows us to examine clustering ‘within chromosomes’ and to compare with random distributions (see Materials and Methods). For chromosome 7 and the foxa3 locus, of the 8 genes that changed on this chromosome (including the mutated gene), all were clustered within 28 Mb of the mutation (P<0.01, χ2 = 31.1 d.f. = 14), despite 11 backcrosses onto C57BL/6 strain . For ercc1 and neb mutants, the changed genes were distributed more widely on chromosome 7 and 2 respectively. Although there was a trend toward clustering with the mutant locus, this distribution was not statistically significant (Table 3): for ercc1 there were 72.7% of significant genes within ∼60 Mb (∼42% of the chromosome); for neb there were 53.3% within ∼75 Mb (∼42% of the chromosome). The inability to identify more conclusive evidence of linkage in ercc1 or neb mutants was probably due to the large number of changed genes: compared to SAP102, PSD-95 and Foxa3 where 21, 12 and 10 genes respectively were changed, 140 genes were changed in ercc1 mutants and 92 in neb mutants (Table 4). In the case of the efemp2 locus on chromosome 19, there were 4 genes changed; however clustering was not statistically significant (Table 3) perhaps because chromosome 19 is too short and only a small portion (∼8% of the chromosome, which was absent of significant genes) was estimated to segregate independently and thus it is possible that the entire chromosome was derived from ES-cell background. Thus using the ‘within chromosome’ clustering does not reveal statistically significant results for all knockout mice and appears to be influenced by chromosome size and number of significantly changed genes. To overcome these limitations we used an alternative approach where we compared the frequency of changed genes and their contribution to deviation from random distributions on the targeted chromosome with other chromosomes (see Materials and Methods). Using this ‘between chromosome’ method we found that of 7 mutant examined, 6 showed evidence of clustering of changed genes to the targeted chromosome (Table 4).
Microarrays are a powerful method for detecting changes in gene expression and here we examine data from a range of different knockout mice. When comparing RNA from mutant mice and wild type controls, we observed a higher than expected frequency of changed genes near to the targeted/knocked-out gene. These genes were either clustered around the targeted allele or over-represented on that chromosome. Recent reports found evidence of linked changes in two knockouts ,  (one of them published during the preparation of this manuscript). In our studies of PSD-95 and SAP102, the gene was disrupted in 129P2 ES cells and the RNA extracted from mice on a hybrid background (since a distinct mouse strain was used during breeding of the mutants). We consider the most likely explanation for the changes in gene expression of the genes clustered around the knockout locus to be that these genes were of ES cell origin (flanking DNA). In this model, the flanking DNA (from ES cells) contains polymorphisms that alter gene expression relative to the genetic background of the backcrossed mice. We found changes in gene expression in SAP102 mice (from 129P2 ES cells) backcrossed onto 129S5 and C57BL/6 strains. Although 129P2 and 129S5 might be closely related genetic backgrounds, a large variability has been reported between 129 substrains , . The notion that flanking DNA can contribute to complications in mouse behavioural studies has been recognised and breeding strategies recommended to minimise its effects . We have reported here that although backcrosses shorten the length of ES cell flanking regions more than ten backcrosses was insufficient to completely eliminate them (see PSD-95 and Foxa3 examples), as recognized in . Marker selection (as the example provided by hprt) can accelerate the identification of mice with shorter flanking DNA.
An alternative model, which we cannot exclude, is that the changes in gene expression around the knockout locus were because of the modifications introduced during gene targeting (e.g. neomycin gene) that produce a cis-acting effect on neighbouring genes. This could arise either by disrupting a functional region in the host genome (i.e., a regulatory element) or interaction of the foreign sequence with the flanking DNA. The number of changing genes in the microarray experiments in the SAP102 and PSD-95 samples was inversely correlated with the number of backcrosses and active marker selection, suggesting the cis-acting effect, if it exists, only partially contributes. To explore a cis effect in more detail, a future approach could consist of analyzing microarray data from knock-in animals, where the insertion is made into a region lacking functional elements (i.e., coding region or a regulatory element).
Microarray analysis in knock-out animals should be carefully examined to avoid misinterpretation in the gene expression profile. Rather than assign the observed changes to the mutation, the genotype of the changed genes needs to be tested. This current study highlights the effects of flanking DNA of ES cell origin, which is selected during breeding of the mutant mice. It may therefore be necessary to genotype all changed genes before studying mutation-dependent effects. Although this may not be practical in all situations, the simplest approach would be to treat those changed genes on the targeted chromosome with extreme caution and only consider changes on other chromosomes or in those portions of the targeted chromosome estimated to segregate independently. Future array based methods could include mRNA profiling and a separate array for profiling the genotype of the corresponding probe sets and their surrounding regions. We would also recommend that when gene expression data is reported on mutant mice, that the chromosome location of the genes encoding those mRNAs is clearly reported. Studies on mutant mice on the inbred background of the ES cells would obviate the need to perform genotyping.
Materials and Methods
For microarray studies, the ages of mice used were as follows: 6 weeks old (129S5 strain); 12–13 weeks old (C57BL/6j strain). Number of mice: five SAP102−/Y and five littermate wild-types (129S5 strain); four SAP102−/Y and three littermate wild-types (C57BL/6j strain); five PSD-95−/− and four littermate wild-types (hippocampus, C57BL/6j strain); two PSD-95−/− and two littermate wild-types (cortex, C57BL/6j strain). Animals were dispatched by cervical dislocation. All dissections were performed consecutively on the same day at 4°C and tissue kept moist with chilled PBS (Gibco). Collected tissue (hippocampus and cortex) was immediately frozen in liquid nitrogen or immersed into RNAlater reagent (Qiagen) on ice for RNA isolation. Animals were treated in accordance with the UK Animal Scientific Procedures Act (1986).
Nucleic acids isolation
RNeasy Mini and Midi columns (Qiagen) were used to purify the RNA from hippocampus and cortex, respectively. On-column DNase treatment was carried out to remove traces of DNA. During all the purifications, manufacturer's instructions were followed. Samples were concentrated with the addition of 1/10 vol of 3 M NaOAc pH 6 and 3 vol of 100% EtOH and precipitated at −20°C. Once resuspended in 12 µl of RNase-free H2O, an aliquot was used to check the RNA integrity in 1% formaldehyde-agarose gels.
To isolate genomic DNA, mouse tails were cut into 4–5 pieces and treated with 500 µl of the Lysis Buffer (10 mM Tris-HCl pH = 7.5, 1 mM EDTA, 1% SDS, 10 mM NaCl) and 2.5 µl Proteinase K (20 mg/ml) during 37 °C O/N. Then, 500 µl buffered phenol was added and the mix was incubated 1–2 h with gentle rotation at 4 °C. Once taking the supernatant after a low-speed spin (5,000 rpm 5 min), 500 µl of isoamylic-chloroform-phenol was used and spin at 13,000 rpm for 5 min. DNA was immediately precipitated by adding 1 ml isopropyl alcohol. After spinning at 13,000 rpm for 5 min, the pellet was washed with 500 µl 70% ethanol and resuspend in 100 µl H2O.
Microarray hybridization, data acquisition and statistical analyses
For the transcriptome profiling, MG-430 2.0 arrays (Affymetrix) were used, that contain 45,101 probe sets including controls, one array per sample. Briefly, 3 µg of total RNA was reverse transcribed, labelled and hybridized using One-cycle Labelling Kit instructions (Affymetrix). Fluidics Station 450 and GCS3000, both from Affymetrix, were used for the washing and scanning steps respectively.
Signal intensities values and detection flags were extracted using GCOS (Affymetrix), using a Target value of 500 in the scaling step. Data were normalized by the median per gene and per array using GeneSpring v7 (Silicon Genetics): i) ANOVA test (p<0.05) without assuming equal variances, ii) signal intensity must be a Present or Marginal call in all wild-type samples when down-regulated and knock-out samples when up-regulated, and iii) fold change >1.8 that was selected arbitrarily for the purposes of the present work. In parallel data was processed using dChip (PM/MM difference model) . Only those probe sets appearing in both analyses were considered.
The significant genes obtained in the microarray analyses were grouped in bins of 10 Mb to simplify both visualization in the figures and χ2 tests. Centromeres were not considered. We compared the distribution of the significant genes along the targeted chromosome to a random distribution. This random distribution was calculated by dividing the number of significant genes by the number of bins. When testing the distribution of significant genes between all the chromosomes, three types of random distribution were analyzed: equal proportion of significant genes in all the chromosomes, proportion normalized by the length of each chromosome and proportion normalized by the number of genes (based upon Ensembl release 45). The microarray data is deposited in the ArrayExpress database, www.ebi.ac.uk/arrayexpress, accession number E-MEXP-1159.
Retrotranscription and PCR
Typically, 1 µg of total RNA was used for the retrotranscription, following the recommendations of Invitrogen (SuperScript II Reverse Transcriptase). PCR reactions were set up per sample using 1 µl cDNA, 0.3 µM each primer, 25 µl 2× Quantitect SYBR Green PCR Master Mix (Qiagen) to a final volume of 50 µl. The sequences of the primers were 5′-GTGGATTCTTTCTCCTTCAGC-3′ and 5′-ATACAACCTCCTCAGCTTTGC-3′ (Fundc2), 5′-TTCTGGTTTGTTTTGTTTTGC-3′ and 5′-TTCATGCCTTTTGTAAGCAAG-3′ (BB315069), 5′-CTCTCCACTTCATGCTCCAC-3′ and 5′-ATGGACATCCAGATTGCTAAG-3′ (Fbxo39), 5′-AATCGTCATAGGGAGGTTCAC-3′ and 5′-AGACCTTGCTATAGGCTCCAG-3′ (BG092359), 5′-ACGAGAGTGGTCAAGGTTAAAG-3′ and 5′-GGAGAGAAGATCATCGTTGG-3′ (PSD-95), 5′-AAAGGACCTCTCGAAGTGTTG-3′ and 5′-ATTTGGCTTTTCCAGTTTCAC-3′ (Hprt), 5′-GTGCTACGGGTGAATGAG-3′ and 5′-TGGTCTATCTGAAGGTGGCC-3′ (Sap102). Reactions were run out in the 7500 Real Time PCR System (Applied Biosystems) with the following conditions: 1 cycle of 95°C for 15 min, 35–40 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 50 s, 1 cycle of 72°C for 5 min. The resulting amplicons were checked for specificity and size in agarose gels. The data generated was analysed in 7500 System SDS v1.2.2 software, that calculated the cycle threshold (CT) and the fold change ( = 2−ΔΔCT). GAPDH was used as endogenous controls. To analyze statistically the quantitative PCR results, Student's t-test was used.
For genotyping, each PCR was performed using 1 µl genomic DNA, 1 U Platinum Taq DNA polymerase (Invitrogen), 5 µl PCR buffer 10× (Invitrogen), 10 mM each dNTP, 50 mM MgCl2 and 0.5 µM each primer in a final volume of 50 µl. For hprt alleles genotyping, the protocol described in  was used with few modifications. Reactions to detect wild-type and mutant alleles were set up in parallel for the same genomic DNA: 1 cycle of 94°C 5 min, 35 cycles of 94°C 30 s, 62°C 30 s for the wild-type allele or 67°C for the null allele, 72°C 60 s and 1 cycle of 72°C 5 min. Addition of 1% DMSO was required for the null allele genotyping. For atp7a allele genotyping, the protocol described in  was used with few modifications, in which the conditions were 1 cycle of 94°C 5 min, 35 cycles of 94°C 30 s, 55°C 30 s, 72°C 60 s and 1 cycle of 72°C 5 min.
We thank Jane Robinson for the maintenance of the animals; Nurudeen Afinowi, Marcelo Coba and Peter Cuthbert for their technical assistance and discussion; Lawrence Humphreys and Paul Charlesworth for providing the material for genotyping; Noboru Komiyama and Douglas Strathdee for reviewing the manuscript; Jane Turner for editorial assistance.
Conceived and designed the experiments: SG LV. Performed the experiments: LV. Analyzed the data: SG LV. Wrote the paper: SG LV.
- 1. Tymms M, Kola I (1999) Gene knockout protocols (Methods in Molecular Biology). Humana Press, Inc. M. TymmsI. Kola1999Gene knockout protocols (Methods in Molecular Biology).Humana Press, Inc431
- 2. Crusio WE (2004) Flanking gene and genetic background problems in genetically manipulated mice. Biol Psychiatry 56: 381–385.WE Crusio2004Flanking gene and genetic background problems in genetically manipulated mice.Biol Psychiatry56381385
- 3. Cuthbert PC, Stanford LE, Coba MP, Ainge JA, Fink AE, et al. (2007) Synapse-associated protein 102/dlgh3 couples the NMDA receptor to specific plasticity pathways and learning strategies. J Neurosci 27: 2673–2682.PC CuthbertLE StanfordMP CobaJA AingeAE Fink2007Synapse-associated protein 102/dlgh3 couples the NMDA receptor to specific plasticity pathways and learning strategies.J Neurosci2726732682
- 4. Migaud M, Charlesworth P, Dempster M, Webster LC, Watabe AM, et al. (1998) Enhanced long-term potentiation and impaired learning in mice with mutant postsynaptic density-95 protein. Nature 396: 433–439.M. MigaudP. CharlesworthM. DempsterLC WebsterAM Watabe1998Enhanced long-term potentiation and impaired learning in mice with mutant postsynaptic density-95 protein.Nature396433439
- 5. Nguyen PV, Abel T, Kandel ER (1994) Requirement of a critical period of transcription for induction of a late phase of LTP. Science 265: 1104–1107.PV NguyenT. AbelER Kandel1994Requirement of a critical period of transcription for induction of a late phase of LTP.Science26511041107
- 6. Linden DJ (1996) A protein synthesis-dependent late phase of cerebellar long-term depression. Neuron 17: 483–490.DJ Linden1996A protein synthesis-dependent late phase of cerebellar long-term depression.Neuron17483490
- 7. Milner B, Squire LR, Kandel ER (1998) Cognitive neuroscience and the study of memory. Neuron 20: 445–468.B. MilnerLR SquireER Kandel1998Cognitive neuroscience and the study of memory.Neuron20445468
- 8. Alberini CM (1999) Genes to remember. J Exp Biol 202: 2887–2891.CM Alberini1999Genes to remember.J Exp Biol20228872891
- 9. Bading H (2000) Transcription-dependent neuronal plasticity the nuclear calcium hypothesis. Eur J Biochem 267: 5280–5283.H. Bading2000Transcription-dependent neuronal plasticity the nuclear calcium hypothesis.Eur J Biochem26752805283
- 10. Qian Z, Gilbert ME, Colicos MA, Kandel ER, Kuhl D (1993) Tissue-plasminogen activator is induced as an immediate-early gene during seizure, kindling and long-term potentiation. Nature 361: 453–457.Z. QianME GilbertMA ColicosER KandelD. Kuhl1993Tissue-plasminogen activator is induced as an immediate-early gene during seizure, kindling and long-term potentiation.Nature361453457
- 11. West AE, Griffith EC, Greenberg ME (2002) Regulation of transcription factors by neuronal activity. Nat Rev Neurosci 3: 921–931.AE WestEC GriffithME Greenberg2002Regulation of transcription factors by neuronal activity.Nat Rev Neurosci3921931
- 12. Tu JC, Xiao B, Naisbitt S, Yuan JP, Petralia RS, et al. (1999) Coupling of mGluR/Homer and PSD-95 complexes by the Shank family of postsynaptic density proteins. Neuron 23: 583–592.JC TuB. XiaoS. NaisbittJP YuanRS Petralia1999Coupling of mGluR/Homer and PSD-95 complexes by the Shank family of postsynaptic density proteins.Neuron23583592
- 13. Naisbitt S, Kim E, Tu JC, Xiao B, Sala C, et al. (1999) Shank, a novel family of postsynaptic density proteins that binds to the NMDA receptor/PSD-95/GKAP complex and cortactin. Neuron 23: 569–582.S. NaisbittE. KimJC TuB. XiaoC. Sala1999Shank, a novel family of postsynaptic density proteins that binds to the NMDA receptor/PSD-95/GKAP complex and cortactin.Neuron23569582
- 14. Husi H, Ward MA, Choudhary JS, Blackstock WP, Grant SG (2000) Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat Neurosci 3: 661–669.H. HusiMA WardJS ChoudharyWP BlackstockSG Grant2000Proteomic analysis of NMDA receptor-adhesion protein signaling complexes.Nat Neurosci3661669
- 15. Husi H, Grant SG (2001) Isolation of 2000-kDa complexes of N-methyl-D-aspartate receptor and postsynaptic density 95 from mouse brain. J Neurochem 77: 281–291.H. HusiSG Grant2001Isolation of 2000-kDa complexes of N-methyl-D-aspartate receptor and postsynaptic density 95 from mouse brain.J Neurochem77281291
- 16. Collins MO, Husi H, Yu L, Brandon JM, Anderson CN, et al. (2005) Molecular characterization and comparison of the components and multiprotein complexes in the postsynaptic proteome. J Neurochem. MO CollinsH. HusiL. YuJM BrandonCN Anderson2005Molecular characterization and comparison of the components and multiprotein complexes in the postsynaptic proteome.J Neurochem
- 17. Pocklington AJ, Armstrong JD, Grant SG (2006) Organization of brain complexity–synapse proteome form and function. Brief Funct Genomic Proteomic 5: 66–73.AJ PocklingtonJD ArmstrongSG Grant2006Organization of brain complexity–synapse proteome form and function.Brief Funct Genomic Proteomic56673
- 18. Husi H, Grant SG (2001) Proteomics of the nervous system. Trends Neurosci 24: 259–266.H. HusiSG Grant2001Proteomics of the nervous system.Trends Neurosci24259266
- 19. Woods DF, Bryant PJ (1991) The discs-large tumor suppressor gene of Drosophila encodes a guanylate kinase homolog localized at septate junctions. Cell 66: 451–464.DF WoodsPJ Bryant1991The discs-large tumor suppressor gene of Drosophila encodes a guanylate kinase homolog localized at septate junctions.Cell66451464
- 20. Komiyama NH, Watabe AM, Carlisle HJ, Porter K, Charlesworth P, et al. (2002) SynGAP regulates ERK/MAPK signaling, synaptic plasticity, and learning in the complex with postsynaptic density 95 and NMDA receptor. J Neurosci 22: 9721–9732.NH KomiyamaAM WatabeHJ CarlisleK. PorterP. Charlesworth2002SynGAP regulates ERK/MAPK signaling, synaptic plasticity, and learning in the complex with postsynaptic density 95 and NMDA receptor.J Neurosci2297219732
- 21. Opazo P, Watabe AM, Grant SG, O'Dell TJ (2003) Phosphatidylinositol 3-kinase regulates the induction of long-term potentiation through extracellular signal-related kinase-independent mechanisms. J Neurosci 23: 3679–3688.P. OpazoAM WatabeSG GrantTJ O'Dell2003Phosphatidylinositol 3-kinase regulates the induction of long-term potentiation through extracellular signal-related kinase-independent mechanisms.J Neurosci2336793688
- 22. Yao WD, Gainetdinov RR, Arbuckle MI, Sotnikova TD, Cyr M, et al. (2004) Identification of PSD-95 as a regulator of dopamine-mediated synaptic and behavioral plasticity. Neuron 41: 625–638.WD YaoRR GainetdinovMI ArbuckleTD SotnikovaM. Cyr2004Identification of PSD-95 as a regulator of dopamine-mediated synaptic and behavioral plasticity.Neuron41625638
- 23. Fagiolini M, Katagiri H, Miyamoto H, Mori H, Grant SG, et al. (2003) Separable features of visual cortical plasticity revealed by N-methyl-D-aspartate receptor 2A signaling. Proc Natl Acad Sci U S A 100: 2854–2859.M. FagioliniH. KatagiriH. MiyamotoH. MoriSG Grant2003Separable features of visual cortical plasticity revealed by N-methyl-D-aspartate receptor 2A signaling.Proc Natl Acad Sci U S A10028542859
- 24. Garry EM, Moss A, Delaney A, O'Neill F, Blakemore J, et al. (2003) Neuropathic sensitization of behavioral reflexes and spinal NMDA receptor/CaM kinase II interactions are disrupted in PSD-95 mutant mice. Curr Biol 13: 321–328.EM GarryA. MossA. DelaneyF. O'NeillJ. Blakemore2003Neuropathic sensitization of behavioral reflexes and spinal NMDA receptor/CaM kinase II interactions are disrupted in PSD-95 mutant mice.Curr Biol13321328
- 25. Kuehn MR, Bradley A, Robertson EJ, Evans MJ (1987) A potential animal model for Lesch-Nyhan syndrome through introduction of HPRT mutations into mice. Nature 326: 295–298.MR KuehnA. BradleyEJ RobertsonMJ Evans1987A potential animal model for Lesch-Nyhan syndrome through introduction of HPRT mutations into mice.Nature326295298
- 26. Hooper M, Hardy K, Handyside A, Hunter S, Monk M (1987) HPRT-deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells. Nature 326: 292–295.M. HooperK. HardyA. HandysideS. HunterM. Monk1987HPRT-deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells.Nature326292295
- 27. Moore RC, Redhead NJ, Selfridge J, Hope J, Manson JC, et al. (1995) Double replacement gene targeting for the production of a series of mouse strains with different prion protein gene alterations. Biotechnology (N Y) 13: 999–1004.RC MooreNJ RedheadJ. SelfridgeJ. HopeJC Manson1995Double replacement gene targeting for the production of a series of mouse strains with different prion protein gene alterations.Biotechnology (N Y)139991004
- 28. Selbert S, Bentley DJ, Melton DW, Rannie D, Lourenco P, et al. (1998) Efficient BLG-Cre mediated gene deletion in the mammary gland. Transgenic Res 7: 387–396.S. SelbertDJ BentleyDW MeltonD. RannieP. Lourenco1998Efficient BLG-Cre mediated gene deletion in the mammary gland.Transgenic Res7387396
- 29. Tsuda H, Maynard-Currie CE, Reid LH, Yoshida T, Edamura K, et al. (1997) Inactivation of the mouse HPRT locus by a 203-bp retroposon insertion and a 55-kb gene-targeted deletion: establishment of new HPRT-deficient mouse embryonic stem cell lines. Genomics 42: 413–421.H. TsudaCE Maynard-CurrieLH ReidT. YoshidaK. Edamura1997Inactivation of the mouse HPRT locus by a 203-bp retroposon insertion and a 55-kb gene-targeted deletion: establishment of new HPRT-deficient mouse embryonic stem cell lines.Genomics42413421
- 30. McEwan C, Melton DW (2003) A simple genotyping assay for the Hprt null allele in mice produced from the HM-1 and E14TG2a mouse embryonic stem cell lines. Transgenic Res 12: 519–520.C. McEwanDW Melton2003A simple genotyping assay for the Hprt null allele in mice produced from the HM-1 and E14TG2a mouse embryonic stem cell lines.Transgenic Res12519520
- 31. Cecchi C, Avner P (1996) Genomic organization of the mottled gene, the mouse homologue of the human Menkes disease gene. Genomics 37: 96–104.C. CecchiP. Avner1996Genomic organization of the mottled gene, the mouse homologue of the human Menkes disease gene.Genomics3796104
- 32. Behr R, Sackett S, Bochkis I, Phillip L, Kaestner K (2007) Impaired male fertility and atrophy of semineforous tubules caused by haploinsufficiency for Foxa3. Dev Biol In press. R. BehrS. SackettI. BochkisL. PhillipK. Kaestner2007Impaired male fertility and atrophy of semineforous tubules caused by haploinsufficiency for Foxa3.Dev Biol In press
- 33. Witt CC, Burkart C, Labeit D, McNabb M, Wu Y, et al. (2006) Nebulin regulates thin filament length, contractility, and Z-disk structure in vivo. Embo J 25: 3843–3855.CC WittC. BurkartD. LabeitM. McNabbY. Wu2006Nebulin regulates thin filament length, contractility, and Z-disk structure in vivo.Embo J2538433855
- 34. Niedernhofer LJ, Garinis GA, Raams A, Lalai AS, Robinson AR, et al. (2006) A new progeroid syndrome reveals that genotoxic stress suppresses the somatotroph axis. Nature 444: 1038–1043.LJ NiedernhoferGA GarinisA. RaamsAS LalaiAR Robinson2006A new progeroid syndrome reveals that genotoxic stress suppresses the somatotroph axis.Nature44410381043
- 35. Hanada K, Vermeij M, Garinis GA, de Waard MC, Kunen MG, et al. (2007) Perturbations of vascular homeostasis and aortic valve abnormalities in fibulin-4 deficient mice. Circ Res 100: 738–746.K. HanadaM. VermeijGA GarinisMC de WaardMG Kunen2007Perturbations of vascular homeostasis and aortic valve abnormalities in fibulin-4 deficient mice.Circ Res100738746
- 36. Yang S, Farias M, Kapfhamer D, Tobias J, Grant G, et al. (2007) Biochemical, molecular and behavioral phenotypes of Rab3A mutations in the mouse. Genes Brain Behav 6: 77–96.S. YangM. FariasD. KapfhamerJ. TobiasG. Grant2007Biochemical, molecular and behavioral phenotypes of Rab3A mutations in the mouse.Genes Brain Behav67796
- 37. Schalkwyk LC, Fernandes C, Nash MW, Kurrikoff K, Vasar E, et al. (2007) Interpretation of knockout experiments: the congenic footprint. Genes Brain Behav 6: 299–303.LC SchalkwykC. FernandesMW NashK. KurrikoffE. Vasar2007Interpretation of knockout experiments: the congenic footprint.Genes Brain Behav6299303
- 38. Simpson EM, Linder CC, Sargent EE, Davisson MT, Mobraaten LE, et al. (1997) Genetic variation among 129 substrains and its importance for targeted mutagenesis in mice. Nat Genet 16: 19–27.EM SimpsonCC LinderEE SargentMT DavissonLE Mobraaten1997Genetic variation among 129 substrains and its importance for targeted mutagenesis in mice.Nat Genet161927
- 39. Threadgill DW, Yee D, Matin A, Nadeau JH, Magnuson T (1997) Genealogy of the 129 inbred strains: 129/SvJ is a contaminated inbred strain. Mamm Genome 8: 390–393.DW ThreadgillD. YeeA. MatinJH NadeauT. Magnuson1997Genealogy of the 129 inbred strains: 129/SvJ is a contaminated inbred strain.Mamm Genome8390393
- 40. Wolfer DP, Crusio WE, Lipp HP (2002) Knockout mice: simple solutions to the problems of genetic background and flanking genes. Trends Neurosci 25: 336–340.DP WolferWE CrusioHP Lipp2002Knockout mice: simple solutions to the problems of genetic background and flanking genes.Trends Neurosci25336340
- 41. Li C, Hung Wong W (2001) Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol 2: RESEARCH0032.C. LiW. Hung Wong2001Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application.Genome Biol2RESEARCH0032