Open Collections

UBC Faculty Research and Publications

Identification of genetic loci that modulate cell proliferation in the adult rostral migratory stream… Poon, Anna; Goldowitz, Daniel Mar 19, 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12864_2013_Article_5968.pdf [ 1.25MB ]
Metadata
JSON: 52383-1.0223207.json
JSON-LD: 52383-1.0223207-ld.json
RDF/XML (Pretty): 52383-1.0223207-rdf.xml
RDF/JSON: 52383-1.0223207-rdf.json
Turtle: 52383-1.0223207-turtle.txt
N-Triples: 52383-1.0223207-rdf-ntriples.txt
Original Record: 52383-1.0223207-source.json
Full Text
52383-1.0223207-fulltext.txt
Citation
52383-1.0223207.ris

Full Text

RESEARCH ARTICLE Open AccessIdentification of genetic loci that modulate cellproliferation in the adult rostral migratory streamusing the expanded panel of BXD miceAnna Poon and Daniel Goldowitz*AbstractBackground: Adult neurogenesis, which is the continual production of new neurons in the mature brain, demonstratesthe strikingly plastic nature of the nervous system. Adult neural stem cells and their neural precursors, collectivelyreferred to as neural progenitor cells (NPCs), are present in the subgranular zone (SGZ) of the dentate gyrus, thesubventricular zone (SVZ), and rostral migratory stream (RMS). In order to harness the potential of NPCs to treatneurodegenerative diseases and brain injuries, it will be important to understand the molecules that regulate NPCs inthe adult brain. The genetic basis underlying NPC proliferation is still not fully understood. From our previous quantitativetrait locus (QTL) analysis, we had success in using a relatively small reference population of recombinant inbred strains ofmice (AXBXA) to identify a genetic region that is significantly correlated with NPC proliferation in the RMS.Results: In this study, we expanded our initial QTL mapping of RMS proliferation to a far richer genetic resource, theBXD RI mouse strains. A 3-fold difference in the number of proliferative, bromodeoxyuridine (BrdU)-labeled cells wasquantified in the adult RMS of 61 BXD RI strains. RMS cell proliferation is highly dependent on the genetic backgroundof the mice with an estimated heritability of 0.58. Genome-wide mapping revealed a significant QTL on chromosome(Chr) 6 and a suggestive QTL on Chr 11 regulating the number of NPCs in the RMS. Composite interval analysis furtherrevealed secondary QTLs on Chr 14 and Chr 18. The loci regulating RMS cell proliferation did not overlap withthe suggestive loci modulating cell proliferation in the SGZ. These mapped loci serve as starting points to identifygenes important for this process. A subset of candidate genes in this region is associated with cell proliferationand neurogenesis. Interconnectivity of these candidate genes was demonstrated using pathway and transcriptionalcovariance analyses.Conclusions: Differences in RMS cell proliferation across the BXD RI strains identifies genetic loci that serve to provideinsights into the interplay of underlying genes that may be important for regulating NPC proliferation in the adultmouse brain.Keywords: Neural progenitor cells, Adult neurogenesis, Rostral migratory stream, Cell proliferation, Recombinant inbredmice, Quantitative trait locus mappingBackgroundThe persistent division of neural progenitor cells (NPCs)and the production of new neurons in the adult brainraise hope for potential therapies targeting the NPCs tocompensate for neuronal loss in injured or disease brains.This process of continual neuron production, also knownas adult neurogenesis, occurs in discrete brain regions thatinclude the subgranular zone (SGZ) of the dentate gyrus,the subventricular zone (SVZ) of the lateral ventricle, andthe rostral migratory stream (RMS) which is the rostralextension of the SVZ [1,2]. Previous studies have detectedincreased NPC proliferation under pathological conditions,and the neural precursor cells generated were recruitedto the affected brain regions [3-5]. To develop effectivestrategies that harness the NPCs as a renewable source forrepair, it will be necessary to understand how neurogenesisis regulated in the mature brain.* Correspondence: dang@cmmt.ubc.caCentre for Molecular Medicine and Therapeutics, Child and Family ResearchInstitute, Department of Medical Genetics, University of British Columbia,Vancouver, BC V5Z 4H4, Canada© 2014 Poon and Goldowitz; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of theCreative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use,distribution, and reproduction in any medium, provided the original work is properly credited.Poon and Goldowitz BMC Genomics 2014, 15:206http://www.biomedcentral.com/1471-2164/15/206There have been considerable advances in knowledgeconcerning the regulation of adult neurogenesis over thepast two decades [6,7]. Adult neurogenesis is a multifac-torial process that encompasses several stages includingproliferation, migration, and then the differentiation andsurvival of new neurons. Each stage is dynamically regu-lated by both extrinsic and intrinsic factors [7,8]. Regulationat early stages of neurogenesis, most notably the prolif-eration of NPCs, is especially complex as a wide rangeof extracellular factors, stimuli, transcription factors, andepigenetic modifiers have been identified [9]. A number ofmorphogens including Wnt, Notch, Sonic hedgehog, andEphrins have also been shown to regulate cell proliferationin the adult SVZ [7]. External stimuli such as age, exercise,sleep, and stress have been shown to influence NPC prolif-eration [10-12]. To control the proliferative behaviourof NPCs, these extrinsic factors must act on an intrinsicsystem where pro- and anti- proliferative genes are differ-entially regulated to provide instructions to the NPCson the appropriate time and frequency to divide. Thegenetic basis of NPC proliferation, however, is not fullyunderstood.It has also been previously shown that adult neurogen-esis is significantly dependent on the genetic background,and the genetic diversity among inbred strains can serveas a reservoir for gene discovery [13-15]. A wide range ofdifferences in the number of proliferative NPC has beenquantified in the RMS of nine inbred strains (includingDBA/2 J, C57BL/6 J, A/J) and 27 AXB/BXA recombinantinbred (RI) strains derived from the initial mating ofC57BL/6 J and A/J mice [14]. Genome-wide mappingof strain differences in RMS cell proliferation using theAXB/BXA panel led to the identification of a significantquantitative trait locus (QTL) on distal chromosome (Chr)11 that accounts for ~20% of the phenotypic differencesobserved among the strains [14]. From this earlier work,we suspected the involvement of more than one locus inregulating NPC proliferation. Therefore, in order to fullyappreciate the complexity of this process, we explored aseparate genetic reference panel called the BXDs. TheBXDs is one of the largest murine mapping panel consistsof 80 unique BXD RI strains, which is three times thesize of the AXB/BXA resource. The BXD RI strains werederived from an initial mating between C57BL/ 6 J andDBA/2 J that was followed by inbreeding F2 progenyfor ≥20 generations. The substantial differences in cellproliferation in the RMS of the BXD RI strains allowed usto detect additional QTLs that modulate NPC prolifera-tion. Functional annotation and gene expression analysesusing pre-existing transcriptome data highlighted a subsetof candidate genes in the mapped chromosomal regions.Common features shared by these candidate genes includeexpression in the RMS, functional implication in cellproliferation/cell cycle progression, and participation insignalling pathways important for neurogenesis. Thesefindings provide insights into the dynamic interplay ofgenetic loci and underlying genes that may modulateNPC proliferation in the adult brain.ResultsSignificant strain-differences in RMS cell proliferationamong the BXDsA substantial range of cell proliferation in the RMS, asdetected by the uptake of BrdU, was identified in 61BXD RI strains, with a 3-fold difference among the strains(Figures 1 and 2). This range extends beyond the differ-ences detected between the two parental strains, C57BL/6 J and DBA/2 J (Figure 1). Significant inter-strain differ-ences in RMS linear density was detected (F60,204 = 4.77 ,P < 0.0001), and heritability of cell proliferation in theRMS is estimated to be 0.58 (P < 0.0001).There is no significant sex effect (t(263) = 0.82, P = 0.4111;females = 63.97 ± 1.66; males = 65.83 ± 1.55) or body weighteffect (R2 = 0.004; P = 0.29) on the number of proliferativecells in the RMS. We also examined the batch effect as theBXD panel has three epoch substructures: i) BXD 1–30generated by Benjamin A. Taylor in the 1970s [16], ii)BXD 33–42 generated by Benjamin A. Taylor in the late1990s [17], and iii) BXD 43–103 generated by Peirce andcolleagues in the early 2000s [18]. No significant batcheffect on RMS linear density was detected (F2, 262 = 0.44;P = 0.65). However, age had a significant effect on RMSlinear density (R2 = 0.015; P = 0.0442).We also correlated our RMS linear density data to 3911traits previously generated using the BXD RI referencepanel. RMS linear density (GeneNetwork Trait ID: 13545)is significantly associated with traits from other brainregions such as the hippocampal volume (Trait ID: 10456,r = −0.45, P = 0.017), dentate gyrus volume (Trait ID:10460, r = −0.45, P = 0.019), hippocampus granule cellnumber (Trait ID: 10378, r = −0.46, P = 0.017), dorsalstriatum volume (Trait ID: 10998, r = −0.30, P = 0.04),and striatal neuron number (Trait ID: 13437, r = −0.44,P = 0.0006). These phenotypic correlations suggest com-mon genetic determinants underlying RMS cell prolif-eration and other traits.QTLs regulating NPC proliferation in the RMSTo gain insights into the complex genetics regulatingadult NPC proliferation, we performed QTL analysis forthe differences in RMS linear density among the BXD RIstrains. We discovered a highly significant locus on Chr6 (76.8-88.8 Mb; P = 0.05) and a suggestive locus on Chr11 (50-58 Mb; P = 0.63) that modulate NPC numbers inthe RMS (Figure 3A). Since age had a significant effecton RMS linear density, we regressed the RMS lineardensity for each animal against age to ensure the observedgenotype-phenotype association is not confounded byPoon and Goldowitz BMC Genomics 2014, 15:206 Page 2 of 14http://www.biomedcentral.com/1471-2164/15/206differences in age. QTL mapping of the adjusted RMSlinear density identified the same significant and suggest-ive QTLs on Chr 6 and Chr 11, respectively (Figure 3B).Separate QTL analyses were also performed to confirmthat mapping results are not confounded by strain epochs.QTL mapping of the old (Taylor) BXD strains and thenew (UTHSC) BXD strains both identified the samesignificant QTL on Chr 6, which again demonstrate therobustness of the Chr 6 QTL (Additional file 1).We used marker regression analysis to estimate the alleliceffect sizes of the mapped QTLs. The genotype of single-nucleotide polymorphism (SNP) and microsatellite markersunderlying the Chr 6 QTL revealed that the C57BL/6 Jallele is associated with ~19 BrdU + cells/mm increase inRMS cell proliferation compared with having a DBA/2 Jallele. Conversely, genetic markers in the Chr 11 QTLregion showed that a DBA/2 J allele has an additiveeffect of ~11 BrdU + cells/mm compared with having aC57BL/6 J allele. These findings suggest the complexgenetic modulation of NPC proliferation with the involve-ment of more than one QTL.Composite interval analysis revealed secondary intervalson Chr 14 (40.3- 49.2 Mb) and Chr 18 (58.2-74.9 Mb)work additively with the Chr 6 QTL in modulated NPCproliferation in the RMS (Figure 3C). One-way ANOVAconfirmed the allelic effects of the Chr 6 (F1, 58 = 39.87;Figure 1 Quantification of the number of proliferating (BrdU+) cells in the RMS of BXD RI strains. RMS linear density (i.e. mean number ofBrdU + cells per mm length of RMS± SEM) of 61 BXD RI strains (white bars) and their parental strains, C57BL/6 J (red bar), and DBA/2 J (green bar). Thesample size per strain is indicated in the bars.Figure 2 Representative sagittal sections of BrdU-labeled RMS of three separate BXD RI lines. All mice received a single pulse of BrdU forone hour. BrdU immunohistochemistry revealed inter-strain differences in BrdU + cell numbers from BXD 39 having high numbers of BrdU + cellsin the RMS to BXD 55 having low numbers of BrdU + cells in RMS. Whereas BXD 77 have intermediate numbers of BrdU + cells in the RMS. Arrowsmark the beginning and end of the RMS.LV, lateral ventricle; scale bar: 200 μm.Poon and Goldowitz BMC Genomics 2014, 15:206 Page 3 of 14http://www.biomedcentral.com/1471-2164/15/206Figure 3 (See legend on next page.)Poon and Goldowitz BMC Genomics 2014, 15:206 Page 4 of 14http://www.biomedcentral.com/1471-2164/15/206P < 0.0001), Chr 14 (F1,58 = 16.01; P = 0.0002), and Chr18 (F1, 58 = 12.56; P = 0.0008) on RMS proliferation.Pair-Scan analysis was performed to assess two-way inter-action between pairs of genetic markers from differentchromosomes. Once again, we found significant associationof markers in Chr 6 QTL with markers in the secondaryChr 14 and Chr 18 intervals (LRS Full > 40.1; P < 0.01).To further assess the type of interaction among theseloci (i.e. additive or epistatic), the BXD strains were splitinto groups according to their genotypes at these QTLintervals, and an average RMS linear density was calcu-lated for each group. The plotted RMS linear densitiesshow how different allele combinations at these loci areassociated with different levels of RMS linear density(Figure 4). The major effect of the Chr 6 QTL is demon-strated where strains carrying the C57BL/6 J allele inthe Chr 6 interval are associated with higher RMS lineardensities compared to having the DBA/2 J allele, irre-spective of the genotypes at the Chr14 or Chr18 QTLintervals. When the genotypes at these secondary QTLregions are taken into account, having the C57BL/6 Jallele at either the Chr 14 (Figure 4A) or Chr 18 intervals(Figure 4B) are associated with even higher RMS lineardensities compared to having the DBA/2 J alleles. TheC57BL/6 J alleles in the Chr 14 and Chr 18 QTL regionsare respectively associated with ~15 BrdU + cells/mmand ~10 BrdU + cells/mm increase in RMS cell prolifer-ation compared with having DBA/2 J alleles. These find-ings suggest the major Chr 6 QTL works additively withloci on Chr 14 and 18 when modulating NPC proliferationin the RMS.Mapped loci harbour genes that may serve as modulatorsof NPC proliferationWe predict the mapped genomic regions harbours genesregulating the number of proliferative NPCs in the RMS.To test our hypothesis, we implemented a combinationof bioinformatics tools including QTLminer at GeneNet-work (http://www.genenetwork.org/webqtl/main.py?For-mID=qtlminer) and DAVID (http://david.abcc.ncifcrf.gov)to obtain information on genes in the QTL regions. Weprioritized genes according to 1) the presence of single-nucleotide polymorphisms (SNPs) and/or insertion/deletions(indels), 2) expression of the gene in the RMS(See figure on previous page.)Figure 3 QTL analyses of cell proliferation in the RMS of BXD RI strains. The x-axis for figures (A-C) represents the chromosomes 1–19, & X(top panel) and their physical maps in megabases (bottom panel). The y-axis and the blue line depict the likelihood ratio statistic (LRS), whichindicates the strength of association between genotypes of markers across the genome and the phenotype (i.e. RMS linear density). The lightred and gray horizontal lines mark the significant (P = 0.05) and suggestive (P = 0.63) threshold, respectively. Whole-genome interval mapping ofunadjusted RMS linear density (A) and for the adjusted RMS linear density corrected for the effects of age (B) have mapped a significant QTL onChr 6 (76.8-88.8 Mb) and a suggestive QTL on Chr 11 (50-58 Mb) regulating RMS linear density. (C) Composite interval mapping revealed an additionalsignificant locus on Chr 14 (39–49.5 Mb) and a suggestive locus on Chr 18 (58–86 Mb) that work additively with the Chr 6 QTL in modulatingRMS linear density.Figure 4 Allelic effects on cell proliferation in the RMS. Plots ofRMS linear density (i.e. # BrdU + cells per mm length of RMS) versusallele genotypes at markers associated with the major Chr 6 QTLs andsecondary QTLs on Chr 14, and Chr 18. BXD strains were divided intodifferent groups based on their genotypes at markers closest to theQTL peaks on Chr 6, 14, and 18. B6 and D2 represent homozygousalleles of C57BL/6 J and DBA/2 J for the markers. Dots represent groupmeans ± SEM. (A) Effect of genotype on RMS linear density (y-axis) atmarkers in the Chr 6 QTL (x-axis) and the Chr 14 QTL intervals (B6 andD2 alleles are represented by red and green colour lines, respectively).(B) Effect of genotype on RMS linear density (y-axis) at markers in theChr 6 QTL (x-axis) and the Chr 18 QTL intervals (B6 and D2 alleles arerepresented by red and green colour lines, respectively).Poon and Goldowitz BMC Genomics 2014, 15:206 Page 5 of 14http://www.biomedcentral.com/1471-2164/15/206neuroblast transcriptome, 3) presence of transcripts inother neurogenic regions (e.g., SVZ or SGZ) in the adultmouse brain, and 4) association with gene ontology (GO)terms such as neurogenesis, cell cycle, or proliferation.In the significant Chr 6 QTL region (76.8-88.8 Mb) thereare 105 genes that have known or predicted functions butonly 4 genes met all our criteria (Table 1). Two of thesegenes, transforming growth factor alpha (Tgfa) andminichromosome maintenance deficient 2 mitotin (Mcm2)have been implicated in adult neurogenesis [19,20].Pathway analysis showed Tgfa to be involved in theErbB signalling pathway, which regulates diverse bio-logical processes such as proliferation, differentiation,and survival. Whole genome mapping and compositeinterval analyses also revealed secondary QTLs influencingcell proliferation in the RMS. To determine whether thesesuggestive loci interact with the major Chr 6 locus at themolecular level, we further examined the candidate genesresiding in the suggestive QTLs. In the suggestive Chr 11QTL region (50-58 Mb), 93 genes have known/predictedfunctions and three genes met our candidate gene criteria(Table 1). One of the candidate genes is secreted acidiccysteine rich glycoprotein (Sparc) which has been previ-ously shown to promote cell proliferation in the SGZof the dentate gyrus [21]. In the Chr 14 QTL region(40.3-49.2 Mb), which was revealed through compositeinterval mapping, there are 46 genes with known/pre-dicted functions and 2 of these met all the criteria listedabove: cyclin-dependent kinase inhibitor 3 (Cdkn3) andglucosamine-phosphate N-acetyltransferase 1 (Gnpnat1).Despite not being directly linked to neurogenesis, both ofthese genes regulate cell cycle progression [22,23]. In theChr 18 QTL region (58.2-74.9 Mb), there are 101 genesthat have known/predicted functions. Five genes met all ofour candidate gene criteria (Table 1). One of these genes,SMAD family member 4 (Smad4) has been directly im-plicated in adult neurogenesis [24]. Pathway analysesrevealed the gene calcium/calmodulin-dependent proteinkinase II alpha (Camk2a) in Chr 18 QTL is involved in theErbB signalling pathway, a pathway also shared by Tgfa inthe major Chr 6 QTL. Camk2a also participates in theWnt signalling pathway. Two other genes Ppp2ca in thesuggestive Chr 11 QTL and Smad4 in the Chr 18 QTL arealso components of the Wnt signaling machinery. Thesesignalling pathways identified from our analyses havebeen shown to regulate NPC proliferation in the adultbrain [25-28].Co-expression of candidate genes further revealedpathway dynamicsWe further explored the interconnectivity of the candidategenes based on their expression level in the adult brain.We suspect polymorphisms in these candidate genesare likely to affect gene expression level, similar to theheritable differences in number of RMS proliferative cellsobserved among the BXD. Since there is no RMS tran-scriptome data available for the BXDs, we used the BXDhippocampal microarray data available at the GeneNet-work [Hippocampus Consortium M430v2 (Jun06) PDNN]to investigate expression co-variation among our candi-date genes. The hippocampal transcriptome was selectedbased on findings from our earlier phenotypic correlationanalysis where cell proliferation in the RMS is signifi-cantly associated with several hippocampal traits includinghippocampus and dentate gyrus volumes and dentategyrus granule cell number. The transcripts levels of eachcandidate genes were extracted from the BXD hippocam-pal expression database. A network graph (Figure 5) wasthen generated showing how the expression of the candi-date genes residing in different loci positively or negativelycorrelated with each other. We found genes participatingin cell proliferation exhibited similar expression levelswhere Gfpt1 (a gene in the significant Chr 6 QTL) ishighly correlated with the expression of two other can-didate genes, Sparc (a gene in the Chr 11 QTL; r = 0.767;P < 1.00E−16) and Ppp2ca (a candidate gene in the suggest-ive Chr 11 QTL; r = 0.706; P = 4.29E−13). Ppp2ca, whichis involved in the Wnt signalling pathway, is also sig-nificantly correlated with Camk2a in the Chr 18 QTL(r = 0.66; P = 6.26E−11). This finding is consistent withthe general assumption that genes in the same pathwayexhibit similar gene expression profiles [30]. In additionto gene expression covariance, we also detected significantcorrelation between gene expression and the phenotype(i.e. RMS linear density). The expression of candidate genesAnxa4 (Chr 6 QTL), Ppp2ca (Chr 11 QTL), Gnpnat1(Chr14 QTL), and Camk2a (Chr18 QTL) varied acrossthe different BXD strains, and their transcriptional dif-ferences are significantly correlated with the phenotypicdifferences in RMS linear density observed among theBXDs (r = −0.58, r = −.50, r = −0.28, r = 0.29, respectively;P < 0.03; N = 56 BXD RI strains; Additional file 2).Cell proliferation in the SGZ is regulated by a different setof QTLsThe high correlation between the RMS linear density andthe hippocampal traits such as volume and cell numbers,which harbours the SGZ of the dentate gyrus, led us tosuspect common genetic determinants underlying thesetwo neurogenic sites. We first assessed cell proliferation inthe SGZ one-hour post BrdU injection. Similar to whatwas observed in the RMS, the number of BrdU + cellsin the SGZ significantly differs among the BXD RIstrains (F60,170 = 2.88, P < 0.0001) (Figure 6A). Heritabilityof cell proliferation in the SGZ is estimated to be 0.5(P < 0.0001). The number of BrdU + cells in the SGZwas 3.5 fold higher in C57BL/6 J compared to DBA/2 J,which was opposite to the parental strain differencesPoon and Goldowitz BMC Genomics 2014, 15:206 Page 6 of 14http://www.biomedcentral.com/1471-2164/15/206Table 1 Strong candidate genes identified in the chromosomes (Chr) 6, 11, 14, and 18 QTL intervalsGene symbol Gene name Chr Gene location:start (Mb)Genelength (kb)SNPs Indels Gene Ontology (GO)annotationImplicated in SVZ-RMSneurogenesis [ref.]Implicated in SGZneurogenesis [ref.]Tgfa Transforming growth factor alpha 6 86.145 79.742 6 3 Positive regulation of cell division,negative regulation of apoptosis✔ [19]Anxa4 Annexin A4 6 86.687 56.745 134 7 Cell growth and survival, cellproliferation, carcinogenesisGfpt1 Glutamine fructose-6-phosphatetransaminase 16 86.993 49.362 48 1 Amino sugar and nucleotide sugarmetabolism, cell regenerationMcm2 Minichromosome maintenancedeficient 2 mitotin6 88.833 15.307 3 0 DNA replication initiation, DNAunwinding during replication✔ [20] ✔ [29]Ppp2ca Protein phosphatase 2 (formerly 2A),catalytic subunit, alpha isoform11 51.912 23.926 1 0 Phosphoprotein phosphatase activity, meiosis,negative control of cell growth and divisionRad50 RAD50 homolog (S. cerevisiae) 11 53.463 57.801 4 0 DNA repair, homologous recombination,cell cycleSparc Secreted acidic cysteine richglycoprotein11 55.208 25.580 83 5 Response to growth factor stimulus,regulation of cell proliferation✔ [21]Cdkn3 Cyclin-dependent kinase inhibitor 3 14 45.692 0.05 3 3 Cell cycle arrest, phosphatase activityGnpnat1 Glucosamine-phosphateN-acetyltransferase 114 45.996 12.376 1 0 Amino sugar and nucleotide sugar metabolism,actin dynamics, cell cycle progressionCamk2a Calcium/calmodulin-dependentprotein kinase II alpha18 61.085 62.521 0 1 G1/S transition of mitotic cell cycle, neuronalsynaptic plasticitySeh1l SEH1-like (S. cerevisiae) 18 67.935 17.718 4 2 Cell division, chromosome segregation, mitosisSmad4 Similar to MAD homolog4 (Drosophila)18 73.799 64.729 160 74 Cell proliferation, tissue morphogenesis ✔ [24]Elac1 elaC homolog 1 (E. coli) 18 73.895 19.442 42 7 tRNA 3'-end processing, cell growthand proliferationMapk4 mitogen-activated protein kinase 4 18 74.088 136.463 141 70 Cell cycle, protein amino acid phosphorylationAbbreviations: SNPs single-nucleotide polymorphisms, indels insertion/deletions.PoonandGoldowitzBMCGenomics2014,15:206Page7of14http://www.biomedcentral.com/1471-2164/15/206observed in the RMS. However this reversal in phenotypicdirection was not observed in the BXD RI panel. Forexample, BXD68 had both high numbers of BrdU + cellsin the RMS and SGZ. Whereas, BXD55 had low numbersof BrdU + cells in the RMS and SGZ.The effects of confounding variables on cell proliferationin the SGZ were also examined. Age had a significant ef-fect on the number of BrdU + cells in the SGZ (R2 = 0.059,P = 0.0002). Strain epoch, sex, body weight differences didnot significantly influence the number of proliferative cellsin the SGZ.QTL mapping of cell proliferation in the adult SGZrevealed suggestive QTLs on Chr 1 (40-59 Mb), Chr 5(104.5-119.5 Mb), and Chr 9 (71–83.7 Mb) and no sig-nificant QTL was identified (Figure 6B). QTL mappingof SGZ proliferation corrected for differences in agerevealed the same suggestive QTLs on Chr 1, 5, and 9(Figure 6C). These SGZ QTLs do not correspond tothe loci associated with cell proliferation in the RMSand therefore suggest a different set of genetic modulatorsregulating the number of proliferative cells in the SGZ.This is further supported by the lack of phenotypic cor-relation between cell proliferation in the SGZ and RMS(r = 0.14, P = 0.2917, N = 61 BXD strains).DiscussionAdult neurogenesis is a multi-stage process that is com-plex in terms of regulation as each stage is modulated bydifferent genetic and environmental factors [7,9]. Theearliest phase of adult neurogenesis is the proliferationFigure 5 Transcriptional co-expression network graph of candidate genes. The transcripts levels of each candidate genes were extractedfrom the BXD hippocampal expression database available at the GeneNetwork. A network graph was then generated showing how the expressionof the candidate genes positively or negatively correlated with each other. Strength of correlation between two connected genes is indicated inthe legend.Poon and Goldowitz BMC Genomics 2014, 15:206 Page 8 of 14http://www.biomedcentral.com/1471-2164/15/206Figure 6 (See legend on next page.)Poon and Goldowitz BMC Genomics 2014, 15:206 Page 9 of 14http://www.biomedcentral.com/1471-2164/15/206of the NPCs in the adult SGZ, SVZ, and RMS. NPCproliferation is highly heritable and dependent on thegenetic background. The effects of genetic variationson NPC proliferation is likely through a network of geneswhich will be difficult to reveal from conventional singlegene study. Here, we used a systems genetics approachwhere we correlate genetic variation and expression datato NPC proliferation in the adult mouse brain. We tookadvantage of the genetic and phenotypic diversity in theBXD reference panel, and through genome-wide QTLmapping, we discovered novel genetic loci that likelyhouse modulators of NPC proliferation in the RMS. Acloser examination of candidate genes in the mapped loci(using public databases on gene ontology, genetic poly-morphisms, and gene expression) identified a subset ofgenes associated with adult neurogenesis and/or cellproliferation. We further demonstrated interconnectivityamong some of these candidate genes by showing they co-vary at the transcriptional level and participate in the samepathways previously shown to control adult neurogenesis.To probe the genetic architecture of a polygenic pheno-type as complex as NPC proliferation requires a largegenetic reference panel that can provide high resolutionand power to identify genes with subtle but significanteffects on NPC proliferation. In this study, we utilizedthe genetic diversity among the BXD RI strains as a toolto identify genes and pathways involved in NPC prolifera-tion. Our BXD panel of 61 strains is the largest mappingpanel employed thus far to study adult neurogenesis, andit is over twice the size the AXB/BXA RI set we havepreviously examined [14]. In addition, over half of theBXDs in our panel (34 out of 61 strains) were from themore recent UTHSC BXD set, which have approxi-mately twice the number of recombinations per straincompared to the older BXD subset produced by Taylor.These features of our BXD panel improved the statisticalpower to map QTLs at high resolution as well as uncoverloci-loci interaction. Normal distribution of the RMSlinear densities in the BXD genetic reference panel in-dicates that more than one genetic locus is involved inNPC proliferation. Subsequent genome-wide intervalmapping provided novel insights into the genetic regionsregulating NPC proliferation. A significant Chr 6 QTLwas identified to account for 19% of the inter-straindifferences, and this locus was found to interact additivelywith two other genomic regions on Chr 14 and Chr 18.Pairwise interaction between Chr 6 QTL and these sec-ondary loci explained ~40% of the inter-strain differences.Genome-wide mapping also revealed a suggestive locuson Chr 11 that accounts for ~11% of the phenotypicvariance in the BXD population and it does not directlyinteract with the Chr 6 QTL. The identification of inter-acting and non-interacting genetic loci suggests the gen-etic network modulating NPC proliferation is comprisedof regulatory pathways of variable interactivity.To better understand the regulatory pathways modulat-ing NPC proliferation in the RMS, we needed to firstidentify the molecular players participating in these path-ways. It is presumed that the mapped loci harbor geneticmodifiers of NPC proliferation. Our rather stringentcandidate gene analyses revealed several genes in thesignificant Chr 6 QTL regions that are associated withneurogenesis and cell proliferation. One of them isTgfa, which is a growth factor that binds to the epider-mal growth factor receptors (EGFR). Upon binding toits receptor, Tgfa activates the ErbB signalling pathway andsubsequently triggers the Mitogen-activated protein kinase(MAPK) signalling cascade. These pathways play crucialroles in regulating cell proliferation and differentiation.The absence of Tgfa resulted in attenuation of NPC prolif-eration in the dorsolateral corner of lateral ventricles andreduction of newborn neurons in the OB [19]. Anothercandidate gene implicated in adult neurogenesis is Mcm2.This gene encodes for a component of the DNA replica-tion machinery for the S phase of cell cycle. Mcm2 hasbeen used as a proliferation marker in adult SVZ andhippocampal neurogenesis [29,31]. Hypomorphic expres-sion of Mcm2 resulted in decreased NPC proliferation inthe SVZ [20]. In addition to genes in the Chr 6 region,genes in the suggestive QTL regions also harbour promis-ing candidates that have been previously implicated inadult neurogenesis. One of which is Sparc in the Chr11 QTL. Sparc encodes a cysteine-rich acidic matrix-associated protein, which is involved in extracellularmatrix synthesis, regulation of cell shape, and cellularresponse to growth factor stimulus. Sparc also affectshippocampal neurogenesis with the knockout of Sparcresulting in decreased cell proliferation in the SGZ of(See figure on previous page.)Figure 6 Cell proliferation in the SGZ of BXD RI strains and QTL analyses. (A) The number of BrdU+ cells in the SGZ (± SEM) of 61 BXD RIstrains (white bars) and their parental strains, C57BL/6 J (red bar), and DBA/2 J (green bar). The sample size per strain is indicated in the bars.Whole-genome scan LRS plot generated from QTL mapping of the unadjusted SGZ cell proliferation data (B) and the adjusted data corrected forage effects (C). The x-axis represents the chromosomes 1–19, & X (top panel) and their physical maps in megabases (bottom panel). The y-axis and theblue line depict the LRS, which indicates the strength of association between genotypes of markers across the genome and the phenotype.The light red and gray horizontal lines mark the significant (P = 0.05) and suggestive (P = 0.63) threshold, respectively. Whole-genome intervalmapping of unadjusted SGZ proliferation data (B) and the SGZ proliferation data adjusted for age (C) identified no significant QTL butrevealed suggestive QTLs on Chr 1 (40-59 Mb), Chr 5 (104.5-119.5 Mb), and Chr 9 (71–83.7 Mb).Poon and Goldowitz BMC Genomics 2014, 15:206 Page 10 of 14http://www.biomedcentral.com/1471-2164/15/206the dentate gyrus [21]. Smad4 in the suggestive Chr 18QTL has been directly implicated in adult neurogenesis[24]. Smad4 encodes for a signal transduction proteinthat participates in the Bone morphogenetic proteinsignaling pathway, Wnt signaling pathway, and theTransforming growth factor beta (TGF-beta) signalingpathway. Conditional knockout of Smad4 in the SVZneural stem cells greatly impaired neuron productionin the OB and induced ectopic oligodendrocytes in thecorpus callosum [24].Candidate gene pathway analysis further provided in-sights into the interconnection among the candidategenes. Both the Tgfa gene in the major Chr 6 QTL and theCamk2a gene in the secondary Chr 18 QTL are involvedin the ErbB signalling pathway, which regulates diversebiological processes such as proliferation, differentiation,and survival. Tgfa serves as one of the extracellular ligandsthat bind to Epidermal growth factor receptors andsubsequently triggers downstream intracellular signal-ling pathways, which is modulated by Camk2a. Camk2aalso participates in the Wnt signalling pathway as wellas Ppp2ca in the suggestive Chr 11 QTL. Ppp2ca is oneof the major Ser/Thr phosphatases which has been pre-viously implicated as a negative regulator of cell prolif-eration. Ppp2ca interacts with Axin, a scaffold proteinthat down-regulates the Wnt signalling pathway by de-stabilizing β-catenin [32]. These findings highlight thecomplexity of NPC regulation where genes at different lociconverge on the same regulatory pathway and possiblyexert their effects in an additive manner. Overlap in genesparticipating in seemingly independent pathways suchas ErbB signalling pathway and Wnt signalling path-ways further suggests crosstalk between these pathways.Previous reviews have described the regulation of adultneurogenesis as a fine balance of the different systemsof networks [9,33]. The interconnectedness between path-ways indicates that the perturbation of one may triggercompensatory changes at others [9].The interconnectivity among candidate genes is furtherreflected at the transcriptional level. Because there is noBXD transcriptome data available for the RMS, we probedthe hippocampal gene expression database for the tran-script levels of our candidate genes. We later confirmedthe expression of these candidate genes in the mouseRMS using the microarray data of laser microdissectedRMS generated by Khodosevich et al. [34]. Our geneco-expression network indicates the candidate genes glu-tamine fructose-6-phosphate transaminase 1 (Gfpt1) inthe significant Chr 6 QTL as a highly-connected node thatpositively co-vary with Ppp2ca and Sparc in the suggestiveChr 11 QTL. Gfpt1 also negatively co-varies Gnpnat1 inthe Chr 14 QTL and Seh1l in the Chr 18 QTL. Gfpt1 isa rate-limiting enzyme of the hexosamine pathway, anddespite not directly implicated in neurogenesis, studieshave shown Gfpt1 to positively regulate cell proliferationand regeneration of other tissues [35,36]. The Gfpt1regulated-hexosamine biosynthesis pathway has beenshown to drive the levels of β-catenin and cell prolifera-tion [37]. The expression of β-catenin is the downstreamtarget of the canonical Wnt pathway (or Wnt/β-cateninpathway), which is regulated by other QTL candidategenes such as Ppp2ca. These findings demonstrate thedynamic interplay of candidate genes and the convergenceof different pathways to influence cell proliferation in theadult brain.The high correlation between cell proliferation in theRMS and several hippocampal traits such as the numberof granule cells in the dentate gyrus led us to suspectshared genetic determinants underlying proliferation inRMS and dentate gyrus. We examined the QTLs modu-lating cell proliferation in SGZ of the dentate gyrus usingour expanded panel of BXDs. The number of proliferativecells in the SGZ differed ~4 fold among the 61 BXDRI strains. A similar fold difference was reported byKempermann and colleagues where they used Ki67 asa proliferative marker and quantified the number ofdividing cells in the SGZ of 29 BXD RI strains [13].The proliferative differences we detected among theBXDs were mapped to three suggestive QTLs on Chr1, Chr 5, and Chr 9. These loci did not overlap withthe QTLs regulating RMS cell proliferation. From thesefindings we concluded that there are separate sets of genesthat differentially modulate NPC proliferation at the RMSand SGZ.ConclusionsThe bottom-up approach, starting with the manipulationof a single gene to induce changes in phenotype, hasyielded valuable insights into a gene’s role in adult neuro-genesis. However, this approach is limited in capturing thefull complexity of adult neurogenesis where several genesand signalling pathways can regulate neurogenesis in theadult brain. In this study, we gained insights into thegenetic architecture underlying NPC proliferation usinga top-down, phenotypic-driven approach.The rich genetic and phenotypic diversity in the ex-panded panel of BXD RI strains has allowed us to probethe complex networks regulating NPC proliferation inthe adult mouse brain. Here, we performed genome-widescans for QTLs associated with the differences in thenumber of NPCs among the BXD RI strains. Novel QTLsand significant loci-loci interaction were detected. Usingbioinformatics resources and transcriptome databases, wefurther identified several candidate genes with correlatingexpression patterns and discovered the convergence ofthese genes on the same/parallel signalling pathwaysthat are known to regulate adult neurogenesis. Thecombinatorial influence of these regulators on NPCPoon and Goldowitz BMC Genomics 2014, 15:206 Page 11 of 14http://www.biomedcentral.com/1471-2164/15/206proliferation remains to be elucidated. Nevertheless, ourfindings provide a firm starting point to unravel modifiergenes of NPC proliferation and offer novel insights intothe dynamic interplay of regulatory pathways controllingthis process.MethodsAnimalsBXD RI strains were obtained from two different sources.Parental strains C57BL/6 J, DBA/2 J, and BXD 1–42 RIstrains were purchased from the Jackson Laboratory (BarHarbor, ME, USA). BXD 43–103 RI strains were providedby Dr. Robert W. Williams and Dr. Lu Lu (University ofTennessee Health Science Center, Memphis, TN, USA).Mice were housed at the University of Tennessee AnimalFacility in a pathogen-free, ~23.5°C, and 45–50 humidityenvironment on a 12 h light-12 h dark cycle. A total of 61BXD RI strains and 265 mice were used in this study.At least one male and one female per BXD strain wereexamined. Mice studied were between 50–85 days old.All experiments were conducted according to the Guidefor the Care and Use of Laboratory Animals (NationalInstitutes of Health, USA) and the Canadian Council ofAnimal Care. Approval was obtained from the Institu-tional Animal Care and Use Committee at the Universityof Tennessee Health Science Center and the Animal CareCommittee at the University of British Columbia on allour animal protocols.BrdU administration and detectionAll mice received a single intraperitoneal injection ofBrdU (50 mg BrdU kg−1) to label actively proliferatingcells in their brains. One-hour post injection, mice wereperfused with acetic acid: 95% ethanol (1:3) as previouslydescribed [14]. Brains were collected and paraffin em-bedded. Embedded brains were sagittally sectioned at8 μm and every 10th section was mounted on Superfrost/Plus slides for anti-BrdU immunohistochemistry. In brief,sections were deparaffinzed, and treated with 1 M HClfor 30 mins at 37°C to denature DNA. Slides were incu-bated with the mouse primary anti-BrdU monoclonalantibody (1:200; BD Biosciences, Mississauga, ON, Canada)overnight at room temperature. The next day, secondarybiotinylated horse anti-mouse IgG (1:200, Vector Labora-tories, Burlingame, CA, USA) was applied to the slidesfor 1 h at room temperature VECTASTAIN Elite ABCkit (Vector Laboratories) and 3, 3′-diaminobenzidine(DAB; Sigma-Aldrich) were subsequently used to revealBrdU immunoreactivity.QuantificationThe number of proliferative (BrdU+) cells in the RMS wassystematically quantified using the single best-sectionquantification method as previously described [14]. Briefly,sagittal sections encompassing the medial to lateral extendof the RMS of the left hemisphere was examined, and thesection containing the RMS exhibiting the fullest extentof the SVZ-OB trajectory was identified. The numberof BrdU + cells in the RMS of this optimal section wasquantified under brightfield illumination (20× objective;Zeiss 200 M Axiovert inverted microscope equippedwith Axiovision 4.6 software). Images of the quantifiedRMS were captured using the same inverted microscope,and the length of the RMS was subsequently measuredwith NIH ImageJ (version 1.42) software. For each animal,the RMS linear density, which is the number of BrdU +cells per mm of RMS length, was calculated. The averageRMS linear density and standard error of the mean (SEM)were determined for each strain. To demonstrate theeffectiveness of our single best-section quantificationmethod, we also quantified the total number of NPCsin the RMS of 20 randomly selected RI mice where wecounted the number of BrdU + cells in every 10th sec-tion throughout the entire medial to lateral extent ofthe RMS. We found the total NPC counts were highlycorrelated with the RMS linear densities of these animals(R2 = 0.7744; P < 0.0001; Additional file 3).Mice analyzed for RMS cell proliferation were alsoused to quantify cell proliferation in another active siteof adult neurogenesis called the SGZ. The SGZ is lo-cated between the granular layer and hilus of the dentategyrus of the hippocampus. The number of proliferative,BrdU + cells in the SGZ was counted throughout the leftdorsal hippocampus for every tenth sagittal section. Countsstopped where the dorsal and ventral components of thehippocampus merged. The average number of BrdU + cellsin the SGZ and SEM were determined for each BXD strain.Statistical analysisOne-way analysis of variance (ANOVA) was performedto detect significant inter-strain differences in the num-ber of proliferative cells in the RMS (JMP 10 statisticalsoftware, SAS Institute, Cary, NC, USA). Tukey’s HonestlySignificant Difference (HSD) post-hoc test was subse-quenlty employed to determine signficant differences be-tween two BXD strains. General linear modeling was usedto determine the contribution of possible confoundingcovariates including age, sex, body weight, and strainepoch effects and their interaction effects on RMS prolif-eration. Residuals from regression fitting these variablesfor RMS linear density were obtained and subsequentlyused to adjust for the mean RMS linear density per strain[14]. Analyses that yielded P ≤ 0.05 were considered sig-nificant. Heritability was estimated in a broad sense wherewe calcualted the ratio of variance that is accounted for bythe differences between strains over the total variance,which includes both between-strain variance and within-strain variance [13].Poon and Goldowitz BMC Genomics 2014, 15:206 Page 12 of 14http://www.biomedcentral.com/1471-2164/15/206QTL mappingCell proliferation data collected from the 61 BXD RIstrains was deposited into the GeneNetwork, which isan open-access online database that contains detailedgenotype information of each BXD RI strain. Genome-wide interval mapping of QTLs regulating NPC prolif-eration was performed using WebQTL, a module of theGeneNetwork. The likelihood ratio statistic (LRS) wascomputed to assess the strength of genotype-phenotypeassociation of the genome scans. Permutation test of2000 permutations was computed to establish the sig-nificance and suggestive thresholds where the LRS valuescorresponded to a genome-wide P value of 0.05 and0.63, respectively. A significant QTL is referred to as achromosomal region with LRS score equal or above thegenome-wide significant level (P = 0.05). A suggestiveQTL is a region of the chromosome with LRS score equalor above the genome-wide suggestive level (P = 0.63). LRSscores of the mapped QTLs were converted to the likeli-hood of the odds (LOD) scores by dividing LRS by 4.61.The confidence limits of each QTL were defined by the1.5 LOD support interval [38].Candidate gene analysisAn integration of bioinformatics strategies and gene ex-pression data were employed to evaluate the underlyinggenes in the mapped QTL intervals. The genetic variationstructure within identified QTL regions were examinedusing the single-nucleotide polymorphism (SNP) and in-sertion/deletion (indel) data available at the GeneNetworkSNP browser (genenetwork.org/webqtl/snpBrowser.py).The numbers of SNPs and indels that are associatedwith each candidate gene, and ones that differ betweenthe two parental inbred strains (i.e., DBA/2 J andC57BL/6 J) were determined. Sequencing data released bythe Mouse Genomes Project (www.sanger.ac.uk/resources/mouse/genomes/) was used to confirm the presences ofSNPs and indels in each of the candidate gene. The expres-sion of each candidate gene in the adult brain is visualizedusing Allen Brain Atlas (http://www.brain-map.org). Micro-array data on laser-microdissected NPCs in the RMS [34]was used to determine the presence/absence and transcriptlevel of a candidate gene in the RMS. Candidate gene werefurther assessed on their associated Gene Ontoloy (GO)terms and pathways using information available at theDatabase for Annotation, Visualization, and IntegratedDiscovery (DAVID, http://david.abcc.ncifcrf.gov) and theQTLMiner module of the GeneNetwork (http://www.gen-enetwork.org/webqtl/main.py?FormID=qtlminer). Geneswere prioritized according to the following criteria: 1)polymorphic (i.e. associated with ≥ 1 SNPs and/or Indels),2) expression in the adult neurogenic sites including theRMS, and 3) associated with GO terms such as neurogen-esis, cell proliferation, and cell cycle progression.Availability of supporting dataBXD data supporting the results of this article are availablein the BXD Published Phenotypes Database at the Gene-Network (www.genenetwork.org) under Record IDs: 13545(RMS linear density, unadjusted data), 13586 (RMS lineardensity data corrected for age effects), 14786 (SGZ BrdU+ cellcounts, unadjusted data), and 16187 (SGZ BrdU+ cell countscorrected for age effects).Other data supporting the results of this article areincluded as additional files (Additional files 1, 2, and 3)within the article.Additional filesAdditional file 1: Whole-genome QTL analyses of the old and newBXD sub-populations. (A) Whole-genome interval mapping of the oldBXD strains 1–42 generated by Benjamin A. Taylor; 27 of the 61 BXD strainsexamined in this study belong to this group (B) Whole-genome interval map-ping of the new BXD strains 43–100 generated at the University of TennesseeHealth Science Center (UTHSC); 34 of the 61 BXD strains examined in thisstudy belong to this group. Despite differences in LRS scores, the same Chr 6QTL (76.8-88.8 Mb) is identified from mapping the two BXD sub-populations.Additional file 2: Correlation between the transcript levels ofcandidate genes and RMS linear density. Inter-strain differences in theexpression of Anxa4 (a candidate gene in the significant Chr 6 QTL interval),Ppp2ca (a candidate gene in the suggestive Chr 11 QTL interval), Gnpnat1(a candidate gene in the suggestive Chr 14 QTL interval), and Camk2a(a candidate gene in the suggestive Chr 18 QTL) were observed in thehippocampi of different BXD RI strains. (A) Scatterplot of the Anxa4 transcriptlevels negatively correlated with the mean RMS linear density. (B) Scatterplotof the Ppp2ca transcript levels negatively correlated with the mean RMSlinear density. (C) Scatterplot of the Gnpnat1 transcript levels negativelycorrelated with the mean RMS linear density. (D) Scatterplot of the Camk2atranscript levels positively correlated with the mean RMS linear density. Eachdot represents a BXD strain.Additional file 3: Correlation between the RMS linear density andtotal number of proliferating NPCs in the RMS. RMS linear density (i.e.the number of BrdU + cells per mm length of RMS; x-axis) was determinedusing the single best-section quantification method, and it is significantlycorrelated with the total BrdU + cell counts (y-axis) which was determinedfrom surveying every 10th section throughout the medial to lateral extent ofthe RMS (P < 0.0001). Each data point represents counts obtained from arandomly selected RI mouse.AbbreviationsBrdU: Bromodeoxyuridine; Chr: Chromosome; NPCs: Neural progenitor cells;RI: Recombinant inbred; RMS: Rostral migratory stream; SGZ: Subgranularzone; SVZ: Subventricular zone; QTL: Quantitative trait loci.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsDG conceived the study. AP collected phenotypic and carried out all dataanalyses including QTL mapping and candidate gene analysis. AP and DGwrote the manuscript. Both authors read and approved the final manuscript.AcknowledgementsWe thank Richard Cushing and Nicholas Folk for their invaluable technicalassistance in tissue collection and sectioning. We also thank Dr. RobertWilliams for his expert assistance on QTL mapping at the GeneNetwork andfor his comments on data interpretation. This work was supported by NIHgrants R01DA020677, R01AG18245, and NeuroDevNet, a Canadian Networkof Centres of Excellence devoted to understand brain development andtranslate knowledge into treatments for disorders of the brain.Poon and Goldowitz BMC Genomics 2014, 15:206 Page 13 of 14http://www.biomedcentral.com/1471-2164/15/206Received: 10 August 2013 Accepted: 10 March 2014Published: 19 March 2014References1. Lois C, Alvarez-Buylla A: Long-distance neuronal migration in the adultmammalian brain. Science 1994, 264:1145–1148.2. Curtis MA, Low VF, Faull RLM: Neurogenesis and progenitor cells in theadult human brain: a comparison between hippocampal andsubventricular progenitor proliferation. Dev Neurobiol 2012, 72:990–1005.3. Zhang RL, LeTourneau Y, Gregg SR, Wang Y, Toh Y, Robin AM, Zhang ZG,Chopp M: Neuroblast division during migration toward the ischemicstriatum: a study of dynamic migratory and proliferative characteristics ofneuroblasts from the subventricular zone. J Neurosci 2007, 27:3157–3162.4. Ziabreva I, Perry E, Perry R, Minger SL, Ekonomou A, Przyborski S, Ballard C:Altered neurogenesis in Alzheimer’s disease. J Psychosom Res 2006,61:311–316.5. Curtis MA, Eriksson PS, Faull RLM: Progenitor cells and adult neurogenesisin neurodegenerative diseases and injuries of the basal ganglia. Clin ExpPharmacol Physiol 2007, 34:528–532.6. Johnson MA, Ables JL, Eisch AJ: Cell-intrinsic signals that regulate adultneurogenesis in vivo: insights from inducible approaches. BMB Rep 2009,42:245–259.7. Ming G-L, Song H: Adult neurogenesis in the mammalian brain: significantanswers and significant questions. Neuron 2011, 70:687–702.8. Khodosevich K, Alfonso J, Monyer H: Dynamic changes in thetranscriptional profile of subventricular zone-derived postnatally bornneuroblasts. Mech Dev 2013, 130:424–432.9. Kempermann G: Seven principles in the regulation of adult neurogenesis.Eur J Neurosci 2011, 33:1018–1024.10. Enwere E, Shingo T, Gregg C, Fujikawa H, Ohta S, Weiss S: Aging results inreduced epidermal growth factor receptor signaling, diminished olfactoryneurogenesis, and deficits in fine olfactory discrimination. J Neurosci 2004,24:8354–8365.11. Jin J, Kang H-M, Park C: Voluntary exercise enhances survival and migrationof neural progenitor cells after intracerebral haemorrhage in mice. Brain Inj2010, 24:533–540.12. Lucassen PJ, Meerlo P, Naylor AS, van Dam AM, Dayer AG, Fuchs E, Oomen CA,Czéh B: Regulation of adult neurogenesis by stress, sleep disruption,exercise and inflammation: implications for depression and antidepressantaction. Eur Neuropsychopharmacol 2010, 20:1–17.13. Kempermann G, Chesler EJ, Lu L, Williams RW, Gage FH: Natural variationand genetic covariance in adult hippocampal neurogenesis. Proc NatlAcad Sci USA 2006, 103:780–785.14. Poon A, Li Z, Wolfe GW, Lu L, Williams RW, Hayes NL, Nowakowski RS,Goldowitz D: Identification of a Chr 11 quantitative trait locus thatmodulates proliferation in the rostral migratory stream of the adultmouse brain. Eur J Neurosci 2010, 32:523–537.15. Poon A, Goldowitz D: Effects of age and strain on cell proliferation in themouse rostral migratory stream. Neurobiol Aging 2013, 34:1712.e15–21.16. Taylor BA, Bedigian HG, Meier H: Genetic studies of the Fv-1 locus of mice:linkage with Gpd-1 in recombinant inbred lines. J Virol 1977, 23:106–109.17. Taylor BA, Wnek C, Kotlus BS, Roemer N, MacTaggart T, Phillips SJ: Genotypingnew BXD recombinant inbred mouse strains and comparison of BXD andconsensus maps. Mamm Genome 1999, 10:335–348.18. Peirce JL, Lu L, Gu J, Silver LM, Williams RW: A new set of BXDrecombinant inbred lines from advanced intercross populations in mice.BMC Genet 2004, 5:7.19. Tropepe V, Craig CG, Morshead CM, van der Kooy D: Transforming growthfactor-alpha null and senescent mice show decreased neural progenitor cellproliferation in the forebrain subependyma. J Neurosci 1997, 17:7850–7859.20. Pruitt SC, Bailey KJ, Freeland A: Reduced Mcm2 expression results insevere stem/progenitor cell deficiency and cancer. Stem Cells 2007,25:3121–3132.21. Campolongo M, Benedetti L, Podhajcer OL, Pitossi F, Depino AM:Hippocampal SPARC regulates depression-related behavior. Genes BrainBehav 2012, 11:966–976.22. Nalepa G, Barnholtz-Sloan J, Enzor R, Dey D, He Y, Gehlhausen JR, LehmannAS, Park S-J, Yang Y, Yang X, Chen S, Guan X, Chen Y, Renbarger J, Yang F-C,Parada LF, Clapp W: The tumor suppressor CDKN3 controls mitosis. J Cell Biol2013, 201:997–1012.23. Boehmelt G, Wakeham A, Elia A, Sasaki T, Plyte S, Potter J, Yang Y, Tsang E,Ruland J, Iscove NN, Dennis JW, Mak TW: Decreased UDP-GlcNAc levelsabrogate proliferation control in EMeg32-deficient cells. EMBO J 2000,19:5092–5104.24. Colak D, Mori T, Brill MS, Pfeifer A, Falk S, Deng C, Monteiro R, Mummery C,Sommer L, Götz M: Adult neurogenesis requires Smad4-mediated bonemorphogenic protein signaling in stem cells. J Neurosci 2008, 28:434–446.25. Gómez-Gaviro MV, Scott CE, Sesay AK, Matheu A, Booth S, Galichet C,Lovell-Badge R: Betacellulin promotes cell proliferation in the neural stemcell niche and stimulates neurogenesis. Proc Natl Acad Sci USA 2012,109:1317–1322.26. Wu MV, Hen R: The young and the restless: regulation of adultneurogenesis by Wnt signaling. Cell Stem Cell 2013, 12:139–140.27. Zhang L, Yang X, Yang S, Zhang J: The Wnt /β-catenin signaling pathwayin the adult neurogenesis. Eur J Neurosci 2010, 33:1–8.28. Logan TT, Villapol S, Symes AJ: TGF-β Superfamily Gene Expression andInduction of the Runx1 Transcription Factor in Adult Neurogenic Regionsafter Brain Injury. PLoS One 2013, 8:e59250.29. Bohlen und Halbach von O: Immunohistological markers for proliferativeevents, gliogenesis, and neurogenesis within the adult hippocampus.Cell Tissue Res 2011, 345:1–19.30. Segal E, Wang H, Koller D: Discovering molecular pathways from proteininteraction and gene expression data. Bioinformatics 2003, 19:i264–i272.31. Maslov AY, Barone TA, Plunkett RJ, Pruitt SC: Neural stem cell detection,characterization, and age-related changes in the subventricular zone ofmice. J Neurosci 2004, 24:1726–1733.32. Nakamura T, Hamada F, Ishidate T, Anai K, Kawahara K, Toyoshima K,Akiyama T: Axin, an inhibitor of the Wnt signalling pathway, interactswith beta-catenin, GSK-3beta and APC and reduces the beta-cateninlevel. Genes Cells 1998, 3:395–403.33. Hsieh J: Orchestrating transcriptional control of adult neurogenesis.Genes Dev 2012, 26:1010–1021.34. Khodosevich K, Seeburg PH, Monyer H: Major signaling pathways inmigrating neuroblasts. Front Mol Neurosci 2009, 2:7.35. Kim J, Song G, Wu G, Bazer FW: Functional roles of fructose. Proc Natl AcadSci USA 2012, 109:E1619–E1628.36. Yang C-T, Hindes AE, Hultman KA, Johnson SL: Mutations in gfpt1 andskiv2l2 cause distinct stage-specific defects in larval melanocyte regener-ation in zebrafish. PLoS Genet 2007, 3:e88.37. Olivier-Van Stichelen S, Guinez C, Mir A-M, Perez-Cervera Y, Liu C, Michalski J-C,Lefebvre T: The hexosamine biosynthetic pathway and O-GlcNAcylation drivethe expression of β-catenin and cell proliferation. Am J Physiol EndocrinolMetab 2012, 302:E417–E424.38. Dogan A, Lasch P, Neuschl C, Millrose MK, Alberts R, Schughart K, NaumannD, Brockmann GA: ATR-FTIR spectroscopy reveals genomic loci regulatingthe tissue response in high fat diet fed BXD recombinant inbred mousestrains. BMC Genomics 2013, 14:386.doi:10.1186/1471-2164-15-206Cite this article as: Poon and Goldowitz: Identification of genetic locithat modulate cell proliferation in the adult rostral migratory streamusing the expanded panel of BXD mice. BMC Genomics 2014 15:206.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitPoon and Goldowitz BMC Genomics 2014, 15:206 Page 14 of 14http://www.biomedcentral.com/1471-2164/15/206

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0223207/manifest

Comment

Related Items