Genomic profiles for human peripheral blood T cells, B cells, natural killer cells, monocytes, and polymorphonuclear cells: Comparisons to ischemic stroke, migraine, and Tourette syndrome

Please download to get full document.

View again

of 11
15 views
PDF
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
Genomic profiles for human peripheral blood T cells, B cells, natural killer cells, monocytes, and polymorphonuclear cells: Comparisons to ischemic stroke, migraine, and Tourette syndrome
Document Share
Document Tags
Document Transcript
  Genomic profiles for human peripheral blood T cells, B cells, natural killer cells, monocytes, and polymorphonuclear cells: Comparisons to ischemicstroke, migraine, and Tourette syndrome XinLi Du  a,b , Yang Tang  a,b , Huichun Xu  a,b , Lisa Lit   a,b , Wynn Walker   a,b , Paul Ashwood  a,d ,Jeffrey P. Gregg  a,c , Frank R. Sharp  a,b, ⁎ a   MIND Institute, University of California at Davis, 2805 50th Street, Sacramento, CA 95817, USA  b  Department of Neurology, University of California at Davis, 2805 50th Street, Sacramento, CA 95817, USA c  Department of Pathology, University of California at Davis, 2805 50th Street, Sacramento, CA 95817, USA d  Department of Medicine, University of California at Davis, 2805 50th Street, Sacramento, CA 95817, USA Received 3 November 2005; accepted 5 February 2006Available online 20 March 2006 Abstract Blood genomic profiling has been applied to disorders of the blood and various organ systems including brain to elucidate disease mechanismsand identify surrogate disease markers. Since most studies have not examined specific cell types, we performed a preliminary genomic survey of major blood cell types from normal individuals using microarrays. CD4 + T cells, CD8 + T cells, CD19 + B cells, CD56 + natural killer cells, andCD14 + monocytes were negatively selected using the RosetteSep antibody cocktail, while polymorphonuclear leukocytes were separated withdensity gradient media. Genes differentially expressed by each cell type were identified. To demonstrate the potential use of such cell subtype-specific genomic expression data, a number of the major genes previously reported to be regulated in ischemic stroke, migraine, and Tourettesyndrome are shown to be associated with distinct cell populations in blood. These specific gene expression, cell-type-related profiles will need to be confirmed in larger data sets and could be used to study these and many other neurological diseases.© 2006 Elsevier Inc. All rights reserved.  Keywords:  Blood; Humans; Gene expression; Microarrays; Genome; T cells; B cells; NK cells; Neutrophils; Migraine; Stroke; Tourette Gene expression profiling of peripheral blood usingmicroarrays has been applied to malignant and immunedisorders, including leukemia, lymphoma, systemic lupuserythematosis, rheumatoid arthritis, and many others [1 – 4].This approach has helped identify important diagnostic and prognostic markers as well as potential therapeutic targets.This approach has also been extended to many diseases of other organ systems. It is likely that many inflammatory,autoimmune, and genetic factors could affect gene expressionof peripheral blood cells without causing overt changes tohematological and immunological phenotypes. Proof-of-prin-ciple blood genomic studies have been performed in animals[5] and humans [6,7]. Subsequent studies have demonstrated characteristic blood genomic patterns for acute ischemic stroke[8], migraine headache [9], Tourette syndrome [10], renal cell carcinoma [11], multiple sclerosis [12], benzene exposure [13], trauma [14], and neurogenetic disorders including neurofibromatosis type I, tuberous sclerosis type II, Downsyndrome [7,15], and Huntington chorea [16]. The study of   blood gene expression profiles appears to be a promisingapproach that may provide mechanistic insights and surrogatemarkers for many diseases.Several blood RNA isolation methods have been used todate. These include methods starting with whole blood,mononuclear cells, and buffy coat  [6,7,11,14,17,18]. How-ever, the RNA isolated using these methods comes fromvarious blood cell subsets that originate from different  Genomics 87 (2006) 693 – 703www.elsevier.com/locate/ygeno ⁎  Corresponding author. MIND Institute, University of California at Davis,2805 50th Street, Sacramento, CA 95817, USA. Fax: +1 916 703 0369.  E-mail address:  frank.sharp@ucdmc.ucdavis.edu (F.R. Sharp).0888-7543/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.ygeno.2006.02.003  developmental lineages, perform separate and distinct bio-logical functions, and, most likely, have very different geno-mic expression signatures. It has been recognized that ageand gender and the different composition of blood cells fromeach individual represent a major source of normal variationof blood gene expression [6,7]. In addition, a disease may predominantly affect one specific blood cell subtype whilesparing others. Therefore, characterizing the contribution of every blood cell subtype to the overall blood genomic patternmay be essential to distinguish significant genomic changesfrom noise, interpret the disease-related patterns, and decide onthe proper blood cell types to perform follow-up confirmatoryanalyses.Expression profiles of blood cells such as T lymphocytes[14,19] and platelets [18,20] have been described. However, studies that compare directly the whole genomic expression profiles of several major blood cell subtypes have not been performed in detail. In this study, we attempted to build a preliminary gene expression database by comparing major leukocyte subsets from three healthy donors, including polymorphonuclear cells (PMN), monocytes, B cells, CD4 + Tcells, CD8 + cytotoxic T cells, and natural killer (NK) cells todetermine whether there is likely to be a unique expressionsignature of each cell type. To demonstrate the utility of theseexpression signatures, we applied these data to the whole bloodgenomic profiles of several neurological diseases that we havestudied previously, including acute ischemic stroke [8],migraine [9], and Tourette syndrome [10], to demonstrate that  the blood genomic signatures of each of these conditions can beascribed to certain blood cell subtypes being affected by eachdisease. Future studies likely could determine not only whether hematological and systemic diseases affect gene expression inspecific subsets of blood cells, but also whether the diseasesaffect specific signaling pathways in specific subsets of cells in blood. Results Qualitative analysis The numbers of   “  present  ”  and  “ unique ”  genes for each celltype are listed in Table 1. Of the 54,675 genes, higher  percentages of the genes are expressed (present) by lympho-cytes, including B (36.6%), CD4 (36.4%), CD8 (35.3%), and NK cells (36.2%), than by monocytes/platelets (31.9%) andPMNs (24.0%). However, there are higher percentages of unique genes for PMNs (1.2%) and monocytes/platelets (0.9%)than for lymphocytes (0 – 0.6%) (Table 1). Among thelymphocytes, B cells have the highest number of characteristicgenes, while unique transcripts for T cells are relatively scarcedue to the largely similar profiles for T cell subsets CD4 + andCD8 + and the profiles for NK cells. Quantitative analysis Among 54,675 genes (probe sets) on the array, a total of 2635 are differentially expressed between the blood cell types(parametric analysis of variance (ANOVA),  p  < 0.05 withBonferroni correction), among which 269 are significant using a Student  –  Newman – Keuls post hoc test. For practicalreasons, we focused on the 269-probe set list since it shouldcontain the most characteristic genes and potentially containgenomic expression markers for each cell lineage. These 269genes/probe sets were mathematically separated into nineclusters of relatively unique expression profiles using ahierarchical algorithm [25] as demonstrated in Fig. 1. The  pattern of expression of each gene in each cluster and the foldchanges of the genes are shown in separate panels on theright side of  Fig. 1. In general the fold changes varied asmuch as 10- to 100-fold. The genes in each cluster are listedin Table 2. The left side of  Fig. 1 not only shows the gene expression (  y  axis) for different blood cell types (  x  axis), but also shows the gene expression of the three individuals performed for each cell type. Note that the expression levels(red — fivefold increase; bright green — fivefold decrease) for each individual are extremely reproducible between cell typesand between genes. This indicates that the microarraytechnological variables have a minimal effect upon theexpression profiles shown in Fig. 1 and indicate that our criteria for selecting genes for each cell type are stringent andnot significantly affected by individual differences at least inthis preliminary study. Cellular origin of blood genes regulated by neurological diseases Fig. 2 represents a melding of the data from the current study with that from our previous disease-specific studies. Thegenes that were most highly regulated in ischemic stroke [8],Tourette syndrome [10], and migraine [9] were selected and the cell-specific expression of each of those genes (from the present study) is shown. As demonstrated in Fig. 2, the major genes up-regulated in whole blood after stroke were expressedmainly by PMNs and monocytes/platelets. The major genesup-regulated by Tourette syndrome were mostly from NK cellsand/or CD8 + T cells. The major genes up-regulated bymigraine were predominantly from platelets/monocytes, Table 1Results of the qualitative analysisCell type Number of  present probesetsPercentage of  present probesets Number of unique probesetsPercentage of unique probesetsPMN 13,139 24.0% 155 1.2%CD14 + monocyteor platelet 17,426 31.9% 152 0.9%CD19 + B cell 20,009 36.6% 120 0.6%CD4 + T cell 19,909 36.4% 24 0.1%CD8 + T cell 19,321 35.3% 8 0.0%CD56 +  NK cell 19,777 36.2% 42 0.2%A total of 54,675 probe sets that examined approximately 39,500 genes weresurveyed on each array.  “ Present  ”  probe sets include probe sets that have 3 present detection calls for a specific cell type regardless of the calls for other celltypes.  “ Unique ”  probe sets include probe sets that have 3 present calls for a celltype and 15 absent calls for every other cell type.694  X. Du et al. / Genomics 87 (2006) 693  –  703  though there were some regulated genes from PMNs, CD4 + ,CD8 + , and NK cells. Discussion This study surveyed the global expression profiles of sixmajor subtypes of blood cells. These data support previousstudies showing that T cells and even platelets have genes that are expressed in common, but also have genes that are fairlyspecific for each cell type and, perhaps more importantly, havedifferent families of genes that tend to be expressed in a specificcell type compared to another cell type [14,18 – 20]. Character-ization of these profiles should help elucidate the molecular andgenomic basis of the development, differentiation, and functionfor each cell type.Genes in cluster 1 are highly enriched in a monocyte/platelet  population compared to other cell types. The recent literatureshows that many genes from this cluster, such as CLU, GP1BB,PF4V1, and others (Table 2), are specifically expressed by platelets [18,20]. Cluster 2 represents genes enriched in PMNs and monocytes, while genes in cluster 3 are expressedexclusively by PMNs. Many genes in these two clusters playcrucial roles in innate immunity. These include receptor molecules such as TREM1 [26], FPRL1 [27], and TLR2 [28], which are involved in microbial recognition and lead to phagocyte activation and the amplification of the inflammatoryresponse. There are effector molecules such as MMP9 [29];S100 proteins P, A9, and A12 [30]; and neutrophil cytosolicfactors 1, 2, and 4, which participate in the neutralization of andaid clearance of microorganisms and foreign materials, andscavenger molecules such as IL1R2 [31] and TNFRSF10 [32] that help suppress excessive and harmful innate immuneresponses. In comparison, genes down-regulated in PMNs(cluster 5) did not provide many functional insights. The lowexpression of several ribosomal proteins and transcriptionelongation factor in this cluster may indicate a slower rate of  protein translation in PMNs and is consistent with somewhat fewer RNA transcripts in this cell type (Table 1).Several molecules expressed by B cells (cluster 4) serveimportant central roles in B cell development, proliferation, anddifferentiation, such as MS4A1 [33], BLNK  [34], and BANK1 [35]. Other molecules, including immunoglobulins and HLAantigens, important for normal B cell functions, were alsoexpressed (Table 2). While there are a few common genes between NK cells and T cells, most notably T cell receptor subunits and lymphocyte-specific tyrosine kinase (LCK) Fig. 1. A total of 269 genes that are differentially expressed between blood cell types (parametric one-way ANOVA followed by Student  –  Newman – Keuls post hoctest,  p  < 0.05, with Bonferroni multiple comparison correction) were subjected to a hierarchical cluster algorithm with Pearson correlation as a measure of similarity.(Left) Clusters of genes (nine clusters) with similar expression patterns are displayed from top to bottom (  y  axis), while cell types are displayed from left to right alongthe  x  axis. For each cell type the results of the three different individuals are shown adjacent to one another. The relative expression of each gene is color coded; redshows a fivefold increase and green shows a fivefold decrease. (Right, 1 – 9) Line graphs of genes segregated in the cluster analysis are shown for each of the nineclusters identified on the left. The  x  axis shows the cell types and the  y  axis shows the relative expression values (log scale) as mean − 1 standard deviation (log ratio).695  X. Du et al. / Genomics 87 (2006) 693  –  703  Table 2Results of the quantitative analysisCommon GenBank DescriptionCluster 1 a1/3GTP AI972498 Clone IMAGE:4812754,mRNAACRBP AB051833 Acrosin-binding proteinARHGAP6 NM_001174 Rho GTPase-activating protein 6C21orf7 NM_020152 Chromosome 21 openreading frame 7CA2 M36532 Carbonic anhydrase IICD163 NM_004244 CD163 antigenCD36 NM_000072 CD36 antigen(collagen type I receptor,thrombospondin receptor)CD9 NM_001769 CD9 antigen (p24)CLEC2 NM_016509 C-type lectin-likereceptor-2CLU M25915 ClusterinCSPG2 BF590263 Chondroitin sulfate proteoglycan 2 (versican)CXCL5 AK026546 Chemokine (C-X-C motif)ligand 5CYP1B1 NM_000104 Cytochrome P450,family 1, subfamily B, polypeptide 1ELOVL7 AW138767 Hypothetical proteinFLJ23563EMS1 NM_005231 EMS1 sequence(mammary tumor andsquamous cellcarcinoma-associated(p80/85 Src substrate)F13A1 NM_000129 Coagulation factor XIII,A1 polypeptideFSTL1 BC000055 Follistatin-like 1GNG11 NM_004126 Guanine nucleotide binding protein(G protein),  γ 11GP1BB NM_000407 Glycoprotein Ib(platelet),  β  polypeptideHIST1H3H NM_003536 Histone 1, H3hITGB3 M35999 Integrin,  β 3(platelet glycoprotein IIIa,antigen CD61)KIAA0626 NM_021647MS4A6A NM_022349 Membrane-spanning4-domains, subfamily A,member 6AMYLK AA526844 MSTP083 mRNA,complete cdsPF4 NM_002619 Platelet factor 4(chemokine (C-X-C motif)ligand 4)PF4V1 NM_002620 Platelet factor 4 variant 1PPBP R64130 Proplatelet basic protein(chemokine (C-X-C motif)ligand 7)PRKAR2B NM_002736 Protein kinase,cAMP-dependent,regulatory, type II,  β PROS1 NM_000313 Protein S ( α )PTGS1 S36219 Prostaglandin – endoperoxidesynthase 1 (prostaglandinG/H synthase andcyclooxygenase)RIN2 AL136924 Ras and Rab interactor 2Table 2 ( continued  )Common GenBank DescriptionCluster 1 SDPR NM_004657 Serum deprivation response(phosphatidylserine binding protein)SDPR BF982174 Serum deprivation response(phosphatidylserine binding protein)SPARC NM_003118 Secreted protein, acidic,cysteine-rich (osteonectin)THBS1 BF055462 Thrombospondin 1TREML1 AF534823 Triggering receptor expressed on myeloidcells-like 1TUBB1 NM_030773 Tubulin,  β 1Cluster 2 ANXA3 M63310 Annexin A3APOBEC3A U03891 Apolipoprotein B mRNAediting enzyme, catalytic polypeptide-like 3AAQP9 NM_020980 Aquaporin 9BASP1 NM_006317 Brain abundant,membrane attachedsignal protein 1CD14 NM_000591 CD14 antigenCLECSF12 AF400600 C-type (calcium-dependent,carbohydrate-recognitiondomain) lectin,superfamily member 12CLECSF9 BC000715 C-type (calcium-dependent,carbohydrate-recognitiondomain) lectin,superfamily member 9CREB5 AI689210 cAMP-responsiveelement binding protein 5CSF3R NM_000760 Colony-stimulatingfactor 3 receptor (granulocyte)DKFZP434B044 AL136861 Hypothetical protein DKFZp434B044DKFZp434H2111 AK026776 Hypothetical protein DKFZp434H2111FCGR2A NM_021642 Fc fragment of IgG,low affinity IIa, receptor for (CD32)FLJ20273 NM_019027 RNA-binding proteinFLJ23091 AL534095 Putative NF- κ Bactivating protein 373FLJ23091 AA775681 Putative NF- κ Bactivating protein 373FLJ23153 AA650281 Likely ortholog of mouse tumor necrosis- α -inducedadipose-related proteinFOS BC004490 v-Fos FBJ murineosteosarcoma viraloncogene homologFPR1 NM_002029 Formyl peptidereceptor 1GALNAC4S-6ST NM_014863GPR86 NM_023914 G-protein-coupledreceptor 86HIST2H2BE NM_003528 Histone 2, H2beHSPC159 AK025603 HSPC159 proteinIL13RA1 NM_001560 Interleukin 13 receptor,  α 1IL1RN U65590MNDA NM_002432 Myeloid cell nuclear differentiation antigen696  X. Du et al. / Genomics 87 (2006) 693  –  703  Table 2 ( continued  )Common GenBank DescriptionCluster 2 NCF1 NM_000265 Neutrophil cytosolicfactor 1 (47 kDa,chronic granulomatousdisease, autosomal 1) NCF2 BC001606 Neutrophil cytosolicfactor 2 (65 kDa,chronic granulomatousdisease, autosomal 2) NCF4 NM_013416 Neutrophil cytosolicfactor 4, 40 kDa NFE2 L13974 Nuclear factor (erythroid-derived 2), 45 kDaPADI4 NM_012387 Peptidyl argininedeiminase, type IVQPCT NM_012413 Glutaminyl-peptidecyclotransferase(glutaminyl cyclase)RGS18 AF076642 Regulator of G-proteinsignaling 18S100A12 NM_005621 S100 calcium-binding protein A12 (calgranulin C)S100A9 NM_002965 S100 calcium-binding protein A9 (calgranulin B)SGK NM_005627 Serum/glucocorticoidregulated kinaseSLC22A4 NM_003059 Solute carrier family 22(organic cation transporter),member 4SNCA BG260394 Synuclein,  α (non-A4 component of amyloid precursor)TLR2 NM_003264 Toll-like receptor 2TLR4 U93091TLR8 AW872374TM6SF1 NM_023003 Transmembrane 6superfamily member 1TMG4 BF905445 Transmembrane γ -carboxyglutamic acid protein 4TREM1 NM_018643 Triggering receptor expressed on myeloidcells 1Cluster 3 ABCA1 NM_005502 ATP-binding cassette,subfamily A (ABC1),member 1ACSL1 NM_001995 Acyl-CoA synthetase long-chain family member 1ADM NM_001124 AdrenomedullinC4BPA NM_000715 Complement component 4 binding protein,  α CCR3 NM_001837 Chemokine (C-C motif)receptor 3CHI3L1 M80927 Chitinase 3-like 1(cartilage glycoprotein-39)CKLFSF2 AA778552 Chemokine-like factor superfamily 2CYP4F3 NM_000896 Cytochrome P450,family 4, subfamily F, polypeptide 3EMR3 AF239764 EGF-like module-containing, mucin-like,hormone receptor-like 3G0S2 NM_015714 Putative lymphocyteG0/G1 switch gene (continued on next page) Table 2 ( continued  )Common GenBank DescriptionCluster 3 GPR109B NM_006018 Putative chemokine receptor HAL NM_002108 Histidine ammonia-lyaseIL1R2 U64094 Human soluble type IIinterleukin-1 receptor mRNA, complete cdsIL8 NM_000584 Interleukin 8IL8RB NM_001557 Interleukin 8 receptor,  β KCNJ15 D87291 Potassium inwardlyrectifying channel,subfamily J, member 15KCNJ2 BF111326 Potassium inwardlyrectifying channel,subfamily J, member 2KRT23 NM_015515 Keratin 23 (histonedeacetylase inducible)MANSC1 NM_018050 Hypothetical proteinFLJ10298MGAM NM_004668 Maltase – glucoamylase( α -glucosidase)MME AI433463 Membranemetalloendopeptidase(neutral endopeptidase,enkephalinase,CALLA, CD10)MMP9 NM_004994 Matrix metalloproteinase 9(gelatinase B, 92-kDagelatinase, 92-kDa type IVcollagenase)MSCP BG251467 Mitochondrial solutecarrier proteinPBEF1 BC020691 Pre-B-cell colonyenhancing factor 1PROK2 AF182069 Prokineticin 2PTGS2 NM_000963 Prostaglandin – endoperoxidesynthase 2 (prostaglandinG/H synthase andcyclooxygenase)S100P NM_005980 S100 calcium-binding protein PSEC14L1 AI017770 SEC14-like 1( Saccharomyces cerevisiae )TNFAIP6 NM_007115 Tumor necrosis factor, α -induced protein 6TNFRSF10C AF012536 Tumor necrosis factor receptor superfamily,member 10c, decoywithout an intracellular domainVNN3 NM_018399 Vanin 3Cluster 4 AKAP2 BG540494 Paralemmin 2ANXA2 NM_004039 Annexin A2ATP1B3 U51478 ATPase, Na + /K  + transporting, β 3 polypeptideBANK1 NM_017935 B-cell scaffold proteinwith ankyrin repeats 1BLNK NM_013314 B-cell linker CCDC6 AK024913 cDNA: FLJ21260 fis,clone COL01441CPVL NM_031311 Carboxypeptidase,vitellogenic-likeCXXC5 BC006428 CXXC finger 5DPYSL2 NM_001386 Dihydropyrimidinase-like 2FCRH3 BF514552 Fc receptor-like protein 3 (continued on next page) 697  X. Du et al. / Genomics 87 (2006) 693  –  703
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks