Regulation of Microrna Expression and Abundance During Lymphopoiesis

Immunity Resource

Regulation of MicroRNA Expression and Abundance during Lymphopoiesis

Stefan Kuchen,1,10 Wolfgang Resch,1,10,* Arito Yamane,1,10 Nan Kuo,1 Zhiyu Li,1 Tirtha Chakraborty,2 Lai Wei,3 Arian Laurence,3 Tomoharu Yasuda,2 Siying Peng,2 Jane Hu-Li,4 Kristina Lu,5 Wendy Dubois,6 Yoshiaki Kitamura,7 Nicolas Charles,7 Hong-wei Sun,8 Stefan Muljo,4 Pamela L. Schwartzberg,5 William E. Paul,4 John O’Shea,3 Klaus Rajewsky,2 and Rafael Casellas1,9,* 1Genomics and Immunity, NIAMS, NIH, Bethesda, MD 20892, USA 2Immune Disease Institute and Department of Pathology, Harvard Medical School, Boston, MA 02115, USA 3Lymphocyte Cell Biology, NIAMS, NIH, Bethesda, MD 20892, USA 4Laboratory of Immunology, NIAID, NIH, Bethesda, MD 20892, USA 5Genetic Disease Research Branch, NHGRI, NIH, Bethesda, MD 20892, USA 6Laboratory of Genetics, NCI, NIH, Bethesda, MD 20892, USA 7Laboratory of Immune Cell Signaling, NIAMS, NIH, Bethesda, MD 20892, USA 8Biodata Mining and Discovery, NIAMS, NIH, Bethesda, MD 20892, USA 9Center of Cancer Research, NCI, NIH, Bethesda, MD 20892, USA 10These authors contributed equally to this work *Correspondence: [email protected] (W.R.), [email protected] (R.C.) DOI 10.1016/j.immuni.2010.05.009

SUMMARY species, suggesting these regulatory RNAs are involved in basic cellular functions across the phyla (Lagos-Quintana et al., 2001; Although the cellular concentration of miRNAs is crit- Lau et al., 2001; Lee and Ambros, 2001; Marson et al., 2008; ical to their function, how miRNA expression and Wightman et al., 1993). This view has been strengthened by abundance are regulated during ontogeny is unclear. the early embryonic lethality of mice deficient in miRNA process- We applied miRNA-, mRNA-, and ChIP-Seq to char- ing factors (Bernstein et al., 2003; Chong et al., 2008; Kanello- acterize the microRNome during lymphopoiesis poulou et al., 2005; Liu et al., 2004). within the context of the transcriptome and epige- In the mammalian genome, miRNAs are encoded within introns of protein-coding genes or as independent entities nome. We show that lymphocyte-specific miRNAs transcribed either by RNA polymerase II (Rodriguez et al., are either tightly controlled by polycomb group- 2004) or RNA polymerase III (Borchert et al., 2006). In some mediated H3K27me3 or maintained in a semi-acti- instances, groups of miRNAs are organized in genomic clusters vated epigenetic state prior to full expression. processed from a single transcript. Because of their palindromic Because of miRNA biogenesis, the cellular concen- nature, miRNAs in nascent primary transcripts (pri-miRNAs) tration of mature miRNAs does not typically reflect display a characteristic stem-loop structure that is recognized transcriptional changes. However, we uncover a and cleaved in the nucleus by the Drosha-DGCR8 complex subset of miRNAs for which abundance is dictated into 60–70 nucleotide (nt) precursor (pre) miRNAs. Once in the by miRNA gene expression. We confirm that concen- cytoplasm, pre-miRNAs are further processed by the RNase III tration of 5p and 3p miRNA strands depends largely endonuclease DICER into mature RNA fragments of 22 nt in on free energy properties of miRNA duplexes. Unex- length, which are loaded into the RNA-induced silencing complex (RISC). Partial sequence complementary between the pectedly, we also find that miRNA strand accumula- 50 end of the mature miRNA (6–8 nt seed region) and its target tion can be developmentally regulated. Our data mRNA leads to downregulation of protein expression (Bartel, provide a comprehensive map of immunity’s micro- 2009; Doench and Sharp, 2004; Kim, 2005). RNome and reveal the underlying epigenetic and As is the case for nonhematopoietic tissues, lymphocytes and transcriptional forces that shape miRNA homeo- other cells of the immune system rely on miRNAs to effect stasis. lineage commitment, proliferation, migration, and differentiation (Taganov et al., 2007; Xiao and Rajewsky, 2009). In most cases, these activities are orchestrated by both ubiquitously expressed INTRODUCTION and hematopoietic-specific miRNA species (Basso et al., 2009; Landgraf et al., 2007; Merkerova et al., 2008; Monticelli et al., MicroRNAs (miRNAs) are noncoding small RNAs that modulate 2005; Neilson et al., 2007; Wu et al., 2007). Deletion or overex- the proteome of the cell by annealing to 30 untranslated regions pression of these miRNAs impairs the immune system at various of cognate mRNAs and inhibiting protein translation and/or developmental stages (Chen et al., 2004; Li et al., 2007; O’Con- promoting mRNA instability (Bartel, 2004). Since their discovery nell et al., 2008; Rodriguez et al., 2007; Thai et al., 2007; Ventura in C. elegans (Lee et al., 1993; Wightman et al., 1993), miRNA et al., 2008; Vigorito et al., 2007; Xiao et al., 2008). Similarly, orthologs and paralogs have been described in a variety of conditional ablation of DICER or other miRNA processing factors

828 Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. Immunity Immunity’s MicroRNome

results in a profound block of both B and T cell development to a lesser extent, in other cells and tissues (Figure S1). Using (Cobb et al., 2005; Koralov et al., 2008; Muljo et al., 2005; O’Car- mice expressing the Em-BclXL transgene, which blocks activa- roll et al., 2007). It is notable that these striking phenotypes are tion-induced cell death in B cells (Fang et al., 1996), we found driven for the most part through small changes in the cellular that scRNAs are probably byproducts of mRNA degradation concentration of key factors. In the B cell compartment for and thus may be analogous to the well-characterized internu- instance, miR-150 curtails the activity of the c-Myb transcription cleosomal DNA fragmentation occurring during apoptosis factor in a dose-dependent fashion over a narrow range of (Figure S1). We conclude that the large majority of sRNAs micro- miRNA and c-Myb concentrations (Xiao et al., 2007). Similarly, sequenced belong to the miRNA family. mass action seems to be the underlying principle behind miR-155 regulation of the B cell mutator AID or miR-17 and The Immune System’s MicroRNome miR-92-mediated inhibition of the tumor suppressor Pten and With the ultimate goal of defining the microRNome of developing the proapoptotic Bim proteins (Dorsett et al., 2008; Teng et al., lymphocytes, we next normalized the entire data set on the basis 2008; Ventura et al., 2008; Xiao et al., 2007; Xiao et al., 2008). of the absolute number of miRNA reads sequenced in each These examples, which in hindsight explain the haploinsuffi- sample. To validate this strategy, we quantified fold changes in ciency observed in AID, cMyb, PTEN, and Bim heterozygous miRNAs miR-191_5p, miR-21_5p, miR-25_3p, miR-128-1_3p, mice (Bouillet et al., 2001; Di Cristofano et al., 1998; Takizawa miR-128-2_3p, let7c-1_5p, let7c-2_5p, and let-7e_5p between et al., 2008; Xiao et al., 2007), clearly demonstrate that the brain and activated B cell samples by using LNA qPCR. Reassur- absolute cellular concentration of miRNAs is crucial at managing ingly, we observed a high degree of correlation between LNA a cell’s proteome. Yet, how specific cell lineages establish and miRNA-seq (R2 = 0.995, Figure S2A). Using TaqMan tech- miRNA concentrations upon differentiation remains to be nology, we also analyzed eight miRNAs in various B and T cell defined. populations as well as HPCs and found again an overall agree-

Here, we use parallel sequencing to define chromatin modifi- ment between TaqMan and deep sequencing (r = 0.76; CI.95 = cations, the transcriptome, and microRNome of developing [0.66, 0.83]; p < 0.0001, Figure S2B). Finally, biological and tech- lymphocytes. This integrative approach reveals the epigenetic, nical replicates from mature resting B cells and neutrophils transcriptional, and, indirectly, posttranscriptional mechanisms respectively showed significant reproducibility between controlling miRNA cellular concentrations. miRNA-seq runs (r > 0.94; p < 0.0001 for all replicates, Fig- ure S2C). These analyses demonstrate that deep sequencing RESULTS can reliably profile miRNA abundance relative to the total pool of miRNAs; thus our normalization protocol permits direct Deep Sequencing of Small RNAs comparison of miRNA expression between different cells and To determine miRNA abundance during lymphopoiesis, we tissues. microsequenced small RNAs (sRNAs) 18–30 nt in length from Of the 600 mature mouse miRNAs (miRBase 13), 353 were mouse hematopoietic progenitor cells (HPCs) and downstream detected at more than 100 sequence tags per million (TPM) in T cell lineages including CD4+CD8+ thymocytes; splenic CD4+ at least one of the 40 tissues and cell types analyzed and CD8+ T cells; ConA activated CD8+ cells; ex-vivo differenti- (Table S2). Moreover, using a modified version of the miRDeep ated Th (T helper type) 1, Th2, Th17, and iTreg (induced regula- algorithm (Friedla¨ nder et al., 2008, see Experimental Proce- tory T) cells, as well as nTreg (natural regulatory T) cells; and Tfh dures), we isolated 18 putative miRNAs, the majority of which (T follicular helper) cells, which were cell sorted from secondary showed phylogenetic conservation (Data S1). Each individual lymphoid organs or spleens from immunized mice respectively tissue or cell type expressed a mean of 102 ± 28 miRNAs at (Table 1). In addition, essentially all stages of B cell ontogeny greater than 100 TPM and 162 ± 43 when the cutoff was set at were examined, from progenitor bone marrow B cells (proB) to 25 TPM. In agreement with previous studies (Landgraf et al., terminally differentiated plasma cells (Table 1). For comparative 2007), hierarchical clustering of miRNA profiling both paralleled purposes, samples from other hematopoietic lineages (mast known developmental histories and clearly separated cells of cells, basophils, neutrophils, dendritic cells, and NK cells), the immune system from other tissues (Figure 2A). Unexpect- mouse embryonic stem cells (ESCs), and fibroblasts (MEFs), edly, expression of the let-7 family of miRNAs also partitioned along with 11 adult tissues, were also included in the survey. hematopoietic from nonhematopoietic cells. For instance, let- Altogether, >300,000,000 sRNAs were sequenced. On average, 7f was predominant in the immune system, comprising in four million sequence tags per library were aligned to the some cases up to 60% of all let-7 sequences, whereas its mouse genome with 100% identity and annotated as either expression in nonhematopoietic cells was statistically lower miRNAs; noncoding (nc) RNAs (rRNAs, tRNAs, snoRNAs, or (p < 0.0001; Wilcoxon rank-sum test, Figure 2B). Conversely, snRNAs); Piwi-interacting RNAs (piRNAs); small-coding RNAs let-7c abundance was substantially greater in nonhematopoietic (scRNAs); highly repetitive RNA sequences; or unknown short cells (Figure 2B). Other let-7 family members also showed differ- RNAs (Figure 1A and Table S1 available online). With the excep- ential expression in these two compartments (data not shown). tion of testes, which harbor a large number of piRNAs (Tam et al., Thus, profiling and hierarchical clustering of miRNAs set apart 2008; Watanabe et al., 2008; Figure 1A), miRNAs were the most hematopoietic lineages from other cells and tissues. abundant RNA species identified, oscillating between 68% in ESCs displayed a large number of specific miRNAs (151), germinal center (GC) B cells and 99.4% in the lung (Figure 1B). followed by brain (52) and testes (41) (Figure 2A; Table S2 and Interestingly, scRNAs, which are presumably mRNA breakdown Data S2). In the immune system, neutrophils, basophils, as well products, were considerably enriched in GC B lymphocytes, and as T and B cells (considered as a single group) also displayed

Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. 829 Immunity Immunity’s MicroRNome

Table 1. Tissues and Cell Types Cell Type Species Organ Sorting, Puriﬁcation, and Culture Conditions ProB cells Mouse Bone marrow B220+IgMCD43+CD25 PreB cells Mouse Bone marrow B220+IgMCD43CD25+ Immature B cells Mouse Spleen B220+IgMhiCD21/CD35lo Mature B cells Mouse Spleen B220+IgMloCD21/CD35lo Marginal zone B cells Mouse Spleen B220+IgMhiCD21/CD35hi B1 B cells Mouse Peritoneal cavity B220loIgMhi Germinal center B cells Mouse Lymph nodes (immunized) CD19+CD95hi Plasma cells Mouse Lymph nodes B220CD138hi from IL6 transgenic mice LPS+IL4 B cells Mouse Spleen CD43 depleted; 50 mg/ml LPS, 2.5ng/ml IL4, 72 hs LPS+a-d-dextran B cells Mouse Spleen CD43 depleted; 50 mg/ml LPS, 2.5ng/ml a-d-dextran, 72 hs Mast cells Mouse Bone marrow 20 ng/ml IL3, 20ng/ml SCF, 8 weeks Basophils Mouse Bone marrow 20 ng/ml IL3, sorted at day 10, CD49b+Fc3RIa+CD11b+Kit Neutrophils Mouse Peritoneal cavity (casein) Percoll gradient, 88% Ly-6G+ purity Dendritic cells Mouse Bone marrow 20 ng/ml GM-CSF, non-adherent, 12 days Macrophages Mouse Bone marrow 10 ng/ml GM-CSF, 10ng/ml IL3, nonadherent, 14 days NK cells Mouse Spleen (RAG1/) 1000 U/ml IL2, adherent, 8 days Hematopoietic progenitor cells Mouse Bone marrow (5-FU) 20 ng/ml IL3, 50 ng/ml IL6, 50 ng/ml SCF, 7 days, 60% c-Kit+ScaI+Lin purity Double-positive T cells Mouse Thymus CD3+CD4+CD8+ Naive CD4+ T cells Mouse Lymph node/Spleen CD4+CD62L+CD44lowCD25- Naive CD8+ T cells Mouse Spleen CD3+CD8+CD4 ConA CD8+ T cells Mouse Spleen 5 mg/ml Concanavalin A, 5 days Th1 T cells Mouse Spleen (sorted CD4+) Protocol from Wei et al. (2009) Th2 T cells Mouse Spleen (sorted CD4+) Protocol from Wei et al. (2009) Th17 T cells Mouse Spleen (sorted CD4+) Protocol from Shi et al. (2008) T follicular helper T cells Mouse Spleen (immunized) CD4+CD44hiCXCR5hi Natural regulatory T cells Mouse Lymph node/spleen CD4+CD25+ Induced regulatory T cells Mouse Spleen (sorted CD4+) Protocol from Wei et al. (2009) ESCs Mouse Blastocysts Described in Marson et al. (2008) Mouse embryonic ﬁbroblasts Mouse E13.5 (C57B6) Passage 2 Naive B cells Human Tonsil CD19+CD38-/loCD27 Memory B cells Human Tonsil CD19+CD38-/loCD27+ Pregerminal Center B cells Human Tonsil CD19+CD38+CD77IgD+ Centroblasts Human Tonsil CD19+CD38+CD77+IgD Centrocytes Human Tonsil CD19+CD38+CD77IgD Plasma cells Human Tonsil CD19-/loCD38hiCD27hiIgD

a substantial number of preferentially expressed miRNAs (Fig- lymphocytes (Figure 2D and Data S1). This feature implies ure 2A). In addition to miR-223, which has been previously that a common regulatory pathway might drive expression of ascribed to the myeloid lineage (Chen et al., 2004), neutrophils these miRNAs. expressed high amounts of miR-26 (a and b), miR-744, A total of 49 miRNAs were preferentially upregulated in miR-340, and eight other miRNAs (Figure 2C). In striking contrast lymphocytes (Figure 3A and Table S3). As formerly shown to ESCs, hematopoietic progenitor cells displayed few unique (Chen et al., 2004), the miR-181 family was highly abundant in miRNAs, illustrated in Figure 2D by miR-193b. From basophils developing bone marrow B cells and thymocytes (Figure 3A). and mast cells we characterized a highly specific miRNA In CD4+CD8+ T cells, miR-181 miRNAs comprised 13.7% (1073496_chr3) located in intron 8 of the CPA3 gene, which (137,024 TPM) of the microRNome, which is in good accord encodes the basophil and mast cell-secreted protease carboxy- with published estimates (15.6%; Neilson et al., 2007). peptidase A (Figure 2D). We also uncovered four miRNAs with miR-128, previously implicated in neural specification and malig- preferential expression in NK cells. Notably, these miRNAs nancy, was even more dominant in CD4+CD8+ T cells (54,134 displayed nearly identical expression profiles, which, in addition TPM compared to 27,568 TPM in the brain, Figure 3B). In B cells, to NK cells, included lower expression in Tfh cells, ConA-acti- the most abundant miRNAs were miR-320 and miR-191, which vated CD8+ T cells, Th1 cells, and LPS+IL-4 stimulated B comprised as much as 10% (109,145 TPM) of all miRNAs in

830 Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. Immunity Immunity’s MicroRNome

Figure 1. sRNA Microsequencing of Hema- topoietic and Nonhematopoietic Cells (A) Pie charts representing the distribution of the six different classes of sRNAs measured from LPS+IL-4 activated B cells, HPCs, heart, and testes. (B) Percentage of miRNAs of total sRNA sequences that had at least one perfect alignment to the mouse reference genome (mm9). Excluding testes, the miRNA mean for all cell types and tissues analyzed was 91.1%. The following abbreviations are used: MEFs, mouse embryonic fibroblasts; S. Glands, salivary glands; ESCs, embryonic stem cells; HPCs, hematopoietic progenitor cells; proB, progenitor B220loCD43+ B cells; preB, B220loCD25+IgM bone marrow B cells; B1, peritoneal B220loIgMhi B cells; MZ, IgMhiCR1CR2hi marginal zone B cells; GC, CD95hiB220+ germinal center B cells; PCs, CD138hiB220 plasma cells; LPS+IL-4, mature B220+CD43 splenic B cells activated ex vivo for 72 hr in the presence of lipopolysaccharide and interleukin 4; LPS+a-d-D, mature B cells activated ex vivo for 72 hr in the presence of lipopolysaccharide and dextran conjugated anti-IgD; ConA, CD8+ T cells activated ex-vivo in the presence of concanavalin A for 72 hr; iTregs, ex-vivo induced regulatory T cells; nTregs, natural regulatory T cells isolated by cell sorting. plasma cells and up to 5% in bone marrow and some peripheral activated B cells (H2AK9Ac, H2BK5Ac, H2BK5me1, H2BK12Ac, B cells, respectively (Figure 3C). By comparison, the highest H2BK20Ac, H2BK120Ac, H3K4me1, H3K4me2, H3K9Ac, expression of miR-155 in activated B cells was 3,417 TPM H3K9me1, H3K9me2, H3K9me3, H3K4Ac, H3K14Ac, (Table S2). Other notable examples were miR-147, miR-31, H3K18Ac, H3K23Ac, H3K27Ac, H3K36Ac, H3K27me1, and miR-15b, which showed predominant expression in Th1, H3K27me2, H3K36me1, H3K36me2, H3K79me1, H3K79me2, Tfh (and basophils), and nTreg cells, respectively (Figure 3B). H3K79me3, H4K5Ac, H4K8Ac, H4K12Ac, H4K16Ac, H4K91Ac, miR-139 and miR-28 were highly exclusive to bone marrow H4K20me1, and H3K20me3), together with the histone variant and GC B lymphocytes respectively (Figure 3C). We also isolated H2A.Z, RNA polymerase II (PolII), and the histone acetyltransfer- several uncharacterized miRNAs in lymphocytes, including ase CBP-p300 (Table S4). The raw epigenetic data of miRNA 482234_chr15, which was confined to IgMhi immature B cells, genes (±5 Kb) is provided as Data S3. This bedgraph file can and 145561_chr11, present in LPS+IL-4-activated B cells, iTreg be loaded and viewed using the UCSC browser (http:// cells, Th17 cells, and ESCs (Figures 3B and 3C). Both miRNAs genome.ucsc.edu/). were phylogenetically conserved and intronic to noncoding On the basis of their expression profiles during B and T cell mRNAs (Figure 3, schematics). On the basis of these results, development, we broadly classified miRNAs as (1) inactive, (2) we conclude that miRNA-Seq reliably identifies the microRNome active, (3) poised, or (4) induced (Figure 4). As expected, miRNA signature of hematopoietic cells, including lymphocytes. genes not expressed during lymphopoiesis (e.g., the heart- specific miR-383) displayed repressive modifications such as Epigenetic Regulation of miRNA Expression H3K27me3, H3K9me3, H3K27me2, and/or H4K20me3 (Wang in Lymphocytes et al., 2008; Figure 4A and Table S4). In keeping with their lack To examine how miRNA expression is regulated during lympho- of transcription, we detected little or no PolII at promoter regions poiesis, we first mapped histone modifications at high resolution of these genes (Figure 4A and Table S4). Alternatively, miRNA across the entire genome of HPCs, proB, preB, mature resting, genes fully active across development (e.g., let-7g) displayed and LPS+IL-4 activated B cells. We concentrated on histone chromatin modifications previously linked to actively transcribed H3, lysine 4 trimethylation (H3K4me3), which demarcates both genes ((Wang et al., 2008), Figure 4B and Table S4). (We point gene promoters and transcriptional activity; H3K27me3, which out that whereas lack of posttranscriptional mechanisms prevent is associated with transcriptional inhibition; and H3K36me3, accumulation of let-7 mature miRNAs in ESCs [Viswanathan which follows transcriptional elongation. These data were et al., 2008 and Figure 4B, bar graph]; the let-7 genes are none- complemented with similar epigenetic analyses of ESCs (Marson theless fully expressed therein.) et al., 2008), as well as naive CD4+ T cells and downstream- Among miRNAs confined to specific stages of B and/or T cell differentiated populations Th1, Th2, Th17, iTreg, and nTreg cells development, our epigenetic analysis revealed at least two dis- (Wei et al., 2009). For a more comprehensive view of the epige- tinct subsets. As exemplified by miR-155 and 145561_chr11, netic landscape at and around miRNA genes, we mapped 32 one group was H3K27 demethylated in HPCs or resting cells, additional histone modifications in mature resting and LPS+IL-4 yet showed little or no expression at these early stages of

Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. 831 Immunity Immunity’s MicroRNome

Figure 2. miRNA Signatures of the Mouse Immune System, Embryonic Cells, and Adult Tissues (A) Heat map showing expression of 375 miRNAs (detected at >100 TPM) in mouse hematopoietic and nonhematopoietic cells based on hierarchical clustering analysis. miRNAs enriched in a particular sample are depicted with red bars, whereas depleted miRNAs are depicted with green bars. B and T cell populations outlined in Table 1 were combined into a single group. (B) Percentage of let-7a (gray line), let-7c (red line), and let-7f (blue line). The total let-7 sequence tags were set to 100%. Let-7f and let-7c miRNA abundance is inversely proportional in hematopoietic cells (light-blue box), whereas roughly equivalent in nonhematopoietic cells and tissues (light-red box). (C) Expression profiles of miRNAs enriched in neutrophils. Two independently processed neutrophil samples (1 and 2) are shown and the highest TPM value of the two is given in parenthesis. (D) Examples of miRNAs predominantly expressed in HPCs, mast cells, and NK cells. For uncharacterized miRNAs, their genomic location is schematized. Orientation of miRNAs and genes is represented with red and black arrowheads, respectively. Mammalian phylogenetic conservation is depicted with dense black bar graphs based on UCSC PhastCons30Way algorithm. development (Figure 4C, Table S4, and Data S3). Cell activation, well as PolII and p300 recruitment. These miRNA genes thus however, brought about a dramatic increase in miRNA expres- appear to be poised for activation early during ontogeny. On sion along with a high degree of activating modifications as the other hand, other hematopoietic-specific miRNAs, illustrated

832 Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. Immunity Immunity’s MicroRNome

Figure 3. miRNA Signatures of the Mouse B and T Cell Compartments (A) Heat map depicts expression of 49 lymphocyte-specific miRNAs. (B and C) Examples of representative miRNAs displaying specific expression in T or B lymphocytes. Only the highest TPM value is provided in each of the graphs. by miR-139 and miR-147, were more tightly regulated in that processing defects in transformed cells (Thomson et al., 2006), they retained H3K27me3 up until full miRNA gene expression it is unclear to what extent transcription per se dictates miRNA was invoked (Figures 4D, Table S4, and Data S3). Transcription cellular abundance in primary cells. To systematically address was then accompanied by an increase in active chromatin marks this issue, we quantified expression of spliced primary tran- such as H3K4me3 and H3K36me3. Importantly, in the case of scripts by mRNA-seq and compared mRNA RPKM to miRNA miR-139, which was expressed in proB and preB cells, transcrip- TPM values in six stages of B cell development (HPCs, proB, tional downregulation in mature resting and activated B cells was preB, resting mature, LPS+IL-4 activated, and GC B cells). On accompanied by a return of H3K27me3 (Figure 4D), strength- the basis of a Spearman’s rank correlation coefficient > 0.7, ening the notion that targeted expression of this miRNA subset we found coordinate expression of spliced primary transcripts is tightly regulated. On the basis of these results, we conclude and mature miRNAs in 15 cases out of 54 (27%, Figure 5B and that lymphocyte-specific miRNAs differ as to when the repres- Table S5). Within this correlated group, we found examples of sive mark deposited by polycomb group proteins, H3K27me3, miRNAs intronic to coding genes illustrated by miR-139 and is selectively lost. The potential functional impact of such distinc- miR-152 (Figure 5C), or embedded within noncoding RNAs, as tion in the context of B and T cell differentiation is discussed in the case of miR-107 and miR-362 (Table S5). We conclude below. that during B cell development, fluctuations in expression of one-quarter of miRNAs are correlated to spliced primary Transcriptional Regulation of miRNA Expression transcript amounts. in Lymphocytes One corollary of our epigenetic analysis is that, at least for Ectopically Expressed mRNA Targets Influence some miRNAs, changes in miRNA cellular concentration during miRNA Abundance lymphopoiesis are probably determined by gene transcription. The above results uncovered examples where one of two miRNA This view is consistent with the observation that clustered strands more closely parallels changes in miRNA gene expres- miRNAs in the genome display fairly similar expression profiles sion (for instance, Figure 5D), implying that mechanisms other (Figure 5A; Baskerville and Bartel, 2005; Ruby et al., 2007). The than transcription and/or posttranscriptional processing may rationale being that proximally located miRNAs are encoded influence final miRNA cellular abundance. To explore this idea, by common primary transcripts. Yet, studies of human tumors we determined the overall 5p-3p distribution for mouse miRNAs. and mouse embryonic lines have revealed considerable discrep- We found that only a minority of miRNAs (38%, or 153 out of 407) ancies between primary transcripts and mature miRNAs (Thom- display highly polarized (>99:1) 5p or 3p strand bias (Figure 6A, son et al., 2006). As such discrepancies likely reflect miRNA boxed data points), a considerable proportion (15%) show less

Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. 833 Immunity Immunity’s MicroRNome

Figure 4. Epigenetic control of miRNA expression in lymphocytes (A) Chromatin modifications associated with promoter and gene body of Sgcz (encoding the heart-specific miR-383) during lymphopoiesis. Silencing is achieved via trimethylation of lysines 27 and 9 of histone H3, and lysine 20 of histone H4. Top bar graph provides expression profiles (in TPM) of miR-383_5p. (B) The ubiquitously expressed let-7g gene displays chromatin modifications associated with active genes, including RNA polII and CBP-p300 recruitment. Bar graph plots let-7g_5p. (C) Example of a hematopoietic miRNA gene, miR-155, induced at late stages of development but depleted of H3K27me3 in progenitor cells. Modification patterns of H3K14Ac, H3K36me3, H3K36Ac, H3K79me2, H2BK12Ac, and H3K27me3, as well as polII and p300 recruitment, are shown for mature resting (black bars) and LPS+IL-4-activated (red bars) B cells. Values represent the total number of sequence tags aligned from 2.5 Kb upstream of the gene transcription start site to its transcription termination site. The bar graph depicts expression profiles of miR-155_5p. (D) Example of a hematopoietic miRNA gene, miR-139, that retains H3K27me3 inhibitory mark up until full gene expression is invoked. Chromatin modifications H3K27me3 and H3K36me3 are shown and bar graph depicts miR-139_5p TPM values. than 3-fold strand bias (Figure 6A, enclosed with dotted lines), tions against the calculated stability of the 50 and 30 ends of and the remaining 47% fall somewhere in between these two the miRNA duplex showed an overall significant correlation extremes. To determine whether this strand distribution was (r = 0.53; CI.95 = [0.42, 0.61]; p < 0.0001, Figure 6C). This finding, evolutionarily conserved, we deep-sequenced the human B which is consistent with previous studies (Schwarz et al., 2003; cell microRNome from six tonsillar populations (Data S4) and Tomari et al., 2004), argues that cellular 5p-3p distributions compared the 5p:3p ratio between mouse and human for can be partially explained by the propensity of the miRNA duplex miRNAs with exact sequence conservation (highlighted in red ends to fray. Yet, we found multiple instances where miRNA in Figure 6A). We found extensive 5p:3p correlation between strands showed differential expression profiles across the the two species (r = 0.94; CI.95 = [0.89, 0.96]; p < 0.0001, various tissues and cell types examined. A case in point is Figure 6B), indicating miRNA strand abundance might be under miR-30c2, whose 3p strand is undetected in hematopoietic similar mechanistic constraints across species. cells, but expressed at levels comparable to its complement, The uneven accumulation of 5p and 3p miRNA strands has miR-30c2_5p, in somatic cells and tissues (Figure 6D, left graph). been explained by the functional asymmetry principle, which Another example is the miR-150_5p _3p pair, which is expressed postulates that during processing of miRNA duplexes, the strand in resting immature and mature B cells as 3p > 5p, whereas this preferentially loaded into the RISC complex is the one whose 50 pattern is reversed in the T cell compartment (Figure 6D, right end is less tightly paired to its complement (Schwarz et al., 2003). graph). Other examples are miR-139, miR-329, and miR-30e The ejected passenger strand or miRNA* is degraded, and thus (Table S2). in general less abundant. Supporting the functional asymmetry At least one way the above observations can be explained is principle, a systematic comparison of 5p-3p strand concentra- through cognate mRNA target abundance, which is expected

834 Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. Immunity Immunity’s MicroRNome

Figure 5. miRNA Abundance in B Cell Development as a Function of Transcription (A) Correlation between expression and chromosome distance separating mouse miRNAs. Data points represent all possible pair combinations for miRNAs located in the same chromosome. The y axis provides the Pearson correlation coefficient calculated for tissue expression patterns, whereas the x axis gives the physical distance between the miRNAs in kilobases (Kb). (B) The transcriptome of HPCs, proB cells, preB cells and mature resting, LPS+IL-4 activated, and germinal center B cells as determined by deep sequencing and normalized as RPKM values. Each data point represents one of 126 mRNA-miRNA pairs (Table S3) at any one of the six B cell developmental stages. (C) Examples of mRNA/miRNA pairs that show correlation (r > 0.7, upper row) or no correlation (r < 0.7, lower row) across the six B cell developmental stages. Black bars represent mRNAs RPKM values (left y axis), whereas magenta bars depict miRNAs TPM values (right y axis). (D) Examples of miRNAs with one strand (5p in both cases) showing greater correlation to mRNA expression profiles than the other. to vary throughout ontogeny and could in principle impact the miR-30a_5p (2840 versus 868 TPM, Figure S4B), indicating stability and/or half-life of miRNA-RISC complexes. To explore that downregulation of miR-30_5p by mRNA targets is con- this possibility, we expressed in A70 proB cells miR-30_5p served between mice and humans. We confirmed these results complementary target sequences (mirTs; Gentner et al., 2009), by targeting 4 additional endogenous miRNAs in 293T cells. which are recognized by the seed domain of the entire miR-30 Interestingly, we found examples of both specific upregulation family of miRNAs. We found that mirT expression affected (miR-744_5p) and downregulation (miR-25_3p, miR-128_3p, in highly specific fashion the steady-state abundance of miR-191_5p) of miRNAs upon cognate mRNA expression miR-30d_5p and miR-30e_5p, the two miR-30 members (Figure S4C). We conclude that the steady-state abundance of expressed at significant amounts in A70 cells (Figure S4A). miRNA strands can be influenced by ectopic expression of Specifically, miR-30_5p amounts declined relative to control complementary mRNA sequences. As proposed below, this upon cognate mRNA expression: 463 versus 3320 TPM for feature might help explain the changes in 3p:5p strand ratios miR-30d_5p and 330 versus 1237 TPM for miR-30e_5p observed during development. (Figure S4A). Attesting to the high specificity of the mirT-miRNA interaction, no other endogenous miRNAs (including miR-30_3p DISCUSSION arms) were drastically affected by miR-30_5pT expression. Analogously, expression of miR-30_5pTs in 293T cells resulted The hematopoietic microRNome can be set apart from that of in specific depletion of 5p arms from the two expressed somatic cells and tissues by a distinctive miRNA signature, miR-30 members: miR-30e_5p (2001 versus 7693. TPM) and differential expression patterns of let7, and a unique 5p:3p

Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. 835 Immunity Immunity’s MicroRNome

Figure 6. 5p:3p Fluctuations during Lymphocyte Development

(A) 5p versus 3p species abundance of mouse miRNAs (graphed as log2 of TPM values). Closed red circles depict miRNAs entirely conserved between mouse and human. Dotted lines delimit all miRNAs that display roughly equal numbers of 5p and 3p species. (B) Correlation of 5p/3p ratios between mouse and human. Human miRNAs were microsequenced from 6 tonsillar B cell populations isolated by cell sorting: naive, pre-germinal center, centrocytes, centroblasts, plasma cells, and memory B cells. For mouse, the entire panel of miRNA samples was used. Only miRNAs entirely conserved between the two species were plotted. Both the Spearman correlation and P values are provided. (C) Comparative analysis of miRNA 5p and 3p strand abundance (y axis) versus dissociation energy of 50 and 30 miRNA duplex ends (measured in Kcal/mol with RNAduplex software). The plot shows that as the relative thermodynamic stability of the miRNA duplex 30 end increases (represented by positive values on the x axis), its 50 end is preferentially unwound, leading to asymmetric uploading of its 5p strand into the RISC complex and an overall increase in 5p cellular abundance (positive end of the y axis). Data at the negative end of the x and y axes provide the counterexample. (D) The left plot shows a relative abundance of miR-30c2_5p (red bars) and 3p (blue bars) strands across all cells and samples examined. Hematopoietic (light-blue box) and nonhematopoietic (light-red box) cells can be subdivided according to the absence or presence of miR-30c2_3p. The right plot shows analysis of miR-150_5p (red bars) and 3p (blue bars) distribution in various B and T cell populations. ‘‘Mature (2)’’ and ‘‘CD8 (2)’’ refer to duplicate samples. distribution for selected miRNAs. Previous studies uncovered basis of a probabilistic model of miRNA biogenesis (Friedla¨ nder miRNAs circumscribed to or enriched in the immune system, et al., 2008; data not shown). Of these, however, only a small such as miR-150, miR-155, miR-223, miR-146, and the fraction (18 in total) was detected at amounts greater than miR-181 family (Barski et al., 2009; Basso et al., 2009; Chen 100 TPM. It is interesting to note therefore that the overwhelming et al., 2004; Landgraf et al., 2007; Merkerova et al., 2008; Mon- majority of putative miRNAs uncovered by our analysis appear to ticelli et al., 2005; Neilson et al., 2007; Taganov et al., 2006; represent low-abundant hairpin species that might on occasion Wu et al., 2007; Zhang et al., 2009). Our comparative cluster be processed by the RNAi pathway. Admittedly, we cannot analysis corroborated these findings and further assigned rule out that some of these putative miRNAs are expressed at previously characterized miRNAs to the hematopoietic system: considerable levels in tissues not examined in our survey. miR-15b (nTreg cells), miR-26a/b (neutrophils), miR-28 (GC B Furthermore, the arbitrary 100 TPM threshold is not a predictor cells), miR-107 (basophils), miR-320 and miR-148a (plasma of physiological activity, which in principle could still be signifi- cells), miR-128 (CD4+CD8+ cells), miR-221 and miR-222 cant at very low levels of miRNA expression. In light of our (Tfh-basophils), miR-147 (Th1 cells), miR-182 (NK cells), and findings however, it is reasonable to propose that most mouse miR-193b (HPCs). Of note, both miR-28 and miR-148a were and human miRNAs have already been characterized. This also highly specific for human GC and plasma cells, respectively idea is supported by the fact that recent large-scale cloning (Figure S6). In addition, we identified 14 mouse miRNAs that efforts yielded few uncharacterized, mostly low-abundant, show preferential expression in various cells of the immune miRNAs (Basso et al., 2009; Landgraf et al., 2007). system In the context of gene discovery, it is important to point The combining high-throughput miRNA-seq, mRNA-seq, and out that of the >300,000,000 sequence reads analyzed, as ChIP-seq provides an unprecedented view of the various regula- many as 2,406 putative mouse miRNAs were predicted on the tory steps that shape miRNA expression and abundance during

836 Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. Immunity Immunity’s MicroRNome

lymphopoiesis. At the epigenetic level, the methylation status of miRNA editing, or miRNA nuclear transport. In addition to these, H3K27 partitioned lymphocyte-specific miRNA genes into two we have found that ectopically expressed mRNA targets can distinct subsets. Those belonging to the miR-139 and miR-147 both increase or decrease a particular miRNA pool. Whether group were found to retain the repressive H3K27me3 mark endogenous mRNAs can likewise influence fluctuations in throughout development up until miRNA gene transcription miRNA abundance remains to be determined. Nevertheless, it was elicited. miRNA genes of the miR-155/145561_chr11 group is reasonable to propose that an increase in the absolute number were H3K27 demethylated at earlier stages of development, and of mRNA targets may impact the balance and/or half-life of thus they appeared to be poised for activation prior to full ‘‘free’’ miRNA, miRNA-RISC, and miRNA-RISC-mRNA com- transcription. The rationale for this subdivision might be provided plexes. We find this scenario intriguing because it raises the by the spatiotemporal context in which these miRNAs are possibility that under physiological conditions, miRNAs and expressed. miR-147 for instance is induced as naive CD4+ mRNAs regulate each other’s homeostasis. T cells differentiate into the Th1 cell lineage. This process of In summary, our studies reveal some of the epigenetic, tran- fate determination relies on positive epigenetic reprogramming scriptional, and posttranscriptional strategies that help orches- and expression of key factors, and negative regulation of trate cellular abundance of miRNAs during lymphopoiesis. The competing pathways driving differentiation to other T helper hematopoietic miRNA signatures provided by the data represent cell types (Wei et al., 2009). Similarly, miR-139 derepression a valuable resource that will help guide future gene-targeting occurs as part of another fate determination step, from pluripo- experiments of individual miRNAs. In this respect, a major chal- tent hematopoietic progenitor cells to pro-B cells. Although the lenge in the field has been the identification of critical miRNAs: precise role of miR-139 and miR-147 remains to be determined, miRNA targets driving developmental decisions. A strategy to the expectation would be that miRNAs involved in symmetric or solve this problem is based on the observation that cellular progressive lymphocyte lineage commitment would be tightly concentration of miRNA targets fluctuates as a function of regulated by robust mechanisms. One such mechanism may cognate miRNA expression (e.g., miR-150:c-Myb [Xiao et al., be polycomb group-mediated H3K27 trimethylation, which 2007]). In principle, by applying microsequencing and bioinfor- might ensure induction of cell-fate determination only at the matics to a large number of developmental stages, it should be appropriate time. possible to predict functional miRNA:miRNA target pairs. Our Our data have also revealed lymphocyte-induced miRNA preliminary studies using the miRNA- and mRNA-Seq data genes that appear to be epigenetically poised for transcription from B cell development support this view. We anticipate that early in development. These miRNA genes are associated with genomic approaches such as this will help unravel how miRNAs some activating histone marks, are H3K27me3 depleted, recruit regulate development and effector functions of the immune low levels of PolII, but lack H3K36me3 or H3K79me2—hallmarks system. of polymerase elongation. This subinduced chromatin state is reflected by the low, but detectable, miRNA expression prior to EXPERIMENTAL PROCEDURES full gene transcription. For miR-155, 145561_chr11, and other miRNAs under this category, full expression occurs as resting RNA Isolation and Quality Control lymphocytes are exposed to mitogens and cytokines typically Sorted or cultured cells were collected by centrifugation, dissolved in Trizol at a concentration of 5 to 20 3 106 cells per milliliter, and stored at 80C. Total released during the immune response. The benefits of maintain- RNA was isolated in accordance with the manufacturer’s recommendations. ing critical miRNA genes like miR-155 in a preexisting subin- The RNA integrity number (RIN) for all samples was assessed from 100 ng of duced state maybe best understood in the cadre of viral or total RNA with Agilent RNA 6000 Nano Kit and Bioanalyzer 2100 (Agilent bacterial infections, which require rapid T and B cell responses. Technologies). The RIN of all mouse and human cell populations was at least On the basis of these considerations, it will be important to deter- 9 and the RIN of tissues was at least 8 except for pancreas (RIN 7.3) and skin mine whether miRNAs strictly dependent on H3K27me3 for (RIN 7). expression are involved in lymphocyte developmental decisions, Small RNA Profiling by Illumina Deep-Sequencing while H3K27me3-independent ones drive B and T cell activation. Small RNA sample preparation was done in accordance with Illumina’s Ongoing studies indicate that this might be indeed the case. protocol. In brief, 50 and 30 adapters were sequentially ligated to small RNA Transcriptionally, we have shown that expression of at least of 18–30 bases gel purified from 5–10 mg total RNA. Adaptor-ligated small one-quarter of mature miRNAs closely follows spliced primary RNA was reverse transcribed, amplified by 15 PCR cycles with high-fidelity transcripts as determined by mRNA-Seq. Because mRNAs are Phusion polymerase (Finnzymes), and sequenced on a Genome Analyzer stabilized by polyadenylation and other mechanisms, this is (Illumina) in accordance with the manufacturer’s instructions. likely to be an underestimate, e.g., the correlation between Epigenetic Analysis mature miRNAs and unspliced pri-miRNAs is expected to be Cells were fixed with 1% paraformaldehyde at 37C for 10 min and either more significant. Regardless of the precise number, the results MNase-treated for ChIP or sonicated for Polymerase II and p300 IP. Precipi- were nonetheless unexpected in light of the fact that studies tated DNA fragments were processed in accordance with Illumina’s protocol with cell lines fail to find any such correlation (Thomson et al., and sequenced on a Genome Analyzer with manufacturer’s instructions. 2006). Transformed cells, however, display a global reduction During analysis, short sequence reads were trimmed to 25 nts and aligned to in mature miRNAs (Lee et al., 2008; Lu et al., 2005), presumably the mouse genome with either ELAND or Bowtie. Comparison data for mESC and T cell populations were obtained from public databases (Marson et al., from a general deficiency in miRNA processing. Alternatively, in 2008; Wei et al., 2009). Uniquely aligned reads were analyzed by SICER primary cells several posttrancriptional mechanisms are bound (Zang et al., 2009) with an expectation value E of 50 in a random background to promote deviations between transcription rate and mature model. Reads on significant islands as defined by SICER were normalized to miRNA expression, such as efficiency of miRNA processing, the total number of reads on islands. Downstream analysis was carried out in R.

Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. 837 Immunity Immunity’s MicroRNome

Quantitative RT-PCR Bouillet, P., Cory, S., Zhang, L.C., Strasser, A., and Adams, J.M. (2001). RNA samples used for quantitative RT-PCR were isolated with the mirVana Degenerative disorders caused by Bcl-2 deficiency prevented by loss of its miRNA isolation kit (ABI, part No: 1561) in accordance with the manufacturer’s BH3-only antagonist Bim. Dev. Cell 1, 645–653. instructions. cDNA generated from 10 ng of total RNA by the TaqMan micro- Chen, C.Z., Li, L., Lodish, H.F., and Bartel, D.P. (2004). MicroRNAs modulate RNA reverse transcription kit (ABI, part No: 4366596) was applied for each hematopoietic lineage differentiation. Science 303, 83–86. TaqMan qRT-PCR reaction with TaqMan 2 3 Universal PCR Master Mix (ABI, part No: 4324018). All TaqMan qRT-PCR reactions were performed Chong, M.M., Rasmussen, J.P., Rudensky, A.Y., Rundensky, A.Y., and and analyzed by StepOnePlus Real-time PCR system (ABI) with the following Littman, D.R. (2008). The RNAseIII enzyme Drosha is critical in T cells for cycle conditions: 95C for 10 min, 1 cycle; 95C for 15 s and 60C for 60 s, preventing lethal inflammatory disease. J. Exp. Med. 205, 2005–2017. 50 cycles. The expression of individual miRNA was normalized and expressed Cobb, B.S., Nesterova, T.B., Thompson, E., Hertweck, A., O’Connor, E., as a percentage relative to U6 with the following formula: fold induction = Godwin, J., Wilson, C.B., Brockdorff, N., Fisher, A.G., Smale, S.T., and [DCt] 2 , where DCt = Ct(target) Ct(U6). Merkenschlager, M. (2005). T cell lineage choice and differentiation in the For a comparison of brain samples to activated B cells, total RNA was sent absence of the RNase III enzyme Dicer. J. Exp. Med. 201, 1367–1373. to Exicon for commercial miRCURY-based microRNA quantification of Di Cristofano, A., Pesce, B., Cordon-Cardo, C., and Pandolfi, P.P. (1998). Pten selected microRNAs (mmu-let-7c, mmu-let-7e, mmu-miR-21, mmu-miR-25, is essential for embryonic development and tumour suppression. Nat. Genet. mmu-miR-128, and mmu-miR-191). 19, 348–355. Doench, J.G., and Sharp, P.A. (2004). Specificity of microRNA target selection ACCESSION NUMBERS in translational repression. Genes Dev. 18, 504–511. Dorsett, Y., McBride, K.M., Jankovic, M., Gazumyan, A., Thai, T.H., Robbiani, All sequence data are available in the Gene Expression Omnibus (GEO) D.F., Di Virgilio, M., San-Martin, B.R., Heidkamp, G., Schwickert, T.A., et al. database (http://www.ncbi.nlm.nih.gov/gds) under the accession number (2008). MicroRNA-155 suppresses activation-induced cytidine deaminase- GSE21630. mediated Myc-Igh translocation. Immunity 28, 630–638. Fang, W., Mueller, D.L., Pennell, C.A., Rivard, J.J., Li, Y.S., Hardy, R.R., SUPPLEMENTAL INFORMATION Schlissel, M.S., and Behrens, T.W. (1996). Frequent aberrant immunoglobulin gene rearrangements in pro-B cells revealed by a bcl-xL transgene. Immunity Supplemental Information includes three figures, five tables, four data collec- 4, 291–299. tions, and Supplemental Experimental Procedures and can be found with this Friedla¨ nder, M.R., Chen, W., Adamidi, C., Maaskola, J., Einspanier, R., article online at doi:10.1016/j.immuni.2010.05.009. Knespel, S., and Rajewsky, N. (2008). Discovering microRNAs from deep sequencing data using miRDeep. Nat. Biotechnol. 26, 407–415. ACKNOWLEDGMENTS Gentner, B., Schira, G., Giustacchini, A., Amendola, M., Brown, B.D., Ponzoni, M., and Naldini, L. (2009). Stable knockdown of microRNA in vivo by lentiviral We thank members of the Casellas lab for discussions; J. Newman and vectors. Nat. Methods 6, 63–66. R. Young for mouse ESCs; J. Simone for cell sorting; G. Gutierrez for technical Kanellopoulou, C., Muljo, S.A., Kung, A.L., Ganesan, S., Drapkin, R., assistance with the genome analyzer; B. Wold for the mRNA-seq protocol; and Jenuwein, T., Livingston, D.M., and Rajewsky, K. (2005). Dicer-deficient L. Naldini for the LV-SFFV lentiviral vector. This work was supported in part by mouse embryonic stem cells are defective in differentiation and centromeric the Intramural Research Program of NIAMS-NIH. S. K. was supported by the silencing. Genes Dev. 19, 489–501. Swiss Foundation for Grants in Biology and Medicine. Kim, V.N. (2005). MicroRNA biogenesis: Coordinated cropping and dicing. Nat. Rev. Mol. Cell Biol. 6, 376–385. Received: October 18, 2009 Revised: March 22, 2010 Koralov, S.B., Muljo, S.A., Galler, G.R., Krek, A., Chakraborty, T., Accepted: April 8, 2010 Kanellopoulou, C., Jensen, K., Cobb, B.S., Merkenschlager, M., Rajewsky, Published online: June 3, 2010 N., and Rajewsky, K. (2008). Dicer ablation affects antibody diversity and cell survival in the B lymphocyte lineage. Cell 132, 860–874.

REFERENCES Lagos-Quintana, M., Rauhut, R., Lendeckel, W., and Tuschl, T. (2001). Identification of novel genes coding for small expressed RNAs. Science 294, Barski, A., Jothi, R., Cuddapah, S., Cui, K., Roh, T.Y., Schones, D.E., and 853–858. Zhao, K. (2009). Chromatin poises miRNA- and protein-coding genes for Landgraf, P., Rusu, M., Sheridan, R., Sewer, A., Iovino, N., Aravin, A., Pfeffer, expression. Genome Res. 19, 1742–1751. S., Rice, A., Kamphorst, A.O., Landthaler, M., et al. (2007). A mammalian Bartel, D.P. (2004). MicroRNAs: Genomics, biogenesis, mechanism, and microRNA expression atlas based on small RNA library sequencing. Cell function. Cell 116, 281–297. 129, 1401–1414. Bartel, D.P. (2009). MicroRNAs: Target recognition and regulatory functions. Lau, N.C., Lim, L.P., Weinstein, E.G., and Bartel, D.P. (2001). An abundant Cell 136, 215–233. class of tiny RNAs with probable regulatory roles in Caenorhabditis elegans. Science 294, 858–862. Baskerville, S., and Bartel, D.P. (2005). Microarray profiling of microRNAs reveals frequent coexpression with neighboring miRNAs and host genes. Lee, R.C., and Ambros, V. (2001). An extensive class of small RNAs in Caeno- RNA 11, 241–247. rhabditis elegans. Science 294, 862–864. Basso, K., Sumazin, P., Morozov, P., Schneider, C., Maute, R.L., Kitagawa, Y., Lee, R.C., Feinbaum, R.L., and Ambros, V. (1993). The C. elegans hetero- Mandelbaum, J., Haddad, J., Jr., Chen, C.Z., Califano, A., and Dalla-Favera, R. chronic gene lin-4 encodes small RNAs with antisense complementarity to (2009). Identification of the human mature B cell miRNome. Immunity 30, lin-14. Cell 75, 843–854. 744–752. Lee, E.J., Baek, M., Gusev, Y., Brackett, D.J., Nuovo, G.J., and Schmittgen, Bernstein, E., Kim, S.Y., Carmell, M.A., Murchison, E.P., Alcorn, H., Li, M.Z., T.D. (2008). Systematic evaluation of microRNA processing patterns in tissues, Mills, A.A., Elledge, S.J., Anderson, K.V., and Hannon, G.J. (2003). Dicer is cell lines, and tumors. RNA 14, 35–42. essential for mouse development. Nat. Genet. 35, 215–217. Li, Q.J., Chau, J., Ebert, P.J., Sylvester, G., Min, H., Liu, G., Braich, R., Borchert, G.M., Lanier, W., and Davidson, B.L. (2006). RNA polymerase III Manoharan, M., Soutschek, J., Skare, P., et al. (2007). miR-181a is an intrinsic transcribes human microRNAs. Nat. Struct. Mol. Biol. 13, 1097–1101. modulator of T cell sensitivity and selection. Cell 129, 147–161.

838 Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. Immunity Immunity’s MicroRNome

Liu, J., Carmell, M.A., Rivas, F.V., Marsden, C.G., Thomson, J.M., Song, J.J., (2008). Pseudogene-derived small interfering RNAs regulate gene expression Hammond, S.M., Joshua-Tor, L., and Hannon, G.J. (2004). Argonaute2 is the in mouse oocytes. Nature 453, 534–538. catalytic engine of mammalian RNAi. Science 305, 1437–1441. Teng, G., Hakimpour, P., Landgraf, P., Rice, A., Tuschl, T., Casellas, R., and Lu, J., Getz, G., Miska, E.A., Alvarez-Saavedra, E., Lamb, J., Peck, D., Papavasiliou, F.N. (2008). MicroRNA-155 is a negative regulator of activa- Sweet-Cordero, A., Ebert, B.L., Mak, R.H., Ferrando, A.A., et al. (2005). tion-induced cytidine deaminase. Immunity 28, 621–629. MicroRNA expression profiles classify human cancers. Nature 435, 834–838. Thai, T.H., Calado, D.P., Casola, S., Ansel, K.M., Xiao, C., Xue, Y., Murphy, A., Marson, A., Levine, S.S., Cole, M.F., Frampton, G.M., Brambrink, T., Frendewey, D., Valenzuela, D., Kutok, J.L., et al. (2007). Regulation of the Johnstone, S., Guenther, M.G., Johnston, W.K., Wernig, M., Newman, J., germinal center response by microRNA-155. Science 316, 604–608. et al. (2008). Connecting microRNA genes to the core transcriptional regulatory Thomson, J.M., Newman, M., Parker, J.S., Morin-Kensicki, E.M., Wright, T., circuitry of embryonic stem cells. Cell 134, 521–533. and Hammond, S.M. (2006). Extensive post-transcriptional regulation of Merkerova, M., Belickova, M., and Bruchova, H. (2008). Differential expression microRNAs and its implications for cancer. Genes Dev. 20, 2202–2207. of microRNAs in hematopoietic cell lineages. Eur. J. Haematol. 81, 304–310. Tomari, Y., Matranga, C., Haley, B., Martinez, N., and Zamore, P.D. (2004). Monticelli, S., Ansel, K.M., Xiao, C., Socci, N.D., Krichevsky, A.M., Thai, T.H., A protein sensor for siRNA asymmetry. Science 306, 1377–1380. Rajewsky, N., Marks, D.S., Sander, C., Rajewsky, K., et al. (2005). MicroRNA Ventura, A., Young, A.G., Winslow, M.M., Lintault, L., Meissner, A., Erkeland, profiling of the murine hematopoietic system. Genome Biol. 6, R71. S.J., Newman, J., Bronson, R.T., Crowley, D., Stone, J.R., et al. (2008). Muljo, S.A., Ansel, K.M., Kanellopoulou, C., Livingston, D.M., Rao, A., and Targeted deletion reveals essential and overlapping functions of the miR-17 Rajewsky, K. (2005). Aberrant T cell differentiation in the absence of Dicer. J. through 92 family of miRNA clusters. Cell 132, 875–886. Exp. Med. 202, 261–269. Vigorito, E., Perks, K.L., Abreu-Goodger, C., Bunting, S., Xiang, Z., Kohlhaas, Neilson, J.R., Zheng, G.X., Burge, C.B., and Sharp, P.A. (2007). Dynamic S., Das, P.P., Miska, E.A., Rodriguez, A., Bradley, A., et al. (2007). microRNA- regulation of miRNA expression in ordered stages of cellular development. 155 regulates the generation of immunoglobulin class-switched plasma cells. Genes Dev. 21, 578–589. Immunity 27, 847–859. Viswanathan, S.R., Daley, G.Q., and Gregory, R.I. (2008). Selective blockade O’Carroll, D., Mecklenbrauker, I., Das, P.P., Santana, A., Koenig, U., Enright, of microRNA processing by Lin28. Science 320, 97–100. A.J., Miska, E.A., and Tarakhovsky, A. (2007). A Slicer-independent role for Argonaute 2 in hematopoiesis and the microRNA pathway. Genes Dev. 21, Wang, Z., Zang, C., Rosenfeld, J.A., Schones, D.E., Barski, A., Cuddapah, S., 1999–2004. Cui, K., Roh, T.Y., Peng, W., Zhang, M.Q., and Zhao, K. (2008). Combinatorial patterns of histone acetylations and methylations in the human genome. Nat. O’Connell, R.M., Rao, D.S., Chaudhuri, A.A., Boldin, M.P., Taganov, K.D., Genet. 40, 897–903. Nicoll, J., Paquette, R.L., and Baltimore, D. (2008). Sustained expression of microRNA-155 in hematopoietic stem cells causes a myeloproliferative Watanabe, T., Totoki, Y., Toyoda, A., Kaneda, M., Kuramochi-Miyagawa, S., disorder. J. Exp. Med. 205, 585–594. Obata, Y., Chiba, H., Kohara, Y., Kono, T., Nakano, T., et al. (2008). Endoge- nous siRNAs from naturally formed dsRNAs regulate transcripts in mouse Rodriguez, A., Griffiths-Jones, S., Ashurst, J.L., and Bradley, A. (2004). oocytes. Nature 453, 539–543. Identification of mammalian microRNA host genes and transcription units. Genome Res. 14(10A), 1902–1910. Wei, G., Wei, L., Zhu, J., Zang, C., Hu-Li, J., Yao, Z., Cui, K., Kanno, Y., Roh, T.Y., Watford, W.T., et al. (2009). Global mapping of H3K4me3 and H3K27me3 Rodriguez, A., Vigorito, E., Clare, S., Warren, M.V., Couttet, P., Soond, D.R., reveals specificity and plasticity in lineage fate determination of differentiating van Dongen, S., Grocock, R.J., Das, P.P., Miska, E.A., et al. (2007). CD4+ T cells. Immunity 30, 155–167. Requirement of bic/microRNA-155 for normal immune function. Science Wightman, B., Ha, I., and Ruvkun, G. (1993). Posttranscriptional regulation of 316, 608–611. the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in Ruby, J.G., Stark, A., Johnston, W.K., Kellis, M., Bartel, D.P., and Lai, E.C. C. elegans. Cell 75, 855–862. (2007). Evolution, biogenesis, expression, and target predictions of a substan- Wu, H., Neilson, J.R., Kumar, P., Manocha, M., Shankar, P., Sharp, P.A., and tially expanded set of Drosophila microRNAs. Genome Res. 17, 1850–1864. Manjunath, N. (2007). miRNA profiling of naı¨ve, effector and memory CD8 Schwarz, D.S., Hutva´ gner, G., Du, T., Xu, Z., Aronin, N., and Zamore, P.D. T cells. PLoS ONE 2, e1020. (2003). Asymmetry in the assembly of the RNAi enzyme complex. Cell 115, Xiao, C., and Rajewsky, K. (2009). MicroRNA control in the immune system: 199–208. Basic principles. Cell 136, 26–36. Shi, G., Cox, C.A., Vistica, B.P., Tan, C., Wawrousek, E.F., and Gery, I. (2008). Xiao, C., Calado, D.P., Galler, G., Thai, T.H., Patterson, H.C., Wang, J., Phenotype switching by inflammation-inducing polarized Th17 cells, but not Rajewsky, N., Bender, T.P., and Rajewsky, K. (2007). MiR-150 controls B by Th1 cells. J. Immunol. 181, 7205–7213. cell differentiation by targeting the transcription factor c-Myb. Cell 131, Taganov, K.D., Boldin, M.P., Chang, K.J., and Baltimore, D. (2006). NF-kap- 146–159. paB-dependent induction of microRNA miR-146, an inhibitor targeted to Xiao, C., Srinivasan, L., Calado, D.P., Patterson, H.C., Zhang, B., Wang, J., signaling proteins of innate immune responses. Proc. Natl. Acad. Sci. USA Henderson, J.M., Kutok, J.L., and Rajewsky, K. (2008). Lymphoproliferative 103, 12481–12486. disease and autoimmunity in mice with increased miR-17-92 expression in Taganov, K.D., Boldin, M.P., and Baltimore, D. (2007). MicroRNAs and lymphocytes. Nat. Immunol. 9, 405–414. immunity: Tiny players in a big field. Immunity 26, 133–137. Zang, C., Schones, D.E., Zeng, C., Cui, K., Zhao, K., and Peng, W. (2009). Takizawa, M., Tolarova´ , H., Li, Z., Dubois, W., Lim, S., Callen, E., Franco, S., A clustering approach for identification of enriched domains from histone Mosaico, M., Feigenbaum, L., Alt, F.W., et al. (2008). AID expression levels modification ChIP-Seq data. Bioinformatics 25, 1952–1958. determine the extent of cMyc oncogenic translocations and the incidence of Zhang, J., Jima, D.D., Jacobs, C., Fischer, R., Gottwein, E., Huang, G., Lugar, B cell tumor development. J. Exp. Med. 205, 1949–1957. P.L., Lagoo, A.S., Rizzieri, D.A., Friedman, D.R., et al. (2009). Patterns of Tam, O.H., Aravin, A.A., Stein, P., Girard, A., Murchison, E.P., Cheloufi, S., microRNA expression characterize stages of human B-cell differentiation. Hodges, E., Anger, M., Sachidanandam, R., Schultz, R.M., and Hannon, G.J. Blood 113, 4586–4594.

Immunity 32, 828–839, June 25, 2010 ª2010 Elsevier Inc. 839