DICER IN IMMUNE CELL DEVELOPMENT AND FUNCTION

Abstract

Dicer is an enzyme of the RNase III endoribonuclease family, which is crucial for RNA interference (RNAi) in eukaryotes. Dicer is a component of the protein machinery (the RNA Induced Silencing Complex [RISC]) which is involved in catalyzing the formation of mature microRNAs from their precursors in the process of microRNA biogenesis. RISC-associated microRNAs bind to specific sequences in the 3’ untranslated region of cognate mRNAs largely through complementary base pairing, resulting in either translational inhibition and/or the degradation of a specific mRNA pool. MicroRNAs epigenetically regulate the cellular levels of receptors, transcription factors and signaling proteins that govern the developmental pathways and functions of multiple cellular processes. The pivotal role played by Dicer in microRNA formation has also piqued the interest of molecular immunologists who have sought to understand the biological relevance of microRNAs in the development and function of the immune system. Here, we review the major findings of these studies and provide an overview of the role of Dicer and microRNAs in immune cell development and function. Additionally, we highlight deficiencies in our knowledge and new research areas that may enhance our understanding of the role of Dicer and microRNAs in immunity.

INTRODUCTION

Dicer is a class III endoribonuclease discovered in the laboratory of Gregory Hannon whose research employed Drosophila cells to identify the factors involved in RNA interference – a process wherein small non-coding RNAs interact with cognate messenger RNAs, resulting in the regulation of gene expression (Bernstein et al., 2001). This work showed that Dicer is integral to the process of RNAi and functions by cleaving double stranded RNAs into small interfering RNA (siRNA) that are 22 nucleotides in length. Moreover, it was demonstrated through phylogenetic analysis that the Dicer protein is well conserved among eukaryotes. Genes encoding Dicer-like proteins that perform similar functions have been found it ciliates, nematodes, arthropods, fungi and plants, indicating the appearance of Dicer early in eukaryotic evolution (Murphy et al., 2008). It is now known that the dicer1 gene, which encodes the Dicer protein, is located on chromosome 14 in humans and on chromosome 12 in mice.

MicroRNAs (miRs) are a family of endogenously derived non-coding RNAs that epigenetically regulate gene expression (He and Hannon, 2004). They were first described by Rosalind Lee in Caenorhabditis elegans while investigating the regulation of the LIN-14 protein by a small RNA derived from the lin-4 gene (Lee et al., 1993). Subsequent studies showed that microRNAs exist across a wide range of phyla and established in the literature as major posttranscriptional gene regulators. It is estimated that ~60% of the human genome may be regulated by microRNAs (Friedman et al., 2009).

The protein machinery that is involved in the formation and functioning of microRNAs incudes the enzyme Dicer which is required for microRNA biogenesis - a process in which mature microRNAs are formed from their immature precursors (Kim et al., 2005). This process begins in the nucleus, wherein RNA polymerase II transcribes genomic DNA containing microRNA sequences, giving rise to pri-microRNAs. Pri-microRNAs are further processed into pre-microRNAs by a nuclear protein complex called the microprocessor complex. Pre-microRNAs are transported from the nucleus to the cytoplasm by Exportin-5. Subsequently, they are loaded onto a protein complex called the RNA Induced Silencing Complex (RISC). RISC is composed of Dicer, Argonaute-2, the Tar RNA Binding Protein (TRBP) as well as other proteins whose functions are yet to be clearly defined (Koscianska et al., 2001). Once pre-microRNAs have been RISC-loaded, they are cleaved to their mature form (~22nt in length) by Dicer. The mature microRNAs, while still associated with the RISC, are capable of binding their cognate mRNA target through microRNA-mRNA interactions. This occurs largely through complementary base pairing between a sequence on the microRNA called the seed region and the 3’ untranslated region on the target mRNA, leading to either translational inhibition and/or mRNA degradation (Krol et al., 2010). It therefore follows that Dicer loss-of-function studies may provide a useful method for analyzing the phenotypic variations, which occur in cells when microRNA production is altered.

DICER LOSS-OF-FUNCTION STUDIES

The biological relevance of Dicer and microRNAs in regulating immune cell functions have been studied in loss-of-function experiments conducted by several research groups (Alemdehy et al., 2012; Cobb et al., 2005; Cobb et al., 2006; Fedeli et al., 2009; Koralov et al., 2008; Kuipers et al., 2010; Liston et al., 2008; Muljo et al., 2005; Sissons et al., 2012; Xu et al., 2012; Zhou et al., 2008; Zhou et al., 2009). These studies however, cannot be pursued through a conventional genetic knockout approach since disrupting the dicer1 gene results in embryonic lethality in mice (Bernstein et al., 2003). In an effort to overcome this limitation, the cre-lox method has been used (Figure 1) to conditionally ablate dicer1 in different immune cell subsets. Studies employing this method have addressed the effect of Dicer and microRNA deficiency in the development and effector capabilities of the immune system. This review will highlight the key findings of these studies including results obtained in our laboratory in an effort to illuminate the potential roles of Dicer and microRNAs in immune cell development and function.

Figure 1.

Cre-lox technique for the conditional deletion of dicer1. In this method, a Cre mouse is generated (parent 1) wherein the gene that codes for the Cre recombinase enzyme is positioned downstream of a Transcription Factor Binding Site (TFBS). A TFBS is carefully chosen such that it recruits a transcription factor that is specifically expressed in a unique cellular subset or at a precise developmental stage. This allows the Cre recombinase to be expressed only when the transcription factor is available. A floxed mouse is also generated (parent 2) wherein the gene (or segement of a gene) targeted for deletion, which in this example is exon 23 of dicer1, is flanked by LoxP sites. This construct also contains a reporter gene, as indicated in the figure. A Cre LoxP mouse is obtained when the Cre mouse is crosses with the Floxed mouse. In a Cre LoxP mouse, the Cre recombinase, when expressed exclusively in cells where the specific transcription factor is available, drives homologous recombination between the loxP sites, leading to the deletion of exon 23 of dicer1 and allows the reporter gene to be expressed. Moreover, in cells that do not contain the specific transcription factor, the dicer1 gene will not be disrupted. TSS: Transcription Start Site, STOP: Stop codon.

Effect in Cells of the Myeloid Lineage

The combined efforts of multiple research groups have shown that microRNAs are important regulators of the early stages of myelopoiesis (El Gazzar et al., 2012). MicroRNAs perform their function by regulating the regulators i.e. by fine-tuning the expression of transcription factors and other developmental genes that play important roles in orchestrating the transition from immature progenitors to mature functional myeloid cells.

Myeloid progenitor cells

During myelopoiesis, hematopoietic stem cells differentiate into myeloid stem cells, which further differentiate into granulocyte-monocyte progenitors (GMPs). While granulocyte progenitors become terminally differentiated granulocytes, the myeloid progenitors further differentiate into circulating monocytes, tissue resident macrophages, myeloid DCs and neutrophils. For example, recent work (Velu et al., 2009; Popovic et al., 2009) has established that miR-196b, which is preferentially expressed in hematopoietic stem cells, targets Hox genes that positively regulate myeloid commitment and survival. In addition, the concerted effects of miRs 17-5p, 20a and 106a negatively regulate monocyte differentiation through their ability to target the M-CSF receptor (Fontana et al., 2007).

Extensive work on characterizing the roles of microRNAs in myeloid development has been performed and the overall biological relevance of Dicer in GMPs and in terminally differentiated cells originating from the myeloid lineage is beginning to be explored in detail (Buza-Vidas et al., 2012). In an effort to further address this issue, various studies employing the cre-lox method to conditionally delete dicer1 in myeloid cells have been conducted (Table 1). One recent study that investigated the effect of deleting the dicer1 gene in myeloid-committed progenitors analyzed their maturation into monocytes, macrophages and neutrophils in vivo (Alemdehy et al., 2012). The experimental model, based on C/EBPA driven deletion of dicer1, showed that although conditional deletion was accomplished, mice lacking dicer1 (dicer1^fl/fl mice) in the myeloid compartment survived for only short durations. This observation was attributed to the fact that C/EBPA is not entirely myeloid restricted and is also expressed in the lung epithelium. In order to circumvent this limitation, livers were isolated from dicer1^fl/fl mice at embryonic day 13 and transplanted into lethally irradiated syngeneic recipients. This study showed that differentiation of GMPs into monocytes, mature macrophages, neutrophils and myeloid DCs was severely impaired in dicer1^fl/fl mice. The occurrence of a developmental block was further confirmed in in vitro studies in which dicer1^fl/fl cells were observed to retain a progenitor-like phenotype and failed to differentiate in response to GM-CSF. The study also analyzed the impact of dicer1 insufficiency on the biology of myeloid cells located at central and peripheral tissue sites. Intriguingly, they found dysplastic neutrophils that were unable to migrate from the bone marrow and a marked decrease in the number of myeloid cells in the spleen and in macrophages present in the abdominal cavity. A functional study revealed that multiple genes involved in the regulation of hematopoiesis were de-repressed in dicer1^fl/fl myeloid progenitors. Computational analysis predicted these genes to be targets of 20 microRNA families and many of these microRNAs were predicted to regulate lineage-affiliated genes. Taken together, this study suggests that Dicer and microRNAs are key mediators involved in the maturation of myeloid progenitor cells into functional monocytes, macrophages, DCs and neutrophils. Importantly, this study also uncovered microRNAs that may be critical to the biology of myeloid cells.

Table 1.

A summary of factors driving Cre expression, their biological rationale and the phenotypic outcome in studies that required myeloid cell specific deletion of dicer1.

Factor driving Cre expression	Biological rationale	Result
C/EBPA	A transcription factor that is strongly expressed in granulocyte-monocyte progenitors and controls the expression of myeloid-specific genes	dicer1 conditionally deleted in granulocyte-monocyte progenitors. (Alemdehy et al., 2012)
Lyz2	A promoter/enhancer element that is restricted to monocytes, mature macrophages and granulocytes	Myeloid cell specific dicer1 deletion (Sissons et al., 2012)
CD11C	A protein expressed in myeloid dendritic cells and other terminally differentiated myeloid cells	dicer1 conditionally deleted in myeloid dendritic cells (Kuipers et al., 2010)

Macrophages

A study by Sissons et al. (2012) explored the role of Dicer and microRNAs in macrophage fusion into multinucleated giant cells. A major function of giant cells is to contain pathogenic invaders. Their formation is driven largely by type-2 cytokines including IL-4 and IL-13. (Brodbeck et al., 2009). M2 macrophages are also active players in wound healing and pathogen defense. Sissons et al., (2012) employed the Lyz2-cre approach to conditionally delete dicer1 in macrophages (Table 1). In dicer1^fl/fl macrophages, an examination of the proteins involved in giant cell formation revealed that Tm7sf4, an IL-4 induced transmembrane protein important in cellular fusion, was upregulated. The authors hypothesized that the upregulation in Tm7sf4 resulted from a loss of negative regulation by microRNAs. They identified miR-7a-1 as a negative regulator of Tm7sf4. Surprisingly, the expression of E-cadherin, also an IL-4 responsive gene, remained unperturbed in dicer1^fl/fl macrophages. Previous reports by other groups have shown that E-cadherin is targeted by microRNAs (Saydam et al., 2009; Ma et al., 2010). It is therefore interesting that only the IL-4 induced expression of Tm7sf4 is perturbed in response to dicer1 deletion. It is therefore plausible that certain microRNA targets, such as Tm7sf4, may have a greater dependency on microRNAs for their regulation compared to other targets such as E-cadherin.

The demonstration of microRNA regulation of giant cell formation in macrophages may lead to further research that focuses on the roles played by Dicer in regulating important disease related inflammatory responses such as the chronic inflammatory disease associated with lupus and rheumatoid arthritis (RA). MiR-146a has been implicated in RA; and elevated levels of miR-146a in synovial tissue have been strongly correlated with the disease score in RA (Abou-Zeid et al., 2011). The consistency of this observation has prompted the suggestion that miR-146a may be a biomarker for RA (Chan et al., 2009). We were however, unable to identify a study involving a quantitative analysis of dicer levels in RA synovial fluid. Given the chronic inflammatory nature of this disease, it is likely that the expression of Dicer and other components of the RNAi pathway may be perturbed in synovial tissues and in resident macrophages and T cells and that these perturbations may in turn influence microRNA biogenesis and RNAi in RA.

Dendritic cells

A study conducted by Kuipers et al. (2010) sought to explore the effect of deletion of dicer1 in dendritic cells (DCs) in vivo (Table 1). They observed only a moderate reduction in DC-specific microRNA expression in dicer1^fl/fl DCs. This observation suggests that either there may have been an inefficient knockdown of dicer1, or that DC-specific microRNAs may persist longer after dicer1 deletion. MicroRNAs have been shown to have varying turnover rates (Gantier et al., 2011). It is therefore plausible that DC-specific microRNAs may have altered turnover rates, thus persisting longer. This theory however, has not been tested thus far. The authors observed that although DC numbers in the lymph nodes and the spleen were unaffected by the dicer1 deletion, there was a progressive decrease in the number of Langerhans cells (LC) with age. Functional studies showed that dicer1 deletion resulted in an impaired capacity to upregulate CD40 and CD86 co-stimulatory molecules in response to a peptide antigen. Moreover, these cells were incapable of presenting MHC-II restricted antigenic peptides to CD4⁺ T cells, while the MHC-I restricted presentation of antigen to CD8⁺ T cells was unaffected. These findings suggest that the regulation of co-stimulatory molecules in LCs and perhaps in their DC precursors is under microRNA regulation. In addition, the differences observed in MCH-II mediated antigen presentation to CD4⁺ T cells suggest that the extrinsic pathway for antigen processing and presentation may have a greater dependency on microRNAs for its regulation. The authors found that dicer1^fl/fl LCs showed reduced levels of the proliferation factor c-myc and increased levels of the pro-apoptotic molecule bim (Kuipers et al., 2010). This pro-apoptotic phenotype may be explained in part by the observed reduction in TGF-β receptor subunit 2; a positive regulator of LC persistence in the epidermis.

Effect in Cells of the Lymphoid Lineage

Dicer loss-of-function studies have been conducted in B cell progenitors, mature B cells, CD8⁺ T cells, Th1, Th2 and Treg sub-lineages of CD4⁺ T cells. Currently, to our knowledge there is no Dicer loss-of-function study in Th17 cells in the literature. However, recently the roles of miRs 155 and Let-7a in regulating the function and differentiation of Th17 cells were described (Zhang et al., 2013; Yao et al., 2012). A more global perspective on Dicer and microRNAs in inflammation and in regulating the Th17 cell phenotype in health and disease could greatly contribute to our understanding of RNAi in conditions such as RA and psoriasis, where Th17 cells have been shown to play an important role (Tokura et al., 2010). This represents a potentially important area for future research.

T cells

Cobb et al. (2005, 2006) conducted detailed studies to assess the role of Dicer in the early stages of T cell differentiation and lineage commitment. They employed the Lck-cre method to conditionally delete dicer1 in thymocytes (Table 2) and observed that dicer1 deletion in thymocytes resulted in fewer αβ T cells while the numbers of γδ T cells were elevated. In addition, in vitro assays with thymocytes showed that overall, dicer1 deletion led to a higher degree of apoptotic cell death. This group further assessed the effect of dicer1 deletion on the development of Treg cells from T helper precursors in the thymus using the Lck-cre method (Table 2). They found that dicer1 deletion resulted in significantly fewer thymic Treg cells and moreover, dicer1^fl/fl Treg cells had reduced Foxp3: the major transcription factor involved in sustaining the Treg cell phenotype. Treg cells may mediate their anti-inflammatory effects through the secretion of soluble mediators such as IL-10 and TGF-β (Vignali et al., 2008; Schmidt et al., 2012). Dicer1 deletion resulted in reduced TGF-β secretion, suggesting that the dicer1^fl/fl Treg cells are functionally defective. A striking observation was that the pathology observed in mice with the dicer1 deletion mirrored that observed with the foxp3 gene deletion (Cobb et al., 2005).

Table 2.

A summary of factors driving Cre expression, their biological rationale and the phenotypic outcome in studies that required lymphoid specific deletion of dicer1.

Factor driving Cre expression	Biological rationale	Result
CD4	A glycoprotein expressed almost exclusively in T helper cells and functions as a co-receptor in establishing the TCR-MHC interaction	dicer1 conditionally deleted in CD4+ T cells (Muljo et al., 2005)
Lck	A thymocyte-specific protein tyrosine kinase. A marker for studying T cell maturation in the thymus	dicer1 conditionally deleted in immature T cells at the double negative (CD4– CD8– stage) (Cobb et al., 2005)
FoxP3	A transcription factor selectively expressed in regulatory T cells and functions in establishing and maintaining the regulatory T cell phenotype	dicer1 conditionally deleted in regulatory T cells (Liston et al, 2008; Zhou et al., 2008)
CD2	A protein expressed in all T cell committed progenitors	dicer1 deleted in T cell committed progenitors (Fedeli et al., 2009)
mb1	A gene encoding the Ig-α co-receptor of the B cell receptor. Mb1 expression begins at the pre-B cell stage	dicer1 conditionally deleted in B cells at the pre-B cell stage (Koralov et al., 2008)
CD19	A protein expressed in B cells, but not in plasma cells. Also expressed in follicular dendritic cells	dicer1 conditionally deleted in mature B cells (Belver et al., 2010; Xu et al., 2012)
AICDA	Activation-induced Citidine Deaminase (AICDA) is an enzyme expressed in B cells and is crucial for somatic hypermutation and class switching of immunoglobulin genes	The timing of AICDA expression drives dicer1 deletion in antigen stimulated B cells (Xu et al., 2012)

Foxp3 knockout mice lack Treg cells and generally do not survive beyond 3 weeks of age due to multi-organ failure; a phenotype resulting from the complete loss of peripheral T cell tolerance. The phenotypic similarities between Foxp3 deficiency and Dicer deficiency in vivo were further explored by Zhou et al. (2008) and Liston et al. (2008), by conditionally deleting dicer1 in Treg cells through the foxp3-cre approach (Table 2). They found that mice with dicer1^fl/fl Treg cells did not survive beyond 4 – 8 weeks of age. Furthermore, lymphocyte infiltration in multiple organs, reduced Treg derived cytokines, and high levels of serum IgE were also observed. These features are consistently seen in foxp3 knockout mice. A microarray experiment revealed that 393 genes were upregulated in dicer1^fl/fl Treg cells. However, only about one-third of these genes were predicted to be targeted by microRNAs present in Treg cells. Intriguingly, a notable overlap was observed between the sets of genes that were deregulated in dicer1^fl/fl Treg cells and those that were deregulated in foxp3 deficient Treg cells, suggesting a molecular interplay between the RNAi pathway and the gene expression program sustained by Foxp3.

Zhou et al. (2008) reported that dicer1 deletion resulted in the increased expression of activation markers CD25 and CD69, as well as the inhibitory ligand CTLA-4 in Treg cells. Contrary to this finding, Liston et al. reported a decrease in CTLA-4 levels in dicer1^fl/fl Treg cells. CTLA-4 is found at low levels on resting T cells and is upregulated following T cell activation (McCoy et al., 1999). It is established that CTLA-4 plays a crucial role in T cell tolerance and is therefore an important molecule for Treg cells to possess. Prior work by others (Fayyad-Kazan et al., 2012) has demonstrated that miR-145 negatively regulates CTLA-4 in healthy adult humans. It follows therefore that a complete loss of microRNA biogenesis would abolish miR-145, leading to an upregulation in CTLA-4. The contrasting results observed by the two aforementioned groups point to a more complex regulatory network governing the expression of CTLA-4 following T cell activation. The precise mechanism(s) through which Dicer is involved in regulating the activation threshold and/or the inhibitory functions of Treg cells remains to be established.

Major alterations in microRNA profiles have been found during the process of T cell differentiation, suggesting that microRNAs may be important in regulating T cell activation and differentiation (Jeker et al., 2013). To gain further insight into the role of Dicer in this process, Muljo et al. (2005) developed a model where dicer1 was conditionally deleted in Th cells using the CD4-cre approach (Table 2). In dicer1^fl/fl mice, a reduction in CD4⁺ T cells in the spleen, lymph nodes and blood was observed, suggesting that Dicer is required for the maintenance of CD4⁺ T cells in the periphery. The homeostatic maintenance of peripheral T cells is dependent on growth promoting cytokines such as IL-7 and IL-15. It is plausible that dicer1 deletion makes peripheral T cells less sensitive to these cytokines, resulting in a reduction in their numbers. Alternatively, dicer1 deletion could have negatively affected both CD4 and CD8 lineages early in T cell development. The study also found dicer1^fl/fl CD4⁺ T cells to have impaired proliferative capacity in response to both Th1 and Th2 skewing conditions. Therefore, Dicer1 deletion did not appear to have an effect on the levels of T-bet and GATA-3; transcription factors that sustain the Th1 and Th2 phenotypes respectively. Intriguingly, the study found that dicer1^fl/fl CD4⁺ T cells produced excess IFN-γ, a Th1 enhancing factor and as a result, these cells have a propensity to attain the Th1 phenotype. There is also recent evidence in the literature suggesting that both IFN-γ expression and signaling are subject to microRNA regulation (Ram Savan et al., 2012, Ma et al., 2011, Banerjee et al. 2010) and that miR-29 plays a dual role in regulating IFN-γ expression. For example, miR-29 stabilizes IFN-γ mRNA and promotes its translation (Ram Savan et al., 2012, unpublished); it also targets and degrades IFN-γ mRNA during bacterial infections (Ma et al., 2011). Furthermore, Banerjee et al. (2010) have demonstrated that miR-155 negatively regulates the alpha chain of the IFN-γ receptor.

CD8⁺ T cells are indispensable in the establishment of long-term immunity to viral and bacterial insults. Zhang et al. (2011) explored the role of Dicer during the early activation and proliferation phases of CD8⁺ T cells by employing the Lck-cre approach (Table 2). They observed that dicer1^fl/fl CD8⁺ T cells proliferated poorly compared to their wild-type counterpart, suggesting that antigen-restricted T cell proliferation was affected in these cells. When an exogenous antigen is encountered, CD8⁺ T cells and antigen presenting cells are momentarily confined to the lymph nodes in order to maximize local antigen presentation (Pellegrini et al., 2003; Zhang et al., 2011). At the end of this phase, activated CD8⁺ T cells migrate into the blood and peripheral sites where they perform important effector functions. Dicer1^fl/fl CD8⁺ T cells are impaired in their ability to reach peripheral sites, suggesting that Dicer is a key factor in regulating the trafficking of CD8⁺ T cells to peripheral sites. Unexpectedly, in vitro studies demonstrated that dicer1^fl/fl CD8⁺ T cells proliferated normally in response to growth promoting cytokines (Zhang et al, 2011). This proliferation was positively correlated with the upregulation of CD69, a proliferation marker. The authors provide evidence showing that members of the miR-130 and miR-301 families target CD69 and that in dicer1^fl/fl CD8⁺ T cells, a loss of microRNA mediated repression led to the observed increase in CD69; an observation suggesting that T cell activation is Dicer dependent.

Invariant NKT (iNKT) cells are present both in the thymus and periphery and recognize antigens presented by the non-canonical MHC class I molecule CD1d (Bendelac et al., 2007). Fedeli and colleagues characterized the effects of dicer1 deletion on iNKT cell development using the CD4-cre approach (Table 2) (Fedeli et al., 2009). Significant reductions in the number of thymic, hepatic and splenic iNKT cells were observed. This was thought to be caused by a developmental block resulting from defective mitosis in Dicer1^fl/fl iNKT cells. A greater degree of apoptotic cell death in dicer1^fl/fl iNKT cells was also observed. Furthermore, deleting dicer1 using an alternative CD2-cre approach (Table 2) resulted in the complete loss of thymic and peripheral iNKT cells without severely affecting the total numbers of peripheral T cells. These data suggest a requirement for Dicer in the early development of the iNKT cell subset.

B cells

The role of Dicer in controlling differentiation and immunoglobulin gene regulation in the B cell lineage was assessed by Koralov et al. (2008) by deleting dicer1 in pro-B cells using the mb1-cre approach (Table 2). This altered the development of pro-B cells and prevented their transition to the pre-B cell stage. Microarray experiments revealed that in dicer1^fl/fl B cells, there was an elevated level of bim, a proapoptotic gene and a validated target of the miR 17~92 family, suggesting that in the B cell lineage, dicer1 deletion tips the balance toward apoptosis. These observations indicate that in B cells, disrupting Dicer function leads to the induction of apoptosis through a microRNA-dependent mechanism.

Belver et al. (2010) and Xu et al. (2012) studied the biological relevance of Dicer and microRNAs in B cell differentiation, and the establishment of tolerance to self-antigens via a CD19-cre approach (Table 2). Their studies showed that although no major developmental blocks were observed during the early stages of B cell differentiation (pro-B, pre-B and IgM⁺ B cell maturation); there was a decrease in circulating B cell numbers. Moreover, the peripheral B cell population was skewed toward the marginal zone B cell phenotype, an effect that was shown to be regulated though miR-185. Furthermore, B cells were partially deregulated, as evidenced by the increased usage of J segments, higher frequencies of arginine and lysine residues in the CDR3 region of the immunoglobulin heavy chain and the presence of self-reactive IgG1 antibodies.

The consequences of dicer1 deletion on antibody production and class switching were further explored by Xu et al. (2012), using the Aicda-cre approach (Table 2). They found that in naive dicer1^fl/fl B cells, there were physiologically normal levels of IgM but lower levels of IgG and IgA. Upon antigenic stimulation however, dicer1^fl/fl B cells failed to produce high affinity class switched antibodies that were specific to the antigen. Furthermore, dicer1^fl/fl B cells were unable to achieve a memory B cell phenotype or differentiate into plasma cells. Collectively, these observations suggest that in the absence of Dicer, B cells retain a naive-like state even after encountering antigen. In addition, Aicda-cre driven dicer1 deletion resulted in poor proliferative potential, higher levels of bim and increased apoptosis; observations consistent with those previously made by Koralov et al. (2008). Xu et al. (2012) also observed that in dicer1^fl/fl B cells, there was in increase in cell cycle inhibitor genes. The authors posit that this increase in cell cycle inhibitor genes may be causing the increase in apoptosis since many cell cycle inhibitor genes are targeted by microRNAs.

CONCLUSIONS AND FUTURE DIRECTIONS

The investigations discussed here have generated valuable data on the critical role of Dicer and microRNAs in immunity. Through the use of conditional knockout models, these studies have shown that in the absence of dicer1, immune cell development and function are severely compromised.

The establishment of either a pre-apoptotic state or the full-fledged induction of apoptosis is a common feature observed in many of the dicer1 conditional knockout studies discussed here, suggesting that Dicer is an important player in the regulation of cell survival. Previous work has shown that genes associated with cell survival, cell cycle regulation and apoptosis are regulated by microRNAs, among other mechanisms (Hermeking et al., 2007; Lynam-Lennon et al., 2009). It is therefore plausible that the manipulation of Dicer levels through drugs, antibodies or other techniques may impact the development and survival of immune cells. These observations are relevant to treatment protocols in cancer, where the survival of cancer cells is detrimental and in neurodegenerative diseases, where the survival of normal host cells may be beneficial. The specific loss of Dicer by natural means such as through aging has previously been reported and has been shown to cause hypersensitivity to oxidative stress in a murine model (Mori et al., 2012). In addition, the loss of dicer has also been reported to occur in human macular degeneration (Kaneko et al., 2011). There is tremendous potential in exploring further the effects of manipulating dicer levels through pharmacological means.

Emerging evidence suggests that Dicer and microRNAs are deregulated in cancer (Karube et al., 2005; Kobel et al., 2009; Chiosea et al. 2006; Ma et al., 2011). These studies have focused on obtaining correlative data between cancer progression and Dicer levels. Some studies have reported a decrease in Dicer levels with cancer progression (Karube et al., 2005), while others have shown an increase (Chiosea et al., 2006). It is still unclear whether dicer1 is an oncogene, a tumor suppressor, or perhaps both depending on the cell type and its environment. The conclusions made by these studies are based largely on data obtained from patient tumor samples. We have not found studies that have specifically analyzed the levels of Dicer in the immune cells that infiltrate solid tumors. Clearly, further work is necessary to gain a better understanding of the role of Dicer and microRNAs at the immune-tumor interface and how this may influence the balance between tumor-promoting and anti-tumor mechanisms. Additionally, there are currently no clinical therapies to our knowledge which specifically target the levels of Dicer, although experimental studies have shown the regulation of dicer by microRNAs (Asirvatham et al., 2008) and chemical inhibitors (Shi et al., 2013).

Work done by our lab suggests that Dicer levels vary considerably in response to environmental factors such as cellular stresses and interferons (Wiesen et al., 2009, Oshlag et al., 2013). We are currently exploring how heat, cold and circadian rhythms may regulate Dicer and potentially microRNAs. An intriguing question at the crossroads of RNAi and immunity is: Do variations in Dicer levels and/or activity caused by environmental factors influence cellular behavior especially in the context of an antigenic challenge? Currently, very little is known of the effects of stresses on the activity of Dicer.

It has been known through scientific inquiry and anecdotal evidence that stressful events such as work-related anxiety, depression, posttraumatic stress disorder and sleep deprivation can influence the biology of immune cells (Atanackovic et al., 2004; Altemus et al., 2006; Moldofsky et al., 1989). Many of these stress-induced effects have been attributed to the hypothalamus-pituitary-adrenal (HPA) axis. The HPA axis can induce the secretion of glucocorticoids and other hormonal mediators from the adrenal cortex in response to the aforementioned stresses. Glucocorticoids have potent anti-inflammatory effects and have recently been shown to alter the circadian rhythm in immune cells (Son et al., 2011). Based on work described in this review, we now know that altering Dicer and microRNA levels in immune cells can profoundly influences their ability to survive, differentiate, become activated and perform effector functions. It is also plausible that the stress-induced secretion of glucocorticoids could be altering immune cell biology in a Dicer-dependent manner. This thesis may help develop an RNAi-based mechanistic explanation of why immune cells show altered phenotypes in stressed individuals. Furthermore, exploring the potential mechanisms could lead to further understanding of how Dicer and microRNAs regulate immune cell functions in (a) bacterial and viral infection, (b) immune cells in the tumor microenvironment and (c) inflammatory sites in individuals suffering from chronic inflammatory conditions and autoimmune illnesses.

In vivo, immune cells are exposed to a varied spectrum of cytokines, pro-inflammatory and anti-inflammatory mediators. It is likely that the cell signaling pathways invoked by such factors would impinge on the RNAi pathway and potentially influence cellular behavior in a Dicer/microRNA dependent manner. This hypothesis is only beginning to be explored and likely represents a fertile area for future research.