Signaling to Translational Control Pathways: Diversity in Gene Regulation in Inflammatory and Vascular Cells
Article Outline
- Abstract
- An Overview of Signal-Dependent Translational Control
- Signal-Dependent Translation in Myeloid Leukocytes: New Inflammatory Pathways, Diverse Regulatory Mechanisms
- Diversity in Signal-Dependent Translation: Molecular Mechanisms and Responding Cells
- Therapeutic Implications
- Acknowledgments
- References
- Copyright
The expression of a subset of genes is strongly controlled at translational checkpoints, a major mechanism of posttranscriptional regulation. Inflammatory and vascular cells receive outside-in signals to specialized pathways that regulate translation of specific messenger RNAs in a transcript-specific fashion and thereby influence key features of cellular phenotype. These pathways and the expression of proteins that they control may be dysregulated in cardiovascular diseases and are therapeutic targets.
Intercellular signaling leading to altered expression of genes in vascular and inflammatory cells is a central feature in current concepts of the pathophysiology of cardiovascular diseases, including atherosclerosis, myocarditis and other conditions of myocardial injury, cardiac allograft rejection, and additional syndromes. In atherosclerosis alone, altered gene expression in key cell types is central to current concepts of atherosclerotic plaque initiation and progression, plaque rupture and its acute and subacute complications, and responses to iatrogenic interventions (Glass and Witztum 2001). Pathways that regulate gene expression in physiologic inflammation, hemostasis, and vasculogenesis become dysregulated in atherosclerosis and other cardiovascular syndromes. Characterization of these pathways and key checkpoints that regulate them will yield insights into fundamental events in these disorders and mechanisms of action of current therapies, and is likely to identify new therapeutic targets.
For many investigators and physicians, transcriptional regulation is synonymous with altered gene expression. Indeed, control of levels of specific messenger RNAs (mRNAs) via new transcription and altered abundance, distribution, and activity of transcription factors is central to vascular and inflammatory responses. Examples of transcriptional regulators specifically relevant to atherosclerosis and vascular syndromes include the nuclear factor-kappa β family of nuclear regulatory factors, peroxisome proliferator-activated receptors, and other members of the nuclear receptor superfamily (Glass and Witztum, 2001, Thurberg and Collins, 1998). Analysis of altered patterns of transcript expression based on DNA chip technology is a topical genomic strategy utilized in efforts to discover new molecular therapies for atherosclerosis (Glass and Witztum 2001). Nevertheless, posttranscriptional pathways are also major sites of control of expression of protein gene products (Brewer, 2001, Calkhoven et al., 2002, Kozak, 1991). This is clear when differences in levels of mRNA transcripts and the proteins they code for, identified by unbiased genomic and proteomic analysis, are considered (Pradet-Balade et al., 2001, Weyrich et al., 2003). Posttranscriptional control offers biologic advantages—including precision in spatial and temporal accumulation of factors with potent biologic activities—and operates in parallel and in series with transcriptional regulation in cell- and transcript-specific fashions (Calkhoven et al., 2002, Huang and Richter, 2004, Kozak, 1991, Matthews et al., 2000). Interacting transcriptional and posttranscriptional pathways control complex biologic responses including circadian oscillations in the vasculature, synaptic plasticity, chemokine synthesis, and other events (Huang and Richter, 2004, McNamara et al., 2001, Yost et al., 2004). Recognition of the diversity of posttranscriptional control is rapidly expanding in the “postgenomic” era (Brewer 2001). Posttranscriptional pathways are, however, largely uncharacterized in inflammatory and vascular cells.
This brief review focuses on cells of the innate immune system, because of their ubiquitous participation in atherosclerosis and other inflammatory diseases (Glass and Witztum 2001) and because they are accessible and informative primary human cells for the study of gene expression in physiologic and pathologic conditions (Figure 1) (Weyrich et al., 1995, Weyrich et al., 1996, and cited references). Recent observations clearly indicate that posttranscriptional regulation also determines key phenotypic features of vascular cells of several types (Kraiss et al., 2000, Maeshima et al., 2002, Weyrich et al., 2003), although space prevents a detailed discussion. Rather than a comprehensive review of posttranscriptional mechanisms, this article also focuses on translational regulation in response to signals from the environment and uses examples in which the gene product has clinical and pathophysiologic relevance. In addition, these examples indicate that understanding mechanisms of translational regulation has practical significance for molecular diagnosis and specific therapeutic interventions in vascular and other diseases.
Figure 1.
Outside-in signals regulate gene expression at transcriptional and posttranscriptional checkpoints in human myeloid leukocytes. Human monocytes and polymorphonuclear leukocytes (PMNs) (not shown) are accessible differentiated primary human cells that provide general models of gene regulation in addition to being specialized inflammatory cells that respond to outside-in signals from the environment. Inflammatory signals are delivered by soluble agonists, juxtacrine agonists that bind to monocyte receptors while anchored on the plasma membranes of interacting cells (platelets, endothelial cells, other leukocytes), and engagement of adhesion molecules such as P-selectin glycoprotein ligand 1 (PSGL-1) on the monocyte surface (McIntyre et al. 2003). Expression of specific genes is regulated at transcriptional and posttranscriptional checkpoints. Synthesis of some proteins with potent biologic activities is under specialized signal-dependent translation control. PSGL-1, certain other adhesion molecules, and some receptors can transmit signals to specialized translational control pathways that act in parallel or in series with transcriptional mechanisms. Transcriptional and posttranscriptional checkpoints and pathways can be examined by a variety of assays that detect specific mRNAs, their distributions and translational efficiencies, and accumulation of their protein products. Some approaches, such as polysome analysis, provide both genomic and proteomic information.
An Overview of Signal-Dependent Translational Control
It is estimated that 5% to 20% of eukaryotic genes are strongly controlled by translational mechanisms, depending on the cell type and specific conditions for their expression; the protein products largely and, perhaps exclusively, have critical biologic activities rather than housekeeping functions (Gingras et al., 2001, Kozak, 1991, Pradet-Balade et al., 2001). Translational regulation often operates in concert with transcriptional control and/or with other posttranscriptional mechanisms for precise and tightly orchestrated synthesis of highly regulated proteins (reviewed in Kozak, 1991, Matthews et al., 2000). Translational control of certain genes is accomplished by specialized mechanisms that are responsive to signals from the environment. Rapid protein synthesis without requirement for new transcription is one of several biologic advantages of “signal-dependent” translation (Weyrich et al. 1998) of mRNA transcripts that are stored or silenced in the basal state (Gingras et al. 2001).
The initiation phase of translation, in which mRNAs associate with ribosomes and form polyribosome (polysome) translational units (Table 1), is rate limiting in most cases and is subject to tight regulation (Calkhoven et al., 2002, Gingras et al., 2001). Analysis of mRNAs preferentially associated with polysomes and those shifting to polysomes from ribonucleoparticles and other sequestered sites generally identifies actively translated transcripts and those with facilitated translation, and can be accomplished using interrogation of arrayed DNA libraries (Gingras et al., 2001, Lindemann et al., 2001, Pradet-Balade et al., 2001). Binding of eukaryotic initiation factor 4E (eIF4E) to the 5′ methyl cap of nuclear-transcribed mRNAs is a critical event in translation initiation. Phosphorylation via a signal-dependent kinase pathway together with other mechanisms regulates eIF4E binding to capped transcripts and translational efficiency (Gingras et al. 2001). eIF4E then facilitates organization of a heteromeric translation initiation complex (eIF4F) that includes a scaffolding protein (eIF4G) and an RNA helicase (eIF4A), which “unwinds” inhibitory secondary structure in the 5′ untranslated region (5′ UTR) of the transcript (Gingras et al., 2001, Huang and Richter, 2004) (Table 1). Complex secondary structure of the transcript 5′ UTR that requires helicase activity of the eIF4F complex and energy input to linearize is a major discriminating feature of mRNAs subject to specialized translational control, whereas mRNAs with short 5′ UTRs and/or those with simple structure are more readily translated (Gingras et al. 2001). Translation of specific mRNAs is also regulated by features of the 5′ UTR other than secondary structure including terminal oligopyrimidine tracts (5′ TOPs), upstream open reading frames, internal ribosome entry sites, and binding sites for transacting factors (Brewer, 2001, Calkhoven et al., 2002).
Table 1. Initiation of cap-dependent mRNA translation
| • Binding of eukaryotic initiation factor 4E (eIF4E) to the 5′ methyl cap (m7GpppN) of the mRNA transcript. |
| • eIF4E organizes a heteromeric complex, eIF4F, consisting of eIF4E, eIF4G (a scaffolding protein), eIF4A (an RNA helicase), and eIF4B (an activator of eIF4A). |
| • Binding of the 43S preinitiation complex (40S ribosomal subunit, eIF3, eIF2-GTP-Met-tRNAi) to eIF4F, forming the 48S complex and anchoring it to the 5′ untranslated region (5′ UTR) of the mRNA transcript. |
| • Unwinding of the 5′ UTR secondary structure and scanning of the 5′ UTR sequence by the 48S complex. |
| • Recognition of an AUG triplet (usually the first one encountered), release of initiation factors, and joining of a 60S ribosome subunit to the 40S ribosomal subunit to form translationally competent 80S ribosomes. |
The 3′ UTR of the mRNA transcript is also a site of translational regulation and contains sequences that bind cytoplasmic transacting proteins. These 3′ UTR motifs include AU-rich elements (AREs) and cytoplasmic polyadenylation sites (Brewer, 2001, Huang and Richter, 2004). Previously unrecognized 3′ UTR sites for transacting proteins that can interact with 5′ UTR regulatory elements and mediate translational control in a signal- and transcript-specific fashion have recently been reported (Mazumder et al., 2003a, Mazumder et al., 2003b, and cited references). In addition, microRNAs—an abundant class of short noncoding RNAs that appear to control translation of target transcripts by binding to sites in their 3′ UTRs—have recently been identified in vertebrates as well as invertebrates (reviewed in Bartel 2004), although their influence on translation in differentiated mammalian cells is not clear. Additional mechanisms yet to be characterized also regulate translation at steps beyond initiation (Huang and Richter, 2004, Lindemann et al., 2001). An emerging concept is that diversity of regulatory pathways and checkpoints, including initiation and postinitiation mechanisms, is a conserved feature of translational control and that dysregulation of posttranscriptional regulation is a mechanism of disease (Brewer, 2001, Calkhoven et al., 2002, Gingras et al., 2001, Huang and Richter, 2004, Macdonald, 2001).
Specialized signal-responsive pathways control translation of specific transcripts by regulating key translational mechanisms (Weyrich et al. 1998). One of these pathways is centered on mammalian target of rapamycin (mTOR), an evolutionarily conserved serine/threonine kinase that may be a master switch in gene regulation in response to signals from the environment (Calkhoven et al., 2002, Gingras et al., 2001, Raught et al., 2001). It is best known for its influences on cell-cycle progression and cell growth in response to mitogens and growth stimuli and was identified using the molecular probe and inhibitor rapamycin, an immunosuppressant and antineoplastic macrolide (Calkhoven et al., 2002, Gingras et al., 2001, Raught et al., 2001). mTOR receives upstream signals from the plasma membrane by incompletely characterized transduction steps that include protein kinase B (PKB/AKT) and phospholinositide 3 kinase (PI3K), depending on the cell system examined, and then controls the phospholylation state and activity of two key downstream effectors, ribosomal S6 kinase (S6K1) and eIF4E-binding protein 1 (4E-BP1) and related family members (Gingras et al., 2001, Schalm et al., 2003). Signaling to these phosphoproteins via mTOR regulates activity of eIF4E and initiation of translation of a subset of highly regulated mRNAs with complex 5′ UTR secondary structures and other features (Table 1) (Gingras et al. 2001). Rapamycin (also called sirolimus) specifically blocks mTOR activity and signaling to S6K1, 4E-BP1, and eIF4E by this pathway (Gingras et al., 2001, Raught et al., 2001). Rapamycin is currently in use for prevention of cardiac allograft rejection and in drug-eluting stents for the prevention of coronary restenosis, where its cellular targets remain to be completely defined (Marks 2003).
Signal-Dependent Translation in Myeloid Leukocytes: New Inflammatory Pathways, Diverse Regulatory Mechanisms
Monocytes and polymorphonuclear leukocytes (PMNs) are key innate immune effector cells of the myeloid series with pleiotropic activities in regulated and dysregulated inflammation. Monocytes have ubiquitous roles at each stage of the atherosclerotic progression from fatty streak to plaque rupture, and monocytes and neutrophils are each activated and mediate inflammatory and thrombotic events at sites of plaque disruption (Figure 2) (Buffon et al., 2002, Freedman and Loscalzo, 2002, Glass and Witztum, 2001). Although monocytes have long been known to express new gene products in response to inflammatory stimuli, PMNs were not thought to synthesize new proteins after leaving the marrow, but it is now recognized that they produce chemokines and other inflammatory proteins when appropriately stimulated (Lindemann et al., 2004, Yost et al., 2004 and cited references). Whereas stimulation of myeloid leukocytes with soluble inflammatory mediators induces altered gene expression, it is also triggered by outside-in signals delivered by engagement of surface adhesion molecules including integrins and the selectin-recognition molecule P-selectin glycoprotein ligand 1 (PSGL-1) (Figure 1, Figure 2) (Galt et al., 2001, Mahoney et al., 2001, Weyrich et al., 1995, Weyrich et al., 1996). Pathophysiologically relevant models for studying these pathways include in vitro interaction of human myeloid leukocytes with activated platelets (Figure 2) (Galt et al., 2001, Mahoney et al., 2001, Weyrich et al., 1996) or endothelial cells (McIntyre et al. 2003), which mimic cell–cell interactions in atherosclerotic progression and acute coronary syndromes (Freedman and Loscalzo, 2002, McIntyre et al., 2003).
Figure 2.
Human cell models yield information relevant to gene regulation in atherosclerosis and its complications. (A) Stable aggregates resulting from adhesion of activated platelets to monocytes and polymorphonuclear leukocytes (PMNs) form in the blood of patients with atherosclerosis and in animal models. Activation of human platelets with thrombin, platelet activating factor (PAF), and other agonists in vitro induces display of P selectin on the platelet surface and engagement of P-selectin glycoprotein ligand 1 (PSGL-1) on monocytes and PMNs, mediating cell–cell adhesion and providing a model for in vivo events. Under these conditions, outside-in signals are delivered via PSGL-1. In addition, outside-in signals are transmitted by surface receptors on monocytes that recognize platelet signaling molecules (the chemokine RANTES, PAF, and others). Altered expression of transcriptionally and posttranscriptionally regulated genes is one functional consequence. (B) An aggregate composed of isolated human monocytes (large arrow) and adherent thrombin-stimulated platelets (small arrows) is shown. Immunostaining was accomplished with an antibody that detects a ribosomal marker, illustrating the translational “machinery” of these cells. (C) Clinical observations and studies in murine models indicate that interactions of platelets with myeloid leukocytes contribute to atherosclerotic progression (McIntyre et al., 2003, Weyrich et al., 2003), in addition to coronary thrombosis and other local events at the time of plaque rupture (Freedman and Loscalzo 2002). (D) Outside-in signals delivered by platelets to human monocytes adherent to collagen induce expression of matrix metalloproteinase 9 (MMP-9) (green fluorescent staining). Engagement of PSGL-1 and integrins regulates synthesis of MMP-9 at transcriptional and posttranscriptional checkpoints in this plaque rupture model (Galt et al. 2001). (E) A monocyte surrounded by platelets in a ruptured plaque removed from a patient at the time of carotid endarterectomy stains positive for MMP-9 (brown reaction product), illustrating relevance of gene regulatory pathways identified in vitro (panel D). Panels C, D, and E were originally published in Galt et al. 2001 and Weyrich et al. 2003, and are reproduced by permission.
Using human cell models, we have characterized signal-dependent pathways for translational control in myeloid leukocytes in addition to transcriptional mechanisms. While examining expression of mRNAs that are induced or amplified when PSGL-1 is engaged on primary human monocytes, we identified a group of transcripts that are candidates for translational control by analysis of the sequences of their 5′ UTRs. One of these codes for urokinase plasminogen activator receptor (UPAR), a key regulator of surface protease activity, integrin activity, adhesion, and migration (Mahoney et al. 2001 and cited references). The mRNA coding for UPAR is constitutively present in freshly isolated monocytes, but there is little or no protein (Figure 3) (Mahoney et al. 2001), classic evidence for posttranscriptional regulation and translational control (Kozak, 1991, Matthews et al., 2000, Pradet-Balade et al., 2001). Adhesion of monocytes to activated platelets or engagement of PSGL-1 by purified P selectin rapidly induced synthesis of UPAR that was not inhibited by blocking transcription, indicating signal-dependent translation of the constitutive mRNA. In parallel, we explored pathways that might regulate its expression and found that mTOR, S6K1, and 4E-BP1 are present and activated in response to outside-in signals in human monocytes (Figure 3) (Mahoney et al. 2001). Rapid synthesis of UPAR was specifically inhibited by blocking mTOR with rapamycin and by PI3K inhibitors, as was UPAR-dependent adhesion of monocytes to immobilized vitronectin (Mahoney et al. 2001). In contrast, synthesis of the chemokine interleukin 8 (IL-8), which is transcriptionally regulated under these conditions, was not (Mahoney et al. 2001). This comparison was chosen because of differences in the 5′ UTRs of their transcripts that predict specialized translational regulation for UPAR but not for IL-8 (see above) (Figure 4).
Figure 3.
The mTOR pathway regulates signal-dependent translation of constitutive mRNAs in human monocytes. (A) Freshly-isolated human monocytes contain urokinase plasminogen activator receptor (UPAR) mRNA in the basal state but little or no UPAR protein, indicating translational silencing. UPAR protein is rapidly synthesized when monocytes are signaled by thrombin-stimulated platelets (see Figure 2) and also by specific engagement of P-selectin glycoprotein ligand 1 (PSGL-1). Rapid synthesis of UPAR is blocked by translational inhibitors but not transcriptional blockade, indicating signal-dependent translation of constitutive mRNA (Mahoney et al. 2001). (B) Mammalian target of rapamycin (mTOR) is present in human monocytes. Freshly isolated human monocytes were incubated in the absence of an agonist (rounded cells at left) or in the presence of the chemokine RANTES (elongate polarized cell at right) and stained with an antibody that identifies mTOR, which revealed a diffuse granular cytoplasmic pattern. (C) The mTOR pathway is activated by signals from receptors and PSGL-1 on human monocytes. This previously unrecognized inflammatory signaling pathway regulates synthesis of UPAR and is inhibited by rapamycin (sirolimus). See also Figure 4 and Mahoney et al. 2001.
Figure 4.
Complex 5′ untranslated region (5′ UTR) secondary structure regulates expression of urokinase plasminogen activator receptor (UPAR) in myeloid leukocytes. (A) The 5′ UTR of UPAR (426 nucleotides) is predicted to have extensive secondary structure requiring significant energy to linearize (ΔG) whereas the interleukin 8 (IL-8) 5′ UTR (101 nt) is simple, requiring little free energy to unwind. Rapid synthesis of UPAR protein by monocytes in response to P-selectin glycoprotein ligand 1 (PSGL-1) engagement is translationally regulated by mammalian target of rapamycin (mTOR) and is specifically inhibited by rapamycin and wortmannin, whereas synthesis of IL-8 is transcriptionally regulated under these conditions. (B) Overexpression of eukaryotic initiation factor 4E (eIF4E) in transfected U937 myeloid cells resulted in a dramatic increase in synthesis of UPAR protein compared with that in control cells treated with an empty vector, consistent with complicated 5′ UTR secondary structure of the UPAR mRNA (see A) and translational regulation. In contrast, synthesis of actin and IL-8 (not shown) were not altered. Panel B is reproduced from Mahoney et al. 2001.
Overexpression of eIF4E facilitates translation of some, but not all, mRNAs with inhibitory secondary structure in the 5′ UTRs in part dependent on cell-specific factors, and is a test for translational regulation by this mechanism (Matthews et al. 2000). UPAR protein was dramatically increased in U937 myelomonocytic leukocytes stably transfected with eIF4E, whereas IL-8 was not, consistent with the differences in the secondary structures of their 5′ UTRs and with differential inhibition by rapamycin in primary monocytes (Figure 4) (Mahoney et al. 2001). We also found that engagement of PSGL-1 on primary monocytes induces phosphorylation of eIF4E (Mahoney et al. 2001), demonstrating an additional pathway to this checkpoint (Gingras et al. 2001). Together, these studies demonstrated differential control of expression of two key monocyte gene products—UPAR and IL-8—in response to PSGL-1 engagement, and provided the first evidence for signal-dependent translation via mTOR in human myeloid cells. Additional studies also indicate gene-specific regulation in these cells. For example, specific outside-in signals in parallel to those delivered by engagement of PSGL-1 are required to induce or amplify synthesis of MMP-9 (Figure 2) (Galt et al. 2001) and cyclo-oxygenase 2 (DA Dixon, SM Phiscott, ND Duey, AS Weyrich, GA Zimmerman, unpublished data) in human monocytes.
As a second example, human PMNs also carry silenced mRNAs in the basal state and rapidly translate them in response to activating signals. Microarray analysis detected multiple transcripts in freshly isolated unstimulated PMNs, and radiolabeling with 2-dimensional (2D) gel electrophoresis demonstrated that multiple protein products are synthesized in response to cellular activation (Lindemann et al. 2004). Platelet activating factor (PAF), a lipid agonist implicated in atherogenesis and acute coronary syndromes (McIntyre et al. 2003), is a potent stimulus for this response but it is also induced by lipopolysaccharide and tumor necrosis factor α indicating that several classes of surface receptors are linked to the relevant intracellular signal transduction cascades. Studies in a model in which HL60 myeloid leukocytic cells were differentiated to surrogate PMNs demonstrated that activation of these cells triggers association of multiple mRNAs with polyribosomes, also consistent with receptor-mediated signaling to translation control pathways (Yost et al. 2004). Furthermore, in primary human PMNs, S6K1 and 4E-BP1 are activated by PAF and other agonists and by adhesion of the leukocytes to P selectin. These responses are specifically inhibited by rapamycin (Lindemann et al. 2004). Treatment of PMNs with rapamycin also selectively inhibited synthesis of a subset of agonist-induced proteins detected by 2D gel electrophoresis (Lindemann et al. 2004). This provides evidence for signal-dependent activation of mTOR and its control of translation of a subpopulation of transcripts in this terminally differentiated nonproliferating cell.
To further establish physiologic relevance of signal-dependent activation for mTOR in PMNs we have identified specific gene products that are under selective translational control by this pathway. One of the newly synthesized proteins in PAF-stimulated PMNs had mobility characteristics on 2D gel electrophoresis suggesting that it is the α subunit of the IL-6 receptor (IL-6Rα), which we chose for validation because of its pathophysiologic significance. IL-6Rα is released from activated human PMNs and can then mediate “trans-signaling” of endothelial cells (Lindemann et al. 2004 and cited references). Trans-signaling also induces vascular smooth muscle activation (Klouche et al. 1999) and therefore could be operative in restenosis (Welt and Rogers 2002) and other vascular syndromes in which IL-6 plays a role. In addition, there is evidence that trans-signaling resulting from the release of IL-6Rα by activated PMNs is a mechanism that mediates the “switch” from the acute to the mononuclear phase of inflammation (reviewed in Kaplanski et al. 2003), and thus may be involved in the progression of atherosclerotic lesions when leukocytes are activated in the coronary or systemic circulations (Buffon et al. 2002).
Freshly isolated human PMNs have the IL-6Rα mRNA, but little or no protein is present in the basal state, analogous to the situation with UPAR in monocytes (see above) and indicating translational control) (Figure 5) (Lindemann et al. 2004). In response to signaling by PAF or other agonists, IL-6Rα protein is rapidly synthesized and released, a response blocked by translational but not transcriptional inhibitors and specifically inhibited by rapamycin (Figure 5) (Lindemann et al. 2004). Analysis of the 5′ UTR of the IL-6Rα transcript in PMNs by rapid amplification of cDNA ends reveals extensive secondary structure that represses expression of a reporter construct in transfected CHO and THP-1 cells (Lindemann et al. 2004, MM Denis, AS Weyrich, GA Zimmerman unpublished data). These results demonstrate a mechanism involving repressed translation of constitutive IL-6Rα mRNA by 5′ UTR secondary structure in the basal state and signal-dependent translation of the protein regulated by mTOR in response to cellular activation. Parallel studies demonstrate that retinoic acid receptor α is also synthesized in activated PMNs by a mechanism that involves translational silencing of constitutive mRNA in the resting state and mTOR-dependent translation in response to agonist stimulation, and provide evidence for novel translational control of a transcriptional regulator under these conditions (Yost et al. 2004).
Figure 5.
Interleukin 6 receptor α (IL-6Rα) is regulated by signal-dependent translation in human polymorphonuclear leukocytes (PMNs). (A) Trans-signaling of human endothelial cells by IL-6Rα released by activated PMNs. Activation of PMNs triggers release of soluble IL-6Rα (step 1), which associates with constitutive GP130 transmembrane homodimers on the surfaces of endothelial cells forming a heterotrimeric signal-transducing receptor. This receptor recognizes IL-6 endogenously produced by endothelial cells or released by other cell types, resulting in induction of genes for adhesion molecules, chemokines, and cytokines (step 2). This leads to recruitment of additional leukocytes and potentially to a switch from PMN accumulation to the mononuclear phase of inflammation (step 3). See Jones and Rose-John, 2002, Kaplanski et al., 2003, Lindemann et al., 2004 for details. (B) Messenger RNA for IL-6Rα is present at baseline in freshly isolated human PMNs (time 0) and does not increase following activation with platelet activating factor (PAF) for the times shown. Transcript levels were assayed by real-time polymerase chain reaction. (C) Western analysis demonstrated that IL-6Rα protein is absent at baseline, but is rapidly synthesized when freshly isolated PMNs are activated. IL-6Rα synthesis is inhibited by rapamycin but not by actinomycin D. Panels B and C are reprinted from Lindemann et al. 2004.
These examples in key innate immune effector cells—monocytes and PMNs—demonstrate that specialized pathways regulate translation and new synthesis of specific proteins in response to outside-in inflammatory signals. In addition, translation control mechanisms can silence synthesis of critical proteins after their expression is induced by inflammatory signals. Ceruloplasmin (CP), an acute-phase protein synthesized by hepatocytes and cytokine-stimulated monocytic cells with incompletely characterized roles in inflammation, is an independent risk factor for atherosclerotic complications and can mediate oxidative modification of low density lipoprotein (Mazumder et al., 1997, Mazumder et al., 2003b). Thus, its uncontrolled synthesis and local release may contribute to atherosclerotic disease. Interferon (IFN)-γ, a mediator with complex roles in atherogenesis (Glass and Witztum 2001), induces expression of CP mRNA and protein in peripheral blood monocytes and U937 myelomonocytic cells in a time-dependent fashion (Mazumder et al. 1997). After a period of several hours of stimulation with IFN-γ, synthesis of CP protein is selectively terminated in the face of persistent abundant CP mRNA (Mazumder and Fox 1999). Additional observations indicated a mechanism of translational silencing involving circularization of the CP mRNA, which allows a cytoplasmic protein that binds to a specific sequence motif in the 3′ UTR—IFN-γ-activated inhibitor of translation (GAIT)—to interact with the 5′ UTR and/or 5′ UTR transacting proteins and block initiation of translation of the CP transcript (Sampath et al. 2003). Recently, ribosomal protein L13a was identified as the essential GAIT transacting protein in U937 cells, indicating a novel mechanism of translational control in which an integral ribosomal protein is released in response to a specific outside-in signal and mediates silencing (Mazumder et al., 2003a, Mazumder et al., 2003b). The intracellular signaling pathway remains to be defined, but appears to require delayed phosphorylation of L13a and dissociation of L13a from the large ribosomal subunit (Mazumder et al. 2003a). Translational repression of other transcripts by a soluble 3′ UTR binding protein has previously been reported in a U937 cell subline (Ch'ng et al. 1990). Furthermore, translational repression of IL-1β at a step beyond initiation has also been reported in activated human monocytic cells (Kaspar and Gehrke 1994), indicating additional diversity in the checkpoints and pathways of translational control in these cells.
Diversity in Signal-Dependent Translation: Molecular Mechanisms and Responding Cells
The previous examples illustrate diverse translational control mechanisms in activated human myeloid leukocytes and strongly suggest that additional variations will be discovered. Diversity of signal-dependent translational control is also evident from the variety of responding cells that display these mechanisms. Proliferating lymphocytes utilize translational regulation for approximately 20% of their genes (Pradet-Balade et al. 2001), an observation with relevance to atherosclerosis, transplantation, and immune cardiovascular disorders. Emerging evidence indicates that human endothelial cells have signal-dependent pathways for translational regulation and that translational checkpoints influence phenotypic changes in response to hemodynamic and humoral stimulation (Kraiss et al., 2000, Maeshima et al., 2002). The human platelet is perhaps the most surprising vascular cell to exhibit signal-dependent translation of physiologically relevant transcripts, providing a unique primary human cell model for examination of posttranscriptional regulation in the absence of a nucleus (Weyrich et al., 1998, Weyrich et al., 2003). It is likely that other cells of the vascular wall and myocardium will also have common and cell-specific mechanisms of signal-dependent translational regulation.
Therapeutic Implications
Because many translationally regulated genes code for proteins with potent biologic activities and key regulatory functions (Gingras et al., 2001, Kozak, 1991), signal-dependent pathways that control their expression may be particularly useful therapeutic targets. In addition, studies of cellular utilization of these pathways may reveal new mechanisms by which known therapeutic agents act. For example, rapamycin (sirolimus) inhibits vascular smooth muscle proliferation (Marks 2003). Nevertheless examples outlined here indicate that it also inhibits synthesis of key gene products controlled by mTOR in PMNs and monocytes, cells that accumulate at the earliest stages in vascular restenosis before intimal proliferation occurs (Welt and Rogers 2002). Thus, rapamycin may have effects that modulate inflammation in these earlier requisite stages, in addition to acting as an antiproliferative agent for smooth muscle cells later in the process. The outcome of such pharmacologic modulation will depend on the profile of gene products whose expression is regulated by mTOR (Lindemann et al., 2004, Mahoney et al., 2001, Weyrich et al., 2003, Yost et al., 2004). In addition to strategies based on inhibition, in some cases translational control mechanisms critically dampen or terminate inflammatory events (Mazumder et al., 2003a, Yost et al., 2004) and thus may be targets for gain-of-function therapeutic manipulation to this end. Additional definition of signal-dependent translation pathways will be required to fully exploit these potentials.
Acknowledgments
The authors thank their technical staff for help with the studies outlined in this review, Michele Czerwinski for preparation of the manuscript, their colleagues and trainees for helpful discussions, and Diana Lim for preparation of the figures. This work was supported by the National Institutes of Health (HL-44525 and P50 HL-50153 to GAZ, HL-66277 to ASW) and the American Heart Association (Established Investigator Award to A.S.W.; Western States Affiliate Postdoctoral Fellowship Award to S.W.L.).
References
- . MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004;116:281–297
- . Misregulated posttranscriptional checkpoints: inflammation and tumorigenesis. J Exp Med. 2001;193:F1–F4
- Widespread coronary inflammation in unstable angina. New Engl J Med. 2002;347:5–12
- . Translational control of gene expression and disease. Trends Mol Med. 2002;8:577–583
- Reversal of creatine kinase translational repression by 3′ untranslated sequences. Science. 1990;248:1003–1006
- . Platelet-monocyte aggregates: bridging thrombosis and inflammation. Circulation. 2002;105:2130–2132
- Differential regulation of matrix metalloproteinase-9 by monocytes adherent to collagen and platelets. Circ Res. 2001;89:509–516
- . Regulation of translation initiation by FRAP/mTOR. Genes Dev. 2001;15:807–826
- . Atherosclerosis: the road ahead. Cell. 2001;104:503–516
- . Regulation of local mRNA translation. Curr Opin Cell Biol. 2004;16:308–313
- . The role of soluble receptors in cytokine biology: the agonistic properties of the sIL-6R/IL-6 complex. Biochimica Biophysica Acta. 2002;1592:251–263
- IL-6: a regulator of the transition from neutrophil to monocyte recruitment during inflammation. Trends Immunol. 2003;24:25–29
- . Peripheral blood mononuclear cells stimulated with C5a or lipopolysaccharide to synthesize equivalent levels of IL-1 beta mRNA show unequal IL-1 beta protein accumulation but similar polyribosome profiles. J Immunol. 1994;153:277–286
- Novel path to activation of vascular smooth muscle cells: up-regulation of gp130 creates an autocrine activation loop by IL-6 and its soluble receptor. J Immunol. 1999;163:4583–4589
- . An analysis of vertebrate mRNA sequences: intimations of translational control. J Cell Biol. 1991;115:887–903
- Fluid flow activates a regulator of translation, p70/p85 S6 kinase, in human endothelial cells. Am J Physiol. 2000;278:H1537–H1544
- Activated platelets mediate inflammatory signaling by regulated interleukin 1 beta synthesis. J Cell Biol. 2001;154:485–490
- Neutrophils alter the inflammatory milieu by signal-dependent translation of constitutive messenger RNAs. Proc Natl Acad Sci USA. 2004;101:7076–7081
- . Diversity in translational regulation. Curr Opin Cell Biol. 2001;13:326–331
- Tumstatin, an endothelial cell-specific inhibitor of protein synthesis. Science. 2002;295:140–143
- Cell adhesion regulates gene expression at translational checkpoints in human myeloid leukocytes. Proc Natl Acad Sci USA 98. 2001;98:10,284–10,289
- . Sirolimus for the prevention of in-stent restenosis in a coronary artery. N Engl J Med. 2003;349:1307–1309
- . Origins and principles of translational control. In: Sonenberg N, Hershey JWB editor. Translational Control of Gene Expression. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 2000;p. 1–31
- . Delayed translational silencing of ceruloplasmin transcript in gamma interferon-activated U937 monocytic cells: role of the 3′ untranslated region. Mol Cell Biol. 1999;19:6898–6905
- Induction of ceruloplasmin synthesis by IFN-γ in human monocytic cells. J Immunol. 1997;159:1938–1944
- Regulated release of L13a from the 60S ribosomal subunit as a mechanism of transcript-specific translational control. Cell. 2003;115:187–198
- . Translational control by the 3′-UTR: the ends specify the means. Trends Biochem Sci. 2003;28:91–98
- Cell–cell interactions: leukocyte–endothelial interactions. Curr Opin Hematol. 2003;10:150–158
- Regulation of CLOCK and MOP4 by nuclear hormone receptors in the vasculature: a humoral mechanism to reset a peripheral clock. Cell. 2001;105:877–889
- Translation control: bridging the gap between genomic and proteomics?. Trends Biochem Sci. 2001;26:225–229
- . The target of rapamycin (TOR) proteins. Proc Natl Acad Sci USA. 2001;98:7037–7044
- . Transcript-selective translational silencing by gamma interferon is directed by a novel structural element in the ceruloplasmin mRNA 3′ untranslated region. Mol Cell Biol. 2003;23:1509–1519
- TOS motif-mediated raptor binding regulates 4E-BP1 multisite phosphorylation and function. Curr Biol. 2003;13:797–806
- . The nuclear factor-kappa B/inhibitor of kappa B autoregulatory system and atherosclerosis. Curr Opin Lipidol. 1998;9:387–396
- . Inflammation and restenosis in the stent era. Arterioscler Thromb Vasc Biol. 2002;22:1769–1776
- Monocyte tethering by P-selectin regulates monocyte chemotactic protein-1 and tumor necrosis factor-alpha secretion. Signal integration and NF-kappa B translocation. J Clin Invest. 1995;95:2387–2394
- Activated platelets signal chemokine synthesis by human monocytes. J Clin Invest. 1996;97:1525–1534
- Signal-dependent translation of a regulatory protein, Bcl-3, in activated human platelets. Proc Natl Acad Sci USA. 1998;95:5556–5561
- The evolving role of platelets in inflammation. J Thromb Haemost. 2003;1:1897–1905
- Activated polymorphonuclear leukocytes rapidly synthesize retinoic acid receptor-α. J Exp Med. 2004;200:671–680
PII: S1050-1738(04)00175-6
doi:10.1016/j.tcm.2004.10.004
© 2005 Elsevier Inc. All rights reserved.
