Atherosclerotic Plaque Macrophage Transcriptional Regulators Are Expressed in Blood and Modulated by Tristetraprolin
http://www.100md.com
Willmar D. Patino, Ju-Gyeong Kang, Satoa
参见附件。
the Cardiovascular Branch (W.D.P., J.-G.K., S.M., O.Y.M, P.M.H.) and the Pulmonary-Critical Care Medicine Branch (B.R.G.), National Heart, Lung
Blood Institute, National Institutes of Health, Bethesda, Md.
Abstract
Circulating monocytes and plaque macrophages mediate inflammation in the pathogenesis of atherosclerosis. We purified these cells from patients undergoing carotid endarterectomy for advanced atherosclerosis and examined their in vivo transcriptomes using the serial analysis of gene expression (SAGE) technique. We observed striking differences in transcriptional regulators as monocytes transformed into plaque macrophages in contrast to monocytes and lung macrophages from normal subjects. Consistent with its role in moderating inflammation, tristetraprolin (TTP, ZFP36) was among the most highly expressed macrophage transcriptional regulators. Interestingly, the mRNAs of a subset of the macrophage transcriptional regulators specifically interacted with TTP, revealing a network of genes that may be important in controlling macrophage inflammatory activity. Giving functional significance to this interaction, the knockdown of TTP increased both cognate macrophage gene mRNAs and inflammatory tumor necrosis factor protein release. In contrast, transient overexpression of TTP resulted in decreased levels of the same genes supporting its role in regulating macrophage gene expression. Together, our results indicate that the in vivo gene expression analyses of cells involved in pathogenesis can provide biological insights for functional studies with potential clinical implications.
Key Words: serial analysis of gene expression macrophage monocyte transcription
Introduction
The chronic vascular inflammation of atherosclerosis initiates early in life, thus understanding its pathogenesis may help us develop strategies to prevent its progression to symptomatic disease with aging.1,2 The clinical efficacy of cholesterol-lowering 3-hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reductase inhibitor (statin), with its fortuitous antiinflammatory activity, has clearly demonstrated the utility of targeting the immune system in atherosclerosis.3–5 Because macrophages mediate the inflammation and instability of atherosclerotic plaques, obtaining new insights into the molecular mechanisms regulating its function may provide opportunities to identify and modulate target genes involved in pathogenesis.
The complexity of the cardiovascular system, the difficulty of obtaining patient vascular tissue samples, and the lack of model systems have limited experimental studies of human atherosclerosis. Our previous approach to these difficulties involved studying purified monocytes of patients with atherosclerosis and using in vitro human monocytic cell lines for functional studies.6 Using this approach, we identified FOS as a diagnostic marker that may also contribute to pathogenesis,6 such as through its role in macrophage activity and calcification.7,8
To identify other genes that may be important in the disease process, we examined the transcriptomes of macrophages purified from atherosclerotic plaques of patients undergoing carotid endarterectomy (CEA) by the serial analysis of gene expression (SAGE) technique.9,10 Our analyses revealed a network of transcriptional regulators that were abundantly expressed in plaque macrophages, but not in lung macrophages, and may be related to their function in vascular tissues. These factors appeared to be increased in the circulating monocytes of patients with advanced carotid stenosis compared with normal subjects, possibly indicative of the systemic inflammation associated with atherosclerosis.
One of the most highly expressed macrophage transcriptional regulators was tristetraprolin (TTP), known to have antiinflammatory activity by binding and destabilizing tumor necrosis factor (TNF) mRNA.11 The targeted disruption of TTP in mice results in diffuse systemic inflammation manifest by arthritis and immunologic abnormalities,12 but to date its cardiovascular effects have not been reported. We report the specific interaction of plaque macrophage transcriptional regulator mRNAs with TTP protein. Giving functional significance to this interaction, the knockdown of TTP increased both cognate macrophage gene mRNAs and TNF protein release. In contrast, transient overexpression of TTP resulted in decreased levels of the same genes supporting its role in regulating macrophage gene expression.
Materials and Methods
Human Subjects
Subjects were recruited after informed consent in accordance with the National Institutes of Health Internal Review Board. Patients were selected from those scheduled to undergo CEA for atherosclerosis. Those selected for the plaque macrophage SAGE library construction also had concomitant coronary heart disease. The normal age-matched control subjects were screened to ensure absence of clinically evident atherosclerosis based on noninvasive cardiac testing, including carotid artery ultrasonogram with intima-media thickness (IMT) measurements. Normal subjects undergoing bronchoalveolar lavage were enrolled for lung macrophage purification. The exclusion criteria for all subjects were as follows: chronic infection, inflammatory disease, neoplasm, immunosuppression, and chemotherapy treatment.
Plaque Macrophage Purification
Monocytes and macrophages to be used for SAGE library synthesis were purified from blood and plaque tissue, respectively, from the same patients undergoing CEA surgery. Carotid artery plaques were processed within 1 hour of surgical resection, as generally described.13,14 The tissue was cubed (0.5 mm) and digested in HBSS (HEPES 4.8 mg/mL) containing collagenase type IV (450 U/mL), DNase I (500 U/mL), and trypsin inhibitor (1 mg/mL) (Worthington Biochemical Co) for 30 minutes to 1 hour at 37°C. From the sequentially filtered cell suspension (600 to 40 μm nylon mesh, Spectrum Laboratories Inc), macrophages were purified by using CD14 Microbeads and Fc Blocking reagent per protocol (Miltenyi Biotec) and resuspended in RNA Lysis/Binding buffer (Dynal Biotech Inc). Lung macrophages were similarly purified using CD14 Microbeads after filtering and washing the bronchoalveolar lavage. Cell viability was >95% by Trypan Blue exclusion. Purified macrophages were stained with CD14 fluorescein isothiocyanate (FITC) antibody (>90% CD14+), and RT-PCR for cell markers was also performed to determine purity (plaque macrophage purity analyses in Figure I in the online data supplement available at http://circres.ahajournals.org).
Blood Monocyte Purification
Blood samples were collected in citrate-containing Vacutainer CPT tubes (Becton Dickinson) from controls and from patients intraoperatively and processed within 1 hour to obtain mononuclear cells (MNC).15 Monocytes were purified from the mononuclear fraction using CD14 MicroBeads for SAGE library synthesis as previously described.6
Cells and Tissue Culture
THP1 human monocytic cell line was purchased from the American Type Culture Collection and maintained as recommended. For activation of THP1 cells, 2 nmol/L phorbol 12-myristate 13-acetate (PMA) was added for the indicated times. Samples were collected by gently washing unattached cells and pooling them with adherent cells released with trypsin-EDTA. Cell viability was determined using Trypan Blue. For the small interfering RNA (siRNA) inhibition and transient TTP expression studies, untouched primary human monocyte were isolated from blood of healthy donors using a negative selection kit (Miltenyi Biotec) and cultured as described.16
Serial Analysis of Gene Expression
The LongSAGE protocol was used to create the libraries; 50 000 to 100 000 tags were sequenced per library as previously described.10 Tag counts were determined using the SAGE2000 software (http://www.sagenet.org). Tag counts by unique sequences or by Unigene clusters were normalized to 100 000 tags per library and simple fold changes were calculated using Microsoft Access software and the Unigene/SAGEmap database (0 tag was arbitrarily assigned a value of 0.5 for fold-change calculation).17 Functional annotation was performed by importing Unigene IDs with the corresponding tag counts and fold changes into the pathway analysis software (Ingenuity Systems).
Quantitative Real-Time RT-PCR
mRNA from lysates were purified by binding to poly(dT) magnetic beads (Dynal Biotech Inc), reverse transcribed using Superscript II (Invitrogen), and quantified by SYBR Green (Molecular Probes) RT-PCR using standard protocols on the 7900HT (Applied Biosystems Inc).18 RT-PCR data were normalized by measuring average cycle threshold (Ct) ratios between candidate genes and the control gene eukaryotic translation initiation factor (EIF3S5, TIF). The formula 2Ct(Candidate)/2Ct(Control) was used to calculate normalized ratios. Color-coded normalized fold changes were generated from log 2–transformed control-normalized ratios (normalized Ct ratio divided by the average Ct ratio of all control samples) using Cluster v2.2 and Treeview Software (http://rana.lbl.gov/EisenSoftware.htm).18 The primer sequences for all the various genes are provided in supplemental Table I. All RT-PCR measurements were done in duplicates and repeated at least 3 times.
Immunohistochemistry, Western Blotting, and TNF Immunoassay
Antibodies were as follows: rabbit polyclonal TTP (sc-14030, Santa Cruz Biotechnology); mouse monoclonal human CD14 (Immunotech); mouse monoclonal GAPD (Ambion); control rabbit IgG (sc-2027, Santa Cruz Biotechnology); and mouse IgG (Biocare). Cryosections (8 to 10 μm) of carotid plaques were immunostained with Vector Blue substrate (Vector Laboratory) and developed by secondary antibody conjugated to alkaline phosphatase.19 Western blotting was performed using standard SDS-PAGE, semidry protein transfer, and chemiluminescent substrate visualization with horse radish peroxidase–conjugated secondary antibody (Pierce Biotechnology). For Western blotting with immunoprecipitated samples, Rabbit IgG TrueBlot secondary antibody (eBioscience) was used to reduce interference by the immunoprecipitating primary antibody. TNF protein release was measured using the Quantikine human TNF- immunoassay kit (R&D Systems).
TTP Protein Immunoprecipitation
Cytoplasmic extracts were prepared by incubating THP1 cells (107, stimulated with 2 nmol/L PMA for 4 hours) in 1 mL of immunoprecipitation buffer (10 mmol/L Tris [pH 7.5], 2 mmol/L MgCl2, 150 mmol/L KCl, 5% [vol/vol] glycerol, 2 mmol/L dithiothreitol, 0.5% [vol/vol] Brij58 [Pierce], and protease inhibitor cocktail [Roche]) for 30 minutes on ice. Protein-G Sepharose 4B FF beads (30 μL; Amersham) and 10 μg of anti-TTP antibody or control rabbit IgG were added and rotated overnight at 4°C. The beads were pelleted by centrifugation at 200g for 1 minute and washed 3 times with immunoprecipitation buffer. Total RNA was isolated from the immunoprecipitate by phenol/chloroform extraction followed by ethanol precipitation.
Transfection of siRNA and Plasmid DNA
Nonspecific and TTP-specific siRNA duplexes were purchased from Dharmacon Research. The pCMV6-TTP expression vector was purchased from Origene Technologies Inc. siRNAs and plasmids were transfected using Nucleofector (Amaxa Inc) according to the protocol of the manufacturer. For THP1 cells, 0.5 μg of siRNAs or plasmid DNA was transfected into 1x106 cells in 100 μL of Nucleofector Solution V (program U-01) and then incubated in standard THP1 medium containing 2 nmol/L PMA for the indicated times. For human monocytes, 0.5 μg of siRNAs or plasmid DNA was transfected into 1.5x106 cells in 100 μL of Human Monocyte Nucleofector Solution (program Y-01) and then incubated in Human Monocyte Nucleofector Medium (Amaxa) supplemented with 2 mmol/L glutamine and 10% FCS.
Statistical Analysis
The 6 SAGE libraries described in this report represent approximately 10 million base pairs of DNA sequencing. Because of the time and resource requirements of the SAGE technique, making multiple SAGE libraries for replication purposes is not feasible with the current technology. Nonetheless, in this study 2 different subjects were used for making the patient macrophage and monocyte libraries with good correlation, and the differential SAGE tags were highly reproducible as confirmed by RT-PCR. Data obtained by Ingenuity analysis software were exported into Excel spreadsheets and probability values for SAGE tag counts calculated accounting for library size differences as previously described.20
Data are expressed as mean±SD. Probability values were calculated with the use of Student’s t test.
Results
SAGE Libraries and Analyses
We created a total of 6 SAGE libraries, 4 from patients with atherosclerosis and 2 from control subjects without any clinical evidence of atherosclerosis (Table 1). From the same 2 patients (P1 and P2) undergoing CEA, paired monocyte and macrophage SAGE libraries (plaque macrophages, P1Mp and P2Mp; blood monocytes, P1Mc and P2Mc) were made to examine gene expression transformation during differentiation. As a screen to identify plaque macrophage enriched genes, we created a lung macrophage (L1Mp) library purified from the bronchoalveolar lavage of a normal subject (L1). A control monocyte (C1Mc) library from a normal age-matched control subject (C1) was also used for comparison (all SAGE libraries available at http://cgap.nci.nih.gov/SAGE).
Using the 464 486 SAGE tags sequenced in the 6 libraries, 14 233 genes were identified by matching them to the Unigene database.17 For the purposes of this study, we focused only on the tags identified by Unigene. SAGE tags representing monocyte and macrophage cell markers were appropriately enriched in the respective libraries (supplemental Figure II). Because transcriptional events are critical for differentiation and the SAGE technique accurately measures transcript levels, we focused on identifying genes involved in macrophage transcriptional regulation using the following scheme: (1) examination of genes classified as transcriptional regulators by global function and pathway analysis software (Ingenuity Systems); (2) selection of genes with high SAGE tag counts in duplicate libraries to increase reproducibility by RT-PCR; and (3) subselection of genes differentially expressed in inflamed plaque macrophages compared with quiescent normal lung macrophages.
Of the 14 233 genes matched to the Unigene database, 3989 were classified into functional families by the Ingenuity program, of which 548 belonged to the transcriptional regulator family (supplemental Tables II and III, respectively). The 548 genes were sorted by fold change (2-fold, plaque versus lung macrophages) and total tag count (12 tags combining all 6 libraries) for increased reproducibility. Applying these criteria resulted in a smaller group of 72 plaque macrophage transcriptional regulators (supplemental Table IV).
Plaque Macrophage Transcriptional Regulators
We selected the most abundantly expressed transcriptional regulators that were relatively specific to plaque macrophages (P<0.05; supplemental Table IV). In agreement with our previous observation, the protooncogene FOS, important in inflammation and myeloid cell differentiation, was among the top genes on the list (Table 2).6 As monocytes differentiated into macrophages, JUN, a FOS partner, which together comprise transcription activating protein-1 (AP-1), and their homologs JUNB and FOSB were coexpressed in plaque macrophages. Cyclin-dependent kinase inhibitor CDKN1A (p21), also important for atherosclerosis and with recently described transcriptional activity, was identified.21,22
A significant fraction of the differentially expressed genes were zinc finger proteins (Table 2 and supplemental Table IV). We observed high levels of early growth response 1 (EGR1), Kruppel-like transcription factors (KLF), and mRNA destabilizing Cys-Cys-Cys-His zinc finger protein 36 (ZFP36), also known as TTP.11,23,24 Of the Kruppel-like factors, KLF4 has been shown to be a proinflammatory factor in macrophages and, along with KLF6, associated with regulating p21.25–27 Although the prototypical mRNA target of TTP is the "AU-rich" element (ARE) of TNF mRNA, we observed similar sequence elements in a number of our macrophage candidate genes raising the possibility of their interaction with TTP.
Expression in Plaque Macrophages and in Circulating Monocytes
We first confirmed the SAGE tag fold changes by testing mRNA levels of the nine selected genes (Table 2) using normal lung macrophages and purified monocytes and plaque macrophage mRNAs from CEA patients (Figure 1A). The differential expression of the genes by RT-PCR corresponded well with the SAGE tag data. These results were also confirmed using PMA-activated THP1 cells for use as an in vitro model of macrophage differentiation (Figure 1B).
We have previously shown that FOS mRNA levels in monocytes may reflect their level of activation and atherosclerosis severity.6 In this study, we wondered whether there may be similar correlations between the levels of our macrophage genes and atherosclerosis. As an exploratory study, we examined purified monocytes from 14 patients and 14 control subjects with the highest and lowest levels of FOS, respectively.6 The patient and control subjects were closely matched except for risk factors inherently associated with atherosclerosis such as male gender, smoking, and prior history of coronary heart disease (Table 1). In support of an inflammatory model of atherosclerosis, the patient group showed an activated monocyte pattern compared with controls (Figure 1C). FOS mRNA levels between the 2 groups showed the most significant difference by probability value as expected based on their selection criteria. However, 3 other genes, FOSB, JUNB, and JUN, were significantly higher in patients compared with controls, P=0.003, P=0.005, and P=0.02, respectively.
TTP Interacts With a Subset of Transcriptional Regulators
Because many of our macrophage candidate genes contained ARE regions and are known to be posttranscriptionally regulated,28 we investigated whether TTP interacted with them in our activated THP1 monocytic cell model. We first coimmunoprecipitated the RNA–TTP complex from the cytosolic extracts of PMA-stimulated THP1 cells using anti-TTP antibody at an optimized potassium concentration of 150 mmol/L (data not shown). These were analyzed by Western blotting to ensure specific pull-down of the protein (Figure 2A). We next checked the immunoprecipitate for the presence of known TTP targets TNF and TTP mRNAs as internal positive controls and observed high levels of specific binding by RT-PCR (Figure 2B). Of our queried genes, FOS, EGR1, and KLF6 were significantly coimmunoprecipitated by TTP-specific antibody compared with control IgG (P<0.05 indicated by asterisks, Figure 2B). KLF4, p21, and JUNB only showed a trend toward TTP binding under our optimized in vitro conditions. FOSB and JUN did not show significant binding, although this could also be explained by their low transcript levels under these conditions (Figure 2B).
Inhibition of TTP by siRNA Increases Putative Target mRNAs
We next examined the functional significance of the interaction between the macrophage transcriptional regulators and TTP using both THP1 and primary human monocytes. We used siRNA to knockdown TTP in these cells and ascertained its decreased protein level by Western blotting, whereas GAPD served as protein loading control (Figure 3A and 3B). TTP siRNA–transfected cells had significantly higher levels of FOS and p21 mRNAs compared with nonspecific siRNA, demonstrating functional interaction between these transcripts and TTP (Figure 3). The relative fold-change increases in the macrophage transcriptional regulators by TTP inhibition were comparable to that observed for the positive control TNF mRNA. EGR1 and KLF6 mRNA levels were not as significantly affected, and FOSB and JUN were too low to be detected at 48 hours, when maximal effects were seen for the other factors (Figure 3, data not shown).
siRNA Inhibition of TTP Increases TNF Protein Release by Human Monocytes
We also investigated whether the increased TNF mRNA by TTP inhibition translated to augmented TNF protein release to functionally confirm TTP RNA interference in our cell system. By Western blotting, we verified decreased TTP protein levels over a 48-hour time period after specific siRNA transfection, both in THP1 cells stimulated with PMA and in cultured primary human monocytes (Figure 4A and 4B, respectively). During this same period, we observed relatively increased TNF protein release that corresponded to TTP protein knockdown compared with control in both cell types (Figure 4A and 4B). These results overtly fit with the expected results of blocking TNF mRNA destabilization, although the regulatory factors involved in TNF protein release are likely to be complex. Overall, the relative increases in released TNF protein were maintained over the time course of TTP RNA interference, validating our experimental system.
Increased Expression of TTP Attenuates Macrophage Transcriptional Regulators
To obtain additional evidence for the interaction between TTP and macrophage transcriptional regulators, we examined the effects of augmenting TTP expression on the mRNA levels of the macrophage genes. The transient expression of TTP increased its proteins levels and decreased the levels of TNF mRNA in both THP1 and primary human monocytes as positive control (Figure 5A and 5B). As predicted from the results of the siRNA experiments inhibiting TTP protein, transiently overexpressing it gave the expected opposite effect of decreasing the transcript levels of the macrophage genes (Figure 5). Similar to the TTP inhibition experiment (Figure 3), the levels of housekeeping gene GAPD were not affected by TTP alteration and served as negative control (Figure 5).
Discussion
We view this first report of in vivo purified human plaque macrophage SAGE database as an important resource for delineating the genetic program of these cells in atherosclerosis. Tissue culture–differentiated human macrophage SAGE libraries have been described,29 but these were poorly correlated to our libraries, highlighting the significance of cellular environment on gene expression. Our observation of circulating monocytes from patients expressing plaque macrophage transcriptional regulators supports a model of systemic inflammation in atherosclerosis and reinforces the utility of focusing on the in vivo transcriptomes of human samples. By comparing purified macrophages from atherosclerotic plaque to lung macrophages and normal monocytes, we have identified plaque-enriched transcriptional regulators. These include additional stress response genes of the AP-1/FOS pathway that previously had not been appreciated by comparing only monocyte expression profiles.6 Our demonstration of an antiinflammatory protein, TTP, functionally interacting with FOS, EGR1, and p21 corroborates our current analyses because these genes have been shown to be important in macrophage differentiation and atherosclerosis.21,23,30
The current analyses reveal networks of genes involved in the activation and inhibition of transcription for macrophage function that may result in complex interactions, such as gene expression oscillations for improved signaling.31 Interestingly, 2 of the 6 atherosclerosis marker genes we previously identified in patient monocytes, FOS and period homolog 1 (PER1), are thought to be important in circadian rhythm and autoregulatory feedback.6,32,33 We also observed an increased pattern of AP-1 gene expression in patients with atherosclerosis that is now further supported by the paired increases in the levels of FOS/JUNB and FOSB/JUN in monocyte and macrophage transcriptomes, respectively (Table 2). Although interaction of AP-1 elements are variable and complex, these expression patterns are likely to be important for macrophage differentiation.28,30,34 The interaction of TTP with FOS or other AP-1 partners may have important consequences, as they can regulate p21 and the differentiation of mononuclear cells into osteoclasts.35,36
Our study not only demonstrates the potential to reveal new insights into pathogenesis but also to clarify the mechanism of clinically important therapies. While screening for factors that may augment TTP levels, we observed that the HMG-CoA reductase inhibitor simvastatin caused an increase in monocyte TTP mRNA and protein levels (data not shown). However, because statins have been shown to augment lipopolysaccharide-induced TNF transcription by inhibiting FOS expression37 and to induce differentiation and apoptosis in human monocytes,38 the functional significance of increased TTP was difficult to interpret in this context. Nonetheless, the observation of TTP upregulation by statins merits further study as a potential mechanism by which this clinically efficacious medication may modulate inflammation.
Macrophages constitute a heterogeneous population of cells that may be subspecialized even within a specific tissue.39 Our plaque and lung macrophage transcriptomes express a markedly different profile of transcriptional regulators likely related to their specific functions. Insights obtained from these differences and from creating other macrophage transcriptomes (such as that of liver Kupffer macrophages, which have recently been shown to affect atherosclerosis in mice40) may yield more useful information. The current study serves as 1 perspective for interpreting our SAGE transcriptomes, but we anticipate that it will be a starting point for examining other genetic networks important in human macrophage function. Using similar analyses for other human diseases may also reveal new genetic insights with potential translational applications.
Acknowledgments
This research was supported by the Division of Intramural Research, National Heart, Lung, and Blood Institutes, the National Institutes of Health. We thank Toren Finkel, Michael N. Sack, and Neal Epstein for advice and critical reading of the manuscript and Perry J. Blackshear for helpful discussions regarding TTP.
Footnotes
Both authors contributed equally to this study.
Original received October 1, 2005; resubmission received February 13, 2006; revised resubmission received March 21, 2006; accepted March 30, 2006.
References
Glass CK, Witztum JL. Atherosclerosis. The road ahead. Cell. 2001; 104: 503–516. [Order article via Infotrieve]
Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med. 2005; 352: 1685–1695.
Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation. 2002; 105: 1135–1143.
Jain MK, Ridker PM. Anti-inflammatory effects of statins: clinical evidence and basic mechanisms. Nat Rev Drug Discov. 2005; 4: 977–987. [Order article via Infotrieve]
Patino WD, Mian OY, Kang JG, Matoba S, Bartlett LD, Holbrook B, Trout HH 3rd, Kozloff L, Hwang PM. Circulating transcriptome reveals markers of atherosclerosis. Proc Natl Acad Sci U S A. 2005; 102: 3423–3428.
Wang ZQ, Ovitt C, Grigoriadis AE, Mohle-Steinlein U, Ruther U, Wagner EF. Bone and haematopoietic defects in mice lacking c-fos. Nature. 1992; 360: 741–745. [Order article via Infotrieve]
Johnson RS, Spiegelman BM, Papaioannou V. Pleiotropic effects of a null mutation in the c-fos proto-oncogene. Cell. 1992; 71: 577–586. [Order article via Infotrieve]
Polyak K, Riggins GJ. Gene discovery using the serial analysis of gene expression technique: implications for cancer research. J Clin Oncol. 2001; 19: 2948–2958.
Saha S, Sparks AB, Rago C, Akmaev V, Wang CJ, Vogelstein B, Kinzler KW, Velculescu VE. Using the transcriptome to annotate the genome. Nat Biotechnol. 2002; 20: 508–512. [Order article via Infotrieve]
Blackshear PJ. Tristetraprolin and other CCCH tandem zinc-finger proteins in the regulation of mRNA turnover. Biochem Soc Trans. 2002; 30: 945–952. [Order article via Infotrieve]
Carballo E, Lai WS, Blackshear PJ. Feedback inhibition of macrophage tumor necrosis factor-alpha production by tristetraprolin. Science. 1998; 281: 1001–1005.
St Croix B, Rago C, Velculescu V, Traverso G, Romans KE, Montgomery E, Lal A, Riggins GJ, Lengauer C, Vogelstein B, Kinzler KW. Genes expressed in human tumor endothelium. Science. 2000; 289: 1197–1202.
Liu-Wu Y, Svenningsson A, Stemme S, Holm J, Wiklund O. Identification and analysis of macrophage-derived foam cells from human atherosclerotic lesions by using a "mock" FL3 channel in flow cytometry. Cytometry. 1997; 29: 155–164. [Order article via Infotrieve]
Holodniy M, Mole L, Yen-Lieberman B, Margolis D, Starkey C, Carroll R, Spahlinger T, Todd J, Jackson JB. Comparative stabilities of quantitative human immunodeficiency virus RNA in plasma from samples collected in VACUTAINER CPT, VACUTAINER PPT, and standard VACUTAINER tubes. J Clin Microbiol. 1995; 33: 1562–1566.
Seager Danciger J, Lutz M, Hama S, Cruz D, Castrillo A, Lazaro J, Phillips R, Premack B, Berliner J. Method for large scale isolation, culture and cryopreservation of human monocytes suitable for chemotaxis, cellular adhesion assays, macrophage and dendritic cell differentiation. J Immunol Methods. 2004; 288: 123–134. [Order article via Infotrieve]
Lash AE, Tolstoshev CM, Wagner L, Schuler GD, Strausberg RL, Riggins GJ, Altschul SF. SAGEmap: a public gene expression resource. Genome Res. 2000; 10: 1051–1060.
Cerutti JM, Delcelo R, Amadei MJ, Nakabashi C, Maciel RM, Peterson B, Shoemaker J, Riggins GJ. A preoperative diagnostic test that distinguishes benign from malignant thyroid carcinoma based on gene expression. J Clin Invest. 2004; 113: 1234–1242.
Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B. PUMA induces the rapid apoptosis of colorectal cancer cells. Mol Cell. 2001; 7: 673–682. [Order article via Infotrieve]
Audic S, Claverie JM. The significance of digital gene expression profiles. Genome Res. 1997; 7: 986–995.
Merched AJ, Chan L. Absence of p21Waf1/Cip1/Sdi1 modulates macrophage differentiation and inflammatory response and protects against atherosclerosis. Circulation. 2004; 110: 3830–3841.
Devgan V, Mammucari C, Millar SE, Brisken C, Dotto GP. p21WAF1/Cip1 is a negative transcriptional regulator of Wnt4 expression downstream of Notch1 activation. Genes Dev. 2005; 19: 1485–1495.
McCaffrey TA, Fu C, Du B, Eksinar S, Kent KC, Bush H Jr, Kreiger K, Rosengart T, Cybulsky MI, Silverman ES, Collins T. High-level expression of Egr-1 and Egr-1-inducible genes in mouse and human atherosclerosis. J Clin Invest. 2000; 105: 653–662.
Feinberg MW, Lin Z, Fisch S, Jain MK. An emerging role for Kruppel-like factors in vascular biology. Trends Cardiovasc Med. 2004; 14: 241–246. [Order article via Infotrieve]
Zhang W, Geiman DE, Shields JM, Dang DT, Mahatan CS, Kaestner KH, Biggs JR, Kraft AS, Yang VW. The gut-enriched Kruppel-like factor (Kruppel-like factor 4) mediates the transactivating effect of p53 on the p21WAF1/Cip1 promoter. J Biol Chem. 2000; 275: 18391–18398.
Narla G, Heath KE, Reeves HL, Li D, Giono LE, Kimmelman AC, Glucksman MJ, Narla J, Eng FJ, Chan AM, Ferrari AC, Martignetti JA, Friedman SL. KLF6, a candidate tumor suppressor gene mutated in prostate cancer. Science. 2001; 294: 2563–2566.
Feinberg MW, Cao Z, Wara AK, Lebedeva MA, Senbanerjee S, Jain MK. Kruppel-like factor 4 is a mediator of proinflammatory signaling in macrophages. J Biol Chem. 2005; 280: 38247–38258.
Hess J, Angel P, Schorpp-Kistner M. AP-1 subunits: quarrel and harmony among siblings. J Cell Sci. 2004; 117: 5965–5973.
Hashimoto S, Suzuki T, Dong HY, Yamazaki N, Matsushima K. Serial analysis of gene expression in human monocytes and macrophages. Blood. 1999; 94: 837–844.
Liebermann DA, Gregory B, Hoffman B. AP-1 (Fos/Jun) transcription factors in hematopoietic differentiation and apoptosis. Int J Oncol. 1998; 12: 685–700. [Order article via Infotrieve]
Lipan O, Wong WH. The use of oscillatory signals in the study of genetic networks. Proc Natl Acad Sci U S A. 2005; 102: 7063–7068.
Hirayama J, Cardone L, Doi M, Sassone-Corsi P. Common pathways in circadian and cell cycle clocks: light-dependent activation of Fos/AP-1 in zebrafish controls CRY-1a and WEE-1. Proc Natl Acad Sci U S A. 2005; 102: 10194–10199.
Brown SA, Ripperger J, Kadener S, Fleury-Olela F, Vilbois F, Rosbash M, Schibler U. PERIOD1-associated proteins modulate the negative limb of the mammalian circadian oscillator. Science. 2005; 308: 693–696.
Shaulian E, Karin M. AP-1 as a regulator of cell life and death. Nat Cell Biol. 2002; 4: E131–E136. [Order article via Infotrieve]
Kardassis D, Papakosta P, Pardali K, Moustakas A. c-Jun transactivates the promoter of the human p21(WAF1/Cip1) gene by acting as a superactivator of the ubiquitous transcription factor Sp1. J Biol Chem. 1999; 274: 29572–29581.
Kwak HB, Jin HM, Ha H, Kang MJ, Lee SB, Kim HH, Lee ZH. Tumor necrosis factor-alpha induces differentiation of human peripheral blood mononuclear cells into osteoclasts through the induction of p21(WAF1/Cip1). Biochem Biophys Res Commun. 2005; 330: 1080–1086. [Order article via Infotrieve]
Matsumoto M, Einhaus D, Gold ES, Aderem A. Simvastatin augments lipopolysaccharide-induced proinflammatory responses in macrophages by differential regulation of the c-Fos and c-Jun transcription factors. J Immunol. 2004; 172: 7377–7384.
Vamvakopoulos JE, Green C. HMG-CoA reductase inhibition aborts functional differentiation and triggers apoptosis in cultured primary human monocytes: a potential mechanism of statin-mediated vasculoprotection. BMC Cardiovasc Disord. 2003; 3: 6. [Order article via Infotrieve]
Ross JA, Auger MJ. The biology of the macrophage. In: Burke B, Lewis CE, eds. The Macrophage. 2nd ed. Oxford, UK: Oxford University Press; 2002: 1–57.
Bradshaw G, Gutierrez A, Miyake JH, Davis KR, Li AC, Glass CK, Curtiss LK, Davis RA. Facilitated replacement of Kupffer cells expressing a paraoxonase-1 transgene is essential for ameliorating atherosclerosis in mice. Proc Natl Acad Sci U S A. 2005; 102: 11029–11034.
the Cardiovascular Branch (W.D.P., J.-G.K., S.M., O.Y.M, P.M.H.) and the Pulmonary-Critical Care Medicine Branch (B.R.G.), National Heart, Lung
Blood Institute, National Institutes of Health, Bethesda, Md.
Abstract
Circulating monocytes and plaque macrophages mediate inflammation in the pathogenesis of atherosclerosis. We purified these cells from patients undergoing carotid endarterectomy for advanced atherosclerosis and examined their in vivo transcriptomes using the serial analysis of gene expression (SAGE) technique. We observed striking differences in transcriptional regulators as monocytes transformed into plaque macrophages in contrast to monocytes and lung macrophages from normal subjects. Consistent with its role in moderating inflammation, tristetraprolin (TTP, ZFP36) was among the most highly expressed macrophage transcriptional regulators. Interestingly, the mRNAs of a subset of the macrophage transcriptional regulators specifically interacted with TTP, revealing a network of genes that may be important in controlling macrophage inflammatory activity. Giving functional significance to this interaction, the knockdown of TTP increased both cognate macrophage gene mRNAs and inflammatory tumor necrosis factor protein release. In contrast, transient overexpression of TTP resulted in decreased levels of the same genes supporting its role in regulating macrophage gene expression. Together, our results indicate that the in vivo gene expression analyses of cells involved in pathogenesis can provide biological insights for functional studies with potential clinical implications.
Key Words: serial analysis of gene expression macrophage monocyte transcription
Introduction
The chronic vascular inflammation of atherosclerosis initiates early in life, thus understanding its pathogenesis may help us develop strategies to prevent its progression to symptomatic disease with aging.1,2 The clinical efficacy of cholesterol-lowering 3-hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reductase inhibitor (statin), with its fortuitous antiinflammatory activity, has clearly demonstrated the utility of targeting the immune system in atherosclerosis.3–5 Because macrophages mediate the inflammation and instability of atherosclerotic plaques, obtaining new insights into the molecular mechanisms regulating its function may provide opportunities to identify and modulate target genes involved in pathogenesis.
The complexity of the cardiovascular system, the difficulty of obtaining patient vascular tissue samples, and the lack of model systems have limited experimental studies of human atherosclerosis. Our previous approach to these difficulties involved studying purified monocytes of patients with atherosclerosis and using in vitro human monocytic cell lines for functional studies.6 Using this approach, we identified FOS as a diagnostic marker that may also contribute to pathogenesis,6 such as through its role in macrophage activity and calcification.7,8
To identify other genes that may be important in the disease process, we examined the transcriptomes of macrophages purified from atherosclerotic plaques of patients undergoing carotid endarterectomy (CEA) by the serial analysis of gene expression (SAGE) technique.9,10 Our analyses revealed a network of transcriptional regulators that were abundantly expressed in plaque macrophages, but not in lung macrophages, and may be related to their function in vascular tissues. These factors appeared to be increased in the circulating monocytes of patients with advanced carotid stenosis compared with normal subjects, possibly indicative of the systemic inflammation associated with atherosclerosis.
One of the most highly expressed macrophage transcriptional regulators was tristetraprolin (TTP), known to have antiinflammatory activity by binding and destabilizing tumor necrosis factor (TNF) mRNA.11 The targeted disruption of TTP in mice results in diffuse systemic inflammation manifest by arthritis and immunologic abnormalities,12 but to date its cardiovascular effects have not been reported. We report the specific interaction of plaque macrophage transcriptional regulator mRNAs with TTP protein. Giving functional significance to this interaction, the knockdown of TTP increased both cognate macrophage gene mRNAs and TNF protein release. In contrast, transient overexpression of TTP resulted in decreased levels of the same genes supporting its role in regulating macrophage gene expression.
Materials and Methods
Human Subjects
Subjects were recruited after informed consent in accordance with the National Institutes of Health Internal Review Board. Patients were selected from those scheduled to undergo CEA for atherosclerosis. Those selected for the plaque macrophage SAGE library construction also had concomitant coronary heart disease. The normal age-matched control subjects were screened to ensure absence of clinically evident atherosclerosis based on noninvasive cardiac testing, including carotid artery ultrasonogram with intima-media thickness (IMT) measurements. Normal subjects undergoing bronchoalveolar lavage were enrolled for lung macrophage purification. The exclusion criteria for all subjects were as follows: chronic infection, inflammatory disease, neoplasm, immunosuppression, and chemotherapy treatment.
Plaque Macrophage Purification
Monocytes and macrophages to be used for SAGE library synthesis were purified from blood and plaque tissue, respectively, from the same patients undergoing CEA surgery. Carotid artery plaques were processed within 1 hour of surgical resection, as generally described.13,14 The tissue was cubed (0.5 mm) and digested in HBSS (HEPES 4.8 mg/mL) containing collagenase type IV (450 U/mL), DNase I (500 U/mL), and trypsin inhibitor (1 mg/mL) (Worthington Biochemical Co) for 30 minutes to 1 hour at 37°C. From the sequentially filtered cell suspension (600 to 40 μm nylon mesh, Spectrum Laboratories Inc), macrophages were purified by using CD14 Microbeads and Fc Blocking reagent per protocol (Miltenyi Biotec) and resuspended in RNA Lysis/Binding buffer (Dynal Biotech Inc). Lung macrophages were similarly purified using CD14 Microbeads after filtering and washing the bronchoalveolar lavage. Cell viability was >95% by Trypan Blue exclusion. Purified macrophages were stained with CD14 fluorescein isothiocyanate (FITC) antibody (>90% CD14+), and RT-PCR for cell markers was also performed to determine purity (plaque macrophage purity analyses in Figure I in the online data supplement available at http://circres.ahajournals.org).
Blood Monocyte Purification
Blood samples were collected in citrate-containing Vacutainer CPT tubes (Becton Dickinson) from controls and from patients intraoperatively and processed within 1 hour to obtain mononuclear cells (MNC).15 Monocytes were purified from the mononuclear fraction using CD14 MicroBeads for SAGE library synthesis as previously described.6
Cells and Tissue Culture
THP1 human monocytic cell line was purchased from the American Type Culture Collection and maintained as recommended. For activation of THP1 cells, 2 nmol/L phorbol 12-myristate 13-acetate (PMA) was added for the indicated times. Samples were collected by gently washing unattached cells and pooling them with adherent cells released with trypsin-EDTA. Cell viability was determined using Trypan Blue. For the small interfering RNA (siRNA) inhibition and transient TTP expression studies, untouched primary human monocyte were isolated from blood of healthy donors using a negative selection kit (Miltenyi Biotec) and cultured as described.16
Serial Analysis of Gene Expression
The LongSAGE protocol was used to create the libraries; 50 000 to 100 000 tags were sequenced per library as previously described.10 Tag counts were determined using the SAGE2000 software (http://www.sagenet.org). Tag counts by unique sequences or by Unigene clusters were normalized to 100 000 tags per library and simple fold changes were calculated using Microsoft Access software and the Unigene/SAGEmap database (0 tag was arbitrarily assigned a value of 0.5 for fold-change calculation).17 Functional annotation was performed by importing Unigene IDs with the corresponding tag counts and fold changes into the pathway analysis software (Ingenuity Systems).
Quantitative Real-Time RT-PCR
mRNA from lysates were purified by binding to poly(dT) magnetic beads (Dynal Biotech Inc), reverse transcribed using Superscript II (Invitrogen), and quantified by SYBR Green (Molecular Probes) RT-PCR using standard protocols on the 7900HT (Applied Biosystems Inc).18 RT-PCR data were normalized by measuring average cycle threshold (Ct) ratios between candidate genes and the control gene eukaryotic translation initiation factor (EIF3S5, TIF). The formula 2Ct(Candidate)/2Ct(Control) was used to calculate normalized ratios. Color-coded normalized fold changes were generated from log 2–transformed control-normalized ratios (normalized Ct ratio divided by the average Ct ratio of all control samples) using Cluster v2.2 and Treeview Software (http://rana.lbl.gov/EisenSoftware.htm).18 The primer sequences for all the various genes are provided in supplemental Table I. All RT-PCR measurements were done in duplicates and repeated at least 3 times.
Immunohistochemistry, Western Blotting, and TNF Immunoassay
Antibodies were as follows: rabbit polyclonal TTP (sc-14030, Santa Cruz Biotechnology); mouse monoclonal human CD14 (Immunotech); mouse monoclonal GAPD (Ambion); control rabbit IgG (sc-2027, Santa Cruz Biotechnology); and mouse IgG (Biocare). Cryosections (8 to 10 μm) of carotid plaques were immunostained with Vector Blue substrate (Vector Laboratory) and developed by secondary antibody conjugated to alkaline phosphatase.19 Western blotting was performed using standard SDS-PAGE, semidry protein transfer, and chemiluminescent substrate visualization with horse radish peroxidase–conjugated secondary antibody (Pierce Biotechnology). For Western blotting with immunoprecipitated samples, Rabbit IgG TrueBlot secondary antibody (eBioscience) was used to reduce interference by the immunoprecipitating primary antibody. TNF protein release was measured using the Quantikine human TNF- immunoassay kit (R&D Systems).
TTP Protein Immunoprecipitation
Cytoplasmic extracts were prepared by incubating THP1 cells (107, stimulated with 2 nmol/L PMA for 4 hours) in 1 mL of immunoprecipitation buffer (10 mmol/L Tris [pH 7.5], 2 mmol/L MgCl2, 150 mmol/L KCl, 5% [vol/vol] glycerol, 2 mmol/L dithiothreitol, 0.5% [vol/vol] Brij58 [Pierce], and protease inhibitor cocktail [Roche]) for 30 minutes on ice. Protein-G Sepharose 4B FF beads (30 μL; Amersham) and 10 μg of anti-TTP antibody or control rabbit IgG were added and rotated overnight at 4°C. The beads were pelleted by centrifugation at 200g for 1 minute and washed 3 times with immunoprecipitation buffer. Total RNA was isolated from the immunoprecipitate by phenol/chloroform extraction followed by ethanol precipitation.
Transfection of siRNA and Plasmid DNA
Nonspecific and TTP-specific siRNA duplexes were purchased from Dharmacon Research. The pCMV6-TTP expression vector was purchased from Origene Technologies Inc. siRNAs and plasmids were transfected using Nucleofector (Amaxa Inc) according to the protocol of the manufacturer. For THP1 cells, 0.5 μg of siRNAs or plasmid DNA was transfected into 1x106 cells in 100 μL of Nucleofector Solution V (program U-01) and then incubated in standard THP1 medium containing 2 nmol/L PMA for the indicated times. For human monocytes, 0.5 μg of siRNAs or plasmid DNA was transfected into 1.5x106 cells in 100 μL of Human Monocyte Nucleofector Solution (program Y-01) and then incubated in Human Monocyte Nucleofector Medium (Amaxa) supplemented with 2 mmol/L glutamine and 10% FCS.
Statistical Analysis
The 6 SAGE libraries described in this report represent approximately 10 million base pairs of DNA sequencing. Because of the time and resource requirements of the SAGE technique, making multiple SAGE libraries for replication purposes is not feasible with the current technology. Nonetheless, in this study 2 different subjects were used for making the patient macrophage and monocyte libraries with good correlation, and the differential SAGE tags were highly reproducible as confirmed by RT-PCR. Data obtained by Ingenuity analysis software were exported into Excel spreadsheets and probability values for SAGE tag counts calculated accounting for library size differences as previously described.20
Data are expressed as mean±SD. Probability values were calculated with the use of Student’s t test.
Results
SAGE Libraries and Analyses
We created a total of 6 SAGE libraries, 4 from patients with atherosclerosis and 2 from control subjects without any clinical evidence of atherosclerosis (Table 1). From the same 2 patients (P1 and P2) undergoing CEA, paired monocyte and macrophage SAGE libraries (plaque macrophages, P1Mp and P2Mp; blood monocytes, P1Mc and P2Mc) were made to examine gene expression transformation during differentiation. As a screen to identify plaque macrophage enriched genes, we created a lung macrophage (L1Mp) library purified from the bronchoalveolar lavage of a normal subject (L1). A control monocyte (C1Mc) library from a normal age-matched control subject (C1) was also used for comparison (all SAGE libraries available at http://cgap.nci.nih.gov/SAGE).
Using the 464 486 SAGE tags sequenced in the 6 libraries, 14 233 genes were identified by matching them to the Unigene database.17 For the purposes of this study, we focused only on the tags identified by Unigene. SAGE tags representing monocyte and macrophage cell markers were appropriately enriched in the respective libraries (supplemental Figure II). Because transcriptional events are critical for differentiation and the SAGE technique accurately measures transcript levels, we focused on identifying genes involved in macrophage transcriptional regulation using the following scheme: (1) examination of genes classified as transcriptional regulators by global function and pathway analysis software (Ingenuity Systems); (2) selection of genes with high SAGE tag counts in duplicate libraries to increase reproducibility by RT-PCR; and (3) subselection of genes differentially expressed in inflamed plaque macrophages compared with quiescent normal lung macrophages.
Of the 14 233 genes matched to the Unigene database, 3989 were classified into functional families by the Ingenuity program, of which 548 belonged to the transcriptional regulator family (supplemental Tables II and III, respectively). The 548 genes were sorted by fold change (2-fold, plaque versus lung macrophages) and total tag count (12 tags combining all 6 libraries) for increased reproducibility. Applying these criteria resulted in a smaller group of 72 plaque macrophage transcriptional regulators (supplemental Table IV).
Plaque Macrophage Transcriptional Regulators
We selected the most abundantly expressed transcriptional regulators that were relatively specific to plaque macrophages (P<0.05; supplemental Table IV). In agreement with our previous observation, the protooncogene FOS, important in inflammation and myeloid cell differentiation, was among the top genes on the list (Table 2).6 As monocytes differentiated into macrophages, JUN, a FOS partner, which together comprise transcription activating protein-1 (AP-1), and their homologs JUNB and FOSB were coexpressed in plaque macrophages. Cyclin-dependent kinase inhibitor CDKN1A (p21), also important for atherosclerosis and with recently described transcriptional activity, was identified.21,22
A significant fraction of the differentially expressed genes were zinc finger proteins (Table 2 and supplemental Table IV). We observed high levels of early growth response 1 (EGR1), Kruppel-like transcription factors (KLF), and mRNA destabilizing Cys-Cys-Cys-His zinc finger protein 36 (ZFP36), also known as TTP.11,23,24 Of the Kruppel-like factors, KLF4 has been shown to be a proinflammatory factor in macrophages and, along with KLF6, associated with regulating p21.25–27 Although the prototypical mRNA target of TTP is the "AU-rich" element (ARE) of TNF mRNA, we observed similar sequence elements in a number of our macrophage candidate genes raising the possibility of their interaction with TTP.
Expression in Plaque Macrophages and in Circulating Monocytes
We first confirmed the SAGE tag fold changes by testing mRNA levels of the nine selected genes (Table 2) using normal lung macrophages and purified monocytes and plaque macrophage mRNAs from CEA patients (Figure 1A). The differential expression of the genes by RT-PCR corresponded well with the SAGE tag data. These results were also confirmed using PMA-activated THP1 cells for use as an in vitro model of macrophage differentiation (Figure 1B).
We have previously shown that FOS mRNA levels in monocytes may reflect their level of activation and atherosclerosis severity.6 In this study, we wondered whether there may be similar correlations between the levels of our macrophage genes and atherosclerosis. As an exploratory study, we examined purified monocytes from 14 patients and 14 control subjects with the highest and lowest levels of FOS, respectively.6 The patient and control subjects were closely matched except for risk factors inherently associated with atherosclerosis such as male gender, smoking, and prior history of coronary heart disease (Table 1). In support of an inflammatory model of atherosclerosis, the patient group showed an activated monocyte pattern compared with controls (Figure 1C). FOS mRNA levels between the 2 groups showed the most significant difference by probability value as expected based on their selection criteria. However, 3 other genes, FOSB, JUNB, and JUN, were significantly higher in patients compared with controls, P=0.003, P=0.005, and P=0.02, respectively.
TTP Interacts With a Subset of Transcriptional Regulators
Because many of our macrophage candidate genes contained ARE regions and are known to be posttranscriptionally regulated,28 we investigated whether TTP interacted with them in our activated THP1 monocytic cell model. We first coimmunoprecipitated the RNA–TTP complex from the cytosolic extracts of PMA-stimulated THP1 cells using anti-TTP antibody at an optimized potassium concentration of 150 mmol/L (data not shown). These were analyzed by Western blotting to ensure specific pull-down of the protein (Figure 2A). We next checked the immunoprecipitate for the presence of known TTP targets TNF and TTP mRNAs as internal positive controls and observed high levels of specific binding by RT-PCR (Figure 2B). Of our queried genes, FOS, EGR1, and KLF6 were significantly coimmunoprecipitated by TTP-specific antibody compared with control IgG (P<0.05 indicated by asterisks, Figure 2B). KLF4, p21, and JUNB only showed a trend toward TTP binding under our optimized in vitro conditions. FOSB and JUN did not show significant binding, although this could also be explained by their low transcript levels under these conditions (Figure 2B).
Inhibition of TTP by siRNA Increases Putative Target mRNAs
We next examined the functional significance of the interaction between the macrophage transcriptional regulators and TTP using both THP1 and primary human monocytes. We used siRNA to knockdown TTP in these cells and ascertained its decreased protein level by Western blotting, whereas GAPD served as protein loading control (Figure 3A and 3B). TTP siRNA–transfected cells had significantly higher levels of FOS and p21 mRNAs compared with nonspecific siRNA, demonstrating functional interaction between these transcripts and TTP (Figure 3). The relative fold-change increases in the macrophage transcriptional regulators by TTP inhibition were comparable to that observed for the positive control TNF mRNA. EGR1 and KLF6 mRNA levels were not as significantly affected, and FOSB and JUN were too low to be detected at 48 hours, when maximal effects were seen for the other factors (Figure 3, data not shown).
siRNA Inhibition of TTP Increases TNF Protein Release by Human Monocytes
We also investigated whether the increased TNF mRNA by TTP inhibition translated to augmented TNF protein release to functionally confirm TTP RNA interference in our cell system. By Western blotting, we verified decreased TTP protein levels over a 48-hour time period after specific siRNA transfection, both in THP1 cells stimulated with PMA and in cultured primary human monocytes (Figure 4A and 4B, respectively). During this same period, we observed relatively increased TNF protein release that corresponded to TTP protein knockdown compared with control in both cell types (Figure 4A and 4B). These results overtly fit with the expected results of blocking TNF mRNA destabilization, although the regulatory factors involved in TNF protein release are likely to be complex. Overall, the relative increases in released TNF protein were maintained over the time course of TTP RNA interference, validating our experimental system.
Increased Expression of TTP Attenuates Macrophage Transcriptional Regulators
To obtain additional evidence for the interaction between TTP and macrophage transcriptional regulators, we examined the effects of augmenting TTP expression on the mRNA levels of the macrophage genes. The transient expression of TTP increased its proteins levels and decreased the levels of TNF mRNA in both THP1 and primary human monocytes as positive control (Figure 5A and 5B). As predicted from the results of the siRNA experiments inhibiting TTP protein, transiently overexpressing it gave the expected opposite effect of decreasing the transcript levels of the macrophage genes (Figure 5). Similar to the TTP inhibition experiment (Figure 3), the levels of housekeeping gene GAPD were not affected by TTP alteration and served as negative control (Figure 5).
Discussion
We view this first report of in vivo purified human plaque macrophage SAGE database as an important resource for delineating the genetic program of these cells in atherosclerosis. Tissue culture–differentiated human macrophage SAGE libraries have been described,29 but these were poorly correlated to our libraries, highlighting the significance of cellular environment on gene expression. Our observation of circulating monocytes from patients expressing plaque macrophage transcriptional regulators supports a model of systemic inflammation in atherosclerosis and reinforces the utility of focusing on the in vivo transcriptomes of human samples. By comparing purified macrophages from atherosclerotic plaque to lung macrophages and normal monocytes, we have identified plaque-enriched transcriptional regulators. These include additional stress response genes of the AP-1/FOS pathway that previously had not been appreciated by comparing only monocyte expression profiles.6 Our demonstration of an antiinflammatory protein, TTP, functionally interacting with FOS, EGR1, and p21 corroborates our current analyses because these genes have been shown to be important in macrophage differentiation and atherosclerosis.21,23,30
The current analyses reveal networks of genes involved in the activation and inhibition of transcription for macrophage function that may result in complex interactions, such as gene expression oscillations for improved signaling.31 Interestingly, 2 of the 6 atherosclerosis marker genes we previously identified in patient monocytes, FOS and period homolog 1 (PER1), are thought to be important in circadian rhythm and autoregulatory feedback.6,32,33 We also observed an increased pattern of AP-1 gene expression in patients with atherosclerosis that is now further supported by the paired increases in the levels of FOS/JUNB and FOSB/JUN in monocyte and macrophage transcriptomes, respectively (Table 2). Although interaction of AP-1 elements are variable and complex, these expression patterns are likely to be important for macrophage differentiation.28,30,34 The interaction of TTP with FOS or other AP-1 partners may have important consequences, as they can regulate p21 and the differentiation of mononuclear cells into osteoclasts.35,36
Our study not only demonstrates the potential to reveal new insights into pathogenesis but also to clarify the mechanism of clinically important therapies. While screening for factors that may augment TTP levels, we observed that the HMG-CoA reductase inhibitor simvastatin caused an increase in monocyte TTP mRNA and protein levels (data not shown). However, because statins have been shown to augment lipopolysaccharide-induced TNF transcription by inhibiting FOS expression37 and to induce differentiation and apoptosis in human monocytes,38 the functional significance of increased TTP was difficult to interpret in this context. Nonetheless, the observation of TTP upregulation by statins merits further study as a potential mechanism by which this clinically efficacious medication may modulate inflammation.
Macrophages constitute a heterogeneous population of cells that may be subspecialized even within a specific tissue.39 Our plaque and lung macrophage transcriptomes express a markedly different profile of transcriptional regulators likely related to their specific functions. Insights obtained from these differences and from creating other macrophage transcriptomes (such as that of liver Kupffer macrophages, which have recently been shown to affect atherosclerosis in mice40) may yield more useful information. The current study serves as 1 perspective for interpreting our SAGE transcriptomes, but we anticipate that it will be a starting point for examining other genetic networks important in human macrophage function. Using similar analyses for other human diseases may also reveal new genetic insights with potential translational applications.
Acknowledgments
This research was supported by the Division of Intramural Research, National Heart, Lung, and Blood Institutes, the National Institutes of Health. We thank Toren Finkel, Michael N. Sack, and Neal Epstein for advice and critical reading of the manuscript and Perry J. Blackshear for helpful discussions regarding TTP.
Footnotes
Both authors contributed equally to this study.
Original received October 1, 2005; resubmission received February 13, 2006; revised resubmission received March 21, 2006; accepted March 30, 2006.
References
Glass CK, Witztum JL. Atherosclerosis. The road ahead. Cell. 2001; 104: 503–516. [Order article via Infotrieve]
Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med. 2005; 352: 1685–1695.
Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation. 2002; 105: 1135–1143.
Jain MK, Ridker PM. Anti-inflammatory effects of statins: clinical evidence and basic mechanisms. Nat Rev Drug Discov. 2005; 4: 977–987. [Order article via Infotrieve]
Patino WD, Mian OY, Kang JG, Matoba S, Bartlett LD, Holbrook B, Trout HH 3rd, Kozloff L, Hwang PM. Circulating transcriptome reveals markers of atherosclerosis. Proc Natl Acad Sci U S A. 2005; 102: 3423–3428.
Wang ZQ, Ovitt C, Grigoriadis AE, Mohle-Steinlein U, Ruther U, Wagner EF. Bone and haematopoietic defects in mice lacking c-fos. Nature. 1992; 360: 741–745. [Order article via Infotrieve]
Johnson RS, Spiegelman BM, Papaioannou V. Pleiotropic effects of a null mutation in the c-fos proto-oncogene. Cell. 1992; 71: 577–586. [Order article via Infotrieve]
Polyak K, Riggins GJ. Gene discovery using the serial analysis of gene expression technique: implications for cancer research. J Clin Oncol. 2001; 19: 2948–2958.
Saha S, Sparks AB, Rago C, Akmaev V, Wang CJ, Vogelstein B, Kinzler KW, Velculescu VE. Using the transcriptome to annotate the genome. Nat Biotechnol. 2002; 20: 508–512. [Order article via Infotrieve]
Blackshear PJ. Tristetraprolin and other CCCH tandem zinc-finger proteins in the regulation of mRNA turnover. Biochem Soc Trans. 2002; 30: 945–952. [Order article via Infotrieve]
Carballo E, Lai WS, Blackshear PJ. Feedback inhibition of macrophage tumor necrosis factor-alpha production by tristetraprolin. Science. 1998; 281: 1001–1005.
St Croix B, Rago C, Velculescu V, Traverso G, Romans KE, Montgomery E, Lal A, Riggins GJ, Lengauer C, Vogelstein B, Kinzler KW. Genes expressed in human tumor endothelium. Science. 2000; 289: 1197–1202.
Liu-Wu Y, Svenningsson A, Stemme S, Holm J, Wiklund O. Identification and analysis of macrophage-derived foam cells from human atherosclerotic lesions by using a "mock" FL3 channel in flow cytometry. Cytometry. 1997; 29: 155–164. [Order article via Infotrieve]
Holodniy M, Mole L, Yen-Lieberman B, Margolis D, Starkey C, Carroll R, Spahlinger T, Todd J, Jackson JB. Comparative stabilities of quantitative human immunodeficiency virus RNA in plasma from samples collected in VACUTAINER CPT, VACUTAINER PPT, and standard VACUTAINER tubes. J Clin Microbiol. 1995; 33: 1562–1566.
Seager Danciger J, Lutz M, Hama S, Cruz D, Castrillo A, Lazaro J, Phillips R, Premack B, Berliner J. Method for large scale isolation, culture and cryopreservation of human monocytes suitable for chemotaxis, cellular adhesion assays, macrophage and dendritic cell differentiation. J Immunol Methods. 2004; 288: 123–134. [Order article via Infotrieve]
Lash AE, Tolstoshev CM, Wagner L, Schuler GD, Strausberg RL, Riggins GJ, Altschul SF. SAGEmap: a public gene expression resource. Genome Res. 2000; 10: 1051–1060.
Cerutti JM, Delcelo R, Amadei MJ, Nakabashi C, Maciel RM, Peterson B, Shoemaker J, Riggins GJ. A preoperative diagnostic test that distinguishes benign from malignant thyroid carcinoma based on gene expression. J Clin Invest. 2004; 113: 1234–1242.
Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B. PUMA induces the rapid apoptosis of colorectal cancer cells. Mol Cell. 2001; 7: 673–682. [Order article via Infotrieve]
Audic S, Claverie JM. The significance of digital gene expression profiles. Genome Res. 1997; 7: 986–995.
Merched AJ, Chan L. Absence of p21Waf1/Cip1/Sdi1 modulates macrophage differentiation and inflammatory response and protects against atherosclerosis. Circulation. 2004; 110: 3830–3841.
Devgan V, Mammucari C, Millar SE, Brisken C, Dotto GP. p21WAF1/Cip1 is a negative transcriptional regulator of Wnt4 expression downstream of Notch1 activation. Genes Dev. 2005; 19: 1485–1495.
McCaffrey TA, Fu C, Du B, Eksinar S, Kent KC, Bush H Jr, Kreiger K, Rosengart T, Cybulsky MI, Silverman ES, Collins T. High-level expression of Egr-1 and Egr-1-inducible genes in mouse and human atherosclerosis. J Clin Invest. 2000; 105: 653–662.
Feinberg MW, Lin Z, Fisch S, Jain MK. An emerging role for Kruppel-like factors in vascular biology. Trends Cardiovasc Med. 2004; 14: 241–246. [Order article via Infotrieve]
Zhang W, Geiman DE, Shields JM, Dang DT, Mahatan CS, Kaestner KH, Biggs JR, Kraft AS, Yang VW. The gut-enriched Kruppel-like factor (Kruppel-like factor 4) mediates the transactivating effect of p53 on the p21WAF1/Cip1 promoter. J Biol Chem. 2000; 275: 18391–18398.
Narla G, Heath KE, Reeves HL, Li D, Giono LE, Kimmelman AC, Glucksman MJ, Narla J, Eng FJ, Chan AM, Ferrari AC, Martignetti JA, Friedman SL. KLF6, a candidate tumor suppressor gene mutated in prostate cancer. Science. 2001; 294: 2563–2566.
Feinberg MW, Cao Z, Wara AK, Lebedeva MA, Senbanerjee S, Jain MK. Kruppel-like factor 4 is a mediator of proinflammatory signaling in macrophages. J Biol Chem. 2005; 280: 38247–38258.
Hess J, Angel P, Schorpp-Kistner M. AP-1 subunits: quarrel and harmony among siblings. J Cell Sci. 2004; 117: 5965–5973.
Hashimoto S, Suzuki T, Dong HY, Yamazaki N, Matsushima K. Serial analysis of gene expression in human monocytes and macrophages. Blood. 1999; 94: 837–844.
Liebermann DA, Gregory B, Hoffman B. AP-1 (Fos/Jun) transcription factors in hematopoietic differentiation and apoptosis. Int J Oncol. 1998; 12: 685–700. [Order article via Infotrieve]
Lipan O, Wong WH. The use of oscillatory signals in the study of genetic networks. Proc Natl Acad Sci U S A. 2005; 102: 7063–7068.
Hirayama J, Cardone L, Doi M, Sassone-Corsi P. Common pathways in circadian and cell cycle clocks: light-dependent activation of Fos/AP-1 in zebrafish controls CRY-1a and WEE-1. Proc Natl Acad Sci U S A. 2005; 102: 10194–10199.
Brown SA, Ripperger J, Kadener S, Fleury-Olela F, Vilbois F, Rosbash M, Schibler U. PERIOD1-associated proteins modulate the negative limb of the mammalian circadian oscillator. Science. 2005; 308: 693–696.
Shaulian E, Karin M. AP-1 as a regulator of cell life and death. Nat Cell Biol. 2002; 4: E131–E136. [Order article via Infotrieve]
Kardassis D, Papakosta P, Pardali K, Moustakas A. c-Jun transactivates the promoter of the human p21(WAF1/Cip1) gene by acting as a superactivator of the ubiquitous transcription factor Sp1. J Biol Chem. 1999; 274: 29572–29581.
Kwak HB, Jin HM, Ha H, Kang MJ, Lee SB, Kim HH, Lee ZH. Tumor necrosis factor-alpha induces differentiation of human peripheral blood mononuclear cells into osteoclasts through the induction of p21(WAF1/Cip1). Biochem Biophys Res Commun. 2005; 330: 1080–1086. [Order article via Infotrieve]
Matsumoto M, Einhaus D, Gold ES, Aderem A. Simvastatin augments lipopolysaccharide-induced proinflammatory responses in macrophages by differential regulation of the c-Fos and c-Jun transcription factors. J Immunol. 2004; 172: 7377–7384.
Vamvakopoulos JE, Green C. HMG-CoA reductase inhibition aborts functional differentiation and triggers apoptosis in cultured primary human monocytes: a potential mechanism of statin-mediated vasculoprotection. BMC Cardiovasc Disord. 2003; 3: 6. [Order article via Infotrieve]
Ross JA, Auger MJ. The biology of the macrophage. In: Burke B, Lewis CE, eds. The Macrophage. 2nd ed. Oxford, UK: Oxford University Press; 2002: 1–57.
Bradshaw G, Gutierrez A, Miyake JH, Davis KR, Li AC, Glass CK, Curtiss LK, Davis RA. Facilitated replacement of Kupffer cells expressing a paraoxonase-1 transgene is essential for ameliorating atherosclerosis in mice. Proc Natl Acad Sci U S A. 2005; 102: 11029–11034.
您现在查看是摘要介绍页，详见ORG附件。