- Email: [email protected]
Global Expression Analysis of Extracellular Matrix–Integrin Interactions in Monocytes
Immunity, Vol. 13, 749–758, December, 2000, Copyright 2000 by Cell Press
Global Expression Analysis of Extracellular Matrix–Integrin Interactions in Monocytes Antonin R. de Fougerolles,*‡ Gloria Chi-Rosso,* Adriana Bajardi,* Philip Gotwals,* Cynthia D. Green,† and Victor E. Koteliansky* * Biogen Inc. 12 Cambridge Center Cambridge, Massachusetts 02142 † CuraGen Corporation 555 Long Wharf Drive New Haven, Connecticut 06511
Summary Central to immune and inflammatory responses are the integrin-mediated adhesive interactions of cells with their extracellular matrix (ECM)-rich environment. Using a comprehensive and quantitative mRNA profiling technique, we analyzed the effect of ECM-induced attachment on monocyte gene expression, its regulation by growth factors, and the integrin specificity of this event. Adhesion of cells to fibronectin resulted in increased expression of a large number of genes, which was strongly potentiated by the presence of growth factors. Adhesion activated both the NF-B and Jak/STAT pathways of gene transcription and increased expression of genes involved in inflammatory and immune responses, revealing the importance of ECM–integrin interactions in these processes. Introduction Macrophages are present in almost every tissue; the tissue macrophage pool is 500–1000 times larger than the bone marrow and blood pools (Cohn, 1983). This wide distribution within the peripheral tissues reflects the fact that these cells play an essential role in the disposal of foreign agents, in the initiation and mediation of immune and inflammatory responses, as well as in the repair process following tissue injury. Human blood monocytes undergo differentiation to macrophages upon migration into the extracellular matrix (ECM)-rich environment of extravascular tissue (Werb, 1987). Attachment of cells to ECM proteins, such as fibronectin, collagen, and laminin, is mediated predominantly by members of the integrin family of adhesion molecules (Hemler, 1990). Among the various ECM proteins, fibronectin plays a key role in promoting cell adhesion and various functions of monocytes and macrophages. Attachment of monocytes to fibronectin, which occurs primarily through the ␣41 and ␣51 integrin heterodimers, promotes cell migration, phagocytosis, and differentiation and modulates the secretion of inflammatory mediators (Kaplan and Gaudernack, 1982; Thorens et al., 1987; Juliano and Haskill, 1993; Wesley et al., 1998). Leukocyte proliferation and cytokine secretion can be ‡ To whom correspondence should be addressed (e-mail: tony_de_
[email protected]).
synergistically regulated by the combination of integrinmediated adhesion and soluble growth factors, demonstrating the intimate collaboration that exists between growth factors and integrins (Schwartz et al., 1995). We investigated on a global and unbiased basis the ability of ECM–integrin interactions to regulate monocyte gene expression using a recently published restriction enzyme–based method (GeneCalling) for identifying differentially expressed genes (Shimkets et al., 1999). GeneCalling couples expression profiling with a database query that utilizes fragment length and end sequence information to provide immediate feedback on expressed genes. As GeneCalling requires no a priori gene sequence, it is an open system that detects both known and novel genes. GeneCalling provides ⬎95% coverage of the expressed genome with a sensitivity of detection greater than 1:100,000 and can detect gene expression changes down to 1.5-fold differences (Shimkets et al., 1999). This approach has been used to identify genes linked to cardiac hypertrophy (Shimkets et al., 1999), obesity (Renz et al., 2000), angiogenesis (Kahn et al., 2000), and hematopoietic differentiation (Ohneda et al., 2000). Using this technology, we investigated the effect of ECM-induced adhesion on monocyte gene expression, its regulation by growth factors, and the integrin specificity of this event. The human monocytic cell line THP-1 is a well-studied model system for examining integrininduced gene expression in circulating leukocytes (Juliano and Haskill, 1993; Meredith et al., 1996). Adhesion of cells to fibronectin resulted in increased expression of a large number of genes, which was strongly potentiated by the presence of growth factors. We confirmed that ECM-induced attachment can result in activation of NF-B (Juliano and Haskill, 1993; Meredith et al., 1996) and demonstrated that the presence of growth factors can synergistically increase integrin-mediated activation of the NF-B pathway and lead to the activation of the Jak/STAT pathway. Comparison among different ECM molecules also revealed that the degree of integrin-specific gene expression is considerable. Integrin-mediated attachment resulted in increased expression of numerous inflammatory and immune response genes, revealing an important role for ECM–integrin interactions in affecting monocyte function and thus impacting on the development of pathologies. This is of particular relevance in the context of immune and inflammatory responses, where integrin-mediated adhesive interactions with the ECM-rich peripheral tissues are central to the localization of both resident and infiltrating monocytes at inflammatory sites. Results Attachment to Extracellular Matrix Is a Strong Regulator of Gene Expression To address the effect of ECM-induced adhesion on monocyte gene expression and its regulation by growth factors, THP-1 cells were either left in suspension or
Immunity 750
Table 1. Global Analysis of Monocyte Gene Expression Changes Mediated through Attachment to Extracellular Matrix Molecules
Type of Subtraction
Time (Hours)
Number of Fragments Detected (All Fragments)
Number of Fragments Differentially Expressed (Fragments 2-Fold)
Change in Gene Expression (%)
Type I collagen versus suspension Laminin versus suspension Fibronectin 120 kDa versus suspension Fibronectin versus suspension Fibronectin ⫹ serum versus suspension ⫹ serum
1 1 1 4 4
26 099 26 443 26 924 22 494 27 253
896 1009 733 281 1189
3.4 3.7 2.7 1.2 4.4
The indicated comparisons were made, and the magnitude of gene expression change is reflected by both the number and percentage of gene fragments that are differentially expressed.
plated on fibronectin for 4 hr in either the presence or absence of 2% fetal calf serum. To examine integrinspecific gene expression, THP-1 cells were left either in suspension or plated on a variety of different ECM proteins (RGD-containing 120 kDa fragment of fibronectin, laminin, collagen type I) for 1 hr in the absence of serum. Differential gene expression was quantified using GeneCalling (Shimkets et al., 1999). The comparisons analyzed in this study are listed in Table 1, with the total number of fragments detected and the number of differentially expressed fragments (⫾ 2-fold) listed. Attachment of cells for 4 hr to fibronectin under serum-containing conditions resulted in 1189 fragments (4.4%) being differentially expressed relative to suspension cells also cultured in the presence of serum (Table 1). The majority of the 1189 differentially expressed fragments are upregulated upon attachment to fibronectin, including over 150 fragments whose expression is increased by ⬎5-fold (Figure 1A). We positively identified the genes corresponding to 113 of the most highly induced fragments using competitive PCR. The 113 fragments represented a sixth of all upregulated fragments and segregated into 87 distinct clusters, representing 67 genes and 20 EST fragments (Tables 2 and 3). These genes and EST fragments represent a 5-fold increase in the number of matrix-induced genes identified to date (Juliano and Haskill, 1993; Fan et al., 1995; Meredith et al., 1996; Marra et al., 1997). The sensitivity and reproducibility of GeneCalling is highlighted by the fact that among the 67 known genes were 14 genes previously identified as being induced upon THP-1 attachment to fibronectin (Juliano and Haskill, 1993; Fan et al., 1995; Meredith et al., 1996; Marra et al., 1997). In addition, all the genes (IL-1␣, IL-8, TNF-␣, and tissue factor) previously identified in human peripheral blood monocytes as being induced by fibronectin attachment were also detected by GeneCalling (Haskill et al., 1988, 1991; Yurochko et al., 1992; Lin et al., 1994; Fan et al., 1995; McGilvray et al., 1997). Using independent experimental samples, real-time PCR results on over a dozen genes (MCP-1, MnSOD, IL-8, Gro-␣, MIP-1␣, IP10, CD44, IL-10 receptor, CD40, glutamate-cysteine ligase catalytic subunit, NF-B1, TNFAIP2, and p62 lck ligand) confirmed in all cases the direction and magnitude of the gene expression changes observed (data not shown). The list of known genes that are induced by attachment to fibronectin include secreted and cell surface molecules, metabolic proteins and intracellular regulatory proteins, and transcription factors (Table 2). Two-
thirds (46 of 67) of the identified genes induced by fibronectin attachment in the presence of serum are known to be linked to or regulated through either the NF-B (28 genes) or Jak/STAT (18 genes) pathways of gene transcription (Table 2) (Baeuerle and Henkel, 1994; Darnell, 1997). Included among the 28 NF-B-regulated genes are 14 genes previously identified as being induced by fibronectin attachment (A20 protein, Gro-␣, Gro-, IB, IL-1, IL-1ra, IL-8, JunB, MCP-1, MIP-1␣, NFB, superoxide dismutase, tissue factor, TNF␣) (Juliano and Haskill, 1993; Fan et al., 1995; Meredith et al., 1996; Marra et al., 1997). We have now discovered an additional 14 NF-B-regulated genes that can be induced by fibronectin attachment (EBI3, RANTES, cartilage gp39, IL-10 receptor, CD40, ferritin heavy chain, GLCLC, BCL3, ATF3, RelB, TNFAIP2, MyD88, TRAF-3, IRAK) (Table 2). The upregulation of numerous regulators of the NF-B activation cascade upon fibronectin binding (IRAK, MyD88, p62 lck ligand) may be suggestive of a role for these proteins in integrin-mediated activation of NF-B. Many genes linked to the Jak/STAT pathway are induced by fibronectin attachment (Tables 2 and 4). Among these genes, only STAT5A has previously been described as being regulated via integrins. Attachment to fibronectin has recently been shown to increase activity and expression of STAT5A in endothelial cells (Brizzi et al., 1999). In addition to STAT5A, we have now identified 17 other fibronectin-inducible genes whose expression is known to be regulated through the Jak/ STAT pathway (Table 2). Among these genes are many well-known interferon-inducible and Jak/STAT-regulated genes, such as spermidine/spermine N1-acetyltransferase, Leu-13 (9–27), IFI56, IFI41, and p59 oligoadenylate synthetase (Darnell, 1997). Potentiation of Matrix-Induced Gene Expression by Growth Factors Given the synergy that exists between growth factor receptor and integrin signaling, we sought to compare the pattern of ECM-induced gene expression seen in the presence of growth factors with that seen in its absence. In contrast to the 1189 fragments differentially expressed by fibronectin attachment in the presence of serum, attachment of cells for 4 hr to fibronectin under serum-free conditions resulted in only 281 fragments being differentially expressed as compared to serumfree suspension cells (Table 1). Similar to that seen previously under serum-containing conditions, attachment to fibronectin under serum-free conditions resulted in
Matrix-Induced Gene Expression in Monocytes 751
Figure 1. Distribution by Fold Change of Differentially Expressed Fragments upon Attachment to Extracellular Matrix Molecules (A–C) THP-1 cells attached to fibronectin (FN), collagen type I (CO I), laminin (LM), or fibronectin 120 kDa fragment (FN120) were compared to cells left in suspension, under conditions listed in Table 1. The number of differentially expressed fragments are grouped by fold change (solid bars, upregulated fragments; gray bars, downregulated fragments). (D) Photomicrograph of THP-1 cells after 1 hr attachment in the absence of serum to various extracellular matrix molecules. Cells attached to either collagen type I (CO I), laminin (LM), or fibronectin 120 kDa fragment (FN120) show similar degrees of attachment and cell spreading on the different substrates; cells in suspension are also shown for comparative purposes.
over 70% of the differentially expressed fragments being upregulated (Figure 1B). Comparison of the fragments differentially regulated by fibronectin under serum-free and serum-containing conditions revealed overlap between the two (Table 3). Out of the 281 fragments regulated by fibronectin attachment in the absence of serum, 79 were also regulated by fibronectin in the presence of serum, and 202 fragments were regulated by fibronectin only in the absence of serum. Synergy between ECM and growth factors was revealed by the large number of fragments (1110) whose fibronectin-specific regulation required the presence of serum. This synergy was also evident in the increased induction of genes attached to fibronectin in the presence versus the absence of serum (Table 3). Matrix-Induced Activation of NF-B and Jak/STAT Pathways and Regulation by Growth Factors Analysis of gene expression patterns revealed that attachment of cells to fibronectin can activate genes regulated by the NF-B transcription pathway independent of growth factors, while ECM-induced activation of the Jak/STAT-regulated genes required the presence of
growth factors (Tables 3 and 4). To extend these results, luciferase reporter, nuclear translocation, and phosphorylation assays were performed on THP-1 cells under adherent and nonadherent conditions in the presence or absence of growth factors. Fibronectin attachment activated NF-B-mediated nuclear translocation and transcription (Figures 2A and 2B). Consistent with the required presence of growth factors in ECM-mediated Jak/STAT pathway activation, attachment of cells to fibronectin in the presence of serum resulted in activation of the Jak/STAT pathway as assessed by phosphorylation of STAT1; no such activation was seen upon fibronectin attachment in the absence of serum (Figure 2C). Effect of Different Integrin–Matrix Interactions on Gene Expression To investigate the effect of different integrin ligands on gene expression, THP-1 cells were left in suspension or adhered for 1 hr on plates coated with ECM proteins specific for ␣21 (type I collagen), ␣51 (RGD-containing 120 kDa fragment of fibronectin), and ␣61 (laminin) integrins (Figure 1D). Inhibition of adhesion with blocking mAbs to the relevant ␣ subunits confirmed the
Immunity 752
Table 2. THP-1 Cell Attachment to Fibronectin Activates the NF-B and Interferon Pathways of Gene Induction
The names of the differentially expressed genes are given with their GenBank accession number, the number of fragments that were identified, and the average fold induction of the gene upon 4 hr attachment to fibronectin versus suspension (done in the presence of 2% serum). Genes linked to the NF-B pathway are shown in green, and those linked to the Jak/STAT pathway are shown in red.
specificity of the integrin–ECM interactions (data not shown). Compared to cells left in suspension, attachment of THP-1 cells to various ECM proteins results in the change of a large number of genes (700–1000 fragments, representing 2.7%–3.7% of the expressed genome) (Table 1). In all cases, integrin engagement resulted in roughly 80% of differentially expressed fragments showing an increase in mRNA expression (Figure 1C). To investigate the importance of integrin specificity in regulating gene expression, we compared the distribution of differentially regulated fragments induced by specific integrin–ECM interactions (Table 5). Among the differentially expressed fragments, roughly half (414 fragments) are common to all three matrix molecules. A strong degree of integrin specificity exists, as most of the rest of the differentially expressed fragments are regulated only through specific integrin–ECM interactions (collagen, 316 fragments; laminin, 416 fragments; 120 kDa fibronectin, 262 fragments) (Table 5). The one notable exception to this is the 144 differentially expressed fragments that are shared between collagen type I and laminin but are not regulated by attachment to the 120 kDa fragment of fibronectin. Gene identification both confirmed genes that had previously been shown to be inducible through attachment to multiple extracellular matrices (IL-1, IL-8, Gro, A20 protein) (Meredith
et al., 1996) and resulted in the discovery of another 37 such genes (Table 5). Many of the genes commonly regulated through engagement of ␣21, ␣51, and ␣61 integrins are genes whose expression is induced through activation of NF-B (Table 5). Discussion Global gene expression analysis has enabled us to quantitate the relative contribution of integrin-mediated cell attachment to changes in gene expression. Our results demonstrate the effect integrin engagement has on gene expression and highlight the ability of growth factors and different matrix molecules to influence integrin-mediated gene expression. The positive identification of over 140 genes and ESTs whose expression in monocytes is induced by attachment to ECM not only represents a significant increase in the number of genes previously identified but also reveals the importance of ECM–integrin interaction in multiple aspects of monocyte biology. Monocyte extravasation and attachment in ECM-rich tissues that occurs during the course of immune and inflammatory responses, and mimicked here by attachment of cells to fibronectin, revealed many genes important in monocyte migration and activation. Monocyte
Matrix-Induced Gene Expression in Monocytes 753
Table 3. The Effect of Serum on Fibronectin-Induced Gene Induction in THP-1 Cells
Differentially expressed genes are given with their GenBank accession number and fold gene induction upon 4 hr attachment to fibronection versus suspension in either the presence or absence of 2% serum. ND, not determined.
attachment to the ECM results in the upregulation of numerous chemokines (IL-8, Gro-␣, Gro-, MCP-1, MIP1␣, RANTES, IP10) and cell adhesion molecules (CD18, CD44) important in attracting other cells to inflammatory sites (Springer, 1994; Baggiolini et al., 1997). Genes important in monocyte activation and effector function are also upregulated by fibronectin attachment and include secreted molecules (IL-1, TNF␣) and cell surface molecules (CD40, CD44) (Webb et al., 1990; Grewal and Flavell, 1998). Fibronectin attachment can also affect macrophage activation and function through increased expression of intracellular molecules (hck, MyD88) that are directly involved in macrophage activation and cyto-
kine production (English et al., 1993; Kawai et al., 1999). Matrix-induced attachment also changes expression of genes (pleckstrin, calpain) that, through alterations to the cytoskeleton, are important in the regulation of monocyte cell spreading, migration, and phagocytosis (Brumell et al., 1999; Kulkarni et al., 1999). Along with induction of numerous genes that increase the cytolytic and cytotoxic effect of macrophages, matrix-induced attachment also results in the induction of genes responsible for protecting the cells from self-lysis. NF-B activation initiates a response that blunts the apoptotic action of several agonists of cell death (Beg and Baltimore, 1996; Van Antwerp et al., 1996). Among
Immunity 754
Table 4. Effect of Growth Factors on Integrin-Mediated Activation of the NF-B and Jak-STAT Pathways of Gene Transcription Fold Induction Encoded Protein NF-B pathway NF-B1 p105/p50 MCP-1 TNFAIP2 (B94) IL-1 CD40 Jak-Stat pathway STAT5A IFI56 ISG15 Spermine/spermidine N1acetyltransferase p59 oligoadenylate synthetase PIM-1 Leu-13 (9-27)
FN (⫺ Serum) FN (⫹ Serum) 2.0 4.2 2.9 2.6 1.6
6.0 38.5 20.0 7.8 9.7
0 0 0 0
17.0 38.0 11.8 8.7
0 0 0
5.9 5.2 3.4
The names of the differentially expressed genes are given with the fold gene induction upon 4 hr attachment to fibronectin versus suspension in either the presence or absence of 2% serum.
the genes induced by fibronectin attachment are several NF-B-regulated genes (MnSOD, A20, Bcl-3) that have been shown to have antiapoptotic function (Fridovich, 1989; Opipari et al., 1992; Rebollo et al., 2000). Other matrix-regulated genes not linked to NF-B activation may also play a role in preventing apoptosis. For instance, enforced expression of PIM-1 kinase enhances growth factor–independent survival and inhibits apoptosis in murine myeloid cells (Lilly and Kraft, 1997). Attachment of monocytes to ECM has been demonstrated to induce differentiation to macrophages (Kaplan and Gaudernack, 1982; Wesley et al., 1998). We found that attachment to fibronectin resulted in the upregulated expression of an array of genes previously associated with differentiation of monocytes to macrophages, including CD44, ferritin, spermidine/spermine acetyltransferase (SSAT), cartilage gp39, and CD82 (Krause et al., 1996; Gingras and Margolin, 2000). Fibronectin-mediated attachment also resulted in increased expression of several key transcriptional regulators of macrophage differentiation, including PU.1, STAT5A, and IRF-1. PU.1 is essential for macrophage development because it activates the expression of the M-CSF receptor and transactivates other genes, such as the adhesion molecules CD11b and CD18, the LPS receptor (CD14), and Fc and scavenger receptors, involved in the acquisition of a functional phenotype (Valledor et al., 1998). STAT5A is induced during monocyte differentiation, and its antiapoptotic activity is essential during terminal stages of myeloid differentiation (Kieslinger et al., 2000). Expression of IRF-1 activates the transcription of interferon, which acts to promote inhibition of monocyte proliferation and can also protect cells from apoptosis (Xaus et al., 1999). Thus, matrix-induced adhesion results in the expression of numerous markers associated with monocyte differentiation and induces expression of transcription factors that can regulate multiple aspects of myelopoiesis. Identification of matrix-induced genes may also help
Figure 2. The Effect of Growth Factors on Fibronectin-Mediated Activation of NF-B and STAT1 Fibronectin-mediated activation of NF-B occurs irrespective of the presence of growth factors, while activation of STAT1 requires the presence of growth factors. (A) Induction of NF-B-mediated transcription. THP-1 cells were transfected with a B luciferase reporter construct (2 ⫻ B sites) and adhered to plates containing immobilized fibronectin for 2 and 4 hr in either the presence of serum-free media or media containing 2% fetal calf serum. Cells were collected and luciferase activity was measured. Results are normalized to luciferase activity seen in transfected cells left in suspension. (B) Induction of NF-B translocation into the nucleus. Coverslips were plated with either fibronectin or poly-lysine control substrate. THP-1 cells were attached for 4 hr in serum-free media or media containing 2% fetal calf serum, washed, and fixed in paraformaldehyde. Fixed cells were then permeabilized using 0.5% Triton X-100 and stained for NF-B p65 using a rabbit anti-NF-B p65 antisera followed by a FITC-donkey anti-rabbit secondary reagent. Slides were mounted and NF-B staining was visualized under fluorescent microscopy. (C) Effect of growth factors on fibronectin-mediated activation of STAT1. THP-1 cells were starved for 72 hr and adhered to plates containing immobilized fibronectin for the indicated times in either the presence of serum-free media or media containing 2% fetal calf serum. Bound cells were lysed and cell extracts were subjected to 8% SDS-PAGE. Proteins were electrophoretically transferred to nitrocellulose filter and immunoblotted with either a rabbit antiphosphoSTAT1 (Tyr 701) antibody to detect activated STAT1 or a rabbit anti-STAT1 polyclonal antisera to detect total STAT1.
Matrix-Induced Gene Expression in Monocytes 755
Table 5. Comparison of Gene Regulation through Attachment to Different Integrin Ligands
Venn diagram illustration of the number of differentially expressed fragments (compared to cells in suspension) induced upon 1 hr attachment to either type I collagen, laminin, or 120 kDa fragment of fibronectin. Differentially expressed genes are given with their GenBank accession number and the fold change in gene expression upon 1 hr attachment to different extracellular matrix proteins (FN120, fibronectin 120 kDa fragment; LM, laminin; and COL, collagen type I). Genes linked to the NF-B pathway are shown in green. Colorimetric scale indicating fold change in gene expression is as shown in Table 3.
in understanding the role of monocytes in disease. Both attachment to extracellular matrix and monocyte activation/differentiation are hypothesized to play an important role in the development of atherosclerosis (Ross, 1999). Atherosclerotic plaque formation is characterized by the accumulation of lipid droplets in macrophagederived foam cells. Fatty acid synthesis requires a source of acetyl CoA and NADPH, and among the genes induced by attachment to fibronectin are several enzymes (phosphogluconate dehydrogenase, kynureninase, LACS1) that promote production of both acetyl CoA and NADPH (Hillgartner et al., 1995; Heyes et al., 1997). In addition, LACS1 is also a key protein that promotes intracellular fatty acid retention (Schaffer and Lodish, 1994). Consistent with this hypothesis is the finding that attachment of peripheral blood monocytes to a collagen matrix not only induces cellular differentiation but also increases intracellular lipid accumulation (Wesley et al., 1998). Lastly, expression of matrix-regulated genes, such as ferritin and adipophilin, was found to be induced in macrophages in early human atherosclerotic plaques, and their high level of expression is associated with the development of plaques and disease progression (Pang et al., 1996; Wang et al., 1999).
Thus, monocyte adherence to ECM proteins, such as fibronectin and collagen, may contribute to the development of atherosclerosis not only through increased expression of numerous cytokines and chemokines already implicated in the development of atherosclerosis, but also through increased expression of numerous metabolic enzymes and proteins that play a key role in lipid metabolism. The identification of adherence-induced genes allowed us to further dissect integrin-mediated signal transduction pathways in monocytes. Using a global approach, we have demonstrated that the majority of matrix-induced genes are ones known to be under the control of the NF-B and Jak/STAT transcriptional pathways. We have confirmed that ECM-induced attachment can result in activation of NF-B (Juliano and Haskill, 1993; Meredith et al., 1996) and have demonstrated that the presence of growth factors can synergistically increase integrin-mediated activation of the NF-B pathway and lead to the activation of the Jak/STAT pathway. Comparison of cell attachment to different ECM molecules allowed us to also investigate the relative importance of integrin specificity in affecting gene expression. The ability of different integrin–ECM interactions to regu-
Immunity 756
late a large number of the same genes is not surprising, given that changes in gene expression are not the consequence of specific integrin engagement alone, but may also require adhesion, cell shape change, and cytoskeletal rearrangement (Juliano and Haskill, 1993). While the finding that distinct integrin engagement can result in different patterns of gene expression has been documented (Huhtala et al., 1995), we have determined, using an unbiased global approach, that the degree of integrin-specific gene expression is considerable. The pattern of gene expression induced upon attachment to extracellular matrix molecules in monocytes does show some similarities to that seen in other cell types. While the lack of comprehensive gene profiling for matrix-induced genes on other cell types makes proper comparisons between cell types impossible, a degree of commonality exists between cell types. Attachment of a variety of cell types (endothelial, smooth muscle, and fibroblast) to fibronectin has been shown to activate the NF-B pathway of gene transcription and to induce some of the same genes (IL-1␣, IL-8, MCP-1, STAT5A) seen in monocytes (Qwarnstrom et al., 1994; Marra et al., 1997; Scatena et al., 1998; Brizzi et al., 1999; GarciaVelasco and Arici, 1999). While a certain degree of commonality exists across cell types with regards to integrin-mediated gene expression, many of the genes induced in monocytes by matrix-induced attachment play specific and important roles in this cell type. Identification and analysis of these genes reveals an important role for ECM–integrin interactions in affecting monocyte function and thus impacting on the development of pathologies, such as atherosclerosis. This is of particular relevance in the context of inflammation and the localization of both resident and infiltrating monocytes at an inflammatory site. The importance of integrin interaction with the ECMrich environment of peripheral tissues was recently demonstrated in numerous in vivo models of inflammation, including asthma, hypersensitivity, and arthritis (Henderson et al., 1997; de Fougerolles et al., 2000). Experimental Procedures Adhesion Assays Adhesion assays were essentially carried out as described previously (Gotwals et al., 1996). Human plasma fibronectin and its 120 kDa human ␣-chymotryptic fragment were purchased from GIBCO– BRL (Rockville, MD); rat tail type I collagen was purchased from Collaborative Research Inc. (Bedford, MA); basement membrane laminin was purchased from Sigma (St. Louis, MO). Tissue culture plates (100 mm) (Becton Dickinson, Franklin Lakes, NJ) were coated overnight at 4⬚C with 5 ml of adhesive substrates (10 g/ml) diluted in PBS. Plates were washed twice with PBS, and nonspecific binding sites were blocked with 1% BSA in PBS for 1 hr at room temperature. Human monocytic THP-1 cells were washed into either serum-free binding buffer (RPMI 1640 and 0.25% BSA) or binding buffer containing 2% fetal calf serum (JRH Biosciences, Lenexa, KS). Cells (1.6 ⫻ 106/ml) were then either left in suspension or adhered to extracellular matrix–immobilized plates (5 ml/plate) for 1 to 4 hr at 37⬚C. Unbound cells were removed by washing the plates twice with PBS. Suspension or extracellular matrix–adhered cells (107 cells minimum) were lysed in Trizol (Life Technologies, Grand Island, NY) and RNA isolated as described (Shimkets et al., 1999). All suspension and extracellular matrix–adhered (fibronectin, fibronectin 120 kDa fragment, type I collagen, and laminin) samples were done in triplicate and processed independently.
Differential Gene Expression Analysis GeneCalling reactions were performed essentially as described (Shimkets et al., 1999). Briefly, following double-stranded cDNA synthesis of poly(A)⫹ RNA, cDNA was separated into 96 different pools and each digested with a different pair of restriction enzymes. The resulting fragments were ligated with complementary adapters, one labeled with biotin and the other with fluorescamine (FAM), and amplified by PCR. After affinity purification on streptavidin and separation by polyacrylamide gel electrophoresis, the FAM-labeled fragments were detected by laser excitation. A composite restriction fragment profile is generated for each sample, based on average peak height and variance of nine separate restriction enzyme digestions (each sample triplicate is subjected to three separate restriction enzyme digestions). The restriction fragment profiles of two samples are compared and differentially expressed fragments are identified. Linkage of a differentially expressed cDNA fragment to a gene is made through knowledge of the restriction enzymes used to generate the fragment ends and the length of the fragment itself. A species-specific query of databases reveals genes that fit these criteria, with gene identification confirmed by competitive PCR using gene-specific oligonucleotides.
NF-B Luciferase and Translocation Assays Luciferase assays on THP-1 cells were performed as described (Rosales and Juliano, 1996). THP-1 cells were transfected by DEAEdextran method with a B luciferase reporter construct. The B luciferase reporter construct contains 2 ⫻ B sites derived from the MHC promoter fused to luciferase and was a gift from Dr. Ilana Stancovski (California Institute of Technology, Pasadena, CA). Transfected cells were resuspended in either serum-free binding buffer or binding buffer containing 2% fetal calf serum and either left in suspension or adhered to plates containing immobilized fibronectin for 2 and 4 hr at 37⬚C as outlined above. Bound cells were collected and luciferase activity was measured using a luciferase assay system (Promega Corp., Madison, WI) according to the manufacturer’s instructions. Results are normalized to luciferase activity seen in transfected cells left in suspension. Nuclear translocation of NF-B was performed as described previously (Scatena et al., 1998). Briefly, THP-1 cells were resuspended in either serum-free binding buffer or binding buffer containing 2% fetal calf serum and immediately plated onto glass coverslips coated with 10 g/ml fibronectin (GIBCO–BRL) or 10 g/ml poly-lysine (Sigma) for 4 hr at 37⬚C. Cells were washed and bound cells fixed for 5 min in 3% paraformaldehyde in PBS (pH 7.4) containing 2% sucrose. Cells were then permeabilized with HEPES–Triton X-100 buffer (20 mM HEPES [pH 7.4], 300 mM sucrose, 50 mM NaCl, 3 mM MgCl2, and 0.5% Triton X-100) and stained for NF-B p65 using a rabbit anti-NF-B p65 polyclonal antisera followed by a FITCdonkey anti-rabbit IgG secondary reagent (Santa Cruz Biotechnology, Santa Cruz, CA). Slides were mounted and NF-B staining was visualized under confocal microscopy (Leica, Wetzlar, Germany).
STAT1 Phosphorylation Assays THP-1 cells were serum starved for 72 hr and adhered to plates containing immobilized fibronectin for the indicated times in either the presence of serum-free binding buffer or binding buffer containing 2% fetal calf serum as outlined above. Unbound cells were removed by washing the plates twice with PBS, and bound cells were lysed in ice-cold lysis buffer containing 1% Triton X-100 (v/v) in 20 mM HEPES (pH 7.8), 150 mM NaCl, 0.1% SDS, 5 mM MgCl2, 1 mM Na4VO3, 50 mM NaF, and complete mini-protease inhibitors (Roche, Mannheim, Germany). Lysates were incubated for 30 min on ice before centrifugation at 15,000 ⫻ g for 15 min at 4⬚C. Total protein concentration of postnuclear lysates was determined by BCA (Pierce, Rockford, IL). Equal amounts of cell lysates were subjected to 8% SDS-PAGE and transferred to nitrocellulose (MSI, Westborough, MA) for immunoblotting. The membranes were blocked with 5% milk in TBS (10 mM Tris-HCl [pH 7.6], 150 mM NaCl) and incubated with either a rabbit anti-phosphoSTAT1 (Tyr 701) antibody or a rabbit anti-STAT1 polyclonal antisera (New England Biolabs, Beverly, MA). Bound antibody was revealed with horseradish peroxidase (HRP)-conjugated anti-rabbit antibody us-
Matrix-Induced Gene Expression in Monocytes 757
ing enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech, Piscataway, NJ). Acknowledgments We would like to thank Andro Hsu and Dr. John Peyman for assistance with gene identification, Akos Szilvasi for assistance with the NF-B translocation assays, and Drs. Joe Rosa and Roy Lobb for support and helpful discussions. Received August 17, 2000; revised October 4, 2000. References Baeuerle, P.A., and Henkel, T. (1994). Function and activation of NFB in the immune system. Annu. Rev. Immunol. 12, 141–179. Baggiolini, M., Dewald, B., and Moser, B. (1997). Human chemokines: an update. Annu. Rev. Immunol. 15, 675–705. Beg, A.A., and Baltimore, D. (1996). An essential role for NF-B in preventing TNF-␣-induced cell death. Science 274, 782–784. Brizzi, M.F., Defilippi, P., Rosso, A., Venturino, M., Garbarino, G., Miyajima, A., Silengo, L., Tarone, G., and Pegoraro, L. (1999). Integrin-mediated adhesion of endothelial cells induces JAK2 and STAT5A activation: role in the control of c-fos gene expression. Mol. Biol. Cell 10, 3463–3471. Brumell, J.H., Howard, J.C., Craig, K., Grinstein, S., Schreiber, A.D., and Tyers, M. (1999). Expression of the protein kinase C substrate pleckstrin in macrophages: association with phagosomal membranes. J. Immunol. 163, 3388–3395. Cohn, Z.A. (1983). The macrophage—versatile element of inflammation. Harvey Lect. 77, 63–80. Darnell, J.E. (1997). STATs and gene regulation. Science 277, 1630– 1635. De Fougerolles, A.R., Sprague, A.G., Nickerson-Nutter, C.L., ChiRosso, G., Rennert, P.D., Gardner, H., Gotwals, P.J., Lobb, R.R., and Koteliansky, V.E. (2000). Regulation of inflammation by collagenbinding integrins ␣11 and ␣21 in models of hypersensitivity and arthritis. J. Clin. Invest. 105, 721–729. English, B.K., Ihle, J.N., Myracle, A., and Yi, T. (1993). Hck tyrosine kinase activity modulates tumor necrosis factor production by murine macrophages. J. Exp. Med. 178, 1017–1022. Fan, S., Mackman, N., Cui, M., and Edgington, T.S. (1995). Integrin regulation of an inflammatory effector gene: direct induction of the tissue factor promoter by engagement of 1 or ␣4 integrin chains. J. Immunol. 154, 3266–3274. Fridovich, I. (1989). Superoxide dismutases: an adaptation to a paramagnetic gas. J. Biol. Chem. 264, 7761–7764. Garcia-Velasco, J.A., and Arici, A. (1999). Interleukin-8 expression in endometrial stromal cells is regulated by integrin-dependent cell adhesion. Mol. Hum. Reprod. 5, 1135–1140. Gingras, M.C., and Margolin, J.F. (2000). Differential expression of multiple unexpected genes during U937 cell and macrophage differentiation detected by suppressive subtractive hybridization. Exp. Hematol. 28, 65–76. Gotwals, P.J., Chi-Rosso, G., Lindner, V., Yang, J., Ling, L., Fawell, S.E., and Koteliansky, V.E. (1996). The a1b1 integrin is expressed during neointima formation in rat arteries and mediates collagen matrix reorganization. J. Clin. Invest. 97, 2469–2477. Grewal, I.S., and Flavell, R.A. (1998). CD40 and CD154 in cell-mediated immunity. Annu. Rev. Immunol. 16, 111–135. Haskill, S., Johnson, C., Eierman, D., Becker, S., and Warren, K. (1988). Adherence induces selective mRNA expression of monocyte mediators and proto-oncogenes. J. Immunol. 140, 1690–1694. Haskill, S., Beg, A.A., Tompkins, S.M., Morris, J.S., Yurochko, A.D., Sampson-Johannes, A., Mondal, K., Ralph, P., and Baldwin, A.S., Jr. (1991). Characterization of an immediate-early gene induced in adherent monocytes that encodes IB-like activity. Cell 65, 1281– 1289. Hemler, M.E. (1990). VLA proteins in the integrin family: structures,
functions, and their role on leukocytes. Annu. Rev. Immunol. 8, 365–400. Henderson, W.R., Chi, E.Y., Albert, R.K., Chu, S.J., Lamm, W.J., Rochon, Y., Jonas, M., Christie, P.E., and Harlan, J.M. (1997). Blockade of CD49d (alpha4 integrin) on intrapulmonary but not circulating leukocytes inhibits airway inflammation and hyperresponsiveness in a mouse model of asthma. J. Clin. Invest. 100, 3083–3092. Heyes, M.P., Chen, C.Y., Major, E.O., and Saito, K. (1997). Different kynurenine pathway enzymes limit quinolinic acid formation by various human cell types. Biochem. J. 326, 351–356. Hillgartner, F.B., Salati, L.M., and Goodridge, A.G. (1995). Physiological and molecular mechanisms involved in nutritional regulation of fatty acid synthesis. Physiol. Rev. 75, 47–76. Huhtala, P., Humphries, M.J., McCarthy, J.B., Tremble, P.M., Werb, Z., and Damsky, C.H. (1995). Cooperative signaling by alpha 5 beta 1 and alpha 4 beta 1 integrins regulates metalloproteinase gene expression in fibroblasts adhering to fibronectin. J. Cell Biol. 129, 867–879. Juliano, R.L., and Haskill, S. (1993). Signal transduction from the extracellular matrix. J. Cell Biol. 120, 577–585. Kahn, J., Mehraban, F., Ingle, G., Xin, X., Bryant, J.E., Vehar, G., Schoenfeld, J., Grimaldi, C.J., Peale, F., Draksharapu, A., et al. (2000). Gene expression profiling in an in vitro model of angiogenesis. Am. J. Pathol. 156, 1887–1900. Kaplan, G., and Gaudernack, G. (1982). In vitro differentiation of human monocytes. Differences in monocyte phenotypes induced by cultivation on glass or on collagen. J. Exp. Med. 156, 1101–1114. Kawai, T., Adachi, O., Ogawa, T., Takeda, K., and Akira, S. (1999). Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity 11, 115–122. Kieslinger, M., Woldman, I., Moriggl, R., Hofmann, J., Marine, J., Ihle, J.N., Beug, H., and Decker, T. (2000). Antiapoptotic activity of Stat5 required during terminal stages of myeloid differentiation. Genes Dev. 14, 232–244. Krause, S.W., Rehli, M., Kreutz, M., Schwarzfischer, L., Paulauskis, J.D., and Andreesen, R. (1996). Differential screening identifies genetic markers of monocytes to macrophage maturation. J. Leukoc. Biol. 60, 540–545. Kulkarni, S., Saido, T.C., Suzuki, K., and Fox, J.E.B. (1999). Calpain mediates integrin-induced signaling at a point upstream of rho family members. J. Biol. Chem. 274, 21265–21275. Lilly, M., and Kraft, A. (1997). Enforced expression of the Mr 33,000 Pim-1 kinase enhances factor-independent survival and inhibits apoptosis in murine myeloid cells. Cancer Res. 57, 5348–5355. Lin, T.H., Yurochko, A., Kornberg, L., Morris, J., Walker, J.J., Haskill, S., and Juliano, R.L. (1994). The role of protein tyrosine phosphorylation in integrin-mediated gene induction in monocytes. J. Cell Biol. 126, 1585–1593. Marra, F., Pastacaldi, S., Romanelli, R.G., Pinzani, M., Ticali, P., Carloni, V., Laffi, G., and Gentilini, P. (1997). Integrin-mediated stimulation of monocyte chemotactic protein-1 expression. FEBS Lett. 414, 221–225. McGilvray, I.D., Lu, Z., Bitar, R., Dackiw, A.P.B., Davreux, C.J., and Rotstein, O.D. (1997). VLA-4 integrin cross-linking on human monocytic THP-1 cells induces tissue factor expression by a mechanism involving mitogen-activated protein kinase. J. Biol. Chem. 272, 10287–10294. Meredith, J.E., Winitz, S., Lewis, J.M., Hess, S., Ren, X., Renshaw, M.W., and Schwartz, M.A. (1996). The regulation of growth and intracellular signaling by integrins. Endocr. Rev. 17, 207–220. Ohneda, O., Ohneda, K., Nomiyama, H., Zheng, Z., Gold, S.A., Arai, F., Miyamoto, T., Taillon, B.E., McIndoe, R.A., Shimkets, R.A., et al. (2000). WECHE: a novel hematopoietic regulatory factor. Immunity 12, 141–150. Opipari, A.W., Hu, H.M., Yabkowitz, R., and Dixit, V.M. (1992). The A20 zinc finger protein protects cells from tumor necrosis factor cytotoxicity. J. Biol. Chem. 267, 12424–12427. Pang, J.H., Jiang, M.J., Chen, Y.L., Wang, F.W., Wang, D.L., Chu,
Immunity 758
S.H., and Chau, L.Y. (1996). Increased ferritin gene expression in atherosclerotic lesions. J. Clin. Invest. 97, 2204–2212. Qwarnstrom, E.E., Ostberg, C.O., Turk, G.L., Richardson, A., and Bomsztyk, K. (1994). Fibronectin attachment activates the NF-kB p50/p65 heterodimer in fibroblasts and smooth muscle cells. J. Biol. Chem. 269, 30765–30768. Rebollo, A., Dumoutier, L., Renauld, J., Zaballos, A., Ayllon, V., and Carlos-Martinez, A. (2000). Bcl-3 expression promotes cell survival following interleukin-4 deprivation and is controlled by AP1 and AP1-like transcription factors. Mol. Cell. Biol. 20, 3407–3416. Renz, M., Tomlinson, E., Hultgren, B., Levin, N., Gu, Q., Shimkets, R.A., Lewin, D.A., and Stewart, T.A. (2000). Quantitative expression analysis of genes regulated by both obesity and leptin reveals a regulatory loop between leptin and pituitary-derived ACTH. J. Biol. Chem. 275, 10429–10436. Rosales, C., and Juliano, R. (1996). Integrin signaling to NF-B in monocytic leukemia cells is blocked by activated oncogenes. Cancer Res. 56, 2302–2305. Ross, R. (1999). Atherosclerosis—an inflammatory disease. N. Engl. J. Med. 340, 115–126. Scatena, M., Almeida, M., Chaisson, M.L., Fausto, N., Nicosia, R.F., and Giachelli, C.M. (1998). NF-B mediates ␣v3 integrin-induced endothelial cell survival. J. Cell Biol. 141, 1083–1093. Schaffer, J.E., and Lodish, H.F. (1994). Expression cloning and characterization of a novel adipocyte long chain fatty acid transport protein. Cell 79, 427–436. Schwartz, M.A., Schaller, M.D., and Ginsberg, M.H. (1995). Integrins: emerging paradigms of signal transduction. Annu. Rev. Cell Dev. Biol. 11, 549–599. Shimkets, R.A., Lowe, D.G., Tai, J.T., Sehl, P., Jin, H., Yang, R., Predki, P.F., Rothberg, B.E.G., Murtha, M.T., Roth, M.E., et al. (1999). Gene expression analysis by transcript profiling coupled to a gene database query. Nat. Biotech. 17, 798–803. Springer, T.A. (1994). Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76, 301–314. Thorens, B., Mermod, J., and Vassalli, P. (1987). Phagocytosis and inflammatory stimuli induce GM-CSF mRNA in macrophages through posttranscriptional regulation. Cell 48, 671–679. Valledor, A.F., Borras, F.E., Cullell-Young, M., and Celada, A. (1998). Transcription factors that regulate monocyte/macrophage differentiation. J. Leukoc. Biol. 63, 405–417. Van Antwerp, D.J., Martin, S.J., Kafri, T., Green, D.R., and Verma, I.M. (1996). Suppression of TNF-␣-induced apoptosis by NF-B. Science 274, 787–789. Wang, X., Reape, T.J., Li, X., Rayner, K., Webb, C.L., Burnand, K.G., and Lysko, P.G. (1999). Induced expression of adipophilin mRNA in human macrophages stimulated with oxidized low-density lipoprotein and in atherosclerotic lesions. FEBS Lett. 462, 145–150. Webb, D.S., Shimuzi, Y., van Seventer, G.A., Shaw, S., and Gerrard, T.L. (1990). LFA-3, CD44, and CD45: physiologic triggers of human monocyte TNF and IL-1 release. Science 249, 1295–1297. Werb, Z. (1987). Phagocytic cells: chemotaxis and effector functions of macrophages and granulocytes. In Basic and Clinical Immunology, D.P. Sites, J.P. Stobe, and J.V. Wells, eds. (Norwalk, CT: Appleton-Century-Crofts), p. 96. Wesley, R.B., Meng, X., Godin, D., and Galis, Z.S. (1998). Extracellular matrix modulates macrophage functions characteristic to atheroma: collagen type I enhances acquisition of resident macrophage traits by human peripheral blood monocytes in vitro. Arterioscler. Thromb. Vasc. Biol. 18, 432–440. Xaus, J., Cardo, M., Valledor, A.F., Soler, C., Lloberas, J., and Celada, A. (1999). Interferon ␥ induces the expression of p21waf-1 and arrests macrophage cell cycle, preventing induction of apoptosis. Immunity 11, 103–113. Yurochko, A.D., Liu, D.Y., Eierman, D., and Haskill, S. (1992). Integrins as a primary signal transduction molecule regulating monocyte immediate-early gene induction. Proc. Natl. Acad. Sci. USA 89, 9034–9038.
Global Expression Analysis of Extracellular Matrix–Integrin Interactions in Monocytes Antonin R. de Fougerolles,*‡ Gloria Chi-Rosso,* Adriana Bajardi,* Philip Gotwals,* Cynthia D. Green,† and Victor E. Koteliansky* * Biogen Inc. 12 Cambridge Center Cambridge, Massachusetts 02142 † CuraGen Corporation 555 Long Wharf Drive New Haven, Connecticut 06511
Summary Central to immune and inflammatory responses are the integrin-mediated adhesive interactions of cells with their extracellular matrix (ECM)-rich environment. Using a comprehensive and quantitative mRNA profiling technique, we analyzed the effect of ECM-induced attachment on monocyte gene expression, its regulation by growth factors, and the integrin specificity of this event. Adhesion of cells to fibronectin resulted in increased expression of a large number of genes, which was strongly potentiated by the presence of growth factors. Adhesion activated both the NF-B and Jak/STAT pathways of gene transcription and increased expression of genes involved in inflammatory and immune responses, revealing the importance of ECM–integrin interactions in these processes. Introduction Macrophages are present in almost every tissue; the tissue macrophage pool is 500–1000 times larger than the bone marrow and blood pools (Cohn, 1983). This wide distribution within the peripheral tissues reflects the fact that these cells play an essential role in the disposal of foreign agents, in the initiation and mediation of immune and inflammatory responses, as well as in the repair process following tissue injury. Human blood monocytes undergo differentiation to macrophages upon migration into the extracellular matrix (ECM)-rich environment of extravascular tissue (Werb, 1987). Attachment of cells to ECM proteins, such as fibronectin, collagen, and laminin, is mediated predominantly by members of the integrin family of adhesion molecules (Hemler, 1990). Among the various ECM proteins, fibronectin plays a key role in promoting cell adhesion and various functions of monocytes and macrophages. Attachment of monocytes to fibronectin, which occurs primarily through the ␣41 and ␣51 integrin heterodimers, promotes cell migration, phagocytosis, and differentiation and modulates the secretion of inflammatory mediators (Kaplan and Gaudernack, 1982; Thorens et al., 1987; Juliano and Haskill, 1993; Wesley et al., 1998). Leukocyte proliferation and cytokine secretion can be ‡ To whom correspondence should be addressed (e-mail: tony_de_
[email protected]).
synergistically regulated by the combination of integrinmediated adhesion and soluble growth factors, demonstrating the intimate collaboration that exists between growth factors and integrins (Schwartz et al., 1995). We investigated on a global and unbiased basis the ability of ECM–integrin interactions to regulate monocyte gene expression using a recently published restriction enzyme–based method (GeneCalling) for identifying differentially expressed genes (Shimkets et al., 1999). GeneCalling couples expression profiling with a database query that utilizes fragment length and end sequence information to provide immediate feedback on expressed genes. As GeneCalling requires no a priori gene sequence, it is an open system that detects both known and novel genes. GeneCalling provides ⬎95% coverage of the expressed genome with a sensitivity of detection greater than 1:100,000 and can detect gene expression changes down to 1.5-fold differences (Shimkets et al., 1999). This approach has been used to identify genes linked to cardiac hypertrophy (Shimkets et al., 1999), obesity (Renz et al., 2000), angiogenesis (Kahn et al., 2000), and hematopoietic differentiation (Ohneda et al., 2000). Using this technology, we investigated the effect of ECM-induced adhesion on monocyte gene expression, its regulation by growth factors, and the integrin specificity of this event. The human monocytic cell line THP-1 is a well-studied model system for examining integrininduced gene expression in circulating leukocytes (Juliano and Haskill, 1993; Meredith et al., 1996). Adhesion of cells to fibronectin resulted in increased expression of a large number of genes, which was strongly potentiated by the presence of growth factors. We confirmed that ECM-induced attachment can result in activation of NF-B (Juliano and Haskill, 1993; Meredith et al., 1996) and demonstrated that the presence of growth factors can synergistically increase integrin-mediated activation of the NF-B pathway and lead to the activation of the Jak/STAT pathway. Comparison among different ECM molecules also revealed that the degree of integrin-specific gene expression is considerable. Integrin-mediated attachment resulted in increased expression of numerous inflammatory and immune response genes, revealing an important role for ECM–integrin interactions in affecting monocyte function and thus impacting on the development of pathologies. This is of particular relevance in the context of immune and inflammatory responses, where integrin-mediated adhesive interactions with the ECM-rich peripheral tissues are central to the localization of both resident and infiltrating monocytes at inflammatory sites. Results Attachment to Extracellular Matrix Is a Strong Regulator of Gene Expression To address the effect of ECM-induced adhesion on monocyte gene expression and its regulation by growth factors, THP-1 cells were either left in suspension or
Immunity 750
Table 1. Global Analysis of Monocyte Gene Expression Changes Mediated through Attachment to Extracellular Matrix Molecules
Type of Subtraction
Time (Hours)
Number of Fragments Detected (All Fragments)
Number of Fragments Differentially Expressed (Fragments 2-Fold)
Change in Gene Expression (%)
Type I collagen versus suspension Laminin versus suspension Fibronectin 120 kDa versus suspension Fibronectin versus suspension Fibronectin ⫹ serum versus suspension ⫹ serum
1 1 1 4 4
26 099 26 443 26 924 22 494 27 253
896 1009 733 281 1189
3.4 3.7 2.7 1.2 4.4
The indicated comparisons were made, and the magnitude of gene expression change is reflected by both the number and percentage of gene fragments that are differentially expressed.
plated on fibronectin for 4 hr in either the presence or absence of 2% fetal calf serum. To examine integrinspecific gene expression, THP-1 cells were left either in suspension or plated on a variety of different ECM proteins (RGD-containing 120 kDa fragment of fibronectin, laminin, collagen type I) for 1 hr in the absence of serum. Differential gene expression was quantified using GeneCalling (Shimkets et al., 1999). The comparisons analyzed in this study are listed in Table 1, with the total number of fragments detected and the number of differentially expressed fragments (⫾ 2-fold) listed. Attachment of cells for 4 hr to fibronectin under serum-containing conditions resulted in 1189 fragments (4.4%) being differentially expressed relative to suspension cells also cultured in the presence of serum (Table 1). The majority of the 1189 differentially expressed fragments are upregulated upon attachment to fibronectin, including over 150 fragments whose expression is increased by ⬎5-fold (Figure 1A). We positively identified the genes corresponding to 113 of the most highly induced fragments using competitive PCR. The 113 fragments represented a sixth of all upregulated fragments and segregated into 87 distinct clusters, representing 67 genes and 20 EST fragments (Tables 2 and 3). These genes and EST fragments represent a 5-fold increase in the number of matrix-induced genes identified to date (Juliano and Haskill, 1993; Fan et al., 1995; Meredith et al., 1996; Marra et al., 1997). The sensitivity and reproducibility of GeneCalling is highlighted by the fact that among the 67 known genes were 14 genes previously identified as being induced upon THP-1 attachment to fibronectin (Juliano and Haskill, 1993; Fan et al., 1995; Meredith et al., 1996; Marra et al., 1997). In addition, all the genes (IL-1␣, IL-8, TNF-␣, and tissue factor) previously identified in human peripheral blood monocytes as being induced by fibronectin attachment were also detected by GeneCalling (Haskill et al., 1988, 1991; Yurochko et al., 1992; Lin et al., 1994; Fan et al., 1995; McGilvray et al., 1997). Using independent experimental samples, real-time PCR results on over a dozen genes (MCP-1, MnSOD, IL-8, Gro-␣, MIP-1␣, IP10, CD44, IL-10 receptor, CD40, glutamate-cysteine ligase catalytic subunit, NF-B1, TNFAIP2, and p62 lck ligand) confirmed in all cases the direction and magnitude of the gene expression changes observed (data not shown). The list of known genes that are induced by attachment to fibronectin include secreted and cell surface molecules, metabolic proteins and intracellular regulatory proteins, and transcription factors (Table 2). Two-
thirds (46 of 67) of the identified genes induced by fibronectin attachment in the presence of serum are known to be linked to or regulated through either the NF-B (28 genes) or Jak/STAT (18 genes) pathways of gene transcription (Table 2) (Baeuerle and Henkel, 1994; Darnell, 1997). Included among the 28 NF-B-regulated genes are 14 genes previously identified as being induced by fibronectin attachment (A20 protein, Gro-␣, Gro-, IB, IL-1, IL-1ra, IL-8, JunB, MCP-1, MIP-1␣, NFB, superoxide dismutase, tissue factor, TNF␣) (Juliano and Haskill, 1993; Fan et al., 1995; Meredith et al., 1996; Marra et al., 1997). We have now discovered an additional 14 NF-B-regulated genes that can be induced by fibronectin attachment (EBI3, RANTES, cartilage gp39, IL-10 receptor, CD40, ferritin heavy chain, GLCLC, BCL3, ATF3, RelB, TNFAIP2, MyD88, TRAF-3, IRAK) (Table 2). The upregulation of numerous regulators of the NF-B activation cascade upon fibronectin binding (IRAK, MyD88, p62 lck ligand) may be suggestive of a role for these proteins in integrin-mediated activation of NF-B. Many genes linked to the Jak/STAT pathway are induced by fibronectin attachment (Tables 2 and 4). Among these genes, only STAT5A has previously been described as being regulated via integrins. Attachment to fibronectin has recently been shown to increase activity and expression of STAT5A in endothelial cells (Brizzi et al., 1999). In addition to STAT5A, we have now identified 17 other fibronectin-inducible genes whose expression is known to be regulated through the Jak/ STAT pathway (Table 2). Among these genes are many well-known interferon-inducible and Jak/STAT-regulated genes, such as spermidine/spermine N1-acetyltransferase, Leu-13 (9–27), IFI56, IFI41, and p59 oligoadenylate synthetase (Darnell, 1997). Potentiation of Matrix-Induced Gene Expression by Growth Factors Given the synergy that exists between growth factor receptor and integrin signaling, we sought to compare the pattern of ECM-induced gene expression seen in the presence of growth factors with that seen in its absence. In contrast to the 1189 fragments differentially expressed by fibronectin attachment in the presence of serum, attachment of cells for 4 hr to fibronectin under serum-free conditions resulted in only 281 fragments being differentially expressed as compared to serumfree suspension cells (Table 1). Similar to that seen previously under serum-containing conditions, attachment to fibronectin under serum-free conditions resulted in
Matrix-Induced Gene Expression in Monocytes 751
Figure 1. Distribution by Fold Change of Differentially Expressed Fragments upon Attachment to Extracellular Matrix Molecules (A–C) THP-1 cells attached to fibronectin (FN), collagen type I (CO I), laminin (LM), or fibronectin 120 kDa fragment (FN120) were compared to cells left in suspension, under conditions listed in Table 1. The number of differentially expressed fragments are grouped by fold change (solid bars, upregulated fragments; gray bars, downregulated fragments). (D) Photomicrograph of THP-1 cells after 1 hr attachment in the absence of serum to various extracellular matrix molecules. Cells attached to either collagen type I (CO I), laminin (LM), or fibronectin 120 kDa fragment (FN120) show similar degrees of attachment and cell spreading on the different substrates; cells in suspension are also shown for comparative purposes.
over 70% of the differentially expressed fragments being upregulated (Figure 1B). Comparison of the fragments differentially regulated by fibronectin under serum-free and serum-containing conditions revealed overlap between the two (Table 3). Out of the 281 fragments regulated by fibronectin attachment in the absence of serum, 79 were also regulated by fibronectin in the presence of serum, and 202 fragments were regulated by fibronectin only in the absence of serum. Synergy between ECM and growth factors was revealed by the large number of fragments (1110) whose fibronectin-specific regulation required the presence of serum. This synergy was also evident in the increased induction of genes attached to fibronectin in the presence versus the absence of serum (Table 3). Matrix-Induced Activation of NF-B and Jak/STAT Pathways and Regulation by Growth Factors Analysis of gene expression patterns revealed that attachment of cells to fibronectin can activate genes regulated by the NF-B transcription pathway independent of growth factors, while ECM-induced activation of the Jak/STAT-regulated genes required the presence of
growth factors (Tables 3 and 4). To extend these results, luciferase reporter, nuclear translocation, and phosphorylation assays were performed on THP-1 cells under adherent and nonadherent conditions in the presence or absence of growth factors. Fibronectin attachment activated NF-B-mediated nuclear translocation and transcription (Figures 2A and 2B). Consistent with the required presence of growth factors in ECM-mediated Jak/STAT pathway activation, attachment of cells to fibronectin in the presence of serum resulted in activation of the Jak/STAT pathway as assessed by phosphorylation of STAT1; no such activation was seen upon fibronectin attachment in the absence of serum (Figure 2C). Effect of Different Integrin–Matrix Interactions on Gene Expression To investigate the effect of different integrin ligands on gene expression, THP-1 cells were left in suspension or adhered for 1 hr on plates coated with ECM proteins specific for ␣21 (type I collagen), ␣51 (RGD-containing 120 kDa fragment of fibronectin), and ␣61 (laminin) integrins (Figure 1D). Inhibition of adhesion with blocking mAbs to the relevant ␣ subunits confirmed the
Immunity 752
Table 2. THP-1 Cell Attachment to Fibronectin Activates the NF-B and Interferon Pathways of Gene Induction
The names of the differentially expressed genes are given with their GenBank accession number, the number of fragments that were identified, and the average fold induction of the gene upon 4 hr attachment to fibronectin versus suspension (done in the presence of 2% serum). Genes linked to the NF-B pathway are shown in green, and those linked to the Jak/STAT pathway are shown in red.
specificity of the integrin–ECM interactions (data not shown). Compared to cells left in suspension, attachment of THP-1 cells to various ECM proteins results in the change of a large number of genes (700–1000 fragments, representing 2.7%–3.7% of the expressed genome) (Table 1). In all cases, integrin engagement resulted in roughly 80% of differentially expressed fragments showing an increase in mRNA expression (Figure 1C). To investigate the importance of integrin specificity in regulating gene expression, we compared the distribution of differentially regulated fragments induced by specific integrin–ECM interactions (Table 5). Among the differentially expressed fragments, roughly half (414 fragments) are common to all three matrix molecules. A strong degree of integrin specificity exists, as most of the rest of the differentially expressed fragments are regulated only through specific integrin–ECM interactions (collagen, 316 fragments; laminin, 416 fragments; 120 kDa fibronectin, 262 fragments) (Table 5). The one notable exception to this is the 144 differentially expressed fragments that are shared between collagen type I and laminin but are not regulated by attachment to the 120 kDa fragment of fibronectin. Gene identification both confirmed genes that had previously been shown to be inducible through attachment to multiple extracellular matrices (IL-1, IL-8, Gro, A20 protein) (Meredith
et al., 1996) and resulted in the discovery of another 37 such genes (Table 5). Many of the genes commonly regulated through engagement of ␣21, ␣51, and ␣61 integrins are genes whose expression is induced through activation of NF-B (Table 5). Discussion Global gene expression analysis has enabled us to quantitate the relative contribution of integrin-mediated cell attachment to changes in gene expression. Our results demonstrate the effect integrin engagement has on gene expression and highlight the ability of growth factors and different matrix molecules to influence integrin-mediated gene expression. The positive identification of over 140 genes and ESTs whose expression in monocytes is induced by attachment to ECM not only represents a significant increase in the number of genes previously identified but also reveals the importance of ECM–integrin interaction in multiple aspects of monocyte biology. Monocyte extravasation and attachment in ECM-rich tissues that occurs during the course of immune and inflammatory responses, and mimicked here by attachment of cells to fibronectin, revealed many genes important in monocyte migration and activation. Monocyte
Matrix-Induced Gene Expression in Monocytes 753
Table 3. The Effect of Serum on Fibronectin-Induced Gene Induction in THP-1 Cells
Differentially expressed genes are given with their GenBank accession number and fold gene induction upon 4 hr attachment to fibronection versus suspension in either the presence or absence of 2% serum. ND, not determined.
attachment to the ECM results in the upregulation of numerous chemokines (IL-8, Gro-␣, Gro-, MCP-1, MIP1␣, RANTES, IP10) and cell adhesion molecules (CD18, CD44) important in attracting other cells to inflammatory sites (Springer, 1994; Baggiolini et al., 1997). Genes important in monocyte activation and effector function are also upregulated by fibronectin attachment and include secreted molecules (IL-1, TNF␣) and cell surface molecules (CD40, CD44) (Webb et al., 1990; Grewal and Flavell, 1998). Fibronectin attachment can also affect macrophage activation and function through increased expression of intracellular molecules (hck, MyD88) that are directly involved in macrophage activation and cyto-
kine production (English et al., 1993; Kawai et al., 1999). Matrix-induced attachment also changes expression of genes (pleckstrin, calpain) that, through alterations to the cytoskeleton, are important in the regulation of monocyte cell spreading, migration, and phagocytosis (Brumell et al., 1999; Kulkarni et al., 1999). Along with induction of numerous genes that increase the cytolytic and cytotoxic effect of macrophages, matrix-induced attachment also results in the induction of genes responsible for protecting the cells from self-lysis. NF-B activation initiates a response that blunts the apoptotic action of several agonists of cell death (Beg and Baltimore, 1996; Van Antwerp et al., 1996). Among
Immunity 754
Table 4. Effect of Growth Factors on Integrin-Mediated Activation of the NF-B and Jak-STAT Pathways of Gene Transcription Fold Induction Encoded Protein NF-B pathway NF-B1 p105/p50 MCP-1 TNFAIP2 (B94) IL-1 CD40 Jak-Stat pathway STAT5A IFI56 ISG15 Spermine/spermidine N1acetyltransferase p59 oligoadenylate synthetase PIM-1 Leu-13 (9-27)
FN (⫺ Serum) FN (⫹ Serum) 2.0 4.2 2.9 2.6 1.6
6.0 38.5 20.0 7.8 9.7
0 0 0 0
17.0 38.0 11.8 8.7
0 0 0
5.9 5.2 3.4
The names of the differentially expressed genes are given with the fold gene induction upon 4 hr attachment to fibronectin versus suspension in either the presence or absence of 2% serum.
the genes induced by fibronectin attachment are several NF-B-regulated genes (MnSOD, A20, Bcl-3) that have been shown to have antiapoptotic function (Fridovich, 1989; Opipari et al., 1992; Rebollo et al., 2000). Other matrix-regulated genes not linked to NF-B activation may also play a role in preventing apoptosis. For instance, enforced expression of PIM-1 kinase enhances growth factor–independent survival and inhibits apoptosis in murine myeloid cells (Lilly and Kraft, 1997). Attachment of monocytes to ECM has been demonstrated to induce differentiation to macrophages (Kaplan and Gaudernack, 1982; Wesley et al., 1998). We found that attachment to fibronectin resulted in the upregulated expression of an array of genes previously associated with differentiation of monocytes to macrophages, including CD44, ferritin, spermidine/spermine acetyltransferase (SSAT), cartilage gp39, and CD82 (Krause et al., 1996; Gingras and Margolin, 2000). Fibronectin-mediated attachment also resulted in increased expression of several key transcriptional regulators of macrophage differentiation, including PU.1, STAT5A, and IRF-1. PU.1 is essential for macrophage development because it activates the expression of the M-CSF receptor and transactivates other genes, such as the adhesion molecules CD11b and CD18, the LPS receptor (CD14), and Fc and scavenger receptors, involved in the acquisition of a functional phenotype (Valledor et al., 1998). STAT5A is induced during monocyte differentiation, and its antiapoptotic activity is essential during terminal stages of myeloid differentiation (Kieslinger et al., 2000). Expression of IRF-1 activates the transcription of interferon, which acts to promote inhibition of monocyte proliferation and can also protect cells from apoptosis (Xaus et al., 1999). Thus, matrix-induced adhesion results in the expression of numerous markers associated with monocyte differentiation and induces expression of transcription factors that can regulate multiple aspects of myelopoiesis. Identification of matrix-induced genes may also help
Figure 2. The Effect of Growth Factors on Fibronectin-Mediated Activation of NF-B and STAT1 Fibronectin-mediated activation of NF-B occurs irrespective of the presence of growth factors, while activation of STAT1 requires the presence of growth factors. (A) Induction of NF-B-mediated transcription. THP-1 cells were transfected with a B luciferase reporter construct (2 ⫻ B sites) and adhered to plates containing immobilized fibronectin for 2 and 4 hr in either the presence of serum-free media or media containing 2% fetal calf serum. Cells were collected and luciferase activity was measured. Results are normalized to luciferase activity seen in transfected cells left in suspension. (B) Induction of NF-B translocation into the nucleus. Coverslips were plated with either fibronectin or poly-lysine control substrate. THP-1 cells were attached for 4 hr in serum-free media or media containing 2% fetal calf serum, washed, and fixed in paraformaldehyde. Fixed cells were then permeabilized using 0.5% Triton X-100 and stained for NF-B p65 using a rabbit anti-NF-B p65 antisera followed by a FITC-donkey anti-rabbit secondary reagent. Slides were mounted and NF-B staining was visualized under fluorescent microscopy. (C) Effect of growth factors on fibronectin-mediated activation of STAT1. THP-1 cells were starved for 72 hr and adhered to plates containing immobilized fibronectin for the indicated times in either the presence of serum-free media or media containing 2% fetal calf serum. Bound cells were lysed and cell extracts were subjected to 8% SDS-PAGE. Proteins were electrophoretically transferred to nitrocellulose filter and immunoblotted with either a rabbit antiphosphoSTAT1 (Tyr 701) antibody to detect activated STAT1 or a rabbit anti-STAT1 polyclonal antisera to detect total STAT1.
Matrix-Induced Gene Expression in Monocytes 755
Table 5. Comparison of Gene Regulation through Attachment to Different Integrin Ligands
Venn diagram illustration of the number of differentially expressed fragments (compared to cells in suspension) induced upon 1 hr attachment to either type I collagen, laminin, or 120 kDa fragment of fibronectin. Differentially expressed genes are given with their GenBank accession number and the fold change in gene expression upon 1 hr attachment to different extracellular matrix proteins (FN120, fibronectin 120 kDa fragment; LM, laminin; and COL, collagen type I). Genes linked to the NF-B pathway are shown in green. Colorimetric scale indicating fold change in gene expression is as shown in Table 3.
in understanding the role of monocytes in disease. Both attachment to extracellular matrix and monocyte activation/differentiation are hypothesized to play an important role in the development of atherosclerosis (Ross, 1999). Atherosclerotic plaque formation is characterized by the accumulation of lipid droplets in macrophagederived foam cells. Fatty acid synthesis requires a source of acetyl CoA and NADPH, and among the genes induced by attachment to fibronectin are several enzymes (phosphogluconate dehydrogenase, kynureninase, LACS1) that promote production of both acetyl CoA and NADPH (Hillgartner et al., 1995; Heyes et al., 1997). In addition, LACS1 is also a key protein that promotes intracellular fatty acid retention (Schaffer and Lodish, 1994). Consistent with this hypothesis is the finding that attachment of peripheral blood monocytes to a collagen matrix not only induces cellular differentiation but also increases intracellular lipid accumulation (Wesley et al., 1998). Lastly, expression of matrix-regulated genes, such as ferritin and adipophilin, was found to be induced in macrophages in early human atherosclerotic plaques, and their high level of expression is associated with the development of plaques and disease progression (Pang et al., 1996; Wang et al., 1999).
Thus, monocyte adherence to ECM proteins, such as fibronectin and collagen, may contribute to the development of atherosclerosis not only through increased expression of numerous cytokines and chemokines already implicated in the development of atherosclerosis, but also through increased expression of numerous metabolic enzymes and proteins that play a key role in lipid metabolism. The identification of adherence-induced genes allowed us to further dissect integrin-mediated signal transduction pathways in monocytes. Using a global approach, we have demonstrated that the majority of matrix-induced genes are ones known to be under the control of the NF-B and Jak/STAT transcriptional pathways. We have confirmed that ECM-induced attachment can result in activation of NF-B (Juliano and Haskill, 1993; Meredith et al., 1996) and have demonstrated that the presence of growth factors can synergistically increase integrin-mediated activation of the NF-B pathway and lead to the activation of the Jak/STAT pathway. Comparison of cell attachment to different ECM molecules allowed us to also investigate the relative importance of integrin specificity in affecting gene expression. The ability of different integrin–ECM interactions to regu-
Immunity 756
late a large number of the same genes is not surprising, given that changes in gene expression are not the consequence of specific integrin engagement alone, but may also require adhesion, cell shape change, and cytoskeletal rearrangement (Juliano and Haskill, 1993). While the finding that distinct integrin engagement can result in different patterns of gene expression has been documented (Huhtala et al., 1995), we have determined, using an unbiased global approach, that the degree of integrin-specific gene expression is considerable. The pattern of gene expression induced upon attachment to extracellular matrix molecules in monocytes does show some similarities to that seen in other cell types. While the lack of comprehensive gene profiling for matrix-induced genes on other cell types makes proper comparisons between cell types impossible, a degree of commonality exists between cell types. Attachment of a variety of cell types (endothelial, smooth muscle, and fibroblast) to fibronectin has been shown to activate the NF-B pathway of gene transcription and to induce some of the same genes (IL-1␣, IL-8, MCP-1, STAT5A) seen in monocytes (Qwarnstrom et al., 1994; Marra et al., 1997; Scatena et al., 1998; Brizzi et al., 1999; GarciaVelasco and Arici, 1999). While a certain degree of commonality exists across cell types with regards to integrin-mediated gene expression, many of the genes induced in monocytes by matrix-induced attachment play specific and important roles in this cell type. Identification and analysis of these genes reveals an important role for ECM–integrin interactions in affecting monocyte function and thus impacting on the development of pathologies, such as atherosclerosis. This is of particular relevance in the context of inflammation and the localization of both resident and infiltrating monocytes at an inflammatory site. The importance of integrin interaction with the ECMrich environment of peripheral tissues was recently demonstrated in numerous in vivo models of inflammation, including asthma, hypersensitivity, and arthritis (Henderson et al., 1997; de Fougerolles et al., 2000). Experimental Procedures Adhesion Assays Adhesion assays were essentially carried out as described previously (Gotwals et al., 1996). Human plasma fibronectin and its 120 kDa human ␣-chymotryptic fragment were purchased from GIBCO– BRL (Rockville, MD); rat tail type I collagen was purchased from Collaborative Research Inc. (Bedford, MA); basement membrane laminin was purchased from Sigma (St. Louis, MO). Tissue culture plates (100 mm) (Becton Dickinson, Franklin Lakes, NJ) were coated overnight at 4⬚C with 5 ml of adhesive substrates (10 g/ml) diluted in PBS. Plates were washed twice with PBS, and nonspecific binding sites were blocked with 1% BSA in PBS for 1 hr at room temperature. Human monocytic THP-1 cells were washed into either serum-free binding buffer (RPMI 1640 and 0.25% BSA) or binding buffer containing 2% fetal calf serum (JRH Biosciences, Lenexa, KS). Cells (1.6 ⫻ 106/ml) were then either left in suspension or adhered to extracellular matrix–immobilized plates (5 ml/plate) for 1 to 4 hr at 37⬚C. Unbound cells were removed by washing the plates twice with PBS. Suspension or extracellular matrix–adhered cells (107 cells minimum) were lysed in Trizol (Life Technologies, Grand Island, NY) and RNA isolated as described (Shimkets et al., 1999). All suspension and extracellular matrix–adhered (fibronectin, fibronectin 120 kDa fragment, type I collagen, and laminin) samples were done in triplicate and processed independently.
Differential Gene Expression Analysis GeneCalling reactions were performed essentially as described (Shimkets et al., 1999). Briefly, following double-stranded cDNA synthesis of poly(A)⫹ RNA, cDNA was separated into 96 different pools and each digested with a different pair of restriction enzymes. The resulting fragments were ligated with complementary adapters, one labeled with biotin and the other with fluorescamine (FAM), and amplified by PCR. After affinity purification on streptavidin and separation by polyacrylamide gel electrophoresis, the FAM-labeled fragments were detected by laser excitation. A composite restriction fragment profile is generated for each sample, based on average peak height and variance of nine separate restriction enzyme digestions (each sample triplicate is subjected to three separate restriction enzyme digestions). The restriction fragment profiles of two samples are compared and differentially expressed fragments are identified. Linkage of a differentially expressed cDNA fragment to a gene is made through knowledge of the restriction enzymes used to generate the fragment ends and the length of the fragment itself. A species-specific query of databases reveals genes that fit these criteria, with gene identification confirmed by competitive PCR using gene-specific oligonucleotides.
NF-B Luciferase and Translocation Assays Luciferase assays on THP-1 cells were performed as described (Rosales and Juliano, 1996). THP-1 cells were transfected by DEAEdextran method with a B luciferase reporter construct. The B luciferase reporter construct contains 2 ⫻ B sites derived from the MHC promoter fused to luciferase and was a gift from Dr. Ilana Stancovski (California Institute of Technology, Pasadena, CA). Transfected cells were resuspended in either serum-free binding buffer or binding buffer containing 2% fetal calf serum and either left in suspension or adhered to plates containing immobilized fibronectin for 2 and 4 hr at 37⬚C as outlined above. Bound cells were collected and luciferase activity was measured using a luciferase assay system (Promega Corp., Madison, WI) according to the manufacturer’s instructions. Results are normalized to luciferase activity seen in transfected cells left in suspension. Nuclear translocation of NF-B was performed as described previously (Scatena et al., 1998). Briefly, THP-1 cells were resuspended in either serum-free binding buffer or binding buffer containing 2% fetal calf serum and immediately plated onto glass coverslips coated with 10 g/ml fibronectin (GIBCO–BRL) or 10 g/ml poly-lysine (Sigma) for 4 hr at 37⬚C. Cells were washed and bound cells fixed for 5 min in 3% paraformaldehyde in PBS (pH 7.4) containing 2% sucrose. Cells were then permeabilized with HEPES–Triton X-100 buffer (20 mM HEPES [pH 7.4], 300 mM sucrose, 50 mM NaCl, 3 mM MgCl2, and 0.5% Triton X-100) and stained for NF-B p65 using a rabbit anti-NF-B p65 polyclonal antisera followed by a FITCdonkey anti-rabbit IgG secondary reagent (Santa Cruz Biotechnology, Santa Cruz, CA). Slides were mounted and NF-B staining was visualized under confocal microscopy (Leica, Wetzlar, Germany).
STAT1 Phosphorylation Assays THP-1 cells were serum starved for 72 hr and adhered to plates containing immobilized fibronectin for the indicated times in either the presence of serum-free binding buffer or binding buffer containing 2% fetal calf serum as outlined above. Unbound cells were removed by washing the plates twice with PBS, and bound cells were lysed in ice-cold lysis buffer containing 1% Triton X-100 (v/v) in 20 mM HEPES (pH 7.8), 150 mM NaCl, 0.1% SDS, 5 mM MgCl2, 1 mM Na4VO3, 50 mM NaF, and complete mini-protease inhibitors (Roche, Mannheim, Germany). Lysates were incubated for 30 min on ice before centrifugation at 15,000 ⫻ g for 15 min at 4⬚C. Total protein concentration of postnuclear lysates was determined by BCA (Pierce, Rockford, IL). Equal amounts of cell lysates were subjected to 8% SDS-PAGE and transferred to nitrocellulose (MSI, Westborough, MA) for immunoblotting. The membranes were blocked with 5% milk in TBS (10 mM Tris-HCl [pH 7.6], 150 mM NaCl) and incubated with either a rabbit anti-phosphoSTAT1 (Tyr 701) antibody or a rabbit anti-STAT1 polyclonal antisera (New England Biolabs, Beverly, MA). Bound antibody was revealed with horseradish peroxidase (HRP)-conjugated anti-rabbit antibody us-
Matrix-Induced Gene Expression in Monocytes 757
ing enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech, Piscataway, NJ). Acknowledgments We would like to thank Andro Hsu and Dr. John Peyman for assistance with gene identification, Akos Szilvasi for assistance with the NF-B translocation assays, and Drs. Joe Rosa and Roy Lobb for support and helpful discussions. Received August 17, 2000; revised October 4, 2000. References Baeuerle, P.A., and Henkel, T. (1994). Function and activation of NFB in the immune system. Annu. Rev. Immunol. 12, 141–179. Baggiolini, M., Dewald, B., and Moser, B. (1997). Human chemokines: an update. Annu. Rev. Immunol. 15, 675–705. Beg, A.A., and Baltimore, D. (1996). An essential role for NF-B in preventing TNF-␣-induced cell death. Science 274, 782–784. Brizzi, M.F., Defilippi, P., Rosso, A., Venturino, M., Garbarino, G., Miyajima, A., Silengo, L., Tarone, G., and Pegoraro, L. (1999). Integrin-mediated adhesion of endothelial cells induces JAK2 and STAT5A activation: role in the control of c-fos gene expression. Mol. Biol. Cell 10, 3463–3471. Brumell, J.H., Howard, J.C., Craig, K., Grinstein, S., Schreiber, A.D., and Tyers, M. (1999). Expression of the protein kinase C substrate pleckstrin in macrophages: association with phagosomal membranes. J. Immunol. 163, 3388–3395. Cohn, Z.A. (1983). The macrophage—versatile element of inflammation. Harvey Lect. 77, 63–80. Darnell, J.E. (1997). STATs and gene regulation. Science 277, 1630– 1635. De Fougerolles, A.R., Sprague, A.G., Nickerson-Nutter, C.L., ChiRosso, G., Rennert, P.D., Gardner, H., Gotwals, P.J., Lobb, R.R., and Koteliansky, V.E. (2000). Regulation of inflammation by collagenbinding integrins ␣11 and ␣21 in models of hypersensitivity and arthritis. J. Clin. Invest. 105, 721–729. English, B.K., Ihle, J.N., Myracle, A., and Yi, T. (1993). Hck tyrosine kinase activity modulates tumor necrosis factor production by murine macrophages. J. Exp. Med. 178, 1017–1022. Fan, S., Mackman, N., Cui, M., and Edgington, T.S. (1995). Integrin regulation of an inflammatory effector gene: direct induction of the tissue factor promoter by engagement of 1 or ␣4 integrin chains. J. Immunol. 154, 3266–3274. Fridovich, I. (1989). Superoxide dismutases: an adaptation to a paramagnetic gas. J. Biol. Chem. 264, 7761–7764. Garcia-Velasco, J.A., and Arici, A. (1999). Interleukin-8 expression in endometrial stromal cells is regulated by integrin-dependent cell adhesion. Mol. Hum. Reprod. 5, 1135–1140. Gingras, M.C., and Margolin, J.F. (2000). Differential expression of multiple unexpected genes during U937 cell and macrophage differentiation detected by suppressive subtractive hybridization. Exp. Hematol. 28, 65–76. Gotwals, P.J., Chi-Rosso, G., Lindner, V., Yang, J., Ling, L., Fawell, S.E., and Koteliansky, V.E. (1996). The a1b1 integrin is expressed during neointima formation in rat arteries and mediates collagen matrix reorganization. J. Clin. Invest. 97, 2469–2477. Grewal, I.S., and Flavell, R.A. (1998). CD40 and CD154 in cell-mediated immunity. Annu. Rev. Immunol. 16, 111–135. Haskill, S., Johnson, C., Eierman, D., Becker, S., and Warren, K. (1988). Adherence induces selective mRNA expression of monocyte mediators and proto-oncogenes. J. Immunol. 140, 1690–1694. Haskill, S., Beg, A.A., Tompkins, S.M., Morris, J.S., Yurochko, A.D., Sampson-Johannes, A., Mondal, K., Ralph, P., and Baldwin, A.S., Jr. (1991). Characterization of an immediate-early gene induced in adherent monocytes that encodes IB-like activity. Cell 65, 1281– 1289. Hemler, M.E. (1990). VLA proteins in the integrin family: structures,
functions, and their role on leukocytes. Annu. Rev. Immunol. 8, 365–400. Henderson, W.R., Chi, E.Y., Albert, R.K., Chu, S.J., Lamm, W.J., Rochon, Y., Jonas, M., Christie, P.E., and Harlan, J.M. (1997). Blockade of CD49d (alpha4 integrin) on intrapulmonary but not circulating leukocytes inhibits airway inflammation and hyperresponsiveness in a mouse model of asthma. J. Clin. Invest. 100, 3083–3092. Heyes, M.P., Chen, C.Y., Major, E.O., and Saito, K. (1997). Different kynurenine pathway enzymes limit quinolinic acid formation by various human cell types. Biochem. J. 326, 351–356. Hillgartner, F.B., Salati, L.M., and Goodridge, A.G. (1995). Physiological and molecular mechanisms involved in nutritional regulation of fatty acid synthesis. Physiol. Rev. 75, 47–76. Huhtala, P., Humphries, M.J., McCarthy, J.B., Tremble, P.M., Werb, Z., and Damsky, C.H. (1995). Cooperative signaling by alpha 5 beta 1 and alpha 4 beta 1 integrins regulates metalloproteinase gene expression in fibroblasts adhering to fibronectin. J. Cell Biol. 129, 867–879. Juliano, R.L., and Haskill, S. (1993). Signal transduction from the extracellular matrix. J. Cell Biol. 120, 577–585. Kahn, J., Mehraban, F., Ingle, G., Xin, X., Bryant, J.E., Vehar, G., Schoenfeld, J., Grimaldi, C.J., Peale, F., Draksharapu, A., et al. (2000). Gene expression profiling in an in vitro model of angiogenesis. Am. J. Pathol. 156, 1887–1900. Kaplan, G., and Gaudernack, G. (1982). In vitro differentiation of human monocytes. Differences in monocyte phenotypes induced by cultivation on glass or on collagen. J. Exp. Med. 156, 1101–1114. Kawai, T., Adachi, O., Ogawa, T., Takeda, K., and Akira, S. (1999). Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity 11, 115–122. Kieslinger, M., Woldman, I., Moriggl, R., Hofmann, J., Marine, J., Ihle, J.N., Beug, H., and Decker, T. (2000). Antiapoptotic activity of Stat5 required during terminal stages of myeloid differentiation. Genes Dev. 14, 232–244. Krause, S.W., Rehli, M., Kreutz, M., Schwarzfischer, L., Paulauskis, J.D., and Andreesen, R. (1996). Differential screening identifies genetic markers of monocytes to macrophage maturation. J. Leukoc. Biol. 60, 540–545. Kulkarni, S., Saido, T.C., Suzuki, K., and Fox, J.E.B. (1999). Calpain mediates integrin-induced signaling at a point upstream of rho family members. J. Biol. Chem. 274, 21265–21275. Lilly, M., and Kraft, A. (1997). Enforced expression of the Mr 33,000 Pim-1 kinase enhances factor-independent survival and inhibits apoptosis in murine myeloid cells. Cancer Res. 57, 5348–5355. Lin, T.H., Yurochko, A., Kornberg, L., Morris, J., Walker, J.J., Haskill, S., and Juliano, R.L. (1994). The role of protein tyrosine phosphorylation in integrin-mediated gene induction in monocytes. J. Cell Biol. 126, 1585–1593. Marra, F., Pastacaldi, S., Romanelli, R.G., Pinzani, M., Ticali, P., Carloni, V., Laffi, G., and Gentilini, P. (1997). Integrin-mediated stimulation of monocyte chemotactic protein-1 expression. FEBS Lett. 414, 221–225. McGilvray, I.D., Lu, Z., Bitar, R., Dackiw, A.P.B., Davreux, C.J., and Rotstein, O.D. (1997). VLA-4 integrin cross-linking on human monocytic THP-1 cells induces tissue factor expression by a mechanism involving mitogen-activated protein kinase. J. Biol. Chem. 272, 10287–10294. Meredith, J.E., Winitz, S., Lewis, J.M., Hess, S., Ren, X., Renshaw, M.W., and Schwartz, M.A. (1996). The regulation of growth and intracellular signaling by integrins. Endocr. Rev. 17, 207–220. Ohneda, O., Ohneda, K., Nomiyama, H., Zheng, Z., Gold, S.A., Arai, F., Miyamoto, T., Taillon, B.E., McIndoe, R.A., Shimkets, R.A., et al. (2000). WECHE: a novel hematopoietic regulatory factor. Immunity 12, 141–150. Opipari, A.W., Hu, H.M., Yabkowitz, R., and Dixit, V.M. (1992). The A20 zinc finger protein protects cells from tumor necrosis factor cytotoxicity. J. Biol. Chem. 267, 12424–12427. Pang, J.H., Jiang, M.J., Chen, Y.L., Wang, F.W., Wang, D.L., Chu,
Immunity 758
S.H., and Chau, L.Y. (1996). Increased ferritin gene expression in atherosclerotic lesions. J. Clin. Invest. 97, 2204–2212. Qwarnstrom, E.E., Ostberg, C.O., Turk, G.L., Richardson, A., and Bomsztyk, K. (1994). Fibronectin attachment activates the NF-kB p50/p65 heterodimer in fibroblasts and smooth muscle cells. J. Biol. Chem. 269, 30765–30768. Rebollo, A., Dumoutier, L., Renauld, J., Zaballos, A., Ayllon, V., and Carlos-Martinez, A. (2000). Bcl-3 expression promotes cell survival following interleukin-4 deprivation and is controlled by AP1 and AP1-like transcription factors. Mol. Cell. Biol. 20, 3407–3416. Renz, M., Tomlinson, E., Hultgren, B., Levin, N., Gu, Q., Shimkets, R.A., Lewin, D.A., and Stewart, T.A. (2000). Quantitative expression analysis of genes regulated by both obesity and leptin reveals a regulatory loop between leptin and pituitary-derived ACTH. J. Biol. Chem. 275, 10429–10436. Rosales, C., and Juliano, R. (1996). Integrin signaling to NF-B in monocytic leukemia cells is blocked by activated oncogenes. Cancer Res. 56, 2302–2305. Ross, R. (1999). Atherosclerosis—an inflammatory disease. N. Engl. J. Med. 340, 115–126. Scatena, M., Almeida, M., Chaisson, M.L., Fausto, N., Nicosia, R.F., and Giachelli, C.M. (1998). NF-B mediates ␣v3 integrin-induced endothelial cell survival. J. Cell Biol. 141, 1083–1093. Schaffer, J.E., and Lodish, H.F. (1994). Expression cloning and characterization of a novel adipocyte long chain fatty acid transport protein. Cell 79, 427–436. Schwartz, M.A., Schaller, M.D., and Ginsberg, M.H. (1995). Integrins: emerging paradigms of signal transduction. Annu. Rev. Cell Dev. Biol. 11, 549–599. Shimkets, R.A., Lowe, D.G., Tai, J.T., Sehl, P., Jin, H., Yang, R., Predki, P.F., Rothberg, B.E.G., Murtha, M.T., Roth, M.E., et al. (1999). Gene expression analysis by transcript profiling coupled to a gene database query. Nat. Biotech. 17, 798–803. Springer, T.A. (1994). Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76, 301–314. Thorens, B., Mermod, J., and Vassalli, P. (1987). Phagocytosis and inflammatory stimuli induce GM-CSF mRNA in macrophages through posttranscriptional regulation. Cell 48, 671–679. Valledor, A.F., Borras, F.E., Cullell-Young, M., and Celada, A. (1998). Transcription factors that regulate monocyte/macrophage differentiation. J. Leukoc. Biol. 63, 405–417. Van Antwerp, D.J., Martin, S.J., Kafri, T., Green, D.R., and Verma, I.M. (1996). Suppression of TNF-␣-induced apoptosis by NF-B. Science 274, 787–789. Wang, X., Reape, T.J., Li, X., Rayner, K., Webb, C.L., Burnand, K.G., and Lysko, P.G. (1999). Induced expression of adipophilin mRNA in human macrophages stimulated with oxidized low-density lipoprotein and in atherosclerotic lesions. FEBS Lett. 462, 145–150. Webb, D.S., Shimuzi, Y., van Seventer, G.A., Shaw, S., and Gerrard, T.L. (1990). LFA-3, CD44, and CD45: physiologic triggers of human monocyte TNF and IL-1 release. Science 249, 1295–1297. Werb, Z. (1987). Phagocytic cells: chemotaxis and effector functions of macrophages and granulocytes. In Basic and Clinical Immunology, D.P. Sites, J.P. Stobe, and J.V. Wells, eds. (Norwalk, CT: Appleton-Century-Crofts), p. 96. Wesley, R.B., Meng, X., Godin, D., and Galis, Z.S. (1998). Extracellular matrix modulates macrophage functions characteristic to atheroma: collagen type I enhances acquisition of resident macrophage traits by human peripheral blood monocytes in vitro. Arterioscler. Thromb. Vasc. Biol. 18, 432–440. Xaus, J., Cardo, M., Valledor, A.F., Soler, C., Lloberas, J., and Celada, A. (1999). Interferon ␥ induces the expression of p21waf-1 and arrests macrophage cell cycle, preventing induction of apoptosis. Immunity 11, 103–113. Yurochko, A.D., Liu, D.Y., Eierman, D., and Haskill, S. (1992). Integrins as a primary signal transduction molecule regulating monocyte immediate-early gene induction. Proc. Natl. Acad. Sci. USA 89, 9034–9038.