Myeloid Leukemia Cells Macrophage Differentiation of Human HL-60

Interaction Between α5β1 Integrin and Secreted Fibronectin Is Involved in Macrophage Differentiation of Human HL-60 Myeloid Leukemia Cells This information is current as of September 26, 2021. Amale Laouar, Frank R. Collart, Cynthia B. H. Chubb, Bei Xie and Eliezer Huberman J Immunol 1999; 162:407-414; ; http://www.jimmunol.org/content/162/1/407 Downloaded from

References This article cites 41 articles, 22 of which you can access for free at: http://www.jimmunol.org/content/162/1/407.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 26, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 1999 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. ␣ ␤ Interaction Between 5 1 Integrin and Secreted Fibronectin Is Involved in Macrophage Differentiation of Human HL-60 Myeloid Leukemia Cells1

Amale Laouar, Frank R. Collart, Cynthia B. H. Chubb, Bei Xie, and Eliezer Huberman2 ␣ ␤ We examined the role of ﬁbronectin (FN) and FN-binding integrins in macrophage differentiation. Increased FN and 5 1 integrin gene expression was observed in phorbol 12-myristate 13-acetate PMA-treated HL-60 cells and PMA- or macrophage-CSF-treated blood monocytes before the manifestation of macrophage markers. After treatment of HL-60 cells and monocytes, newly synthesized FN was released and deposited on the dishes. An HL-60 cell variant, HL-525, which is deﬁcient in the protein kinase C␤ (PKC-␤) and resistant to PMA-induced differentiation, failed to express FN after PMA treatment. ␤ Transfecting HL-525 cells with a PKC- expression plasmid restored PMA-induced FN gene expression and macrophage Downloaded from ␣ ␤ differentiation. Untreated HL-525 cells (which have a high level of the 5 1 integrin) incubated on FN differentiated into macrophages. The percentage of cells having a macrophage phenotype induced by PMA in HL-60 cells, by FN in HL-525 ␣ ␤ cells, or by either PMA or macrophage-CSF in monocytes was reduced in the presence of mAbs to FN and 5 1 integrin. The integrin-signaling nonreceptor tyrosine kinase, p72Syk, was activated in PMA-treated HL-60 and FN-treated HL-525 cells. We suggest that macrophage differentiation involves the activation of PKC-␤ and expression of extracellular matrix

␣ ␤ http://www.jimmunol.org/ proteins such as FN and the corresponding integrins, 5 1 integrin in particular. The stimulated cells, through the integrins, attach to substrates by binding to the deposited FN. This attachment, in turn, may through integrin signaling activate nonreceptor tyrosine kinases, including p72Syk, and later lead to expression of other genes involved in evoking the macrophage phenotype. The Journal of Immunology, 1999, 162: 407–414.

he extracellular matrix (ECM)3 is an intricate assembly of phosphorylation in hemopoietic cells (11–14). These changes in proteins that includes collagen, laminin, and fibronectin tyrosine phosphorylation are likely caused by the activation of T (FN) (1). Cells interact with these proteins and with each nonreceptor tyrosine kinases, such as p125FAK or p72Syk (9, 11, 14, other via specific receptors located on their surface. A major class 15). Subsequent to such a tyrosine phosphorylation event, there is of these receptors is the integrins, each of which is composed of a rapid induction of immediate-early genes, including transcription by guest on September 26, 2021 two distinct ␣ and ␤ transmembrane glycoprotein subunits that are factors such as c-fos,c-jun, IkB, and MAD-6, as well as cytokines noncovalently linked (2). The ␣␤ associations determine the li- such as IL-1␤, IL-8, and TNF-␣ (16). gand-binding specificities of the integrin heterodimers for various Macrophages secrete FN, bind FN, and migrate in response to ECM proteins (3). In some cases, two integrins that share a ligand FN (17–20). In addition, FN provides signals that lead to en- recognize different regions of the ligand molecule, as is true of the hanced TNF-␣ production, respiratory burst activity, and ␣ ␤ ␣ ␤ 5 1 and 4 1 FN receptors, that bind to the Arg-Gly-Asp-Ser phagocytosis of foreign microorganisms (18, 21–23), and (RGDS) and Leu-Asp-Val (LDV) motif of FN, respectively (4). promote the differentiation of blood monocytes into tissue Alternatively, two distinct integrins can bind to the same region of macrophages (24–26). The process of FN secretion and adhe- the same ligand, such as the ␣ ␤ and ␣ ␤ integrins, which both 5 1 v 3 sion was shown to be under the control of protein kinase C recognize the RGDS site of FN (5). (PKC) (15, 27, 28). In addition to being adhesion receptors, it has become clear that The human HL-60 myeloid leukemia cell line is often used as integrins are also signal-transducing receptors (6) that regulate cell growth (7) and apoptosis (8), influence gene expression (9), and a model system to study terminal differentiation in myeloid modulate tumor behavior (10). Several reports have shown that cells (29, 30). Activation of PKC by PMA is well known to binding of adhesive ligands to integrins can induce protein tyrosine cause the induction of macrophage markers, such as adhesion and spreading, in myeloid cells (31–33). It is clear that HL-60 cells and blood monocytes adhere and spread on plastic tissue Center for Mechanistic Biology and Biotechnology, Argonne National Laboratory, culture plates in response to PMA treatment; however, the pre- Argonne, IL 60439 cise mechanism by which signals are transmitted from PMA to Received for publication June 8, 1998. Accepted for publication August 31, 1998. the intracellular machinery that controls cell adhesion and The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance spreading, and thus differentiation, remains largely unknown. with 18 U.S.C. Section 1734 solely to indicate this fact. The present study was initiated to test the hypothesis that 1 This work was supported by the U.S. Department of Energy, Office of Biological PMA-induced macrophage differentiation in HL-60 cells and and Environmental Research, under Contract W-31-109-ENG-38. peripheral blood monocytes may involve interaction between 2 Address correspondence and reprint requests to Dr. Eliezer Huberman, Center for secreted FN and its integrin receptors; validation of this hy- Mechanistic Biology and Biotechnology, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439-4833. E-mail address: [email protected] pothesis would allow the delineation of the signaling role that ␤ Syk 3 Abbreviations used in this paper: ECM, extracellular matrix; FN, fibronectin; PKC- and nonreceptor tyrosine kinase p72 play in this RGDS, Arg-Gly-Asp-Ser; PKC, protein kinase C; M-CSF, macrophage-CSF. differentiation.

Materials and Methods entiation induction, cells were incubated with the mAb or peptide for 20 Reagents min before as well as during treatment with the inducers. Isotypic controls (IgG1, IgG2a, and IgG2b), were purchased from Sigma Immunofluorescence (St. Louis, MO) as were dialyzed murine mAbs to human fibronectin (FN- The immunostaining procedures were conducted at 4°C by using either 15, IgG1), an ␣-naphthyl acetate esterase assay kit, RGDS and Gly-Pro- ␤ 96-microwell plates or tissue culture chamber slides (Nunc). The cells were Arg-Pro (GPRP) peptides, and Ficoll-Hypaque. mAbs to the human 1 washed twice with PBSA (PBS containing 1% BSA and 0.1% NaN ) and ␣ ␣ 3 (K20, IgG2a), 4 (HP211, IgG1), and 5 (SAM1, IgG2b) were from Im- FAK incubated for 45 min with the appropriate primary mAb under saturating munotech (Westbrook, ME). The mAb to focal adhesion kinase (p125 ) conditions. The cells were then washed twice with PBSA and incubated for was from Transduction Laboratories (Lexington, KY) and Upstate Bio- Syk an additional 45 min with the secondary Ab CY3. After a wash with PBSA, technology (Lake Placid, NY), and the mAb to p72 was from Wako the slides were mounted with phosphate-buffered Gelvatol (Becton Dick- Chemicals (Richmond, VA). Indocarcyanine-conjugated anti-murine goat inson, Sunnyvale, CA). Fluorescence was examined by using a Vaytek Ig (CY3) was purchased from Jackson ImmunoResearch (West Grove, ␣ ␤ digital confocal microscope. To examine the blocking effects of the protein PA), and the mAb to the human V 3 integrin mAb (23CG, IgG1) was ␤ ␣ kinase inhibitors, cells were incubated with the inhibitors 20 min before as bought from PharMingen (San Diego, CA). The anti-human 4 6 integrin well as during treatment with the inducers. mAb was kindly provided by Dr. S. Kennel of Oak Ridge National Lab- oratory. The macrophage-CSF (M-CSF) was from Biosource International RT-PCR analysis (Camarillo, CA), and plates precoated with mouse laminin, collagen type I, or collagen type IV were from Becton Dickinson (Bedford, MA). Human RNA was purified by centrifugation through a CsCl cushion as previously ␣ ␤ described (36). cDNA was synthesized from total cellular RNA using 5- and 1-chain integrin cDNAs were purchased from Life Technologies (Grand Island, NY). SuperScript II reverse transcriptase (Life Technologies) under the conditions recommended by the supplier. The reverse transcriptase reaction used Cells and cell culture 2 ␮g of total RNA and either 100 ng of oligo(dT) primer or 2 pmol of a Downloaded from gene-specific primer. PCR amplification used the Tfl polymerase (Epicentre The human myeloid HL-60 leukemia cell line was originally obtained from Technologies, Madison, WI) under conditions recommended by the sup- R. C. Gallo of the National Cancer Institute. The HL-525 cells were es- plier. The FN template primers, F1F/F2R nucleotides 3945–3966 and tablished in our laboratory and have been described previously (31). The 4325–4346; 396-bp product) and F5F/F6R (nucleotides 3981–4001 and cells were incubated in tissue culture plates with RPMI 1640 medium sup- 4706–4727; 746-bp product), were derived from the human sequence plemented with 15% heat-inactivated FCS (Intergen, Purchase, NY), pen- (GenBank accession number X02761). The combination of primer F1F and icillin (100 ␮g/ml), streptomycin (100 ␮g/ml), and 2 mM L-glutamine (Life

F6R resulted in a 782-bp PCR product. The template primers for human http://www.jimmunol.org/ Technologies) at 37°C in a humidified atmosphere containing 8% CO2. glyceraldehyde-3-phosphate dehydrogenase, G1F/G2R (nucleotides 19–39 Human peripheral blood leukocytes were obtained from heparinized whole and 713–734; 715-bp product), were derived from the human sequence venous blood. Monocytes were separated by Ficoll-Hypaque density gra- (GenBank accession number X01677). One set of cycle parameters was dient (1.077 g/ml) centrifugation. The mononuclear cells concentrated at used for all primers (denaturation at 94°C for 50 s; annealing at 63°C for the surface were collected and washed twice in RPMI 1640 medium. 1 min; extension at 73°C for 1 min), with the total number of cycles (25– Monocytes (Ͼ95%) were then isolated by selective adherence in tissue 40) tailored to the specific primer pair. For all reactions, various amounts culture dishes (Nunc, Naperville, IL) for 90 min at 37°C. of the reverse transcriptase reaction were used to ensure correspondence between the amount of amplified product and the input cDNA. For the FN Stable transfection of cells amplification reactions, at least three independent primer pairs were used All transfections were performed as previously described (32), by electro- for each set of reverse transcriptase products to validate the amplification by guest on September 26, 2021 poration. A Bio-Rad gene pulser apparatus with a capacitance extender in pattern. 0.4-cm gap electroporation cuvettes (Eppendorf Scientific, Madison, WI) Immunoprecipitation and in vitro kinase activity was used. The pMV7-RP58 plasmid (kindly provided by Dr. I. B. Wein- stein, Columbia University, New York, NY) contained both the full-length Pellets containing 109 cells were lysed in buffer (1% Triton X-100; 0.1% rat PKC-␤1 cDNA and the bacterial neomycin phosphotransferase (neo) SDS, 1% deoxycholate sodium, 50 mM NaCl, 20 mM Tris-HCl (pH 7.4), gene that confers resistance to the antibiotic G418 (geneticin, Sigma). The 5 mM EDTA, 5 mM EGTA, 1 mM PMSF, 20 ␮g/ml leupeptin, 20 ␮g/ml pMV7 plasmid contained the neomycin gene only. For each transfection, aprotinin, and 1 mM sodium o-vanadate) for 30 min (37). Nuclei and 5 ϫ 106 cells were mixed with 10 ␮g of supercoiled plasmid DNA and 0.2 cellular debris were removed by centrifugation at 16,000 ϫ g for 15 min. Syk ml of phosphate-buffered sucrose (272 mM sucrose, 7 mM Na2HPO4,pH Supernatants were incubated with anti-p72 mAbfor2hat4°C, and the 7.4) in a total volume of 0.5 ml. The cells were electroporated at 250 V and immunocomplex was recovered by further incubation with 50 ␮l of protein allowed to recover in 10 ml of serum-supplemented RPMI medium for 24 h G-Sepharose for1hat4°C. Immunoprecipitates were washed in 1 ml of before selection in a medium containing 0.5 mg/ml geneticin. The geneti- lysis buffer and divided into two aliquots. One was subjected to an in vitro cin-resistant transfectants were obtained by limited dilution in 24-well kinase assay, and the other was subjected to immunoblotting analysis. For plates and tested for PKC-␤ expression and PMA inducibility of macro- immunoblotting analysis, immunoprecipitates were resuspended in Laem- phage markers. The selected clones were maintained in a geneticin-con- mli’s sample buffer, boiled, and subjected to SDS-PAGE (7.5% acryl- taining medium. amide). Proteins were transferred onto polyvinylidene fluoride membranes. Immunoblots were incubated with 1% BSA in TBST (20 mM Tris-HCl Differentiation markers (pH 7.5), 150 mM NaCl, and 0.05% Tween 20) for 30 min. Then anti- p72Syk mAb was added for 1 h. After being washed in TBST, the immu- To determine the percentage of adherent and spread cells, we inoculated ϫ 5 noblots were incubated with alkaline phosphatase-conjugated anti-IgG for 5 10 cells in 0.5 ml of medium into each well of a 24-well tissue culture 30 min, washed twice in TBS buffer (20 mM Tris-HCl (pH 7.5) and 150 plate in the presence or absence of PMA. For some experiments, we pre- mM NaCl), and then incubated with alkaline phosphatase substrate. When coated the surface of the wells with either FN or BSA by overnight incu- color reaction developed to the desired intensity, the immunoblots were bation at room temperature with 0.5 ml of PBS solution containing protein washed in deionized water and photographed. For in vitro kinase activity, at 20 ␮g/ml. The nonspecific sites were blocked with 1% BSA in PBS for ␮ one-half of each immunoprecipitate was incubated with an equal volume of 30 min. The wells were then washed with 3 M MnCl2, inoculated with the kinase buffer (25 mM HEPES (pH 7.5), 10 mM MnCl2, and 5 mM MgCl2) cells, and incubated for different time intervals. The percentage of cell containing 0.5 ␮Ci of [32P]ATP and 0.2 mg/ml histone for 30 min. Proteins attachment and spreading was determined as previously described (34). were then resuspended in sample buffer for SDS-PAGE and separated on The percentage of attached cells was up to 20% higher than that of spread Syk ␣ 12% acrylamide gels. p72 activity was measured by the phosphorylation cells. Nonspecific esterase activity was determined using the -naphthyl of histone (32P-labeled histone). acetate esterase assay kit as prescribed by the manufacturer. Phagocytosis was detected as described previously (32), by the ability of cells to ingest 1.7-␮m-diameter Fluoresbrite beads (Polysciences, Warrington, PA) that Results had been sterilized and opsonized by incubation in 70% ethanol for 20 min FN gene expression during macrophage differentiation of HL-60 and then in RPMI supplemented with 20% PBS (not heated) for 18 h at myeloid leukemia cells 37°C; cells were considered positive if they engulfed 20 beads/cell. Ly- sozyme activity was determined as previously described (35). To examine We examined FN gene expression during macrophage differenti- the blocking effect of mAbs, peptides, or other kinase inhibitors on differ- ation of HL-60 cells treated with PMA because this ECM protein The Journal of Immunology 409

Table I. Immunostaining of cell surface integrins on untreated or PMA-treated HL-60 and HL-525 cells a

HL-60 HL-525

Integrin Untreated ϩPMA Untreated ϩPMA ␣ ␤ ϩ ϪϩϩϪϩ v 3 / / ␣ ␤ ϩ ϪϩϩϪϩ 6 4 / / ␤ ϩ Ϫ ϩϩϩ ϩϩϩ ϩϩϩϩ 1 / ␣ ϩ Ϫ ϩϩϩ ϩϩϩ ϩϩϩϩ 5 / ␣ ϩ Ϫ Ϫ ϩϩϩ ϩ 4 / a The HL-60 cells were either untreated or treated with 3 nM PMA. The HL-525 cells were either untreated or treated with 30 nM PMA. After 48 h of incubation, the ␤ ␣ ␣ ␣ ␤ ␣ ␤ ϩ viable cells were reacted with mAb to the 1, 4, 5, 6 4, and v 3 integrins. indicates the intensity of immunostaining; ϩ/Ϫ indicates little or no immunostaining; ϩϩϩ indicates intense immunostaining, which was visualized by a digital confocal microscope (see Fig. 2).

nostaining in the intercellular spaces of the culture dishes contain-

ing HL-60 macrophages but not in those containing control cells Downloaded from (Fig. 1B). Little or no immunostaining with the FN mAb was observed in untreated or PMA-treated HL-525 cells, or with the isotypic control Abs in untreated or in PMA-treated HL-60 and HL- 525 cells. These results indicate that PMA treatment of HL-60 cells results

in increased FN gene expression and cell surface manifestation of http://www.jimmunol.org/ FN within 4 h after PMA treatment, a time frame that precedes the expression of the macrophage phenotype. These results also show that PMA treatment causes the release and deposition of FN.

FIGURE 1. A, RT-PCR analysis of FN mRNA levels in untreated and Induction of macrophage differentiation in HL-525 cells by 3 nM PMA-treated HL-60 cells and in untreated and 30 nM PMA-treated exogenous FN HL-525 cells. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) hy- Our analysis of FN gene expression suggested that FN might be bridization was used to demonstrate equal loading of RNA samples. B involved in the signaling pathway of PMA-induced macrophage

Manifestation of the FN Ag on the surface of untreated (control) and 3 nM by guest on September 26, 2021 differentiation in HL-60 cells. It was therefore of interest to deter- PMA-treated HL-60 cells. Immunostaining was visualized by a digital confocal microscope. C, Attachment and spreading of HL-525 cells (1 ϫ 105/ mine whether supplying FN exogenously would promote macro- ml) incubated on BSA- or FN-precoated culture dishes. The pattern of FN phage differentiation in the FN-deficient HL-525 cells. For this on the precoated dishes in the absence of cells was visualized after immu- reason, we cultured HL-525 cells on FN-precoated dishes in the nostaining with anti-FN mAb. presence and absence of 30 nM PMA. The presence of this ECM protein was in itself sufficient to cause Ͼ75% of the HL-525 cells to manifest macrophage markers such as cell adherence and has been reported to be produced by blood macrophages (18–20). spreading (Fig. 1C), and PMA treatment further increased this per- An attempt to assess FN mRNA levels by standard Northern hy- centage to 90%. No significant cell attachment and spreading was bridization to total RNA or poly(A)-enriched RNA yielded incon- observed on insolubilized FN when HL-60 cells were used or sistent results. To circumvent this problem, we introduced semi- when HL-60 and HL-525 were cultured on dishes precoated with quantitative RT-PCR analysis. By using this approach with three BSA or other ECM proteins such as laminin or collagen type I or different sets of primers (including one that codes for RGDS), we type IV (data not shown). Taken together, our results implicate FN observed barely detectable FN-specific amplification products in in the induction of macrophage differentiation and indicate that untreated HL-60 cells (Fig. 1A). FN steady-state mRNA as as- exogenous FN can substitute for the insufficient FN gene expres- sessed by RT-PCR was induced at 4 h, and its level was further sion needed for such a differentiation in the PMA-resistant HL-525 increased at 24 h after 3 nM PMA treatment. We also examined cells. FN gene expression in cells from an HL-60 variant, HL-525. These ␣ ␤ cells, unlike the parental cells, are resistant to PMA-induced mac- Essential role of secreted FN and 5 1 integrin in macrophage rophage differentiation (31, 32). Contrary to what was observed in differentiation ␤ the HL-60 cells, no FN gene expression was observed in either Because FN has been reported to bind to a large degree to 1 and ␤ untreated or PMA-treated HL-525 cells (Fig. 1A). 3 integrins (reviewed in Refs. 2, 5, and 38), it was of interest to The cell surface levels of FN protein corresponded to the levels determine whether HL-60 and HL-525 cells manifest these recep- of FN gene expression in HL-60 and HL-525 cells. Immunostain- tors on their cell surfaces. For comparison and as a control, we also ␣ ␤ ing of viable cells with specific mAbs indicated that untreated tested for 6 4 integrin, which binds laminin but not FN (39). HL-60 cells exhibit little or no FN (Fig. 1B). Treatment with 3 nM Immunostaining of viable cells showed that HL-60 and HL-525 PMA caused the cells to display an increase of FN immunostaining cells exhibit barely detectable immunofluorescence when reacted ␣ ␤ as early as 4 h. Prolonging this treatment to 2 days, which caused with Mabs specific to v 3 FN-binding integrin (2, 5, 38), as well ␣ ␤ the HL-60 cells to acquire a macrophage phenotype (32), resulted as to 6 4, a laminin-binding integrin (39). PMA treatment yielded in scattered FN immunostaining across the surface of the cells. A only some immunostaining for these Ags (Table I). Immunostain- ␣ ␤ ␣ noteworthy observation was the abundance of speckled FN immu- ing with mAbs to 5, 1, and 4 integrin chains, which was also ␣ ␤ 410 INTERACTION BETWEEN 5 1 INTEGRIN AND SECRETED FIBRONECTIN

hesion molecules for their ability to alter such a differentiation. ␣ ␤ Our results indicated that the mAbs to FN, 5, and 1 integrin chains and the RGDS peptide attenuated the manifestation of mac- ␤ rophage markers; the combination of the anti- 1 and anti-FN mAbs reduced the manifestation of several differentiation markers by Ͼ80% (Table II) (cell spreading, nonspeciﬁc esterase activity, phagocytosis, and lysozyme activity). Moreover, this Ab combination also abated the inhibited cell proliferation induced in HL-60 cells by PMA and in HL-525 cells by FN. ␣ ␤ In addition, we tested an mAb to the v 3 integrin that binds to ␤ FN through an RGDS site but does not belong to the 1 integrin family. This mAb reduced the manifestation of macrophage markers in the PMA-treated HL-60 cells and FN-treated HL-525 cells (Table II) by only ϳ20%, a result we attribute to the low abundance of this integrin on these cell lines (Table I). The mAb to the ␣ 4 chain integrin, which does not bind to the RGDS motif of FN, was ineffective in altering macrophage differentiation in either FIGURE 2. Manifestation of ␤ -chain integrin on the cell surface of 1 HL-60 or HL-525 cells (Table II), implicating RGDS-dependent Downloaded from untreated and 3 nM PMA treated HL-60 and HL-525 cells 48 h. Viable cell adhesion and spreading in the macrophage differentiation pro- cells (1 ϫ 105) were reacted with 10 ␮g/ml anti-␤ Mab. Immunostaining 1 ␣ ␤ was visualized by a digital confocal microscope. cess. As a negative control, we used the mAb to the 6 4 integrin, which does not bind to FN (2); this mAb failed to affect these markers in treated cells (Table II). Control Abs and control peptide barely detectable in HL-60 cells, was intense on HL-525 cells (Ta- GPRP had no effect on the expression of the macrophage markers ble I, Fig. 2). PMA treatment caused both cell lines to display an ␣ ␤ ␣ or on cell proliferation. All mAbs and peptides, when used alone http://www.jimmunol.org/ increase in 5 and 1 and a decrease in 4 immunostaining (Table ␣ ␤ or in combination, did not affect the replication or expression of I, Fig. 2). Moreover, the similar intensity of 5 and 1 immunostaining in PMA-treated HL-60 cells and untreated HL-525 cells maturation markers in untreated HL-60 cells and HL-525 cells (Table I, Fig. 2) can explain the ability of HL-525 cells to adhere cultured on dishes precoated with BSA (Table II). These results to and differentiate on FN-precoated dishes, a result that sug- indicate that the acquisition of a macrophage phenotype in HL-60 ␣ ␤ cells involves the production, release, and deposition of FN and its gests a role for the FN-binding integrin, 5 1, in macrophage ␣ ␤ differentiation. binding to the 5 1 integrin. ␤ To demonstrate an active role for FN and integrin binding in To ensure that the involvement of an FN/ 1-integrins interaction

PMA- and FN-induced macrophage differentiation in HL-60 and is not restricted to macrophage differentiation in the HL-60 cell by guest on September 26, 2021 HL-525 cells, respectively, we tested Mabs speciﬁc to these ad- system, we included human peripheral blood monocytes in our

Table II. Inhibitory effect of speciﬁc mAbs and peptides on macrophage differentiation of HL-60 and HL-525 cells a

No. of Cells Adherent and Cells with Nonspeciﬁc Phagocytizing Lysozyme Activity (ϫ105 cells/ml)b Spread Cells (%)c Esterase Activity (%)d Cells (%)e (␮g/107 cells)f

HL-60 HL-525 HL-60 HL-525 HL-60 HL-525 HL-60 HL-525 HL-60 HL-525 Inhibitors PMA on FN PMA on FN PMA on FN PMA on FN PMA on FN

Control 2.6 Ϯ 0.3 3.3 Ϯ 0.6 91 Ϯ 770Ϯ 10 68 Ϯ 565Ϯ 183Ϯ 976Ϯ 43554 Control isotype IgG 2.4 Ϯ 0.4 3.5 Ϯ 0.4 87 Ϯ 473Ϯ 965Ϯ 662Ϯ 784Ϯ 11 84 Ϯ 73444 mAb to FN 4.4 Ϯ 0.3 4.0 Ϯ 0.7 27 Ϯ 729Ϯ 318Ϯ 227Ϯ 12 27 Ϯ 11 25 Ϯ 16 7 10 ␣ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ 4 2.2 0.1 2.8 0.5 88 5752NTNT799817NTNT ␣ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ 5 4.1 0.3 3.9 0.2 32 7283294NT3862811 29 NT ␤ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ 1 3.8 0.3 3.9 0.3 35 43423433513 33 52612 13 25 FN ϩ ␤ 5.2 Ϯ 0.3 5.9 Ϯ 0.4 19 Ϯ 311Ϯ 411Ϯ 79Ϯ 217Ϯ 713Ϯ 1 Յ0.01 0.1 ␣ ␤ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ v 3 3.0 0.1 3.7 0.4 77 2603514481623598NTNT ␣ ␤ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ Ϯ 6 4 2.2 0.1 3.2 0.6 89 4745735676913822NTNT Peptides GPRP 2.9 Ϯ 0.2 3.2 Ϯ 0.5 88 Ϯ 870Ϯ 12 76 Ϯ 160Ϯ 15 73 Ϯ 878Ϯ 5NTNT RGDS 3.5 Ϯ 0.1 4.1 Ϯ 0.1 45 Ϯ 10 23 Ϯ 12 40 Ϯ 515Ϯ 10 42 Ϯ 530Ϯ 15 NT NT

a The HL-60 cells (1.5–2 ϫ 105/ml) were incubated for 2 days with or without 3 nM PMA, and HL-525 cells (1.51–2 ϫ 105/ml) were incubated in culture dishes precoated ␮ ␤ ␣ ␣ ␤ ␤ with FN or BSA for 2 days in the presence or absence of the Abs or peptides. The cells were incubated with 70 g/ml anti-FN, anti- 1, anti- 5, anti- v 3, or anti- 4 mAb (or isotypic control mAb, IgG) or 700 ␮g/ml RGDS peptide (or control peptide, GPRP). With the exception of lysozyme activity data, the results are the mean Ϯ SD of three to ﬁve independent experiments. The mAbs or peptides alone had little or no effect on the cell proliferation or expression of macrophage markers in 2-day untreated HL-60 cells and 2-day HL-525 cells incubated on culture dishes that were precoated with BSA. NT, not tested. b The number (ϫ 105 cells/ml) of untreated HL-60 cells was 8.2 Ϯ 0.5, and the number of HL-525 cells incubated on a BSA-precoated dish was 9.2 Ϯ 0.8. c The percentage of adherent and spread cells was Յ3 for untreated HL-60 cells and for HL-525 cells incubated on a BSA-precoated dish. d The percentage of cells showing nonspeciﬁc esterase activity was Յ3 for untreated HL-60 cells and for HL-525 cells incubated on a BSA-precoated dish. e The percentage of phagocytosing cells was Յ5 for untreated HL-60 cells and HL-525 cells incubated on a BSA-precoated dish. f Lysozyme activity of cell lysates was Յ0.01 ␮g/107 cells for untreated HL-60 cells and HL-525 cells incubated on a BSA-precoated dish. The Journal of Immunology 411 Downloaded from

FIGURE 3. A, Manifestation of the FN Ag on the surface of untreated (control) or blood monocytes treated with PMA or M-CSF. Cells (5 ϫ 105/ml) were incubated for 48 h in either the presence or absence of 3 nM PMA or 240 U/ml M-CSF. The viable cells were then reacted with anti-FN mAb, and the immunostaining was visualized by a digital confocal micro- http://www.jimmunol.org/ scope. B, Inhibitory effect of specific mAb on PMA- or M-CSF-induced blood monocyte adherence and spreading. Monocytes obtained from four individuals were either untreated (control) or treated for 2 days with 3 nM PMA or 240 U/ml M-CSF in the presence or absence of Abs. The cells ␮ ␤ were incubated with 70 g/ml anti-FN, anti- 1 integrin, or isotypic mAb. The results are the mean Ϯ SD of four independent experiments. by guest on September 26, 2021 study. Our results showed that a 2-day stimulation of blood monocytes by either of two macrophage inducers, PMA or M-CSF (9, 14, 33), increased the amount of speckled FN immunostaining across the cells and in the intercellular spaces (Fig. 3A). As was the case for PMA-treated HL-60 cells, the addition of anti-FN or anti- ␤ 1 mAbs reduced the number of spread cells by 60% in PMA- and M-CSF-treated monocytes (Fig. 3B). Taken together, our findings ␮ indicate that an interaction between deposited FN and ␤ integrins FIGURE 4. A, Northern blot analysis of total RNA samples (40 g/line) 1 ␤ is also involved in PMA- or M-CSF-induced macrophage differ- of PKC- steady-state mRNA levels in parental HL-525 cells and in HL- 525 cells transfected with the PMV7 plasmid containing either the bacterial entiation in blood monocytes. neomycin (neo) phosphotransferase gene or both the full-length PKC-␤ cDNA and the neomycin gene. Stable neomycin (HL-525/neo) and PKC-␤ ␤ ␤ ␤ (HL-525/ 3-2, HL-525/ 3-30) transfectants were selected and maintained Essential role of PKC- in PMA-induced FN gene expression in the presence of 0.4 mg/ml geneticin. HL-60 cells were included for during macrophage differentiation comparison. B, RT-PCR analysis of FN mRNA levels in untreated and Since the HL-525 cells exhibited markedly diminished PKC-␤ HL-525/neo and HL-525/␤3–2 transfectants treated for 2 days with 30 nM gene expression (Fig. 4A), it is possible that this PKC isoenzyme PMA. C, Induction of macrophage differentiation in HL-525 cells trans- ␤ ␤ ␤ is critical for FN gene expression. To test for this possibility, we fected with PKC- cDNA. HL-525/neo, HL-525/ 3-2, or HL-525/ 3-30 cells (0.5 ϫ 106 cells) were incubated in the presence or absence of 30 nM transfected the HL-525 cells with an expression plasmid contain- ␤ PMA. After 2 days of incubation, the percentage of adherent and spread ing the full-length PKC- cDNA and the neomycin gene that con- cells, phagocytizing cells, and lysozyme activity were determined as pre- fers geneticin resistance. As a control, the HL-525 cells were trans- viously described in Materials and Methods. The determinations of cell fected with a plasmid containing only the neomycin gene. Our adherence and spreading and of phagocytosis were performed on trans- findings showed that PKC-␤ gene expression in HL-525 cells fected HL-525 cell cultures at an early (third) passage, whereas lysozyme transfected with PKC-␤ expression plasmid was restored to levels assay was performed on extracts from such cultures at a late (ninth) pas- similar to those of HL-60 cells (Fig. 4A). These cells also regained sage, which contained a large fraction of revertants. The results are the susceptibility to PMA-induced FN gene expression (Fig. 4B) and mean Ϯ SD of three independent experiments. GAPDH, glyceraldehyde- macrophage differentiation, as manifested by a significant increase 3-phosphate dehydrogenase. in adherent and spread cells, in phagocytizing cells, and in lysozyme activity after PMA treatment (Fig. 4C). No restoration of (Fig. 4, B and C). These results implicate the activation of PKC-␤ PMA-induced FN gene expression and macrophage differentiation in FN gene expression during PMA-induced macrophage differ- was observed in HL-525 cells transfected with the control vector entiation in the HL-60 cell system. ␣ ␤ 412 INTERACTION BETWEEN 5 1 INTEGRIN AND SECRETED FIBRONECTIN

FIGURE 5. p72Syk level and activity (32P-labeled histone) in untreated or 30 nM PMA-treated HL-60 and HL-525 cells, and in HL-525 cells cultured on FN-precoated culture dishes. After 15 or 30 min of treatment, the cells were lysed with Triton X-100, and p72Syk was immunoprecipitated as described in Materials and Methods. After washing, the immunoprecipitates were divided into two aliquots. One was dissolved in an SDS- PAGE sample buffer, subjected to electrophoresis, and blotted to polyvinylidene ﬂuoride membranes. Protein levels of p72Syk were analyzed by Western blotting as described in Materials and Methods. The other half of the immunoprecipitates were incubated for 30 min with a kinase buffer containing histone and [32P]ATP, after which they were subjected to SDS-

PAGE. After electrophoresis, the gel was air dried and the ﬁlm was ex- Downloaded from posed for 24 h. The kinase activity is deﬁned by the degree of phosphorylation of histone, the kinase substrate (32P-labeled histone).

Activation of the p72Syk tyrosine kinase during macrophage differentiation http://www.jimmunol.org/ Because the activation of nonreceptor tyrosine kinases such as FIGURE 6. Sequence of events leading to HL-60 macrophage differ- p125FAK and p72Syk has been implicated in FN-evoked integrin- entiation. The cells were either untreated (C) or treated with 3 nM PMA for mediated signal transduction (11, 15), we examined the role of the indicated time period. A, RT-PCR analysis of mRNA levels of FN and ␮ ␤ these kinases in macrophage differentiation in HL-60 and HL-525 GAPDH. B, Northern blot analysis of total RNA sample (20 g/line) for 1 Syk FAK Syk and ␣ subunit integrin. C, p72 level and activation (32P-labeled cells. The p125 and p72 kinases were selected because they 5 histone). are found in some myeloid cell types (9, 11–13). By using specific mAbs, we detected the p72Syk but not the p125FAK kinase, with the Syk level of p72 protein varying only slightly among untreated and by guest on September 26, 2021 PKC-␤ activation and binding of secreted FN to the ␣ ␤ integrin treated HL-60 and HL-525 cells (Fig. 5). p72Syk activation was 5 1 are essential steps in the signaling pathway leading to acquisition observed within 15 min after treatment of HL-60 cells with 30 nM of the macrophage phenotype. PMA and after incubation of HL-525 cells on the dishes precoated Previously, it was found that macrophages secrete FN (17–20). with FN (Fig. 5). Interestingly, treatment with 30 nM PMA did not Although much is known about the role of FN in inflammation and activate p72Syk in the treated HL-525 cells (Fig. 5), indicating that phagocytosis (21–23), little is known about the influence that this this event may require PKC-␤ activity. To further investigate the released ECM protein may have on early stages of macrophage role that PKC-␤ plays in activating of p72Syk, we examined the differentiation. Here we provide explicit evidence that FN is one of activation of this tyrosine kinase in PKC-␤-transfected HL-525 the key elements in the macrophage differentiation signaling cells. Results were inconsistent due to the high reversion fre- cascade. quently of these cells after six cell passages (32). In our initial experiments, we determined the level of gene ex- Sequential events leading to macrophage differentiation in pression and cell surface manifestation of FN in HL-60 cells. HL-60 cells Whereas untreated cells exhibited little or no FN-specific RT-PCR amplification products, treatment with PMA markedly increased Taken together, our results support the hypothesis that PMA-in- FN gene expression at as early as 15 min and in a time-dependent duced HL-60 macrophage differentiation occurs in a coherent se- manner. Moreover, we found that increased FN gene expression quence of events initiated by the activation of PKC-␤. This acti- was followed by FN secretion and deposition on the tissue culture vation evokes a series of chronological changes in gene expression dishes. Next, we examined the ability of blood monocytes to pro- (depicted in Fig. 6), leading to the development of a macrophage duce and secrete FN. Treatment with either PMA or M-CSF phenotype. Induction of FN steady-state mRNA was observed caused the cells to differentiate into macrophages and to display an within 15 min and was maintained for up to 24 h (Fig. 6A). Up- increased level of FN, which was detected not only across the cells regulation of ␣ and ␤ integrin gene expression, appeared to begin 5 1 but also in the intercellular spaces. Additional support for an as- at 4 h and steadily increased thereafter for up to 24 h after PMA sociation between macrophage differentiation and FN came from treatment (Fig. 6B). Activation of p72Syk tyrosine kinase, which our study with an HL-60 variant, HL-525 cells, which, unlike the was also observed at 4 h, decreased markedly at 24 h after PMA parental cells, are resistant to PMA-induced macrophage matura- treatment (Fig. 6C). tion (31, 32, 40). Contrary to results found in HL-60 cells, little to no FN gene expression was observed in PMA-treated HL-525 Discussion cells. This result raised the possibility that FN might play a critical The central finding in this study is that the acquisition of a mac- role in inducing macrophage differentiation. If so, one might spec- rophage phenotype in myeloid precursor cells occurs in a sequen- ulate that supplying FN exogenously would, in itself, promote tial process during which induction of FN gene expression by macrophage differentiation in the FN-deficient HL-525 cells. This The Journal of Immunology 413 notion was successfully proved, in that incubating HL-525 cells on (11–13, 15). Many reports demonstrate that ligation of integrins dishes precoated with FN but not with laminin or collagen resulted leads to enhanced activation of nonreceptor tyrosine kinases (11, in the manifestation of macrophage markers, including cell spread- 15). Although much attention has been paid to p125fak, it is not the ing, lysozyme production, phagocytosis, and nonspecific esterase only nonreceptor tyrosine kinase that is activated by cell adhesion activity. These findings implicate secreted FN as a signaling pro- (2, 9). For monocytic cells, evidence for whether these cells ex- tein in the macrophage differentiation pathway. press p125FAK is contradictory (9, 11, 14). In our experiments, It is well established that interactions between cells and ECM neither untreated nor PMA-stimulated HL-60 or HL-525 cells cul- proteins are often mediated by integrin receptors (2, 5). By using tured in dishes precoated with FN or not expressed detectable ␣ ␤ FAK anti- 5 and anti- 1 mAbs, we found that whereas HL-60 cells p125 protein. Recently, integrin ligation in monocytes was ␣ ␤ Syk displayed weak immunostaining of 5 1 integrin, HL-525 cells found to lead to the activation of p72 , a member of the ZAP70 exhibited an intense immunostaining of this adhesion molecule. family of tyrosine kinase (9). Moreover, cross-linking of phago- Because expression of integrins is reportedly regulated by various cytic Fc␥ receptors on macrophages or cultured monocytes was growth factors and other stimuli (34, 41), it was of interest to shown to induce a rapid phosphorylation of p72Syk, suggesting that examine whether integrin cell surface manifestation was affected this enzyme may play a role in the phagocytic process (43). In our by treatment with PMA. We found that PMA induced HL-60 and experimental cell system, the interaction of integrins (most likely ␣ ␤ ␣ ␤ Syk untreated HL-525 cells to display an increase in 5 1 integrin 5 1) with secreted FN could activate p72 . This possibility was ␣ ␤ Syk immunostaining. Moreover, the intensity of 5 1 integrin immu- investigated through an in vitro kinase assay on p72 immuno- nostaining in PMA-treated HL-60 cells was similar to that in un- precipitates. We found that this tyrosine kinase is present in HL-60 treated HL-525 cells. This result can explain why HL-525 cells and HL-525 cells and that its protein level varies only to a limited Downloaded from adhere and differentiate on dishes precoated with FN, while HL-60 degree between these two cell lines and between untreated and cells do so only after treatment with PMA. We are currently ex- treated cells. We also found that p72Syk was activated in HL-525 tending the study reported here to investigate the factors that ac- and HL-60 cells under conditions that bring about a macrophage count for the difference between the HL-60 and HL-525 cell sur- phenotype, specifically, after treatment with FN and PMA, respec- face manifestations of the integrins. tively. A noteworthy observation was that p72Syk activity was in-

One of the major observations we want to communicate is that duced 4 h after PMA treatment and decreased 24 h after PMA http://www.jimmunol.org/ the interaction of FN with FN-binding integrins is intimately in- treatment in HL-60 cells, suggesting that this enzyme is activated volved in the induction of the macrophage differentiation process. during the induction of differentiation for only a limited time frame We demonstrated this with two experimental approaches. In one that is downstream of FN gene expression. This result can explain approach, we tested the ability of anti-FN and anti-FN-binding why in a previous study, HL-60 cells stimulated with PMA for 24 integrin mAbs to block macrophage differentiation in HL-60 cells to 72 h did not display p72Syk activation (12). At present, we can- and peripheral blood monocytes. We assumed that these Abs not corroborate that p72Syk activation is indeed a downstream sig- would prevent the interaction of the integrins with the deposited naling protein of FN that is involved in the pathway leading to FN and that if this interaction was involved in macrophage differ- macrophage phenotype. For this reason, we simply raise the pos- entiation, then the presence of the mAbs would inhibit such dif- sibility that this kinase has properties that make it a probable can- by guest on September 26, 2021 ␣ ␤ ferentiation. Our studies indicated that the mAbs to FN and 5 1 didate for this function. integrin (which bind to the RGDS motif on FN (4)) were able to We suggest that a pathway leading to macrophage differentia- reduce the percentage of cells showing macrophage phenotypes in tion in human progenitor cells is followed in a chronological and ␣ ␤ induced HL-60 cells and monocytes but that the mAb to 4 1 coherent sequence that involves the activation of PKC; the pro- integrin (which binds to the LDV motif on FN (4)) was not. Our duction, release, and deposition of FN and the up-regulation of ␣ ␤ results also showed a decrease in HL-60 cell differentiation when 5 1 gene expression. The stimulated cells, through their integrins ␣ ␤ the RGDS peptide was used. Other investigators have found that ( 5 1 in particular), adhere to the deposited FN (most likely synthetic peptides containing the RGDS motif promote cell adhe- through its RGDS site). This adhesion, in turn, leads to the acti- sion when immobilized on suitable substrates and inhibit cell ad- vation, through integrin signaling, of downstream kinases (6, 44), hesion to tissue culture dishes precoated with FN (17). Taken to- including p72Syk, and later to the expression of genes involved in gether, our results indicate that induction of macrophage evoking the macrophage phenotype. ␣ ␤ differentiation involves the interaction of 5 1 integrin with, most likely, the RGDS site of the deposited FN. Acknowledgments In another approach we used HL-525 cells, which exhibit little or no PKC-␤ nor FN gene expression. We showed that when these We thank Dr. S. Kennel (Oak Ridge National Laboratory) for the generous gift of the mAb against the human ␣ ␤ integrin, Drs. M. Bhattacharyya and cells are transfected with PKC-␤ cDNA they regain PKC-␤ ex- 6 4 D. Glesne for critical review of the manuscript, and K. Nobles for secretarial pression and susceptibility to PMA-induced differentiation, as well expertise. as FN gene expression. Our results suggest that PKC-␤ plays an essential role in FN gene expression during PMA-induced macrophage maturation. We also found that macrophage differentiation References induced in HL-60 by PMA and in HL-525 by FN was inhibited by 1. Mosher, D. F., J. Sottile, C. Wu, and J. A. McDonald. 1992. Assembly of ex- anti-␣ or anti-␤ integrins, anti-FN mAbs, and RGDS peptide. tracellular matrix. Curr. Opin. Cell Biol. 4:810. 5 1 2. Hynes, R. O. 1992. Integrin: versatility, modulation and signaling in cell adhe- These lines of evidence further support the hypothesis that the sion. Cell 69:11. ␣ ␤ interaction of FN and its 5 1 receptor is involved in attaining of 3. Juliano, R. L., and S. Haskill. 1993. Signal transduction from the extracellular a macrophage phenotype. This conclusion is similar in concept to matrix. J. Cell Biol. 120:577. 4. Guan, J. L., and R. O. Hynes. 1990. Lymphoid cells recognize an alternatively that proposed some years ago by Utsumi et al. (42), who found that ␣ ␤ spliced segment of ﬁbronectin via the integrin receptor 4/ 1. Cell 60:53. adhesion of immature thymocytes to thymic stromal cells through 5. Aota, S.-I., and K. M. Yamada. 1995. Fibronectin and cell adhesion: speciﬁcity ␤ of integrin-ligand interaction. Adv. Enzymol. Relat. Areas Mol. Biol. 70:1. FN- 1 integrin interaction induces T lymphocyte differentiation. 6. Chen, Q., T. H. Lin, C. J. Der, and R. L. Juliano. 1996. Integrin-mediated acti- Evidence from several laboratories suggests that integrins are vation of MEK and mitogen-activated protein kinase is independent of Ras. capable of transmitting signals from the ECM to the cell interior J. Biol. Chem. 271:18122. ␣ ␤ 414 INTERACTION BETWEEN 5 1 INTEGRIN AND SECRETED FIBRONECTIN

␣ ␤ 7. Schneller, M., K. Vuori, and E. Ruoslahti. 1997. v 3 integrin associates with 28. Scheidl, H., T. Chandra, B. Gmeiner, G. Zerlauth, G. Scita, and G. Wolf. 1991. activated insulin and PDGF ␤ receptors and potentiates the biological activity of The effect of diacylglycerols on fibronectin release and its reversal by retinoic PDGF. EMBO J. 16:5600. acid in cell culture. Carcinogenesis 12:1791. ␤ 8. Miao, J. Y., S. Araki, K. Kaji, and H. Hayashi. 1997. Integrin 4 is involved in 29. Harris, P., and P. Ralph. 1985. Human leukemic models of myelomonocytic apoptotic signal transduction in endothelial cells. Biochem. Biophys. Res. Com- development: a review of the HL-60 and U937 cell lines. J. Leukoc. Biol. 37:407. mun. 233:182. 30. Laouar, A., Collart, F. R., and Huberman, E. 1998. Induction of differentiation in 9. Lin, T. H., C. Rosales, K. Mondal, J. B. Bolen, S. Haskil, and R. L. Juliano. 1995. human myeloid HL-60 leukemia cells. In: Cell Biology Handbook, Vol. 1. Ac- Integrin-mediated tyrosine phosphorylation and cytokine message induction in ademic Press, San Diego, pp. 233–238. monocytic cells: a possible signaling role for the Syk tyrosine kinase. J. Biol. 31. Homma, Y., C. B. Henning-Chubb, and E. Huberman. 1986. Translocation of Chem. 270:16189. ␣ ␤ protein kinase C in human leukemia cells susceptible or resistant to differentiation 10. Varner, J. A., D. A. Emerson, and R. L. Juliano. 1995. Integrin 5 1 expression negatively regulates cell growth: reversal by attachment to fibronectin. Mol. Biol. induced by phorbol 12 myristate 13 acetate. Proc. Natl. Acad. Sci. USA 83:7316. Cell 6:725. 32. Tonetti, D. A., C. Henning-Chubb, D. T. Yamanishi, and E. Huberman. 1994. ␤ 11. Lin, T. H., A. Yurochko, L. Kornberg, J. Morris, J. J. Walker, S. Haskill, and Protein kinase C- is required for macrophage differentiation of human HL-60 R. L. Juliano. 1994. The role of protein tyrosine phosphorylation in integrin- leukemia cells. J. Biol. Chem. 269:23230. mediated gene induction in monocytes. J. Cell Biol. 126:1585. 33. Laouar, A., J. Wietzerbin, and B. Bauvois. 1993. Divergent regulation of cell 12. Qin, S., and H. Yamamura. 1997. Up-regulation of Syk activity during HL60 cell surface protease expression in HL-60 cells differentiated into macrophages with differentiation into granulocyte but not into monocyte/macrophage-lineage. Bio- granulocyte macrophage colony stimulating factor or neutrophils with retinoic chem. Biophys. Res. Commun. 236:697. acid. Int. Immunol. 5:965. 13. Melford, S. K., M. Turner, S. J. Briddon, V. L. Tybulewicz, and S. P. Watson. 34. Bauvois, B., D. Rouillard, J. Sanceau, and J. Wietzerbin. 1992. IFN-␥ and trans- ␤ 1997. Syk and Fyn are required by mouse megakaryocytes for the rise in intra- forming growth factor 1 differently regulate fibronectin and laminin receptors of cellular calcium induced by a collagen-related peptide. J. Biol. Chem. 272:27539. human differentiation monocytic cells. J. of Immunol. 148:3912. 14. Kharbanda, S., A. Saleem, Z. Yuan, Y. Emoto, K. V. Prasad, and D. Kufe. 1995. 35. Richards, P. G., D. J. Walton, and J. Heptinstall. 1996. The effect of tyrosine Stimulation of human monocytes with macrophage colony-stimulating factor in- nitration on the structure and function of hen egg-white lysozyme. Biochem. J. duces a Grb2-mediated association of the focal adhesion kinase pp125fak and

315:473. Downloaded from dynamin. Proc. Natl. Acad. Sci. USA 92:6132. 36. Chirgwin, J. M., A. E. Przybyla, A. E. MacDonald, and W. J. Rutter. 1979. 15. Vuori, K., and E. Ruoslahti. 1993. Activation of protein kinase C precedes ␣ ␤ 5 1 Isolation of biological active ribonucleic acid from sources enriched in ribonu- integrin-mediated cell spreading on ﬁbronectin. J. Biol. Chem. 268:21459. clease. Biochemistry 18:5294. 16. Rosales, C., V. O’Brien, L. Kornberg, and R. Juliano. 1995. Signal transduction by cell adhesion receptors. Biochim. Biophys. Acta 1242:77. 37. Ichinohe, T., H. Takayama, Y. Ezumi, M. Arai, N. Yamamoto, H. Takahashi, and 17. Hershkoviz, R., R. Alon, D. Gilat, and O. Lider. 1992. Activated T lymphocytes M. Okuma. 1997. Collagen-stimulated activation of Syk but not c-Src is severely and macrophages secrete ﬁbronectin which strongly supports cell adhesion. Cell. compromised in human platelets lacking membrane glycoprotein VI. J. Biol. 18. Hynes, R. O., and K. M. Yamada. 1982. Fibronectins: multifunctional modular Chem. 272:63.

glycoproteins. J. Cell Biol. 95:369. 38. Yamada, K. M. 1991. Adhesive recognition sequences. J. Biol. Chem. http://www.jimmunol.org/ 19. Alitalo, K., T. Hovi, and A. Vaheri. 1980. Fibronectin is produced by human 266:12809. macrophages. J. Exp. Med. 151:602. 39. Mainiero, F., A. Pepe, K. K. Wary, L. Spinardi, M. Mohammadi, J. Schlessinge, ␣ ␤ ␤ 20. Johansson, S., K. Rubin, M. Hook, T. Ahlgren, and R. Seljelid. 1979. In vitro and F. G. Giancotti. 1995. Signal transduction by the 6 4 integrin: distinct 4 biosynthesis of cold insoluble globulin (fibronectin) by mouse peritoneal macro- subunit sites mediate recruitment of Shc/Grb2 and association with the cytoskel- phages. FEBS Lett. 105:313. eton of hemidesmosomes. EMBO J. 14:4470. 21. Mosesson, M. W., and D. L. Amrani. 1980. The structure and biologic activities 40. Yang, K. D., T. Mizobuchi, S. M. Kharbanda, R. Datta, E. Huberman, of plasma fibronectin. Blood 56:145. D. W. Kufe, and R. M. Stone. 1994. All-trans retinoic acid reverses phorbol ester 22. Peat, E. B., N. H. Augustine, W. K. Drummond, J. F. Bohnsack, and H. R. Hill. resistance in a human myeloid leukemia cell line. Blood 83:490. ␣ 1995. Effects of fibronectin and group B streptococci on tumour necrosis factor- 41. Bellon, T., C. Lopez-Rodriguez, M. A. Rubio, G. Jochems, C. Bernabeu, and production by human culture-derived macrophages. Immunology 84:440. ␣ ␤ A. L. Corbi. 1994. Regulated expression of p150,95 (CD11c/CD18; X/ 2) and 23. Yang, K. D., N. H. Augustine, M.-F. Shaio, J. F. Bohnsack, and H. R. Hill. 1994. VLA-4 (CD49d/CD29; ␣ /␤ ) integrins during myeloid cell differentiation. Eur.

4 1 by guest on September 26, 2021 Effects of fibronectin on actin organization and respiratory burst activity in neu- J. Immunol. 24:41. trophils, monocytes, and macrophages. J. Cell. Physiol. 158:347. 24. Rein, G., A. Abraham, S. P. Raychadauri, and H. Weltman. 1994. Monocyte 42. Utsumi, K., M. Sawada, S. Narumiya, J. Nagamine, T. Sakata, S. Iwagami, adhesion to fibronectin in psoriasis. Int. J. Dermatol. 33:374. Y. Kita, H. Teraoka, H. Hirano, M. Ogata, T. Hamaoka, and H. Fujiwara. 1991. 25. Armstrong, J. W., and S. K. Chapes. 1994. Effects of extracellular matrix proteins Adhesion of immature thymocytes to thymic stromal cells through fibronectin on macrophage differentiation, growth, and function: comparison of liquid and molecules and its significance for the induction of thymocyte differentiation. agar culture systems. J. Exp. Zool. 269:178. Proc. Natl. Acad. Sci. USA 88:5685. 26. Lundahl, J., A. Eklund, G. Hallden, and C. M. Skold. 1996. Adhesion properties 43. Greenberg, S., P. Chang, and S. C. Silverstein. 1994. Tyrosine phosphorylation or Syk of human alveolar macrophages with respect to extracellular matrix components the ␥ subunit of Fc␥ receptors, p72 , and paxillin during Fc receptor-mediated and chemotactic agonists. Immunol. Lett. 53:121. phagocytosis in macrophages. J. Biol. Chem. 269:3897. 27. Pacifici, R., J. Roman, R. Kimble, R. Civitelli, C. M. Brownfield, and C. Bizzarri. 44. Miyamoto, S., H. Teramoto, O. A. Coso, J. S. Gutkind, P. D. Burbelo, ␣ ␤ ␣ ␤ 1994. Ligand binding to monocyte 5 1 integrin activates the 2 1 receptor via S. K. Akiyama, and K. M. Yamada. 1995. Integrin function: molecular hierar- ␣ the 5 subunit cytoplasmic domain and protein kinase C. J. Immunol. 153:2222. chies of cytoskeletal and signaling molecules. J. Cell Biol. 131:791.