Neutrophils Chemotaxis and Phagocytosis in Human Coronin Function Is Required

Coronin Function Is Required for Chemotaxis and Phagocytosis in Human Neutrophils

This information is current as Ming Yan, Caterina Di Ciano-Oliveira, Sergio Grinstein and of September 23, 2021. William S. Trimble J Immunol 2007; 178:5769-5778; ; doi: 10.4049/jimmunol.178.9.5769 http://www.jimmunol.org/content/178/9/5769 Downloaded from

References This article cites 25 articles, 11 of which you can access for free at: http://www.jimmunol.org/content/178/9/5769.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 23, 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 © 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Coronin Function Is Required for Chemotaxis and Phagocytosis in Human Neutrophils1

Ming Yan,2 Caterina Di Ciano-Oliveira, Sergio Grinstein, and William S. Trimble3

Coronins are a family of conserved actin-associated proteins that have been implicated in a variety of cellular processes dependent on actin rearrangements. In this study, we show that in primary human neutrophils, coronins-1–4 and -7 are expressed. Coronin-1 accumulates at the leading edge of migrating neutrophils and at the nascent phagosome. Inhibition of coronin function by transduction of a dominant-negative form of the protein leads to inhibition of chemotaxis and a reduction in neutrophil spreading and adhesion. This inhibition appears to correlate with changes in the distribution of F-actin structures within the cell. In addition, phagocytosis is inhibited, but neither secretion nor activation of the NADPH oxidase appears to be affected. Together, these results show that coronins are required for actin-dependent changes in cell morphology that lead to migration and phagocytosis. The

Journal of Immunology, 2007, 178: 5769–5778. Downloaded from

hemotaxis is the directed migration of cells toward a gra- sion of coronin-1 includes a coiled coil domain implicated in tri- dient of a chemical attractant. In mammals, chemotaxis mer formation (6). ␤-propeller domains often represent interaction C plays an important role in a host of biological processes surfaces for protein-protein interactions, although it is not cur- ranging from development to innate immunity. In the case of the rently known what proteins bind to the coronin ␤-propeller do- immune system, many cells, but most notably neutrophils, efﬁ- mains. Coronin-1 has been shown to associate with the Arp2/3 http://www.jimmunol.org/ ciently sense chemical gradients to migrate toward the sites of complex (7) and to F-actin (8, 9). In yeast, the Arp2/3 binding was injury or infection. This migratory response requires the develop- mapped to C-terminal coiled-coil region (10) and electron micros- ment of polarity within the neutrophil and unique actin structures copy indicated that this region bound near the p35 subunit of are assembled at the leading edge and rear of the cell to achieve Arp2/3 to cause a conformational change in the complex (11). The movement. Although many of the regulatory proteins contributing Arp2/3 binding region of mammalian coronin-1 has yet to be de- to the polarized accumulation of actin have been described, addi- termined. The regions of coronin-1 responsible for actin binding tional factors are likely to also contribute to this regulation. have been more controversial. Initial studies suggested that the One such actin-associated protein that may be important for strongest actin binding site was in the N-terminal 34 aa, while the

neutrophil chemotaxis is coronin. Coronin was first described in second and third WD domains also had weak actin-binding capac- by guest on September 23, 2021 the slime mold Dictyostelium discoideum as a soluble protein that ity (9). More recently, studies have suggested that the C-terminal bound to actin-myosin complexes (1). Loss of the coronin gene half of the protein bound and cross-linked actin filaments (12) product results in cells with impaired chemotaxis and phagocytosis while deletion of residues 400–416 resulted in a protein that did (2). Coronins are conserved from yeast to humans, with at least not bind to the actin cytoskeleton (6). seven isoforms being expressed in mammals (3), but relatively The precise role of coronin in actin assembly also remains un- little is known about the specific roles of the mammalian forms or clear. In yeast, the coronin homolog Crn1p enhances barbed-end their functional relationship to the Dictyostelium form. assembly, apparently by reducing the lag phase of polymerization Recent structural studies of murine coronin-1 have revealed that (13). In contrast, Dictyostelium coronin associates with the entire the protein forms a seven-bladed ␤ propeller structure similar to length of actin filaments and it has been suggested to speed up the ␤ subunit of heterotrimeric G proteins (4). ␤ propellers are depolymerization (14). It has also been suggested that the phys- comprised of a protein domain called the WD domain, so-named ical association of coronin with the Arp2/3 complex alters the due to the presence of conserved tryptophan and aspartate residues, latter’s activity (10), by causing a conformational change in the and the WD domain folds such that it contributes to the formation complex (11). of two adjacent blades of the propeller (5). The C-terminal exten- Recently, we examined the role of coronin-1 in phagocytosis in the murine macrophage cell line RAW 264.7. Using a previously described dominant-negative fragment of coronin-1 (15) linked to Programme in Cell Biology, Hospital for Sick Children, Toronto, Ontario, Canada, the protein transducing domain of TAT, we were able to acutely and Department of Biochemistry, University of Toronto, Toronto, Ontario, Canada inhibit the function of coronin (16). We showed that phagocytosis Received for publication May 5, 2006. Accepted for publication February 13, 2007. in this model system depended on functional coronin-1. Moreover, The costs of publication of this article were defrayed in part by the payment of page while binding to and signaling from the FcRs appeared normal, charges. This article must therefore be hereby marked advertisement in accordance and both actin and full-length (FL)4 coronin-1 accumulated at the with 18 U.S.C. Section 1734 solely to indicate this fact. phagocytic cup, the dominant-negative fragment prevented accu- 1 This work was supported by a grant from the Canadian Institutes of Health Re- search. S.G. is the current holder of the Pitblado Chair in Cell Biology. W.S.T. is a mulation of Arp2/3 and completion of phagocytosis. recipient of a Canada Research Chair in Molecular Cell Biology. 2 Current address: Division of Nephrology, Duke University Medical Center, Durham, NC 27708. 4 Abbreviations used in this paper: FL, full length; fMLF, N-formyl-Met-Leu-Phe; 3 Address correspondence and reprint requests to Dr. William S. Trimble, Programme RT, room temperature. in Cell Biology, Hospital for Sick Children, 555 University Avenue, Toronto, Ontario M5G 1X8, Canada. E-mail address: [email protected] Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 www.jimmunol.org 5770 CORONIN IS REQUIRED FOR NEUTROPHIL CHEMOTAXIS

In this study, we set out to examine the role of coronin in the (Qiagen). All RT-PCR products were detected on a 1% agarose gel and con- process of neutrophil chemotaxis. We first show that neutrophils firmed by DNA sequence analysis. PCR primers of all seven human coronin express five of the seven human coronin genes. We then demon- genes were follows as (using the nomenclature in Ref. 5; coronin-1 sense 5Ј-GCTGCACGAGCGGAGGTGTG-3Ј and antisense 5Ј-ATCCGAGCTG strate that coronin-1 transiently accumulates at the leading edge of GGAGTGCCAC-3Ј; coronin-2 sense 5Ј-TCCTTCCGCAAAGTGGTCCG-3Ј migrating neutrophils similar to F-actin. Moreover, by introducing and antisense 5Ј-AGGCTGTCCAGGCGGTACAG-3Ј; coronin-3 sense 5Ј-CA the dominant-negative fragment of coronin-1 into neutrophils, we GATTTGTTGCCATAATCATAGAG-3Ј and antisense 5Ј-TGATAATGGC observed significant changes in their adhesion properties and a ATTATCACAGCC-3Ј; coronin-4 sense 5Ј-ACAAAGGAGTCTGTCATCAC AAG-3Ј and antisense 5Ј-GCCATGGAATTGAAGATAGG-3Ј; coronin-5 nearly complete block in chemotaxis. Phagocytosis was also im- sense 5Ј-CCATCACCAAGAATGTGCAC-3Ј and antisense 5Ј-GCAGTCAA paired in these cells, but several other functions such as secretion TCATCTTCACCG-3Ј; coronin-6 sense 5Ј-ATCCGGCCAGGACGCCGAA and NADPH oxidase activation were unaffected. Our results dem- CC-3Ј and antisense 5Ј-ACCAGCTCGCACAGCATGTTC-3Ј; coronin-7 onstrate that coronin-1 plays a critical role in neutrophil adhesion sense 5Ј-GCCAGCTGCTCCTATATGAAG-3Ј and antisense 5Ј-TAGTC Ј and cell migration. CCACTCGTCCTCGTC-3 . Measurement of endogenous coronin concentration by Materials and Methods immunoblotting Reagents and Abs A total of 1.1 ϫ 106 neutrophils was lysed in 1 ml of boiling 1% SDS lysis All restriction enzymes were from New England Biolabs. DMEM, FBS, buffer, and passed five-time through a 27-gauge needle. A total of 2.5 ␮lof PBS, trypsin-EDTA, penicillin/streptomycin, and HEPES-buffered RPMI the neutrophil lysate (roughly 3000 cell equivalents) was subjected to SDS- 1640 were purchased from Wisent. Paraformaldehyde was obtained from PAGE on a 10% polyacrylamide gel. To generate a standard curve, purified

Canemco. Human IgG, human plasma fibronectin, and PMA were pur- GST-coronin from 0 to 10 ng was electrophoresed the same gel. Western Downloaded from chased from Sigma-Aldrich. Human coronin-1 cDNA was a gift from Dr. blotting was then conducted to detect endogenous human neutrophil coro- S. Toyoshima (National Institute of Health Science, Tokyo, Japan). Poly- nin-1 and purified GST-coronin using a polyclonal coronin-1 antiserum clonal anti-coronin-1 and N-terminal 60 coronin-1 antiserum were pro- (1/10,000). Multiple exposures were generated to ensure that signals were vided by Dr. J. Pieters (University of Basel, Basel, Switzerland). The plas- obtained within a linear range. Densitometric scans of bands were per- mids of pTAT, pTAT-␤-galactosidase, and ␤-galactosidase (non-TAT) formed and the concentration of endogenous coronin was interpolated from were provided by Dr. S. F. Dowdy (University of California, San Diego, within the linear range of recombinant GST-coronin protein. CA). Cy3-and Cy5-conjugated secondary Abs were obtained from Jackson ImmunoResearch Laboratories. Dihydrorhodamine 1,2,3 (DHR 1, 2, 3), Transduction with TAT-␤-gal or TAT-WD protein http://www.jimmunol.org/ rhodamine-phalloidin, and calcein AM were obtained from Molecular All TAT-WD and TAT-␤-galactosidase transductions were performed as Probes. N-formyl-Met-Leu-Phe (fMLF) was obtained from Fluka Chemie. ␮ described previously (16). Briefly, neutrophils and HeLa cells were first Collagen-coated Transwell chambers (3.0 m pore size) were purchased washed in PBS, treated with TAT-␤-gal or TAT-WD protein at a final from Corning. Polymorphprep was obtained from Accurate Chemical ␮ concentration of 200 nM in PBS containing 1 mM calcium chloride, 1 mM and Scientific. Polystyrene beads (3 m diameter) were obtained from magnesium chloride, and 10 mM glucose, then incubated at 37°C for 30 Bangs Laboratory. Fluorescent mounting medium was obtained from ␤ min, without CO2. To monitor -galactosidase activity, the cells treated DakoCytomation. with TAT-␤-gal were fixed for 5 min at 4°C and then incubated for 24 h Isolation of human neutrophils at 37°C in a histochemical staining solution containing 3 mM ferrocyanide, 3 mM potassium ferricyanide, 10% DMSO, 2 mM MgCl2 and 1 mg/ml by guest on September 23, 2021 Human blood was obtained from healthy volunteers. Donations were ob- 5-bromo-4-chloro-3-indolyl ␤-D-galactoside in PBS. Samples were exam- tained following guidelines approved by the Hospital for Sick Children ined using an Axiovert 200 M light microscope (Carl Zeiss) with a Mi- Research Ethics Board. The neutrophils were purified using a Polymor- crocolor RGB filter. phprep gradient separation procedure according to the manufacturer’s in- structions. Isolated neutrophils were resuspended in PBS containing 1 mM Cell spreading calcium chloride, 1 mM magnesium chloride, and 10 mM glucose at a ␤ ϳ ϫ 6 Neutrophils, either treated with TAT- -gal, TAT-WD, or untreated, were concentration of 1 10 cells/ml and kept at 4°C before use normally ␮ Ϫ7 Ͼ plated on 20 g/ml fibronectin-coated coverslips with or without 10 M within 1–2 h (and never 6 h) of isolation. To ensure that cell viability was fMLF for 3 min at room temperature (RT) to allow cell spreading. The cells not compromised during preparation, the responsiveness to fMLF (chemo- were then fixed in 4% paraformaldehyde for 1 h and washed three times taxis and spreading, described below) was tested in all cases and nonre- with PBS. The area of each cell was outlined and measured using Image J sponsive preparations were discarded. Giemsa stain assay was used to eval- software. Data of the two-dimensional surface areas per cell are means Ϯ uate the purity of the neutrophils, according to the manufacturer’s protocol. ␮ SE of three independent experiments with at least 30 cells counted in each Briefly, 30 l of freshly isolated neutrophils were loaded on the glass slide case. The t tests were used for pairwise comparisons and p Ͻ 0.05 was and a thin film was made using an applicator stick. Air-dried films were considered to be statistically significant. fixed by dipping the film in a Coplin jar containing absolute methanol for 10 min. Staining with Giemsa solution (1/20 dilution, v/v) was conducted Adhesion for 20 min. Films were rinsed with distilled water and allowed to air dry. The slide was mounted and the neutrophils and other cell types were ex- Adhesiveness of neutrophils was measured by allowing ϳ10,000 neutro- amined using a Leica DB LB2 microscope. phils to briefly adhere at 37°C/well of a fibronectin (20 ␮g/ml) coated 96-well U-shaped plate. Purified human neutrophils were labeled with cal- DNA constructs, protein purification, and RT-PCR amplification cein AM for 10 min following three washes with PBS. Neutrophils were then pretreated for 30 min with TAT-WD or TAT-␤-gal as control, and TAT-WD and TAT protein purification were described previously ( 16). To allowed to adhere in PBS containing 1 mM calcium chloride, 1 mM mag- generate the construct GST-WT-coronin, the WT-cor-GFP construct was nesium chloride, and 10 mM glucose for 10 min at 37°C before removing Xho Age digested with I and I, subcloned into the pTAT vector, and then nonadhesion cells by spinning plate upside down at 100 ϫ g for 1 min. The Xho Nco digested with I and I and resubcloned into the pGEX-KG vector bound cells were quantified in a fluorescence plate reader (Spectra MSX containing the GST tag. GST-coronin was expressed and affinity purified Gemini EM) at Ex 497/Em 517, then measured again after removing the by a standard procedure. Briefly, the pGEX-coronin construct was trans- nonadherent cells by centrifugation. The adhesive index was calculated as formed into DH5␣, and grown in Luria Bertani broth to OD ϭ 0.6 at 37°C Ϯ ␤ a ratio of the second measurement to the first. Data are means SE of indicates p Ͻ 0.05 with at least (ء) and induced with 0.2 mM isopropyl -D-thiogalactoside (IPTG) for 4 h. three independent experiments. Asterisk Cell pellets were lysed in a French Press in STE (150 mM NaCl, 50 mM 6 wells counted in each. Tris-HCl, 1 mM EDTA, 1 mM DTT, 5 ␮g/ml leupeptin (pH 8.0)), and centrifuged at 12,000 ϫ g for 30 min. The supernatant was applied to a glutathi- Immunofluorescence and phalloidin staining one-Sepharose 4B (Qiagen) column. The GST-coronin fusion protein was eluted with STE containing 20 mM reduced glutathione. Total RNA from To detect endogenous coronin-1 protein distribution, neutrophils treated white blood cells or neutrophils was extracted using Qiamp RNA blood mini with 10Ϫ7 M fMLF were allowed to spread on 25-mm fibronectin-coated kit (Qiagen) according to the manufacturer’s protocol. RT-PCR was used to coverslips at 37°C. The cells were fixed after 0 and 1 min in 4% parafor- amplify partial cDNAs of all seven coronins by using the one-step RT-PCR kit maldehyde for 1 h, and permeabilized for 15 min using 0.1% Triton X-100 The Journal of Immunology 5771

FIGURE 2. Distribution and concentration of coronin-1 in human neutrophils. Neutrophils treated with 10Ϫ7 M fMLF were loaded on 25-mm fibronectin-coated coverslips. The cells were fixed at 0 (a) and 1 min (b) following adhesion with 4% paraformaldehyde and permeabilization and immunological staining with rabbit anti-coronin-1. A projection of 35 Downloaded from 0.15-␮m Z-stack images collected using a Quorum-spinning disk confocal microscope is shown, with a single middle slice shown in the insets (a1, FIGURE 1. Comparison of RNA expression of all white blood cells and b1). Volocity software was used to measure total or lamellipodial volume neutrophils in seven coronin genes. Total RNA from white blood cells (a) of each cell and the sum pixel intensity. Scale bar in a,5␮m. c, The or neutrophils (b) was extracted using the QiAmp RNA blood mini kit endogenous neutrophil coronin-1 (indicated by open arrow) was quantified (Qiagen). RT-PCR were used to amplify partial cDNAs of all seven coro- by immunoblotting lysate from ϳ3000 cells (first lane, c) on blots that nins in human neutrophils by using the one step RT-PCR kit (Qiagen).

included defined amounts of rGST-coronin-1 (closed arrow) and probed http://www.jimmunol.org/ RT-PCR products of neutrophils and whole white blood cells were frac- with Ab specific to coronin-1. A representative of three independent ex- tionated verified on a 1% agarose gel and verified by DNA sequencing. periments is shown. Products were electrophoresed on the same gels, but separated in the figure for presentation in order. Molecular mass markers are indicated at the right. Coronin numbering follows that of Rybakin and Clemen (5). ber was then moved to another well for the next time period and the number of calcein-AM positive cells that migrated into the lower well was summed. The data represent the mean value (SEM) from at least three containing 100 mM glycine. The neutrophils were immunologically independent experiments. To determine the concentration dependence of stained with rabbit anti-coronin-1 (1/5000) for 1 h, washed, then incubated this effect, varying concentrations of fMLF ranging from 2 ϫ 10Ϫ5 to 2 ϫ Ϫ8 with Cy3 donkey anti-rabbit secondary Ab (1/2000) for another 1 h. To 10 M were coated on the bottom wells of the chamber and neutrophils by guest on September 23, 2021 determine cell volumes, a z-series of images were captured using a Quorum were allowed to migrate at 37°C for 2 h. spinning disk confocal microscope (Quorum Technologies) and Volocity software (Improvision) was used to determine total and lamellipodium vol- Oxidase analysis ume and sum pixel intensity for each cell. In all cases, subthreshold in- ␤ tensities were used to ensure that the signals were not saturated. Neutrophils were treated either with TAT- -gal, TAT-WD protein, or left To label F-actin, rhodamine-phalloidin was diluted to 0.4 U/ml in PBS untreated (control) for 30 min with rotation at room temperature. To measure NADPH oxidase activity, DHR 1,2,3 (up to 2 ␮M) was added for and incubated with cells for 30 min. Coverslips were washed in PBS and ␮ mounted using DakoCytomation fluorescent mounting medium. Cells were another 20 min while rotating at room temperature. PMA (2 M) or noth- imaged using a Zeiss LSM510 laser scanning confocal microscope. ing (control) was added and cells were incubated for 10 min at 37°C. Neutrophils were then fixed in 4% paraformaldehyde for 1 h and washed Phagocytosis three times with PBS. Oxidase activity was measured as the average fluorescence of 10,000 cells (at least 104 gated events were recorded in each Suspended neutrophils were left untreated (control) or either treated with experiment) by using flow cytometric analysis. The index data are means Ϯ TAT-␤-gal or TAT-WD protein for 30 min before exposure to polystyrene SE of three independent experiments. beads opsonized with human IgG and prestained with Cy2-conjugated anti- Alternatively, superoxide (the initial product formed by the NADPH human Ab. Cells were then allowed to adhere for 5 min in ice-cold HEPES- oxidase) was measured over time by determining superoxide dismutase- buffered RPMI to coverslips coated with fibronectin. Unbound beads were inhibitable cytochrome c reduction (17). Briefly, a total of 125 ␮l of neu- washed away with ice-cold PBS. The binding index was calculated, using trophils in each assay well were pretreated either with TAT-␤-gal, or a phase contrast microscopy, as the average number of remaining beads TAT-WD protein (ϳ10,000 cells), then resuspended in PBS supplemented associated with each cell. Alternatively, neutrophils were warmed to 37°C with divalent cations and glucose (0.9 mM CaCl2, 0.5 mM MgCl2, and 7.5 and allowed to undergo phagocytosis for 5 min. After phagocytosis, neu- mM glucose) (PBSG). 75 ␮M of cytochrome C with or without superoxide trophils were fixed in 4% paraformaldehyde at 4°C for 1 h without per- dismutase (60 ␮g/ml) was then added for 10 min at 37°C. The neutrophils meabilization and labeled with Cy3 Ab. The phagocytic index was calcu- were then incubated with or without PMA (2 ␮M) before placing the 96- lated as the number of internalized beads, which are Cy2 positive and Cy3 well microplate into the reading chamber with agitation at 37°C. The ab- negative. The index data are means Ϯ SE of three independent experi- sorbance at 550 nm was quantified in a microplate reader (Versa Max) in ments, with at least 30 cells counted in each case. kinetic mode for 20 min, acquiring readings at 10-s intervals. Neutrophil migration Secretion assay The calcein AM-labeled neutrophils were treated with either TAT-␤-gal, Neutrophils were treated either with TAT-␤-gal, TAT-WD protein, or left TAT-WD protein for 20 min with rotation at room temperature. 50 ␮lof untreated (control) for 30 min with rotation at room temperature. To mea- suspended cells (ϳ10,000 cells) in RPMI 1640 without serum were loaded sure CD63 externalization as a measure of secretion, the cells were treated on a Transwell membrane (3-␮m pore size and 6-mm diameter) in a 12- with fMLF (10Ϫ7 M), Ionomycin (5 ␮M), cytochalasin D (1 ␮M) and well plate. To keep a stable chemotaxis gradient, 10 ␮l of low melting fMLF (10Ϫ7 M) in combination, or control in PBS containing 1 mM CaCl, agarose gel containing 10Ϫ6 M fMLF was coated on the bottom wells of 1 mM MgCl, and 10 mM glucose for a further 10-min incubation at 37°C. chemotaxis chamber. Neutrophils were allowed to migrate across the Neutrophils were then fixed in 4% paraformaldehyde for 1 h and washed Transwell membrane into the lower well along a gradient of chemoattrac- three times with PBS. The neutrophils were incubated with mouse anti- tant in RPMI 1640 without serum at 37°C for 30 min intervals. The cham- CD63 (1/100) for 1 h, washed, then incubated with goat anti-mouse Alexa 5772 CORONIN IS REQUIRED FOR NEUTROPHIL CHEMOTAXIS

Table I. Determination of coronin-1 concentration in neutrophilsa

Total Cell Lamellipodia

Mean Ϯ SE Mean Ϯ SE

Volume ␮m3 (n ϭ 10 cells) 369.2 Ϯ 17.95 31.95 Ϯ 3.49 Sum of pixel intensity 6.83 ϫ 107 Ϯ 3.9 ϫ 106 1.26 ϫ 107 Ϯ 1.86 ϫ 106 Coronin concentrationb 0.135 ng/369 ␮m3 (0.366 pg/␮m3) 0.025 ng/31.95 ␮m3 (0.806 pg/␮m3)

a The volume of fMLF-stimulated human neutrophils was determined by measuring the volume of 10 cells stained with anti-coronin-1 Ab following three-dimensional image reconstruction. Similarly, the average volume of the lamellipodium at the leading edge of cells was also measured in this way. For cells in which the projected image signal intensity was not saturated, the average pixel intensity over the entire cell was compared with that found only at the ruffle, to determine the proportion of coronin-1 in the cell that was distributed at the ruffle. The coronin-1 concentration was measured by comparing the intensity of the protein signal from a known number of cells with that of defined amounts of recombinant protein, as shown in Fig. 2c. Densitometric scans from a variety of blots exposed for different times were used to ensure that the signal was in the linear range of the film. b Coronin-1 concentration is about 0.135 ng/cell and 1.56% of total protein in human neutrophils.

488 secondary Ab (1/1000) for another 1 h. CD63 secretion was measured by using ﬂow cytometric analysis as above. Downloaded from Puriﬁcation of GST-FL and WD coronin Bl21(DE3) LysS cells transformed with either GST, GST-FL-coronin, or GST-WD-coronin were induced with 0.5 mM IPTG for4hat30°C. Pel- leted cells were frozen at Ϫ80°C and then lysed in STE buffer (10 mM Tris-Cl (pH 8.0), 150 mM NaCl, 1 mM EDTA) with the following addi- tions: 1% Triton X-100, 1% L-sarcosine, 5 mM DTT, 200 ␮g/ml lysozyme, 5 mM benzamidine, 1 mM PMSF, 10 ␮g/ml leupeptin, and 10 ␮g/ml http://www.jimmunol.org/ pepstatin. The lysate was passed through a French Press (1000 ␺ ϫ 2) and subsequently centrifuged at 12,000 rpm for 10 min. The supernatant was then passed through a glutathione-Sepharose 4B column (Amersham) and GST fusion proteins were eluted with 10 mM reduced glutathione in 50 mM Tris-Cl (pH 8.0) buffer (elution buffer). by guest on September 23, 2021

FIGURE 4. Effects of TAT-WD on migration of neutrophils through Transwell filters. a, The calcein AM (1:1000) labeled neutrophils were treated with either TAT-␤-gal or TAT-WD protein for 20 min with rotation at room temperature. A total of 50 ␮l of suspended cells in RPMI 1640 were loaded in each Transwell suspended in a well of a 12-well plate. Cells FIGURE 3. Accumulation of coronin and actin at the phagocytic cup. were allowed to migrate across the Transwell membrane into a lower well Suspended neutrophils were mixed with human IgG-opsonized beads and chamber along a gradient of fMLF in RPMI 1640 at 37°C for four 30-min allowed to adhere to 20 ␮g/ml fibronectin-coated coverslips for 5 min in intervals. The number of calcein-positive cells that had migrated into the ice-cold RPMI 1640. The neutrophils were warmed to 37°C to allow lower wells was counted using a ϫ40 objective on an inverted fluorescent phagocytosis to proceed for 1 min (a–c), 3 min (d–f), or 5 min (g–i), then microscope after moving the Transwell into the next well. The data rep- were fixed. After fixation, neutrophils were stained with rhodamine-phal- resent the mean value (SEM) from three independent experiments. b, loidin to detect F-actin (a, b, and g) and antiserum for coronin-1 (b, e, and Similar experiments were performed using different concentrations of h). c, f, and i, Differential interference contrast images of the cells. Filled fMLF. In this case the cells were collected at a single time point (2 h). arrowheads indicate the beads at unsealed phagocytic cups, open triangles Data are from three independent experiments. In the presence of TAT- indicate recently sealed phagosomes, and open arrowheads indicate phago- WD, no significant migration was measured at any concentration of somes that no longer stain for coronin-1. Scale bar, 5 ␮m. fMLF used. The Journal of Immunology 5773 Downloaded from

FIGURE 5. Effects of TAT-WD on adhesion of neutrophils. Calcein AM-labeled neutrophils were treated with TAT-WD or TAT-␤-gal for 30 min at 37°C, and were loaded on ﬁbronectin (20 ␮g/ml) coated 96-well

U-shape wells for 10 min at 37°C to allow cell spreading. The number of http://www.jimmunol.org/ bound cells was first measured by putting the plate into a fluorescence plate reader (Spectra MSX Gemini EM) to measure the basal fluorescence signal of each well at Ex 497/Em 517. Adhesion was then measured by measuring the fluorescent signal under the same conditions after removing nonadher- FIGURE 6. Effects of TAT-WD on neutrophil spreading. Neutrophils, ent cells by spinning the plate upside down at 100 ϫ g for 1 min. The untreated control (a and d), or treated with either TAT-␤-gal (b and e)or adhesive index was calculated at a ratio of second to first measurement. TAT-WD (c and f), were plated on 20 ␮g/ml fibronectin-coated coverslips p Ͻ 0.05 with Ϫ ,ء .Data are means Ϯ SE of three independent experiments with 10 7 M fMLF (a–c) or without fMLF (d–f) for 3 min at RT to allow at least six wells counted in each. cell spreading. Scale bar, 20 ␮m. To determine the extent of spreading, the

area of each cell was outlined by using Image J software and tabulated (g). by guest on September 23, 2021 These data are means Ϯ SE of three independent experiments with at least .p Ͻ 0.05 ,ء .Pyrene actin assay 30 cells counted in each case Purified Arp2/3 complex, a GST-fusion of the verprolin-cofilin-acidic (VCA) domain of N-WASp, and rabbit skeletal-muscle G-actin labeled with pyrene with 20% efficiency were obtained from Cytoskeleton Inc. Actin polymerization was analyzed by the method of Uruno et al. (18). can be seen in Fig. 1a, PCR products can be seen for all isoforms Briefly, G-actin was diluted to 0.4 mg/ml (10 ␮M) in G-buffer (5 mM of coronin in a mixture of human white blood cells. The identity of Tris-Cl (pH 8.0), 0.2 mM CaCl2, 0.2 mM ATP, and 0.5 mM DTT) and all of the products was subsequently confirmed by DNA sequenc- nucleation centers were removed by centrifugation at 200,000 ϫ g for2h. ␮ ␮ ing. In contrast to the white blood cell mixture, neutrophils purified Polymerization was initiated by adding 100 l of pyrene G-actin to 200 l Ͼ of 1.5ϫ polymerization buffer (7.5 mM Tris-Cl (pH 7.5), 1.5 mM EGTA, to 97% based on Giemsa staining (data not shown) expressed

0.15 mM CaCl2, 4.5 mM NaN3, 75 mM KCl, 3 mM MgCl2, 0.75 mM DTT, transcripts for five of the isoforms, while coronins-5 and -6 were and 0.3 mM ATP). The final actin concentration was ϳ3.3 ␮M. Polymer- not detected (Fig. 1b). Unfortunately, due to the lack of specific ization was then monitored by measuring the increase in fluorescence using Abs, we were unable to determine whether the protein is expressed a SpectraMAX Gemini EM fluorometer (Molecular Devices) with 365 Ϯ 9 nm excitation and 407 Ϯ 9 nm emission filters. Where indicated, elution for each coronin isoform. However, because coronin-1 has been buffer or purified proteins (Arp2/3, 30 nM; GST-VCA, 100 nM; GST-FL the best characterized to date, and because reagents exist for its coronin, 300 nM; GST-WD coronin, 300 nM; and GST, 300 nM) were analysis, we subsequently focused on this isoform. added to 1.5ϫ polymerization buffer before the addition of pyrene G-actin. Coronin-1 accumulates at the leading edge of migrating human Results neutrophils Several isoforms of coronin are expressed in neutrophils If coronin-1 is involved in chemotaxis in neutrophils, we would Coronin is an actin-binding protein that has been implicated in expect that it would undergo changes in distribution during cell chemotaxis in Dictyostelium. However, its role in this process in migration. We therefore set out to define the distribution of coro- other organisms has not been defined. We were particularly inter- nin-1 in resting human neutrophils and in neutrophils that have ested in investigating the role of coronin proteins in the chemotaxis been exposed to the chemoattractant peptide fMLF. As shown in of mammalian cells, and primary human neutrophils provide an Fig. 2a, in resting neutrophils that have recently adhered to glass excellent model of chemotactic migration in vitro. To understand coverslips, coronin-1 appears evenly distributed around the cell coronin function in neutrophils, we first set out to characterize the with higher concentrations near the plasma membrane. Following expression patterns of seven coronin genes in the human neutro- exposure to fMLF, the neutrophils become highly polarized and phil. To do so, we created pairs of primers specific for each human coronin-1 concentrates in the lamellipodium at the leading edge of isoform and performed RT-PCR on mRNA derived from human the cell (Fig. 2b) where actin accumulation is also found (not neutrophils or from pooled white cells from blood as a control. As shown). 5774 CORONIN IS REQUIRED FOR NEUTROPHIL CHEMOTAXIS

FIGURE 7. Effect of TAT-WD on actin distribution of neutrophils. Downloaded from Neutrophils were treated with either TAT-␤-gal (a and b) or TAT-WD (c and d) protein for 30 min with rotation at room temperature, and allowed to adhere to 20 ␮g/ml ﬁbronectin-coated coverslips with (a and c) or without (b and d)10Ϫ7 M fMLF for 3 min in RPMI 1640. The cells were ﬁxed, permeabilized, and stained with rhodamine-phalloidin to detect F-actin. TAT-␤-gal-treated cells appeared normal with most actin associated with the basal surface of the cell (a and b). The TAT-WD-transduced cell had http://www.jimmunol.org/ dramatically altered actin morphologies in both fMLF and control conditions. TAT-WD-treated cells were rounder and actin was associated with the membrane at all planes of the cell (a section through the middle of the cell is shown in c and d). Inset images taken at the basal surface of the cell (c1, d1) revealed few adhesion structures. Scale bar, 5 ␮m.

To determine the degree of recruitment of coronin-1 to the leading edge of the cell, we ﬁrst estimated the concentration of coro- FIGURE 8. Phagocytosis was inhibited by TAT-WD in neutrophils.

␤ by guest on September 23, 2021 nin-1 in neutrophils. A protein lysate from 3000 neutrophils was Neutrophils treated with either TAT- -gal (a and c) or TAT-WD (b and d) electrophoresed and blotted together with defined amounts of re- protein were mixed with human IgG-opsonized beads, which was prela- beled with Cy2 Ab, and allowed to adhere to 20 ␮g/ml fibronectin-coated combinant coronin-1 and probed with anti-coronin-1 antiserum coverslips for 5 min in ice-cold HEPES-buffered RPMI 1640. Unbound (Fig. 2c). Densitometric scans of the recombinant proteins were beads were washed away with ice-cold PBS. The binding index was cal- plotted, and the amount of coronin-1 protein in the lysate was culated using a phase contrast microscope to count the average number of interpolated from the linear portion of curves derived from several remaining beads bound per cell (e). Alternatively, neutrophils were repeats of this experiment. Based on this approach we were able to warmed to 37°C and allowed to undergo phagocytosis for 5 min. After estimate that each neutrophil contains ϳ0.83 pg of coronin-1 pro- phagocytosis, neutrophils were fixed in 4% paraformaldehyde at RT for 1 h tein (Table I). We then estimated cell volumes using three-dimen- without permeabilization and labeled with Cy3 Ab. a and b, External and sional reconstruction of confocal images and determined that the internal beads (Cy2 positive and Cy3 negative), while c and d show ex- average volume of a resting neutrophil was ϳ360 ␮m3. Taken ternal beads (Cy2 positive and Cy3 positive). The phagocytic index was together with the protein estimates, this indicates that the concen- calculated by counting the number of Cy2-positive and Cy3-negative beads associated with cells (e). The index data are means Ϯ SE of three inde- tration of coronin-1 in neutrophils is ϳ42 ␮M. p Ͻ ,ء .pendent experiments, with at least 30 cells counted in each case To determine the concentration of coronin-1 within leading edge 0.05. Scale bar, 5 ␮m. lamellipodia we determined the volume of this portion of the cell, again by three-dimensional reconstruction of confocal slices. We then estimated the proportion of the coronin-1 that was localized to nized RBC (16), and in both the neutrophil model system HL-60 the lamellipodium. Cells that had been immunostained with anti- cells (19) and in primary neutrophils (20) to opsonized zymosan. coronin-1 Ab were analyzed and the sum pixel intensities of those However, it has also been noted that immunoreactivity to coro- cells in which the signal was not saturating were measured. The nin-1 (also called TACO) can be retained at the phagosome fol- sum pixel intensity of the entire cell was then compared with that lowing internalization of Mycobacterium species and it has been of the lamellipodium to determine the fraction of coronin-1 that suggested that this retention may contribute to the inhibition of was recruited to this structure. We approximate that 31% of the phagosome maturation (21). We therefore set out to determine coronin-1 is localized within the lamellipodia, and these structures whether coronin-1 accumulates at the forming phagosome of ad- average a volume of ϳ35 ␮m3, resulting in a local coronin-1 con- herent neutrophils exposed to opsonized latex beads, and how long centration of 146 ␮M (Table I). it is retained. As can be seen in Fig. 3 (top panels) both F-actin and coronin-1 appear at the phagosomal cup within 1 min of interac- Coronin-1 is also transiently recruited to the nascent tion with opsonized particles, even before it has sealed (arrows). phagosome Within 3 min, many of the particles have been fully engulfed and Coronin-1 has been previously shown to be transiently recruited to by this time some particles have begun to shed the coronin-1 from phagosomes in Dictyostelium (2), in macrophages ingesting opso- the inner surface of the phagosome (open arrowheads). By 5 min, The Journal of Immunology 5775

FIGURE 9. Oxidase activity is not affected by TAT-WD. a, Neutrophils were treated with either TAT-␤Gal or TAT-WD protein for 30 min before incubation with dihydrorhodamine 1,2,3 (2 ␮M). The cells were then activated PMA (2 ␮M) or vehicle for a further 10 min at 37°C, ﬁxed in 4% Downloaded from paraformaldehyde, and measured for oxidase activity by ﬂow cytometric analysis (at least 104 gated events were recorded in each experiment). The index p Ͻ 0.05. b, The superoxide dismutase-sensitive component of the reduction of cytochrome c ,ء .data are means Ϯ SE of three independent experiments was monitored spectroscopically as an increase in absorbance at 550 nm, and presented as nanomoles of superoxide using Beer’s law, as described previously (17). Values plotted are the means Ϯ SE of three independent experiments.

many of the particles are no longer decorated with coronin-1 (open ment using different starting concentrations of fMLF and found http://www.jimmunol.org/ arrows). In all cases the time course of coronin-1 accumulation and that while the control cells were sensitive to fMLF over a defined release mirrored that of F-actin. concentration range, the TAT-WD-treated neutrophils were unable to chemotax at any of the fMLF concentrations tested (Fig. 4b). TAT transduction of dominant-negative coronins inhibits chemotaxis The distribution of coronin-1 at the leading edge of migrating cells TAT-WD inhibits adhesion and spreading of neutrophils and at the phagocytic cup suggests that it may participate in reg- To gain insights into the mechanisms behind the inhibition of che- ulating actin turnover in these dynamic processes. However, neu- motaxis, we determined whether the neutrophils were altered in trophils are refractory to transfection, limiting the number of ap- their ability to adhere to the substratum. We had previously shown by guest on September 23, 2021 proaches that can be used to gain insights into protein function. We that inhibition of coronin-1 function in RAW 264.7 cells resulted have previously described a dominant-negative form of coronin-1 in their rounding and detachment from the substratum. To measure that inhibited the function of this protein during phagocytosis in adherence, we allowed neutrophils to bind to fibronectin-coated the murine macrophage RAW 264.7 cell line. This involved the multiwell plates for 30 min, then dislodged weakly adherent cells transduction of a domain of coronin-1 containing the WD repeat with a light centrifugal force away from the substratum. As shown sequences into cells acutely through the addition of a membrane in Fig. 5, TAT-WD-treated neutrophils were much less able to permeable TAT transduction domain (TAT-WD) (16). To first adhere tightly to the substratum. confirm that neutrophils would be efficiently transduced, we incu- Because the TAT-WD-treated cells were less adherent, we bated neutrophils or HeLa cells in the presence recombinant forms asked whether they were less able to spread on the substrate. Neu- of either ␤-galactosidase or ␤-galactosidase fused to the TAT trophils treated with either TAT-␤-gal, TAT-WD or left untreated, transduction peptide (TAT-␤-gal) and found that both cell types were allowed to spread on fibronectin-coated coverslips in the ab- were readily transduced by the TAT-␤-gal but not by ␤-galacto- sence or presence of fMLF for 3 min, then photographed (Fig. 6, sidase alone (data not shown). a–f). The surface areas of the cells were then measured and, as We then assayed the ability of dominant-negative coronin-1 to shown in Fig. 6g, there was little difference in area of cells that had inhibit cell migration. Neutrophil migration was measured across a not been exposed to fMLF, but spreading in the presence of fMLF permeable Transwell membrane. Transwell chambers were placed was significantly reduced in TAT-WD-treated cells. in culture dishes containing fMLF and the number of neutrophils To understand the mechanism responsible for the defect in able to migrate across the Transwell membrane into the dish was spreading and adhesion, we examined the actin distribution in counted. Transwell chambers were moved to fresh dishes after TAT-transduced neutrophils that had been incubated with or with- each 30-min interval to ensure a gradient was maintained, and the out fMLF for 3 min. In cells transduced with the control TAT-␤- cumulative number of cells that migrated was determined. As gal protein, the majority of F-actin was detected at the bottom of shown in Fig. 4a, treatment of neutrophils with TAT-WD signif- the cell in punctate structures (Fig. 7a), and in the presence of icantly inhibited their ability to migrate across the membrane, fMLF F-actin was also recruited to ruffles at the periphery of the compared with control cells that had been treated with TAT-␤-gal. cell membrane (Fig. 7b). In contrast, in TAT-WD-treated cells, the Because we coated the base of the dishes with fMLF embedded in F-actin was found around the entire cell membrane, often in blobs soft agar to maintain a standing gradient, we could not be sure of or associated with membrane blebs (Fig. 7, c and d) and very little the precise concentration of fMLF that the cells received, and be- remained associated with the bottom of the cell (Fig. 7, c1 and d1). cause sensitivity to fMLF follows a bell-shaped curve, it remained It therefore appears that TAT-WD altered actin assembly, prevent- possible that TAT-WD did not inhibit chemotaxis, but simply al- ing the formation of adhesive structures needed for binding, tered the sensitivity to fMLF. We therefore repeated the experi- spreading, and chemotaxis. 5776 CORONIN IS REQUIRED FOR NEUTROPHIL CHEMOTAXIS

FIGURE 10. Effect of GST-coronin-1 proteins on Arp2/3-dependent actin polymerization. a, Coomassie staining of GST-FL coronin-1 (cor) and GST-WD cor purified from bacterial lysates. b, The polymerization of pyrene-labeled G-actin (final concentration, 3.3 ␮M) was as- sessed by measuring the increase in fluorescence intensity over time. Where indicated, Arp2/3 (30 nM) and GST-VCA (100 nM) were added alone (Ⅺ) or together with either buffer control (vehicle, छ, GST (300 nM, E), GST-FL cor (300 nM, F) or GST-WD cor (300 nM, Œ).

TAT-WD also inhibits phagocytosis, but not secretion or oxidase phagosome in RAW 264.7 macrophage cells was inhibited by activation TAT-WD treatment and this correlated with a reduced accumula- Downloaded from We then examined the effect of TAT-WD on other cellular pro- tion of actin (15). To determine whether the WD domain of coro- cesses in which actin polymerization may play a role. To examine nin-1 inhibited Arp2/3, we tested its effect on a pyrene actin po- the role of coronin-1 at the phagocytic cup, neutrophils treated lymerization assay. Rather than using the TAT-tagged constructs, with either TAT-␤-gal or TAT-WD were incubated in the presence we used GST-fusions of the WD domain and FL coronin-1, be- of opsonized beads on ice to allow binding to the neutrophils, then cause the FL protein was extremely insoluble and the GST domain contributed to its solubility. As shown in Fig. 10a, FL coronin-1 is warmed to 37°C for 5 min to allow phagocytosis to proceed. The http://www.jimmunol.org/ efficiency of phagocytosis was determined by staining the external poorly expressed in bacteria, compared with the WD domain particles of nonpermeabilized neutrophils with anti-human IgG alone. Ab. As can be seen in Fig. 8, a–d, there were no external beads As has been previously seen, actin alone can polymerize but associated with neutrophils incubated with TAT-␤-gal whereas does so with a long lag phase due to its weak nucleating capacity those cells treated with TAT-WD had ingested significantly fewer (Fig. 10b). When incubated in the presence of Arp2/3 and its ac- particles. These data are presented quantitatively in Fig. 8e. The tivator, the VCA domain of WASp, nucleation is accelerated, lead- number of beads associated with each cell was measured as the ing to a leftward shift of the curve. As expected, based on the yeast binding index, and as shown in Fig. 8e, this was not affected by studies, FL GST-coronin1 significantly inhibited Arp2/3-depen- TAT-WD treatment. dent nucleation, although not to the extent seen for the yeast pro- by guest on September 23, 2021 In contrast to phagocytosis and chemotaxis, other properties of tein. The inability of this protein to completely inhibit actin nu- the neutrophils were unaffected by TAT-WD treatment. For ex- cleation is likely due to the insolubility of the mammalian protein ample, CD63 externalization as a marker of secretion was mea- and the difficulty expressing it at high concentrations. In contrast, sured by flow cytometry and we found that control cells, TAT-␤- the GST-WD domain had no effect on the ability of Arp2/3 to gal and TAT-WD-treated cells had similar surface levels of CD63 promote nucleation, despite its much greater solubility. Thus, con- following treatment with fMLF alone, with cytochalasin D and sistent with the yeast coronin protein, the domains of coronin that fMLF, or following ionomycin treatment (data not shown). inhibit Arp2/3 function lie outside the WD repeats. The NADPH oxidase was previously shown to associate in a complex with coronin-1 during phagocytosis (20, 22), raising the Discussion possibility that coronin-1 participates in NADPH oxidase assem- In this report, we have shown that human neutrophils transcribe bly and recruitment to the phagosome. We therefore measured five of the seven coronin genes in contrast to whole white blood oxidase activity by labeling cells with the nonfluorescent precursor cell fractions in which transcripts for all seven genes can be de- dihydrorhodamine 1,2,3, which is converted to a fluorescent prod- tected. This was somewhat surprising because coronins-1 and -7 uct, rhodamine 1,2,3, by reactive oxygen species. Formation of were thought to be primarily expressed in hemopoietic tissue, co- rhodamine 1,2,3 was quantified by flow cytometry. No difference ronins-5 and -6 were thought to be mainly in the brain and coro- in oxidase activity was observed following activation of the neu- nins-2–4 were thought to be broadly expressed (5). The neutro- trophils with phorbol esters in the presence or absence of phils transcribe the hemopoietic and generally expressed forms, TAT-WD (Fig. 9a). In addition, we used a colorimetric assay to but fail to transcribe the two forms abundantly expressed in the measure the kinetics of superoxide production, based on its ability brain. It would be of interest to know which cells in the white to reduce cytochrome c. As shown in Fig. 9b, no significant dif- blood cell fraction transcribe coronins-5 and -6. Our PCR analysis ference was noted in the production of superoxide following phor- was not quantitative, so there could be relatively little mRNAs of bol ester activation in the presence of TAT-WD or the control some isoforms, but clearly multiple coronins mRNAs are present. TAT-␤-gal. Together, these data indicate that the function of co- If each of these transcripts corresponds to a functional protein ronin is required for some but not all of the dynamic changes product, then it is possible that coronin function may be redundant associated with neutrophil activation. in neutrophils, although the sequence similarity between the different isoforms is relatively low. Although they may have over- FL coronin-1, but not TAT-WD, inhibits Arp2/3 function lapping functions, at least some coronins are thought to have dis- In vitro studies had previously shown that a small portion of the C tinct distributions. For example, mammalian coronin-7 was found terminus of yeast coronin inhibits the activity of Arp2/3 (9). We to be localized to the Golgi and its localization did not depend on had previously shown that Arp2/3 accumulation at the nascent actin polymerization (23). The Journal of Immunology 5777

In the case of coronin-1, we have shown that it undergoes re- remodeling. However, it remains possible that the association of organization during the polarization of the neutrophil, accumulat- endogenous coronin-1 to p40phox is not altered in the presence of ing at the leading edge of the migrating cells. We noted that the TAT-WD. concentration increased more than three fold at the leading edge of Although we cannot be certain at this time that all of the coronin the cell, potentially concentrating coronin-associated molecules isoforms are expressed as proteins in neutrophils, their potential that could participate in actin remodeling. This is a region known plurality raises the possibility that the dominant-negative form of to be undergoing dynamic changes in actin structure and a great the protein is not only inhibiting coronin-1, but could be generally deal of actin nucleation and branching occurs through the action of blocking common functions of all coronins. In the case of macro- the Arp2/3 complex. Coronin-1 has previously been shown to as- phages, we were able to use small interfering RNA techniques that sociate with Arp2/3 and in yeast the complex is thought to be allowed us to conclude that coronin-1 function alone was required regulated by its association with crn1p (10). Indeed, we show here for phagocytosis (16). Unfortunately, because neutrophils are that FL recombinant coronin-1 also inhibits Arp2/3-stimulated ac- short-lived and refractory to transfection, we are unable to use tin nucleation, although not as efficiently as was observed for the similar approaches here. Long-term studies of the functions of yeast ortholog. This inefficiency is likely to be due to with the coronin isoforms in neutrophils will require either the use of limited solubility of the recombinant mammalian coronin-1 pro- specific dominant-negative forms for each isoform or the gen- tein, rather than to distinct properties of the protein. However, the eration of tissue-specific isoform knockouts to address their in- WD domains had no effect on this activity, suggesting that their dividual roles. role may be limited to recruitment of the complex. Hence, the recruitment of coronin to the leading edge of the cell could be Acknowledgments Downloaded from important for the actin nucleation and branching activities con- We thank Dr. Satoshi Toyoshima for the coronin-1 cDNA, Dr. Jean Pieters ducted by Arp2/3. for polyclonal anti-coronin-1 Abs, and Dr. Steve Dowdy for pTAT, pTAT- Coronin-1 has also been suggested to serve as a bridge between ␤-galactosidase, and ␤-galactosidase (non-TAT) constructs. the actin cytoskeleton and the plasma membrane due to its membrane association (6). It should be noted that the surface area of the Disclosures leading edge of neutrophils is also vastly increased to extensive The authors have no financial conflict of interest. http://www.jimmunol.org/ ruffling and invaginations of this area (24, 25). We can therefore not distinguish whether the concentration reflects an enrichment of References the protein within the small cytoplasmic volume of the lamellipod 1. de Hostos, E., B. Bradtke, F. Lottspeich, R. Guggenheim, and G. Gerisch. 1991. or if its concentration reflects the enrichment of membrane do- Coronin, an actin-binding protein of Dictyostelium discoideum localized to cell mains to which it may be associated. In either case, the polarized surface projections, has sequence similarities to G-protein ␤ subunits. EMBO J. 10: 4097–4104. accumulation of coronin-1 would be important in ensuring the ac- 2. de Hostos, E., C. RehfeuB, B. Bradtke, D. Waddell, R. Albrecht, J. Murphy, and tin polymerization also occurs in a polarized manner, facilitating G. Gerisch. 1993. Dictyostelium mutants lacking the cytoskeletal protein coronin cell migration. When cells were pretreated with coronin their at- are defective in cytokinesis and cell motility. J. Cell Biol. 120: 163–173.

3. de Hostos, E. 1999. The coronin family of actin-associated proteins. Trends Cell by guest on September 23, 2021 tachment to, and spreading on, the substratum was also impaired Biol. 9: 345–350. because these processes would require dynamic remodeling of ac- 4. Appleton, B. A., P. Wu, and C. Wiesmann. 2006. The crystal structure of murine tin structures. coronin-1: a regulator of actin cytoskeletal dynamics in lymphocytes. Structure 14: 87–96. We had previously used the dominant-negative TAT-WD coro- 5. Rybakin, V., and C. S. Clemen. 2005. Coronin proteins as multifunctional reg- nin-1 fragment to inhibit coronin function in macrophage cells and ulators of the cytoskeleton and membrane trafficking. BioEssays 27: 625–632. 6. Gatfield, J., I. Albrecht, B. Zanolari, M. O. Steinmetz, and J. Pieters. 2005. As- found that this protein inhibited the phagocytosis at a stage con- sociation of the leukocyte plasma membrane with the actin cytoskeleton through sistent with a role in actin polymerization (16). We show here that coiled coil-mediated trimeric coronin 1 molecules. Mol. Biol. Cell 16: phagocytosis in neutrophils is also arrested by treatment of cells 2786–2798. 7. Machesky, L. M., E. Reeves, F. Wientjes, F. J. Mattheyse, A. Grogan, with this protein. Hence, coronins appear to have a conserved role N. F. Totty, A. L. Burlingame, J. J. Hsuan, and A. W. Segal. 1997. Mammalian in phagocytosis by a variety of cell types. actin-related protein 2/3 complex localizes to regions of lamellipodial protrusion A failure to observe any effect of the TAT-WD on oxidase ac- and is composed of evolutionarily conserved proteins. Biochem. J. 328(Pt. 1): 105–112. tivation was somewhat surprising given that coronin-1 was previ- 8. Suzuki, K., J. Nishihata, Y. Arai, N. Honma, K. Yamamoto, T. Irimura, and ously shown to associate directly with the p40phox subunit in a S. Toyoshima. 1995. Molecular cloning of a novel actin-binding protein, p57, phox phox with a WD repeat and a leucine zipper motif. FEBS Lett. 364: 283–288. complex with p47 and p67 (22). In addition, it has been 9. Oku, T., S. Itoh, M. Okano, A. Suzuki, K. Suzuki, S. Nakajin, T. Tsuji, shown that upon treatment of neutrophils with PMA, both p40phox W. M. Nauseef, and S. Toyoshima. 2003. Two regions responsible for the actin and coronin showed redistribution to perinuclear regions and this binding of p57, a mammalian coronin family actin-binding protein. Biol. Pharm. Bull. 26: 409–416. was not observed in patients of chronic granulomatosis disease 10. Humphries, C. L., H. I. Balcer, J. L. D’Agostino, B. Winsor, D. G. Drubin, lacking p47phox or p67phox (22). This led the authors to speculate G. Barnes, B. J. Andrews, and B. L. Goode. 2002. Direct regulation of Arp2/3 that the phox proteins may contribute to the regulation of the actin complex activity and function by the actin binding protein coronin. J. Cell Biol. 159: 993–1004. cytoskeleton through their interaction with coronin. Allen et al. 11. Rodal, A. A., O. Sokolova, D. B. Robins, K. M. Daugherty, S. Hippenmeyer, (20) demonstrated that in patients lacking p47phox, coronin-1 was H. Riezman, N. Grigorieff, and B. L. Goode. 2005. Conformational changes in phox the Arp2/3 complex leading to actin nucleation. Nat. Struct. Mol. Biol. 12: 26–31. still recruited to the phagosome in the absence of p67 and in 12. Liu, C. Z., Y. Chen, and S. F. Sui. 2006. The identification of a new actin-binding time course studies showed that in neutrophils from chronic gran- region in p57. Cell Res. 16: 106–112. ulomatous disease patients, there was accumulation of p47phox and 13. Goode, B., J. Wong, A.-C. Butty, M. Peter, A. McCormack, J. Yates, D. Drubin, phox and G. Barnes. 1999. Coronin promotes the rapid assembly and cross-linking of p67 in the periphagosomal area, but this association was only actin filaments and may link the actin and microtubule cytoskeletons in yeast. transient and was lost upon dissociation of the actin and coronin-1. J. Cell Biol. 144: 83–98. This raised the possibility that coronin-1 could participate in the 14. Gerisch, G., R. Albrecht, C. Heizer, S. Hodgkinson, and M. Maniak. 1995. Che- moattractant-controlled accumulation of coronin at the leading edge of Dictyo- stabilization of the NADPH oxidase complex in conjunction with stelium cells monitored using a green fluorescent protein-coronin fusion protein. p91phox and p22phox. Although the present study does not directly Curr. Biol. 5: 1280–1285. 15. Mishima, M., and E. Nishida. 1999. Coronin localizes to leading edges and is address this possibility, it shows that the activation of the NADPH involved in cell spreading and lamellipodium extension in vertebrate cells. J. Cell oxidase does not depend on the ability of coronin to mediate actin Sci. 112(Pt. 17): 2833–2842. 5778 CORONIN IS REQUIRED FOR NEUTROPHIL CHEMOTAXIS

16. Yan, M., R. F. Collins, S. Grinstein, and W. S. Trimble. 2005. Coronin-1 function neutrophils from patients with X-linked chronic granulomatous disease. Blood is required for phagosome formation. Mol. Biol. Cell 16: 3077–3087. 93: 3521–3530. 17. Cohen, H. J., P. E. Newburger, M. E. Chovaniec, J. C. Whitin, and E. R. Simons. 21. Ferrari, G., H. Langen, M. Naito, and J. Pieters. 1999. A coat protein on phago- 1981. Opsonized zymosan-stimulated granulocytes-activation and activity of the somes involved in the intracellular survival of mycobacteria. Cell 97: 435–447. superoxide-generating system and membrane potential changes. Blood 58: 22. Grogan, A., E. Reeves, N. Keep, F. Wientjes, N. Totty, A. Burlingame, J. Hsuan, 975–982. and A. Segal. 1997. Cytosolic phox proteins interact with and regulate the as- 18. Uruno, T., J. Liu, P. Zhang, Y. Fan, C. Egile, R. Li, S. C. Mueller, and X. Zhan. sembly of coronin in neutrophils. J. Cell Sci. 110: 3071–3081. 2001. Activation of Arp2/3 complex-mediated actin polymerization by cortactin. 23. Rybakin, V., M. Stumpf, A. Schulze, I. V. Majoul, A. A. Noegel, and A. Hasse. Nat. Cell Biol. 3: 259–266. 2004. Coronin 7, the mammalian POD-1 homologue, localizes to the Golgi ap- 19. Itoh, S., K. Suzuki, J. Nishihata, M. Iwasa, T. Oku, S. Nakajin, W. M. Nauseef, paratus. FEBS Lett. 573: 161–167. and S. Toyoshima. 2002. The role of protein kinase C in the transient association 24. Keller, H., and P. Eggli. 1998. Actin accumulation in pseudopods or in the tail of of p57, a coronin family actin-binding protein, with phagosomes. Biol. Pharm. polarized walker carcinosarcoma cells quantitatively correlates with local folding Bull. 25: 837–844. of the cell surface membrane. Cell Motil. Cytoskeleton 40: 342–353. 20. Allen, L. A., F. R. DeLeo, A. Gallois, S. Toyoshima, K. Suzuki, and 25. Pytowski, B., F. R. Maxﬁeld, and J. Michl. 1990. Fc and C3bi receptors and the W. M. Nauseef. 1999. Transient association of the nicotinamide adenine dinu- differentiation antigen BH2-Ag are randomly distributed in the plasma membrane cleotide phosphate oxidase subunits p47phox and p67phox with phagosomes in of locomoting neutrophils. J. Cell Biol. 110: 661–668. Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021