QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 18, Issue 12, 5069-5080, December 2007

This Article Abstract Full Text (PDF) Supplemental Materials All Versions of this Article: E07-05-0428v1 18/12/5069    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lokuta, M. A. Articles by Huttenlocher, A. Search for Related Content PubMed PubMed Citation Articles by Lokuta, M. A. Articles by Huttenlocher, A.

Type I PIP Kinase Is a Novel Uropod Component that Regulates Rear Retraction during Neutrophil Chemotaxis

Mary A. Lokuta^*,, Melissa A. Senetar^,, David A. Bennin^*, Paul A. Nuzzi, Keefe T. Chan, Vanessa L. Ott^*, and Anna Huttenlocher^*,

Departments of Medical Microbiology and Immunology and *Pediatrics, University of Wisconsin, Madison, WI 53706

Submitted May 10, 2007; Revised September 21, 2007; Accepted September 28, 2007
Monitoring Editor: Josephine Adams

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cell polarization is necessary for directed migration and leukocyte recruitment to inflamed tissues. Recent progress has been made in defining the molecular mechanisms that regulate chemoattractant-induced cell polarity during chemotaxis, including the contribution of phosphoinositide 3-kinase (PI3K)-dependent phosphatidylinositol (3,4,5)-trisphosphate [PtdIns(3,4,5)P₃] synthesis at the leading edge. However, less is known about the molecular composition of the cell rear and how the uropod functions during cell motility. Here, we demonstrate that phosphatidylinositol phosphate kinase type I (PIPKI661), which generates PtdIns(4,5)P₂, is enriched in the uropod during chemotaxis of primary neutrophils and differentiated HL-60 cells (dHL-60). Using time-lapse microscopy, we show that enrichment of PIPKI661 at the cell rear occurs early upon chemoattractant stimulation and is persistent during chemotaxis. Accordingly, we were able to detect enrichment of PtdIns(4,5)P₂ at the uropod during chemotaxis. Overexpression of kinase-dead PIPKI661 compromised uropod formation and rear retraction similar to inhibition of ROCK signaling, suggesting that PtdIns(4,5)P₂ synthesis is important to elicit the backness response during chemotaxis. Together, our findings identify a previously unknown function for PIPKI661 as a novel component of the backness signal that regulates rear retraction during chemotaxis.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Neutrophils are critical participants in the innate immune response to inflammatory stimuli such as tissue injury and infection. The rapid recruitment of neutrophils to inflammatory sites requires a highly specialized form of directed cell migration in which shallow gradients of chemoattractant are translated into intracellular signals that establish polarized protrusion in the direction of migration (Devreotes and Janetopoulos, 2003; Niggli, 2003; Parent, 2004). A key component of this process is the asymmetric recruitment of phosphatidylinositol (3,4,5)-trisphosphate [PtdIns(3,4,5)P₃] to the membrane adjacent to the highest concentration of chemoattractant (Meili et al., 1999; Servant et al., 2000). PtdIns(3,4,5)P₃ and the RhoA family GTPases Cdc42 and Rac mediate a positive feedback mechanism that promotes actin polymerization and the formation of a dominant pseudopod at the cell front (Weiner et al., 1999; Servant et al., 2000; Van Keymeulen et al., 2006). In contrast, actin-based protrusions are inhibited at the cell rear, where RhoA-stimulated actomyosin contractility mediates rear retraction (Niggli, 1999; Eddy et al., 2000). Research into the molecular events that regulate chemotaxis has progressed rapidly; however, there is still limited understanding of the molecular components that define the cell rear and mediate retraction during directed cell migration.

Phosphatidylinositol (4,5)-bisphosphate [PtdIns(4,5)P₂] is a membrane phospholipid that regulates a variety of cellular functions including actin cytoskeleton and focal adhesion dynamics, cell signaling, and vesicular trafficking (Ling et al., 2006). PtdIns(4,5)P₂ binds to and regulates the function of many cytoskeleton-associated proteins including talin, vinculin, -actinin, WASP, ezrin-radixin-moesin (ERM) proteins, and cofilin (Ling et al., 2006). The localized synthesis of PtdIns(4,5)P₂ has also been implicated in modulating cell motility by functioning as a substrate for phosphoinositide 3-kinase (PI3K) and phospholipase C to generate PtdIns(3,4,5)P₃ and inositol (1,4,5)-trisphosphate, respectively (Ling et al., 2006). There has been substantial interest in understanding the temporal and spatial regulation of PtdIns(4,5)P₂ synthesis during cell migration. However, because of the relative abundance of PtdIns(4,5)P₂ and the difficulty in detecting localized changes in its production, the temporal and spatial dynamics of PtdIns(4,5)P₂ synthesis during chemotaxis have remained largely undefined.

One mechanism by which PtdIns(4,5)P₂ is generated is through type I phosphatidylinositol phosphate kinases (PIPKI), which produce PtdIns(4,5)P₂ through the phosphorylation of the fifth hydroxyl of phosphatidylinositol (4)-phosphate (PI4P) (Doughman et al., 2003). The PIPKI family includes three isoforms: , β, and . Furthermore, PIPKI mRNA is alternatively spliced to form PIPKI635 and PIPKI661, which differ by a 26-amino acid C-terminal extension. Previous work has shown that the PIPKI family of lipid kinases display distinct subcellular distributions in fibroblasts (Di Paolo et al., 2002; Ling et al., 2002; Doughman et al., 2003). PIPKI targets to the leading edge of cells where it regulates membrane ruffling (Doughman et al., 2003), whereas PIPKI661 targets to focal adhesions and may regulate integrin-mediated adhesions during cell migration via an interaction with the cytoskeletal protein talin (Di Paolo et al., 2002; Ling et al., 2002). However, PIPKI635 does not target to focal adhesions or interact with talin (Ling et al., 2002). Although PIPKI family members have been implicated in cell motility (Ling et al., 2006), the generation of localized PtdIns(4,5)P₂ and functions of PIPKI family members during neutrophil chemotaxis have not been explored.

In this report, we demonstrate that the type I phosphatidylinositol phosphate kinase (PIPKI661), which generates PtdIns(4,5)P₂, is enriched in the uropod during directed migration. Using primary murine neutrophils that express green fluorescent protein (GFP)-tagged PIPKI661 and DMSO-differentiated HL-60 cells (dHL-60) retrovirally infected with GFP-tagged PIPKI661, we show that enrichment of PIPKI661 at the cell rear occurs early during chemoattractant-induced cell polarization and that cells overexpressing PIPKI661 form a prominent uropod. Our data indicate that although lipid kinase activity of PIPKI661 is not required for targeting to the rear, overexpression of kinase-dead PIPKI661 compromised uropod formation, and retraction of the cell rear in dHL-60 cells. Finally, we were able to detect enrichment of the PIPKI661 product, PtdIns(4,5)P₂, at the uropod during chemotaxis, and this enrichment was impaired by expression of the kinase-dead PIPKI661. Taken together, our findings identify PIPKI661 as a novel component of the uropod and suggest that PIPKI661 may regulate rear release via the localized generation of PtdIns(4,5)P₂ and contribute to backness signaling during chemotaxis.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Isolation of Primary Neutrophils and Nucleofection
C57Bl/6 mice (Harlan Sprague-Dawley, Madison, WI) were killed in accordance with an institution approved protocol, and the long bones were removed and flushed with cold PBS/HSA/hep (Dulbecco's phosphate-buffered saline without Ca²⁺ or Mg²⁺ containing 0.15% human serum albumin and 100 U/ml preservative-free heparin) using a 30-gauge needle. Cells were dispersed using an 18-gauge needle, pelleted, and resuspended in PBS/HSA/hep. Subsequently, cells were passed through a 70-µm cell sieve, overlayed onto a discontinuous gradient of Histopaque 1083 and Histopaque 1119 (Sigma Aldrich, St. Louis, MO), and spun for 30 min at 700 x g with no brake. Red blood cells were lysed using ACK buffer (155 mM NH₄Cl, 10 mM KHCO₃, and 127 µM EDTA) and washed with PBS/HSA/hep. Cells were resuspended in PBS/HSA/hep and held at 4°C until use. Murine bone marrow–derived neutrophils were nucleofected as previously described (Kunisaki et al., 2006) and maintained at 37°C at 5% CO₂ in the presence of 25 ng/ml recombinant murine GM-CSF (R&D Systems, Minneapolis, MN) for 24 h. The transfection efficiency was generally between 10 and 15% at 24 h, with good viability after 16–24 h. Primary human neutrophils used in RT-PCR and immunoblot analysis were obtained from healthy donors as described previously (Lokuta et al., 2003).

Cell Culture and Retroviral Infection
HL-60 cells (UCSF tissue culture facility) were cultured and differentiated as previously described (Nuzzi et al., 2007b). Phoenix viral packaging cells were transiently transfected by calcium-phosphate precipitation (Nuzzi et al., 2007a). Viral supernatant was harvested and used to retrovirally infect HL-60 cells as previously described (Nuzzi et al., 2007a). Populations of GFP-positive cells were obtained by fluorescence-activated cell sorting (FACS) and verified for expression by immunoblotting.

Time-Lapse Microscopy of Focal Adhesion Dynamics
HeLa cells (ATCC, Manassas, VA) were maintained according to the manufacturer's instructions, transiently transfected with GFP-PIPKI661 and dsRed-paxillin using Lipofectamine (Invitrogen, Carlsbad, CA), plated on 35-mm glass-bottom dishes coated with 10 µg ml^–1 fibronectin, and allowed to adhere for 1 h. Focal adhesion dynamics of GFP-PIPKI661 and dsRed-paxillin was recorded using a Nikon Eclipse TE300 inverted fluorescence microscope (Melville, NY) with a cooled charge-coupled device video camera (Hamamatsu Photonics, Bridgewater, NJ) using a 60x objective and captured into Metamorph V 7.0r2 (Universal Imaging, West Chester, PA) at 2-min intervals for 60 min. Image processing was performed as described (Franco et al., 2004). The results are representative of three separate experiments.

RNA Isolation and RT-PCR
Total RNA was extracted from primary human neutrophils, undifferentiated and differentiated HL-60 cells, HEK 293 cells, Jurkat T-cells, and D10 T-cells using RNA STAT-60 (Iso-Tex Diagnostics, Friendswood, TX). After treatment with DNase (Promega, Madison, WI), RT-PCR was performed with 1 µg of RNA and 40 U RNAsin (Promega) using the One Step RT-PCR kit (Qiagen, Chatsworth, CA) and gene-specific primers as follows: PIPKI forward, 5'-CGCCCAGAGCACCTCAGATG-3'; reverse, 5'-GCTCCACCTGCACTGTAATCTG-3'; PIPKI661 forward, 5'- CGCCCAGAGCACCTCAGATG-3'; reverse, 5'-GCGGGGAGTACACCCAGCTCCTCT-3'; GAPDH forward, 5'-GAGTCAACGGATTTGGTCGTAT-3'; and reverse, 5'-AGTCTTCTGGGTGGCAGTGAT-3'. The following thermal cycling parameters were used: 50°C for 30 min, 95°C for 15 min, 94°C for 1 min, 55°C for 30 s, 72°C for 1 min (35 cycles), and a final extension at 72°C for 10 min. Each RT-PCR sample was resolved by electrophoresis on a 4.25% nondenaturing polyacrylamide gel, stained with ethidium bromide, and analyzed using a UVP Bio-dock system (UVP, Upland, CA). Data are representative of RT-PCR from multiple separate RNA preparations.

Protein Extraction, Antibodies, and Immunoblots
Primary human neutrophils, undifferentiated and differentiated HL-60 cells, and HEK 293 cells were lysed (Nuzzi et al., 2007b), samples were clarified by centrifugation, and protein concentration was determined using a bicinchoninic acid (BCA) protein assay (Pierce, Rockford, IL). Equal amounts of total protein were denatured in SDS sample buffer, resolved on 4–20% gradient SDS-PAGE gels, and transferred to nitrocellulose. Immunoblots were performed using standard conditions with polyclonal antibodies against PIPKI (Ling et al., 2002) or GFP (BD Biosciences, San Jose, CA). IRDye 800CW goat--rabbit IgG (Rockland Immunochemicals, Gilbertsville, PA) and Alexa Fluor 680 goat--mouse IgG (Invitrogen) were used as the secondary antibodies. Immunoblots were imaged on an Odyssey Infrared Imaging System with the Odyssey v2.0 software (LI-COR Bioscience, Lincoln, NE). Data are representative of immunoblots from multiple separate lysate preparations.

Immunofluorescence
Primary murine neutrophils were resuspended in Dulbecco's phosphate-buffered saline (DPBS) alone, or DPBS containing 100 nM formylmethionylleucylphenylalanine (fMLP) and were allowed to adhere for 10 min to glass coverslips coated with 10 µg ml^–1 fibrinogen. Cells were fixed and permeabilized with 6.6% paraformaldehyde, 0.05% glutaraldehyde, and 0.25 mg ml^–1 saponin in PBS, pH 7.2, for 15 min and then quenched with 0.15 M glycine for 15 min. Nonspecific binding was blocked with PBS containing 10% heat-inactivated fetal bovine serum and 0.25 mg/ml saponin at 4°C overnight. Cells were then stained for 30 min with -PIPKI (Ling et al., 2002) and costained with rhodamine-labeled phalloidin (Molecular Probes, Eugene, OR) and DAPI (Molecular Probes). FITC goat -rabbit IgG (Jackson ImmunoResearch, West Grove, PA) was used as the secondary antibody. Detection of active RhoA was performed as previously described (Berdeaux et al., 2004) using GST-rhotekin (Cytoskeleton, Denver, CO). D10 T-cells expressing GFP-PIPKI661 were plated on poly-L-lysine–coated coverslips, stimulated with LTB₄, fixed, and stained with -ERM (Cell Signaling Technology, Beverly, MA). Cells were mounted in mounting media and viewed on a Nikon Eclipse TE300 inverted fluorescence microscope (Melville, NY) using a 100x oil immersion DIC objective. Fluorescent images were digitally acquired using a cooled charge-coupled device video camera (Hamamatsu Photonics) and processed with Metamorph V 7.0r2 (Universal Imaging). Data are representative of a minimum of 50 cells from at least two independent experiments performed in duplicate.

Expression Constructs
Wild-type GFP-PIPKI661 was generated by PCR amplification of a murine PIPKI661 cDNA (Ling et al., 2002; a generous gift from Dr. Richard Anderson) and subcloning into pEGFP (Clontech, Palo Alto, CA) as an N-terminal GFP fusion, which was then subcloned into the empty pMX retroviral vector (a generous gift from Dr. Clive Svendsen). Kinase-dead GFP-PIPKI661 (D253A; Ling et al., 2002) was generated by mutating wild-type pMX-GFP-PIPKI661 with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). GFP-PIPKI635 was generated by PCR amplification of the PIPKI661cDNA (in which a stop sequence was inserted after amino acid 635) and subcloning into the pMX vector. Wild-type and kinase-dead (D253A) GFP-PIPKI661-FLAG were generated by PCR amplification of GFP-PIPKI661 and subcloning into pCDNA3.1(+) (Invitrogen) containing a C-terminal FLAG tag. Wild-type and kinase-dead (D253A) mCherry-PIPKI661 were generated by PCR amplification of PIPKI661 cDNA and subcloning into the pmCherry-C1 vector. pmCherry-C1 was generated by PCR amplification of pRSET-B-mCherry (a generous gift from Dr. Roger Tsien) with primers containing a 5' NheI site and 3' BglII site and subcloning into pEGFP-C1 (Clontech) to replace the enhanced GFP (EGFP) sequence. GFP-PH-PLC and GFP-PHAKT constructs were a generous gift from Dr. Tamas Balla. GFP-Ezrin was kindly provided by Monique Arpin and Richard Lamb. dsRed-paxillin was a kind gift from Rick Horwitz. The accuracy of all constructs was verified by DNA sequencing.

Chemotaxis Assay
For each experiment, 5 x 10⁵ cells were plated in Gey's media (dHL-60) or EGM-2MV (Cambrex Bioscience, Walkersville, MD; neutrophils) for 10 min on a glass-bottom dishes coated with 2.5 µg ml^–1 fibrinogen (Sigma) and 10 µg ml^–1 fibronectin and either left untreated or pretreated for 30 min with 10 µM ROCK inhibitor (Y-27632, Sigma). An Eppendorf Femtotip was loaded with 58 µM C5a, and a chemotactic gradient was formed by slow release of the chemoattractant from the tip into the media using an Eppendorf FemtoJet microinjection system (Westbury, NY) as described (Servant et al., 1999; Lokuta et al., 2003). Chemotaxis was recorded using a Nikon Eclipse TE300 inverted fluorescence microscope with a cooled charge-coupled device video camera (Hamamatsu Photonics) using a 60x oil immersion DIC objective and captured into Metamorph v7.0r2 (Universal Imaging) at 10-s ('') intervals for 10 min. Localization studies were performed on at least three independent samples from multiple cell lines or neutrophil preparations.

Kinase Activity Assays
dHL60 cells that express either GFP, wild-type GFP-PIPKI661, or kinase-dead GFP-PIPKI661 were washed in PBS and resuspended in lysis buffer (0.2% NP-40, 142.5 mM KCL, 5 mM MgCl₂, 10 mM HEPES, pH 7.4, supplemented with protease inhibitor cocktail [P-8340; Sigma], phosphatase inhibitor cocktail [P-5726; Sigma], 2 mM phenylmethylsulfonyl fluoride, 100 mM sodium orthovanadate, 900 mM benzamidine, and 1 mM phenanthroline). Lysates were cleared by centrifugation, and protein concentrations were determined by BCA assay (Pierce). -GFP (Molecular Probes) was used to immunoprecipitate from 200 µg total lysate at 4°C for 3 h with tilting. Immune complexes were washed twice with lysis buffer and once in kinase buffer without ATP or substrate. Reactions were carried out for 15 min at 32°C in 50 µl volume containing 50 mM Tris, pH 7.5, 10 mM MgCl₂, 0.25 mM EGTA, 25 µM PtdIns₄P (Echelon Biosciences, Salt Lake City, UT; P-4016), 50 µM ATP, and 10 µCi [³²P]ATP. PtdIns [³²P]4,5-P₂ generation was terminated by the addition of 100 µl 1 N HCl and extracted with 200 µl chloroform/methanol (1:1). The organic phase containing radiolabeled lipids was washed one time with 80 µl methanol, 1 N HCl (1:1) and spotted on thin-layer chromatography (TLC) plates as described previously (Zhang et al., 1997). [³²P]PIP₂ generation was analyzed by PhosphorImager (Molecular Dynamics, Sunnyvale, CA). The results presented are representative of three separate experiments.

Quantification
MetaMorph software version 7.1.2 (Universal Imaging) was used to analyze fluorescence localization and shape. Localization was determined by drawing a line down the center of the cell from the rear of the cell to the front of the cell. The fluorescence intensity along that line was then obtained using the linescan function of MetaMorph. Peak fluorescence intensities for each cell were taken as 1 and minima as 0. These relative fluorescence values were then averaged for 15–45 cells in each condition. The difference in fluorescence shape between the GFP-PIPKI661 and GFP-PIPKI635 was determined through the integrated morphometry shape factor analysis in Metamorph. This algorithm assigns a value from 0 to 1, depending upon how closely the object represents a circle, with 1 being a perfect circle. Metamorph cell tracking was used to determine the speed of the cell front and the cell rear. Only cells that were actively chemotaxing for at least 2 min and both the front and rear of the cells were fully visible for the entire time were tracked. Speeds were averaged for the trackable time interval for each cell front and cell rear. Based on this exclusion criteria a total of 10 cells per condition from three separate experiments were tracked.

Online Supplementary Material
Dual color time-lapse microscopy of HeLa cells coexpressing GFP-PIPKI661 and dsRed-paxillin (see Figure 2D) is shown in Supplementary Video S1 (scale bar, 5 µm). Time-lapse microscopy of primary neutrophils (see Figure 3) that express GFP-PIPKI661 (Supplementary Video S2), GFP-ezrin (Supplemementary Video S3), or GFP-PIPKI635 (Supplementary Video S4) migrating toward a micropipette tip releasing C5a are shown (scale bars, 10 µm). Time-lapse microscopy of primary neutrophils migrating toward a micropipette tip releasing C5a (see Figure 5) is shown in Supplementary Videos S5 (GFP-PH-PLC), and S6 (GFP-PHAKT; scale bars, 10 µm). Time-lapse microscopy of primary neutrophils that express wild-type GFP-PIPKI661 (see Figure 6B) left either untreated (Supplementary Video S7) or pretreated with ROCK inhibitor (Supplementary Video S8) migrating toward a micropipette tip are shown. Time-lapse microscopy of dHL-60 cells (see Figure 7) that express either GFP (Supplementary Video S9), wild-type GFP-PIPKI661 (Supplementary Video S10), or kinase-dead GFP-PIPKI661 (Supplementary Videos S11 and S12) migrating toward a pipette tip are shown. Images for Video S1 were captured at 2-min intervals for 60 min. All other images were captured at 10-s ('') intervals for 10 min. All time-lapse microscopy images were captured using a 60x objective with a Nikon Eclipse TE300 inverted fluorescence microscope with a cooled charge-coupled device video camera (Hamamatsu Photonics) using differential interference contrast microscopy (DIC) and captured into Metamorph v7.0r2 (Universal Imaging). Time-lapse series were converted into QuickTime movie format with CinePak compression and were prepared using MetaMorph v7.Or2 (Universal Imaging).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Expression of PIPKI661 in Neutrophils and HL-60 Cells
The role of PIPKI661 in the regulation of adhesions in fibroblasts makes PIPKI661 an attractive candidate to contribute to localized PtdIns(4,5)P₂ production during neutrophil motility. To determine expression of the type I PIPKI in neutrophils and the neutrophil-like HL-60 cell line, we used a series of isoform specific reagents for RT-PCR and immunoblot analysis. The type I PIPKI, specifically PIPKI661, was readily detected in primary neutrophils by RT-PCR (Figure 1A) and immunoblotting (Figure 1B). PIPKI and PIPKIβ were also detected by RT-PCR in neutrophils but were not readily detected by immunoblotting (data not shown). Additionally, we observed an increase in PIPKI expression upon DMSO-induced differentiation of neutrophil-like HL-60 cells, suggesting that PIPKI may be specifically up-regulated in neutrophil-like cells (Figure 1B).

View larger version (53K): [in this window] [in a new window]   Figure 1. Endogenous PIPKI is expressed in neutrophils and localizes toward the cell rear upon chemoattractant stimulation. (A) RT-PCR was used to determine the expression of PIPKI in primary neutrophils (PMN), undifferentiated and differentiated HL-60 cells (uHL and dHL), and control HEK-293 cells (HEK) using gene-specific primers against human PIPKI (P) and the PIPKI661 splice variant (P661). GAPDH loading controls indicate equal sample was loaded. (B) Representative immunoblots show endogenous PIPKI (P) expression in primary neutrophils, HEK cells, and HL-60 cells in separate blots (boxes). (C and D) Representative images of neutrophils demonstrate localization of endogenous PIPKI (P; C) and active RhoA (rhotekin; D) to the cell rear upon fMLP stimulation. Neutrophils were plated on coverslips coated with 10 µg ml–1 fibrinogen in the presence of 100 nM fMLP for 15 min, fixed, and stained for PIPKI or GST-rhotekin (green), rhodamine phalloidin for F-actin (red), and DAPI (blue). Note enrichment of PIPKI at the cell rear. Shown are representative images from at least three experiments with more than 50 cells examined. Scale bars, 5 µm.

Asymmetric Distribution of PIPKI in Neutrophils

GFP-PIPKI661 Retains Lipid Kinase Activity and Focal Adhesion Targeting
To define the dynamics of PIPKI localization in live cells, we constructed a GFP-PIPKI fusion protein for both wild-type and kinase-dead PIPKI661 (Figure 2). Expression of the GFP-fusions in differentiated HL-60 cells (dHL-60) was confirmed by immunoblotting (Figure 2B), and the activity of the GFP-tagged kinase was measured (Figure 2C). We found that wild-type GFP-PIPKI661 retained kinase activity when expressed in dHL-60 cells and that the D253A kinase-dead mutant PIPKI661 showed markedly reduced kinase activity (Figure 2C). Previous work has shown that PIPKI661 targets to focal adhesions (Di Paolo et al., 2002; Ling et al., 2002). To determine if GFP-PIPKI661 retained its focal adhesion targeting properties, GFP-PIPKI661 was expressed in HeLa cells and analyzed by live fluorescence imaging. We found that GFP-PIPKI661 targeted to focal adhesions where it colocalized with paxillin (Figure 2D), displaying dynamic assembly and disassembly at both leading and trailing edge adhesion complexes (Supplementary Video S1). Taken together, we demonstrate that GFP-PIPKI661 is functional and shows appropriate targeting to focal adhesions in HeLa cells.

View larger version (40K): [in this window] [in a new window]   Figure 2. GFP-PIPKI661 is functional and targets to focal adhesion in HeLa cells. (A) Schematic representation of the PIPKI661 constructs. The domains and active site residues of PIPKI661 are noted. The C-terminal 26-amino acid extension that defines PIPKI661 is shaded (dark gray). The kinase-dead construct was generated by mutating the active site aspartate (D253A denoted with an asterisk). (B) Stable HL-60 cell lines were generated that express N-terminal GFP-tagged versions of PIPKI661. Expression of GFP-PIPKI661 wild-type (WT) and PIPKI661 kinase-dead (KD) in dHL-60 cells was detected by immunoblot with -GFP. (C) Lysates from dHL-60 cells that express either low or high levels of GFP-PIPKI661 wild-type (WT) and kinase-dead (KD) were immunoprecipitated with -GFP antibody and assayed for PI4P kinase activity using TLC. Generation of PtdIns(4,5)P2 was detected by autoradiography. Shown is a representative experiment of more than three replicates. (D) Representative images of HeLa cells that express GFP-PIPKI661 demonstrate regions of colocalization of PIPKI661 with the focal adhesion protein paxillin. HeLa cells that express GFP-PIPKI661 and dsRed-paxillin were plated on 35-mm glass-bottom dishes coated with 10 µg ml–1 fibronectin and analyzed by time-lapse fluorescence microscopy. Shown is a representative of three separate experiments with more than six cells imaged. Scale bar, 5 µm. Corresponding time-lapse microscopy is shown in Supplementary Video S1. Scale bar, 5 µm.

PIPKI661 Is a Novel Component of the Uropod in Polarized Leukocytes

View larger version (66K): [in this window] [in a new window]   Figure 3. GFP-PIPKI661 targets to the uropod of primary neutrophils. (A) Representative DIC and fluorescent images of neutrophils that express either GFP-PIPKI661 or GFP-ezrin. Merged image shows enrichment of GFP-PIPKI661 at the cell rear similar to GFP-ezrin in response to a chemotactic gradient. Primary murine neutrophils that express either GFP-PIPKI661 or GFP-ezrin were plated onto 2.5 µg ml–1 fibrinogen and 10 µg ml–1 fibronectin and exposed to a chemotactic gradient generated by the slow release of C5a from a Femtotip micropipette. DIC and fluorescent time-lapse images were taken at 100-s ('') intervals. The tip of the micropipette is marked with an asterisk (*). Scale bars, 5 µm. Corresponding time-lapse microscopy is shown in Supplementary Videos S2 (GFP-PIPKI661) and S3 (GFP-ezrin). Scale bars, 10 µm. (B) Time-lapse sequence of neutrophils that express either GFP-PIPKI661 or GFP-ezrin. Note enrichment of both GFP-PIPKI661 and GFP-ezrin in the uropod. The direction of the chemoattractant source is denoted with a filled white circle (). Scale bars, 5 µm. (C) Fluorescence intensities from the cell rear to the cell front were determined for at least 45 cells for each condition from four different experiments. Shown are the average relative fluorescence intensities for GFP, GFP-PIPKI661, and GFP-ezrin. Note the rear localization of GFP-PIPKI661 and GFP-ezrin. These differences in localization between GFP, GFP-PIPKI661, and GFP-ezrin at 2 µm from the cell rear were confirmed using repeated measures ANOVA and subsequent Bonferroni's multiple comparison test and were found to have a p < 0.05. Similarly, GFP-PIPKI661 and GFP-ezrin fluorescence intensities are reduced compared with GFP at the point 15 µm from the cell rear, as indicated by repeated-measures ANOVA and Bonferroni's multiple comparison test with a p < 0.05. Shown are the SEM for 2 and 15 µm from the cell rear. (D) Representative DIC and fluorescent images of neutrophils that express GFP-PIPKI661. Magnified merged image (right) shows localization of GFP-PIPKI661 (middle) at the cell rear in response to uniform chemoattractant stimulation. Scale bar, 5 µm. (E) Representative DIC and fluorescent images of neutrophils that express GFP-PIPKI661 or GFP-PIPKI635. Merged image (right) shows enrichment of GFP-PIPKI661 (top middle) in the uropod, whereas GFP-PIPKI635 (bottom middle) is localized at the cell membrane, at the front of the cell, and the uropod. The direction of the chemoattractant source is denoted with a filled white circle (). Scale bars, 2 µm. Corresponding time-lapse microscopy of neutrophils expressing GFP-PIPKI635 is shown in Supplementary Video S4. Scale bar, 10 µm. (F) The shapes of the fluorescent signals in neutrophils expressing GFP-PIPKI661 or GFP-PIPKI635 were analyzed using the integrated morphometry shape factor analysis of Metamorph. This algorithm assigns a value from 0 to 1, describing the shape of the fluorescence signal where a perfect circle attains a value of 1 and a line is assigned a 0. At least 15 cells from three different experiments were analyzed, and the SD is shown. Two-tailed Student's t test, p < 0.001.

View larger version (31K): [in this window] [in a new window]   Figure 4. PIPKI is expressed in T-cells and localizes to the uropod. (A) RT-PCR was used to determine the expression of PIPKI in primary human T-cells, Jurkat T-cells, and control HEK-293 cells (HEK) using gene-specific primers against human PIPKI (P). (B) Stable D10 T-cell lines were generated that express GFP control and wild-type GFP-PIPKI661. Expression of GFP-PIPKI661 wild-type (WT) in D10 T-cells was detected by immunoblot with -GFP. (C) Representative images of D10 T-cells demonstrate colocalization of GFP-PIPKI661 at the cell rear with ERM proteins upon LTB4 stimulation. D10 T-cells that express GFP-PIPKI661 were plated on coverslips coated with poly-L-lysine for 30 min, treated with LTB4, fixed, stained with -ERM (middle). Merged image (overlay) shows colocalization of GFP-PIPKI661 with ERM proteins at the cell rear.

View larger version (65K): [in this window] [in a new window]   Figure 5. Asymmetric distribution of GFP-PH-PLC and GFP-PHAKT in primary neutrophils exposed to a gradient of chemoattractant. (A) Representative DIC and fluorescent images of neutrophils that express either GFP-PH-PLC or GFP-PHAKT. Bone marrow derived murine neutrophils were nucleofected with either GFP-PH-PLC or GFP-PHAKT and plated on 35-mm glass-bottom dishes coated with a mixture of 2.5 µg ml–1 fibrinogen and 10 µg ml–1 fibronectin and exposed to a chemotactic gradient generated by the slow release of C5a from a micropipette. DIC and fluorescent time-lapse images were taken at 10-s ('') intervals. The tip of the micropipette is marked with an asterisk (*). Merged image shows enrichment of PHAKT at the leading edge, whereas PH-PLC is periodically enriched at both the cell front and cell rear. Scale bars, 5 µm. Corresponding time-lapse microscopy is shown in Supplementary Videos S5 (GFP-PH-PLC) and S6 (GFP-PHAKT). Scale bars, 10 µm. (B) Quantification of fluorescence localization in neutrophils expressing either GFP-PH-PLC or GFP-PHAKT. Fluorescence intensities from the cell rear to the cell front were determined for 17 cells per condition. Shown are the numbers of cells and the distance from the cell rear of their peak fluorescence intensities for GFP-PH-PLC or GFP-PHAKT. (C) Time-lapse sequence of neutrophils that express GFP-PH-PLC. Note periodic enrichment of PH-PLC at the cell rear (arrows). (D) Time-lapse sequence of neutrophils that express GFP-PHAKT. Note enrichment of PHAKT at the leading edge. The direction of the chemoattractant source is denoted with a filled white circle (). Scale bars, 5 µm.

PIPKI661 Uropod Localization Is Independent of ROCK Signaling

View larger version (74K): [in this window] [in a new window]   Figure 6. PIPKI661 retains uropod localization upon ROCK inhibition. Representative DIC and fluorescent images of neutrophils pretreated with ROCK Y-27632 inhibitor that express wild-type GFP-PIPKI661 (P661 WT). Bone marrow–derived murine neutrophils were nucleofected with P661 wild type (WT), plated on 35-mm glass-bottom dishes coated with a mixture of 2.5 µg ml–1 fibrinogen and 10 µg ml–1 fibronectin, pretreated for 30 min with Y-27632 or vehicle control, and subsequently exposed to a chemotactic gradient generated by the slow release of C5a from a micropipette. DIC and fluorescent time-lapse images were taken at 10-s ('') intervals as described in Materials and Methods. The tip of the micropipette is marked with an asterisk (*) or the direction of chemoattractant source is denoted with a filled white circle (). Merged image shows enrichment of PIPKI661 in the cell rear despite elongated morphology observed upon inhibition of ROCK. Scale bars, 10 µm. Corresponding time-lapse microscopy for cells that express wild-type GFP-PIPKI661 are shown in Supplementary Videos S7 (control) and S8 (Y-27632). Scale bars, 10 µm.

GFP-PIPKI Localizes to the Uropod and Regulates Rear Retraction in Neutrophil-like HL-60 Cells

View larger version (64K): [in this window] [in a new window]   Figure 7. PIPKI expression modulates cell morphology and rear retraction. (A) Representative DIC and fluorescent images of dHL-60 cells that express GFP control, GFP-PIPKI661 wild-type (WT), or GFP-PIPKI661 kinase-dead (KD)show enrichment of GFP-PIPKI661 WT in the uropod and GFP-PIPKI661 KD at the cell rear in elongated tails. Differentiated HL-60 cells (dHL-60) that express GFP-PIPKI661 WT or KD were plated onto 2.5 µg ml–1 fibrinogen and 10 µg ml–1 fibronectin andexposed to a chemotactic gradient generated by the slow release of C5a from a Femtotip micropipette. DIC and fluorescent time-lapse images were taken at 10-s ('') intervals. The tip of the micropipette is marked with an asterisk (*) or the direction of the chemoattractant source is denoted with a filled white circle (). Scale bars, 5 µm. Corresponding time-lapse microscopy is shown in Supplementary Videos S9 (GFP), S10 (GFP-PIPKI661 WT), and S11–12 (GFP-PIPKI661 KD). (B) Time-lapse sequence of dHL-60 cells that express control GFP (top), GFP-PIPKI661 WT (middle), or GFP-PIPKI661 KD (bottom). Scale bars, 10 µm. Note prominent uropod in cells that express GFP-PIPKI661 WT compared with the elongated morphology and reduced rear retraction in cells that express GFP-PIPKI661 KD. Scale bars, 10 µm. The tip of the micropipette is marked with an asterisk (*) or the direction of the chemoattractant source is denoted with a filled white circle (). Scale bars, 5 µm. (C) Fluorescence intensities from the cell rear to the cell front were determined for at least 25 cells in each condition from four different experiments. Shown are the average relative fluorescence intensities for GFP, GFP-PIPKI661 WT, and GFP-PIPKI661 KD. Note the rear localization of both GFP-PIPKI661 WT and GFP-PIPKI661 KD compared with the GFP control. These differences in localization at 2 µm from the cell rear were confirmed using repeated-measures ANOVA and subsequent Bonferroni's multiple comparison test and were found to have a p < 0.05. Similarly, GFP-PIPKI661 WT and GFP-PIPKI661 KD fluorescence intensities are reduced at the point 25 µm from the cell rear as compared with GFP as indicated by repeated-measures ANOVA and Bonferroni's multiple comparison test with a p < 0.05. Shown are the standard deviations for 2 and 25 µm from the cell rear. (D) Speeds of the cell front and cell rear for dHL-60 cells expressing GFP, GFP-PIPKI661 WT, or GFP-PIPKI661 KD were determined for 10 cells in each condition from three different experiments. Shown are the average speeds ± SEM. The ratio of the speed of the cell front to the speed of the cell rear is also shown. Note the reduced speed of the cell rear in dHL-60 cells expressing GFP-PIPKI661 KD compared with those expressing GFP. These differences were confirmed using repeated measures ANOVA and subsequent Bonferroni's multiple comparison test and found to have a p < 0.05. (E) Cell length of dHL-60 cells expressing GFP, GFP-PIPKI661 WT, or GFP-PIPKI661 KD. Cell length was measured for at least 30 cells per condition from four different experiments. Shown are the averages ± the SD. Note the increase in cell length in the dHL-60 cells expressing GFP-PIPKI661 KD compared with those expressing GFP. These differences were confirmed using repeated measures ANOVA and subsequent Bonferroni's multiple comparison test and found to have a p < 0.05.

PIPKI Regulates the Localized Generation of PtdIns(4,5)P₂ at the Uropod during Chemotaxis
To determine whether PIPK is necessary for the localized generation of PtdIns(4,5)P₂ at the uropod during chemotaxis, we examined the effect PIPKI661 expression on the distribution of PtdIns(4,5)P₂ in primary neutrophils using the GFP-PH-PLC probe. mCherry-PIPKI661 and kinase-dead mCherry-PIPKI661 were coexpressed with GFP-PH-PLC in primary murine bone marrow neutrophils. Our findings indicate that expression of kinase-dead PIPKI661 impaired the production of PtdIns(4,5)P₂ at the uropod compared with expression of either mCherry alone or wild type mCherry-PIPKI661 (Figure 8). Taken together, our findings identify PIPKI661 as a novel uropod component that regulates rear release via the localized generation of PtdIns(4,5)P₂ and suggest that PIPKI661-mediated PtdIns(4,5)P₂ synthesis and periodic accumulation is important for backness signaling and rear release during neutrophil chemotaxis.

View larger version (41K): [in this window] [in a new window]   Figure 8. Kinase-dead PIPKI661 expression reduces PtdIns(4,5)P2 localization to the cell rear. (A) Representative DIC and fluorescent images of bone marrow-derived murine neutrophils nucleofected with mCherry-PIPKI661 wild type (mCh-PIPKI661 WT) and GFP-PH-PLC or with mCherry-PIPKI661 kinase-dead (mCh-PIPKI661 KD) and GFP-PH-PLC. Cells were plated on 35-mm glass-bottom dishes that were coated with a mixture of 2.5 µg ml–1 fibrinogen and 10 µg ml–1 fibronectin and exposed to a chemotactic gradient generated by the slow release of C5a from a micropipette. DIC and fluorescent time-lapse images were taken at 15-s ('') intervals as described in Materials and Methods. The direction of chemoattractant source is denoted with a filled white circle (). Merged image shows enrichment of PIPKI661 WT and KD in the cell rear and reduced GFP-PH-PLC at the cell rear in cells coexpressing mCh-PIPKI661 KD compared with those coexpressing mCh-PIPKI661 WT. Scale bars, 5 µm. (B) Fluorescence intensities from the cell rear to the cell front were determined for at least 20 cells in each condition from three different experiments. Shown are the average relative fluorescence intensities for GFP-PH-PLC when coexpressed with mCherry control, mCh-PIPKI661 WT, or mCh-PIPKI661 KD. Note the reduction in rear localization of GFP-PH-PLC when coexpressed with GFP-PIPKI661 KD compared with coexpression with GFP-PIPKI661 WT or mCherry control. These differences in localization at 2 µm from the cell rear were confirmed using repeated-measures ANOVA and subsequent Bonferroni's multiple comparison test and found to have a p < 0.05. Shown are the standard deviations for 2 µm from the cell rear.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

During chemotaxis, spatial information about the chemotactic gradient is translated into an internal asymmetry of signaling molecules that elicit either front- or back-specific responses. For example, leading edge components such as Rac, Cdc42, and PI3K are thought to promote the frontness response by triggering accumulation of actin and the phosphoinositide PtdIns(3,4,5)P₃ at the leading edge (Meili et al., 1999; Weiner et al., 1999; Servant et al., 2000; Van Keymeulen et al., 2006). In contrast, components of the cell rear such as Rho and ROCK are thought to promote backness by the generation of actomyosin-based contraction and regulation of rear release (Niggli, 1999; Eddy et al., 2000; Xu et al., 2003; Wong et al., 2006). Accordingly, spatial restriction of PIPKI661 to the cell rear suggests that PIPKI661 plays a role in backness signaling at the trailing edge during neutrophil chemotaxis. Further analysis of PIPKI661 function during chemotaxis suggests that PIPKI661 promotes backness via localized PtdIns(4,5) P₂ production to regulate rear release (Figures 7 and 8), because expression of kinase-dead PIPKI661 compromised rear release resulting in the formation of abnormal, elongated, or bicuspid tails, whereas cells that overexpress wild-type PIPKI661 generate a more prominent uropod and exhibit efficient rear release.

Previous work has shown that the backness response is mediated by RhoA via downstream signaling to ROCK to initiate actomyosin contraction (Kimura et al., 1996; Yoshinaga-Ohara et al., 2002; Xu et al., 2003). Interestingly, the rear retraction defect noted upon overexpression of kinase-dead GFP-PIPKI661 phenocopies previous results observed upon inhibition of RhoA (Alblas et al., 2001; Worthylake et al., 2001), ROCK (Somlyo et al., 2000; Alblas et al., 2001; Worthylake et al., 2001), and myosin IIa (Eddy et al., 2000). Previous work has also shown that PIPKI interacts with RhoA (Chong et al., 1994; Ren et al., 1996) and that this interaction increases PtdIns(4,5)P₂ production by PIPKI family members (Chong et al., 1994; Weernink et al., 2004). Therefore, PIPKI661 may promote rear retraction via integration of PtdIns(4,5)P₂ with the backness pathway through the association of PIPKI661 with RhoA. Previous work has also shown that PIPKI661 associates with AP2 to regulate endocytosis (Bairstow et al., 2006). As endocytosis and recycling of integrins at the cell rear plays an important role in neutrophil rear release (Lawson and Maxfield, 1995), it is intriguing to speculate that PIPKI661 may also promote rear detachment via endocytosis of surface receptors at the cell rear. Alternatively, PIPKI661 may promote rear detachment via the synthesis of PtdIns(4,5)P₂, which regulates the activity of cytoskeleton-associated ERM proteins (Yoshinaga-Ohara et al., 2002; Fievet et al., 2004). A challenge for future investigation will be to define the mechanisms by which PIPKI661 and the localized production of PtdIns(4,5)P₂ within the uropod regulates rear retraction.

Recent evidence suggests cross talk between front and rear signaling modules in that PtdIns(3,4,5)P₃ and Cdc42 signaling at the leading edge feedback to affect RhoA signaling at the back of the cell (Van Keymeulen et al., 2006). However, less is known about how the back of the cell may regulate signaling at the front and modulate cell polarity. There is evidence to support a role for the uropod in regulating cell polarity (Lee et al., 2004), and recent studies suggest that the uropod provides an area where contractile stresses are concentrated (Smith et al., 2007) and may play a more active role in cell motility by regulating integrin clustering through WASP (Zhang et al., 2006). This raises the intriguing possibility that the cell rear may be an active signaling module that regulates cell polarity and provides positive feedback to the front of the cell during directed cell migration. On the basis of our findings, we propose that the uropod component PIPKI and its lipid product PtdIns(4,5)P₂ play an active role in establishing backness possibly via RhoA/ROCK-mediated signaling during chemotaxis. Future studies should shed light on the role of these pathways in regulating cell polarity and their contribution to positive feedback mechanisms that optimize directed cell migration.

	ACKNOWLEDGMENTS

 
We gratefully acknowledge Jun Zhu for construction of the GFP-PIPKI661 fusion constructs; Tamas Balla (NIH, Bethesda, MD) for the GFP-PH-PLC and the GFP-PHAKT constructs; Monique Arpin (Institute Curie, France) and Richard Lamb (Institute of Cancer Research, London, United Kingdom) for the GFP-ezrin construct; Rick Horwitz (University of Virginia, Charlottesville, Virginia) for the dsRed-paxillin construct; Richard Anderson and Kun Ling (University of Wisconsin, Madison, WI) for the C-terminal PIPKI antibody, PIPKI661 cDNA; Garry Nolan (Stanford University, Stanford, CA) for Phoenix cells; Clive Svendsen (University of Wisconsin, Madison, WI) for the pMX retroviral vector; Roger Tsien (University of California at San Diego) for the pRSET-B-mCherry construct; and Kathy Schell, Joel Puchalski, and Dagna Sheerar for their expertise at the Flow Cytometry Facility (University of Wisconsin, Madison, WI). This work was supported by National Institutes of Health Grants R01 GM074827 (A.H.), American Heart Association grant-in-aid (A.H.), and National Institutes of Health Grant RO1 AI68062 (V.O.). P.N. was supported by an American Heart Association predoctoral fellowship, and M.A.S. is supported by a postdoctoral fellowship from the American Heart Association (0625751Z).

	Footnotes

 
This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E07-05-0428) on October 10, 2007.

The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org).

These authors contributed equally to this work.

Address correspondence to: Anna Huttenlocher (huttenlocher{at}wisc.edu).

Abbreviations used: C5a, Complement factor 5a; DIC, differential interference contrast microscopy; dHL-60, differentiated HL-60; fMLP, formyl-Met-Leu-Phe; PMN, polymorphonuclear cells; PI3K, phosphoinositide 3-kinase; PtdIns(3,4,5)P₃, phosphatidylinositol (3,4,5)-trisphosphate; PtdIns(4,5)P₂, phosphatidylinositol (4,5)-bisphosphate; PIPKI661, Type I661 phosphatidylinositol phosphate kinase; KD, kinase-dead; GFP, green fluorescent protein.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Alblas, J., Ulfman, L., Hordijk, P., and Koenderman, L. (2001). Activation of Rhoa and ROCK are essential for detachment of migrating leukocytes. Mol. Biol. Cell 12, 2137–2145.[Abstract/Free Full Text]

Alonso-Lebrero, J. L. et al. (2000). Polarization and interaction of adhesion molecules P-selectin glycoprotein ligand 1 and intercellular adhesion molecule 3 with moesin and ezrin in myeloid cells. Blood 95, 2413–2419.[Abstract/Free Full Text]

Bairstow, S. F., Ling, K., Su, X., Firestone, A. J., Carbonara, C., and Anderson, R. A. (2006). Type Igamma661 phosphatidylinositol phosphate kinase directly interacts with AP2 and regulates endocytosis. J. Biol. Chem 281, 20632–20642.