Preservation of the pattern of tyrosine phosphorylation

in human neutrophil lysates

II. A sequential lysis protocol for the analysis of tyrosine

phosphorylation-dependent signalling

Caroline Gilbert 1, Emmanuelle Rollet-Labelle 1, Paul H. Naccache *

Centre de Recherche en Rhumatologie et Immunologie, CIHR group on the Molecular Mechanisms of Inflammation,

Centre de Recherche du CHUL, and Department of Medicine, Faculty of Medicine, Laval University, Ste-Foy, Quebec, Canada

Received 17 September 2001; received in revised form 6 November 2001; accepted 6 November 2001

Abstract

In stimulated neutrophils, themajority of tyrosine phosphorylated proteins are concentrated in TritonX-100 or NP-40 insoluble

fractions. Most immunobiochemical studies, whose objective is to study the functional relevance of tyrosine phosphorylation are,

however, performed using the supernatants of cells lysed in non-ionic detergent-containing buffers (RIPA lysis buffers). This

observation prompted us to develop an alternative lysis protocol. We established a procedure involving the sequential lysis of

neutrophils in buffers of increasing tonicities that not only preserved and solubilized tyrosine phosphorylated proteins but also

retained their enzymatic activities. The sequential lysis of neutrophils in hypotonic, isotonic and hypertonic buffers containing

non-ionic detergents resulted in the solubilisation of a significant fraction of tyrosine phosphorylated proteins. Furthermore, we

observed that in monosodium urate crystals-stimulated neutrophils, Lyn activity was enhanced in the soluble fraction recovered

from the hypertonic fraction, but not from that of the first hypotonic lysis. The distribution of tyrosine phosphorylated proteins

between the NP-40 soluble and insoluble fractions was both substrate- and agonist-dependent. In neutrophils stimulated with

fMet-Leu-Phe, MSU crystals or by CD32 ligation, the tyrosine phosphorylated proteins were mostly insoluble. On the other hand,

in GM-CSF-treated cells, the phosphoproteins were more equally distributed between the two fractions. The results of this study

provide a new experimental procedure for the investigation of tyrosine phosphorylation pathways in activated human neutrophils

which may also be applicable to other cell types. D 2002 Elsevier Science B.V. All rights reserved.

Keywords: Cbl; CD32; Detergent insolubility; Fcg receptors; fMet-Leu-Phe; GM-CSF; Kinase activity; Lyn; Monosodium urate crystals;

Neutrophil; Phagocytosis; Signal transduction; Syk; Tyrosine phosphorylation

1. Introduction

Neutrophils play important roles in host defense

against injury and infection as well as in the inflam-

matory response (Smith, 1994). In order to accomplish

their function, neutrophils store in their granules var-

ious proteases and microbicidal agents (Borregaard

0022-1759/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.

PII: S0022-1759 (01 )00553 -1

Abbreviations: HLB, hypotonic buffer; HyperLB, hypertonic

buffer; ILB, isotonic buffer; LB, lysis buffer; MSU, Monosodium

urate; RIPA, rapid immunoprecipitation assay; SB, sample buffer.* Corresponding author. CHUL, Room T1-49, 2705 Boulevard

Laurier, Ste-Foy, Quebec, Canada G1V 4G2. Tel.: +1-418-654-

2772; fax: +1-418-654-2765.

E-mail address: paul.naccache@crchul.ulaval.ca

(P.H. Naccache).1 The authors contributed equally to this study.

www.elsevier.com/locate/jim

Journal of Immunological Methods 261 (2002) 85–101

and Cowland, 1997) and are able to produce toxic

oxygen-derived metabolites (Rosen and Michel, 1997)

and lipids mediators (McDonald et al., 1991; Pouliot et

al., 1998) as well as to synthesize and secrete various

proteins involved in the regulation of the inflammatory

response (Hachicha et al., 1995; Cassatella, 1996).

Neutrophils respond to a wide variety of soluble

factors including cytokines (McDonald et al., 1991;

Pouliot et al., 1998) and chemoattractants (Torres et

al., 1993) as well as to particulate agonists (Gaudry et

al., 1993; Roberge et al., 1993). One of the earliest

events observed upon stimulation of neutrophils is an

increase in the level of tyrosine phosphorylation of a

number of proteins (Rollet et al., 1994). Several

neutrophil functional responses including adherence

(Gaudry et al., 1992; Naccache et al., 1994), locomo-

tion (Gaudry et al., 1992), and superoxide production

(Naccache et al., 1990) are inhibited by tyrosine kinase

inhibitors or enhanced by tyrosine phosphatase inhib-

itors (Grinstein et al., 1990; Cui et al., 1994). The

activity of multiple tyrosine kinases including Fgr

(Gutkind and Robbins, 1989; Hamada et al., 1993;

Krump et al., 1997), Lyn (Gaudry et al., 1995; Yan et

al., 1996; Ibarrola et al., 1997; Welch and Maridon-

neau-Parini, 1997), Hck (Welch et al., 1996), Jak2 (Al-

Shami et al., 1998), Pyk2 (Yan and Novak, 1999) and

Syk (Asahi et al., 1993; Yan and Berton, 1996; Yan et

al., 1997; Fernandez and Suchard, 1998) is modulated

after the stimulation of neutrophils. The presence of

tyrosine phosphatases such as SHP-1 (Brumell et al.,

1997), SHP-2 (Tidow et al., 1999), CD148 (Hundt and

Schmidt, 1997) and CD45 (Cui et al., 1996; Fialkow et

al., 1997) has been observed in human neutrophils.

Several tyrosine phosphorylated substrates such as Cbl

(Naccache et al., 1997; Willeke et al., 2000), CD32

(Rollet-Labelle et al., 2000), CD50 (Skubitz et al.,

1995), members of the MAP kinase family (Grinstein

and Furuya, 1992; Torres et al., 1993; Nahas et al.,

1996), paxillin (Fuortes et al., 1994; Fernandez and

Suchard, 1998), and SAM68 (Gilbert et al., 2001) have

also been identified.

The mechanisms linking neutrophil surface recep-

tors to tyrosine kinase signalling are, however, incom-

pletely understood. Immunoprecipitation coupled with

SDS-polyacrylamide gel electrophoresis is one of the

most commonly used immuno-chemical techniques

for studying tyrosine phosphorylation-dependent

pathways. This technique can potentially determine

several important characteristics of the antigen under

examination such as its presence and quantities, its

relative molecular weight, its rate of synthesis or

degradation, and the presence of certain post-transla-

tional modifications. It can also provide information

about interactions with proteins, nucleic acids and

other ligands. These protocols depend on lysing the

cells under mild conditions in RIPA buffers containing

cocktails of protease and phosphatase inhibitors.

These lysates are then centrifuged and the super-

natants are used for subsequent immunoprecipitations.

It is commonly observed, however, that the profile of

tyrosine phosphorylated proteins of whole neutrophils

is rapidly lost or artificially increased upon lysing the

cells in classical RIPA buffers. An alternative assay

involving lysis under denaturing conditions was pre-

viously developed to preserve the phosphorylation

levels in neutrophil lysates (Al-Shami et al., 1997b).

Although this protocol allows the identification of

tyrosine phosphorylated proteins (Al-Shami et al.,

1997b; Barabe et al., 1998; Khamzina and Borgeat,

1998; Rollet-Labelle et al., 2000; Gilbert et al., 2001),

it disrupts protein–protein interactions and inactivates

most enzymatic activities.

Most studies of the type discussed above analyze

the soluble fractions following lysis in RIPA buffers

while discarding the insoluble fractions. Several stud-

ies, however, have identified and focussed on the

formation of detergent-resistant membrane structures

(DRM, DIGs or lipids rafts) into which tyrosine

phosphorylated proteins and tyrosine kinases tend to

concentrate after receptor stimulation (Zhou et al.,

1995; Yan et al., 1996). The function of these deter-

gent-insoluble structures appears to be related to

signal transmission and/or membrane trafficking. Sev-

eral studies have shown that tyrosine kinase activities

were increased and concentrated in these insoluble

fractions upon ligation of phagocytic receptors in

adherent human neutrophils (Zhou et al., 1995; Yan

et al., 1996). Analysis of tyrosine phosphorylation

events in these structures depends on the preparation

of stable cell lysates that can be used as starting

material for their analysis by immunoprecipitation.

In the present study, the distribution and preserva-

tion of tyrosine phosphorylated substrates and of

several signalling-associated molecules was moni-

tored in human neutrophil lysates prepared under a

variety of native conditions. The results obtained

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–10186

illustrate the inadequacy of standard cell lysates

prepared under native conditions for use as starting

material in subsequent immunoprecipitation or co-

immunoprecipitation protocols. An alternative method

based on sequential lysis under conditions of increas-

ing tonicity is described. The use of this alternative

protocol underlined the critical importance of the

complete monitoring of the tyrosine phosphorylation

status as well as of the solubility and, when appli-

cable, the enzymatic activity of each individual target

molecule, in response to each and every agonist.

2. Materials and methods

2.1. Antibodies

The polyclonal antibodies, anti-Cbl (sc-170), anti-

Lyn (sc-15) and anti-SHP-1 (sc-287, 1/1000) were

purchased from Santa Cruz Biotechnology (Santa

Cruz, CA, USA). The monoclonal antibodies anti-

Syk (MAB88906), anti-phosphotyrosine (UBI 05-

321, clone 4G10) were purchased from Chemicon

International (Missisauga, Ontario, Canada) and Up-

state Biotechnology (Lake Placid, NY, USA) respec-

tively. An anti-FcgRII polyclonal antiserum against the

cytoplasmic tail of CD32 (CT-10) was generated in our

laboratory as described previously (Ibarrola et al.,

1997). Monoclonal anti-FcgRII antibodies (IV.3) werepurified from the ascitic fluids of mice inoculated with

hybridoma HB 217 which was obtained from the

American Type Culture Collection (ATCC) (Rock-

ville, MD, USA). F(abV)2 fragments of the IV.3 anti-

body were prepared essentially as described in the

Pierce catalog (Rockford, IL, USA). Briefly, the anti-

bodies were digested with pepsin and intact antibodies

were eliminated by adding protein A and protein G

beads. The integrity and purity of the F(abV)2 frag-

ments was verified by their ability to label intact hu-

man neutrophils as determined by flow cytometry as

well as by their neutrophil activating properties (cal-

cium mobilization, superoxide generation, tyrosine

phosphorylation).

2.2. Reagents

Triton X100, trypsin inhibitor soybean, fMet-Leu-

Phe, and sodium orthovanadate (Na2VO4) were pur-

chased from Sigma Chemical (St-Louis, MO, USA).

Di-isopropylfluorophosphate (DFP), NP-40 10% sol-

ution and NP-40 were obtained from Calbiochem (La

Jolla, CA, USA). P-Nitrophenylphosphate, aprotinin

and leupeptin were purchased from ICN Pharmaceut-

icals (Costa Mesa, CA, USA). GM-CSF was supplied

by the Genetics Institute (Cambridge, MA, USA).

Triclinic MSU crystals were kindly provided by Drs.

R. de Medicis and A. Lussier (Universite of Sher-

brooke, Sherbrooke, Quebec, Canada) and prepared as

previously described (Gaudry et al., 1993). Dextran T-

500, Ficoll Paque, sephadex G-10, protein A sephar-

ose were purchased from Pharmacia Biotech (Dorval,

Quebec, Canada).

2.3. Cells

Neutrophils were obtained from healthy adult vol-

unteers as previously described (Rollet et al., 1994)

and were resuspended at a concentration of 4� 107

cells/ml in Hanks’ balanced salt solution (HBSS)

containing 1.6 mM calcium and no magnesium, pH

7.4.

2.4. Cell stimulation

Before all stimulations, neutrophil suspensions

(4� 107 cells/ml) were pre-incubated at room temper-

ature with 1 mM DFP (to minimize proteolysis) for 5

min. Neutrophils were then stimulated at the same

concentration at 37 �C with 3 mg/ml of MSU crystals

or 3 nM of GM-CSF for 10 min or with 100 nM of

fMet-Leu-Phe for 1 min. For CD32 cross-linking, the

cell suspensions were pre-incubated with 2.0 mg/ml of

anti-FcgRII (IV.3) antibodies for 1 min at 37 �C and

stimulated by cross-linking the cell-bound antibodies

with 20 mg/ml of F(abV)2 goat anti-mouse Fc antibody

(Jackson Immune Research, Mississauga, ON, Can-

ada) for 1 min at 37 �C or the indicated periods of

time. For immunoprecipitations, the F(abV)2 fragment

of IV.3 was used instead of the whole antibody.

2.5. Preparation of total cell lysates under denaturing

conditions

After stimulation, the reactions were stopped by

transferring 100 ml of the cell suspensions to an equal

volume of boiling 2� Laemmli sample buffer (SB)

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 87

(1� is 62.5 mM Tris–HCl, pH 6.8, 4% SDS, 5% b-mercaptoethanol, 8.5% glycerol, 2.5 mM orthovana-

date, 10 mM p-nitrophenylphosphate, 10 mg/ml leu-

peptin, 10 mg/ml aprotinin, 0.025% bromophenol

blue) and boiled for 7 min.

2.6. Hypotonic cell lysis

Neutrophils were pre-incubated and stimulated as

described above. After stimulation, the reactions were

rapidly stopped by transferring the cell suspensions in

pre-cooled (� 20 �C) 1.5-ml microcentrifuge tubes.

The cells were then centrifuged for 5–10 s at

6000� g in a micro-centrifuge and the pellets resus-

pended at a final concentration of 4� 107 cells/ml in a

hypotonic lysis buffer (HLB) (final concentrations:

0.1% NP-40, 20 mM Tris–HCl pH 7.5, 10 mM NaCl,

1 mM EDTA, 2 mM orthovanadate, 10 mg/ml aproti-

nin, 10 mg/ml leupeptin, 2 mM PMSF, 50 mg/ml

trypsin inhibitor soybean, and 3 mM DFP). For better

inhibition of phosphatases, p-nitrophenylphosphate 10

mM may be added in the lysis buffer. After 5 min of

incubation at 4 �C, the lysates were centrifuged at

600� g for 10 min at 4 �C. Aliquots of the super-

natants were added to an equal volume of boiling

2� SB or used for immunoprecipitation under native

or denaturing conditions or for the determination of

tyrosine kinase activities. The pellets were resus-

pended in HLB and then diluted in the same volume

of 2� SB or in an isotonic lysis buffer (ILB) or a

hypertonic lysis buffer (HyperLB) (see below for

composition) for immunoprecipitation or determina-

tion of tyrosine kinase activity. Nuclear integrity and

total cell lysis were routinely verified by light micro-

scopy.

2.7. Isotonic cell lysis

Neutrophils were pre-incubated and stimulated as

described above. After stimulation, the reactions were

rapidly stopped by transferring the cell suspensions

into precooled (� 20 �C) 1.5-ml microcentrifuge

tubes. The cell suspensions were then centrifuged

for 5–10 s at 6000� g in a micro-centrifuge. The

supernatants were removed and the cell pellets were

resuspended and incubated for 5 min at 4� 107 cells/

ml at 4 �C in cold isotonic lysis buffer (ILB) (final

concentrations: 1% NP-40, 20 mM Tris–HCl pH 7.5,

137 mM NaCl, 1 mM EDTA, 2 mM orthovanadate,

10 mg/ml aprotinin, 10 mg/ml leupeptin, 2 mM PMSF,

50 mg/ml trypsin inhibitor soybean, and 3 mM DFP).

The lysates were centrifuged at 600� g (Fig. 1) or

13,000� g (Fig. 5) for 10 min at 4 �C. Aliquots of thesupernatants were added to an equal volume of boil-

ing 2� SB. The pellets were resuspended in ILB and

then diluted in the same volume of 2� SB.

2.8. Sequential solubilisation

After stimulation, the soluble and insoluble frac-

tions of the hypotonic lysis were individually pro-

cessed for subsequent analysis. The soluble fraction

was centrifuged at 180,000� g at 4 �C for 45 min.

The supernatant was analyzed by transfer of an aliquot

into 2� SB. The pellet was dissolved in 1 volume of

HLB and transferred to boiling 2� SB. The insoluble

fraction of HLB was resuspended in ILB and centri-

fuged for 10 min at 13,000� g. The soluble material

of this second lysis was transferred into 2� SB and

the pellet was resuspended in the hypertonic lysis

buffer (HyperLB) (final concentrations: 1% NP-40,

1% Triton X100, 20 mM Tris–HCl pH 7.5, 400 mM

NaCl, 1 mM EDTA, 2 mM orthovanadate, 10 mg/ml

aprotinin, 10 mg/ml leupeptin, 2 mM PMSF, 50 mg/ml

trypsin inhibitor soybean, and 3 mM DFP). Two 5-s

sonication steps were necessary for maximal pellet

extraction. The lysates were centrifuged at 13,000� g

for 10 min at 4 �C. Aliquots of the supernatants weretransferred to an equal volume of boiling 2� SB. The

pellets were resuspended in HLB and then diluted in

the same volume of 2� SB. The samples were then

electrophoresed as described above.

Fig. 1. Distribution and preservation of tyrosine phosphorylation profiles and signalling molecules in HLB and ILB. Neutrophils (4� 107 cells/

ml) were stimulated by cross-linking CD32 as indicated in Materials and methods. Briefly, the reactions were stopped by a rapid transfer of cell

aliquots into a precooled Eppendorf tube and a rapid centrifugation for lanes 3–10 or in the boiling 2� SB for lanes 1–2. The appropriate lysis

buffer was then added to the cell pellet, HLB for lanes 3–6 and ILB for lanes 7–10. Aliquots of soluble (lanes 3, 4 and 7, 8) and insoluble

(lanes 5, 6 and 9, 10) fractions were diluted in an equal volume of 2� SB. The samples were analysed by coomassie blue protein staining (panel

A) or by Western blot with anti-phosphotyrosine, anti-Cbl, anti-Syk, anti-SHP-1 or anti-CD32 antibodies (panels B–F, respectively).

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–10188

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 89

The soluble fractions of HyperLB were also used

under native conditions for immunoprecipitation and

tyrosine kinase activity. These fractions (soluble

HyperLB) were also obtained by direct addition of

HyperLB to the HLB pellets.

2.9. Immunoprecipitation under denaturing condi-

tions

Neutrophils were incubated and stimulated as

described above. Aliquots (500 ml) of the cells were

lysed by direct transfer to an equal volume of boiling

2� lysis buffer (2�LB) (1� is 62.5 mM Tris–HCl,

pH 6.8, 2% SDS, 1.5% b-mercaptoethanol, 8.5%

glycerol, 2.5 mM orthovanadate, 10 mg/ml leupeptin,

10 mg/ml aprotinin, 0.025% bromophenol blue) and

boiled for 7 min. Immunoprecipitations were per-

formed as previously described (Al-Shami et al.,

1997b). Briefly, lysates were filtered through sepha-

dex G-10 columns to remove the denaturing agents.

The filtered lysates were precleared with protein A-

sepharose at 4 �C for 30 min in the presence of 1%

NP-40, 2 mM orthovanadate, 10 mg/ml leupeptin, and

10 mg/ml aprotinin. The samples were then immuno-

precipitated using 2 mg of anti-Cbl, 2 mg of anti-Syk or

6 mg of anti-CD32 previously bound to protein A-

sepharose for 90 min at 4 �C on a rotator platform

with constant end-over-end mixing. The beads were

collected and washed four times with a lysis buffer

containing 137 mM NaCl, 1% NP-40 but no SDS, b-mercaptoethanol or bromophenol blue. Sample buffer

(40 ml, 2� ) was added to the beads which were

boiled for 7 min. The proteins in the samples were

then separated by electrophoresis as described above.

The membranes were blotted with anti-phosphotyro-

sine or with the immunoprecipitating antibodies (anti-

Syk, anti-Cbl or anti-CD32) for visualizing the

amounts of precipitated protein.

2.10. Native immunoprecipitation and tyrosine kinase

activity

The aliquots (200 ml or 8� 106 cell equivalents) of

HLB or HyperLB supernatant lysates were mixed

with HyperLB or HLB, respectively, to bring back

the NaCl, NP-40 and Triton X-100 concentrations to

137 mM, 0.75%, 0.75%, respectively. These isotonic

lysates were immunoprecipitated using 1.5 mg of anti-

Lyn for 90 min at 4 �C on a rotator platform with

constant end-over-end mixing. Fifty microliters (30%

slurry) of protein A-sepharose was then added and the

samples were incubated for 1 h at 4 �C. The beads

were collected and washed four times with ILB buffer

without EDTA. The beads were incubated at 4 �C in

kinase buffer (1�KB, final concentrations: 500 mMHepes pH 7.5, 10 mM MgCl2, 3 mM MnCl2 with or

without to 50 mM ATP) and transferred to 37 �C for

the indicated times. The reactions were stopped by a

quick spin and the supernatants were removed. Sam-

ple buffer (40 ml, 2�) was added to the beads which

were boiled for 7 min. The proteins in the samples

were then separated by electrophoresis as described

above. The membranes were blotted with the anti-

phosphotyrosine, or with the immunoprecipitating

antibodies for visualizing the amounts of precipitated

protein. Immunoprecipitation with control rabbit IgG

followed by kinase activity was simultaneously under-

taken to verify the specificity of the precipitations.

2.11. Electrophoresis and immunoblotting

Before electrophoresis, all samples in 2� SB were

boiled for 7 min. The samples (equivalent cell num-

bers) were then subjected to 7.5–20% SDS-PAGE

and the proteins transferred to Immobilon PVDF

membranes (Millipore, Bedford, MA, USA). Immu-

noblotting was performed using the appropriate anti-

bodies (anti-Cbl (1/1000), anti-Lyn (1/2000), anti-

SHP-1 (1/1000), anti-CD32 (1/1000), anti-Syk (1/

2000) and anti-phosphotyrosine (1/4000)) and re-

vealed using the renaissance detection system (NEN

Life Science, Boston, MA.) using HRP-conjugated

secondary anti-mouse or anti-rabbit antibodies (Jack-

son, Immune Research, Mississauga, ON, Canada) at

a dilution of 1/20,000 as previously described (Al-

Shami et al., 1997a).

3. Results

3.1. Effect of tonicity of the lysis buffers on the

preservation and distribution of tyrosine phosphor-

ylation patterns and signalling-associated molecules

The difficulties associated with the preservation of

phosphotyrosine profiles and protein integrity in

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–10190

human neutrophils have recently been described (Al-

Shami et al., 1997b; Naccache et al., 1997). The

extraction of tyrosine phosphorylated proteins is

known to be affected by the composition of the lysis

buffer (Woods Ignatoski, 1996). The pH, presence of

phosphate, salt concentration, cell concentration, pres-

ence of cations and protease and phosphatase inhib-

itors can all modify the detergent behavior and the

solubilisation of individual proteins. Previous studies

have reported the solubilisation of proteins without

affecting the structure of the nuclei using a hypotonic

lysis buffer coupled with low concentrations of deter-

gents (0.1% NP-40) (Pouliot et al., 1996; McDonald

et al., 1998; Gilbert et al., 2001). The data in Fig. 1

compare this hypotonic buffer (HLB) with a more

classical isotonic RIPA buffer (ILB). The soluble and

insoluble fractions of the two lysis buffers (lanes 3–

10) were compared to the whole cell lysates (lanes 1–

2). The protein and tyrosine phosphorylation profiles

as well as the preservation of several proteins are

illustrated in panels A–F for the soluble (lanes 3–4,

7–8) and insoluble (lanes 5–6, 9–10) fractions

derived from these cell lysis protocols. To study the

tyrosine phosphorylation profile, neutrophils were sti-

mulated by cross-linking CD32 for 1 min at 37 �C as

described in Materials and methods. The samples

were analyzed by Coomassie blue protein staining

(panel A) or by immunoblotting with anti-phospho-

tyrosine, anti-Cbl, anti-Syk, anti-SHP-1 or anti-CD32

antibodies (panels B–F, respectively).

The percentage of lysed cells and the integrity of

the nuclei were routinely verified by light microscopy

(data not shown) or by immunoblot analysis with

nuclear markers such as p62 nucleoporin (a constitu-

ent of nuclear membranes). This confirmed that the

cells were completely lysed in the two buffers and that

nuclear integrity was better preserved in HLB than in

ILB (data not shown). Coomassie blue staining

(panel A) of these samples indicated that the amount

of protein extracted with the two lysis buffers was

roughly similar.

The tyrosine phosphorylation profile induced by

CD32 cross-linking, as monitored after the direct

transfer of the cells into sample buffer, is illustrated

in the first two lanes of panel B and served as a

reference point for the evaluation of the adequacy of

the native cell lysates. Two major conclusions can be

drawn from monitoring the tyrosine phosphorylation

profiles in the various fractions (panel B). Firstly,

tyrosine phosphorylated proteins were concentrated

in the insoluble fractions of both lysis protocols (lanes

5–6 and 9–10). Secondly, the tyrosine phosphoryla-

tion profiles were significantly better preserved in

HLB than in ILB (lanes 4–6 vs. lanes 8–10).

Immunoblotting of the various fractions with anti-

Cbl, anti-Syk, anti-SHP-1 or anti-CD32 antibodies

(panels C–F, respectively) further illustrated the dif-

ferences between the HLB and ILB lysis procedures.

The anti-Cbl blots showed that Cbl was better pre-

served in the soluble and insoluble fractions of HLB

(lanes 3–6) than ILB (7–10). Additionally, CD32

cross-linking induced a translocation of Cbl to the

insoluble fractions which was more evident in HLB

than ILB. The tyrosine kinase Syk was somewhat

more soluble in ILB than HLB and was relatively

insensitive to proteolysis. SHP-1 was also well-pre-

served in both lysis buffers though it was more easily

extracted in ILB than in HLB. Ligation of CD32

induced a translocation of CD32 from the soluble to

the insoluble fractions which was more readily appa-

rent in HLB than ILB. Hence, ILB is adequate for

studies requiring an optimal solubilisation of Syk and

SHP-1. However, based on the three criteria, (i)

prevention of proteolysis, (ii) preservation of tyrosine

phosphorylation and (iii) day-to-day reproducibility of

tyrosine phosphorylation levels (data not shown), the

data in Fig. 1 indicate that HLB is the preferred buffer

for the preparation of starting material from which to

study tyrosine phosphorylation-dependent signalling

events in human neutrophils.

3.2. The presence of tyrosine kinase cofactors in the

lysis buffers artefactually increases the tyrosine

phosphorylation levels in the soluble and insoluble

fractions

The composition of a lysis buffer can modify the

solubility, activity and phosphorylation status of pro-

teins and their intramolecular associations. Mg2 + is a

known tyrosine kinase cofactor which is present in the

lysis buffers used in various studies. The effects of the

addition of Mg2 + to HLB were examined and the

results of these experiments are summarized in Fig. 2.

Panel A presents the patterns of tyrosine phosphor-

ylation observed in whole cell lysates (lanes 1 and 2),

and in the soluble (lanes 3–4, 7–8) and insoluble

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 91

(lanes 5–6, 9–10) fractions derived from cells stimu-

lated by CD32 cross-linking and lysed in HLB. The

addition of Mg2 + (lanes 7–10) to the lysis buffer

increased the tyrosine phosphorylation level in frac-

tions from unstimulated as well as stimulated cells.

Under these conditions, the increases in the basal

levels of tyrosine phosphorylation make it more

difficult to distinguish the stimulatory effects of

CD32 cross-linking (see lanes 1 and 2 for compar-

ison).

Since Mg2 + significantly affected the tyrosine

phosphorylation profiles, the effect of this cation on

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–10192

the activity of a tyrosine kinase known to be activated

by CD32 cross-linking, namely Lyn, was examined.

The data in panels B and C show that while equal

amounts of Lyn were solubilized in the absence or

presence of Mg2 + (panel B), the kinase recovered by

immunoprecipitation was more heavily tyrosine phos-

phorylated when Mg2 + was present during lysis and

precipitation (panel C). Hyperphosphorylation of

SHP-1 was also observed in the presence of Mg2 +

in the lysis buffer (data not shown). The effect of

Mg2 + on the catalytic activity of Lyn was then tested

in an in vitro kinase assay. The results obtained (panel

D) show that Lyn was able to autophosphorylate itself

under both sets of conditions but that the autophos-

phorylation capacity of Lyn was increased when

Mg2 + was present in the lysis buffer.

3.3. Tyrosine phosphorylated proteins are concen-

trated in the insoluble fractions

The results described in Figs. 1 and 2 revealed a

heterogeneous distribution of phosphorylated proteins

among the soluble and insoluble fractions derived

from the lysis protocols after CD32 engagement. We

therefore monitored more closely three major tyrosine

phosphorylated substrates previously identified fol-

lowing CD32 cross-linking, namely, Cbl, Syk and

CD32 itself (Rollet-Labelle et al., 2000) (Fig. 3).

Briefly, aliquots of the soluble fractions derived from

the lysis of control and CD32-stimulated cells in HLB

were resuspended in an equal volume of boiling

2�LB as described in Materials and methods. The

insoluble material was resuspended in the same vol-

ume of HLB as the soluble fractions and subsequently

in an equal volume of boiling 2�LB. Cbl, Syk and

CD32 were immunoprecipitated from both fractions

Fig. 3. Immunoprecipitation of tyrosine phosphorylated proteins

from the HLB soluble and insoluble fractions. Neutrophils (4� 107

cells/ml) were stimulated by cross-linking CD32 as indicated in

Materials and methods. The reactions were stopped by a rapid

transfer of cell aliquots in a precooled Eppendorf tube and a rapid

centrifugation. The cell pellets were resuspended in HLB and the

soluble and insoluble fractions were diluted by 1/2 in boiling

2�LB. Immunoprecipitations under denaturing conditions were

performed as described in Materials and methods with the anti-Cbl,

anti-Syk or anti-C32 antibody. Phosphorylated proteins were

revealed with an anti-phosphotyrosine antibody (upper panels)

and the amount of immunoprecipitated protein was visualized by

immunoblot with the respective antibodies (lower panels).

Fig. 2. Effect of the presence of Mg2 + in HLB on tyrosine phosphorylation and tyrosine kinase activity levels. Neutrophils (4� 107 cells/ml)

were stimulated by cross-linking CD32 as indicated in Materials and methods. The reactions were stopped by transferring cell aliquots into the

boiling 2� SB for lanes 1–2 or a rapid transfer into a precooled Eppendorf tube and a rapid centrifugation for lanes 3–10. HLB with (lanes 3–

6) or without (lanes 7–10) MgCl2 was then added to the cell pellet. Aliquots of soluble (lanes 3–4 and 7–8) and insoluble (lanes 5–6 and 9–

10) fractions were diluted in an equal volume of 2� SB. Samples were analysed by immunoblot with an anti-phosphotyrosine antibody. In

panels B–D, the unstimulated cells were lysed with the appropriate HLB buffer. In panel B, the soluble fraction or starting material of both lysis

protocols was immunoblotted with anti-Lyn antibody. For immunoprecipitation in panels C and D, 200 ml of soluble material (8� 106 of cells)

was pre-incubated with an anti-Lyn antibody. Lyn activity was determined in the presence of ATP for 1 or 5 min in a kinase buffer. Precipitates

were analysed by Western blot with an anti-phosphotyrosine antibody.

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 93

as described in Materials and methods. The precip-

itates were probed with antiphosphotyrosine antibod-

ies as well as with anti-Cbl, anti-Syk or anti-CD32

antibodies (panels A, B, C, respectively). The results

obtained showed that phosphorylated Cbl, Syk and

CD32 were highly concentrated in the insoluble frac-

tions in agreement with the observations of Fig. 1.

Reblots with anti-Cbl, anti-Syk or anti-CD32 anti-

bodies (lower panels) indicated that cross-linking of

CD32 did not grossly modify the amounts of Cbl and

Syk present in each fraction. On the other hand,

significant decreases of CD32 were observed in the

soluble fractions which were compensated, in part at

least, by increases of CD32 in the insoluble frac-

tions.

3.4. Distribution of tyrosine phosphorylated proteins

between the soluble and insoluble fractions in

neutrophils stimulated with various agonists

The whole cell tyrosine phosphorylation profiles

observed in response to soluble and particulate ago-

nists in human neutrophils are characteristic of each

class of agonist used (Rollet et al., 1994). The dis-

tribution of phosphorylated proteins in the soluble and

insoluble fractions following lysis in HLB of cells

stimulated with a representative chemotactic factor

(fMet-Leu-Phe), a growth factor (GM-CSF) and a

particulate agonist (MSU crystals) was studied next

and the results are illustrated in Fig. 4. A comparison

of panels A, B and C indicates that the distribution of

the tyrosine phosphorylated substrates is both agonist-

and protein-dependent. For example, in response to

stimulation with fMet-Leu-Phe (panel A) and MSU

crystals (panel C), most of the tyrosine phosphory-

lated proteins were found in the insoluble fraction

except for the 42-kDa protein (most likely a MAP

kinase(s)) which was recovered in the soluble fraction.

In the case of stimulation by GM-CSF (panel B), the

tyrosine phosphorylated proteins were distributed

among the soluble and the insoluble fractions. Thus,

the majority, but not all of the tyrosine phosphorylated

Fig. 4. Distribution of tyrosine phosphorylated proteins in the

soluble and insoluble fractions after lysis in HLB in response to

various agonists. Neutrophils (4� 107 cells /ml) were stimulated at

37 �C with fMet-Leu-Phe 100 nM for 1 min (panel A) or with 3 nM

GM-CSF for 10 min (panel B) or with 3 mg/ml MSU for 10 min

(panel C). The reactions were stopped by transferring the cell

aliquots into boiling 2� SB for lanes 1–2 or a rapid transfer into a

precooled Eppendorf tube and a rapid centrifugation for lanes 3–6.

HLB was added to the cell pellet for lanes 3–6 (HLB) and the lysis

was carried out processed as described in Materials and methods.

Aliquots of soluble (lanes 3, 4) and insoluble (lanes 5, 6) fractions

were diluted in an equal volume of 2� SB. The samples were

analysed by immunoblot with an anti-phosphotyrosine antibody.

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–10194

proteins in response to the four agonists investigated

(CD32, fMet-Leu-Phe, GM-CSF and MSU crystals),

were recovered in the insoluble fraction. Furthermore,

the stimulatory effects of these agonists were pre-

served in this lysis protocol as the sum of the tyrosine

phosphorylated proteins present in the soluble and the

Fig. 5. Sequential solubility and analysis of hypotonic soluble and insoluble fractions. Neutrophils (4� 107 cells/ml) were stimulated with 3 mg/

ml of MSU crystals for 10 min at 37 �C. The reactions were stopped by a transfer of cell aliquots in the boiling 2� SB for lanes 1–2 or a rapid

transfer in a precooled Eppendorf tube and a rapid centrifugation for lanes 3–18. The cell pellets were resuspended in HLB and after a

centrifugation step, the soluble fraction (lanes 3–8) was transferred into a precooled ultracentrifuge tube and centrifuged for 45 min at

180,000� g as described in Materials and methods. Aliquots of the starting material of this ultracentrifugation (HLB soluble fraction, lanes 3–

4) and the supernatant (lanes 5–6) or pellet (lanes 7–8) ultracentrifugation and also aliquots of the hypotonic insoluble fractions (lanes 9–10)

were diluted 1/2 in a boiling 2� SB. The pellet of the hypotonic lysis was incubated in a second lysis buffer (ILB). After centrifugation, the

soluble material of this second lysis (lanes 11–12) was diluted in a 2� SB. The pellet of this isotonic lysis was incubated in a third lysis buffer

(HyperLB) and sonicated 2� 5 s. After centrifugation, the soluble materials of the third lysis (lanes 15–16) and the insoluble materials (lanes

17–18) of this lysis were diluted in a 2� SB. The samples were analysed by Western blot with anti-phosphotyrosine.

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 95

insoluble fractions closely corresponded to those

detected in the whole cell lysates.

3.5. Sequential solubilisation of the tyrosine phos-

phorylated proteins

Since most of the tyrosine phosphorylated proteins

were recovered in the pellets of the hypotonic lysis, a

further extraction of this pellet was warranted in order

to be able to carry out immunoprecipitations from this

fraction. On the other hand, we also wanted to better

characterize the soluble fraction of the hypotonic

lysis. The soluble and insoluble fractions of the

hypotonic lysis buffer were therefore individually

processed for subsequent analysis as indicated in

panel A of Fig. 5.

After stimulation with MSU crystals, the soluble

fraction (lanes 3–4) of the hypotonic lysis was cen-

trifuged for 45 min at 180,000� g. The resulting

soluble and insoluble fractions (lanes 5–8) were

analysed by immunoblotting with an anti-phosphotyr-

osine antibody. Two conclusions can be derived from

this analysis. First, the tyrosine phosphorylated pro-

teins were concentrated in the pellets of the ultra-

centrifugation step (lanes 7–8 vs. 5–6). Secondly,

these data confirm the stability of the tyrosine phos-

phorylation signals in the HLB soluble fractions.

The HLB insoluble fraction was then sequentially

lysed in buffers of increasing tonicities, the composi-

tion of which is described in Materials and methods.

Briefly, the HLB insoluble material was resuspended

in ILB and an aliquot was removed for direct analysis

(lanes 9–10). ILB soluble and insoluble fractions

were then prepared (lanes 11–14). The ILB insoluble

material was then resuspended in HyperLB and solu-

ble and insoluble fractions were isolated (lanes 15–

18). The relevant fractions were analyzed by immu-

noblotting with anti-phosphotyrosine antibodies

(panel B). The whole cell lysates (lanes 1–2) served

as the reference point for the extent of stimulation as

far as the global tyrosine phosphorylated profile was

concerned. As shown in Fig. 5, the majority of the

tyrosine phosphorylated proteins present in the pellet

of the HLB step remain insoluble in ILB (lanes 11 and

12 vs. lanes 13 and 14). It should be noted that the

profile of tyrosine phosphorylated proteins in the ILB

Fig. 6. Tyrosine kinase activity in the hypotonic and hypertonic soluble fractions. Neutrophils (4� 107 cells /ml) were stimulated by 3 mg/ml of

MSU crystals for 10 min at 37 �C. The reactions were stopped by a rapid transfer in a precooled eppendorf and a rapid centrifugation and the

HLB was added to the cell pellet. The aliquots (200 ml or 8� 106 equivalent cells) of the HLB lysate supernatants were diluted with 200 ml ofHyperLB and 400 ml of HLB. The pellet of the hypotonic lysis was directly incubated in a second lysis buffer (HyperLB) and sonicated 2� 5 s.

The soluble material (200 ml or 8� 106 equivalent cells) of this lysis was diluted in 3 volumes of HLB and incubated with the anti-Lyn

antibodies and the kinase activity was performed as described in Materials and methods. Autophosphorylation was visualised by Western blot

with anti-phosphotyrosine antibody (upper panel) and the amount of Lyn immunoprecipitated was monitored by Western blot with an anti-Lyn

antibody (lower panel).

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–10196

pellet still closely resembled that of the original

material (lanes 1 and 2). Finally, resuspension in

HyperLB did result in the solubilisation of some

proteins in the ILB pellet (lanes 15 and 16). However,

several tyrosine phosphorylated proteins remained

insoluble even under these more stringent conditions

(lanes 17–18).

A partial characterization (by immunoblotting) of

the fractions obtained during the sequential lysis

protocol followed herein was carried out (data not

shown). Cytosolic markers (myeloid related protein-

14 (MRP-14) and lactate dehydrogenase (LDH)) were

found predominantly in the soluble fraction of the first

lysis with HLB and in the soluble fractions of the

ultracentrifugation steps. The distribution of VLA2,

CD16, CD11b, CD45 or CD32 between the soluble

and insoluble fractions of HLB and HyperLB lysis

step was specific for each protein tested. Of relevance,

the receptor for cholera toxin (GM1) (Hart, 1975), a

major constituent of lipid rafts, was totally insoluble

in HLB buffer but was solubilized in HyperLB. Early

endosome-associated protein-1 (EEA-1), an endoso-

mal marker, was equally distributed between the

soluble and insoluble fractions of HLB. The HyperLB

step completely solubilized EEA-1. Lysosome-asso-

ciated membrane protein-1 (LAMP-1) was entirely

recovered in the HLB soluble fraction. Cytoskeletal

markers were also analyzed. A significant transloca-

tion of actin to the insoluble fraction was evident after

CD32 cross-linking as monitored by Coomassie blue

staining (Fig. 1). Paxillin and ezrin were equally

distributed between the soluble and the insoluble

fractions of the hypotonic lysis. Extraction with the

hypertonic lysis buffer solubilized these proteins.

Finally, nuclear markers (p62 nucleoporin) remained

insoluble and were found in the pellets of the

HyperLB step.

3.6. Lyn activity following its extraction from Hy-

perLB

In order to assess the state of the proteins extracted

in HyperLB, the immunoreactivity and enzymatic

activity of the Src kinase Lyn were tested next in

the soluble fractions of the HLB and HyperLB lysates.

We previously showed that Lyn is activated in re-

sponse to MSU crystals (Gaudry et al., 1995). We

wanted to confirm this result following the sequential

lysis described here. Neutrophils were exposed to 3

mg/ml of MSU crystals for 10 min before being lysed

in HLB. HLB soluble and insoluble fractions were

prepared as described in Fig. 5. Since we observed in

Fig. 5 that the HLB insoluble material was also

insoluble in ILB, the HLB insoluble fractions were

directly resuspended in HyperLB. Soluble and insolu-

ble fractions were prepared and Lyn was immuno-

precipitated from the HLB and HyperLB soluble

fractions. The results shown in Fig. 6 demonstrate

that Lyn immunoprecipitated from the HyperLB solu-

ble fractions retained in vitro kinase activity. How-

ever, a stimulatory effect of MSU crystals on the auto-

kinase activity of Lyn was only detectable in the

HyperLB lysates (Fig. 6, upper panel, right side).

Reblots indicated that equal amounts of Lyn were

immunoprecipitated and loaded in the control and

MSU-stimulated cells (Fig. 6, lower panel). The

results of preliminary experiments (data not shown)

indicated that the auto-kinase activity of Hck was

also increased in the soluble material of the hyper-

tonic lysis following stimulation with MSU crys-

tals.

4. Discussion

The results obtained during the course of these

studies provide a framework for the immunobiochem-

ical investigation of tyrosine phosphorylation-depend-

ent signalling pathways in human neutrophils. A

sequential lysis protocol in buffers of increasing toni-

cities is described that allows the preparation of

neutrophil lysates that retain their characteristic tyro-

sine phosphorylation profile and enzymatic activities.

This procedure also established that the detergent

solubility of individual tyrosine phosphorylated sub-

strates differed, not only according to the protein

under investigation, but also in response to the spe-

cific agonist used.

The basis of this method is an initial lysis in a

hypotonic buffer. Under these conditions, the overall

tyrosine phosphorylation profile is preserved to a

major extent. The two fractions thus obtained (soluble

and insoluble) can then be further analyzed, either

directly (in the case of the soluble fraction) or follow-

ing its solubilization in buffers of increasing tonicities

(in the case of the original insoluble fraction). Addi-

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 97

tionally, an ultra-centrifugation step can be utilized to

further characterize the fraction soluble in the hypo-

tonic lysis buffer. Each of the fractions can be ana-

lyzed by immunoblotting or by monitoring enzymatic

activities. This general protocol can be adapted to suit

individual experimental requirements.

The severe experimental problems associated with

the preparation of neutrophil lysates (proteolysis,

dephosphorylation or hyperphosphorylation) are

widely acknowledged but poorly documented. The

data presented in this report confirm and extend

previously published observations concerning the

lability of neutrophil lysates (Al-Shami et al., 1997b;

McDonald et al., 1998). The characteristic tyrosine

phosphorylation profiles observed upon direct transfer

of suspensions of neutrophils into boiling sample

buffers are grossly altered upon lysis in classical RIPA

buffers (see Al-Shami et al., 1997b; and also Fig. 1).

A previous attempt to preserve the original patterns of

tyrosine phosphorylation relied on a denaturing lysis

protocol which, while effective, eliminated the possi-

bility of studying protein–protein interactions and

enzymatic activities (i.e., kinases and phosphatases)

in the lysates (Al-Shami et al., 1997b). The sequential

lysis protocol described in the present manuscript

overcomes these limitations in that lysates prepared

under native conditions maintain to a significant

extent their profile of tyrosine phosphorylation as well

as their enzymatic activities.

Furthermore, whereas tyrosine phosphorylation is

transient in neutrophils, we observed that tyrosine

phosphorylation in the insoluble fraction is well con-

served even 60 min following cell lysis. This can be

explained first by the technique used to stop the

stimulation, i.e. rapid transfer of the cells in pre-

cooled tubes followed by rapid centrifugation. This

approach immediately eliminates the incubation me-

dium which may contain degradative products (pro-

teases, phosphatases) released by the cells during

stimulation.

Cross-linking of CD32 induces its insolubility as

described in Figs. 1 and 3. Insolubility in non-ionic

detergent coupled with concentration of tyrosine

phosphorylated specific substrates (Cbl, Syk, and

CD32; Figs. 1 and 3) in the HLB insoluble fractions

suggests the implication of the lipid raft structures or

detergent resistant membranes (DRMs). Insolubilisa-

tion of membrane receptors coupled to kinase activ-

ities was also observed in T cell lines (Solomon et al.,

1998) and in various others cells (Zhou et al., 1995).

One of the main findings of the present investiga-

tion was that the detergent extractability of the tyro-

sine phosphorylated substrates was both substrate and

agonist dependent. Nevertheless, a large percentage of

the tyrosine phosphorylated proteins was insoluble in

regular RIPA buffers, of hypo- or iso-tonicity. This

point raises questions about the interpretation of

immunoprecipitation studies in which the soluble

fraction is used as starting material in the absence of

a detailed picture of the distribution of the protein

being examined. Furthermore, the level of phospho-

rylation of various proteins can be artefactually in-

creased by the presence of Mg2 + in the lysis buffers, a

procedure sometimes adopted to help preserve mem-

brane and cytoskeletal integrity. Once again, this may

impact on the interpretation of the functional signifi-

cance of apparent stimulation, or lack of stimulation,

of the tyrosine phosphorylation of specific substrates

in those cases where the profiles of tyrosine phos-

phorylation of the lysates used as starting material for

the immunoprecipitation differ significantly from that

of whole cells.

The importance of knowing the distribution of

proteins in detergent extracts of the cells is strikingly

illustrated by the behavior of CD32, Cbl and Syk in

response to stimulation by CD32 ligation. All three

proteins were rapidly tyrosine phosphorylated follow-

ing CD32 cross-linking (Marcilla et al., 1995; Nacc-

ache et al., 1997; Rollet-Labelle et al., 2000) and, as

shown in Fig. 3, they were recovered in both the

soluble and the insoluble fractions. However, in all

three cases, the tyrosine phosphorylated proteins were

highly concentrated in the detergent-insoluble frac-

tions. The physiological relevance of the soluble form

of these proteins is therefore likely to differ from that

present in the insoluble fraction.

The functional relevance of the above considera-

tions is also highlighted by the results of the experi-

ments in which the kinase activity of Lyn was

monitored in the soluble fractions of both the hypo-

tonic and hypertonic lysis buffers. The enzyme

remained active in both fractions, an indication of

preservation of structure and function in the lysates.

Of more functional relevance, however, was the

observation that the stimulatory effects of MSU crys-

tals on the activity of Lyn were only readily detectable

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–10198

in the hypertonic buffer lysate. A study of the soluble

fraction of the hypotonic lysate would have missed

this effect, while a lysis in an isotonic buffer would be

likely to result in a significant level of signal distor-

tion (proteolysis, dephosphorylation). The sequential

lysis protocol characterized in the present study over-

comes several of these problems by preserving the

original phosphorylation status and by sequentially

giving access to the various fractions. These observa-

tions indicate that both fractions must always be

examined in this kind of studies. These data are

consistent with the results of Zhou et al. (1995) who

reported previously a partitionning of Src kinase

activities between the soluble and the insoluble frac-

tions derived from adherent neutrophils. These en-

zymes are involved in the early steps of several neu-

trophil responses and are differentially regulated

depending on their intracellular distribution (Welch

and Maridonneau-Parini, 1997). Our results also indi-

cate that, depending on the agonist used, the distribu-

tion of the tyrosine phosphorylated substrates varies

and must be individually characterized. Furthermore,

the addition of cofactors in the lysis buffer must be

carefully controlled.

In conclusion, we demonstrated that the detergent

insoluble fractions cannot be excluded in the bio-

chemical studies of stimulated human neutrophils.

The concentration of signalling molecules in this

fraction may explain some of the controversial results

concerning tyrosine phosphorylated proteins, tyrosine

kinase activities or protein associations which have

been reported in stimulated neutrophils. The present

approach, which is based on the examination of all

cell fractions, may help to understand the biochemical

mechanisms of neutrophil responses. This method

may also be extended to the monitoring of phospha-

tase activities if NaVO4 is omitted from the lysis

buffers allowing the evaluation of the balance of

kinase and phosphatase activities in stimulated cells

which is most likely to be a key to the regulation of

several functional responses in many cells.

Acknowledgements

Supported in part by grants and fellowships from the

Canadian Institutes of Health Research and from the

Arthritis Society of Canada. C. Gilbert was supported

by fellowships from the K.M. Hunter Charitable

Foundation, the Canadian Institutes of Health Research

and the Fonds pour la Formation de Chercheurs et

l’Aide a la Recherche and the Fonds de la Recherche en

Sante du Quebec.

The authors wish to thank Mr. Sylvain Levasseur

for his expert help with the hybridoma cultures and

the purification of antibodies and Dr. Maria Fernandes

for carefully reading this manuscript.

References

Al-Shami, A., Bourgoin, S.G., Naccache, P.H., 1997a. Granulocyte-

macrophage colony-stimulating factor-activated signaling path-

ways in human neutrophils: I. Tyrosine phosphorylation-depend-

ent stimulation of phosphatidylinositol 3-kinase and inhibition

by phorbol esters. Blood 89, 1035–1044.

Al-Shami, A., Gilbert, C., Barabe, F., Gaudry, M., Naccache, P.H.,

1997b. Preservation of the pattern of tyrosine phosphorylation in

human neutrophil lysates. J. Immunol. Methods 202, 183–191.

Al-Shami, A., Mahanna, W., Naccache, P.H., 1998. Granulocyte-

macrophage colony-stimulating factor-activated signaling path-

ways in human neutrophils. Selective activation of Jak2, Stat3,

and Stat5b. J. Biol. Chem. 273, 1058–1063.

Asahi, M., Taniguchi, T., Hashimoto, E., Inazu, T., Maeda, H.,

Yamamura, H., 1993. Activation of protein-tyrosine kinase

p72syk with concanavalin A in polymorphonuclear neutrophils.

J. Biol. Chem. 268, 23334–23338.

Barabe, F., Gilbert, C., Liao, N., Bourgoin, S.G., Naccache, P.H.,

1998. Crystal-induced neutrophil activation: VI. Involvement of

FcgammaRIIIB (CD16) and CD11b in response to inflammatory

microcrystals. FASEB J. 12, 209–220.

Borregaard, N., Cowland, J.B., 1997. Granules of the human neu-

trophilic polymorphonuclear leukocyte. Blood 89, 3503–3521.

Brumell, J.H., Chan, C.K., Butler, J., Borregaard, N., Siminovitch,

K.A., Grinstein, S., Downey, G.P., 1997. Regulation of Src

homology 2-containing tyrosine phosphatase 1 during activation

of human neutrophils. Role of protein kinase C. J. Biol. Chem.

272, 875–882.

Cassatella, M.A., 1996. Cytokines Produced by Polymorphonuclear

Neutrophils: Molecular and Biological Aspects. Chapman &

Hall, Austin.

Cui, Y., Harvey, K., Alcard, L., Jansen, J., Hughes, C., Siddiqui,

R.A., English, D., 1994. Regulation of neutrophil responses by

phosphotyrosine phosphatase. J. Immunol. 152, 5420–5428.

Cui, Y., Harvey, K.A., Siddiqui, R.A., Jansen, J., Akard, L.P.,

Thompson, J.M., Garcia, J.G., English, D., 1996. Cytosolic in-

activation of translocated neutrophil plasma membrane protein

tyrosine phosphatase. Blood 87, 341–349.

Fernandez, R., Suchard, S.J., 1998. Syk activation is required for

spreading and H2O2 release in adherent human neutrophils. J.

Immunol. 160, 5154–5162.

Fialkow, L., Chan, C.K., Downey, G.P., 1997. Inhibition of CD45

during neutrophil activation. J. Immunol. 158, 5409–5417.

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 99

Fuortes, M., Jin, W.W., Nathan, C., 1994. Beta 2 integrin-dependent

tyrosine phosphorylation of paxillin in human neutrophils trea-

ted with tumor necrosis factor. J. Cell Biol. 127, 1477–1483.

Gaudry, M., Caon, A.C., Gilbert, C., Lille, S., Naccache, P.H., 1992.

Evidence for the involvement of tyrosine kinases in the loco-

motory responses of human neutrophils. J. Leukocyte Biol. 51,

103–108.

Gaudry, M., Roberge, C.J., de Medicis, R., Lussier, A., Poubelle,

P.E., Naccache, P.H., 1993. Crystal-induced neutrophil activa-

tion: III. Inflammatory microcrystals induce a distinct pattern

of tyrosine phosphorylation in human neutrophils. J. Clin. Invest.

91, 1649–1655.

Gaudry, M., Gilbert, C., Barabe, F., Poubelle, P.E., Naccache, P.H.,

1995. Activation of Lyn is a common element of the stimulation

of human neutrophils by soluble and particulate agonists. Blood

86, 3567–3574.

Gilbert, C., Barabe, F., Rollet-Labelle, E., Bourgoin, S.G., McColl,

S.R., Damaj, B.B., Naccache, P.H., 2001. Evidence for a role for

SAM68 in the responses of human neutrophils to ligation of

CD32 and to monosodium urate crystals. J. Immunol. 166,

4664–4671.

Grinstein, S., Furuya, W., 1992. Chemoattractant-induced tyrosine

phosphorylation and activation of microtubule-associated pro-

tein kinase in human neutrophils. J. Biol. Chem. 267, 18122–

18125.

Grinstein, S., Furuya, W., Lu, D.J., Mills, G.B., 1990. Vanadate

stimulates oxygen consumption and tyrosine phosphorylation

in electropermeabilized human neutrophils. J. Biol. Chem. 265,

318–327.

Gutkind, J.S., Robbins, K.C., 1989. Translocation of the FGR pro-

tein-tyrosine kinase as a consequence of neutrophil activation.

Proc. Natl. Acad. Sci. U. S. A. 86, 8783–8787.

Hachicha, M., Naccache, P.H., McColl, S.R., 1995. Inflammatory

microcrystals differentially regulate the secretion of macrophage

inflammatory protein 1 and interleukin 8 by human neutrophils:

a possible mechanism of neutrophil recruitment to sites of in-

flammation in synovitis. J. Exp. Med. 182, 2019–2025.

Hamada, F., Aoki, M., Akiyama, T., Toyoshima, K., 1993. Associ-

ation of immunoglobulin G Fc receptor II with Src-like protein-

tyrosine kinase Fgr in neutrophils. Proc. Natl. Acad. Sci. U. S. A.

90, 6305–6309.

Hart, D., 1975. Evidence for the non-protein nature of the receptor

for the enterotoxin in Vibrio cholerae on murine lymphoid cells.

Infect. Immun. 11, 742–747.

Hundt, M., Schmidt, R.E., 1997. Functional characterization of re-

ceptor-type protein tyrosine phosphatase CD148 (HPTP eta/

DEP-1) in Fc gamma receptor IIa signal transduction of human

neutrophils. Eur. J. Immunol. 27, 3532–3535.

Ibarrola, I., Vossebeld, P.J., Homburg, C.H., Thelen, M., Roos, D.,

Verhoeven, A.J., 1997. Influence of tyrosine phosphorylation on

protein interaction with FcgammaRIIa. Biochim. Biophys. Acta

1357, 348–358.

Khamzina, L., Borgeat, P., 1998. Correlation of alpha-fetoprotein

expression in normal hepatocytes during development with ty-

rosine phosphorylation and insulin receptor expression. Mol.

Biol. Cell 9, 1093–1105.

Krump, E., Nikitas, K., Grinstein, S., 1997. Induction of tyrosine

phosphorylation and Na + /H + exchanger activation during

shrinkage of human neutrophils. J. Biol. Chem. 272, 17303–

17311.

Marcilla, A., Rivero-Lezcano, O.M., Agarwal, A., Robbins, K.C.,

1995. Identification of the major tyrosine kinase substrate in

signaling complexes formed after engagement of Fc gamma

receptors. J. Biol. Chem. 270, 9115–9120.

McDonald, P.P., McColl, S.R., Naccache, P.H., Borgeat, P., 1991.

Studies on the activation of human neutrophil 5-lipoxygenase

induced by natural agonists and Ca2 + ionosphere A23187. Bio-

chem. J. 280, 379–385.

McDonald, P.P., Bovolenta, C., Cassatella, M., 1998. Activation of

distinct transcription factors in neutrophils by bacterial LPS,

interferon gamma and GM-CSF and necessity to overcome the

action of endogenous proteases. Biochemistry 37, 13173–

13265.

Naccache, P.H., Gilbert, C., Caon, A.C., Gaudry, M., Huang, C.K.,

Bonak, V.A., Umezawa, K., McColl, S.R., 1990. Selective in-

hibition of human neutrophil functional responsiveness by erb-

statin, an inhibitor of tyrosine protein kinase. Blood 76, 2098–

2104.

Naccache, P.H., Jean, N., Liao, N.W., Bator, J.M., McColl, S.R.,

Kubes, P., 1994. Regulation of stimulated integrin surface ex-

pression in human neutrophils by tyrosine phosphorylation.

Blood 84, 616–624.

Naccache, P.H., Gilbert, C., Barabe, F., Al-Shami, A., Mahana, W.,

Bourgoin, S.G., 1997. Agonist-specific tyrosine phosphoryla-

tion of Cbl in human neutrophils. J. Leukocyte Biol. 62, 901–

910.

Nahas, N., Molski, T.F., Fernandez, G.A., Sha’afi, R.I., 1996. Ty-

rosine phosphorylation and activation of a new mitogen-acti-

vated protein (MAP)-kinase cascade in human neutrophils

stimulated with various agonists. Biochem. J. 318, 247–253.

Pouliot, M., McDonald, P.P., Krump, E., Mancini, J.A., McColl,

S.R., Weech, P.K., Borgeat, P., 1996. Colocalization of cytosolic

phospholipase A2, 5-lipoxygenase, and 5- lipoxygenase-activat-

ing protein at the nuclear membrane of A23187-stimulated hu-

man neutrophils. Eur. J. Biochem. 238, 250–258.

Pouliot, M., Gilbert, C., Borgeat, P., Poubelle, P.E., Bourgoin, S.,

Creminon, C., Maclouf, J., McColl, S.R., Naccache, P.H., 1998.

Expression and activity of prostaglandin endoperoxide synthase-

2 in agonist-activated human neutrophils. FASEB J. 12, 1109–

1123.

Roberge, C.J., Gaudry, M., de Medicis, R., Lussier, A., Poubelle,

P.E., Naccache, P.H., 1993. Crystal-induced neutrophil activa-

tion: IV. Specific inhibition of tyrosine phosphorylation by col-

chicine [see comments]. J. Clin. Invest 92, 1722–1729.

Rollet, E., Caon, A.C., Roberge, C.J., Liao, N.W., Malawista, S.E.,

McColl, S.R., Naccache, P.H., 1994. Tyrosine phosphorylation

in activated human neutrophils. Comparison of the effects of

different classes of agonists and identification of the signaling

pathways involved. J. Immunol. 153, 353–363.

Rollet-Labelle, E., Gilbert, C., Naccache, P.H., 2000. Modulation of

human neutrophil responses to CD32 cross-linking by serine/

threonine phosphatase inhibitors: cross-talk between serine/

threonine and tyrosine phosphorylation. J. Immunol. 164, 1020–

1028.

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101100

Rosen, H., Michel, B.R., 1997. Redundant contribution of myelo-

peroxidase-dependent systems to neutrophil-mediated killing of

Escherichia coli. Infect. Immun. 65, 4173–4178.

Skubitz, K.M., Ahmed, K., Campbell, K.D., Skubitz, A.P., 1995.

CD50 (ICAM-3) is phosphorylated on tyrosine and is associated

with tyrosine kinase activity in human neutrophils. J. Immunol.

154, 2888–2895.

Smith, J.A., 1994. Neutrophils, host defense, and inflammation: a

double-edged sword. J. Leukocyte Biol. 56, 672.

Solomon, K.R., Mallory, M.A., Finberg, R.W., 1998. Determination

of the non-ionic detergent insolubility and phosphoprotein asso-

ciations of glycosylphosphatidylinositol-anchored proteins ex-

pressed on T cells. Biochem. J. 334, 325–333.

Tidow, N., Kasper, B., Welte, K., 1999. SH2-containing protein

tyrosine phosphatases SHP-1 and SHP-2 are dramatically in-

creased at the protein level in neutrophils from patients with

severe congenital neutropenia (Kostmann’s syndrome). Exp.

Hematol. 27, 1038–1045.

Torres, M., Hall, F.L., O’Neill, K., 1993. Stimulation of human

neutrophils with formyl-methionyl-leucyl- phenylalanine indu-

ces tyrosine phosphorylation and activation of two distinct mi-

togen-activated protein-kinases. J. Immunol. 150, 1563–1577.

Welch, H., Maridonneau-Parini, I., 1997. Lyn and Fgr are activated

in distinct membrane fractions of human granulocytic cells. On-

cogene 15, 2021–2029.

Welch, H., Mauran, C., Maridonneau-Parini, I., 1996. Nonreceptor

protein-tyrosine kinases in neutrophil activation. Methods 9,

607–618.

Willeke, T., Behrens, S., Scharffetter-Kochanek, K., Gaehtgens, P.,

Walzog, B., 2000. Beta2 integrin (CD11/CD18)-mediated sig-

naling involves tyrosine phosphorylation of c-Cbl in human

neutrophils. J. Leukocyte Biol. 68, 284–292.

Woods Ignatoski, K.M., Verderame, M., 1996. Lysis buffer compo-

sition dramatically affects extraction of phospho-tyrosine-con-

taining proteins. BioTechniques 20, 794–796.

Yan, S.R., Berton, G., 1996. Regulation of Src family tyrosine

kinase activities in adherent human neutrophils. Evidence that

reactive oxygen intermediates produced by adherent neutrophils

increase the activity of the p58c-fgr and p53/56lyn tyrosine

kinases. J. Biol. Chem. 271, 23464–23471.

Yan, S.R., Novak, M.J., 1999. Beta2 integrin-dependent phosphor-

ylation of protein-tyrosine kinase Pyk2 stimulated by tumor

necrosis factor alpha and fMLP in human neutrophils adherent

to fibrinogen [In Process Citation]. FEBS Lett. 451, 33–38.

Yan, S.R., Fumagalli, L., Berton, G., 1996. Activation of SRC

family kinases in human neutrophils. Evidence that p58C-

FGR and p53/56LYN redistributed to a Triton X-100-insoluble

cytoskeletal fraction, also enriched in the caveolar protein cav-

eolin, display an enhanced kinase activity. FEBS Lett. 380,

198–203.

Yan, S.R., Huang, M., Berton, G., 1997. Signaling by adhesion in

human neutrophils: activation of the p72syk tyrosine kinase and

formation of protein complexes containing p72syk and Src fam-

ily kinases in neutrophils spreading over fibrinogen. J. Immunol.

158, 1902–1910.

Zhou, M.J., Lublin, D.M., Link, D.C., Brown, E.J., 1995. Distinct

tyrosine kinase activation and Triton X-100 insolubility upon Fc

gamma RII or Fc gamma RIIIB ligation in human polymorpho-

nuclear leukocytes. Implications for immune complex activation

of the respiratory burst. J. Biol. Chem. 270, 13553–13560.

C. Gilbert et al. / Journal of Immunological Methods 261 (2002) 85–101 101