Experimental Cell Research 279, 141–152 (2002)
doi:10.1006/excr.2002.5581
CD4 Is Active as a Signaling Molecule on the Human
Monocytic Cell Line Thp-1
Gina M. Graziani-Bowering,* Lionel G. Filion,* Pierre Thibault,† and Maya Kozlowski* ,‡ ,1
*Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ottawa, Canada K1H 8M5; †Institute for Biological
Sciences, National Research Council, Ottawa, Canada K1A 0R6; and ‡Biologics and Genetics Therapies Directorate,
Centre for Biologics Research, Health Canada, Ottawa, Canada K1A 0L2
CD4 is a 56-kDa membrane glycoprotein expressed
by a subset of T cells, by cells of the monocyte/macro-
phage lineage, and by eosinophils and dendritic cells.
CD4 serves as a coreceptor for HIV and IL-16. T cell
CD4 mediates signal transduction by associating with
the protein tyrosine kinase p56 lck; this interaction does
not exist in monocytes. We wished to elucidate the
mechanism(s) by which monocyte CD4 transduces sig-
nals. Stimulation of CD4 on Thp-1 monocytic cells in-
duced a Ca 2ⴙ flux and the time-dependent activation of
phosphotyrosine proteins ranging from 35 to 180 kDa.
We identified the 140- and 85-kDa proteins as phospho-
lipase C gamma (PLC-␥) and the regulatory subunit of
phosphatidylinositol 3-kinase (PI-3K), respectively.
Using immunoprecipitation/Western immunoblotting
however, we were unable to show any direct associa-
tion between CD4 and PLC-␥, PI-3K, or other known
signaling proteins. To identify proteins capable of as-
sociating with the cytoplasmic tail of CD4, we fused it
with gluthatione S-transferase and used the fusion
protein in far Western and pull-down experiments. In
both types of experiments, the fusion protein rou-
tinely associated with 45- and 55-kDa proteins. Mass
spectrometry analysis of the tryptic peptides gener-
ated from these two proteins indicated novel se-
quences. © 2002 Elsevier Science (USA)
Key Words: CD4; monocytes; Thp-1; signal transduc-
tion; Ca 2ⴙ flux; PI3-K; PLC-␥1.
INTRODUCTION
The CD4 molecule is a 56-kDa membrane glycopro-
tein [1]. In humans, CD4 is expressed by a subset of T
cells [1], cells of the monocyte/macrophage lineage [2],
dendritic cells [2], and eosinophils [3]. The CD4 mole-
cule has been the subject of intense study in recent
years given its function as a HIV coreceptor [4 – 6], in
1
To whom reprint requests should be addressed at Health Can-
ada, Biologics and Genetics Therapies Directorate, Centre for Bio-
logics Research, Tunney’s Pasture, Ottawa, ON, Canada K1A 0L2,
Postal Locator 2201C. E-mail: Maya_kozlowski@hc-sc.gc.ca
141
conjunction with the seven-transmembrane G-protein-
coupled chemokine receptors CCR5 and CXCR4 [7–9].
CD4 is also a coreceptor [10], apparently in conjunction
with these same chemokine receptors [11, 12], for the
chemokine IL-16. Additional functions for CD4 have
also been identified for T cell CD4, whereas very little
is known about CD4 in human monocytes/macro-
phages and other CD4 ⫹ cell populations.
In CD4 ⫹ T cells, CD4 plays a crucial role in the signal
transduction response of the cells to antigens pre-
sented to them in the context of MHC class II mole-
cules (reviewed in [13]). This signal is mediated by
virtue of the association of CD4 with the Src-related
protein tyrosine kinase p56 lck, which functions to ty-
rosine-phosphorylate proteins associated with the T
cell antigen receptor complex [14, 15]. The end result is
the induction of the T cell proliferation, differentiation,
and cytokine synthesis and release required to mount
an effective immune response to the triggering antigen
(reviewed in [16]). Furthermore, in T cells, HIV-in-
duced apoptosis is mediated by the CD4-p56 lck complex;
the downstream signaling events are unknown [17].
Such a process may contribute to the depletion of CD4 ⫹
T cells, a hallmark of AIDS pathogenesis [17].
Unlike T cell CD4, monocytic CD4 has no associated
kinase activity [18 –20]. Some investigators, however,
have observed CD4-mediated signaling in monocyte/
macrophages. For example, a complex inositol poly-
phosphate response and a Ca 2⫹ flux were induced in
monocytes in which CD4 and Fc␥ receptors were co-
cross-linked [21]. IL-16 stimulation of various CD4 ⫹
monocytic cell lines induces a translocation of PKC
from the cytosol to the membrane [22], as well as
activation of the SAPK 2 and p38 MAPK signaling path-
ways [23]. Finally, gp120 of HIV can induce a CD4-
2
Abbreviations used: Ab, antibody; APS, ammonium persulfate;
Ca 2⫹, calcium ion; CD4 cyt, cytoplasmic tail of CD4; DAG, diacylglyc-
erol; DTT, dithiothreitol; ERK, extracellular signal-regulated kinase;
FBS, fetal bovine serum; FITC, fluorescein isothiocyanate; GST,
glutathione S-transferase; HCK, hematopoietic cell kinase; Hepes,
N-2-hydroxethylpiperazine-N-2-ethanesulfonic acid; IL-1␤, interleukin
1 beta; IL-16, interleukin 16; InsP3, inositol 1,4,5-trisphosphate; IPTG,
isopropylthio-␤-D-galactoside; 2ME, 2-mercaptoethanol; MAPK, mito-
0014-4827/02 $35.00
© 2002 Elsevier Science (USA)
All rights reserved.
142 GRAZIANI-BOWERING ET AL.
dependent down-regulation of the CCR5 chemokine
receptor in monocytes [24].
The goals of this study were to expand on earlier
observations that monocyte CD4 was capable of signal
transduction and to begin to elucidate the mechanism
by which this process occurs. A better understanding of
this process may provide us with further insights into
both the normal physiological role of monocyte CD4
and the role of this molecule in HIV/AIDS pathogene-
sis; in contrast to CD4 ⫹ T cells, which can receive an
apoptotic signal in response to HIV, HIV-infected
monocytes/macrophages persist, although functionally
impaired, and serve as HIV reservoirs [25, 26].
MATERIALS AND METHODS
Antibodies and sCD4. Rabbit anti-mouse Ig and rabbit anti-goat
IgG were obtained from Sigma (St. Louis, MO), murine antiphospho-
tyrosine mAb clone 4G10 and murine anti-PI3-K (p85) mAb were
obtained from Upstate Biotechnology Inc. (Lake Placid, NY), murine
anti-GST mAb, rabbit polyclonal Ab to PLC-␥1, and goat polyclonal
Ab to actin were obtained from Santa Cruz Biotechnology (Santa
Cruz, CA), and FITC-conjugated goat anti-rabbit IgG was obtained
from Cedarlane Laboratories (Toronto, ON). Rabbit serum to Shp-1
was generated in-house as previously described [27]. T4-4 rabbit sera
to sCD4 and sCD4 were obtained through the AIDS Research and
Reference Reagent Program, Division of AIDS, NIAID, NIH (Rock-
ville, MD), courtesy of Dr. R. Sweet, SmithKline Beecham Pharma-
ceuticals [28].
Production of rabbit anti-CD4 sera and F(ab⬘) 2 fragment prepara-
tion. New Zealand White rabbits were obtained from Charles River
Laboratories (Montreal, QC), pre-bled, and immunized with 20 ␮g of
sCD4 employing Ribi (Cedarlane Laboratories) as an adjuvant. The
experimental protocol was pre-approved by the University of Ottawa
Animal Care Committee and was in accordance with the guidelines
of the Canadian Council of Animal Care. The production of our own
sera was necessary due to the volumes of sera required to generate
the F(ab⬘) 2 fragment. The sCD4 antigen employed was the same as
that used by the NIH AIDS depository for the production of their
T4-4 sera. The rabbits were boosted with the same antigen prepara-
tion and adjuvant at monthly intervals. Test bleeds were obtained 2
weeks after each immunization and were tested by flow cytometry
for anti-CD4 reactivity using Thp-1 cells. LF T4-4 sera from different
months were pooled and the IgG fraction was isolated by affinity
chromatography on a protein A–Sepharose column. The digestion of
the Fc portion was performed with pepsin. Briefly, 0.1 mg/ml of
pepsin in acetate buffer was added to purified IgG to give an enzyme/
antibody ratio of 1:20, and the reaction mixture was incubated for 5 h
at 37°C. The reaction was stopped by adding Tris base (2.0 M, pH
10.5), and the mixture was small-volume dialyzed against PBS (0.01
M, pH 8.0) at 4°C. The separation of intact IgG molecules from the
F(ab⬘) 2 fragments was performed using a protein A–Sepharose col-
gen-activated protein kinase(s); mAb, monoclonal antibody; MIP-1␤,
macrophage inflammatory protein-1 beta; NaCl, sodium chloride; NaF,
sodium fluoride; NaN3, sodium azide; NHS, N-hydroxysuccinimide;
PI3-K, phosphatidylinositol 3-kinase; PIP2, phosphatidylinositol 4,5-
bisphosphate; PKC, protein kinase C; PLC-␥1, phospholipase C-gamma
one; RT, room temperature; SAPK, stress-activated protein kinase;
sCD4, soluble CD4; SDF-1␣, stromal-derived factor-1␣; SLP-76, Src
homology 2 domain-containing leukocyte protein of 76 kDa; TCR, T cell
receptor; TEMED, N,N,N,N⬘-tetramethylethylenediamine; TNF-␣, tu-
mor necrosis factor alpha.
umn [29]. The putative fragment was passed three times over the
protein A–Sepharose column to ensure that no intact IgG molecules
were present in the eluate of the column. Isolated F(ab⬘) 2 fragment
was electrophoresed on a 7.5% SDS–PAGE to determine if there was
free intact IgG present in the preparation; intact IgG was also
electrophoresed as a control.
Verification of LF T4-4 F(ab⬘) 2 binding to CD4 on Thp-1. The LF
T4-4 F(ab⬘) 2 was tested by flow cytometry for reactivity to CD4.
Thp-1 cells (10 5) were incubated for 15 min with LF T4-4, LF T4-4
F(ab⬘) 2, or prebleed sera, each used at various dilutions to determine
maximal antibody binding. Cells were washed with PBS containing
0.01 M NaN 3 and resuspended in 100 ␮l of PBS, NaN 3 containing
10% human AB serum. Cells were incubated for 15 min with 1 ␮l of
FITC-conjugated goat antirabbit IgG, washed, resuspended in PBS,
NaN 3, and analyzed by flow cytometry (data not shown).
Cell culture and stimulation. Thp-1 cells [30] were cultured at
37°C, 5% CO 2, in RPMI 1640 (Gibco BRL, Grand Island, NY) sup-
plemented with 10% (v/v) FBS (Cansera, Rexdale, ON), 10 units/ml
of penicillin (Gibco BRL), and 10 ␮g/ml of streptomycin (Gibco BRL).
Prior to stimulation with T4-4 rabbit serum to sCD4, the cells were
cultured overnight in medium containing 2% (v/v) FBS, and washed
the next morning for 5 min at RT, and centrifuged at 200g with PBS
(Gibco BRL) prior to being further starved for 3 to 4 h by culture in
serum-free medium. Following serum starvation, the cells were col-
lected by centrifugation for 5 min at RT, 200g. Cell numbers and
viability were determined using trypan blue and volumes of 10 ⫻ 10 6
cells/ml were aliquoted. Cells were stimulated by the addition of 25
␮g/ml of T4-4 serum and subsequent incubation for 3 min at 37°C. As
a negative control, starved but unstimulated Thp-1 cells were also
incubated at 37°C for 3 min. In some experiments, the following
negative control stimulations were also included: 25 ␮g/ml of rabbit
serum to Shp-1 and 25 ␮g/ml of T4-4 serum which had been pread-
sorbed by incubation with either sCD4 or BSA immobilized on NHS-
activated Sepharose beads (Pharmacia Biotech, Baie D’Urfé, PQ),
according to the manufacturer’s instructions. Following stimulation,
the cells were washed with cold PBS for 5 min at 4°C, 200g, and the
pellets were stored at ⫺80°C until further processing.
In experiments in which cells were stimulated with LF T4-4
F(ab⬘) 2, a plate-bound method was used. Briefly, Costar six-well
cluster plates were coated overnight at 4°C with 10 ␮g/ml of LF T4-4
F(ab⬘) 2. Prior to use, the plates were washed three times with sterile
HBSS, pH 7.2. For cell stimulation, 5 ⫻ 10 6 Thp-1 cells were added
to the F(ab⬘) 2-coated wells and the plates were incubated for 3 min at
37°C. To stop the stimulation, the plates were placed on ice, the cells
were collected and washed at 4°C, and the lysates were prepared as
described below.
Preparation of cell lysates. Unstimulated and stimulated cell pel-
lets were lysed in buffer consisting of 20 mM Tris–HCl, pH 8 (Sigma),
150 mM NaCl (BDH, Toronto, ON), 1% (v/v) NP-40 (Sigma), 10%
(v/v) glycerol (BDH), 10 mM NaF (Sigma), 25 ␮g/ml of aprotinin
(Sigma), 25 ␮g/ml of leupeptin (Sigma), and 1 mM sodium orthovana-
date (Sigma). The resuspended pellets were incubated for 30 to 60
min on ice, and the lysates were centrifuged for 20 min at 4°C, 9850g,
to separate insoluble material from the lysates. Protein concentra-
tion was determined by a colorimetric assay (Bio-Rad, Mississauga,
ON) according to the manufacturer’s directions.
Immunoprecipitation. Immunoprecipitation reactions consisted
of 1.5 to 2 mg of Thp-1 protein lysate, 50 ␮l of a 50% (v/v) protein A-
or protein G–Sepharose 6MB Bead slurry (Pharmacia Biotech), and
the recommended amount of antibody. Lysis buffer was added to
bring the immunoprecipitation reactions to a final volume of 1 ml,
and the reactions were incubated by inversion for 2 h at 4°C. The
immune complexes were collected by centrifugation for 5 min at 4°C,
9850g, and washed three times for 5 min at 4°C, 9850g, with 1 ml
lysis buffer. The final pellets were resuspended with 25 ␮l of
Laemmli sample buffer (Bio-Rad).
 CD4 SIGNAL TRANSDUCTION IN MONOCYTIC CELLS
Western blotting. Immune complexes were electrophoresed at
107 V (at RT) through SDS–polyacrylamide gels. The resolving gels
consisted of 8 to 10% (w/v) acrylamide (Bio-Rad) in 375 mM Tris–
HCl, pH 8.1 (Gibco BRL), 0.1% (w/v) SDS (Bio-Rad), 0.15% (w/v) APS
(Bio-Rad), and 0.2% (v/v) TEMED (Sigma). The stacking gels con-
sisted of 5% (w/v) acrylamide in 127 mM Tris–HCl, pH 6.8, 0.1%
(w/v) SDS, 0.2% (w/v) APS, and 0.2% (v/v) TEMED. The electropho-
resed proteins were electroblotted for 1 h at 100 V (at RT) onto
Immobilon-P PVDF Transfer membranes [Millipore (Canada) Ltd.,
Nepean, ON], and the membranes were blocked with 5% (w/v) BSA
(ICN Biomedicals, Montreal, PQ) in 1⫻ TBST (10 mM Tris–HCl, pH
8, 150 mM NaCl, 0.05% (v/v) Tween 20). Blocked membranes were
incubated with primary Ab diluted in 5% (w/v) BSA in 1⫻ TBST,
washed with 1⫻ TBST, incubated with 125I-labeled protein A (NEN
Life Science Products, Inc., Boston, MA), washed with 1⫻ TBST, and
exposed to BioMax X-ray film (Kodak, New Haven, CT) at ⫺80°C.
For some experiments, a secondary Ab was required prior to incu-
bation with 125I-labeled protein A. When required, blots were
stripped by incubation for 30 min in a 50°C water bath with 0.7 mM
DTT (Gibco BRL), 62.5 mM Tris–HCl, pH 6.7 (Sigma), 100 mM 2ME
(Sigma), and 2% (v/v) SDS (Bio-Rad). Following stripping, the mem-
branes were washed extensively with 1⫻ TBST and exposed to film
to ensure efficient stripping prior to subsequent reblotting with other
antibodies.
Production of GST-CD4 cyt fusion protein. GST-CD4 cyt fusion pro-
tein was generated by subcloning PCR-amplified human CD4 cyto-
plasmic tail (residues 396 to 433) into pGEX-2T (Pharmacia Biotech).
BL21 bacterial cells transformed by pGEX-2T expression plasmids
were induced with IPTG (Gibco BRL), and the fusion proteins were
purified with glutathione Sepharose 4B beads (Pharmacia Biotech).
To elute GST and GST-CD4 cyt fusion proteins immobilized onto glu-
tathione Sepharose 4B Beads, the complexes were incubated for 10
min at RT with a 0.5⫻ volume of glutathione elution buffer consist-
ing of 10 mM reduced glutathione (Sigma) and 50 mM Tris–HCl, pH
8 (Sigma). The complexes were centrifuged for 5 min at 4°C, 9850g,
and the supernatants containing the eluted fusion proteins were
collected for use in far Western blots.
Far Western blots. Unstimulated and T4-4-stimulated Thp-1 cell
lysates were electrophoresed in duplicate, and the proteins trans-
ferred to membranes as described above. Blocked membranes were
incubated with 1 ␮g/ml of eluted GST or GST-CD4 cyt protein. Bound
GST or GST-CD4 cyt was detected by incubation of the membranes
with anti-GST murine mAb, followed by rabbit anti-mouse Ig Ab,
and finally with 125I-labeled protein A.
GST-CD4 cyt pull-down experiments. For each pull-down reaction,
3.5 to 5 mg of Thp-1 protein lysate was resuspended to a 1-ml volume
with lysis buffer. The lysates were precleared with 30 ␮g of GST
protein (immobilized to glutathione Sepharose 4B Beads) by inver-
sion for 1 h at 4°C. Following preclearing, the lysates were centri-
fuged for 5 min at 4°C, 9850g, and the precleared supernatants were
transferred to fresh Eppendorf tubes for a second round of preclear-
ing with GST protein. The lysates were centrifuged for 5 min at 4°C,
9850g, and the twice-precleared supernatants were incubated by
inversion at 4°C for 2 to 18 h with 30 ␮g of GST or GST-CD4 cyt fusion
protein (immobilized to glutathione Sepharose 4B Beads). The im-
mune complexes were pelleted by centrifugation for 5 min at 4°C,
9850g, and the complexes were washed three times for 5 min at 4°C,
9850g, with 1 ml of lysis buffer. The final pellets were resuspended
with 25 ␮l of Laemmli sample buffer. The complexes were electro-
phoresed as described above and proteins were visualized by staining
the gels with Coomassie brilliant blue.
Calcium flux experiments. A 1 mM Fluo-3/AM (Molecular Probes,
Eugene, OR) solution in 1 mM DMSO (Sigma) and 3.75% (w/v)
Pluronic F-127 (Sigma) was prepared and used to load Thp-1 cells
[31]. Thp-1 cells were washed with Ca 2⫹-free PBS for 5 min at RT,
200g. The cells were resuspended with Buffer A (RPMI 1640 con-
taining 20 mM Hepes, pH 7) and the cells were counted using trypan
143
blue. Viable cells (5 ⫻ 10 6) were washed twice with Buffer A for 10
min at RT, 400g. The cells were resuspended with Buffer A to a final
volume of 5 ml, and Fluo3/AM was added to a final 1 ␮M concentra-
tion. The cells were vortexed gently and then incubated in the dark
for 45 min in a 37°C shaking H 2O bath at 50 rpm. Buffer B (5 ml)
(Buffer A containing 5% (v/v) heat-inactivated FBS, pH 7.4) was
added to the cells and the cells were incubated for 15 min in a 37°C
shaking H 2O bath at 50 rpm. An additional 30 ml of Buffer B was
added to the cells and the cells were washed for 10 min at RT, 400g.
The cells were resuspended with 10 ml of Buffer B to give a final cell
concentration of 0.5 ⫻ 10 6 cells/ml. The cells were aliquoted into 1-ml
volumes and calcium levels were determined using the FACScan
flow cytometer (Becton–Dickinson, Franklin Lakes, NJ) equipped
with CellQuest software, Version 3.2.1fl. Cell samples were main-
tained at 37°C during data acquisition. Baseline intracellular cal-
cium levels were measured for 2 min, the acquisition of data was
stopped momentarily for the addition of various stimuli, and acqui-
sition was resumed for an additional 6 min. Calcium ionophore
A23187 (20 ␮M) (Sigma) and 5 mM EGTA (Sigma) were used as
positive and negative controls, respectively.
RESULTS
Stimulation of Thp-1 Cells through the CD4 Receptor
Results in Calcium Flux
To determine if CD4 expressed on the surface of the
human monocytic cell line Thp-1 was capable of trans-
ducing extracellular signals, Thp-1 cells were stimu-
lated with T4-4 rabbit serum to sCD4 and CD4-medi-
ated changes in intracellular Ca 2⫹ levels were
measured. Stimulation of Thp-1 cells with T4-4 in-
duced a Ca 2⫹ flux (Fig. 1B) compared to the intracellu-
lar baseline Ca 2⫹ levels observed for unstimulated cells
(Fig. 1A). Ca 2⫹ flux occurred within 2 min of stimula-
tion. No Ca 2⫹ flux was observed, however, upon the
addition of T4-4 serum which had been pre-adsorbed
with a sCD4-NHS-activated Sepharose bead complex
(Fig. 1C) or upon the addition of an irrelevant rabbit
serum generated against Shp-1, an intracellular ty-
rosine phosphatase (Fig. 1D).
Stimulation of Thp-1 Cells through the CD4 Receptor
Results in the Time-Dependent Induction of Protein
Tyrosine Phosphorylation
Anti-phosphotyrosine immunoblotting analysis of
unstimulated and T4-4-stimulated Thp-1 lysates re-
vealed a time-dependent tyrosine phosphorylation of
various proteins; the phosphorylation peaked at 1 to 3
min and declined gradually thereafter. A representa-
tive experiment is shown in Fig. 2, top. To ensure that
equal amounts of protein lysate were loaded in each
lane (as determined by the relative amounts of CD4),
the membrane was stripped and blotted with T4-4 se-
rum (Fig. 2, bottom).
The Tyrosine Phosphorylation Induced by T4-4
Serum to sCD4 Is Mediated by CD4
To control for the possibility that the use of T4-4, a
polyclonal rabbit serum to CD4, may have induced
144 GRAZIANI-BOWERING ET AL.
FIG. 1. Stimulation of Thp-1 monocytic cells with T4-4 rabbit
serum to sCD4 induces Ca 2⫹ flux. Thp-1 cells loaded with Fluo3-AM
were stimulated with various stimuli (indicated by arrows) and the
resulting Ca 2⫹ flux was measured by flow cytometric analysis. (A)
Baseline Ca 2⫹ levels in unstimulated cells. (B) Stimulation with T4-4
rabbit serum. (C) Stimulation with T4-4 rabbit serum preadsorbed
with sCD4 (conjugated to NHS-activated Sepharose beads). (D)
Stimulation with rabbit serum to Shp-1. (E) Stimulation with the
Ca 2⫹ ionophore A23187 (first arrow) followed by the addition of the
metal ion chelator EGTA (second arrow) (n ⫽ 3).
some non-CD4-specific signaling, we performed exper-
iments in which a polyclonal rabbit serum to Shp-1, a
protein not expressed on the Thp-1 cell surface, was
used to stimulate the cells. The use of rabbit serum to
Shp-1 (used at the same concentration as T4-4 serum)
induced less tyrosine phosphorylation compared to the
T4-4 stimulation, although the level of tyrosine phos-
phorylation was enhanced compared to that of the un-
stimulated control; Fig. 3A, top (n ⫽ 4).
To further confirm that the induction of tyrosine
phosphorylation observed is actually mediated by T4-4
binding to CD4, we attempted to adsorb the CD4-spe-
cific reactivity of the T4-4 serum. T4-4 serum preincu-
bated with sCD4 prior to use in the stimulation reac-
tion resulted in decreased levels of tyrosine
phosphorylation compared to the levels observed for
the use of T4-4 serum alone (Fig. 3A, top). The de-
creased tyrosine phosphorylation observed for stimula-
tions performed with serum to Shp-1 and with T4-4
serum adsorbed with sCD4 was not due to decreased
amounts of protein lysate loaded in the wells, as indi-
cated by the levels of CD4 detected in this blot (Fig. 3A,
bottom); the prominent, slightly lower molecular mass
band observed in lane 3 is the sCD4 used to adsorb the
T4-4 serum prior to stimulation.
As a third strategy, we attempted to adsorb the
CD4-specific reactivity of the T4-4 serum by preincu-
bation with sCD4 coupled to NHS-activated Sepharose
beads. Following the incubation, the T4-4 serum was
recovered and used for subsequent Thp-1 stimulation.
A negative control in the form of T4-4 serum incubated
with the irrelevant protein BSA (coupled to NHS-acti-
vated Sepharose beads) was also included. The T4-4
serum adsorbed with the sCD4-NHS-activated Sepha-
rose bead complex induced lower levels of tyrosine
phosphorylation compared to the use of T4-4 alone
(Fig. 3B, top); the level of tyrosine phosphorylation for
the sCD4-adsorbed T4-4 stimulation was comparable
to that observed for the unstimulated control. The
stimulation of Thp-1 cells with T4-4 serum incubated
with BSA (coupled to NHS-activated Sepharose beads)
resulted in greater levels of tyrosine phosphorylation
compared to stimulation with T4-4 serum preadsorbed
with the sCD4/bead complex and was comparable to
the levels observed for stimulation with T4-4 alone.
The differences in tyrosine phosphorylation levels ob-
served for the various stimuli were not due to different
amounts of protein lysate loaded per lane since com-
parable levels of CD4 were detected following the strip-
ping and reblotting of the membrane with T4-4 serum
(Fig. 3B, bottom).
Finally, to exclude the possibility that the tyrosine
phosphorylation observed in response to T4-4 stimula-
tion is not due to activation of Fc receptors on Thp-1
cells by the Fc portion of T4-4, we generated a F(ab⬘) 2
fragment of LF T4-4 and used it to stimulate Thp-1
cells. Induction of phosphoproteins by plate-bound
F(ab⬘) 2 was comparable to that induced by plate-bound
T4-4 or LF T4-4, strongly suggesting the lack of in-
volvement of FcR on Thp-1 cells (Fig. 3C).
PLC-␥1 and PI3-K Undergo Tyrosine Phosphorylation
in Response to T4-4 Stimulation
T4-4 stimulation of Thp-1 cells induced the tyrosine
phosphorylation of proteins of specific molecular
masses. Among the higher molecular mass species was
a protein of approximately 140 kDa (Figs. 2 and 4, top).
Thp-1 cells stimulated with T4-4 serum undergo a cal-
cium flux (Fig. 1), a response mediated by phos-
 CD4 SIGNAL TRANSDUCTION IN MONOCYTIC CELLS 145

FIG. 2. Stimulation of Thp-1 monocytic cells with T4-4 rabbit serum to sCD4 results in the time-dependent tyrosine phosphorylation of
various proteins. (Top) Serum-starved Thp-1 cells (10 ⫻ 10 6) were stimulated for various times at 37°C with 25 ␮g/ml of T4-4 rabbit serum
to sCD4. Total protein lysate (100 ␮g) was electrophoresed and transferred. The blocked membranes were incubated with 1 ␮g/ml of
anti-phosphotyrosine mAb, followed by a 1:2000 dilution of rabbit anti-mouse Ab, and detected using 125I-labeled protein A. Induced
tyrosine-phosphorylated proteins (pp) having molecular masses of 180, 140, 120, 110, 85, 65, 55, 50, and 35 kDa are indicated by asterisks.
(Bottom) The blot was stripped, incubated with 10 ␮g/ml of T4-4 serum, and detected using 125I-labeled protein A; the position of CD4 is
indicated by an arrow. Mobilities of the molecular mass (MW) standards are shown on the left (n ⫽ 3). Autoradiographs were scanned using
the HP ScanJet4C scanner and HP DeskScanII software, and figures were generated using Microsoft PowerPoint97 software.

phoinositol-specific phospholipase C [32]. The molecu-
lar mass of PLC-␥1 is approximately 148 kDa [33],
which corresponds to the approximately 140-kDa ty-
rosine-phosphorylated protein observed following T4-4
stimulation. Furthermore, it has been reported that
PI3-K forms a complex with CD4 and p56 lck in human
T cells [34]. We speculated, therefore, that the approx-
imately 85- to 90-kDa Thp-1 protein observed to un-
dergo tyrosine phosphorylation in response to T4-4
stimulation (Figs. 2 and 4, top) might be the regulatory
subunit of PI3-K (p85). To determine if PLC-␥1 and
PI3-K were two of the proteins that become tyrosine
phosphorylated in response to T4-4 stimulation, anti-
phosphotyrosine immunoprecipitations of unstimu-
lated and T4-4-stimulated Thp-1 cells were performed.
Following electrophoresis and transfer of the immune
complexes, the membranes were blotted with rabbit
antibody to PLC-␥1. The level of PLC-␥1 immunopre-
cipitated and detected upon T4-4 stimulation was in-
creased (by 3.8-fold, as determined by densitometry)
compared to the unstimulated condition (Fig. 4, mid-
dle). Given that PLC-␥1 (p148) and the regulatory sub-
unit (p85) of PI3-K have significantly different molec-
ular masses, and given the absence of any background
in the 85-kDa range following blotting with antibody to
PLC-␥1, the membrane was reblotted, without strip-
ping, for PI3-K (p85). The level of immunoprecipitated
PI3-K was consistently increased upon T4-4 stimula-
tion, compared to the unstimulated condition (Fig. 4,
middle). The membranes were subsequently stripped
and blotted with T4-4 to control for the relative
amounts of proteins present in the various lysate prep-
arations (Fig. 4, bottom); slightly decreased levels of
CD4 were present in the T4-4-stimulated lysate com-
pared to the unstimulated lysate, suggesting that the
protein concentration of this lysate may have been
slightly lower. These results suggest that the anti-
phosphotyrosine immunoprecipitation performed with
the T4-4-stimulated lysate contained relatively lower
amounts of protein, and yet, the highest levels of phos-
phorylated PLC-␥1 and PI3-K were nonetheless ob-
served for this condition. To confirm that the enhanced
detection of PLC-␥1 and PI3-K was in fact due to success-
ful immunoprecipitation of tyrosine-phosphorylated pro-
teins, the membrane was reprobed with anti-phosphoty-
rosine mAb (Fig. 4, top). In the anti-phosphotyrosine blot,
the highest amount of immunoprecipitated tyrosine-
phosphorylated protein was observed for the immunopre-
cipitation reaction performed with the T4-4-stimulated
lysate, as expected.
146 GRAZIANI-BOWERING ET AL.

FIG. 3. The tyrosine phosphorylation induced by stimulation of Thp-1 monocytic cells with T4-4 rabbit serum to sCD4 is mediated by
CD4. (A) Serum-starved Thp-1 cells (10 ⫻ 10 6) were stimulated for 3 min at 37°C with various stimulants; 100 ␮g of total protein lysate was
electrophoresed and analyzed by immunoblotting. (Top) The blocked membrane was incubated with 1 ␮g/ml of anti-phosphotyrosine mAb,
followed by a 1:2000 dilution of rabbit anti-mouse Ab, and detected using 125I-labeled protein A. (Bottom) The blot was stripped and incubated
with 10 ␮g/ml of T4-4 serum and detected using 125I-labeled protein A; the position of CD4 is indicated by an arrow (n ⫽ 3). (B) Serum-starved
Thp-1 cells (10 ⫻ 10 6) were stimulated for 3 min at 37°C with various stimulants; 100 ␮g of total protein lysate was electrophoresed and
analyzed by immunoblotting. (Top) The blocked membrane was incubated with 1 ␮g/ml of anti-phosphotyrosine mAb followed by a 1:2000
dilution of rabbit anti-mouse Ab, and detected using 125I-labeled protein A. (Bottom) The blot was stripped and incubated with 10 ␮g/ml of
T4-4 serum and detected using 125I-labeled protein A; the position of CD4 is indicated by an arrow (n ⫽ 2). (C) Serum-starved Thp-1 cells
(5 ⫻ 10 6) were stimulated for 3 min at 37°C with various stimulants; 100 ␮g of total protein lysate was electrophoresed and analyzed by
immunoblotting. (Top) The blocked membrane was incubated with 1 ␮g/ml of anti-phosphotyrosine mAb, followed by a 1:2000 dilution of
rabbit anti-mouse Ab, and detected using 125I-labeled protein A. (Bottom) The blot was stripped and incubated with a 1:1000 dilution of goat
anti-actin Ab, following by a 1:500 dilution of rabbit anti-goat IgG and detected using 125I-labeled protein A (n ⫽ 3). Mobilities of the
molecular mass (MW) standards are shown on the left. Autoradiographs were scanned using the HP ScanJet4C scanner and HP DeskScanII
software, and figures were generated using Microsoft PowerPoint97 software.

CD4 Does Not Directly Interact with Either Tyrosine-
Phosphorylated PLC-␥1 or PI3-K
T4-4 stimulation of Thp-1 cells was shown to induce
the tyrosine phosphorylation of PLC-␥1 and PI3-K. To
determine if either PI3-K or PLC-␥1 was directly asso-
ciated with CD4, unstimulated and T4-4-stimulated
Thp-1 lysates were used to immunoprecipitate CD4
using T4-4 serum. The electrophoresed immune com-
plexes were subsequently blotted for the presence of
PI3-K, PLC-␥1, and a number of other known signaling
molecules. Despite the successful detection of PLC-␥1
in the lysate controls, and despite the prolonged expo-
sure of the blots (up to 1 week), no PLC-␥1 was de-
tected in the CD4 immunoprecipitation reactions (data
not shown). A reprobing of the membrane with T4-4
serum confirmed that CD4 had been successfully im-
munoprecipitated, and anti-phosphotyrosine Western
blots of unstimulated and T4-4-stimulated Thp-1 ly-
sates were also performed to confirm that tyrosine
phosphorylation had been induced. Attempts to detect
PI3-K in CD4 immunoprecipitation reactions were also
unsuccessful (data not shown). In addition, we per-
formed similar experiments to determine if CD4 di-
rectly associated with a number of other known signal-
ing molecules. These molecules, partially chosen based
on the molecular masses of Thp-1 proteins shown to
undergo tyrosine phosphorylation in response to T4-4
stimulation, included kinases (HCK, PKC), G proteins
(G ␣, G ␤, and Rap-1), adaptor molecules (NCK, SLP-76),
and VAV, an exclusively hematopoietic GDP/GTP ex-
change factor. None of these molecules could be de-
tected in the CD4 immunoprecipitation reactions (data
not shown).
GST-CD4 cyt Interacts with 35-, 45-, and 55-kDa
Proteins, Irrespective of T4-4 Stimulation
The failure to observe direct association between
CD4 and any of the candidate signaling molecules for
which we screened necessitated a different strategy to
determine which proteins interact with the cytoplas-
mic tail of CD4. To this end, we produced a GST-CD4 cyt
fusion protein (Fig. 5A) for use in Far Western blot
analysis and in pull-down experiments.
Far Western blot analysis of unstimulated and T4-
 CD4 SIGNAL TRANSDUCTION IN MONOCYTIC CELLS
FIG. 4. T4-4 stimulation induces tyrosine phosphorylation of
PLC-␥1 and PI3-K. Tyrosine-phosphorylated proteins were immuno-
precipitated from 2 mg of unstimulated and T4-4-stimulated Thp-1
total protein lysates using 10 ␮g of anti-phosphotyrosine mAb. The
immunoprecipitation reactions, as well as 100-␮g lysate controls,
were electrophoresed and transferred. The same membrane was
subjected to four rounds of immunoblotting. (Top) The blocked mem-
brane was incubated with 1 ␮g/ml of anti-phosphotyrosine mAb
followed by a 1:2000 dilution of rabbit anti-mouse Ab and detected
using 125I-labeled protein A. (Middle) The membrane was incubated
with 1 ␮g/ml of rabbit polyclonal Ab to PLC-␥1 and detected using
125
I-labeled protein A, followed by incubation with a 1:1000 dilution
of murine mAb to PI3-K (p85 subunit) and with a 1:2000 dilution of
rabbit anti-mouse Ab and detected using 125I-labeled protein A. (Bot-
tom) The blot was stripped, blocked, and incubated with 10 ␮g/ml of
T4-4 rabbit serum and detected using 125I-labeled protein A. The
positions of PLC-␥1, PI3-K, and CD4 are indicated by arrows. Mo-
bilities of the molecular mass standards are shown on the left (n ⫽
4). Autoradiographs were scanned using the HP ScanJet4C scanner
and HP DeskScanII software, and figures were generated using
Microsoft PowerPoint97 software.
4-stimulated Thp-1 lysates revealed that the GST-
CD4 cyt fusion protein interacted with phosphotyrosine
proteins having molecular masses of approximately
140, 110, 90, 85, 55, and 45 kDa; these molecular
masses are similar to the molecular masses of proteins
observed to undergo increased tyrosine phosphoryla-
147
tion in response to T4-4 stimulation (Fig. 2). In con-
trast, the GST control protein bound weakly, if at all, to
the proteins bound by GST-CD4 cyt. Furthermore, the
banding patterns obtained for the GST-CD4 cyt-blotted
membranes were the same for both unstimulated and
T4-4-stimulated lysates, suggesting that the CD4 por-
tion of the GST-CD4 cyt fusion protein bound to these
proteins constitutively. A representative experiment is
shown in Fig. 5B.
The GST-CD4 cyt fusion protein construct was also
used in a series of pull-down experiments. Proteins of
45 and 55 kDa were routinely observed in the GST-
CD4 cyt pull-down reaction; these proteins were not ob-
served in the GST pull-down reaction (Fig. 6). In some
experiments (n ⫽ 3), an enhanced 35-kDa protein was
FIG. 5. GST-CD4 cyt fusion protein binds to numerous Thp-1 pro-
teins as determined by far Western blot analysis. (A) Schematic of
GST-CD4 cyt fusion protein: GST-CD4 cyt fusion protein was generated
by subcloning PCR-amplified human CD4 cytoplasmic tail (residues
396 to 433) into the pGEX-2T expression plasmid. BL21 bacterial
cells transformed by pGEX-2T expression plasmids were induced
with IPTG. The fusion proteins were purified with glutathione
Sepharose 4B beads and used in pull-down experiments. Alterna-
tively, the fusion proteins were eluted from the beads for use in far
Western blots. (B) Far Western blot analysis: 100 ␮g of unstimulated
and T4-4-stimulated Thp-1 protein lysates was electrophoresed in
duplicate and analyzed by immunoblotting. The blocked membranes
were incubated with 1 ␮g/ml of eluted GST or GST-CD4 cyt protein
and detected with a 1:200 dilution of murine anti-GST mAb, followed
by a 1:2000 dilution of rabbit anti-mouse Ab, and detected using
125
I-labeled protein A. GST-CD4 cyt-binding Thp-1 proteins having
molecular masses of 140, 100, 90, and 85 kDa are indicated asterisks,
and 55- and 45-kDa proteins are indicated by arrows. Mobilities of
the molecular mass standards are shown on the left (n ⫽ 3). Auto-
radiographs were scanned using the HP ScanJet4C scanner and HP
DeskScanII software, and figures were generated using Microsoft
PowerPoint97 software.
148 GRAZIANI-BOWERING ET AL.
FIG. 6. GST-CD4 cyt fusion protein immunoprecipitates three pro-
teins from Thp-1 total protein lysate, irrespective of CD4 stimula-
tion. Unstimulated and T4-4-stimulated lysates (3.5 mg) were pre-
cleared twice with 30 ␮g of GST control protein. The precleared
lysates were subjected to an overnight immunoprecipitation with 30
␮g of either GST or GST-CD4 cyt protein. The Thp-1 proteins directly
associated with GST or GST-CD4 cyt proteins were visualized by elec-
trophoresing the immune complexes and staining the gel with Coo-
massie blue. Electrophoresed GST and GST-CD4 cyt proteins (in the
absence of Thp-1 lysates) were included as controls. GST-CD4 cyt-
binding Thp-1 proteins having molecular masses of 35, 45, and 55
kDa are indicated by arrows. Mobilities of the molecular mass stan-
dards are shown on the left (n ⫽ 9). Autoradiographs were scanned
using the HP ScanJet4C scanner and HP DeskScanII software, and
figures were generated using Microsoft PowerPoint97 software.
also observed for the GST-CD4 cyt pull-down reactions
which was only weakly visible in the GST pull-down
reaction (Fig. 6). The association between GST-CD4 cyt
and the 35-, 45-, and 55-kDa proteins occurred irre-
spective of whether the lysates were from unstimu-
lated or T4-4-stimulated cells; the ability of GST-CD4 cyt
to associate with various Thp-1 proteins irrespective of
stimulation was consistent with the observations made
in the far Western blot experiments. Equal amounts of
GST and GST-CD4 cyt proteins were used for the pull-
down reactions, as indicated by the similar intensities
of these bands (30 and 33 kDa, respectively) (Fig. 6).
DISCUSSION
Whereas the role of CD4-mediated signal transduc-
tion has been well established for CD4 ⫹ T cells, less is
known about the role of this molecule in monocytes. It
has been shown that in primary monocytes, monocytic
cell lines, and macrophages, CD4 has no associated
p56 lck nor any other kinase activity [18 –20]. Using
Western blot analysis, we confirmed the lack of p56 lck
expression in U937, HL-60, and Thp-1 monocytic cell
lines (data not shown). Species heterogeneity with re-
spect to CD4 expression by macrophages has been ob-
served; while human and rat macrophages express
CD4, murine macrophages do not, despite the expres-
sion of CD4 by a subset of murine T cells [35]. The lack
of CD4 expression by murine macrophages has been
attributed to the production of a truncated CD4 tran-
script [36]. To determine if CD4 expressed in human
monocytes differed from human T cell CD4, we se-
quenced RT–PCR products generated using primers
spanning the transmembrane and cytoplasmic do-
mains of CD4. No differences were found between the
CD4 transcripts from human CD4 ⫹ T cells, primary
monocytes, or the U937 cell line (data not shown).
Despite the failure to detect any direct interaction
between CD4 and kinases [18 –20], there is evidence
that CD4 is not an inert remnant originating from the
pluripotent hematopoietic stem cell precursor shared
by both lymphoid and myeloid cells. Rather, CD4 has
been shown to be an active signaling molecule capable
of responding to various stimuli. For example, a com-
plex inositol polyphosphate response and a Ca 2⫹ flux
were induced in monocytes in which CD4 and Fc␥
receptors were co-cross-linked [21]. IL-16 stimulation
of various CD4 ⫹ monocytic cell lines induces a trans-
location of PKC from the cytosol to the membrane [22],
as well as activation of the SAPK and p38 MAPK
signaling pathways [23]. Finally, HIV gp120 can in-
duce a CD4-dependent down-regulation of the CCR5
chemokine receptor in monocytes [24].
The observations made in the studies cited, however,
must be interpreted cautiously with respect to the ex-
act role played by CD4. In two of the studies, CD4-
mediated signal transduction involved a coreceptor
consisting of either Fc␥ [21] or CCR5 [24] receptors. In
the case of human cell studies performed with IL-16, it
has been shown that while CD4 is required for IL-16 to
mediate its effects [10], it remains to be determined
whether IL-16 also requires interaction with other cell
surface molecules, such as chemokine receptors [11,
12]. Although various studies have demonstrated IL-
16-induced down-regulation of CCR5 and CXCR4 on
monocyte-derived macrophages [37], as well as desen-
sitization of CCR5 to MIP-1␣ [38] and of CXCR4 to
SDF-1␣ [12] in T cells, no CD4/chemokine receptor/
IL-16 complex has been demonstrated to date [12].
Since monocyte CD4-mediated signaling may be mod-
ulated by the participation of either Fc␥ receptors, or
possibly chemokine receptors, it is difficult to dissect
out the exact role of CD4 in transducing signals in each
of these studies.
We have established that monocyte CD4 is capable of
transducing signals independently of other surface re-
ceptors. Thp-1 monocytic cells loaded with Fluo-3/AM
and stimulated with T4-4 serum underwent a dramatic
calcium flux; this response was not observed when T4-4
was adsorbed with sCD4 prior to stimulation, nor did
 CD4 SIGNAL TRANSDUCTION IN MONOCYTIC CELLS
calcium flux occur when the cells were stimulated with
an equal amount of an irrelevant rabbit serum to
Shp-1. These results establish that the response ob-
served upon T4-4 stimulation is mediated by CD4 and
not by Fc receptors. Signaling through CD4 requires
effective receptor multimerization [39]. Thus, mAb
alone may not be able to induce the degree of CD4
multimerization induced by mAb cross-linked with a
secondary Ab or by polyclonal Ab. This may explain the
failure of various soluble ␣-CD4 mAb and soluble
F(ab⬘) 2 fragments of LF T4-4 to induce tyrosine phos-
phorylation and/or calcium flux compared to the more
effective induction of phosphotyrosine by these Ab
preparations when these were immobilized onto plastic
plates.
This scenario would also parallel the observation
that activation of T cells through CD3 or the TCR is
strongest when antibodies to these receptors are cou-
pled to solid matrices to achieve infinite cross-linking,
as opposed to the very weak to nonexistent activation
observed when intact antibodies or Fab fragments are
used [40]. Similarly, in the case of T cell CD4, the
coreceptor function of CD4 requires the formation of
highly cross-linked lattices between CD4 oligomers
and TCR oligomers on the surface of the T cell, with
oligomers of MHC class II molecules complexed with
antigen on the surface of the antigen-presenting cell
[40]. Thus it is possible that the degree of CD4 mul-
timerization required to induce signals cannot be
achieved by soluble ligands such as mAb.
In addition to a calcium flux, we also observed a
time-dependent tyrosine phosphorylation of many pro-
teins in response to CD4-mediated stimulation. The
CD4 specificity of our phosphotyrosine response was
again established by using T4-4 adsorbed with sCD4,
which resulted in a decreased level of tyrosine phos-
phorylation; this dramatic decrease did not occur if
T4-4 was preincubated with BSA. The use of the irrel-
evant rabbit serum to Shp-1 also resulted in a de-
creased level of tyrosine phosphorylation. Further-
more, the observation that F(ab⬘) 2 fragments of LF
T4-4 could also induce tyrosine phosphorylation sug-
gests that FcR are not involved in T4-4-induced signal-
ing. Finally, immune complexes have a higher affinity
for Fc receptors than Ab alone [41, 42]. That immune
complexes consisting of sCD4/T4-4 did not induce ty-
rosine phosphorylation also argues against the involve-
ment of FcR in the induction of phosphotyrosine in
response to T4-4.
To elucidate the mechanism by which tyrosine phos-
phorylation was induced by T4-4 stimulation, we at-
tempted to identify the proteins which underwent ty-
rosine phosphorylation. It has been reported that
PI3-K forms a complex with CD4 and p56 lck in human
T cells [34]. We hypothesized that PI3-K might form a
similar complex with monocyte CD4 and some other, as
149
yet unidentified, protein(s) and that PI3-K might,
therefore, be one of the phosphoproteins activated in
monocytes following CD4-mediated stimulation. The
tyrosine phosphorylation of PI3-K was confirmed based
on our ability to detect it among the immunoprecipi-
tated proteins that were tyrosine phosphorylated in
response to T4-4 stimulation.
Based on our observations that T4-4 stimulation re-
sulted in a CD4-specific calcium flux, we proposed that
another of the tyrosine-phosphorylated proteins ob-
served in the phosphotyrosine blots might be PLC-␥1.
The detection of tyrosine-phosphorylated PLC-␥1 was
not as strong as the detection of tyrosine-phosphory-
lated PI3-K. This may reflect technical factors (such as
the larger size of PLC-␥1), which may have influenced
the efficiency of the antiphosphotyrosine immunopre-
cipitation reaction.
Despite our findings that PLC-␥1 and PI3-K were
tyrosine phosphorylated in response to CD4 stimula-
tion, we were unable to show any direct interaction
between either of these signaling molecules with im-
munoprecipitated CD4. These findings were not unex-
pected, however, since the SH2 domains of these and
other signaling proteins recognize tyrosine-phosphory-
lated residues [43], and the cytoplasmic tail of CD4
does not contain any tyrosine residues [44]. Attempts
to immunoprecipitate CD4 and detect numerous other
known signaling molecules were also unsuccessful.
As an alternative strategy for identifying proteins
linking CD4 to the PLC-␥1 and PI3-K signaling path-
ways, we generated a GST-CD4 cyt fusion protein for use
in far Western blots and pull-down experiments. In
both types of experiments, 45- and 55-kDa proteins
were routinely found to associate with the GST-CD4 cyt
fusion protein, but not with the GST control protein.
This association occurred regardless of whether the
cells had been stimulated with T4-4. Such a constitu-
tive association of CD4 with these proteins is reminis-
cent of the constitutive association of CD4 and p56 lck in
T cells [14, 15]. Furthermore, in some of the pull-down
experiments, a 35-kDa protein was also observed to be
constitutively associated with the GST-CD4 cyt protein.
The p45 and p55 proteins routinely isolated using the
GST-CD4 cyt affinity matrix were subjected to mass
spectrometry using the nanoLC-MS-MS system that
combines micromass modular capLC and Q-TOF-2
quadrupole/time-of-flight. The list of tryptic peptide
masses was used to obtain potential protein candidates
in a database search using a peptide mass fingerprint
algorithm (PeptideSearch, EMBL). The search was
conducted against a nonredundant protein sequence
database containing over 257,000 entries downloaded
from the European Bioinformatic Institute web site
(ftp://ftp.ebi.ac.uk/pub/databases/peptidesearch/). How-
ever, no matches were made between the tryptic peptides
generated from the p45 and p55 proteins and any of the
150 GRAZIANI-BOWERING ET AL.
FIG. 7. Proposed model of CD4 signaling in Thp-1 monocytic
cells. Based on our observations, we propose that in Thp-1 monocytic
cells, the cytoplasmic tail of CD4 is constitutively associated with
35-, 45-, and/or 55-kDa proteins. Upon stimulation with T4-4 serum,
a signal is transduced by CD4, resulting in the tyrosine phosphory-
lation, and therefore activation, of PLC-␥1 and PI3-K; we propose
that the p35, p45, and/or p55 molecules may function as linkers/
adaptors between CD4 and these signaling molecules. The activated
PLC-␥1 will in turn cleave PIP 2, thus activating PKC and inducing
Ca 2⫹ flux.
proteins in the database (data not shown), suggesting
that the p45 and p55 proteins are novel.
Based on our observations, we propose that in Thp-1
monocytic cells, the cytoplasmic tail of CD4 is consti-
tutively associated with 35-, 45-, and/or 55-kDa pro-
teins. Upon stimulation with T4-4 serum, a signal is
transduced by CD4, resulting in the tyrosine phosphor-
ylation, and therefore activation, of PLC-␥1 and PI3-K;
we propose that the p35, p45, and/or p55 molecules
may function as linkers/adaptors between CD4 and
these signaling molecules. The activated PLC-␥1 will
in turn cleave PIP 2, thus activating PKC and inducing
Ca 2⫹ flux. Our proposed signaling model is depicted in
Fig. 7.
Our study has provided insight into the initial events
that occur upon engagement of the CD4 receptor in
monocytic cells. The ultimate outcome of this engage-
ment has yet to be determined. Many different signal-
ing pathways have been characterized. Central among
these are the serine/threonine kinases that participate
in the mitogen-activated protein kinase (MAPK) path-
ways. The MAPK pathways can be divided into the
ERK (extracellular signal-regulated kinase) pathway,
the SAPK (stress-activated protein kinase) pathway,
and the p38 MAPK pathway [45]. The ERK pathway
mediates signals in response to mitogens and growth
factors, whereas the SAPK and p38 MAPK pathways
are activated by stress factors, such as inflammatory
cytokines (IL-1␤, TNF-␣), ionizing radiation, osmotic
shock, heat shock, oxidative stress, and chemical and
physiological stresses. It has been shown that IL-16
stimulation of CD4 ⫹ macrophages does not activate the
ERK pathway, but does activate p46 and p54 of the
SAPK pathway [23]. Similarly, in our own work, we
were unable to detect any activation of ERK-1 and
ERK-2 (data not shown), and we found proteins of 45
and 55 kDa to be constitutively associated with our
GST-CD4 cyt fusion protein. The identities of these 45-
and 55-kDa proteins are unknown, but we do not ex-
pect them to correspond to the p46 and p54 proteins of
the SAPK pathway. These proteins are found down-
stream in the SAPK pathway, whereas we expect our
p45 and p55 molecules to be upstream in their respec-
tive pathway(s), given their ability to interact with the
cytoplasmic tail of CD4, a membrane glycoprotein. Fur-
thermore, p46 and p54 of the SAPK pathway are ki-
nases, and no CD4-associated kinase activity has been
detected in monocytes [18 –20]. Further studies will be
required in order to identify the p35, p45, and p55
proteins which were found to associate with our GST-
CD4 cyt fusion protein, as well as to determine if T4-4
stimulation of Thp-1 cells results in the activation of
the SAPK pathway.
We thank Drs. William Casley, Ashok Kumar, and John Webb for
critically reviewing the manuscript, Alexander Hayes for technical
assistance, and Ms. Monique Parenteau for her expert operation of
the FACScan flow cytometer. This work was performed at Health
Canada (Biologics and Genetics Therapies Directorate) and at the
University of Ottawa, Department of Biochemistry, Microbiology,
and Immunology. G.M.G.-B. was supported by a National Health
Ph.D. Fellowship (AIDS) from Health Canada. L.G.F. was supported
by an AIDS grant from Health Canada and the Ontario Ministry of
Health. M.K. was supported by a grant from the National Biotech-
nology Strategy Fund.
REFERENCES
1. Reinherz, E. L., Kung, P. C., Goldstein, G., and Schlossman,
S. F. (1979). Separation of functional subsets of human T cells
by a monoclonal antibody. Proc. Natl. Acad. Sci. USA 76, 4061–
4065.
2. Wood, G. S., Warner, N. L., and Warnke, R. A. (1983). Anti-Leu-
3/T4 antibodies react with cells of monocyte/macrophage and
Langerhans lineage. J. Immunol. 131, 212–216.
3. Lucey, D. R., Dorsky, D. I., Nicholson-Weller, A., and Weller, P. F.
(1989). Human eosinophils express CD4 protein and bind human
immunodeficiency virus 1 gp120. J. Exp. Med. 169, 327–332.
4. Klatzmann, D., Champagne, E., Chamaret, S., Gruest, J.,
Guetard, D., Hercend, T., Gluckman, J. C., and Montagnier, L.
(1984). T-lymphocyte T4 molecule behaves as the receptor for
human retrovirus LAV. Nature 312, 767–768.
5. Dalgleish, A. G., Beverley, P. C., Clapham, P. R., Crawford,
D. H., Greaves, M. F., and Weiss, R. A. (1984). The CD4 (T4)
antigen is an essential component of the receptor for the AIDS
retrovirus. Nature 312, 763–767.
 CD4 SIGNAL TRANSDUCTION IN MONOCYTIC CELLS
6. McDougal, J. S., Mawle, A., Cort, S. P., Nicholson, J. K., Cross,
G. D., Scheppler-Campbell, J. A., Hicks, D., and Sligh, J. (1985).
Cellular tropism of the human retrovirus HTLV-III/LAV. I.
Role of T cell activation and expression of the T4 antigen. J.
Immunol. 135, 3151–3162.
7. Feng, Y., Broder, C. C., Kennedy, P. E., and Berger, E. A.
(1996). HIV-1 entry cofactor: Functional cDNA cloning of a
seven-transmembrane, G protein-coupled receptor [see com-
ments]. Science 272, 872– 877.
8. Deng, H., Liu, R., Ellmeier, W., Choe, S., Unutmaz, D.,
Burkhart, M., Di Marzio, P., Marmon, S., Sutton, R. E., Hill,
C. M., Davis, C. B., Peiper, S. C., Schall, T. J., Littman, D. R.,
and Landau, N. R. (1996). Identification of a major co-receptor
for primary isolates of HIV-1 [see comments]. Nature 381, 661–
666.
9. Dragic, T., Litwin, V., Allaway, G. P., Martin, S. R., Huang, Y.,
Nagashima, K. A., Cayanan, C., Maddon, P. J., Koup, R. A.,
Moore, J. P., and Paxton, W. A. (1996). HIV-1 entry into CD4⫹
cells is mediated by the chemokine receptor CC-CKR-5 [see
comments]. Nature 381, 667– 673.
10. Cruikshank, W. W., Center, D. M., Nisar, N., Wu, M., Natke, B.,
Theodore, A. C., and Kornfeld, H. (1994). Molecular and func-
tional analysis of a lymphocyte chemoattractant factor: associ-
ation of biologic function with CD4 expression. Proc. Natl. Acad.
Sci. USA 91, 5109 –5113.
11. Cruikshank, W. W., Kornfeld, H., and Center, D. M. (2000).
Interleukin-16 [Review]. J. Leukocyte Biol. 67, 757–766.
12. Van Drenth, C., Jenkins, A., Ledwich, L., Ryan, T. C., Mash-
ikian, M. V., Brazer, W., Center, D. M., and Cruikshank, W. W.
(2000). Desensitization of CXC chemokine receptor 4, mediated
by IL-16/CD4, is independent of p56lck enzymatic activity. J.
Immunol. 165, 6356 – 6363.
13. Ravichandran, K. S., Collins, T. L., and Burakoff, S. J. (1996).
CD4 and signal transduction [Review]. Curr. Top. Microbiol.
Immunol. 205, 47– 62.
14. Rudd, C. E., Trevillyan, J. M., Dasgupta, J. D., Wong, L. L., and
Schlossman, S. F. (1988). The CD4 receptor is complexed in
detergent lysates to a protein-tyrosine kinase (pp58) from hu-
man T lymphocytes. Proc. Natl. Acad. Sci. USA 85, 5190 –5194.
15. Veillette, A., Bookman, M. A., Horak, E. M., and Bolen, J. B.
(1988). The CD4 and CD8 T cell surface antigens are associated
with the internal membrane tyrosine-protein kinase p56lck.
Cell 55, 301–308.
16. van Leeuwen, J. E. and Samelson, L. E. (1999). T cell antigen-
receptor signal transduction [Review]. Curr. Opin. Immunol.
11, 242–248.
17. Corbeil, J., Tremblay, M., and Richman, D. D. (1996). HIV-
induced apoptosis requires the CD4 receptor cytoplasmic tail
and is accelerated by interaction of CD4 with p56lck. J. Exp.
Med. 183, 39 – 48.
18. Pelchen-Matthews, A., Armes, J. E., Griffiths, G., and Marsh,
M. (1991). Differential endocytosis of CD4 in lymphocytic and
nonlymphocytic cells. J. Exp. Med. 173, 575–587.
19. Vinante, F., Pizzolo, G., Rigo, A., Vincenzi, C., Sorio, C., Cas-
satella, M., and Berton, G. (1992). The CD4 molecule belongs to
the phenotypic repertoire of most cases of acute myeloid leuke-
mia. Leukemia 6, 1257–1262.
20. Sorio, C., Saggioro, D., Chieco-Bianchi, L., and Berton, G.
(1993). The monosialoganglioside GM1 induces internalisation
and degradation of the CD4 antigen in U937 cells: Evidence for
a novel mechanism of CD4 down-modulation in a p56lck-nega-
tive cell line, which is independent of protein kinase C activa-
tion. Biochem. Biophys. Res. Commun. 191, 1105–1110.
151
21. Guse, A. H., Roth, E., Broker, B. M., and Emmrich, F. (1992).
Complex inositol polyphosphate response induced by co-cross-
linking of CD4 and Fc gamma receptors in the human monocy-
toid cell line U937. J. Immunol. 149, 2452–2458.
22. Parada, N. A., Cruikshank, W. W., Danis, H. L., Ryan, T. C.,
and Center, D. M. (1996). IL-16- and other CD4 ligand-induced
migration is dependent upon protein kinase C. Cell. Immunol.
168, 100 –106.
23. Krautwald, S. (1998). IL-16 activates the SAPK signaling path-
way in CD4⫹ macrophages. J. Immunol. 160, 5874 –5879.
24. Wang, J. M., Ueda, H., Howard, O. M., Grimm, M. C., Chertov,
O., Gong, X., Gong, W., Resau, J. H., Broder, C. C., Evans, G.,
Arthur, L. O., Ruscetti, F. W., and Oppenheim, J. J. (1998).
HIV-1 envelope gp120 inhibits the monocyte response to che-
mokines through CD4 signal-dependent chemokine receptor
down-regulation. J. Immunol. 161, 4309 – 4317.
25. Crowe, S. M. (1995). Role of macrophages in the pathogenesis of
human immunodeficiency virus (HIV) infection [Review]. Aust.
N. Z. J. Med. 25, 777–783.
26. Filion, L. G. (1993). HIV immunocytopathology in the develop-
ment of opportunistic infections: Cellular dysfunctions in mac-
rophages. In “Master Lectures In Opportunistic Infections.”
Intramed, Milan, Italy.
27. Kon-Kozlowski, M., Pani, G., Pawson, T., and Siminovitch,
K. A. (1996). The tyrosine phosphatase PTP1C associates with
Vav, Grb2, and mSos1 in hematopoietic cells. J. Biol. Chem.
271, 3856 –3862.
28. Deen, K. C., McDougal, J. S., Inacker, R., Folena-Wasserman,
G., Arthos, J., Rosenberg, J., Maddon, P. J., Axel, R., and Sweet,
R. W. (1988). A soluble form of CD4 (T4) protein inhibits AIDS
virus infection. Nature 331, 82– 84.
29. Coligan, J. E., Kruisbeek, A. M., Marguilies, D. H., Shevach,
E. M., and Strober, W. (2001). Induction of immune responses.
In “Current Protocols In Immunology.” Wiley, New York.
30. Tsuchiya, S., Yamabe, M., Yamaguchi, Y., Kobayashi, Y.,
Konno, T., and Tada, K. (1980). Establishment and character-
ization of a human acute monocytic leukemia cell line (THP-1).
Int. J. Cancer 26, 171–176.
31. Brousseau P, Payette Y, Tryphonas H, Blakely B, Boermans H,
Flipo D, and Fournier M (1998). “Intracellular Levels of Cal-
cium Assay Using Flow Cytometry.” CRC Press, Boca Raton
FL.
32. Katan, M. (1998). Families of phosphoinositide-specific phos-
pholipase C: structure and function [Review]. Biochim. Bio-
phys. Acta 1436, 5–17.
33. Burgess, W. H., Dionne, C. A., Kaplow, J., Mudd, R., Friesel, R.,
Zilberstein, A., Schlessinger, J., and Jaye, M. (1990). Charac-
terization and cDNA cloning of phospholipase C-gamma, a ma-
jor substrate for heparin-binding growth factor 1 (acidic fibro-
blast growth factor)-activated tyrosine kinase. Mol. Cell. Biol.
10, 4770 – 4777.
34. Prasad, K. V., Kapeller, R., Janssen, O., Repke, H., Duke-
Cohan, J. S., Cantley, L. C., and Rudd, C. E. (1993). Phospha-
tidylinositol (PI) 3-kinase and PI 4-kinase binding to the CD4-
p56lck complex: The p56lck SH3 domain binds to PI 3-kinase
but not PI 4-kinase. Mol. Cell. Biol. 13, 7708 –7717.
35. Crocker, P. R., Jefferies, W. A., Clark, S. J., Chung, L. P., and
Gordon, S. (1987). Species heterogeneity in macrophage expres-
sion of the CD4 antigen. J. Exp. Med. 166, 613– 618.
36. Moore, S. C., Anderson, S. J., and Walker, W. S. (1992). Mouse
macrophages contain a truncated CD4 transcript. J. Leukocyte
Biol. 52, 128 –129.
37. Hermann, E., Darcissac, E., Idziorek, T., Capron, A., and Bahr,
G. M. (1999). Recombinant interleukin-16 selectively modulates
152 GRAZIANI-BOWERING ET AL.
surface receptor expression and cytokine release in macro-
phages and dendritic cells. Immunology 97, 241–248.
38. Mashikian, M. V., Ryan, T. C., Seman, A., Brazer, W., Center,
D. M., and Cruikshank, . W. (1999). Reciprocal desensitization
of CCR5 and CD4 is mediated by IL-16 and macrophage-inflam-
matory protein-1 beta, respectively. J. Immunol. 163, 3123–
3130.
39. Center, D. M., Kornfeld, H., Ryan, T. C., and Cruikshank,
W. W. (2000). Interleukin 16: implications for CD4 functions
and HIV-1 progression [Review]. Immunol. Today 21, 273–280.
40. Sakihama, T., Smolyar, A., and Reinherz, E. L. (1995). Molec-
ular recognition of antigen involves lattice formation between
CD4, MHC class II and TCR molecules [Review]. Immunol.
Today 16, 581–587.
41. Schifferli, J. A., Ng, Y. C., and Peters, D. K. (1986). The role of
complement and its receptor in the elimination of immune
complexes [Review]. N. Engl. J. Med. 315, 488 – 495.
Received October 12, 2001
Revised version received May 22, 2002
Published online August 2, 2002
42. Ravetch, J. V., and Clynes, R. A. (1998). Divergent roles for Fc
receptors and complement in vivo [Review]. Annu. Rev. Immu-
nol. 16, 421– 432.
43. Moran, M. F., Koch, C. A., Anderson, D., Ellis, C., England, L.,
Martin, G. S., and Pawson, T. (1990). Src homology region 2
domains direct protein-protein interactions insignal transduc-
tion. Proc. Natl. Acad. Sci. USA 87, 8622– 8626.
44. Maddon, P. J., Littman, D. R., Godfrey, M., Maddon, D. E.,
Chess, L., and Axel, R. (1985). The isolation and nucleotide
sequence of a cDNA encoding the T cell surface protein T4: A
new member of the immunoglobulin gene family. Cell 42, 93–
104.
45. Kyriakis, J. M. (1999). Making the connection: Coupling of
stress-activated ERK/MAPK (extracellular-signal-regulated ki-
nase/mitogen-activated protein kinase) core signalling modules
to extracellular stimuli and biological responses [Review]. Bio-
chem. Soc. Symp. 64, 29 – 48.