Activation of γδ T Cells in Malaria: Interaction of Cytokines and a Schizont‐Associated Plasmodium falciparum Antigen

Abstract

A soluble Plasmodium falciparum antigen that specifically stimulates γδ T cells has been found associated predominantly with schizonts rather than ring forms, trophozoites, or gametocytes. This
schizont-associated antigen (SAA) is resistant to protease digestion, is anionic at pH 8.5, is heat- and pH-resistant, and
contains a phosphate group(s) that is crucial for biologic activity. Partially purified SAA induced proliferative responses
and interferon-γ production by γδ T cells. These stimulatory effects were greatly enhanced by monocyte-derived cytokines,
interleukin (IL)-10, IL-12, and IL-1β, but not by tumor necrosis factor-α. Taken together, these results suggest that concurrent
stimulation of αδ T cells by SAA and by cytokines released from activated monocytes (IL-10, IL-12, IL-1β) may represent the
major mechanism underlying the selective activation of γδ T cells that is consistently observed in clinical cases of P. falciparum infection.

Full text

233
Activation of
gd
T Cells in Malaria: Interaction of Cytokines and a Schizont-
Associated Plasmodium falciparum Antigen
S. Pichyangkul, P. Saengkrai, K. Yongvanitchit,
Department of Immunology and Medicine, US Army Medical
Component, AFRIMS, Bangkok, Thailand
A. Stewart, and D. G. Heppner
A soluble Plasmodium falciparum antigen that specifically stimulates
gd
T cells has been found
associated predominantly with schizonts rather than ring forms, trophozoites, or gametocytes. This
schizont-associated antigen (SAA) is resistant to protease digestion, is anionic at pH 8.5, is heat-
and pH-resistant, and contains a phosphate group(s) that is crucial for biologic activity. Partially
purified SAA induced proliferative responses and interferon-
g
production by
gd
T cells. These
stimulatory effects were greatly enhanced by monocyte-derived cytokines, interleukin (IL)-10, IL-
12, and IL-1
b
, but not by tumor necrosis factor-
a
. Taken together, these results suggest that
concurrent stimulation of
gd
T cells by SAA and by cytokines released from activated monocytes
(IL-10, IL-12, IL-1
b
) may represent the major mechanism underlying the selective activation of
gd
T cells that is consistently observed in clinical cases of P. falciparum infection.
Healthy individuals have a small subpopulation of T lympho- limited role in protection but could contribute to the pathophys-
iologic changes observed in severe malaria through a systemiccytes with T cell receptors (TCRs) made up of a
gd
-chain pair
rather than an
ab
-chain pair [1].
gd
T cells differ from
ab
T release of cytokines [14, 15].
The mechanisms underlying the marked
gd
T cell responsecells in several parameters, including ontogenic appearance,
tissue distribution, and antigen recognition [2]. Although in- to P. falciparum and the specific antigens involved still remain
unknown. To elucidate these issues, we describe a series ofcreased numbers of
gd
T cells have been noted in several
infectious disease processes, including tuberculosis, listeriosis, experiments to investigate the interaction of
gd
T cells with
malaria antigen and with host cytokine mediators.leishmaniasis, trypanosomiasis, and malaria [3 –9], the physio-
logic role of
gd
T cell activation in these infectious diseases
is not fully understood and may vary. Materials and Methods
Infection with Plasmodium falciparum can cause a number
of immunologic disturbances in the host, and among these is
Parasite cultures and lysate preparations. AP. falciparum
clone from Thailand, TM267R, was cultured in group O
/
human
a marked increase in circulating
gd
T cells. This induction can
red blood cells (RBCs) suspended in RPMI 1640 (GIBCO Labora-
be readily observed both in vivo and in vitro [7 –11]. The early
tories, Grand Island, NY) with 10% heat-inactivated human serum
and consistent appearance of activated
gd
T cells in natural
(medium). Parasite cultures were synchronized by two treatments
infections suggests a possible role of these cells in the immune
with 5% sorbitol. Cultures were harvested when the parasites were
response to malaria. In vitro evidence of a protective role of
in specific stages (ring, trophozoite, and schizont). Parasites were
human
gd
T cells has been reported by demonstrating inhibition
further purified by flow cytometric sorting of hydroethidine-stained
of P. falciparum growth in culture by
gd
T cells from naive
parasite-infected RBCs (FACStar Plus2; Becton Dickinson, Moun-
donors [12]. In an in vivo mouse model using mice congenitally
tain View, CA). Hydroethidine (Sigma, St. Louis) is a vital dye
deficient in either
gd
or
ab
T cells,
gd
T cells have been
that is metabolically converted to ethidium and then binds to nu-
shown to mediate protection against liver stages but not blood
cleic acids, allowing parasite-infected RBCs to be distinguished
stages of Plasmodium yoelii [13].
from normal RBCs by flow cytometry, as described by Elloso et
al. [12]. Using these techniques, we normally obtained 95%–98%
gd
T cells are also known to produce a variety of proin-
homogenous populations of ring, trophozoite, and schizont forms.
flammatory cytokines [14]. It has been speculated that a mas-
Hydroethidine staining did not have any observable effect on sub-
sive increase of
gd
T cells in peripheral blood might have a
sequent parasite growth or on the ability of harvested parasites to
stimulate lymphoid cells (compared with unstained parasites, data
not shown).
Received 7 October 1996; revised 27 February 1997.
P. falciparum V13 isolate from a Vietnamese patient was used
Presented in part: 45th annual meeting of the American Society of Tropical
for gametocyte preparation. This isolate readily produces gameto-
Medicine Hygiene, Baltimore, Maryland, December 1996 (abstract #14).
cytes in culture with RPMI 1640 supplemented with 10% heat-
Financial support: United States Army Medical Research & Materiel Com-
inactivated human serum. Parasite cultures were synchronized for
mand.
The authors’ views do not reflect the position of the United States Depart-
schizonts, as described earlier. Thirty-percent schizonts in human
ment of the Army or of the Department of Defense.
group O
/
RBCs were cultured for 7–10 days at 377C, 5% CO
2
Reprints or correspondence: Dr. Sathit Pichyangkul, Dept. of Immunology
without adding new human O
/
RBCs. The medium was changed
and Medicine. US Army Medical Component, AFRIMS, APO AP 96546.
every 24 h. Gametocyte formation was determined by examination
The Journal of Infectious Diseases 1997;176:233 – 41
of Giemsa-stained blood smears. Gametocyte cultures were har-
This article is in the public domain.
0022– 1899/97/7601–0030
vested when asexual-stage parasites became difficult to detect mi-
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

234 Pichyangkul et al. JID 1997;176 (July)
croscopically. Gametocytes were stained with hydroethidine and Enzyme treatment characterization of antigen. Supernatants
from schizont lysates, prepared as described earlier, were observedthen further purified by flow cytometric sorting as described. Purity
of 70%–80% was normally achieved in the gametocyte prepara- to maintain their ability to stimulate
gd
T cells after heat treatment
at 1007C for 10 min. These schizont lysate supernatants weretions.
To prepare the lysate of each purified parasite stage, parasite treated with a combination of nonspecific proteases in an attempt
to eliminate their stimulatory activity. Proteinase K (Sigma) waspellets were resuspended in TRIS-HCl buffer (0.01 M, pH 8.5) at
10
7
cells/mL and subjected to 4–5 freeze-thaw cycles (immersion added at 2
m
g/100
m
L for 4 h at 377C, followed by heat inactivation
at 1007C for 3 min. Then, protease XIV (Sigma) was added at 2of tube in ethanol– dry ice for 3 min, followed by a 377C water
bath for 3 min). Parasite lysate supernatants were obtained by
m
g/100
m
L for another4hat377C, followed by a second heat
inactivation at 1007C for 3 min.filtering the lysates through a 0.22-
m
m filter unit (Millipore, Bed-
ford, MA). For lymphoid cell stimulation, parasite lysates or super- In a separate experiment, schizont lysate supernatants were
treated with 10 U/100
m
L calf intestinal alkaline phosphatasenatants were used at a final concentration of 10% (vol/vol).
Purification of lymphocyte subpopulations. Peripheral blood (Sigma) for 24 h at 377C. All enzyme-treated samples were diluted
1:10 with medium, filtered through a 0.22-
m
m filter, and used tomononuclear cells (PBMC) were prepared by centrifugation sepa-
ration using Ficoll-Hypaque (Histopaque 1.077; Sigma). Blood stimulate
gd
T cells at a final concentration of 5% (vol/vol). As
a control, heat-inactivated alkaline phosphatase was prepared bywas taken from healthy, malaria-naive 25- to 40-year-old Thai
men. Lymphocytes were enriched using the sheep erythrocyte ro- heating the enzyme to 1007C for 3 min. Supernatants were also
dialyzed in 1000 Da cutoff dialysis bags (Spectrum Medical Indus-sette method. The resulting lymphocytes were further fractionated
by negative selection using immunomagnetic beads (Dynal, Lake tries, Los Angeles) at pH 2 (0.1 Mglycine HCl buffer, pH 2.0)
and pH 10 (0.1 Mcarbonate buffer, pH 10) for 24 h each.Success, NY). To enrich for the
ab
T cell population, lymphocytes
were stained with anti–TCR-
g
/
d
-1 and anti–CD56 (Becton Dick- Fractionation of schizont lysate supernatants by anion-exchange
chromatography. For high-performance liquid chromatographyinson), and the stained cells were removed with magnetic beads
coated with goat anti–mouse IgG, according to the manufacturer’s (HPLC) fractionation, schizont lysate supernatants, prepared as
described earlier, were heated at 1007C for 3 min to eliminateinstructions. To enrich for a population of
gd
T cells and NK
cells, lymphocytes were stained with anti –TCR-
a
/
b
-1 (Becton hemoglobin. Supernatants were then lyophilized and reconstituted
with TRIS-HCl buffer (0.01 M, pH 8.5) to achieve a 10-fold in-Dickinson), and the stained cells were removed with magnetic
beads coated with anti–mouse IgG. The enriched populations of crease in concentration. Samples (100
m
L) of concentrated material
were loaded on a DEAE-5-PW (75 17.5 mm) HPLC column.
ab
T cells, and of
gd
T cells and NK cells were further purified by
positive or negative selection by flow cytometry (FACStar Plus2). They were eluted with a linear gradient of 0 –0.5 MNaCl in a
TRIS-HCl buffer (0.01 M, pH 8.5), at a flow rate of 1 mL/min.Antibodies used for cell selection were obtained from Becton
Dickinson. Cells were positively selected using anti– TCR-
a
/
b
-1 One-milliliter fractions were collected, diluted 1:10 with medium,
filtered, and assayed for stimulating activity on
gd
T cells at afluorescein isothiocyanate (FITC) for
ab
T cells, anti–TCR-
g
/
d
-1 FITC for
gd
T cells, or anti–CD16 phycoerythrin (PE) in final concentration of 5% (vol/vol). Active fractions were then
pooled and later used for studying
gd
T cell activation.combination with anti–CD56 PE for NK cells. In negative selec-
tion experiments,
ab
T cells were purified by the removal of cells Lymphoproliferation as measured by [
3
H]thymidine incorpora-
tion. Highly purified
gd
T cells (10
5
cells/well, 96-well plates)stained with anti –CD14 FITC (monocytes), anti– CD20 FITC (B
cells), anti–TCR-
g
/
d
-1 PE, anti–CD16 PE, and anti – CD56 PE. were stimulated with DEAE-purified schizont-associated antigen
(SAA; 5% concentration, vol/vol) and with cytokines, includingSimilarly,
gd
T cells were negatively selected by removal of cells
stained with anti–CD14 FITC, anti – CD20 FITC, anti–TCR-
a
/
b
- interleukin (IL)-1
b
, IL-10, IL-12, and tumor necrosis factor (TNF-
a
), each at 100 ng/mL (R&D Systems, Minneapolis). After 3 days1 FITC, anti–CD16 PE, and anti –CD56 PE. Negative selection
of NK cells was achieved by removal of cells stained with anti– of incubation, [
3
H]thymidine (0.5
m
Ci/250
m
L/well) was added,
and cultures were incubated for another 24 h. Cells were thenCD14 FITC, anti – CD20 FITC, anti – TCR-
a
/
b
-1 FITC, and anti–
TCR-
g
/
d
-1 FITC. harvested onto glass fibers (Mach II harvester; Tomtec, Orange,
CT), and radioactivity uptake was measured using a liquid scintilla-We observed that positively and negatively selected cells
showed similar responses to P. falciparum stimulation. Further- tion counter (Beta plate; Wallac, Turku, Finland).
Cytokine production. Highly purified
gd
T cells were stimu-more, the antibody staining used in this study did not have any
observable effect on cell activation (data not shown). Thus, we lated with DEAE-purified SAA and with cytokines (IL-1
b
, IL-10,
IL-12, and TNF-
a
), using the conditions described earlier. Cultureused positively selected cells for all experiments. The populations
of
ab
T cells,
gd
T cells, and NK cells obtained by these purifica- supernatants were harvested after 24 h of incubation and then
assayed for cytokine production using commercial ELISA kits fortion protocols were normally 90% – 95% pure.
Flow cytometry analysis of cell activation (CD69 and CD25 IFN-
g
and IL-4 (Genzyme, Cambridge, MA).
expression). Parasite lysates or lysate supernatants derived from
ring, trophozoite, schizont, or gametocyte stages were added to
Results
highly purified
ab
T cells,
gd
T cells, and NK cell culture (10
5
cells/well) at a final concentration of 10% (vol/vol) in 96-well
Activation of lymphocyte subpopulations by schizont-stage
plates. After 24 h of incubation, stimulated cells were stained with
parasites. It has been reported that in PBMC cultures, the
anti–CD25 FITC and anti – CD69 PE (Becton Dickinson). Ten
percentage of
gd
T cells increased significantly after stimula-
thousand stained cells, gated on forward and side scatter, were
tion with schizont-stage P. falciparum [10 –15]. Highly purified
analyzed for log fluorescence intensity by flow cytometry (FACS-
can; Becton Dickinson).
(ú90%) lymphocyte subsets consisting of
ab
T cells, NK cells,
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

235JID 1997;176 (July)
gd
T Cell Activation in Malaria
Figure 1. CD69 and CD25 expression on lym-
phocyte subpopulation in response to P. falcipa-
rum schizont stimulation. Highly purified
ab
T
cells, NK cells, and
gd
T cells were stimulated
with schizont lysates or schizont lysate superna-
tants at final 10% (vol/vol) concentration. After 24
h of incubation, cells were harvested, stained with
anti–CD69 PE in combination with anti – CD25
fluorescein isothiocyanate (FITC), and then ana-
lyzed by flow cytometry. Gates were set with iso-
type control antibodies conjugated with either phy-
coerythrin or FITC. Nos. in upper right corner of
each histogram represent % of cells that express
CD69
/
/CD25
/
; nos. in upper left corners represent
% of cells that express CD69
/
/CD25
0
. Similar
results were obtained in 3 experiments.
or
gd
T cells were obtained by rosette formation. Immunomag- Activation of
gd
T cells by antigen prepared from different
life-cycle stages of P. falciparum. We have previously re-netic separation and flow cytometric sorting were used to char-
acterize the response to schizont antigen stimulation in different ported that schizont-stage parasites, compared with trophozo-
ite- and ring-stage parasites, strongly induce activation of thelymphocyte subpopulations. The purified lymphocyte subsets
were cultured with schizont lysates for 24 h. As shown in figure monocyte/macrophage cell type [16]. In the current experi-
ments, we assessed antigen preparations from different life-
1, Ç50% of the
gd
T cells became activated after exposure to
cycle stages of P. falciparum parasites for their ability to induce
schizont lysate, as monitored by the expression of CD69 (an
gd
T cell activation. Purified
gd
T cells were cocultured with
early activation antigen) and CD25 (an IL-2 receptor). No con-
parasite lysates derived from either ring, trophozoite, or schiz-
comitant activation of
ab
T cells was observed after similar
ont stages. After 24 h, expression of CD69 was measured by
exposure to schizont lysate. A small number of positively se-
flow cytometry. The results in figure 2A show that the antigen
lected NK cells were activated in response to stimulation with
that selectively activates
gd
T cells was mainly and consistently
schizont lysate; the results are largely due to
gd
T cell contami-
associated with schizont-stage parasites. The stimulating activ-
nation, since we have observed that Ç5%–10% of
gd
T cells
ity for
gd
T cells gradually increased as the growth of the
also express NK markers (CD56) (data not shown). A similar
parasite advanced from ring to schizont. In a similar compari-
degree of
gd
T cell activation was also observed when cells
son of parasite lysates derived from schizont and gametocyte
were stimulated with filtered supernatant of schizont lysate
stages, schizont lysate was found to contain more stimulating
rather than entire lysate (figure 1). No monocytes were present
activity for
gd
T cells than gametocyte lysate did (figure 2B).
in any purified lymphocyte subsets, as confirmed both by mor-
phology with Giemsa staining and by anti–CD14 immunostain- Physiochemical characteristic of the SAA that activates
gd
T cells.
gd
T cell–specific antigen was present in the solubleing (data not shown). Taken together, these results suggest that
a soluble schizont antigen is directly stimulatory for
gd
T cells, fraction of schizont lysate. The results of attempts to character-
ize the nature of this
gd
T cell– specific SAA are presented inwithout a requirement for antigen presentation by monocytes.
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

236 Pichyangkul et al. JID 1997;176 (July)
Figure 2. CD69 expression on
gd
T cells in response to parasite lysate prepared from different life-cycle stages of P. falciparum. Highly
purified
gd
T cells were stimulated with parasite lysates prepared from ring, trophozoite, and schizont stages at final 10% (vol/vol) concentration
(A). In separate experiment,
gd
T cells were stimulated with schizont and gametocyte lysates prepared from P. falciparum isolate (V13) that
readily produces gametocytes in culture medium (B). After 24 h of incubation, cells were harvested and stained with anti– CD69 fluorescein
isothiocyanate (FITC) and analyzed by flow cytometry. Gates were set with isotype control antibody conjugated with FITC. M1 region consists
of CD69
0
cells; M2 region consists of CD69
/
cells. Nos. in each histogram represent % of
gd
T cells that express CD69. Similar results were
obtained in 3 experiments.
table 1. The enzyme alkaline phosphatase completely inhibited of SAA-induced
gd
T cells. P. falciparum is known to induce
gd
T cell expansion [10 –15]. The mechanism underlying
gd
the stimulating activity for
gd
T cells of SAA. This suggests
T cell proliferation, however, remains elusive. The monocyte-
that SAA contains one or more phosphate groups that are im-
derived cytokines IL-10 and IL-12 are known to have growth-
portant for its biologic activity. The
gd
T cell–specific antigen
promoting activity on preactivated
gd
T cells [17, 18]. Our
is protease-resistant in nature since the majority of its activity
study showed that both IL-10 and IL-12 (at concentrations of
was maintained after protease digestion. Similarly, it is heat-
1–100 ng/mL) specifically stimulated resting
gd
T cells, but
and pH-resistant.
not resting
ab
T cells, to express the CD69 antigen (data not
The results obtained from anion-exchange chromatography
shown). These results led to a series of experiments designed
indicates that SAA bound to a DEAE column at pH 8.5 and
to determine the monocyte-derived cytokines that might be
could be eluted at a low salt concentration (figure 3).
involved in SAA-induced
gd
T cell proliferative response and
Role of monocyte-derived cytokines (IL-1
b
, IL-10, IL-12,
IFN-
g
production. The effects of cytokines IL-1
b
, IL-10, IL-
and TNF-
a
) on proliferative response and IFN-
g
production
12, and TNF-
a
, which are present in supernatants of P. falcipa-
rum schizont–stimulated PBMC cultures (data not shown),
were evaluated on SAA-induced
gd
T cell proliferation.
Table 1. Effects of various treatments on the schizont lysate super-
[
3
H]thymidine uptake analysis after 4 days in culture revealed
natant used to stimulate
gd
T cell activation.
minimal proliferation in
gd
T cell cultures stimulated with
Treatment %
gd
T cells expressing CD69
DEAE-purified SAA alone. Greatly increased labeling by
[
3
H]thymidine was observed in
gd
T cell cultures stimulated
None 34
with SAA in combination with either IL-10, IL-12, or IL-1
b
/Heat (1007C, 10 min) 36
but not TNF-
a
(figure 4).
/pH 2 37
We next studied the effect of these cytokines on SAA-in-
/pH 10 37
/Protease 29
duced IFN-
g
production. As shown in figure 5,
gd
T cell
/Alkaline phosphatase 9*
cultures stimulated with DEAE-purified SAA produced small
/Heat-inactivated alkaline phosphatase 35
amounts of IFN-
g
. However, a substantial increase in IFN-
g
production (2- to 3-fold) was consistently observed in all
* Nonstimulated cell cultures have a baseline expression of 9% of
gd
T
cells expressing CD69.
experiments when
gd
T cells were cultured with both SAA
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

237JID 1997;176 (July)
gd
T Cell Activation in Malaria
Figure 3. Anion-exchange chromatography of P. fal-
ciparum schizont–associated antigen. One hundred mi-
croliters of schizont lysate supernatants were loaded onto
a DEAE–high-performance liquid chromatography col-
umn. Samples were eluted with TRIS-HCl buffer (0.01
M, pH 8.5) in a linear gradient of 0–0.5 MNaCl. Flow
rate was 1 mL/min; 1-mL fractions were collected. Eluted
samples were analyzed for ability to induce CD69 expres-
sion on
gd
T cells. lindicates % of
gd
T cells expressing
CD69 after 24 h in culture and are plotted against A
280
of individual fraction numbers.
Figure 4. Proliferative response of schizont-associated antigen (SAA)– reactive
gd
T cells. Highly purified
gd
T cells were cultured in flat-
bottom microculture plates and stimulated with DEAE-purified SAA (10% vol/vol) and with interleukin (IL)-1
b
, IL-10, IL-12, or tumor
necrosis factor (TNF)-
a
at 100 ng/mL each. [
3
H]thymidine was added at day 3 and incubated for another 18 –24 h. Cells were harvested, and
[
3
H]thymidine uptake was measured. Results are mean {SE of 3 separate experiments. CPM, counts per min.
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

238 Pichyangkul et al. JID 1997;176 (July)
Figure 5. Interferon (IFN)-
g
production from schizont-associated antigen (SAA)–reactive
gd
T cells. Highly purified
gd
T cells were
cultured and stimulated with DEAE-purified SAA (5% vol/vol) and with interleukin (IL)-1
b
, IL-10, IL-12, or tumor necrosis factor (TNF)-
a
(100 ng/mL each). Culture supernatants were harvested after 24-h incubation and assayed for IFN-
g
production. Results are mean {SE of 3
separate experiments.
and IL-10 or IL-12. IFN-
g
production was not enhanced in
gd
Taken together, these data indicate that SAA-reactive
gd
T
cells are the major producer cells of the early release of IFN-T cells cultured with SAA and IL-1
b
or TNF-
a
. IL-4 was
not produced in any experimental condition studied (data not
g
seen in schizont-stimulated PBMC cultures.
shown).
SAA-reactive
gd
T cells as major producer cells of an early Discussion
released IFN-
g
in P. falciparum schizont–stimulated PBMC
cultures. Preliminary experiments had indicated that schiz- One of the most consistent and striking responses to falcipa-
onts or schizont lysate induced early IFN-
g
production in rum malaria infection is the early expansion of
gd
T cells [7 –
PBMC cultures. To further evaluate the contribution of SAA- 11]. The nature of the
gd
T cell–specific antigen and the
reactive
gd
T cells to the observed IFN-
g
production of PBMC mechanism underlying the observed activation of
gd
T cells
stimulated with schizont lysate, we measured the IFN-
g
pro- have not been well studied. The present investigation has docu-
duction in PBMC cultures after positive removal of
gd
T cells mented the presence of a
gd
T cell–specific malaria antigen
by flow cytometry. The results in figure 6 show that
gd
T cell that is protease-resistant in nature and mainly associated with
depletion diminishes the early production of IFN-
g
by ú90% parasites in the schizont stage (SAA). SAA is anionic at pH
compared with that of control undepleted populations. These 8.5 and is heat- and pH-resistant. The presence of a phosphate
undepleted populations had been stained and run through the group or groups, critical for its biologic activity in stimulating
flow cytometer without sorting as a control.
gd
T cells, is inferred from its inactivation after alkaline phos-
An additional experiment used schizont lysate that had been phatase treatment. SAA shows some chemical similarities with
treated with alkaline phosphatase to inhibit the stimulating ac- mycobacterial antigen, which is also specific for
gd
T cells
tivity of SAA for
gd
T cells. Alkaline phosphatase treatment [19]. These apparent similarities include the probable nonpro-
of schizont lysate caused a substantial reduction in production teinaceous nature, the tendency to be anionic at high pH, and
of IFN-
g
in PBMC cultures (figure 7). However, no reduction the presence of phosphate groups that are critical for their
was seen when the alkaline phosphatase was heat-inactivated biologic activities. A recent report has shown that the mycobac-
terial antigen that activates
gd
T cells is a derivative of thymi-prior to incubation with the schizont lysate.
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

239JID 1997;176 (July)
gd
T Cell Activation in Malaria
Figure 6. Depletion of
gd
T cells in peripheral blood mononuclear cell (PBMC) cultures abolished interferon (IFN)-
g
production in response
to schizont lysate stimulation. PBMC isolated from malaria-naive individuals (4 110
6
cells) were stained with anti–TCR-
g
/
d
-1 fluorescein
isothiocyanate (FITC), washed, and analyzed by flow cytometry. Negatively stained cells were then sorted and used as
gd
T cell–depleted
PBMC. Control undepleted populations were stained with isotype control antibody FITC and then negatively sorted by flow cytometry, under
same conditions used to obtain
gd
T cell–depleted PBMC. Both
gd
T cell –depleted PBMC and control undepleted PBMC were stimulated
with schizont lysate at final 10% (vol/vol) concentration. Supernatants were harvested after 24 h of incubation and assayed for IFN-
g
production.
Results are mean {SE of 3 separate experiments.
dine triphosphate [20]. Our findings agree with a recent study in these studies await further investigation. In our study, the
observed recognition of SAA by
gd
T cells and their subse-by Berh et al. [21], which also suggests that phosphorylated
antigens from P. falciparum stimulate V
g
9/V
d
2
gd
T cells. quent activation and proliferation do not require monocytes as
antigen-presenting cells. This finding supports the notion thatThe details of the molecular structure of the SAA observed
Figure 7. Inhibition of schizont lysate– induced interferon (IFN)-
g
production by alkaline phosphatase (AP) treatment. PBMC isolated from
malaria-naive individuals (4 110
6
cells/mL) were stimulated with schizont lysates, AP-treated schizont lysates, or heat-inactivated AP-treated
schizont lysates at final 20% (vol/vol) concentration. Culture supernatants were harvested after 24 h of incubation and assayed for IFN-
g
production. Results are mean {SE of 3 separate experiments.
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

240 Pichyangkul et al. JID 1997;176 (July)
gd
T cells recognize antigens without a requirement for major blood-stage malaria infection appears to be limited. In the Plas-
modium chabaudi mouse model, mice that lack
ab
T cellshistocompatibility complex –restricted presentation [22, 23].
A stimulatory effect of monocyte cytokines on NK and
gd
lose the ability to control primary infections with blood-stage
parasites; mice lacking
gd
T cells remain able to resolve theT cell activation has been reported [24, 25]. Examination of
the
gd
T cell proliferative response to P. falciparum schizont infection [31]. These findings suggest that
gd
T cells and their
cytokines have limited importance in protective immunityantigen revealed an important enhancing effect of the activated
monocyte cytokines IL-10, IL-12, and IL-1
b
. Each of these against blood-stage parasites in a murine model. We speculate
that the predominance of an SAA-specific response by
gd
Tcytokines, in combination with SAA, resulted in augmentation
of the proliferation of
gd
T cells compared with that elicited cells, which is associated with high levels of cytokine produc-
tion, may provide an environment that favors development ofby each type of stimulus alone. The synergistic effect among
IL-10, IL-12, and IL-1
b
on the proliferative response of SAA- immunopathology rather than immunoprotection. For example,
the observed increase in
gd
T cells in the circulation is knownreactive
gd
T cells is under investigation.
P. falciparum is known to induce IFN-
g
production [26 – to correlate with gastrointestinal pathology in malaria [29]. The
consequences of the
gd
T cell response in malaria warrant28]. The precise mechanism and the nature of the IFN-
g
–
producing cells, however, remains unknown. In these experi- further investigation.
In conclusion, we have identified a P. falciparum antigen,ments, we have shown that stimulation of purified populations
of
gd
T cells with SAA resulted in early IFN-
g
production. SAA, which is specifically recognized by
gd
T cells but not
by
ab
T cells or by most NK cells. The antigen is resistant toThis production was found to be greatly enhanced by the pres-
ence of the monocyte cytokines IL-10 or IL-12. However, IL- protease digestion, and it contains a phosphate group(s) that is
crucial for its biologic activity. Optimal activation of SAA-1
b
, which enhanced SAA-induced
gd
T cell proliferation,
failed to enhance SAA-induced IFN-
g
production of
gd
T reactive
gd
T cells is dependent on the presence of cytokines
released from activated monocytes, IL-10, IL-12, and IL-1
b
.cells. In control experiments using unfractionated populations
of PBMC, either depletion of
gd
T cells from PBMC cultures Upon exposure to SAA in the presence of these cytokines,
purified
gd
T cells both proliferate and produce large amountsor removal of the stimulating activity for
gd
T cells of SAA
from the schizont lysate significantly reduced production of of IFN-
g
. SAA may prove to be a significant mediator of the
immunopathology of human falciparum malaria.IFN-
g
. These results strongly suggest a crucial role for
gd
T
cells, SAA, and possibly IL-10 and IL-12 in the early IFN-
g
production observed in cultures of schizont lysate–stimulated
References
PBMC. The precise roles of IL-10 and IL-12 and their effect
on P. falciparum schizont –induced IFN-
g
production require
1. Brenner MB, Mclean J, Dialynas DP, et al. Identification of a putative
second T cell receptor. Nature 1986;322:145– 9.
further study.
2. Haas W, Pereira P, Tonegawa S.
g
/
d
cells. Annu Rev Immunol 1993; 11:
Recent observations in patients infected with Plasmodium
637–85.
vivax showed that the increase in circulating
gd
T cells is
3. Modlin RL, Pirmez C, Hofman FM, et al. Lymphocytes bearing antigen-
associated with the onset of schizont rupture [29]. Malaria
specific
gd
T-cell receptor accumulate in human infections disease le-
sion. Nature 1989;339:544– 8.
pigment is known to be a potent inducer of monocyte/macro-
4. Ohga S, Yoshikai Y, Takeda Y, Hiromatsu K, Nomoto K. Sequential
phage activation and cytokine production [16]. When seques-
appearance of
g
/
d
and
a
/
b
bearing T cells in the peritoneal cavity
tered schizont-stage parasites rupture, there will be a simultane-
during an i.p. infection with Listeria monocytogenes. Eur J Immunol
ous release of SAA and parasite pigment. The phagocytosis of
1990;20:533– 8.
particulate pigment products by macrophages will lead to the
5. Uyemura K, Klotz J, Pirmez C, et al. Microanatomic clonality of
g
/
d
T
cells in human leishmaniasis lesions. J Immunol 1992;148:1205– 11.
release of cytokines, including IL-10, IL-12, and IL-1
b
.We
6. Minoprio P, Bandeira A, Mota Santos T, Coutinho A. Preferential expan-
theorize that these monocyte products then may in turn augment
sion of Ly-1-B and CD40CD80T cells in the polyclonal lymphocyte
the activation of SAA-reactive
gd
T cells, resulting in their
responses to murine T. cruzi infection. Int Immunol 1989;1:176– 84.
selective proliferation and the production of large amounts of
7. Roussilhon C, Agrapart M, Ballet JJ, Bensussan A. T lymphocytes bearing
IFN-
g
.
the
gd
T cells receptors in patients with acute Plasmodium falciparum
malaria. J Infect Dis 1990;162:283– 5.
It is not clear whether this activation of
gd
T cells in response
8. Ho M, Webster HK, Tongtawe P, Pattanapanyasat K, Weidanz WP. In-
to SAA has a beneficial or a deleterious effect on the patient.
creased
gd
T cells in acute Plasmodium falciparum malaria. Immunol
Tsuji et al. [13] recently demonstrated that
gd
T cells are
Lett 1990;25:139– 41.
crucial for protection against hepatic-stage malarial parasites
9. Goodier M, Fey P, Eichmann K, Langhorne J. Human peripheral blood
and postulated that self-stress proteins (heat-shock proteins)
gamma delta T cells respond to antigen of Plasmodium falciparum. Int
Immunol 1992;4:33– 41.
may be expressed by parasitized hepatocytes and act as target
10. Goerlich R, Ha
¨cker G, Pfeffer K, Heeg K, Wagner H. Plasmodium falcipa-
antigens responsible for
gd
T cell activation. Activated
gd
T
rum merozoites primarily stimulate the V gamma 9 subset of human
cells produce IFN-
g
[14], which suppresses the growth of he-
g
/
d
T cells. Eur J Immunol 1991;21:2613– 6.
patic-stage malarial parasites by stimulating nitric oxide pro-
11. Behr C, Dubois P. Preferential expansion of V
g
9V
d
2 T cells following
duction [30]. Although
gd
T cells can be cytolytic against
stimulation of peripheral blood lymphocytes with extracts of Plasmo-
dium falciparum. Int Immunol 1992;4:361– 6.
merozoites in vitro [12], the in vivo role of
gd
T cells against
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from

241JID 1997;176 (July)
gd
T Cell Activation in Malaria
12. Elloso MM, van der Heyde HC, vande Waa JA, Manning DD, Weidanz 23. Sciammas R, Johnson RM, Sperling AI, et al. Unique antigen recognition
WP. Inhibition of Plasmodium falciparum in vitro by human
gd
T cells. by herpes virus–specific TCR-
gd
cell. J Immunol 1994;152:5392– 7.
J Immunol 1994;153:1187– 94. 24. Tripp CS, Wolf SF, Unanue ER. Interleukin 12 and tumor necrosis factor
13. Tsuji M, Mombaerts P, Lefrancois L, Nussenzweig RS, Zavala F, Tone-
a
are costimulators of interferon-
g
production by natural killer cells in
gawa S.
gd
T cells contribute to immunity against the liver stages of severe combined immunodeficiency mice with listeriosis, and interleu-
malaria in
ab
T-cell–deficient mice. Proc Natl Acad Sci USA 1994;kin 10 is a physiologic antagonist. Proc Natl Acad Sci USA 1993; 90:
91:345–9. 3725 –9.
14. Goodier MR, Lundqvist C, Hammarstrom ML, Troye-Blomberg M, 25. Skeen MJ, Zeigler HK. Activation of
gd
T cells for production of IFN-
g
Langhorne J. Cytokine profiles for human V
g
9
/
T cells stimulated by is mediated by bacteria via macrophage-derived cytokines IL-1 and IL-
Plasmodium falciparum. Parasite Immunol 1995;17:413– 23. 12. J Immunol 1995;154:5832 –41.
15. Langhorne J, Goodier M, Behr C, Dubois P. Is there a role for
gd
T cells 26. Troye-Blomberg M, Anderson G, Stoczkowska M, et al. Production of IL
in malaria? Immunol Today 1992;13:298– 300. 2 and IFN
g
by T cells from malaria patients in response to Plasmodium
16. Pichyangkul S, Saengkrai P, Webster K. Plasmodium falciparum pigment falciparum or erythrocyte antigens in vitro. J Immunol 1985; 135:3498–
induces monocytes to release high levels of tumor necrosis factor-
a
and 504.
interleukin-1
b
. Am J Trop Med Hyg 1994;51:430– 5. 27. Chizzolini C, Grau GE, Geinoz A, Schrijvers D. T lymphocyte interferon-
17. Perussia B, Chan SH, D’Andrea A, et al. Natural killer (NK) cell stimula- gamma production induced by Plasmodium falciparum antigen is high
tory factor or IL-12 has differential effects on the proliferation of TCR- in recently infected non-immune and low in immune subjects. Clin Exp
ab
/
, TCR-
gd
/
T lymphocytes and NK cells. J Immunol 1992; 149: Immunol 1990;79:95 –9.
3495–502. 28. Riley EM, Jepsen S, Anderson G, Otoo LN, Greenwood BM. Cell-medi-
18. Schlaak JF, Hermann E, Gallati H, Meyer-Zum-Buschenfelde KH. Differ- ated immune response to Plasmodium falciparum antigens in adult Gam-
ential effect of IL-10 on proliferation and cytokine production of human bians. Clin Exp Immunol 1988; 71:377– 82.
g
/
d
and
a
/
b
T cells. Scan J Immunol 1994;39:209– 14. 29. Perera MK, Carter R, Goonewardene R, Mendis KN. Transient increase
19. Tanaka Y, Sano S, Nieves E. Nonpeptide ligands for human
gd
T cells. in circulating
g
/
d
T cells during Plasmodium vivax malarial paroxysms.
Proc Natl Acad Sci USA 1994;91:8175– 9.
J Exp Med 1994;179:311– 5.
20. Constant P, Davodeau F, Peyrat MA, et al. Stimulation of human
gd
T
30. Sedegah M, Finkelman F, Hoffman SL. Interleukin 12 induction of inter-
cells by nonpeptidic mycobacterial ligands. Science 1994;264:267 –70.
feron
g
–dependent protection against malaria. Proc Natl Acad Sci USA
21. Berh C, Poupot R, Peyrat MA, et al. Plasmodium falciparum stimuli for
1994;91:10700– 2.
human
gd
T cells are related to phosphorylated antigens of Mycobac-
31. Langhorne J, Mombaerts JP, Tonegawa S.
ab
and
gd
T cells in the
teria. Infect Immun 1996;64:2892– 6.
immune response to the erythrocytic stages of malaria in mice. Int
22. Schild H, Mavaddat N, Litzenberger C, et al. The nature of major histocom-
patibility complex recognition by
gd
T cells. Cell 1994;76:29– 37. Immunol 1995;7:1005– 11.
/ 9d2b$$jy26 05-29-97 15:28:28 jinfal UC: J Infect
by guest on July 21, 2011jid.oxfordjournals.orgDownloaded from