© 1992 Oxford University Press

International Immunology, Vol. 4, No.3, pp. 361 -366

Preferential expansion of VT9 V^2 T cells following stimulation of peripheral blood lymphocytes with extracts of Plasmodium falciparum Unite de Parasitologie Experimental, Departement d'lmmunoiogie, 28 rue du Docteur Roux, 75724 Paris Cedex 15, France Key words T cells, VT9, Vj2, T cells, peripheral Wood lymphocytes, Plasmodium falciparum

A Plasmodium falciparum schizont lysate has been previously described as being a powerful inducer of proliferation for human peripheral T lymphocytes. In this report we study the phenotype of cycling T cells from unexposed donors and examine how the P. falciparum lysate compares with the conventional T cell mitogen phytohemagglutlnln (PHA), a known superantigen staphylococcal enterotoxln B (SEB), and a classical antigen pure protein derivative (PPD). We show that for this lymphoprollferative activity interaction with the MHC class II molecule is required and that In the presence of P. falciparum the great majority of the cycling cells at day 6 are yS T cells, all of them bearing VT9 V62. Our results suggest that P. falciparum induces a T cell proliferative response that resembles a response of human peripheral blood y6 T cells to superantigen. This observation Is In agreement with the elevated level of peripheral yd lymphocytes observed during and after malaria acute infection. Introduction Plasmodium falciparum infection is accompanied by several perturbations of the host immune system. One of the best known is the polyclonal activation of B lymphocytes which contribute to hypergammaglobulmemia, immune complex formation, and autoantibody production (1,2). At the T cell level a decrease of total number of peripheral T cells and some perturbations of the CD8/CD4 ratio (3-5) have been described. More recently, an augmentation of the relative number of yS T lymphocytes in the peripheral blood has been observed during and after acute malaria (6,7). In this context it was of interest to analyze how human peripheral blood T cells respond to a crude lysate of P. falciparum schizonts. We and others (8-12) have previously shown that a crude extract of P. falciparum can induce in vitro a strong proliferation of peripheral blood T cells from unexposed donors. These observations suggested that P. falciparum could act as a mitogen. Nevertheless, the kinetics of the proliferative response, as well as the inhibition by an anti-CD3 antibody (12), were in favor of an antigenic response. In order to further characterize this activity, we analyzed the phenotype of the cycling T lymphocyte population using a simultaneous double surface and DNA staining. We show that most of the T cells which are in cycle at the day of maximal

proliferation, as measured by [3H]thymidine incorporation, are yd T cells all bearing VT9 Vj2 and that this proliferation is inhibited when anti-class II antibodies are added at the beginning of the culture period.

Methods Cell preparation PBMC from heparinized Wood of healthy adults who have never been in contact with malaria were isolated by Ficoll-Hypaque density centrifugation. Stimuli Phytohemagglutinin (PHA) and staphyfococcal enterotoxin B (SEB) from Sigma (St Louis MO) were added to the cells at final concentrations of 2.5 and 10 /*g/ml respectively. Pure protein derivative (PPD) (Statens Serum Institute) was used as 5 /ig/ml. P. falciparum schizont lysate (PFSL) was obtained from a continuous culture in O + human erythrocytes of the knob + variant of the P. falciparum isolate FUP-1 (13). Schizont enrichment was performed by Plasmagel sedimentation (Laboratoire Roger Bellon, Neuilly sur Seine, France), giving preparations

Correspondence to: C. Behr Transmitting editor: V. Nussenzweig

Received 18 August 1991, accepted 22 November 1991

Downloaded from http://intimm.oxfordjournals.org/ at University of California, Santa Barbara on September 1, 2015

Charlotte Behr and Philippe Dubois

362

yS T cell activation by P. falciparum

containing >95%schizont-infected erythrocytes. Both parasitized and control uninfected erythrocytes from the same donor treated as above were washed three times in RPMI 1640 (Flow Laboratory). The final pellet was resuspended in 4 volumes of distilled water, aliquoted, and frozen at - 80°C until use. The lysate was used in the different tests at a final protein concentration of 37.5 Lymphocyte culture and proliferation test

Antibodies Anti-CD3, anrj-CD4, anti-CD8, and anti-HLA-DR-DP-DQ (13) were purchased from Coultronics (Hialed, USA); anti-a/3 (WT 31) and anti-CD25 from Becton-Dickinson (USA); anti-p75 from Endogen (USA); anti-76 (TCR61), anti-V41 (TCS61), and anti-V79 (V?9 (2A)) from T Cell Sciences (USA). Anti-VT9 (Ti V / ) and anti-Vj2 (1H4) were a gift from Dr Th. Hercend. Immunofluorescence test Cultivated cells were centrifugated on Ficoll-Hypaque gradient before staining. Cells were incubated at 4°C with each mAb for 30 min, washed, and resuspended in PBS BSA 3% (BSA from Sigma, St Louis, MO). For indirect staining the cells were further

Results and discussion Unlike a red Wood cell lysate, a PFSL has been previously shown to stimulate human T cells from non-exposed donors to undergo brisk proliferation (9,10,12). This proliferative response is maximal at day 6, as measured by [3H]thymidine incorporation and is accessory cell (AC) dependent. Antigens, in general, are processed by ACs and presented to the T cells in the context of MHC class II elements, whereas mitogens do not require processing and presentation by the ACs. To distinguish between these possibilities, we tested the presentation requirements for PFSL proliferative activity. Unlike PHA, both SEB- and PFSLinduced T cell proliferation require presentation by class II molecules. Indeed, monoclonal anti-HLA-DR and anti-HLA-DRDP-DQ antibodies inhibited the response of T lymphocytes to PFSL and to the known superantigen SEB, but had little effect on the response to PHA (Fig. 1). To characterize the phenotype of the cells which proliferate

cpm 100000 80000 60000

40000

20000

o-1-

NoAg

PHA

SEB

PFSL

Fig. 1. Inhibition of PFSL-induced proliferation by anti-HLA class II antibodies. PBMC were stimulated with PHA, SEB, and PFSL (as described in Methods) in the presence of mAbs anti-HLA-DR-DP DO at 10 P g/ml. Proliferation was assessed after 3 days for PHA and SEB stimulation and after 6 days for PFSL stimulation by measuring [3HJthymidine uptake. To control for non-specific inhibition, we used an irrelevant mAb of the same isotype (lgG2A) at the same concentration. Results are representative of six different experiments with cells from different donors.

Downloaded from http://intimm.oxfordjournals.org/ at University of California, Santa Barbara on September 1, 2015

For assessment of T cell proliferation, 2 x i O 5 PBMC per well were incubated in triplicate for 3 days (PHA, SEB) or 6 days (PPD, PFSL) at 37°C (95% humidity and 5% CO2) with the indicated stimuli, in 96-well round-bottomed plastic micro-culture plates (Costar). [3H]thymidine (1 /xCi) was added per well 14 h before harvesting. For phenotyping and cycle studies, bulk cultures were initiated in 24-well flat-bottomed plastic culture plates (Costar) with 2 x 106 PMBC per well for 3 days (PHA, SEB) or 6 days (PPD, PFSL).

incubated with the conjugated goat antj-mouse-PE (Immunotech, Marseille, France) and after with the mAb-FITC conjugated in the same way. After staining the cells were either analyzed directly or further stained with 7-aminoactinomycin D (7-AAD; Sigma) following a modification of Rabinovitch et al. (14): briefly, surfacestained cells were fixed overnight in 0.25% paraformaldehyde and, after washing, incubated with PBS-0.3% saponine and 7-AAD at 20 /tg/ml for a minimum of 2 h at 4°C before analysis. Fluorescence analysis was performed with a flow cytometer FACScan analyzer (Becton-Dickinson) with a angle laser excitation of 488 nm. Green (FITC) and red (PE) fluorescences were analyzed on a logarithmic scale, the 7-AAD fluorescence on a linear scale.

yS T cell activation by P. falciparum in vitro, PBMC were isolated from six healthy donors who had never been exposed to malaria and cultivated for a period of 3 days in the presence of SEB, or for 6 days in the presence of either PPD or PFSL. At the end of the culture period, we analyzed the surface phenotype, TCR expression and, in each subset, the proportion of cells in S and G2 + M phases of the cell cycle. As shown in a representative experiment (Fig 2) the stimulation of PBMC with PFSL, but not with SEB, induces an increased percentage of TCR76 versus TCRa/3 lymphocytes. To ensure that this increase corresponds to a real expansion of the TCR75 cells, we analyzed the TCR expression of the cycling cells at the day

363

of the maximum proliferation. As shown in Fig. 3 and Table 1 the great majority of cells in S and G 2 + M phases of cell cycle expressed TCR76. Furthermore, the TCR75 cells present all the characteristics of activated lymphocytes: (i) the forward scatter analysis shows that they correspond to enlarged blastic lymphocytes; (ii) they expressed the p55 and p75 chains of IL-2 receptor as determined by a double staining with TCR51 and anti-CD25 or anti- p75 (YAT1) antibodies (data not shown), and (iii) they expressed HLA-class II molecules as determined by a double staining with TCR51 and an anti-monomorphic class II antibody (13) (data not shown). The phenotype of the cells

fiim'Q •

u

--ii:" " " " I ' ' ' ' I '

Side Scatter

7 AAD fluorescence

WT31 (FITC)

TCR5 (FITC) 1

B 1-

Scat

u

U.

3

m

O

0:.-

-

1

Side Scatter

Side Scatter

|«'

""iB*

' i l *

TCR5 (FITC)

WT31 (HTC)

7 AAD fluorescence

TCR8 (FITC) 1

WT31 (FITC)

TCR5 (FITC)

l

Fig. 2. Analysis of TCR expression on T cells stimulated by PFSL compared to SEB of a representative donor. Cell cycle and expression of T cell receptor were analyzed on PBMC directly after isolation at day 0 (A), or after a culture period of 3 days in the presence of SEB (B) or after a period of 6 days in the presence of PFSL (C). The cells were doubte-stained with anfrCD3 [CD3{PE)] and anti-a^ [WT31(FITC)] or anti-^ [TCR61 (F1TC)] mAbs and analyzed without further treatment for the simultaneous analysis of the cell size (forward scatter) and the expression of different T cell receptors. For the cell cycle analysis, the same stainings were performed and the cells further treated with 7-AAD. For DNA histograms, 7-AAD fluorescence is analyzed on T cells [gate on positive cells for CD3(PE)] for (A), or on T cells expressing TCRa0 T (gate on double positive cells for CD3(PE) and WT31(FITQ] for (B)].

Downloaded from http://intimm.oxfordjournals.org/ at University of California, Santa Barbara on September 1, 2015

99-810.21

w &

364

yd T cell activation by P. falciparum

1

' ft taa

-

'

•

"

JB°

"

"

.

-

•

'l

•'>'"

' l »

a

* 1 B

3

TCR^FUO

.#

IB»-

»B'-

!•»IB

TIV«A(FITO

B

'|B>

'•»=

' | B

3

TTVIACFITO

Fig. 4. Characterization of VT and V6 chains of 76 T cells induced by PFSL. After 6 days of stimulation PBMC were stained simultaneously with anti-CD4 [CD4 (PE)], anti-CD8 [CD8 (PE)], and anti-76 [TCR61(FITC)] antibodies (A), and either wrth anti-76 (TCRS1) or anti-Vj2 (1H4) antibody with a second layer PE-labdled goat-anti-mouse antibody and anti-V79 [TTV^ (FITC)] or anti-Vs1 [TCSii(FITC)] antibodies (B-D).

and Va1 were performed on day 6. As shown in Fig. 4 B - D , all TCFfyS-positive cells are 1H4 + , TiVTA+, TCS 4 1" and subsequently belong to the VT9 Vj2 subset. This result was obtained with all donors studied. This selective outgrowth of VT9 Vj2 TCR T cells combined with the requirement for class II molecules suggests that the P. falciparum lysate does not stimulate T cells like a classical mitogen. However, the high level of proliferation observed in unexposed donors is difficult to explain by a conventional antigendriven response. The term superantigen was first proposed for exoproteins produced by bacteria such as Staphylococcus, Streptococcus and Mycoplasma which stimulate T cells bearing a particular Vfl element of the TCRa/3 (18-20). Nevertheless, staphylococcus enterotoxin A (SEA) also seems to act as a superantigen for T cells bearing VT9 (21,22). More recently retroviral gene products have been shown to be involved in the superantigen activity associated with the Mis locus in mice (23 - 25). No such activity has so far been described for parasites either in the mouse or human. The proliferate activity described in this report has most of the characteristics of that induced by superantigens. Here,

however, the proliferate response induced by P. falciparum lysate peaks at day 6, whereas bacterial superantigens induce the maximum response at day 3 or 4. This suggests that the P. falciparum lysate contains some products which act as a novel type of superantigen. This possibility has been already proposed for a GroEL-homologous molecule on Daudi cdls, which induces the same outgrowth of VT9 Vj2 T cells with the same retardation kinetics (26,27). Recently V79 Vj2 T cell clones were suggested to present two 'superspecificities', one of them being non MHC-restricted, and directed against Mycobacteria-pulsed antigen presenting cells (APC) (28). P. falciparum could induce a ligand on APCs related to that common to Mycobacterium tuberculosis and other microorganisms described by these authors. In our hands, however, the activation of V79 Vj2 T cells from normal peripheral blood lymphocytes appears to depend on MHC molecules. We are now investigating the biochemical characteristics of the activity and the mechanisms of induction of this VT9 Vj2 proliferate response. Acknowledgements We thank the blood donors. We thank Dr Th. Hercend and Dr F. Triebel for the kind gift of anti-76 antibodies. We also thank H. Jouin for helpful

Downloaded from http://intimm.oxfordjournals.org/ at University of California, Santa Barbara on September 1, 2015

1B»-

B-.

•.-.