plasmodium yoelii in mice: differential induction of cytokine gene expression during...

CELLULAR IMMUNOLOGY 160,79-90 (19%)

Plasmodium yoelii in Mice: Differential Induction of Cytokine Gene Expression during Hyporesponsiveness Induction and Restimulation’

BERNADETTE LUCAS, KENDALL SMITH,* AND AZIZUL HAQUE’

Unite’Mixte INSERM U 167, CNRS 624, Centre d’lmmunologie et de Biologie Parasitaire, Znstitut Pasteur, 59019 Lille, France; and *Division of Allergy and Immunology, Cornell University Medical College, New York, New York 10021

Received September 1,1994; accepted September 19,1994

Acute Plasmodium yoelii murine malaria is associated with a marked depression of splenic T cell responses. The present study was undertaken to address the question if a defect in T cell proliferation results from a relative increase of a non-T cell population in the spleen or real biological changes occurring in T cells of the spleen after infection. When animals were acutely infected, the splenic cells responded poorly to cross-linked anti-CD3 mAb, Con A, and PWM stimulation. At this stage, a very limited array of cytokine was expressed. We failed to detect the transcripts for IL-2R ~55, IL-2, IL-6, IL- 10, and IFN--, in mice with acute P. yoelii malaria irrespective of the number of splenocytes subjected to RT-PCR. In contrast, late in the infection when mice cleared the parasites and became resistant to reinfection, mRNAs for the above cytokines as well as for IL-4, IL-5, GM-CSF, and TNF-a were detectable. During this late phase of infection, lympho- cytes proliferated vigorously in response to TCR- and T cell mitogen-mediated stimulation. Surprisingly, during an early phase (as early as 3 days postinfection) with low parasitemia, before the establishment of T cell unresponsiveness, a broad array of cytokine expression including IL-2 and IFN-7 expression as well as marked lymphoproliferative response upon T cell mitogen- and TCR-mediated stimulation was observed. When the expression of cytokine gene in freshly isolated (ex uiuo) splenocytes from P. yoelii-infected animals was investigated, a similar pattern of cytokine profile was detected. We devised a methodology in which RNA from an increasing number of splenocytes (ranging from 1 to 16 million) was used to compensate for any difference in the frequency of splenic T cells between immune and acutely infected mice and to augment target molecules which could be measured simultaneously by PCR. The data presented in this study led us to speculate that “an-

1 This work was supported by CNRS, INSERM/Institut Pasteur de Lille.

* To whom correspondence should be addressed at Immunologie et G&&tique des Maladies Parasitaires, U INSERM 399, Faculti! de Mid- ecine, 27 Bd. Jean Moulin, 13385 Marseilles Cedex 05, France.

ergy” or relative increase of a non-T cell population cannot account solely for the T cell unresponsiveness in the acute phase of infection. We suggest that inactiva- tion or/and ablation of reactive T cells may explain T cell hyporesponsiveness during acute malaria. o 1995

Academic Press, Inc.

INTRODUCTION

The inoculation of Balb/c mice with RBC infected with 17XNL Plasmodium yoelii malarial parasites pro- duces an early phase of infection with a low parasitemia and then rapidly develops to an acute stage, and the en- suing late phase is associated with the elimination of parasites from the circulation. The processes behind the resolution of parasitemia can be assumed to reflect the generation of a protective immune response since this ordinarily self-limiting disease is fatal in B- or T-cell- deficient mice (l-3). Little is known about the immune responses elicited during the initial stage of this infection.

We and others have reported severe impaired cellular response to T cell mitogens and to parasite antigens during the acute phase of infection (4-6). Due to the spleno- megaly which takes place at this stage of infection (7, Lucas et al., submitted for publication), whether a defect in T cell proliferation reflects the relative increase of the non-T cell population or some important qualitative changes occur in T cells after the infection remains to be defined.

In this study, we have investigated some important as- pects of T cell biology (e.g., cytokine gene expression, and proliferative response to TCR3 and mitogenic stimulation) in order to determine the kinetics of changes

3 Abbreviations used: PCR, Polymerase chain reaction; mRNA, messenger RNA, IL, interleukin; IL-BR, interleukin 2 receptor; IFN- y, interferon y; APC, antigen-presenting cells; Con A, concanavalin A; PWM, pokeweed mitogen; TCR, T cell receptor; RT, reverse transcription; NO, nitric oxide.

79 000%8749/95 $6.00

Copyright 0 1995 by Academic Press, Inc.

All rights of reproduction in any form reserved.

80 LUCAS, SMITH, AND HAQUE

that occur in T cell during murine malaria. Our results demonstrated that the lymphocyte hyporesponsiveness to cross-linked anti-CD3 mAb, PWM, and Con A in acutely infected mice was associated with nondetection of transcripts for IL-2 and IFN-y by PCR. The gene expression for these two important cytokines was also not detectable in ex: viuo collected spleen cells from mice with acute P. yoelii malaria. It is of significance that a broad repertoire of cytokine is expressed during the late stage of infection when animals become resistant to reinfection. There was a good correlation between the expression of mRNAs for IL-2 and IFN-7 and the development of immunity to blood-stage parasites, which indi- cated the important effector role of these two cytokines in this malarial infection. Surprisingly, we detected an early primary T cell response manifested by the proliferation and expression of a whole panel of cytokines at Day 3 postinfection in mice before the establishment of unresponsiveness in acute phase of infection. To our knowledge, there was no previous report in malarial infections that directly posed the question of whether primed T cells in hyporesponsive state differ from the primed T cells in a responding condition qualitatively by their ability or inability to express multiple effector lymphokine genes.

MATERIALS AND METHODS

Mice

Female Balb/cJ mice 6-8 weeks old (Jackson Labora- tories, Bar Harbor, ME) were used for experimental infections. The mice were documented to be infection-free and were maintained under virus-free conditions in the Animal Research Facility, Dartmouth Medical School (Hanover, NH).

Parasite and Infection

The nonlethal strain of P. yoelii (17XNL) was obtained from Dr. Martha Sedegah (Malaria Program, Na- val Medical Research Institute, Bethesda, MD). In uiuo parasites were maintained by weekly passage in Balb/c mice. All the experiments were performed with animals exposed to parasites which had undergone three in viva passages. Parasitized RBC was suspended in RPM1 1640 and was adjusted to the desired concentration, and injected intraperitoneally into passage or experimental mice. A dose of lo6 PRBC was inoculated for passage mice as well as for experimental mice. The percentage of parasitemia was determined by blood smear examination as described previously (6). When parasites were not detectable at Day 25 or 30 postinfection by our blood smear examination, the mice were left for another 10 days to clear the residual parasites before exposure to a second infection.

Antibodies

The anti-CD3 mAbs was purchased from PharMingen (San Diego, CA). A goat antibody against hamster IgG used in immobilizing anti-CD3 mAb was obtained from Jackson Immunology Research Laboratories (West Grove, PA).

Antigen Preparations

P. yoelii antigen was prepared as described previously (8). Briefly, mice with parasitemia reaching its peak were bled under anesthesia. Blood was pooled in hepa- rinized tubes containing 50 mM PBS, pH 7.2. The blood was then passed through a column of glass beads to elim- inate white blood cells. The red cells were washed two times with PBS by centrifugation. The red cells were then lysed by treatment with 0.06% saponin (Merck, Germany) in PBS by incubating at room temperature for 10 min. The liberated parasites were washed twice with PBS and the pellet was frozen at -20°C. The parasites were then lysed through three cycles of freezing and thawing. This was followed by disruption with an ultra- sonic probe for 1 min (30 set X 2) at 80 W on ice (Sonics & Materials Inc., Danburry, CT). Nonsoluble material was then separated by centrifugation at 12,000 rpm for 30 min. The quantity of protein in soluble antigenic material was then determined with a kit (Bio-Rad, Rich- mond, CA).

In Vitro Stimulation of Splenocytes

The protocol of in vitro proliferation was reported elsewhere (6). Briefly, cell suspensions after perfusion of spleens were centrifuged at 350g for 10 min. RBCs were lysed with lysis buffer containing 155 mM NH&l, 10 mM KHCOB, 0.11 mM EDTA. The bulk cultures were set up in 24-wells plates (Nunc, NJ) with varying concentration of cells, in a final volume of 1.2 ml of complete medium which consisted of RPM1 1640,10% heat-inactivated fetal calf serum, Hepes (25 mM) buffer, L-gluta- mine (2 mM), 2-mercaptoethanol (1.10p5 M), and genta- micin (50 pg/ml). Cultures were also set up in 96-wells plates (2.5 X lo5 cells/well) in a final volume of 0.2 ml. The doses of mitogens or of parasite antigen and the du- ration of culture for optimal stimulation were predeter- mined in preliminary experiments. The cross-linking of anti-CD3 mAb was carried out in the following way: since anti-CD3 mAb was produced in hamster (Phar- Mingen) the culture plates were coated with an goat antibody anti-hamster IgG (Jackson Immunology Re- search Laboratories). The 24-well culture plates were coated with cold anti-hamster IgG antibody (14 pg of anti-hamster IgG antibody in 200 ~1 of PBS) and were incubated overnight at 4°C. The following morning, the antibody was removed and the wells were washed with cold PBS and then 5 pg of anti-CD3 mAb was added to

KINETIC CHANGES IN T CELL DURING MURINE MALARIA 81

the well in a volume of 200 ~1 of PBS. After incubating the culture plates at 37°C for 2 to 3 hr, the antibody was removed from the wells. The spleen cells were then added to the wells and incubated for 42 hr. The cells were carefully put into suspension and 100 ~1 of the suspension was transferred to each well of 96-well plates. Then, 0.5 &i of [methyZ--3H]thymidine (sp act, 74 GBq/mmol, 2 Ci/mmol, Amersham, UK) was added to each well and incubated for another 6 hr. The splenic cells from infected or control mice were incubated in the presence P. yoelii blood stage antigen (20 pg/ml) for 5 days. Pulsed cells were harvested with an automated multiple sample cell harvester. Incorporated radioactivity was measured in a liquid scintillation counter (LS-1800 Beckman Scintillation Counter).

RNA Extraction and Reverse Transcription (RT)

Extraction of RNA was performed by a modified guanidium thiocyanate method (9). Briefly, spleens were removed and single cell suspensions were prepared and counted (Coulter Electronics Inc., Hialeah, FL). A part of spleen cell population was also activated in the presence of different stimuli. A different number of freshly collected spleen cells or splenocytes that were cultured for 18 hr were lysed in 500 ~1 lysis solution, consisting of guanidium isothiocyanate, 4 M guanidium isothiocyanate (Fisher), 0.5% N-lauroylsarcosine (Sigma Chemi- cal Co., St. Louis, MO), 25 mM sodium citrate, pH 7.0, 0.1 M 2-ME (Sigma). Lysates were prepared at least in duplicate, vortexed, and stored at -70°C until RNA extraction. After thawing, 50 ~1 of 2 M Na acetate (pH 4.0), 500 ~1 of water-saturated acid phenol (GIBCO/BRL), and 100 ~1 of chloroform:isoamyl alcohol (49:l) were added to the lysates with thorough vortexing after each addition. The mixture was then chilled on ice for 15 min and spun at 10,OOOg for 15 min at 4°C. The acqueous phase was obtained and RNA was precipitated in an equal volume of 2-propanol at -70°C for at least 3 hr. Precipitates were pelleted at 4°C washed once with 75% ethanol in diethylpyrocarbonate-treated double-dis- tilled water (DEPC-ddH20) and repelleted at 4°C at 1OOOg for 15 min. After drying in vacuum, the precipitate was dissolved in 10 ~1 DEPC-ddH20 containing 2 pg oli- go(dT) (12-18 mer; United States Biochemical Corp., OH) and incubated at 70°C for 10 min. After cooling on ice, the mixture was incubated with 10 ~1 of solution containing 4 ~1 of 5 X RT buffer (375 mM KCl, 250 mM Tris/HCl, pH 8.3, 15 mM MgC12), 2 ~1 DTT (0.1 M, GIBCO/BRL), 2 ~1 dNTP (10 mM, United States Bio- chemical Corp.), 0.5 ~1 RNasin (2000 U/40 ~1, Boeh- ringer Mannheim Biochemicals, Indianapolis, IN), and 0.5 ~1 of acetylated BSA (Sigma). The samples were then vortexed, briefly centrifuged, and incubated for 2 min at 37°C. Then, lpl(200 U) of MMLV reverse transcriptase (GIBCO/BRL) was added and the reaction mix was in-

cubated at 37°C for 1 hr. To stop the enzyme activity, the samples were then heated to 95°C for 5-10 min, and 50 ~1 dHzO was added to the 20 ~1 reaction mixture. The cDNA samples were stored at 4°C until further use.

Polymerase Chain Reaction (PCR)

Cytokine-specific primer pairs were designed and assessed for specificity by computer-assisted search of up- dated versions of GenBank. The 5’ and 3’ primers are complementary to sequences in different exones or spanned exon-exon junctions and are therefore mRNA/ cDNA specific. The 5’ primers were designed to all recognize sequences approximately 1000 bp upstream of the poly(T) tract of the cDNA to minimize potential errors caused by differential reverse transcription. Cytokine- specific primers, IL-2R p55 primer, and P-actin primer pairs were synthesized on a Cyclone DNA synthesizer (Bioresearch, San Rafael, CA), and purified on Nensorb Prep columns (DuPont Co., Wilmington, DE). The DNA synthesizer facility was provided by Dr. Tom Ciardelli (Dept. of Toxicology, Dartmouth Medical School, Hano- ver, NH). These primers presented in Table 1 were found to be the same as described previously (10). Five microli- ters of cDNA prepared from varying number of splenocytes was amplified in 0.5 ml GeneAmp reaction tubes (Cetus Corp., Emeryville, CA) in the presence of 2 PM final concentration of 5’ and 3’ primers, 200 WM dNTP, 0.5 U of Taq polymerase (Perkin-Elmer), and PCR buffer containing 2.5 mM MgC12, 50 mM KCl, 10 mM Tris-Cl, and 0.001% gelatin in a final volume of 25 ~1. The reaction mixture was overlaid with a drop of light mineral oil, and PCR was performed in a DNA thermal cycler (Perkin- Elmer-Cetus) for 30 cycles. Each cycle consisted of dena- turation at 94°C for 1 min, annealing at 60°C for 1 min 30 set, and extension at 72°C for 1 min 30 sec. The terminal extension was allowed to go for 7 min. Concentrations of primers and numbers of cycles had been optimized for the whole panel of cytokines tested. The PCR amplification products obtained were analyzed by 2% agarose gel (Sigma) electrophoresis in Tris acetate/EDTA buffer, visualized with ethidium bromide staining under uv transillumination, and photographed. One microgram of a HaeIIIdigested $X174 DNA was run in parallel as mo- lecular weight markers (bands at 1353, 1078, 872, 603, 310, 281, and 234 bp) (Clontech, Palo Alto, CA). The specificity of the amplified bands was validated by their predicted size and restriction enzyme digests giving ap- propriately sized fragments. PCR-assisted mRNA amplification was repeated at least twice for at least three sep- arately prepared cDNA samples.

Preliminary PCR experiments were performed with the cells cultured for 6, 18, and 24 hr to determine the optimum time period for the expression of cytokines tested in the present study. In general, the 18-hr time point was observed to be the optimum when splenic cells


TABLE 1 Oligonucleotide Primer Sequences Used for PCR-Assisted mRNA Amplification

@-Actin (345 bp)” Sense 5’.TGG-AAT-CCT-GTG-GCA-TCC-ATG-AAA-C-3 Antisense 5’-AAC-GCA-GCT-CAG-TAA-CAG-TCC-GCC-TA-3’

IL-2Rp55 (501 bp) Sense 5’-TCC-TGG-AGC-AAC-TGC-CAG-TGC-ACC-AGC-A-3 Antisense 5’-CTT-ATA-CTC-CAT-TGT-GAG-CAC-AAA-TGT-CT-3’

IL-2 (320 bp) Sense 5’-ATG-TAC-AGG-ATG-CAG-CTC-GCA-TCC-TGT-GTC-A-3’ Antisense 5’-AGT-CAA-ATC-CAG-AAC-ATG-CCG-CAG-AGG-TCC-A-3’

IL-3 (592 bp) Sense 5’-ATG-GTT-CTT-GCC-AGC-TCT-ACC-ACC-A-3’ Antisense 5’-GAT-AAG-ACA-TTT-GAT-GGC-ATA-AAG-GA-3’

IL-4 (351 bp) Sense 5’-ACA-AAA-ATC-ACT-TGA-GAG-AGA-TCA-T-3 Antisense 5’-AGT-AAT-CCA-TTT-GCA-TGA-TGC-TCT-T-3’

IL-5 (242 bp) Sense 5’-ATG-ACT-GTG-CCT-CTG-TGC-CTG-GAG-C-3” Antisense 5’-CTG-TTT-TTC-CTG-GAG-TAA-ACT-GGG-G-3’

IL-6 (600 bp) Sense 5’-CTG-GTG-ACA-ACC-ACG-GCC-TTC-CCT-A-3 Antisense 5’-ATG-CTT-AGG-CAT-AAC-GCA-CTA-GGT-T-3

IL-10 (237 bp) Sense 5’-ACC-TGG-TAG-AAG-TGA-TGC-CC-AGG-CA-3’ Antisense 5’-CTA-TGC-AGT-TGA-TGA-AGA-TGT-CAA-A-3

INF-7 (366 bp) Sense 5’-GAA-AGC-CTA-GAA-AGT-CTG-AAT-AAC-T-3 Antisense 5’-ATC-AGC-AGC-GAC-TCC-TTT-TCC-GCT-T-3’

GM-CSF (435 bp) Sense 5’-ATG-TGC-CTG-CAG-AAT-TTA-CTT-TTC-CT-3 Antisense 5’-TGG-GCT-TCC-TCA-TTT-TTG-GCC-TGG-T-3

TNF-ol (446 bp) Sense 5’-AGC-CCA-CGT-CGT-AGC-AAA-CCA-CCA-A-3 Antisense 5’-ACA-CCC-ATT-CCC-TTC-ACA-GAG-CAA-T-3’

’ Sizes of amplified fragments are shown in parentheses.

were stimulated with immobilized anti-CD3 mAb, Con A, or PWM. In the case of antigen, a 48-hr culture period was found to be sufficient for simultaneous detection of transcripts of cytokines tested.

Statistical Analysis

Data were analyzed for differences by the Student t test. Probability significance was determined by two- tailed Student’s t test assuming equal variances. Statis- tical significance was set at P < 0.05 for all comparisons. The interassay variations were within 15%.

RESULTS

TCR-Mediated or Mitogen-Driven Proliferative Response of Splenocytes Was Markedly Diminished in Mice with Acute P. yoelii Malaria

The lymphoproliferative response to cross-linked anti-CD3 mAb and to T cell mitogens (e.g., Con A,

PWM) was evaluated in mice at different time periods after primary infection, and in animals given a challenge infection following elimination of parasites from the circulation. In the early phase of infection, i.e., at Day 3 postinfection, lymphoproliferative response toward cross-linked anti-CD3 mAb or to the mitogens used (PWM, Con A) was at the level demonstrated by normal spleen cells. On Day 3 postinfection, the parasitemia ranged between 1 and 2.8%. The lymphoproliferative response remained almost at the same level until Day 6 postinfection. As shown in Table 2, the proliferative response of splenocytes reached its nadir in animals on Day 12 postinfection when parasitemia was reaching its peak. When the parasites were cleared from the circulation at Day 30 postinfection, the lymphoproliferative response returned to the normal level. The animals cured from the primary infection were challenged with a secondary infection. These animals did not show the presence of parasites in the peripheral blood after reinfection. The lymphoproliferative response in these animals

KINETIC CHANGES IN T CELL DURING MURINE MALARIA

TABLE 2 Lymphocyte Proliferation to Different Stimuli during the Course of Primary and Secondary P. yoelii Infection

83

Proliferation cpm (mean + SEM) Stimulator

Animals Parasitemia (%) Anti-CD3 Con A PWM Medium

After first infection 3 days 6 days 12 days 30 days

After second infection

12 days

Controls (noninfected)

-2 65,146 + 259 57,528 + 706 51,201 f 424 3,928 + 187 -8 81,525 f 1403 127,866 f 1059 30,647 f 782 4,267 + 187

-28 5,378 + 432 7,219 + 297 2,296 f 77 1,627 f 523 -0 105,619 f 5325 165,591 f 6919 74,828 f 3460 3,006 f 202

-0 95,389 f 2963 91,601 !z 634 73,289 + 11,699 2,812 f 205

None 68,502 rfr 465 92,612 f 3531 52,029 + 1766 1,977 f 67

Note. Proliferation of splenocytes was obtained from animals at different time periods after primary and secondary infection. Spleen cells were stimulated with cross-linked anti-CD3 mAb, with Con A (5 @g/ml) or with PWM (2.5 ag/ml). [3H]TdR incorporation was assessed 48 hr after activation during the last 6 hr of culture. The level of parasitemia was determined at each time period when lymphocyte proliferative response was assessed. This experiment was repeated five times.

at Day 12 after secondary infection toward cross-linked anti-CD3 mAb, PWM, or Con A was at least comparable to or better than that in normal mice (Table 2).

Expression of Cytokine Genes in Splenic Cells from Mice Acutely Infected with or Immune to P. yoelii after in Vitro Stimulation with Cross-Linked Anti-CD3 mAb, Con A, or P WM

RNAs were prepared from 4 million cells after 18 hr culture with cross-linked anti-CD3 mAb or with mitogens such as Con A or PWM. When normal mice splenocytes were stimulated by cross-linked anti-CD3 mAb (Fig. la) or PWM (Fig. lc), mRNAs for IL-2, IL-2R ~55, IL-4, IL-5, IL-6, IL-lo, IFN-7, GM-CSF, and TNF-a (faintly) were expressed. The message for IL-3 was not detectable in PWM- or in anti-CD3 mAb-stimulated normal cells. In the case of Con A-stimulated normal cells (Fig. lb), the transcripts for IL-2R ~55, IL-3, and IL-10 were not detected. However, genes for the rest of the cytokines tested were expressed. When splenic cells were obtained from the mice at an early phase of infection, i.e., at Day 3 after primary infection, and were stimulated with these three stimulators, an almost identical cytokine expression profile as of normal mice was observed. However, the transcripts for IL-2 and IL-2R p55 were not demonstrable in the case of anti-CD3 mAb stimulation, whereas the message for both of them were detectable in the splenic cells stimulated with PWM. In the case of Con A stimulation, the transcript for IL-2 but not for IL-2R p55 was detectable (Fig. lb). At Day 3 postinfection, the message for IL-3 was undetectable whether the cells were stimulated with anti-CD3 mAb, PWM, or Con A. When splenocytes were obtained from mice at Day 6 postinfection and activated with Con A, the genes for IL-2, IL-4, IL-5, IL-6 (faintly), IL-lo, IFN-

y, and GM-CSF were expressed. The mRNA for IL-2 was not detectable in the case of anti-CD3 mAb or PWM stimulation. On Day 6 postinfection, the genes for IL- 6 and IL-10 were expressed in anti-CD3 mAb-activated splenic cells but not in PWM-stimulated cells. At Day 12 when parasitemia was approaching its peak, mRNAs for IL-2R ~55, IL-2, IL-3, IL-6, IL-lo, IFN-7, and TNF- (Y were not detectable in cross-linked anti-CD3 mAb- stimulated splenocytes. On Day 12 postinfection, genes for IL-2R ~55, IL-2, IL-6, and IL-10 were not demonstrable in PWM-stimulated splenocytes. The band seen with the primers for IL-3 in the splenic cells from mice infected for 12 days did not correspond to the predicted size. Also mRNAs for IL-2R ~55, IL-2, IL-3, IL-6, IL- 10, and IFN-y were not detectable in Con A-stimulated spleen cells. The splenocytes from mice infected for 12 days demonstrated the presence of message for TNF-a after Con A or PWM stimulation but not after anti-CD3 mAb activation. At Day 30 when animals became cured after primary infection, mRNAs for IL-2R ~55 and all the cytokines tested except IL-3 were detected in cross- linked anti-CD3 mAb-stimulated splenocytes. On Day 30 postinfection, in Con A-stimulated splenocytes, the transcripts for IL-2R ~55 and all the cytokines tested were detectable except IL-3 and TNF-ol. When splenic cells from the same animals were stimulated with PWM, genes for IL-2, IL-4, IL-5, IL-6, IFN-7, and GM-CSF were expressed but not for IL-2R ~55, IL-3, IL-lo, and TNF-a. Splenic cells were obtained from P. yoelii-immune mice at Day 12 after challenge infection and were analyzed by PCR for the messages of IL-2R ~55 and of different cytokines. When these cells were activated with cross-linked anti-CD3 mAb, mRNAs for IL-2R ~55, IL-2, IL-4, IL-5, IL-6, IL-lo, IFN-7, GM-CSF, and TNF-cu were expressed. At Day 12 after secondary infec-

a

LUCAS, SMITH, AND HAQUE

FIG. 1. Cytokine gene expression detected by PCR-assisted mRNA amplification in spleen cells during the course of I? yoelii infection. Splenocytes were activated with cross-linked anti-CD3 mAb (a), Con A (b), or PWM (c). Splenocytes were obtained from noninfected mice (A), from mice primarily infected for 3 days (B), 6 days (C), 12 days (D), and 30 days (E), and from animals at Day 12 (F) after secondary infection. Splenic cells (4 X 10’) were stimulated with immobilized anti-CD3 mAb or mitogens for 18 hr in bulk cultures.

tion, transcripts for IL-2R p55 and all the cytokines tested (including IL-3) were detectable in Con A-stimulated immune cells. Genes for IL-2R p55 and all the cytokines (IL-10 being faintly expressed) except for IL-3 were expressed in PWM-stimulated splenocytes. These results are presented in Figs. la-lc.

Expression of Cytokine Genes in ex Vivo Collected Splenocytes from Mice after Primary and Secondary Infection with P. yoelii

Total RNA was prepared from 8 million spleen cells which were collected from the mice immediately after sacrifice at different days after infection. Some cytokines were expressed in normal mice splenocytes-these are IL-5, TNF-a, and GM-CSF. mRNAs for IL-2R ~55, IL-2, IL-5, IL-6, IL-lo, IFN--/, TNF-a, and GM-CSF were expressed in splenic cells from mice which were infected for 3 days. IL-4 message was very faintly expressed in these splenic cells. On Day 3 postinfection, the parasitemia varied between 1 and 2.8%. Transcripts for IL-4, IL-5, IFN-7, GM-CSF, and TNF-(Y were detectable in the splenocytes from 6 days infected mice.

The messages for IL-2 and IL-10 were faintly expressed in these splenic cells, and the bands seen with primers for IL-2R ~55, IL-3, and IL-6, respectively, were not of the predicted size. At Day 6 after primary infection, the parasitemia ranged between 6 and 8.5%. Genes for IL-2 and IFN-7 were not expressed in splenocytes from mice which were infected for 12 days and displayed an ascend- ing parasitemia varying between 22 and 32%. However, messages for IL-4, GM-CSF, and TNF-a were expressed in splenic cells isolated from mice on Day 12 postinfection. The gene for IL-5 was faintly expressed in these splenocytes. At Day 30 after primary infection when the parasites disappeared from the circulation from P. yoelii-infected mice, the expression for IL-2 and IFN-7 was again detected in the splenic cells. Transcripts for IL-4, IL-5, GM-CSF, and TNF-a were also observed in the splenocytes of these mice. On Day 30 postinfection, the message for IL-2 was very faintly expressed in splenic cells, and the band that appeared with primers for IL-2R p55 did not correspond to the predicted size. Mice after recovering from primary infection became resistant to reinfection. The expression of different cytokines was


F

G

H

I

J

K

FIG. 2. PCR-assisted mRNA amplification in 8 million splenocytes from naive animals (A), and from mice infected with P. yoelii for 3 days (B), 6 days (Cl, 12 days (D), 30 days (El; from animals at 1 hr (F), 3 hr (Gl, 6 hr (H), 12 hr (I), 24 hr (J), and at Day 12 (K) after exposure to a secondary infection. The splenic cells were freshly obtained and were not subjected to in vitro stimulation. Balb/c mice were infected with 1 X

10s P. yoelii blood-stage parasites and were challenged with the same number of parasites at Day 40 after primary infection.

examined in freshly obtained splenocytes from mice at different time periods following secondary infection. mRNAs for IL-5, IFN-7, GM-CSF, and TNF-a were expressed in the splenocytes isolated from mice at 1, 3, 6, 12, or 24 hr or at Day 12 after reinfection. The transcript for IL-2 was detectable in immune cells obtained at 24 hr or at 12 days after secondary infection. The bands observed after the amplification with the primers for IL- 2R p55 at 1,3,6, and 12 hr after reinfection did not correspond to the predicted size. The expression of mRNAs for IL-4 and IL-6 was erratic, the former was expressed in splenic cells at 3, 6, 12, and 24 hr after challenge infection whereas the latter being expressed at 1 (faintly), 6, and 12 hr (faintly). These results are shown in the Fig. 2.

Expression of Cytokine Genes in Different Number of Splenocytes from Acutely Infected or from P. yoelii- Immune Mice after Stimulation with Con A

Whatever the number of cells stimulated with Con A, there was always expression of more cytokines in normal

and in immune cells than in acutely infected mice. When only 1 million cells were analyzed, the gene expression was already observed in the immune animals for IFN--r whose mRNA was not detectable in acutely infected mice at this cell concentration, or even in the case when the cell number was increased 16-fold (Fig. 3). When 4 million cells were activated with Con A, genes for both IL-2 and IFN-y were expressed in splenic cells from immune and normal animals. Extraction of RNA from splenic cells, ranging in number from 1 to 16 million, from the mice with acute P. yoelii malaria did not result in the expression of IL-2. When the cell number was aug- mented to 16 million, fewer cytokines were expressed in all cases indicating the threshold of detection of some of the cytokines mRNAs by PCR exceeded. mRNAs for IL- 2R ~55, IL-2, IL-4, IL-5, IL-6, IL-lo, IFN--y, GM-CSF, and TNF-cu were detected when 4 million splenocytes from immune animals were used, whereas mRNAs for only IL-4, IL-5, IFN--y, and TNF-a! were expressed when 16 million cells from the same animals were analyzed. These results are presented in Fig. 3.


1x106

2x108

4x106

FIG. 3. Cytokine gene expression detected by PCR-assisted mRNA amplification of different number of splenocytes from normal (A), acutely P. yoelii-infected (B), and immune (C) mice. Spleen cells were stimulated with Con A (5 +g/ml) in bulk culture for 18 hr.

Expression of Cytokine Genes in Different Number of Splenocytes from Acutely Infected or from P. yoelii- Immune Mice after Stimulation with P. yoelii Antigens

Three concentrations of cells, i.e., 4, 8, and 16 million cells, were activated with the parasite antigens for 48 hr. When samples from 4 million spleen cells were used for RNA extraction, the genes for IL-2R ~55, IL-2, IL-4, IFN-7, and GM-CSF were expressed in the splenic cells from immune mice, whereas mRNAs for only IL-5 (faintly) and IL-6 were detected in the cells from mice with acute P. yoelii malaria. When samples from 8 million splenic cells were analyzed, the transcripts for IL- 2R ~55 and all the cytokines tested except for IL-3 were detected in the splenocytes from immune animals. In contrast, mRNAs for only IL-5, GM-CSF, and TNF-a were detectable in the splenocytes from acutely infected mice when the same cell concentration was used. With the cells from age-matched normal mice at the same concentration, genes for IL-5 (faintly), IL-6, IFN-7, GM- CSF, and TNF-a were expressed. When 16 million spleen cells were activated with P. yoelii antigens, the mRNAs for IL-2R ~55, IL-2, IL-3 (faintly), IL-4, IL-5, IL-6, IL-lo, IFN-7, GM-CSF, and TNF-a were demon-

strable in the immune mice, whereas transcripts for only IL-5, GM-CSF, and TNF-a were detected in the mice with acute P. yoelii malaria. Interestingly, when the samples from the same number of splenocytes from normal mice were stimulated with the parasite antigens, the genes for IL-2, IL-4, IL-5, IL-6, IL-lo, IFN-7, GM-CSF, and TNF-a were expressed. These results are shown in the Fig. 4.

Expression of Cytokine Genes in Different Concentration of ex Vivo Collected Splenocytes from Acutely Infected and P. yoelii-Immune Mice

Spleen cells were collected immediately after sacrific- ing of normal, acutely infected, and immune mice. When samples from 4 million cells were analyzed, the genes for IFN-7 and GM-CSF were detected in the immune mice, whereas mRNAs for IL-5 and TNF-7 were expressed in the splenic cells from acutely infected mice. When the specific primers for IL-PR ~55 was used, the band seen with the cells from acutely infected mice was not of the predicted size. The mRNA for IFN-7 was not demonstrable when 8 million splenic cells from normal or acutely infected animals were analyzed. mRNAs for IFN-7 and IL-2 (faintly) were detected in the cells from


8x

FIG. 4. Cytokine gene expression detected by PCR-assisted mRNA amplification of different number splenocytes from normal (A), acutely P. yoelii-infected (B), and immune (C) mice. Spleen cells were stimulated with P. yoelii antigens (20 pg/ml) in bulk culture for 48 hr.

immune mice when samples from 16 million cells were analyzed. The transcripts for IL-2 and IFN-y were not detected in the cells from acutely infected mice at the same cell concentration. However, transcript for IFN-7 but not for IL-2 was expressed in normal splenocytes at this cell concentration (Fig. 5).

DISCUSSION

For many years, mechanisms underlying development of immunodepression in infectious diseases are poorly

4x108

8x108

A

understood and biological relevance of this phenomenon was questioned because in many instances the pathogen is ultimately controlled by the immune system of the host. But recent progress in this field has shown that many of the infectious organisms are indeed able to ma- nipulate and subvert the immune system of hosts (11).

Our results clearly demonstrate that a period of immunodepression occurs in P. yoelii infection accompanied with a high parasitemia, which probably ensures successful continuation of cycle (transmission by mos- quitoes). We have detected an early phase of primary

16x106

FIG. 5. Cytokine gene expression detected by PCR-assisted mRNA amplification of different number of el: uiuo collected spleen cells from normal (A), acutely I? yoelii-infected (B), and immune (C) mice. These splenocytes were freshly collected and were not subjected to any activation.


response which was marked by lymphocyte reactivity toward cross-linked anti-CD3 mAb, Con A, or to PWM. This initial phase is short-lived emerging during the period from 3 to 6 days postinfection with low parasitemia. The infection then progressed to an acute stage (parasitemia reaching its peak) by Days 10 to 12. At this stage, lymphoproliferative response to these stimuli was markedly depressed. Between Days 25 and 30 postinfection when parasites in the peripheral blood were cleared, a second phase of cell-mediated response was identified in P. yoelii-infected mice, which was more pronounced in animals with secondary (challenged) infection. Data from a separate study (Lucas et al, submitted for publication) demonstrated that the level of antigen reactivity which was markedly high in the early phase of infection reached its nadir during acute stage of infection. Taken together these data indicate that the memory T cells generated early in the infection become inactivated and reverted to nonresponsive ones. Such refractoriness of reactive T cells can indeed occur after primary response, and this interesting observation was recently reported by Benacerraf and his colleagues in ovalbumin-induced tolerance (12).

It is now well established that many of the actions of T cells are dependent on the production by these cells of lymphokines, and yet the nature of cytokines elaborated during the immune response to malarial parasites is not clear. This fact has led us to investigate cytokine gene expression associated with responding and nonresponding periods in P. yoelii malaria. Our results demonstrated that primary response during early stage was associated with IL-2R and a broad repertoire of cytokines (IL-2, IL-4, IL-5, IL-6, IL-lo, IFN-7, GM-CSF, and TNF-a) expression after TCR-mediated or mitogenic stimulation. Similar cytokine profile was observed in the late phase of infection. In splenocytes from reinfected (challenged) mice, in addition to above cytokine transcripts, mRNA for IL-3 was expressed after Con A (Fig. lb) or antigen stimulation (Fig. 4). Most interestingly, during the acute phase of infection when lymphocyte proliferative response reached its nadir, transcripts for only IL-4, IL-5, GM-CSF, and TNF-(Y were detectable, whereas mRNAs for IL-2R ~55, IL-2, IL-3, IL-6, IL-lo, and IFN--, were not observed even at high cell numbers. The nondetectability of IL-2R p55 in the acutely infected animals was remarkable since antigen-experi- enced T cells express IL-2R ~55 (13). Although the mRNA for IFN-7 was not detectable consistently in the splenocytes from acutely infected mice when splenic cells were stimulated with Con A or parasite antigen, the transcript for IFN-7 was detectable in the cells stimulated with PWM. Several workers previously reported that PWM may induce lymphokine production in a somewhat different way than other mitogens (14, 15). When the expression of cytokine gene in ex vivo collected splenocytes from responding and nonresponding

phases of infection was analyzed, an almost identical pattern of cytokine profile was observed as in the case of in vitro stimulation.

The results of this study did not show a coordinated Thl or Th2 type of response in this murine model of malaria, although a sequential Thl and Th2 response was described in Plasmodium chabaudi chabaudi murine model (16). The response operating in the early phase of infection, and in the cured or reinfected mice, did not fall into a Thl response which complied more with the ThO response (17, 18). During the period of acute infection accompanied by striking lymphocyte hyporesponsiveness, uniform nondetectability of transcripts for IL-2 and IFN-7 and the presence of IL-4 expression present more of a Th2 type response. In addition, during this stage of infection, the mRNAs for other cytokines such as IL-6 and IL-lo, which might be potentially produced both by T cells and macrophages, were also not detectable. Apparently, the expression of other important cytokines such as TNF-a, IL-4, IL-5, and GM-CSF was not sufficient to effect immunity against this parasite. However, they may serve as well-organized redundancy of control mechanisms in the cytokine system (19, 20). It is of note that mRNAs for IL-2 and IFN--/ were detectable in splenic cells from normal mice only when 16 million spleen cells were stimulated with parasite antigen. It may be possible that the P. yoelii antigenic prep- aration contains determinants which cross-reacts with environmental antigens to which noninfected normal mice are exposed. A level of proliferative response by normal splenocytes in the presence of parasite antigen was observed, however, only when higher concentrations of cells were assayed (Lucas et al., submitted for publication) .

It is of significance that mRNAs for IL-2 and IFN-7 upon T cell mitogen- or TCR-mediated stimulation were seen in splenocytes from P. yoelii-infected mice before the establishment of cellular hyporesponsiveness. These data led us to speculate that anergy cannot solely account for T cell unresponsiveness in acute phase of infection. APCs following binding of TCR to the Ag/MHC might provide costimulatory signals (21) since whole splenic cell population was used in our assay system. However, if this mechanism was the sole mechanism of the T cell unresponsiveness associated with lack of detection of IL-2 mRNA, a transient early T cell primary response should have been absent. In addition, we have used immobilized anti-CD3 mAb for stimulation, which should circumvent the requirement by APCs (22); thus, the impairment observed indicates more to defect in TCR function than in the costimulatory signaling path- way provided by APCs. This supports the recently described notion that the activated T cells need only a TCR stimulus to produce IL-2, whereas naive T cells re- quire at least two signals (23).

We have not yet determined the number of IL-2- or


IFN--, producing cells in nonimmune vs immune mice. Expressions of these two cytokines were observed in the 4 million Con A-stimulated cells from mice infected for 3 days, whereas their expressions were undetectable with the same or an increased number of cells from the animals that progressed to an acute stage by Day 12 postinfection. These data led us to conclude that either some of the IL-2- and/or IFN-y-producing cells were ab- lated or they became refractory by abandoning some or all their functional activities.

Another interesting aspect of the present study is the demonstration of correlation between the expression of IL-2 and IFN--r and the development of immunity toward blood stages in this murine malaria. IL-2 is important for clonal expansion of antigen-reactive T cells, generating natural killer cells, for induction of other cytokines and also for production of certain isotype of antibody (24,25). IFN-7 has been implicated as a mediator of protection against both blood and liver stages of Plm- nodium infections (26-28). It appears that T cells medi- ate blood stage immunity by the production of IFN-y which can activate macrophages to produce metabolites toxic for intraerythrocytic forms (29, 30). This process almost certainly occurs in the spleen (31). Also, striking in uiuo effects are induced by the injection or depletion of IFN-7 (32,33). IFN--y may also play a role in the elab- oration of IgG2a antibody (34) which may be implicated in the resolution of P. yoelii primary infection (35). Our results showed that expression of IL-4 was associated with nondetection of mRNAs for IL-2 and IFN--, in acute P. yoelii infection. We do not yet know if IL-4 is implicated in the regulation of expression of IL-2 and IFN-7 in this infection. However, it has previously been reported that IL-4 can downregulate the expression of IL-2 and IFN-7 (36-38). Transcripts of IL-6 and IL-10 were detectable in the spleen cells from immune animals but not in the splenocytes from acutely infected mice. The antiparasitic role of IL-6 on the hepatic stage in malaria has been suggested (39). The functions of IL-10 in malarial infections has not been defined, although cross- regulation of this cytokine has been described (40). mRNA for TNF-a was frequently detected in ex viva or in vitro stimulated splenocytes during acute phase of P. yoelii infection. Some of the pathological manifestations such as fever and cerebral lesions may result from over- production of TNF (41, 42). TNF has also been implicated in blood-stage immunity (43).

We have in a separate report described sevenfold less T cells (Thy 1.2-positive cells) in a given splenic cell population from acutely infected mice (Lucas et al., submitted for publication). For this reason, in this study the cytokine gene expression was evaluated in splenocytes ranging from 1 to 16 million which should largely cover the difference. This increase in the number of splenocytes has allowed to increase number of target molecules which could be measured simultaneously (Figs. 3 and 4).

Indeed, in our experimental system, the product DNA signal is averaged for the number of input cells that are lysed, as has recently been shown with HIV-infected heterogeneous cell population (44). It should be noted that the PCR results presented in this study are princi- pally qualitative. We recognize that the results presented do not prove that the respective cytokine proteins were released from spleen cells during P. yoelii infection in Go, but the good correlation between in vitro and in viuo gene expression of cytokines (45, 46) points to the likelihood of presence of these cytokines in the most important site of malarial infection.

In the present study we have used unseparated cells rather than purified T cells because the costimulatory function of APCs (macrophages and B cells) or the in- hibitory role of some of the metabolites from macrophages (IL-10 or NOB) was less likely to be compromised in our assay system. Use of total splenocytes has allowed us to study the expression of a number of cytokines simultaneously, both of T cell and monocyte origin. More- over, the effects with the unsorted cells would be closer to the events occurring in the physiologic milieu. In view of the fact that the spleen, where malarial parasites filter through a network of T cells as well as monocyte macrophages, plays a significant role in mediating resistance to malaria infection (47,48), we focused on the response of splenocytes in regard to proliferation and cytokine expression.

We believe that impairment in cytokine gene expression and diminished T cell proliferation are not solely related to lower number of T cells in splenic population because these defects still persist even though the cell number was sufficiently increased to compensate the difference. In addition, lower concentration of cells, obtained from acutely infected animals injected with rIL- 2 or when nonresponsive cells tested in the presence of exogenous rIL-2, resulted in the significant restoration of proliferation as well as expression of a whole panel of cytokines (Lucas et al., submitted for publication).

Cytokine response in an infection situation would probably be occurring in a cascade, and also the redundancy mechanism would be operating (19,20). Determi- nation of preeminent role of a cytokine(s) in a given infection would only be possible if we study the response of a number of cytokines simultaneously. Our experimental approach of analysis of a variety of cytokines simultaneously in varying number of cells, both of T cell and monocyte origin during various stages of murine malaria sheds insight as to which of the cytokines is related to hyporesponsiveness or to protection.

ACKNOWLEDGMENTS

The authors thank Dr. Tom Ciardelli for his help in synthesizing cytokine-specific primers, and Dr. Carol Beading for discussion.

LUCAS, SMITH, AND HAQUE 90

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

REFERENCES

Weinbaum, F. I., Evans, C. B., and Tigelaar, R. E., J. Immunol. 117,1999,1976.

Jayawardena, A. N., Targett, G. A. T., Carter, R. L., Leuchars, E., and Davis, A. J., Immunology 32,849,1977.

Grun, J. L., and Weidanz, W. P., Infect. Zmmun. 41, 1197, 1983.

Weinbaum, F. I., Weintraub, J. F., Nkrumah, N., Evans, C. B., Tifelaar, R. E., and Rosenberg, Y. H., J. Zmmunol. 121, 629, 1978.

Lelchuk, R., Rose, G., and Playfair, J. H. L., Cell. Zmmunol. 84, 253,1984.

Lucas, B., Engel, A., Camus, D., and Haque, A., Cell. Zmmunol. 150,59,1993.

Weiss, L., Johnson, J., and Weidanz, W. Am. J. Trop. Med. Hyg. 8,135,1989.

Lucas, B., Engels, A., Camus, D., and Haque, A., Infect. Zmmun. 61,4863,1993.

Chomczynski, P., and Sacchi, N., Anal. Biochem. 162,156,1987.

Murray, L. J., Lee, R., and Martenes, C., Eur. J. Zmmunol. 20, 163,199O.

Marrack, P., and Kappler, J., Cell 76,323,1994.

Vidard, L., Colarusso, L. J., and Benacerraf, B., Proc. Natl. Acad. Sci. USA 91,5627,1994.

Grant, A. J., Roessler, E., Ju, G., Tsudo, M., Sugamura, K., and Waldman, T. A., Proc. Natl. Acad. Sci. USA 59,2165,1992.

Sander, B., Cardell, S., Moller, G., and Moller, E., Eur. J. Zmmu- nol. 21,1887,1991.

Butch, A. W., Chung, G-H., Hoffmann, J. W., and Nahm, M. H., J. Zmmunol. 150,39,1993.

Langhorne, J., Meding, S. J., Eichman, K., and Gillard, S., Zm- munol. Reu. 112,71,1989.

Swain, S. L., Weinberg, A. D., and English, M., J. Zmmunol. 144, 1788,199o.

Street, N. E., Schumacher, J. H., Fong, T. A., Bass, H., Fiorenti- ono, D. F., Leverah, J. A., and Mosmann, T. R., J. Zmmunol. 144,1629,1990.

Paul, W. E., Cell 57,521,1989.

Kroemer, G., de Alboran, I., Gonzalo, J. A., and Martinez, A. C., Crit. Rev. Zmmunol. 13, 163,1993.

Ding, L., Linsley, P. S., Huang, L-Y., Germain, R. N., and Shev- ach,E. M., J. Zmmunol. 151,1224,1993.

Geppert, T. D., and Lipsky, P. E., J. Zmmunol. 138,1660,1987.

Sagerstrom, C. G., Kerr, E. M., Allison, J. P., and Davis, M. M., Proc. Natl. Acad. Sci. USA 90,8987,1993.

Smith, K. A., Science 240,1169,1988.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

Waldman, T. A., Zmmunol. Today 14,264,1993.

Ferreira, A., Schofield, L., Enea, V., Schellenkens, H., Van der Meide, P., Collins, W. E., Nussenzweig, R. S., and Nussenzweig, V., Science 232,881, 1986.

Schofield, L., Ferreira, A., Altszuler, R., Nussenzweig, V., and Nussenzweig, R. S., J. Zmmurwl. 139,2020, 1987.

Shear, H. L., Srinivasan, R., Nolan, T., and Ng, C., J. Zmmunol. 143,2038,1989.

Playfair, J. H., Zmmunol. Lett. 25, 173,199O.

Ockenhouse, C. F., Schulman, S., and Shear, H. L., J. Zmmunol. 133,1601,1984.

Grun, J. L., Long, C. A., and Weidanz, W. P., Infect. Zmmun. 48, 853,1985.

Clark, I. A., Hunt, N. H., Butcher, G., and Cowden, W. B., J. Zmmunol. 139,3493,1987.

Stevenson, M. M., Tam, M. F., and Nowotarski, M., Zmmunol. L&t. 25, 115, 1990.

Snapper, C. M., and Paul, W. E., Science 236,944,1987.

White, W. I., Evans, C. B., and Taylor, T. W., Infect. Zmmun. 59, 3547,199l.

Peleman, R., Wu, J., Fargeas, C., and Delespesse, G., J. Exp. Med. 170,1751,1989.

Bello-Fernandez, C., Oblakowski, P., Meager, A., Duncombe, A. S., Rill, D. M., Hollbrand, A. V., and Brenner, M. K., Zmmu- nology 72,161,1991.

Tanaka, T., Hu-li, J., Seder, R. A., St. Groth, B. F., and Paul, W. E., Proc. Natl. Acad. Sci. USA 90,5914,1993.

Pied, S., Civas, A., Berlot-Picard, F., Renia, L., Miltgen, F., Gen- tilini, M., Doly, J., and Mazier, D., J. Zmmunol. 148, 197, 1992.

Fiorentino, D. F., Bond, M. W., and Mosmann, T. R., J. Exp. Med. 170,2081,1989.

Kwiatkowski, D., Zmmunol. Lett. 25,213,1990.

Grau, G. E., Fajardo, L. F., Piguet, P. F., Allet, B., Lambert, P. H., and Vassalli, P., Science 237,1210,1987.

Taverne, J. W., Fiers, W., and Playfair, J. H. L., Clin. Exp. Zm- munol. 67, 1, 1987.

Patterson, B. K., Till, M., Otto, P., Goolsby, C., Furtado, M. R., McBride, L. J., and Wolinsky, S. M., Science 260,976, 1993.

Kelso, A., Troutt, A. B., Maraskovsky, E., Gough, N. M., Morris, L., Pech, M. H., and Tompson, J. A., Zmmunol. Reu. 123, 85, 1991.

Herold, K. C., Lancki, D. W., Dunn, D. E., Arai, K., and Fitch, F. W., Eur. J. Zmmunol. 16,1533, 1986.

Weiss, L., Geduldig, U., and Weidanz, W., Am. J. Anat. 176, 251,1986.

Kumar, S., Good, M. F., Dontfraid, F., Vine@ J. M., and Miller, L. H., J. Zmmunol. 143,2017,1989.

plasmodium yoelii in mice: differential induction of cytokine gene expression during...

Documents