Deletion of T Cells Bearing the V{beta}8.1 T-Cell Receptor following Mouse Mammary Tumor Virus 7 Integration Confers Resistance to Murine Cerebral Malaria

Plasmodium berghei ANKA induces a fatal neurological syndromeknown as cerebral malaria (CM) in susceptible mice. Host geneticelements are among the key factors determining susceptibilityor resistance to CM. Analysis of mice of the same H-2 haplotyperevealed that mouse mammary tumor virus 7 (MTV-7) integrationinto chromosome 1 is one of the key factors associated withresistance to neurological disease during P. berghei ANKA infection.We investigated this phenomenon by infecting a series of recombinantinbred mice (CXD2), derived from BALB/c (susceptible to CM)and DBA/2 (resistant to CM) mice, with P. berghei ANKA. We observeddifferences in susceptibility to CM induced by this Plasmodiumstrain. Mice with the MTV-7 sequence in their genome were resistantto CM, whereas those without integration of this gene were susceptible.Thus, an integrated proviral open reading frame or similar genomicsequences may confer protection against neuropathogenesis duringmalaria, at least in mice.

INTRODUCTION

Top

Introduction

Cerebral malaria (CM) is a neurological syndrome caused by theasexual blood stage of Plasmodium falciparum in humans and Plasmodiumberghei ANKA in experimental rodent models. Numerous parameters,such as the diversity of parasite populations and genetic polymorphismsin the vertebrate host, influence host-parasite interactionsand pathogenesis during malaria infection (6, 22, 27, 29). Itis important to increase our understanding of the mechanismsby which these factors contribute to the development of neuropathogenesisbecause CM continues to kill 2 to 2.5 million people annually,mostly children in areas where the parasite is endemic (2).

Several studies with humans have shown that host immunogeneticfactors have a significant effect on susceptibility to severeforms of malaria (13, 20, 23). Genes encoding certain allelesof HLA molecules are known to be associated with protectionagainst pathogenesis during malaria (14, 15). Alleles of thetumor necrosis factor alpha (TNF- ${alpha}$ ) gene have also been shownto be associated with protection against severe malaria (21).TNF- ${alpha}$ plays a role in CM development by inducing the expressionof adhesion molecules on the surface of endothelial cells ofbrain microvessels, leading to the sequestration of parasitizederythrocytes due to their cytoadhesion within the vascular systemof the brain (7).

In experimental models, mice lacking interferon (IFN) regulatoryfactor-1 (IRF-1) have been found to be protected against deathdue to CM (24). IRF-1 has been associated with malarial neuropathogenesisvia the high level of production of type-1 cell-derived cytokinessuch as TNF- ${alpha}$ and IFN- ${gamma}$ (8-12). Moreover, inbred mice with thesame H-2 haplotype show various degrees of susceptibility toCM and can be divided into three categories: highly susceptible,such as B10.D2 mice; weakly susceptible, such as BALB/c mice;and resistant, such as DBA/2 mice. This suggests that H-2 locusdoes not play a predominant role in determining susceptibilityor resistance to neuropathogenesis during malaria in the mousemodel (4).

Attempts to estimate the relationship between the immune responseand disease severity have shown that ${alpha}$ ßT cells playa direct role in the complex pathogenesis of malaria (4). Ina previous study, we demonstrated that the development of neurologicalsyndromes in B10.D2 mice infected with P. berghei ANKA was associatedwith an increase in the number of CD8Vß8.1⁺ and CD8Vß8.2⁺T cells in peripheral blood. Strong evidence for a causal linkbetween the expansion of populations of T cells bearing Vß8.1and/or Vß8.2 and the pathological consequences ofP. berghei ANKA infection has been provided by the significantlylower levels of CM observed in mice treated with a monoclonalantibody eliminating all populations of cells bearing T-cell-receptor(TCR) Vß8.1 and Vß8.2 chains. In addition,BALB.D2 mice, which are congenic for a mouse mammary tumor virus(MTV-7) that constitutively depletes Vß6, Vß7,Vß8.1, and Vß9 populations, is susceptibleto P. berghei ANKA infection but does not develop CM (4).

The open reading frame in the 3' long terminal repeat of theMTV-7 provirus, which is integrated into chromosome 1 in severalmouse strains, encodes a superantigen, Mls1^a. Like all superantigens,Mls1^a elicits a very powerful response from mature T cells bearingthe appropriate TCR Vß chains, and during T-cell maturationin the thymus, this response results in clonal elimination orclonal inactivation of the responding cells (3, 28). Superantigensmay therefore have significant deleterious effects on the T-cellrepertoire in the periphery (19).

These observations indicate that there is an association betweenthe presence of certain TCR-restricted lymphocyte subpopulationsand the development of neuropathogenesis in experimental modelsof malaria. In the present study, we identified a specific exampleof such an association based on the observation that BALB/cmice (MTV-7^-) are susceptible to CM induced by P. berghei ANKAclone 1.49L, whereas BALB.D2 and DBA/2 mice (MTV-7⁺) are resistant.We analyzed the susceptibility and/or resistance to CM of variousrecombinant inbred (RI) lines of the CXD2 series derived fromBALB/c x DBA/2 crosses to investigate the possible role of MTV-7integration in determining susceptibility or resistance to CM.

MATERIALS AND METHODS

Top

Materials and Methods

Mice. We used 6- to 8-week-old male and female mice of various CXD2RI (n > 20) lines: (CXD2) D (n = 32), (CXD2) E (n = 30),(CXD2) F (n = 23), (CXD2) J (n = 10), (CXD2) K (n = 25), (CXD2)L (n = 26), and (CXD2) N (n = 22). These lines were developedand maintained at the Pasteur Institute (5). Eight-week-oldBALB/c J Rj and DBA/2 mice were obtained from Janvier (Le GenestSt. Isle, France) and Charles River (Les Elboeuf, France), respectively.

Parasites and induction of cerebral malaria. Erythrocytic stages of clone 1.49L of P. berghei ANKA strain,kindly provided by D. Walliker (Institute of Genetics, Edinburgh,United Kingdom), were maintained in C57BL/6 mice. This clonewas selected for its great capacity to induce CM, with neurologicalsigns (ataxia, paralysis, deviation of the head, and convulsion)appearing 6 to 8 days after infection (1). Blood stages of theclone were used as stabilates (10⁷ parasitized red blood cells[pRBC]/ml in Alsever's solution supplemented with 10% glycerol)stored in liquid nitrogen. CM was induced by the intraperitonealinjection of 10⁶ PbA-infected RBCs (pRBCs). Parasitemia wasmonitored by evaluating daily blood smears for parasites afterGiemsa staining.

Southern blotting. Genomic DNA samples were isolated from the tail (25), and Southernblot analysis was carried out as previously described (16).Probes were derived from MMTV-C3H (18) and consisted of a 1.8-kbBglII/HindIII fragment encompassing the 3' long terminal repeatand a 2.3-kb HindIII/BglII fragment encompassing the env region.

Antibodies and flow cytometry. Fluorescein isothiocyanate-conjugated monoclonal antibodiesspecific for the mouse CD3 ${varepsilon}$ chain (145-2C11) and biotin-conjugatedmonoclonal antibodies against various Vß TCR families(Vß4 [KT4]; Vß6 [44-22-1]; Vß7[RM4415]; Vß8.1,2 [KJ16]; Vß8.1, Vß8.2,and Vß8.3 [F23.1]; and Vß9 [MR10]) wereproduced in the laboratory.

Lymphocyte preparation and staining. Peripheral blood lymphocytes (PBL) were isolated from uninfectedcontrol mice by centrifugation in Ficoll-Hypaque (Pharmacia)and washed twice with 3% fetal calf serum (Gibco-BRL) in phosphate-bufferedsaline. For cytofluorometry, lymphocytes were incubated firstwith biotinylated monoclonal antibody directed against the variousVßs and subsequently with anti-CD3 fluorescein-labeledmonoclonal antibody in the presence of phycoerythrin-conjugatedstreptavidin. Analysis was carried out with a FACScan cytofluorometer(Becton Dickinson, Grenoble, France) and CellQuest software.Viable lymphocytes were gated on the basis of light scattering(FSC/SSC) and for all Vß analyses (5,000 to 10,000events were acquired and recorded per sample). The percentageof fluorescent cells was determined by integrating profilesaccording to the number of viable lymphocytes.

Statistical analysis. Statistical analysis was carried out with StatView version 4.5software. Kaplan-Meier tests were used to assess survival. Differencesbetween the survival curves of MTV-7⁺ and MTV-7^- mice were analyzedby the Logrank (Mantel-Cox) test. P values of <0.05 wereconsidered to be significant. Parasitemia is expressed as themean percentage of all erythrocytes that were infected.

RESULTS AND DISCUSSION

Top

Results and Discussion

Malaria induced by P. berghei ANKA in susceptible mice typicallypresents neurological signs (ataxia, paralysis, deviation ofthe head, and convulsions), which appear 6 to 13 days afterinfection. In a previous study, we showed that mice of the samemajor histocompatibility complex haplotype (H-2^d)—B10.D2,BALB/c, and DBA/2—displayed different phenotypes in termsof susceptibility to neuropathogenesis induced by intraperitonealinfection with 10⁶ P. berghei ANKA-parasitized erythrocytes.All of the B10.D2 mice and 40% of the BALB/c mice were susceptibleto CM, whereas DBA/2 mice presented a resistant phenotype (4).We also demonstrated that the immunopathogenesis of CM in susceptibleH-2^d mice was associated with the expansion before death ofa restricted T-cell subpopulation bearing Vß8.1 orVß8.2 receptor chains. Strong evidence supportingthe existence of a link between the increase in number of Tcells bearing the Vß8.1 or Vß8.2 segmentand the pathological consequences of P. berghei ANKA infectionwas provided by the observation that the incidence of CM wassignificantly lower in BALB.D2 mice congenic for MTV-7, resultingin the depletion of Vß8.1⁺ T cells through the openreading frame superantigen and in the (BALB.D2 x B10.D2)F₁ mice(4). We investigated whether MTV-7 was sufficient to induceresistance to CM by using (CXD2) RI strains derived from BALB/cmice bearing MTV-6, -8, and -9 (susceptible to CM) and DBA/2mice bearing MTV-1, -6, -7, -8, -11, -13, -14, and -17 (resistantto CM). We determined the MTV integration profile in the various(CXD2) RI strains by Southern blotting (Fig. 1). Of the sevenstrains analyzed, only the (CXD2) E and F strains had not inheritedthe MTV-7 locus (Table 1). The Mtv-7 gene encodes a superantigenknown to deplete all T-cell subpopulations bearing TCR Vß6,Vß7, Vß8.1, and Vß9 (26). We thereforeassessed the ability of MTV-7 to act as a superantigen by analyzingthe TCR Vß repertoire of PBL by flow cytometry todetect the clonal depletion of cells bearing the reactive TCRVßs. We observed a loss of TCR Vß6-, Vß7-,Vß8.1-, and Vß9-bearing T cells among thePBL of the (CXD2) D, J, K, L, and N mouse strains, all of whichare MTV-7⁺. In contrast, we observed no depletion in (CXD2)E and F mice, which are MTV-7^- (Fig. 2). We also analyzed neuropathogenesisin the various (CXD2) RI strains after intraperitoneal inoculationwith 10⁶ P. berghei ANKA-infected erythrocytes. Only (CXD2)E and F mice remained as susceptible to CM as their relativeBALB/c parents, whereas (CXD2) D, N, J, K, and L mice displayeda resistant phenotype similar to that of DBA/2 mice (Fig. 3).All of the mice that developed CM died 7 to 13 days after parasiteinoculation; these mice displayed 10 to 15% parasitemia andsevere neurological symptoms associated with CM, including hemiplegiaand coma. The resistance to neuropathogenesis of MTV-7⁺ CXD2mice does not seem to be directly due to resistance to infectionby the parasite, since we observed no significant differencein susceptibility, as shown by the parasitemia curves of theCXD2 strains during the first 2 weeks of infection (Fig. 4).We also observed no correlation between the level of parasitemiaand neuropathogenesis. The mice that escaped CM died 20 dayslater from anemia due to hyperparasitemia.

View larger version (31K):
[in this window]
[in a new window]
FIG. 1. Profiles of MTV integration in the various (CXD2) RI strains. Genomic DNA was digested with EcoRI, subjected to electrophoresis in a 0.7% agarose gel, and Southern blotted. The membrane was then probed with envelope (A) or long terminal repeat (B) probes. Lambda HindIII DNA was used as a molecular mass marker.

View this table:
[in this window]
[in a new window]
TABLE 1. MTV integration profiles of the various (CXD2) RI strains^a

View larger version (16K):
[in this window]
[in a new window]
FIG. 2. Fluorescence-activated cell sorting analysis of PBL from the various (CXD2) RI strains. Data indicate the mean percentage ± the standard deviation of Vß⁺ cells among CD3⁺ cells from two separate experiments. Mice (n = 3) were tested individually for each RI strain.

View larger version (23K):
[in this window]
[in a new window]
FIG. 3. Kaplan-Meier graph of survival from CM of (CXD2) RI strains of mice. All mice were infected intraperitoneally with 10⁶ P. berghei ANKA clone 1.49L pRBCs. Neurological manifestations appeared 7 to 13 days after parasite inoculation. Mice that escaped CM died 20 to 28 days after inoculation from anemia due to hyperparasitemia. Results for BALB/c (n = 20), DBA/2 (n = 19), (CXD2) D (n = 17), (CXD2) E (n = 20), (CXD2) F (n = 8), (CXD2) J (n = 7), (CXD2) K (n = 23), and (CXD2) N (n = 8) mice are shown. The results were obtained from two separate experiments.

View larger version (29K):
[in this window]
[in a new window]
FIG. 4. Parasitemia curve for the various (CXD2) RI strains of mice infected with P. berghei ANKA clone 1.49L. Results for BALB/c (n = 20), DBA/2 (n = 19), (CXD2) D (n = 17), (CXD2) E (n = 20), (CXD2) F (n = 8), (CXD2) J (n = 7), (CXD2) K (n = 23), and (CXD2) N (n = 8) mice are shown. The results are expressed as the mean parasitemias per day for two separate experiments.

We compared the incidence of CM in MTV-7⁺ and MTV-7^- mice (Fig.5). A strong correlation was observed between the presence ofMtv-7 in the mouse genome and protection against neuropathogenesisin experimental malaria infection. A significantly larger percentageof MTV-7^- mice (58.5%) than of MTV-7⁺ mice (2.6%) developedCM ( ${chi}$ ² = 59.4, 1 df, and P < 0.001). We observed no differencesin the timing of clinical signs and death of mice that developedCM, regardless of whether they were MTV-7⁺ or MTV-7^- (data notshown). Again, the difference in susceptibility to CM cannotbe related to parasite growth because no difference in parasitemiawas observed between MTV-7⁺ and MTV-7^- mice during the first2 weeks after infection, as confirmed by the Mantel-Cox logrank test ( ${chi}$ ² = 12.47, 1 df, and P = 0.0004; Fig. 6). Thesedata suggest that the viral genome product has no direct effecton parasite development and confirm our previous results obtainedwith BALB.D2 and BALB.SW, which display integration of the endogenousviral genome (Mtv-7) and of its exogenous counterpart MMTV-SW,which encodes the same superantigen, respectively (4).

View larger version (23K):
[in this window]
[in a new window]
FIG. 5. Frequency of CM in MTV-7⁺ and MTV-7^- (CXD2) RI mice. The results are expressed as the percent occurrence of CM in MTV-7⁺ (n = 115) and MTV-7^- (n = 53) mice.

View larger version (15K):
[in this window]
[in a new window]
FIG. 6. Parasitemia curve for MTV-7⁺ (n = 63) and MTV-7^- (n = 20) RI mice infected with P. berghei ANKA clone 1.49L. The results are expressed as the mean percentage of erythrocytes infected from two separate experiments. D13, day 13 after parasite inoculation.

We observed a difference in the sensitivities to CM of RI strainsbelonging to the MTV-7^- group. We found that 76% of (CXD2) Emice presented with CM versus only 26% of the (CXD2) F mice.We investigated the possible direct or indirect involvementof MTVs other than MTV-7 in the mechanisms underlying susceptibilityor resistance to neuropathogenesis during P. berghei ANKA infection.We compared the MTV integration profiles of (CXD2) E and (CXD2)F mice with each other and with those of all of the other CXD2mice studied. The only difference found between these two strainswas that MTV-11 was absent from (CXD2) E (Table 1). This observationraised questions about possible interaction between mechanismsinduced by MTV-7 and by MTV-11, conferring resistance to neurologicaldisease in (CXD2) F mice. The log rank test showed a significantdifference in CM incidence between MTV-11⁺ and MTV-11^- micefor the MTV-7⁺ (CXD2) E and F mouse lines ( ${chi}$ ² = 7.79, 1 df, andP = 0.0053). However, MTV-11 itself had no direct effect onthe incidence of CM. MTV-7 probably plays a complex role indetermining susceptibility to Plasmodium infection, dependingon the parasite species and the strain of mouse and probablyalso the developmental stage at which the mice were experimentallyinfected. Our results show that MTV-7 does not affect susceptibilityto P. berghei ANKA infection.

In agreement with results presented here and our results demonstratingthat the development of CM in H-2^d haplotype (B10.D2) mice iscorrelated with an increase of PBL bearing Vß8.1 TCRbefore death and that all mouse strains lacking Vß8.1⁺T cells due to antibody depletion or deletion and anergy resultingfrom superantigen activity are resistant to CM (4), we thereforeconclude that MTV-7, whether provided endogenously or exogenously,is sufficient to confer resistance to CM in laboratory micestrains with the H-2^d haplotype via the deletion of Vß8.1⁺cells. It is still unclear how Vß8⁺ cells affect neuropathogenesis.These cells may release IFN- ${gamma}$ and TNF, two major mediators ofmalaria pathogenesis, by inducing the expression of adhesionmolecules on the endothelial cells of brain capillaries responsiblefor the sequestration of infected erythrocytes, leukocytes,and monocytes (27).

MTV has also been shown to protect against pathogenesis in Leishmaniamajor infection. The mice of most inbred strains are resistantto infection by L. major, whereas the BALB/c strain is unableto control the infection and develops progressive disease. Inthis experimental model, Vß4V ${alpha}$ 8 TCR⁺ CD4 T cells havebeen demonstrated to be responsible for the early burst of interleukin-4involved in the Th2 response that confers susceptibility todisease (17). Moreover, it has been shown that BALB/c mice infectedwith MMTV-SIM, resulting in the depletion of cells expressingVß4 TCR, become specifically resistant to infectionby L. major, whereas BALB/c mice infected with MMTV-SW, resultingin the depletion of Vß6⁺ T cells, do not become resistant.

In conclusion, this study, as well as previously published data(4), provides a defined genetic and cellular basis for susceptibilityand resistance in experimental murine CM involving modificationof the T-cell repertoire by depletion of a precise T-cell subset.Future studies should seek to determine the underlying mechanismand, perhaps most important of all, whether similar mechanismsoperate in P. falciparum infection in humans.

ACKNOWLEDGMENTS

We thank Jack Formentin and Maurel Tefit for animal care, MoniqueBauzou for parasite detection on blood smears, DanièleVoegtle for excellent technical assistance, and Wendy Houssinfor help in preparing the manuscript.

This work was supported by the "Programme de Recherche Fondamentaleen Microbiologie Maladies Infectieuses et Parasitaires" fromthe French Ministry of Research and was conducted as part ofthe CNRS Laboratoire Européen Associé "Génétiqueet Développement de la Tolérance Naturelle." M.I.B.was supported by the Institut Lilly. S.B. is a recipient ofa fellowship from the Fondation pour la Recherche Médicale.S.P. and P.-A.C. were recipients of the Visiting Scientist fellowshipfrom the FCT (Portugal).

O.G. and A.E. contributed equally to this study.

FOOTNOTES

* Corresponding author. Mailing address: Unité d'Immunophysiopathologie Infectieuse, Département d'Immunologie, 25 Rue du Dr Roux, 75724 Paris Cedex 15, France. Phone: 33-1-40-61-35-75. Fax: 33-1-40-61-30-66. E-mail: spied{at}pasteur.fr.

Editor: S. H. E. Kaufmann

REFERENCES

Top