CD1d-restricted NKT cells contribute to malarial splenomegaly and enhance parasite-specific antibody responses

2588

D. S. Hansen et al.

Eur. J. Immunol. 2003. 33: 2588–2598

CD1d-restricted NKT cells contribute to malarial splenomegaly and enhance parasite-specific antibody responses Diana S. Hansen, Mary-Anne Siomos, Tania de Koning-Ward, Lynn Buckingham, Brendan S. Crabb and Louis Schofield The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia

CD1d-restricted NKT cells are a novel T cell lineage with unusual features. They co-express some NK cell receptors and recognize glycolipid antigens through an invariant T cell receptor (TCR) in the context of CD1d molecules. Upon activation through the TCR, NKT cells produce large amounts of IFN- + and IL-4. It has been proposed that rapid cytokine output by activated NKT cells may induce bystander activation of other lymphoid lineages. The impact of CD1d-restricted NKT cell activation in the induction of B cell-mediated immune responses to infection is still unclear. We show here that CD1-restricted NKT cells contribute to malarial splenomegaly associated with expansion of the splenic B cell pool and enhance parasitespecific antibody formation in response to Plasmodium berghei infection. The increased B cell-mediated response correlates with the ability of NKT cells to promote Th2 immune responses. Additionally, antibody responses against the glycosylphosphatidylinositol (GPI)anchored protein merozoite surface protein 1 (MSP-1) were found to be significantly lower in CD1–/– mice compared to wild-type animals. P. berghei-infected MHC class II (MHCII)–/– mice also generated antibodies against MSP-1, suggesting that antibody production against GPIanchored antigens in response to malaria infection can arise from both MHCII-dependent and independent pathways. Key words: Malaria / Antibody responses / NKT cells / Splenomegaly / CD1

1 Introduction CD1d-restricted NKT cells are a novel T cell lineage with unusual features [1]. They co-express some NK cell receptors and recognize glycolipid antigens through an invariant TCR in the context of CD1d molecules. Upon antigen-specific or polyclonal stimulation through the TCR, CD1d-restricted NKT cells produce large amounts of IL-4 and IFN- + with rapid kinetics [2]. They have therefore been postulated to have an immunoregulatory role and influence Th1/Th2 differentiation of the acquired immune system [3]. It has also been proposed that the rapid production of IL-4 and IFN- + by activated NKT cells may induce bystander activation of other lymphoid lineages. In fact, in vivo stimulation of CD1d-restricted NKT cells with the glycolipid antigen § -galactosylceramide ( § -GalCer) induces NK cells to proliferate [4] and secrete

Received Revised Accepted

5/11/02 7/7/03 22/7/03

IFN- + [5], generates CD69+ CD8+ T lymphocytes exerting both cytotoxic and IFN- + producing activities [6] and induces proliferation of CD4+ and CD8+ memory T cells [7]. The impact of CD1d-restricted NKT cell activation in the induction of B cell-mediated immune responses is not fully characterized. It has been shown that V § 14transgenic mice, which display a seven- to tenfold increased number of NKT cells, produce enhanced levels of Th2-controlled Ig isotypes such as IgE and IgG1 [8]. In addition, in vivo administration of § -GalCer induces up-regulation of early activation markers on B lymphocytes through IL-4 production by CD1-restricted NKT cells [9]. The rapid transactivation of B cells by CD1restricted NKT cells might suggest that antigenstimulated NKT cells could play a role in early B cellmediated immunity against infection. However, this interesting proposition has not been explored.

[DOI 10.1002/eji.200323666] Abbreviations: P.i: Post-infection GPI: Glycosylphosphatidylinositol GST: Glutathione S Transferase MSP-1: Merozoite surface protein-1 HMS: Hyperreactive malarious splenomegaly 0014-2980/03/0808-2588$17.50 + .50/0

We have recently shown that CD1d-restricted NKT cells influence cytokine responses, pathogenesis and fatality in the Plasmodium berghei model of cerebral malaria [10]. In this study we sought to investigate the contribu© 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Eur. J. Immunol. 2003. 33: 2588–2598

tion of the CD1/NKT cell pathway to induction of B cellmediated immune responses during malaria infection. We found that CD1d-restricted NKT cells contribute to the splenomegaly associated with expansion of the splenic B cell population and enhance parasite-specific antibody formation in response to infection. The increased B cell-mediated response correlated with the ability of CD1-restricted NKT cells to influence the immune response towards Th2. Antibody responses require MHC-restricted recognition of peptide antigens by CD4+ T helper cells. It has also been shown that antibody formation against lipidated or non-peptide antigens can occur in a CD1-restricted manner [11, 12]. Consistent with those observations, we also found that antibody production against the GPI-anchored protein MSP-1 in response to malaria infection is reduced in CD1d deficient mice and that antibody responses against GPI-anchored antigens may arise from both MHCII-dependent and independent pathways.

2 Results 2.1 Reduced splenomegaly and absolute number of B cells in CD1–/– mice during P. berghei infection To study the contribution of the CD1/NKT cell pathway in the induction of B cell-mediated immune responses, BALB/c wild-type and CD1–/– mice were infected with P. berghei and various immune responses were analyzed at different time points post-infection (p.i.). Spleen enlargement or splenomegaly is a clinically defined disorder reported to occur both in human and murine malaria infections. Fig. 1 shows that CD1–/– mice display reduced splenomegaly in response to P. berghei infection compared to wild-type animals. Spleens from non-infected mice from both strains were of similar size, indicating that the reduced splenomegaly in CD1–/– animals results from a differential response to infection. The spleens of infected mice are highly hyperemic and this is a typical feature of P. berghei infections. The absolute number of viable splenocytes was identical in uninfected wild-type and CD1–/– mice (Fig. 2A). In both mouse strains, the total number of splenocytes increases as the infection progresses reaching a peak at day 7 p.i. However, this increase was significantly higher in wild-type animals (Fig. 2A). To analyze the cell composition in spleens of malaria-infected mice, splenocytes from wildtype and CD1–/– mice were stained with anti-CD4, antiCD8 and anti-B220 antibodies. The cells were analyzed by flow cytometry and the absolute number of B cells and T cells calculated. The total number of CD4+ and CD8+ T cells did not significantly change during the first week of infection (Fig. 2C, D). After day 10 p.i. the abso-

Regulation of B cell responses in malaria by NKT cells

2589

Fig. 1. Reduced splenomegaly in CD1–/– mice during infection with P. berghei. Spleens from wild-type (A) or CD1–/– (C) mice collected at day 7 p.i. with P. berghei. Spleens from non-infected wild-type (B) or CD1–/– (D) mice are shown for comparison.

lute numbers of both CD4+ and CD8+ cells were significantly reduced, confirming previous observations [13] of T cell depletion during P. berghei infection. No significant differences in the number of T cells were found between wild-type and CD1–/– mice. In contrast, the absolute number of B lymphocytes increased in response to malaria infection. However, this response was significantly lower in CD1–/– animals compared to wild-type mice (Fig. 2B). To determine whether the decreased B cell population observed in CD1–/– mice reflected a reduction in B lymphocyte proliferation, wild-type and CD1–/–- mice were fed the thymidine analog 5-bromo-2´deoxyruridine (BrdU) and B cell proliferation in response to P. berghei infection was analyzed in vivo. As shown in Fig. 3A, B both the percentage and the absolute number of BrdU+ B220+ cells were significantly reduced in CD1ddeficient mice compared to wild-type controls. To further analyze whether the expression of CD1d on B lymphocytes had an impact on the decreased proliferation rates observed during malaria infection, B lymphocytes from wild-type mice were labeled with 5-(and-6)carboxyfluorescein diacetate succinimidyl ester (CFSE) and adoptively transferred into both wild-type and CD1–/– recipient mice. Four hours later the animals were infected with P. berghei and CFSE staining was evaluated on splenic B220+ cells harvested at day 6 p.i. Fig. 3C shows that B cells from wild-type mice proliferated more when adoptively transferred into wild-type recipients compared to CD1–/– mice. Taken together, these results suggest that the CD1/NKT cell pathway contributes to the expansion of the splenic B cell population that is associated with splenomegaly during infection with P. berghei.

2590

D. S. Hansen et al.

Eur. J. Immunol. 2003. 33: 2588–2598

Fig. 2. Reduced absolute number of splenocytes and B lymphocytes in CD1–/– mice in response to infection with P. berghei. Spleen cells from wild-type or CD1–/– mice were stained with anti-B220, anti-CD4 and anti-CD8 antibodies. The absolute number of splenocytes (A), B cells (B), CD4+ T cells (C) and CD8+ T cells (D) was calculated. Each experiment is representative of four separate infections. Each point represents the mean of three samples ± SE. *p X 0.05 between wild-type and CD1–/– mice.

Fig. 3. B cell proliferation in response to P. berghei is reduced in CD1–/– mice. P. berghei-infected wild-type and CD1–/– mice were fed BrdU. The percentage (A) of BrdU+ cells was analyzed on gated B220+ lymphocytes at different times post-infection. Representative histograms are shown. The absolute number (B) of BrdU+ B220+ cells in spleens of P. berghei-infected mice. Each point represents the mean of three samples ± SE *p X 0.05, **p X 0.05 between wild-type and CD1–/– mice. (C) Adoptive transfer of CFSE-labeled B lymphocytes from wild-type mice. Wild-type or CD1–/– mice were adoptively transferred with CFSE-labeled B cells from wild-type animals. Mice (n=3) were then infected with 1×106 P. berghei-infected red blood cells. Splenocytes were harvested at day 6 p.i and CFSE staining was assessed on gated B220+ cells by flow cytometry. Representative histograms are shown.

Eur. J. Immunol. 2003. 33: 2588–2598

2.2 Lack of CD1high B cells does not account for the reduced B cell expansion in CD1–/– mice CD1d is constitutively expressed in all mouse B lymphocytes [14, 15]. A subset of splenic B cells, representing 10% of the total B cell pool, displays higher levels of CD1 (CD1high B cells) than the bulk of B lymphocytes [15]. These cells are situated in the marginal zone of the spleen and are therefore good candidates to interact with blood-borne antigens. To study whether this cell population was preferentially expanded during malaria infection, explaining the increased cellularity observed in wild-type mice compared to CD1d-deficient animals, splenocytes from BALB/c wild-type mice collected at various times p.i. were stained with anti-B220 and antiCD1d antibodies and analyzed by flow cytometry. Fig. 4 shows that both the percentage and the absolute number of CD1high B cells significantly decrease as malaria infection develops. By day 7 p.i. only 0.74% of CD1high B cells were found in spleens of BALB/c infected mice. Thus, lack of CD1high B cells cannot explain the reduced B cell expansion in response to malaria infection in CD1–/– mice. It remains to be determined whether CD1high B cells are deleted, down-regulate CD1d expression or migrate out of the spleen in response to malaria infection.

Regulation of B cell responses in malaria by NKT cells

2591

2.3 Reduced parasite-specific IgM and IgG antibody formation in CD1–/– mice in response to malaria infection The antibody response to P. berghei was studied in wildtype and CD1–/– mice. To that end, mice were infected and sera were prepared at different times p.i. The parasite-specific IgM titers were significantly lower in CD1–/– mice compared to wild-type animals (Fig. 5A). IgG responses to P. berghei total lysate were also reduced in CD1–/– animals compared to wild-type controls (Fig. 5B). Thus the CD1/NKT cell pathway promotes the formation of parasite-specific antibody response during malaria infection. The IgG subclasses produced in response to infection were also analyzed. The parasite-specific IgG1 antibody titers were significantly higher in wild-type compared to CD1–/– mice (Fig. 5C). In contrast, IgG2a responses were more elevated in CD1–/– animals (Fig. 5D). IgG2b and IgG3 levels were very low and no significant differences were found between the two groups at the time points tested (data not shown).

Fig. 4. Lack of CD1high B cells does not account for the reduced splenomegaly in CD1–/– mice. Spleen cells from BALB/c wild-type mice were stained with anti-B220 and anti-CD1d antibodies at various time points p.i. with P. berghei. The percentage (A) of double positive cells from total splenocytes was analyzed by flow cytometry. Representative histograms are shown. The absolute number (B) of CD1high B cells in spleens of P. berghei infected mice. Each point represents the mean of at least three samples.

2592

D. S. Hansen et al.

Eur. J. Immunol. 2003. 33: 2588–2598

Fig. 5. The antibody response to P. berghei ANKA in wild-type and CD1–/– mice. Wild-type and CD1–/– mice were infected with P. berghei. Serum samples were collected at different time points p.i. and the parasite-specific IgM (A), IgG (B), IgG1 (C) and IgG2a (D) antibody responses were analyzed by ELISA against total parasite lysate. Each point represents the mean of eight-to-ten samples ± SE. *p X 0.05, **p X 0.01, ***p X 0.005 between wild-type and CD1–/– mice.

2.4 Lack of CD1d does not impair B cell activation in response to malaria infection Fig. 1–5 indicate that B cell-mediated immune responses to malaria are significantly reduced in CD1ddeficient mice. To investigate whether lack of CD1d expression on B lymphocytes had an impact on the level of cell activation, wild-type and CD1–/– mice were infected with P. berghei and the expression of the early activation marker CD69 and the costimulatory molecules B7–1 and B7–2 on B lymphocytes was analyzed by FACS. The expression of CD69, B7–1 and B7–2 increased greatly in B cells of infected wild-type and CD1–/– mice reaching a peak at day 5 p.i. No significant differences were found between wild-type and CD1–/– mice (Fig. 6), suggesting that lack of CD1d does not impair activation of B lymphocytes in response to infection.

2.5 Differential Th1/Th2 bias in wild-type and CD1–/– mice in response to malaria infection B cell-mediated immune responses including antibody secretion and immunoglobulin isotype switching can be influenced by cytokines. To investigate whether the differential B cell-mediated immune response to P. berghei in wild-type and CD1–/– mice reflected the ability of CD1d-restricted NKT cells to influence the Th1/Th2 balance of conventional T cell populations in response to

infection, we examined IL-4 and IFN- + production in response to anti-CD3 antibody by total splenocytes isolated from infected wild-type and CD1–/– mice. Spleen cells from both wild-type and CD1–/– mice produced high levels of IFN- + at early stages of the infection. From day 7 p.i, IFN- + levels decreased in wild-type but remained high in spleen cells from CD1–/– mice (Fig. 7A). In contrast, IL-4 production was initially down-regulated but significantly increased from day 7 p.i, suggesting a switch from Th1 to Th2 immune response in wild-type animals (Fig. 7B). In CD1–/– mice, IFN- + production was not down-regulated and no switch to IL-4 production was observed, indicating BALB/c CD1d-restricted NKT cells provide help for the development of Th2 immune responses. Thus in wild-type mice, increased B cell mediated-immune responses to infection correlate with the ability of CD1-restricted NKT cells to promote development of Th2 responses among conventional T lymphocyte populations.

2.6 Antibody formation against GPI-anchored proteins during malaria infection in wild-type, CD1–/– and MHCII–/– mice Fig. 1–7 suggest that IL-4 output by CD1d-restricted NKT cells may lead to a generalized enhancement of B cell-mediated immune responses. In addition, it has been shown that CD1d-restricted NKT cells may contribute to the development of B cell responses by providing

Eur. J. Immunol. 2003. 33: 2588–2598

Regulation of B cell responses in malaria by NKT cells

2593

suggest that the CD1/NKT cell contribution to antibody formation might operate early after exposure to antigen (Fig. 8A). MHCII–/– mice, which lack CD4+ T lymphocytes and are completely unable to produce antibodies against T-dependent antigens [16], produced antibodies against MSP-119 in response to infection with P. berghei (Fig. 8B). Moreover, a pool of sera from MHCII–/– mice recognized the 38 kDa MSP-119-GST fusion protein in Western blot but not GST alone confirming the specificity of the response (Fig. 8C). Thus, antibody formation against GPI-anchored proteins in response to malaria infection can arise by both MHCII-dependent and independent pathways. We also sought to determine whether MHCII–/– mice were able to respond to non-GPI anchored malaria antigens. Because no recombinant proteins of non-GPI anchored P. berghei antigens are available, we utilized Triton X-114 (TX-114) phase separation of total parasite lysates to exclude comprehensively all integral membrane proteins. Sera from infected wild-type and CD1–/– mice showed strong reactivity to a range of malarial cytosolic antigens present in the TX114-depleted fraction. In contrast sera from MHCII–/– mice were unable to recognize these antigens.

3 Discussion

Fig. 6. Lack of CD1d does not impair B cells to become activated in response to malaria infection. Spleen cells from naive or P. berghei-infected wild-type and CD1–/– mice were stained with PE-conjugated anti-B220 together with FITCconjugated anti-CD69 (top panels), FITC-conjugated antiB7–1 (middle panels) or FITC-conjugated anti-B7–2 (bottom panels) antibodies. The percentage of double positive cells was analyzed by flow cytometry.

help for antibody formation against GPI-anchored antigens [11]. It is therefore possible that the reduced antibody response found in CD1–/– mice results in part from the lack of the CD1/NKT cell pathway of antibody production to GPI-anchored proteins. To investigate that possibility, wild-type, CD1–/– mice and MHC class II-/mice were infected with P. berghei, drug-cured and rechallenged 3 weeks after the first infection and the antibody responses against a MSP-119 glutathione S transferase (GST) fusion protein were assessed by ELISA. Sera collected from wild-type mice had modestly but significantly higher anti-MSP-119 antibody levels compared to CD1–/– mice 7 days after the first infection and 14 days after the second challenge (day 35). These data

Splenomegaly is a common phenomenon associated with malaria infections both in humans and experimental rodent models. Hyperreactive malarious splenomegaly (HMS) in humans is a clinically and serologically defined anomaly, which is believed to represent an aberrant or excessive immunological response to malarial parasites and is a prognostic indicator of poor outcome [17]. HMS is associated with high titers of anti-malarial antibodies and elevated IgM serum levels. A high rate of B cell proliferation has been demonstrated in HMS resulting in an expanded population of activated B cells [18]. Moreover, it has been proposed that HMS predisposes to the development of splenic lymphoma with villous lymphocytes, a B cell lymphoproliferative disorder arising in the splenic marginal zone [19]. The molecular etiology of this disorder is not understood. In this study we have shown that CD1d deficient mice display a considerably reduced splenomegaly associated with a reduced expansion of the splenic B cell population in response to infection, indicating that the CD1/NKT cell pathway contributes to the development of this disorder in murine malaria. Whilst the precise relationship to HMS of the rodent splenomegaly described here remains to be elucidated, it is clear that this murine model recapitulates several of the distinctive features associated with malarious splenomegaly in humans.

2594

D. S. Hansen et al.

Eur. J. Immunol. 2003. 33: 2588–2598

Fig. 7. Differential Th1/Th2 bias in P. berghei-infected wild-type and CD1–/– mice. Spleen cells from P. berghei-infected wild-type or CD1–/– mice were stimulated for 3 days with anti-CD3 antibody. Cells cultured in medium alone were used as background controls. IFN- + (A) and IL-4 (B) levels in cell culture supernatant were assessed by capture ELISA. Each experiment is representative of four separate infections. Each point represents the mean of three samples ± SE. *p X 0.05, **p X 0.01, ***p X 0.005 between wild-type and CD1–/– mice.

The spleen and in particular the splenic marginal zone is an important presentation site for blood-borne antigens such as malaria products associated with erythrocytic stage infection. A subset of B cells, representing 10% of

the total splenic B cell pool, expresses high levels of CD1 and can be found in the splenic marginal zone [20]. This B cell subset is an important source of antibodies to Tindependent Ag [21] and is postulated to interact with

Fig. 8. IgG responses against MSP-119 in P. berghei-infected wild-type, CD1–/– and MHCII–/– mice. (A, B) Groups of eight wildtype, CD1–/– and MHCII–/– mice were infected with P. berghei, drug-cured and re-challenged 3 weeks after the first infection. Serum samples were collected at different time-points p.i and the IgG titers against an MSP-119-GST fusion protein were assessed by ELISA. *p X 0.05 between wild-type and CD1–/– mice; *p X 0.05, **p X 0.005 between MHCII–/– at day 0 and after infection. (C) Reactivity of sera from wild-type, CD1–/– and MHCII–/– mice with MSP-119-GST fusion protein in Western blot. MSP-119GST fusion protein or GST were run in an SDS-PAGE using 12% gels and transferred onto nitrocellulose. A TX-114-depleted phase containing only P. berghei soluble and non-GPI anchored proteins (cytosolic fraction) was included as control. After incubation with anti-GST monoclonal antibody or pooled sera from P. berghei-infected mice, protein bands were detected with an anti-mouse IgG conjugated to alkaline-phosphatase followed by development with NBT and BCIP.

Eur. J. Immunol. 2003. 33: 2588–2598

CD1-restricted NKT cells [15]. As shown in Fig. 4, this population markedly decreases in response to malaria infection, suggesting that the absence of CD1high B in CD1–/– mice was not responsible for the reduced splenomegaly and B cell expansion. It remains to be to determined whether CD1high B cells are deleted or migrate out of the spleen to the periphery. Alternatively, the levels of CD1d expression on B cells could be down-regulated. In support of that proposition, recent gene expression studies from our laboratory have indicated that CD1d is down-regulated in spleens of P. berghei-infected mice (Sexton et al., submitted for publication). Downregulation of CD1 expression on the surface of antigen presenting cells has been documented in infection with Mycobacterium tuberculosis and has been postulated as an escape mechanism utilized by the intracellular pathogen to prevent immune recognition [22]. Lack of CD1d did not diminish the capacity of B cells to become activated in response to infection. Rather than a deficiency in B cell activity, the reduced B cell-mediated immune response observed in CD1–/– mice appears to reflect a functional involvement of CD1d-restricted NKT cells in response to infection. In fact, it has been proposed that the rapid cytokine production by CD1-restricted NKT cells may induce bystander activation of other lymphoid lineages, including B cells. In support of that proposition, we have previously found [10] and confirmed in the present study that BALB/c CD1d-restricted NKT cells provide help for the development of Th2 responses. Murine malaria infections are characterized by an initial Th1 response which switches to Th2 after 7 to 10 days of infection [22, 23] and our data indicates that this process is regulated by CD1d-restricted NKT cells [10]. Thus in wild-type mice, it is reasonable to postulate that the increased B cell mediated-immune responses to infection correlate with the ability of CD1-restricted NKT cells to promote development of Th2 responses. Consistently, CD1d-deficient mice had enhanced parasite-specific IgG2a antibodies correlating with increased production of IFN- + , which is a critical factor in the development of IgG2a [24]. We have recently shown that the immunological properties of CD1-restricted NKT cells vary depending on the expression of molecules encoded by the natural killer complex (NKC) [10]. NKT cells from congeneic BALB.B6Cmv1r mice, in which the region of chromosome 6 containing the NKC from C57BL/6 has been introduced onto the BALB/c background, produce more IFN- + [10] and show reduced splenomegaly and parasite-specific antibody responses compared to wild-type mice (unpublished observations). This evidence further supports the view that it is the immunological environment resulting from cytokine production by NKT cells rather than expression of CD1 which is involved in the regulation of B cellmediated immune responses to infection.

Regulation of B cell responses in malaria by NKT cells

2595

We have previously shown that CD1d-deficient mice have reduced antibody responses against P. berghei circumsporozoite (CS) protein [11]. However, results obtained by others [25] have failed to support our observations. In the present study we have found that antibody production against the GPI-anchored protein MSP1 is significantly reduced in CD1–/– mice only during early time points after parasitic challenge whereas no differences between CD1–/– and wild-type mice were found when the anti-MSP-1 responses were evaluated after 3 or more weeks of infection. This result suggests that the CD1/NKT cell contribution to antibody formation operates early after exposure to antigen and kinetic differences in the evaluation of antibody response might explain the contrasting evidence on the response to CS protein. In addition, it has also been found that CD1–/– mice showed increased susceptibility to Borrelia burgdorferi infection which correlates with impaired antibody responses to lipoproteins responsible for mediating disease [26]. Moreover, human double-negative T cells from systemic lupus erythematosus patients provide help for IgG formation in a CD1c-restricted fashion [12] and CD1high B cells have been identified as the main source of rapid IgM auto-antibody production playing a role in the pathogenesis of hereditary lupus in NZB/NZW mice [27]. Together, all these lines of evidence suggest an alternative mechanism to MHCII-mediated T cell help for immunoglobulin production against non-peptide antigens or lipidated antigens. Mice lacking MHCII molecules are reported to be incapable to mount any IgG responses to T-dependent antigens [16]. In the present study we have found that MHCII–/– mice produced specific antibodies against MSP-1 during P. berghei infection. These data support our previous observations [11] which suggest that antibody production against GPI-anchored proteins can arise by MHCII-dependent and MHCII-independent pathways. Previous studies have shown that § -GalCer activated NKT cells induce rapid activation of other cell lineages including NK cells [5], CD4+ T cells [7], CD8+ T [7] cells and B lymphocytes [9]. Therefore, NKT cells have been proposed to act as a bridge between the innate and acquired immune system [6]. In the present study we have extended those observations utilizing an infection model of rodent malaria. We provide evidence that the CD1/NKT cell pathway might promote the development of B cell-mediated response to malaria infection by at least two different mechanisms: it induces a generalized enhancement of B cell-mediated responses by inducing a Th2 polarization of the immune system and appears to contribute to antibody formation against GPI-anchored proteins. Further studies are required to fully elucidate the significance of the CD1/NKT cell pathway of antibody formation in response to infection, particularly

2596

D. S. Hansen et al.

whether this arises through cognate or non-cognate mechanisms.

4 Materials and methods 4.1 Mice and infections Eight-to-twelve-week-old BALB/c wild-type, BALB/c CD1–/– (F10 generation) [28] and MHC class II–/– (F15 generation) [29] mice were injected i.p with 1×106 P. berghei ANKA-infected erythrocytes. In some experiments mice were treated with chloroquine (10 mg/kg) and pyrimethamine (10 mg/kg) for 5 to 7 days and re-challenged with 1×106 P. berghei ANKAinfected erythrocytes.

4.2 Flow cytometry Spleen cells from BALB/c wild-type or CD1–/– mice were incubated with anti-CD16 antibody, washed and stained for 1 h on ice with anti-B220-CyChrome or PE-conjugated alone or together with anti-CD1d-PE, anti-CD69-FITC, antiB7-1-FITC or anti-B7-2-FITC for detection of B cells. AntiCD8-FITC and anti-CD4-PE were used for detection of T lymphocytes (all antibodies from PharMingen, San Diego, CA). After washing twice, the cells were resuspended in PBS and analyzed in a FACScan cytofluorometer (Becton Dickinson, Grenoble, France) using CellQuest software. Viable lymphocytes were gated by forward and side scatter.

4.3 BrdU incorporation assay P. berghei-infected BALB/c wild-type and CD1–/– mice were maintained ad libitum for 2 weeks on light-shielded, BrdUsupplemented drinking water (0.8 mg/ml) replaced every 2–3 days. Splenocyte suspensions from individual mice collected at different time points post-infection were incubated with anti-CD16, washed and stained with anti-B220-PEconjugated antibody as described above. For staining of incorporated BrdU, cells were resuspended in ice-cold 0.15 M NaCl and fixed by incubation with 95% ethanol for 30 min on ice. Cells were then washed and incubated with 1% paraformaldehyde, 0.01% PBS-Tween for 30 min at room temperature. After washing, the cells were treated with 50 Kunitz U/ml DNase I (Sigma) and stained with anti-BrdUFITC conjugated antibody (BD Bioscience, San Diego, CA). Cells were then resuspended in PBS and analyzed by FACS.

4.4 CFSE staining and adoptive transfer Splenocytes from BALB/c wild type mice were incubated Tris-NH4Cl buffer to lyse red blood cells. Splenic B cells were then purified by depletion of T cells with anti-CD4, anti-CD8 and anti-Thy1.2 Dynabeads following the manufacturer’s instructions (Dynal Biotech, Norway). Adherent cells were

Eur. J. Immunol. 2003. 33: 2588–2598 then depleted by binding to plastic. The resulting population was 93–95% B220+ as assessed by FACS. Cells were resuspended at a density of 1×107 cells/ml in PBS containing 10 ? M CFSE (Molecular Probes, Eugene, OR). Cells were incubated at 37°C for 30 min and washed three times with complete RPMI-1640 medium, 5% fetal calf serum. Recipient mice were then injected i.v with 5×107 cells. Four hours later mice were infected with P. berghei ANKA (1×106). CFSE staining was assessed by flow cytometry at day 6 p.i on gated B220+ cells.

4.5 P. berghei ANKA lysate preparation Blood collected from P. berghei ANKA-infected mice (10 ml) was diluted 1:2 in RPMI-1640 medium and passed through a Whatman CF-11 cellulose column. The erythrocytes were eluted by washing the column with two volumes of RPMI1640 medium. The purified erythrocytes were centrifuged at 2,000 rpm for 5 min and trypsinized for 10 min at 37°C to remove mouse antibodies bound to cell membranes. After washing three times with RPMI-1640 medium, the erythrocytes were lysed with PBS-0.05% saponin and centrifuged at 10,000 rpm for 10 min. The pellet was washed and resuspended in PBS. The parasites were disrupted by five cycles of freezing-thawing and centrifuged for 5 min at 2,000 rpm. The supernatant was stored at –20°C until use.

4.6 Purification of P. berghei cytosolic proteins by TX114 partitioning A P. berghei lysate was resuspended in ice-cold TBS containing protease inhibitors (Complete™, Boheringer Mannheim). Precondensed TX-114 was added at a final concentration of 2%. The lysate was incubated for 15 min on ice with occasional mixing. The mixture was then centrifuged 10 min at 10,000×g at 4°C and the supernatant was warmed to 37°C in a water bath until the solution became cloudy. The mixture was centrifuged for 10 min at 1,000×g at room temperature. The detergent-depleted phase containing soluble proteins was collected and stored at –20°C until use.

4.7 Generation of MSP-119 glutathione S transferase fusion protein (MSP-119-GST) The DNA sequence corresponding to the MSP-119 fragment lacking the GPI anchor sequence (amino acid Gly 1672-Ser 1766) was amplified from P. berghei ANKA genomic DNA using the oligonucleotides PbM19eF (5’-CGCGGATCCGGTATAGACCCTAAGCATGTATG) and PbM19eR (5’-GGAAGATCTTAGCTACAGAATACACCATCATAAT). The resulting PCR product was ligated into the BamHI site of pGEX-4T-1, expressed as a GST fusion protein in Escherichia coli and purified using glutathione-Sepharose as described by the manufacturer (Amersham Pharmacia Biotech).

Eur. J. Immunol. 2003. 33: 2588–2598 4.8 ELISA for detection of malaria-specific antibodies Microtiter plates were coated with P. berghei ANKA lysate (5 ? g/ml) or MSP-119-GST fusion protein (2 ? g/ml) in carbonate buffer pH 9.6 by overnight incubation at 4°C. Empty sites were blocked with 5% skim milk for 1 h at 37°C. After washing three times with 0.05% Tween 20 in PBS, plates were incubated with different antisera in serial 1:2 dilutions for 1 h at 37°C. The plates were washed three times and incubated with a peroxidase-conjugated rabbit antimouse antibody (Pierce, Rockford, IL). The isotype titers were determined by incubating for 1 h at 37°C with rabbit anti-mouse antibodies against IgM, IgG1, IgG2a, IgG2b, IgG3. The plates were then washed three times and incubated with a peroxidase-conjugated goat anti-rabbit antibody (Pierce). In all ELISA, the bound complexes were detected by reaction with tetramethy-benzidine (KBL, Maryland) and H2O2. Absorbance was read at 450 nm.

4.9 Western blotting P. berghei MSP-119-GST fusion protein, GST and a P. berghei cytosolic fraction were run in a SDS-PAGE using 12% gels and transferred onto nitrocellulose filter. The filter was blocked with 5% skimmed milk for 1 h at 20°C and incubated for 1 h with the different antisera or with an anti-GST monoclonal antibody. After washing three times with PBS 0.05% Tween-20, bound antibody was detected with a rabbit anti-mouse alkaline phosphatase conjugate (Pierce). Bands were developed with p-nitro blue tetrazolium chloride (NBT) and 5-bromo-4-chloro-3-indoylphosphate p-toluidine salt (BCIP) (Bio-Rad Laboratories, Richmond, CA). The reaction was stopped with distilled water.

4.10 In vitro T cell stimulation Spleen cells from BALB/c wild-type and CD1–/– mice (n=3) were collected at different times p.i. with P. berghei ANKA. The cells were suspended in complete RPMI-1640 medium, 5% fetal calf serum and seeded in 96-well plates at a density of 2×106 cells/ml. The cells were then stimulated in triplicates for 3 days with anti-CD3 (5 ? g/ml, PharMingen, San Diego, CA). Cells cultured in medium alone were used as background controls. The cell culture supernatants were collected to measure the production of IL-4 and IFN- + by capture ELISA.

4.11 ELISA for IL-4 and IFN- .+ detection The following pairs of antibodies were used: 11B11 for capture and BVD6–24G2 for detection of IL-4 and R4–6A2 for capture, and XMG1–2 for detection of IFN- + (all antibodies from PharMingen, San Diego, CA). Ninety-six-well plates were coated with capture antibody by overnight incubation at 4°C in phosphate buffer pH 9. Plates were blocked with

Regulation of B cell responses in malaria by NKT cells

2597

1% BSA for 1 h at 37°C. Splenocyte culture supernatants were tested in duplicate by overnight incubation at 4°C. The plates were then incubated for 3 h at 20°C with the biotinylated antibody followed by a 2 h incubation at 20°C with streptavidin-peroxidase conjugate (Pierce). Bound complexes were detected by reaction with tetramethyl-benzidine (KBL, Maryland) and H2O2. Absorbance was read at 450 nm. The cytokine concentration in samples was calculated as pg/ml using recombinant cytokines (PharMingen) for the preparation of standard curves.

4.12 Statistical analysis A paired-sample Student’s t-test was used for data evaluation.

Acknowledgements: Supported by the UNDP/World Bank/WHO Special Program for Research and Training in Tropical Diseases, NIH grant AI-45548, the NH&MRC, and the HFSP. L. S. and B. S. C. are International Research Scholars of the Howard Hughes Medical Institute.

References 1 Arase, H., Arase, N., Ogasawara, K., Good, R. A. and Onoe, K., An NK1.1+ CD4+8– single-positive thymocyte subpopulation that expresses a highly skewed T cell antigen receptor V g family. Proc. Natl. Acad. Sci. USA 1992. 89: 6505–6510. 2 Yoshimoto, T. and Paul, W. E., CD4pos, NK1.1pos T cells promptly produce interleukin 4 in response to in vivo challenge with antiCD3. J. Exp. Med. 1994. 179: 1285–1295. 3 Porcelli, S. A., The CD1 family: a third lineage of antigenpresenting molecules. Adv. Immunol. 1995. 59: 1–98. 4 Eberl, G. and MacDonald, H. R., Selective induction of NK cell proliferation and cytotoxicity by activated NKT cells. Eur. J. Immunol. 2000. 30: 985–992. 5 Carnaud, C., Lee, D., Donnars, O., Park, S.-H., Beavis, A., Koezuka, Y. and Bendelac, A., Cutting edge: cross-talk between cells of the innate immune system: NKT cells rapidly activate NK cells. J. Immunol. 1999. 163: 4647–4650. 6 Nishimura, T., Kitamura, H., Iwakabe, K., Yahata, T., Ohta, A., Sato, M., Takeda, K., Okumura, K., Van Kaer, L., Kawano, T., Taniguchi, M., Nakui, M., Sekimoto, M. and Koda, T., The interface between innate and acquired immunity: glycolipid antigen presentation by CD1d-expressing dendritic cells to NKT cells induces the differentiation of antigen-specific cytotoxic T lymphocytes. Int. Immunol. 2000. 7: 987–994. 7 Eberl, G., Brawand, P. and MacDonald, H. R., Selective bystander proliferation of memory CD4+ and CD8+ T cells upon NK T or T cell activation. J. Immunol. 2000. 165: 4305–4311. 8 Bendelac, A., Hunziker, R. D. and Lantz, O., Increased interleukin 4 and immunoglobulin E production in transgenic mice overexpressing NK1 T cells. J. Exp. Med. 1996. 184: 1285–1293. 9 Kitamura, H., Ohta, A., Sekimoto, M., Sato, M., Iwakabe, K., Nakui, M., Yahata, T., Meng, H., Koda, T., Nishimura, S., Kawano, T., Taniguchi, M. and Nishimura, T., Alphagalactosylceramide induces early B-cell activation through IL-4 production by NKT cells. Cell Immunol. 2000. 199: 37–42.

2598

D. S. Hansen et al.

10 Hansen, D. S., Siomos, M. A., Buckingham, L., Scalzo, A. A. and Schofield, L., Regulation of murine malarial pathogenesis by CD1d-restricted NKT cells and the natural killer complex. Immunity 2003. 18: 391–402. 11 Schofield, L., McConville, M. J., Hansen, D., Campbell, A. S., Fraser-Reid, B., Grusby, M. J. and Tachado, S. D., CD1drestricted immunoglobulin G formation to GPI-anchored antigens mediated by NKT cells. Science 1999. 283: 225–229. 12 Sieling, P. A., Porcelli, S. A., Duong, B. T., Spada, F., Bloom, B. R., Diamond, B. and Hahn, B. H., Human double-negative T cells in systemic lupus erythematosus provide help for IgG and are restricted by CD1c. J. Immunol. 2000. 165: 5338–5344. 13 Hirunpetcharat, C. and Good, M. F., Deletion of Plasmodium berghei-specific CD4+ T cells adoptively transferred into recipient mice after challenge with homologous parasite. Proc. Natl. Acad. Sci. USA 1998. 95: 1715–1720. 14 Brossay, L., Jullien, D., Cardell, S., Sydora, B. C., Burdin, N., Modlin, R. L. and Kronenberg, M., Mouse CD1 is mainly expressed on hemopoietic-derived cells. J. Immunol. 1997. 159: 1216–1224. 15 Makowska, A., Faizunnessa, N. N., Anderson, P., Midtvedt, T. and Cardell, S., CD1high B cells: a population of mixed origin. Eur. J. Immunol. 1999. 29: 3285–3294. 16 Cosgrove, D., Gray, D., Dierich, A., Kaufman, J., Lemeur, M., Benoist, C. and Mathis, D., Mice lacking MHC class II molecules. Cell 1991. 66: 1051–1066. 17 Bryceson, A., Fakunle, Y. M., Fleming, A. F., Crane, G., Hutt, M. S., de Cock, K. M., Greenwood, B. M., Marsden, P. and Rees, P., Malaria and splenomegaly. Trans. R. Soc. Trop. Med. Hyg. 1983. 77: 879. 18 Bates, I. and Bedu-Addo, G., Chronic malaria and splenic lymphoma: clues to understanding lymphoma evolution. Leukemia 1997. 11: 2162–2167. 19 Wallace, S., Bedu-Addo, G., Rutherford, T. R. and Bates, I., Serological similarities between hyperreactive malarial splenomegaly and splenic lymphoma in west Africa. Trans. R. Soc. Trop. Med. Hyg. 1998. 92: 463–467. 20 Amano, M., Baumgarth, N., Dick, M. D., Brossay, L., Kronenberg, M., Herzenberg, L. A. and Strober, S., CD1 expression defines subsets of follicular and marginal zone B cells in the spleen: beta 2-microglobulin-dependent and independent forms. J. Immunol. 1998. 161: 1710–1717.

Eur. J. Immunol. 2003. 33: 2588–2598 21 Martin, F., Oliver, A. M. and Kearney, J. F., Marginal zone and B1 B cells unite in the early response against T- independent blood-borne particulate antigens. Immunity 2001. 14: 617–629. 22 Stenger, S., Niazi, K. R. and Modlin, R. L., Down-regulation of CD1 on antigen-presenting cells by infection with Mycobacterium tuberculosis. J. Immunol. 1998. 161: 3582–3588. 23 Taylor-Robinson, A. W., Phillips, R. S., Severn, A., Moncada, S. and Liew, F. Y., The role of TH1 and TH2 cells in a rodent malaria infection. Science 1993. 260: 1931–1934. 24 Finkelman, F. D., Katona, I. M., Mosmann, T. R. and Coffman, R. L., IFN-gamma regulates the isotypes of Ig secreted during in vivo humoral immune responses. J. Immunol. 1988. 140: 1022–1027. 25 Molano, A., Park, S. H., Chiu, Y. H., Nosseir, S., Bendelac, A. and Tsuji, M., Cutting edge: the IgG response to the circumsporozoite protein is MHC class II-dependent and CD1dindependent: exploring the role of GPIs in NK T cell activation and antimalarial responses. J. Immunol. 2000. 164: 5005–5009. 26 Kumar, H., Belperron, A., Barthold, S. W. and Bockenstedt, L. K., Cutting edge: CD1d deficiency impairs murine host defense against the spirochete, Borrelia burgdorferi. J. Immunol. 2000. 165: 4797–4801. 27 Zeng, D., Lee, M. K., Tung, J., Brendolan, A. and Strober, S., Cutting edge: a role for CD1 in the pathogenesis of lupus in NZB/ NZW mice. J. Immunol. 2000. 164: 5000–5004. 28 Smiley, S. T., Kaplan, M. H. and Grusby, M. J., Immunoglobulin E production in the absence of interleukin-4-secreting CD1dependent cells. Science 1997. 275: 977–979. 29 Köntgen, F., Süss, G., Stewart, C., Steinmetz, M. and Bluethmann, H., Targeted disruption of the MHC class II Aa gene in C57BL/6 mice. Int. Immunol. 1993. 5: 957–964.

Correspondence: Diana S. Hansen, The Walter and Eliza Hall Institute of Medical Research, 1G Royal Parade, Parkville, Victoria 3050, Australia Fax: +61-3-93470852 e-mail: hansen — wehi.edu.au