Background
Non-lethal malaria infections in mice directly infected with blood stage parasites are characterised by parasitaemia sometimes exceeding 40% of infected erythrocytes and an acute inflammatory response [
1]. Much of pathology at this time is thought to be a consequence of the production of pro-inflammatory cytokines [
2,
3]. These cytokines can be induced by direct interaction between the parasite and dendritic cells, monocytes and macrophages [
4,
5] resulting in NK, γδ and Th1 CD4
+ T cell activation and the further release of cytokines such as IFN-γ, TNF-α and LT [
2,
6]. However, it is not known whether these strong pro-inflammatory responses are, in part, a result of high initial parasitaemia that may not occur when the infection is initiated by the natural route of mosquito infection, and also whether the pre-existing sporozoite and pre-erythrocytic forms affect in any way the blood stage infection or the host's immune response to it.
Sporozoites migrate rapidly to the liver where they invade hepatocytes and initiate pre-erythrocytic schizont development. A blood stage infection begins approximately two days later, after rupture of the mature liver schizont, and release of merozoites, which then invade erythrocytes and establish the erythrocytic cycle. This exposure of the host to malarial antigens and parasite Pathogen-associated Molecular Patterns (PAMPs) [
7], in an environment such as the liver, before the erythrocytic stage of the infection may well have an impact on the subsequent innate and acquired immune response to the blood stages. Although the liver is not a secondary lymphoid organ, it is likely to be a site where phagocytic cells, such as Kuppfer (cells (KC) and dendritic cells (DC), encounter and take up sporozoites. It does enlarge with multiple infections and is a site of phagocytosis of infected and uninfected red cells [
8]. The liver environment is considered to be tolerogenic [
9] and could, therefore, influence APC activation and presentation, and thus the nature and magnitude of the CD4
+ T cell response to those antigens seen later in the blood stages. The interactions of DC from the liver with malaria parasites have not been studied, but naïve KC are not activated by infectious sporozoites to produce IL-12p40 and antigen-presenting capacity is impaired [
10]. Since CD4
+ T cells are important for the development of protective immunity and contribute to pathology during blood stage infection, it is important to know if pre-erythocytic stage infections affect this response.
A primary blood stage infection of a clone of Plasmodium chabaudi, (ER), following direct injection of infected red blood cells (pRBC) was compared with a blood stage infection after natural transmission via inoculation of sporozoites by Anopheles stephensi mosquitoes. Blood parasitaemia and cytokine production was measured in splenic CD4+ T cells at early and later stages of a primary blood stage infection. Overall, the parasitaemia following sporozoite challenge were lower than those observed after direct blood inoculation. Despite this, there was an early IFN-γ response from CD4+ T cells, albeit reduced in comparison with the response to direct blood challenge. Interestingly, uninfected mosquito bites given prior to exposure to infected mosquitoes resulted in a delayed and lower parasitaemia accompanied by an increased IFN-γ response from CD4+ T cells.
Materials and methods
Mice and parasites
Female C57BL/6 mice, aged 6–8 weeks from Charles River (Margate, Kent, U.K.) or from the specific pathogen free breeding facilities at The National Institute for Medical Research (NIMR), London, were infected with P. c. chabaudi (ER) parasites, originally obtained from David Walliker (University of Edinburgh, UK). Parasitaemia were monitored by Giemsa-stained thin blood films.
Maintenance of mosquitoes
Anopheles stephensi (strain SD500) female mosquitoes bred at the insectary of the Cell and Molecular Biology Division, Biology Department at Imperial College for Science Technology and Medicine (London,) maintained as described [
11,
12] were kindly provided by Jacqui Mendoza. Groups of 50 mosquitoes in paper cups were given a solution of 80 g/l D (-) fructose (Sigma, U.K.) and 0.5 g/l p-amino benzoic acid (PABA) (Sigma, U.K.) in H
2O, placed inside a cooled incubator with the temperature maintained between 24–27°C, relative humidity higher than 50 % and a light cycle of 12 hours light/12 hours dark.
Infection of mosquitoes
Mice (group size of 5) were injected
i.p. with 10
5 parasitized erythrocytes. Six and fourteen days later, the presence of gametocytes in mice was determined through Giemsa-stained thin blood films. Mice were anaesthetized by ip injection of a mixture of Rompum (Bayer, Suffolk, UK) and Vetalar (Parke-Davis Veterinary, Gwent, UK) [
13], and placed on the net covering the top of the paper cups, ventrally exposed to the mosquitoes. The feeding session lasted for 20 min and every 5 min mice were transferred to other paper cups, to ensure that all the cups of mosquitoes would be exposed to all the mice and vice-versa. At the end of the feeding session, after removal of those mosquitoes that did not feed, mosquitoes were returned to the incubator.
Infection of mice by mosquito bite
To verify if the mosquitoes were infected, on day 8 and 10 after feeding mosquitoes on infected mice, the stomachs from 10 mosquitoes from several paper cups were dissected and examined under a microscope for the presence of oocysts. Fourteen days after feeding mosquitoes on infected mice, transmission of the parasites from those mosquitoes to uninfected mice was carried out, but only after confirming that sporozoites were present in the salivary glands of a sample of 10 mosquitoes by mosquito dissection. The feeding session proceeded as described above.
Dissection of salivary glands and sporozoite injection
Mosquito salivary glands, isolated under a binocular magnifier were pooled and disrupted using a glass homogeniser (Merck, UK). The number of sporozoites was calculated using an Improved Neubauer chamber (Weber Scientific International Ltd, UK) and different numbers of sporozoites were then injected into the tail vein of mice.
Preparation of cell suspensions for surface and intracellular staining
Single cell suspensions of spleens were added erythrocyte lysis buffer containing 0.16 M NH4Cl (Sigma, UK) and 0.17 M Tris (Sigma) pH 7.65, at a final cell concentration of approximately 107 cells/ml, and incubated for 10 min at room temperature. Cells were then centrifuged and the total number of cells obtained was counted using an Improved Neubauer chamber. the cells were then placed in 96 wells plates (NUNC, UK) at a concentration of 2 × 105 cells per well.
Stimulation of cells
Cells were stimulated using Phorbol 12-myristate 13-acetate (PMA) and ionomycin (Calbiochem, UK) at final concentrations of 50 ng/ml and 500 ng/ml respectively and incubated for two hours at 37°C and 5 % CO2. Cells were then further incubated for two hours with Brefeldin A (Sigma) at a final concentration of 10 μg/ml
Surface and intracellular staining of cells
Antibodies to CD4, IL-2, IL-4, IL-10, IFN-γ and appropriate isotype controls labeled with FITC or PE were obtained from BD Bioscience, UK. Surface and intracellular staining was carried out in the presence of normal rat IgG and anti-FcR block (BD Bioscience, UK) to prevent nonspecific and FcR-mediated binding of Ig as described previously [
14]. Cells were acquired on a FacsCalibur, (BD Bioscience). 50,000 viable cells (gated on 90° and forward light scatter for viable lymphocytes) were acquired and analysed using Cell Quest software (BD Bioscience).
Discussion
Transmission of P. c. chabaudi (ER) through the natural route of infection results in an erythrocytic stage infection with a delay in patency, lower peak parasitaemia and shorter duration of an acute infection compared with an infection initiated by direct injection of pRBC. Despite these differences, increases in spleen cellularity, and induction of IFN-γ producing CD4+ T cells, albeit at lower levels, were still observed in the acute infection.
The delay in patency and peak parasitaemia was, in general, consistent with the length of time needed to complete the hepatic cycle, which has been shown to last approximately 52 hours in
P. chabaudi (AS) [
16]. The appearance of parasites in the blood circulation later than two days is likely to have been due to injection of the lower numbers of sporozoites, as the numbers of sporozoites present in infected mosquitoes were very variable.
As direct blood challenge is the accepted route of infection for the study of host responses to blood-stage malaria infections, it is important that this reflects the natural infection as far as possible. Splenomegaly is a characteristic of malaria infections, observed in human infections and rodent models [
17‐
19]. In mice, this is thought to be due in part to the haematopoetic response to the high numbers of RBC lost through infection, and to the trapping of leukocytes in the spleen [
20‐
23]. Despite the reduced numbers of pRBC and the shorter duration of the infection, the increase in the cellularity of the spleen was also observed in mice submitted to infectious mosquito bites. This increase was comparable with that seen in mice injected with pRBC, suggesting that factors other than or, in addition to, parasite load play a role in this process.
Interestingly, increases in numbers of CD4+ T cells did not follow the same pattern and were somewhat reduced in infections initiated by mosquito transmission, perhaps reflecting the reduced PAMP or antigenic stimulus of lower parasitaemia. The lower numbers of CD4+ T cells producing IL-2 would support this possibility. In addition to the smaller increase in CD4+ T cells, more importantly there was a greater reduction in CD4+ T cells producing IFN-γ compared with IL-4 producing cells. This suggests that there may be less of a bias towards Th1 responses in these mice, than is seen in mice infected directly with pRBC.
Sporozoites injected by mosquitoes develop initially in the liver, an organ often described as tolerogenic or immunoprivileged [
9]. It is possible that uptake of parasite material by dendritic cells or Kupffer cells occurs in the relative absence of inflammation resulting in immuno-regulation rather than activation. A population of B220+ DC have been identified in liver that may have such immunoregulatory properties [
24,
25] and Kupffer cells from naïve and sporozoite-infected mice have been shown to produce low or negligible amounts of IL-12p40 [
10]. Although antigen-presentation and T cell activation of liver-residing pathogens is likely to take place in lymphoid organs in primary infections, there are increased numbers of Kupffer cells and lymphocytes in the liver during infection, and phagocytic cells of the liver could transport parasite antigens to lymph nodes. This may well influence subsequent presentation to, and activation of CD4
+ T cells, impacting on those CD4
+ T cell responses to antigens shared between liver and blood stages of
Plasmodium. A detailed analysis of the cytokine profiles of CD4
+ T cells specific for parasite antigens expressed at both liver and blood stages, such as MSP-1 [
26], which are activated after mosquito-transmitted infections and blood stage infections would address this. In addition to the lower numbers of IFN-γ
+ CD4
+ T cells, there were also fewer IL-10
+ CD4
+ T cells. IL-10 is known to reduce the inflammatory response in this infection, mainly characterized by the production of IFN-γ [
15]. These results suggest that the lower parasitaemia stimulates less of an inflammatory response and consequently less of a counter-regulatory response. Since much of the pathogenesis of acute blood stage malaria in this model has been ascribed to pro-inflammatory responses induced in part by IFN-γ from Th1 CD4
+ T cells [
1], it will also be important to determine the extent of the pathological sequelae in mosquito-transmitted infections.
In this study, mosquito bites of uninfected
Anopheles on their own exert an effect over the development of the parasite subsequently injected into the mouse via infectious mosquitoes. This is reflected in later patency of infection, lower peak parasitaemia, and increases in the number of IFN-γ
+ and IL-4
+ CD4
+ T cells in the blood stage of infection in mice given first non-infectious bites. These results indicate that uninfected mosquito bites can potentiate the immune response elicited by the later infectious mosquito bites. These observations are in line with a previous study showing that there is an increased antibody response in animals bitten by anopheles mosquitoes [
27].
These effects may be exerted by components within mosquito saliva injected while blood feeding either by stimulating innate inflammatory responses that enhance T cell activation or by inducing host responses that prevent mosquitoes from remaining at the site of the bite. In line with the study described here, saliva and salivary gland proteins from ticks and sand flies have been shown to interfere with the host immune response taking place during a new parasite infection [
28‐
30]. Hyperimmune serum containing high levels of antibodies against uninfected tick saliva components injected into an animal, also prevented ticks from sticking to the skin and led to the inhibition of tick-borne encephalitis virus replication in them [
31]. In the same way, mice previously exposed to bites by uninfected sand flies are significantly protected against
Leishmania infection [
32].
However, this is not a general finding that can be extended to all insect/tick borne infections. In contrast to the observations in this study, salivary gland extracts from
Aedes aegypti have been shown to cause a reduction in splenic cell proliferation and in the production of Th1 and Th2 cytokines [
33]. Tick, sand fly and black fly (
Simulium vittatum) salivary components also can have suppressive effects on the host immune and inflammatory response [
34‐
38]. It was proposed that this allows the vectors to remain on the host for long periods of time and enhances transmission and establishment of the pathogens, although such reductions have not been observed using Culicinae mosquito
Culex quinquefasciatus [
39].
Saliva from blood sucking arthropod vectors has several components, such as anti-platelet aggregatory, anticoagulatory and anti-vasoconstrictory substances [
40,
41]. The differences in the effects of bites on the host response may therefore be because the relevant components are not present in all types of insect or tick vectors. This certainly requires further investigation.
In summary, direct blood challenge may not fully reflect the early host responses that take place after mosquito bites, or after repeated bites by uninfected mosquitoes, the normal situation for people living in endemic malaria areas. Therefore, it is important that detailed studies of early host responses should include experiments where transmission by the natural route takes place, especially when testing or comparing antigens as potential vaccine candidates.
Authors' contributions
LF carried out and monitored all infections, participated in experimental design and planning, performed the flow cytometry assays and helped draft manuscript.
ES participated in the flow cytometry assays
GB was responsible for the maintenance of mosquitoes and developing the mosquito transmission studies.
JL conceived the study, participated in its design and drafted the final manuscript.
All authors read and approved the manuscript before submission.