Parasite load stemming from immunization route determines the duration of liver-stage immunity

Abstract

Immunization with radiation-attenuated Plasmodium sporozoites (RAS) induces sterile and long-lasting protective immunity. Although intravenous (IV) route of RAS-immunization is reported to induce superior immunity compared to intradermal (ID) injection, its role in the maintenance of sterile immunity is yet to be understood. We investigated if the route of homologous sporozoite challenge of Plasmodium berghei (Pb) RAS-immunized mice would influence the longevity of protection. C57BL/6 mice immunized with Pb-RAS by IV were 100% protected upon primary IV/ID sporozoite challenge. In contrast, ID-immunization resulted in 80% protection, regardless of primary challenge route. Interestingly, the route of primary challenge was found to bring difference in maintenance of sterile protection. While IV Pb-RAS immunized mice remained protected at all challenges regardless of the route of primary-challenge, ID Pb-RAS immunized mice receiving ID primary-challenge became parasitemic upon secondary IV-challenge. Significantly, primary IV-challenge of Pb-RAS ID immunized mice resulted in 80% and 50% survival at secondary and tertiary challenge, respectively. According to phenotypically diverse liver CD8⁺ T cells, the percentages and the numbers of both CD8⁺ T effector memory and resident memory cells were significantly higher in IV than in ID Pb RAS immunized mice. IFN-γ producing CD8⁺ T cells specific to Pb TRAP₁₃₀ and MIP-4-Kb-17 were also found significantly higher in IV-mice than in ID-mice. The enhanced T cell generation and the longevity of protection appear to be dependent on the parasite-load during challenge when infection is tolerated under suboptimal CD8⁺T-cell response.

Introduction

Residents of malaria endemic areas typically do not develop sterilizing and lasting immunity despite repeated exposures to bites from Plasmodium infected mosquitoes. By contrast, multiple exposures of mice ^1–3 and humans ⁴ to Plasmodium berghei (Pb) or Plasmodium falciparum (Pf) radiation-attenuated sporozoites (RAS), respectively, induce sterile and lasting protection against homologous sporozoite (spz) challenge. Like infectious spz, RAS invade hepatocytes where they express liver stage antigens (LS Ag) but fail to develop into blood stage parasites. Although the mechanisms of protection are not fully understood, the accumulation of LS Ags is thought to be essential for the induction and maintenance CD8⁺ T cells with effector and memory attributes required for lasting sterile protective immunity ². The availability of mouse models of sterile and lasting protective immunity induced with both Pb RAS ^2,3 and P. yoelii (Py) RAS ⁵ have been quite useful for investigating the various parameters of RAS immunization regimen including the routes of vaccination, duration of protection, as well as immune responses that correlate with protection.

Protective immunity induced by multiple IV immunizations of C57BL/6 mice with Pb RAS ^2,3 renders sterile and protracted protection against multiple challenges. In contrast to the IV RAS immunization regimen, which results in nearly 100% protection, exposure of mice to RAS by other routes including intradermal (ID) or subcutaneous (SC), yields much lower and varying degrees of protection ^6–9. Similar results were observed with ID administration of Pf RAS to human subjects ¹⁰. The dose as well as the volume of RAS administered by IV or ID immunization also impacts the degree of protection ^9,11,12. The mechanisms responsible for the varying levels of protection induced by RAS administered by the different routes is not fully understood. Results from a recently published study demonstrate that a single RAS dose by ID route induces increased level of regulatory responses that might be interfering with the induction of protection ¹³. According to other studies ⁶, it is likely that in contrast to the IV route, other routes of immunization fail to deliver a sufficient number of spz for adequate accumulation of LS Ags load needed for the induction of 100% sterile protection. Guided by this observation, in the present study we asked whether the challenge route with infectious spz administered to RAS immunized mice would affect the degree ⁹ as well as the longevity of protection. Our present study expands these observations by showing the induction of phenotypically distinct liver CD8⁺ T cell such as effector memory (EM) and tissue resident memory (RM) T cells as well as Ag-specific IFN-γ producing CD8⁺ T cells are clearly associated with the protective immunity induced by Pb RAS. On the basis of our results, optimized challenge dose of infectious sporozoite delivered through IV but not ID route of inoculation enhances the longevity of sterile protection against Pb infection.

Results and Discussion

The route of RAS immunization and the timing of challenge affect protection

It has been accepted that RAS IV immunization of humans and mice induces sterile immunity against spz challenge, whereas RAS ID immunization results in significantly decreased protective efficacy or fails to induce even short-lived protection ^6,9,10. In our preliminary experiments, we immunized C57BL/6 mice with 3 biweekly doses of Pb RAS (75K, 20K, 20K) by either ID or IV routes, and 9 days after the last boost immunization we challenged mice with 10K infectious spz by IV inoculation (Fig.1A) and determined parasitemia by thin blood smear taken daily for 15 days. Mice that received Pb RAS by IV were fully (100%) protected, while protection decreased to 80% in mice that received Pb RAS by ID route (Fig. 1B). This observation partly confirms findings of others in that RAS vaccination administered by ID results in a decreased protective efficacy. In contrast to the small decrease in protection we observed here, others have reported much reduced protection in that only 7% −13% mice survived spz challenge after ID RAS immunization. This difference in protection may stem from a longer interval 3 – 4 weeks between last boost immunization and spz challenge ⁶. It is entirely possible that innate immune responses induced within a short period between boost and challenge contributed to the insignificant differences in protection that we observed when mice were challenged 9 days after last boost with Pb RAS. Therefore, in the next set of experiments we delayed infectious spz challenge of both IV and ID Pb RAS immunized mice to examine if any changes in level of sterile protection could be observed. As shown in Fig. 1C, we delayed challenging the mice until day 18 post Pb RAS vaccination. As observed previously, all Pb RAS IV immunized mice were protected (100%) upon delayed challenge, while mice that received 3 doses of Pb RAS by the ID route succumbed to parasitemia on day 7 after the challenge (Fig.1D). Naïve control mice that were challenged with infectious spz became parasitemic on day 4 after the challenge (Fig.1D). Similar to observations made by others, delaying the challenge resulted in parasitemia only amongst mice vaccinated with Pb RAS by ID, hence confirming that ID delivery of Pb RAS is inferior to the IV route of immunization for inducing sterile protection ⁶.

Figure 1: The route of Pb RAS administration and the timing of challenge dictate the level of protection.

C57BL/6 mice were immunized thrice with Pb RAS (75K, 20K, 20K) through ID or IV route at 2 weeks interval and challenged with 10K infectious sporozoites on day 9 (A, B) or 18 post last-vaccination (C, D). Protection was determined by measuring blood-stage parasitemia until day 15 post challenge. Kaplan-Meier survival analyses show % mice with/out parasitemia. n=10–20 mice per group. The data are representative of three independent experiments.

The parasite-load in liver regulates magnitude and longevity of protection

RAS-induced lasting and sterile protection is linked to the dose and the number of immunizations as shown in humans ^14,15 and in mice ³ (Dalai unpublished ¹⁶) as well as a mouse strain ¹². For C57BL/6 mice this requirement constitutes 3 doses of Pb RAS against challenge with 1–10K infectious Pb spz. Lower Pb RAS doses induce but a transient protection (UK personal communication), likely because protection is dependent on the level of LS Ag available from the partly developed RAS. The confirmation that protection is dependent on the LS Ag depot came from observations made by one of us ² that protection induced by Pb RAS wanes in mice treated with primaquine, a drug that interferes with the formation of liver stage parasites and thus LS Ags. Moreover, long-term maintenance of Pb RAS-induced protection has been shown to depend on the persistence of liver central memory (CM) CD8⁺ T_CM cells ¹⁷ that upon challenge differentiate into CD8⁺ T_EM cells and swiftly produce IFN-γ.

Repeated exposure to bites from P. falciparum (Pf) infected mosquitoes assures some level of protective immunity against severe malaria infection amongst residents of malaria endemic areas, as maintenance of immunity is thought to be dependent on repeated antigenic challenge ¹⁸. Moreover, experimentally infectious spz challenge has also been shown to extend the vaccine induced sterilizing immunity in mice ^16,19. Interestingly, individuals living in high-parasite-transmission regions maintain more effective immunity than persons residing in moderate or low-transmission-regions^20–22. These reports indicate that parasite Ag load derived from exposure to infectious spz or blood stage Ags plays an instrumental role in the induction and the longevity of protection. We hypothesized that the maintenance of Pb RAS-induced protection may also be influenced by the liver Ag load that is enhanced by exposure of Pb RAS immunized mice to infectious challenge. We tested this hypothesis by challenging Pb RAS immunized mice through ID or IV routes. In the first group, mice that received Pb RAS via the IV route were challenged on day 9 post vaccination through either IV or ID routes and none developed parasitemia. This group of mice remained protected during secondary (day 226) and tertiary challenge (day 471) administered by IV (Fig.2A). Thus, IV vaccinated mice maintained 100% sterile protection regardless of their route of primary challenge (Fig.2B). Amongst the mice that received Pb RAS through ID vaccination, 80% were parasitemia free upon primary challenge administered either by IV or ID. Interestingly, while all the mice that received primary challenge by the IV route continued to maintain protection at secondary IV challenge, mice receiving primary challenge by ID became parasitemic upon secondary challenge (day 226) given by IV (Fig.2C). Amongst the surviving ID vaccinated and IV challenged mice, 50% remained sterile protected upon tertiary challenge on day 471 (Fig.2C). Our results suggest that the differential load of Ags derived from infectious spz challenge modulates the pre-existing Ag-specific response mounted to Pb RAS vaccination. Unlike IV challenge, liver Ag load resulting from ID challenge may be insufficient to amplify already suboptimal ID Pb RAS-induced CD8⁺ T cell response to the level needed to ensure complete protection.

Figure 2: The magnitude of parasite liver load at challenge affects protection.

C57BL/6 mice were immunized through ID or IV route as described in Figure 1. Primary sporozoite challenge was given through either IV or ID routes with 10K infectious sporozoites on day 9 post last boost vaccination. The surviving mice in each group, IV vaccinated (A, B) or ID vaccinated (A, C) received secondary and tertiary challenge sporozoite challenges (10K) only by IV on days 226 and 471, respectively. The mice that were negative for blood-stage parasitemia until day-15 post challenge were considered protected. Results (Fig. B, C) show % protected mice post challenge or re-challenge. n=10 mice per group. The data represent the mean of two independent experiments that yielded similar results.

Route of spz administration regulates the parasite-load in liver

Clearly, the route of delivery of infectious spz challenge affected the longevity of protection amongst IV versus ID RAS immunized mice. Pervious study analysing parasite bioluminescent signal from the parasite load present in the liver has shown that the difference in the efficacy of protection between IV and ID immunized mice is associated with number of spz that reach the liver¹¹. Because in this study the only a single dose of infectious spz was used for ID vs IV delivery, in our study we tested if delivery of different doses of infectious spz by IV and ID would reflect the dose dependent differences in LS parasite load.

Naïve mice were inoculated with increasing doses, 100, 500, 1000 and 10,000 infectious spz through IV or ID, and 48 hr later liver parasite load was measured by q-PCR for 18S rRNA expression. According to the results shown in figure 3, increasing spz inoculum (100, 500, 1000, 10,000) through either ID or IV corresponded to a gradual increase of liver stage parasite load, hence decreased C_T value (Fig.3A). Importantly, mice receiving spz by IV showed significantly lower C_T value as compared to ID inoculated mice, irrespective of the inoculum size, suggesting that IV inoculated spz significantly contribute to the higher parasite-load in liver than ID delivered spz (Fig.3B). Interestingly, the ΔC_T value of mice receiving 1000 spz via IV route was comparable with the mice receiving 10,000 spz through ID route (Fig. 3A). It suggests that inoculation through ID route requires 10-fold higher number of spz than that of IV route to get comparable parasite liver load. Furthermore, the higher parasite liver load in IV inoculated mice was corroborated with higher blood-stage parasitemia. The IV inoculated mice showed significantly greater % of parasitized red blood cells (RBCs) that increased up to day 8 post-infection (Fig.3C). Our results confirm observations made by others ^6,9 and thus validate that spz infection of hepatocytes through ID route results in a lower liver parasite load as compared to with IV route of immunization/infection.

Figure 3: The route of infectious sporozoite inoculation modulates parasite load in liver.

Naive C57BL/6 mice were inoculated with 100, 500, 1000, and 10,000 infectious Pb sporozoites by either IV or ID route. Forty-eight hours post inoculation, parasite load in liver was determined by Pb 18s rRNA using q-PCR. The expression of β-actin gene in individual sample was used as an internal control. The data are presented as (A) ΔC_T (C_T of 18S rRNA -C_T of β-actin) and (B) Fold expression (2^-ΔCT). The C_T value is inversely proportional to the expression of specific gene. Means were compared by non-parametric Mann-Whitney’s U test (C) Percent parasitemia was determined by counting infected RBCs from Giemsa stained blood smears (from at least 30 microscopic field) during day 3 to 8 post-inoculation with infectious sporozoites. Means were compared by two-way ANOVA test followed by Sidak’s multiple comparisons test n=3 mice per group. p<0.05 is considered as a significant. The results represent one of the two independent experiments with similar results.

It is possible that in mice protected by IV RAS immunization, the infectious spz challenge itself supplied additional LS Ags that provided enhanced stimulus, be it to the innate cells or cells of the adaptive immune responses, mobilizing and activating the cellular network to engage in an effector function and memory responses. In instances of ID immunized mice, the number of spz reaching the liver may be decreased owing to spz death enroute to the liver. This might have been the cause of the failure of ID vaccination strategy during clinical trials with Pf RAS ¹⁰. In brief, on the basis of our findings it is possible that low level of liver parasite Ags in ID vaccinated mice may have contributed to the ineffectiveness of CD8⁺ T cell responses to maintain durable protection.

ID RAS immunization is inferior to IV route for the induction of liver CD8⁺ T cells

In general, sterile protection induced against pre-erythrocytic stage Ags e.g., circumsporozoite protein (CSP) or thrombospondin-related antigens protein (TRAP) is considered to be based on antibody ²³, CD4 T cell responses ^24–26 and as has been shown in murine models ^27–29 and non-human primates on CD8⁺ T cells ^14,30. Antibodies specific against CSP are crucial for neutralizing spz before they invade hepatocytes and enter the liver^6,7. An extensive analysis on the efficacy of RTS,S vaccine trial suggests that CSP-specific antibody-mediated response prevents more than 95% spz from entering the liver ³¹. Therefore, antibody mediated response may be insufficient to induce sterile protection against LS infection, because even a single surviving spz that enters the liver could result in the establishment of complete blood-stage infection³¹. Because antibodies do not typically access the spz after they have invaded the liver, other mechanism must control the development or even invasion of hepatocytes by spz ^2,14,15.

It is well established in the mouse model that Pb RAS delivered by IV induces CD8⁺ T cells in the liver ^{2, 32}, where they perform the key function which involves elimination of LS parasites, hence rendering sterile and lasting protection. The fine specificity of some of the CD8⁺ T cells involved in protective immunity against the liver stage have been identified. Accordingly, TRAP-₁₃₀ peptide-specific CD8⁺ T cells eliminate liver stage parasite possibly by in vivo cytolytic activity ²⁹, whereas Kb-17-specific CD8⁺ T cells appear to associate with protection induced by Pb RAS via IFN-γ production³³. We show here that both Kb-17 and TRAP-₁₃₀ recalled significantly stronger INF-γ responses in livers and spleens from IV than ID Pb RAS immunized mice (Fig.4A,B). Kb-17-recalled responses in the spleen of IV immunized mice were somewhat lower than in the liver, nonetheless, they were significantly higher than in ID immunized mice (Fig.4B). It is possible that decreased protection observed in ID immunized mice stems in part from significantly lower numbers of IFN-γ producing cells specific for Kb-17 and TRAP_130. Although the TRAP₁₃₀- and Kb-17-specific CD8⁺ T cells may not represent the fundamental mechanism responsible for Pb RAS-induced protection, they are likely to be involved in the network of cellular protective immunity.

Figure 4. ID Pb RAS immunization induces inferior Ag specific IFN-γ responses in livers and spleens.

(A) Mice were immunized thrice with Pb RAS by IV or ID at 2 weeks interval as described in the legend for Figure 1. Three weeks after the last boost, livers and spleens from immunized and naïve mice were harvested, mononuclear cells were isolated and stimulated with TRAP-₁₃₀ or Kb-17 peptides, and IFN-γ responses were measured in the ELISPOT assay. The results expressed as the mean +SEM of responses from 5 individual mice/group and show the number of spots per 10⁶ cells. Cells stimulated with PMA/ionomycin served as a positive control. Two-way ANOVA test followed by Tukey’s multiple comparisons test were used. ****p<0.0001. Results are representative of one out of three experiments.

In a previously published study, it was shown that IV immunization with Pb RAS induces significantly higher level of CD8⁺ T_EM in livers and spleens than ID immunization, nonetheless, spz-specific IFN-γ response between IV and ID immunized mice were not significant ⁶. Having observed significantly different levels of IFN-γ between IV and ID immunized mice, we extended our analyses of splenic and liver CD8⁺ T cells arising in IV and ID Pb RAS immunized mice by evaluating CD8⁺ T cell subsets on the basis of phenotypic profiles. The expansion of CD8⁺ T cells expressing the T_EM phenotype (CD44^hiCD62L^lo) and to some extent the central memory T_CM phenotype (CD44^hiCD62L^hi) reflected the different routes of immunization and consequently the level of accumulated LS Ag load. The expansion of the CD8⁺ T cell subsets in the liver and the spleen was assessed at 3 weeks post RAS vaccination, thus corresponding closely to the time point of the delayed challenge. According to our collective results, the percentage of liver CD8⁺ T_EM cells were 2 -fold higher in IV vaccinated mice (p<0.0001) as compared to mice that received Pb RAS via the ID route (Fig.5A and Supplemental Figure S1A). This difference in IV vs. ID immunization is also reflected in the total number of CD8⁺ T_EM cells in the liver (Fig.5A). We noted no difference in the percentages, but did note a significant difference in the number of CD8⁺ T_CM in the liver of IV immunized mice (Fig.5C). In contrast, no significant differences in the % or the number of spleen CD8⁺T_EM cells (Fig.5B) or CD8⁺T_CM cells (Fig. 5D) were noted between IV or ID Pb RAS immunized mice (Supplemental Figure S2). These results align with previously observed levels of phenotypically diverse liver and spleen CD8⁺ T cells following Pb RAS immunization. They also underscore that either the absence of activated liver CD8⁺ T_EM cells or if activated, the level of liver CD8⁺T_EM cells was insufficient in ID immunized mice to mediate protection^3,6.

Figure 5: The route of Pb RAS immunization modulates the formation of CD8⁺ memory T cells in the liver.

C57BL/6 mice were immunized as described in Figure 1. (A and C) Liver and (B and D) splenic CD8⁺ T_EM cells (CD44^hiCD62L^lo) and CD8⁺ T_CM cells (CD44^hiCD62L^hi) were analysed 3 weeks post last vaccination. In addition, (E) liver and (F) spleen CD8⁺ T_EM cells were phenotypically stained for CD11a and KLRG1. Results are expressed as % of cells of CD8⁺ T_EM cells (CD44^hiCD62L^lo) and as number of liver and splenic CD3⁺CD8⁺ T_EM cells expressing CD11a and KLRG1. Data are presented as means ± SEM of 4–5 mice/group. Means were compared by non-parametric Mann-Whitney’s U test or Tukey’s multiple comparisons tests. **** P<0.0001, *** P= 0.0001, **P= 0.0018, * P=0.0165. P value <0.05 is considered statistically significant. The data represents one of two independent experiments yielding similar results.

In separate experiments, we further characterized CD8⁺T_EM cells in IV and ID immunized mice. Splenic and hepatic CD8⁺T_EM cells were analysed for the expression of CD11a and KLRG-1, markers that indicate Ag encounter and functional attribute, respectively ³⁴. According to our results, a significantly higher % of CD11a⁺ and KLRG1⁺ CD8⁺T_EM cells was found in the spleens of IV immunized mice than in ID immunized mice (Fig.5F and Supplemental Fig. 2). Interestingly, there was no difference between the two groups of mice in regard to % of liver KLRG1⁺ CD8⁺ T_EM cells (Fig. 5E), although on the basis of cell numbers IV immunized mice had significantly more cells in the liver and the spleen than ID immunized mice (Fig. 5E, F). There is a critical threshold (~1% of PBMCs) of CSP-specific CD8⁺ T memory cells required to achieve and maintain sterile protection ³⁵, and we made similar observation but with polyclonal LS specific CD8⁺ T cells ³.

We also asked if the different routes of Pb RAS delivery may affect the formation of CD8⁺T_RM cells. It has been shown that a substantially increased population of T_RM cells appears to be related to the prolong history or a level of antigen exposure ³⁶. Because we are hypothesizing that the IV delivery of Pb RAS results in a larger Ag depot in the liver than ID immunization, it would be expected that more CD8⁺ T_RM cells would accumulate in the livers of IV but not ID immunized mice. Analysis of CD8⁺CD44^hi T cells for the expression of CD69 and CXCR6 phenotypic markers, indicators of CD8⁺ T_RM cells ^37,38, shows that indeed significantly higher percentage and number of CD8⁺ T_RM cells accumulated in the livers of IV immunized mice than in the livers of ID Pb RAS immunized mice (Fig.6A and Supplemental S3). Our observations are supported by finding from an elegantly conducted study that large numbers of liver CD8⁺ T_RM cells correlate with protection in mice immunized by IV with Pb RAS ³⁹. In support of our hypotheses that liver stage Ags derived from the Pb RAS do accumulate and persist in the liver and likely are responsible for the maintenance of liver CD8⁺ T_RM cells, we did not observe any differences in the level of CD8⁺ T_RM cells in the spleens of mice from the two groups (Fig.6B). Apart from inflammatory stimuli, liver CD8⁺ T_RM cells also require a local encounter with the Ag ³⁹, which in our case were likely provided by the LS Ags expressed by the liver parasites. According to our results as well as results published by others ⁶, the levels of CD8⁺ T cells observed in the livers of mice immunized with Pb RAS by the ID route were likely ineffective in preventing LS parasite development. The 80% protection observed amongst the ID immunized mice challenged on day-9 (Fig.1B) after the last boost immunization may be attributed primarily to innate immune responses involving NK/NKT cells ⁴⁰. Since the innate immune responses are typically short-lived, mice receiving delayed challenge (18 days) may not tolerate parasite load and therefore succumb to malaria pathology (Fig.1D).

Figure 6. Pb RAS immunization by IV route induces CD8⁺ T_RM cells in the liver but not in the spleen.

Mice were immunized as described for the previous figures. Three weeks after the last immunization with Pb RAS, CD8⁺CD44^hi IHMC and spleen cells were characterized for T_RM cells using CD69 and CXCR6 mAbs. Results are expressed as % and cell number of CD69⁺CXCR6⁺ cells of total CD8⁺CD44^hi T cells. Results from a representative experiment show the mean+ SEM of responses of 5 individual mice per group. Two-way ANOVA test followed by Tukey’s multiple comparisons test were used. **** P<0.0001

Antigenic load versus induction of immune response: a model.

A mathematical model ⁴¹ proposes that effective anti-malaria interventions will initially result in a rapid decrease in clinical disease as population-level immunity is maintained, but may subsequently result in an increase of clinical disease as immunity is lost. According to this model, the incidence of clinical diseases vis-à-vis transmission intensity suggests that immune population encountering the reduced parasite exposure would lose their protection over period of time. Simply stated, this model predicts that maintenance of protection requires intermittent parasite exposure. In a previous study, Haeberlein et al had examined the effect of route of administration on immune response in BALB/c mice using P. y, when there is no difference in parasite liver load between ID and IV immunized mice. To have a comparable parasite liver load in both IV and ID mice groups, they inject 5 time more spz using ID (50k) than IV (10K). They suggested that the difference in protection that have been seen between ID and IV immunization is due to that ID immunization induced a regulatory immune response in the liver and lymph node of immunized mice, and conclude that difference in protection between IV and ID immunized mice is independent of parasite liver load ¹³. However, our study shows that IV vaccinated mice receiving intermediate challenges could maintain sterile protection for as many as 471 days, whereas ID vaccinated mice saw a gradual loss of protective immunity. This loss appears to be affected by the parasite load in the liver during vaccination as well as challenge. Since parasite load in liver during vaccination is shown to control the size of memory CD8⁺ T cell generated against LS infection ², we conceived that parasite load during infectious challenge also affected the maintenance of CD8⁺ T cell frequency in liver. The pre-existing smaller frequency of CD8⁺ T cells generated after ID vaccination would be expected to expand to the desired level upon challenge, and to promote protection. This would be the case with the mice receiving IV challenge, but it appears not to have been achievable in the mice receiving primary ID challenge. Spz entry to liver is not optimal when administered through ID route (Fig.3) resulting in expression of low level of Ags which is not be sufficient to expand the pre-existing CD8⁺ T cells to the level that is needed to maintain protection. The notion for this interpretation is based on the observation that ID vaccinated mice receiving IV challenge survived re-challenge on day-226 post-vaccination, presumably owing to the maintenance of adequate frequency of LS specific CD8⁺ memory T cells (Fig. 6). The IV vaccinated mice, however, did not seem to be affected by the route of challenge. Mice vaccinated through IV route generated high frequency of CD8⁺ T cells post-vaccination (Fig.4), and presumably rendered the host protected from challenges given on day 9 or 18 post-last vaccination. These observations led us to consider these frequencies bettered the required protective-threshold ³. These cells may have further expanded upon ID or IV-challenge, and subsequently maintained more than what is required protective-threshold with intermittent re-challenges. This phenomenon may be seen as the maintenance of complete sterile protection (Fig. 2B).

Conclusion:

Induction of liver-stage specific protective response requires an effective delivery of spz to liver. It is natural to have differential immune response, including low or sub-optimal, induced by vaccine or in response to parasite-exposure among endemic population. If the host possessing suboptimal immune response tolerates the infectious-challenge, the required level of antigen specific-T cell response may be achieved and maintained for conferring protection. Considering the precarious malaria-transmission in the endemic areas due to various vector control measures, the vaccination strategy for anti-malarial vaccines including attenuated whole spz should be rationally designed to achieving the protective-threshold of antigen specific lymphocytes.

Materials and Methods

Ethics statement.

All procedures were reviewed and approved by WRAIR/NMRC Institutional Animal Care and Use Committee (protocol # 15-MVD-30) and were performed in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International in compliance with the Animal Welfare Act and in accordance with the principles set forth in the “Guide for the Care and Use of Laboratory Animals”, Institute of Laboratory Animals Resources, National Research Council, National Academy Press, 1996.

At Nirma University, all the animal experiments were reviewed and approved by the Institutional Animal Ethical Committee (CPCSEA Reg. No: 883/PO/ ReBi/S/05/CPCSEA) (Project Number: IS/PHD/15–1/015). C57BL/6 mice (6–8 weeks) were obtained from ZRC (Zydus Research Centre, Ahmedabad) and housed in pathogen free conditions at the central facility of Nirma University. Animals were acclimated for a week before studies begun and food and water were provided ad libitum.

Parasites

Plasmodium berghei (Pb) (ANKA strain) ⁴² sporozoites were maintained at WRAIR by cyclical transmission in mice and Anopheles stephensi mosquitoes. Briefly, spz were dissected from the salivary glands of mosquitoes 17–21 days after an infective blood meal, as described previously²⁸. For challenge and immunization, spz were counted microscopically, adjusted to a given concentration in RPMI 1640 (Life Technologies, Grand Island, NY) with 1% normal mouse serum, and used immediately after dissection to ensure maximal infectivity.

Mice.

Six- to 8-week old female C57Bl/6J (B6) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Animals were housed under specific-pathogen-free conditions at the Walter Reed Army Institute of Research animal facility and handled according to institutional guidelines. Autoclaved food and water were provided ad libitum.

Female C57BL/6 mice (6–8 weeks) were brought and housed at the central animal facility of the Institute at Nirma University. Animals were handled as per the institutional guidelines. All animal studies were approved by Institutional Animal Ethics Committee (IAEC). Autoclaved food and water were provided ad libitum.

Immunizations and challenge.

For immunizations at WRAIR, Pb spz were attenuated by γ-irradiation (15,000 rad) using a cobalt-60 source (Cobalt-60 Model 109; JL Shepard & Associates, San Fernando, CA). Mice were primed intravenously (i.v.) (200 μl) into the tail vein or intradermal (ID) at the base of the tail (50 μl) with 75K Pb RAS followed by two homologous boost immunizations of 20K RAS given one week apart (Figure 1). At indicated time points after the last Pb RAS immunization, mice were challenged i.v. with 10K infectious Pb spz or Pb-luc spz.

For immunizations at Nirma University, Plasmodium berghei (ANKA stain) sporozoites were harvested from the salivary glands of infected Anopheles stephensi mosquitoes on day 18 after the infected blood meal as described previously ^{4, 3}. Sporozoites were placed in RPMI-1640 medium without serum and attenuated with gamma radiation delivering 13,600 rads from a Cobalt-60 source, at PERD centre or for vaccination. For challenge or infection studies spz were directly used after dissection and counting.

Mice were immunized with 3 doses of Pb RAS (75,000; 20,000; 20,000) through either intradermal (ID) or intravenous (IV) route at an interval of 2 weeks. Mice were challenged with infectious spz (10K) by either IV or ID routes. Parasitemia was monitored daily on Giemsa stained blood smear under the microscope (minimum 15 microscopic fields) till day15 post challenge.

Intrahepatic mononuclear cells (IHMC) isolation.

At the indicated time points after immunization mice were euthanized by CO₂ inhalation. Livers were perfused with 10ml PBS, removed and ground through 70μM nylon cell strainer (BD Labware, Franklin Lakes, NJ), and the cell suspension was processed as previously described ². Briefly, cells were resuspended in PBS containing 37.5% Percoll (Amersham Pharmacia Biotec, Uppsala, Sweden) and centrifuged at 850g for 30 min at room temperature. Red blood cells (RBC) were lysed with RBC lysis buffer (Sigma) and the remaining IHMC were resuspended in complete RPMI 1640 medium containing 10% heat-inactivated fetal bovine serum (FBS) (Hyclone), 1% Penicillin/Streptomycin, 1% GlutaMAX, 1% HEPES, 1% non-essential amino acids (Gibco), 50μM 2-mercaptoethanol (Sigma).

Splenocyte isolation.

Spleens harvest 3 weeks after the last immunization for analysis. Spleens were ground through 70 μM nylon cell strainer (BD Labware, Franklin Lakes, NJ, USA), and the cell suspension was processed as previously described (13). Briefly, Pellet was collected and red blood cells (RBC) were lysed with RBC lysis buffer (Sigma) for 5 min. Splenocytes were resuspended in complete RPMI 1640 medium containing 10% heat-inactivated FBS (Hyclone), 1% Penicillin/ Streptomycin, 1% GlutaMAX, 1% HEPES, 1% non-essential amino acids (Gibco), 50 μM 2-mercaptoethanol (Sigma).

Phenotypic characterization of effector, memory and resident memory CD8⁺ T cells

CD8⁺ T cell response in liver and spleen were measured by flow-cytometry at 21 days post last vaccination (before challenge) ¹⁷ using fluorescent labelled antibodies; CD3-PE (clone 17A2), CD8-Pacific Orange (clone 53–6.7), CD44-Alexa Flour −700 (clone IM-7), CD62L-PE-Cy7 (clone MEL-14) ), CD11a-FITC (clone 2D7), KLRG1-PerCP-cy5.5 (clone 2F1), CD69-BV-605 (clone Hi2F3) (BD Bioscience), CXCR6-BV421 (clone CD186) (Biolegend). Flow cytometer data was analysed using FlowJo ver. 7 software. Data are presented as means ± standard errors of the mean (SEM). Means were compared by non-parametric Mann-Whitney’s U test or Tukey multiple comparison test. P value<0.05 is considered statistically significant.

IFN-γ ELISPOT assay.

Liver and splenic lymphocytes were isolated, washed and resuspended in complete RPMI medium (see above). BD ELISPOT plates were prepared using the mouse IFN-γ ELISPOT kit (BD Biosciences) according to the manufacturer’s instruction. Briefly, ELISPOT plates were coated with anti-IFN-γ abs overnight at 4ºC, subsequently washed with PBS and blocked using RPMI 1640 + 10% FBS for 2hrs at room temperature and washed prior to use. Cells were plated into ELISPOT plates at a concentration of 200–300×10³ per well in 200 μl and stimulated with 1 μg/ml of Pb Kb-17 (IVSFSFQNM) or Pb TRAP₁₃₀ (SALLNVDNL) ²⁹ peptides or medium alone for 42hrs at 37ºC. Plates were developed using Mouse IFN-γ ELISPOT Kit (BD Biosciences) according to manufacturer’s instructions. Results were quantified as the number of IFN-γ-specific spots per 10⁶ cells after subtracting results from medium control wells.

Determination of infectious load of parasite in liver and blood

Mice were inoculated with 100, 500, 1000, and 10,000 infectious sporozoites through IV or ID route. Forty-eight hours later liver was perfused and total RNA was extracted from lower left liver-lobe of individual mouse using RNA isolation kit (Thermo scientific, USA). rDNA was prepared from 1 μg RNA using random hexamer provided in RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific). Briefly, quantitative PCR (qPCR) of 18s rRNA gene was performed using SYBR® Premix Ex Taq™ II (Takara) on QuantStudio-3 Real-Time PCR system (Thermo Scientific). Briefly, 1μl rDNA (1:100 dilution) was used to amplify the P. berghei 18s rRNA using 5’-AAGCATTAAATAAAGCGAATACATCC TTAC-3’ (Forward primer) and 5’- GGAGATTGGTTTTGACGTTTATGTG-3’ (Reverse primer), and the mouse β-actin gene using 5’-GGATGCAGAAGGAGATCACTG-3’ (Forward primer), and 5’-CGATCCACACGGAGTACTTG-3’ (Reverse primer). Here, we took β-actin as an internal control. The parasite load in liver (18S rRNA) was represented as a ΔC_T (C_T of 18S rRNA -C_T of β-actin) value and fold expression (2^-ΔCT). The C_T value is inversely proportional to the expression of specific gene ⁴³.

Blood-stage parasitemia was measured by counting infected RBCs in minimum 30 microscopic fields of Giemsa-stained blood smears during day 4 – 8 post inoculation with infectious sporozoite.

Statistical analysis:

The data are presented as the means ± standard errors of the mean (SEM). Statistical analysis was performed in GraphPad Prism 6, and means were compared by two-tailed non-parametric Mann-Whitney U test or two-way ANOVA test followed by Tukey’s/Sidak’s multiple comparisons test. A P value of <0.05 was considered statistically significant.

Figure 7: Proposed mechanism of CD8⁺ T cell generation after RAS vaccination, and maintenance upon challenge.

RAS vaccination though IV route generates a threshold frequency of CD8⁺ T cells which are adequate to induce sterile protection, whereas ID vaccination induces meagre CD8⁺ T cell response leaving mice susceptible to infection. Early challenge to ID vaccinated mice with infectious sporozoites expands pre-existing CD8⁺ T cells generated after vaccination. The expansion of CD8⁺ T cells in ID vaccinated mice would be affected by the parasite antigen load in liver during infectious challenge. The expression of liver-stage antigen in IV challenge model would be high, whereas the same in ID model could be restricted. The low CD8⁺ memory T cells generated in ID vaccinated mice might be improved with intermittent challenge. The variation in the maintenance of sterile protection in ID vaccinated mice would depend upon the parasite antigen load in the liver during challenge. The IV vaccinated mice shown to have maintained protection irrespective of the load of challenge. Mice immunized through IV route generated high frequency of CD8⁺ T cells which may be more than the required protective threshold. These cells might be expanded upon ID or IV challenge, and subsequently maintained more than the required protective threshold with intermittent re-challenges.