Analysis of the role of TpUB05 antigen from Theileria parva in immune responses to malaria in humans compared to its homologue in Plasmodium falciparum; UB05 antigen

Background: Despite the amount and there is still no effective vaccines against malaria. Immunity to either malaria or East Coast fever is usually seen as species- and/or strain-specific. But there is growing body of evidence suggesting the possibility of the existence of cross strain, cross species and cross genus immune responses in apicomplexans. The principle of gene conservations indicates that homologues play similar role in closely related organisms. UB05 antigen (XP_001347656.2) from P. falciparum is part of chimeric UB05-09 antigen; a potential vaccine candidate has been demonstrated to be a marker of protective immunity in malaria. The homologue of UB05 in T. parva is TpUB05 (XP_763711.1) which was also tested and shown to be a potential marker of protective immunity in ECF as well. In a bid to identify potent markers of protective immunity to aid malaria vaccine development, TpUB05 was tested in malaria caused by P. falciparum . Results: It was observed that TpUB05 provoked stronger immune responses in malaria compared to UB05 antigen as tested using ELISA, ex-vivo ELISpot assay and in vitro growth inhibition assay. Conclusion: This study suggests for the first time that TpUB05 from T. parva is a better marker of protective immunity in malaria compared to its homologue UB05 from P. falciparum . Now we present the and compare the performance of from falciparum vis-à-vis that of TpUB05 from in falciparum TpUB05 similarity was observed. The in silico analysis predicted presence of human T cell and antibody epitopes on TpUB05 and the cross-reactivity observed was expected in view of the significant homology between the two antigens. However, based on the ELISA and GIA data, TpUB05 may contain more potent B-cell epitopes compared to UB05 that are yet to be identified.

P. falciparum cDNA library to produce a 243 bp DNA fragment (Dinga et al., 2016). Both fragments were cloned into a pET32a+ expression vector and overexpressed E. coli cells yielding recombinant fusion proteins migrating at 28 KDa and 26 KDa, respectively in SDS-PAGE performed in 15% polyacrylamide. The recombinant fusion proteins contained a 6xHis Tag, a S Tag domain and a 109 amino acid thioredoxin fusion protein partner. However, it was easier to overexpress and purify TpUB05 as compared to UB05 antigen (Dinga et al., 2015 and2016).
TpUB05 antigen from T. parva possesses human B cell epitopes which are associated with protective immunity against malaria Plasma from the study subjects were tested ex-vivo with TpUB05 using ELISA and it was shown that SIS possessed more anti-TpUB05 antibody than FSS. Adults (SIS + FSS) produced more antigenspecific antibody to TpUB05 (p = 0.0001) than children (SC) (Fig. 1A). When the antibody response to TpUB05 was compared to that of UB05 in the malaria study, it was observed that the former produced a significantly higher antibody response (p = 0.006) compared to the latter (Fig. 1B).

Human antibody levels to TpUB05 negatively correlates with fever and malaria parasitaemia
When the antibody levels to TpUB05 were plotted as a function of parasitaemia, it was observed that that there was a negative correlation between parasitaemia and antibody levels. This implied that the higher the specific antibody level the lower the parasite density in the blood. In other experiments, the antibody levels were shown to be statistically higher in subjects with no fever and then in subjects with fever ( Fig. 2A, 2B). Taken together these results suggested that high anti-TpUB05 antibody levels was associated with a state of protective immunity to malaria.

TpUB05 is a better reagent to distinguish between SIS and FSS compared to UB05
There is currently no simple assay to distinguish between the semi-immune (SIS) and the frequently sick status (FSS) in malaria. We therefore compared the diagnostic potential of TpUB05 and its homologue UB05 in distinguishing between these two clinical groups. The results are presented in Table 3 showing that rTpUB05 is a relatively better as a better reagent than UB05 for diagnosing acquisition of some immunity against malaria. While the specificity of the assay with UB05 was 85% that of TpUB05 was 92.5%. and within the cut off range for a good diagnostic test The positive predictive value for UB05 was 57% as compared to 90% by TpUB05. Similar trend was observed for the negative predictive value which was for UB05 was 39.1% and this was increased to 52.1% when using TpUB05 (Table 1). However, the sensitivity of the TpUB05 assay was still too low for TpUB05 to be considered for routine usage.

Human PBMCs recognize T-cell epitopes in TpUB05 in a manner that correlates with immune protection against malaria
With the presence of human B-cell epitopes in TpUB05, it was worthwhile to test for the possibility of the presence of human T-cell epitopes in TpUB05. Since TpUB05 is the T. parva homologue of UB05 found in P. falciparum, and it has been shown that UB05 stimulates T-cell proliferation in persons exposed to malaria, we decided to test the ability of TpUB05 to recall T-cell function in humans. To do this, PBMCs from 63 subjects were isolated and stimulated with r-TpUB05 to produce IFN-γ and measured using human ELISpot assay. Twenty-seven subjects recognized TpUB05 of which SIS make up 85.2% (Fig. 3A). When the recognition of TpUB05 was compared to that of UB05 (Fig. 3B), there was no significant difference (Fig. 3C).
We also found that there was a significantly higher stimulatory index (SI) of IFN-γ production between subjects with no fever (p = 0.002) compared to those who had fever (Fig. 4A). The same trend was observed when the absence or presence of parasitaemia was taken into consideration (Fig. 4B).
These results suggested that TpUB05 stimulated human PBMCs to produce IFN-γ in a manner that correlated with protection against malaria.

Online prediction indicates the presence of human epitopes on TpUB05
Online prediction of human T cell and antibody epitopes revealed the presence of epitopes that could bind to human MHC on TpUB05. The MHC I and II binding peptides with a percentile rank below 1.0 were considered as those with very high affinity for the MHC molecules (Table 2). Using algorithms at www.iedb.org also predicted the presence of human antibody epitopes in TpUB05 (Table 2).

Rabbit polyclonal antibody raised against recombinant TpUB05 performs better in inhibiting P. falciparum parasite growth in vitro than antibodies to recombinant UB05
The ability of the TpUB05-specific antiserum to inhibit parasite growth in vitro was tested using the growth inhibition assay (GIA) as described in Materials and Methods and the data compared with that obtained with UB05-specific antiserum. Rabbit anti-TpUB05 antiserum was able to inhibit parasite growth in vitro in a way that was significantly higher than the performance of anti-UB05 antiserum (p = 0.0001) ( Figure 5). This inhibition was observed for all the parasite strains tested ( Figure 5).
In positive control experiments we tested in parallel two monoclonal antibodies available from Bioresources, namely, anti AMA1 and anti EBAI1. Amongst the parasite strains tested, only the P.
falciparum HB3 strain was not inhibited by the positive control antibodies (anti-AMA1 and anti-EBA175) indicating that these two anti-sera are less cross-reactive that those direct against TpUB05 and UB05. The results presented in figure 5 showed that Anti-TpUB05 antiserum did not significantly inhibited parasite growth more than anti-AMA1 antibody. However, it performed far better than anti-EBA175 antibody in inhibiting in vitro parasite growth (p = 0.0001).

Discussion
Malaria vaccine development has been mostly focused on single antigens. But the complex nature of the malaria parasite life cycle and the mechanisms it uses to evade the human immune responses implies the ideal malaria vaccine should target several antigens expressed in different stages of the parasite's development (Titanji et al., 2017) as well as homologous antigens from related species in a bid to circumvent the setbacks observed with strain-specific or species-specific protective immune responses. The first step towards this would be the identification and characterization of homologous antigens from phylogenetically related organisms as better or more potent markers of protective immunity against malaria. Here we show for the first time that TpUB05 antigen from T. parva is a better diagnostic reagent for semi immune status in malaria as compared to UB05 antigen from P.
falciparum. Antibodies to TpUB05 were more effective than those directed against its homologue UB05 in inhibiting P. falciparum growth in culture.
The principle of the conservation of gene function shows that most orthologous gene products play similar role in closely related organisms (Gabaldón and Koonin 2013). We had earlier shown that the T. parva homologue of UB05; TpUB05, is a marker of protective immunity in ECF. We then decided to characterize TpUB05 in malaria caused by P. falciparum. This observation of effective cross immunisation with homologous antigens has been shown before (Gabaldón and Koonin 2013). In other words, it might be more useful to search for antigens that induce protection against ECF and test in malaria caused by a distant relative like P. falciparum. To verify this hypothesis, we carried out simultaneous testing of immune responses in semi-immune and malaria susceptible subjects and showed TpUB05 is a more potent marker of immune protection in malaria compared to UB05.
Although B cells have been seen as contributing little to resistance and protective immunity and resistance to infections with Apicomplexa, many studies have shown that hosts infected with these parasites are capable of producing parasite-specific immunoglobulins which are protective, after recovering from an infection (Hogh 1996). The present study shows that there was a statistically significant higher amount of antibodies in human plasma from SIS that recognised TpUB05 as compared to plasma from FSS, hence an indication of correlation with protection from malaria caused by P. falciparum. When this was compared to the ELISA results obtained with UB05 antigen (Dinga et al., 2017) under the same experimental conditions, a similar trend was observed. However, TpUB05 appeared to be a better marker of protective immunity to malaria than UB05 using ELISA, as TpUB05 was able to detect a higher amount of antigen-specific antibody in human plasma in a manner that was statistically significant.
Studies to elucidate the mechanism(s) of the protective immune response to apicomplexan parasites implicate the role of the production of gamma interferon (IFN-γ) amongst other cytokines and chemokines (Doolan and Martinez-Alier. 2006), which control parasite infectivity and interfere with parasite development. To find out whether TpUB05 from T. parva can induce immune response similar to those occurring during malaria caused by P. falciparum, we tested the ability of TpUB05 to recall T cell function in PBMCs from subjects in a malaria endemic region. Biostatistics analyses revealed that there are T cell epitopes in TpUB05 that could bind and be recognized by human T cells leading to the production of IFN-gamma. This production of IFN-gamma occurred in a manner that correlated to immune protection against malaria as it was preferentially recognized by semi-immunes as compared to frequently sick subjects. When these data were compared to that obtained with UB05 antigen from P. falciparum (Dinga et al., 2016), which were obtained under the same experimental conditions, the same trend of correlation with protection against malaria was observed. There was no significant difference between these two antigens in stimulating the production of IFN-gamma from human PBMCs suggesting that they could stimulate similar reactions in vivo; this however remains to be shown.
The hallmark of an effective malaria subunit vaccine would be its ability to stimulate the cellular as well as the antibody components of the immune system that are protective. This implies that an antigen's ability to preferentially detect antigen-specific antibodies and recall T cell function in people who have acquired limited protective immunity to malaria is an indication that the protein is involved in the immune protection against malaria parasites. TpUB05 was able to recall SIS's T-cells' ability to produce IFN-gamma and detect antigen-specific antibodies in these semi-immunes, hence a marker of protective immunity in malaria.
Contamination of antigen preparations with lipopolysaccharide (LPS) have been shown to confound Tcell response assay results to malaria antigens. However, it is unlike there was a contribution of possible contaminating bacterial LPS to the observed responses in the present study as the responses were specific to each of the groups. The cytokines whose production appears to be influenced by LPS include IL-1beta and IL-6 but not IFN-gamma which was studied herein (Jansk et al., 2003).
Some schools of thought have it that an association is not necessarily causation and as such should not be interpreted in isolation. But when the data obtained in this study is associated with fever and parasitaemia it implies causation. Normal body temperature with little or no parasites in the blood as well as higher IFN-gamma production and antibody amounts have been shown to correlate with immune protection against clinical malaria (Titanji et al., 2009, Doolan et al., 2009, Greenhouse et al., 2011, Dinga et al., 2016. The data obtained in this study strongly suggest that TpUB05 is a better marker of protective immunity in clinical malaria compared to UB05.
Having shown that TpUB05 possesses T cell and B cell epitopes that are bind and recognize human MHC molecules and antibodies, respectively, it was necessary to find out if polyclonal antibodies raised against TpUB05 from T. parva has any effect on malaria parasite development by employing the in vitro growth inhibition assay. Growth inhibition assay which involves impaired merozoite invasion and subsequent development of parasites in erythrocytes is currently being considered as one of most relevant assays to screen potential blood-stage vaccine candidates prior to moving to the stage of clinical development. Purified polyclonal total IgG raised in rabbits against TpUB05 from T.
parva was able to significantly inhibit malaria parasite growth in vitro. This inhibition was statistically higher than that observed with anti-UB05 polyclonal antibody as well as anti-EBA175 monoclonal antibody but not anti-AMA1 monoclonal antibody. Hence TpUB05 should be considered a potent and better marker of protective immunity against malaria as compared to UB05 from P. falciparum.
Bioinformatics comparison of TpUB05 and UB05 shows that they exhibit a degree of sequence homology: 43.3% identity and 67% similarity was observed. The in silico analysis predicted presence of human T cell and antibody epitopes on TpUB05 and the cross-reactivity observed was expected in view of the significant homology between the two antigens. However, based on the ELISA and GIA data, TpUB05 may contain more potent B-cell epitopes compared to UB05 that are yet to be identified.
Put together the results from this study implies TpUB05 is a more potent marker of protection against malaria and confirms the notion that homologues could play similar or better roles in related organisms. This is in line with previous studies where Mycobacterium bovis (BCG vaccine) was used to vaccinate humans against M. tuberculosis (Fine et al., 1999) and humans exposed to Onchocerca ochengi were protected against O. volvulus infection (Wahl et al., 1998). The potential role of TpUB05 in inducing a more effective immune protection against malaria infection and disease warrants more investigation.
While species-specific vaccine development against either malaria or ECF is expected to effectively reduce transmission, targeting cross-reactive epitopes in potent vaccine candidates may lead to protective immunity that cuts across multiple apicomplexan parasites. The importance of such immune cross-reactivity may even insinuate enhancement of protective immune responses amongst distantly related apicomplexans like P. falciparum and T. parva. The results obtained in the current study permits the evaluation of such a potential cross-species malaria vaccine candidate at an earlier stage, with the potential to save time and resources for development decisions.

Study site
The study was carried out in Buea, which is endemic for malaria. Buea is a multi-ethnic town found along the flanks of Mount Cameroon in the South West region of the Republic of Cameroon. The prevalence of malaria in Buea varies between a meso-endemic (dry season) and a hyperendemic (rainy season) and) zone with perennial malaria transmission (Nkuo Akenji et al., 2005).

Study population
A general invitation was given to the people staying in the Buea municipality and those who visited the Buea Regional Hospital for medical attention in order to recruit subjects for this study (Dinga et al., 2016). After explaining the objectives of the project to voluntary individuals, informed consent was obtained from each participant above 18 years of age or parents/guardians of sick children below the age of 5. The medical history of each subject was obtained and recorded by a collaborating Medical Nurse who also assisted in completing a questionnaire form. Blood was then drawn from the subjects by a Medical Laboratory Technician for the diagnosis of malaria, peripheral blood mononuclear cells isolation, and serology as described below.
The following criteria were used to screen and recruit subjects into the study; (i) subjects who were aged 18 years of age or older and who had been living in the study site for at least 3 years; (ii) subjects with no history of a malaria episode in the last 12 months, no fever and parasitaemia at sample collection, no use of mosquito bed net, hence had been exposed to mosquito bites and was on no prophylaxis were designated semi-immune subjects (SIS); (iii) subjects who had had at least one malaria episode in the last 12 months and had fever and parasitaemia at sample collection referred to as frequently sick subjects (FSS); and (iv) children aged 5 or below who had fever and parasitaemia at the time of sample collection were referred to as sick children (SC). This cohort of individuals was highly selected and described previously (Titanji et al., 2009, Dinga et al., 2014. The analyses for fever, parasitaemia and blood sample collection for ELISpot assay and ELISA took place between March-June 2014(Dinga et al., 2016.

Haemoglobin measurement
In order to assess the level of anaemia which be due to malaria, Haemoglobin level of the volunteers was measured using a haemoglobinometer (STAT-Site MHgb metre, Stanbio Laboratory Texas, USA).

Assessment of nutritional status
It has been shown that being in good health reduces susceptibility to non-communicable and infectious diseases including malaria (Suhrcke et al., 2011). Assessment of the nutritional status of included in the assay. After the 3 hours' incubation at room temperature, the plates were washed with Wash buffer three times and 100 μL of anti-rabbit IgG-HRP conjugate (Sigma) diluted at 1:10 000 in Wash buffer containing 1% skimmed milk and incubated for 1 hour at room temperature. After which the plates were washed with Wash buffer and 100 μL of substrate added. Optical density (OD) of the wells were obtained at 405 nm using a microplate reader (LabsystemsMultiskan MCC 340,Helsinki,Finland). A subject was considered positive if its OD value was equal to or greater than the mean control OD +2SD after subtracting background. The experiment was run in duplicates.

Growth Inhibition Assay
The ability of purified rabbit IgGs raised against r-TpUB05 and r-UB05 to inhibit the replication of P.
falciparum in vitro, was tested by measuring parasite lactate dehydrogenase (pLDH) in late trophozoite/early schizont stage cultures as described in Methods_In_Malaria_Research 6th Edition (Moll et al., 2013, Dinga et al., 2018. Briefly, enrichment of late-stage infected erythrocytes was done by performing three 5% sorbitol synchronizations and 60% Percoll gradient centrifugations. The assay was done using late trophozoite or schizont stages. This was ensured by starting and stopped (≈96 hours) the experiment when most of the parasites were at those stages. The start parasitaemia of the culture was at 0.1%-0.3% parasitaemia and 1% haematocrit in a CO 2 incubator for 2 cycles. Polyclonal antibodies raised against r-TpUB05 or r-UB05 was added at the start of the cultured at an optimized dilution of 1:10 (final concentration of anti-TpUB05 total IgG = 0.48 mg/mL and anti-UB05 total IgG = 0.63 mg/mL). The monoclonal antibodies, anti-EBA-175 RII (final concentrations; R217 at 0.133 mg/mL and R218 at 0.153 mg/mL) and anti-AMA1 (final concentration of 0.1 mg/mL) and were used as positive controls in the inhibition assays, while negative control wells contained pre-immune sera (final concentration of 0.7 mg/mL) or no antiserum. Control wells were included in the assay that contained either only parasitized RBCs (pRBC only) or normal RBCs. Percentage growth inhibition was calculated by first of all subtracting the average OD from the normal RBCs wells from all the other OD values obtained before proceeding to the calculation proper.

Immune epitope prediction using online resources
In silico prediction of immune epitopes from TpUB05 and UB05 was done using algorithms in IEDB Analysis Resources (www.immuneepitope.org). This was already described in Dinga et al., 2015Dinga et al., , 2016Dinga et al., , and 2017 Figure 1 Human ELISpot assay using TpUB05 to stimulate human PBMCs for IFN-gamma production.
A: r-TpUB05 from T. parva was used to stimulate human T-cells from 63 subjects. B: Human ELISpot assay using r-UB05 from P. falciparum. C: Combining and comparing SI values obtained using r-TpUB05 and r-UB05.

Figure 2
Relationship between T-cell responses (IFN-γ production) to recombinant TpUB05, fever and parasitaemia. A: subjects with no fever (SIS) appear to produce more IFN-gamma in response to stimulation by TpUB05 compared to those with fever (FSS) (p = 0.002) B: production of IFN-gamma correlated with protective immune response against malaria.   Comparing anti-TpUB05 and anti-UB05 polyclonal antiserum in Growth Inhibition Assay.
Rabbit antisera against rTpUB05 and rUB05 were used in vitro to test for their ability to inhibit parasite growth. This was done using P. falciparum laboratory strains; 3D7, FCR-1/FV0 and HB3 and 2 field isolates; GH01 and SC01. They were tested at 1:10 dilution.
Standard deviation is indicated with error bars. The experiment was done twice in triplicates.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.