Immunogenicity Studies with Microbial Fractions of M . tuberculosis H 37 Rv Total Culture Filtrate

Current study investigates the whole secretory proteome of Mycobacterium tuberculosis as culture filtrate fractions to identify immunoprotective protein antigens on the basis of protection studies in animal (mouse and guinea pig) models. Secretory culture filtrate proteins (CFPs) of M. tuberculosis H37Rv were fractionated into fifteen narrow molecular mass fractions in the order of increasing molecular size (F1-F15) by electroelution. Immunization studies revealed proteins in the molecular weight range of 20-24kDa (F7), 25-30kDa (F8) and 37-42kDa (F11) as key protective fractions against experimental tuberculosis in both the animal (mice and guinea pig) models. Amongst these fractions, F7 imparted even better protection as compared to BCG. Immunological studies with all the fractions demonstrated that although selected three protective fractions were able to induce significant immune responses in both short term culture filtrate (STCF) immunized and Mtb infected animals, there were number of other non-protective fractions also that were inducing higher immune responses either in immunized animals (e.g.F12-F15) or in Mtb challenged animals (e.g.F1-F6). These results demonstrate that only those mycobacterial proteins that are recognized by the host immune system both during immunization and infection can induce significant protection against experimental tuberculosis, however there is no direct correlation between the level of immune responses and degree of protective efficacy.


Introduction
Tuberculosis (TB) is second only to HIV/AIDS as the greatest killer worldwide due to a single infectious agent.In 2013, 9 million people fell ill with TB and 1.5 million died from the disease (http://www.who.int/mediacentre/factsheets/fs104/en/).Currently, only vaccine available to prevent tuberculosis (TB) is the live attenuated M. bovis Bacille Calmette-Guérin (BCG) vaccine.BCG vaccination protects children against TB meningitis and against disseminated, miliary disease, but confers a variable protection (ranging from 0% to 80%) against pulmonary TB (Fine, 1995).In the last decade, much progress has been made and several candidates TB vaccines have entered clinical trials (Brennan & Thole, 2012).The efficacy study with a new generation TB vaccine (MVA85A) showed the absence of efficacy against TB over that conferred by BCG alone (Tameris et al., 2013).Recently, the TB vaccine, H4:IC31 demonstrated an acceptable safety profile and was immunogenic in South African adults (Geldenhuys et al., 2015).Rapid vaccine development and improvement in currently available vaccine require reliable correlates of protection against tuberculosis.In spite of intensive efforts in experimental models, there are few guidelines for rational selection of vaccine candidates for tuberculosis.Proteins secreted from live mycobacteria may have an advantage in triggering an earlier immune response as compared to those released by mycobacterial lysis (Skeiky & Sadoff, 2006).
Cell-mediated immune response, involving mainly the CD4+ and CD8+ T-cell subsets, plays an essential role in the pathogenesis of TB (Aktas et al., 2009).In mouse TB challenge studies, IFN-γ-secreting CD4+T cells are reported to be important mediators of protection (Kaufmann, Cole, Mizrahi, Rubin, & Nathan, 2005) and attempts to induce TB-antigens specific IFN-γ-secreting CD4+ T cells has been the dominant theme of most TB vaccine research studies.Various attempts have been made to identify key antigens of importance for protective immunity against M. tuberculosis in number of studies (Boesen, Jensen, Wilcke, & Andersen, 1995;Haslov et al., 1995;Olsen, Hansen, Holm, & Andersen, 2000;Sable, Verma, Behera, & Khuller, 2005) and demonstrated that no correlation exists between high Th1 response, at least high IFN-γ, induced during natural infection and protection imparted after immunization.Thus, still there are no immunological biomarkers that predict vaccine efficacy to tuberculosis is available.Since, culture filtrate of Mtb contains a mixture of proteins ranging from 3kDa to more than 100 kDa, evaluation of each and every protein present in culture filtrate for the protective efficacy in experimental TB models is not a feasible approach.Culture filtrate fractions have also been used earlier, for the identification of Immunodominant antigens in mice during infection with Mtb (Andersen et al., 1992) and for their ability to induce IFN-γ responses in advanced TB patients as well as healthy TB contacts and community controls (Boesen et al., 1995).The findings from our laboratory indicated that the CTL response of healthy contacts was found to be directed towards multiple antigens fractions with prominent recognition of proteins below 15 kDa and those present in the region of 30 kDa (Sable, Kaur, Verma, & Khuller, 2005).However no report is available regarding the nature of protective efficacy of Mtb culture filtrate fractions.Thus, for the identification of key protective antigens, the present study was initiated with the objective to evaluate the culture filtrate fractions of Mycobacterium tuberculosis H 37 Rv for protective efficacy in two animal models of experimental tuberculosis i.e. mouse/guinea pig.Based on the direct evidence of protective potential against Mtb challenge studies in these animal models, our data suggests proteins in the region of 20-42kDa in mycobacterial secretory proteome are suitable candidates for the development of future antituberculous multivalent subunit vaccine and warrants further exploration.

Mycobacterial Cultures and Growth Conditions
Mycobacterium tuberculosis H 37 Rv and Mycobacterium bovis BCG originally obtained from National Collection of Type Cultures (NCTC), London and being maintained on Lowenstein Jensen's (LJ) medium in the laboratory were used in the study.

Animals
Four to five weeks old female BALB/c (H-2 d ) mice weighing 15-20 gmeach, obtained from Central Research Institute, Kasauli, India were used in this study.Mice were housed in cages contained within a negative pressure regulated animal isolator and were fed on standard pellet diet [Hindustan Lever Ltd., Mumbai] and water ad libitum.
Dunkin Hartley outbred female guinea pigs weighing 250-350 gm each, were procured from Haryana Agriculture University (HAU), Hisar, India.All animal experiments were carried out in accordance with the rules and regulations set forth by the Institute Animal Ethics Committee (IAEC), Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, India.

Isolation and Fractionation of Total Short-Term Culture Filtrate Proteins (STCF) of M. tuberculosis H 37 Rv in to Narrow Molecular Mass Fractions (n=15)
Culture filtrate proteins were isolated by growing M. tuberculosis H 37 Rv in liquid synthetic Youman's medium for four weeks as a stationary pellicle culture at 37 °C (Haslov et al., 1995).The culture supernatant was filter sterilized and concentrated initially using tangential flow system with a cartridge of 3 kDa cut-off membrane (Millipore).The concentrated culture filtrate was desalted by extensive washings with distilled water and finally exchanged with PBS.This preparation was referred to as STCF.The protein profile was analysed by 12% sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) followed by silver staining (Morrissey, 1981).Aliquots of culture filtrate proteins were finally stored at -20 °C till further use.
A panel of molecular weight fractions was obtained from STCF by electro-elution (Boesen et al., 1995).Briefly, Five-milligram of the above STCF was separated on 16% preparative SDS-PAGE overnight on protein II xi cell (Bio-Rad) having gel dimensions of 16/20 cm inner plate and 18.3/20 cm outer plate.Constant current of 10mA was applied for stacking of protein samples followed by resolution at 30 mA.After the electrophoresis run, the gel was pre equilibrated in tris borate buffer pH 8.7 (50mM tris and 250 mM boric acid) and transferred to whole gel elutor (Bio-Rad) and electroeluted at 250 mA for 30 min and the current was reversed for 10 seconds at the end of the elution period to loosen molecules sticking to the cellophane membrane at the bottom of the wells.Fractions were harvested from the unit with the help of vacuum pump and adjacent fractions were pooled to make easily manageable numbers and designated as F1 to F15 in the order of increasing molecular size and kept frozen at -20 °C until further use.All fractions were analysed by separation on SDS-PAGE (16%) followed by silver staining (Figure 1).The proteins in the fractions were identified on the basis of reactivity with available monoclonal / polyclonal antibodies by ELISA as represented in Table 1.The purified proteins were also analysed for reactivity with other antibodies (corresponding to other proteins) and the OD at 492nm was found to be below 0.1.

Mice
BALB/c mice (5 animals per group) were immunized with Mtb culture filtrate fractions (F1 to F15 individually in parallel) emulsified with 250 μg of dimethyl dioctadecyl ammonium bromide (DDA) as adjuvant.DDA emulsion was prepared as described previously (Brandt, Elhay, Rosenkrands, Lindblad, & Andersen, 2000).Immunizations were performed with 25 μg of each culture filtrate fraction individually (F1 to F15) or 25 μg of total short term culture filtrate (STCF) protein adjuvanted with 250 μg of DDA in a volume of 0.2ml/dose/animal.Animals in each group were injected subcutaneously with three biweekly doses of the above-mentioned immunogens.For comparison, group of mice were also immunized with BCG (10 5 CFUs/animal) subcutaneously as a single dose.Control animals were sham-immunized with adjuvant only.

Guinea Pigs
Guinea pigs (5 animals per group) were immunized with F7, F8 and F11 fractions individually along with 500 µg of DDA adjuvant.A group of animals were also immunized with a single dose of BCG (10 5 CFUs/animal) administered subcutaneously in parallel.Immunization was carried out with 100 μg of each culture filtrate fractions or with 100 μg total short term culture filtrate (STCF) protein. .A group of guinea pigs were also immunized with combination of F7 and F11 (50 μg each fraction, total 100 μg) culture filtrate fraction.The guinea pigs were immunized three times with the designated dose per animal, at biweekly interval, using 500 μg of DDA as adjuvant in a volume of 0.2ml subcutaneously on the back.Control animals were sham-immunized with adjuvant only.

Splenocyte Proliferation Assay
Spleen cells from mice and guinea pigs immunized with STCF were suspended in RPMI-1640 medium supplemented with 10% heat inactivated fetal bovine serum.The concentration of splenocytes was adjusted to ~2x10 5 cells/well in 96 well culture plates and cultured in the presence of 2 μg/ml of STCF or culture filtrate fractions.The plates were incubated for 4 days at 37 °C in 5% Co 2 followed by incubation with [ 3 H]-thymidine (0.25 μCi/well) for 18 hour.The cells were harvested using a cell harvester, and the incorporated radioactivity was measured and stimulation indices (SI) were calculated by dividing mean counts per minute in antigen-stimulated wells by mean counts per minute in unstimulated wells.

IFN-γ and IL-12 (p40) Assay
Levels of IFN-γ and IL-12 induced in culture supernatants of mice splenocytes in response to antigens were estimated after 72 hours by ELISA using commercially available kits (Opt EIA™ Set BD Pharmingen, CA, and USA) following manufacturer's instructions.

IgG ELISA
IgG titers were determined in the sera of STCF immunized animals by ELISA.Briefly, the wells of 96 well ELISA plates were coated with proteins antigens (STCF/ culture filtrate fraction) at a concentration of 2 μg/ml in carbonate buffer (pH 9.6) overnight at 4 °C.Plates were washed with PBS containing 0.05% Tween-20 (PBS-T) and blocked with 3% BSA at 37 °C for 2h.Diluted sera (1:100 dilution in PBS-T having 1% BSA) were added and bound antibodies were detected using HRP-conjugated anti-mouse/anti-guinea pig IgG.The absorbance was read at 492 nm.

Protection Studies
Mice were challenged with 0.1 ml (10 5 bacilli) suspension of M. tuberculosis H 37 Rv per animal intravenously at 3 weeks p.im, while guinea pigs were challenged with 0.1 ml (10 5 bacilli) suspension of M. tuberculosis H 37 Rv per animal intramuscularly (i.m) in the thigh muscle at 6 weeks p.im.Four weeks after the challenge, the mice were sacrificed and lungs and spleen were aseptically isolated.Guinea pigs were sacrificed after 6 weeks of challenge and spleen and the right lung were aseptically isolated.The organs were homogenized and appropriate serial dilutions were spread on Middle brook 7H11 agar plates supplemented with oleic acid-albumin-dextrose-catalase (OADC).To selectively inhibit the growth of the residual BCG bacteria in the test organs, 2-thiophenecarboxylicacid hydrazide (2 μg/ml) was added to media.Colonies were counted after 3-4 weeks of incubation at 37 °C and bacterial counts per organ were determined.

Statistical Methods
Two-way comparison between test and control groups was performed using Student's t-test.Multiple comparisons amongst different groups were performed by (one way-Anova with tukey's multiple comparision test.).The statistical analyses were made using GraphPad prism 5.A p value of <0.05 was considered significant.

Protective Efficacy of Culture Filtrate Fractions in Mouse Model
The log 10 CFUs in the lungs of BCG and STCF immunized animals were significantly lower (p≤0.001)as compared to unimmunized animals (Figure 2A) whereas; no significant difference in CFUs was observed in animals immunized with BCG and STCF+DDA.The log 10 CFUs in the lungs of animals immunized with fractions F2, F3, F7, F8, F10, and F11 were significantly lower (p<0.001)as compared to the unvaccinated animals.When the experimental subunit vaccine preparations based on culture filtrate fractions were compared with BCG, amongst all the fractions, F7 showed significantly higher (P<0.05)reduction in CFUs in the lungs by an average of 0.52 logs followed by F11 and F8 respectively (Figure 2A).
Further evaluation of CFUs in the spleens of Mtb challenged mice indicated that log 10 CFUs in animals immunized with F2, F3, F6, F7, F8, F10, F11, STCF and BCG vaccinated groups were significantly lower as compared to unvaccinated animals (Figure 2B).However, similar to that observed in the lungs, in the spleens also, animals immunized with F7 and F11 fractions showed significant reduction (p<0.05) in CFUs as compared to that observed in BCG vaccinated group.Taken together, these findings indicate that culture filtrate fractions comprising F7, F8 and F11 are the most immunoprotective and suggested their further evaluation in another animal model i.e. guinea pig.3.2 Culture Filtrate Fractions, F7, F8 and F11 Induced Significant Protection against TB in Guinea Pig Model BCG immunized animals had a significantly (p<0.001)lower bacterial burden both in the lungs and spleens as compared with sham-immunized guinea pigs (Figure 3A, 3B).When the vaccine potential of experimental subunit vaccines were compared, it was observed that all the vaccine formulations consisting of STCF, F7, F8 and F11 showed significant reduction (p<0.001) in terms of log 10 CFU as compared with sham immunized group Similar to mouse model, F7 immunised guinea pigs also showed significant reduction of bacterial burden in the lungs (p<0.01) and the spleen (p<0.05) as compared to BCG immunized animals and the protection induced by F7 was even better than that produced by STCF consisting of whole secretory proteome.
On the whole, culture filtrate fraction, F7 adjuvanted with DDA induced significantly higher protection followed by F11 and F8 in both mouse and guinea pig models of experimental tuberculosis

Immunoreactivity of Culture Filtrate Fractions in STCF Immunized Mice
The specificity of the T-cell responses induced by the immunization with STCF was evaluated by stimulating splenic T-cells with narrow molecular mass fractions obtained by the multielution technique.Immune responses in terms of T cell proliferation, cytokine release and serum total IgG were studied at three weeks post immunization.The proliferation represented as stimulation indices (Figure 4A) of splenocytes from STCF immunized animals were found to be significantly higher (p≤ 0.001) than those of sham immunized group (S.I <1 and data not shown) indicating a broader recognition of secreted antigen fractions.When the lymphocyte proliferation induced by protective fractions F7, F8 and F11 were analysed, higher stimulation indices (S.I >6) were observed thus indicating high immunogenicity of these fractions.
As observed in case of stimulation indices, culture filtrate fractions elicited significantly higher (p<0.001)levels of IFN-γ (Th1 cytokine) as compared to unvaccinated animals.The levels were notably higher (≥500pg/ml) in all the fractions except fraction number F1 (Figure 4B).Another Th1 cytokine important for the protective immunity against Mtb is IL-12p40and its secretion was also found to be significantly higher (p<0.001) in the immunized animals as compared to unvaccinated animals.However, prominent secretion of IL-12p40 (>500pg/ml) was observed with proteins in the molecular weight range of 21-45kDa comprising fraction F7 to F12 (Figure 4C).IFN-γ and IL-12p40 secretion in response to STCF was higher as compared to the individual culture filtrate fractions in the immunized group.Further, amongst fractions also, F12-F15 were although not providing significant protection against experimental tuberculosis but were able to induce prominent cytokine release (Figure 3b, 3c).Titers of antigen-specific IgG were evaluated in the sera of immunized as well as control mice by indirect ELISA to study the humoral responses.The animals immunised with STCF elicited substantial levels of IgG response to all the culture filtrate fractions as compared to the sham-immunized animals (data not shown).).However, higher levels of IgG in terms of absorbance (OD 492 nm ≥0.5) were observed with protein fractions F7 to F15.

Immunoreactivity of Culture Filtrate Fractions in STCF Immunized Guinea Pigs.
Immune responses in terms of T cell proliferation and serum IgG levels were also studied at three weeks post immunization of guinea pigs with STCF.The stimulation indices of splenocytes of immunized groups were found to be significantly higher (p ≤ 0.001) than those of sham immunized group (SI < 1) (Figure 5A).However, as observed in mouse model, higher responses (SI ≥ 6) were also observed with proteins in the molecular weight range of 21-45kDa comprising fractions F7 to F12 in STCF immunized guinea pigs.The immunization also resulted in higher IgG response (OD 492nm ≥ 0.2) for all the culture filtrate fractions as compared to the sham-immunized animals (OD 492nm <0.1).However, among the fifteen fractions, F6-F15 showed higher IgG responses (OD 492nm ≥ 0.5) (Figure 5B).

Immunoreactivity of Culture Filtrate Fractions in Mtb Infected Mice
To evaluate the total secretory proteome for its recognition by the immune system during experimental mycobacterial infection, unimmunized mice were challenged with 0.1 ml (10 5 bacilli) suspension of M. tuberculosis H 37 Rv intravenously, and two weeks post infection splenocytes were isolated and IFN-γ and IL-12 cytokine secretions were measured in response to culture filtrate fractions.The response was directed to multiple secreted antigens and significantly higher (p<0.001)levels of IFN-γ were elicited by culture filtrate fractions in infected animals as compared to uninfected animals (≤ 500pg/ml) (Fig. 6A).IL-12p40 secretion (Fig. 6B) was also significantly (p<0.001) higher in the infected animals as compared to uninfected animals (undetectable).However, amongst all the fractions, F1 to F11 having proteins in the molecular weight range of 3-42kDa induced both high IFN-γ and IL-12p40 levels.On contrary, the proteins corresponding to fractions F12-F15, which were inducing strong cytokine, release in STCF immunized animals were not being recognized by immune system during experimental infection.
Thus, based on protection studies in two experimental models (mice and guinea pigs), three culture filtrate fractions of STCF i.e. fraction F7, F8 and F11 comprising the proteins in the molecular weight range of 20 to 42kDa (Table 1) were found to be most protective.Further, analysis of the immune responses to all the fifteen fractions of STCF indicated that although these three protective fractions were also able to elicit prominent immune responses both during Mtb challenge and after immunization in two different animal models, there were other fractions also which were inducing strong Th1 immune responses, not correlating with the protection induced by these fractions..

Discussion
Recent advances in mycobacterial genome characterization, identification of protective antigens, understanding the molecular basis of protective immune response, adjuvant design and development have reinforced the idea of a feasible subunit vaccine for tuberculosis (TB).The development of new antituberculous vaccine(s) requires an understanding of the protective immune response against TB.Despite recent advances, relatively little is known regarding the parameters involved in the protective immunity to tuberculosis in animal species.Based on the experimental studies in mice infected with Mtb, the cytokines considered to be involved in resistance to tuberculosis are interferon-γ produced by T cells (Orme, 2003) and IL-12 from antigen presenting cells that further leads to the induction of IFN-γ (Flynn & Chan, 2001).However, it has been demonstrated that no correlation exists between high Th1 response, at least high IFN-γ, induced during natural infection and protection imparted after immunization (Hovav et al., 2003;Olsen et al., 2000;Sable et al., 2005).As there is no good in vitro correlate of protective immunity, one is left with the daunting task of screening each potential mycobacterial antigen for its protective efficacy.
Excretory/secretory polypeptides of mycobacteria released in culture filtrates have raised much interest as a source of protective antigens owing to their ability to induce a protection at a similar level as did vaccination with BCG (Brandt et al., 2000;Demissie et al., 1999;Sable et al., 2005).So, to identify protective antigen(s) from Mtb culture filtrate with potential use as TB vaccine candidates, a simple and effective method was employed.The Mtb total culture filtrate was fractionated into fifteen non-overlapping narrow molecular mass fractions consisting of 2-5 protein bands by whole gel elutor (Figure 1).The elutor acts as an electrodialyzer and removes SDS from proteins, leaving culture filtrate protein fractions in a non-toxic physiological buffer, which can be used, directly in a various cellular assays.Till date, various narrow molecular mass fractions obtained by similar fractionation technique have been evaluated for immunological reactivity and shown to stimulate lymphocytes of humans (Demissie et al., 1999;Wilkinson, Belisle, Mincek, Wilkinson, & Toossi, 2000).Studies based on human immune recognition have shown that narrow molecular mass culture filtrate fractions of Mtb are recognised by healthy tuberculosis contacts and not by TB patients (Boesen et al., 1995).Successful protection against Mtb infection in an animal model has long been regarded as the best test of a vaccine's efficacy.Therefore, a direct comparative analysis of individual narrow molecular mass fractions adjuvanted with Dimethyl dioctyldecylammonium bromide (DDA) was carried out.To the best of our knowledge, this study is the first report in which the protective efficacy of culture filtrate fractions of whole secretory proteome of Mtb was carried out against experimental tuberculosis in two different animal models using DDA as adjuvant.The adjuvant DDA was used because it is known to induce strong cell mediated immune responses to various mycobacterial antigens without any toxic effects (Brandt et al., 2000;Hilgers & Snippe, 1992;Leal, Smedegard, Andersen, & Appelberg, 1999;Lindblad, Elhay, Silva, Appelberg, & Andersen, 1997).Out of the fifteen fractions of mycobacterial secretory proteome, the mice immunized with fraction F7 (20-24kDa) showed highest protection followed by fraction F11 (37-42kDa) and then by F8 (25-30kDa) (Fig. 2).The mouse is the most commonly used experimental animal model to study human tuberculosis.Initial screening is usually done in the mouse, which has the advantage of being relatively inexpensive, infection with Mtb induced lesions like those seen in the natural disease in humans and availability of a wide variety of immunological reagents such as monoclonal antibodies against surface markers and cytokines (Gupta & Katoch, 2009;Orme, McMurray, & Belisle, 2001).
To further evaluate the protective potential of fractions F7, F8 and F11, the guinea pig model of tuberculosis was tested.The guinea pig model is especially relevant to human tuberculosis clinically, immunlogically and pathologically (Grover et al., 2009;Orme et al., 2001).In addition, guinea pigs are outbred and, therefore, successful vaccination reduces concerns that protective immunity could be mouse strain specific or MHC specific (Skeiky et al., 2005).Significant protection was observed against intramuscular challenge with Mtb H 37 Rv by immunization with fraction F7 followed by fraction F11 and F8 as compared to sham-immunized animals (Figure 3).These results are in concordant with earlier studies in which immunization with secretory proteins of Mtb have been shown to induce protective immunity in guinea pigs (Jain et al., 2008).These studies clearly demonstrate that in the mycobacterial secretory proteome, there are certain regions (21-42kDa), which are able to induce protection both in mouse and guinea pig models even better than BCG.Therefore, subunit vaccine development should be focused on the proteins of this region only.
The induction of Type-1 response dominated by IFN-γ and IL-12 following subunit vaccine immunization is usually considered essential for protection against tuberculosis (Leal et al., 1999;Wu et al., 2008).Experimentally, mice deficient in IFN-γ or in IL-12 have been reported to be highly susceptible to challenge with Mtb (Cooper, Magram, Ferrante, & Orme, 1997).Screening of narrow molecular mass fractions in mice immunized with STCF representing the whole secretory proteome of Mtb revealed highest release of IFN-γ by antigens with molecular masses >42kDa comprising fractions F12 to F15 whereas, higher secretion of IL-12 was elicited by fraction F7 and F8 (20-30kDa) (Fig. 4B and C).Taking the cut off 500pg/ml for both IFN-γ and IL-12, it was observed that these cytokines were mainly induced by the proteins in the molecular weight range of 20-42kDa comprising fractions F7 to F11 (Fig. 6).Surprisingly, the negligible protection observed in the case of animals immunized with fraction F12 to F15 was associated with high concentration of IFN-γ secretion in response to in vitro stimulation with these proteins.These observations can be explained on the findings of Majlessi et al (Majlessi et al., 2006), who recently demonstrated that induction of strong Th1 response (IFN-γ) to Ag85A and ESAT-6 did not confer any protection against Mtb challenge in C 57 BL/6J mice.Moreover, various researchers have also found a positive correlation between the levels of the IFN-γ and the extent of pathology (Hovav et al., 2003;Turner et al., 2000;Vordermeier et al., 2002;Yeremeev et al., 2000).The possibility of the fractions F12 to F15 comprising decoy proteins diverting immune responses towards biased high Th1 type resulting in pathology cannot be ruled out.Since histopathological studies were not carried out in the present study, pathological changes in the vaccinated animals cannot be commented upon.Screening of culture filtrate fractions for "built-in" immunogenicity in guinea pigs immunized with total culture filtrate also yielded predominant recognition of proteins in the molecular weight range of 18-45kDa, comprising fractions F6 to F12 (Figure 5A and B).Due to the lack of specific reagents for use in guinea pigs, immune responses only in terms of splenocyte proliferation and humoral IgG only were monitored.
Considering the similar immune recognition of STCF fractions in both mice and guinea pigs following immunisation with STCF, all the fifteen fractions of secretory proteome were evaluated for their immunoreactivity during early stage of mycobacterial infection in mouse model only.In contrast to immunogenicity studies, the cytokine release in infected animals was found to be directed towards F1-F11 indicating that there are certain proteins in the mycobacterial secretory proteome which have although good antigenicity but have poor in-built immunogenicity.Earlier we have suggested that it is important to consider the in-built immunogenicity for the vaccine design (Sable, Kalra, Verma, & Khuller, 2007).The protective fractions F7, F8 and F11 were found to be recognized by host immune system both during infection as well as immunization (Figure 4B and 4C;Figure 6A and 6B).On the other hand, there were certain non-protective fractions, which were being either recognized during infection (F1) or during immunization (F15).The poor protective potential of F1 comprising low molecular weight proteins is supported by earlier studies demonstrating that in spite of unique ability to produce marked levels of IFN-γ, majority of low molecular mass proteins identified as suitable vaccine candidates in T cell assays, failed to impart significant protection against experimental tuberculosis as compared to BCG (Kamath, Feng, Macdonald, Briscoe, & Britton, 1999;Morris, Kelley, Howard, Li, & Collins, 2000).If such an assumption is true, the progress of these low molecular weight polypeptides as useful vaccine candidates for new prophylactic interventions in long run, from animal experiments to clinical trials and for actual human use seems doubtful.The observations of the present study suggest that one cannot solely rely on the identification of candidate molecules on the basis of T cell antigenicity in existing infection models.Our initial characterizations on the basis of available monoclonal / polyclonal antibodies against well-characterised Mtb proteins reactivity (Sable et al., 2005;Sonnenberg & Belisle, 1997) with purified protein from these protective fractions are represented in Table 1.
Hence, Mtb antigens comprising protein fractions F7, F8 and F11 in the molecular weight range of 20-24 kDa, 25-30 kDa and 37-42 kDa respectively, which showed protective efficacy in two animal models, can be considered as ideal candidates for the constitution of a subunit vaccine against tuberculosis.These results pinpoint the influence of antigen selection and emphasize the necessity of pursuing such kind of more studies to clearly delineate the markers of protection against TB for the selection of successful antituberculous vaccine candidates.

Figure 1 .
Figure 1.Fractionated culture filtrate of M. tuberculosis H 37 Rv The culture filtrate was separated by 16% SDS-PAGE and fractionated into narrow molecular mass fractions by electroelution.The fractions were analysed by SDS-PAGE and silver staining.Lane 1-STCF, Lane 2-molecular weight markers (kilo dalton) and lane 3 to 17-fifteen culture filtrate fractions

Figure 2 .
Figure 2. Comparative account of viable M.tuberculosis H 37 Rv (Log 10 units) observed in the organs of Balb/C mice vaccinated with BCG or different experimental subunit vaccines at 4 weeks post infection The results are expressed as mean Log 10 CFUs ± SD of five animals per group.Significant differences were determined by ANOVA and student 't' test.*** p<0.001, ** p<0.01 w.r.t naive controls.$ p<0.05 w.r.t.BCG immunized group.

Figure 3 .
Figure 3. Comparative account of viable M. tuberculosis H 37 Rv bacilli (Log 10 units) observed in the lungs (A) and spleen (B) of guinea pigs vaccinated with BCG or different experimental subunit vaccines at 6 weeks post infection Results are expressed as Log 10 CFUs (mean ± SD) of five animals per group.The experiment was repeated two times with similar results.Significant differences were determined by ANOVA and student 't' test.*** p<0.001 as compared to naive controls.$ p<0.05 as compared to BCG immunized group.

Figure 4 .
Figure 4. Immune responses to narrow molecular mass culture filtrate protein fractions in STCF immunized mice after three weeks of immunization A) Splenocyte proliferation response.Results are presented as stimulation indices and each value has been calculated as mean ± standard deviation (error bars) of stimulation indices obtained in all the five animals in each group.B) IFN-γ levels and C) IL-12p40 levels.Results represent mean ± S.D values of five mice tested individually The IFN-γ and IL-12p40 levels in unstimulated wells (culture without antigens) were below 100 and 50 pg/ml, respectively.

Figure 5 .
Figure 5. Lymphocyte proliferation (A) and serum IgG (B) responses towards narrow molecular mass culture filtrate fractions of STCF in guinea pigs immunized with STCF a) The results are expressed as SI (mean ± SD) of five immunized animals per group.b) Results represent antigen specific IgG in term of OD at 492nm (mean ± S.D) in the sera of five animals tested individually

Figure 6 .
Figure 6.Cellular immune responses towards narrow molecular mass culture filtrate fractions of STCF in mice challenged with M. tuberculosis H 37 Rv IFN-γ Levels (A) and IL-12p40 levels (B) were measured in the supernatant of splenocytes in vitro stimulated with culture filtrate fractions or STCF.The IFN-γ and IL-12 production in unstimulated wells (culture without antigens) was below 250 and 120 pg/ml, respectively.Results represent mean ± S.D values of five mice from each group tested individually.

Table 1 .
Panel of monoclonal/polyclonal antibodies used to identify purified polypeptides/antigens from protective fractions of Mtb H 37 Rv by ELISA Antibodies against CFPs were provided by Dr John T. Belisle, Colorado State University, Fort Collins, Colorado, USA or kind gift from Dr. Peter Andersen and Dr. I. Rosenkrands, Statens Serum Institute, Copenhagen, Denmark. a