Immunomodulatory effects of recombinant BCG expressing MSP-1C of Plasmodium falciparum on LPS- or LPS+IFN-γ-stimulated J774A.1 cells

Abstract

Macrophage phagocytosis is the first line of defense of the innate immune system against malaria parasite infection. This study evaluated the immunomodulatory effects of BCG and recombinant BCG (rBCG) strains expressing the C-terminus of the merozoite surface protein-1 (MSP-1C) of Plasmodium falciparum on mouse macrophage cell line J774A.1 in the presence or absence of lipopolysaccharide (LPS) or LPS + IFN-γ. The rBCG strain significantly enhanced phagocytic activity, production of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, nitric oxide (NO), and inducible nitric oxide synthase (iNOS) as compared with parental BCG strain, and these activities increased in the presence of LPS and LPS+IFN-γ. Furthermore, the rBCG strain also significantly reduced the macrophage viability as well as the rBCG growth suggesting the involvement of macrophage apoptosis. Taken together, these data indicate that the rBCG strain has an immunomodulatory effect on macrophages, thus strengthen the rational use of rBCG to control malaria infection.

Introduction

Malaria infection, particularly from Plasmodium falciparum, is a leading cause of mortality and morbidity in African and Southeast Asian countries because of the parasite’s ability to adapt to a wide range of conditions inside and outside of the host.^1,2 Various intervention and eradication programs have been implemented by the World Health Organization (WHO) and non-governmental organizations (NGOs), but the prevalence of malaria is increasing, especially in young children. This problem might be due to various possible contributing factors such as genetic diversity,^3,4 the emergence of multidrug-resistant strains^2,5-7 and environmental factors, including climate change.^8,9 Knowledge regarding the mechanisms by which malaria parasites are eliminated by the host immune system is still not fully understand and sometimes controversial. Therefore, a full understanding of protection against parasites by the immune system will provide information for improved malaria prevention and the development of an effective vaccine.

Innate immunity is important in the early control of malaria infection because it restricts parasite replication and impedes the progression of severe and fatal disease.^10,11 Macrophages are a major type of phagocytic cell involved in innate immune protection against malaria. Activated macrophages secrete pro-inflammatory cytokines such as tumor necrosis factor (TNF)-α and interleukin (IL)-1β to stimulate the function of other immune cells and mediate the release of toxic metabolites such as nitric oxide (NO), an unstable free radical gas produced by inducible nitric oxide synthase (iNOS). TNF-α and IL-1β are important in killing parasites and inhibiting parasite replication.^12-14 Furthermore, these cytokines have been reported to protect against the development of cerebral malaria and control parasitemia in animal and human models.^15,16 NO, on the other hand has potent parasiticidal properties against P. falciparum¹⁷ and is an important effector in limiting parasitemia.¹⁸ These mechanisms are vital in controlling malaria infection in both animals and humans.^10,13,19,20 However, overproduction of inflammatory cytokines and NO leads to malarial anemia.^20-22

Development of recombinant DNA using live vectors from viruses^23,24 and bacteria^25-28 to deliver foreign antigens to the immune system has become a popular technique nowadays for developing new generation vaccines. Mycobacterium bovis bacille Calmette–Guérin (BCG), the only vaccine currently available for the prevention of tuberculosis, is among the most extensively used vector for developing recombinant vaccines for other diseases, including malaria.^29-31 Using this strategy, our group previously cloned and expressed a synthetic gene encoding the C-terminus of the merozoite surface protein-1 (MSP-1₁₉; referred in this study as MSP-1C) in a recombinant BCG (rBCG016; referred in this study as rBCG) construct.^32,33 MSP-1C is a 19 kDa blood-stage antigen produced by proteolysis of a high molecular weight precursor, 195 kDa MSP-1 protein. During merozoite invasion of red blood cells, the protein is processed by proteases and released from the parasite surface except for a 19 kDa C-terminal region of MSP-1 which remain on the surface of the invading merozoites.³⁴ This protein is responsible for protective immunity against malaria infection,^35,36 and is one of the most promising malaria vaccine candidates.^29,37,38

We have previously described antibodies produced against the rBCG vaccine inhibited P. falciparum 3D7 merozoite invasion of red blood cells in vitro.³³ Moreover, the rBCG strain also stimulated higher cellular and humoral immune responses in animal model.³³ However, the innate immune response to this strain has not been characterized fully. Previously, we showed that the rBCG strain capable of stimulating phagocytic activity and pro-inflammatory cytokines production in macrophages at different incubation times, 24 h, 48 h, and 72 h.³⁹ In this report, we further investigated the immunomodulatory ability of the rBCG strain in macrophages in the absence or presence of lipopolysaccharides (LPS) alone or in combination with interferon gamma (IFN-γ).

Results

Detection of MSP-1C in rBCG-infected J774A.1 cells

The parental BCG and rBCG strains were subjected to immunocytochemistry analysis using SuperPicture^TM 3rd Gen IHC detection kit probed with specific MSP-1C antibody. As indicated in Figure 1, MSP-1C protein expression was detected in the cytoplasm of rBCG-infected cells (Fig. 1C) but not in BCG-infected cells (Fig. 1B) or uninfected cells (Fig. 1A), indicating that the MSP-1C is stable in the rBCG strain.

Figure 1. Immunocytochemistry detected the expression of MSP-1C protein in macrophage cells. Comparison of macrophage morphology: (A) Uninfected cells, (B) BCG-infected cells, (C) rBCG-infected cells. The positive cytoplasmic staining appears as a dark brown. 20× magnification.

Phagocytic activity of J774A.1 cells against BCG and rBCG

In order to determine whether the BCG and rBCG strains can enhance phagocytosis of macrophage, phagocytic activity was performed. The number of mycobacterium per macrophage was assessed in the absence and presence of LPS or/and IFN-γ and the phagocytic activity of macrophages was expressed as relative population of mycobacterium in 100 counted macrophages. Figure 2A shows the representative microscopic images of J774A.1 cells that phagocytosed BCG or rBCG. The positive staining of mycobacterium appears as red. Microscopic examination of the infected macrophages indicated that both BCG and rBCG strains are efficiently phagocytosed by the macrophages. No obvious changes were detected in the morphology of the mycobacteria as well as the infected macrophages.

Figure 2. Phagocytic activity of J774A.1 cells infected with BCG or rBCG at 48 h incubation with or without LPS and LPS + IFN-γ stimulation. (A) Macrophage morphology, 100X magnification; (I) uninfected, (II) 48 h post- infection with BCG, and (III) 48 h post-infection with rBCG. The positive staining for Mycobacterium appears red. (B) Phagocytic activity of infected J774A.1 cells. Data are presented as the mean phagocytic index ± SEM from 3 independent experiments (n = 3). Data were analyzed by RM ANOVA and *P < 0.05 significantly different.

As shown in Figure 2B, both BCG and rBCG infected cells stimulated higher phagocytic activity in the infected macrophages in response to LPS, and further increased when the cells stimulated with LPS and IFN-γ. However, in all conditions, the phagocytic activity of rBCG-infected cells were significantly higher (27.27 ± 0.37 PI in the absence of LPS; 64.2 ± 0.8 PI in the presence of LPS; 102.0 ± 0.41 PI in the presence of LPS+IFN-γ) than BCG-infected cells (16.8 ± 0.23 PI in the absence of LPS; 52.4 ± 0.76 PI in the presence of LPS; 74.47 ± 0.33 PI in the presence of LPS+IFN-γ) (P < 0.05). These observations suggest that the rBCG strain is more potent in inducing phagocytosis of macrophages than BCG, LPS alone or in combination with IFN-γ. The presence of LPS and LPS+IFN-γ synergistically enhanced the phagocytosis activity of the BCG and rBCG-infected macrophages.

J774A.1 cell viability in response to BCG and rBCG

To investigate whether phagocytosis of BCG and rBCG by macrophages influence the growth of the cells, cell viability assay was conducted. As indicated in Figure 3, macrophage viability was reduced when the cells infected with BCG and rBCG and the reduction was significantly higher in the presence of LPS and LPS+IFN-γ. In the absence of LPS stimulation, the viability of rBCG infected cells was significantly lower (56.1 ± 0.6%) than those infected with BCG (64.76 ± 0.3%) and uninfected cells (100 ± 0.0%) (P < 0.05). In the presence of LPS, the viability of BCG- and rBCG-infected cells decreased significantly compared with uninfected cells (P < 0.05) (51.06 ± 1.3% for rBCG; 48.33 ± 1.3% for BCG and 77.21 ± 0.7% for uninfected cells). A similar pattern was observed when the cells stimulated with LPS+IFN-γ. The reduction in cell viability was significantly higher in BCG- and rBCG-infected cells (47.05 ± 0.1% for rBCG and 55.06 ± 0.1% for BCG) compared with those observed in uninfected cells. However the viability of the BCG-infected cells and the rBCG-infected cells was not significantly different in the presence of LPS or LPS+IFN-γ stimulation. Overall, this finding indicates that the phagocytosis activity of macrophages influences the viability of the cells.

Figure 3. Viability of J774A.1 cells infected with BCG and rBCG at 48 h incubation with or without LPS and LPS + IFN-γ stimulation. Data are presented as the mean viability ± SEM from 3 independent experiments (n = 3). Data were analyzed by RM ANOVA and *P < 0.05 significantly different.

Intracellular growth of BCG and rBCG in J774A.1 cells

To determine whether BCG and rBCG infections influence the growth of the mycobacteria in the infected macrophages, the CFU of the BCG and rBCG strains was calculated. As shown in Table 1, the mean CFU of rBCG was significantly lower (1.24 ± 0.011 × 10⁷ CFU) than control BCG (1.46 ± 0.013 × 10⁷ CFU) in the absence of LPS stimulation. It is observed that the growth of BCG and rBCG reduced markedly in response to LPS and LPS+IFN-γ. In the presence of LPS, the rBCG growth was significantly lower (1.08 ± 0.008 × 10⁷ CFU) than BCG (1.27 ± 0.007 × 10⁷ CFU). However, no significant difference was observed for BCG and rBCG growth in LPS+IFN-γ-stimulated cells.

Table 1.

Treatment		BCG	rBCG
	Mean ± SEM (×107 CFU)			Mean ± SEM (×107 CFU)
Unstimulated		1.46 ± 0.013	1.24 ± 0.011*
LPS		1.27 ± 0.007	1.08 ± 0.008*
LPS + IFN-γ		0.90 ± 0.008	0.87 ± 0.014

Intracellular growth of BCG and rBCG in J774A.1 cells 48 h post-infection with or without LPS and LPS + IFN-γ stimulation. Data are presented as the mean CFU ± SEM from three independent experiments (n=3). Data were analyzed by RM ANOVA and * p < 0.05 significantly different from BCG.

IL-1β and TNF-α production in infected J774A.1 cells

To test the hypothesis that phagocytosis of BCG and rBCG results in an inflammatory cytokine production, the presence of IL-1β and TNF-α was investigated in the culture supernatants of the macrophages by ELISA. Our data showed that normal cells produced higher basal levels of IL-1β and TNF-α, and the production of these cytokines increased when the cells stimulated with LPS or LPS+IFN-γ. As expected, both IL-1β and TNF-α were detected in the culture supernatants of infected cells, and the production of these cytokines was significantly higher in rBCG infected cells than that of BCG infected cells. However, TNF-α concentration was detected much higher in the supernatant of the cells (>600 pg/mL) in comparison to IL-1β concentration (<230 pg/mL) respectively (Fig. 4A and 4B).

Figure 4. Production of (A) IL-1β and (B) TNF-α in J774A.1 cells infected with BCG and rBCG after 48 h incubation with or without LPS and LPS + IFN-γ stimulation. Data are presented as the mean concentration of cytokine ± SEM from 3 independent experiments (n = 3). Data were analyzed by RM ANOVA and *P < 0.05, significantly different.

Cells infected with rBCG strain produced significant higher IL-1β levels (39.82 ± 4.5 pg/mL) than in BCG-infected cells (21.43 ± 4.5 pg/mL) or in uninfected cells (10.76 ± 1.4 pg/mL) (P < 0.05) and the production of this cytokine increased when the cells stimulated with LPS (109.69 ± 9.0 pg/mL for rBCG infected cells; 72.91 ± 9.0 pg/mL for BCG-infected cells; 42.39 ± 1.8 pg/mL for uninfected cells) and further increased in response to LPS+IFN-γ (227.37 ± 4.5 pg/mL for rBCG-infected cells; 186.92 ± 4.5 pg/mL for BCG-infected cells; 68.53 ± 4.5 pg/mL for uninfected cells).

Similar to the IL-1β production, rBCG infection exhibited higher TNF-α production in the culture supernatant of the macrophages (998.36 ± 25.97 pg/mL) than BCG infection (622.72 ± 27.75 pg/mL) or uninfected cells (345.26 ± 5.6 pg/mL) in the absence of LPS. Furthermore, the TNF-α production was increased when the cells stimulated with LPS (2084.74 ± 38.42 pg/mL for rBCG; 1523.41 ± 42.84 pg/mL for BCG; 1627.99 ± 47.53 pg/mL for uninfected cells) and further increased when stimulated with LPS+IFN-γ (2430.5 ± 70.43 pg/mL for rBCG; 1965.21 ± 2.13 pg/mL for BCG; 1845.69 ± 29.65 pg/mL for uninfected cells). These results suggest that rBCG enhances LPS and LPS+ IFN-γ -induced macrophage cytokine secretion more potent than parent BCG control.

IL-1β and TNF-α mRNA expression in infected J774A.1 cells

To assess the effects of BCG and rBCG infections on IL-1β and TNF-α mRNA expression, the infected macrophages were analyzed semi-quantitatively using RT-PCR. As shown in Figure 5A, normal cells produced a low basal expression of IL-1β and TNF-α and elevated IL-1β mRNA and TNF-α mRNA were observed when the cells infected with BCG and rBCG. The expression of these mRNA cytokines increased significantly in response to LPS and LPS+IFN-γ similar to that of cytokine secreted in cell supernatant (P > 0.05).

Figure 5. Expression of IL-1β and TNF-α mRNA in J774A.1 cells infected with BCG and rBCG at 48 h incubation with or without LPS and LPS + IFN-γ stimulation. (A) Representative result for IL-1β, TNF-α, and β-actin mRNA. Quantification of intensity of (B) IL-1β and (C) TNF-α mRNA. Data are presented as the mean relative intensity of mRNA ± SEM from 3 independent experiments (n = 3). Data were analyzed by RM ANOVA and *P < 0.05, significantly different.

In the absence of LPS stimulation, the IL-1β mRNA expression in rBCG-infected cells was 0.48 ± 0.005 while in BCG-infected cells was 0.42 ± 0.006 and in uninfected cells was 0.37 ± 0.001 respectively. The IL-1β mRNA expression was further increased when the cells stimulated with LPS (1.09 ± 0.007 in rBCG-infected cells; 0.91 ± 0.005 in BCG-infected cells; 0.42 ± 0.005 in uninfected cells). The highest IL-1β mRNA expression was detected in cells stimulated with LPS+IFN-γ (1.11 ± 0.007 in rBCG-infected cells; 0.99 ± 0.005 in BCG-infected cells; 0.8 ± 0.006 in uninfected cells) (Fig. 5B).

Similarly, TNF-α mRNA expression was also increased when the cells stimulated with LPS and LPS+IFN-γ. In the absence of LPS, the TNF-α mRNA expression was 0.29 ± 0.011 in rBCG infected cells, 0.2 ± 0.007 in BCG-infected cells and 0.21 ± 0.006 in uninfected cells. In the presence of LPS, TNF-α mRNA expression was significantly increased to 0.58 ± 0.015 in rBCG-infected cells, 0.39 ± 0.012 in BCG-infected cells and 0.49 ± 0.011 in uninfected cells. This expression was further increased to the maximum level in response to LPS+IFN-γ (0.71 ± 0.018 in rBCG-infected cells, 0.69 ± 0.018 in BCG-infected cells and 0.57 ± 0.01 in uninfected cells). However, the TNF-α mRNA expression in rBCG- and BCG-infected cells was not significantly different.

Nitric oxide production in infected J774A.1 cells

As shown in Figure 6, NO production was detected in all culture supernatants of infected and uninfected cells. In the absence of LPS, NO production in rBCG-infected cells were significantly higher (5.11 ± 0.21 µM) then in BCG-infected cells (2.73 ± 0.15 µM) or in uninfected cells (1.22 ± 0.13 µM) (P < 0.05). It is noted that overall NO production by infected and uninfected cells increased in response to LPS and LPS+IFN-γ. In the presence of LPS, NO production by rBCG-infected cells was significantly greater (36.65 ± 0.56 µM) than BCG-infected cells (22.73 ± 0.73 µM) or uninfected cells (7.88 ± 0.64 µM) (P < 0.05). As expected, NO production was further increased when the cells stimulated with LPS+IFN-γ. In the presence of LPS+IFN-γ, rBCG-infected cells exhibited the highest NO production (84.92 ± 0.17 µM) than those in BCG infected cells (81.03 ± 0.1 µM) and uninfected cells (79.52 ± 0.15 µM) (P < 0.05).

Figure 6. Production of nitric oxide in J774A.1 cells infected with BCG and rBCG at 48 h incubation with or without LPS and LPS + IFN-γ stimulation. Data are presented as the mean concentration of nitric oxide ± SEM from 3 independent experiments (n = 3). Data were analyzed by RM ANOVA and *P < 0.05, significantly different.

iNOS protein expression in infected J774A.1 cells

Western blot analysis showed an appearance of a band at approximately 130 kDa (the expected size of iNOS) in both BCG- and rBCG-infected cells as well as in uninfected cells when the blots were incubated with anti-iNOS (Fig. 7A) indicating that the iNOS protein expression was detected in all infected and uninfected cells in the absence or presence of LPS or LPS+IFN-γ. Densitometric analysis of the result revealed that iNOS protein expression was significantly higher in rBCG-infected cells (0.9 ± 0.02) than in BCG-infected cells (0.66 ± 0.01) or in uninfected cells (0.34 ± 0.01) (P < 0.05). In the presence of LPS, iNOS protein expression in rBCG-infected cells were significantly increased (1.16 ± 0.02) than in BCG-infected cells (0.88 ± 0.01) or in uninfected cells (0.41 ± 0.01). Similarly, in the presence of LPS+IFN-γ, iNOS protein expression in rBCG-infected cells were significantly higher (1.24 ± 0.02) compared with uninfected cells (1.1 ± 0.01) (P < 0.05) but not significant when compared with BCG-infected cells (1.2 ± 0.02) (Fig. 7B).

Figure 7. Expression of iNOS protein in J774A.1 cells infected with BCG and rBCG at 48 h incubation with or without LPS and LPS + IFN-γ stimulation. (A) Representative result for iNOS and β-actin. (B) Quantification of intensity of iNOS protein. Data are presented as the mean relative intensity of iNOS ± SEM from 3 independent experiments (n = 3). Data were analyzed by RM ANOVA and *P < 0.05, significantly different.

Discussion

Innate immunity is important in vaccine-induced protection against malaria infection. Innate immunity not only crucial for the first line of defense against invading malaria parasites but it also plays a vital role in the activation of adaptive immunity. Macrophages are one of the important innate immune cells that have pivotal roles in malaria infection.⁴⁰ They act as phagocytes to destroy the parasites, and produce pro-inflammatory cytokines, oxygen radicals, and NO, which allow them to eliminate the parasites from infected red blood cells. At the same time, activated macrophages also act as antigen presenting cells by expressing major histocompatibility complex and co-stimulatory molecules that further activates T cells to induce cell-mediated immunity for elimination of the parasites.⁴¹

Several lines of evidence suggest that mycobacterium such as BCG capable to stimulate macrophage activation.^42-45 Indeed, knowledge regarding the ability of a recombinant BCG to stimulate macrophage activation is well established. For example Luo et al.⁴⁶ have demonstrated that a recombinant BCG expressing mouse IL-18 (rBCG-mIL-18) strain significantly increased macrophage cytotoxicity against bladder cancer MBT-2 cells and induced higher levels of IFN-γ and TNF-α in the culture supernatant of the cells than control BCG. Xu et al.⁴⁷ also demonstrated that a recombinant BCG strain that secretes the chimeric protein of Ag85B and ESAT-6 (rBCG-A(N)-E-A(C)) induced higher expression levels of CD86, CD80, CD40, and HLA-DR as well as increased in TNF-α production of THP-1 cells. However, information concerning a possible mechanism on how rBCG is more potent than parent BCG is limited and not fully understood.

Most of malaria vaccine studies focus on the ability of the vaccine candidates to stimulate adaptive immunity although both of innate and adaptive responses are equally important for protection against malaria. To support our previous data concerning that the rBCG strain expressing a synthetic gene of the P. falciparum MSP-1C is a good vaccine candidate for blood-stage malaria infection, we evaluated the ability of the rBCG strain in stimulating an innate immune response in macrophages in the presence and absence of classical activators, LPS and IFN-γ. We demonstrate here for the first time that the rBCG strain increased macrophage phagocytic activity and induced higher production of NO, iNOS, and pro-inflammatory cytokines such as TNF-α and IL-1β than those produced by parent BCG-infected macrophages. The results are consistent with our previous finding³⁹ and other studies^46-50 demonstrating that a recombinant BCG is more potent in activating macrophages than parental BCG. These findings support our previous data showing that the rBCG vaccine capable of inducing a strong cell mediated and humoral immune responses in animal model. Macrophage is a professional antigen presenting cell. After internalizing antigen through phagocytosis, it presented the processed antigen to T cell through the class II MHC molecule. This interaction stimulated co-stimulatory signal leading to activation of the T cell.^51,52 Therefore, enhanced macrophage phagocytosis will definitely increase adaptive immune responses in the system. Our previous data also proved that the antibodies produced against rBCG strain exhibited a strong protection against merozoite invasion of erythrocytes.³³ Thus, these data strengthen the justification on the use of rBCG as a good vaccine candidate against blood-stage malaria infection. A strong innate immune response is important in early blood-stage malaria infection to prevent the parasite from entering human red blood cells and cause disease, while strong adaptive immune responses may protect against malaria by inhibiting merozoite invasion of erythrocytes as well as enhance clearance of infected erythrocytes from the circulation.^53,54

The mechanism by which rBCG induces more inflammatory response than parental BCG and LPS is not yet known. A previous report showed that glycosylphosphatidylinositol (GPI) anchors of merozoite membrane proteins released from rupturing malaria-infected erythrocytes capable to induce low levels of TNF-α production in macrophages.⁵⁵ In addition, Arama et al.⁵⁰ also demonstrated that rBCG expressing the circumsporozoite antigen of Plasmodium is more potent in inducing TNF-α production than the parent BCG strain. Based on these evidences, we speculated that the presence of MSP1-C antigen of P. falciparum in the rBCG strain increased the ability of the mycobacterium to stimulate higher inflammatory response in macrophages. MSP-1C and BCG alone are capable to induce macrophage activation, and the combination of MSP-1C and BCG in rBCG strain increased the activation. This hypothesis is also supported by Makino et al.⁵⁶ who postulated that increased in GM-CSF cytokine production by macrophages infected with a recombinant BCG expressing a major membrane protein (MMP)-II of Mycobacterium leprae (BCG-SM) was associated with the secretion of MMP-II protein from BCG-SM. However, the actual mechanism on how the presence of MSP-1C in the rBCG strain enhances innate immune response in macrophage will be determined in future studies.

The findings also showed that the production of inflammatory response by the infected macrophages or uninfected macrophages increased in the presence of LPS and LPS+IFN-γ stimulation, which indicated that LPS and LPS+IFN-γ capable to stimulate macrophage activation either individually or synergistically with BCG and rBCG infections. This observation is in agreement with previous data showing that macrophage can be activated not only by infection, but also by immunostimulatory cytokines such as IFN-γ, IL-1β, IL-6, or TNF-α, and bacterial products such as LPS.⁵⁷

Activated macrophages produce large amount of NO, a reactive mediator produced via the NADPH- and l-arginine-dependent enzyme nitric oxide synthase as a defense strategy against infections.^58,59 The role of NO in the host’s response to malaria infection has been increasingly investigated. Previous reports have suggested that NO is essential in preventing or moderating the severe manifestations of malaria infection. Moreover, NO has also been reported to have direct anti-parasitic activity that can kill P. falciparum in vitro.^17,18,59-62 Therefore, enhanced induced NO production in macrophages infected with rBCG vaccine indicating the ability of the rBCG strain to protect against malaria infection.

NO production in macrophages is catalyzed by a family of enzymes called nitric oxide synthases that include eNOS, iNOS, and nNOS. To determine whether iNOS is responsible for stimulating the production of NO in response to BCG and rBCG strains, the expression of iNOS protein was detected in cells using western blot. As expected, a strong iNOS protein expression was observed in rBCG-infected macrophages compared with BCG-infected macrophages or uninfected macrophages, suggesting that iNOS enzyme was involved in regulating NO production by the macrophages.^63,64

NO and iNOS productions have been reported to be correlated with TNF-α and IL-1β productions since both processes involved NF-κB and MAPKinases regulation.^65,66 In this study, both BCG and rBCG infections also stimulated higher production of TNF-α and IL-1β, suggesting the involvement of these cytokines in stimulating the production of NO and iNOS by the macrophages. Although TNF-α and IL-1β are associated with severe malaria and cerebral malaria in children,^67,68 they are also important for parasite killing during malaria infection.^12-16,69-72 Similar to other results, the production of TNF-α and IL-1β by infected macrophages increased with LPS and LPS+IFN-γ. These results also consistent with several other studies showing that BCG stimulates the production of TNF-α and IL-1β by macrophages,^73-75 and the stimulation is enhanced in the presence of LPS⁷³ and further enhanced with the presence of LPS+IFN-γ.^76,77

We also demonstrated that phagocytosis of the BCG and rBCG strains not only reduced the viability of the macrophages but also resulted in a substantial reduction in replication of the mycobacteria. These reductions were significantly greater in macrophages infected with rBCG than those of the parent BCG. This finding is in agreement with a previous work by Venketaraman et al.⁷⁸ who demonstrated that infection of macrophages with BCG reduced the replication of the mycobacterium. However, the actual mechanism of how BCG and rBCG strains kill infected macrophages and mycobacteria, and how this mechanism affects the immune response against the vaccines is unclear. Previous studies with other types of mycobacterium such as M. avium and M. tuberculosis have suggested that this phenomenon might be due to the generation of reactive oxygen intermediates, reactive nitrogen intermediates, and/or acidification of phagosomes by the infected macrophages.^79,80 On the other hand, other studies also suggested that this mechanism might involve macrophage apoptosis.^81,82 Indeed, Grode et al.⁴⁸ also showed that a recombinant BCG secreting the listeriolysin (Hly) of Listeria monocytogenes capable of inducing apoptosis of the infected macrophages. Apoptosis involves the packaging of intracellular bacilli into membrane-bound vesicles, which facilitates antigen presentation and possibly enhances the antimicrobial efficacy of newly recruited macrophages to ensure efficient control of the spread and multiplication of the pathogen.⁸³ This mechanism is important for parasite clearance and could contribute to innate immunity during malaria infection. However, the ability of the rBCG in stimulating macrophage apoptosis is yet to be determined. It would be interesting to investigate whether the reduction in macrophage viability correlates with macrophage apoptosis.

A number of studies have reported that macrophage apoptosis and inhibition of BCG growth in macrophages increase when cells are stimulated with LPS or LPS+IFN-γ.^78,80,84-86 This was also observed in our study. The reduction in macrophage viability as well as BCG and rBCG replication was significantly increased when the infected macrophages were stimulated with LPS or LPS+IFN-γ which indicated that LPS and IFN-γ can work synergistically with bacterial infection to promote more efficient apoptosis in the infected macrophages.

In conclusion, these in vitro data indicated that the rBCG candidate vaccine has proven to be capable of inducing higher inflammatory responses of the mouse macrophage cell line J774A.1. The candidate vaccine also stimulated the reduction of macrophage viability as well as the mycobacterium which suggesting the involvement of macrophage apoptosis. The data presented here strengthen the evidence that the rBCG strain is a good vaccine candidate against malaria. However, the mechanism how the rBCG strain modulates the innate immune response in macrophages and its relationship to protection from malaria need to be addressed in the future. This information will aid in understanding host-disease interactions for the development of an efficient vaccine against the malaria parasite.

Methods

rBCG strain expressing P. falciparum MSP-1C

An rBCG strain (referred in our previous study as rBCG016) expressing a synthetic gene encoding MSP-1C of P. falciparum was constructed previously from a series of oligonucleotides using assembly PCR.³²

Preparation of BCG and rBCG cultures

BCG and rBCG strains (Japan) were grown in 7H11 medium (Becton Dickinson) supplemented with oleic acid, albumin, dextrose, and catalase (OADC) (Becton Dickinson, USA) and kanamycin (Sigma, USA) (15 μg/mL of kanamycin for rBCG or no kanamycin for BCG) and incubated at 37 °C for 2 wk. Mycobacteria were transferred to 7H9 broth (Becton Dickinson) supplemented with OADC and kanamycin as above for 1 wk (A₆₀₀ = ~0.8). For co-culture with macrophages, bacterial suspensions were centrifuged at 1500 × g for 10 min at room temperature. Supernatant was then removed, and the pellet was resuspended in 1 mL of DMEM.

Preparation of murine macrophage cell line J774A.1 for infection

Murine macrophages (J774A.1) were obtained from the American Type Culture Collection (ATCC). Cells were cultured in complete DMEM medium (Sigma) supplemented with 100 U/mL penicillin, 100 mg/mL streptomycin (Invitrogen), and 10% fetal bovine serum (Invitrogen, USA) at 37 °C and 5% CO₂ in a humidified incubator.

Infection of J774A.1 cells with BCG and rBCG

After seeding overnight, 1 × 10⁶ cells/mL were infected with BCG or rBCG at multiplicity of infection (MOI) of 1:20 in complete DMEM culture medium. Cells were incubated at 37 °C in 5% CO₂ for 48 h in the presence or absence of 100 ng/mL LPS or combination of LPS with 10 ng/mL recombinant mice (rm) IFN-γ. Uninfected cells were used as a control.

Detection of MSP-1C by immunocytochemistry

Immunocytochemistry was performed on infected or uninfected cells using a SuperPicture^TM 3rd Gen IHC detection kit (Invitrogen) according to the manufacturer’s instruction. Briefly, infected cells cultured on chamber slides (Nunc) were rinsed in PBS and fixed in ice-cold methanol for 15 min at room temperature. After blocking nonspecific proteins with peroxidase-quenching solution for 5 min, cells were incubated in diluted MSP-1C antibody (1:50) (National Institute for Medical Research, United Kingdom) in a humidified chamber for 1 h at room temperature. After washing with PBS, cells were incubated with secondary antibody conjugated to horseradish peroxidase (HRP) for 10 min in the dark at room temperature. Bound peroxidase was detected with 3,3-diaminobenzidine tetrahydrochloride (DAB) (Sigma) and counterstained with Mayer’s hematoxylin (Sigma).

Phagocytic assay

The phagocytic activity of macrophages was determined by calculating the number of ingested mycobacteria per macrophage. Briefly, extracellular bacilli were removed and infected cells were washed twice with PBS at 800 × g for 5 min. The pellet was suspended in 1 mL of PBS, and 10 μL of cell suspension was layered and fixed on a slide for staining for acid-fast bacilli (Sigma). Slides were examined by microscopy, and the number of phagocytosed mycobacterium determined. Phagocytic index (PI) was the relative population of mycobacteria in 100 macrophages.

Cell viability assay

The ability of macrophages to proliferate in response to BCG and rBCG strains in the presence or absence of LPS or LPS + IFN-γ was assessed by a colorimetric MTT assay (Promega). Briefly, 100 μL macrophages were seeded at a density of 1 × 10⁶ cells/mL and incubated with BCG or rBCG at MOI 1:20, 10 μL of filter-sterilized stock 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) solution (5 mg/mL in PBS) was added to each well of macrophage cultures for 4 h at 37 °C. Isopropanol was then added and to each well. After 2 h of incubation at 37 °C, the formazan produced was solubilized with dimethylsulfoxide and the absorbance of the formazan solution was measured with a microplate reader at 570 nm (Bio-Rad). DMEM was used as a blank and uninfected cells were used as a control.

Intracellular bacilli survival assay

The growth of the rBCG and BCG in macrophages was determined by mycobacterial colony-forming units (CFUs) in infected macrophages. Briefly, extracellular bacilli were removed at 48 h post infection. Cells were lysed with 0.05% SDS to release bacilli. Bacilli were resuspended in 7H9 broth and plated on 7H11 medium supplemented with OADC. Plates were incubated at 37 °C for 14 d. Intracellular persistence of both rBCG and BCG was analyzed by counting viable CFU.

Cytokine assay

The production of TNF-α and IL-1β was measured in culture supernatants of uninfected and infected macrophages using ELISA Quantikine mouse TNF-α and IL-1β (R&D Systems). Briefly, 50 μL of culture supernatants was incubated in 96-well microtiter plate (Nunc) for 2 h at room temperature. The plate was washed 5 times with 300 μL of wash buffer. Following washes, 100 μL of anti-mouse TNF-α or IL-1β was added to each well, and the plates were incubated for 2 h at room temperature. Plates were washed 5 times with 300 μL of wash buffer and 100 μL of substrate solution was added. Plates were incubated for 30 min at room temperature. Reactions were stopped after 30 min with stop solution, and optical density (OD) was measured with a microplate reader (Bio-Rad) at 450 nm. Uninfected cells were used as a control. Cytokine concentrations (pg/mL) were determined using a standard curve based on the positive control provided by the manufacturer.

mRNA assay

Total RNA was extracted from infected and uninfected cells using Rneasy Plus Mini Kit (Qiagen) according to the manufacturer’s instructions. RNA samples were treated with DNAase to degrade any traces of genomic DNA before quantified using a Nanodrop spectrophotometer (Thermofisher) and was reverse transcribed to prepare cDNA using Omniscript® Reverse Transcription. Reverse transcription (RT) of RNA was performed at 37 °C for 60 min in a total volume of 20 µl containing RT buffer, 5 mM dNTP mix, 10 µM Oligo-dT primer, 10 units/µl RNase inhibitor, and Omniscript Reverse Transcriptase.

Reverse transcriptase PCR (RT-PCR) was performed in 25 μL PCR mix containing 2.5 U/µl of Taq polymerase, 200 µM dNTP mix, 1 µM of each primer, and 10× PCR buffer containing 2.5 mM MgCl₂. The amplification of TNF-α and IL-1β mRNA was performed using specific primers for mTNF-α (Forward primer: 5′-ATGAGCACAG AAAGCATGAT C-3′; Reverse primer: 5′- TACAGGCTTG TCACTCGAAT T-3′) and mIL-1β (Forward primer: 5′-GCAACTGTTC CTGAACTCAA-3′; Reverse primer: 5′-CTCGGAGCCT GTAGTGCAG-3′)⁸⁷ with the following conditions: 94 °C for 5 min followed by 35 cycles of 94 °C for 1 min, 60 °C for 1 min, and 72 °C for 30 s with a final extension at 72 °C for 7 min. PCR products were visualized using agarose gel electrophoresis. β-actin housekeeping gene primer (Forward primer: 5′-TGGAATCCTG TGGCATCCAT GAAAC-3′; Reverse primer: 5′-TAAAACGCAG CTCAGTAACA GTCCG-3′)⁸⁸ was used as an internal control. The intensity of each mRNA cytokine expression was determined using ImageJ version 1.47 (National Institutes of Health, USA).⁸⁹ Briefly, integrated density of each band was measured and then normalized with the integrated density of β-actin.

Nitric oxide assay

NO concentrations in culture supernatants of infected cells were determined using the Griess method⁹⁰ (Promega). Briefly, culture supernatants were added to 96-well microtiter plate (Nunc). The plate was incubated with sulfanilamide solution at room temperature for 7 min in the dark. N-(1-naphthyl)-ethylenediamine dihydrochloride solution was added to each well followed by incubation at room temperature for 7 min in the dark. Absorbance at 540 nm was measured by a microplate reader (Bio-Rad). NO concentrations were calculated using a standard curve as described by the manufacturer.

iNOS Detection

The production of iNOS was determined by western blot. Macrophages were lysed with RIPA buffer (0.4 M NaCl, 50mM Tris/HEPES pH 7.5, 1% NP-40, 0.1% SDS, 1mM EDTA, 1mM phenylmethylsulfonyl fluoride, and 0.05% protease inhibitor). Total lysates were denatured with Laemmli buffer (62.5 mM Tris pH 6.8, 10% glycerol, 2% SDS, 0.003% bromophenol blue and 5% 2-mercaptoethanol). Lysates were electrophoresed on 8% polyacrylamide before transfer to polyvinylidene difluoride membranes (GE Healthcare). Membranes were blocked with10% blocking solution for 1 h before incubation with rabbit anti-mouse iNOS antibody (BD PharMingen) for 1 h at 37 °C followed by incubation with goat anti-mouse antibody conjugated to HRP for 1 h at 37 °C (Dako). Chemiluminescence was generated by an ECL western blot detection reagent, as recommended by the manufacturer (GE Healthcare). An antibody specific to β-actin was used as the control (Sigma). Intensity of iNOS protein expression was determined using ImageJ version 1.47 (National Institute of Health, USA).⁸⁹ Briefly, the film image was inverted before the integrated density of each band was measured. The integrated density of iNOS was then normalized with the integrated density of β-actin.

Statistical analysis

All experiments were performed 3 times independently (n = 3). Data were presented as the arithmetic mean ± standard error of the mean (SEM). Repeated measures analysis of variance (RM ANOVA) was applied to compare within groups of infection and between groups of stimulation followed by pairwise comparison with confidence interval adjustment by Bonferroni correction. Differences were considered significant at a P < 0.05.

10.4161/hv.28695