Lactoferrin modulation of BCG-infected dendritic cell functions

Introduction

Tuberculosis (TB) remains one of the leading causes of human mortality and morbidity due to an infectious disease. The World Health Organization estimates that nearly one-third of the world is infected with the causative pathogen, Mycobacterium tuberculosis (MTB), an obligate intracellular bacterium (1,2). The current vaccine for TB is an attenuated strain of Mycobacterium bovis Bacillus Calmette Guerin (BCG) that is effective in preventing childhood manifestation of disease, but varies greatly in vaccine efficacy for adults (3,4). Although there are multiple efforts towards developing novel vaccine candidates for TB, the clinical application of these remain in the distant future (5–7). An alternative method involves manipulation of adjuvant components to increase efficacy of the BCG vaccine (8–13). One such proposed adjuvant component is lactoferrin.

Lactoferrin is an 80-kDa iron-binding protein found primarily in mucosal secretions and secondary granules of neutrophils (14,15). Among its many functions, lactoferrin has the ability to modulate innate and adaptive immune responses, specifically to modulate macrophage cytokine production (16–18), promote maturation of T cells (19) and B cells (20) and enhance delayed-type hypersensitivity response against defined antigens (18,21,22). Most critical to this discussion, mice immunized with BCG and lactoferrin display a reduction in MTB-induced pulmonary pathology, exhibiting limited destructive inflammation (23,24). The addition of lactoferrin to the BCG vaccine enhanced generation of BCG-specific T_h1 responses, demonstrated by increased IFN-γ production (23–25).

It is postulated that lactoferrin functions to enhance efficacy of the BCG vaccine through modulation of antigen-presenting cells (APCs). Indeed, T-cell activation and generation of the T_h1 immune response are, in part, regulated by APC cytokines released upon encounter with antigen (26–28) and subsequent interaction with presentation and co-stimulatory molecules. It was demonstrated that lactoferrin can modulate the production of IL-12 and its negative regulator IL-10 in macrophages (18,25). Treatment of BCG-infected murine macrophages with lactoferrin effectively increased the ratio of IL-12:IL-10 in a concentration-dependent manner (25,29). Moreover, lactoferrin treatment of cultured BCG-infected bone marrow-derived macrophages led to higher surface expression of MHC II and increased the ratio of CD86:CD80, changes which directly led to elevated IFN-γ production from BCG-sensitized CD3 ^{splenocytes (}30). However, very little has been accomplished to identify the effects of lactoferrin on dendritic cells (DCs), the critical cell phenotype required for initiation and polarization of adaptive responses (31) after infection or vaccination.

Differential effects that lactoferrin can exert on DCs include modulating Langerhans cell (32–35) and monocyte-derived DC migration, maturation and activation (36), suppressing pro-inflammatory cytokine production [tumor necrosis factor (TNF)-α and IL-1β] (32,35,37) and promoting a cytokine environment favoring T_h1 development through increasing IL-12p70 and decreasing IL-10 (38). The ability of DCs to stimulate antigen-specific CD4 ^{T cells depends on surface expression of the MHC II and co-stimulatory molecules CD80, CD86 and, to a lesser extent, CD40.}

Mycobacteria species, including BCG, suppression of antigen presentation functions on macrophages is well documented (39–41). Lactoferrin, an immune regulatory factor (42), has the potential to overcome these evolved processes (43). The studies presented here define the ability of lactoferrin to affect bone marrow-derived dendritic cells (BMDCs) to increase the expression of presentation and co-stimulatory molecules, leading to enhance stimulation of antigen-specific T cells. Overall, these studies explore the ability of lactoferrin to promote efficacious antigen presentation of DCs on exposure to BCG, thus enabling formulation of a mechanism for development of effective memory immunity and increase host protection against MTB challenge following vaccination with lactoferrin adjuvant.

Materials and methods

Animals

Female C57BL/6 mice (6 weeks, Jackson Laboratories, Bar Harbor, ME, USA) with 20–25 g initial body weight were used to generate primary DCs and macrophages and isolated T-cell populations. All experiments involving mice were conducted under approved guidelines of the animal ethics committee at the University of Texas, Health Science Center at Houston, protocols HSC-AWC-05-060 and HSC-AWC-06-100.

Lactoferrin and BCG

Low endotoxin bovine milk lactoferrin (<0.2 EU mg ^{by LAL assay, <20% iron saturated, >95% purity) was provided by PharmaReview Corporation (Houston, TX, USA). To examine the potential effects caused by the presence of endotoxin, lactoferrin-treated samples were compared with denatured lactoferrin (heated 1 mg ml lactoferrin in 1× PBS at >80°C for 5 min and the resulting solution filtered through 0.22-μm filter) (}38). Mycobacterium bovis Bacillus Calmette Guerin, Pasteur strain (TMC 1011, American Type Cell Culture), was grown in Dubos base (without addition of glycerol) with 10% supplement on an orbital shaker at 37°C for 2 weeks before use. BCG was diluted with 1× PBS to 3×10 ^{bacteria ml, estimated using McFarland standards (Sigma) and confirmed by plating dilutions onto 7H11 agar plates (Remel, Lenesa, KS, USA). Plates were incubated at 37°C for 3–4 weeks, and colonies were enumerated.}

Bone marrow-derived dendritic cells

Bone marrow cells were isolated from the femur and tibia of C57BL/6 mice. Collected cells were treated with ACK buffer (Cambrex Biosciences, East Rutherford, NJ, USA) to lyse the RBCs. Resulting cells were differentiated for 7 days at 1 × 10 ^{cells ml in McCoy's medium, supplemented with sodium bicarbonate (2.2 g l), 10% FBS, antibiotics (100 μg ml Penicillin G and 50 μg ml Gentamycin), mouse recombinant granulocyte macrophage colony-stimulating factor (GM-CSF) (10 ng ml) and IL-4 (10 ng ml) (Cell Sciences, Canton, MA, USA) (}44). At day 7, non-adherent cells were collected and positively selected for CD11c ^{cells using CD11c MicroBeads (Miltenyi Biotec), resulting in >90% CD11c population. BMDCs were cultured in DMEM complete medium supplemented with 10% FBS (Sigma) at 37°C with 5% CO}₂.

Intracellular BCG proliferation

BMDCs (5 × 10 ^{cells ml per well) were infected with BCG at a range MOI of 1:1 to 5:1 with or without lactoferrin (100 μg ml). For studies examining the response of BMDCs to IFN-γ activation, similarly treated BMDCs, at day 3 post-infection, were stimulated with 10 ng ml recombinant IFN-γ (Cell Sciences). At various time points post-infection, cells were lysed with 500 μl per well of 0.05% SDS, incubated at 37°C, neutralized with 500 μl per well of 1% BSA. Cell lysates were serially diluted in 1× PBS, and 100 μl was plated onto 7H11 agar plates and incubated at 37°C. Colony-forming units (CFUs) were enumerated after 2–3 weeks.}

DC cytokine and cell marker assays

BMDCs were cultured in DMEM complete medium supplemented with 10% FBS as described above. BMDCs remained non-infected or infected with BCG with and without lactoferrin. At 72 h post-infection, supernatants were collected for cytokine determination by enzyme-linked immuno-sorbent assay (ELISA) and BMDCs were isolated for analysis of surface marker expression by FACS. To examine BMDC responses to IFN-γ stimulation, BMDCs at 72 h post-infection were washed with 1× PBS and cultured continually with 10 ng ml ^{mouse recombinant IFN-γ (Cell Sciences). Supernatants and cells were then collected at 72 h post-IFN-γ stimulation for analysis of cytokine production and surface marker expression. Supernatants were assayed for cytokine production using the DuoSet ELISA kits (R&amp;D Systems, Minneapolis, MN, USA) according to manufacturer's instructions. Supernatants were assayed for the presence of T-cell cytokines, IFN-γ, IL-2 and IL-4; pro-inflammatory mediators, TNF-α and IL-6, and the T}_h1 mediator, IL-12p40 and IL-10. Lowest limits of detection for all cytokine and chemokines were between 16 and 32 pg ml^.

BMDC stimulation of BCG antigen-specific T cells

BMDCs remained non-infected or infected with BCG with and without lactoferrin. At 72 h post-infection, fixed DC presenters (4% PFA at 4°C for 15 min) or live DC presenters were co-cultured with purified CD3^{, CD4 or CD8 splenocytes. Splenocytes from naive or BCG (1 × 10 CFU per mouse, boosted at 2 weeks) -immunized mice were isolated and treated with ACK buffer to lyse RBCs, as previously described (}45). CD3 ^{T cells were purified from whole splenocyte cultures by negative selection using Pan T-cell MicroBeads (Miltenyi Biotec). The CD3 splenocytes (>90% CD3) were then positively selected for CD4 and CD8 populations using CD4 and CD8 MicroBeads (Miltenyi Biotec), resulting in 85–90% CD3CD4 and CD3CD8 cell populations. Purified T-cell populations were re-suspended at 1 × 10 cells ml in DMEM complete medium supplemented with 10% FBS, 0.005% (vol/vol) 2-mercaptoethanol (Gibco™, Invitrogen, Grand Island, NY, USA) and antibiotics (100 μg ml penicillin G and 50 μg ml gentamycin sulfate, Sigma). Supernatants were collected at 72 h from the T-cell/BMDC co-culture and stored at −20° for ELISA analysis.}

Generation and identification of BCG antigen-specific T cells

C57BL/6 mice were immunized subcutaneously at the base of the tail with BCG (1 × 10 ^{CFU per mouse) or BCG and lactoferrin (100 μg per mouse) in 1× PBS (100 μl per mouse) and boosted with the same formulation at 2 weeks. One groups served as the non-immunized control. At 6–8 weeks post-boost, mice were sacrificed and spleen cells isolated and purified for CD3CD4 cells. Purified splenocytes (1 × 10 to 2 × 10 cells ml) were overlaid onto BCG-infected bone marrow-derived macrophages (5 × 10 cells ml) (}43) and cultured for 72 h. For the last 6 h of incubation, 1 μl per well of BD GolgiPlug (BD Biosciences, San Diego, CA, USA) and ConA (2 μg ml^{) were added. Re-stimulated splenocytes were isolated and stained for CD62L–FITC or CD44–FITC on ice for 30 min. Splenocytes were then fixed with 4% PFA (Fisher Scientific, Fair Lawn, NJ, USA) at 4°C for 15 min, permeabilized with BD perm/wash and stained for intracellular expression of IFN-γ–PE.}

Enzyme-linked immuno-sorbant assay

Supernatants were assayed for cytokine production using the DuoSet ELISA kits (R&amp;D Systems, Minneapolis, MN, USA) according to manufacturer's instructions. Supernatants were assayed for production of T-cell cytokines, IFN-γ, IL-2 and IL-4; pro-inflammatory mediators, TNF-α and IL-6; the T_h1 mediator IL-12p40, IL-12p70 and IL-10, and chemokines, macrophage inflammatory protein (MIP)-1α, MIP-2 and monocyte chemotactic protein (MCP)-1. To analyse for production of transforming growth factor (TGF)-β1, collected supernatants were activated with 20% (vol/vol) 1N HCl at room temperature for 10 min and neutralized with equal volume of 1.2N NaOH/0.5M HEPES. All TGF-β1 samples were corrected for background TGF-β1 due to FBS by subtracting the media control (no cells). Lower limits of detection were between 15 and 32 pg ml^.

Flow cytometric analysis

Antibodies were diluted to working concentration of 1 μg per 10 ^{cells in staining buffer (1% BSA in 1× PBS). Unspecific staining sites were blocked with Fc Block™ (CD16/32, BD Biosciences PharMingen, San Diego, CA, USA) on ice for 5 min. DCs were stained (50 μl diluted antibody per 10 cells per marker) in the dark with anti-mouse F4/80–FITC (Cell Sciences), CD11c–PE, I-A–FITC, H-2k–FITC, CD80–PE, CD86–PE, CD40–PE, CD62L–FITC and CD44–FITC (BD Biosciences PharMingen) on ice for 30 min. Intracellular staining of splenocytes for IFN-γ–PE (BD Biosciences PharMingen) isolated from immunized mice was performed according to manufacturer's instructions as described above (}43). Stained cells were washed with staining buffer and fixed with 4% PFA. Fixed cells were washed with staining buffer and stored at 10 ^{per 500 μl in staining buffer at 4°C. Analysis was performed using Coulter FlowCentre™ (EPICS XL-MCL). Graphs were generated with WinMDI 2.8.}

Statistics

All experiments were repeated, at least, in triplicate. Data in text are reported as mean ± SD. Changes in surface expression of CD11c, I-A^{, H-2k, CD80, CD86 and CD40 were compared across repeated experiments and significance was determined by paired} t-test between groups. Mean fluorescent intensity (MFI) refers to only positive cells. Data are graphed in standard box and whisker format, showing mean, first and third quartile and range. CFUs were enumerated from triplicate plates per organ. ELISA analysis was averaged from triplicate wells. Statistical analysis were carried out using one-way analysis of variance (ANOVA), two-way ANOVA or Student's t-test (paired or unpaired) where appropriate, and differences were considered significant at P < 0.05.

Results

Direct effects of lactoferrin on naive and BCG-infected DCs

Lactoferrin, as an immune modulator, has direct effects on cytokine production from macrophage cells lines and primary macrophages. These first experiments examine the changes in naive and BCG-infected DC cytokine and chemokine production stimulated by lactoferrin. C57BL/6-derived murine BMDCs were treated with or without BCG in the presence or absence of lactoferrin for 72 h and analysed for pro-inflammatory cytokine and chemokine production by ELISA.

In naive BMDCs, addition of lactoferrin alone stimulated increased production in TNF-α, MIP-1α, MIP-2 and TGF-β1. This effect was directly attributed to lactoferrin, as culturing with denatured lactoferrin resulted in cytokine and chemokine levels that are not significantly different from the media control (Table 1). BMDCs infected with BCG increased in all pro-inflammatory mediators examined, although addition of lactoferrin significantly decreased TNF-α, IL-6, IL-12p40 and MIP-1α levels (Fig. 1). Contrast to the trend observed in pro-inflammatory cytokines and chemokines, production of TGF-β1 was increased in BCG-infected BMDCs cultured in the presence of lactoferrin (Fig. 1). These trends in cytokine and chemokine production were not observed in BCG-infected BMDCs treated with denatured lactoferrin (data not shown). No IL-12p70 or IL-10 production was observed and production of MCP-1 remained at the limit of detection of the assay (data not shown).

Table 1.

Effect of lactoferrin on BMDC cytokine and chemokine production

pg ml−1	Media control	Lactoferrin	Denatured lactoferrin
TNF-α	14 ± 1	154 ± 3*	14 ± 4
IL	10 ± 1	11 ± 1	10 ± 1
IL-12p40	72 ± 3	85 ± 2	72 ± 4
MIP-1	2 ± 1	31 ± 1*	1 ± 1
MIP-2	55 ± 1	380 ± 7*	42 ± 5
TGF-β1	260 ± 8	702 ± 119*	102 ± 1

BMDCs were cultured with or without lactoferrin (100 μg ml^{) or denatured lactoferrin (100 μl ml) for 72 h. Supernatants were collected and analysed by ELISA. For analysis of TGF-β1, samples were subtracted against media background (no cells). Limits of detection for assays range from 15 to 32 pg ml. Average values are shown (mean ± SD) in picogram per milliliter. Significance is compared between the lactoferrin-treated group against both the media control and denatured lactoferrin groups. There were no significant differences between media control and denatured lactoferrin.}

*P < 0.05.

Fig. 1.

Effect of lactoferrin on modulating BCG-infected BMDC cytokine and chemokine production. BCG-infected BMDCs (5 × 10 ^{cells ml) were cultured in the presence (black bars) or absence (white bars) of lactoferrin (100 μg ml) for 72 h. Supernatants were isolated and examined by ELISA for cytokines, TNF-α, IL-6 and IL-12p40; chemokines, MIP-1α and MIP-2, as well as for TGF-β1. For analysis of TGF-β1, samples were subtracted against media background controls (no cells). Lowest limits of detection ranged from 15 to 32pg ml. Data are represented as mean ± SD, with statistical analysis performed using the Student's} t-test. *P < 0.05, ***P < 0.001.

With significant decreases in pro-inflammatory cytokines observed from BCG-infected BMDCs cultured with lactoferrin, the next set of analysis focused on intracellular BCG proliferation. BMDCs were infected with BCG in the presence or absence of lactoferrin for up to 9 days. At various time points, cells were lysed and lysates plated for CFU. No changes were observed in intracellular BCG between the lactoferrin-treated and non-treated BMDCs for up to 9 days post-infection (Fig. 2A).

Fig. 2.

Lactoferrin does not affect intracellular proliferation of BCG. (A) BCG-infected BMDCs (5 × 10 ^{cells ml) were cultured in the presence (dotted line) or absence (solid line) of lactoferrin (100 μg ml). (B) At 72 h post-infection, BCG or BCG-/lactoferrin-treated BMDCs were further activated with IFN-γ (10 ng ml). At various time points, cells were washed and lysed with 0.05% SDS, and lysates plated onto 7H11 plates for enumeration of CFU. Data are represented as mean ± SD (error bars).}

The above analysis focused primary responses of uninfected or BCG-infected BMDCs to lactoferrin treatment. To examine the consequence of lactoferrin on downstream activation of DCs by IFN-γ, BCG-infected BMDCs were cultured in the presence or absence of lactoferrin. At 72 h post-infection, BMDCs were activated with rIFN-γ (10 ng ml^{) and examined for intracellular proliferation of BCG. As expected, growth of BCG decreased in activated DCs, and addition of lactoferrin did not alter or improve BMDC control of organism expansion (}Fig. 2B).

Exogenously activated uninfected and BCG-infected BMDCs cultured with or without lactoferrin were also examined for production of TGF-β1. Similar to the trends observed from non-IFN-γ-activated BMDCs, culturing in the presence of lactoferrin significantly (P < 0.001) increased production of TGF-β1 from naive BMDCs upon IFN-γ activation. BCG infection also stimulated a significant (P < 0.05) increase in TGF-β1 compared with the IFN-γ-activated uninfected BMDCs. While culturing BCG-infected BMDCs with lactoferrin further significantly (P < 0.05) elevated levels of TGF-β1 compared with the infected BMDCs not exposed to lactoferrin, there was no significant difference between the lactoferrin and BCG/lactoferrin groups (Table 2).

Table 2.

Lactoferrin effect on TGF-β1 production (pg ml^{) from activated BMDCs}

TGF-β1	Media control	Lactoferrin	BCG	BCG/lactoferrin
Mean	158	1076***	498	856*
Standard deviation	61	246	34	187

Uninfected and BCG-infected BMDCs were cultured with or without lactoferrin (100 μg ml^{) for 72 h and then activated with exogenous IFN-γ (10 ng ml) for further 72 h. Supernatants were collected and analysed for production of TGF-β1 (pg ml). Samples were subtracted against media background (no cells). Limit of detection for assay is 15 pg ml. Significance is compared between the media control and lactoferrin or BCG and BCG/lactoferrin groups.}

*P < 0.05, ***P < 0.001.

At 72 h post-IFN-γ activation, levels of the other pro-inflammatory cytokines and chemokines remained relatively low, near the limit of detection of the assay. The low concentrations of pro-inflammatory cytokine and chemokines observed from BCG-infected BMDCs at day 3 post-IFN-γ stimulation prompted examination of these soluble mediators at later time points, through 7 days post-IFN-γ activation. Examination of supernatants from BCG-infected BMDCs cultured through day 7 post-IFN-γ activation demonstrated low but steady increase for most cytokines examined (Fig. 3), beginning near day 3. BCG-/lactoferrin-treated BMDCs stimulated with IFN-γ produced significantly lower levels of TNF-α and IL-6 compared with the non-lactoferrin control (Fig. 3A and B). Both IL-12p40 and IL-10 concentrations remained limited, near the lower detection limit of the assay, and no IL-12p70 was observed at day 7 post-IFN-γ stimulation (data not shown). Production of MIP-1α and MIP-2 was similarly decreased with BCG-/lactoferrin-treated and IFN-γ-activated BMDCs having significantly lower concentrations compared with the non-lactoferrin control (Fig. 3C and D). Only production of MCP-1 was significantly increased from BCG-/lactoferrin-treated and IFN-γ-stimulated BMDCs compared with the non-lactoferrin group, even as early as day 1 post-IFN-γ addition, and the elevated levels remained up to day 7 post-IFN-γ (Fig. 3E).

Fig. 3.

Lactoferrin-cultured BMDCs show decreased inflammatory cytokines production upon IFN-γ stimulation. BCG-infected BMDCs (5 × 10 ^{cells ml) were cultured in the presence (dotted line) or absence (solid line) of lactoferrin (100 μg ml). At day 3 post-infection, cells were washed and stimulated with mouse recombinant IFN-γ (10 ng ml). At various time points, supernatants were collected and analysed for (A) TNF-α, (B) IL-6, (C) MIP-1α, (D) MIP-2 and (E) MCP-1, represented as mean ± SD (error bars). Statistical analysis was done using two-way ANOVA followed by Student's} t-test for comparison at each time point. Significance is compared between the (−) lactoferrin and (+) lactoferrin groups at a single time point. *P < 0.05.

Lactoferrin promotes expression of surface molecules involved in antigen presentation to T cells

BMDC expression of surface molecules involved in antigen presentation was examined after treatment with lactoferrin alone, BCG alone or BCG and lactoferrin in combination. None of the treatment criteria promoted full maturation of BMDCs as expression of MHC II, CD80 and CD86 remained low. No differences were observed among the uninfected or BCG-infected BMDCs cultured with or without lactoferrin in percent positive event or in MFI of MHC I, MHC II, CD80, CD86 or CD40 (data not shown). Only expression of CD11c was affected. Lactoferrin-cultured non-infected BMDCs (46.42 ± 17.42%) and BCG-infected BMDCs (40.44 ± 20.86%) have a significantly (P < 0.05) greater population of CD11c ^{cells when measured against BMDCs cultured in the absence of lactoferrin; non-infected (35.53 ± 16.18%) and BCG-infected (29.84 ± 19.10%). When BMDCs were examined for intensity of CD11c expression by MFI, no significant changes were observed between uninfected and BCG-infected BMDCs cultured in the presence or absence of lactoferrin (data not shown).}

Subsequent analysis concentrated on examining surface marker expression from BCG- and/or lactoferrin-pre-treated DCs further stimulated with IFN-γ (10 ng ml^{) for 72 h. Surface molecules examined included MHC I (H-2k), MHC II (I-A), CD80, CD86, CD40 and CD11c. Panel (A) in} Fig. 4 and Fig. 5 shows representative histograms from experimentation. Both uninfected and BCG-infected BMDCs cultured with lactoferrin, upon stimulation with IFN-γ, significantly (P < 0.001) increased the percentage of MHC II (I-A^{)-positive cells (53.2 ± 3.6% and 50.2 ± 4.7%, respectively) compared with similarly treated non-lactoferrin-exposed BMDCs (40.5 ± 3.9% uninfected and 38.1 ± 5.7% BCG infected). No differences were observed in MFI (}Fig. 4B). Expression of MHC I (H2-k^{) of IFN-γ-stimulated uninfected BMDCs resemble the trend observed above, as pre-treatment with lactoferrin significantly (}P < 0.001) increased the percentage of MHC I ^{cells (62.1 ± 9.1%) compared with BMDCs cultured without lactoferrin (41 ± 10.7%). Both IFN-γ-activated BCG and BCG-/lactoferrin-cultured DCs exhibited similar percentages of the BMDC population positive for MHC I expression (}Fig. 4C). When the expression of MHC I was compared by MFI, BCG- and lactoferrin-treated BMDCs (MFI 286 ± 29) are significantly (P < 0.05) increased compared with the BCG-treated BMDCs (MFI 226 ± 41). This suggests that the presence of lactoferrin during BCG infection of BMDCs supports increased expression of MHC I on a per cell basis under subsequent IFN-γ activation.

Fig. 4.

BMDCs cultured in the presence of lactoferrin increase MHC II and MHC I expression in response to IFN-γ stimulation. Surface expression of MHC II (B) and MHC I (C) were assessed on BMDCs uninfected (media) or infected with BCG (MOI 1-5:1) cultured in the presence (light gray bars) or absence (white bars) of lactoferrin (100 μg ml^{). At 72 h post-infection, cells were stimulated with recombinant IFN-γ (10 ng ml) for another 72 h. Cells were assessed for surface expression of I-A (MHC II) and H-2k (MHC I). Background fluorescence was gated to isotype-matched controls (gray histogram). Box and whisker graphs represent mean, first and third quartile and range of data sets, which were generated by WinMDI. Histograms (A) represent one typical experimental set, and percent positive events and MFI were obtained from four to six experimental repeats, each repeated data set collected 10 000–20 000 events. Statistical analysis was done by paired} t-test. *P < 0.05, ***P < 0.001.

Fig. 5.

Lactoferrin increases the MFI ratio of CD86:CD80 in BMDCs upon IFN-γ stimulation. Surface expression of (B) CD80, (C) CD86 and (D) CD40 were assessed on BMDCs uninfected (media) or infected with BCG cultured in the presence (light gray bars) or absence (white bars) of lactoferrin (100 μg ml^{). At 72 h post-infection, cells were stimulated with IFN-γ (10 ng ml) for another 72 h. Cells were assessed for surface expression of CD80, CD86 and CD40. Background fluorescence was gated to isotype-matched controls (gray). Box and whisker graphs (representing mean, first and third quartile and data range) were generated by WinMDI. Histograms (A) represent one typical experimental set, and percent positive events and MFI were obtained from four to six experimental repeats, each repeated data set collected 10 000–20 000 events. Statistical analysis was done using a paired} t-test. **P < 0.01.

Analysis of co-stimulatory molecules focused on CD80 and CD86, the necessary secondary signals for T-cell activation, and CD40, hypothesized to promote T-cell differentiation into the T_h1 subtype (46). Across all experimental groups, only a small percentage of IFN-γ-stimulated BMDCs express CD80 (1–9%), although there was a detectable but non-significant decrease in CD80 MFI from BMDCs pre-cultured with lactoferrin (Fig. 5B). In contrast, the majority of IFN-γ-stimulated BMDCs are positive for CD86 expression (40–60%), and there was a trend of increasing positive CD86 BMDCs when pre-exposed to lactoferrin (52 ± 4% uninfected and 53.6 ± 4.9% BCG infected) compared with DCs pre-cultured in the absence of lactoferrin (37.4 ± 7.3% uninfected and 44.6 ± 5.9% BCG infected). There was no difference in CD86 MFI among all groups examined (Fig. 5C). When the ratio of the CD86 MFI was compared with the CD80 MFI, IFN-γ-stimulated BMDCs pre-cultured in the presence of lactoferrin (lactoferrin only 0.25 ± 0.09%; BCG/lactoferrin 0.28 ± 0.11%) significantly (P < 0.05) increased the ratio of CD86:CD80 compared with those pre-cultured in the absence of lactoferrin (control 0.11 ± 0.05%; BCG only 0.18 ± 0.06%). There was no significant change in the percent positive events for CD40 when comparing uninfected and BCG-infected BMDCs upon stimulation with IFN-γ, and no differences when the DCs were pre-cultured in the presence or absence of lactoferrin (Fig. 5D).

BMDC presenting cells cultured with lactoferrin enhance production of IFN-γ from BCG-sensitized T cells

Bone marrow-derived macrophages treated with BCG and lactoferrin were shown to stimulate increased IFN-γ production from macrophages when cultured with BCG-sensitized CD3 ^{splenocytes (}43). The above increased surface expression of MHC II, MHC I and CD86 from BCG-infected BMDCs cultured with lactoferrin suggested that there may also be an increase in the ability of these DCs to present antigen and stimulate antigen-specific T cells.

BMDCs were infected with BCG, with or without lactoferrin, for 72 h, fixed with 2% PFA at 4°C for 15 min and subsequently co-cultured for 72 h with CD3 ^{splenocytes from BCG-sensitized mice. Production of pro-inflammatory cytokines, TNF-α and IL-6, were significantly (}P < 0.001) increased from BCG-sensitized CD3 ^{splenocytes cultured with BCG-/lactoferrin-treated BMDCs (99 ± 3 pg ml and 24 ± 1 pg ml, respectively) compared with CD3 splenocytes stimulated with BCG only-treated BMDC (2 ± 1 pg ml and 1 ± 0.5 pg ml, respectively). No production of IFN-γ, IL-2 or IL-4 was detected from the naive or BCG-sensitized CD3 splenocytes (data not shown).}

Full activation of T cells, including production of T-cell-related cytokines, requires participation by presenters, whose function may have been impaired by fixation in PFA. Live BMDCs pre-treated with or without BCG and/or lactoferrin were co-cultured with BCG-sensitized CD3 ^{splenocytes. Production of IFN-γ and IL-2 from purified CD3 cells increased when co-cultured with BMDC presenters infected with BCG and treated with lactoferrin compared with those only infected with BCG only (}Fig. 6A). Purified CD4 ^{splenocytes from BCG-sensitized mice were further co-cultured with similarly pre-treated BMDC presenters. CD4 splenocytes stimulated with BMDCs presenters pre-treated with BCG/lactoferrin exhibited a significant (}P < 0.001) increase in IFN-γ, compared with BMDC presenters previously treated with BCG alone (Fig. 6B). The response was limited to CD4 ^{cells. No production of IL-4 from CD3 or CD4 splenocyte co-cultures was observed. No detectable levels of IFN-γ or IL-2 were measured from CD8 splenocytes co-cultured with similarly treated BMDCs.}

Fig. 6.

Lactoferrin enhances the ability of BCG-infected BMDCs to stimulate pre-sensitized CD3 ^{and CD4 splenocytes. (A) CD3 and (B) CD4 splenocytes isolated from C57BL/6 mice vaccinated with BCG (1 × 10 CFU per mouse) were co-cultured with the BMDCs that were uninfected (media) or infected with BCG (MOI 1-5:1). BMDCs were cultured in the pre-treated in the presence (black bars) or absence (white bars) of lactoferrin (100 μg ml) for 72 h prior to assessment of response, and then washed prior to overlay with CD3 or CD4 cells. Subsequent production of IFN-γ and IL-2 were assessed by ELISA. The assay's lower limits of detection were ∼32 pg ml. Data are represented as mean ± SD (error bars). Statistical analysis was done by one-way ANOVA followed by Tukey post-hoc test. *}P < 0.05, ***P < 0.001.

The T_h1-inducing cytokine (IL-12), and its negative regulator (IL-10), were also measured. There were no significant differences in IL-12p40 from either CD3 ^{or CD4 BCG-sensitized splenocytes co-cultured with BMDCs pre-treated with BCG (34 ± 1 pg ml and 110 ± 4 pg ml, respectively) or with BCG and lactoferrin (33 ± 2 pg ml and 102 ± 1 pg ml, respectively). Levels of IL-10 were <20 pg ml from all samples examined.}

BCG/lactoferrin vaccination leads to an increase in generation of CD3^{CD4CD62LIFN-γ antigen-specific T cells}

The ability of lactoferrin to enhance DC stimulation of CD4 ^{splenocytes suggests that lactoferrin has the potential to promote generation of cellular immune responses during immunization. Reported vaccination with BCG/lactoferrin led to changes in immune responses during subsequent MTB infection and an increase in antigen-specific cytokine production of IFN-γ (}24), suggesting an augmentation in antigen-specific T_h1 responses. To further delineate antigen-specific responsive cells generated through immunization, CD3^{CD4 splenocytes from non-immunized, BCG or BCG-/lactoferrin-immunized mice were overlaid onto BCG-pulsed macrophages for 72 h and analysed for intracellular IFN-γ production and surface expression of CD62L.}

Mice immunized with BCG/lactoferrin significantly increased the percent of IFN-γ producing, BCG antigen-specific CD3^{CD4CD62 splenocytes compared with both the BCG and non-immunized groups (}Fig. 7). In contrast, there was no significant elevation in IFN-γ production observed in CD62L ^population.

Fig. 7.

Mice vaccinated with BCG/lactoferrin increase BCG antigen-specific IFN-γ-producing CD3^{CD4CD62L cells. C57BL/6 mice were immunized and boosted at 2 weeks with BCG (MOI 1 × 10 CFU per mouse) or BCG/lactoferrin (100 μg per mouse), or remained non-immunized. At 6–8 weeks post-vaccination, splenocytes were isolated and purified for CD3CD4 cells and co-cultured with BCG-pulsed bone marrow-derived macrophages. At day 3, splenocytes were re-stimulated with ConA (2 μg ml) for 6 h in the presence of BD GolgiPlug (1 μl ml). Splenocytes were collected, blocked with FcBlock and stained for CD62L–FITC or CD44–FITC on ice for 30 min. The stained cells were fixed with 4% PFA, permeablized and stained for IFN-γ–PE on ice for 2 h. Percent of IFN-γ positive events were compared against splenocytes co-cultured with uninfected macrophages. Data represent three separate experimental sets, and each experiment collected 20 000 data events. Data are represented as mean ± SD (error bars). Statistical analysis was done using two-way ANOVA followed by Bonferroni post-test. *}P < 0.05, **P < 0.01.

Direct effects of lactoferrin on naive and BCG-infected DCs

Lactoferrin, as an immune modulator, has direct effects on cytokine production from macrophage cells lines and primary macrophages. These first experiments examine the changes in naive and BCG-infected DC cytokine and chemokine production stimulated by lactoferrin. C57BL/6-derived murine BMDCs were treated with or without BCG in the presence or absence of lactoferrin for 72 h and analysed for pro-inflammatory cytokine and chemokine production by ELISA.

In naive BMDCs, addition of lactoferrin alone stimulated increased production in TNF-α, MIP-1α, MIP-2 and TGF-β1. This effect was directly attributed to lactoferrin, as culturing with denatured lactoferrin resulted in cytokine and chemokine levels that are not significantly different from the media control (Table 1). BMDCs infected with BCG increased in all pro-inflammatory mediators examined, although addition of lactoferrin significantly decreased TNF-α, IL-6, IL-12p40 and MIP-1α levels (Fig. 1). Contrast to the trend observed in pro-inflammatory cytokines and chemokines, production of TGF-β1 was increased in BCG-infected BMDCs cultured in the presence of lactoferrin (Fig. 1). These trends in cytokine and chemokine production were not observed in BCG-infected BMDCs treated with denatured lactoferrin (data not shown). No IL-12p70 or IL-10 production was observed and production of MCP-1 remained at the limit of detection of the assay (data not shown).

Fig. 1.

Effect of lactoferrin on modulating BCG-infected BMDC cytokine and chemokine production. BCG-infected BMDCs (5 × 10 ^{cells ml) were cultured in the presence (black bars) or absence (white bars) of lactoferrin (100 μg ml) for 72 h. Supernatants were isolated and examined by ELISA for cytokines, TNF-α, IL-6 and IL-12p40; chemokines, MIP-1α and MIP-2, as well as for TGF-β1. For analysis of TGF-β1, samples were subtracted against media background controls (no cells). Lowest limits of detection ranged from 15 to 32pg ml. Data are represented as mean ± SD, with statistical analysis performed using the Student's} t-test. *P < 0.05, ***P < 0.001.

With significant decreases in pro-inflammatory cytokines observed from BCG-infected BMDCs cultured with lactoferrin, the next set of analysis focused on intracellular BCG proliferation. BMDCs were infected with BCG in the presence or absence of lactoferrin for up to 9 days. At various time points, cells were lysed and lysates plated for CFU. No changes were observed in intracellular BCG between the lactoferrin-treated and non-treated BMDCs for up to 9 days post-infection (Fig. 2A).

Fig. 2.

Lactoferrin does not affect intracellular proliferation of BCG. (A) BCG-infected BMDCs (5 × 10 ^{cells ml) were cultured in the presence (dotted line) or absence (solid line) of lactoferrin (100 μg ml). (B) At 72 h post-infection, BCG or BCG-/lactoferrin-treated BMDCs were further activated with IFN-γ (10 ng ml). At various time points, cells were washed and lysed with 0.05% SDS, and lysates plated onto 7H11 plates for enumeration of CFU. Data are represented as mean ± SD (error bars).}

The above analysis focused primary responses of uninfected or BCG-infected BMDCs to lactoferrin treatment. To examine the consequence of lactoferrin on downstream activation of DCs by IFN-γ, BCG-infected BMDCs were cultured in the presence or absence of lactoferrin. At 72 h post-infection, BMDCs were activated with rIFN-γ (10 ng ml^{) and examined for intracellular proliferation of BCG. As expected, growth of BCG decreased in activated DCs, and addition of lactoferrin did not alter or improve BMDC control of organism expansion (}Fig. 2B).

Exogenously activated uninfected and BCG-infected BMDCs cultured with or without lactoferrin were also examined for production of TGF-β1. Similar to the trends observed from non-IFN-γ-activated BMDCs, culturing in the presence of lactoferrin significantly (P < 0.001) increased production of TGF-β1 from naive BMDCs upon IFN-γ activation. BCG infection also stimulated a significant (P < 0.05) increase in TGF-β1 compared with the IFN-γ-activated uninfected BMDCs. While culturing BCG-infected BMDCs with lactoferrin further significantly (P < 0.05) elevated levels of TGF-β1 compared with the infected BMDCs not exposed to lactoferrin, there was no significant difference between the lactoferrin and BCG/lactoferrin groups (Table 2).

At 72 h post-IFN-γ activation, levels of the other pro-inflammatory cytokines and chemokines remained relatively low, near the limit of detection of the assay. The low concentrations of pro-inflammatory cytokine and chemokines observed from BCG-infected BMDCs at day 3 post-IFN-γ stimulation prompted examination of these soluble mediators at later time points, through 7 days post-IFN-γ activation. Examination of supernatants from BCG-infected BMDCs cultured through day 7 post-IFN-γ activation demonstrated low but steady increase for most cytokines examined (Fig. 3), beginning near day 3. BCG-/lactoferrin-treated BMDCs stimulated with IFN-γ produced significantly lower levels of TNF-α and IL-6 compared with the non-lactoferrin control (Fig. 3A and B). Both IL-12p40 and IL-10 concentrations remained limited, near the lower detection limit of the assay, and no IL-12p70 was observed at day 7 post-IFN-γ stimulation (data not shown). Production of MIP-1α and MIP-2 was similarly decreased with BCG-/lactoferrin-treated and IFN-γ-activated BMDCs having significantly lower concentrations compared with the non-lactoferrin control (Fig. 3C and D). Only production of MCP-1 was significantly increased from BCG-/lactoferrin-treated and IFN-γ-stimulated BMDCs compared with the non-lactoferrin group, even as early as day 1 post-IFN-γ addition, and the elevated levels remained up to day 7 post-IFN-γ (Fig. 3E).

Fig. 3.

Lactoferrin-cultured BMDCs show decreased inflammatory cytokines production upon IFN-γ stimulation. BCG-infected BMDCs (5 × 10 ^{cells ml) were cultured in the presence (dotted line) or absence (solid line) of lactoferrin (100 μg ml). At day 3 post-infection, cells were washed and stimulated with mouse recombinant IFN-γ (10 ng ml). At various time points, supernatants were collected and analysed for (A) TNF-α, (B) IL-6, (C) MIP-1α, (D) MIP-2 and (E) MCP-1, represented as mean ± SD (error bars). Statistical analysis was done using two-way ANOVA followed by Student's} t-test for comparison at each time point. Significance is compared between the (−) lactoferrin and (+) lactoferrin groups at a single time point. *P < 0.05.

Lactoferrin promotes expression of surface molecules involved in antigen presentation to T cells

BMDC expression of surface molecules involved in antigen presentation was examined after treatment with lactoferrin alone, BCG alone or BCG and lactoferrin in combination. None of the treatment criteria promoted full maturation of BMDCs as expression of MHC II, CD80 and CD86 remained low. No differences were observed among the uninfected or BCG-infected BMDCs cultured with or without lactoferrin in percent positive event or in MFI of MHC I, MHC II, CD80, CD86 or CD40 (data not shown). Only expression of CD11c was affected. Lactoferrin-cultured non-infected BMDCs (46.42 ± 17.42%) and BCG-infected BMDCs (40.44 ± 20.86%) have a significantly (P < 0.05) greater population of CD11c ^{cells when measured against BMDCs cultured in the absence of lactoferrin; non-infected (35.53 ± 16.18%) and BCG-infected (29.84 ± 19.10%). When BMDCs were examined for intensity of CD11c expression by MFI, no significant changes were observed between uninfected and BCG-infected BMDCs cultured in the presence or absence of lactoferrin (data not shown).}

Subsequent analysis concentrated on examining surface marker expression from BCG- and/or lactoferrin-pre-treated DCs further stimulated with IFN-γ (10 ng ml^{) for 72 h. Surface molecules examined included MHC I (H-2k), MHC II (I-A), CD80, CD86, CD40 and CD11c. Panel (A) in} Fig. 4 and Fig. 5 shows representative histograms from experimentation. Both uninfected and BCG-infected BMDCs cultured with lactoferrin, upon stimulation with IFN-γ, significantly (P < 0.001) increased the percentage of MHC II (I-A^{)-positive cells (53.2 ± 3.6% and 50.2 ± 4.7%, respectively) compared with similarly treated non-lactoferrin-exposed BMDCs (40.5 ± 3.9% uninfected and 38.1 ± 5.7% BCG infected). No differences were observed in MFI (}Fig. 4B). Expression of MHC I (H2-k^{) of IFN-γ-stimulated uninfected BMDCs resemble the trend observed above, as pre-treatment with lactoferrin significantly (}P < 0.001) increased the percentage of MHC I ^{cells (62.1 ± 9.1%) compared with BMDCs cultured without lactoferrin (41 ± 10.7%). Both IFN-γ-activated BCG and BCG-/lactoferrin-cultured DCs exhibited similar percentages of the BMDC population positive for MHC I expression (}Fig. 4C). When the expression of MHC I was compared by MFI, BCG- and lactoferrin-treated BMDCs (MFI 286 ± 29) are significantly (P < 0.05) increased compared with the BCG-treated BMDCs (MFI 226 ± 41). This suggests that the presence of lactoferrin during BCG infection of BMDCs supports increased expression of MHC I on a per cell basis under subsequent IFN-γ activation.

Fig. 4.

BMDCs cultured in the presence of lactoferrin increase MHC II and MHC I expression in response to IFN-γ stimulation. Surface expression of MHC II (B) and MHC I (C) were assessed on BMDCs uninfected (media) or infected with BCG (MOI 1-5:1) cultured in the presence (light gray bars) or absence (white bars) of lactoferrin (100 μg ml^{). At 72 h post-infection, cells were stimulated with recombinant IFN-γ (10 ng ml) for another 72 h. Cells were assessed for surface expression of I-A (MHC II) and H-2k (MHC I). Background fluorescence was gated to isotype-matched controls (gray histogram). Box and whisker graphs represent mean, first and third quartile and range of data sets, which were generated by WinMDI. Histograms (A) represent one typical experimental set, and percent positive events and MFI were obtained from four to six experimental repeats, each repeated data set collected 10 000–20 000 events. Statistical analysis was done by paired} t-test. *P < 0.05, ***P < 0.001.

Fig. 5.

Lactoferrin increases the MFI ratio of CD86:CD80 in BMDCs upon IFN-γ stimulation. Surface expression of (B) CD80, (C) CD86 and (D) CD40 were assessed on BMDCs uninfected (media) or infected with BCG cultured in the presence (light gray bars) or absence (white bars) of lactoferrin (100 μg ml^{). At 72 h post-infection, cells were stimulated with IFN-γ (10 ng ml) for another 72 h. Cells were assessed for surface expression of CD80, CD86 and CD40. Background fluorescence was gated to isotype-matched controls (gray). Box and whisker graphs (representing mean, first and third quartile and data range) were generated by WinMDI. Histograms (A) represent one typical experimental set, and percent positive events and MFI were obtained from four to six experimental repeats, each repeated data set collected 10 000–20 000 events. Statistical analysis was done using a paired} t-test. **P < 0.01.

Analysis of co-stimulatory molecules focused on CD80 and CD86, the necessary secondary signals for T-cell activation, and CD40, hypothesized to promote T-cell differentiation into the T_h1 subtype (46). Across all experimental groups, only a small percentage of IFN-γ-stimulated BMDCs express CD80 (1–9%), although there was a detectable but non-significant decrease in CD80 MFI from BMDCs pre-cultured with lactoferrin (Fig. 5B). In contrast, the majority of IFN-γ-stimulated BMDCs are positive for CD86 expression (40–60%), and there was a trend of increasing positive CD86 BMDCs when pre-exposed to lactoferrin (52 ± 4% uninfected and 53.6 ± 4.9% BCG infected) compared with DCs pre-cultured in the absence of lactoferrin (37.4 ± 7.3% uninfected and 44.6 ± 5.9% BCG infected). There was no difference in CD86 MFI among all groups examined (Fig. 5C). When the ratio of the CD86 MFI was compared with the CD80 MFI, IFN-γ-stimulated BMDCs pre-cultured in the presence of lactoferrin (lactoferrin only 0.25 ± 0.09%; BCG/lactoferrin 0.28 ± 0.11%) significantly (P < 0.05) increased the ratio of CD86:CD80 compared with those pre-cultured in the absence of lactoferrin (control 0.11 ± 0.05%; BCG only 0.18 ± 0.06%). There was no significant change in the percent positive events for CD40 when comparing uninfected and BCG-infected BMDCs upon stimulation with IFN-γ, and no differences when the DCs were pre-cultured in the presence or absence of lactoferrin (Fig. 5D).

BMDC presenting cells cultured with lactoferrin enhance production of IFN-γ from BCG-sensitized T cells

Bone marrow-derived macrophages treated with BCG and lactoferrin were shown to stimulate increased IFN-γ production from macrophages when cultured with BCG-sensitized CD3 ^{splenocytes (}43). The above increased surface expression of MHC II, MHC I and CD86 from BCG-infected BMDCs cultured with lactoferrin suggested that there may also be an increase in the ability of these DCs to present antigen and stimulate antigen-specific T cells.

BMDCs were infected with BCG, with or without lactoferrin, for 72 h, fixed with 2% PFA at 4°C for 15 min and subsequently co-cultured for 72 h with CD3 ^{splenocytes from BCG-sensitized mice. Production of pro-inflammatory cytokines, TNF-α and IL-6, were significantly (}P < 0.001) increased from BCG-sensitized CD3 ^{splenocytes cultured with BCG-/lactoferrin-treated BMDCs (99 ± 3 pg ml and 24 ± 1 pg ml, respectively) compared with CD3 splenocytes stimulated with BCG only-treated BMDC (2 ± 1 pg ml and 1 ± 0.5 pg ml, respectively). No production of IFN-γ, IL-2 or IL-4 was detected from the naive or BCG-sensitized CD3 splenocytes (data not shown).}

Full activation of T cells, including production of T-cell-related cytokines, requires participation by presenters, whose function may have been impaired by fixation in PFA. Live BMDCs pre-treated with or without BCG and/or lactoferrin were co-cultured with BCG-sensitized CD3 ^{splenocytes. Production of IFN-γ and IL-2 from purified CD3 cells increased when co-cultured with BMDC presenters infected with BCG and treated with lactoferrin compared with those only infected with BCG only (}Fig. 6A). Purified CD4 ^{splenocytes from BCG-sensitized mice were further co-cultured with similarly pre-treated BMDC presenters. CD4 splenocytes stimulated with BMDCs presenters pre-treated with BCG/lactoferrin exhibited a significant (}P < 0.001) increase in IFN-γ, compared with BMDC presenters previously treated with BCG alone (Fig. 6B). The response was limited to CD4 ^{cells. No production of IL-4 from CD3 or CD4 splenocyte co-cultures was observed. No detectable levels of IFN-γ or IL-2 were measured from CD8 splenocytes co-cultured with similarly treated BMDCs.}

Fig. 6.

Lactoferrin enhances the ability of BCG-infected BMDCs to stimulate pre-sensitized CD3 ^{and CD4 splenocytes. (A) CD3 and (B) CD4 splenocytes isolated from C57BL/6 mice vaccinated with BCG (1 × 10 CFU per mouse) were co-cultured with the BMDCs that were uninfected (media) or infected with BCG (MOI 1-5:1). BMDCs were cultured in the pre-treated in the presence (black bars) or absence (white bars) of lactoferrin (100 μg ml) for 72 h prior to assessment of response, and then washed prior to overlay with CD3 or CD4 cells. Subsequent production of IFN-γ and IL-2 were assessed by ELISA. The assay's lower limits of detection were ∼32 pg ml. Data are represented as mean ± SD (error bars). Statistical analysis was done by one-way ANOVA followed by Tukey post-hoc test. *}P < 0.05, ***P < 0.001.

The T_h1-inducing cytokine (IL-12), and its negative regulator (IL-10), were also measured. There were no significant differences in IL-12p40 from either CD3 ^{or CD4 BCG-sensitized splenocytes co-cultured with BMDCs pre-treated with BCG (34 ± 1 pg ml and 110 ± 4 pg ml, respectively) or with BCG and lactoferrin (33 ± 2 pg ml and 102 ± 1 pg ml, respectively). Levels of IL-10 were <20 pg ml from all samples examined.}

BCG/lactoferrin vaccination leads to an increase in generation of CD3^{CD4CD62LIFN-γ antigen-specific T cells}

The ability of lactoferrin to enhance DC stimulation of CD4 ^{splenocytes suggests that lactoferrin has the potential to promote generation of cellular immune responses during immunization. Reported vaccination with BCG/lactoferrin led to changes in immune responses during subsequent MTB infection and an increase in antigen-specific cytokine production of IFN-γ (}24), suggesting an augmentation in antigen-specific T_h1 responses. To further delineate antigen-specific responsive cells generated through immunization, CD3^{CD4 splenocytes from non-immunized, BCG or BCG-/lactoferrin-immunized mice were overlaid onto BCG-pulsed macrophages for 72 h and analysed for intracellular IFN-γ production and surface expression of CD62L.}

Mice immunized with BCG/lactoferrin significantly increased the percent of IFN-γ producing, BCG antigen-specific CD3^{CD4CD62 splenocytes compared with both the BCG and non-immunized groups (}Fig. 7). In contrast, there was no significant elevation in IFN-γ production observed in CD62L ^population.

Fig. 7.

Mice vaccinated with BCG/lactoferrin increase BCG antigen-specific IFN-γ-producing CD3^{CD4CD62L cells. C57BL/6 mice were immunized and boosted at 2 weeks with BCG (MOI 1 × 10 CFU per mouse) or BCG/lactoferrin (100 μg per mouse), or remained non-immunized. At 6–8 weeks post-vaccination, splenocytes were isolated and purified for CD3CD4 cells and co-cultured with BCG-pulsed bone marrow-derived macrophages. At day 3, splenocytes were re-stimulated with ConA (2 μg ml) for 6 h in the presence of BD GolgiPlug (1 μl ml). Splenocytes were collected, blocked with FcBlock and stained for CD62L–FITC or CD44–FITC on ice for 30 min. The stained cells were fixed with 4% PFA, permeablized and stained for IFN-γ–PE on ice for 2 h. Percent of IFN-γ positive events were compared against splenocytes co-cultured with uninfected macrophages. Data represent three separate experimental sets, and each experiment collected 20 000 data events. Data are represented as mean ± SD (error bars). Statistical analysis was done using two-way ANOVA followed by Bonferroni post-test. *}P < 0.05, **P < 0.01.

Discussion

Lactoferrin demonstrated defined effects on murine BMDCs, culminating in promotion of surface molecules that potentially direct development and stimulation of antigen-specific T cells. The increase in surface expression of presentation and co-stimulatory molecules was associated with an increase in IFN-γ production from BCG-sensitized CD4 ^{T cells. Vaccination with BCG and lactoferrin promoted generation of a larger pool of BCG-responsive CD3CD4CD62L IFN-γ-producing cells. All this suggests a mechanism by which lactoferrin can serve as an adjuvant to promote efficacy of the BCG vaccine, able to promote host protections against subsequent challenge with virulent organisms, as previously demonstrated (}24). These studies illustrate that lactoferrin possesses a previously unreported immune modulatory activity on DCs that may explain promotion of antigen-specific cell-mediated immunity when utilized as a vaccine adjuvant.

Lactoferrin is an iron-binding protein. The molecule used in these studies was only partially saturated (25%), which poses the potential to affect BMDCs by sequestration of free iron. Schaible et al. (47) showed that mice deficient in β2-microglobulin are highly susceptible to MTB infection when excess iron is present in the host, and that addition of lactoferrin significantly decreased bacterial burden through reduction of host environmental iron. In the studies described here, BCG intracellular proliferation was not affected by culturing infected BMDCs with lactoferrin, even upon IFN-γ stimulation, suggesting that potential for lactoferrin to sequestering excess iron in the cell culture environment did not affect BCG survival. This may be a specific response by DCs, as Cumberbatch, et al. showed that both iron-depleted and iron-saturated lactoferrin similarly inhibited DC migration (32), suggesting that in context of DC function, lactoferrin iron binding may play a limited role.

DCs are the main APC responsible for initial directing of adaptive immune responses, in part through activation of CD4 ^{T cells (}31,48). Bone marrow cells differentiated with GM-CSF and IL-4 generate a population of CD11c ^{cells that lose expression of CD11c when cultured} in vitro without exogenous stimulation. In the presence of lactoferrin, both naive and BCG-infected BMDCs express higher surface CD11c and maintain morphological appearances that are characteristic of DCs. However, there were no concurrent changes in MHC II, CD80, CD86 or CD40 expression, suggesting that BCG or lactoferrin by themselves does not generally push for maturation of murine BMDCs. Experiments designed to examine changes in matured DCs demonstrated lactoferrin's ability to increased expression of MHC I, MHC II and CD86:CD80 ratio. This data indicate that while lactoferrin does not directly push for full maturation of BCG-infected DCs, exposure of DCs to lactoferrin led to increase of ‘maturation markers’ upon activation with IFN-γ.

The overall hypothesis for increased expression of presentation and co-stimulatory molecules is that these changes will lead to an enhancement in activation of antigen-specific T-cell populations. Whereas the increase in MHC I and MHC II directly enhances the availability of antigen for T-cell stimulation, the role of co-stimulatory molecules, CD80 and CD86, regulates T-cell activation during antigen presentation. While it has been hypothesized that the difference in CD80 and CD86 is related to generation of CD4 T_h subtypes, published data has failed to link CD80 or CD86 to favor either a T_h1 or T_h2 immune response (49,50). Recently, biochemical studies show that CD86 may preferentially recognize CD28 to promote T-cell activation, whereas CD80 preferentially binds to CTLA-4 and directs T-cell anergy (51–53). Additionally, in vivo graft versus host disease model studies further strengthens this functional dichotomy, with blocking of CD86 inhibiting activation of CD4 ^{T cell generating either a T}_h1- or T_h2-driven response, dominated by production of IFN-γ or IL-4, respectively (54). Thus, lactoferrin may increase the ability of BCG-infected BMDCs to stimulate antigen-specific T cells through enhancement of the ratio of CD86:CD80.

The changes lactoferrin induced in DC surface presentation and co-stimulatory molecules were associated with increased IFN-γ production from BCG pre-sensitized CD3 ^{and CD4 splenocytes. Although lactoferrin-treated human DCs were shown to stimulate IFN-γ production in an allogeneic T-cell response (}38), this is the first antigen-specific example that lactoferrin cultured-infected BMDCs are capable of enhancing CD4 ^{T-cell responses culminating in increased IFN-γ production. This increase in IFN-γ production from BCG-sensitized CD3CD4 in the absence of increased IL-12 production from BMDCs suggests that lactoferrin can enhance efficient antigen presentation. This effect of lactoferrin to promote antigen presentation events from APCs was also observed in BCG-/lactoferrin-treated bone marrow-derived macrophages (}30). Additionally, it should be noted that production of IFN-γ by T cells is antigen specific, and that while lactoferrin also increased surface expression of presentation and co-stimulatory molecules, it did not lead to general T-cell stimulation.

The ability of lactoferrin to increase stimulation of antigen-specific T cells suggests that it can function as an adjuvant to enhance development of cell-mediated immunity during vaccination. Previous in vivo studies demonstrated that lactoferrin admixed to the BCG vaccine protected C57BL/6 mice upon challenge with virulent M. tuberculosis, as observed by early decreases in lung and splenic bacterial loads and reduced pulmonary-related pathology. In addition, splenic recall analysis using BCG antigens indicated an enhancement in antigen-specific T_h1 cytokine responses, as observed by increased production of IFN-γ and IL-2 (24). Here, a further delineation of BCG antigen-responsive T-cells was investigated. Indeed, BCG-/lactoferrin-vaccinated mice generated a higher pool of IFN-γ-producing CD3^{CD4CD62L T cells, suggesting an enhanced pool of effector T cells (}55).

The effect of lactoferrin to promote cytokines that generate host protective immunity and augment BCG-specific T-cell activation was previously demonstrated in macrophages. In both murine and human macrophage cell lines, addition of lactoferrin during BCG infection increased the ratio of IL-12:IL-10 production, generating a cytokine environment that was favorable for development of T_h1 (25,29). In this study, uninfected DCs cultured with lactoferrin showed increases in production of TNF-α, MIP-1α, MIP-2 and TGF-β1. However, in BCG-infected and IFN-γ-activated DCs, the presence of lactoferrin decreased the secretion of TNF-α, IL-6, IL-12p40 and MIP-1α. Lactoferrin may therefore have an anti-inflammatory affect on BCG-infected DCs. Similar anti-inflammatory effects of lactoferrin on DC cytokine production have been reported (36,38). It is interesting to note that exposure of BCG-infected BMDCs to IFN-γ stimulation limits early production of TNF-α and IL-6, which typically peaks around 24 h (18,25), suggesting that the observed lack of increase in IL-12 may be time dependent. In the studies described here, IFN-γ-activated BCG-infected BMDCs required a longer stimulation time for production of IL-12. The discrepancy between the effects of lactoferrin on macrophages and DCs is most likely due to inherent innate cellular immune responses. Additionally, pro-inflammatory cytokine production by DCs may be more tightly controlled, and the cells can be exhausted quickly (<20 h) rendering only low levels of cytokine secretion (31,48,56).

As mentioned above, lactoferrin similarly decreases the production of chemokines from uninfected and BCG-infected DCs, with the exception of MCP-1. Secretion of MCP-1 was not observed from non-IFN-γ-activated DCs; neither lactoferrin nor BCG or in combination stimulated MCP-1 production. Upon IFN-γ activation, lactoferrin-pre-treated BCG-infected DCs up-regulated the production of MCP-1, which acts to attract monocytes and T cells (57). Coupled with the decrease in MIP-1α and MIP-2, which predominately attracts neutrophils (58), these data may indicate that lactoferrin enables BCG-infected DCs, upon T-cell stimulation, to recruit additional monocytes and T cells to the site of antigen deposition. In the context of vaccination, this may be advantageous to enhance generation of antigen-specific T cells.

With the decrease in pro-inflammatory cytokine production from BCG-infected DCs in the presence of lactoferrin, there is an indication that lactoferrin is exerting its anti-inflammatory effects. Uninfected or BCG-infected DCs cultured with lactoferrin, with or without IFN-γ activation, increased the production of TGF-β1, a cytokine with extensive regulatory functions. Production of TGF-β1 has been normally associated with suppression of DC maturation and activation (59). This suggests that production of TGF-β1 may be responsible for the low expression of surface presentation and co-stimulatory molecules of BMDCs cultured with lactoferrin. Interestingly, TGF-β1 is also noted to be a regulator of T-cell differentiation and a suppressor of IFN-γ production (59). However, BCG-/lactoferrin-treated DCs enhanced production of antigen-specific IFN-γ. The lack of IL-12 and increased TGF-β1 suggests that the secretion of cytokines from BCG-/lactoferrin-treated BMDCs may play a minor, as yet undefined, role in influencing antigen-specific IFN-γ production from T –cells, rather surface expression of presentation and co-stimulatory molecules more readily control this response.

It is well documented that mycobacteria can influence host presentation of antigens in multiple ways, including down-regulation of surface MHC (60,61), which decreases availability of bacterial antigens. Indeed, it could be argued that the adjuvant effects of lactoferrin in this model system may simply reflect control of intracellular BCG proliferation, and therefore may not be transferrable to other adjuvant systems. Lactoferrin did not directly affect surface expression of presenting molecules, rather there was a requirement for low level stimulation of the BMDCs for lactoferrin to exert potentiating responses. However, lactoferrin does not possess any direct mycobacteriocidal or mycobacteria-static activity; there were no differences in intracellular BCG CFU in BMDC cultured with or without lactoferrin. Recently, there is evidence that antigen presentation is not affected by the inability of host cells to process the entire mycobacterium organism (62). So while the increase in activation of antigen-specific T cells by BCG- and lactoferrin-treated BMDCs is not related to intracellular survival, it appears that some activation is required for lactoferrin to function to enhance presentation of antigens. This is a natural response during all vaccination protocols, and, therefore, should be similarly effective when lactoferrin is translated for use with other antigens.

This study illustrate the effect of lactoferrin on BCG-infected DCs to promote CD3^{CD4 T-cell activation, and that this activation is associated with increased surface expression of molecules known to be critical in antigen presentation. A previously published report on lactoferrin to affect BCG-infected bone marrow-derived macrophages showed similar results. In that study, there was increased surface expression of antigen presentation molecules (MHC II, CD86), promotion of higher levels of IFN-γ production from CD3 splenocytes but no affect on intracellular proliferation of BCG (}30). The novel aspects of this study show that lactoferrin can directly influence DC function, leading to increased antigen presentation and T-cell stimulation that can also promote development of IFN-γ effector T cells in vivo. Lactoferrin possesses adjuvant characteristics that enhance efficacy of the BCG vaccine, and should be considered a unique adjuvant candidate to augment vaccine efficacy. The mechanisms of lactoferrin action may involve modulation of DCs, allowing them to become more efficient presenters to assist in development of strong memory T-cell responses culminating in increased host protection upon subsequent challenge with virulent mycobacterium.

Funding

National Institutes of Health grants 1R41GM079810-01 and R42-AI051050-03.