Oral Pretreatment of Mice with CpG DNA Reduces Susceptibility to Oral or Intraperitoneal Challenge with Virulent Listeria monocytogenes

Abstract

Listeria monocytogenes is an enteroinvasive intracellular bacterial pathogen that infects humans and other animals, including mice, sometimes resulting in severe systemic infections. Previous studies showed that intraperitoneal (i.p.) pretreatment of susceptible BALB/c mice with immune-stimulatory CpG DNA 48 to 96 h prior to i.p. challenge with virulent L. monocytogenes reduces bacterial numbers in livers by greater than 100-fold, correlating with recovery from infection. Here we show that oral pretreatment of BALB/c mice with CpG DNA results in decreased susceptibility to either oral or i.p. challenge with L. monocytogenes. A single dose of 200 μg of CpG DNA administered to BALB/c mice orally by gavage 48 h or 7 days before oral challenge with virulent L. monocytogenes reduces bacterial numbers approximately 10- to 100-fold in livers and spleens. Lymphotoxin alpha knockout mice lacking Peyer's patches (PPs) and pretreated orally with CpG DNA 48 h prior to oral challenge with L. monocytogenes also have reduced susceptibility to infection, suggesting that PPs are required neither for oral infection nor for CpG-induced resistance against oral infection with L. monocytogenes. Surprisingly, 48-h oral pretreatment of BALB/c mice with 100 to 200 μg of CpG DNA results in approximately 100-fold-decreased bacterial numbers in livers following i.p. challenge with L. monocytogenes, suggesting, along with other data in this report, that orally delivered CpG DNA induces systemic resistance to infection. These results indicate that oral administration of CpG DNA induces systemic innate immune defenses against either oral or systemic infection with virulent L. monocytogenes.

Listeria monocytogenes is a gram-positive, facultative intracellular, enteroinvasive bacterial pathogen. Consumption of contaminated food by humans as well as other animals can cause severe systemic infections, sometimes resulting in bacterial meningitis, especially in immunocompromised hosts (7, 22). Therefore, the natural route of infection with L. monocytogenes is oral, although oral infection of mice is difficult to establish and requires a large dose of bacteria (28). There is also debate over whether the bacteria enter mucosally through Peyer's patch (PP) cells (21, 31) or through enterocytes or epithelial cells (17, 28, 36). The innate immune response against L. monocytogenes involves macrophages (Kupffer cells in the liver), NK cells, and neutrophils (43). Systemic immune responses to L. monocytogenes occur mainly in the spleen and liver, where the organism replicates primarily in macrophages (22).

CpG DNA or DNA containing CpG dinucleotides within specific flanking bases (CpG motifs) is immunostimulatory (25, 27). CpG dinucleotides are underrepresented and selectively methylated in vertebrate DNA but are present at the expected frequency and are unmethylated in bacterial DNA (5). The recognition of CpG motifs, which requires toll-like receptor 9 (18), is thought to be an ancestral nonself pattern sensing mechanism used by the innate immune system to detect intracellular microbial DNA (25). DNA containing CpG motifs activates murine macrophages (6, 13, 40, 41), dendritic cells (20, 39), NK cells (3, 6, 8), and B cells (44, 45, 46). CpG DNA directly activates murine macrophages to secrete interleukin-12 (IL-12) and tumor necrosis factor alpha, enhances levels of gamma interferon (IFN-γ) produced by NK cells in response to IL-12, and stimulates B cells to secrete IL-6 (3, 8, 14, 20, 23, 35, 40, 44). Overall, CpG DNA induces innate defenses with a predominantly Th1 pattern of immune activation, which is important for protection against intracellular pathogens such as L. monocytogenes (10, 11, 15, 16, 19, 22, 24, 26, 37, 42).

We previously reported that intraperitoneal (i.p.) CpG DNA pretreatment with a synthetic CpG oligodeoxynucleotide (ODN) reduced the susceptibility of BALB/c mice to i.p. administered L. monocytogenes (26). Pretreating mice (i.p.) with a single injection of 10 to 30 μg of CpG ODN 1758, 48 to 96 h prior to lethal (i.p.) challenge (10⁵ CFU, or ∼10 50% lethal doses [LD₅₀]) with the virulent 10403s strain of L. monocytogenes, provided up to a greater-than-2-log-unit (100-fold) reduction in the number of L. monocytogenes organisms, which correlates highly with recovery from L. monocytogenes (15). Reduced susceptibility to L. monocytogenes following CpG DNA (i.p.) pretreatment also correlated with sustained IL-12 expression in vivo and was dependent on IFN-γ secretion (26). As mentioned above, the natural route of infection with L. monocytogenes in humans and other animals is oral, and therefore we challenged mice orally with L. monocytogenes to examine whether orally administered CpG DNA would have any immune effect and would reduce susceptibility to L. monocytogenes. Although it has been reported that oral infection of mice with L. monocytogenes is difficult (28), we were able to infect mice to high enough levels to demonstrate that oral pretreatment of mice with CpG DNA can substantially reduce susceptibility to both oral and systemic infection with L. monocytogenes. This is the first report, to our knowledge, of oral CpG DNA administration by itself inducing resistance to an enteric pathogen.

MATERIALS AND METHODS

CpG DNA.

The immunostimulatory, nuclease-resistant phosphorothioate ODN 1826 was provided by the Coley Pharmaceutical Group (Wellesley, Mass.). ODN 1826 has the sequence TCCATGACGTTCCTGACGTT and was diluted in endotoxin-free (EF) sterile saline (0.9% sodium chloride; Abbott Laboratories, North Chicago, Ill.).

Bacteria.

L. monocytogenes strain 10403s was a kind gift from John Harty (University of Iowa). Bacteria used for inoculations were grown at log phase in tryptic soy broth (Becton Dickinson, Sparks, Md.) supplemented with 50 μg of streptomycin sulfate (Sigma, St. Louis, Mo.) per ml in a shaking incubator at 37°C. Virulence was maintained in BALB/c mice, and bacterial stocks were stored in aliquots that were maintained at −70°C.

Mice.

Female BALB/c and C57BL/6 mice were obtained from the National Cancer Institute (Frederick, Md.) at 4 to 8 weeks of age and were maintained under specific-pathogen-free conditions, according to all specific federal and institutional animal care guidelines, at the University of Iowa Animal Care Unit. Lymphotoxin alpha knockout (LTA^−/−) mice were a kind gift from Thomas Waldschmidt (University of Iowa) and were maintained under specific-pathogen-free conditions on a C57BL/6 background. Female age-matched mice were used for experiments.

CpG DNA pretreatment.

Mice were treated i.p. or orally either with ODN 1826 in EF saline (see above) or with EF saline alone. i.p. administrations were performed with 0.3-ml solutions, using 1-ml tuberculin syringes (Becton Dickinson) and 27G¹/₂“ needles (Becton Dickinson). Oral administrations were performed by gavage with 0.1-ml solutions, using 1-ml tuberculin syringes and animal feeding needles (Popper and Sons, Inc., New Hyde Park, N.Y.).

Bacterial infections.

Mice were inoculated i.p. with 10⁵ CFU of L. monocytogenes (∼10 LD₅₀ in BALB/c mice [26]) diluted in EF saline in 0.3-ml volumes or orally with 3 × 10⁹ to 6 × 10⁹ CFU in 0.1-ml volumes, using a 1-ml tuberculin syringe and either a 27G¹/₂“ needle or a feeding needle, respectively. For oral infection experiments, bacteria were cultured in 500 ml of tryptic soy broth containing streptomycin (50 μg/ml) and concentrated by centrifugation. Inoculation titers were checked by plating 10-fold serial dilutions for each experiment to confirm the number of viable bacteria inoculated.

Colony-forming assay.

At 4 to 5 days postinoculation with L. monocytogenes, the level of infection in each mouse was determined by enumerating bacteria in liver and spleen organ homogenates as described previously (26). Briefly, spleens and livers from individually inoculated mice (three to four mice per experimental group) were homogenized in sterile distilled water with 0.2% IGEPAL (Sigma), diluted serially 10-fold, and then cultured on agar plates containing tryptic soy broth with streptomycin (50 μg/ml), with incubation at 37°C overnight. The bacterial colonies were counted, and the numbers of CFU per spleen and per gram of liver used were calculated for each mouse. The means and standard deviations were then determined for each group.

Statistical Analysis.

Statistical comparisons of mean values were performed with the Instat program version 3.0a (GraphPad Software, Inc., San Diego, Calif.). The P values given in figure legends are single-tailed P values; adjustments for multiple comparisons are described in the figure legends.

RESULTS

Oral infection kinetics in mice without CpG pretreatment.

Because it has been reported that mice cannot be readily infected orally with L. monocytogenes (28), we tested a high dose of L. monocytogenes (3 × 10⁹ CFU), which was administered by gavage to increase the chances of establishing infection. Three mice were then sacrificed every day postinfection, and the CFU per spleen and gram of liver were quantitated. As shown in Fig. 1, the mean number of CFU in the spleens of the mice peaked at close to 7 log units by 5 days postinoculation and peaked at above 8 log units in the livers by 4 days postinoculation. By 8 days postinoculation, neither the spleens nor the livers had detectable bacteria. Therefore, we chose days 4 and 5 postchallenge with L. monocytogenes to examine peak bacterial numbers in spleens and livers and to allow the greatest potential differences with CpG treatment.

FIG. 1.

Oral infection kinetics in mice without CpG DNA pretreatment. Twenty-four BALB/c mice were inoculated orally by gavage with 3 × 10⁹ CFU of L. monocytogenes. Three mice per time point were then sacrificed every day postinoculation, and a CFU assay was performed for spleens and livers. Mean CFU values and standard deviations for spleens (per organ) and for livers (per gram) are shown. Results are representative of those from two separate experiments.

Reduced susceptibility to oral challenge: 48 h of CpG DNA pretreatment versus 7 days of pretreatment.

To determine whether oral inoculation of CpG DNA could reduce susceptibility to L. monocytogenes, we pretreated BALB/c mice orally by gavage with CpG ODN 1826 48 h prior to challenging mice orally by gavage with 4 × 10⁹ CFU of L. monocytogenes. We previously used CpG ODN 1758 for i.p. pretreatment of BALB/c mice with CpG DNA (26). In the present experiments, we chose CpG ODN 1826 for oral pretreatment because it stimulates spleen cells in vitro to secrete even higher levels of Th1-like cytokines, such as IFN-γ and IL-12, that correlate with protection against L. monocytogenes (10, 35, 42). We examined the effect of various doses of CpG DNA, starting with a relatively high dose (100 μg) of CpG ODN 1826 since most of the orally administered DNA is probably not systemically bioavailable and since DNA degradation is likely to occur in the intestinal tract even with use of nuclease-resistant ODN with phosphorothioate backbones (1), such as ODN 1826. The number of bacteria in spleens decreased by approximately 100-fold, and the number in the livers decreased by approximately 10-fold, in mice pretreated at the 200-μg dose (Fig. 2). However, the bacterial colony counts then began to increase in the organs from the mice pretreated at the highest dose of 400 μg, indicating a loss of CpG-induced resistance to L. monocytogenes (Fig. 2). This opposite effect induced by high-dose CpG has also been detected following i.p. administration (26). To test whether increasing the length of pretreatment with CpG DNA would enhance the reduction in the number of L. monocytogenes organisms detected in spleens and livers, a 7-day single-dose pretreatment of CpG DNA was also tested. CpG DNA pretreatment has been shown to be effective in inducing Th1-like responses and in protecting mice against L. monocytogenes and Leishmania major up to 1 to 2 weeks before challenge (26, 29). A 7-day pretreatment with CpG DNA in mice was shown to induce Th1-like systemic responses detected by IFN-γ and IL-12 cytokine production from lymph nodes (29). The dose response was examined, and as shown in Fig. 3, lower doses (50 to 200 μg) were immunostimulatory, with the optimal dose of 200 μg resulting in an approximately 10-fold reduction of bacteria in spleens. In livers, a single dose of 100 to 200 μg of CpG DNA 7 days prior to challenge resulted in an approximately 100-fold reduction of bacteria. The high dose (400 μg) again resulted in increased numbers of bacteria in livers and spleens, similar to those seen in untreated mice.

FIG. 2.

Reduced susceptibility to oral challenge: 48-h oral pretreatment dose response. Four BALB/c mice per group were treated orally by gavage with either EF saline (0 μg of CpG) or increasing doses of ODN 1826 in EF saline 48 h prior to oral challenge by gavage with 4.0 × 10⁹ CFU of L. monocytogenes. Mice were sacrificed at 4 days postchallenge, and a CFU assay was performed for spleens (A) and livers (B). Mean values and standard deviations are shown. Results are representative of those from two separate experiments.

FIG. 3.

Reduced susceptibility to oral challenge: 7-day oral pretreatment dose response. Four BALB/c mice per group were treated once orally by gavage with either EF saline (0 μg of CpG) or increasing doses of ODN 1826 in EF saline 7 days prior to oral challenge with 6 × 10⁹ CFU of L. monocytogenes. Mice were sacrificed at 4 days postchallenge, and a CFU assay was performed for spleens (A) and livers (B). Mean values and standard deviations are shown. Results are representative of those from two separate experiments.

Comparison of oral and i.p. CpG DNA pretreatment for reducing susceptibility to oral challenge.

To determine whether local mucosal CpG delivery is required or whether systemic immune activation by i.p. CpG pretreatment also could reduce susceptibility to oral infection, we orally challenged mice that had been pretreated with CpG DNA either orally or i.p. 48 h earlier and compared the protective efficacies of the two routes of administration. As shown in Fig. 4, i.p. pretreatment of mice with CpG DNA resulted in a reduction in the numbers of bacteria in spleens and livers of mice comparable to the reduction resulting from oral pretreatment of mice, suggesting that i.p. CpG-induced innate immunity alone can reduce susceptibility to oral infection with L. monocytogenes. As can be seen, an approximately 10-fold reduction in bacterial numbers in livers of mice treated orally with 100 μg of CpG DNA was found, which was a greater CpG-induced bacterial reduction than seen under similar conditions (Fig. 2). This greater reduction may be a result of decreased replication of L. monocytogenes due to the slightly decreased amount of inoculum or may be a result of variability in infection that is sometimes detected, possibly because of small differences in animal susceptibility or inoculation efficiency.

FIG. 4.

Oral pretreatment versus i.p. pretreatment: reduced susceptibility to oral challenge. Four BALB/c mice per group were treated orally by gavage with EF saline or 100 μg of CpG ODN 1826, or were treated i.p. with EF saline or 30 μg of CpG ODN 1826, 48 h prior to oral challenge by gavage with 3 × 10⁹ CFU of L. monocytogenes. Mice were sacrificed at 4 days postchallenge, and a CFU assay was performed for spleens (A) and livers (B). Mean values and standard deviations are shown.

CpG DNA pretreatment of PP-negative mice.

Because CpG DNA stimulates spleen cells in vitro and in vivo (35, 39), we hypothesized that orally delivered CpG DNA might stimulate similar secondary lymphoid tissue such as PPs. This stimulation might then lead to mucosal immunity, systemic immunity, or both, due to the fact that PP cells migrate out of mucosal areas and cytokines released from PP cells can have systemic effects (2, 12). To test this hypothesis, we examined immune responses in LTA^−/− knockout mice (on a C57BL/6 background), which do not have PPs (4, 9). Because C57BL/6 mice are known to be more resistant to L. monocytogenes than BALB/c mice, probably due to Th1-predominant immune responses in these mice (26, 30), we tested whether the Th1-inducing effects of CpG DNA would enhance the innate resistance of C57BL/6 mice to L. monocytogenes. As shown in Fig. 5, C57BL/6 mice treated with CpG DNA i.p. 48 h prior to i.p. challenge with L. monocytogenes did have reduced susceptibility to infection, with approximately 100-fold-fewer colonies in livers. There was no significant difference in spleens; however, innate immune responses against L. monocytogenes in spleens are normally lower than those in livers (7). The level of CpG-induced resistance to infection in these mice was comparable to that seen in BALB/c mice, which are more susceptible to L. monocytogenes infection (26). This indicates that CpG DNA can reduce susceptibility to infection with L. monocytogenes even in resistant animals that already have tendencies towards Th1 immune responses.

FIG. 5.

i.p. pretreatment of C57BL/6 and LTA^−/−mice reduces susceptibility to i.p. challenge. Four mice per group were treated i.p. with either EF saline or 30 μg of CpG ODN 1826 48 h prior to i.p. challenge with 10⁵ CFU of L. monocytogenes. Mice were sacrificed at 4 days postchallenge, and a CFU assay was performed for spleens (A) and livers (B). Mean values and standard deviations are shown. Results are representative of those from two separate experiments. Statistical analysis was done by a one-way analysis of variance followed by a Tukey-Kramer multiple-comparison test. *, P < 0.05; **, P < 0.01. Outliers that were in the direction of increasing the difference between groups unreasonably were adjusted by winsorization (38).

We also tested LTA^−/− mice with 48 h of i.p. CpG DNA pretreatment and i.p. L. monocytogenes challenge to determine whether these mice would respond similarly to wild-type C57BL/6 mice. LTA^−/− mice lack mesenteric lymph nodes and splenic architecture (4, 9) and therefore may not respond well to CpG DNA due to resultant immunodeficiencies. As shown in Fig. 5, the i.p. CpG pretreatment of LTA^−/− mice provided an approximate 10-fold reduction in liver bacterial counts compared to saline-treated mice, which was less than the approximately 100-fold reduction observed in CpG-treated compared to saline-treated wild-type mice, and there was a significant difference in bacterial numbers in spleens of CpG-treated and saline-treated LTA^−/− mice. The bacteria also replicated to higher levels in LTA^−/− mice, since there were approximately 10-fold-higher counts of bacteria in spleens and livers of LTA^−/− mice compared to wild-type mice, possibly due to decreased splenic development and lymph node-related immunodeficiencies in these mice. However, the approximately 10-fold decrease in the number of bacteria in the livers from i.p. CpG DNA-treated LTA^−/− mice indicates that even these somewhat immunodeficient mice can respond to CpG DNA pretreatment.

To address the question of whether PPs are required for CpG DNA-induced oral resistance to L. monocytogenes, we pretreated LTA^−/− mice orally with CpG DNA 48 h prior to oral challenge with L. monocytogenes. As can be seen in Fig. 6, oral CpG DNA pretreatment reduced bacterial counts approximately 10-fold in livers, similar to i.p. CpG DNA pretreatment of these mice, suggesting that PPs are not required for CpG DNA-induced oral protection against oral challenge. The fact that PP-negative mice could be infected with L. monocytogenes also demonstrates that PPs are not required for oral infection with L. monocytogenes.

FIG. 6.

Oral pretreatment of LTA^−/−, PP-negative mice reduces susceptibility to oral challenge. Six mice per group were treated orally by gavage with EF saline or 100 μg of CpG ODN 1826 48 h prior to oral challenge with 3 × 10⁹ CFU of L. monocytogenes. Mice were sacrificed at 4 days postchallenge, and a CFU assay was performed for spleens (A) and livers (B). Mean values and standard deviations are shown. Results are representative of those from two separate experiments. Statistical analysis was done by an unpaired, nonparametric Mann-Whitney test. *, P < 0.05.

Oral CpG DNA pretreatment reduces susceptibility to i.p. administered L. monocytogenes.

Oral pretreatment of mice with CpG DNA could reduce susceptibility to oral L. monocytogenes infection either by inducing mucosal immune activation that prevents bacterial entry or by inducing systemic immunity that prevents replication of L. monocytogenes in, or dissemination to, the livers and spleens. To determine whether oral pretreatment with CpG DNA would be effective at inducing resistance to i.p. administered L. monocytogenes, BALB/c mice were administered orally a single dose of either 100 or 200 μg of CpG DNA prior to i.p. inoculation. Oral pretreatment of mice with either dose of CpG ODN 1826, 48 or 96 h prior to i.p. challenge with a lethal dose (10⁵ CFU, or ∼10 LD₅₀) of virulent L. monocytogenes, reduced the bacterial colony counts (CFU) by at least 100-fold in livers and up to 10-fold in spleens (data not shown). As in our previously described oral experiments, we used optimal oral doses of CpG DNA (100 to 200 μg) for these oral pretreatment experiments in which we challenged with L. monocytogenes systemically (i.p.). This is compared to the 10 to 30 μg of CpG DNA i.p. pretreatment of BALB/c mice that is effective in reducing susceptibility to L. monocytogenes administered i.p. (26) and to the 30 μg of CpG ODN 1826 used for i.p. pretreatment of C57BL/6 mice in experiments described in this report. Although oral pretreatment with both doses of CpG DNA used (100 and 200 μg) reduced the number of colony counts approximately 100-fold in livers, where L. monocytogenes replicates more readily (7), a dose-response experiment was performed to determine whether lower or even higher doses of CpG DNA would be as effective or more effective against i.p. challenge. As shown in Fig. 7, a dose of 100 μg of CpG DNA administered orally 48 h before i.p. challenge again reduced colony counts approximately 100-fold in livers, with a trend of reduction from 50 to 100 μg (or 200 μg [data not shown]) of CpG DNA that began to reverse at the highest dose of 400 μg of CpG DNA, as seen previously. This result and the other data described above indicate that orally administered CpG DNA can induce systemic innate immune responses that cause resistance to systemic L. monocytogenes infection.

FIG. 7.

Oral pretreatment dose response to i.p. challenge. Four BALB/c mice per group were treated orally by gavage with either EF saline (0 μg of CpG) or increasing doses of ODN 1826 in EF saline 48 h before i.p. challenge with 10⁵ CFU of L. monocytogenes. Mice were sacrificed at 5 days postchallenge, and a CFU assay was performed for spleens (A) and livers (B). Mean values and standard deviations are shown. Results are representative of those from two separate experiments.

DISCUSSION

The oral route is the natural route of infection for the human and animal pathogen L. monocytogenes. Oral administration of CpG DNA as a vaccine adjuvant in combination with antigens has been used effectively to induce antigen-specific immune responses against pathogens, including influenza virus (32) and hepatitis B virus (32, 33, 34). In this report, we demonstrate for the first time that orally administered CpG DNA by itself without a vaccine can induce resistance to infection by an oral pathogen. Although it has been reported that oral infection of mice with L. monocytogenes results in low levels of bacterial translocation across the intestinal barrier (28), we detected 7 to 8 log units of bacteria in spleens and livers after oral infection with high doses (3 × 10⁹ to 6 × 10⁹ CFU) of L. monocytogenes. The infection results that we obtained are broadly similar to those described for a recent study in which a slightly lower dose (10⁹ CFU) of L. monocytogenes (strain P-14B) was administered to mice in their drinking water (31). The level of infection reported in that study, however, was at least 1,000-fold lower than we observed, and the kinetics were accelerated by 2 days in the spleens and livers such that the peaks of bacterial infection were 4 log units at 3 days postinoculation in spleens and 4 log units at 2 days postinoculation in livers. By 8 days postinoculation, as in our study, both the spleens and livers had no detectable colonies. The difference in kinetics and levels of infection between these studies may be due to the differences in doses and routes of oral inoculation used as well as to the different strains of L. monocytogenes used. We chose oral administration by gavage for a more direct, controlled delivery that was not dependent on how much or how often the mice drank. The optimal dose of orally delivered CpG DNA for reducing the amount of bacteria in spleens following oral challenge with L. monocytogenes was 200 μg if administered 48 h or 7 days before challenge. The optimal oral CpG dose for reducing the amount of bacteria in livers was 100 to 200 μg if administered 7 days before challenge, which induced a greater decrease in bacteria than the optimal dose of 200 μg of CpG DNA administered 48 h before challenge. This is the greatest length of protection we have detected to date for single-dose oral CpG pretreatment, and this result suggests that longer time periods of pretreatment are better at inducing immune responses in the liver, where the bacteria replicate more readily (7), possibly because oral CpG DNA-induced activation of Kupffer cells requires longer periods of time.

We have previously shown that i.p. administered CpG DNA induces resistance to i.p. challenge with L. monocytogenes (26). Here we have shown that CpG DNA administered orally reduces susceptibility to either oral or i.p. challenge with L. monocytogenes, suggesting that orally delivered CpG DNA can induce systemic innate immune defenses. The optimal dose of orally administered CpG DNA for reducing susceptibility to systemic infection with L. monocytogenes was 100 to 200 μg if given 48 h prior to challenge, resulting in a 100-fold reduction of bacteria in livers. CpG DNA given i.p. 48 h prior to challenge was as effective as CpG DNA given orally 48 h prior to oral challenge in reducing the numbers of bacteria in both spleens and livers of BALB/c mice, suggesting that the induction of systemic immunity by i.p. administered CpG DNA is sufficient for reducing the susceptibility of mice to oral infection with L. monocytogenes. It remains unclear whether the protection afforded by i.p. and oral CpG administration is mediated through the same or different mechanisms. Mucosal administration of CpG DNA either could prevent infection at the level of the gastrointestinal mucosa or could prevent dissemination from the mucosal surface to the liver and spleen.

One possible mechanism of action for the oral CpG is that it could activate innate immune cells in the PPs and thereby prevent infection and/or dissemination of the L. monocytogenes. To test whether the protective effect of CpG requires the presence of PPs, LTA^−/− mice lacking PPs were pretreated with CpG DNA orally 48 h prior to oral L. monocytogenes challenge. LTA^−/− mice that had been pretreated with oral CpG had reduced susceptibility to oral infection with L. monocytogenes, suggesting that PPs are not required to induce systemic resistance against L. monocytogenes. Furthermore, this experiment demonstrated that PPs were not required for oral infection with L. monocytogenes, since substantial numbers of bacteria (greater than 10⁶ CFU per spleen or per gram of liver) were detected in both spleens and livers of infected mice. This result suggests that L. monocytogenes oral infection in mice occurs via other mucosal cells such as epithelial cells, as reported previously (28), perhaps along with PP cell infection (21, 31). Because CpG DNA administered orally was also able to induce resistance to L. monocytogenes in LTA^−/− mice lacking PPs, it is likely that mucosal cells other than PP cells, such as epithelial cells or intraepithelial lymphocytes, perhaps in addition to PPs, are stimulated by orally administered CpG DNA and that this may lead to the induction of systemic innate immune defenses against L. monocytogenes. Alternatively, it is possible that orally administered CpG DNA passes through the mucosal barrier and enters systemically through the bloodstream to the liver and spleen, where immune responses are stimulated. However, it has been reported that absorption of ODN across the gastrointestinal tract is a slow process and that degradation of DNA may occur even with relatively stable phosphorothioate ODN (1), such as CpG 1826. Regardless of the mechanism, CpG DNA given orally can induce substantially decreased susceptibility to L. monocytogenes, which suggests that oral administration of CpG DNA may be effective against other intracellular pathogens in which Th1-related innate immune responses are important for protection. If oral CpG DNA delivery was optimized, such as by encapsulation of CpG DNA administered in an enterically coated pill form, it might lead to a safe, easy, and inexpensive method of inducing innate immune defenses against such pathogens.