Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Microbiol Immunol. Author manuscript; available in PMC May 1, 2012.
Published in final edited form as:
PMCID: PMC3082616

Attenuated Listeria monocytogenes vaccine vectors expressing Influenza A nucleoprotein: preclinical evaluation and oral inoculation of volunteers


Listeria monocytogenes vectors have shown promise for delivery of viral and tumor antigens in animals. We used 2 mutant vector strains deleted for actA/plcB (BMB72) and actA/inlB (BMB54) and engineered both strains to secrete a heterologous nucleoprotein antigen from the Influenza A virus. Strains were evaluated in vitro and in mice. Twenty-two healthy volunteers received single oral doses of either strain in a physiological study of safety, shedding immunogenicity. Volunteers were observed in the hospital for 7 days and had daily blood cultures, routine safety labs, and fecal cultures; none had fever, positive blood cultures, prolonged shedding, or serious or unexpected events. Four of 12 volunteers who received the actA/plcB-deleted strain had minor, transient, asymptomatic serum transaminase elevations (maximum increase 1.4× upper normal). Six of 6 volunteers who received ≥4×109 CFU (colony forming units) had detectable mucosal immune responses to listerial antigens, but not to the vectored influenza antigen. Approximately half the volunteers had modest Interferon-γ ELISpot responses to a complex listerial antigen, but none had increases over their baseline responses to the influenza antigen. Comparison with prior work suggests that foreign antigen expression and perhaps also freezing may adversely affect the organisms' immunogenicity.

Keywords: attenuated vaccine vector, ELISpot, human study, Listeria monocytogenes

2. Introduction

Live attenuated bacteria expressing foreign antigens present a promising approach in the development of human immunotherapies and vaccines. Favorable attributes of Listeria monocytogenes include its innate immunostimulatory properties, efficient cytosolic entry of antigen presenting cells and its ability to stimulate vigorous CD4 and CD8 T cell responses. L. monocytogenes antigen processing and presentation via the major histocompatability complex (MHC) class I pathway makes it an attractive vector for delivery of viral and cancer antigens. Animal studies show that L. monocytogenes vectors can generate T cell-mediated immune responses to lymphocytic choriomeningitis virus, influenza virus, HIV-1 Gag, and human papilloma virus-16 antigens(15). Despite these promising results, there are safety concerns related to using recombinant wild type L. monocytogenes as a vaccine vector due to the high degree of morbidity and mortality that results from naturally acquired infection. Several groups have developed strategies for attenuating L. monocytogenes by deletion of specific virulence genes(6, 1, 7, 8) and 2 strains have been evaluated in published human studies: ΔactA/plcB (9) and ΔprfA (10)via oral and parenteral routes respectively.

In our prior oral clinical study, 20 healthy adult volunteers received escalating single doses of live attenuated L. monocytogenes ΔactA/plcB safely. No individual experienced a serious adverse event. Three of 20 individuals had mild elevations in serum transaminases (maximum 2.5× upper normal) that were temporally associated with vaccination and could not be otherwise explained. Subsequently another biotechnology group developed an L. monocytogenes ΔactA/inlB strain, specifically demonstrated to minimize invasion of hepatocytes(6). In order to evaluate these two vector candidates, we further engineered these otherwise isogenic strains to express an identical influenza A heterologous antigen from a chromosomally located gene fusion. The intent of the study was to evaluate the safety of the vectors by the oral route, and determine in a translational study whether human immune responses to a vectored viral antigen could be detected.

Influenza A nucleoprotein (NP) was chosen as a model viral antigen, as it has been evaluated previously as a conserved, and potentially cross-protective vaccine antigen for influenza(1113). Influenza A NP has been successfully expressed in L. monocytogenes (14, 2) and, as a component of both live and killed influenza vaccines given to millions, is likely safe to administer to volunteers. An Influenza A NP gene segment was chosen to include known human T cell epitopes (15, 16). Additionally a well-studied 9 amino acid epitope of the Influenza A M1 matrix protein recognized by HLA-A2 humans was included, GILGFVFTL,(17), as HLA-A2 is a frequent haplotype in our North American Caucasian volunteer population. We report here the preclinical and clinical evaluation of the 2 vector strains BMB72 (ΔactA/plcB-NP) and BMB54 (ΔactA/inlB-NP). This Phase 1 clinical study was performed to further evaluate and compare two listerial vectors, and not intended as a step towards commercialization of these vaccine strains or generation of an oral influenza vaccine.

3. Materials and Methods

3.1 Bacterial Strains and Recombinant Vectors

All the L. monocytogenes strains used in this study are derived from the streptomycin-resistant L. monocytogenes strain 10403S (18).Table 1 contains a list of the bacterial strains used to engineer the recombinant strains and their origins. The Influenza A gene fusions were constructed by generating a synthetic polynucleotide coding for the GILGFVFTL epitope of the influenza A M1 protein which was ligated to DNA encoding a 297 amino acid portion of the Influenza A nucleoprotein (NP) and cloned into the pEJ140PhoA vector (a gift from Jeff F. Miller PhD, UCLA). The Influenza A nucleoprotein segment was constructed by PCR amplification from a L. monocytogenes strain (DPL1659 - a gift from Daniel Portnoy PhD at UC Berkeley) which expresses amino acids 1–480 of the Influenza A nucleoprotein (Influenza A/PR/8/34), using primers (5'-to 3') TTGGATCCCCAGGGTTCGACTCCT and GGGCGCGCCGGAGGCCCTCTGTTG. The modified pEJ140PhoA plasmid was then digested with NotI and the fragment containing the Influenza A NP fusion protein was ligated into the NotI site of a modified pPL2 site specific integration vector (19). The resulting plasmid was then transformed into E. coli SM10(20) and subsequently mated into, and then plasmid sequences cured from, the attenuated background L. monocytogenes strains. Three nested segments of nucleoprotein of increasing size were evaluated for expression. As there was no obvious difference in expression, the largest of the three, encompassing 7 known human epitopes(15, 16) was carried on for further study. Chromosomal deletions and the foreign antigen cassette insertion were confirmed by PCR sequencing. The final foreign antigen cassette is shown graphically in Figure 1. Gel electrophoresis and Western blotting to nitrocellulose was performed using standard methods. A commercially available rabbit polyclonal antibody to E. coli alkaline phosphatase (Abcam, Cambridge MA) was used with a goat anti-rabbit peroxidase secondary antibody (KPL, Gaithersburg MD) and a chemiluminescent substrate (LumiGlo, KPL).

Table 1
Strain Table/ Murine Lethal Dose(50)

3.2 Macrophage survival and L929 Plaquing Assays

Bacterial cultures were grown ~16 h in trypticase soy broth (TSB). J774A.1 murine macrophage monolayers (ATCC, Manassas VA) in 24-well plates were infected at a multiplicity of infection (MOI) of 20:1 and gentamicin exclusion assays for intracellular survival were performed as previously described (21). L929 murine fibroblast monolayers (ATCC) were infected with L. monocytogenes (MOI 1:50) and plaques measured 5 days later(22).

3.3 Murine experiments

Animal experiments were reviewed and approved by the IACUC at Massachusetts General Hospital and 8–12 week female BALB/c mice from Charles River Laboratories (Wilmington MA) were used for all experiments. L. monocytogenes strains were grown overnight in TSB containing streptomycin(100 ug/ml). Cultures were pelleted, washed once with normal saline and resuspended in sterile normal saline. Serial 10-fold dilutions were made and groups of 6 mice were injected intraperitoneally (i.p.) in a 300μl volume. In addition to the vaccine strains expressing the influenza antigen, 3 groups of control animals received either the wild type, BMB72 parent strain, or BMB54 parent strain. Animal health was assessed several times daily and the LD50 was calculated (23). For visceral persistence studies mice were inoculated i.p. once with 0.1 LD50 of either wild type (WT), BMB54, or BMB72. Mice were sacrificed at days 1, 3, 7, and 11, spleens and livers homogenized for one minute in 2ml buffered saline, serially diluted and plated in triplicate on TSB plates. For ELISpot studies mice received ~0.1 LD50 of relevant strains and were sacrificed 7 days later. Murine spleens were pooled by vaccine strain (3 animals/group), processed with mesh strainers and red cells were lysed using ammonium chloride buffer. ELISpot experiments were performed using a pair of monoclonal antibodies (one biotinylated) directed against mouse IFN-γ (Pierce, Rockford IL). Plates were then washed and developed with streptavidin-alkaline phosphatase conjugate and nitroblue tetrazolium and 5-bromo-4-chloro-3indolyl phosphate (BCIP) (Bio-Rad, Hercules CA). The lectin control stimulus for murine ELISpot studies was concanavalin A. Test peptides included 11–13mer overlapping peptides based upon the nucleoprotein sequence (3 different influenza peptide pools of approximately 16 peptides each, synthesized by the Partners AIDS Research Center, Boston, MA). Three pools were made to avoid excessive relative dilution of individual peptides. Pool #1 additionally included the matrix protein epitope GILGFVFTL (a human but not a murine epitope). A single large pool of listeriolysin peptides (15mers overlapping by 11) was the kind gift of Cerus Corporation (Concord CA).

3.4 Volunteer study

3.4a Human subjects and volunteer screening

The study was reviewed and supervised by a local IRB (Massachusetts General Hospital) and the NIH and Biosafety Committees(Harvard Medical School Committee on Microbiological Safety). The study was also reviewed and approved by NIH/NIAID Prevention Science Review Committee, an independent Data Safety Monitoring Board (DSMB) (3 physicians with expertise in listeriosis, clinical trials and enteric infections), an NIH physician medical monitor and the US Food and Drug Administration (FDA, BB IND # 12760). All of these groups appreciated the goals of the study to be further evaluation of safety and physiological and immunological responses to listerial vectors and not as a prelude to development of a new influenza vaccine, and found it ethically acceptable. All subjects provided written informed consent to participate.

Healthy men and women 18 to 45 years old were recruited by advertising and underwent a complete screening physical exam and standard laboratory procedures as described (9). Potential volunteers must have previously tolerated a course of therapy with penicillin or ampicillin. In addition to standard clinical screening laboratories, volunteers were required to have normal iron studies, normal liver function tests, and a pre-study stool sample which was negative for routine enteric pathogens, ova and parasites, and L. monocytogenes. Subjects were not HLA haplotyped, as this would have been prohibitively limiting and expensive. Subject were also not screened in any way for previous exposure to L. monocytogenes or for previous influenza exposure, expecting that most would have been previously exposed, especially to influenza. L. monocytogenes are ubiquitous organisms despite best food safety efforts. It was hypothesized that existing immunity to influenza or listerial antigens might be “boosted” by this oral vaccination.

3.4b Inoculum preparation

Frozen inocula (0.9% w/v saline with 20% glycerol, 1.3 ml/vial) were produced utilizing good manufacturing practices (GMP) by contract with the Walter Reed Army Institute of Research Pilot Bioproduction Facility (Silver Spring MD), a requirement of the funder. Vials were thawed at 4°C for 15 minutes and diluted into 0.9% saline for administration to volunteers in 30 ml. Based upon spread plate cultures prior to freezing and multiple assessments of thawed vials, each inoculation of a given number of live colony forming units(CFU) also contained approximately 2-fold greater dead organisms, or residual thereof. Volunteers ingested 2 g of sodium bicarbonate in 150 ml distilled water just prior to inoculation. Actual doses administered were confirmed by serial dilution and counting colonies from triplicate spread plate cultures.

3.4c Clinical assessments and microbiology

Subjects were admitted to the Clinical Research Center at Massachusetts General Hospital for 7 days and had frequent clinical exams, with vital signs taken at least 4 times a day. Volunteers had routine safety blood tests (complete blood count with differential, hepatic and renal function) done on study days 0, 4, 7, 10, 14, and 28, and additionally as deemed appropriate. Peripheral blood mononuclear cells (PBMC) were isolated from heparinized blood via Ficoll gradient separation on days 0, 7, 10, 14, and 28. After discharge, volunteers returned weekly for 6 weeks for a clinical check, stool culture, and immunology samples. A clinical check and blood sampling for serum occurred on days 0, 4, 7, 10, 14, 21, 28, 56, and 168 (6 months after vaccination).

Volunteers had daily blood cultures (Bactec system 9240). All stools passed were graded(24); up to three stools per day were directly cultured (9)for L. monocytogenes on Brucella agar plates with horse blood and on Oxford L. monocytogenes agar plates (both containing streptomycin). Stool samples were also heavily inoculated overnight into University of Vermont (UVM) L. monocytogenes enrichment broth (Difco, Sparks MD) and subsequently an aliquot of suspension was then inoculated onto the same selective agar plates. If no stools were passed by 8 p.m. on a given day, a rectal swab was obtained and incubated overnight in UVM enrichment broth. Quantitative colony counts were not performed. Bacterial isolates from fecal samples were confirmed to be L. monocytogenes by morphology and standard phenotypic tests (beta hemolysis, Gram stain, catalase and motility tests). The last fecal isolate obtained on each subject was also identified by automated biochemical assay (VITEK BioMerieux, Hazelwood MO) and tested for antimicrobial sensitivity to penicillin and streptomycin.

3.5 Immunological assessments

3.5a Antigens

A recombinant 6-histidine-tagged listeriolysin (25) was purified from E. coli via nickel affinity chromatography from a clone generously provided by Daniel Portnoy PhD (UC Berkeley) (26). A soluble sonicate suspension was prepared from L. monocytogenes 10403S as described previously (27) and considered a complex antigen of listerial components. A recombinant N-terminal his-tagged Influenza A nucleoprotein derived from strain A/PR/8/34 (HON1) was cloned into pET30a expression vector (Novagen/EMD, Darmstadt, Germany)/E. coli BL21, and subsequently purified by nickel affinity chromatography on a large scale by the New England Regional Center of Excellence/Biodefense and Emerging Infections (Boston MA).

3.5b IgA and Interferon-γ ELISpot studies

Both IgA ELISpot and Interferon-γ (IFN-γ) ELISpot studies were performed as described (28, 29) using freshly isolated PBMC maintained in R10 medium with fetal calf serum. These primary measures of mucosal and cellular immune responses respectively. Millipore HA cellulose ester (MAHAS4510) were used for IgA ELISpot assays and Immobilon P (MAIPS4510) membrane-bottom plates were used for and IFN-γ ELISpot assays. For IgA ELISpots, antigens used were recombinant nucleoprotein, recombinant listeriolysin, and sonicated WT listerial antigen. Antibodies in cultured lymphocyte supernatants were also harvested for soluble vaccine-specific immunoglobulins by ELISA, as described(25), an assay also known as the ALS assay (30). For IFN-γ ELISpots control wells included PHA (phytohemagglutinin) and “CEF” a commercially available standard peptide pool including 32 CMV, EBV and influenza virus peptides 8–12 amino acids in length (AnaSpec, San Jose CA). Test peptides included the same 3 influenza peptide pools and listeriolysin O (25) peptide pool described above. The complex whole listerial antigen was also used in IFN-γ ELISpot studies. Spots were counted by an automated reader (Immunospot; CTL, Shaker Heights OH). Low level spot counts in unstimulated medium only control wells were subtracted from the test wells. The IFN-γ ELISpot results are presented as mean values of duplicate wells per condition as spot-forming cells (SFC)/106 PBMC. A positive response for an individual was defined as more than 2-fold greater than baseline results for that antigen and over 100 SFC/106 PBMC (31, 32). Because IFN-γ responses did not appear related to oral dose given, results were also analyzed as a whole by organism given, comparing pre-immune with peak values (see Figure 7).

3.5c Seroconversion/ELISA

Serum samples were studied by ELISA to quantify immunoglobulin G and A directed against sonicated listerial antigens, recombinant his-tagged listeriolysin and Influenza A nucleoprotein over time. Antigens were suspended in PBS and used to coat Nunc-Immuno Maxisorp 96-well plates (Nalge Nunc International, Roskilde Denmark). Assays were performed as described (9) and read on a Vmax kinetic microplate reader (Molecular Devices, Sunnyvale CA). Endpoint dilutions are reported as the highest dilution at which a serum sample was ≥0.14 OD units at 405 nm, an arbitrarily chosen cutoff value. Fourfold or greater increases in endpoint titer were considered a positive result. The differences in geometric means between groups were compared statistically with the Mann-Whitney test.

4. Results

4.1 In vitro characterization

Both vaccine strains were demonstrated by sequencing to contain the expected deletions and heterologous fusion antigen. The introduction of the attenuating mutations ΔactA/plcB and ΔactA/inlB did not significantly alter growth kinetics in TSB broth as measured by optical density, nor did the incorporation of the “empty” integration vector, pPL2. Introduction of the foreign antigen fusion cassette did moderately alter growth kinetics; both the rate of growth and the final density of growth were slightly depressed (OD600 nm ~1.5 to 2.0 vs. ~2.3 to 2.7)). Both strains were found to express and secrete modest amounts of a 77-kDa fusion protein antigen and significant amounts of degradation products into culture supernatants by Western blot analysis (Figure 2) using an antibody directed against the alkaline phosphatase tag. Because bacterial vectors are intended to survive and secrete antigens over time intracellularly, antigen load and stability in vitro may not correlate with immunogenicity in vivo. All commercially available antibodies directed against Influenza A nucleoprotein failed to detect the limited section of influenza NP included. The NP region included was engineered to include known human T-cell epitopes, not antibody epitopes. Larger fusion antigens were not easily cloned or secreted in our system (data not shown). We concluded that commercial antibodies were directed at NP epitopes not included in the fusion antigen.

Because intracellular survival and inter-cell spread are important correlates of in vivo virulence in many bacteria, these phenotypes were studied. We found no significant differences in intracellular survival of the vaccine organisms within J774 murine macrophages over 6 hours as compared to either the parental mutants lacking the NP fusion antigen, or WT organisms (data not shown). The ability to plaque (generate a cleared area of dead cells lysed by L. monocytogenes) in L929 murine fibroblasts is used as a marker of cell-to-cell spread. Both the parental mutants and attenuated vaccine strains both had severe defects in plaquing capability, as expected for ΔactA mutants which cannot polymerize actin and move intracellularly(33). On average (20 plaques, mean +/− SD), WT organisms generated a plaque size of 1.48 +/−0.23 mm. The mutant strains, BMB07 and BMB16 generated plaques sizes of 0.58 +/−0.13 mm and 0.56 +/− 0.10 mm respectively. The vaccine strains, BMB72 and BMB54 generated even smaller plaques of 0.45 +/− 0.13 mm and 0.43 +/− 0.11 mm respectively.

4.2 Murine studies

BMB54 and BMB72 were evaluated in mice by i.p. inoculation to quantify mammalian virulence in comparison with WT organism and with our vector strain previously evaluated in humans (9). Table 1 shows that the parental mutant strains BMB07 and BMB16 are much less virulent than wild type organisms with the LD50 of these strains differing from wild type by approximately 3 log10 colony forming units (CFU). The addition of the Influenza A NP antigen cassette in strains BMB54 and BMB72 results in modest further attenuation by approximately 0.5 log10 CFU when compared “head-to-head.”

Others have shown that the BMB54 parental strain is cleared more rapidly from the spleens and livers of mice than wild type (WT) organisms, suggesting that this strain might have an improved clinical safety profile (6). We compared the visceral clearance of the investigational vaccine strains BMB54 and BMB72 and found that splenic and hepatic clearance was synchronous, and therefore we present data from the liver only. As shown in Figure 3, both BMB54 and BMB72 were cleared from the liver by 7 days, but WT organisms were present in large numbers at 11 days. The overall kinetics of bacterial persistence are strikingly different. The WT organisms undergo initial growth through day 3 (~2 log CFU increase), while vaccine organisms undergo continuing reductions in visceral counts.

Murine experiments were performed to document that the vaccine strains could stimulate detectable cellular responses directed against nucleoprotein peptides and listerial peptides, as that was the planned immunological readout of the clinical study. As the heterologous antigen insert was explicitly engineered to include human T-cell epitopes and not include murine T-cell epitopes, there was no attempt made to optimize or maximize murine immune responses. Figure 4 shows that animals receiving vaccine strains had increases in nucleoprotein-specific IFN-γ spots as compared with animals inoculated with saline or background vector strains lacking the NP fusion antigen. Spots in concanavalin A control wells were too numerous to count (TNTC, confluent). All groups receiving any L. monocytogenes strain had strong responses to the listeriolysin peptide pool (over 300 spots/106 splenocytes; not shown in Figure 4).

4.3 Clinical study

4.3a Subjects and doses

A total of 225 people were screened by phone to find 54 to undergo full screening, of whom 22 qualified and provided written informed consent to participate (17 men, 5 women; 16 Caucasian, 3 African American, 2 Hispanic, 1 Asian American). Doses planned/given are shown in Table 2, and actual CFU delivered as measured by plating of each inoculum were within 15% of the planned dose as anticipated. An independent safety monitoring board required an interim dose escalation step of 4 × 109 for strain BMB72 because of small increases in liver function tests observed in a few subjects at lower doses (see below).

Table 2
Clinical Results and Seroconversion

4.3b Clinical responses

All volunteers completed the 7 day hospital stay uneventfully. No volunteer had fever, positive blood cultures, prolonged shedding, or serious or unexpected problems or laboratory findings. One volunteer (No. 2) vomited approximately 16 hours after receiving the oral vaccine. He felt well afterwards and had no associated fever, constitutional or additional gastrointestinal symptoms. One volunteer (No.11) had an isolated headache during hospitalization which resolved. One volunteer receiving the highest dose (No.21) had transient diarrhea on day 2 of his inpatient stay, but experienced no other symptoms over the course of his stay. This volunteer also received a 3 day course of oral amoxicillin upon leaving the hospital for a preliminarily positive stool culture at the time of discharge, per protocol. This culture was ultimately finalized as negative for the vaccine organism. One subject (No. 5) could not complete follow-up through day 56, ending instead at day 35; three additional subjects could not attend their day 168 visit (all a result of change of residence).

4.3c Laboratory safety data

Serial monitoring of complete blood count (CBC), differential, platelets and blood urea nitrogen/creatinine (BUN/Cr) (days 0, 4, 7, 10, 14, 28) revealed no significant abnormalities or safety signals. Several subjects with low baseline leukocyte counts had values slightly below the hospital lower normal range of 4.4 × 103/mm3 over the course of the study which were deemed not clinically significant. No left shifts on differential or thrombocytopenia was observed.

Four of 12 volunteers who received BMB72 (subjects 3, 10, 11 and 20) had minor, asymptomatic abnormalities in serum transaminases during the study that could not be definitively attributed to other causes (maximum 1.4× upper normal). In three, these abnormalities peaked on day 7 or 10 and resolved by day 14. In subject #20 studies were normal on days 7 and 10, and a single isolated serum glutamic oxaloacetic transaminase (SGOT) elevation was noted on day 14. Other measures of liver function were normal (bilirubin, alkaline phosphatase) in these volunteers. Due to these abnormalities, the DSMB required that 2 subjects receive strain BMB72 at a dose of 4×109 CFU before proceeding to 1×1010 CFU (see Table 2). Transaminase elevations appeared idiosyncratic rather than dose-dependent as 2 subjects receiving the highest dose of BMB72 (1×1010 CFU) had no transaminase abnormalities. Interestingly, γ-glutamyltransferase (GGT) remained normal throughout in all subjects. Though apparently more specific for hepatic injury than other transaminases, it did not appear a more sensitive marker here. One subject who received BMB54 (No.5) had abnormal transaminases associated with a markedly elevated serum creatinine phosphokinase (CPK) and muscle soreness in the setting of traumatic recreational activities; this was deemed unrelated by the investigators and the independent DSMB.

4.3d Clinical bacteriology

“Naturally occurring” wild type L. monocytogenes was not detected in any fecal samples before inoculation. After inoculation, L. monocytogenes was detected in fecal samples from the majority of subjects (19 of 22), in a pattern comparable to our previously published study. As shown in Table 2, all subjects shed organisms for 5 days or less, and 18 of 22 shed the organism for 2 days or less. In many samples the strain was only detected after incubation in UVM enrichment broth, indicating a low organism burden. Three subjects who received the lowest dose (108 CFU) never had a positive stool culture. The Brucella/blood agar plates were more likely than the Oxford agar plates to detect either organism when present at low numbers.

4.4 Immune responses

4.4a Mucosal Immune Responses

IgA secreting cells in peripheral blood are a sensitive, simply-assayed correlate of fecal IgA. Surprisingly, IgA-secreting cells directed against listerial or influenza antigens were not detected on days 7 or 10 after vaccination in any volunteer. In addition to direct IgA ELISpot studies, PBMC obtained before and on days 7 and 10 after vaccination were cultured for 48 hours, and tissue culture medium was assayed for soluble immunoglobulins directed against listerial antigens - the ALS assay (30, 25). Subjects receiving 4 × 109 CFU or more of either strain had convincing increases in IgA (but not IgG) directed against sonicated listerial antigens on day 10 (see Figure 5). Only the two subjects who received 1010 BMB72 had IgA responses against listeriolysin (data not shown). Responses to influenza nucleoprotein were not detected in these assays. These results were interpreted to represent low level mucosal immune response against the listerial vector only.

4.4b Humoral immune responses

Serological immune responses were modest at best, with isolated individuals having 4-fold or greater titer increases in ELISAs directed against listeriolysin or sonicated listerial antigen (denoted in Table 2 as 1 or 2 assays positive). No individual seroconverted to the recombinant nucleoprotein antigen. Virtually all individuals had relatively high titers directed against recombinant nucleoprotein at baseline which did not change over time (i.e.≥1:640). Sera from other species (mouse and rabbit) studied similarly in ELISAs did not have similarly high baseline values so these were interpreted to represent bona fide pre-existing immune responses to this influenza protein, rather than inadequate blocking or another technical problem with the assay. A high baseline is not unexpected as most subjects had evidence of cellular immunity to influenza A, and it is expected that most healthy young adults would have been exposed to influenza. Grouped by vaccine given, there was no statistically significant increase in IgG mean titers (GMT, pre-immune to peak value) directed against listerial sonicate, listeriolysin or nucleoprotein, as exemplified in Figure 4, panel B (for listeriolysin). Baseline listeriolysin titers were high, which is not unexpected. Antibodies to streptolysins present in commensal and pathogenic streptococci crossreact with listeriolysin(34). Our volunteers were required to have previously received penicillin or ampicillin, commonly administered to treat Group A streptococcal pharyngitis. Overall, mean serum IgA titers did increase modestly when considered as a group for both vaccine organisms (Figure 6, panel A).

4.4c IFN-γ ELISpot results

All subjects had positive control responses to the lectin phytohemagglutinin (PHA) (usually TNTC), and all but one to the CEF control pool (subject #11 had both robust PHA responses and responses to sonicated listerial antigen, but no apparent response to CEF or influenza nucleoprotein peptides). Most subjects (17/22) had convincing baseline immune responses to at least one of the Influenza A nucleoprotein peptide test pools (tens to many hundreds of spots per million PBMC, see exemplary data in panel A, Figure 7). About two thirds of the subjects (14/22) had some baseline responses to the listeriolysin peptide pools, with mean baseline value 21, range 0 to 205 SFC/106 PBMC, comparable to others' published work (35). ELISpot data were analyzed by individual and as a group by vaccine administered, irrespective of dose, as responses overall did not appear dose-related. Values were analyzed as pre-immune vs. peak value, with peaks typically occurring on days 14 or 28. As shown in Figure 7 Panel B, IFN-γ secreting cells responsive to the complex L. monocytogenes sonicate antigen increased overall after vaccination, but no significant changes were detected in response to any of the 3 nucleoprotein pools (pool #1, which includes the peptide GILGFVFKL, is shown as an exemplar in Figure 7, Panel A) or listeriolysin O peptides (Figure 7, Panel C). Responses to the control CEF pool were strong and unchanged over time (Figure 7 panel D), suggesting that the responses to listerial antigens are real, if modest, increases. There was no significant difference between strains in the proportion responding: 6/10 recipients receiving BMB54 and 6/12 receiving BMB72 had significant increases directed against the listerial sonicate antigen, defined as 2-fold over baseline and >100 SFC/106(p=0.69, not significant, NS). Only 2 of 22 subjects overall had an increase in response to LLO peptides by this definition (one receiving each strain). As positive controls and comparators for the nucleoprotein peptide pool ELISpot studies, we also studied 6 healthy adults who received the standard killed parenteral influenza vaccine (before and after vaccination) and 2 individuals moderately ill with outpatient Influenza A diagnosed by direct antigen testing of nasal swabs. Vaccinees had baseline IFN-γ responses comparable to the 22 healthy volunteers studied here, which did not increase after killed vaccine at all. Influenza patients were studied at the time of presentation and diagnosis and 2–3 weeks later, and had marked increases in IFN-γ spots responsive to nucleoprotein peptide pools (5 to 10-fold over baseline). These results demonstrate that we could have detected increases in IFN-γ spots, had they been present.

5. Discussion

This work compares two L. monocytogenes vaccine vector strains expressing a clinically relevant model viral antigen. Both were derived from the same commonly used laboratory L. monocytogenes strain designated 10403S. The BMB72 parental strain was previously evaluated by us in humans(9); the BMB54 parental strain was independently generated and selected by other investigators as a cancer vaccine vector strain based upon its decreased invasion of hepatic cells and favorable inmmunogenicity profile when administered intravenously (i.v.) in mice (6). Secretion of the Influenza A nucleoprotein antigen fusion appeared to result in an in vitro bacterial growth defect in both strain backgrounds, though a growth defect was not appreciated intracellularly in macrophage-like cells over short term experiments. Both strains were markedly attenuated in mice and in their ability to move intercellulary as measured by plaquing.

Both strains were remarkably safe in small numbers of humans when administered orally even at very large doses (1010 CFU). Fecal shedding was comparable to that observed in an earlier study of the BMB72 parent strain, with a trend toward longer shedding at the highest doses. The only safety signal, transaminase elevations, occurred at a rate, and with timing comparable to that in our prior clinical study (9). The elevations were more modest (<1.5× upper normal values here vs. 2 to 3-fold previously), not associated with symptoms, and not notably dose-related. We speculate that some bacteria may translocate the intestinal wall and be transported systemically, but at too low a level to generate strong systemic immune responses or be detected in blood cultures. No subject receiving BMB54 had abnormal transaminases, suggesting that as demonstrated in vitro (6), this organism may have decreased tropism for hepatic cells in humans. Other published murine studies in which the BMB54 parent strain vs. WT organisms were injected i.v. showed that transaminases peaked approximately 1 and 4 days after intravenous administration respectively (6). In that study, the BMB54 mutant caused much lower peak transaminase values, likely because of the lack of replication within the liver. After intravenous administration of a prfA-defective L. monocytogenes vaccine strain to humans, 1.5 to 3.5-fold elevations in both GGT and transaminases were reported at 8 days after administration, but these tests were apparently not performed during days 1 through 7 after iv administration (10). No clinical data suggest these transaminase elevations are harbingers of prolonged or serious hepatic dysfunction. One murine cancer immunotherapy study using an inlB-positive L. monocytogenes strain exploited this hepatotropism. Hepatic metastases were more efficiently eliminated and survival was prolonged when attenuated L. monocytogenes were used as adjuvant/adjunctive therapy for an irradiated tumor cell vaccine expressing granulocyte-macrophage colony-stimulating factor (36), though that study did not include a comparator strain lacking inlB with decreased liver tropism.

We undertook this study to evaluate the physiological effect of two L. monocytogenes organisms as vectors for delivery of viral antigens. Oral delivery was hoped to stimulate or at least “prime” the mucosal immune system, as many viruses enter through mucosal sites. Bulk IFN-γ ELISpot studies performed on freshly isolated PBMC were chosen as a simple, reproducible measure of systemic cellular immunity increasingly used in studies of viral vaccines. Our earlier human study was limited by a lack of immunological reagents, especially peptide reagents for ELISpots – here we were able to test synthesized overlapping peptide pools for both listeriolysin, and the foreign antigen. A recent study of PBMC derived from approximately 80 healthy human donors showed that bulk IFN-γ ELISpot responses to this same listeriolysin peptide pool also correlated in magnitude and incidence with IL-2 ELISpot responses to the pool (35), so this is likely a reasonable measure of cellular immunogenicity.

Overall, mucosal responses were detected and comparable to our prior oral study – present at low levels at the highest doses administered. We had expected greater mucosal responses at the highest doses administered here. Even at 1010 CFU, we only detected soluble immunoglobulin A directed against sonicated L. monocytogenes via the ALS assay; no convincing IgA ELISpot responses were seen. Serum IgA titers directed against the vector were significantly increased overall, though the significance of this finding is uncertain. IgA ELISpots were the best correlates of luminal intestinal (fecal) antibody in earlier assessments of live attenuated Salmonella vaccines where this was carefully studied (37).

Systemic humoral immune responses to vector and the foreign antigen were not detected. Although antibodies may play some role in protection against listeriosis (38), in general listerial vectors are engineered and studied with the goal of stimulating cellular immunity. All of our volunteers had high baseline antibody titers directed against the nucleoprotein antigen, likely a result of prior influenza infection, which did not change over time.

We were somewhat encouraged by an overall statistically significant increase in IFN-γ spot-forming cells responsive to the complex listerial sonicate antigen, if not the listeriolysin peptides, nor the nucleoprotein antigen as shown graphically in Figure 7. In our and others hands, the listeriolysin peptide pool engendered strong ELISpot responses in mice inoculated parenterally with L. monocytogenes expressing listeriolysin. It was expected that these LLO peptides would be strong, sensitive and specific test peptides in humans, which proved to be incorrect. Our data suggest that humans may preferentially respond to other listerial antigens. We had hoped that existing and measureable IFN-γ ELISpot immune responses to influenza nucleoprotein peptides would be “boosted” by presentation of the nucleoprotein by a live listerial vector, but this could not be demonstrated. Based upon our ELISpot and ELISA data, virtually all volunteers had strong existing immune responses to the nucleoprotein. In retrospect, perhaps an antigen to which humans are naïve might have presented a lower bar with which to evaluate these vectors. It is possible that we might have detected greater cellular responses to both vector and heterologous antigens by using more sophisticated T-cell studies with re-stimulation in vitro, but we doubt such results would be clinically meaningful.

In summary, oral administration of these 2 vaccine organisms resulted in modest mucosal and cellular immune responses to a complex listerial antigen, but not to a secreted viral foreign antigen. The strains were comparable, immunologically. In our prior study, there were more robust mucosal and humoral immune responses to both sonicated L. monocytogenes and LLO in subjects receiving 109 CFU of the BMB72 parental strain orally. We had hoped that higher doses and improved peptide reagents would allow us to detect cellular responses, but this was not the case. Overall the vaccine preparations seemed more attenuated than that previously studied based upon the magnitude of mucosal immune responses and liver function test abnormalities. This could be attributable to the foreign antigen expression. Another important difference in the current study is the utilization of a frozen inoculum, mandated by the NIH; our prior study utilized freshly grown organisms centrifuged from stationary phase broth cultures. Virulence factors are regulated by temperature in a complex fashion in L. monocytogenes, and its ability to adapt to and grow at low temperatures is of importance for food safety, as reviewed recently (39). Some strains have a greater “growth lag phase” after cold storage (40). Cryotolerance (freeze/thaw tolerance) appears to be strain dependent, and growth temperatures may affect this (41). It is beyond the scope of this paper to further examine reasons for the poor immune responses observed. Live attenuated bacterial vectors for oral delivery of vaccine antigens have unfortunately not been highly successful in this or other human studies. Perhaps these highly attenuated, safe strains could be used in other applications requiring transient delivery of other molecules or pharmacologic “payloads” to the gut lumen.

6. Acknowledgments

This work was supported in part by NIH/NIAID-NERCE/BEID Career Development Fellowship 5 U54 A1057159-03 (BMB), NIH/NIAID R01 AI51206 (ELH) and grants M01-RR-01066 (Massachusetts General Hospital GCRC) and UL1 RR025758-01 (Harvard Clinical and Translational Science Center) from the National Center for Research Resources. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health. We acknowledge the generosity of the Cerus Corporation, Concord, CA for providing us with the L. monocytogenes ΔactA/inlB strain as well as the LLO peptide pool. We especially thank our volunteers, and the inpatient clinical research center nursing staff and clinical microbiology laboratory at Massachusetts General Hospital.

List of Abbreviations

5-bromo-4-chloro-3-indolyl phosphate
Blood urea nitrogen/creatinine
Complete blood count
CMV, EBV, influenza peptide pool
colony forming units
creatinine phosphokinase
Data Safety Monitoring Board
interferon γ
lymphocytic chorimeningitis virus
Listeriolysin O
major histocompatability complex
not significant
peripheral blood mononuclear cells
alkaline phosphatase
spot forming cells
Serum glutamic oxaloacetic transaminase
too numerous to count
trypticase soy broth
wild type


1. Frankel FR, Hegde S, Lieberman J, Paterson Y. Induction of cell-mediated immune responses to human immunodeficiency virus type 1 Gag protein by using Listeria monocytogenes as a live vaccine vector. J Immunol. 1995;155:4775–82. [PubMed]
2. Ikonomidis G, Portnoy DA, Gerhard W, Paterson Y. Influenza-specific immunity induced by recombinant Listeria monocytogenes vaccines. Vaccine. 1997;15:433–40. [PubMed]
3. Mustafa W, Maciag PC, Pan ZK, Weaver JR, Xiao Y, Isaacs SN, Paterson Y. Listeria monocytogenes delivery of HPV-16 major capsid protein L1 induces systemic and mucosal cell-mediated CD4+ and CD8+ T-cell responses after oral immunization. Viral Immunol. 2009;22:195–204. [PMC free article] [PubMed]
4. Pan ZK, Ikonomidis G, Lazenby A, Pardoll D, Paterson Y. A recombinant Listeria monocytogenes vaccine expressing a model tumour antigen protects mice against lethal tumour cell challenge and causes regression of established tumours. Nat Med. 1995;1:471–7. [PubMed]
5. Shen H, Slifka MK, Matloubian M, Jensen ER, Ahmed R, Miller JF. Recombinant Listeria monocytogenes as a live vaccine vehicle for the induction of protective anti-viral cell-mediated immunity. Proc Natl Acad Sci U S A. 1995;92:3987–91. [PMC free article] [PubMed]
6. Brockstedt DG, Giedlin MA, Leong ML, Bahjat KS, Gao Y, Luckett W, Liu W, Cook DN, Portnoy DA, Dubensky TW., Jr. Listeria-based cancer vaccines that segregate immunogenicity from toxicity. Proc Natl Acad Sci U S A. 2004;101:13832–7. [PMC free article] [PubMed]
7. Goossens PL, Milon G, Cossart P, Saron MF. Attenuated Listeria monocytogenes as a live vector for induction of CD8+ T cells in vivo: a study with the nucleoprotein of the lymphocytic choriomeningitis virus. Int Immunol. 1995;7:797–805. [PubMed]
8. Li Z, Zhao X, Higgins DE, Frankel FR. Conditional lethality yields a new vaccine strain of Listeria monocytogenes for the induction of cell-mediated immunity. Infect Immun. 2005;73:5065–73. [PMC free article] [PubMed]
9. Angelakopoulos H, Loock K, Sisul DM, Jensen ER, Miller JF, Hohmann EL. Safety and shedding of an attenuated strain of Listeria monocytogenes with a deletion of actA/plcB in adult volunteers: a dose escalation study of oral inoculation. Infect Immun. 2002;70:3592–601. [PMC free article] [PubMed]
10. Maciag PC, Radulovic S, Rothman J. The first clinical use of a live-attenuated Listeria monocytogenes vaccine: a Phase I safety study of Lm-LLO-E7 in patients with advanced carcinoma of the cervix. Vaccine. 2009;27:3975–83. [PubMed]
11. Epstein SL, Tumpey TM, Misplon JA, Lo CY, Cooper LA, Subbarao K, Renshaw M, Sambhara S, Katz JM. DNA vaccine expressing conserved influenza virus proteins protective against H5N1 challenge infection in mice. Emerg Infect Dis. 2002;8:796–801. [PMC free article] [PubMed]
12. Taylor PM, Askonas BA. Influenza nucleoprotein-specific cytotoxic T-cell clones are protective in vivo. Immunology. 1986;58:417–20. [PMC free article] [PubMed]
13. Epstein SL. Prior H1N1 influenza infection and susceptibility of Cleveland Family Study participants during the H2N2 pandemic of 1957; an experiment of nature. Journal of Infectious Disease. 2006;193:49–53. [PubMed]
14. Ikonomidis G, Paterson Y, Kos FJ, Portnoy DA. Delivery of a viral antigen to the class I processing and presentation pathway by Listeria monocytogenes. J Exp Med. 1994;180:2209–18. [PMC free article] [PubMed]
15. Bui HH, Peters B, Assarsson E, Mbawuike I, Sette A. Ab and T cell epitopes of influenza A virus, knowledge and opportunities. Proc Natl Acad Sci U S A. 2007;104:246–51. [PMC free article] [PubMed]
16. Jameson J, Cruz J, Ennis FA. Human cytotoxic T-lymphocyte repertoire to influenza A viruses. J Virol. 1998;72:8682–9. [PMC free article] [PubMed]
17. Sauma SY, Gammon MC, Bednarek MA, Cunningham B, Biddison WE, Hermes JD, Porter G, Tamhankar S, Hawkins JC, Bush BL, et al. Recognition by HLA-A2-restricted cytotoxic T lymphocytes of endogenously generated and exogenously provided synthetic peptide analogues of the influenza A virus matrix protein. Hum Immunol. 1993;37:252–8. [PubMed]
18. Bishop DK, Hinrichs DJ. Adoptive transfer of immunity to Listeria monocytogenes. The influence of in vitro stimulation on lymphocyte subset requirements. J Immunol. 1987;139:2005–9. [PubMed]
19. Lauer P, Chow MY, Loessner MJ, Portnoy DA, Calendar R. Construction, characterization, and use of two Listeria monocytogenes site-specific phage integration vectors. J Bacteriol. 2002;184:4177–86. [PMC free article] [PubMed]
20. Simon R, Priefer U, Puhler A. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria Biotechnology. 1982;1:784–91.
21. Angelakopoulos H, Hohmann EL. Pilot study of phoP/phoQ-deleted Salmonella enterica serovar typhimurium expressing Helicobacter pylori urease in adult volunteers. Infect Immun. 2000;68:2135–41. [PMC free article] [PubMed]
22. Sun AN, Camilli A, Portnoy DA. Isolation of Listeria monocytogenes small-plaque mutants defective for intracellular growth and cell-to-cell spread. Infect Immun. 1990;58:3770–8. [PMC free article] [PubMed]
23. Reed L, Muench H. A simple method of estimating fifty percent endpoints. American Journal of Hygiene. 1938;27:493–7.
24. Levine MM, Kaper JB, Herrington D, Ketley J, Losonsky G, Tacket CO, Tall B, Cryz S. Safety, immunogenicity, and efficacy of recombinant live oral cholera vaccines, CVD 103 and CVD 103-HgR. Lancet. 1988;2:467–70. [PubMed]
25. Dipetrillo MD, Tibbetts T, Kleanthous H, Killeen KP, Hohmann EL. Safety and immunogenicity of phoP/phoQ-deleted Salmonella typhi expressing Helicobacter pylori urease in adult volunteers. Vaccine. 1999;18:449–59. [PubMed]
26. Gedde MM, Higgins DE, Tilney LG, Portnoy DA. Role of listeriolysin O in cell-to-cell spread of Listeria monocytogenes. Infect Immun. 2000;68:999–1003. [PMC free article] [PubMed]
27. Manohar M, Baumann DO, Bos NA, Cebra JJ. Gut colonization of mice with actA-negative mutant of Listeria monocytogenes can stimulate a humoral mucosal immune response. Infect Immun. 2001;69:3542–9. [PMC free article] [PubMed]
28. Hohmann EL, Oletta CA, Killeen KP, Miller SI. phoP/phoQ-deleted Salmonella typhi (Ty800) is a safe and immunogenic single-dose typhoid fever vaccine in volunteers. J Infect Dis. 1996;173:1408–14. [PubMed]
29. Lalvani A, Brookes R, Hambleton S, Britton WJ, Hill AV, Mcmichael AJ. Rapid effector function in CD8+ memory T cells. J Exp Med. 1997;186:859–65. [PMC free article] [PubMed]
30. Chang HS, Sack DA. Development of a novel in vitro assay (ALS assay) for evaluation of vaccine-induced antibody secretion from circulating mucosal lymphocytes. Clin Diagn Lab Immunol. 2001;8:482–8. [PMC free article] [PubMed]
31. Lawn SD, Bangani N, Vogt M, Bekker LG, Badri M, Ntobongwana M, Dockrell HM, Wilkinson RJ, Wood R. Utility of interferon-gamma ELISPOT assay responses in highly tuberculosis-exposed patients with advanced HIV infection in South Africa. BMC Infect Dis. 2007;7:99. [PMC free article] [PubMed]
32. Sander CR, Pathan AA, Beveridge NE, Poulton I, Minassian A, Alder N, Van Wijgerden J, Hill AV, Gleeson FV, Davies RJ, Pasvol G, Mcshane H. Safety and immunogenicity of a new tuberculosis vaccine, MVA85A, in Mycobacterium tuberculosis-infected individuals. Am J Respir Crit Care Med. 2009;179:724–33. [PMC free article] [PubMed]
33. Skoble J, Portnoy DA, Welch MD. Three regions within ActA promote Arp2/3 complex-mediated actin nucleation and Listeria monocytogenes motility. J Cell Biol. 2000;150:527–38. [PMC free article] [PubMed]
34. Gholizadeh Y, Poyart C, Juvin M, Beretti JL, Croize J, Berche P, Gaillard JL. Serodiagnosis of listeriosis based upon detection of antibodies against recombinant truncated forms of listeriolysin O. J Clin Microbiol. 1996;34:1391–5. [PMC free article] [PubMed]
35. Leong ML, Hampl J, Liu W, Mathur S, Bahjat KS, Luckett W, Dubensky TW, Jr., Brockstedt DG. Impact of preexisting vector-specific immunity on vaccine potency: characterization of listeria monocytogenes-specific humoral and cellular immunity in humans and modeling studies using recombinant vaccines in mice. Infect Immun. 2009;77:3958–68. [PMC free article] [PubMed]
36. Yoshimura K, Jain A, Allen HE, Laird LS, Chia CY, Ravi S, Brockstedt DG, Giedlin MA, Bahjat KS, Leong ML, Slansky JE, Cook DN, Dubensky TW, Pardoll DM, Schulick RD. Selective targeting of antitumor immune responses with engineered live-attenuated Listeria monocytogenes. Cancer Res. 2006;66:1096–104. [PubMed]
37. Forrest BD. Identification of an intestinal immune response using peripheral blood lymphocytes. Lancet. 1988;1:81–3. [PubMed]
38. Edelson BT, Unanue ER. Intracellular antibody neutralizes Listeria growth. Immunity. 2001;14:503–12. [PubMed]
39. Chan YC, Wiedmann M. Physiology and genetics of Listeria monocytogenes survival and growth at cold temperatures. Crit Rev Food Sci Nutr. 2009;49:237–53. [PubMed]
40. Buncic S, Avery SM. Relationship between variations in pathogenicity and lag phase at 37 degrees C of Listeria monocytogenes previously stored at 4 degrees C. Lett Appl Microbiol. 1996;23:18–22. [PubMed]
41. Azizoglu RO, Osborne J, Wilson S, Kathariou S. Role of growth temperature in freeze-thaw tolerance of Listeria spp. Appl Environ Microbiol. 2009;75:5315–20. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...