- Email: [email protected]
Potential live vaccines for HIV
Vaccine 19 (2001) 735 – 742 www.elsevier.com/locate/vaccine
Potential live vaccines for HIV Mary Susan Burnett 1, Ning Wang, Matthias Hofmann 2, G. Barrie Kitto * Department of Chemistry and Biochemistry, Institute of Molecular and Cellular Biology, The Uni6ersity of Texas at Austin, Austin, TX 78712, USA Received 23 June 1999; received in revised form 31 July 2000; accepted 31 July 2000
Abstract Potential live vaccines for HIV were developed using an Lpp-OmpA system to target an HIV antigen, reverse transcriptase, or an immunodominant epitope of this enzyme, to the outer membrane of an attenuated strain of Salmonella SL3261. These live vaccines were administered orally to mice, and fecal IgA and helper T cell responses were measured. Results indicated a fecal IgA response specific to HIV reverse transcriptase, as well as a reverse transcriptase-specific helper T cell response, as measured by proliferation assays. Additionally, tests with the epitope vaccines showed a selective cytotoxic CD8 T cell response. These results suggest that this method of antigen targeting to the outer membrane of attenuated bacterial vectors is very promising not only for HIV vaccine development, but also for antigens from other viral or bacterial pathogens, which could be inserted into the Lpp-OmpA protein construct, to elicit mucosal IgA and T cell responses. © 2000 Published by Elsevier Science Ltd. Keywords: Salmonella; Reverse transcriptase; Mucosal IgA
1. Introduction Despite the recent advances with antiviral drugs for treating HIV-positive patients, the new triple drug therapy is not a global solution. While these new antiviral drugs are having very positive results with many patients, they do not work for everyone. Additionally, the treatment program is a very demanding regimen, requiring patients to take up to 70 pills per day [1]. Not all patients can adhere to such a strict schedule, which basically excludes them from treatment. Perhaps the greatest problem with these drugs is that they are not available to patients in the developing world. The annual cost for treatment with these drugs may total $30 000. In developing countries, where the vast majority of new cases are diagnosed, this price tag makes treatment essentially unavailable. Historically, vaccines are the most cost-effective method of handling a disease * Corresponding author. Tel.: +1-512-4713279; fax: + 1-5124718696. E-mail address: [email protected] (G. Barrie Kitto). 1 Present address: Cardiovascular Research Foundation, Washington Hospital Center, Washington, DC 20010, USA. 2 Present address: Institute for Cell Biology, Division of Immunology, University of Tuebingen, 72076 Tuebingen, Germany.
of this magnitude. When one considers the estimation that 40 million people will be infected by the year 2000 [2], the development of an effective vaccine for HIV is imperative. In designing a vaccine for HIV it is currently thought that humoral and cell-mediated immune responses will both play a role in providing protection against viral infection [3–5]. Recent results have indicated the presence of HIV specific IgA in cervical swabs or urine samples from exposed seronegative individuals [6]. These individuals had participated in unprotected sexual intercourse with their infected partners for periods ranging from 2 to 8 years in length and remained uninfected with HIV. This data suggests that the HIV-specific mucosal IgA may be affording protection from viral infection. For this reason, attenuated Salmonella are an attractive vehicle for delivering HIV antigens to the immune system, as such bacteria have been shown to elicit humoral, cell-mediated and mucosal responses to expressed antigens, as well as to the wild type of their species [7–9]. While it has traditionally been difficult to elicit an immune response at the mucosal surface, the use of Salmonella as a live vaccine vector is ideal because it can be administered orally, allowing the Salmonella to infect the mucosal membranes of the gastrointestinal tract
0264-410X/00/$ - see front matter © 2000 Published by Elsevier Science Ltd. PII: S 0 2 6 4 - 4 1 0 X ( 0 0 ) 0 0 3 0 3 - 0
736
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
and reproductive tracts, entry points for the majority of HIV infections. An oral live vaccine would not only be inexpensive to produce and store, but also easy to administer, particularly in developing countries. In an attempt to elicit a stronger immune response than that obtained with previous Salmonella vaccines, we have explored the method of surface targeting for delivering the HIV antigens to the immune system [10]. Plasmids were constructed that resulted in the expression of HIV antigens fused to an Lpp-OmpA sequence [11]. This method of expression resulted in the HIV antigens being targeted to the bacterial outer membrane. Recent results from Richins et al. [12] describe the surface-expression of organophosphorus hydrolase using this same Lpp-OmpA fusion system. Several factors led to the decision to use the HIV antigen reverse transcriptase in this vaccine construct. HIV reverse transcriptase has been shown to elicit cytotoxic T lymphocytes in both infected patients [13,14] and exposed, but seronegative, individuals [15]. Since a strong cell-mediated response is desired in an HIV vaccine, this fact made reverse transcriptase a good candidate. Additionally, the HIV-1 pol gene, which encodes the reverse transcriptase protein, has been found to be more highly conserved than other HIV genes among different primary isolates [16]. Other studies have determined that pol-specific humoral cross-reactivity exists between HIV-1 and HIV2, as well as considerable sequence homology [17,18]. Additionally, recent work has focused on using internal viral proteins as antigens, as opposed to envelope proteins, in the hopes of inducing a cell-mediated response [19,20]. In this study we describe the construction of this live vaccine, and the testing of this vaccine in mice. BALB/c mice were inoculated orally with recombinant Salmonella and the immune responses were monitored. Specifically, mucosal IgA levels were measured, as was helper T cell proliferation. We also constructed an Lpp-OmpA fusion protein system carrying a class Irestricted epitope consisting of residues 203 – 219 from HIV-1 reverse transcriptase. This relatively conserved HIV-1 reverse transcriptase epitope can be recognized by murine and human cytotoxic T cells [21]. Our reasons for doing so were as follows: this construct was more likely to facilitate external expression on the surface of Salmonella and such a construct would allow for ready manipulation of the peptide sequence to determine the most effective vaccine. Recent studies have shown that the flanking residues of CTL epitopes influence the immune response [22,23]. We therefore established a plasmid construct with one lysine residue between the LppOmpA gene and the RT epitope gene as a vaccine candidate.
2. Materials and method
2.1. Recombinant DNA techniques Using polymerase chain reaction techniques, new restriction enzyme sites were engineered on the ends of the reverse transcriptase gene (supplied from NIH AIDS Reagent Bank, in plasmid pKRT2). The first primer, 5% GCG AAG CTT CAT GGC CAT TAG CCC TAT TGA 3%, resulted in the addition of a HindIII site at the 5% end of the reverse transcriptase gene. The second primer, 5% GCC GCT GGA TCC TAG TAT TTT CCT GAT TCC 3% resulted in the addition of a BamHI site at the 3% end of the reverse transcriptase gene. This amplified 1.7-Kb gene was then inserted into the plasmid pSP72 (Promega), which confers ampicillin resistance. The plasmid pTX101, containing the Lpp-OmpA sequences required for outer membrane targeting, along with the lpp and lac promoters, was obtained from Charles Earhart (University of Texas at Austin). This sequence was first inserted into a pCRII vector (Invitrogen), following amplification using the polymerase chain reaction. The desired sequence was then amplified using polymerase chain reaction, to contain HindIII sites on either end, using the following primers, 5% ACA TGC GAA GCT TCA GTT GTC CGG ACG AGT 3% and 5% CGC CTG AAG CTT TAT CTA AGA TGA ATC CGA TG 3%. The 850-bp Lpp-OmpA sequences, along with the lpp and lac promoters, were then inserted into the recombinant plasmid pSP72 containing the HIV reverse transcriptase gene. The resulting recombinant plasmid was called pHART. A plasmid containing the Lpp-OmpA sequence followed by the HIV tat gene was also constructed, and is currently still under study.
2.2. Recombinant plasmid for HIV-re6erse transcriptase epitope A DNA fragment encoding the selected HIV reverse transcriptase peptide was amplified by PCR using primers: 5%-GAATTCGGAAAGAAATCTGTACCGAAATG-3% and 5%-GGATCCTACGGACCGATCTTGGAG-3%. This amplified 100-bp gene was then inserted into a plasmid containing the Lpp-OmpA sequence required for outer membrane targeting, along with the lpp and lac promoters, as described previously. The resulting recombinant plasmid was called pRT. Another recombinant plasmid with one lysine residue inserted between Lpp-OmpA sequence and reverse transcriptase peptide sequence was constructed with the same method and the recombinant plasmid was called pRTLys.
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
2.3. Fusion protein expression The recombinant plasmid pHART was then electroporated into E. coli DH5a, and expression of the HIV reverse transcriptase protein was verified using Western blotting techniques of whole cell lysates. Cells were grown overnight with 500 mg/ml ampicillin where necessary. Cells were pelleted in the presence of Complete™ Protease Inhibitor (Boehringer Mannheim), then boiled with 2X Tris – Glycine sample buffer (Novex) containing DTT for 5 min. Samples were loaded onto a Novex 10% Tris – Glycine SDS gel. Samples were transferred from gel to PVDF membrane, and blocked using 3% BSA in PBS. Affinity purified polyclonal anti reverse transcriptase antibody was prepared by James Caras. Western blotting was performed using the chemiluminescence SuperSignal system (Pierce). The pHART plasmid was then used to transform the Aro− SL3261 (generously donated by B. Stocker). Expression of reverse transcriptase was verified using Western blotting, as described above, of both whole cell lysates and membrane fractions. Bacterial inner and outer membranes were separated using a sucrose gradient.
2.4. Fusion protein expression for the HIV-re6erse transcriptase epitope The recombinant plasmids pRT and pRTLys were electroporated into E. coli DH5a and Salmonella Aro− SL3621 and the expression of the HIV reverse transcriptase peptide was verified using Western blotting techniques with whole cell lysates. Cells were grown overnight with 100 mg/ml ampicillin. Cells were pelleted and then boiled with 2X Tris – Glycine sample buffer (Novex) for 5 min. Samples were loaded onto a Novex 10% Tris–Glycine SDS gel, then transferred from gel to PVDF membrane and blocked using 1% BSA in PBS. Polyclonal anti-reverse transcriptase peptide antisera were purified by protein A column and Western blotting was performed as described above.
2.5. Trypsin digestion of Salmonella to determine cell surface expression Salmonella cells containing pRT and pRTLys plasmid were grown overnight. One ml of cells was pelleted at 14 000 rpm for 5 min at room temperature. The cells were resuspended in 50 ml of 25 mM HEPES buffer (pH 8.0) and incubated on ice for 30 min. Trypsin was added to obtain a final concentration of 0.1 mg/ml and the mixture was incubated at 37°C for 5 min. At this time, 1% Triton X-100 was added and the reaction was terminated by addition of 55 ml 2X SDS Tris–Glycine sample buffer. The samples were then boiled for 5 min. Samples were loaded onto a Novex 10% Tris–Glycine SDS gel. Samples were then
737
transferred from the gel to PVDF membrane. Western blotting was performed.
2.6. Immunization of mice with Salmonella expressing the intact HIV re6erse transcriptase Bacterial cultures were grown in 50 ml of LB medium with or without 500 mg/ml ampicillin as appropriate. Cultures were shaken for 22 h at 37°C. Cells were pelleted at 5000 rpm for 10 min, then resuspended in 500 ml of PBS. Thirty-ml aliquots were placed in labeled 0.5-ml centrifuge tubes and kept on ice until feeding. For the IgA studies and proliferation assays to measure the helper T cell activity, 5-week-old BALB/c mice were obtained from Jackson Laboratories. Mice were divided into three groups of ten and housed five per cage. Mice were aged 2 weeks prior to the start of the experiment. Food and water were removed 4 h prior to vaccination. Prior to bacterial dose, mice were fed 10 ml of 6% sodium bicarbonate using a pipette with disposable yellow tips. After a waiting period of 10 min, mice were fed the appropriate bacteria by the same method. One group of mice was fed 109 CFU of SL3261-pHART, a second group was fed 109 CFU of SL3261 as a control, a third group was fed PBS as a control. Mice were given doses on day 0, 14 and 28. The group of mice fed SL3261-pHART was given ampicillin in their drinking water, at a dose of 1 g/l in an attempt to increase the plasmid stability. One cage of control PBS fed mice were also given ampicillin in their drinking water.
2.7. Immunization of mice with Salmonella expressing the HIV-re6erse transcriptase epitopes Bacterial cultures were grown in 50 ml of LB medium with the addition of 100 mg/ml ampicillin. Cultures were shaken for 22 h at 37°C. Cells were pelleted at 5000 rpm for 10 min, then resuspended in 500 ml of PBS. Thirty-ml aliquots were placed in labeled 0.5-ml centrifuge tubes and kept on ice until used for inoculation of mice. Food and water were removed 4 h prior to vaccination. Prior to bacterial dose, mice were fed 10 ml of 6% sodium bicarbonate using a pipette. After a waiting period of 10 min, mice were fed the appropriate bacteria by the same method. One group of mice was fed 109 CFU of Salmonella containing a plasmid that contained an unrelated Sendai virus epitope in place of the HIV reverse transcriptase epitope as a control. A second group was fed 109 CFU of SL3261-pRT, a third group was fed SL3621-pRTlys. Mice were given doses on day 0, 14. All groups of mice were given ampicillin in their drinking water, at a dose of 1 g/l in an attempt to increase the plasmid stability.
738
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
2.8. IgA studies To monitor the mucosal IgA response, fresh feces were collected weekly from each cage of mice from the experiment above. Samples (0.5 g) were extracted with 1 ml of PBS for 1 – 2 h, with occasional vortexing. Samples were centrifuged at 14 000 rpm for 15 min. Supernatant was collected and stored at − 80°C until assayed. IgA ELISAs were performed as follows: Nunc 96 well polystyrene plates were pre-coated with 200 ng of recombinant reverse transcriptase in 50 ml of PBS. Plates were kept at 4°C in a moist container overnight. Antigen was poured off, and plates were washed three times with wash buffer (0.05% Tween-20 in PBS). Plates were then blocked for 2 h with 200 ml of 3% BSA in PBS at room temperature. Plates were then washed three times, and freshly diluted (1:10 and 1:50), recentrifuged samples were added at 100 ml per well. Plates were incubated at room temperature for 2.5 h, then washed four times. A 100 ml volume of goat-anti-mouse IgA labeled with HRP (Sigma) diluted 1:500 in blocking buffer was added to each well and incubated for 1 h. Plates were then washed four times and 100 ml of ABTS was added to each well. Reaction was stopped with 100 ml of oxalic acid, and plates were then read at 414 nm.
2.9. Proliferation assays Mice were sacrificed by cervical dislocation in sterile conditions. Spleens were removed, and splenocytes were isolated and resuspended in RPMI medium containing 2 mM L-glutamine, 50 mg/ml gentamycin, 50 mM bmercaptoethanol, and 10% fetal calf serum. One× 105 responder cells were added to each well in 96-well round bottomed plates, along with 1× 105 antigen presenting cells. Antigen presenting cells were isolated from nonvaccinated syngenic mice and treated with mitomycin C to inhibit proliferation. Cells were plated with RPMI medium, heat killed SL3261, and either 10 or 50 mg/ml of recombinant reverse transcriptase. All samples were done in triplicate. Plates were covered and incubated at 37°C with 5% CO2 for either 3 or 5 days. Wells were pulsed with 1 mCi of 3H-thymidine for 18 h. Cells were harvested onto glass filters using a Brandel M-24 cell harvester, and the incorporated radioactivity was measured using a Beckman LS6000SC.
(Gibco) with 10% heat inactivated FCS (Gibco), 50 ml/ml penicillin, 50 mg/ml streptomycin, 2 mM L-glutamine and 5× 10 − 5 M b–mercaptoethanol in 24-well cell culture plates (Becton Dickinson, UK). To stimulate T cell growth, ConA supernatant (0.5%), methyl a–D-mannopyranoside (50 mM) and HIV-1 reverse transcriptase peptide (0.1 mM) were added to the culture. Spleen cells were stimulated for 5 days. Target cells (RMA/S cells) were incubated with 1 mM HIV-1 reverse transcriptase peptide overnight and then were labeled with 0.2 mCi 51Cr for 45 min, washed and resuspended in RPMI, 10% FCS. Target cells were incubated with spleen cells at an effector:target ratio of 10:1. After 4 h incubation at 37°C, cell culture supernatants were collected and radioactivity was evaluated by a g counter. Specific CTL lysis was expressed as: percentage specific lysis= 100×(experimental release− spontaneous release)/(maximum release− spontaneous). Maximum release was induced by adding 5% Triton X-100 to wells containing 51Cr-labeled target cells only, while spontaneous release was measured from wells containing target cells in added medium alone. All the CTL assays were performed in triplicate.
3. Results and discussion
3.1. Expression of Lpp-OmpA-HIV re6erse transcriptase fusion protein A plasmid was constructed in which the HIV reverse transcriptase protein was inserted following an LppOmpA sequence, allowing for surface targeting of the HIV protein. Following construction, the recombinant
Xho I H ind III SP6 Xba I Amp lpp-om pA pH A R T 5012 pb
H ind III T7
2.10. CTL assay Bam HI
Spleens were removed from immunized mice and single cell suspensions were prepared. The red blood cells were depleted with lysis buffer (NH4Cl 8.3 g, KHCO3 1.0 g, EDTA 0.0372 g/l, pH 7.4). Approximately 5×106 cells were suspended in Iscoves medium
RT
Fig. 1. Recombinant plasmid pHART. This plasmid contains the Lpp-OmpA targetting sequence followed by the HIV-1 reverse transcriptase gene.
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
739
and outer membrane fractions were separated by sucrose gradient and analyzed by Western blot, a unique band was visible in the outer membrane fraction of the SL3261-pHART. This band was not present in the control SL3261 or the inner membrane fraction of the SL3261-pHART. Expression of the HIV-epitope fusion proteins was evaluated by Western blots of whole cell lysates using a polyclonal antibody preparation to the RT-peptide. A unique band with a molecular weight of 17 000, corresponding to the fusion protein, was clearly observed in the DH5a-RTpeptide and Salmonella-Rt-peptide cells but not in the control DH5a and Salmonella 3621 preparations. The fusion protein was over-expressed in IPTG induced cells.
3.2. Cell surface expression of Lpp-OmpA-HIV re6erse transcriptase CTL epitope fusion protein
Fig. 2. Western blot of E. coli-pHART. In the above figure, lane 1 contains molecular weight standards. Lane 2 contains whole cell lysates of induced E. coli DH5a-pHART. Lane 3 contains whole cell lysates of uninduced E. coli DH5a-pHART. Lane 4 contains whole cell lysates of control E. coli DH5a. The membrane is stained with anti-reverse transcriptase antibodies. Unique bands in lanes 2 and 3 are indicated with arrows.
plasmid, pHART (Fig. 1), was first electroporated into E. coli DH5a, as the Lpp-OmpA system is native to E. coli. Expression of the HIV reverse transcriptase protein was verified using Western blotting techniques (Fig. 2). The membrane was stained with an affinity purified polyclonal anti-HIV-1 reverse transcriptase antibody. Some non-specific binding is visible on the blot and may be due to the fact that the recombinant reverse transcriptase was purified from E. coli. Although background bands are present, a unique band, corresponding to a protein of the expected molecular weight of the reverse transcriptase fusion protein, is visible in the lane containing the DH5a-pHART whole cell lysate, and not in the control DH5a lane. Once reverse transcriptase was found to be expressed in the E. coli, the recombinant plasmid was then electroporated into an attenuated Aro− strain of Salmonella SL3261 [21]. Expression was verified using Western blotting techniques, as described above, of both whole cell lysates (Fig. 3), and membrane fractions (not shown). A unique band is visible in the lanes containing the recombinant SL3261 containing pHART that is not visible in control SL3261. The molecular weight of this protein corresponds to the expected molecular weight of the tripartite reverse transcriptase fusion protein. When inner
Trypsin digestion was used to demonstrate whether the reverse transcriptase peptide expressed on the cell surface. Salmonella cells containing either the pRT or pRTLys plasmid were digested with trypsin, along with the Salmonella control cells. Western blots were used to verify the cleavage of the RT-CTL epitope, using polyclonal antibodies to the reverse transcriptase peptide. After trypsin digestion, the density of the 17-kDa band was significantly reduced in the Salmonella cell lysates containing either the pRT or pRTLys plasmid, due to the cleavage of the RT peptide (data not shown), while no 17-kDa band was observed in the controls. The 1
2
3
64 k D
50 k D
36 k D
Fig. 3. Western blot of SL3261-pHART. In the above figure, lane 1 contains molecular weight standards. Lanes 2 and 4 contains whole cell lysates of SL3261-pHART. Lanes 3 and 5 contain whole cell lysates of control SL3261. Unique bands in lanes 2 and 4 are indicated with arrows.
740
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
can successfully infect the host [22]. In this case, high levels of anti-Salmonella antibodies may be serving to essentially lower the effective dose of the vaccination. A method for possibly overcoming this problem is to use slightly different attenuated bacterial strains for delivering the antigens to the mucosa with each successive dose. Overall, the results obtained for the reverse transcriptase vaccine were positive, indicating that measurable levels of antigen-specific IgA were produced in the appropriately vaccinated mice.
Fig. 4. Graph of reverse transcriptase specific IgA measured in SL3261-pHART vaccinated mice at 9 weeks. The above graph shows the data obtained from a reverse transcriptase specific IgA ELISA. Bar R represents the samples taken from SL3261-pHART vaccinated mice. Bar S represents the samples taken from SL3261 vaccinated mice. Bar C-1 represents the samples taken from control mice vaccinated with PBS. Bar C-2 represents the samples taken from control mice vaccinated with PBS, kept on ampicillin. The SL3261-pHART vaccinated mice had a much greater reverse transcriptase specific IgA response than the other mice in the study.
trypsin digestion showed that the fusion protein was very accessible to the action of the enzyme and therefore on the cell surface.
3.3. Fecal IgA le6els in BALB/c mice immunized with the intact HIV-RT construct A live vaccine consisting of SL3261-pHART was given orally to mice on days 0, 14, and 28. A second group of mice was orally vaccinated with control SL3261 that did not contain the recombinant plasmid. A third group of mice was vaccinated with PBS. Beginning with week 3, mice were assayed for secretory IgA immune responses to the HIV reverse transcriptase antigen. Fresh fecal samples were collected from all three groups of mice. Samples were extracted with PBS and assayed by ELISA for reverse transcriptase specific IgA. The results from an assay for reverse transcriptase specific fecal IgA 9 weeks after initial vaccination are shown in Fig. 4. These data indicate a strong reverse transcriptase-specific IgA response in the vaccinated mice, compared to the unvaccinated mice. Fig. 5 shows a graph of the reverse transcriptase-specific IgA response obtained in vaccinated mice over a period of 10 weeks. The response appears to decline over time, following a peak at 3 weeks from the first vaccination. This decrease in reverse transcriptase specific antibody could possibly be explained by a phenomenon often observed with live vaccine constructs. Additional doses administered over time to increase the immune response to the specific antigen are often less effective, as the immune system has raised antibodies against the bacterial vector, decreasing the number of bacterial cells that
3.4. Cellular proliferation assays in immunized BALB/c mice In the second series of experiments, assaying cellular proliferation in response to reverse transcriptase antigen, results were measured at 12 weeks following initial vaccination. The same mice were used for both the IgA studies and the proliferation assays. Results obtained from the 12-week studies are shown in Fig. 6. Here, splenocytes isolated from four out of the five mice vaccinated with the SL3261 containing the reverse transcriptase construct show a positive response to the two concentrations of recombinant reverse transcriptase antigen. All of the five mice show a positive response to the heat-killed Salmonella used to stimulate the splenocytes. This positive response to the Salmonella is expected, as the mice are exposed not only to the reverse transcriptase antigen but also to the Salmonella carrier. Splenocytes from all five mice showed background levels when incubated with RPMI-1640 medium alone. These results are very positive, indicating that when immunized with the live vaccine expressing the HIV reverse transcriptase antigen, these mice will develop a helper T cell response specific to the reverse transcriptase.
Fig. 5. Graph of reverse transcriptase specific IgA measured in SL3261-pHART mice over 10 weeks. The above graph shows a time course of the levels of reverse transcriptase specific IgA. The R bars represent the samples taken from SL3261-pHART vaccinated mice who were kept on ampicillin. The C-2 bars represent the samples taken from control mice, fed PBS, and kept on ampicillin. There is a definite production of reverse transcriptase specific IgA in the mice vaccinated with SL3261-pHART that is not seen in the control mice.
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
Fig. 6. Proliferation results seen at 12 weeks in mice vaccinated with SL3261-pHART. The above graph shows the data obtained from proliferation assays done 12 weeks after initial vaccinations. The mice were all vaccinated with SL3261-pHART. Isolated splenocytes were incubated with either RPMI-1640 (control medium) as shown in the first bar (RPMI), heat killed SL3261 (SL bar), 2 mg of reverse transcriptase (RT 2 mg bar) or 10 mg of reverse transcriptase (RT 10 mg bar). There appears to be proliferation occurring with both the SL and RT antigens at 12 weeks following vaccination. The above values were adjusted by subtracting the average values obtained for the control mice vaccinated with PBS.
The positive helper T cell response seen in the mice vaccinated with the SL3261-pHART live vaccine is very promising, and indicates, along with the positive results obtained with the reverse transcriptase specific IgA response, that this method of vaccination is worthy of further study.
3.5. Cytotoxic T lymphocyte response In order to determine whether the attenuated Salmonella with the HIV epitope-containing plasmids can cause the induction of a cytotoxic CD8 T cell response, mice were orally immunized twice at 2-week intervals and the cytotoxic response was examined. The results are shown in Fig. 7. The control group of mice, which were immunized with Salmonella containing an unrelated virus epitope plasmid construct, gave negative results. All the mice immunized with Salmonella containing the plasmids pRT and pRTLys gave positive results (Fig. 7). While for budgetary reasons the number of mice tested was small, in all cases the test mice gave a measurably higher degree of lysis of the target cells than did the controls. These results indicate that the live attenuated Salmonella vaccines expressing the HIV reverse transcriptase peptide can induce a cytotoxic T-cell response. Interestingly, the plasmid with one lysine residue between Lpp-OmpA and RT peptide did not stimulate a greater CTL response than the peptide lacking this lysine. Work is currently underway to address the issue of plasmid stability, by integrating the genetic construct
741
for the fusion protein into the bacterial chromosome. Outer membrane targeting of an antigenic peptide from HIV reverse transcriptase, rather than the whole enzyme, is also being explored as a method of inducing mucosal and cytotoxic T cell responses. This method of vaccine construction, using outer membrane targeting of antigens in attenuated bacterial vectors, could also be useful against other pathogens. Additional live vaccines against bacterial or viral pathogens could be developed using this Lpp-OmpA fusion construct to direct specific antigens to the bacterial outer membrane. Recent work using a similarly attenuated strain of S. typhimurium describes the oral genetic immunization of mice, where the plasmid DNA was transferred from the bacterial vector to the host [24]. The described genetic vaccination resulted in cellular and humoral immune responses against the recombinant genes. Efforts are underway to explore this concept of oral genetic vaccination using the Salmonella vector. Successful design of such oral live vaccines would be highly advantageous, due to the low cost of production and storage, and the ease of administration, particularly in the developing world.
Acknowledgements This work was supported in part by a grant from the Foundation for Reserach, Carson City, NV, and also in part by a NIH Biotechnology Training Grant, from the National Institutes of Health, Bethesda, MD. We thank G. Georgiou and C. Earhart for plasmids pTX101 and pTX215, Anneke Metz and Maria Svinth for assistance
Fig. 7. Cytotoxic T lymphocyte response. Target cells were incubated with spleen cells at an effector:target ratio of 10:1 for a 4-h period. Two groups of control mice were immunized with Salmonella containing plasmids expressing an unrelated peptide (S). Triplicate assays are shown for mice immunized with Salmonella expressing either the HIV-pRT (SL3621-RT) or HIVpRTLys (SL3621-RTL) epitopes.
742
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
with molecular biology and tissue culture, Maha Golestaneh, James Camp and Mark Nester for technical assistance, Victoria Kutilek and Susan Fullilove for assistance with preparation of the manuscript.
References [1] Johnston MI. HIV vaccines: problems and prospectives. Hosp Pract 1997;May 15:125–40. [2] Stott EJ, Schild GC. Strategies for AIDS vaccines. Br Soc Antimicrob Chemother 1996:185–198. [3] Haynes BF, Pantaleo G, Fauci AS. Towards and understanding of the correlates of protective immunity to HIV infection. Science 1996;271:324 – 8. [4] Haynes BF. Scientific and social issues of human immunodeficiency virus vaccine development. Science 1993;260:1279– 86. [5] Salk J, Bretscher PA, Salk PL, Clerici M, Shearer GM. A strategy for prophylactic vaccination against HIV. Science 1993;260:1270– 2. [6] Clerici M. Uninfected sex partners: saved by mucosal immunity to HIV? Vaccine Weekly 1997;June 2:7–8. [7] Curtiss R, III, Kelly SM, Gulig PA, Nakayama K. Selective delivery of antigens by recombinant bacteria. Curr Top Microbiol Immunol 1989;146:35–49. [8] Dougan G, Smith L, Heffron F. Live bacterial vaccines and their application as carriers for foreign antigens. Adv Vet Sci Comp Med 1989;33:271 – 300. [9] Charles C, Dougan G. Gene expression and the development of live enteric vaccines. Trends Biotechnol 1990;8:117–21. [10] Georgiou G, Poetschke HL, Stathopoulos C, Francisco JA. Practical applications of engineering gram-negative bacterial cell surfaces. Trends Biotechnol 1993;11:6–10. [11] Francisco JA, Earhart CF, Georgiou G. Transport and anchoring of b-lactamase to the external surface of Eschericia coli. Proc Natl Acad Sci USA 1992;89:2713–7. [12] Richins RD, Kaneva I, Mulchandani A, Chen W. Biodegredation of organophosphorus pesticides by surface-expressed organophosphorus hydrolase. Nat Biotechnol 1997;15:984–7. [13] Walker BD, Flexner C, Paradis TJ, Fuller TC, Hirsch MS, Schooley RT, Moss B. HIV-1 reverse transcriptase is a target for cytotoxic T lymphocytes in infected individuals. Science 1988;240:64 – 6.
.
[14] Lieberman J, Fabry JA, Kuo MC, Earl P, Moss B, Skolnik PR. Cytotoxic T lymphocytes from HIV-1 seropositive individuals recognize immunodominant epitopes in gp160 and reverse transcriptase. J Immunol 1992;148:2738– 47. [15] Rowland-Jones S, Sutton J, Ariyoshi K, Dong T, Gotch F, McAdam S, Whitby D, Sabally S, Gallimore A, Corrah T, Takiguchi M, Schultz T, McMichael A, Whittle H. HIV-specific cytotoxic T-cells in HIV-exposed but uninfected Gambian women. Nat Med 1995;1:59 – 64. [16] Hahn BH, Gonda MA, Shaw GM, Popovic M, Hoxie JA, Gallo RC, Wong-Staal F. Genomic diversity of the acquired immune deficiency syndrome virus HTLV-III: different viruses exhibit greatest divergence in their envelope genes. Proc Natl Acad Sci USA 1985;82:14813– 7. [17] Clavel F, Guetard D, Brun-Vezinet F, Charmaret S, Rey MA, Santos-Ferreira M, Laurent AG, Daugeuet C, Katlama C, Rouziouz C. Isolation of a new human retrovirus from West African patients with AIDS. Science 1986;233:343 – 6. [18] Guyader M, Emerman M, Sonigo P, Clavel F, Montagnier L, Alizon M. Genome organization and transactivation of the human immunodeficiency virus type 2. Nature 1987;326:662–9. [19] Schultz AM. Changing paradigms for an HIV vaccine. In: Cohen S, Shafferman A, editors. Novel Strategies in Design and Production of Vaccines. New York: Plenum, 1996:79–90. [20] Hilleman MR. Strategies for the achievement of prophylactic vaccination against HIV. Antibiotics Chemother 1996;48:161– 72. [21] Hosmalin A, Clerici M, Houghten R, Pendleton C, flexner C, Lucey DR, Moss B, Germain RN, Shearer GM. An epitope in human immunodeficiency virus 1 reverse transcriptase recognized by both mouse and human cytotoxic T lymphocytes. Proc Natl Acad Sci USA 1990;87:2344. [22] Bergmann CC, Yao Q, Ho CK, Buckwold SL. Flanking residues alter antigenicity and immunogenicity of multi-unit CTL epitopes. J Immunol 1996;157:3242. [23] Verma NK, Ziegler HK, Wilson M, Khan M, Safley S, Stocker BAD, Schoolnik GLK. Delivery of Class I and Class II MHCrestricted T-cell epitope of listeriolysin of listeria monocytogenes by attenuated Salmonella. Vaccine 1995;13:142. [24] Darji A, Guzman CA, Gerstel B, Wachholz P, Timmis KN, Wehland J, Chakraborty T, Weiss S. Oral somatic transgene vaccination using attenuated S. typhimurium. Cell 1997;91:765– 75.
Potential live vaccines for HIV Mary Susan Burnett 1, Ning Wang, Matthias Hofmann 2, G. Barrie Kitto * Department of Chemistry and Biochemistry, Institute of Molecular and Cellular Biology, The Uni6ersity of Texas at Austin, Austin, TX 78712, USA Received 23 June 1999; received in revised form 31 July 2000; accepted 31 July 2000
Abstract Potential live vaccines for HIV were developed using an Lpp-OmpA system to target an HIV antigen, reverse transcriptase, or an immunodominant epitope of this enzyme, to the outer membrane of an attenuated strain of Salmonella SL3261. These live vaccines were administered orally to mice, and fecal IgA and helper T cell responses were measured. Results indicated a fecal IgA response specific to HIV reverse transcriptase, as well as a reverse transcriptase-specific helper T cell response, as measured by proliferation assays. Additionally, tests with the epitope vaccines showed a selective cytotoxic CD8 T cell response. These results suggest that this method of antigen targeting to the outer membrane of attenuated bacterial vectors is very promising not only for HIV vaccine development, but also for antigens from other viral or bacterial pathogens, which could be inserted into the Lpp-OmpA protein construct, to elicit mucosal IgA and T cell responses. © 2000 Published by Elsevier Science Ltd. Keywords: Salmonella; Reverse transcriptase; Mucosal IgA
1. Introduction Despite the recent advances with antiviral drugs for treating HIV-positive patients, the new triple drug therapy is not a global solution. While these new antiviral drugs are having very positive results with many patients, they do not work for everyone. Additionally, the treatment program is a very demanding regimen, requiring patients to take up to 70 pills per day [1]. Not all patients can adhere to such a strict schedule, which basically excludes them from treatment. Perhaps the greatest problem with these drugs is that they are not available to patients in the developing world. The annual cost for treatment with these drugs may total $30 000. In developing countries, where the vast majority of new cases are diagnosed, this price tag makes treatment essentially unavailable. Historically, vaccines are the most cost-effective method of handling a disease * Corresponding author. Tel.: +1-512-4713279; fax: + 1-5124718696. E-mail address: [email protected] (G. Barrie Kitto). 1 Present address: Cardiovascular Research Foundation, Washington Hospital Center, Washington, DC 20010, USA. 2 Present address: Institute for Cell Biology, Division of Immunology, University of Tuebingen, 72076 Tuebingen, Germany.
of this magnitude. When one considers the estimation that 40 million people will be infected by the year 2000 [2], the development of an effective vaccine for HIV is imperative. In designing a vaccine for HIV it is currently thought that humoral and cell-mediated immune responses will both play a role in providing protection against viral infection [3–5]. Recent results have indicated the presence of HIV specific IgA in cervical swabs or urine samples from exposed seronegative individuals [6]. These individuals had participated in unprotected sexual intercourse with their infected partners for periods ranging from 2 to 8 years in length and remained uninfected with HIV. This data suggests that the HIV-specific mucosal IgA may be affording protection from viral infection. For this reason, attenuated Salmonella are an attractive vehicle for delivering HIV antigens to the immune system, as such bacteria have been shown to elicit humoral, cell-mediated and mucosal responses to expressed antigens, as well as to the wild type of their species [7–9]. While it has traditionally been difficult to elicit an immune response at the mucosal surface, the use of Salmonella as a live vaccine vector is ideal because it can be administered orally, allowing the Salmonella to infect the mucosal membranes of the gastrointestinal tract
0264-410X/00/$ - see front matter © 2000 Published by Elsevier Science Ltd. PII: S 0 2 6 4 - 4 1 0 X ( 0 0 ) 0 0 3 0 3 - 0
736
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
and reproductive tracts, entry points for the majority of HIV infections. An oral live vaccine would not only be inexpensive to produce and store, but also easy to administer, particularly in developing countries. In an attempt to elicit a stronger immune response than that obtained with previous Salmonella vaccines, we have explored the method of surface targeting for delivering the HIV antigens to the immune system [10]. Plasmids were constructed that resulted in the expression of HIV antigens fused to an Lpp-OmpA sequence [11]. This method of expression resulted in the HIV antigens being targeted to the bacterial outer membrane. Recent results from Richins et al. [12] describe the surface-expression of organophosphorus hydrolase using this same Lpp-OmpA fusion system. Several factors led to the decision to use the HIV antigen reverse transcriptase in this vaccine construct. HIV reverse transcriptase has been shown to elicit cytotoxic T lymphocytes in both infected patients [13,14] and exposed, but seronegative, individuals [15]. Since a strong cell-mediated response is desired in an HIV vaccine, this fact made reverse transcriptase a good candidate. Additionally, the HIV-1 pol gene, which encodes the reverse transcriptase protein, has been found to be more highly conserved than other HIV genes among different primary isolates [16]. Other studies have determined that pol-specific humoral cross-reactivity exists between HIV-1 and HIV2, as well as considerable sequence homology [17,18]. Additionally, recent work has focused on using internal viral proteins as antigens, as opposed to envelope proteins, in the hopes of inducing a cell-mediated response [19,20]. In this study we describe the construction of this live vaccine, and the testing of this vaccine in mice. BALB/c mice were inoculated orally with recombinant Salmonella and the immune responses were monitored. Specifically, mucosal IgA levels were measured, as was helper T cell proliferation. We also constructed an Lpp-OmpA fusion protein system carrying a class Irestricted epitope consisting of residues 203 – 219 from HIV-1 reverse transcriptase. This relatively conserved HIV-1 reverse transcriptase epitope can be recognized by murine and human cytotoxic T cells [21]. Our reasons for doing so were as follows: this construct was more likely to facilitate external expression on the surface of Salmonella and such a construct would allow for ready manipulation of the peptide sequence to determine the most effective vaccine. Recent studies have shown that the flanking residues of CTL epitopes influence the immune response [22,23]. We therefore established a plasmid construct with one lysine residue between the LppOmpA gene and the RT epitope gene as a vaccine candidate.
2. Materials and method
2.1. Recombinant DNA techniques Using polymerase chain reaction techniques, new restriction enzyme sites were engineered on the ends of the reverse transcriptase gene (supplied from NIH AIDS Reagent Bank, in plasmid pKRT2). The first primer, 5% GCG AAG CTT CAT GGC CAT TAG CCC TAT TGA 3%, resulted in the addition of a HindIII site at the 5% end of the reverse transcriptase gene. The second primer, 5% GCC GCT GGA TCC TAG TAT TTT CCT GAT TCC 3% resulted in the addition of a BamHI site at the 3% end of the reverse transcriptase gene. This amplified 1.7-Kb gene was then inserted into the plasmid pSP72 (Promega), which confers ampicillin resistance. The plasmid pTX101, containing the Lpp-OmpA sequences required for outer membrane targeting, along with the lpp and lac promoters, was obtained from Charles Earhart (University of Texas at Austin). This sequence was first inserted into a pCRII vector (Invitrogen), following amplification using the polymerase chain reaction. The desired sequence was then amplified using polymerase chain reaction, to contain HindIII sites on either end, using the following primers, 5% ACA TGC GAA GCT TCA GTT GTC CGG ACG AGT 3% and 5% CGC CTG AAG CTT TAT CTA AGA TGA ATC CGA TG 3%. The 850-bp Lpp-OmpA sequences, along with the lpp and lac promoters, were then inserted into the recombinant plasmid pSP72 containing the HIV reverse transcriptase gene. The resulting recombinant plasmid was called pHART. A plasmid containing the Lpp-OmpA sequence followed by the HIV tat gene was also constructed, and is currently still under study.
2.2. Recombinant plasmid for HIV-re6erse transcriptase epitope A DNA fragment encoding the selected HIV reverse transcriptase peptide was amplified by PCR using primers: 5%-GAATTCGGAAAGAAATCTGTACCGAAATG-3% and 5%-GGATCCTACGGACCGATCTTGGAG-3%. This amplified 100-bp gene was then inserted into a plasmid containing the Lpp-OmpA sequence required for outer membrane targeting, along with the lpp and lac promoters, as described previously. The resulting recombinant plasmid was called pRT. Another recombinant plasmid with one lysine residue inserted between Lpp-OmpA sequence and reverse transcriptase peptide sequence was constructed with the same method and the recombinant plasmid was called pRTLys.
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
2.3. Fusion protein expression The recombinant plasmid pHART was then electroporated into E. coli DH5a, and expression of the HIV reverse transcriptase protein was verified using Western blotting techniques of whole cell lysates. Cells were grown overnight with 500 mg/ml ampicillin where necessary. Cells were pelleted in the presence of Complete™ Protease Inhibitor (Boehringer Mannheim), then boiled with 2X Tris – Glycine sample buffer (Novex) containing DTT for 5 min. Samples were loaded onto a Novex 10% Tris – Glycine SDS gel. Samples were transferred from gel to PVDF membrane, and blocked using 3% BSA in PBS. Affinity purified polyclonal anti reverse transcriptase antibody was prepared by James Caras. Western blotting was performed using the chemiluminescence SuperSignal system (Pierce). The pHART plasmid was then used to transform the Aro− SL3261 (generously donated by B. Stocker). Expression of reverse transcriptase was verified using Western blotting, as described above, of both whole cell lysates and membrane fractions. Bacterial inner and outer membranes were separated using a sucrose gradient.
2.4. Fusion protein expression for the HIV-re6erse transcriptase epitope The recombinant plasmids pRT and pRTLys were electroporated into E. coli DH5a and Salmonella Aro− SL3621 and the expression of the HIV reverse transcriptase peptide was verified using Western blotting techniques with whole cell lysates. Cells were grown overnight with 100 mg/ml ampicillin. Cells were pelleted and then boiled with 2X Tris – Glycine sample buffer (Novex) for 5 min. Samples were loaded onto a Novex 10% Tris–Glycine SDS gel, then transferred from gel to PVDF membrane and blocked using 1% BSA in PBS. Polyclonal anti-reverse transcriptase peptide antisera were purified by protein A column and Western blotting was performed as described above.
2.5. Trypsin digestion of Salmonella to determine cell surface expression Salmonella cells containing pRT and pRTLys plasmid were grown overnight. One ml of cells was pelleted at 14 000 rpm for 5 min at room temperature. The cells were resuspended in 50 ml of 25 mM HEPES buffer (pH 8.0) and incubated on ice for 30 min. Trypsin was added to obtain a final concentration of 0.1 mg/ml and the mixture was incubated at 37°C for 5 min. At this time, 1% Triton X-100 was added and the reaction was terminated by addition of 55 ml 2X SDS Tris–Glycine sample buffer. The samples were then boiled for 5 min. Samples were loaded onto a Novex 10% Tris–Glycine SDS gel. Samples were then
737
transferred from the gel to PVDF membrane. Western blotting was performed.
2.6. Immunization of mice with Salmonella expressing the intact HIV re6erse transcriptase Bacterial cultures were grown in 50 ml of LB medium with or without 500 mg/ml ampicillin as appropriate. Cultures were shaken for 22 h at 37°C. Cells were pelleted at 5000 rpm for 10 min, then resuspended in 500 ml of PBS. Thirty-ml aliquots were placed in labeled 0.5-ml centrifuge tubes and kept on ice until feeding. For the IgA studies and proliferation assays to measure the helper T cell activity, 5-week-old BALB/c mice were obtained from Jackson Laboratories. Mice were divided into three groups of ten and housed five per cage. Mice were aged 2 weeks prior to the start of the experiment. Food and water were removed 4 h prior to vaccination. Prior to bacterial dose, mice were fed 10 ml of 6% sodium bicarbonate using a pipette with disposable yellow tips. After a waiting period of 10 min, mice were fed the appropriate bacteria by the same method. One group of mice was fed 109 CFU of SL3261-pHART, a second group was fed 109 CFU of SL3261 as a control, a third group was fed PBS as a control. Mice were given doses on day 0, 14 and 28. The group of mice fed SL3261-pHART was given ampicillin in their drinking water, at a dose of 1 g/l in an attempt to increase the plasmid stability. One cage of control PBS fed mice were also given ampicillin in their drinking water.
2.7. Immunization of mice with Salmonella expressing the HIV-re6erse transcriptase epitopes Bacterial cultures were grown in 50 ml of LB medium with the addition of 100 mg/ml ampicillin. Cultures were shaken for 22 h at 37°C. Cells were pelleted at 5000 rpm for 10 min, then resuspended in 500 ml of PBS. Thirty-ml aliquots were placed in labeled 0.5-ml centrifuge tubes and kept on ice until used for inoculation of mice. Food and water were removed 4 h prior to vaccination. Prior to bacterial dose, mice were fed 10 ml of 6% sodium bicarbonate using a pipette. After a waiting period of 10 min, mice were fed the appropriate bacteria by the same method. One group of mice was fed 109 CFU of Salmonella containing a plasmid that contained an unrelated Sendai virus epitope in place of the HIV reverse transcriptase epitope as a control. A second group was fed 109 CFU of SL3261-pRT, a third group was fed SL3621-pRTlys. Mice were given doses on day 0, 14. All groups of mice were given ampicillin in their drinking water, at a dose of 1 g/l in an attempt to increase the plasmid stability.
738
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
2.8. IgA studies To monitor the mucosal IgA response, fresh feces were collected weekly from each cage of mice from the experiment above. Samples (0.5 g) were extracted with 1 ml of PBS for 1 – 2 h, with occasional vortexing. Samples were centrifuged at 14 000 rpm for 15 min. Supernatant was collected and stored at − 80°C until assayed. IgA ELISAs were performed as follows: Nunc 96 well polystyrene plates were pre-coated with 200 ng of recombinant reverse transcriptase in 50 ml of PBS. Plates were kept at 4°C in a moist container overnight. Antigen was poured off, and plates were washed three times with wash buffer (0.05% Tween-20 in PBS). Plates were then blocked for 2 h with 200 ml of 3% BSA in PBS at room temperature. Plates were then washed three times, and freshly diluted (1:10 and 1:50), recentrifuged samples were added at 100 ml per well. Plates were incubated at room temperature for 2.5 h, then washed four times. A 100 ml volume of goat-anti-mouse IgA labeled with HRP (Sigma) diluted 1:500 in blocking buffer was added to each well and incubated for 1 h. Plates were then washed four times and 100 ml of ABTS was added to each well. Reaction was stopped with 100 ml of oxalic acid, and plates were then read at 414 nm.
2.9. Proliferation assays Mice were sacrificed by cervical dislocation in sterile conditions. Spleens were removed, and splenocytes were isolated and resuspended in RPMI medium containing 2 mM L-glutamine, 50 mg/ml gentamycin, 50 mM bmercaptoethanol, and 10% fetal calf serum. One× 105 responder cells were added to each well in 96-well round bottomed plates, along with 1× 105 antigen presenting cells. Antigen presenting cells were isolated from nonvaccinated syngenic mice and treated with mitomycin C to inhibit proliferation. Cells were plated with RPMI medium, heat killed SL3261, and either 10 or 50 mg/ml of recombinant reverse transcriptase. All samples were done in triplicate. Plates were covered and incubated at 37°C with 5% CO2 for either 3 or 5 days. Wells were pulsed with 1 mCi of 3H-thymidine for 18 h. Cells were harvested onto glass filters using a Brandel M-24 cell harvester, and the incorporated radioactivity was measured using a Beckman LS6000SC.
(Gibco) with 10% heat inactivated FCS (Gibco), 50 ml/ml penicillin, 50 mg/ml streptomycin, 2 mM L-glutamine and 5× 10 − 5 M b–mercaptoethanol in 24-well cell culture plates (Becton Dickinson, UK). To stimulate T cell growth, ConA supernatant (0.5%), methyl a–D-mannopyranoside (50 mM) and HIV-1 reverse transcriptase peptide (0.1 mM) were added to the culture. Spleen cells were stimulated for 5 days. Target cells (RMA/S cells) were incubated with 1 mM HIV-1 reverse transcriptase peptide overnight and then were labeled with 0.2 mCi 51Cr for 45 min, washed and resuspended in RPMI, 10% FCS. Target cells were incubated with spleen cells at an effector:target ratio of 10:1. After 4 h incubation at 37°C, cell culture supernatants were collected and radioactivity was evaluated by a g counter. Specific CTL lysis was expressed as: percentage specific lysis= 100×(experimental release− spontaneous release)/(maximum release− spontaneous). Maximum release was induced by adding 5% Triton X-100 to wells containing 51Cr-labeled target cells only, while spontaneous release was measured from wells containing target cells in added medium alone. All the CTL assays were performed in triplicate.
3. Results and discussion
3.1. Expression of Lpp-OmpA-HIV re6erse transcriptase fusion protein A plasmid was constructed in which the HIV reverse transcriptase protein was inserted following an LppOmpA sequence, allowing for surface targeting of the HIV protein. Following construction, the recombinant
Xho I H ind III SP6 Xba I Amp lpp-om pA pH A R T 5012 pb
H ind III T7
2.10. CTL assay Bam HI
Spleens were removed from immunized mice and single cell suspensions were prepared. The red blood cells were depleted with lysis buffer (NH4Cl 8.3 g, KHCO3 1.0 g, EDTA 0.0372 g/l, pH 7.4). Approximately 5×106 cells were suspended in Iscoves medium
RT
Fig. 1. Recombinant plasmid pHART. This plasmid contains the Lpp-OmpA targetting sequence followed by the HIV-1 reverse transcriptase gene.
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
739
and outer membrane fractions were separated by sucrose gradient and analyzed by Western blot, a unique band was visible in the outer membrane fraction of the SL3261-pHART. This band was not present in the control SL3261 or the inner membrane fraction of the SL3261-pHART. Expression of the HIV-epitope fusion proteins was evaluated by Western blots of whole cell lysates using a polyclonal antibody preparation to the RT-peptide. A unique band with a molecular weight of 17 000, corresponding to the fusion protein, was clearly observed in the DH5a-RTpeptide and Salmonella-Rt-peptide cells but not in the control DH5a and Salmonella 3621 preparations. The fusion protein was over-expressed in IPTG induced cells.
3.2. Cell surface expression of Lpp-OmpA-HIV re6erse transcriptase CTL epitope fusion protein
Fig. 2. Western blot of E. coli-pHART. In the above figure, lane 1 contains molecular weight standards. Lane 2 contains whole cell lysates of induced E. coli DH5a-pHART. Lane 3 contains whole cell lysates of uninduced E. coli DH5a-pHART. Lane 4 contains whole cell lysates of control E. coli DH5a. The membrane is stained with anti-reverse transcriptase antibodies. Unique bands in lanes 2 and 3 are indicated with arrows.
plasmid, pHART (Fig. 1), was first electroporated into E. coli DH5a, as the Lpp-OmpA system is native to E. coli. Expression of the HIV reverse transcriptase protein was verified using Western blotting techniques (Fig. 2). The membrane was stained with an affinity purified polyclonal anti-HIV-1 reverse transcriptase antibody. Some non-specific binding is visible on the blot and may be due to the fact that the recombinant reverse transcriptase was purified from E. coli. Although background bands are present, a unique band, corresponding to a protein of the expected molecular weight of the reverse transcriptase fusion protein, is visible in the lane containing the DH5a-pHART whole cell lysate, and not in the control DH5a lane. Once reverse transcriptase was found to be expressed in the E. coli, the recombinant plasmid was then electroporated into an attenuated Aro− strain of Salmonella SL3261 [21]. Expression was verified using Western blotting techniques, as described above, of both whole cell lysates (Fig. 3), and membrane fractions (not shown). A unique band is visible in the lanes containing the recombinant SL3261 containing pHART that is not visible in control SL3261. The molecular weight of this protein corresponds to the expected molecular weight of the tripartite reverse transcriptase fusion protein. When inner
Trypsin digestion was used to demonstrate whether the reverse transcriptase peptide expressed on the cell surface. Salmonella cells containing either the pRT or pRTLys plasmid were digested with trypsin, along with the Salmonella control cells. Western blots were used to verify the cleavage of the RT-CTL epitope, using polyclonal antibodies to the reverse transcriptase peptide. After trypsin digestion, the density of the 17-kDa band was significantly reduced in the Salmonella cell lysates containing either the pRT or pRTLys plasmid, due to the cleavage of the RT peptide (data not shown), while no 17-kDa band was observed in the controls. The 1
2
3
64 k D
50 k D
36 k D
Fig. 3. Western blot of SL3261-pHART. In the above figure, lane 1 contains molecular weight standards. Lanes 2 and 4 contains whole cell lysates of SL3261-pHART. Lanes 3 and 5 contain whole cell lysates of control SL3261. Unique bands in lanes 2 and 4 are indicated with arrows.
740
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
can successfully infect the host [22]. In this case, high levels of anti-Salmonella antibodies may be serving to essentially lower the effective dose of the vaccination. A method for possibly overcoming this problem is to use slightly different attenuated bacterial strains for delivering the antigens to the mucosa with each successive dose. Overall, the results obtained for the reverse transcriptase vaccine were positive, indicating that measurable levels of antigen-specific IgA were produced in the appropriately vaccinated mice.
Fig. 4. Graph of reverse transcriptase specific IgA measured in SL3261-pHART vaccinated mice at 9 weeks. The above graph shows the data obtained from a reverse transcriptase specific IgA ELISA. Bar R represents the samples taken from SL3261-pHART vaccinated mice. Bar S represents the samples taken from SL3261 vaccinated mice. Bar C-1 represents the samples taken from control mice vaccinated with PBS. Bar C-2 represents the samples taken from control mice vaccinated with PBS, kept on ampicillin. The SL3261-pHART vaccinated mice had a much greater reverse transcriptase specific IgA response than the other mice in the study.
trypsin digestion showed that the fusion protein was very accessible to the action of the enzyme and therefore on the cell surface.
3.3. Fecal IgA le6els in BALB/c mice immunized with the intact HIV-RT construct A live vaccine consisting of SL3261-pHART was given orally to mice on days 0, 14, and 28. A second group of mice was orally vaccinated with control SL3261 that did not contain the recombinant plasmid. A third group of mice was vaccinated with PBS. Beginning with week 3, mice were assayed for secretory IgA immune responses to the HIV reverse transcriptase antigen. Fresh fecal samples were collected from all three groups of mice. Samples were extracted with PBS and assayed by ELISA for reverse transcriptase specific IgA. The results from an assay for reverse transcriptase specific fecal IgA 9 weeks after initial vaccination are shown in Fig. 4. These data indicate a strong reverse transcriptase-specific IgA response in the vaccinated mice, compared to the unvaccinated mice. Fig. 5 shows a graph of the reverse transcriptase-specific IgA response obtained in vaccinated mice over a period of 10 weeks. The response appears to decline over time, following a peak at 3 weeks from the first vaccination. This decrease in reverse transcriptase specific antibody could possibly be explained by a phenomenon often observed with live vaccine constructs. Additional doses administered over time to increase the immune response to the specific antigen are often less effective, as the immune system has raised antibodies against the bacterial vector, decreasing the number of bacterial cells that
3.4. Cellular proliferation assays in immunized BALB/c mice In the second series of experiments, assaying cellular proliferation in response to reverse transcriptase antigen, results were measured at 12 weeks following initial vaccination. The same mice were used for both the IgA studies and the proliferation assays. Results obtained from the 12-week studies are shown in Fig. 6. Here, splenocytes isolated from four out of the five mice vaccinated with the SL3261 containing the reverse transcriptase construct show a positive response to the two concentrations of recombinant reverse transcriptase antigen. All of the five mice show a positive response to the heat-killed Salmonella used to stimulate the splenocytes. This positive response to the Salmonella is expected, as the mice are exposed not only to the reverse transcriptase antigen but also to the Salmonella carrier. Splenocytes from all five mice showed background levels when incubated with RPMI-1640 medium alone. These results are very positive, indicating that when immunized with the live vaccine expressing the HIV reverse transcriptase antigen, these mice will develop a helper T cell response specific to the reverse transcriptase.
Fig. 5. Graph of reverse transcriptase specific IgA measured in SL3261-pHART mice over 10 weeks. The above graph shows a time course of the levels of reverse transcriptase specific IgA. The R bars represent the samples taken from SL3261-pHART vaccinated mice who were kept on ampicillin. The C-2 bars represent the samples taken from control mice, fed PBS, and kept on ampicillin. There is a definite production of reverse transcriptase specific IgA in the mice vaccinated with SL3261-pHART that is not seen in the control mice.
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
Fig. 6. Proliferation results seen at 12 weeks in mice vaccinated with SL3261-pHART. The above graph shows the data obtained from proliferation assays done 12 weeks after initial vaccinations. The mice were all vaccinated with SL3261-pHART. Isolated splenocytes were incubated with either RPMI-1640 (control medium) as shown in the first bar (RPMI), heat killed SL3261 (SL bar), 2 mg of reverse transcriptase (RT 2 mg bar) or 10 mg of reverse transcriptase (RT 10 mg bar). There appears to be proliferation occurring with both the SL and RT antigens at 12 weeks following vaccination. The above values were adjusted by subtracting the average values obtained for the control mice vaccinated with PBS.
The positive helper T cell response seen in the mice vaccinated with the SL3261-pHART live vaccine is very promising, and indicates, along with the positive results obtained with the reverse transcriptase specific IgA response, that this method of vaccination is worthy of further study.
3.5. Cytotoxic T lymphocyte response In order to determine whether the attenuated Salmonella with the HIV epitope-containing plasmids can cause the induction of a cytotoxic CD8 T cell response, mice were orally immunized twice at 2-week intervals and the cytotoxic response was examined. The results are shown in Fig. 7. The control group of mice, which were immunized with Salmonella containing an unrelated virus epitope plasmid construct, gave negative results. All the mice immunized with Salmonella containing the plasmids pRT and pRTLys gave positive results (Fig. 7). While for budgetary reasons the number of mice tested was small, in all cases the test mice gave a measurably higher degree of lysis of the target cells than did the controls. These results indicate that the live attenuated Salmonella vaccines expressing the HIV reverse transcriptase peptide can induce a cytotoxic T-cell response. Interestingly, the plasmid with one lysine residue between Lpp-OmpA and RT peptide did not stimulate a greater CTL response than the peptide lacking this lysine. Work is currently underway to address the issue of plasmid stability, by integrating the genetic construct
741
for the fusion protein into the bacterial chromosome. Outer membrane targeting of an antigenic peptide from HIV reverse transcriptase, rather than the whole enzyme, is also being explored as a method of inducing mucosal and cytotoxic T cell responses. This method of vaccine construction, using outer membrane targeting of antigens in attenuated bacterial vectors, could also be useful against other pathogens. Additional live vaccines against bacterial or viral pathogens could be developed using this Lpp-OmpA fusion construct to direct specific antigens to the bacterial outer membrane. Recent work using a similarly attenuated strain of S. typhimurium describes the oral genetic immunization of mice, where the plasmid DNA was transferred from the bacterial vector to the host [24]. The described genetic vaccination resulted in cellular and humoral immune responses against the recombinant genes. Efforts are underway to explore this concept of oral genetic vaccination using the Salmonella vector. Successful design of such oral live vaccines would be highly advantageous, due to the low cost of production and storage, and the ease of administration, particularly in the developing world.
Acknowledgements This work was supported in part by a grant from the Foundation for Reserach, Carson City, NV, and also in part by a NIH Biotechnology Training Grant, from the National Institutes of Health, Bethesda, MD. We thank G. Georgiou and C. Earhart for plasmids pTX101 and pTX215, Anneke Metz and Maria Svinth for assistance
Fig. 7. Cytotoxic T lymphocyte response. Target cells were incubated with spleen cells at an effector:target ratio of 10:1 for a 4-h period. Two groups of control mice were immunized with Salmonella containing plasmids expressing an unrelated peptide (S). Triplicate assays are shown for mice immunized with Salmonella expressing either the HIV-pRT (SL3621-RT) or HIVpRTLys (SL3621-RTL) epitopes.
742
M.S. Burnett et al. / Vaccine 19 (2001) 735–742
with molecular biology and tissue culture, Maha Golestaneh, James Camp and Mark Nester for technical assistance, Victoria Kutilek and Susan Fullilove for assistance with preparation of the manuscript.
References [1] Johnston MI. HIV vaccines: problems and prospectives. Hosp Pract 1997;May 15:125–40. [2] Stott EJ, Schild GC. Strategies for AIDS vaccines. Br Soc Antimicrob Chemother 1996:185–198. [3] Haynes BF, Pantaleo G, Fauci AS. Towards and understanding of the correlates of protective immunity to HIV infection. Science 1996;271:324 – 8. [4] Haynes BF. Scientific and social issues of human immunodeficiency virus vaccine development. Science 1993;260:1279– 86. [5] Salk J, Bretscher PA, Salk PL, Clerici M, Shearer GM. A strategy for prophylactic vaccination against HIV. Science 1993;260:1270– 2. [6] Clerici M. Uninfected sex partners: saved by mucosal immunity to HIV? Vaccine Weekly 1997;June 2:7–8. [7] Curtiss R, III, Kelly SM, Gulig PA, Nakayama K. Selective delivery of antigens by recombinant bacteria. Curr Top Microbiol Immunol 1989;146:35–49. [8] Dougan G, Smith L, Heffron F. Live bacterial vaccines and their application as carriers for foreign antigens. Adv Vet Sci Comp Med 1989;33:271 – 300. [9] Charles C, Dougan G. Gene expression and the development of live enteric vaccines. Trends Biotechnol 1990;8:117–21. [10] Georgiou G, Poetschke HL, Stathopoulos C, Francisco JA. Practical applications of engineering gram-negative bacterial cell surfaces. Trends Biotechnol 1993;11:6–10. [11] Francisco JA, Earhart CF, Georgiou G. Transport and anchoring of b-lactamase to the external surface of Eschericia coli. Proc Natl Acad Sci USA 1992;89:2713–7. [12] Richins RD, Kaneva I, Mulchandani A, Chen W. Biodegredation of organophosphorus pesticides by surface-expressed organophosphorus hydrolase. Nat Biotechnol 1997;15:984–7. [13] Walker BD, Flexner C, Paradis TJ, Fuller TC, Hirsch MS, Schooley RT, Moss B. HIV-1 reverse transcriptase is a target for cytotoxic T lymphocytes in infected individuals. Science 1988;240:64 – 6.
.
[14] Lieberman J, Fabry JA, Kuo MC, Earl P, Moss B, Skolnik PR. Cytotoxic T lymphocytes from HIV-1 seropositive individuals recognize immunodominant epitopes in gp160 and reverse transcriptase. J Immunol 1992;148:2738– 47. [15] Rowland-Jones S, Sutton J, Ariyoshi K, Dong T, Gotch F, McAdam S, Whitby D, Sabally S, Gallimore A, Corrah T, Takiguchi M, Schultz T, McMichael A, Whittle H. HIV-specific cytotoxic T-cells in HIV-exposed but uninfected Gambian women. Nat Med 1995;1:59 – 64. [16] Hahn BH, Gonda MA, Shaw GM, Popovic M, Hoxie JA, Gallo RC, Wong-Staal F. Genomic diversity of the acquired immune deficiency syndrome virus HTLV-III: different viruses exhibit greatest divergence in their envelope genes. Proc Natl Acad Sci USA 1985;82:14813– 7. [17] Clavel F, Guetard D, Brun-Vezinet F, Charmaret S, Rey MA, Santos-Ferreira M, Laurent AG, Daugeuet C, Katlama C, Rouziouz C. Isolation of a new human retrovirus from West African patients with AIDS. Science 1986;233:343 – 6. [18] Guyader M, Emerman M, Sonigo P, Clavel F, Montagnier L, Alizon M. Genome organization and transactivation of the human immunodeficiency virus type 2. Nature 1987;326:662–9. [19] Schultz AM. Changing paradigms for an HIV vaccine. In: Cohen S, Shafferman A, editors. Novel Strategies in Design and Production of Vaccines. New York: Plenum, 1996:79–90. [20] Hilleman MR. Strategies for the achievement of prophylactic vaccination against HIV. Antibiotics Chemother 1996;48:161– 72. [21] Hosmalin A, Clerici M, Houghten R, Pendleton C, flexner C, Lucey DR, Moss B, Germain RN, Shearer GM. An epitope in human immunodeficiency virus 1 reverse transcriptase recognized by both mouse and human cytotoxic T lymphocytes. Proc Natl Acad Sci USA 1990;87:2344. [22] Bergmann CC, Yao Q, Ho CK, Buckwold SL. Flanking residues alter antigenicity and immunogenicity of multi-unit CTL epitopes. J Immunol 1996;157:3242. [23] Verma NK, Ziegler HK, Wilson M, Khan M, Safley S, Stocker BAD, Schoolnik GLK. Delivery of Class I and Class II MHCrestricted T-cell epitope of listeriolysin of listeria monocytogenes by attenuated Salmonella. Vaccine 1995;13:142. [24] Darji A, Guzman CA, Gerstel B, Wachholz P, Timmis KN, Wehland J, Chakraborty T, Weiss S. Oral somatic transgene vaccination using attenuated S. typhimurium. Cell 1997;91:765– 75.