Heterosexual transmission accounts for most new cases of HIV-1 infection each year. Recent reports from the World Health Organization estimate that a total of 39.4 million people are now infected with HIV-1 worldwide, 90% of whom live in developing countries.1 As of 2004, nearly 50% of all infected individuals were women.1 The increased incidence of HIV-1 infection in women, particularly those of child-bearing age, underscores the urgent need for effective preventative and therapeutic options that are safe, affordable, and culturally accepted. Because HIV-1 infection rates exceed 35% in parts of sub-Saharan Africa, emphasis has been placed on vaccine development, and new research initiatives have produced several candidate HIV vaccines. Most are still in the early stages of clinical testing, however, and have yet to demonstrate an acceptable level of efficacy.
In the absence of an effective preventative vaccine, topical microbicides may provide a useful and cost-effective means for preventing new infections.2,3 In the last several years, a number of small peptides and proteins have been identified with activity against HIV-1, including molecules that inhibit membrane fusion and viral entry.4-7 A novel antiviral is the naturally occurring compound cyanovirin (CV-N). CV-N is an 11-kd protein originally purified from extracts of cultured cyanobacterium (Nostoc ellipsosporum).8,9 CV-N exerts its antiviral activity by binding to high-mannose residues in the HIV-1 viral envelope and prevents virus entry by blocking fusion with the cell membrane and cell-to-cell transmission of HIV-1 infection.10-13 CV-N has potent neutralizing activity at nanomolar concentrations against a broad range of HIV-1 isolates in vitro, including M-tropic strains that are most commonly isolated early after sexual transmission.14 Tsai and colleagues15,16 demonstrated the efficacy of CV-N in pathogenic SHIV89.6P vaginal and rectal challenge experiments involving cynomolgus macaques.
Although in vivo studies using CV-N establish the principle that protein microbicides can prevent viral transmission, their translation into new therapeutic paradigms is made difficult by the cost of manufacturing enough product for widespread clinical use. Furthermore, the limited half-life of proteins after direct mucosal application is perhaps a greater barrier to the widespread use of these agents.
In this report, we describe an alternative system for the synthesis and mucosal delivery of CV-N based on genetically engineered lactic acid bacteria (LAB). We demonstrate that CV-N secreted by LAB exhibits potent antiviral activity against laboratory and primary patient-derived HIV-1.
MATERIAL AND METHODS
Bacterial Strains, Media, and Growth Conditions
Escherichia coli strain DH5α (Invitrogen, Carlsbad, CA) and 2 strains of LAB, Lactococcus lactis MG1363 and Lactobacillus plantarum NCIMB8826, were used as hosts. E. coli was grown in Luria-Bertani (LB) broth (Difco Laboratories, Detroit, MI) at 37°C. L. lactis and L. plantarum were grown in M17 medium (Difco) in which lactose was replaced by 0.5% glucose at 30°C and Man, Rogosa and Sharpe (MRS) medium (Difco) at 37°C, respectively. Antibiotics were added at the given concentrations: 5 μg/mL of erythromycin for L. lactis and L. plantarum and 100 μg/mL of ampicillin for E. coli.
Transformation Procedures of Lactic Acid Bacteria
Electroporation of L. lactis was performed as described previously.17 For L. plantarum, the following protocol was used. An overnight culture was prepared in MRS with 1% glucose and 0.1% glycine. The preculture was diluted 1:40 (vol/vol) in the same medium, grown to an optical density at 600 nm (OD600) of 0.6, and chilled on ice. Bacteria were harvested by centrifugation for 10 minutes at 6800g and were washed once in 1 culture volume of ice-cold MgCl2 solution (1 mM) and 2 times in 30% polyethylene glycol 3000 plus 10% glycerol. Cells were resuspended in a 1:100 culture volume of 30% polyethylene glycol 3000 plus 10% glycerol. Electroporation was performed using 50 μL of competent cells and a maximum of 5 μL of plasmid DNA in 0.2-cm electroporation cuvettes. Settings of 1.5 kV, 25 μF, and 400 Ω were used with the Gene Pulser (Bio-Rad, Hercules, CA). Electroporated cells were transferred into 0.5 mL of MRS containing 0.5 M of sucrose plus 0.1 M of MgCl2 and incubated for 2 hours at 37°C before plating on selective media.
Design of an Expression System for Cyanovirin Synthesis and Secretion in Lactic Acid Bacteria
To enable expression and secretion of CV-N by L. lactis and L. plantarum, we used derivatives of the broad Gram-positive host range vector pTREX1, which has been used previously for heterologous protein expression and secretion in L. lactis.18,19 The 2 expression plasmids used in this study, designated pTSV1 and pTSV2, contain the pUC origin of replication and the ampicillin resistance gene, enabling them to be used as shuttle vectors in E. coli. Plasmids pTSV1 and pTSV2 also contain expression cassettes that enable heterologous proteins to be expressed intracellularly or secreted by the construction of gene fusion to the lactococcal signal leader derived from the usp45 gene of L. lactis20 (K. Schofield and J.M. Wells, unpublished data). Amplification of DNA fragments by polymerase chain reaction (PCR) was performed using Platinum Pfx DNA Polymerase (Invitrogen) according to standard procedures and the reaction conditions recommended by the manufacturer. Routine molecular biology techniques for cloning and plasmid DNA isolation were performed by standard methods described previously.21 All new vector constructions were confirmed by sequencing.
Intracellular Expression of Cyanovirin
To achieve intracellular production of CV-N, the structural gene for CV-N was amplified and introduced into the Bcl I site of pTSV1, thereby giving rise to pTSV1-CVN (Fig. 1C). PCR primers used for amplification of the CV-N gene were CVN80S (5′-ACT AGT GGA TCC AAA TGA AGC TTG GTA AAT TCT CCC AGA CC-3′) with CVN80AS (5′-GCA TTG TGA TCA TTA TTC GTA TTT CAG GGT ACC G-3′). Plasmid pET (CV-N) was used as a DNA template for CVN amplification.9
Extracellular Secretion of Cyanovirin
Secretion of CV-N was accomplished by cloning the CV-N gene in pTSV2 to create gene fusion, with the secretion signal leader peptide from usp45 encoding an abundant lactococcal secreted protein (see Fig. 1). To facilitate cloning of the CV-N gene into pTSV2, restriction endonuclease sites for Nae I and BamH I were incorporated into the 5′ ends of the CV-N sense primer (5′-ATG CAT GCC GGC CTT GGT AAA TTC TCC CAG ACC TG-3′) and antisense primer (5′-TAC AAT GGA TCCTTA TTC GTA TTT CAG GGT ACC-3′), respectively. In the resulting construct, the amino acid sequence MKKKIISAILMSTVILSAAAPLSGVYA lies immediately upstream of the CV-N coding sequence.20 It has been previously demonstrated that the secretion efficiency of heterologous proteins by LAB can be enhanced by adding negatively charged amino acids after the signal peptide cleavage site at the N-terminus of the mature protein.22,23 To determine whether this modification improved the secretion of CV-N, a 5-amino acid propeptide sequence, DTNSD (D), was introduced into pTSV2-CVN at the N-terminus of the CV-N coding sequence. The introduced propeptide sequence DTNSD, which represents the N-terminus of the lactococcal secreted protein Usp45, contributes to additional negatively charged amino acid sequences at the N-terminus of the mature protein [net charge of (−)2]. To make this construction, the nucleic acid sequence encoding DTNSD was incorporated in the CV-N sense primer downstream of the Nae I site (5′-ATG CAT GCC GGC GAC ACA AAC TCA GAT CTT GGT AAA TTC TCC CAG AC-3′) and used in combination with the CV-N reverse primer indicated previously to amplify the DTNSD-CVN gene fusion product, which was subsequently directionally cloned between the Nae I and Bam H I sites in pTSV2 to generate pTSV2-D-CVN (see Fig. 1). To enhance the translational efficiency of CV-N in L. plantarum, we synthesized a codon-optimized CV-N gene by “gene assembly” PCR using the following oligonucleotides: C1, 5′-GTT ATA ATA CGA GTA GTA TTG ATT TAA ATA GTG TTA TTG AAA ATG TTG ATG GTA GTT TAA AAT GGC AAC CAA GTA ATT TT-3′; C2, 5′-GTT TTA CAT TCG GCA GCT AAT TCA CTA CTA CCA GCT AAT TGC GTA TTC CGA CAC GTT TCA ATA AAA TTA CTT GGT TGC CA-3′; C3, 5′-ACG TGT TAT AAT AGT GCT ATT CAA GGT AGT GTT TTA ACG AGT ACG TGT GAA CGG ACG AAT GGT GGT TAT AAT ACG AGT AG-3′; C4, 5′-CCA TCA ATA TTA GCA ATA TGA TCA TCT AAA TTA ATT TTC GTA CTA ACA AAT TGT TGA GCC CGC GTT TTA CAT TCG GCA GC-3′; C5, 5′-ATG CAT GCC GGC GAC ACA AAC TCA GAT TTA GGT AAA TTT AGT CAA ACG TGT TAT AAT AGT G-3′; and C6, 5′-GCA TTT GGA TCC TTA TTC ATA TTT TAA CGT ACC ATC AAT ATT AGC AAT ATG-3′. The PCR product was gel-purified, digested with Nae I and BamH I, and cloned into pTSV2, resulting in plasmid pTSV2-D-CVNco (see Fig. 1).
Western Blot Analysis of Cyanovirin Expression and Secretion
Preparation of protein extracts from bacterial cells and supernatant was performed as follows. Five milliliters of overnight culture (OD600 = 1) was centrifuged at 3000g for 10 minutes at 4°C. Culture supernatant was removed and passed through a 0.45-μm filter. Ten microliters of supernatant was used for Western blot analysis. A cell pellet was washed 3 times in Tris-buffered saline and resuspended in 200 μL of 20 mM of Tris-HCl, pH 7.5, 10% wt/vol sucrose and 10 mg/mL of lysozyme and was incubated for 1 hour at 37°C. After the cell pellet was washed once with 20 mM of Tris-HCl, pH 7.5, 10% wt/vol sucrose and resuspended in 200 μL of phosphate-buffered saline (PBS), an equal amount of sterile acid-washed glass beads (0.10-11 mm) was added. To break up cells, samples were shaken 3 times for 1 minute with 1 minute intervals, during which the suspension was kept on ice. To separate cellular material from glass beads and unbroken cells, tubes were centrifuged for 2 minutes at 1000g. Protein concentrations were determined by the Bradford procedure. Supernatant of L. lactis and L. plantarum overnight cultures was passed through a 0.45-μm filter before Western blot analyses. Proteins were resolved on 15% sodium dodecyl sulfate (SDS)-polyacrylamide gels and transferred onto Hybond Polyvinylidene fluoride membrane (PVDF) membranes (Biorad). The blot was probed with a rabbit polyclonal antibody against CV-N. Purified recombinant CV-N (rCV-N; 10 μM) from E. coli was loaded at a concentration of 10 ng per lane as a quantification standard. Immunodetection was performed using the peroxidase-based ECL Plus detection system (Amersham Biosciences, Piscataway, NJ) according to the manufacturer's instructions.
HIV-1 Infection of H9 Cells
Immortalized CD4+ H9 cells were maintained in complete RPMI medium containing 10% fetal bovine serum (FBS), 2 mM of glutamine, penicillin (100 U/mL), and streptomycin (100 μg/mL) at 37°C in 5% CO2. Five million H9 cells were incubated with the infectious molecular clone HIV-1NL4-3 for 3 hours at 37°C with an inoculum of 300 TCID50 (50% tissue culture infectious dose) of virus per 106 cells. The infection was performed in the presence of different amounts of bacterial cellular protein extract, and as a control, infection was performed in the presence of unused culture medium. Antiviral activity of intracellular CV-N was analyzed using 10 and 30 μg of cellular protein extract. Protein samples were prepared from overnight cultures (5 mL) of L. lactis MG1363 strains transformed with the empty control vector (pTSV1) or a construct driving intracellular CV-N expression (pTSV1-CVN). Cell-free supernatant was assayed for HIV-1 p24 antigen on day 10 after infection. HIV-1 p24 antigen levels were quantified by enzyme-linked immunosorbent assay (ELISA; Beckman-Coulter, Fullerton, CA) according to the manufacturer's instructions.
Single-Cycle HIV-1 Infection of MT4 Cells and Peripheral Blood Mononuclear Cells
MT4 cells were maintained in complete RPMI medium at 37°C in 5% CO2. HIV-1 Env-pseudotyped and luciferase-expressing reporter virus was produced by cotransfecting HEK293T cells with the envelope-deficient HIV-1 construct pNL-Luc and a vector expressing viral envelope glycoprotein derived from HIV-1HXB2. Env-pseudotyped reporter virus NL-Luc/HXB2env was titered according to the method of Connor et al24 and quantified by p24 antigen ELISA (Beckman-Coulter). Culture supernatant of L. lactis and L. plantarum was harvested from an overnight grown culture and filter-sterilized. Culture supernatant of L. plantarum was adjusted to pH 7.0 before infection experiments. A viral inoculum of 20 ng of p24 equivalent of NL-Luc/HXB2env was incubated with 12.5 μL (L. lactis) and 10 μL (L. plantarum) of bacterial supernatant in a total volume of 100 μL of RPMI medium for 10 minutes at 37°C. Subsequently, the virus in the reaction was used to infect 2 × 105 MT4 cells seeded in a 24-well plate for 3 hours at 37°C. Unbound virus was removed by washing with PBS, and fresh medium was added to the cells. As controls, purified rCV-N produced in E. coli was included at concentrations ranging from 1.5 to 150 nM in the bacterial supernatant of L. lactis and L. plantarum transformed with the empty control vector (pTSV2). Forty-eight hours after infection, cells were washed once in PBS and collected for the luciferase reporter assay. The HIV-1 infection assay of peripheral blood mononuclear cells (PBMCs) was carried out as follows. Briefly, PBMCs were cultured for 48 hours in the presence of phytohemagglutinin (PHA) and recombinant human interleukin-2 (IL-2) and were subsequently seeded at a density of 2 × 105 per well in a 24-well tissue culture plate. Stimulated cells were infected for 3 hours at 37°C with a 100-μL reaction mixture containing 5 μL of filter-sterilized and pH 7-adjusted L. plantarum supernatant and 50 ng of p24 equivalent of NL-Luc/HXB2env, both preincubated for 10 minutes at 37°C. Infected PBMCs were washed once with PBS and maintained in growth medium (RPMI, 10% fetal bovine serum [FBS], IL-2, glutamine, and penicillin/streptomycin) for 48 hours. Cells were harvested by centrifugation and washed once with PBS. MT4 cells and PBMCs were then lysed with 100 μL of reporter lysis buffer (Promega, Madison, WI), and luciferase activity was measured in relative light units (RLU) in a mixture of 20 μL of cell lysate and 100 μL of Luciferase Assay Reagent (Promega).
Infection With Primary HIV-1 Isolates
TZM-bl cells stably expressing CD4− CXCR4-CCR5 and β-galactosidase/luciferase under the control of the HIV-1 LTR were cultured at 37°C in 5% CO2 in complete Dulbecco modified Eagle medium (DMEM) consisting of DMEM supplemented with 10% FBS, penicillin (100 U/mL), and streptomycin (100 μg/mL). One day before infection, TZM-bl cells were plated at 1 × 104 in a 96-well tissue culture plate. Cells were infected in a total volume of 200 μL containing 5 to 25 ng of p24 equivalent of primary HIV-1 isolate and 20 μL of filter-sterilized and pH-adjusted supernatant of L. plantarum. The antiviral activity of rCV-N was tested against primary isolate P1 and was included at concentrations ranging from 1.5 to 150 nM. β-galactosidase activity was quantified 48 hours after infection using the Galacto-Star System (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions.
Recombinant Cyanovirin Expressed by L. lactis Retains Antiviral Activity
We began our experimental work using L. lactis, given that the pTSV vectors were developed primarily for use in this organism. We found that transformation of L. lactis with pTSV1-CVN resulted in intracellular accumulation of CV-N, as shown by Western blot analysis of bacterial cell extracts (Fig. 2A). Intracellular synthesis of CV-N led to no discernible toxic effects on bacterial cell viability or growth kinetics compared with a control strain transformed with the empty vector. The antiviral activity of CV-N was verified by infection of H9 lymphocytic cells with the infectious molecular clone HIV-1NL4-3 in the presence of cellular extracts of L. lactis transformed by plasmids encoding CV-N (pTSV1-CVN) or empty control plasmids (pTSV1). HIV-1 infection of H9 cells in the presence of L. lactis cell extracts containing rCV-N inhibited viral infection in a dose-dependent manner quantified by HIV-1 p24 antigen levels in cell culture supernatant (see Fig. 2B). Ten micrograms of cell extract of pTSV1-CVN-transformed bacteria reduced viral infection by 83%, and 30 μg resulted in complete inhibition of HIV-1 infection. These experiments confirmed that CV-N expressed by L. lactis was biologically active in vitro.
Engineering Cyanovirin Secretion by L. lactis
To enable CV-N secretion by L. lactis, we introduced the gene encoding CV-N into the pTSV2 vector. We found that recombinant bacteria secreted only low amounts of CV-N into the supernatant. Previous work has demonstrated that efficient protein translocation across the membrane requires an acidic N-terminus of the mature protein. In contrast, proteins with N-terminal basic residues are poorly secreted.22,23 Given that the first 5 amino acids of CV-N (LGFKS) exhibit a net global charge of (+1), we hypothesized that fusion of an acidic propeptide sequence DTNSD with a net negative charge of (−2) to the N-terminus of the mature protein CV-N would enhance secretion efficiency. The DTNSD peptide sequence represents the first 5 amino acids of the lactococcal major secreted protein Usp45. To determine whether this strategy increases translocation efficiency of CV-N across the cytoplasmic membrane, we assessed CV-N secretion levels of L. lactis transformed with pTSV2-D-CVN secreting the DTNSD-CV-N fusion protein and L. lactis transformed with pTSV2-CVN expressing CV-N fused with the Usp45 leader sequence. The addition of the DTNSD propeptide sequence in pTSV2-D-CVN resulted in an approximately 6- to 8-fold increase of CV-N secretion compared with pTSV2-CVN (Fig. 3). The amount of secreted rCV-N increased from 35 to 250 ng/mL. These results are consistent with earlier studies, which have shown that the addition of an anionic propeptide spacer sequence at the N-terminus of the mature moiety of secreted proteins not only improves protein translocation but prevents nonspecific association with the cell wall and possible degradation.
Codon Optimization Improves Expression of Cyanovirin in Lactic Acid Bacteria
To enhance translation efficiency further and increase the amount of CV-N expressed, we adjusted the codon use of the CV-N gene to match that of the most frequently used codons in L. plantarum. The codon-optimized sequence of CV-N displays only 65.7% homology to the naturally occurring CV-N of cyanobacterium (blue green alga), Nostoc ellipsosporum, and the A+T content was increased from 48% to 68%. L. lactis transformed with pTSV2-D-CVNco incorporating the codon-optimized CV-N gene secreted 8- to 10-fold higher levels of protein compared with cells expressing unmodified CV-N (pTSV2-D-CVN) (see Fig. 3). The amount of secreted protein was increased up to 2 to 2.5 μg/mL, as determined by Western blot analysis using reference standards of known quantities of rCV-N. Thus, we were able to show that the inclusion of the DTNSD sequence adjacent to the signal peptide combined with the codon optimization of the CV-N gene resulted in significantly enhanced expression and secretion.
Antiviral Activity of Recombinant Cyanovirin Secreted by L. lactis
To verify the biologic activity of secreted rCV-N, a single-cycle HIV-1 infection assay was performed using the envelope-deficient HIV-1 luciferase reporter construct pNL-Luc, pseudotyped with a CXCR4 tropic HIV-1 envelope (HXB2env). Infection of lymphocytic MT4 cells with reporter virus NL-Luc/HXB2env was performed in the presence of bacterial culture supernatant. Purified rCV-N produced in E. coli was included at concentrations ranging from 1.5 to 150 nM in the supernatant of L. lactis transformed with empty control vector.
We found a dose-dependent antiviral effect on HIV-1 infection with maximal suppression of HIV-1 infection using the construct with the highest secretion efficiency (pTSV2-D-CVNco) (see Fig. 3; Fig. 4). Accordingly, the construct pTSV2-D-CVN, which secreted much lower amounts of CV-N, inhibited HIV-1 infection far less effectively. The antiviral activity obtained with pTSV2-D-CVN is equivalent to the activity of purified E. coli-derived rCV-N at a concentration of 1.5 nM, however, which is the reported 50% effective concentration (EC50) of that protein.8 Likewise, the antiviral effect of 15 nM of purified rCV-N corresponds to the viral suppression achieved by pTSV2-D-CVNco, reflecting the approximately 10-fold increased secretion efficiency of that construct compared with pTSV2-D-CVN (see Fig. 4). In summary, the calculated inhibitory concentrations of lactococcal rCV-N were comparable to those obtained with purified E. coli-derived rCV-N, suggesting full biologic activity of rCV-N generated in L. lactis.
Secretion of Cyanovirin by L. plantarum
Although initial experiments used CV-N expressed in L. lactis, we also evaluated protein expression in another strain of LAB, L. plantarum (strain NCIMB8826). Our interest in this strain derives from recent published studies demonstrating that on direct inoculation, L. plantarum persists longer (time (t)½ of 1 week) in mucosal sites than other strains of lactobacilli.25 Similar to the experiments in L. lactis, the inclusion of the propeptide sequence DTNSD in combination with the codon use adjustment was again critical for efficient secretion of CV-N (Fig. 5). These modifications resulted in a concentration of secreted rCV-N of approximately 4 μg/mL, exceeding the amount of CV-N secreted by L. lactis, with a maximum concentration of approximately 2.5 μg/mL. The codon optimization of CV-N enhanced secretion efficiency approximately 10-fold, similar to the increase observed in L. lactis. The difference in the amount of secreted rCV-N can be explained by the property of L. plantarum to grow to higher density than L. lactis, reflecting higher amounts of accumulated rCV-N in the culture medium.
These findings suggest that pTSV plasmids can be successfully introduced into other LAB (other than L. lactis) to express and secrete heterologous proteins such as CV-N.
Antiviral Activity of Recombinant Cyanovirin Secreted by L. plantarum
The antiviral activity of CV-N secreted by L. plantarum was assessed with 2 different HIV-1 infection assays: a single-cycle infection assay using HIV-1 reporter virus NL-Luc/HXB2env to infect MT4 cells and PBMCs and infection of CD4-CXCR4-CCR5 expressing TZM-bl cells with 5 primary patient isolates of HIV-1. The antiviral activity of CV-N secreted by lactobacilli transformed with pTSV2-D-CVN and pTSV2-D-CVNco corresponded to the degree of viral suppression obtained with purified rCV-N at concentrations of 1.5 and 15 nM, respectively (Fig. 6A). Culture media from lactobacilli transformed with the codon-optimized construct pTSV2-D-CVNco reduced HIV-1 infection by 83%, which corresponds to the degree of inhibition (85%) obtained with L. lactis supernatant (see Fig. 4). The same bacterial supernatant inhibited viral infection of primary cells (PBMCs) infected with NL-Luc/HXB2env by 76% (see Fig. 6B). Finally, we used 5 different primary isolates, including a multidrug-resistant strain, to infect TZM-bl cells in the presence of culture supernatant of L. plantarum transformed with pTSV2-D-CVNco and pTSV2. Secreted CV-N reduced viral infection of primary HIV-1 isolates by 56% to 75% (see Fig. 6C). The results of the conducted infection assays confirm the full biologic activity of CV-N produced in L. lactis and L. plantarum and show that it exhibits antiviral activities similar to purified rCV-N generated by E. coli.
Several challenges confront the development of an effective HIV-1 microbicide. The product must target conserved elements found in a genetically diverse swarm of viruses. The eventual formulation must ensure adequate mucosal levels of drug over extended periods. Mucosal administration must not be associated with adverse effects such as mucosal irritation in the host or sexual partner. Microbicides that require active and repeated application before sexual contact may increase toxicity, lower the frequency of compliance, and complicate evaluations of efficacy. In many cases, cultural mores prevent women from taking the necessary steps to apply a vaginal microbicide immediately before intercourse. As a result, a topical microbicide may be of limited value despite having high efficacy. Thus, an urgent need exists to devise efficient systems to allow the delivery of microbicides to mucosal surfaces.
We focused on the antiviral compound CV-N, given recent data demonstrating efficacy in preventing simian HIV (SHIV) transmission in vivo after direct mucosal application.15,16 Our rationale for choosing strains of LAB as mucosal delivery vehicles for CV-N derives from the well-accepted biosafety profile of LAB. LAB are extensively used as starters of food fermentation and are widely consumed by humans.26 Some strains of LAB colonize mucosal surfaces after a single administration.27,28 Thus, LAB-based microbicides may potentially be administered well before sexual contact and may serve as an in situ locus of mucosal microbicide production. This strategy would be especially useful for protein microbicide products by overcoming the 2 major challenges confronting their clinical development: synthesis of enough product for clinical trials and ensuring adequate levels of mucosal drug after direct application.
LAB constitute a normal part of the vaginal microflora and may protect against acquisition of sexually transmitted diseases (STDs), including HIV.29-31 Certain strains of LAB produce hydrogen peroxide (H202), a known antimicrobial compound that exerts antiviral and antibacterial activity in the cervicovaginal tract.28,32 Recent studies have demonstrated that disturbances in the vaginal flora, particularly the loss or absence of LAB, are associated with increased STD incidence and HIV-1 prevalence in African women.33,34 Taken together, these findings indicate a protective role for LAB species that may reduce or prevent transmission of potential pathogens, including HIV-1. Microbicide strategies that use these bacteria as a delivery vehicle are therefore viewed as particularly beneficial.
In recent studies, work by Chang and colleagues35 has shown that the natural human vaginal isolate L. jensenii can be genetically modified to secrete a 2-domain molecule of CD4 (2D CD4), a derivative of the HIV-1 receptor. Secreted 2D CD4 binds HIV-1 gp120 in vitro and also modestly decreases infectivity of a primary HIV-1 isolate (HIV-1JRFL). These promising results, in conjunction with our findings, prove the principle that genetically altered strains of commensal bacteria can be used as vehicles for the delivery of microbicidal compounds.
Can bioengineered LAB be used to prevent the mucosal transmission of HIV-1? Our results demonstrate that different strains of LAB can synthesize and secrete CV-N in sufficient quantity to exert an anti-HIV-1 effect in vitro. Although the pTSV series of plasmids was developed for use in L. lactis, we found that these plasmids were equally functional in L. plantarum. L. lactis and L. plantarum differ significantly in terms of mucosal retention after oral administration or direct application, with a half-life of approximately 6 hours and 7 days, respectively. In choosing a LAB strain for future microbicide development, mucosal retention is likely to be a critical parameter in attaining antiviral efficacy. Several investigators have demonstrated that natural vaginal isolates of lactobacilli can be engineered to secrete heterologous proteins.35,36 It is likely that these strains may persist and colonize the human vagina, thus decreasing the frequency of administration.
Our report describes the optimization of the secretory capacity of LAB by inserting a net negative amino acid charge at the N-terminus of the mature protein and lactobacillus-specific codon optimization. Both manipulations seem to be critical for maximizing expression and secretion in these hosts and need to be employed for use in future in vivo studies. Ultimately, however, the antiviral efficacy of LAB-based microbicides is dependent on the absolute amount of drug expressed and retained in vivo in the vaginal or rectal cavity after mucosal administration.
Currently, we have generated in vitro preliminary data showing the expression and secretion of CV-N by recombinant LAB and have demonstrated the in vitro antiviral activity of secreted CV-N against HIV-1. A similar approach may be applied to the mucosal delivery of other antiviral compounds, including HIV-1 fusion inhibitors. In conclusion, our results suggest that recombinant LAB may serve as effective microbicidal compounds and deserve in vivo testing in simian immunodeficiency virus models of mucosal virus transmission.
The authors thank K. Schofield for assistance with the construction of expression cassettes for use in L. lactis. This research has been facilitated by the infrastructure and resources provided by the Lifespan/Tufts/Brown Center for AIDS Research (NIHP30 AI42853).
1. Piot P, Feachem RG, Lee JW, et al. Public health. A global response to AIDS: lessons learned, next steps. Science
2. D'Cruz OJ, Uckun FM. Clinical development of microbicides for the prevention of HIV infection. Curr Pharm Des
3. Turpin JA. Considerations and development of topical microbicides to inhibit the sexual transmission of HIV. Expert Opin Investig Drugs
4. Cervia JS, Smith MA. Enfuvirtide (T-20): a novel human immunodeficiency virus type 1 fusion inhibitor. Clin Infect Dis
5. Cole AM, Lehrer RI. Minidefensins: antimicrobial peptides with activity against HIV-1. Curr Pharm Des
6. Wild C, Greenwell T, Matthews T. A synthetic peptide from HIV-1 gp41 is a potent inhibitor of virus-mediated cell-cell fusion. AIDS Res Hum Retroviruses
7. Wild C, Oas T, McDanal C, et al. A synthetic peptide inhibitor of human immunodeficiency virus replication: correlation between solution structure and viral inhibition. Proc Natl Acad Sci USA
8. Mariner JM, McMahon JB, O'Keefe BR, et al. The HIV-inactivating protein, cyanovirin
-N, does not block gp120-mediated virus-to-cell binding. Biochem Biophys Res Commun
9. Mori T, Boyd MR. Cyanovirin
-N, a potent human immunodeficiency virus-inactivating protein, blocks both CD4-dependent and CD4-independent binding of soluble gp120 (sgp120) to target cells, inhibits sCD4-induced binding of sgp120 to cell-associated CXCR4, and dissociates bound sgp120 from target cells. Antimicrob Agents Chemother
10. Bewley CA, Otero-Quintero S. The potent anti-HIV protein cyanovirin
-N contains two novel carbohydrate binding sites that selectively bind to Man(8) D1D3 and Man(9) with nanomolar affinity: implications for binding to the HIV envelope protein gp120. J Am Chem Soc
11. Bolmstedt AJ, O'Keefe BR, Shenoy SR, et al. Cyanovirin
-N defines a new class of antiviral agent targeting N-linked, high-mannose glycans in an oligosaccharide-specific manner. Mol Pharmacol
12. Boyd MR, Gustafson KR, McMahon JB, et al. Discovery of cyanovirin
-N, a novel human immunodeficiency virus-inactivating protein that binds viral surface envelope glycoprotein gp120: potential applications to microbicide
development. Antimicrob Agents Chemother
13. Shenoy SR, O'Keefe BR, Bolmstedt AJ, et al. Selective interactions of the human immunodeficiency virus-inactivating protein cyanovirin
-N with high-mannose oligosaccharides on gp120 and other glycoproteins. J Pharmacol Exp Ther
14. Dey B, Lerner DL, Lusso P, et al. Multiple antiviral activities of cyanovirin
-N: blocking of human immunodeficiency virus type 1 gp120 interaction with CD4 and coreceptor and inhibition of diverse enveloped viruses. J Virol
15. Tsai CC, Emau P, Jiang Y, et al. Cyanovirin
-N inhibits AIDS virus infections in vaginal transmission models. AIDS Res Hum Retroviruses
16. Tsai CC, Emau P, Jiang Y, et al. Cyanovirin
-N gel as a topical microbicide
prevents rectal transmission of SHIV89.6P in macaques. AIDS Res Hum Retroviruses
17. Reuter MA, Hanniffy S, Wells JM. Expression and delivery of heterologous antigens using lactic acid bacteria. Methods Mol Med
18. Robinson K, Chamberlain LM, Lopez MC, et al. Mucosal and cellular immune responses elicited by recombinant Lactococcus lactis strains expressing tetanus toxin fragment C. Infect Immun
19. Steidler L, Hans W, Schotte L, et al. Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science
20. van Asseldonk M, Rutten G, Oteman M, et al. Cloning of usp45, a gene encoding a secreted protein from Lactococcus lactis subsp. lactis MG1363. Gene
21. Gilbert C, Robinson K, Le Page RW, et al. Heterologous expression of an immunogenic pneumococcal type 3 capsular polysaccharide in Lactococcus lactis. Infect Immun
22. Le Loir Y, Gruss A, Ehrlich SD, et al. A nine-residue synthetic propeptide enhances secretion efficiency of heterologous proteins in Lactococcus lactis. J Bacteriol
23. Le Loir Y, Nouaille S, Commissaire J, et al. Signal peptide and propeptide optimization for heterologous protein secretion in Lactococcus lactis. Appl Environ Microbiol
24. Connor RI, Sheridan KE, Ceradini D, et al. Change in coreceptor use correlates with disease progression in HIV-1-infected individuals. J Exp Med
25. Mercenier A, Pavan S, Pot B. Probiotics as biotherapeutic agents: present knowledge and future prospects. Curr Pharm Des
26. Dieye Y, Usai S, Clier F, et al. Design of a protein-targeting system for lactic acid bacteria. J Bacteriol
27. Antonio MA, Hillier SL. DNA fingerprinting of Lactobacillus crispatus strain CTV-05 by repetitive element sequence-based PCR analysis in a pilot study of vaginal colonization. J Clin Microbiol
28. Patton DL, Cosgrove Sweeney YT, Antonio MA, et al. Lactobacillus crispatus capsules: single-use safety study in the Macaca nemestrina model. Sex Transm Dis
29. Martin HL, Richardson BA, Nyange PM, et al. Vaginal lactobacilli, microbial flora, and risk of human immunodeficiency virus type 1 and sexually transmitted disease acquisition. J Infect Dis
30. Sewankambo N, Gray RH, Wawer MJ, et al. HIV-1 infection associated with abnormal vaginal flora morphology and bacterial vaginosis. Lancet
31. Silvester ME, Dicks LM. Identification of lactic acid bacteria isolated from human vaginal secretions. Antonie Van Leeuwenhoek
32. Rabe LK, Hillier SL. Optimization of media for detection of hydrogen peroxide production by Lactobacillus species. J Clin Microbiol
33. Mbizvo EM, Msuya SE, Stray-Pedersen B, et al. HIV seroprevalence and its associations with the other reproductive tract infections in asymptomatic women in Harare, Zimbabwe. Int J STD AIDS
34. van de Wijgert J, Coggins C. Microbicides to prevent heterosexual transmission of HIV: ten years down the road. BETA
35. Chang TL, Chang CH, Simpson DA, et al. Inhibition of HIV infectivity by a natural human isolate of Lactobacillus jensenii engineered to express functional two-domain CD4. Proc Natl Acad Sci USA
36. Kahala M, Palva A. The expression signals of the Lactobacillus brevis slpA gene direct efficient heterologous protein production in lactic acid bacteria. Appl Microbiol Biotechnol