Second-generation HIV microbicides: Continued development of griffithsin

We are now more than 25 years into the AIDS pandemic and our main technological intervention for preventing vaginal and rectal transmission of HIV remains the condom. Although capable of providing significant protection if used regularly, condoms (male and female) are used inconsistently, and there is a clear need for additional prevention technologies. Vaginal microbicides are one such woman-controlled technology that may have higher acceptability, and consequently be used more regularly, than condoms. The report by O'Keefe et al. (1) in this issue of PNAS describes a promising, highly potent microbicide candidate, griffithsin, a red algal lectin that acts as an HIV entry inhibitor. The authors also report successful use of a versatile, rapid, and scaleable Nicotiana benthamiana-based manufacturing system for recombinant proteins, a critical step, because any microbicide would be used by millions of people.

Large-scale HIV prevention clinical trial results have been disappointing for the first generation of microbicide candidates (Fig. 1): nonxoynol-9, cellulose sulfate, and Savvy showed a potential increase in HIV transmission (2); Carraguard and BufferGel showed no effect (3, 4). Only PRO 2000 demonstrated a trend toward reduction in HIV transmission (30%; ref. 4). These gels were applied temporally close to intercourse, and it is well known from the contraceptive field that user acceptability is lower for products that must be used immediately before intercourse (e.g., condom, spermicide) compared with products whose use is noncoitally related (e.g., the oral contraceptive). Because it is challenging to ensure consistent use in clinical trials (3), it is difficult to determine whether these microbicides were ineffective because of insufficient activity of the microbicide or because they were used inconsistently. The potency and specificity of griffithsin (and other second-generation microbicides) may enable formulations for noncoital use (e.g., once-a-day vaginal tablets and films, or sustained release vaginal rings), which may improve consistent microbicide use and efficacy.

Fig. 1.

Activity and specificity of vaginal microbicides in clinical development. ¹Safe but ineffective for reducing HIV transmission in clinical trials; ²Showed a potential increase in HIV transmission in clinical trials; ³30% reduction in HIV transmission in a clinical trial (not statistically significant).

The report by O'Keefe et al. (1) is a significant advancement in the development of griffithsin as a candidate microbicide. The authors have shown that N. benthamiana-produced griffithsin potently neutralizes HIV-1 isolates from Clades A, B, and C (IC₅₀ values ranging from <0.003 to 0.15 μg/mL and IC₉₀ values from <0.003 to 7.0 μg/mL). Further, the authors found that exposure of human cervical explants to griffithsin (25 μg/mL) did not induce proinflammatory cytokines—a concern because these cytokines could potentially recruit HIV target cells to the mucosa. Similarly, these concentrations of griffithsin did not have proliferative effects on peripheral blood mononuclear cells, potential target cells for HIV. The authors also tested griffithsin in the rabbit vaginal irritation test, a required test before phase 1 safety testing of microbicides (5). When tested at concentrations up to 1 mg/mL, griffithsin showed no significant signs of toxicity in this model. Griffithsin has already been shown to be stable at vaginal and rectal pH and to function in vaginal fluid (6, 7). In vitro studies confirming the absence of inhibition of Lactobacilli growth and retention of activity in seminal plasma, as well as in vivo studies demonstrating protective efficacy in a non-human primate model are warranted while proceeding to phase 1 safety testing in human volunteers (2, 8).

Despite the promise recombinant proteins offer as microbicides, none are currently being evaluated clinically. One major hurdle to clinical development has been the lack of a cost-effective manufacturing system. O'Keefe et al. (1) have demonstrated rapid and scaleable manufacturing of griffithsin by using a versatile N. benthamiana production platform that has previously been used to generate clinical supply for a NHL vaccine (9). Transient N. benthamiana production systems (10–12) allow for rapid (1–2 weeks) and efficient production of recombinant proteins. Indeed, griffithsin was expressed at over 1 g/kg of leaf tissue after 12 days, and was processed to high purity with a yield of 0.3 g/kg. An environmentally controlled greenhouse producing 3,000 kg of leaf tissue per acre could yield 90,000 × 1 mL doses of 1 mg/mL griffithsin for each 12-day manufacturing run—or over 2 million doses per year. Although griffithsin is not glycosylated, one of the historic concerns with plant manufacturing has been the potential immunogenicity of plant glycans, specifically xylose and a nonmammalian linkage of fucose. Now with the availability of xylosyl- and fucosyltransferase knockout N. benthamiana transgenics (13), biologics with glycans indistinguishable from mammalian production systems can be produced.

The majority of second-generation microbicide agents in clinical development are HIV specific, but every year 340 million people acquire one of the four primary curable sexually transmitted infections (STIs), and unknown numbers acquire chronic viral and bacterial infections (14). In addition to causing significant morbidity on their own, many non-HIV STIs are cofactors for transmission of HIV (15, 16). Intriguingly, griffithsin also has in vitro activity against herpes simplex virus (K. Palmer, personal communication). The potential low cost and versatility of the N. benthamiana manufacturing systems may encourage production of highly potent recombinant drug combinations targeting HIV (e.g., griffithsin, neutralizing monoclonals, RANTES analogs)—thereby reducing selection of escape mutants (2), along with drugs targeting other STIs, for a broad-spectrum microbicide product (Fig. 1). It may even be desirable to combine vaginal microbicides with cervical barrier devices (17) to further improve protective efficacy. Because unsafe sex is one of the highest risk factors for disability and death worldwide (14), there is an urgent need to develop safe, effective, and accessible multipurpose technologies that will prevent pregnancy, sexually transmitted infections, and other common reproductive tract infections. The production and evaluation of griffithsin in a rapid, versatile, scaleable, low-capital-cost manufacturing system is one important step on the critical path to safe and effective microbicides.