Many novel therapeutic molecules are based on antibodies directed to molecular targets, including infectious pathogens. These therapeutic antibodies can be produced in a number of ways while maintaining their antigen affinity properties. However, conventional antibodies are frequently difficult to produce because of their complex structure. Short versions, such as single-chain antibodies, comprising the antibody-binding domains, lose some antibody properties and present reduced affinity and stability. Single-domain antibodies produced by camelidae have very interesting properties compared to classical mammalian antibodies as they have only one variable region (heavy chain), with a reduced size, and show greater stability under a range of conditions (Conrath et al., 2005). By a mechanism that is not fully elucidated and probably related to the small size of 3B2 and 2KD1, the interaction of these antibodies with their target antigen (the highly antigenically preserved viral protein VP6) in RV particles differs to that of conventional immunoglobulins. Thus, 3B2 and 2KD1 produce broad neutralization activity in vitro and are protective against diarrhea and viral shedding in a neonatal mouse model. These properties confer these antibodies a high potential therapeutic value. Experiments to demonstrate the molecular basis of this mechanism of interaction are currently under way.
Monoclonal antibodies are crucial in biopharmaceutical development and account for 42% of the total market of biologics. The annual growth of market requirements is estimated to increase to as much as 25 tons of therapeutic antibodies per year in 2013. This value would correspond to an annual growth rate of 37% [30
]. This scenario has driven pharmaceutical companies to identify novel production alternatives to conventional fermentation technologies that facilitate speed of development and production scale-up at reduced costs. In the last 15
years, living biofactories such as plants, transgenic mammalians and insects have been adapted to produce recombinant proteins of all kinds, providing excellent results. Today, some of these methodologies represent true alternatives to the fermentation of bacteria, yeast, and mammalian or insect cells.
Insect larvae frequently scale up the recombinant protein production mediated by baculovirus vectors to levels of grams/L or hundreds of grams/L instead of tens of milligrams/L usually obtained in insect cells under fermentation. This technology dramatically reduces the production costs and fixed investments while also offering other advantages over classical insect cell fermentation approaches. First, insects are complex organisms that contain multiple cell types in perfect physiological conditions for productivity. Complexity frequently increases productivity in any expression system. Second, the scale-up of production based on insects (lepidopters) is unlimited and fast. Tons of silk thread, composed by two proteins, are produced by a lepidopter every year and companies using insects to produce baculoviruses as biopesticides generate millions of infected insects in short periods of time by simple procedures. Third, given the flexibility of baculovirus vectors, the development times implied by use of insects as living biofactories is very short when compared to the generation of productive transgenic mammalian cell clones.
The first descriptions of insect larvae use as living biofactories emerged in the early ‘90s [31
] and several companies in Europe, USA and Asia currently exploit this technology using Trichoplusia ni
or Bombyx mori
lepidopters with their respective baculovirus vectors. Various kinds of protein have been successfully expressed in larvae, including diagnostic reagents [13
], vaccines [14
], therapeutic proteins [27
], enzymes [32
], and conventional complete or fragmented antibodies [34
]. However, to the best of our knowledge, this is the first description that evaluates the efficiency of insects as living biofactories to express recombinant single-domain antibodies (VHH) derived from llama. Our results demonstrate that insects are an efficient alternative to produce fully functional VHHs, which, in this case, can be used for diagnostic and therapeutic purposes. The antibodies 3B2 and 2KD1 are directed to unknown epitopes within the VP6 inner capsid protein, which is highly immunogenic and conserved among all Group A RVs, including several subgroup specificities, various G/P-type combinations, and distinct species of origin, as previously demonstrated [5
These VHH molecules were previously generated functionally in E. coli
. However, the use of insect host cells to produce VHHs would provide further advantages, such as the possibility to make posttranslational modifications, including the formation of disulphide bonds [36
]. In this case, llama VHHs, which have an extended CDR3 that is often stabilized by an additional disulphide bond with a cysteine residue in CDR2 [1
], would be more efficiently produced.
The VHHs expressed in this study were cloned transcriptionally in phase with the signal peptide of honey bee mellitin. The incorporation of this signal peptide to the construct increased the expression yields, as previously reported with other proteins in the baculovirus expression system [38
], in comparison to the use the KDEL ER retention sequence (at least 200% increase). This signal peptide would allow the glycosylation of the protein. Although an anomalous migration in SDS-PAGE was observed in the VHH 3B2, this characteristic does not seem to be caused by this modification more than by another post-transductional modification since no glycosylation was detected when using the GLYCOPRO Detection Kit (Sigma, USA) (data not shown).
The binding capacity of the larva-derived 3B2 and 2KD1 VHHs was compared in a sandwich ELISA with the same antibodies expressed in E. coli
. The binding behavior was similar in all cases with respect to the RV viral particles (Figure ). In addition, both molecules neutralized Group A RV strain Wa (SbI, P [8
]G1) when they were incubated with the virus (Figure ). The capacity of the monomeric VHH to present neutralizing activity could be related to the small size of these molecules [5
]; however, research into the mechanism behind the RV virus neutralization is still underway.
Furthermore, the intragastric administration of the VHH 3B2 or 2KD1 in a neonatal mouse model exerted a protective effect against RV A-induced diarrhea caused by a highly infectious murine RV, ECw. Fifty-three percent of the pups treated with purified 3B2 were protected after 48 hpc and at 96 hpc about 35% of the animals were still protected, and none of these protected animals presented viral excretion. With the purified 2KD1, 61% of the animals were protected against diarrhea at 48 hpc and only one pup showed viral shedding 96 hpc, resulting in the group with the highest protection rates against RV infection and disease. Using non-purified larvae crude extracts containing the VHHs object of this study (TSP), protection levels were lower compared with their purified counterparts. However, 66.7% and 91.7% of the animals presented no viral excretion at 96 hpc by ELISA when treated with 3B2 TSP or 2KD1 TSP, respectively (Figure ). These results were very similar to those previously obtained when using the same VHHs expressed in E. coli
]. The most interesting protective results were obtained when the 3B2 and 2KD1 VHHs were combined, since 60% of the animals remained protected at 78 hpc and 50% after 96 hpc. Viral shedding was also dramatically reduced to 30% in this group (Figure ). Our results encourage us to continue to evaluate the protective value of these VHHs in a larger animal model of RV infection and disease that more closely resembled the physiology of human infants. Experiments with the larva-derived VHHs described here in gnotobiotic pigs are in course.
In conclusion, the monovalent VHHs were efficiently expressed in larvae and presented the same functionality as those expressed in E. coli
, showing polyreactive properties against several RV strains (INDIANA (SbI, P[5
]G6), Wa (SbI, P[8
]G1) and Ecw) both in ELISA and VN. These properties were shown to be associated with protection in a neonatal mouse model. Our results open up the possibility of using insects as living biofactories for the cost-efficient production of these and other VHHs to be used for diagnostic or therapeutic purposes, thereby eliminating concerns regarding the use of bacterial or mammalian cells.