Heterologous production of antimicrobial peptides has been attempted in several hosts
over the last few years. After the bacterium Escherichia coli, yeasts are the second most used system for heterologous peptide production (Parachin
et al. 2012]). In this study, heterologous production of a modified clavanin fused to thioredoxin,
clavMO-thio, was initially attempted in different E. coli strains. Nevertheless, expression of the encoding gene could never be confirmed (Additional
file 2: Figure S2, Additional file 3: Figure S3 and Additional file 4: Figure S4). The same has been previously observed when the production of SPE10 isolated
from the Pachyrrhizus erosus peptide was attempted in both E. coli and P. pastoris where heterologous peptide production could only be confirmed when produced in yeast
(Song et al. 2005]).
Although Sacharomyces cerevisiae is the most common yeast utilized for biopharmaceutical production, the yeast P. pastoris was chosen as the host in this study for having a GRAS (Generally Regards As Safe)
status, the ability to grow in high cell density cultures (since it does not present
fermentative behavior), and its reported high levels of secreted recombinant protein,
which simplifies downstream purification processes (Ahmad et al. 2014]). Finally, most recombinant AMPs produced in yeast use P. pastoris as a host (Parachin et al. 2012]).
Although some companies claim the production of recombinant clavanins isoforms such
as B, C, D and E using both S. cerevisiae and E. coli as hosts, herein we report for the first time the heterologous production of clavanin
using P. pastoris as a host to express clavMO. Synthetic ClavMO is reported to have higher antibacterial
activity against both Gram-positive (e.g., 78.75 µM against S. aureus ATCC29213; 2.5-fold higher than synthetic clavA) and Gram-negative bacteria (e.g.
39.40 µM against K. pneumoniae—ATCC13885; 2.5-fold higher than synthetic clavA). Furthermore, clavMO has also presented
immunomodulatory, antitumor, antiviral and insecticide activities (Silva et al. 2011b]).
In this study Clav-MO fused to thioredoxin in its N-terminal presented antibacterial
activity. Although one could argue that thioredoxin could have inhibitory activity
by itself its gene sequence was from the E. coli genome where it has been previously described its role in defense against oxidative
stress or in control of apoptosis (Arnér and Holgren 2000]). Moreover thioredoxin is frequently used as a carrier protein for production of
recombinant antimicrobial peptide representing more than 20% of all reported fusion
expressions of antimicrobial peptides (Li 2009]). Finally, in a study for heterologous production of viscotoxin where 13 fusion proteins
were tested, thioredoxin gave the highest yield of soluble protein (Bogomolovas et
al. 2009]). For all those reasons we claimed that the activity against the microorganisms tested
in our work is derived from the activity of the peptide clav-MO.
Here we demonstrate the production of a heterologous AMP using both constitutive and
inducible promoters. Few of the reported studies used a constitutive promoter to express
gene encoding for AMPs (Guo et al. 2012]; Hong et al. 2007]; Yu et al. 2010]). Constitutive AMP production is advantageous as the fermentation process is facilitated
because there is no need for additional inductor or media exchange. Nevertheless,
if the AMP has antifungal activity, the use of such a promoter is not advisable, for
it impairs yeast growth and consequently heterologous AMP production. In all reported
cases, the amount of heterologous peptide produced using constitutive promoter was
not reported. Moreover, in this study the strain with constitutive cassette had a
negative impact on final OD, which corroborates previous studies and reinforces the
utilization of inducible constructions for heterologous production of AMP.
Regarding clavanin mode of action, it is known that its bactericide activity is related
to membrane stability in a pH dependent form (van Kan et al. 2001], 2002], 2003a]). Furthermore, clavanin A has been shown to interact with lipid bi-layers, resulting
in drastic changes in membrane morphology (van Kan et al. 2003b]). Recently, our group has shown that nanoformulated clavanin A inhibit bacterial
growth of S. aureus, K. pneumoniae and Pseudonomas aeruginosa, being an excellent candidate for treating patients contaminated with antibiotic-resistant
bacteria (Saúde et al. 2014]). Another work form our group has shown that clavanin A is effective in treatments
of wound and sepsis infections by avoiding the beginning of sepsis, and as consequence,
it reduces mortality (Silva et al. 2015]).
Nevertheless, the chemical synthesis of clavanin aiming at nanoformulation is not
cost-effective. In this study, recombinant clavMO was shown to inhibit both Gram-positive
and Gram-negative bacteria. Thus, it presents an initial step for the development
of cost-effective, large-scale production of this AMP.
In this study, for the first time, the modified version of clavanin A, clavMO was
heterologously produced in P. pastoris using both constitutive and inducible expression cassettes. Both systems yielded
protein detection in yeast supernatant by Western blot assays. The strain with integrated
constitutive construction resulted in lower final OD when compared to the strain with
the inducible construction integrated in its genome. Therefore, the strain producing
clavMO after induction with metanol was chosen for the following experiments. ClavMO
was produced in a 5Â L scale followed up by purification using gel filtration. Finally,
antimicrobial assays showed that recombinant clavMO could inhibit up to 56 and 89%
Gram-negative and Gram-positive bacteria, respectively. Conclusively, it is evident
that P. pastoris is an excellent host for the functional production of clavMO, and that this system
may be utilized for further scale-up production of AMPs.