Exploration and exploitation of the gut microbiota of black soldier fly larvae (Hermetia illucens)

During the last decade, there is a revived interest in the use of insects in food and feed applications because of its potential as a sustainable protein source. One of the most widely reared edible insects is the black soldier fly (Hermetia illucens, BSF). The voracious larvae of the black soldier fly (BSFL) can convert low-value organic substrates into their high-value larval biomass, thereby making BSFL highly suitable as an feed ingredient. An increasing number of studies underline the importance of the insect’s microbiota in this conversion process, but so far information on the microbiota of BSFL is scarce. Therefore, the overall goal of this PhD study was to gain more insight into the microbiota of H. illucens larvae.

First, we described the dynamics of the bacterial community composition of BSFL during six consecutive rearing cycles in three BSFL-producing companies. Samples of the BSFL and the corresponding substrates with which they were fed, were investigated by amplicon sequencing of partial 16S ribosomal RNA genes. The results showed that within one production cycle there was a clear shift in the community composition between the two time points. Within one rearing facility, the bacterial community compositions were quite stable over time, leading to a low inter-cycle variability. Moreover, the bacterial communities of the larvae for all companies were more similar to each other than these sampled at the first time point. This suggests that the community composition was the result of a selective process affecting the capability of specific bacterial species to colonize the larval gut. Additionally, some bacterial genera including Enterococcus and Providencia were found in each rearing facility.

In the second part of this research, we aimed to establish a representative culture collection of abundant microorganisms associated with BSFL as a starting point to investigate the functions of BSFL core gut bacteria. To particularly isolate the most abundant strains, only high serial dilutions of BSFL extracts were plated on a general agar medium. This resulted in a culture collection of 177 microorganisms, among which 171 bacteria and 6 fungi. A literature search showed that the established culture collection largely reflects the microbial communities of BSFL. Since larval performance on diets with a high fiber content is known to be low, another culture collection of microorganisms with potential to increase the digestibility of such substrates was established. To this end, five types of fibers (i.e., cellulose, hemicellulose, keratin, lignin, and pectin) were selected, and BSFL were reared on feeding substrates with a high content of one of those fibers. Homogenized larvae were then plated on minimal agar medium containing that type of fiber. This resulted in a total of 469 isolates, among which 67, 108, 77, 101, and 116 for the cellulose, hemicellulose, keratin, lignin, and pectin treatments, respectively. The majority of the isolates (> 75%) belonged to the genus Klebsiella. Furthermore, isolates from the cellulose and hemicellulose treatments were subjected to an evaluation of cellulase and hemicellulase activity, respectively. Based on the results from this screening, four isolates with the greatest enzyme activity (and all belonging to the genus Klebsiella) were selected for in vivo experiments. Specifically, these isolates and a Pediococcuspentosaceus strain (isolated from the guts of Tenebriomolitor) were added during rearing of BSFL to improve larval performance. To maximize the probability of obtaining probiotic effects, several modifications to the set-up were made in terms of mono- or multispecies inocula, time point of application, frequency of inoculation, inoculum level, and feeding regime. Despite those modifications, no improvement was observed in terms of larval growth, survival, feed conversion ratio, bioconversion efficiency, substrate reduction, and maximum larval weight.

Finally, it was assessed if the food pathogen Staphylococcus aureus could be taken up by BSFL when present in the insects feeding substrate (horizontal transmission). To this end, S. aureus was artificially inoculated into chicken feed at two different inoculum levels (3 and 7 log CFU/g) and its presence in the chicken feed and larvae was monitored over a course of 6 days by plating samples on a selective agar medium. S. aureus counts of the feeding substrates were found to decrease faster when BSFL were present in the feeding substrate compared to when no larvae were present. After 6 days past inoculation, the counts of S. aureus in the larval samples were below the detection limit for both inoculum levels. To explore whether the gut microbiota could play a role in this reducing effect, antimicrobial activity of the culture collection of abundant microorganisms against S. aureus was evaluated. While a majority (58%) of these isolates showed signs of low to moderate activity against S. aureus, the highest activity was observed for the fungal Trichosporon isolates. It is hypothesized that they play a role in the observed pathogen reduction, which should be further investigated in the future.