Introduction
Dead bodies and detritus provide nutrient sources and an oviposition site for some insect species. In such environments, the newly hatched larvae are exposed to microorganisms, both immediately and as decomposition progresses. Such insects are under strong selective pressure to resist infection and maintain an effective immune response.
In addition to internal immune resources such as coagulation, phagocytosis, encapsulation and nodule formation (Sadd & Siva-Jothy, Reference Sadd and Siva-Jothy2006), some insects are equipped with external antimicrobial defences. An example is provided by Hoback et al. (Reference Hoback, Bishop, Kroemer, Scalzitti and Shaffer2004), who showed oral secretions from several Nicrophorus beetle species and anal secretions from one species of Silphinae were active against Vibrio fischerii bacteria. Early twentieth century studies demonstrated antibacterial activity in external excretions/secretions (ES) from diptera such as Lucilia sericata (Meigen), Calliphora species and Phormia terraenovae (Robineau-Desvoidy) (Simmons, Reference Simmons1935; Gwatkin & Fallis, Reference Gwatkin and Fallis1938; Pavillard & Wright, Reference Pavillard and Wright1957), species which are carrion feeders. Recent work has extended this knowledge in the context of L. sericata and shown that ES from sterile larvae is antibacterial against organisms such as MRSA (Thomas et al., Reference Thomas, Andrews, Hay and Bourgoise1999; Bexfield et al., Reference Bexfield, Nigam, Thomas and Ratcliffe2004; Kerridge et al., Reference Kerridge, Lappin-Scott and Stevens2005; Huberman et al., Reference Huberman, Gollop, Mumcuoglu, Block and Galun2007a). The antimicrobial activity has been identified from two fractions of the ES, one between 0.5 and 10 KDa and a second at less than 500 Da. The antimicrobial constituent from the latter fraction has been patented under the name ‘Seraticin®’ (Bexfield et al., Reference Bexfield, Bond, Roberts, Dudley, Nigam, Thomas, Newton and Ratcliffe2008).
The success of this recent work signals the need to review the antimicrobial capabilities of other insects inhabiting carrion and detritus in order to determine their potential for controlling a wider range of microorganisms. The present study investigated the antibacterial activities of un-induced ES from larvae of a comparable initial coloniser, Calliphora vicina (Robineau-Desvoidy) and two beetle species modelling colonisers of later stages of decomposition, Dermestes maculatus (DeGeer) and Tenebrio molitor (Linnaeus). The antibacterial activities of their un-induced ES are compared to that of ES from L. sericata larvae.
Materials and methods
Bacterial viable counts were used to assess the antibacterial potency of ES from each of the four larval species against five bacterial reference species: Staphylococcus aureus (ATCC 25923), Escherichia coli (ATCC 25922), Pseudomonas aeruginosa (ATCC 27853), Proteus mirabilis (ATCC 43071) and Bacillus cereus (NCIMB 3329). On at least two separate occasions, each experiment was replicated three times.
Marker insect species
Lucilia sericata, a member of the Calliphoridae, was used as the marker species as it has been studied intensively for decades as a result of its use in maggot debridement therapy (Simmons, Reference Simmons1935; Thomas et al., Reference Thomas, Andrews, Hay and Bourgoise1999; Mumcuoglu et al., Reference Mumcuoglu, Miller, Mumcuoglu, Friger and Tarshis2001; Lerch et al., Reference Lerch, Linde, Lehn and Grifka.2003; Bexfield et al., Reference Bexfield, Nigam, Thomas and Ratcliffe2004, Reference Bexfield, Bond, Roberts, Dudley, Nigam, Thomas, Newton and Ratcliffe2008; Kerridge et al., Reference Kerridge, Lappin-Scott and Stevens2005; Reinecke et al., Reference Reinecke, Mayer and Heisig2005; Daeschlein et al., Reference Daeschlein, Mumcuoglu, Assadian, Hoffmeister and Kramer2007; Huberman et al., Reference Huberman, Gollop, Mumcuoglu, Breuer, Bhusare, Shai and Galun2007b; Jaklic et al., Reference Jaklic, Lapanje, Zupancic, Smrke and Gunde-Cimerman2008; van der Plas et al., Reference van der Plas, Jukema, Wai, Dogterom-Ballering, Lagendijk, van Gulpen, van Dissel, Bloemberg and Nibbering2008). Calliphora vicina, also an initial coloniser, was used as a comparison to L. sericata since less is known about this species' antimicrobial excretion/secretion (Gwatkin & Fallis, Reference Gwatkin and Fallis1938). Dermestes maculatus colonises a corpse during the decay stages of decomposition (Kulshrestha & Satpathy, Reference Kulshrestha and Satpathy2001; Oliva, Reference Oliva2001; Grassberger & Frank, Reference Grassberger and Frank2004), and Tenebrionidae, such as T. molitor, have been associated with a corpse during the dry stages of decay (Mégnin, Reference Mégnin1894; Arnaldos et al., Reference Arnaldos, Garcia, Romera, Presa and Luna2005; Gennard, Reference Gennard2007).
Larval ES preparation
Colonies of the Calliphoridae and T. molitor were maintained under a lighting regime of 16:8 h (L:D), whilst cultures of D. maculatus were maintained in complete darkness. Lucilia sericata and C. vicina larvae were reared on a diet of ad libitum porcine liver and D. maculatus and T. molitor on dried porcine cubes supplemented with working dog biscuits. All colonies were maintained at a temperature of 25°C±3°C.
The excretion/secretion (ES) was collected from each species by adding a standardised amount of sterile, deionised water to a sample of larvae which were weighed and the numbers adjusted to give a fixed ratio of maggots to water (1 g ml−1). (Blowfly larvae were collected whilst in the third instar and beetle larvae from actively feeding stages with a standard length of 12±1 mm for D. maculatus and 26±2 mm for T. molitor). Larvae were incubated at 30°C for 60 min (a method adapted from Bexfield et al., Reference Bexfield, Nigam, Thomas and Ratcliffe2004), after which ES was collected and micro-centrifuged at 7826×g for five minutes and filtered (0.20 μm) to remove large particles and bacteria. The ES from each insect species was then stored frozen at −20°C until required.
Representative bacterial species
The bacterial species were chosen on the basis that they were representative of the changing micro-organism populations on a body during decomposition. There has been very limited research on the succession of microbial species on a corpse during decomposition, and information about species, location and growth requirements on live bodies and habitats were the criteria used to determine which bacteria were to be chosen: Staphylococci as common skin bacteria; E. coli and P. mirabilis from the gut; and B. cereus, P. mirabilis and P. aeruginosa associated with the environment, being present in soil and decomposing materials.
Preparation of bacteria
One colony was removed from a stock plate of nutrient agar (Oxoid Ltd, Basingstoke, Hampshire, England) and was inoculated into 20 ml sterile tryptone soya broth (TSB) (Oxoid Ltd, Basingstoke, Hampshire, England). The broth was incubated at 37°C with shaking for 17 h. A sample of 0.1 ml of the overnight bacterial culture was transferred to 10 ml TSB broth and incubated at 37°C with shaking until the optical density reading at 600 nm was in the range of 0.24–0.25.
Confirmation of antibacterial activity in ES from the marker species, Lucilia sericata
The antibacterial activity of L. sericata ES was quantified using liquid culture assays employing 10% TSB and dH2O or ES to confirm its antibacterial activity before being used as the marker species in the experiments reported in this paper. Results demonstrated that L. sericata ES had a significant inhibitory action and that there was >99% less bacterial growth in ES than controls throughout a 24-h period (data not shown).
Sample preparation
To investigate the antibacterial activity of ES from C. vicina, D. maculatus and T. molitor against the known activity of L. sericata, defrosted insect excretion/secretion (ES) from each species was separated into 4 ml aliquots. The purpose of this study was to model the antibacterial activity of ES in the insects' natural environment. Therefore, additional media was not employed in this assay, particularly as its effect on antibacterial activity of ES was not known. In order to demonstrate the pattern of normal bacterial growth, 4 ml aliquots of TSB/dH2O were utilised as controls to replace the ES. Twenty μl of bacteria in TSB was added to all aliquots and the universals incubated at 37°C, with aeration, for a 24-h period.
Effectiveness of bacterial control
Viable counts were prepared on nutrient agar at zero, four, eight and 24 h after inoculation for each of the four insect species and the controls. The agar plates were incubated at 37°C for 20–24 h and viable counts (CFU ml−1) were used to provide a quantitative determination of bacterial growth.
Confirmation of absence of contamination
To confirm the sterility of samples of ES prior to bacterial inoculation, a loop (10 μl) of ES from each insect, TSB, and sterile dH2O were spread separately onto nutrient agar and incubated at 37°C for 24 h. This procedure was repeated at four, eight and 24 h to ensure the media used to prepare dilutions for viable counts was contamination-free.
Statistical analyses
Statistical analyses were performed using SPSS (version 14.0) on log10 transformed mean count data sets to compare bacterial growth in ES from the three insect species in relation to growth in ES from L. sericata. The Bonferroni test was used as a post hoc test on those significantly different data sets confirmed by ANOVA. The results are presented as the mean bacterial count over a 24-h period.
The pH of insect ES
The pH of each aliquot of ES was recorded prior to each experiment in order to observe possible effects on antibacterial activity. No attempt was made to alter the pH of the ES as the intention was to observe its antibacterial activity in its natural state.
Results
A repeated measures ANOVA indicated that there was a significant difference between the mean bacterial counts in ES from each insect species (F3,27=5.43, P=0.005). A Bonferroni post hoc test revealed no significant difference between the antibacterial effectiveness of ES from L. sericata and C. vicina against the five bacterial species (P=1.00). Nor was any significant difference noted between the antibacterial activity of L. sericata and T. molitor (P=0.128). However, L. sericata was significantly more effective at inhibiting the growth of the five bacterial species than D. maculatus (P=0.006).
Staphylococcus aureus (ATCC 25923)
All insect species allowed some growth of S. aureus (fig. 1). Of the species compared, L. sericata was shown to be the most effective source of antibacterial ES for controlling S. aureus.
Fig. 1. A typical 24-h growth curve with aeration of Staphylococcus aureus in ES from Lucilia sericata, Calliphora vicina, Dermestes maculatus and Tenebrio molitor (error bars indicate 95% confidence intervals). The control represents normal bacterial growth (
, L. sericata ES; 
, C. vicina ES; –•–, D. maculates ES; 
, T. molitor ES; – –, control).
Escherichia coli (ATCC 25922)
Both Calliphorids exhibited good bactericidal activity against E. coli. Lucilia sericata and C. vicina ES inhibited E. coli growth over the 24-h period (fig. 2), reducing the mean bacterial count by 84% and 63%, respectively. Over the first four hours, C. vicina ES reduced the initial count of E. coli by 55%; over an 8-h period, the bacterial count was 91% less than the original inoculum. In contrast, bacteria treated with ES from T. molitor and D. maculatus showed exponential growth so that over eight hours there was a 102 increase in bacterial population, indicating that they were not an effective means of control of this species.
Fig. 2. A typical growth curve over a 24-h period with aeration of Escherichia coli in ES from Lucilia sericata, Calliphora vicina, Dermestes maculatus and Tenebrio molitor (error bars indicate 95% confidence intervals). The control reflects normal bacterial growth (, L. sericata ES; , C. vicina ES; –•–, D. maculates ES; , T. molitor ES; – –, control).
Bacillus cereus (NCIMB 3329)
There was also a significant difference in the antibacterial effectiveness of the four insects' ES in controlling bacterial growth over a 24-h period. In the first four hours, both L. sericata and C. vicina ES (fig. 3) demonstrated a 100% bacterial count reduction of B. cereus. They were, therefore, much more efficient at inhibiting the growth of B. cereus than the ES of either D. maculatus or T. molitor, although the latter inhibited bacterial growth in the first four hours.
Fig. 3. A typical Bacillus cereus growth curve over a 24-h period with aeration in ES from Lucilia sericata, Calliphora vicina, Dermestes maculatus and Tenebrio molitor (error bars indicate 95% confidence intervals). The control represents normal bacterial growth (, L. sericata ES; , C. vicina ES; –•–, D. maculates ES; , T. molitor ES; – –, control).
Pseudomonas aeruginosa (ATCC 27853)
P. aeruginosa was effectively inhibited over 24 h by L. sericata ES (fig. 4) and ES from both blowflies was bactericidal during the first four hours of growth. Over the first four hours, C. vicina ES was the more effective and reduced the mean bacterial count by 85%, whilst L. sericata ES only reduced it by 58%. However, over 24 h, L. sericata ES maintained its bacteriostatic activity (55% mean bacterial count reduction), whereas C. vicina ES allowed re-growth to occur in the period between eight and 24 h. In contrast, ES from both D. maculatus and T. molitor were an ineffective means of controlling P. aeruginosa since both allowed growth throughout the 24-h period.
Fig. 4. A typical growth curve of Pseudomonas aeruginosa in ES from Lucilia sericata, Calliphora vicina, Dermestes maculatus and Tenebrio molitor over a 24-h period with aeration (error bars indicate 95% confidence intervals). The control reflects normal bacterial growth (, L. sericata ES; , C. vicina ES; –•–, D. maculates ES; , T. molitor ES; – –, control).
Proteus mirabilis (ATCC 43071)
Interestingly, beetle ES was much more effective against P. mirabilis, whilst blowfly ES had no effect (fig. 5). By the end of a four-hour period, there was a 60% mean bacterial count reduction in D. maculatus ES and a 66% mean bacterial count reduction in T. molitor ES although, after this point, D. maculatus ES allowed bacterial re-growth. Tenebrio molitor ES continued to inhibit the growth of P. mirabilis for a further four hours (77% mean bacterial count reduction). Therefore, T. molitor ES was more efficient at inhibiting the growth of P. mirabilis than ES from L. sericata, C. vicina or D. maculatus.
Fig. 5. Line graph showing a typical growth curve of Proteus mirabilis in ES from Lucilia sericata, Calliphora vicina, Dermestes maculatus and Tenebrio molitor over a 24-h period with aeration (error bars indicate 95% confidence intervals). The control represents normal bacterial growth (, L. sericata ES; , C. vicina ES; –•;–, D. maculates ES; , T. molitor ES; – –, control).
The effect of pH on antibacterial activity of ES
The pH of ES from each insect had different ranges: L. sericata (8.67–8.82); C. vicina (8.53–8.68); D. maculatus (6.00–6.14); and T. molitor (5.14–5.28). However, a two-tailed Pearson's test demonstrated that there was no significant linear correlation between the pH of insect ES and its antibacterial potency against the five bacterial species tested. This was true for L. sericata ES (r=−0.09, n=10, P=0.815), C. vicina ES (r=0.07, n=10, P=0.859), D. maculatus ES (r=0.109, n=10, P=0.776) and T. molitor ES (r=−0.448, n=10, P=0.194). Therefore, pH did not influence antibacterial activity of ES from these species.
Discussion
All four insect species demonstrated differences in their potency against each of the five species of bacteria. However, viable counts showed that, at the concentration tested, ES from all species was capable of reducing bacterial numbers for at least part of a 24-h period (figs 1–5); no lag phase was observed during this period.
The viable count results for L. sericata ES confirm those reported by Bexfield et al. (Reference Bexfield, Bond, Roberts, Dudley, Nigam, Thomas, Newton and Ratcliffe2008), who showed a 97.4% decrease in E. coli populations over a six-hour period. In contrast, research by Jaklic et al. (Reference Jaklic, Lapanje, Zupancic, Smrke and Gunde-Cimerman2008) established a prolonged lag phase of 5–6 h for E. coli, after which exponential growth occurred. He demonstrated this same lag phase for P. aeruginosa, which is also in contrast to the results reported here.
However, an equivalent level of antibacterial potency against S. aureus reported by both Bexfield et al. (Reference Bexfield, Bond, Roberts, Dudley, Nigam, Thomas, Newton and Ratcliffe2008) and Jaklic et al. (Reference Jaklic, Lapanje, Zupancic, Smrke and Gunde-Cimerman2008) was not demonstrated in this study. Bexfield et al. (Reference Bexfield, Bond, Roberts, Dudley, Nigam, Thomas, Newton and Ratcliffe2008) showed that, after a six-hour period, L. sericata ES was able to reduce the population of S. aureus (106) by 91.5%, and Jaklic et al. (Reference Jaklic, Lapanje, Zupancic, Smrke and Gunde-Cimerman2008) demonstrated a 100% reduction of S. aureus after eight hours. These differences may be due to variation in concentration of ES, the types of media used for the liquid culture assay or the duration of the experimental period between these three studies.
Previous researchers have suggested that the pH of insect ES influenced its antibacterial activity (Gwatkin & Fallis, Reference Gwatkin and Fallis1938; Thomas et al., Reference Thomas, Andrews, Hay and Bourgoise1999; Bexfield et al., Reference Bexfield, Nigam, Thomas and Ratcliffe2004). Therefore, the origin of the antibacterial activity demonstrated in the insect ES was explored by investigating the pH ranges of ES from each insect species. Typically, the ES from L. sericata and C. vicina was alkaline (pH 8–9) compared to the ES from D. maculatus and T. molitor, which was found to be acidic (pH 5–6). However, there was no significant correlation between the pH of insect ES and bacterial growth, indicating pH had no effect on antibacterial potency. These results support work conducted by Bexfield et al. (Reference Bexfield, Nigam, Thomas and Ratcliffe2004), who showed that antibacterial activity in L. sericata ES was not dependant on the alkaline condition, by changing the pH of L. sericata ES and demonstrating the same antibacterial activity existed in aliquots that were altered to pH 6 and 7 as was in native ES (pH 8–9). Therefore, it appears that the effectiveness of insect ES relates to agents other than pH.
The fact that larval ES did not always maintain an antibacterial effect over the 24-h experimental period does not minimise its potential as an antibacterial agent. Data from several experiments demonstrated that larval ES is produced continuously (unpublished data); therefore, even short-lived antibacterial activity would still be an effective defence against microbes on a corpse.
The difference in antibacterial potency and spectrum between the blowflies and the beetles is not altogether surprising and may be explained ecologically by the fact that blowflies tend to frequent habitats with potentially more diverse species of bacteria present. As the initial colonisers of a corpse, blowflies encounter bacteria originating from the human body. As a corpse decomposes, production of waste gases extends the body until, eventually, the skin is broken, allowing the gut bacteria (and gases) to escape. It is assumed that most obligate anaerobes will die off at this stage; but facultative anaerobes, such as Proteus and E. coli, will survive and be in direct competition with the insects now feeding on the corpse.
Beetle larvae are generally present on a corpse in the later stages of decomposition (advanced and post decay), when much of the soft tissue has been removed. It is assumed that most human-derived bacteria have been reduced in number by earlier insect colonisers and that bacteria from the environment probably dominate. Bacterial species, such as P. mirabilis, against which the beetle but not blowfly larvae ES in this study were active, may potentially be present at the time of coleopteran colonisation.
The blowflies, L. sericata and C. vicina, produced excretion/secretion (ES), which was active against both the Gram-positive and Gram-negative bacteria tested in this study. This secretion/excretion had a broader spectrum and a more potent antibacterial activity than the ES of either of the beetles, D. maculatus and T. molitor. The blowflies were able to inhibit or reduce growth of four out of the five bacterial species used, whereas the beetles were only effective against one, P. mirabilis. These results indicate the potential of using ES from carrion-feeding coleoptera as antibacterial agents, as well as those of Calliphoridae, and that the poorer antibacterial capacity may reflect the bacterial contact in the habitat to which the beetles, in contrast to the Calliphorids, are exposed.
Acknowledgements
We would like to thank John Flynn for statistical advice and support.
