Enterobacteriaceae were isolated from 61 of the 94 arthropods: 3 of 3 Crustacea samples, 24 of 32 Myriapoda samples and 34 of 49 Hexapoda samples (Table 1). In total, 186 isolates were obtained: 114 from Hexapoda, 64 from Myriapoda, and 8 from Crustacea (Table 3). The most frequent bacterial species was Enterobacter spp. (n = 61), followed by Pantoea spp. (n = 26), Cedecea spp. (n = 18), Serratia spp. (n = 18), Hafnia spp. (n = 16), and Klebsiella spp. (n = 13). In this study, diverse Enterobacteriaceae were isolated from the arthropods. Enterobacter spp., the predominant bacteria isolated, have been responsible for intraperitoneal infection in humans [10]. The next most frequently isolated Enterobacteriaceae — Pantoea spp., Hafnia spp., Serratia spp., Cedecea spp., and Klebsiella spp. have been associated with opportunistic infections in humans. K. pneumoniae and K. oxytoca are known to cause urinary tract infections, pneumonia, and sepsis in patients with compromised immunity [11]. In contrast, some arthropods are known to form symbiotic relationship with these Enterobacteriaceae species to enable the breakdown of ingested food such as cellulose for easy digestion and nutrient assimilation [22].
Table 3
Bacterial species isolated from arthropods
Bacteria | | Subphylum: Sample No. (Isolate No.) | |
Genus | Species | | Hexapoda | Myriapoda | Crustacea | Total |
Enterobacter | cloacae | | 13(24) | 2(4) | 1(1) | 15(29) |
amnigenus | | 4(6) | 12(20) | 0 | 16(26) |
aerogenes | | 1(1) | 2(2) | 0 | 3(3) |
asburiae | | 2(3) | 0 | 0 | 2(3) |
Subtotal | | 19(34) | 16(26) | 1(1) | 36(61) |
Pantoea | Pantoea spp. | | 11(17) | 4(5) | 2(4) | 17(26) |
Cedecea | lapagei | | 5(9) | 1(2) | 1(1) | 7(12) |
davisae | | 3(6) | 0 | 0 | 3(6) |
Subtotal | | 8(15) | 1(2) | 1(1) | 10(18) |
Serratia | marcescens | | 6(10) | 1(2) | 0 | 7(12) |
liquefaciens | | 0 | 3(5) | 0 | 3(5) |
fonticola | | 0 | 1(1) | 0 | 1(1) |
Subtotal | | 6(10) | 4(8) | 0 | 11(18) |
Hafnia | Alvei | | 6(13) | 2(3) | 0 | 8(16) |
Klebsiella | pneumoniae | | 3(3) | 4(5) | 0 | 7(8) |
oxytoca | | 5(5) | 0 | 0 | 5(5) |
Subtotal | | 8(8) | 4(5) | 0 | 12(13) |
Escherichia | vulneris | | 2(2) | 3(3) | 0 | 5(5) |
Coli | | 0 | 1(2) | 0 | 1(2) |
hermannii | | 0 | 1(1) | 0 | 1(1) |
Subtotal | | 2(2) | 5(6) | 0 | 7(8) |
Rahnella | aquatilis | | 3(5) | 1(1) | 0 | 4(6) |
Citrobacter | freundii | | 2(3) | 0 | 0 | 2(3) |
youngae | | 1(1) | 2(2) | 0 | 3(3) |
Subtotal | | 3(4) | 2(2) | 0 | 5(6) |
Cronobacter | Cronobacter spp. | | 2(5) | 0 | 0 | 2(5) |
Raoutella | ornitholytica | | 0 | 1(1) | 1(1) | 2(2) |
planticola | | 0 | 1(1) | 0 | 1(1) |
Subtotal | | 0 | 2(2) | 1(1) | 3(3) |
Kluyvera | intermedia | | 1(1) | 2(2) | 0 | 3(3) |
Leclercia | adecaboxylata | | 0 | 1(1) | 1(1) | 2(2) |
Buttiauxella | agrestis | | 0 | 1(1) | 0 | 1(1) |
Total | | 34(114) | 24(64) | 3(8) | 61(186) |
Arthropod-derived Enterobacteriaceae isolated in this study were susceptible to all antimicrobial agents tested except CST and KAN. CST resistance was observed in 39 (29%) of the 134 isolates, excluding Cedecea spp., Hafnia spp. and Serratia spp. which have intrinsic resistance. The 39 isolates resistant to CST were found in Enterobacter spp. (25/61, 41.0%), Pantoea spp. (6/26, 23.1%), Klebsiella spp. (7/13, 53.8%) and Escherichia spp. (1/8, 12.5%). In Enterobacter spp., CST resistance was found in Hexapoda─ grasshopper (7), cricket (2), mealworm (1), dung beetle (1), butterfly (1) and Myriapoda─ millipede (1), centipede (1). In Pantoea spp., CST resistance was found in Hexapoda─ grasshopper (2), earwig (2). In Klebsiella spp., CST resistance was found in Hexapoda─ cricket (1), mealworm (1), assassin bug (1), earwig (1) and Myriapoda─ millipede (2), centipede (1). In Escherichia spp., CST resistance was only found in Hexapoda─ earwig (1), but no CST resistance in Crustacea (Table 4). The percentage of CST resistance was significantly high in Enterobacter isolates from Hexapoda (19/34, 55.9%) than Myriapoda (6/26, 23.1%: P < 0.05). Hexapods are obligate hosts of bacterial species, more so than any other group of arthropods [23], and as such, these bacterial species have developed resistance to survive the defense mechanism produced by the host [24].
Table 4
Colistin resistance observed in Enterobacteriaceae isolated from arthropods
Subphylum | English name | Enterobacteriaceae species: No. of Positive Samples (Isolate No.) | |
| | Enterobacter spp. | | Pantoea spp. | | Klebsiella spp. | | Escherichia spp. | |
| | No. tested | CST resistant | % | No. tested | CST resistant | % | No. tested | CST resistant | % | No. tested | CST resistant | % |
Hexapoda | Grasshopper | 9(19) | 7(11) | 57.9 | 5(9) | 2(3) | 33.3 | 0 | 0 | 0 | 1(1) | 0 | 0 |
| Cricket | 2(3) | 2(2) | 66.7 | 1(1) | 0 | 0 | 1(1) | 1(1) | 100 | 0 | 0 | 0 |
| Mealworm | 1(1) | 1(1) | 100 | 0 | 0 | 0 | 1(1) | 1(1) | 100 | 0 | 0 | 0 |
| Dung beetle | 2(2) | 1(1) | 50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Click beetle | 1(2) | 1(1) | 50 | 0 | 0 | 0 | 1(1) | 0 | 0 | 0 | 0 | 0 |
| Stag beetle | 2(2) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Lady beetle | 0 | 0 | 0 | 1(1) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Butterfly | 1(4) | 1(3) | 75 | 1(1) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Assassin bug | 1(1) | 0 | 0 | 0 | 0 | 0 | 3(3) | 1(1) | 33.3 | 0 | 0 | 0 |
| Earwig | 0 | 0 | 0 | 3(5) | 2(3) | 60 | 2(2) | 1(1) | 50 | 1(1) | 1(1) | 100 |
| Subtotal | 19(34) | 13(19) | 55.9 | 11(17) | 4(6) | 35.3 | 8(8) | 4(4) | 50 | 2(2) | 1(1) | 50 |
Myriapoda | Millipede | 12(18) | 1(3) | 16.7 | 2(3) | 0 | 0 | 3(4) | 2(2) | 50 | 1(2) | 0 | 0 |
| Centipede | 4(8) | 1(3) | 37.5 | 2(2) | 0 | 0 | 1(1) | 1(1) | 100 | 4(4) | 0 | 0 |
| Subtotal | 16(26) | 2(6) | 23.1 | 4(5) | 0 | 0 | 4(5) | 3(3) | 60 | 5(6) | 0 | 0 |
Crustacea | Pill bugs | 1(1) | 0 | 0 | 2(4) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Total | | 36(61) | 15(25) | 41 | 17(26) | 4(6) | 23.1 | 12(13) | 7(7) | 53.8 | 7(8) | 1(1) | 12.5 |
CST- colistin |
The mcr-1-5 genes were not detected in any of the CST-resistant isolates. The absence of mcr-1-5 gene in the CST-resistant isolates observed in this study prompted us to investigate the PhoPQ/PmrAB two-component system. We examined amino acid substitution in the PhoPQ/PmrAB two-component system in CST-resistant Enterobacter isolates and compared them to those of CST-susceptible Enterobacter strains. Our investigation showed various amino acid substitutions: three amino acid substitution in phoP (L129I, F141L, H207Q), five in phoQ (V102I, L133I, M298L, S448A, G464S), four in pmrA (A19G, S21A, N89T, L146Q) and four in pmrB (H132S, A172T, Q271V, R276Q). The amino acid substitutions observed in phoP, phoQ, and pmrA in this study corresponded with those reported by Uechi et al. [25]. On the other hand, the amino acid substitutions observed in this study were different from those reported in another study by Nawfal Dagher et al. [26]. We could not clarify the specific amino acid substitution responsible for the CST resistance observed. Arthropods rarely have direct contact with CST because of restrictions on CST usage for humans and pigs in Japan. Hence, the observed CST resistance was unexpected and as such it is not plausible to attribute this CST resistance to selective pressure or exposure to CST in the environment. Most arthropods have antimicrobial peptides in their hemolymph that serve as a defense mechanism and plays an important role in fostering symbiotic relationships with beneficial bacteria [6]. It therefore stands to reason that symbiotic bacteria may have developed resistance to antimicrobial peptides. The initial mode of action of antimicrobial peptides and CST against gram-negative bacteria involves binding to the LPS [6], this may have led to the development of cross-resistance to CST [27].
One isolate of K. oxytoca from a butterfly in this study showed resistance to KAN. WGS analysis revealed that the resistance gene aph(3')-Ia was located on the chromosome of K. oxytoca. The aph(3′)-Ia gene was first discovered on transposon Tn903 in Escherichia coli [28] and has subsequently been found on plasmids and chromosomes of clinical and veterinary Enterobacteriaceae isolates [29–30]. The presence of aph(3')-Ia in arthropods may suggest the possibility that arthropods received this bacterium from human and/or domestic animals or vice versa.
In previous studies, Enterobacteriaceae isolated from cockroaches showed resistance to more than two antimicrobial agents [9, 31]. In Japan, Enterobacter spp. isolated from companion animals showed resistance to CTX (33.3%), GM (23.3%), TC (40%), CPFX (43.3%), and CP (46.7%) and were reported to be extended spectrum beta-lactamase (ESBL) producers [32]. In addition, Klebsiella spp. isolated from companion animals were resistant to aminoglycosides and quinolones and were reported to be extended beta-lactamase producers [33]. However, susceptibility to antimicrobial agents were high, as expected in this study. The high susceptibility observed indicate an absence or low prevalence of AMR bacteria and minimal antimicrobial agents’ pollution in the immediate environment of the arthropods investigated.
The present study did not compare the isolation rate of bacteria on the external surface and alimentary tract of the arthropods. However, there was no significant difference between the isolation rate of bacteria on the external surface and the alimentary tract of cockroaches in a previous report [9].