Partitioning the roles of insect and microbial enzymes in the metabolism of the insecticide imidacloprid in Drosophila melanogaster

Resistance to insecticides through enhanced metabolism is a worldwide problem. The Cyp6g1 gene of the vinegar fly, Drosophila melanogaster, is a paradigm for the study of metabolic resistance. Constitutive overexpression of this gene confers resistance to several chemical classes of insecticides, including the neonicotinoids exemplified by the insecticide imidacloprid (IMI). The metabolism of IMI in this species has been previously shown to yield oxidative and nitro-reduced metabolites. While levels of the oxidative metabolites are correlated with CYP6G1 expression, nitro-reduced metabolites are not, raising the question of how these metabolites are produced. Some IMI metabolites are known to be toxic, making their fate within the insect a second question of interest. These questions have been addressed by coupling the genetic tools of gene overexpression and CRISPR gene knock-out with the sensitive mass spectrometric technique, the Twin-Ion Method (TIM). Analysing axenic larvae indicated that microbes living within D. melanogaster are largely responsible for the production of the nitro-reduced metabolites. Knock-out of Cyp6g1 revealed functional redundancy, with some metabolites produced by CYP6G1 still detected. IMI metabolism was shown to produce toxic products that are not further metabolized but readily excreted, even when produced in the Central Nervous System (CNS), highlighting the significance of transport and excretion in metabolic resistance.

this strain upon exposure to IMI and IMI-5-OH respectively ( Figure 3).

2 4
A similar metabolic trend is observed with the over-expression of Cyp6g1 in the CNS Elav_Cyp6g1 larvae indicate that IMI can reach the CNS in a relatively short time where it is 3 2 7 metabolised by CYP6G1 into IMI-5-OH. While the detection of IMI-5-OH in the media in strains, the significantly higher levels of IMI-5-OH and IMI-Ole observed after 60 mins 3 3 0 indicate that metabolites produced in the CNS are excreted out of the larvae. For this to 3 3 1 occur, metabolites produced in the CNS must cross the blood brain barrier, diffuse into the 3 3 2 hemolymph and reach the malpighian tubules for excretion into the media. As the production 3 3 3 of IMI-Ole is not catalysed by CYP6G1, it is not clear whether it is produced in the CNS or 3 3 4 not.

5
Although resistance is attributed to the over-expression of Cyp6g1, CYP6G1 does not 3 3 6 actually detoxify IMI. It rather produces IMI-5-OH, which leads to the production of IMI- Ole. Both metabolites are toxic in D. melanogaster and are likely to bind to at least one have a lower binding affinity for nAChRs and/or if they are more readily excreted. That 3 4 1 excretion plays a vital role in the capacity of an insect to survive to insecticide exposure is 3 4 2 particularly evident with IMI-Ole, a metabolite produced at high concentration and known to 3 4 3 be more toxic than IMI in pest species like M. persicae, A. gossypii 38 and Bemisia tabaci 37 , 3 4 4 and in the honeybee Apis mellifera 36 . While the basis of these differences in toxicity between 3 4 5 species have not been established, they can be explained by mechanisms canvassed here. In IMI-NNO and IMI-NH. The capacity to transport and excrete these metabolites is crucial to strain) to IMI can be explained by its reduced capacity of eliminate the toxic metabolites IMI-3 5 1 5-OH and IMI-Ole, that accumulate within the insect body. CYP6G1 metabolises IMI, but 3 5 2 detoxification depends on the excretion of IMI and the metabolites produces. Our data, 3 5 3 combined with evidence linking candidate genes such as ABC transporters to insecticide 3 5 4 resistance 50,51 , suggest that the sequential activity of DMEs and transporters may be  Finally, the analysis of a Cyp6g1 knock-out mutant revealed the presence of show that the increased susceptibility to IMI observed with the RAL_517-Cyp6g1KO  Figure S4). However, the persistence of IMI-5-OH and IMI-Ole metabolites 3 6 6 in the RAL_517-Cyp6g1KO strain, albeit at lower levels, indicates that one or more 3 6 7 metabolic enzymes can contribute to the metabolism of IMI. That the same metabolites 3 6 8 normally produced by Cyp6g1 are detected suggest that a P450 may be involved in this 3 6 9 enzymatic conversion, possibly one closely related to Cyp6g1 52 . Cyp6g2 is one candidate. leads to both elevated levels IMI-5-OH and IMI-Ole and IMI resistance. Any mutation that 3 7 3 leads such a gene to be over-expressed in a similar tissue-specific pattern to Cyp6g1 or in the 3 7 4 CNS would be likely to confer insecticide resistance, in the absence of significant fitness  Chemicals. IMI (N-[1-[(6-Chloro-3-pyridyl)methyl]-4,5-dihydroimidazol-2-yl]nitramide) 3 8 0 and [ 13 C 6 ]-IMI (>99% isotopic purity; >97% total purity) were obtained from AK Scientific 3 8 1 (Union City, CA, USA) and IsoSciences (King of Prussia, PA, USA), respectively. Authentic and  differing only in Cyp6g1 tissue-specific expression levels, was generated by crossing the 3 9 8 5'HR_GAL4 and Elav_GAL4 driver lines to the Φ86FB genotype 54 referred to as 3 9 9 HR_Φ86FB and Elav_Φ86FB respectively. Flies were reared at 25 °C and raised in bottles  Table S1).  were collected and sterilized using a dilute solution of bleach (2.5% v/v) for 2 mins, followed  Insecticide exposure and metabolite extraction process. were gently transferred onto a mesh, rinsed with more sucrose solution and counted under a 4 1 7 microscope to ensure that only early third instar larvae were used in the experiments. For prepared. Larvae were exposed to a fresh solution of 50:50 ratio of 12 C 6 and 13 C 6 -IMI (total Bretonneux, France) operated at -10°C. IMI and its metabolites were then extracted using Clara, CA, USA). The ionisation of IMI and its metabolites was achieved using a dual- (v/v) mixtures of acetonitrile:H 2 O:formic acid, respectively. IMI and its metabolites were 4 4 0 analysed in both ionisation modes. Only IMI and its major metabolites IMI-5-OH and IMI- Ole were quantified. This was achieved using negative ionisation mode and through external See Volaric for technical assistance with the analytical instrumentation. Supplementary information accompanies this paper.

6 6
Competing financial interests: The authors declare no competing financial interests.

6 7
References  3  3  T  o  m  a  l  s  k  i  ,  M  .  ,  L  e  i  m  k  u  e  h  l  e  r  ,  W  .  ,  S  c  h  a  l  ,  C  .  &  V  a  r  g  o  ,  E  .  L  .  M  e  t  a  b  o  l  i  s  m  o  f  i  m  i  d  a  c  l  o  p  r  i  d  i  n   5  4  7   w  o  r  k  e  r  s  o  f   R  e  t  i  c  u  l  i  t  e  r  m  e  s  f  l  a  v  i  p  e  s   (  I  s  o  p  t  e  r  a  :  R  h  i  n  o  t  e  r  m  i  t  i  d  a  e