Disruption of the odorant coreceptor Orco impairs foraging and host finding behaviors in the New World screwworm fly

The evolution of obligate ectoparasitism in blowflies (Diptera: Calliphoridae) has intrigued scientists for over a century, and surprisingly, the genetics underlying this lifestyle remain largely unknown. Blowflies use odors to locate food and oviposition sites; therefore, olfaction might have played a central role in niche specialization within the group. In insects, the coreceptor Orco is a required partner for all odorant receptors (ORs), a major gene family involved in olfactory-evoked behaviors. Hence, we characterized the Orco gene in the New World screwworm, Cochliomyia hominivorax, a blowfly that is an obligate ectoparasite of warm-blooded animals. In contrast, most of the closely related blowflies are scavengers that lay their eggs on dead animals. We show that the screwworm Orco orthologue (ChomOrco) is highly conserved within Diptera, showing signals of strong purifying selection. Expression of ChomOrco is broadly detectable in chemosensory appendages, and is related to morphological, developmental, and behavioral aspects of the screwworm biology. We used CRISPR/Cas9 to disrupt ChomOrco and evaluate the consequences of losing the OR function on screwworm behavior. In two-choice assays, Orco mutants displayed an impaired response to floral-like and animal host-associated odors, suggesting that OR-mediated olfaction is involved in foraging and host-seeking behaviors in C. hominivorax. These results broaden our understanding of the chemoreception basis of niche occupancy by blowflies.

(A) Multiple sequence alignment of Orco orthologues in Diptera. Codonbased alignment was constructed using the Muscle algorithm implemented in MEGA7 1 . Black arrows indicate the locations of the six introns identified in the ChomOrco genomic region. Purple boxes indicate the positions of the seven transmembrane domains (TM1 to TM7), as predicted by TOPCONS 2 . Alignment is shown in the context of the evolutionary relationships among Diptera species, according to Junqueira et al. 3 . Species abbreviations and accession numbers are given in Supplementary Table S3 Table S4).

Supplementary Fig. S4
CRISPR/Cas9 dual-targeting approach. In addition to the PCR-based genotyping methods used, we evaluated the possibility of inducing large deletion events between Cas9 targeted regions by injecting RNPs multiplexed with both sgRNAs. (A) The goal was to introduce a ~2.2 Kb deletion at the ChomOrco loci between sgR-Orco-E1 and sgR-Orco-E2b targeted sites. (B) Large deletions would be easily distinguished from the wildtype (wt) allele by routine PCR, and electrophoresis. A degree of deletion events was observed from G0 individuals developing from dual-targeting microinjections, but with low efficiency. Complete ˜2.2 kb deletions were rarely seen, and none large deletions were observed after crossing dual-targeting survivors with wt flies. These results may reflect the unbalanced efficiency of the designed sgRNAs (referent to results in Fig. 3). Abbreviations used: wt = wildtype sample; Ldd = 100 bp DNA Ladder (NEB). Primers genOrco-F1 and genOrco-R2 were used (see Supplementary Table  S4). (C) Sequencing confirmation of target-specific deletions between exons E1 and E2b of ChomOrco.  3A) were conducted using the primers genOrco-F1 along with genOrco-R1, and genOrco-F2 along with genOrco-R2, respectively (Supplementary Table S4).

Supplementary Fig. S6
Non-lethal tissue sampling for DNA isolation. (A) Upon emergence, adult flies are individually captured in a 30 mL glass vial (panels I and II). The captured flies are gently held and turned upside down to expose their legs (panels III and IV). A single midleg is then sectioned at the proximal coxa using a fine forceps (panel IV). Legs are transferred to a 1.5 mL centrifuge tube and maintained at -80° until DNA extractions (panel V), while adults are transferred to an individual plastic cage provided with food and water (panel VI). (B) To evaluate the amputation effects on survival and fertility, dissected adult flies were transferred from the individual cages to mating assay cages (three days after leg amputation). Test cages (n = 6) were founded with ten amputated individuals each (males and females, ratio 1:1), while sham individuals (flies that passed through the same sampling procedure but dissection) were used in control cages (n = 6). Survival rates were evaluated six days after adult emergence, and females were individually stimulated to oviposit inside 30 mL glass vials. Viable offspring production was checked on the next day. Bars marked with the same letters are not significantly different as given by two-tailed Student's t-test. Results showed that the non-lethal tissue sampling has little or no hazardous effects on screwworm survival and fertility. Abbreviations used: s (screwworm); g (glass vial); i (index finger); f (forceps); m (midleg); t (tube); d (diet); w (water); p (gauze perch); h (holes). Photo credit to Stephanie Mladinich.  Table S4). Amplifications are resolved in a high-resolution 5 -6% Agarose-1000 TM (Invitrogen), prepared in 1x Tris-Borate-EDTA (TBE) buffer and stained with 3 μl of ethidium bromide (10 mg/mL). Samples are mixed with 3 μl of Gel Loading Dye (NEB) and loaded into the gel. Electrophoresis is performed for at least 2 h at 5 volts/cm in a cold 1x TBE buffer. The gel image displays an example of the expected results from the HRMob analysis using control samples. Only the informative area of the gel is shown; from the top wells of the gel to the last band in the molecular marker (0.1 Kb in the ladder well). Authors were unable to provide a full-length image for this gel.

Supplementary Fig. S9
Control slides for immunostaining of ChomOrco. Wildtype (wt; leftmost) samples showed that ChomOrco protein is found in the cell body (white arrows) and dendrites (black arrows) of olfactory sensory neurons (OSNs). Control slides (rightmost) showed no specific labeling, but some nonspecific background in the absence of primary OR83b-IC3 antibody (same gain/exposition settings were used for these images and samples shown in Fig. 3F).

Supplementary Fig. S10
Response of wt screwworms (ChomOrco wt ) to honey (honey jar) or glycerol (drop) in two-choice trap assays. Screwworm adults distribute evenly between traps when both contain honey (Student's t-test; p = 0.42, n = 11), and show little or no response to glycerol (n = 9), indicating that they don't display preferences to any cage sides in our setup. However, flies show a strong preference for honey in opposition to glycerol (p < 0.001, n = 12). Preference Index was calculated as: PI = (ntrap / ntotal) * 100, where ntrap is the number of flies captured in a given trap, and ntotal the total of flies in the test cage.

Supplementary Table S1
Sequence divergence between ChomOrco and other dipteran's orthologues. Uncorrected evolutionary distances (p-distance) based on nucleotide (nt), and amino-acid (aa) sequences were estimated using pairwise comparisons with pairwise sitedeletions in MEGA7 1 . Results are shown as Avg ± SE.

Supplementary Table S2
Mutational inheritance of Orco mutant alleles. Injected screwworm males (putative founders) were genotyped using our non-lethal DNA extraction method and T7EN1 cleavage assay. Ten confirmed heterozygous flies (Orco wt/-) were randomly selected and individually backcrossed to wt screwworm females. For each crossing, another 8 males (on average) were randomly sampled and genotyped as before. Transmission efficiency was found to be 90% (9 out of 10 crosses produced heterozygous offspring), while inheritance success (germline transmission) ranged from 14.3% to 88.9%. In total, 41% (n = 38 / 81) of the genotyped flies were found to be heterozygous.

Supplementary Table S3
Specification of species and Orco sequences used in the evolutionary analyses.

Supplementary Table S4
Specifications for all the primers used in this study. a These genotyping primers were also used for Illumina library preparation by including Multiplexing Read 1 (5' -CAC TCT TTC CCT ACA CGA CGC TCT TCC GAT CT -3') and Read 2 (5' -GTG ACT GGA GTT CAG ACG TGT GCT CTT CCG ATC T -3') sequences at the 5'-end of the forward and reverse genotyping primers, respectively. B These primers were previously evaluated by Cardoso et al. 7 , and have been widely accepted for functional studies in blowflies due to their consistent results (e.g., Ct values) within and among different species of blowflies and their developmental stages. c sgRNAs used in this study, PAM motif in red. Syntheses were performed as described by Bassett 10 . We considered sgRNAs: (1) targeting ChomOrco exons 1 and 2b, aiming the early disruption of the Orco gene in order to avoid truncated protein activity; (2) with the smaller number of potential off-targets, with preferentially ≥ 3 mismatches between the sgRNA and non-specific genome sequences; (3) with the maximum amount of mismatches present on the 5 first bases of the sgRNA directly upstream to PAM motif, which constitute the so called "seed" region; (4) starting with an "G" base, to improve T7 transcription initiation, and; (5) with an optimal genomic location for the design of genotyping primers.

Long-PCR Product Sequencing
Total DNA was extracted from frozen adult flies using the DNeasy Blood and Tissue Kit (Qiagen).
Long-PCRs were performed in a 50 µl reaction, containing 0.2 µM of each forward and reverse primers (see Supplementary Table S4)

Intrapuparial Development
Screwworm pupae were collected three, six and eight days after pupation and fixed in Carnoy solution (EtOH: Acetic Acid: Chloroform, 6:3:1 (v/v)) for 48 h. Samples were then transferred to 5% formic acid for 48 h and transferred to 70% EtOH until imaging. Specimens were dissected with fine forceps and photographed with a Nikon P-FLAP 2 stereomicroscope system.

Electron Microscopy
Three-day-old adult females were freshly collected, fixed in absolute EtOH, and dehydrated through a graded EtOH series substituted with isoamyl acetate. Specimens were dried on a

Illumina Sequencing
Genomic DNA from G0 flies transiently expressing the ZsGreen marker ( Supplementary Fig. S2) were used as template in PCR amplifications spanning the Cas9-targeted sites. Amplicons were gel extracted using the QIAquick Gel Extraction Kit (Qiagen) and used as templates in a second PCR to incorporate Illumina adapters and barcodes (Supplementary Table S4). Pooled libraries were sequenced on an Illumina MiSeq (250 bp, paired-end run). Resulting raw reads were cleaned using Trimmomatic 12 , connected with COPE 13 , and aligned to the wt reference sequence using BWA-MEM 14 . Alignments were inspected for the presence of indels using CRISPResso 15 and CrispRVariants 4 pipelines.

Non-lethal DNA isolation
Genomic DNA extractions were carried out from a single midleg of adult flies (Supplementary Fig.   S4). Dissected legs were frozen at -80 °C for at least 20 min and homogenized in 50 μl of fresh prepared Proteinase K solution (0.5 mg/mL Proteinase K, 10 mM Tris-HCL, 1 mM EDTA, and 25 mM NaCL in ddH2O). The mixtures were incubated at 37 °C for 60 min followed by enzyme inactivation at 95 °C for 10 min. Extractions were kept at -20 °C until genotyping assays.

Trap's crafting
Traps were handmade of plastic soft drink bottles and contained two parts (Fig. 4B). A collection chamber was made of the upper cone-shaped funnel part of the bottle (60 mm height and 70 mm Ø), and a bait chamber made with the bottom part of the bottle (30 mm height). Two entries (35 mm apart from each other) were made in the collection chamber with an "X" shape cut. The resulting triangular portions were folded-in and used to support a 1 mL pipette tip cuted at the proximal end, which allows flies to enter the traps but prevents them from escaping. Baits were placed in the center of plastic dishes in the inner bottom of the bait chamber (35 mm Ø).

Two-choice assay movie
Supplementary Movie S1 shows that screwworm Orco mutants display Impaired host-seeking behavior (Related to Fig. 4). The movie displays two-choice trap assays showing side-by-side the response of 6-to-9-days-old female wt flies (leftmost) and ChomOrco 16 mutants (rightmost) to oviposition media. Orco mutants display a lack of decision-making and impaired flight orientation towards the stimuli source, while wt female flies are strongly attracted to the odors released from the oviposition device. The assay was recorded using a Canon EOS Rebel T3i under a monochrome mode. For illustration purposes, the movie plays only the first 5 min of each assay at 16X speed.