Insect rearing

M. sexta was reared in our laboratory on an artificial diet (Grosse-Wilde et al. 2011). Larvae were kept in a climate chamber with a 14-h light/10-h dark cycle at 26 °C during the light cycle and 24 °C during the dark cycle and a relative humidity of 60%. After pupation, male and female pupae were kept in separate climate chambers, with a 16-h light/8-h dark cycle at 25 °C and relative humidity of 60% during the light cycle and 70% during the dark cycle. Emerging adults were collected daily and individually housed in brown paper bags (17 cm × 26 cm) in the pupal chambers.

Scanning electron microscopy (SEM)

The epiphysis was cut from the tibia of the foreleg with a pair of curved microscissors, and the scales covering the epiphysis were removed with a toothpick. The samples were dehydrated by repeated washing in 70% ethanol, then placed in 1.5 ml Eppendorf tubes on tissue and air-dried for at least 24 hs. The epiphyses were mounted on a holder covered with adhesive tape and sputter-coated with gold before examination with a scanning electron microscope (LEO 1450 VP, Zeiss, Germany).

Test for antennal grooming

To evaluate whether the epiphysis of M. sexta plays a role in antenna cleaning, we applied a small amount of fluorescent powder (DayGlo, USA) with a toothpick to the epiphysis of the right foreleg of a 3-day-old male moth, leaving the left leg untreated as a control. The moth was then placed in a mesh cage (40 cm x 40 cm x 55 cm) in a climate chamber with a 16-h light/8-h dark cycle at 25 °C and relative humidity of 60% during the light cycle and 70% during the dark cycle. After 24 h, the moth was placed in a freezer for 2 days, and then both antennae and epiphyses were examined under a microscope (Axio Zoom, Zeiss, Germany).

Electrophysiology

The electroepiphysisogram (EEG) was developed as an adaptation of the electroantennogram (EAG). For this recording technique, an epiphysis was dissected from the foreleg and attached to two steel electrodes (‘recording fork’) with conductive gel (Spectra 360 electrode gel, Parker Laboratories) after cutting off a small portion of the epiphysis tip. In addition, we performed single sensillum recordings (SSR) from individual sensilla on the epiphysis. For this purpose, the epiphysis was cut together with a small part of the tibia to provide an area for the grounding electrode. The tungsten recording electrode was inserted into the base of a sensillum.

A constant flow (0.5 l/min) of charcoal-filtered and humidified air was delivered through an aluminum tube (length: 11 cm), with the outlet positioned 1–2 cm from the epiphysis. Ten µl of the odor stimulus was pipetted onto a filter paper disk (diameter: 1.2 cm) placed inside a glass Pasteur pipette. The tip of this Pasteur pipette was inserted into a small hole in the aluminum tube. For odor stimulation, an airstream (0.4 l/min) was delivered through the Pasteur pipette into the continuous airstream for 200 ms (CS-55 Stimulus Controller, Syntech, Germany). The signals were digitally converted (IDAC-4 USB, Syntech, Germany), visualized, and recorded on a PC using the software Autospike (Syntech, Germany).

To analyze the EEG and EAG data, the maximum deviation from baseline, i.e., the amplitude, was determined for each experiment. At the beginning and end of the sequence of odor stimuli tested with an epiphysis or antenna, a control stimulus using solvent was done. The average amplitude elicited by these two control stimulations was calculated and subtracted from the amplitude elicited by each odor stimulus. This provided the solvent-subtracted EEG or EAG response.

Odor stimuli

Twenty-five synthetic odorants (Table S1) were diluted in hexane to a concentration of 10−2.

To prepare a female pheromone gland extract, we dissected the glands of ten 3-day-old virgin females at five hours into the scotophase, when females are most attractive to males (Allen and Hodge 1955). The glands were immersed in 500 µl hexane and placed on a shaker for 1 h. Ten µl of the supernatant was used per stimulation, corresponding to 0.2 female gland equivalents (FGE), a concentration within the range of behavioral attractiveness (0.002 to 2 FGE) (Doolittle et al. 1991).

To collect plant headspace, a non-flowering D. wrightii plant (potted) or a single flower from a potted D. wrightii plant was enclosed in a polyethylene terephthalate bag (Toppits, Germany). Charcoal-filtered air was pumped into the bag through a silicone tube connected to a custom-made pump. The odor-enriched air exited the bag through a second silicone tube that passed through a volatile collection trap (Porapak-Q 25 mg, https://www.volatilecollectiontrap.com). Volatile collection was done in a climate chamber with a 14-h light/10-h dark cycle at 25 °C (day) and 22 °C (night) and relative humidity of 57% during the light cycle and 65% during the dark cycle. After 24 h, the traps were removed and eluted with 400 µl hexane.

Tissue collection and RNA extraction

We studied the expression of chemosensory receptor genes in the epiphyses of male and female moths, both virgin and mated, on day 3 after eclosion. Mating took place on day 2 after eclosion. For RNA extraction, three pairs of epiphyses, i.e., epiphyses from three animals, were pooled per sample, and three samples were prepared for each experimental group (virgin males, mated males, virgin females, and mated females). Tissues were cut from the forelegs and immediately pestled in liquid nitrogen using a mortar containing 1.5 ml of TRI Reagent (Sigma Aldrich, USA). The resulting mixture was transferred to a 2 ml Eppendorf tube. After this step, we followed the manufacturer’s protocol (Direct-zol RNA Miniprep Kits). The total RNA concentration per sample was 30–40 ng/µl. The number of ORs, IRs, and GRs detected in the epiphysis is higher than in a recent study that examined the expression of chemosensory receptor genes in the foreleg of M. sexta using the same technique (Tom et al. 2022), probably because the former study extracted RNA from the entire leg, thereby diluting the mRNA copies of these receptors. However, all chemosensory receptor genes (except MsexOR8) previously found in the entire foreleg were also expressed in the epiphysis.

NanoString gene expression assay

We used the nCounter XT CodeSet gene expression assay (NanoString Technologies, Inc., USA). The custom CodeSet (Zhang et al. 2022) contained 268 probes targeting 71 ORs, 29 IRs, 49 GRs, 47 odorant-binding proteins, 5 pickpocket, 3 sensory neuron membrane proteins, and 62 candidate reference gene transcripts. Some receptor sequences had high homology with duplicates, preventing the design of unique probes, and therefore had to be excluded (Tom et al. 2022). We followed the standard protocol described in the nCounter XT Gene Expression Assay User Manual (MAN-10023-11, page 16). Since MsexABPx, MsexOBP1, MsexOBP5, and MsexOBP6 had very high expression levels, an attenuation mix (Eurofins Genomics, Germany) was used to suppress these counts.

For the hybridization step, we prepared a master mix of 42 µl Reporter CodeSet, 28 µl Reporter-Plus reagent, and 70 µl nCounter SPRINT hybridization buffer. Each hybridization reaction combined 10 µl master mix, 5 µl total RNA (30–40 ng/µl RNA), 1 µl attenuation mix, and 3 µl of a mixture consisting of Capture ProbeSet and Capture-Plus reagent. Hybridization was done at 65 °C for 22 h, after which 16 µl Merck water was added to the sample. The total volume was loaded onto the nCounter SPRINT Cartridge and processed on the nCounter SPRINT Profiler. Raw data was processed using nSolver4.0. Quality control of the mRNA data was done using default parameters for the nCounter SPRINT Profiler according to the NanoString Gene Expression Data Analysis Guidelines (MAN-C0011-04). The parameters were Imaging QC: 75; Binding Density QC: 0.1–1.8; Positive Control Linearity QC: 0.95; Positive Control Limit of Detection QC: 2 standard deviations. Two normalization steps were then performed, first using the geometric mean counts of six external positive control probes and second using the geometric mean counts of at least three endogenous reference genes selected based on their coefficient of variation (CV). The endogenous reference genes used for the second normalization step were msex02_01637RB, msex02_11794RA and msex02_13396RA with CV < 40%. After these two normalization steps, the minimum normalized value for each sample was defined as background, and any chemosensory receptor gene with values above this background in at least two of the three samples was considered to be expressed in that experimental group.

Reverse-transcriptase PCR

To clarify the expression of the OR co-receptor ORCo and the IR co-receptors IR8a, IR25a, and IR76b in the epiphysis, RNA extracted from female and male epiphyses and from male antennae (positive control) was used to synthesize cDNA with the Superscript III Reverse Transcriptase Kit (Thermo Fisher Scientific, Germany). To amplify the genes, PCR was performed with Phusion™ High-Fidelity DNA Polymerase (New England Biolabs, Germany) according to the manufacturer’s protocol and the primers in the table below at an annealing temperature of 60 °C. The size of the PCR products was visualized and analyzed with gel electrophoresis.

Gene

Forward primer

Reverse primer

MsexORCo

ATGATGGCCAAAGTGAAAACACAGG

CTATTTCAGCTGCACCAACACCATG

MsexIR8a

AAGAGCAGTGAAAGAGAAGTTAGTGCGC

TCCACACCCTGTAAAGTGTGTCTTCTG

MsexIR25a

ATGTTATCAGCGAAAAAGACTCCTCACGTC

TCAAAATTTAGGTTTCAAATTAGATAAACCTAAATTTC

MsexIR76b

ATGGCCGGGATCGAGCTCATTATATC

TTATCGATACAGAAAAGCAGAAGGCGCTC

Mating experiments

To test the effect of the epiphysis on mating success, both epiphyses of one sex were removed during the inactive (light) phase on the day of eclosion. The control animals were handled in the same way and for the same amount of time, but without removing the epiphyses (“mock surgery”). On the third day after eclosion, individuals of the epiphysectomized moths were then allowed to mate with a control animal of the opposite sex in a Plexiglas mesh cage (30 cm x 30 cm x 30 cm) during the active (dark) phase. Females were placed in the cages at the beginning of the dark phase, and males were added 4 h later. Cages were then observed for 60 min, and the time of onset of copulation was noted.

Wind tunnel experiments

To test the effect of the epiphysis on feeding and oviposition behavior, we removed the moths’ epiphyses on the day of eclosion during the light phase. The control animals were handled in the same way and for the same amount of time, but without removing the epiphyses (“mock surgery”). Experiments were performed on the third day after eclosion, either with virgin moths (males and females) or after mating on the second day (females). We conducted the experiments during the active phase of the moths in a Plexiglas wind tunnel (250 cm long x 90 cm wide x 90 cm high) at 25 °C, 70% relative humidity, and a wind speed of 44 cm/s. Individual moths were transferred to a plastic mesh cylinder (15 cm x 14 cm) and placed in an acclimation chamber with conditions similar to those in the wind tunnel for at least 1 h prior to the start of the experiment. At the downwind end of the wind tunnel, a moth was placed on a 40 cm platform, while at the upwind end, either a single D. wrightii flower attached to a 50 cm pole (feeding experiments), or a pot with a non-flowering three-leafed D. wrightii plant (oviposition experiments) was placed. Moths that were unable to fly or did not initiate wing beats within 2 min were excluded. Flying moths were observed for 3 min and filmed using a Sony Handycam DCR-SR35 in night shot mode. We counted the number of moths that touched the flower with the tip of their proboscis (feeding experiments) and the number of moths that touched a leaf with their tarsi (oviposition experiments), and calculated the total duration of these contacts per animal. In oviposition experiments, we counted the number of eggs laid on the leaves.

Statistics and figures preparation

Sample size and statistical tests are described in both the text and figure legends. Statistical analyses were done with GraphPad InStat (version 3.10, GraphPad Software, San Diego, CA, https://www.graphpad.com). Figures were generated using PAST (version 3.26, http://folk.uio.no/ohammer/past/), RStudio, GraphPad Prism9 (https://www.graphpad.com), and edited using Adobe Illustrator CS5.