Rearing O. biroi pupae in social isolation and collecting pupal fluid
In O. biroi colonies, larvae and pupae develop in discrete and synchronized cohorts26. Ten days after the first larvae had entered pupation in a large stock colony, the entire colony was anaesthetized using a CO2 pad, and white pupae were separated using a paintbrush. Pupae were individually placed in 0.2 ml PCR tubes with open lid. These tubes were then placed inside 1.5 ml Eppendorf tubes with 5 µl sterile water at the bottom to provide 100% relative humidity. The outer tubes were closed and kept in a climate room at 25 °C. The inner tube in this design prevents the pupa from drowning in the water reservoir. The outer tubes were kept closed throughout the experiment, except for once a day when the tubes were opened to remove pupal social fluid. Pulled glass capillaries were prepared as described elsewhere29, and used to remove and/or collect secretion droplets. We were careful to leave no remains of the secretion behind on the pupae or the inside of the tubes. To ensure that all secretion had been removed, pupae were taken out of the tube after fluid collection and briefly placed on a tissue paper to absorb any excess liquid. The inner tubes were replaced if needed—for example, if fluid traces were visible on the old tube after collection. Each pupa was checked daily for secretion (absent or present), onset of melanization and eclosion, and whether the pupa was alive (responding to touch). Control groups of 30 pupae and 30 adult ants from the same stock colony and cohort as the isolated pupae were placed in Petri dishes with a plaster of Paris floor, and the same parameters as for the isolated pupae were scored daily. Experiments ended when all pupae had either eclosed or died. Newly eclosed (callow) workers moved freely inside the tube and showed no abnormalities when put in a colony. A pupa was declared dead if it did not shed its pupal skin and did not respond to touch three days after all pupae in the control group had eclosed.
To calculate the average secretion volume per secreting pupa (Fig. 1d), the total volume collected daily from a group of isolated pupae (142–166 pupae) was divided by the number of pupae from which fluid had been collected that day. The total volume was determined by multiplying the height of the fluid’s meniscus in the capillary by πr², where r is the inner radius of the capillary (0.29 mm). While pupae were secreting, pupal whole-body wash samples were collected daily. The pupae were removed from colonies with adults and washed promptly with 1500 µl LC–MS grade water. Whole-body wash samples were lyophilized and reconstituted in 15 µl LC–MS grade water.
Collecting additional ant species and honeybees, rearing pupae in social isolation, and collecting pupal fluids
Colonies of the ants N. flavipes, T. sessile, P. pennsylvanica and Lasius neoniger were collected in NY state, USA (Central Park, Manhattan; Pelham Bay Park, Bronx; Prospect Park, Brooklyn; and Woodstock). Solenopsis invicta colonies were collected in Athens, GA, USA. M. mexicanus colonies were collected in Piñon Hills, CA, USA. Colonies comprised of queens, workers and brood were maintained in the laboratory in airtight acrylic boxes with plaster of Paris floors. Colonies were fed a diet of insects (flies, crickets and mealworms). White pupae were socially isolated, cocoons were removed in the case of P. pennsylvanica, and secretion droplets were collected from melanized pupae as described for O. biroi. A. mellifera pupae of unknown age were socially isolated from hive fragments (A&Z Apiaries, USA) and reared as described for O biroi, except that the rearing temperature was set to 32 °C. Relative humidity was set to either 100% to replicate conditions used for the different ant species, or to 75% as recommended in the literature30.
Injecting dye and tracking pupal fluid
Injection needles were prepared as in previous studies31. Injections were performed using an Eppendorf Femtojet with a Narishige micromanipulator. The Femtojet was set to Pi 1000 hPa and Pc 60 hPa. Needles were broken by gently touching the capillary tip to the side of a glass slide. To inject, melanized pupae were placed on ‘Sticky note’ tape (Post-it), with the abdomen tip forward and the ventral side upward. Pupae were injected with blue food colouring (McCormick) into the exuvium for 1–2 s by gently piercing the pupal case at the abdominal tip with the needle. During successful injections, no fluid was discharged from the pupa when the needle was removed, and the moulting fluid inside the exuvium was immediately stained. Pupae were washed in water three times to remove any excess dye. Following injections, 10 pupae were reared in social isolation to confirm the secretion of dyed droplets. For experiments, injected pupae were transferred to colonies with adult ants (Figs. 1f and 4c) or to colonies with adult ants and larvae (Figs. 3b and 4c) to track the distribution of the pupal social fluid.
After spending 24 h with dye-injected pupae, adults were taken out of the colony, briefly immersed in 95% ethanol, and transferred to PBS. Digestive systems were dissected in cold PBS and mounted in DAKO mounting medium. Crop and stomach images (Fig. 1f, inset and Fig. 4c, inset) were acquired with a Revolve microscope (Echo). Larvae are translucent, and the presence of dye in the digestive system can be assayed without dissection. Whole-body images of larvae were acquired with a Leica Z16 APO microscope equipped with a Leica DFC450 camera and Leica Application Suite version 4.12.0 (Leica Microsystems). In the experiment on larval growth (Fig. 3c), larval length was measured from images using ImageJ32.
Occluding pupae
Ten pupae were placed on double-sided tape on a glass coverslip with the ventral side up. The area between the pupae was covered with laser-cut filter paper to prevent adults from sticking to the tape. The pupae were then placed in a 5 cm diameter Petri dish with a moist plaster of Paris floor. To block pupal secretion, the tip of the gaster was occluded with a drop of oil-paint (Uni Paint Markers PX-20), which has no discernible toxic effect7. Secreting pupae received a drop of the same paint on their head to control for putative differences resulting from the paint. Pupae were left in isolation for one day before adults were added to the assay chamber.
Behavioural tracking of adult preference assay
Videos were recorded using BFS-U3-50S5C-C: 5.0 MP, 35 FPS, Sony IMX264, Colour cameras (FLIR) and the Motif Video Recording System (Loopbio). To assess adult preference (Fig. 1g), physical contact of adults with pupae was manually annotated for the first 10 min after the first adult had encountered (physically contacted) a pupa.
Protein profiling
We extracted 30 µl of pupal social fluid and whole-body wash samples with 75:25:0.2 acetonitrile: methanol: formic acid. Extracts were vortexed for 10 min, centrifuged at 16,000g and 4 °C for 10 min, dried in a SpeedVac, and stored at −80 °C until they were analysed by LC–MS/MS.
Protein pellets were dissolved in 8 M urea, 50 mM ammonium bicarbonate, and 10 mM dithiothreitol, and disulfide bonds were reduced for 1 h at room temperature. Alkylation was performed by adding iodoacetamide to a final concentration of 20 mM and incubating for 1 h at room temperature in the dark. Samples were diluted using 50 mM ammonium bicarbonate until the concentration of urea had reached 3.5 M, and proteins were digested with endopeptidase LysC overnight at room temperature. Samples were further diluted to bring the urea concentration to 1.5 M before sequencing-grade modified trypsin was added. Digestion proceeded for 6 h at room temperature before being halted by acidification with TFA and samples were purified using in-house constructed C18 micropurification tips.
LC–MS/MS analysis was performed using a Dionex3000 nanoflow HPLC and a Q-Exactive HF mass spectrometer (both Thermo Scientific). Solvent A was 0.1% formic acid in water and solvent B was 80% acetonitrile, 0.1% formic acid in water. Peptides were separated on a 90-minute linear gradient at 300 nl min−1 across a 75 µm × 100 mm fused-silica column packed with 3 µm Reprosil C18 material (Dr. Maisch). The mass spectrometer operated in positive ion Top20 DDA mode at resolution 60 k/30 k (MS1/MS2) and AGC targets were 3 × 106/2 × 105 (MS1/MS2).
Raw files were searched through Proteome Discoverer v.1.4 (Thermo Scientific) and spectra were queried against the O. biroi proteome using MASCOT with a 1% FDR applied. Oxidation of M and acetylation of protein N termini were applied as a variable modification and carbamidomethylation of C was applied as a static modification. The average area of the three most abundant peptides for a matched protein33 was used to gauge protein amounts within and between samples.
Functional annotation and gene ontology enrichment
To supplement the current functional annotation of the O. biroi genome34, the full proteome for canonical transcripts was retrieved from UniProtKB (UniProt release 2020_04) in FASTA format. We then applied the EggNog-Mapper tool35,36 (http://eggnog-mapper.embl.de, emapper version 1.0.3-35-g63c274b, EggNogDB version 2) using standard parameters (m diamond -d none –tax_scope auto –go_evidence non-electronic –target_orthologs all –seed_ortholog_evalue 0.001 –seed_ortholog_score 60 –query-cover 20 –subject-cover 0) to produce an expanded annotation for all GO trees (Molecular Function, Biological Process, Cellular Components). The list of proteins identified in the pupal fluid was evaluated for functional enrichment in these GO terms, P-values were adjusted with an FDR cut-off of 0.05, and the network plots were visualized using the clusterProfiler package37.
Metabolite profiling
For bulk polar metabolite profiling, we used 10 µl aliquots of pupal social fluid and whole-body wash (pooled samples). For the time-series metabolite profiling, 1 µl of pupal social fluid and whole-body wash was used. Samples were extracted in 180 µl cold LC–MS grade methanol containing 1 μM of uniformly labelled 15N- and 13C-amino acid internal standards (MSK-A2-1.2, Cambridge Isotope Laboratories) and consecutive addition of 390 µl LC–MS grade chloroform followed by 120 µl of LC–MS grade water.
The samples were vortexed vigorously for 10 min followed by centrifugation (10 min at 16,000g and 4 °C). The upper polar metabolite-containing layer was collected, flash frozen and SpeedVac-dried. Dried extracts were stored at −80 °C until LC–MS analysis.
LC–MS was conducted on a Q-Exactive benchtop Orbitrap mass spectrometer equipped with an Ion Max source and a HESI II probe, which was coupled to a Vanquish UPLC system (Thermo Fisher Scientific). External mass calibration was performed using the standard calibration mixture every three days.
Dried polar samples were resuspended in 60 µl 50% acetonitrile, and 5 µl were injected into a ZIC-pHILIC 150 × 2.1 mm (5 µm particle size) column (EMD Millipore). Chromatographic separation was achieved using the following conditions: buffer A was 20 mM ammonium carbonate, 0.1% (v/v) ammonium hydroxide (adjusted to pH 9.3); buffer B was acetonitrile. The column oven and autosampler tray were held at 40 °C and 4 °C, respectively. The chromatographic gradient was run at a flow rate of 0.150 ml min−1 as follows: 0–22 min: linear gradient from 90% to 40% B; 22–24 min: held at 40% B; 24–24.1 min: returned to 90% B; 24.1 −30 min: held at 90% B. The mass spectrometer was operated in full-scan, polarity switching mode with the spray voltage set to 3.0 kV, the heated capillary held at 275 °C, and the HESI probe held at 250 °C. The sheath gas flow was set to 40 units, the auxiliary gas flow was set to 15 units. The MS data acquisition was performed in a range of 55–825 m/z, with the resolution set at 70,000, the AGC target at 10 × 106, and the maximum injection time at 80 ms. Relative quantification of metabolite abundances was performed using Skyline Daily v 20.1 (MacCoss Lab) with a 2 ppm mass tolerance and a pooled library of metabolite standards to confirm metabolite identity (via data-dependent acquisition). Metabolite levels were normalized by the mean signal of 8 heavy 13C,15N-labelled amino acid internal standards (technical normalization).
The raw data were searched for a targeted list of ~230 polar metabolites and the corresponding peaks were integrated manually using Skyline Daily software. We were able to assign peaks to 107 compounds based on high mass accuracy (<2 ppm mass deviation) and retention time accuracy (<12 s deviation from known standards). A pool of all the biological samples was used for quality control and analysed using a data-dependent Top2 MS/MS scan (with polarity switching) to acquire MS/MS data and further validate metabolite identity. The data-dependent MS/MS scans were acquired at a resolution of 17,500, 1 × 105 AGC target, 50 ms max injection time, 1.6 Da isolation width, stepwise normalized collision energy (NCE) of 20, 30, 40 units, 8 s dynamic exclusion, and loop count of 2.
Scanning electron microscopy
All sample preparation steps for scanning electron microscopy were performed under a fume hood at room temperature, unless indicated otherwise. Scintillating vials containing O. biroi pupae were placed on a rotator to ensure uniform treatment. One- and fifteen-days old pupae (n = 9 each) were incubated overnight at 4 °C in a fixative solution containing 2% glutaraldehyde, 2% paraformaldehyde in 0.1 M sodium cacodylate buffer, pH 7.2. Pupae were then washed 4 times with 0.1 M sodium cacodylate, pH 7.2, for 5 min each, followed by incubation for 2.5 h at 4 °C in a 1% osmium tetroxide solution in 0.1 M sodium cacodylate buffer, pH 7.2. Subsequently, pupae were washed four times with Milli-Q purified water. A graded series of ethanol was used to dehydrate the washed samples. The dehydration steps until 70% ethanol were performed at 4 °C, and thereafter the samples were dehydrated (90% and 100% ethanol) at room temperature. After complete dehydration, the pupae were dried with hexamethyldisilazane as described previously38. Afterwards, samples were mounted on a carbon tape placed on a flat stub and coated with a 12 nm thick layer of iridium nanoparticles using a Leica EM ACE600 sputter coater. The coated samples were imaged with a Jeol JSM-IT500HR scanning electron microscope. Cocoons of P. pennsylvanica were directly mounted on a carbon tape, coated, and imaged as described above for O. biroi pupae.
Instant structured illumination microscopy
For imaging, 13- to 17-day-old melanized pupae were placed in the microwell of a 35 mm glass bottom Petri dish (MatTek) and covered with a coverslip. Abiotic conditions (100% humidity, 25 °C) were maintained throughout the recording using an environmental stage chamber (Okolab). Bright-field images of pupal fluid droplets secreted from the abdominal tip were acquired with a DMi8 inverted microscope (Leica), an Orca fusion CMOS camera (Hamamatsu) and the VisiTech InstantSIM (iSIM) (VisiTech International) real-time superresolution system with a 20×/0.75 water objective at a resolution of 3.08 pixels per µm. Imaging was performed at different depths. The z-plane depths were selected to optimize visualization of secretion droplets, the rectal invagination, and the genital opening. Images were acquired using VisiView acquisition software version 4.5.0.13 (VisiTech International) and image processing was performed using FIJI/ImageJ version 1.52p32.
Statistics and reproducibility
Statistical analyses were performed in MATLAB R2020b (MathWorks) and R version 3.6.3. In datasets used for ANOVA or ANCOVA, normality was assessed by examining Q–Q plots of model residuals and Box–Cox transformation was applied when needed. For heteroskedasticity, White adjustment was applied when appropriate. ANOVA and ANCOVA models were followed by two-sided Games–Howell post hoc tests for multiple comparisons and the P-values obtained were adjusted for multiple testing using Tukey’s method. A nonparametric repeated-measures ANOVA model with aligned rank transformation was used to analyse the larvae placement data. The model included treatment, day, and treatment-by-day interaction. Estimated marginal means were computed and used for two-sided post hoc pairwise comparisons with FDR = 0.05. The time-series metabolomic dataset was analysed with MetaboAnalyst version 4.039 using one-way repeated-measures ANOVA with auto-scaling normalization and FDR = 0.05. Details of statistical tests are given in the respective figure legends.
Figure 1b shows a representative pupa with a secretion droplet out of n = 1,368 pupae observed under identical conditions. Figure 1f shows a representative pupa and the crop of a representative adult out of a total of n = 10 replicate colonies of 10 adults and 10 pupae each. The inset of Fig. 1h shows a representative pupa out of a total of n = 90 pupae.
Figure 3c shows a representative larva that has ingested dyed pupal secretion out of n = 3 replicate colonies with 10 larvae, adults and pupae each.
Experimental ants were collected haphazardly from stock colonies and distributed across experimental colonies or setups at random. The order of collection and place in which the colonies or setups were arranged were randomized. Blinding was not applicable in this study due to the descriptive nature of the experiments. Sample size calculations were not performed for the experiments. Appropriate sample sizes were estimated based on preliminary data that showed strong effects with small variations. The study included one- to several-month-old females, as well as females of different developmental stages (larvae and pupae). Only female ants display social behaviour, and all worker ants are female. In O. biroi in particular, males are only produced sporadically and do not partake in the social life of the colony. Conducting experiments with males is therefore not relevant in the context of the current study.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Source: Ecology - nature.com
