The presence of resilin, an elastomeric protein, in insect vein joints provides the flexible, passive deformations that are crucial to flapping flight. This study investigated the resilin gene expression and autofluorescence dynamics among Apis mellifera (honey bee) worker age classes and drone honey bees. Resilin gene expression was determined via ddPCR on whole honey bees and resilin autofluorescence was measured in the 1m-cu, 2m-cu, Cu-V, and Cu2-V joints on the forewing and the Cu-V joint of the hindwing. Resilin gene expression varied significantly with age, with resilin activity being highest in the pupae. Autofluorescence of the 1m-cu and the Cu-V joints on the ventral forewing and the Cu-V joint on the ventral hindwing varied significantly between age classes on the left and right sides of the wing, with the newly emerged honey bees having the highest level of resilin autofluorescence compared to all other groups. The results of this study suggest that resilin gene expression and deposition on the wing is age-dependent and may inform us more about the physiology of aging in honey bees.

Specimen Collection

Honey bees were collected from Bountiful Blossoms Bee Company, LLC in Glenwood, IA, USA from May–June, 2019. Biological age class was estimated on-site using class-specific characteristics and behavioral norms. Specifically, purple-eyed pupae were collected for the pupae class, representing a distinct age class of approximately 7–10 days post-oviposition, workers actively emerging out of cells as the newly emerged class, workers actively tending to the nest as nest honey bees, and workers returning to the hive with pollen on their legs as foragers. Drones at the entrance of the hive were collected for drone specimens. Honey bees were collected from 5 different hives for each class. For gene expression analysis, honey bees were immediately placed in RNAlater and then stored at −80°C until further use. For autofluorescent analysis of resilin on wings, honey bees were collected in wooden cages and then frozen at −20 °C until further use.

Resilin Autofluorescence

Dissection and Slide Preparation

For each honey bee, the wings were dissected along the axillary sclerites. Each wing was first swabbed with 70% ethanol and then secondly with a solution of 50% Triton X and 50% distilled water to clean off pollen and other particles on the wing. The fluorescence of resilin is sensitive to pH: in neutral and alkaline media, the excitation/excitation maxima are 320/420 nm, while in an acidic media, they are 285/420 nm. The wash protocol of ethanol, Triton X-100, and distilled water resulted in wings being exposed to a primarily neutral solution (pH 7.2–7.3), which was taken into account when selecting the best filter for autofluorescence detection. After being swabbed, the wings were covered and dried. The microscope slides were cleaned using 70% ethanol and wiped to prevent dust contamination. The wings were arranged, polyurethane adhesive was swabbed along the edges of the microscope slide, and another microscope slide was placed on top. 

Microscopy

Wings were observed using fluorescence microscopy on an Olympus BX61 microscope (Olympus Scientific Solutions Americas Corp., Waltham, MA, USA) equipped with an aniline blue filter (excitation maxima: 377 nm; emission maxima: 409 nm). Images were taken with an Olympus SC100 Color Camera that was connected to the microscope. Initial images of resilin were taken with samples and underwent tests to determine the exposure that allowed for the greatest clarity of resilin across multiple joints and focal planes without saturation. All images were taken at 20× magnification and 450 ms exposure. The dorsal, ventral, left, and right sides of all wings were imaged at the first medio-cubital (1m-cu), second medio-cubital (2m-cu), cubito-vannal (Cu-V), and second cubito-vannal (Cu2-V) joints on the forewing and the cubito-vannal (Cu-V) joint on the hindwing.

Image Analysis

All images were blinded by a third party and uploaded to Image J version 15.2 (U.S. National Institutes of Health, Bethesda, Maryland, USA). The scale was set and the following measurements were selected: area, minimum and maximum gray value, mean gray value, and integrated density. The picture identification code, honey bee and hive identification code, and orientation were recorded, and scale bars were burned into each image.

The autofluorescent resilin was manually traced three times and the average for each measurement was found. Then three selections of the background were taken and the average for each measurement was found. The corrected total cell fluorescence (CTCF) equation, a quantitative fluorescence imaging analysis, was used to calculate the final measurement: CTCF = Mean Integrated Density − (Mean area of the cell x Mean Gray Value of the background). CTCF can be used for any fluorescent material within a restricted area. CTCF calculates the fluorescence through the integrated density measurement, which finds the summation of gray pixel values within a selection and multiplies it by the area of the selection. Average CTCF was calculated because it incorporates the area and intensity of resilin fluorescence into one standard number while simultaneously subtracting out background fluorescence. Average CTCF was measured in relative fluorescence units (RFU). RFU is a standard measure of fluorescence intensity.

ddPCR

Droplet digital PCR was completed according to the manufacturer’s protocol using the Bio-Rad QX200 ddPCR System. Pro-resilin gene expression analysis was used to make inferences about resilin deposition timing as it related to worker age classes. Pro-resilin is the precursor protein that gets transported to the subcuticular space before crosslinking to form a functional resilin protein network. The concentration of cDNA was normalized to a housekeeping gene, eukaryotic translation initiation factor 3 subunit C (eIF3-S8).

Primers and probes were utilized at concentrations of 300 nM and 150 nM, respectively. A PCR was performed using the C1000 Touch thermocycler (Bio-Rad Laboratories). The thermocycler was programmed with the following settings: (1) 10 min at 95 °C, (2) 30 s at 94 °C, (3) 1 min at 57.5 °C, (4) repeat steps 2 and 3 for a total of 39 cycles, (5) 10 min at 98°C, and (6) 20 min at 4 °C. Additionally, a ramp rate of 2.5 °C was applied to Steps 2–4. Samples were then transferred to the Bio-Rad QX200 Droplet Reader. The output from the droplet reader was analyzed using Quantasoft software version 1.7 (Bio-Rad Laboratories, Hercules, CA, USA)to determine the copy number of target sequences.