Collecting fluid and particle tracking microrheology (PTM)
To collect fluid for viscosity measurements, we removed the female reproductive tract and submersed it in mineral oil at 4ºC until fluid extraction, which took place within 24 hours. To visually distinguish the mineral oil from biological fluid, we dyed the oil blue using a colored gel dye (Wilton Candy Colors, USA) in a 1:75 dye:oil ratio. Under 0.63x magnification (Zeiss Stemi 508, USA), we trimmed the fat surrounding the reproductive tract, unraveled the coiled oviducts, severed the uterus from the oviduct at the utero-tubal junction (UTJ), divided the oviduct in half to separate the lower and upper regions, and submersed in dyed mineral oil. We used glass Pasteur pipettes bent into a u-shape under a flame to push down from one end of the tissue to the other to squeeze fluid within the tissue out into the oil, then collected and centrifuged the samples, removed the oil supernatant and stored fluids at -80ºC (Yuana et al. 2015; Patczai et al. 2017).
To obtain sufficient volume for downstream methods, we pooled the samples for each region from at least ten individuals per species, then warmed pools to 37ºC to simulate natural physiological values, and again centrifuged at 3000 rpm for 3 min to ensure full separation of the mineral oil from the reproductive fluid. We then combined 2µL of fluid with 0.5µL of ~0.002% w/v suspension of fluorescent nanoparticles (PEG-coated polystyrene particles, PS-PEG). PS-PEG were prepared by coating red fluorescent carboxylate-modified PS spheres (PS-COOH), 500 nm in diameter (ThermoFisher FluoSpheres Carboxylate-Modified Microspheres, 0.5 µm, red fluorescent (580ex/605em), 2% solids, USA) with 5-kDa methoxy-PEG-amine (Creative PEG-Works, USA) via NHS-ester chemistry as previously described (Joyner et al. 2019). We gently reverse-pipetted the mixture to make sure the nanoparticles were homogeneously scattered throughout and pipetted 2.0µL into a 1-mm ID Viton O-ring microscopy chamber (McMaster Carr, USA) and covered with a small circular glass coverslip, both of which were sealed with vacuum grease (Dow Corning, USA) to prevent fluid flow and evaporation, and equilibrated for 30 minutes prior to imaging to reduce dynamic error.
To measure viscosity, we recorded a minimum of three videos of the suspended fluorescent nanoparticles within each fluid sample at a frame rate of 33.33 Hz for 300 frames (10 sec) using an EMCCD camera (Axiocam 702; Zeiss, Germany) attached to an Zeiss 800 LSM inverted microscope and x63/1.20 W Korr UV VIS IR water-immersion objective with image resolution of 0.093μ per pixel. To avoid edge effects on nanoparticle movement, we randomly selected central locations within the chamber for our video recordings. All samples remained at 37ºC during imaging using a stage incubator (PM 2000 Rapid Balanced Temperature, PeCon, Germany). To track the diffusion of PS-PEG nanoparticles in each sample, we used particle tracking data analysis using automated software custom-written in MATLAB (Mathworks, USA). Based on a previously developed algorithm (Crocker and Grier 1996), the program determined the x andy positions of nanoparticle centers based on an intensity threshold and then constructed particle trajectories by connecting particle centers between sequential images given an input maximum moving distance between frames. Finally, the program calculated the time-averaged mean squared displacement [MSD (τ)] as
〈Δr2( τ)〉 = 〈x (t+ τ) – x (t )]2 + [t (t + τ )+ y (t )]2
where τ is the time lag between frames and angle brackets denote the average over the time points. The MSD of PS-PEG nanoparticles is directly proportional the viscosity of the surrounding fluid. A fast-moving particle (high MSD) reflects a low viscosity fluid whereas a slow-moving particle (low MSD) reflects a high viscosity fluid. Using the generalized Stokes–Einstein relation, measured MSD values were used to compute viscoelastic properties of the hydrogels (Joyner et al. 2020). The Laplace transform of 〈Δr 2(τ )〉, 〈Δr 2(s )〉, is related to viscoelastic spectrum G (s ) using the equation G (s ) = 2k BT /[πas 〈Δr 2(s ) 〉], where k BT  is the thermal energy, a  is the particle radius, s  is the complex Laplace frequency. The complex modulus can be calculated as G *(ω ) = G ′(ω ) + G ′′( ), with   being substituted for s , where i  is a complex number and ω  is frequency. We pooled these data across all particles to characterize female reproductive fluid viscosity per species. Due to technical difficulties, we were unable to collect viscosity data from the upper oviduct for two of the polyandrous species (P. maniculatus and P. gossypinus ). For this reason, we combined viscosity data for both the lower and upper oviducts into a single measure for every focal species.