For the present study, the Ethical Committee of the Department of Veterinary Medicine, University of Teramo - Italy, was consulted, and the response (protocol n. 16571) was that the experimental procedures did not fall within the ambits of Directive 63/2010 of the European Parliament and of the Council on the protection of animals used for scientific purposes (transposed into Italian law by Legislative Decree 26/2014). For this reason, the Ethical Committee did not consider necessary any authorization.
Preliminary procedures
In this study, volume-variable micropipettes (model Acura 825, volume range: 0.5–10 µL; 10–100 µL; 100–1000 µL. Socorex Isba SA, Ecuclens, Swiss) were used to perform dilution of the samples. All the micropipettes used in this study were validated gravimetrically before the experiments for the specific volume they displaced. The medium used for the gravimetric validation was bi-distilled water, prepared a few minutes before the test. For each micropipette, the volume to check was dispensed on a certified analytic balance (Explorer, Ohaus Corp., Parsippany, NJ, USA). The precision, as the degree of closeness between repeated measurements of the same volume, was calculated on 20 measures in triplicate, by the calculation of the coefficient of variation (CV) and the intraclass correlation coefficient (ICC – see the statistical evaluation section for more details).
The threshold for acceptable precision was 2% of CV and R = 0.98 for ICC: values beyond the thresholds required calibration. In this study, no micropipette was found beyond the threshold set-up.
The standardization of the procedures of dilution was increased as much as possible. Briefly, the tip was systematically dried externally to remove eventual external micro drops. After dilution, the sample was mixed using a different micropipette able to displace 90% of the final volume, pipetting several times (20 folds) and avoiding bubble formation. In this study, vortex mixing was not performed, since seminal plasma and media created bubble formation in the sample.
The flow cytometric analyses were performed using the CytoFLEX (Beckman Coulter Inc., Brea, CA, USA) equipped with three lasers (wavelength of 405 nm – 80 mW; 488 nm – 50 mW; 638 nm – 50 mW). Data were managed by the CytExpert software version 2.1 (Beckman Coulter). For all the trials, volumetric flow cytometric (volFC) analyses were performed at a flow rate standardized at 10 µL/min, and the analysis was stopped using the volumetric limit of 20 µL. Events were recorded systematically 30 s after the beginning of the analysis, to allow the stabilization of the fluidics.
Preliminary trial
The VolFC analysis was preliminary verified using standard Flow-Count fluorospheres (Batch 7548247F) at certificated concentration, estimated by the manufacturer in 1014 beads/µL (for the batch used in this study). In detail, fluorospheres were diluted 1:2, 1:4, 1:8, and 1:16 (v:v) with bi-distilled water. A sample was not diluted and raw-analysed. After an appropriate resuspension, performed as above described, the samples were analysed using the flow cytometer CytoFLEX (Beckman Coulter). Samples with progressive fluorosphere dilution were prepared 3 times, as replicates. Total events (beadsT), and events excited with 488 nm and recorded with the 585/42 band-pass filter (beads488), and the original concentration of beads was calculated with the following formula:
Concentration = number of events * 50 * dilution
Thus, total beads concentration (CbeadsT) and beads concentration at 488 nm (Cbeads488) were calculated.
To verify the precision of fluorosphere concentration, 6 aliquots of the same batch were undiluted (1:0) or diluted 1:4 and analysed in 6 replicates.
To estimate if the worn-out fluidic components could affect concentration estimation, 6 samples of diluted beads (1:4, v:v) were compared using a worn-out (serial number 20192241; approximately 600 working hours) or new peristaltic pump tube (serial number 20210758).
Animals and semen collection
A total of 12 mature bulls were included in the trials. The bulls (2 to 7 year olds) were housed in the Alpenseme Artificial Insemination Centre of the Provincial Breeders Federation of Trento (Ton, Trento, Italy). All the animals were routinely collected twice a week within the artificial insemination program and aliquots of the ejaculates were removed to carry out the trials. For this reason, no semen collections were performed specifically for this study. The animals were managed according to the European Commission Directive for Farm Animal Welfare (Directive 98/58/EC) and the National Law for Animal Welfare and Protection (Italy). Collections were performed by the same operator using a pre-warmed artificial vagina.
After collection, an aliquot was removed and used for the experiments (fresh samples – trials 1–3). Then, the semen was processed to produce cryopreserved artificial insemination doses, part of which was used for the procedures on cryopreserved samples (trial 4). Briefly, volume was estimated by weight using a precision balance CP6201 (Sartorius AG, Gottingen, Germany), assuming 1 mL = 1 g 19. Concentration was evaluated photometrically using bovine-specific Accucell (IMV Technologies, L’Aigle, France) after dilution 1:100 with saline solution 20. Semen was diluted at approximately 100 × 106 sperm/mL with Bioxcell (IMV Technologies) and equilibrated for 3 h at 5°C in a passive refrigerator. After equilibration, diluted semen was packaged in 0.25-mL straws, frozen with a programmable nitrogen freezer (Digicool 5300, IMV Technologies), following a cooling curve previously described 21, and plugged in liquid nitrogen for storage. Seven days after freezing, 8 straws/ejaculate were thawed in the water bath at 38°C for 1 min and evaluated for concentration (trial 4).
Trial 1. Effect of dilution on concentration estimated with a volumetric flow cytometer and NucleoCounter
To verify the effect of the sample dilution on the sample concentration, progressive dilutions (1:12.5; 1:25; 1:50; 1:100; 1:200; 1:400; 1:800) of the sample were prepared and analysed using the volFC and NC SP-100 (ChemoMetec, Allerod, Denmark). The micropipettes used for sampling were the same between methods. Samples for volFC were diluted with phosphate-buffered saline (PBS) sterilized by filtration (Filtropur S, pore size 0.2 µm, Sarstedt AG & Co, Nümbrecht, Germany) added with Hoechst 33342 (ENZ-52401, Enzo Life Sciences AG, Lausen, Switzerland) at the final concentration of 1.69 µM (PBS-H).
The volFC samples were incubated for 15 min at 37°C 22 and analysed using the CytoFLEX (Beckman Coulter) using the 405 nm laser (excitation). Events were detected using the 450/45 band-pass filter. The flow rate was standardized at 10 µL/min, and the analysis was stopped using the volumetric limit of 20 µL. Events were recorded systematically 30 s after the beginning of the analysis, to allow the stabilization of the fluidics. A morphological gate was created on the forward scatter x side scatter plot to detect only sperm-referred events. A 450-fluorescent gate was created on the event histogram, to detect only events positive for the Hoechst 33342 (H+). Finally, a double morphological and fluorescent gate was created. Total events (evT), morphological gated events (evM), fluorescent gated events (evH+), and events selected by morphologically and fluorescence (evM/H+) were recorded. The original concentration of the sample was calculated with the following formula:
Concentration = number of events * 50 * dilution
Samples for NC were diluted with SP-100 diluent (ChemoMetec). Close before the analysis, the sample was systematically mixed using a different micropipette able to displace 90% of the final volume, pipetting several times (20 folds) and avoiding bubbles formation. Then, the resuspended sample was loaded in SP1-cassette with a waiting period standardized at 10 seconds before the analysis, inserted in the device taking care to adjust the dilution factor, and analysed. The concentration that appeared on the display, was recorded and used for the statistical analysis. The prevalence of the error messages provided by the NucleoCounter (Error: Sample could not be analysed) at the different dilutions was also recorded.
Trial 2. The repeatability of volumetric flow cytometry compared with NucleoCounter to estimate sperm concentration
Aliquots of fresh semen from the 12 bulls were pooled (4 bulls/pool) to create three pooled samples. A total of 10 repeated measurements were performed on each pooled sample using the volFC and NC, to estimate the repeatability of both methods. For the volFC, samples were prepared at dilution 1:400 with sterilized PBS-H and incubated for 15 min at 37°C and analyzed flow-cytometrically as above described. Total events (evT), morphological gated events (evM), fluorescent gated events (evH+), and events selected by morphologically and fluorescence (evM/H+) were recorded. The original concentration of each category in the sample was calculated with the following formula:
Concentration = number of events * 50 * 400
For the NC, repeated analyses were performed on the same samples after dilution 1:401 using the SP-100 diluent (ChemoMetec), following the Manufacturer’s instruction, selecting bull as species and 401 as dilution factor. In brief, 25 µL of fresh semen was transferred in a sterile cup, and 10 mL of SP-100 was dispensed (Dispensette III bottle-top dispenser, Brand, Wertheim, Germany) in the cap. Then, SP1-cassettes were prepared as above described and analysed.
Trial 3. Comparison of different methods to estimate the concentration
Fresh samples from each of the 12 bulls were prepared in triplicate and analysed with different methods to estimate concentration: i) haemocytometer (Hem); ii) NucleoCounter; iii) flow cytometry with fluorospheres; iv) volumetric flow cytometry.
Evaluation of concentration by Hem was performed using a Bürker counting chamber (Merck, Leuven, Belgium) after dilution 1:1000 with 0.9% NaCl solution with 3% glutaraldehyde to ensure sperm immobilization. At least 400 spermatozoa in two chambers were counted to estimate concentration. If the difference of spermatozoa in each chamber exceed 10% compared to the other of the same slide, the sample was re-prepared. Each sample was prepared in triplicate.
The NC was used as recommended by the manufacturer’s instructions. The bull was selected as species, and the dilution factor inserted was 401. Samples were prepared as above mentioned. After loading in the SP1-cassette, a waiting period of 10 s was observed, and then the sample was analysed. Each sample was prepared and analysed in triplicate. The concentration reported on the display was recorded and used for statistical analysis.
The estimation of sperm concentration by spFC was performed using Flow-Count Fluorospheres (Beckman Coulter) at certified concentration, using the batch above mentioned. The sample was diluted 1:200 using sterilized and stained PBS (Hoechst 33342, dilution 1:1000 - PBS-H). Then, the sample was further diluted 1:1 (v:v) with fluorospheres and analyzed with CytoFLEX (Beckman Coulter). Each sample was prepared in triplicate. Events excited at 405 nm and recorded at 450/45 nm (spermatozoa) and events excited at 488 nm and recorded at 585/42 nm (fluorospheres) were used to estimate sperm concentration using the following formula:
spFC concentration = [((405-events*200) / (488-events))*fluorospheres concentration]
The analysis, performed at the flow rate of 10 µl/minute, was stopped at 20,000 events.
For the volFC, concentration was calculated after dilution of the semen at 1:400 with sterilized PBS-H. Samples were incubated for 15 min at 37°C, mixed as above mentioned soon before the analysis, performed with the CytoFLEX flow cytometer (Beckman Coulter). Each sample was prepared in triplicate. The flow rate was standardized at 10 µL/min, and the analysis was stopped using the volumetric limit of 20 µL. Events were recorded systematically 30 s after the beginning of the analysis, to allow the stabilization of the fluidics. A morphological gate, 405-fluorescent gate, and double morphological plus fluorescent gate were created and events in the respective gate were recorded as total events (evT), morphological gated events (evM), fluorescent gated events (evH+), and events selected by both morphological and fluorescent gate (evM/H+). The original concentration of the sample was calculated with the following formula:
Concentration = number of events * 50 * 400
Trial 4. Evaluation of cryopreserved sperm concentration with volumetric flow cytometry and NucleoCounter
A total of 8 straws for each bull included in the present study were thawed in a waterbath (38°) for 60 seconds, then samples were transferred in a sterile 5-ml tube. For volFC, samples were diluted 1:40 with sterilized PBS-H and incubated for 15 min at 37°C. After proper resuspension, the sample was analysed flow-cytometrically as above described. Each sample was prepared, for the analysis, in triplicate. Total events (evT), morphological gated events (evM), fluorescent gated events (evH+), and events selected by morphologically and fluorescence (evM/H+) were recorded and used to calculate the original concentration, using the formula:
Concentration = number of events * 50 * 40
For the NC, samples were diluted 1:51 using the SP-100 diluent (ChemoMetec) in a sample cup, according to the Manufacturer’s instruction. The dilution factor of 51 was set on the device. Then, SP1-cassettes were prepared as above described. The dilution of each sample was performed in triplicate.
Statistical analysis
Data were presented as mean ± standard deviation (SD). The normal distribution of the data was tested using the Shapiro-Wilk test. Furthermore, homoscedasticity between groups was tested by Levene’s test. Data in the present study were normally and homogeneously distributed.
Preliminary trial
Differences in the recalculated concentration between different dilutions were performed using a general linear model (GLM) based on ANOVA, in which the dilution factor was considered a fixed variable. A Scheffè post-hoc test was performed when appropriate. The proximity between the concentration calculated at the different dilutions and the certified concentration of the fluorospheres was calculated as a percentage value (100% = certified concentration), and the GLM was used to verify the effect of the dilution, followed by the Scheffè test for the post-hoc analysis. The repeatability of the analysis in the undiluted (1:0) and diluted samples (1:4) was estimated by the ICC 23. Finally, the effect of worn-out fluidic components on concentration estimation was checked by the t-Student test for paired values. In all the cases, significance was set at P < 0.05.
Trial 1
In this trial, the concentration of spermatozoa at increasing dilution was estimated with volFC for each sample using the following flow cytometric populations: evT, evM, evH+, and evM/H+. Differences between concentrations in the different populations of the volFC and NucloCounter, at correspondent dilution (1:12.5 to 1:800), were tested using the GLM. Dilution and concentration with the different techniques (NC, CevT, CevM, CevH+, and CevM/H+) were considered fixed factors. The general linear model was followed by the post hoc Scheffè test when appropriate. Significant differences were considered with P < 0.05.
Trial 2
The repeatability of concentration was measured with volFC, using the different populations (evT; evM; evH+; evM/H+), and NucleoCounter was estimated in three pools of bovine fresh semen by calculating the ICC on 10 replicates, as above mentioned 23.
Trial 3
Differences in the concentration estimated with the different techniques, namely Hem, NC, spFC, and volFC (calculated on the different populations - evT; evM; evH+; evM/H+) were tested using a general linear model (GLM) based on ANOVA. The reproducibility was quantified by the variation between replicates, using the coefficient of variation 24. The agreement between the different techniques was calculated by Lin’s concordance correlation coefficient (CCC - ρc) with a strength of agreement as follows: > 0.99, almost perfect; between 0.95 to 0.99, substantial; between 0.90 to 0.95, moderate; < 0.90 poor 25. Graphical agreement between techniques was performed using the Passing-Bablok regression plot 26.
Trial 4
The concentrations of cryopreserved semen from the 12 bulls included in the study estimated using volumetric flow cytometry (different populations) and NucleoCounter were compared. Differences between concentrations in the different populations of the volFC and NucloCounter were tested using the GLM. The different techniques or populations considered (NC, CevT, CevM, CevH+, and CevM/H+) were considered fixed factors. The general linear model was followed by the post hoc Scheffè test. Significant differences were considered with P < 0.05. The agreement between the different techniques to estimate sperm cryopreserved concentration was tested by calculating Lin’s CCC. Graphical agreement between techniques was performed using the Passing-Bablok regression plot 26.