Relationship of residual feed intake with semen parameters and testicular ultrasound of Nellore bulls

ABSTRACT The objective of this study was to evaluate characteristics of the testicular parenchyma and vascular parameters of the pampiniform plexus obtained by ultrasound, semen quality parameters, and sperm freezability in Nellore bulls classified based on residual feed intake (RFI). Twenty-seven bulls (21.82±0.88 months of age) evaluated for feed efficiency were sampled for the study, including 15 with low RFI (−0.592±0.09 kg dry matter/day) and 12 with high RFI (0.792±0.10 kg dry matter/day). In ultrasound and Doppler assessment, the most efficient animals (low RFI) showed higher pulsatility and resistive indexes, as well as a tendency towards greater heterogeneity of the testicular parenchyma (0.625±0.032 vs. 0.508±0.032, 1.012±0.072 vs. 0.802±0.072, and 12.9±0.96 vs. 10.2±0.96, respectively, for low vs. high RFI). However, these animals tended to have lower peak diastolic velocity (5.19±0.50 for low RFI vs. 6.54±0.50 for high RFI). Analysis of fresh semen showed a lower percentage of minor defects in low RFI animals (2.67±1.19%) compared with high RFI animals (8.10±1.19%), without differences in the other parameters in fresh or thawed semen and after thermoresistance testing. Evaluation of flow cytometry parameters showed a higher quality of mitochondrial respiration in semen samples of low RFI animals (22.04±2.50%) compared with high RFI animals (12.29±2.71%). Therefore, although RFI exerts an effect on the Doppler parameters of the pampiniform plexus, it is not sufficient to affect the quality of fresh or thawed semen.


Introduction
Investments in feed account for more than half of the total cost of cattle production; consequently, improvements in feed efficiency are important for the growth of the livestock sector, directly affecting the profitability of livestock farming (Forbes, 2007).Growing attention has been given to the development of strategies that can increase herd efficiency, reducing both feed intake and the environmental impact of methane production by cattle (Herd et al., 2002;Sakamoto et al., 2021).Studies have been conducted over the past decades to better understand residual feed intake (RFI), which has proved to be an important tool that can be used in the genetic improvement of beef cattle (Grion et al., 2014;Ceacero et al., 2016).
Studies have reported an unfavorable relationship between fertility indices and feed efficiency in young cattle (Awda et al., 2013).Within this context, reports suggest that young bulls with low RFI (more efficient) produce ejaculates characterized by lower sperm motility and lower semen quality based on sperm morphology (Awda et al., 2013).However, Rossi (2017) and Kowalski et al. (2017) found no effect of RFI on the reproductive parameters of young bulls.Similarly, Ferreira Júnior et al. (2018) observed a low genetic and phenotypic correlation between RFI and scrotal circumference (SC), indicating a low or zero effect of feed efficiency on fertility traits of bulls.
Sperm quality is very important for the reproductive success of bulls, with direct effects on the results of field and in vitro fertility, which can even cause changes in the quality of the produced embryos (Saacke, 2008).Therefore, studies have investigated sperm function in detail to estimate the fertilization potential of bulls and to predict long-term male fertility from the collected semen samples (Freitas-Dell' Aqua et al., 2009).
Ultrasonography is a complementary technique to andrological examination and can be an excellent tool to examine the integrity of the testicular parenchyma, in addition to being a rapid and non-invasive method.Spectral Doppler ultrasound can be used to assess blood flow indices in the testicular arteries and pampiniform plexus and thus estimate blood perfusion in this region (Ortiz-Rodriguez et al., 2017;Claus et al., 2019;Rodrigues et al., 2020).Studies have evaluated the relationship of these parameters with the semen quality of various species based on the testicular health of individuals (Ahmadi et al., 2013;Tomlinson et al., 2017;Camela et al., 2019;Gloria et al., 2018).Within this context, some studies observed lower echogenicity in low RFI animals when compared with high RFI animals, a finding that may indicate lower cellularity (Fontoura et al., 2016;Bourgon et al., 2018).On the other hand, Kowalski et al. (2017) found no difference between animals with distinct RFI.Therefore, we tested the hypothesis that selection for RFI affects the reproductive parameters of Nellore bulls.The objective of the present study was to evaluate the characteristics of the testicular parenchyma obtained by B-mode ultrasound, vascular parameters of the pampiniform plexus measured by spectral Doppler, semen parameters, and sperm freezability in Nellore bulls classified for feed efficiency.

Study location and animals
The study was conducted in Sertãozinho, São Paulo, Brazil (21°10' south latitude and 48°5' west longitude, with altitude of 579 m and predominant tropical condition).The local Animal Use Ethics Committee approved the project, and the study was conducted in accordance with the Ethical Guidelines on Animal Experimentation adopted by the National Council for the Control of Animal Experimentation (Protocol 08791/19).
Twenty-seven young Nellore males (17 born in 2016 and 10 born in 2017) were evaluated.The animals had participated in postweaning (seven months) feed efficiency tests, including 145 animals in 2017 and 128 animals in 2018.In the performance test, the animals were fed in collective pens equipped with electronic feed bunks for the recording of feed intake (GrowSafe ® Systems Ltd.; Airdrie, Alberta, Canada), with 28 days of adaptation and 77±10 days of test.
The RFI was calculated as the residual of the regression equation according to the model proposed by Koch et al. (1963) within the contemporary group: DMI = β0 + βW*BW 0.75 + βG*ADG + ε, in which DMI is the mean dry matter intake during the test; β0 is the intercept of the equation; BW 0.75 is the mean metabolic body weight; ADG is the average daily weight gain; βW and βG are the regression coefficients of DMI on BW 0.75 and ADG, respectively; and ε is the residual of the equation, corresponding to RFI.
Relationship of residual feed intake with semen parameters and testicular ultrasound of Nellore bulls Borges et al. 3 For this study, animals with extreme low (more efficient animals) and high RFI (less efficient animals) were chosen to compose the total samples.After the performance test, the animals were kept on Brachiaria brizantha pasture with free access to water and proteinated salt until the start of the experimental period, when the animals were 21.5±0.886months old and had reached body weight of 495±62.2kg.
The experiment lasted 21 days to obtain the ultrasonographic parameters and to perform the sperm evaluation of fresh and thawed semen from low and high RFI animals (Figure 1).

Testicular ultrasound
Two testicular ultrasound assessments were performed, one at the beginning of the experimental period and one after 21 days.A Z5 Vet ultrasound apparatus (Mindray, Shenzhen, China) coupled to a linear 7.5-MHz transducer was used.Scans were performed in the longitudinal and transverse planes of the right and left testes.The images were analyzed using the Image Pro Plus 7.01 software (Media Cybernetics Inc.; San Diego, CA, USA), with numerical grayscale pixel values ranging from 0 (absolute black) to 255 (absolute white) (Giffin et al., 2009).
Doppler ultrasound was applied to the region of the spermatic cord to determine the mean diameter of the testicular artery.Spectral Doppler was used for the measurement of vascular parameters using three waves for calculation: peak systolic velocity (PSV), end-diastolic velocity (EDV), vascular resistive index [RI = (PSV -EDV)/PSV], and pulsatility index [PI = (PSV -EDV)/M, in which M is the mean PSV and EDV] (Wood et al., 2010;Feliciano et al., 2012).

Andrological evaluation
Scrotal circumference was measured with a millimeter tape measure as recommended by the CBRA (2013).The animals were subjected to andrological evaluation every seven days, totaling four assessments, with the first two for standard sperm evaluation (Figure 1).In the last two assessments, semen was also cryopreserved.Semen samples were collected with an Autojac ® electroejaculator (Neovet, Brazil).
For the analysis of sperm morphology, the semen samples were stored in 500 µL of 4% saline-buffered formalin for examination by differential interference contrast (DIC) microscopy (Eclipse Ni-U, Nikon ® , Tokyo, Japan).A total of 200 cells were counted and defects in the head, midpiece, tail, and acrosome were recorded.Anomalies were classified as major, minor, and total sperm defects (CBRA, 2013).

Post-thaw semen evaluation
Two cryopreserved straws of each sample per bull were thawed in a water bath at 37 ℃ for 30 s.Samples were subjected to thermoresistance testing (TRT) in a water bath at 46 ℃ for 30 min (Barnabé et al., 1981;Crespilho et al., 2006).After thawing and after TRT, the semen samples were analyzed by CASA as described above.
Mitochondrial membrane potential and plasma membrane stability were assessed using the combination of MitoStatusRed (MST; mitochondrial potential) and Yo-Pro ® (YP; Y3603 Life Technologies, Darmstadt, Germany).For this purpose, YP (25 nM) and MST (20 μM) were added to the 500-μL semen aliquots extended in TALP-PVA, and the samples were incubated for 20 min at 37 ℃.

Statistical analysis
Results were subjected to analysis of variance considering repeated measures using the PROC MIXED procedure of the SAS program (Statistical Analysis System, version 9.4.).The following statistical model was adjusted: in which y is the response variable, µ is the overall mean, RFI is the effect of RFI class (low, high), evaluation is the effect of evaluation class (i = 1, 2 or i = 1,..., 4, depending on the variable), age is the linear effect of the covariate age of animal at evaluation, and e is the error.The repeated measures of the same animal were modeled considering compound symmetry (CS) as residual (co)variance structure.
Means were adjusted by the least squares method (LSMEANS) and compared by the probability of difference (PDIFF), when necessary.Statistical significance was set at P<0.05 and a tendency was considered when 0.05>P<0.1.

Results
Animals classified as low RFI exhibited higher RI and PI (P = 0.019 and P = 0.049, respectively) than high RFI animals.However, EDV tended to be lower in low RFI animals compared with high RFI (P = 0.065).In addition, a tendency towards greater testicular heterogeneity was observed in low RFI animals (P = 0.061).There were no differences in the other ultrasound variables evaluated between low and high RFI animals (Table 1).
No differences were observed in the characteristics of fresh semen between RFI groups, except for minor defects whose percentage was lower in semen of low RFI animals compared with high RFI animals (Table 2).Similarly, there were no differences in the sperm kinetic parameters evaluated between more and less efficient animals after thawing and after rapid TRT (Table 3).The percentages of cells with a stable plasma membrane (PMStable, Figure 2A) and MPAI (Figure 2B) were similar between low and high RFI animals.There was also no difference in the percentage of cells with high mitochondrial potential between low and high RFI animals (HMP, Figure 2C).However, the quality of high mitochondrial potential of stable cells (HMPStable, Figure 2D) was greater in low RFI animals than in high RFI animals (P = 0.013).P-value: RFI -between RFI classes in the analysis of thawed semen; Time -between time 0 (thawed semen) and TRT (30 min at 46 ℃); RFI×TRTinteraction between RFI and TRT.

Discussion
Spectral Doppler ultrasound analysis of the testicular artery showed a tendency towards a lower EDV in low RFI animals when compared with high RFI animals.The PSV and EDV represent the velocity at which blood flows through the analyzed blood vessel and reaches the tissue.According to Ortiz-Rodriguez et al. ( 2017), the higher the velocity at which blood passes through the testicular artery, the better the blood perfusion in the testis.Studies have reported a relationship between blood flow velocity and male reproductive capacity in different animal species (Ortiz-Rodriguez et al., 2017;Gacem et al., 2020;Lemos et al., 2020).
In the present study, higher RI and PI were observed in low RFI animals compared with high RFI animals.According to Pozor and McDonnell (2004), PI and RI are more sensitive markers of blood flow than EDV and PSV, since these indexes provide not only information about velocity but also about vascular impedance.The PI and RI represent the difficulty of blood to flow through the vessel; the higher these indexes in the testicular artery, the lower the testicular tissue perfusion and, consequently, the supply of oxygen and nutrients to the testes (Strina et al., 2016;Fávaro et al., 2020).
A: Percentage of spermatozoa with a stable plasma membrane (PMStable); B: percentage of cells with intact plasma and acrosome membranes (MPAI); C: percentage of cells with high mitochondrial potential (HMP); D: quality of the mitochondrial potential of stable cells (HMPStable).
The asterisk indicates a significant difference (P<0.05).
Figure 2 -Flow cytometry analysis of cryopreserved semen from bulls with low and high residual feed intake (RFI).Parenchymal organs require continuous blood flow, and the arteries that supply these structures typically have low resistance (Carvalho et al., 2008).
Analysis of pixel intensity of the testicular parenchyma revealed no differences between high and low RFI animals.This result corroborates the findings of Kowalski et al. (2017), who did not observe a difference between RFI classes in young developing Purunã bulls.On the other hand, Fontoura et al. (2016) and Bourgon et al. (2018) found higher maximum pixel intensities in animals with lower feed efficiency (high RFI).This difference might be related to the age of the animals in the cited studies since pixel intensity of the testicular parenchyma is higher before and during puberty (Brito et al., 2004;Rodrigues et al., 2020).The hypothesis to explain this difference is that spermatogenesis starts at a certain stage of development of the testicular parenchyma during puberty (Kastelic and Brito, 2012).Furthermore, the breed may also be a determinant factor, as reported by Rodrigues et al. (2020), who observed differences in testicular pixel intensity between zebu (Nellore) and taurine (Caracu) animals.
Observing the ultrasound results, the tendency towards a difference (P = 0.061) between RFI classes might be related to differences in the number or diameter of the seminiferous tubules, which could affect the heterogeneity of the testicular parenchyma (Brito et al., 2012).The testicular tissue is homogenous and moderately echogenic.This state can change during puberty or in the presence of testicular pathology that can alter homogeneity and increase the pixel intensity as a result of fibrotic processes (Kastelic and Brito, 2012).Despite these differences in the ultrasound parameters of the testicular artery and vascular parameters of the pampiniform plexus between low and high RFI animals, they were not sufficient to cause differences in SC or in the quality of sperm motility in these animals.
The mean SC did not differ between low and high RFI animals.Similar results have been reported in previous studies comparing bulls with distinct RFI values (Wang et al., 2012;Fontoura et al., 2016;Kowalski et al., 2017).On the other hand, Awda et al. (2013) and Bourgon et al. (2018) observed a greater SC in high RFI animals, but the difference decreased when the animals received better-quality diet, suggesting that this difference in reproductive parameters between low and high RFI animals is due to the energy distribution for maintenance and production and reproductive traits.The authors suggested that, in animals with low RFI, reproductive parameters may have a lower priority, a fact delaying sexual maturity.
Although several studies have associated vascular parameters with semen quality (Gloria et al., 2018;Hedia et al., 2019;Gacem et al., 2020), the higher EDV in less efficient animals and the higher RI and PI in more efficient animals observed in the present study were not sufficient to cause alterations in the seminal parameters studied.Evaluation of the parameters of fresh and thawed semen and after TRT showed that feed efficiency did not affect sperm kinetics, since no differences in CASA parameters were detected between low and high RFI animals.The results of the present study corroborate other studies that evaluated sperm motility or progressive motility in low and high RFI animals (Awda et al., 2013;Fontoura et al., 2016;Bourgon et al., 2018).However, some authors changed the division of low and high RFI classes and reported different results.Including body composition traits in the equation for calculating RFI, Fontoura et al. (2016) found higher total and progressive sperm motility in less efficient (high RFI) animals when compared with more efficient (low RFI) animals, which was not observed in the present study.On the other hand, Wang et al. (2012) observed lower sperm motility in low RFI animals.However, this difference was not sufficient to reduce the fertility of breeding animals, with the most efficient animals having a larger number of offspring.It should also be noted that the mean values reported by the cited authors are considered excellent for andrological examination (Penny, 2010;CBRA, 2013;Chenoweth and McPherson, 2016).
When the sperm morphology of fresh semen was analyzed, we found only differences in the percentage of minor defects, with more efficient (low RFI) animals exhibiting a smaller number of defects than less efficient (high RFI) animals.This result contradicts most of the studies that compared sperm morphology between low and high RFI bulls and did not observe any difference (Wang et al., 2012;Bruinjé et al., 2019) or observed higher percentages of sperm pathologies in low RFI animals (Fontoura et al., 2016;Bourgon et al., 2018).Despite the significant difference in sperm morphology observed here between animals with distinct RFI, the mean values are within the range recommended by different andrology handbooks (Penny, 2010;CBRA, 2013;Chenoweth and McPherson, 2016).
The percentages of spermatozoa with a stable plasma membrane, with intact plasma and acrosome membranes, and with high mitochondrial potential were similar between low and high RFI animals.These results corroborate some of the findings reported by Bruinjé et al. (2019); however, these authors reported a higher percentage of mitochondrial respiration activity in spermatozoa from low RFI animals, while the proportion of cells with low mitochondrial potential was higher in these animals.Other studies observed higher rates of mitochondrial activity in liver tissue (Lancaster et al., 2014), in longissimus dorsi muscle (Kolath et al., 2006), and in lymphocytes (Ramos and Kerley, 2013) of more efficient animals (low RFI).
In the present study, although there was no difference in the proportion of cellular respiration of sperm between the two RFI classes, more efficient animals (low RFI) exhibited a better quality of cellular respiration of stable cells in cryopreserved semen.These results may explain why the lower blood flow observed in the testicular artery of low RFI animals was not sufficient to change the sperm kinetics of fresh or thawed semen.One hypothesis would be that sperm cells of low RFI bulls are more efficient in energy production, requiring less blood supply, or that they are adapted to a lower nutritional demand, since mitochondria can be partially influenced by the surrounding environment, particularly by other organelles (Keil et al., 2011).
The results suggest that RFI does not influence sperm kinetics nor the sensitivity of sperm to cryopreservation; however, feed efficiency influences blood flow in the vascular cone, increasing the difficulty of blood to pass through the testicular arteries and to reach the testes in more efficient animals.However, cellular metabolism may have compensated for the lower availability of nutrients for sperm cells.

Conclusions
The lower blood flow in the pampiniform plexus of low RFI bulls possibly resulted in greater heterogeneity of the testicular parenchyma evaluated by B-mode and Doppler ultrasound.On the other hand, the reduced blood flow in the pampiniform plexus of low RFI bulls was not sufficient to change sperm kinetics, indicating that the RFI class does not affect the quality of fresh semen, thawed semen, or semen after rapid thermoresistance testing.

Figure 1 -
Figure 1 -Flow diagram of the experimental period.

Table 1 -
Least square means of ultrasound-measured testicular traits of Nellore bulls according to residual feed intake (RFI) class

Table 2 -
Least square means of sperm kinetic parameters of fresh semen from Nellore bulls according to residual feed intake (RFI) class SC -scrotal circumference; TM -total motility; PM -progressive motility; RAP -rapid motility; VAP -average path velocity; VSL -straight line velocity; VCL -curvilinear velocity; ALH -amplitude of lateral head displacement; BCF -beat cross frequency; STR -straightness; LIN -linearity; SEM -standard error of the mean.

Table 3 -
Least square means of sperm kinetic parameters after thawing and rapid thermoresistance testing of Nellore bulls according to residual feed intake (RFI) class