Seasonal variation in bull semen quality demonstrates there are heat-sensitive and heat-tolerant bulls

Intra-bull semen quality variation

To understand variation in bull semen quality, we assessed 1271 ejaculates from 79 different bulls (11 different breeds) housed at Rockhampton stud farm, in the state of Queensland, Australia, over a period of 5 years (2014–2018). The raw data, together with the semen analysis and when the samples for each individual bull were collected is available in Supplementary 1. The climate in this area (23.3786° S, 150.5089° E) is considered sub-tropical, ranging from 16 °C in winter to over 30 °C in summer. A comprehensive semen analysis was undertaken, including sperm morphology and motility. To determine the variation in semen quality, we plotted the percentage of sperm normal forms for each bull that had 5 or more ejaculates taken annually. Morphology was used as a measure of sperm quality, as Söderquist et al.¹⁷ demonstrated that sperm motility is heavily influenced by the collection/collectors and, therefore potentially unreliable and irreproducible. This resulted in the analysis of 1178 ejaculates from 50 bulls, with an average of 23 ejaculates per bull. The percentage of sperm normal forms as a box and whiskers plot for each bull is given (Fig. 1). As shown, many bulls demonstrated extremely high variation between ejaculates, with several males ranging from < 50% normal forms (considered an outright “fail” in terms of cryopreservation potential) to > 70% (considered an outright “pass” in terms of cryopreservation potential) of normal sperm morphology. On the contrary, some bulls appeared to produce consistent semen samples across the year.

To determine the amplitude and the proportion of bulls demonstrating variation in the number of normal sperm forms, we measured the difference between the maximum and the minimum values recorded for each animal. From this analysis we found that: 9 (18%) bulls showed less than 20% variation in normal forms; 15 (30%) bulls had between 20–40% variation; 13 (26%) bulls were between 40–60% and for 13 (26%) bulls this number was over 60%. These data have major implications when interpreting semen analysis, since a bull could be classified as either fertile or infertile depending on which ejaculate was considered. This data also sheds light into why correlations between the vBBSE parameters such as morphology and the bull fertility are so variable.

Seasonal effect on semen quality

Several sources of environmental influence have been suggested to affect bull sperm quality. These include feed availability (i.e., higher conception rates in rainy seasons)²⁷, excessive protein intake²⁸, day length²⁹, thermal heat stress and age^30,31. To better understand the dynamics of semen quality variation within our samples, we plotted sample “pass” and “fail” cryopreservation criteria against the month of collection. A raw bull semen sample is classified as “pass” when motility is above 60% and normal forms greater than 70%. When samples were between 30 and 60% motility and 50–70% normal forms, they were classified as a “compensatory” (or qualified) pass (q-pass). The compensatory pass relied on there being the ability to have at least 10 million motile normal forms of spermatozoa in each straw to allow for conception. An outright failure was given to any sample with less than 30% progressive motility or 50% normal forms. This allowed each ejaculate to be placed into a binary “pass” or “fail”.

The data for the percentage of total males that “failed” within each month (1271 ejaculates) is shown (Fig. 2A). Clearly, there is a seasonal pattern, with over 90% pass rate in winter (June–August) that fell to 50% or lower in summer (Dec-Feb). Considering that all bulls were greater than 4 years old, housed on the same stud farm and received the same dietary supplement we found no relationship in terms of “pass” or “fail” rates to these parameters. Thus, the data clearly suggested that Temperature/Temperature-Humidity or day length were responsible for the increased failure rates seen during Summer. Therefore, to understand if there was any causal relationship, we correlated either the average monthly temperature (Fig. 2B) or daylight (Fig. 2C) with monthly failure rates. The data showed a correlation with monthly temperature (r² = 0.55; and temperature-humidity index – see further modelling below) but not with daylight hours (r² = 0.05). Combined, these data suggest that temperature was the most likely reason for increased failure rates during the warm/hot months.

Changes in normal sperm forms categorised by breed

The present study investigated 11 different breeds of cattle, and we reasoned that maybe one, or more breed(s) contributed to failure rates more than others. Therefore, we plotted the percentage of normal forms for every ejaculate against the breed (Fig. 3). All breeds showed similar variation except for the Belmont Red, Boran and Wagyu. However, a relatively small number of bulls from the Belmont Red and Boran breeds were assessed in this study, therefore, it is unclear if they are indeed more resistant to heat. In the case of the Wagyu, it is worth mentioning that only one animal exhibited poor sperm morphology in several ejaculates (Fig. 3 circled) during winter. A close inspection of the records showed that during this time the animal had a fever episode, with body temperature reaching 39.4 °C, and that the sperm morphology returned to normal in approximately 70 days.

Some bulls are heat-sensitive, whilst others are heat-tolerant

Analysis of the present data clearly illustrated that some bulls showed marked variation in terms of their semen quality throughout the year (Fig. 1). Meanwhile, others demonstrated much less variation, and were reasonably consistent. To further clarify these differences, we closely analysed the percentage of sperm morphology from two bulls, both of whom had several ejaculates were taken throughout the year, including during and after summer (Fig. 4). There was a clear pattern, and evidence of two types of bulls. Prior to the summer season, bull 1 (Fig. 4, red), designated here as “heat-sensitive”, exhibited > 70% normal forms of spermatozoa. This value decreases dramatically, reaching its lowest point (10%) mid-January, before undergoing a recovery by April (> 70%). In contrast, bull 2 (Fig. 4, green) showed a consistent semen profile throughout the year. The data suggest this bull was more “heat-tolerant”.

To further explore the concept of “heat-tolerant” and “heat-sensitive” bulls, we subjected 20 Wagyu bulls to a single event of controlled heat stress (40 °C, 12 h). This experiment was performed during Winter, at Singleton (New South Wales, Australia, 32.5695° S, 151.1788° E), where the average temperature was 17 °C and never exceeded 18 °C. Prior to the heat stress event, baseline semen samples were taken from each animal. After heat stress, semen samples were taken every week for 11 weeks. During the experiment, two bulls were removed from the program due to infection and sickness whilst a 3rd bull was removed as it refused to co-operate with electroejaculation procedure. From the remaining bulls, we were able to reproduce the heat-sensitive and heat-tolerant bull phenomenon. The raw data from this work is given in Supplementary 1, and an example of the data is shown (Fig. 5). For 14 bulls, we found no difference in terms of their baseline samples, which were between 70–90% normal forms. This is consistent with the Wagyu bull characteristics and their heat-tolerance (Fig. 5, yellow, green, blue lines). Within these “heat-tolerant” bulls, there was a variation of 16–22% sperm normal forms. For the other three bulls, two of them showed a decline in sperm quality, which began 2–3 weeks after the heat event, dropping from a baseline of 85% and 90% normal forms to 55% and 59%, respectively (30–31% variation in normal forms; Fig. 5, grey and orange line). The third bull showed a greater degree of heat-sensitivity. Starting at 77% morphologically normal sperm, the spermiogram of this bull illustrated a rapid decrease in normal forms in a short time (2 weeks), reaching around 40% after 4–5 weeks. Sperm morphology remained at this level (37% variation in normal form) for four weeks, before recovery. These data show that under experimental condition, the phenomenon of heat-sensitive and heat-tolerant animals can be reproduced. Further, it appears that there are degrees of heat-sensitivity.

Environmental heat stress leads to poor sperm quality 17 days later

Similar to previous reports, we noted that sperm quality does not begin to deteriorate until 2–3 weeks after the heat stress event of the bulls³². Based on the timing of spermatogenesis, this is consistent with reports that meiotic cells are more susceptible to heat stress following a heating event, with poor quality spermatozoa appearing in the ejaculate around 2–3 weeks later. To better understand the relationship between a “heat-event” and the production of poor-quality spermatozoa, we modelled both maximum temperature and maximum temperature humidity index (THI) and their relationship to the proportion of morphologically normal spermatozoa. The THI is an index representing the effect of humidity on the heat stress of an animal. THI was obtained using the following formula:

where T_max = maximum temperature, (^oF), and H = relative humidity.

We plotted the correlation between semen quality and _Tmax on the day, and every day prior (up to 40 days) to semen collection (Fig. 6). This modelling demonstrated that poor semen quality was due to maximum daytime temperature 17 days prior (Fig. 6a, arrow). Notably, 1 day of heat-stress appears to be sufficient to cause poor sperm quality, since if we take the average of 2 (Fig. 6b) or 3-day maximal temperatures prior to collection (Fig. 6c) the correlation patterns were similar. Supplementary 3 shows further modelling for T_max and THI using between 1 and 5-day average temperatures prior to semen collection.

Understanding the temperatures at which heat-sensitive bulls fail

To determine the T_max at which bulls in the paddock begin to produce poor quality spermatozoa, we modelled data using both parameters measured at 17 days prior to the heat event, and plotted samples from 12 heat-sensitive bulls (6 Brahmans, 4 Drought Masters and 2 Santa Gertrudis). The relationship between sperm morphology and T_max 17 days prior to heat even was plotted, with a spline smoothing cure to show the mean quality as a function of T_max (Fig. 7a). As the temperature increase, so the quality of sperm morphology decreases as expected. To gain further clarity, we next fitted a nominal logistic regression analysis to model the proportion of spermatozoa that would either pass, Q-pass or fail sperm cryopreservation criteria as a function of T_max 17 days prior. T_max effect was highly significant for both outcome categories, with both p < 0.001. The predicted proportions from the model for each pass condition at varying T_max were plotted (Fig. 7B). At lower temperatures (< 22 °C), the proportion of samples that would be expected to pass (0.90, solid line) or qualified-pass (Q-pass) (0.08, dotted line) summates to 0.98. This is similar to the actual number of samples that pass during winter (Fig. 2). However, as temperatures rise, the number of samples designated as “pass” begins to drop, while the number of qualified-pass samples increase up to 30.5 °C. These data show that a “q-pass” is likely reflecting the beginnings of the effect of heat stress. At this point, the proportion of samples with a q-pass has reached 0.38, whilst the proportion of samples with a direct “pass” has dropped to 0.22, giving and overall, 0.6 probable pass rate. At 34 °C, the proportion with an overall pass (pass 0.07 + q-pass 0.30) drops to 0.37 and continues to drop as temperatures rise above 34 °C. From a practical perspective, this data show that for heat-sensitive bulls, with temperature histories 17 days earlier of around 31 °C, 40% of the sperm samples collected will fail a spermiogram assessment and once the temperature exceeds 34 °C the failure rate will exceed 63%. As such, these data suggest that within heat-sensitive bulls, there are clearly some bulls that are more sensitive than others.

Climate change modelling shows bulls are subjected to more heat events per decade

To understand the impact of global warming, we downloaded historical temperature data from the Bureau of Meteorology using data from the Rockhampton site (bom.gov.au). We then counted the number of days/decade the maximum temperature exceed 32 °C–45 °C respectively. These data are shown in Supplementary Table 3. To give an example of “extreme” climate that would likely impact all heat-sensitive bulls, we plotted the number days/decade which were higher than 38 °C, 39 °C and 40 °C (Fig. 8) from the area in which the bulls were housed. As shown, the number of days per decade over these extreme temperatures jumps from 19, 7 and 1 (1939–1949; 38 °C, 39 °C and 40 °C respectively) to 75, 39 and 15 in the last decade. These data show the urgent need to identify heat-tolerant bulls as future sires.

Source: Ecology - nature.com