Key messages

Novel, heritable traits that are measured earlier than current re-calving and mating traits could increase the accuracy of estimated breeding values (BV) for fertility and facilitate faster genetic gain in dairy cattle.

Using a unique research herd, we demonstrated that dairy cows with elite parent average Fertility BV had earlier puberty, shorter anogenital distance (AGD), more condensed calving, less time between calving and first ovulation, stronger and longer heats, earlier submission, and greater pregnancy rates than animals with an inferior parent average Fertility BV.

The suitability of age at puberty, AGD, oestrous traits (timing, heat strength and length), and timing of conception for predicting Fertility BV are now being investigated in a large-scale trial involving 5,000 dairy replacements across 54 herds.

The estimated heritability for age at puberty (about 30%) and its moderate genetic correlation with pregnancy rate during first lactation (about 0.45) support its further investigation as an early predictor trait for genetic improvement of fertility.

Background

Herd reproductive performance is a key determinant of farm productivity in seasonal-calving, pasture-based dairying systems, such as those operated in New Zealand, Australia, and Ireland. In New Zealand, herd reproductive performance is evaluated by the 6-week in-calf rate, defined as the percentage of the herd that conceives during the first 42 days of the seasonal breeding period; however, the national average 6-week in-calf rate of 68% (New Zealand Dairy Statistics 2020–21) is well below the industry target of at least 78%.

Many complex, interacting factors affect reproduction, including genetic merit for fertility. The Fertility Breeding Value (BV) was included in the New Zealand national breeding objective in the early 2000s to halt declining cow fertility, but subsequent genetic gain has been low. This is partially due to the low heritability (i.e.,≈5%) of re-calving and mating traits used to estimate the Fertility BV as well as the fact that sires have already been selected and widely used when daughter fertility phenotypes become available. The Fertility BV reflects the ability of sires to produce daughters that re-calve within the first 42 days of the herd’s planned start of seasonal calving (CR42) during second and subsequent lactations. Being presented for mating during the first 21 days of seasonal breeding (PM21) in first and subsequent lactations are also included in the Fertility BV as predictor traits.

A research programme commenced in 2013 to identify novel, more heritable traits that are collected earlier than the current re-calving and mating traits, which will increase the accuracy of the Fertility BV and facilitate a faster rate of genetic improvement in fertility. Firstly, we demonstrated that a superior Fertility BV results in improved reproductive performance, thereby increasing farmer confidence in using genetic selection for fertility. Secondly, we aimed to identify underlying predictors for genetic variation in fertility. The methodology used was to establish a research herd of ~550 Holstein Friesians with high/positive (POS; +5%) and low/negative (NEG; –5%) parent average Fertility BV (Meier et al., 2021a). To put divergence of these groups into context, the national average Fertility BV was +0.2% (standard deviation (SD) +2.0%) for all heifers born in 2015 (n561,675; www.dairynz.co.nz/ animal/animal-evaluation; accessed 17 Aug 2021). This herd, known as the ‘Fertility Animal Model’, enabled us to identify key points of difference in the biology underpinning various reproductive measures between cows that are genetically fertile or sub-fertile but otherwise similar in other traits such as live weight and milk production. Thirdly, we selected the most promising earlier-in-life novel traits and tested the practicality of measuring these phenotypes at scale and, moreover, their suitability as genetic predictors of subsequent reproductive success. This work, consisting of the ‘Puberty Scale-Up (PSU)’ and the ‘Fertility Scale-Up (FSU)’ trials, is enabling us to provide recommendations on which novel phenotypes to pursue for further investigation as predictor traits in the Fertility BV. Here, we summarise the key findings to date.

Fertility Animal Model

In spring 2015, the POS and NEG Fertility BV heifers, born from contract matings in commercial herds, were collected at ~9 days old to establish the research herd. We collected samples and data from these animals as they progressed through the heifer rearing phase (2015–17; Meier et al., 2021a), and during their first (2017–18) and second (2018–19) lactations (Meier et al., 2021b).

Heifer performance

The Fertility BV was associated with differences in the onset of puberty and heifer reproductive performance (Table 1; Meier et al. 2021a). The POS Fertility BV heifers reached puberty earlier and at a lower live weight and percentage of mature live weight than the NEG heifers, which meant that more POS heifers had reached puberty by the start of seasonal breeding. Consequently, the POS heifers conceived, on average, 3.6 days earlier, with a higher pregnancy rate (PR) during their first breeding season. Our results indicated that the timing of puberty and conception in heifers are potential early predictor traits of genetic merit for fertility.

Table 1. Heifer reproductive parameters in animals with a POS (+5%) relative to a NEG (-5%) Fertility Breeding Value. PRx = pregnancy rate after x days of breeding. Adapted from Meier et al. (2021a).

	POS	NEG	Difference	P-value
Eligible heifers (n)	275	248	+27	-
Age at puberty (d)	358	385	-27	<0.001
Live weight at puberty (d)	274	294	-20	<0.001
% mature live weight at puberty (%)	51	55	-4	<0.001
Pubertal at mating start date (%)	94	82	+12	<0.001
Interval from mating start date to conception (d)	13.0	16.6	-3.6	<0.001
PR21 (%)	75	62	+13	<0.05
PR42 (%)	90	82	+8	<0.05
PR98_final (%)	98	94	+4	<0.05

 

Calving pattern

The earlier puberty onset and conception of the POS heifers advanced the calving pattern during first lactation by an average of 4 days relative to the NEG heifers (Meier et al., 2021b). By lactation 2, the difference was 12 days earlier, with 16% more POS animals re-calved within the first 42 days of the seasonal calving period. Hence, timing of the calving that initiates first lactation is a useful, earlier trait than PM21 or CR42.

Resumption of cycling, submission rates and reproductive treatments

Substantially more POS than NEG cows were submitted during the first 21 and 42 days of artificial breeding during both first and second lactations (Table 2; Meier et al., 2021b). The extremely poor submission rates for NEG Fertility BV cows were due to the large percentage of NEG cows still anoestrous after 42 days of breeding, which required CIDR hormonal interventions (Table 2). Furthermore, the NEG Fertility BV cows that cycled spontaneously postpartum (i.e., without interventions) still had a 9 day longer anoestrous interval in both lactations. The ability to resume oestrous cyclicity postpartum is a key driver of reproduction and is captured in the current PM21 and CR42 traits included in the Fertility BV. Nevertheless, the inclusion of additional measures, such as timing of first heat postpartum, pre-mating cycling rate or accounting for hormonal interventions may result in more accurate Fertility BV.

Table 2. First and second lactation reproductive parameters in animals with either a POS (+5%) or NEG (-5%) Fertility Breeding Value. SRx = submission rate after x days of breeding; PRx = pregnancy rate after x days of breeding. All data are corrected for calving season day relative to 1^st June each year. Final PR are corrected for use of CIDR hormonal treatments for anoestrus after 42 days of breeding. Adapted from Meier et al. (2021b).

	POS	NEG	Difference	P-value
1st lactation
Cows (n)	249	216	+33	-
SR21 (%)	87	49	+38	<0.001
SR42 (%)	95	55	+40	<0.001
PR21 (%)	54	26	+28	<0.001
PR42 (%)	67	34	+33	<0.001
PR98_final (%)	78	68	+10	0.294
Conception rate to 1st service (%)	57	52	+5	0.499
CIDRs after 42 days (%)	5	46	-41	<0.001
Interval from mating start date to conception (d)	20	31	-11	<0.05
2nd lactation
Cows (n)	204	121	+83	-
SR21 (%)	88	63	+25	<0.001
SR42 (%)	94	73	+21	<0.001
PR21 (%)	45	29	+16	0.245
PR42 (%)	74	44	+30	0.132
PR76_final (%)	86	72	+14	0.238
Conception rate to 1st service (%)	54	51	+3	0.4814
CIDRs after 42 days (%)	6	30	-24	<0.001
Interval from mating start date to conception (d)	22	34	-12	<0.005

 

Timing of conception and pregnancy rates

On average, the POS Fertility BV cows conceived 12 d earlier (relative to the mating start date) than the NEG Fertility BV cows (Table 2; Meier et al., 2021b). Differences in PR in both lactations 1 and 2 were similar, with approximately 30% more POS cows pregnant by 42 days of breeding compared with NEG. The superior reproductive performance of the POS cows was driven by their greater ability to resume cycling post calving, which could not be explained by differences in milk production or metabolic status. Furthermore, the high percentage of NEG cows that failed to get in calf resulted in high culling rates and, therefore, poor survival rates between lactations. There is an opportunity to use foetal aged pregnancy diagnosis records, which are now routinely collected on 2 million cows annually, and are available earlier than calving records, to improve the accuracy of the Fertility BV.

Oestrous activity characteristics

Oestrous strength and duration, measured using activity monitoring devices, are candidate traits worthy of further investigation. Cows with POS Fertility BV had longer (~1 h), more active oestrous events than NEG cows, but the inter-oestrous interval was similar between the two groups (Reed et al., 2022). These stronger, longer heat events may support easier heat detection, enabling more animals to be bred at the correct time. Similar differences in oestrous activity between the POS and NEG animals were evident as maiden heifers (Reed et al. unpublished data).

Anogenital distance

The POS cows had a shorter anogenital distance (AGD; the distance between the anus and the genitals) than NEG cows when measured at about 29 months old during first lactation (Grala et al., 2021). Furthermore, reproductive performance of both POS and NEG heifers was substantially less in those with longer AGD (Grala et al., 2021), indicating that AGD is a promising trait for improving accuracy of Fertility BV.

Puberty/Fertility Scale-Up Trials

We are currently testing novel fertility traits in a 2018-born population of 5,000 Holstein-Friesian and Holstein Friesian x Jersey crossbred animals across 54 herds. Our first objective is to determine the phenotypic variability and heritability of our measures of age at puberty, AGD, and oestrous characteristics. Our second objective is to determine if these novel, earlier-in-life traits are predictive of reproductive success during first and second lactation.

Age at puberty and anogenital distance

Herds were visited on three occasions to collect blood samples when the average (±SD) ages of the heifers in each herd were 299±15 days old (visit 1; V1), 327±15 days (V2) and 355±15 days (V3). The visits were timed to capture variation in puberty onset between animals. Age at puberty was defined as the age of the animal at the visit when their blood progesterone concentration reached a threshold of ≥1 ng/mL for the first time, or their age at V3 plus 31 days if they did not exceed this threshold in any collected samples. Live weight, stature (height and length) and AGD were measured at V2.

The average age at puberty was 352±35 d, with 20% of heifers having attained puberty by V1, 39% by V2 and 56% by V3. We observed a large variation in the rates of puberty among herds at each visit as well as in the age at puberty among individual heifers, which were phenotypically associated with body measures (i.e., a younger age at puberty was detected in animals with a heavier live weight, both absolute and as a % mature live weight, and a larger stature at V2). Furthermore, heifers with a greater % Jersey reached puberty earlier, whereas those with a greater % overseas Holstein ancestry reached puberty later, indicating that routine phenotyping strategies will need to consider these breed differences.

Using our scalable measure of puberty onset, a moderate heritability of age at puberty of about 30% was estimated from pedigree or genomic data. This is considerably larger than the ~5% heritability of either PM21 or CR42. Furthermore, our results demonstrate that just two, or even one, herd visits to obtain blood progesterone tests could provide sufficient data to rank animals accurately for age at puberty, which has important implications for routine data collection. Likewise, AGD is about 25% heritable. Reasonably rapid gain can, therefore, be made through genetic selection for these traits.

Pregnancy during first lactation and its correlation with heifer traits

Foetal-aged pregnancy diagnosis was undertaken between 11 to 14 weeks after the herd’s start of mating during the first lactation. Several different traits defining reproductive outcomes were investigated with PR42 (a binary PR trait indicating the ability to conceive during the first 42 days of breeding) being the most robust for genetic parameterisation purposes. Current analyses indicate that the genetic correlation between age at puberty and PR42 is about 0.45, with a credibility interval of about 0.25 to 0.60. This moderate genetic relationship indicates that around 20% of the genetic variance in PR42 can be explained by age at puberty. We are now following PSU/FSU trial animals through their second lactation to determine how these relationships hold with successive parities.

Oestrous activity derived traits

A subset of 2,000 animals across 17 herds had activity monitoring devices attached to a hind leg during the PSU trial and again during the FSU trial from 2–3 weeks before calving until pregnancy diagnoses. These data are being analysed to derive oestrous activity traits, including age at first detected oestrus as a proxy for age at puberty, interval from calving to first heat, pre-mating cycling rate, and oestrous strength and duration. Wearable devices offer opportunities to collect fertility phenotypes across large numbers of animals.

Conclusion

Overall, the POS relative to NEG Fertility BV was associated with earlier puberty, shorter AGD, more condensed calving, reduced time between calving and first ovulation, stronger and longer heats, earlier submission, and greater PR. Our results should increase farmer confidence that selection for cow fertility will lead to tangible improvements in herd reproductive performance and have supported a revision of the Fertility BV, launched December 2021, to better reflect the timing of calving from first lactation onwards. Importantly, we have determined biological differences controlling reproduction that could be used to develop novel, early-in-life traits to accelerate genetic gain enabling faster improvements in a herd’s inherent fertility. Measures related to age at puberty, AGD, oestrous traits (timing, heat strength and length) and timing of conception are now being tested at scale. Preliminary results suggest age at puberty is a suitable early predictor trait of reproductive success for genetic evaluations.

Acknowledgements

This work was conducted under ‘Pillars of a New Dairy System’ (www.dairynz.co.nz/pillars), an 8-year partnership programme funded by dairy farmers through DairyNZ Inc. and the Ministry of Business, Innovation and Employment (DRCX1302), with aligned core funding for fertility from AgResearch. Industry support provided by Fonterra, LIC, and CRV is gratefully acknowledged.

References

Grala, TM., Price, MD., Kuhn-Sherlock, B., Burke, CR and Meier, S. (2021) Investigating anogenital distance and antral follicle count as novel markers of fertility within a herd of cows with positive or negative genetic merit for fertility traits. Journal of Dairy Science, 104: 12939–12952.

Meier, S., McNaughton, LR., Handcock, R., Amer, PR., Beatson, P., Bryant, JR., Dodds, KG., Spelman, R., Roche, JR. and Burke, CR. (2021a) Heifers with positive genetic merit for fertility traits reach puberty earlier and have a greater pregnancy rate than heifers with negative genetic merit for fertility traits. Journal of Dairy Science, 104: 3707–3721.

Meier, S., Kuhn-Sherlock, B., Amer, PA., Roche, JR. and Burke, CR. (2021b) Positive genetic merit for fertility traits is associated with superior reproductive performance in pasture-based dairy cows with seasonal calving. Journal of Dairy Science, 104: 10382–10398.

New Zealand Dairy Statistics 2020–2021, http://nz-dairy-statistics-2020-21-dnz.pdf (dairynz.co.nz), accessed 15 Dec 2021

Reed, CB., Kuhn-Sherlock, D., Burke, CR. and Meier, S. (2022) Estrous activity in lactating cows with divergent genetic merit for fertility traits. Journal of Dairy Science, 105: in press.