Subfertility, sometimes associated with negative energy balance (NEB), has a detrimental impact on the productivity of dairy systems. Lipid supplementation in early lactation may partially attenuate the NEB; however, the positive influence of lipid supplementation on reproduction is independent of the cow’s energy balance [1 ] . [2] showed that the type of fatty acid (FA) that is offered to animals through the diet, and not the addition of extra energy, stimulates the ovary and generates the development of follicular structures of large size. In this regard, it has been suggested that long-chain dietary FA, particularly polyunsaturated FA (PUFA), can improve the reproductive performance of dairy cattle.

There are two main families of essential PUFA called omega-3 and omega-6 that could affect fertility. Omega-3 is a precursor of the trienoic (3 series) prostaglandins and omega-6 of the dienoic (2 series) prostaglandins. The main source of omega-6 FA is linoleic acid from the diet (C18:2n-6), predominant in sunflower and soybean oils, and can be converted to arachidonic acid (C20:4 n-6), which is a precursor of dienoic prostaglandin (PG), such as PGF2α. The same elongase and desaturase enzymes, also convert the main omega-3 FA of the diet (α-linolenic acid; C18:3n-3), abundant in linseed oil, to eicosapentaenoic acid (EPA; C20:5 n-3), which is a precursor of the trienoic PG, such as the PGF3α [3] that has less biological activity on the corpus luteum (CL). There is an enzymatic competition between omega-3 and omega-6 for the desaturation and elongation, as well as for the PG synthetase site. The composition of phospholipids in cell membranes is related to the lipids from the diet, which determines the amount and type of prostaglandins synthesized from them [4] [5] . Therefore, an increase in the supply of omega-3 favors the synthesis of PGF3α instead of PGF2α in endometrial cells and contributes to a reduction in embryonic mortality [6] [7] by reducing luteolytic activity.

Most embryonic losses in cattle occur between 8 and 16 days after AI [8] due to some embryos may not reach a sufficient size at that time to inhibit the synthesis of PGF2α and, therefore, luteolysis ensues [9] . The loss of embryos in such cases may be associated with their inability to inhibit the luteolytic action of PGF2α during the critical period of maternal recognition of pregnancy [10 ] . In this context, it’s proposed that the inhibition of PGF2α synthesis could increase embryonic survival rate and pregnancy rate [11] .

On the other hand, diet lipids could stimulate ovarian activity by improving follicular development, increasing progesterone production, and/or decreasing progesterone clearance, which could lead to higher pregnancy rates [12] .

Even though the supply of omega-3 FA has a significant positive effect on the reproductive performance of dairy cows in intensive production systems (100% TMR diets), studies referring to the effect of omega-3 FA on reproductive and fertility indicators in pasture-based production systems are very scarce [13] .

This experiment aimed to evaluate the effects of C18:3n-3 supplementation on energy balance and ovarian dynamics in early lactation-grazing dairy cows. We hypothesized that the supplementation with rumen-protected FA rich in C18:3n-3 improves ovarian function without affecting energy balance in grazing dairy cows.

Treatment did not affect total DMI (22.7 vs. 22.1 kg∙DM∙day⁻¹ in ALA and CON cows, p = 0.61, Figure 1 ). However, treatment had an effect on concentrate intake given that ALA consumed less than CON (3.9 vs. 5.9 kg∙DM∙day⁻¹, respectively, p < 0.01, Figure 1 ). Treatment tended to affect the intake of pasture given that ALA consumed more than control (5.6 vs. 3.4 kg∙DM∙day⁻¹, p = 0.06, Figure 1 ) and the intake of PMR (13.1 vs. 12.8 kg∙DM∙day⁻¹ for ALA and CON respectively, p = 0.08, Figure 1 ). Treatment did not affect total ME intake (62.6 vs. 60.8 Mcal∙day⁻¹ in ALA and CON cows, p = 0.44, Figure 1 ), but it had a trend to affect the intake of ME from pasture (14.6 vs. 9.3 kg∙Mcal∙day⁻¹ for ALA and CON respectively, p = 0.06, Figure 1 ), the intake of ME from concentrate (13.7 vs. 18.0 kg∙Mcal∙day⁻¹ for ALA and CON respectively, p < 0.01, Figure 1 ) and, also, the intake of ME from PMR given that ALA consumed more than CON (34.3 vs. 33.4 kg∙DM∙day⁻¹, p = 0.08, Figure 1 ).

Treatment did not affect milk production (33.4 vs. 33.8 kg∙day⁻¹ in ALA vs. CON cows, p = 0.72). Regarding milk fatty acids composition, milk from cows supplemented with calcium salts of linseed oil had practically two times the concentration of α-linolenic-acid (ALA, C18:3n-3) to control group (1.0 vs. 0.48 g/100 g of FA; p < 0.01). This difference could be explained by the supplementation and may demonstrate C18:3n-3 intestinal absorption. For further details see González et al. (2020).

Our findings that supplementation with protected linseed oil in grazing dairy increases the size of CL, and blood progesterone concentration without affecting BW, BCS, glucose, BHBA, insulin, GH and IGF1, total DMI, total ME intake, and milk production supports our working hypothesis stating that rumen protected FA rich in C18:3n-3 improve ovarian function without affecting energy balance in grazing dairy cows. In this experiment, there was no effect on the total DMI and ME intake in cows supplemented with ALA, which may explain the absence of differences in milk production between treatments.

It is known that hormones may play an important role in coordinating the partition of dietary fatty acids between milk fat secretion, deposition in adipose tissue, and body lipid mobilization. Some hormones like GH, IGF-1, and insulin, are crucial when evaluating energy balance. At the time of the experiment, cows were in 58.0 ± 17.0 DIM, so we considered that the concentrations of these hormones should be within the normal range and treatment would not have a major effect on them. There are also no differences in BCS that suggest a different anabolic state between treatments.

The reduction of starch consumption in the ALA group could lead to a reduction in glycemia and insulinemia, but this did not happen in this trial. Blood samples were taken after ingestion of the supplement during milking time, but no differences were observed in these two parameters. The explanation for this may be that fat supplementation reduces glucose oxidation [22] .

Plasmatic metabolites such as βHBA and glucose and regulatory hormones such as insulin, GH, and IGF-I act as dynamic or short-term indicators of energy balance [23] . Cows in a negative energy balance show a decrease in plasmatic glucose, insulin, and IGF-I concentrations and an increase in βHBA and GH levels [24] . Taken together, the results obtained in this study (similar plasma concentrations of βHBA, glucose, insulin, GH, and IGF-I) suggest that supplementation with rumen-protected linseed oil in replacement of corn as an energy source did not affect the energy metabolism of dairy cows in early lactation. This lack of ALA effect on βHBA was in accordance with the absence of variations in metabolic parameters and with a similar GH/insulin ratio in both groups ( Table 2 ). In this sense, [24] , suggested that the hormonal profile may play a more important role than the type of diet (fat) in regulating body lipid mobilization in the early-lactation dairy cow.

In cattle, GH is the main hormone stimulating milk production through direct and indirect mechanisms (IGF system and interactions with insulin) [25] . Hepatic production of IGF-I is positively correlated with both, energy intake and plasma GH and insulin levels [26] . The lack of a significant effect of fat on IGF-I concentrations in the present experiment is consistent with similar GH and insulin levels observed ( Table 2 ) and with the similar (p = 0.44) energy intake of cows ( Figure 1 ).

Even though the sizes of preovulatory FOL were not different between groups, the size of the main CL at day 6 was significantly greater in cows of the ALA group. This could be explained by some effect generated by lipids on the development of the CL acting on ovarian intrinsic or extrinsic factors. In the endometrium, gene expression associated with different patterns of prostaglandin synthesis is a consequence of the n-6: n-3 ratio influencing the luteal lifespan [5] . There is no evidence of any luteotropic influences between days 5 to 7 of gestation. However, bovine embryos generated an anti-luteolytic stimulus on cocultured luteal cells [27] and this could also explain the major sizes of CL reported in ALA treatment that had greater conception rates.

The greater size of the main CL together with the higher progesterone concentration in ALA cows would explain the finding that ALA tended to increase pregnancy rate at 200 DIM by 26.3 percentage points (81.69% vs. 55.43%). A greater mean size of the CL has been reported for cows receiving omega-3 FA in the form of either formaldehyde-treated linseed [28] or whole linseed [29] and is a factor that contributes to an increase in the plasmatic concentration of progesterone [30] . [6] also found no effect on follicular dynamics, but the CL was larger in cows fed whole flaxseed treated with formaldehyde or a mixture of fish oil and formaldehyde-treated whole flaxseed.

[1] proposed the hypothesis that dietary fats can modulate the survival of the CL by reducing PGF2α secretion to improve pregnancy. One study [31] indicates that omega-3 FA causes disruption in the lipid microdomain structure and alters lateral mobility of the PGF2α receptor in bovine luteal cells. This could alter luteal sensibility to PGF2α and may be the cause of the difference in fertility presented between treatments. Even when the ratio of omega-6: omega-3 supplied through the diet changes, the pattern of PGF2α secretion changes as well [5] . Finally, the change in the fatty acid profile of the diet modifies the fatty acid composition of the follicular fluid. [32] found that the proportion of ALA in the follicular fluid in cows fed encapsulated flaxseed oil was 2.5 times higher than in control cows fed encapsulated saturated fatty acids. Therefore, the fatty acid profile could play a key role in the lifespan of the CL acting through intrinsic factors of the ovary.

Additionally, the greater progesterone concentration observed in ALA cows at day 21 coincides with the greater proportion of pregnant cows and greater size of CL at day 6 in this group. Progesterone concentration could be explained by the higher levels of cholesterol detected in these cows, given that cholesterol is a precursor for the synthesis of progesterone in luteal cells [1] . The concentration of plasma cholesterol is increased consistently under regimens of supplemental fat [33] . [34] proposed that an increase in dietary lipids would stimulate intestinal biosynthesis of cholesterol to meet the increased demand for fat absorption and transport, which would help explain the higher plasma cholesterol concentration in ALA ( Figure 2 ). In our study, the size of preovulatory FOL and the level of circulating IGF-1 are not affected by treatment, but, conversely, the size of CL at day 6 is ( Table 3 ). These findings are good because it has been reported that low IGF-1 concentrations could lead to low steroidogenesis [35] and, that small CL with poor progesterone secretion could lead to small follicle size and poor conception rates [36] . Also, low progesterone concentration could partially be responsible for reduced fertility [30] and increased pregnancy loss [37] in dairy cows.

We also found that ALA supplementation did not decrease the lipid mobilization and BW loss characteristic of early lactation cows [38] . Similar results were observed in early lactation grazing dairy cows [20] [39] [40] .

Finally, it is known that ALA supplementation improves fertility by reducing endometrial inflammation. However, further research is needed to investigate how lipids participate in the formation and function of the CL to enhance conception rates. As pointed out by [41 ] , the mechanisms through which fatty acids influence follicular development remain unknown.

As the supplementation with protected linseed oil in grazing dairy increases the size of CL, blood progesterone concentration, and it tended to increase conception rate by 200 DIM without affecting BW, BCS, glucose, BHBA, insulin, GH and IGF1, total DMI, total ME intake, and milk production we concluded that supplementation with rumen-protected FA rich in C18:3n-3 improves ovarian function without affecting energy balance in grazing dairy cows. The use of this supplement seems promissory; however, additional studies with greater numbers of animals are necessary to confirm its effect on reproductive performance.

This work was supported by the National Institute of Agricultural Technology (INTA). The authors thank Gabriel Schmidt, Ariel Plattner, and Dario Arias for their assistance with the fieldwork, and Tecnuar SRL.