At mating, as well as during conventional artificial insemination (AI), semen is deposited into the cervical canal of the female. Semen transport along the uterine horns is elicited by myometrial contractions and a certain number of spermatozoa reach the uterotubal junction (UTJ) and the first part of isthmus. In isthmus a sperm reservoir forms in order to (i) escape reflux and phagocytosis in the uterine lumen and (ii) ensure that sufficient numbers of potentially fertile spermatozoa are present at the site of fertilization when oocytes are released 1. Prostaglandin (PG) F_2α is known to be involved in myometrial contractions 234. Langendijk et al. 5 demonstrated that both suppression and stimulation of uterine contractions adversely affect fertilization in sows. Measurement of cortisol in blood is a traditional methodology for monitoring and evaluating stress in pigs, even though it is known that cortisol is not solely a stress-mediated hormone. Glucocorticoids are complex anti-inflammatory drugs and are thought to alter gene expression by as much as 1% in their target cells 6. An increase of cortisol in response to mating appears to be a normal part of the mating process in the female pig 7. Our previous studies indicated an interaction between cortisol and prostaglandin F_2α metabolite (PG-metabolite) in sows during AI depending on the type of housing 8. Madsen et al. 9 reported that AI resulted in an increase of PG-metabolite in sows and that no increases were seen in sows mated by boar. Recently, the elevation of PG-metabolite has been seen also in sows inseminated IUI 10. The aims of the present work were to study whether there are any relationships between cortisol and PG-metabolite in mated sows or inseminated with the intrauterine technique and compare these to changes occurring in conventionally inseminated sows.

Twenty-five out of 33 sows, which were used in this study, became pregnant and farrowed at full term. The evaluation of the results is from these 25 animals, 8 sows in Boar-group, 9 sows in IUI-Group and 8 sows in AI-Group. There were no differences in the number of total born piglets per litter (mean ± SEM) from sows mated, intrauterinally inseminated and conventionally inseminated (13.9 ± 0.5, 15.6 ± 1.3 and 13.1 ± 1.0, respectively).

This study demonstrates that cortisol concentrations are higher in mated sows than in sows conventionally inseminated. On the other hand, both types of artificial insemination induce a dramatic increase of PG-metabolite, which is not seen in mated sows. It is known that mating of gilts results in a significant rise in plasma cortisol levels with a duration of approximately 30 min 13. A significant transient increase in plasma concentrations of cortisol was also seen in oestrous gilts after introduction to the boar but not in gilts during the back pressure test 14. In our previous study, we have shown that in sows housed in individual pens, concentrations of cortisol did not change after service and were significantly higher than in sows housed in crates before service 8. Even though all sows were kept in crates in the present study, we have seen that sows in IUI-group had higher concentrations of cortisol before service than mated sows. The reason for high concentrations of cortisol might be due to fear of humans 15. Thus, it is plausible that sows, which were randomly chosen for intrauterine insemination were more susceptible for handling before service. This could result in an early activation of the hypothalamic-pituitary-adrenal axis. Macaulay et al. 16 reported that in heifers an increase in plasma cortisol after mating was higher than after AI, with peaks after 15 and 60 minutes respectively. On the other hand in ewes there was no increase in cortisol during mating 17. Houdeau et al. 18 made a comparable study on ewes, where plasma cortisol levels were measured during difficult and easy AI procedures. Their results showed the same outline independent of procedures, a quick but diminutive rise in plasma cortisol, reaching its maximum after 10 minutes, and with 30 minutes duration.

In inseminated sows from our previous study 8, basal level of PG-metabolite around 400 pmol/l and maximum values around 1000 pmol/l were seen. In the present study we found that the basal level of PG-metabolite was around 200 pmol/l and maximum values were around 560 pmol/l. We have no reasonable explanation for this discrepancy. However, it is important to know that the relative increase of PG-metabolite after service is similar i.e. approximately 2.5 fold. Claus et al. 19 reported that mating immediately increased PG-metabolite concentrations to a maximum 10 min after the onset of ejaculation, and this maximum coincided with maximal oestradiol concentrations in peripheral blood plasma. These authors suggested that oestrogens in seminal plasma, after natural service, lead to release of PGF_2α from the endometrium and this might explain the increase of myometrial contractions. The finding of PG-metabolite elevations in sows after mating is in contrast to our present, as well as previous results 89. At present, we have no reasonable explanation for this discrepancy. However, a dramatic increase of PG-metabolite in inseminated sows seen in our present, as well as previous studies 89 is in agreement with studies performed by Willenburg et al 4. The inclusion of oestradiol in the extended semen dose prior to insemination did not result in a higher production of PG-metabolite (1 h after AI) as did vaginal deposition of oestradiol in corn oil compared to control inseminated sows. Taken together, there is a causal link between oestradiol inducing a uterine release of PGF_2α, and an enhanced uterine contractility 4. Wongtawan et al. 20 suggested that cervical stimulation during intrauterine insemination trigger the release of other hormones than oxytocin such as prostaglandins. In earlier studies, Langendijk et al. 21 measured myometrial activity in oestrous sows during infusion of different substances. These authors could not see any differences in contractions between saline and seminal plasma, but oestrogens, cloprestenol (prostaglandin analogue) and clenbuterol altered the contractions significantly. Cloprostenol increased whereas clenbuterol decreased both frequency and amplitude of uterine contractions. Both suppression and stimulation of uterine contractility might adversely affect fertilization in sows 5. These authors concluded that new studies are required to establish an optimal level of uterine contractility and to investigate the importance of the timing of uterine contractility relative to the moment of insemination. This is in line with studies by Cao et al. 2, who reported that low concentrations of prostaglandins stimulate and high concentrations inhibit contractions of porcine myometrium. It is also plausible to speculate that these effects depend on the presence of contractile and relaxant prostanoid receptors in the cornua of the porcine uterus 22.

Semen transport to the site of fertilization in pigs is not only dependent on myometrial contractility but also on the spontaneous motility of the oviduct 23. Intraluminal pressure in the porcine oviduct was not affected by saline, oestrogen, boar seminal plasma or boar semen. However, after deposition of oestrogen solution in the uterus and AI with 100 ml of boar semen, a clear increase of PG-metabolite could be seen 2425. An exogenous administration of PGF_2α increased the peripheral plasma levels of PGF_2α metabolite and produced an increase in the frequency of the phasic pressure fluctuations in the porcine oviductal isthmus 26. Interestingly, Rodriguez-Martinez et al. 27 reported that concentrations of PGF_2α in the oviductal fluid were high during the first day and reached a maximum during the 2^nd day of oestrus in gilts. These authors concluded that the collection catheter and/or manipulation at surgery have a clear effect on the prostaglandin F_2α production and release. Therefore, a dramatic increase of PG-metabolite seen in intrauterine and conventionally inseminated sows might be associated with the insertion of the catheter. The other plausible explanation for such dramatic increase of PG-metabolite might be the effect of semen extender, for example, as EDTA used in our study. Findings of Willenburg et al. 4 might indicate that the extender that was used in their study also stimulate production of PG-metabolite in inseminated sows.

An intravenous injection of CRH results in elevated cortisol levels, with a peak reached at 20 min without any effect on plasma PG-metabolite levels in cycling gilts or in castrated boars 28. In contrast to CRH-treated pigs, an increase of PG-metabolite followed by a cortisol peak approximately 40 min later were seen after the administration of ACTH in ovariectomized gilts, cycling gilts and castrated boars 928. In the present study cortisol levels were elevated earlier than PG-metabolite in AI-groups, which might suggest that ACTH is not involved in elevating cortisol levels during AI. Furthermore, a study by Montgomery et al. 29 showed that injection of ACTH gives a peak in cortisol after 4 hours and it takes 9–10 hours before cortisol levels are back to a basal stage. Lang et al. 30 reported that it takes one hour for the cortisol to reach a maximum and one hour more for cortisol to return to basal levels after ACTH injections in sows. However, our results demonstrate a rise in plasma cortisol level almost immediately, with a peak 15 minutes after service. Observations have been made that CRH, can possibly act directly or indirectly to increase cortisol secretion further than that accomplished by ACTH on the adrenal gland 31. Furthermore, the splanchnic nerve is capable of increasing the cortisol levels. In a well-designed study on calves, Edwards and Jones 32 found roughly the same pattern and duration of the cortisol release when they stimulated the splanchnic nerve as we found during AI and mating, probably due to a CRH-like peptide released from the adrenal medulla. These authors have also shown that the production of cortisol during intravenous infusion of ACTH could be further increased after stimulation of the splanchnic nerve 33. Van Oers et al. 34 demonstrated that administrated CRH most likely increase the adrenal cortex sensitivity to ACTH due to a higher blood flow through the adrenal gland, and these authors hypothesized that intra adrenal CRH is controlled by the central nervous system.

It might be of importance to look upon our results with some knowledge about corticosteroid-binding globulin (CBG) and the relation between the total and free cortisol. Both total and free corticosteroid plasma levels were increased after mating in gilts, with significant changes in the maximum corticosteroid binding capacity 5. Nyberg and Madej 35 reported that an anovulatory oestrus in sows seemed to be associated with high CBG binding capacity and high cortisol level in blood plasma, which might result in pronounced reduction of circulating free cortisol. Furthermore, studies of Klemcke et al. 36 demonstrated the occurrence and changes of CBG-like activity and cortisol content in porcine uterine lumen during the oestrous cycle. Endometrial CBG-like activity and free cortisol could indicate their biological role on uterine function in pigs.

At the present time, we have no obvious explanation for the biological effect of cortisol during the very short time for mating or insemination. Because of the well-known anti-inflammatory effects of glucocorticoids 37, it is plausible that the significant elevation of cortisol concentrations in mated sows may have an impact on the inhibition of prostaglandin synthesis. Cortisol metabolism and differential regulation can occur within different peripheral target tissues 38 and glucocorticoids are known to be involved in female reproductive function 39. In a study on bovine endometrium by Lee et al. 39, it is concluded that cortisol inhibits both TNFα-stimulated PGE₂ and PGF_2α as well as basal PGF_2α production in stromal cells. Interestingly, cortisol does not affect PGE₂ production in stromal cells, nor OT-stimulated PG production in epithelial cells. Since the endometrium consists of many more stromal than epithelial cells, cortisol-inhibited PGF_2α production could be of physiological relevance. With this in mind it could be of importance to look at 11β-hydroxysteroid dehydrogenase (11-HSD), an enzyme with two isoforms that facilitate a conversion between the active cortisol and inactive cortisone 38. At least in heifers the isoform 11-HSD1, converts cortisone to cortisol, and has greater activity during the late follicular stage and oestrus 39. Since our results indicate a hormonal difference between mating and insemination, either intrauterinally or conventionally, the reproductive outcome could be affected by the service technique, as it was shown by Tummaruk et al. 40 with data from almost 7000 sows that naturally mated sows had 0.3 more piglets born alive per litter than AI sows.

We have shown that mating gives a greater increase of cortisol than AI without the amplified PG-metabolite plasma level seen in both the inseminated groups. It was also demonstrated that IUI-group had an earlier significant increase of PG-metabolite concentrations than sows inseminated conventionally. Further investigation using different semen extenders or even different type of insemination catheters might be helpful in understanding the reason for an immediate increase of PG-metabolite after insemination but not after mating.

The author(s) declare that they have no competing interests.

MN was responsible for hormones and data analyses, and manuscript preparation. MTM was responsible for the study design and collection of samples. CBA participated in the study design and collection of samples. HK contributed with expertise in prostaglandins. AM was involved in the study design and responsible for revision of the manuscript. All authors read and approved the final manuscript.