The Effect of Soybean-Derived Phytoestrogens on Concentrations of Plasma Isoflavones , 15-keto-13 , 14-dihydroprostaglandin F 2 α and Progesterone in Dairy Cows

The objective of the study was to determine the effect of soybean-derived phytoestrogens and their metabolites on the activity of sex hormones during the oestrous cycle in multiparous lactating dairy cows. The experiment was carried out on 4 multiparous lactating Holstein cows in the form of replicated Latin square in double reversal design. The experiment in the total length of 168 days was divided into 4 periods of 42 days, each consisting of a 21-day preliminary period and a 21-day collecting period. Cows were divided into 2 groups of 2 cows. The control group (C) was fed a diet based on extruded rapeseed cake while the experimental group (S) was fed a diet containing extruded full-fat soya. The intake of total isoflavones was 3297 mg/d in S and 58.0 mg/d in C (P < 0.001). The concentrations of individual isoflavones, it is daidzein, genistein and equol in plasma were significantly higher in the experimental group S (49.3, 78.7 and 218.8 ng/ml, respectively) than in the control group C (13.5, 42.9 and 18.3 ng/ml, respectively, P < 0.001). Plasma concentration of progesterone throughout the oestrous cycle was not influenced by the diet used (P > 0.05). Plasma concentration of prostaglandine PGFM throughout the oestrous cycle in the experimental group (S) tended to be higher (P = 0.095) than in the control group (C). No differences in the length of the oestrous cycle between the cows fed different diets were observed. Isoflavones, prostaglandins, bovine, oestrous cycle Soybeans or soybean products are used in diets of high yielding dairy cows, especially during early lactation, as an excelent source of energy and high-quality protein (Chouinard et al. 1997). Nevertheless, soybeans are known as the major dietary source of phytoestrogens (Bingham et al. 1998). Phytoestrogens may induce various pathologies in the female reproductive tract, such as disorders in the ovarian and uterine function (Rosselli et al. 2000) resulting in disruption of the reproductive processes, decreased fertility or even temporal infertility in cows (Woclawek-Potocka et al. 2005b). Recently, several studies have been conducted to evaluate the effect of soybean phytoestrogens on the bovine reproductive tract, oestrous cycle and hormonal profile of fertile cows (e. g. WoclawekPotocka et al. 2005ab, 2006, 2008; Piotrowska et al. 2006). In ruminants, endogenous estrogens are known to control the oestrous cycle by influencing prostaglandin synthesis (Goff 2004). Furthermore, in bovines, prostaglandine PGF2α is the major luteolytic agent (McCracken et al. 1999), whereas prostaglandine PGE2 has luteoprotective and antiluteolytic properties (Kennedy 1983; Asselin et al. 1996). Optimal PGF2α to PGE2 ratio is essential for endometrial receptivity, maintenance of corpus luteum (CL), and progesterone (P4) secretion (Milvae et al. 1996). Although it has been recently shown that soy-bean derived phytoestrogens regulate both PGF2α and PGE2 secretion (Woclawek-Potocka et al. 2005b), in the subsequent study of WoclawekPotocka et al. (2005a) it has been demonstrated that phytoestrogens preferentially ACTA VET. BRNO 2010, 79: 525-532; doi:10.2754/avb201079040525 Address for correspondence: Ing. Jarmila Watzková Agriresearch Rapotin Ltd. Department of Animal Nutrition and Quality of Livestock Products Vídeňská 699, 691 23 Pohořelice, Czech Republic Phone: + 420 519 426 002 E-mail: Jarmilaa@email.cz http://www.vfu.cz/acta-vet/actavet.htm stimulated PGF2α synthesis in the epithelial cells of bovine endometrium. Based on the stronger phytoestrogen-dependent stimulation of PGF2α compared to PGE2 production in epithelial cells, it appears that phytoestrogens and their metabolites mainly modulate the PGF2α to PGE2 ratio that cause premature luteolysis. The objective of the study was to determine the effect of soybean-derived phytoestrogens and their metabolites on the activity of sex hormones during the oestrous cycle in multiparous lactating dairy cows. Materials and Methods Animals and diets The experiment was carried out on 4 high-yielding lactating Holstein cows (lactation 3–4, week 16–46 of lactation) with similar milk production (27.3 kg + 1.7) that were divided into 2 groups according to milk yield. The control group of animals was fed a diet based on extruded rapeseed cake (C); the experimental group was fed a diet based on extruded full-fat soya (S). The experiment was carried out in the form of replicated Latin square in double reversal design (Tempelman 2004). The trial in the total length of 168 days was divided into 4 periods of 42 days. Each period consisted of a preliminary period (21 d) and a collecting period (21 d). Before the beginning of the experiment, the oestrus was synchronized by application of PGF2α (Oestrophan, Bioveta a. s., CR). The day of oestrus was considered to be day 0 of the collecting period. The length of oestrous cycle was determined on the basis of signs of the next oestrus and by the veterinarian via rectal examination. The animals were under regular veterinary rectal examination during the whole experiment. Sampling and analysis Cows were fed individually twice daily (6.30 and 16.30 h) the diet based on maize silage, meadow hay and supplemental mixture (Table 1). Samples of feed and feed refusals were taken twice a week in the collecting periods and used for the determination of the content of basal nutrients and phytoestrogens (daidzein, genistein). Dry matter (DM) was determined by drying at 103 oC for 4 h. The content of PDIN, PDIE (digestible protein in the intestine when rumen fermentable N supply or energy supply are limited) and NEL (net energy of lactation) was calculated according to Sommer (1994). The contents of daidzein and genisten in feed and feed refusals were analysed using HPLC-DAD as given in details below. Blood samples were taken into heparinised tubes from vena jugularis after morning milking (8:00 h) during the collecting period three times a week. The blood plasma was separated by centrifuge (1500 × g for 15 min, 4 °C) and stored at -20 °C until analyses. For determination of prostaglandine PGFM (15-keto-13,14-dihydroprostaglandin F2α) in blood plasma, prostaglandine stabiliser was added to each tube with blood and mixed thoroughly prior centrifugation. The content of phytoestrogens and their metabolites was analysed using HPLC-MS/MS. The level of P4 in blood plasma was measured by imuno-chemiluminescence method in automatic analyser IMMULITE (DPC). The PGFM concentration in blood plasma was determined with an EIA using HRP-labeled PGFM and anti-PGFM serum (Skarzynski et al. 2003). Determination of phytoestrogens and their metabolites

Soybeans or soybean products are used in diets of high yielding dairy cows, especially during early lactation, as an excelent source of energy and high-quality protein (Chouinard et al. 1997).Nevertheless, soybeans are known as the major dietary source of phytoestrogens (Bingham et al. 1998).Phytoestrogens may induce various pathologies in the female reproductive tract, such as disorders in the ovarian and uterine function (Rosselli et al. 2000) resulting in disruption of the reproductive processes, decreased fertility or even temporal infertility in cows (Woclawek-Potocka et al. 2005b).Recently, several studies have been conducted to evaluate the effect of soybean phytoestrogens on the bovine reproductive tract, oestrous cycle and hormonal profile of fertile cows (e. g.Woclawek-Potocka et al. 2005ab, 2006, 2008;Piotrowska et al. 2006).
In ruminants, endogenous estrogens are known to control the oestrous cycle by influencing prostaglandin synthesis (Goff 2004).Furthermore, in bovines, prostaglandine PGF 2α is the major luteolytic agent (McCracken et al. 1999), whereas prostaglandine PGE 2 has luteoprotective and antiluteolytic properties (Kennedy 1983;Asselin et al. 1996).Optimal PGF 2α to PGE 2 ratio is essential for endometrial receptivity, maintenance of corpus luteum (CL), and progesterone (P 4 ) secretion (Milvae et al. 1996).Although it has been recently shown that soy-bean derived phytoestrogens regulate both PGF 2α and PGE 2 secretion (Woclawek-Potocka et al. 2005b), in the subsequent study of Woclawek-Potocka et al. (2005a) it has been demonstrated that phytoestrogens preferentially stimulated PGF 2α synthesis in the epithelial cells of bovine endometrium.Based on the stronger phytoestrogen-dependent stimulation of PGF 2α compared to PGE 2 production in epithelial cells, it appears that phytoestrogens and their metabolites mainly modulate the PGF 2α to PGE 2 ratio that cause premature luteolysis.
The objective of the study was to determine the effect of soybean-derived phytoestrogens and their metabolites on the activity of sex hormones during the oestrous cycle in multiparous lactating dairy cows.

Animals and diets
The experiment was carried out on 4 high-yielding lactating Holstein cows (lactation 3-4, week 16-46 of lactation) with similar milk production (27.3 kg + 1.7) that were divided into 2 groups according to milk yield.The control group of animals was fed a diet based on extruded rapeseed cake (C); the experimental group was fed a diet based on extruded full-fat soya (S).The experiment was carried out in the form of replicated Latin square in double reversal design (Tempelman 2004).The trial in the total length of 168 days was divided into 4 periods of 42 days.Each period consisted of a preliminary period (21 d) and a collecting period (21 d).Before the beginning of the experiment, the oestrus was synchronized by application of PGF 2α (Oestrophan, Bioveta a. s., CR).The day of oestrus was considered to be day 0 of the collecting period.The length of oestrous cycle was determined on the basis of signs of the next oestrus and by the veterinarian via rectal examination.The animals were under regular veterinary rectal examination during the whole experiment.

Sampling and analysis
Cows were fed individually twice daily (6.30 and 16.30 h) the diet based on maize silage, meadow hay and supplemental mixture (Table 1).Samples of feed and feed refusals were taken twice a week in the collecting periods and used for the determination of the content of basal nutrients and phytoestrogens (daidzein, genistein).Dry matter (DM) was determined by drying at 103 ºC for 4 h.The content of PDIN, PDIE (digestible protein in the intestine when rumen fermentable N supply or energy supply are limited) and NEL (net energy of lactation) was calculated according to Sommer (1994).The contents of daidzein and genisten in feed and feed refusals were analysed using HPLC-DAD as given in details below.
Blood samples were taken into heparinised tubes from vena jugularis after morning milking (8:00 h) during the collecting period three times a week.The blood plasma was separated by centrifuge (1500 × g for 15 min, 4 °C) and stored at -20 °C until analyses.For determination of prostaglandine PGFM (15-keto-13,14-dihydroprostaglandin F 2α ) in blood plasma, prostaglandine stabiliser was added to each tube with blood and mixed thoroughly prior centrifugation.The content of phytoestrogens and their metabolites was analysed using HPLC-MS/MS.The level of P 4 in blood plasma was measured by imuno-chemiluminescence method in automatic analyser IMMULITE (DPC).
The PGFM concentration in blood plasma was determined with an EIA using HRP-labeled PGFM and anti-PGFM serum (Skarzynski et al. 2003).
Determination of phytoestrogens and their metabolites Contents of target compounds were determined after their releasing from bonded forms.Hydrolysis of glycosides was carried out by acidic hydrolysis from feed samples and by enzymatic hydrolysis from biotic samples.HPLC/DAD method was used for determination of phytoestrogens in plant materials and LC-MS/MS method was employed for determination of estrogenic compounds in biotic samples.
Feed samples Homogenised samples of feed were hydrolysed with 6 mol/l hydrochloric acid and ethanol under the reverse condenser at the boiling point of ethanol.After hydrolysis the extract was clean up by SPE procedure on Oasis HLB, Waters (UK) cartridges.
The HPLC analysis was carried out on an HP 1200 liquid chromatograph coupled with a diode array detector (DAD) (Hewlett Packard, USA).
The limit of detection for total isoflavones obtained under the described method was 3.5 mg/kg for daidzein and 2.8 mg/kg for genistein.The repeatability expressed as relative standard deviation (n = 6) was 4.1% and 4.9%, respectively.

Plasma samples
Target analytes were hydrolysed from possible conjugates by enzymatic hydrolysis with Helix pomatia enzyme β-glucuronidase/sulfatase in sodium acetate buffer at 37 °C.After hydrolysis the analytes were extracted by ethylacetate.
Liquid chromatograph HP 1100, (Hewlett Packard, USA) coupled with mass spectrometry detector -ion trap, Finnigan LCQ Deca, (Finnigan, USA) operated in selected reaction monitoring (SRM) mode was used for analysis.
The limit of detection obtained under the described method was 2.5 ng/ml for daidzein and equol, and 5 ng/ml for genistein.The repeatability expressed as relative standard deviation (n = 6) was 5.0% for daidzein, 6.8% for genistein and 3.5% for equol in plasma samples.

Statistical analyses
Data obtained in the experiment were analysed using GLM procedure of the SAS/STAT, Version 8 according to the following model: Y ijk = µ + T i + C j + D k + (T x C) ij + ε ijk , where µ= general mean, T i = treatment effect (i = 2), C j = cow effect (j = 4), D k = day (replication) effect (k = 8), (T x C) ij = interaction effect, ε ijk = residual error.Results are expressed as means with standard error of the mean (SE)

Results
The nutrient intake is presented in Table 2.The intake of DM, PDIN, PDIE and NEL did not differ significantly between groups (P > 0.05).Average concentration of total isoflavones (aglycones and glycones) daidzein and genistein determined in extruded fullfat soya was 150.9 mg/kg and 222.6 mg/kg, respectively, resulting in the average total isoflavones intake of 3297 mg/d in S. The concentration of total daidzein and genistein in extruded rapeseed cake was under the limit of detection of used analytical method.Low intake 58.0 mg/d of total isoflavones in C was found.
Concentrations of isoflavones and studied hormones in plasma are given in Table 3.The concentrations of daidzein, genistein and equol in plasma were significantly higher in S than in C (P < 0.001).Plasma concentration of prostaglandine PGFM throughout the experiment in S tended to be higher (P = 0.095) than that in C. The concentration of P 4 in the plasma in S did not differ significantly throughout the experiment in comparison to C (P > 0.05).
As shown in Fig. 1 higher plasma concentrations of daidzein were found in S (P < 0.001) compared to C. Similar tendencies between S and C were observed in concentration of genistein (Fig. 2).High plasma concentrations of equol were found in S (P < 0.001) compared to C with almost undetectable levels of equol (Fig. 3).The p-ethyl phenol in the plasma was not detected.The plasma concentrations of PGFM in S varied with the days of oestrous cycle with the highest concentration on Day 15 (0.349 ± 0.259 ng/ ml, Fig. 4).The plasma concentration of PGFM in C varied only little throughout the experiment.As shown in Fig. 5

Discussion
Extruded full-fat soya used in this experiment contained 150.9 mg/kg of daidzein and 222.6 mg/kg of genistein.Thus, average total isoflavones intake in S was 3297 mg/d.Although the concentration of isoflavones in extruded rapeseed cake was under the limit of 528  In the present experiment, cows fed S diet had significantly higher concentrations of daidzein, genistein and equol in plasma in comparison to C (P < 0.001), and p-ethylphenol was not detected.On the other hand, Woclawek-Potocka et al. (2005b) or Piotrowska et al. (2006) found significant concentrations of p-ethyl-phenol and equol in the plasma of cows fed soybeans but they detected neither daidzein nor genistein.In contrast to our findings, in the above mentioned studies no phytoestrogens or their metabolites in plasma of cows fed the control diet were detected.This discrepancy can be caused by differences in dynamics of isoflavones and their metabolites in the plasma after administration of soybeans.As described in the study of Woclawek-Potocka et al. (2008), daidzein concentration in plasma of cycling heifers increased after single dose of soybeans during 0.5 h (P < 0.05) and then remained constant until 2 h post feeding.Three hours after soybean feeding the daidzein concentration started to decrease to same levels as at the beginning of the experiment.Genistein concentration in blood plasma increased during 0.5 h after soybean feeding (P < 0.05) and then remained constant.Similarly, the equol concentration in the blood plasma was increasing up to 4 h after soybean administration (P < 0.05) and then remained at an approximately constant level until the end of the experiment.On the other hand, the p-ethyl phenol concentration in the blood plasma remained constant during first 3 h of the experiment (P > 0.05) and started to increase (P < 0.05), beginning at 4 h after feeding.Blood samples for determination of isoflavones and their metabolites in our experiment were taken approximately 1 h after feeding, thus our findings are in accordance with the above-mentioned study.
Endogenous estrogens (E2) are known to modulate the oestrous cycle in ruminants by influencing the synthesis of prostaglandins (Goff 2004), as demonstrated e.g. in the study of Zhang et al. (1991).E2 has been proved to stimulate cell viability and proliferation in the female reproductive tract of many species (Reynolds and Redmer 1998).Due to the fact that phytoestrogens are structurally similar to E2 (Middleton and Kandaswami 1986) and act as agonists or antagonists of endogenous E2 (Rosselli et al. 2000, Dubey et al. 2000), it can be assumed that they influence the same reproductive processes regulated by E2.As already mentioned, in bovines, PGF 2α is the major luteolytic agent (McCracken et al. 1999) whereas PGE 2 exerts luteoprotective properties (Asselin et al. 1996).Thus, the development and maintenance of the CL whether during the oestrous cycle or for establishment of pregnancy depends on the optimum PGF 2α : PGE 2 ratio (Milvae et al. 1996;Arosh et al. 2004).In the endometrium, the PGF 2α is synthetised mainly by the epithelial cells and PGE 2 by the stromal cells (Skarzynski et al. 2000;Asselin et al. 1996).In their recent in vivo and in vitro studies, Woclawek-Potocka et al. (2005a,b) have shown that soy-bean derived phytoestrogens regulate both PGE 2 and PGF 2α secretion in endometrium of cycling and early pregnant heifers or cows with preferential increases in PGF 2α secretion (Woclawek-Potocka 2005a,c).Furthermore, equol and p-ethyl-phenol, via the preferential decrease of the viability of stromal cells (Woclawek-Potocka 2005c), can negatively influence the production of PGE 2 and may cause additional disturbances in prostaglandins synthesis and disrupt the physiologic ratio between PGE 2 and PGF 2α .
Due to the fact that PGF 2α is rapidly metabolised in the organism, a metabolite of 15-keto-13,14-dihydroprostaglandin F 2α (PGFM) was determined in blood plasma of cows in our experiment as described by Skarzynski et al. (2003).Plasma concentration of PGFM throughout the experiment in S tended to be higher (P = 0.095) than in C.This is in disagreement with Woclawek-Potocka et al. (2005b) who found significantly higher (P < 0.01) plasma concentrations of PGFM in soybean-fed heifers than in control animals.This discrepancy can be caused by a higher variability in PGFM concentrations determined in their study with the highest level reaching up to 1.6 ng/ml while the highest concentration of PGFM in our experiment with multiparous cows was 0.349 ng/ml.The plasma concentrations of PGFM in C varied only little throughout the experiment and were similar to those determined by Woclawek-Potocka et al. (2005b).
The concentration of P 4 in the plasma in S did not differ significantly throughout the experiment in comparison to C (P > 0.05).This is in agreement with Piotrowska et al. (2006) or Woclawek-Potocka et al. (2005b).Similarly to our findings, in both of the mentioned studies the P 4 concentrations increased from Day 0 to 12-15 and then started to decline.In the present study, no significant difference was detected between P 4 concentrations in S compared to C in dependence on days of oestrous cycle.On the other hand, Piotrowska et al. (2006) found that the P 4 concentration in soybean-fed animals on Day 15 and 18 of oestrous cycle was significantly lower than in standard diet-fed heifers.
In our experiment, the length of the oestrous cycle was not affected by the treatment (P > 0.05).This is in agreement with Piotrowska et al. (2006).On the other hand, the average length of the oestrous cycle in our experiment was shorter than that reported in the abovementioned study.This shortage was probably caused by regular per rectum examinations that were used to examine the ovaries.
The length of the oestrous cycle in cattle is commonly 18-24 days with 21 days considered the average (Rioux and Rajotte 2004).Nevertheless, the intensity and duration of oestrous behaviours during the oestrous cycle is highly variable among individuals.Social interactions, housing and management factors, the physical environment impinging on the animal, nutritional status age and physiological state, genetic factors and presence of the bull can influence oestrus in cattle (Orihuela 2000).The oestrous cycle length in the cows in this experiment was probably influenced by the frequent interventions of a veterinarian (three times a week in the collecting period).Frequent rectal examinations could cause congestion of the endometrium and consequent shortening of the oestrous cycle in the cows.
In conclusion, the effect of soybean-derived phytoestrogens and their metabolites on the activity of sex hormones during the oestrous cycle was studied on multiparous (lactation 3-4) lactating dairy cows.Inclusion of extruded full-fat soya to their diet resulted in a significant increase in daidzein, genistein and, in particular, equol concentrations blood plasma.Although these compounds tended to increase the plasma concentration of PGFM throughout the oestrous cycle (P = 0.095), plasma concentration of P 4 throughout the cycle was not influenced by the treatment.Further, no differences in the length of the oestrous cycle were observed.These data suggest that if cows are continuously exposed to a diet that include soya, phytoestrogen active metabolites may act chronically and locally on the reproductive tract.

Fig. 1 .
Fig. 1.Concentration of daidzein in the peripheral blood plasma of high-yielding lactating Holstein cows fed extruded full-fat soya diet (S) and extruded rapeseedcake diet (C) during the oestrous cycle Fig.2. Concentration of genistein in the peripheral blood plasma of high-yielding lactating Holstein cows fed extruded full-fat soya diet (S) and extruded rapeseed-cake diet (C) during the oestrous cycle

Fig. 3 .
Fig. 3. Concentration of equol in the peripheral blood plasma of high-yielding lactating Holstein cows fed extruded full-fat soya diet (S) and extruded rapeseedcake diet (C) during the oestrous cycle

Table 1 .
Composition of diet in g/kg of dry matter