DISCUSSION

In the current experiment, a significant increase of total cholesterol in blood serum was observed in both experimental groups of animals, which were administered S.officinalis root powder in their diet, when compared with the control group. The current results contradicted some results obtained in non-ruminants (Shim Reference Shim2010; Wu et al. Reference Wu, Leung, Kwan, Sham, Tang, Wang, Wan, Lee and Chan2013; Weng et al. Reference Weng, Yu, Cui, Zhu, Guo, Wei, Duan, Yin, Guan, Wang, Yang, Xi and Wen2014); however others confirm a lack of saponin effects on cholesterol concentration (Calvert et al. Reference Calvert, Blight, Illman, Topping and Potter1981) or even increased cholesterol synthesis after dietary saponin administration (Jenkins & Atwal Reference Jenkins and Atwal1994). Many studies on monogastric animals have revealed that saponins decreased total blood cholesterol and lowered LDL-cholesterol (Shim Reference Shim2010; Wu et al. Reference Wu, Leung, Kwan, Sham, Tang, Wang, Wan, Lee and Chan2013; Weng et al. Reference Weng, Yu, Cui, Zhu, Guo, Wei, Duan, Yin, Guan, Wang, Yang, Xi and Wen2014). In those studies, the metabolism of cholesterol after saponin treatment was augmented. One of the reasons for this effect was the formation of large mixed micelles by saponins with bile acids (Oakenfull & Sidhu Reference Oakenfull and Sidhu1990), which were excreted, so serum cholesterol concentration was decreased (Oakenfull & Sidhu Reference Oakenfull and Sidhu1990). Moreover, in the intestine, saponins could interact with cholesterol and decrease intestinal cholesterol absorption with no change in bile absorption (Harwood et al. Reference Harwood, Chandler, Pellarin, Bangerter, Wilkins, Long, Cosgrove, Malinow, Marzetta, Pettini, Savoy and Mayne1993). However, these kinds of interactions of saponins are rather implausible in ruminants, because of rumen metabolism. Several studies have demonstrated that saponins are transformed by microbes in the rumen (Gutierrez & Davis Reference Gutierrez and Davis1962; Meagher et al. Reference Meagher, Smith and Wilkins2001; Flåøyen et al. Reference Flåøyen, Wilkins and Sandvik2002) and undergo hydrolysis, epimerization and hydrogenation (Gutierrez & Davis Reference Gutierrez and Davis1962; Meagher et al. Reference Meagher, Smith and Wilkins2001). The derivatives of saponin degradation in the rumen, sapogenins, can manifest other properties concerning interaction with cholesterol. Lorent et al. (Reference Lorent, Le Duff, Quetin-Leclercq and Mingeot-Leclercq2013) studied the interactions of triterpenoid monodesmosidic saponins (α-hederin, a derivative that lacks one sugar, δ-hederin and another derivative, an aglycone form, i.e. hederagenin with cholesterol) and their effect on membrane permeabilization. They found that these interactions are highly dependent on sugars linked to the saponin aglycone form (Lorent et al. Reference Lorent, Le Duff, Quetin-Leclercq and Mingeot-Leclercq2013) and α-hederin was the most effective, whereas hederagenin manifested the smallest activity (Lorent et al. Reference Lorent, Le Duff, Quetin-Leclercq and Mingeot-Leclercq2013). Thus, in the current experiment, the effect of saponins on serum cholesterol concentration, different to that observed in non-ruminants, could be partly the result of their rumen modification and increased sapogenin production. After post-ruminal absorption, the interaction of sapogenins with cholesterol is less active and therefore their influence on blood concentration of LDL-cholesterol could be weaker. The concentration of LDL-cholesterol concentration was observed to be unchanged after saponin administration. However, total cholesterol was increased, probably as a result of augmented HDL-cholesterol, since LDL-cholesterol remained unchanged. It has been reported that elevated blood cholesterol in cattle is related to increased concentration of HDL-cholesterol (Grummer & Carroll Reference Grummer and Carroll1988; Hocquette & Bauchart Reference Hocquette and Bauchart1999). A probable reason for elevated HDL-cholesterol after saponin treatment in the current experiment will be discussed below.

There are many other factors modifying blood cholesterol concentration. Thyroid hormones are responsible for cholesterol metabolism and limit its concentration in blood (Ness Reference Ness1991). Insufficient level of thyroid hormones in blood leads to hypercholesterolaemia (Roberts & Ladenson Reference Roberts and Ladenson2004; Duntas & Wartofsky Reference Duntas and Wartofsky2007). In the current experiment, the increased blood cholesterol was independent of thyroid hormones, because their concentrations after saponin treatment were increased.

As for the other active compounds, fatty acids, depending on the number of saturated bonds and isomeric form, can improve or impair cholesterol metabolism due to its concentration in blood (Fernandez & West Reference Fernandez and West2005; Ascherio Reference Ascherio2006; Aronis et al. Reference Aronis, Khan and Mantzoros2012). Unsaturated fatty acids improve both cholesterol metabolism and its elimination from blood (Fernandez & West Reference Fernandez and West2005). Although the concentrations of FFA were unaltered in the current experiment, their profile – proportions and composition – varied significantly, suggesting an important influence of saponins on fatty acid synthesis/distribution in the organism. However, other than volatile fatty acid (VFA) modification in the rumen (Wina et al. Reference Wina, Muetzel and Becker2005), this effect has not so far been studied in animals. The only report concerns the effect of Q. saponaria L. saponin on meat fatty acid composition in lambs – an increase in C20 : 4n-6 amount was observed, as well as reduced ratio between α-linolenic and linoleic acids (Brogna et al. Reference Brogna, Nasri, Salem, Mele, Serra, Bella, Priolo, Makkar and Vasta2011). It was established in in vivo (Wallace et al. Reference Wallace, Arthaud and Newbold1994) and in vitro (Makkar et al. Reference Makkar, Sen, Blümmel and Becker1998) experiments that saponins can modify the balance of rumen microorganisms through their antiprotozoal activity and due to this fact an increase of rumen bacteria was observed. These microbes participate in dietary fatty acid metabolism. Considering the above data, it can be assumed that altered concentrations of fatty acids in the current experiment could be the result of changes in rumen microflora after saponin action.

Unsaturated fatty acids improved both cholesterol metabolism and its elimination from the blood (Fernandez & West Reference Fernandez and West2005). Dietary saturated fatty acids (especially C16 : 0), as well as trans isomers of unsaturated fatty acids, evoked deleterious effects on cholesterol metabolism (Fernandez & West Reference Fernandez and West2005; Brouwer et al. Reference Brouwer, Wanders and Katan2010; Ooi et al. Reference Ooi, Ng, Watts and Barrett2013), increasing concentrations of LDL-cholesterol and lowering HDL-cholesterol. Moreover, there is no doubt that in monogastric animals and humans, dietary polyunsaturated fatty acids (PUFA), especially n-3 but also n-6, favour the metabolism of cholesterol augmenting HDL-cholesterol and reducing LDL-cholesterol blood concentrations (Fernandez & West Reference Fernandez and West2005). In cattle, dietary PUFA undergo biohydrogenation in the rumen (Demeyer & Doreau Reference Demeyer and Doreau1999), however, this process can be limited by various factors (Petersen & Jensen Reference Petersen and Jensen2014). Therefore, these fatty acids can be absorbed from the rumen into blood and then be present in milk. Taking into account the data above, the elevated blood concentration of n-6 - C20 : 4n-6 observed in the current experiment after saponin administration at the higher dose could be one of the reasons for elevated HDL-cholesterol concentration in blood. Hocquette & Bauchart (Reference Hocquette and Bauchart1999) found that diets enriched in PUFA can stimulate synthesis of HDL and may contribute to higher blood cholesterol concentration. This effect may be considered as protective due to prevention of atherogenesis. Other parameters of significant health importance evaluated in the current experiment – atherogenic or thrombogenic indices – were unchanged, suggesting rather harmless saponin action.

The substantial elevation of serum T3 and numerically increased T4 were observed after saponin administration. So far, there is no evidence confirming any possible effect of saponins on the thyroid gland and thyroid hormone secretion in dairy cows. However, as was mentioned above, numerous reports from in vivo and in vitro studies have indicated that saponins, through inactivation of protozoa in rumen, contribute to increases in the bacterial population (Makkar et al. Reference Makkar, Sen, Blümmel and Becker1998). Therefore, an increase of microbial protein synthesis is observed (Makkar & Becker Reference Makkar, Becker, Waller and Yamasaki1996; Wang et al. Reference Wang, McAllister, Newbold, Cheeke and Cheng1997; Patra & Yu Reference Patra and Yu2014), as well as bacterial nitrogen flow to the duodenum. Thus, the elevation of serum thyroid hormone concentration observed in the current experiment may result from improved protein turnover after saponins’ action. It was reported by Hammond et al. (Reference Hammond, Huntington and Bitman1984) that an increase in T4 concentration in growing steers was related to an increased protein intake. Moreover, Gerrits et al. (Reference Gerrits, Decuypere, Verstegen and Karabinas1998) found elevated T3 and T4 in blood after increased protein intake in calves. According to Gerrits et al. (Reference Gerrits, Decuypere, Verstegen and Karabinas1998) the augmented T3 concentration resulted from the increased T4 concentration (Gerrits et al. Reference Gerrits, Decuypere, Verstegen and Karabinas1998). There is also a report about a stimulatory effect of soyasaponins on thyroid hormone concentration in mice (Ishii & Tanizawa Reference Ishii and Tanizawa2006).

Besides thyroid hormones, glucagon plays a pivotal role in lactating ruminants (Danfaer et al. Reference Danfaer, Tetens and Agergaard1995). Glucagon concentration in portal blood can be increased as a consequence of elevated amino acids or propionate, and their availability seems to be an important factor limiting the process (Danfaer et al. Reference Danfaer, Tetens and Agergaard1995). In the present experiment, saponins administered at a higher dose evoked a substantial rise in glucagon concentration. The glucagon increase after saponin treatment could be related to their action in the rumen. It is commonly known that the major effect of saponins on rumen fermentation is an increased production of propionate and decreased acetate to propionate ratio (Wina et al. Reference Wina, Muetzel and Becker2005). Moreover, Budan et al. (Reference Budan, Bellenot, Freuze, Gillmann, Chicoteau, Richomme and Guilet2014) recently observed in an in vitro experiment that S. officinalis root extract increased total VFA concentration in the rumen and markedly decreased acetate to propionate ratio. In an earlier experiment, Cieslak et al. (Reference Cieslak, Zmora, Stochmal, Pecio, Oleszek, Pers-Kamczyc, Szczechowiak, Nowak and Szumacher-Strabel2014) obtained similar results using S. officinalis root powder in in vitro ruminal fermentation. Although in the present experiment the exact content of VFA in serum was not analysed, it can be supposed that elevated propionate production in rumen could be one possibility for the increased glucagon concentration.

Unlike glucagon, blood insulin concentration was only slightly increased after feeding with S. officinalis root powder. There are no other data concerning the effect of saponins on insulin concentration in ruminants; however, rodent studies revealed that saponins augmented insulin secretion (Lee et al. Reference Lee, Kao, Liu and Cheng2006; Park et al. Reference Park, Ha and Chung2008) and showed antidiabetic properties, decreasing disaccharidase activities in the small intestine (Lee et al. Reference Lee, Choi, Seo, Lee, Lee, Lee, Kim and Lee2014), stimulating insulin signalling (Kwon et al. Reference Kwon, Kim, Hong and Park2009) and lowering glucose concentration (Singh et al. Reference Singh, Farswan, Ali, Afzal, Al-Abbasi, Kazmi and Anwar2014; Weng et al. Reference Weng, Yu, Cui, Zhu, Guo, Wei, Duan, Yin, Guan, Wang, Yang, Xi and Wen2014).

Saponins are known to evoke some deleterious effects on blood cells. In many studies performed on rodents and humans, it was established that saponins exhibited a lytic action on erythrocyte membranes, probably connected to their affinity for membrane sterols (Bomford Reference Bomford1980). For human erythrocytes, triterpenoid saponins appeared to be highly haemolytic (Sindambiwe et al. Reference Sindambiwe, Calomme, Geerts, Pieters, Vlietinck and Vanden Berghe1998). In vitro ruminant studies also proved their haemolytic properties (Voutquenne et al. Reference Voutquenne, Guinot, Froissard, Thoison, Litaudon and Lavaud2005; Budan et al. Reference Budan, Tessier, Saunier, Gillmann, Hamelin, Chicoteau, Richomme and Guilet2013); however, in vivo ruminant studies did not confirm the detrimental effects of saponins on blood cell parameters (Hess et al. Reference Hess, Beuret, Lötscher, Hindrichsen, Machmüller, Carulla, Lascano and Kreuzer2004; Sliwinski et al. Reference Sliwinski, Kreuzer, Sutter, Machmuller and Wettstein2004). In dairy cows, no effects of Yucca saponin were observed on haematocrit readings or on haemoglobin concentration (Sliwinski et al. Reference Sliwinski, Kreuzer, Sutter, Machmuller and Wettstein2004). The lack of haemolytic effect was also noticed in sheep fed with S. saponaria (Hess et al. Reference Hess, Beuret, Lötscher, Hindrichsen, Machmüller, Carulla, Lascano and Kreuzer2004). In the current experiment, saponins exerted a different effect; the red blood cell number in dairy cows was significantly elevated. It is difficult to explain this result, particularly since literature data concerning this problem in ruminants are very scarce. Further experiments are needed to explain this phenomenon.