**4. Discussion**

Some plants containing saponins, polysaccharides and flavonoids have shown positive effects on antioxidant capacity and immune status in ruminants [8,41]. In addition, saponins have been reported to improve the energy utilization efficiency and increase the duodenal flux of amino acids and microbial protein [11]. Consequently, lambs supplemented with herbal products containing saponins, polysaccharides, and flavonoids would be expected to have higher growth rates. However, although in our study FBW and DWG were not affected by HM, a linear reduction trend was observed in DWG of lambs fed the HM3 diet, which could be a consequence of the lower dry matter digestibility observed with HM3. This suggests that high doses of HM in the diet could affect the growth rate of lambs when it is used for prolonged periods. Similar results were previously reported by Liu et al. [42] in lambs supplemented with *Medicago sativa* saponin extracts (0, 0.5, 1, 2 and 4 g kg−<sup>1</sup> DM for 90 days); and by Nasri et al. [43] who examined the effects of increasing doses of *Quillaja saponaria* saponin extracts (0, 30, 60 and 90 mg kg−<sup>1</sup> DM for 57 days) in lambs fed high concentrate diets. In the latter investigation, BW and DWG was similar among treatments, regardless of the dose of saponins used. In another study, Wang et al. [9] investigated the effects of supplementing lambs with *Astragalus membranaceus* roots (0, 20, 50 and 80 g kg−<sup>1</sup> DM for 56 days) containing saponins, polysaccharides and flavonoids. In that study, BW was not affected, but DWG was higher in the treatments supplemented with *Astragalus membranaceus*, perhaps as a consequence of the beneficial effects that the saponins, flavonoids and polysaccharides of the plant had on the antioxidant and immune status, and on the serum concentration of growth hormone in the animals.

Some plants containing saponins, polysaccharides and flavonoids increase the relative abundance of fibre-degrading bacteria in the rumen [41]. This could result in higher fibre and feed digestibility and could also increase ruminal passage rate and dry matter intake. However, in our study, DMI was similar among lambs of all treatments during the experimental period. Although HM could increase the rate of passage, saponins are natural surfactant glycosides, which may have a bitter and astringent taste for animals [44]. This can cause low palatability of the diet, which would partially explain the absence of changes observed in DMI. In a similar study, Liu et al. [42] investigated the effects of extracts of *Medicago sativa* saponins (0, 0.5, 1, 2 and 4 g kg−<sup>1</sup> DM for 90 days) on the productive performance of lambs. In that study, DMI increased linearly as the dose of saponins in the diet increased. This suggests that the lambs are able to adapt to consume saponins, but this adaptation could require long periods of supplementation.

Some plant-extracted saponins have shown promising effects on improving feed utilization efficiency because they can suppress enteric methane emissions through direct effects on ruminal microorganisms [12,14]. In the present study, FCR was similar among treatments, suggesting that HM did not affect feed utilization efficiency. The absence of significant changes in FCR could be explained by the fact that DMI and DWG were also not affected by the treatments. Similar results were previously reported by Mandal et al. [18] in goats supplemented with 5 g d−<sup>1</sup> of *Acacia concinna* pods for 90 days, and by Liu et al. [42] in lambs supplemented with alfalfa saponin extracts (0.5, 1, 2 and 4 g kg−<sup>1</sup> DM for 90 days). In their study, they observed that FCR was similar among treatments, even though feed digestibility was higher in lambs supplemented with saponins.

Previous studies have reported that digestion and utilization of nutrients in the diet of ruminants could be improved by dietary supplementation of saponins [12,45], and plants containing saponins, polysaccharides and flavonoids [41]. However, in our study, a negative effect of HM on DM digestibility was observed. Similar results were previously reported by Nasri et al. [43] in lambs supplemented with saponin extracts from *Quillaja saponaria* at dietary concentrations of 30, 60 and 90 mg kg−<sup>1</sup> DM; and by Nasehi et al. [21] in lambs supplemented with increasing doses (0, 6.1, 8.7 and 11.3 g kg−<sup>1</sup> DM) of saponins from the green tea plant (*Camellia sinensis)*. Their results showed that saponins reduced the digestibility of DM of the lambs but did not affect their productive performance.

Regarding carcass characteristics, HCW and HCY were not affected by dietary supplementation of HM. No information is available on the effects of HM containing saponins, polysaccharides and flavonoids on sheep or goa<sup>t</sup> carcass characteristics. However, results that are congruen<sup>t</sup> with our findings were previously reported by Nasri et al. [43] on lambs supplemented with increasing doses of saponin extracts from *Quillaja saponaria* (0, 30, 60 and 90 mg kg−<sup>1</sup> DM for 57 days); and by Abdallah et al. [46] on sheep supplemented with 10 and 15% dried *Astragalus membranaceus* roots containing saponins, flavonoids and polysaccharides. Their results showed that HCW and HCY were not affected by dietary supplementation of saponins, and neither were they affected by the mixture of saponins, polysaccharides and flavonoids from *Astragalus membranaceus*. The limited information on the effects of HM on ruminant carcass characteristics makes it difficult to explain the results observed in this and other studies. However, the similarity of BFT in the carcass of lambs from all treatments may partially explain the absence of changes in HCY in the present study.

BFT and LMA were also not affected by the HM dietary supplementation. The mechanism of action of herbal products and their bioactive compounds on lipogenesis has not been studied in lambs [4]. However, Liang et al. [47] observed that, in beef cattle fed with high-grain rations, supplementation of flavonoid extracts in the diet increased BFT through changes in the differential expression of genes involved in lipid metabolism. In the present study BFT was not affected by the inclusion of HM in the diet, even though it contains parts of the plant *Balanites roxburghii*, which contains flavonoids [25]. This suggests that the effects of flavonoids on BFT are dependent on botanical origin. Given that fat deposition, physical and chemical carcass characteristics of lambs are influenced by breed, sex, age and weight [48,49], the homogeneity of these characteristics in the lambs used in the present study partially explains the absence of changes in LMA and BFT.

Regarding the internal and external organs of lambs, similar results were previously reported by Hundal et al. [19] in goats supplemented with 2% of *Macrotyloma uniflorum* seeds containing saponins; and by Abdallah et al. [46] in sheep supplemented with 10 and 15% of *Astragalus membranaceus* roots containing saponins, flavonoids and polysaccharides. They observed that the weight of the kidneys on sheep supplemented with the highest dose of saponins, flavonoids and polysaccharides from *A. membranaceus* was higher than that on sheep from the other treatments, but there was no effect on the other internal organs. Information on the effects of herbal products or their bioactive compounds on the size and weight of internal organs in ruminants is still limited, which makes it difficult to explain the results observed in this study. However, differences in the internal organs of sheep are

influenced by the breed, sex and age of the animals [50], and by the feeding regime [51]. In the present study all these factors were controlled, which would partially explain the absence of significant changes.

The lowest pH of the meat was observed in the lambs with the HM1 treatment, while in the animals of the other treatments the pH was similar, within the normal range of 5.5 to 5.8 suggested by Sañudo et al. [52]. Abdallah et al. [46] did not observe pH changes in the meat of lambs supplemented with 10 and 15% *Astragalus membranaceus* roots containing saponins, flavonoids and polysaccharides. In another study, Nasri et al. [43] also did not observe pH changes in the meat of lambs supplemented with saponin extracts from *Quillaja saponaria* at concentrations of 30, 60 and 90 mg kg−<sup>1</sup> DM. However, it was observed that the pH of the meat in lambs of all treatments was below the normal range, similar to what was observed in our study with the HM1 treatment. Therefore, the effects of HM on the pH of the meat observed in the present study could be related to the presence of bioactive compounds. The pH is important for preserving meat during storage. A low pH has a bacteriostatic effect, while a pH above the normal range favours the growth of proteolytic microorganisms [53,54]. This suggests that supplementation of low doses of HM in the diet could promote favourable bacteriostatic effects in lamb meat, and thus increase its shelf life.

Ponnampalam et al. [55] mentioned that an ultimate pH > 5.8 is associated with alterations in drip loss and WBSF. In addition, in sheep meat, Watanabe et al. [56] reported a curvilinear association between ultimate pH and WBSF values, with a toughness peak at pH around 6.0 and improvements in tenderness at pH below and above 6.0. In our study, WBSF decreased as the dose of HM increased; however, this result must be carefully interpreted considering the low number of replicates used and the high coefficient of variation observed (30.74%, data not shown). Similar results were previously reported by Qin et al. [57] in lambs fed pomace (7.8 and 16% for 80 days) obtained from *Hippophae rhamnoides* fruits, which contained 0.69 and 1.02% flavonoids, respectively. In that experiment, WBSF decreased when the flavonoid dose increased. In another study, Abdalla et al. [46] observed no significant changes in WBSF of meat from lambs supplemented with saponins, flavonoids and polysaccharides from *Astragalus membranaceus* roots. WBSF is a well-known method for estimating the meat tenderness [57], consequently, the lower WBSF observed in the present study suggests that dietary supplementation of HM could improve the lamb meat tenderness. Although the exact mechanism is unknown, the changes in WBSF observed in this and other studies sugges<sup>t</sup> that bioactive compounds contained in some plants facilitate the activation of some peptidases such as calpains and cathepsins, which help prevent and delay post-mortem muscle fibre stiffening [58]. It is also possible that these bioactive compounds act by reducing calpastatin activity, allowing a higher rate of myofibril protein degradation [59]. This hypothesis is supported by the observed linear increase in drip loss as WBSF decreased, because drip water losses may increase when calpastatin activity decreases [60]. Furthermore, Webb and Agbeniga [61] reported a linear relationship between WBSF and drip loss, in which higher drip loss was associated with rapid tenderisation and lower WBSF of the meat.

Drip loss is associated with the capacity to retain water in the muscle, with the juiciness and the tenderness of the meat [46,49]. In the present study, the drip loss of meat increased when the dose of HM increased, indicating that high doses of HM could affect the water retention capacity, tenderness and juiciness of meat. Abdallah et al. [46] investigated the effects of dietary supplementation with dried *Astragalus membranaceus* roots containing saponins, flavonoids, and polysaccharides, and observed that meat drip loss decreased in response to *A. membranaceus* supplementation. However, WBSF was similar in the meat of lambs from all treatments. Although the exact mechanism involved is unknown, the higher drip loss observed in the meat analysed in the present study could be related to the observed changes in WBSF, as previously discussed.

Colour is an important attribute of meat quality because it is the first aspect that attracts consumers when choosing fresh meat [62]. A variety of secondary compounds from plants can improve oxidative stability and prevent discolouration of meat of small ruminants [2]. In the present study, HM did not affect the values of L\*, a\*, Chroma and Hue ◦. However, b\* decreased in response to supplementation of HM in the diet. This result could be positive because consumers generally do not expect to find high b\* in fresh meat [63]. Similarly, previous studies [46,64] reported that supplementation with medicinal plants containing saponins, polysaccharides, and flavonoids also did not affect the colouration of meat from lambs and goats.

There is little information on the use of HM containing saponins, polysaccharides and flavonoids as a colour preservative in ruminant meat. The pigment content of meat can modify its colouration [35]. Likewise, the inclusion of some medicinal plants containing flavonoids increases the hypertrophy of muscle fibres in lambs [57], which could dilute the content of muscle pigments and consequently alter meat colour [65,66]. These findings sugges<sup>t</sup> that the HM used could increase muscle hypertrophy, which would partially explain the observed reduction in b\*. On the other hand, Luo et al. [64] reported that dietary supplementation of medicinal plants containing saponins, polysaccharides and flavonoids altered the pigment content on the meat of small ruminants. Similar effects of consumption of these metabolites would partially explain the b\* changes in the meat of lambs supplemented with HM in the present study.

In the present study, the chemical composition of lamb meat was similar in all treatments, perhaps as a consequence of the low impact of HM supplementation on the nutritional composition of the diet. In a similar study, Abdallah et al. [46] investigated the effects of *Astragalus membranaceus* roots (0, 10 and 15% for 47 days) containing saponins, polysaccharides and flavonoids on sheep meat quality. In that research, the moisture, protein and ash content of the meat was similar among treatments. However, they observed that fat content decreased in sheep that ate *A. membranaceus* roots. Furthermore, in our study, HM supplementation had little impact on the final BW of the lambs, all being of the same breed and age, which partially explains the absence of significant changes in the chemical composition of the meat [48,49,54].
