2.1. Preparation of Experimental Rations

The sweet potato plant was grown on an experimental farm, and the vines were collected after separating them from the stem. The SPV was then spread on a plastic sheet for drying in an area protected from direct sunlight. After the drying process was completed, the SPV was ground using mortar and pestle, packed in bags of 100 kg, and stored in a cool and dry place until used. The CM was prepared from wheat bran, ground maize, linseed cake, and salt (Table 1). The CM was formulated to contain CP and energy concentrations to meet the minimum recommendation reported in [16] for intensive small ruminant feeding (i.e., 17% CP and 9 MJ ME/kg DM). The SPV was mixed with the CM when preparing the experimental diets.

2.2. Experimental Design and Treatment Diets

Forty yearling local rams with an initial average body weight of 18.5 ± 0.7 kg were allowed to acclimatize to the experimental environment for 2 weeks. During the adaptation period, all the rams were sprayed with acaricide and drenched with the recommended dose of anthelmintics. During the acclimatization period, rams were identified with numbered ear tags and fed with a CM and grass hay ad libitum. At the end of the acclimatization period, individual animals were weighed and blocked by the similarity of their live weight. Then, individual rams from each block were randomly assigned into five treatment diets with eight rams each. The individual pens were fitted with feeders and watering troughs. The CM/SPV mixture was offered in combination twice a day in equal portions while the grass hay was provided to all treatment groups ad libitum. The supply of grass hay was measured, and adjustments were made when the feed refusal was less than 10% of the offered.

The yearling rams were assigned into five dietary groups: SPV0, control diet without SPV; SPV10, a diet with 10% replacement of CM by SPV; SPV20, a diet with 20% replacement of CM by SPV; SPV30, a diet with 30% replacement of CM; and SPV40, a diet with 40% replacement of CM by SPV. Accordingly, the ratios of CM: SPV mixture offered to the SPV0, SPV10, SPV20, SPV30, and SPV40 treatment groups were 385:0, 347:38.5, 308:77, 270:116, and 231:154 g/head day, respectively. The experiment lasted 82 days, excluding the acclimatization period of 15 days.

2.3. Data Collection Procedures

2.4. Protocols of the In Vitro Gas Production (GP) and Methane Production

2.5. Chemical Analysis

Analyses of proximate nutrients and minerals were performed as outlined in [20] at the Institute of Animal Science, University of Hohenheim, Germany. The samples were analyzed for DM (Method 3.1), crude ash (Method 8.1), CP (Method 4.1.1), and petroleum EE (Method 5.1.1). Neutral detergent fiber (aNDFom) was assayed on OM basis after amylase treatment and acid detergent fiber (ADFom) on an OM basis (Methods 6.5.1 and 6.5.2, respectively). Acid detergent lignin (ADL) was analyzed according to Method 6.5.3. The analysis of calcium (Ca), phosphorous (P), magnesium (Mg), potassium (K), sodium (Na), iron (Fe), copper (Cu), and manganese (Mn) were determined according to Methods 10 and 11 of [20] using an inductively coupled plasma spectrometer (ICP-OES). All analyses were performed in duplicate and averaged.

2.6. Statistical Analysis

Results of the proximate and mineral compositions are expressed as means for duplicate analysis of feed samples. Data on feed intake, body weight, carcass parts, in vitro GP and CH₄ production, ME, and dOM were subjected to one-way ANOVA by fitting treatment diets as a single factor. The mean comparisons were conducted using Tukey’s Studentized range test. In addition, the orthogonal polynomial contrast analysis was performed to determine the linear, quadratic, and cubic trends due to the substitution of CM with SPV on feed intake, weight gain, FCR, and carcass components. All analyses were conducted by using the Statistical Analysis System (SAS, 2016, Ver. 9.4). The effects were considered significant at p < 0.05, and variability in the data was expressed as the pooled standard error of the mean (SEM).

3.1. Nutrient Compositions of SPV and Experimental Diets

The analyzed proximate compositions of treatment diets, grass hay, and SPV are presented in Table 2. The CP and ash contents in SPV were higher than those in any of the treatment diets. Similarly, the SPV had higher values of NDFom, ADFom, and ADL than the experimental diets. However, the SPV was low in EE content. Although the grass hay had the lowest values of CP and EE, it contained higher NDFom, ADFom, and ADL as compared with the SPV and experimental diets. The ash content in grass hay was also higher than that in the SPV0, SPV10, and SPV20 diets. Among treatment diets, the ash and CP contents improved with increased substitution levels of CM with SPV. Similarly, the contents of structured carbohydrates consistently elevated with increased substitution levels of SPV. On the other hand, the EE content was reduced with increased substitution levels with SPV.

As shown in Table 3, the SPV had the highest concentrations of Ca and K as compared with grass hay and treatment diets, the latter mineral being exceptionally high. As a result, the concentration of both minerals specifically that of Ca consistently increased across treatment diets with increased replacement of CM by SPV. The concentration of P and Mg was higher in the SPV than in the grass hay, and the former slightly decreased with increased substitution levels. On the other hand, the SPV was deficient in some other minerals, notably Na. The concentration of Cu showed a decreasing trend with increased substitution levels of SPV. The highest Fe concentration was observed in grass hay and showed an inconsistent pattern among treatment diets.

3.2. Intake of Nutrients, Body Weight Gain, and Feed Conversion Efficiency

The substitution of CM by SPV up to 40% (SPV40) did not affect the final body weight, weight gain, feed intake, and FCR of the yearling rams (Table 4). Accordingly, the final body weight and the total body weight gain values were similar across treatment diets. The total feed intake and FCR values were also similar across treatment diets. Rams fed with the control diet and the SPV30 diet had similar body weight and weight gain values. Moreover, rams fed on the SPV20 and SPV30 diets showed similar FCR values. None of the parameters studied showed a significant linear, quadratic, or cubic trend with increasing substitution levels of SPV.

As shown in Table 5, the intake of nutrients from hay showed very similar patterns and did not differ among treatment diets. However, the intake of all nutrients from CM significantly (p < 0.001) reduced with increased substitution levels of CM by SPV. On the other hand, the intake of all nutrients from SPV consistently increased as the substitution levels of SPV increased. The CP intake from SPV was tripled in the SPV30 and SPV40 diets along with increased substitution levels of CM with SPV. Regarding the total nutrient intake, no difference was observed among treatment diets for DM and NDF intake. As the substitution levels of SPV increased, there was a significant (p < 0.001) increase in the total intake of CP and ADF. Conversely, the total EE intake decreased (p < 0.001) across treatment diets as the substitution of CM with SPV increased.

3.3. Carcass Components

The preslaughter weight has differed (p < 0.05) among rams being higher in the SPV0 diet than among those reared in the SPV10, SPV20, and SPV30 diets, while it was similar to those fed on the SPV40 diet (Table 6). Moreover, the preslaughter weight showed linear (p = 0.025) and quadratic (p = 0.007) effects with increased substitution levels of CM with SPV. Rams fed with the SPV0 diet had also higher (p < 0.05) dressed hot carcass weight than those reared in the SPV10, SPV20, and SPV40 diets. Significant linear (p = 0003) and cubic (p = 0.013) effects were observed in hot carcass weight with increased substitution levels of CM with SPV. The weight of the thorax and foreleg was higher (p < 0.05) in rams fed with the SPV0 diet than in those reared in different substitution levels of CM with SPV. The linear, quadratic, and cubic effects were significant for thorax weight with increased substitution levels of CM by SPV. Similarly, significant linear (p = 0.002) and cubic (p = 0.031) trends were noted for foreleg with increased substitution levels.

Rams fed on the SPV0 diet had higher (p < 0.05) hindleg and liver values than those fed with the SPV40 diet. Likewise, significant linear and cubic trends were observed for both carcass cuts (Table 6). The average lumbar weight was highest in rams fed with SPV20 followed by SPV0 though the difference was insignificant. The dressing percentage ranged from 37.6 in rams fed with the SPV0 diet to 39.7% in those reared in the SPV30 diet and was not affected by the substitution of CM with SPV. However, a significant quadratic (p = 0.022) trend was observed for dressing percentage with increased substitution levels of CM with SPV. No significant difference was observed for the lumbar, neck, heart, and REA between rams fed with the control diet (SPV0) and the other treatment diets. Similarly, the orthogonal contrast effects were insignificant for these carcass parameters. Although insignificant, rams fed on the SPV0 diet had the highest REA, while those reared in the SPV20 diet had the lowest value.

3.4. In Vitro GP and CH₄ Production

The in vitro 24-h GP was higher (p < 0.001) for the SPV0 diet than for the SPV30 and SPV40 diets (Table 7). The GP was similar among the SPV20, SPV30, and SPV40 diets. The SPV showed the lowest (p < 0.001) 24-h GP as compared to treatment diets. The CH₄ production of the SPV0 diet was similar to that of the SPV10, SPV20, and SPV30 diets. However, the SPV40 diet and SPV had the lowest CH₄ production and differed (p < 0.001) from that of SPV0. The pCH₄ did not differ among treatment diets but was higher (p < 0.01) than that observed in SPV. As shown in Figure 1, the in vitro CH₄ production consistently and significantly reduced across the treatment diets as the substitution levels of CM with SPV increased. The ME in the SPV0 diet was similar to that in the SPV10 and SPV20 diets but was higher (p < 0.001) than that in the SPV30 and SPV40 diets and SPV. Both the SPV30 and SPV40 diets had higher (p < 0.001) ME values than the SPV. The dOM did not differ between treatment diets (Figure 2). Although SPV had the lowest dOM, it did not differ (p > 0.05) from the SPV30 and SPV40 diets (Table 7). The SPV had the lowest GP, CH₄, pCH4, and ME values than all other treatment diets.

4.1. Nutrient Composition of Experimental Diets and SPV

As the substitution level of CM with SPV increased, the concentrations of crude ash, CP, NDFom, ADFom, and ADL improved which is attributed to the high corresponding values of SPV. The CP value of SPV was comparable to the values reported in [10, 21] for the leaf part of the plant but was higher than that reported in [8] which was reported for local sweet potato leaves. According to [21], the CP values ranged from 20% to 28% for the leaves of four varieties of sweet potato with an average value of about 24%. In another study, the CP content of different SPV varieties (which included both the leaves and stem) was reported to be in the range of 15.5%–18.7% on DM basis [5].

The NDF and ADF values observed in the current study were higher than those reported in [3, 10]. The nutrient content of the SPV could be affected by the type of variety and soil in which it has been cultivated. Moreover, such variations in the concentrations of structured carbohydrates might be associated with the maturity stage of the vine and the type of variety used. High level of ADF is particularly associated with low digestibility of the feed which would apparently reduce the feed intake by animals.

The SPV used in this experiment showed higher values for the trace elements of Fe (1211 mg/kg DM) and Mn (114 mg/kg DM), and this observation is consistent with those reported in [8]. Both elements play key functions in rumen digestion, performance, fertility, and fitness [22, 23]. The current results may not necessarily show an oversupply of both microelements considering the fact that grazing ruminants in smallholder settings are not adequately supplemented with such minerals [24]. Thus, the current observation indicates that SPV could serve as a reliable source of trace mineral supplementation for grazing ruminants. With increased substitution levels of CM with SPV, the Ca concentration enhanced while that of P slightly reduced and this observation is in line with that of [8], which reported a relatively high Ca concentration with low p value in SPV.

4.2. Intake of Nutrients, Weight Gain, and Carcass Components

Although total DM intake was similar in all treatment groups, that of CP and ADF consistently and significantly enhanced with increased substitution levels of CM with SPV whereas the EE intake decreased. In agreement with the current observation, [25] reported that the intake of CP, NDF, and ADF improved with the increased supplementary levels of sweet potato silage.

The lack of significant difference in final body weight and body weight gain among rams fed with different substitution levels of CM with SPV may suggest that the treatment diets were comparable in the supply of nutrients during the growth of the yearling rams. However, the pattern of the weekly body weight (data not shown) showed a linear increase which suggests that replacing CM by SPV has enhanced the growth performance of yearling rams. This observation is similar to that reported in [25], which reported that improved growth performance was observed in lambs fed with various levels of SPV silage containing 70% vine and 30% tuber. In another study, the authors in [3] reported that SPV silage has improved diet digestibility and the retention of nitrogen in tropical cattle. Similarly, the authors in [25] reported a decreasing trend of FCR with increasing supplementation levels of SPV silage in local goats which might be associated with the quality of the feed which is composed of not only potato vines but also the tuber part that might be rich in carbohydrates.

Although the slaughter and hot carcass weight was higher in rams fed with the control diet (SPV0) than in those fed with different levels of SPV, it did not show a clear trend among treatment diets. For instance, the slaughter weight of rams fed with SPV0 did not differ from those of SPV40, and the hot carcass weight of SPV0 was not different (p > 0.05) from that of SPV30. Such observations are indeed very difficult to justify based on previous works in the literature. The weight of the lumbar, dressing percentage, and rib-eye muscle area were similar across treatment diets, which suggests that replacement of CM with SPV might be an option to feedlots where conventional feeds may not be affordable. Likewise, the weight of hindleg was similar among all rams fed with treatment diets except those in SPV40, which suggests the significance of SPV supplementation to poor-quality forages in smallholder settings.

The dressing percentage was not affected by the replacement of CM with the SPV. Consistent with the current findings, the authors in [12, 25] reported a significant improvement in dressing percentage in goats and lambs supplemented with SPV. Improvement in carcass traits is very important for the smallholders because the commercial value of carcass components from sheep is determined by a range of different features, notably the rib-eye area (area of the Longissimus dorsi muscle), which has been a research focus on a molecular level by using advanced technologies [19, 26, 27]. The rib-eye muscle area is significantly associated with the amount of muscle in the carcass lean meat yield [28, 29] indicating its significance in commercial marketing.

4.3. In Vitro GP and Methane Production

To the authors’ knowledge, there is only limited information available reporting on the CH₄ production potential of SPV or other parts of the plant, making it difficult to discuss the current findings with a view to published works in the literature. The CH₄ production (mL/g DM) of the SPV30 and SPV40 diets was (p < 0.05) lower than that of the control diet (SPV0) which indicates that the replacement of CM with SPV has considerably reduced the CH₄ production. The CH₄ reduction ability of SPV was reported in [8] and described as a potential candidate in mitigating CH₄ emissions from tropical livestock. The potential of the SPV in reducing CH₄ as compared to CM might be associated with the higher amount of fiber fractions present in SPV. On the other hand, many researchers [30–32] have reported that high concentrations of bioactive plant secondary metabolites such as polyphenolic compounds may have the potential to hinder the CH₄ production in the rumen though the possible mechanisms of their effects on rumen methanogenesis have not yet fully understood [33].

On top of that, the CH₄ emissions from the ruminants can be mitigated through the proper selection of feed ingredients including unconventional feeds to be used in the formulation of diets by giving emphasis to their suitable combination ratios [34, 35]. In the current study, the CH₄ production was consistently reduced with increased substitution levels of CM with SPV. This observation could be explained by the fact that within the concentrate feeding, soluble sugars produce more gas and CH₄ than feeding with SPV that would contain low sugar as it has been evident in its relatively low energy density.

The present result suggests that the use of SPV as a supplement of tropical forages could play a pivotal role in mitigating CH₄ production from the ruminants with positive effects on voluntary feed intake, growth performances, feed conversion efficiency, and some carcass components. This observation is supported by the findings of [3] who reported that feeding the SPV silage to the heifers has improved the diet digestibility as well as the retention of nitrogen by reducing the CH₄ emissions per unit of digested feed.

With increased substitution levels of CM with SPV, the contents of NDF, ADF, and ADL increased while that of GP, ME, and dOM reduced. These observations are consistent with those of [36] who reported a negative correlation of NDF and ADF with ME and dOM in legume plants. Although no significant difference was noted in the growth of yearling rams, the ME concentration has significantly decreased with increased substitution of CM with SPV notably in the SPV30 and SPV40 diets which may suggest the necessity of provision of good energy source feeds. Moreover, using SPV as a sole feed resource or increasing its substitution levels in the CM diet may affect its palatability with possible determinantal effects on ME digestibility specifically with high rates of inclusion as observed in rams fed with the SPV40 diet. Although the SPV could be supplied to animals in the form of silage, the CP content could be reduced substantially [3]. Therefore, the dried form of SPV might be preferred for feeding ruminants to maintain its CP content and prevent the possible loss of other nutrients during the process of silage making. Furthermore, the dOM did not differ among treatment diets, suggesting that feeding the dried form of SPV could be beneficial in small ruminant nutrition.