Evaluating nutritional value of processed potato vines by in vitro gas production

ABSTRACT

This study aimed at determining the in vitro nutritive value of treated potato vines and evaluating their potential to replace lucerne in ruminant diets. Potato vines were treated as air-dried (H); ensiled without additive (S), with 10% molasses (SM) or with 10% molasses plus 4% urea (SMU); pre-dried and ensiled either with 10% molasses (HSM), or with 10% molasses plus 4% urea (HSMU). Gas production kinetics, and ruminal digestibility and fermentation of the vines were determined in vitro alone or as 20% or 40% of a mixed ration with lucerne. The SMU and HSMU had the highest asymptotic gas production, ruminal digestibility, total volatile fatty acid (TVFA) and microbial protein (MP). Including SMU and HSMU at 40% in the mixed ration resulted also in the highest digestibility, TVFA and MP. These findings indicate the great potential of additive-treated potato vine silages to replace conventional fodder in the diets of ruminants.

KEYWORDS:

• Silage treatment
• food by-products
• feeding value
• ruminant nutrition
• non-conventional forages

Introduction

By-products from the agro-food industry are of increasing importance, which can alleviate part of feed scarcity and thereby contribute to produce cost-effective animal products (Grasser et al. 1995; Valizadeh & Sobhanirad 2009).

The vines of potato (Solanum tuberosum) are one of these promising by-products, which have great potential to replace fodder in ruminant diets. Potatoes are one of the world’s most important food crops with an annual yield of 365 million tons worldwide (FAO 2013). Correspondingly, there is a large quantity of potato vines produced annually (38–58 tons/ha), which could be used as animal feed (Parfitt et al. 1982). However, considered as waste, potato vines are often removed and destroyed some days before tubers are harvested (Nicholson et al. 1978). This is due mainly to their glycoalkaloid content, which is toxic to mammals (Morris & Lee 1984; Walker 1997). From limited data on ruminants, the toxic dose of α-solanine (one of the main components of potato glycoalkaloids) has been reported to be 225 mg/kg body weight for sheep (Konig 1953). Although animals are considerably less susceptible to glycoalkaloids than humans (Friedman et al. 1997), fresh potato vines with glycoalkaloid content of 60–300 mg/100 g dry matter (DM) may be harmful to ruminants when fed in large quantities (Nicholson et al. 1978; Parfitt et al. 1982; Walker 1997). However, processing such as air-drying or ensiling of potato vines may reduce their glycoalkaloid content (Nicholson et al. 1978), potentially making them safe to be used in ruminant nutrition. Similarly, Parfitt et al. (1982) and Dijkstra (1945) reported that potato vine silages can be used safely in the feeding of ruminants.

While the nutritive value of potato tuber or its by-products from the agro-food industry, as animal feeds, has been documented (Rooke et al. 1997; Eriksson & Murphy 2004; Eriksson et al. 2009), there is very limited information on the nutritive value of different forms of potato vines. Therefore, the objectives of the present study were to determine the in vitro nutritive value and rumen fermentation characteristics of processed potato vines and evaluate the impact of their inclusion into a mixed ration on its ruminal digestion and fermentation.

Materials and methods

Potato vines harvesting and preliminary processing

Potato vines of ‘Marfona’ cultivar (Seed and Plant Improvement Institute, Karaj, Iran) were harvested at 90 days after planting from three potato farms in Hamedan province and immediately transported to the research centre laboratory (Faculty of Agriculture, 34°–47′–53.7″ N, 48°–28′–49.5″). The potato vines were mixed and then chopped (to about 1 cm) with a commercial vegetable chopper (AKA Electric Ltd. Tehran). One part was frozen at −20°C for chemical analysis and the remainder was divided into two parts; one part was air-dried in the shade to a constant weight over two weeks and the other part was wilted for two days and then ensiled. Lucerne (Medicago sativa) hay (second cut) which is commonly used in ruminant diets was used as a control. It was harvested in the early blooming phase from three farms in Hamedan province and air-dried to a constant weight.

Treatments and incubations

The current in vitro experiment consisted of three sets of incubations: the first set of incubations of 144 h was used to determine the gas production kinetics of the treated potato vines compared with that of lucerne hay. Ruminal digestibility and fermentation of the treated potato vines and lucerne hay were determined using the second set of incubations of 24 h. The third set of incubations of 24 h was used to evaluate the ruminal digestibility and fermentation of a practical mixed ration (typical for fattening lambs) containing different levels of treated potato vines substituting lucerne hay. Each set of incubations was repeated twice (n = 2 runs) on separate days.

In the first and second sets of incubations, treated potato vines and lucerne hay were considered as the treatments; they consisted of (1) potato vines hay, dried to a constant weight (H); (2) potato vines wilted during 2 days and subsequently ensiled without additives (S, considered as untreated potato vine silage); (3) potato vines wilted during 2 days and subsequently ensiled with 10% (W/W DM) molasses (SM); (4) potato vines wilted during 2 days and then ensiled with 10% molasses and 4% urea (SMU); (5) potato vine hay soaked gradually with water (1:1.5, W/W) and mixed thoroughly (to reach approximately the same DM content as that of wilted vines used in SM and SMU) and ensiled with 10% molasses (HSM); (6) potato vine hay soaked and ensiled with 10% molasses and 4% urea (HSMU) and 7) lucerne hay.

Potato vines intended for ensiling (1 kg) were filled into plastic buckets of 1 l, pressed, covered by plastic cellophane and sealed tightly. The buckets were then placed in black plastic bags, sealed and left at room temperature to be ensiled for two months. At the end of the fermentation period, the buckets were opened and pH of the silages was determined using a portable pH metre (WTW multilab 540 Ionalyser, Weilheim, Germany). For this purpose, 5 g fresh silage was weighed into a beaker containing 100 ml distilled water and mixed thoroughly; after standing for 30min at room temperature, pH was measured.

In the third set of incubations, H, SMU and HSMU were included in a mixed ration to evaluate in vitro their impact on the ruminal digestibility and fermentation of the ration. For this purpose, a basal ration composed of (DM basis) 40% lucerne hay, 40% barley, 13% wheat straw and 7% soybean meal, providing 10.8 MJ metabolisable energy (ME) and 140 g crude protein (CP) per kg of the ration, was considered as the control (CTRL). Lucerne in the basal ration was replaced by 50 and 100% with H, SMU and HSMU in H₅₀, H₁₀₀, SMU₅₀, SMU₁₀₀, HSMU₅₀ and HSMU₁₀₀ treatments, respectively. The rations were formulated to be iso-energetic and iso-nitrogenous.

In vitro gas production experiments

The gas production procedure was conducted as described by Menke and Steingass (1988). Rumen liquor for incubations was collected before the morning feeding from three ruminally fistulated Mehraban rams fed a maintenance diet composed of (per kg DM of the diet) 700 g lucerne hay and 300 g barley, providing 9.58 MJ ME and 137.4 g CP. Rumen liquors were pooled, strained through four layers of cheesecloth into a pre-warmed (38–39°C) insulated flask and immediately transported to the laboratory. For the incubation of potato vine silages, they were oven dried at 55°C for 48 h and ground to pass through a 1 mm sieve. In the first set of incubations, subsamples of 200 mg (DM basis) of the processed vines and lucerne hay were incubated in triplicate in 100 ml glass syringes with 30 ml of buffered rumen inoculum under continuous flow of CO₂. The inoculum was prepared by mixing rumen liquor with the buffer at a ratio of 1:2 (v/v). The buffer composed of 237 ml macro-minerals solution, 0.12 ml micro-minerals solution, 237 ml bicarbonate buffer solution, 1.22 ml Resazurin solution, 50 ml reducing solution and 474.7 ml distilled water per litre. Three syringes containing buffered rumen inoculum without fermentation substrate were used as blanks. The syringes were then placed in a water bath at 39°C and gas volume was measured at 2, 4, 6, 8, 12, 24, 48, 72, 96, 120 and 144 h of incubation.

Data on the gas produced at different times during 144 h of incubation in Exp. 1 were fitted to the model proposed by France et al. (1993) as shown in the below equation, by the nonlinear procedure of SAS (SAS 2002).where GP (ml) is the gas produced at the time t, ‘a’ (ml) is the asymptote of gas production, ‘b’ and ‘c’ are constants and L (h) is the lag time. T_1/2 (h, time to half asymptote of gas production) and µ (/h, fractional rate of gas production at T_1/2) were calculated using the following equations (France et al. 1993): Metabolisable energy of processed potato vines and lucerne hay was estimated according to Menke and Steingass (1988), using the following equation:where ME is the metabolisable energy (MJ/kg DM), GP is the gas produced after 24 h of incubation (ml/200 mg substrate), CP is the crude protein content (% of DM) and EE is the ether extract (% of DM).

In the second and third sets of incubations of 24 h, 500 mg of the samples was incubated anaerobically in triplicate with 40 ml buffered rumen inoculum in glass syringes during 24 h (Makkar et al. 1995). In the second incubations, potato vine silages were also used simultaneously in one other set of incubations of 24 h to measure immediate gas produced from the buffering of fermentation acids (FAs) (Grings et al. 2005). For this, the potato vine silages and rumen inoculum were first autoclaved at 121°C for 20 min and subsequently the vines were incubated in triplicate with 40 ml autoclaved rumen inoculum in the syringes. These incubations were used to correct the gas produced after 24 h of incubation (GP₂₄) to the indirect immediate gas produced from FAs buffering in the silages. Additionally, three syringes containing buffered rumen inoculum without the fermentation substrate were used as blanks. At the end of the incubation, the syringe contents were transferred into centrifuge tubes and placed immediately in cold water at 4°C to stop the fermentation. The tubes were then centrifuged at 15,000 × g for 20 min at 4°C, aliquots of 4 ml of the supernatant were mixed with 1 ml of 25% metaphosphoric acid and kept frozen at −20°C until subsequent analysis for VFA and ammonia. The remaining supernatant was discarded and the pellet was oven dried at 55°C for 48 h to estimate in vitro apparent dry matter degradability (IVADMD). Fermentation residues were subsequently refluxed with a neutral detergent solution (NDS) for 1 h to estimate in vitro true dry matter degradability (IVTDMD). In vitro true organic matter degradability (IVTOMD) of the samples was determined by subtracting the ash content of the neutral detergent fibre (NDF) fraction and that of the substrate.

Partitioning factor (PF) was estimated as the ratio of the organic matter (OM) truly degraded (mg) to the gas produced (ml) after 24 h of incubation (Blummel et al. 1997). Microbial protein (MP) production (mg) was estimated by nitrogen balance method using the following equation according to Grings et al. (2005):

where MP is microbial N × 6.25; diet N is the nitrogen content of the substrate; ΔNH₃–N = NH₃–N in blanks – NH₃–N in the cultures at the end of incubation; NDFN is the nitrogen content remaining in the residue after NDS refluxing.

Chemical analyses

Standard methods as described in Association of Official Agricultural Chemists (AOAC 2000) were used to determine DM, total ash, CP and EE. Crude protein was determined using the Kjeldahl method (ID no. 984.13). NDF and acid detergent fibre (ADF) were determined according to the amylase-free method described by Van Soest et al. (1991), and expressed exclusive of residual ash. Non-fibrous carbohydrate (NFC) was calculated using the following equation:

Ammonia concentration in the supernatant was determined using the phenol-hypochlorite method as illustrated by Broderick and Kang (1980). For this purpose, ammonia standard stock (100 mM) was prepared by dissolving 0.6607 g ammonium sulphate in 0.1 N HCl made up to 100 ml; working standards (1–32 mM) were made by diluting appropriate aliquots of stock standard. Phenol reagent solution was made by dissolving 0.15 g sodium nitroferricyanide in 1.5 l distilled water, then 33 ml phenol solution (90%) was added, mixed and made up to 3 l by distilled water. Hypochlorite reagent solution was prepared by dissolving 15 g NaOH in 2 l distilled water, then 113.6 g Na₂HPO₄.7H₂O was added into the solution and mixed with mild heating. After cooling, 150 ml commercial bleach (5.25% sodium hypochlorite) was added and the solution was made to 3 l by distilled water. This solution was then filtered through Whatman #1 paper and stored. For determining ammonia concentration in the samples, 50 µl of the rumen inoculum or standard solutions was pipetted into test tubes, then 2.5 ml phenol reagent and 2 ml hypochlorite reagent were added into each tube, vortexed and incubated at 37°C in a water bath for 10min. From the tube content, 300 µl was subsequently pipetted into spectrophotometer plates and absorbance was recorded at 630 nanometres. Ammonia concentration was determined using the equation calculated from standard curves. TVFA concentration was determined by a Markham apparatus as described by Barnett and Reid (1956). Briefly, 2 ml of the rumen inoculum plus 1 ml 10% potassium oxalate buffer and 1 ml oxalic acid was injected to the apparatus and a distillate of 100 ml was collected and subsequently titrated against a standard of 0.01 N NaOH using phenolphthalein as the indicator. TVFA concentration was then calculated using the following equation:

Statistical analysis

All the incubations were repeated twice (runs) on two different days. Data (7 treatment levels × 3 replications = 21 data for each variable or parameter) on estimated gas production kinetic parameters in the first set of incubations and those of digestibility and fermentation variables in the second and third sets of incubations were averaged across runs (days) and subjected to analysis of variance by the General Linear Model procedure of SAS (SAS 2002) using the following model:where Yij is the observation, µ is the overall mean for each parameter, Ti is the effect of treatments and eij is the residual error. In the case of statistical significance (declared at P < .05), the means for each parameter or variable were compared among the treatments using Duncan’s multiple range test.

Results

Chemical composition

Chemical composition of the samples is summarised in Table 1. Both ensiled and hay forms of potato vines had lower OM content compared with lucerne (P < .05). Among the potato vines, the silages had higher OM than H (P < .05) and the OM content of S was highest among the silages. Potato vines hay had slightly lower CP than lucerne hay (P < .05), but the CP content of S was comparable to that of lucerne hay. Among the potato vine silages, molasses + urea-treated silages (SMU and HSMU) had higher CP content than those treated only with molasses (SM and HSM) (P < .05) and untreated potato vine silages (S) had the lowest CP. Potato vines had lower NDF and ADF, and higher NFC than lucerne hay (P < .05). Among the potato vines, NDF and ADF contents of the silages were lower than those of H (P < .05). The pH value of potato vine silage made without additives (S) was higher than that of other silages made with additives (P < .05).

Table 1. Chemical composition of processed potato vines and lucerne hay.

Items	Treatments							SEM	P-value
	S	SM	SMU	HSM	HSMU	H	Lucerne
DM (%)	37.0^g	39.8^f	40.5^e	41.7^d	42.6^c	93.9^b	96.9^a	0.10	<.001
OM (% of DM)	87.4^b	85.8^cd	86^cd	85.4^d	86.2^c	83.8^e	90.4^a	0.22	<.001
NDF (% of DM)	37.0^c	34.7^d	33.6^d	35.1^d	34.3^d	41.6^b	44.8^a	0.45	<.001
ADF (% of DM)	27.8^c	26.6^d	24.7^e	26.9^cd	25.8^de	32.3^b	36.3^a	0.35	<.001
CP (% of DM)	14.9^de	16.3^c	22.2^a	16.2^c	21.2^b	14.4^e	15.5^cd	0.24	<.001
pH (% of DM)	5.36^a	4.12^b	4.15^b	4.14^b	4.21^b	–	–	0.059	<.001
EE (% of DM)	2.05	2.15	2.15	2.15	2.15	1.85	1.90	0.243	.925
NFC (% of DM)	33.6^a	32.7^ab	28.1^d	32.0^b	28.5^d	26.0^e	30.2^c	0.46	<.001

Notes: DM, dry matter (%); OM, organic matter (% of DM); CP, crude protein (% of DM); NDF, neutral detergent fibre (% of DM); ADF, acid detergent fibre (% of DM); EE, ether extract (% of DM); NFC, non-fibrous carbohydrates; S, untreated potato vine silage; SM, pre-wilted potato vines ensiled with 10% molasses; SMU, pre-wilted potato vines ensiled with 10% molasses plus 4% urea; HSM, pre-dried potato vines ensiled with 10% molasses; HSMU, pre-dried potato vines ensiled with 10% molasses plus 4% urea; H, potato vines hay.

SEM, standard error of the means. a,b: different superscript letters indicate differences (P < 0.05) among the treatments.

Gas production kinetics

The asymptotic gas production (a) was affected by treatment (P < .001), with the highest amount obtained in HSMU (Table 2). Preparing the potato vine silages with molasses increased their asymptote of gas production compared with S (P < .05); however, treating the silages with both molasses and urea increased further this parameter. Among the potato vines, the asymptotic gas production of H and HSMU was comparable to that of lucerne hay. There was also a significant difference among the treatments for lag phase (L) (P < .001). Lucerne hay and H had shorter lag phase than the silages (P < .05). Among the silages, urea treatment resulted in a shorter lag phase; however, treating the silages with molasses shortened the lag phase only in the silages made from pre-dried potato vines (HSM and HSMU). The variation of T_1/2 was the same as that observed for the lag phase, as H had the shortest T_1/2. Among the silages, S had longer T_1/2 than additives-treated silages (P < .05), and molasses plus urea-treated silages had shorter T_1/2 than the silages treated only with molasses (P < .05). The fractional gas production rate (µ) was lower in S than in other treatments (P < .05). The highest ME was estimated for HSMU, followed by SMU and H, respectively, with the latter having comparable ME with lucerne hay. Untreated potato vine silage had the lowest ME.

Table 2. Gas production kinetic parameters of processed potato vines and lucerne hay (the first set of incubations).

Parameters	Treatments							SEM	P-value
	S	SM	SMU	HSM	HSMU	H	Lucerne
A (ml)	51.3^e	59.0^d	68.0^c	59.1^d	73.6^a	69.8^bc	71.8^ab	0.98	<.001
L (h)	0.58^a	0.52^a	0.33^b	0.28^b	0.18^c	0.01^e	0.01^e	0.026	<.001
T1/2(h)	11.3^a	9.2^b	8.8^c	9.4^b	8.3^d	5.5^e	5.5^e	0.11	<.001
µ (/h)	0.070^c	0.086^ab	0.088^a	0.080^b	0.090^a	0.083^ab	0.088^ab	0.0027	<.002
ME (MJ)	7.4^d	8.7^c	9.9^b	8.3^c	10.5^a	9.5^b	9.9^b	0.15	<.001

Notes: A, asymptote of gas production (per 200 mg DM); L, lag time; T_1/2, half time to asymptote; µ, fractional rate of gas production. S, untreated potato vine silage; SM, pre-wilted potato vines ensiled with 10% molasses; SMU, pre-wilted potato vines ensiled with 10% molasses plus 4% urea; HSM, pre-dried potato vines ensiled with 10% molasses; HSMU, pre-dried potato vines ensiled with 10% molasses plus 4% urea; H, potato vines hay.

SEM, standard error of the means. a,b: different superscript letters indicate differences (P < .05) among the treatments.

24-h Digestibility and fermentation variables

The second set of incubations

In vitro apparent and true dry matter degradabilities (IVADMD, IVTDMD) and IVTOMD differed among the treatments (P < .001). The potato vines hay had the lowest ruminal degradabilities followed by S (Table 3). Treatment of the silages with the additives improved their degradabilities, with the highest amounts obtained in the silages prepared with molasses and urea (SMU and HSMU), being 8.6% higher than S. The GP₂₄ was also affected by the treatments (P < .001), as in Exp. 1, H exhibited higher gas production than S, and among the silages, those prepared with molasses and urea (SMU and HSMU) produced higher GP₂₄ than S (P < .05). TVFA concentration was highest in SMU and HSMU (increased by 36.8% compared with S), followed by SM and HSM; this variable was higher in lucerne hay than in H and S (P < .05). Ammonia concentration was highest in the molasses plus urea-treated silages (SMU and HSMU), followed by the silages treated only with molasses (SM and HSM). Untreated potato vine silage and H had the lowest ammonia concentrations. There was also a significant difference among the treatments for PF (P < .001). Untreated potato vine silage had the highest PF among the silages, but the treatment of the silages with the additives decreased their PF. MP also differed among the treatments (P < .001), as it was lower in S than in H (P < .05). Supplementing the potato vine silages with the additives enhanced their MP production, with the highest amounts observed in the silages made with molasses plus urea (increased by 179% compared with S). The MP of lucerne hay was similar to that of SM and HSM.

Table 3. In vitro ruminal digestibility and fermentation variables of processed potato vines and lucerne hay (the second set of incubations).

Variables	Treatments							SEM	P-value
	S	SM	SMU	HSM	HSMU	H	Lucerne
IVADMD (%)	65.1^e	69.9^c	73.5^a	68.5^d	73.1^a	60.9^f	71.4^b	0.34	<.001
IVTDMD (%)	77.7^e	80.5^c	84.9^a	80.2^c	82.9^b	72.2^f	79.1^d	0.29	<.001
IVTOMD (%)	81.2^e	83.4^d	87.7^a	83.4^d	86.0^b	77.9^f	85.0^c	0.30	<.001
GP24 (ml)	58.9^f	67.7^e	80.2^c	73.8^d	83.1^c	88.1^b	92.7^a	1.46	<.001
TVFA (mM)	58.5^d	70.3^b	80.0^a	67.8^bc	77.2^a	56.0^d	64.4^c	1.12	<.001
NH3 (mM)	14.1^c	16.1^b	20.9^a	16.3^b	20.4^a	13.3^c	15.9^b	0.32	<.001
PF	5.80^a	5.09^b	4.43^cd	4.56^c	4.15^d	3.59^e	4.16^d	0.087	<.001
MP (mg)	19.9^e	34.3^c	55.6^a	33.1^c	50.7^b	26.1^d	32.7^c	1.12	<.001

Notes: IVADMD, in vitro apparent dry matter degradability; IVTDMD, in vitro true dry matter degradability; IVTOMD, in vitro true organic matter degradability; GP₂₄, gas produced after 24 h of incubation (/500 mg DM); TVFA, total VFA concentration; PF, partitioning factor (mg OM degraded/ml gas produced); MP, microbial protein; S, untreated potato vine silage; SM, pre-wilted potato vines ensiled with 10% molasses; SMU, pre-wilted potato vines ensiled with 10% molasses plus 4% urea; HSM, pre-dried potato vines ensiled with 10% molasses; HSMU, pre-dried potato vines ensiled with 10% molasses plus 4% urea; H, potato vines hay.

SEM, standard error of the means, a,b: different superscript letters indicate differences (P < .05) among the treatments.

The third set of incubations

In the last series of incubations, replacing lucerne hay in the basal ration with H did not affect the ruminal degradability of the ration, neither at 50% nor at 100% substitution levels (Table 4). However, the inclusion of SMU and HSMU improved IVTDMD and IVTOMD at both substitution levels compared with the CTRL (P < .05). Replacing lucerne hay with H decreased GP₂₄ at both substitution levels, while SMU and HSMU lowered GP₂₄ only at the high substitution level compared with the CTRL (P < .05). As for GP₂₄, TVFA decreased also when H substituted lucerne hay at both levels; in contrast, SMU and HSMU enhanced TVFA at the high substitution level compared with the CTRL (P < .05). Ammonia concentration increased when lucerne hay was replaced by SMU and HSMU at both 50 and 100% levels (P < .05). PF was higher in all the rations, in which lucerne was replaced totally by the processed potato vines compared to the CTRL (P < .05). The variation of MP was approximately the same as that of PF among the treatments, as it increased when lucerne hay was replaced by processed potato vines, especially at the 100% substitution level.

Table 4. In vitro ruminal digestibility and fermentation of a mixed ration containing processed potato vines substituted to lucerne (the third set of incubations).

Variables	Treatments							SEM	P-value
	CTRL	S50	S100	SMU50	SMU100	HSMU50	HSMU100
IVADMD (%)	67.7^c	67.1^cd	66.2^d	70.5^b	75.1^a	70.8^b	73.9^a	0.43	<.001
IVTDMD (%)	76.6^c	76.3^c	75.1^c	80.9^b	83.1^a	79.6^b	83.0^a	0.60	<.001
IVTOMD (%)	80.1^dc	81.0 ^c	79.0^d	84.6^b	87.9^a	84.2^b	87.5^a	0.55	<.001
GP24 (ml)	109.0^a	103.0^c	97.5^d	110.3^a	105.5^bc	109.2^a	106^b	0.87	<.001
TVFA (mM)	55.5^c	48.5^d	45.8^e	56.3^bc	58.8^a	56.2^bc	58.3^ab	0.73	<.001
NH3(mM)	19.4^c	19.4^c	18.8^c	20.6^b	22.8^a	21.1^b	23.7^a	0.33	<.001
PF	3.51^c	3.70^bc	3.85^ab	3.67^cd	3.94^a	3.64^cd	3.92^a	0.049	<.001
MP (mg)	26.5^c	28.7^bc	29.8^b	30.0^b	33.9^a	29.2^b	33.6^a	0.73	<.001

Notes: IVADMD, in vitro apparent dry matter degradability (%); IVTDMD, in vitro true dry matter degradability; IVTOMD, in vitro true organic matter degradability (%); GP₂₄, gas produced after 24 h of incubation (ml/500 mg DM); TVFA, total VFA concentration (mmol/l); NH3 (mmol/l); PF, partitioning factor (mg OM degraded/ml gas produced); MP, microbial protein (mg); CTRL, Control (basal ration); S₅₀, 50% replacement of lucerne in basal ration with untreated potato vine silage; S₁₀₀, 100% replacement of lucerne in basal ration with untreated potato vine silage; SMU₅₀, 50% replacement of lucerne in basal ration with potato vines ensiled with 10% molasses plus 4% urea; SMU₁₀₀, 100% replacement of lucerne in basal ration with potato vines ensiled with 10% molasses plus 4% urea; HSMU₅₀, 50% replacement of lucerne in basal ration with pre-dried potato vines ensiled with 10% molasses plus 4% urea; HSMU₁₀₀, 100% replacement of lucerne in basal ration with pre-dried potato vines ensiled with 10% molasses plus 4% urea.

SEM, standard error of the means. a,b: different superscript letters indicate differences (P < .05) among the treatments.

Discussion

Air-drying and ensiling are two conventional methods commonly used to conserve forages for long-term storage. Further, ensilage of potato vines has been found to reduce their glycoalkaloids content (Nicholson et al., 1978). Air-drying is a cost-effective way to conserve forages in arid and semi-arid regions, which can also be used in the case of potato vines with high moisture content. This practice may also reduce the glycoalkaloids content of potato vines. There is no published data in this regard in the literature; however, in an experiment conducted in our laboratory, air drying resulted in a decrease of 32% in α-solanine content of potato vines (Behzadi et al., Bu-Ali Sina University, unpublished data). This may be a result of the partial degradation of these substances by plant enzymes (Friedman 2006) early in the drying period. Regarding this argument, the silage made from pre-dried potato vines might have lower glycoalkaloids content compared with the silage made from fresh vines.

Chemical composition

In the current study, a higher ash content observed in potato vines compared with lucerne hay is likely mainly due to the soil contamination of potato vines in the farms. Likewise, Salehi et al. (2014) reported a lower OM content in potato vines compared with lucerne. The lower ash content of potato vine silages compared with their hay form seems to be a result of the loss of minerals in effluent (McDonald et al. 1991), although the molasses and urea included in additive-treated silages could also affect their OM content. When accounting for ash contamination, potato vines seem to have comparable CP with lucerne hay. These results are in line with those found by Salehi et al. (2014) and Parfitt et al. (1982) who reported a relatively comparable CP content for potato vines with lucerne. A relatively higher CP in S than H is probably a result of ash loss during ensiling in S; however, excess N provided by the additives (urea and molasses) has also enhanced CP content in additive-treated silages. It should also be noted that a rapid fermentation favoured by the additives, especially by molasses, might have shortened the fermentation period and restricted protein degradation, enhancing CP content in molasses-treated silages (Davies et al. 1998). A low cell wall content observed in potato vines compared with lucerne hay is also in agreement with previous published observations (Salehi et al. 2014). Partial degradation of cell walls during ensiling may be the origin of lower NDF and ADF contents in potato vine silages compared with that in their hay form. Consistent with our results, Nicholson et al. (1978) found that the ADF content of 6 cultivars from 12 studied cultivars of potato vines decreased during ensiling. This is most likely due to the partial degradation of cell wall by silage microorganisms as well as the hydrolysis of hemicellulose and cellulose (to a lesser extent) by organic acids produced during fermentation (Dewar et al. 1963; McDonald et al. 1991). However, in the forages with high water-soluble carbohydrates (WSCs) content, it is possible that a substantial fermentation of WSCs leads to their increased NDF and ADF contents (Anderson et al. 2009). A high pH observed in untreated potato vine silages may be indicative of the fact that potato vines content of WSCs is insufficient to make a silage of good quality. Nicholson et al. (1978) reported WSC contents of 3.7–8.6% for potato vines of different cultivars, which are less than the minimum recommended levels for ensiling forages of high buffering capacity such as legumes (being 10% and 14% for forages of 40% and 35% DM content, respectively) (Haigh & Parker 1985).

In general, legumes (including potatoes) are considered more difficult to ensile because of their lower WSC content and higher buffering capacity (caused by a relatively high CP content) compared with grasses (McDonald et al. 1991; McAllister & Hristov 2000). A high buffering capacity accompanied with a low WSC causes a gradual decline in pH during ensiling, resulting in a final high pH in legume silages (McDonald et al., 1991). One of the common effective approaches to overcome this problem is the use of additives with high soluble carbohydrates such as molasses. Supplementing legume silages with WSC sources provides readily fermentable carbohydrates for lactic acid bacteria to produce lactic acid, which prevents adverse microbial activity and restricts protein degradation (Heron et al. 1989; Davies et al. 1998). Thus, potato vines seem to need additives of high WSC to ensile properly.

Gas production kinetics

Several factors appear to affect the gas production kinetics in the current study, with the most important being the fermentable organic matter (FOM) content of the samples. A lower asymptotic gas production observed in S compared with H could be a result of its lower FOM content. This is due to the partial fermentation of OM to FAs during ensiling, which contribute slightly to the direct gas produced during ruminal fermentation, though part of silage FAs contributes indirectly to gas production by neutralising with bicarbonate buffer in the culture (Grings et al. 2005).

Glycoalkaloids and other secondary metabolites, including phenolic compounds, protease inhibitors and phytoalexins contained in the potato vines (Friedman 2006), may also affect gas production kinetics. Phenolic compounds are characterised by their antimicrobial nature (Puupponen-Pimia et al. 2001; Akyol et al. 2016), which may affect rumen microbial activity. Although ensiling potato vines has been found to substantially reduce their glycoalkaloid content (Nicholson et al. 1978), the remaining glycoalkaloid, accompanied with other secondary metabolites, might be toxic to some rumen microorganisms, resulting in low gas production in the silages. However, it seems that the inclusion of molasses and urea has improved gas production by compensating the negative impacts of glycoalkaloids and a low FOM in intact potato vine silages. This is likely mediated by supplying FOM to rumen microorganisms directly, or indirectly through efficient degradation of cell walls (Ohyama et al. 1975; Hassanat et al. 2007). In addition, it is probable that an efficient degradation of glycoalkaloids during ensiling in additive-treated silages has resulted in their higher asymptotic gas production compared with the untreated potato vine silage (S). There are also data indicating that supplementing fermentation substrates with nitrogen improves their ruminal fermentation in vitro because of generally insufficient N supply by rumen fluid to support fully the growth of rumen microorganisms (Dryhurst & Wood 1998; Mould et al. 2005). Thus, the higher asymptotic gas production observed in additive-treated silages seems to be a result of a better supply of nutrients (energy and N) to rumen microorganisms in these silages. Secondary metabolites accompanied with a low OM content are likely the principal causes of slightly lower asymptotic gas production observed in H compared with lucerne.

Lag phase represents the period when feed particles are hydrated and colonised by rumen microorganisms (Dehority 2003). The existence and length of the lag phase are affected by different factors, including the nature of substrate incubated, the amount of inoculum used and the microbial species present in the inoculum (France et al. 2005). Anti-nutritive factors may extend the lag phase by inhibiting rumen bacteria and slowing their attachment to feed particles (McAllister et al. 1994). Compared with H and lucerne hay, a longer lag phase observed in the silages may be related to their lower WSCs, or other readily fermentable nutrients being partially consumed during ensiling. The same argument could be explicative among the silages, where the additive-treated silages had a shorter lag phase than the silage made without additives. Indeed, using WSCs in preparing silages increases their residual WSCs directly, or indirectly by increasing the fermentation rate and thereby shortening the fermentation period (Gordon 1989).

Although a high proportion of cell wall to WSCs seems to be the principal cause of lower fractional gas production rate in the untreated silage, it is possible that a greater glycoalkaloid content or other gas production inhibitors, such as phenolic compounds in S, contribute to its lower gas production rate.

Overall, gas production profiles indicated that despite a lower OM content, potato vine hay has comparable energetic values with lucerne hay. In contrast, the potato vine silage made without additives had low ruminal fermentation indices. Thus, chemical composition and secondary metabolites seem to be two principal determinants of potato vine fermentation potential. In the current study, unfortunately, there was no possibility of determining the glycoalkaloids content in the potato vines; thus, the estimated kinetic parameters should be interpreted with caution. To achieve conclusive results on the nutritive values of potato vines, it would be necessary to discriminate the impacts of glycoalkaloids and phenolic compounds from those of chemical composition on gas production profiles.

24-h Digestibility and fermentation variables

The second set of incubations

Cell wall content and secondary metabolites seem to play an important role in the ruminal degradability and fermentation of the substrates in the present study. Part of lower DM and OM degradabilities observed in H, despite its lower cell wall content compared with lucerne hay, is likely due to its glycoalkaloids and other biologically active compounds such as protease inhibitor and phenolic compounds (Akyol et al. 2016). Among the silages, improvement of IVTDMD and IVTOMD in additive-treated silages, especially those made with molasses plus urea, could be a result of more efficient degradation of cell walls (McDonald et al. 1991; Yahaya et al. 2002), and glycoalkaloids during ensiling (Nicholson et al. 1978)

A higher OM content, as stated earlier, as well as a greater OM degradability in lucerne hay is presumably the origin of its higher GP₂₄ compared with H; however, it is possible that secondary metabolites contained in H affect adversely its GP₂₄. In the incubations of 24 h, the silages produced proportionally lower GP₂₄ than hays (H and lucerne hay). In addition to a low FOM content in the silages, described previously, a relatively low GP₂₄ in the silages is due partly to subtracting immediate gas production (the gas originated from FAs buffering in the silages) from GP₂₄. Among the silages, there was a good correlation between IVTOMD and GP₂₄; however, treatment source of variation was wider for GP₂₄ than IVTOMD. This may be attributed to the different extent of the negative impact of potato vine antimicrobial agents on ruminal digestibility and gas production. In this connection, it has been indicated that the negative impact of anti-nutritive factors such as secondary metabolites on ruminal degradability is less than that on gas production (Khazaal et al. 1993).

Despite producing lower GP₂₄, S had comparable TVFA with H, indicating that all FAs produced during ensiling in the silages have not been converted to CO₂. Indeed, all the VFAs in H-containing cultures have been produced during ruminal fermentation, which contribute to both direct and indirect production of the gas, whereas at least part of the VFAs in S-containing cultures have been produced previously during ensiling, which have a minor contribution to direct gas production. Moreover, there are indications that the fermentation of feeds with low cell wall content is associated with a relatively low gas production because of the shift in the VFA profile from acetate to propionate and lactate production, which is accompanied with a decrease in direct gas production (Russell & Wallace 1997; Mould et al. 2005).

PF, measured as the ratio of OM degraded to GP₂₄, is an index of partitioning digested OM between the fermentation pathway (producing gas and VFA) and the microbial biomass production pathway; thus, a higher PF denotes a higher proportion of digested OM incorporated into rumen microbial biomass than into the fermentation pathway (Blummel et al. 1997). Compared with H, the potato vine silages had higher PF because of proportionally lower gas production than the corresponding amount of OM disappeared. However, additive treatment of the silages reduced their PF by a greater increase in GP₂₄ than IVTOMD. These results reflect the higher sensitivity of gas production to anti-nutrients (secondary metabolites of potato vines) and fermentable nutrients (molasses and urea addition) than ruminal digestibility. This is supported by the findings of Getachew et al. (2000) who reported that the PF of tannin-rich feeds could be much higher than the theoretical range of PF (2.75–4.41). It has been due to the more pronounced inhibitory impact of tannins on gas production than OM degradability (part of previously dissolved OM is not fermented by inhibited microbes). Thus, in these cases, PF cannot be a reliable indicator of MP synthesis. Therefore, regarding the above-mentioned argument, MP was rather determined using a nitrogen balance approach (Grings et al. 2005), instead of being estimated from the amount of OM degraded and the volume of the gas produced, as proposed by Blummel (2000). The additive treatment of the silages improved substantially their ruminal MP synthesis. In general, the efficiency of ruminal MP synthesis is low in silage-based diets (Givens & Rulquin 2004). This is due, in part, to an imbalance between nitrogen and energy supplied to rumen bacteria (Chamberlain & Choung 1995; Dewhurst et al. 2000). These diets are characterised typically by a high rumen degradable protein and low fermentable energy, limiting MP synthesis. Therefore, supplementing silages with WSCs enhances their potential to support the efficiency of rumen MP synthesis (Miller et al. 2001; Lee et al. 2002). As mentioned earlier, treating silages with WSCs increases their residual WSCs; data in the literature indicate a strong linkage between residual WSC content in silages and the amount of rumen MP synthesis (Jaakkola et al. 2006), as well as rumen microbial N use efficiency (Merry et al. 2006), reducing N excretion in the environment. In fact, readily fermentable organic matter accelerates the growth of rumen microorganisms by increasing their efficiency of protein synthesis, which may result subsequently in higher digestion and fermentation of the substrate (Sampath et al. 1995). This may be a partial explanation for a higher MP, GP₂₄ and degradability in additive-treated silages compared with the untreated silage.

The third set of incubations

A higher ruminal degradability of DM and OM in the diets containing SMU and HSMU is due partly to a lower cell wall content in SMU and HSMU. In addition, a better supply of nutrients (including mainly N and fermentable carbohydrates) to rumen microorganisms, as a result of additives treatment (Aksu et al. 2006; Yitbarek & Tamir 2014) in SMU and HSMU, may also contribute to a higher digestibility of the diets containing these silages compared with the CTRL. In contrast, a low GP₂₄ in SMU₁₀₀ and HSMU₁₀₀ is due to the low gas production potential of SMU and HSMU, as stated earlier, which is a general feature of silages (Grings et al. 2005). Furthermore, a high PF and MP in the same diets are probably other causes lowering GP₂₄ in these diets, indicating a higher partitioning of digested OM to the microbial biomass production pathway than to that of gas and VFA production (Blummel 2000). Although a high PF and MP in the diets containing H could result in partly a lower GP₂₄ and TVFA compared with CTRL; however, secondary metabolites, including glycoalkaloids, might also contribute in lowering GP₂₄ and TVFA in these diets. In addition to a generally high non protein nitrogen content of silages (McDonald et al. 1991), urea treatment of the silages included in the diets has had a great contribution to a high ammonia concentration in SMU- and HSMU-containing diets. Thus, a high ammonia concentration, synchronised with readily fermentable carbohydrates (from residual WSCs), in the SMU- and HSMU-containing diets might have contributed to a higher MP in these diets (Dewhurst et al. 2000).

Conclusion

Based on the findings of this study, potato vines seem to have a great potential to be used as fodder in the feeding of ruminants. However, a relatively high ash content caused by soil contamination adversely affects their nutritive value, especially in the hay form. In addition, these findings showed that intact potato vines do not have the potential to be ensiled properly alone because of their low WSC content and a high buffering capacity. Hence, they should be ensiled with additives rich in WSCs such as molasses, although supplementing the potato vine silages with urea improved their fermentation capacity in this experiment, likely by providing a balanced supply of nutrients to silage-fermenting bacteria. Ruminal fermentation indices indicated that processed potato vines, especially additive-treated silages, can replace lucerne in ruminant diets. However, the effects of potato vine glycoalkaloids and their other secondary metabolites such as phenolic compounds on rumen microbes need to be clarified and the use of ensiled potato vines needs to be tested in vivo.

Disclosure statement

No potential conflict of interest was reported by the authors.