Nutrient composition and in vitro digestion parameters of Jerusalem artichoke ( Helianthus tuberosus L . ) herbage at different maturity stages in horse and ruminant

Jerusalem artichoke (Helianthus tuberosus L.) is a tuberous plant belonging to Asteraceae family and originating from North America that can be cultivated in various areas of the world (Gunnarsson et al., 2014). It is a 3–4 m long perennial plant. Its length, colour of tubers, numbers of branches and stems, and leaf ratio are dependent on variety, soil type and climatic conditions (Kays and Nottingham, 2007; Szpunar-Krok et al., 2016). It can be cultivated on various soil types (pH 4.5–8.2 and salinity) and different climatic conditions (in average 6.3–26.6 °C). In the leaves there are natural substances like β-bisabolene, ABSTRACT. The aim of the study was to determine the nutrient composition and in vitro digestion parameters of Jerusalem artichoke (Helianthus tuberosus L.) herbage at different maturity stages (vegetative, early flowering, full flowering and early seeding) in horses and ruminants. The crude protein (CP), ether extract (EE), ash, non-fibrous carbohydrate (NFC), neutral detergent fibre assayed with a heat stable amylase and expressed exclusive of residual ash (aNDFom), acid detergent fibre expressed exclusive of residual ash (ADFom), acid detergent lignin (ADL), total condensed tannins (TCT), total saponin (TSP) and carotenoids (lutein, zeaxanthin, lycopene and α-,β-,γ-carotenes) content in plant samples were analysed. The in vitro total gas and methane production, metabolizable energy (ME), true dry matter disappearance (T-DMd), true organic matter disappearance (T-OMd), gas yield (GY24), partial factor (PF24) and microbial crude protein production (MCP) values and volatile fatty acids (VFAs) concentration in fermentation fluid for herbage samples were determined with the in vitro digestion techniques carried out using horse faeces and ruminant fluids. The aNDFom, ADFom, ADL, EE and TCT content increased with plant maturation (P < 0.05). Zeaxanthin, lycopene and β-carotene content was the highest in full flowering and the lowest in early seeding plant (P < 0.05). With the plant maturation the following parameters decreased (P < 0.05): CP, NFC, ash and TSP content in plant samples, and in vitro total gas and methane production, T-DMd, T-OMd, ME values and molarities of individual VFAs for both horses and ruminants. Consequently, Jerusalem artichoke herbage, especially at vegetative stage, has the potential to be used as quality forage in terms of high/moderate nutrient composition and satisfactory digestion values for both horses and ruminants. Received: 25 November 2016 Revised: 19 June 2017 Accepted: 8 September 2017


Introduction
Jerusalem artichoke (Helianthus tuberosus L.) is a tuberous plant belonging to Asteraceae family and originating from North America that can be cultivated in various areas of the world (Gunnarsson et al., 2014).It is a 3-4 m long perennial plant.Its length, colour of tubers, numbers of branches and stems, and leaf ratio are dependent on variety, soil type and climatic conditions (Kays and Nottingham, 2007;Szpunar-Krok et al., 2016).It can be cultivated on various soil types (pH 4.5-8.2 and salinity) and different climatic conditions (in average 6.3-26.6 °C).In the leaves there are natural substances like β-bisabolene, ABSTRACT.The aim of the study was to determine the nutrient composition and in vitro digestion parameters of Jerusalem artichoke (Helianthus tuberosus L.) herbage at different maturity stages (vegetative, early flowering, full flowering and early seeding) in horses and ruminants.The crude protein (CP), ether extract (EE), ash, non-fibrous carbohydrate (NFC), neutral detergent fibre assayed with a heat stable amylase and expressed exclusive of residual ash (aNDFom), acid detergent fibre expressed exclusive of residual ash (ADFom), acid detergent lignin (ADL), total condensed tannins (TCT), total saponin (TSP) and carotenoids (lutein, zeaxanthin, lycopene and α-,β-,γ-carotenes) content in plant samples were analysed.The in vitro total gas and methane production, metabolizable energy (ME), true dry matter disappearance (T-DMd), true organic matter disappearance (T-OMd), gas yield (GY 24 ), partial factor (PF 24 ) and microbial crude protein production (MCP) values and volatile fatty acids (VFAs) concentration in fermentation fluid for herbage samples were determined with the in vitro digestion techniques carried out using horse faeces and ruminant fluids.The aNDFom, ADFom, ADL, EE and TCT content increased with plant maturation (P < 0.05).Zeaxanthin, lycopene and β-carotene content was the highest in full flowering and the lowest in early seeding plant (P < 0.05).With the plant maturation the following parameters decreased (P < 0.05): CP, NFC, ash and TSP content in plant samples, and in vitro total gas and methane production, T-DMd, T-OMd, ME values and molarities of individual VFAs for both horses and ruminants.Consequently, Jerusalem artichoke herbage, especially at vegetative stage, has the potential to be used as quality forage in terms of high/moderate nutrient composition and satisfactory digestion values for both horses and ruminants.
Jerusalem artichoke has a number of advantages: low input cultivation, high crop yield and strong resistance to pests and plant diseases (Yang et al., 2015).It was demonstrated that some bioactive compounds (such as sesquiterpene lactones) present in its leaves have antimicrobial effects.The Jerusalem artichoke tuber or tuber meal is commonly used in animal and human nutrition as inulin source (Helmi et al., 2014).Its areal parts (leaf, stem) were used in domestic animal nutrition in the previous century (Konopiński and Bormann, 1937;Becker and Nehring, 1969), and forgotten for many years now are again investigated in herbivore nutrition (Razmkhah et al., 2017).It is known that areal parts of Jerusalem artichoke include 35-37% neutral detergent fibre assayed with a heat stable amylase and expressed exclusive of residual ash (aNDFom), 24-25% acid detergent fibre expressed exclusive of residual ash (ADFom), 37-41% nonfibrous carbohydrate (NFC) and about 18% crude protein (CP) (Stauffer et al., 1980;Karsli and Bingöl, 2009); in ruminants the in vitro organic matter digestibility (OMD) and metabolizable energy (ME) value of these green mass ranged from 59 to 63% and about 9.6 MJ • kg −1 , respectively (Karsli and Bingöl, 2009).These features make the Jerusalem artichoke a good forage for herbivores perfectly digesting fibre like ruminants and horses.However, it contains antinutritional components as saponins and tannins which at high concentration can cause some negative effects on animals.
Grazing herbivores are characterized by a great variety of mechanisms and anatomical formations that use the chemical energy locked up in the plant structural carbohydrates (aNDFom, ADFom).Some parts of the gastrointestinal tract in these animals are bigger than in others in order to accommodate the fermented forage digested by microorganisms producing volatile fatty acids (VFAs) and lactate (Allen, 1997;Frape, 2004).The most important sections were fibre, i.e. cellulose, hemicellulose and lignin, is digested are caecum and ventral colon (in horses) and rumen (in ruminants) (NRC, 2001;Frape, 2004).Dietary fibre stimulates chewing activity and saliva flow to the rumen in ruminants.Bicarbonate and phosphate buffers in saliva neutralize acidity increased by fermentation of organic matter in the rumen.Inadequacy of fibre or forage in ruminant diet may decrease the chewing movements, ruminal pH, fibre digestibility, microbial yield, and lead to metabolic disorders like acidosis, laminitis, displaced abomasum (Shaver, 1997;NRC, 2001).Similarly, fibrous feeds in horse diet are important for intestinal peristalsis, chewing movements, hindgut volume and fluid/electrolyte retention, and they prevent some metabolic upsetsglycaemic disorders, metabolic acidosis and laminitis (Frape, 2004).Besides the fact that such forages have beneficial effects on digestion in horses and ruminants, they may be also a source of protein, soluble carbohydrate, energy and carotenoids.
Nowadays, the countries in the southern parts of Europe and in the Eastern Mediterranean region cope with changing climate conditions.The increased temperature makes the production of forage plants more difficult.Therefore, studies on the alternative forage sources, which are suitable for arid and semi-arid climatic conditions, have been intensified (Kara et al., 2015(Kara et al., , 2016)).
The aim of the present study was to determine the nutrient composition and in vitro digestion values of Jerusalem artichoke (Helianthus tuberosus L.) herbage in two herbivores -horses and ruminants whose diet is characterised by high forage:concentrate ratio.In the study, Jerusalem artichoke herbage was harvested at different maturity stages in Central Anatolia (Turkey) where arid climatic conditions are dominant.Thus, it was hypothesized that Jerusalem artichoke herbage can be used as an alternative forage source.

Samples collection area
Plant samples were collected in Karaman, Central Anatolia region (Turkey), 1038 m a.s.l.(36°33'50"N, 32°56'52"E) where steppe and dry forests dominate.The mean January temperature is 0 °C and in July -above 20 °C.The mean annual rainfall amount is below 400 mm.Rain type is convectional and frontal.These features are typical for arid conditions (Altın et al., 2012).
The carotenoids (α-carotene, β-carotene, γ-carotene, lutein, lycopene and zeaxanthin) were extracted from herbage samples and determined according to the method of Kara and Baytok (2017) in cool and sunlight free laboratory conditions.About 2.0 g of dried plant sample was weighed into a 15-ml conic centrifuge tube (Isolab, Wertheim, Germany).The methanol (5 ml) and 30% of methanolic potassium hydroxide (1 ml) were added into tube and mixed using a vortex at 2000 rpm for 30 min (Velp Vortex ZXClassic, Velp Scientifica, Usmate, Italy), and then incubated in thermostatically controlled cabinet at 4 °C (Lovibond, Dortmund, Germany).After incubation, tubes were centrifuged in a cooling centrifuge (Nüve NF800R, Nüve, Ankara, Turkey) at 4000 rpm and at 4 °C for 5 min.Supernatant was decanted into a 50-ml conic centrifuge tube (LP Italiana SPA, Milano, Italy).Extraction procedure was repeated twice with 8 ml of a solvent (hexane, ethanol or petroleum ether), and extracts were pooled.On top of pure and separate extracts, 25 ml of saturated aqueous sodium chloride was added and the mixture shaken.Pooled extracts were read exactly for volume determination.The absorbance values of extracts were measured using a UviLine 9100 spectrophotometer (SI Analytics -Xylem Analytics Germany Sales GmbH & Co. KG, Mainz, Germany).The α-carotene content was determined at 445 nm and 2710 A 1% 1cm using hexane as solvent.The β-carotene content was estimated at 450 nm and 2620 A 1% 1cm using ethanol as solvent.The γ-carotene content was determined at 462 nm and 2760 A 1% 1cm using hexane as solvent.The lycopene content was determined at 470 nm and 3450 A 1% 1cm using petroleum ether as solvent.The lutein content was estimated at 445 nm and 2550 A 1% 1cm using ethanol as solvent.The zeaxanthin content was determined at 449 nm and 2348 A 1% 1cm using petroleum ether as solvent (Rodriguez-Amaya, 1999).
Total saponin (TSP) content of herbage samples was determined according to the modified technique of Vador et al. (2012).The approximately 50 mg of milled herbage samples were extracted using a 5 ml of 99.9% methanol.The TSP content determination was carried out using p-anisaldehyde reagent.The p-anisaldehyde reagent was prepared in 1:1 of distilled water:absolute methanol.The 2 ml of sample extract was mixed with 500 μl of 0.5% p-anisealdehyde reagent at 2000 rpm for 2 min using a vortex (Velp Mixer Classic, Velp Scientifica, Usmate, Italy), and kept aside for 10 min at room conditions.The 2 ml of 50% sulphuric acid reagent was added

In vitro digestion technique for ruminants
Fresh rumen fluid was used as inoculum.Rumen fluid was obtained from two beef cattle (Simmental 16 months of age and about 600 kg body weight) fed diet containing 80% concentrated mixture feed and 20% forage in dry matter (DM) basis (3.78 kg flaked maize grain, 3.60 kg barley grain, 1.30 kg maize silage, 0.63 kg wheat straw, 30 g salt and 20 g vitaminmineral premix as DM for a day; in total 9.36 kg DM/ cattle/day) applied an intensive fattening.An approximately 1 l of rumen fluid was received using a stomach tube after the morning feeding.It was collected into a thermos under CO 2 gas, and filtered with four layers of cheesecloth in the laboratory.The technique was carried out according to the procedures of Menke and Steingass (1988).One litre of buffer mixture included 474 ml of bi-distilled water, 237.33 ml of macro-mineral solution (5.7 g Na 2 HPO 4 , 6.2 g KH 2 PO 4 and 0.6 g MgSO 4 in 1 l bi-distilled water), 237.33 ml buffer solution (35 g NaHCO 3 and 4 g NH 4 HCO 3 in 1 l bidistilled water), 0.12 ml trace mineral solution (13.2 g CaCl 2 *2H 2 O, 10 g MnCI 2 *4H 2 O, 1 g CoCl 2 *6H 2 O and 0.8 g FeCl 3 -*6H 2 O in 100 ml bi-distilled water), 1.22 ml resazurin solution (0.1 g resazurin in 100 ml bi-distilled water) and 50 ml reducing solution (285 mg Na 2 S*7H 2 O and 4 ml 1 N NaOH in 96 ml bi-distilled water).Dried plant samples (200 ± 10 mg) milled through a 1 mm sieve were incubated in rumen fluid and buffer mixture in 100 ml glass syringes (Fortuna ® , Poulten & Graf Ltd., Wertheim, Germany) (n = 6; three syringes for cumulative gas production and three syringes for dry matter-organic matter loss).Thirty millilitres of the rumen fluid plus buffer mixture at a 1:2 (v/v) ratio was added into each syringe.In addition, three blank syringes (no template; rumen fluid plus buffer mixture) were incubated to calculate the total gas production.After closing the clips on the inlet silicon tube of the syringe, the syringes were manually shaken and the clips were opened to remove gas by pushing the piston upwards to achieve complete gas removal.The clips were closed, the initial volume recorded and the syringes were incubated in a water bath at 39 °C for up to 96 h.

In vitro digestion technique for horses
Fresh faeces samples were used as inoculum and were obtained from two thoroughbred horses (6-7 years of age, 480-500 kg body weight) fed diet containing 96% forage and 4% concentrate, as DM basis (5.4 kg meadow hay, 2.25 kg wheat straw, 0.9 kg lucerne herbage, 0.36 kg crushed oat grain, 20 g salt and 10 g vitamin-mineral premix; totally 8.94 kg DM/ horse/day).Faeces samples were collected immediately after defecation and transferred into a thermos containing water at 39 °C under CO 2 gas and transferred to the laboratory within 1 h.Faeces samples were diluted at a 1:10 ratio with 0.9% sterile serum physiologic solution (Polifleks, Polifarma, İstanbul, Turkey) using a laboratory type blender (Waring Products Division, Torrington, CT, USA).Diluted faeces inoculum was filtered through four layers of cheesecloth under constant CO 2 gas (anaerobically) and used in the in vitro digestion technique.

Determination of total gas and methane production
After measuring the total gas volume at 24 h, the tubing of the plastic syringe outlet was inserted into the inlet of the methane analyser (Sensor Europe GmbH, Erkrath, Germany) and the piston was pushed to insert the accumulated gas into the analyser (Kara et al., 2015).

Determination of volatile fatty acids (VFAs), in vitro true dry matter disappearance and in vitro true organic matter disappearance
Three of the fermentation syringes for both ruminants and horses were stopped after 24 h, and then the VFAs concentrations in fermentation fluids and in vitro true dry matter disappearance (T-DMd) and the in vitro true organic matter disappearance (T-OMd) were analysed.
The in vitro dry matter and organic matter disappearance was determined by filtering the fermentation residues using a vacuum unit (Velp Dietary Fibre Analyzer, Velp Scientifica, Usmate, Italy) on pre-weighed glass crucibles (porosity #2, Velp Scientifica, Usmate, Italy) dried at 105 °C and burning the residual at 550 °C.
In vitro T-DMd was calculated as: OM)] × 100.After measuring the total gas volume at 24 h of in vitro incubation, approximately 10 ml digestion fluid in each syringe was collected in Falcon tubes.The digestion fluid was filtered through four layers of cheesecloth, and 2 ml digestion fluid was mixed with 0.5 ml 25% (w/v) meta-phosphoric acid and kept frozen (−20 °C) for the analysis of VFAs in microcentrifuge tubes.The frozen samples were thawed at 4 °C and centrifuged at 15 000 g for 15 min using a microcentrifuge (Gyrozen 1524, Daejeon, South Korea).After, the supernatants were filtered using a syringe-driven filter unit with a 0.22 mm pore diameter (Millex Filter Unit, Merck Millipore Ltd., Tullagreen, Ireland), the filtrates were transferred into the vials (Chromacol, Thermo Fisher Scientific, Orlando, FL, USA).Analysis of the VFAs was carried out using a gas chromatograph (TRACE TM 1300, Thermo Fisher Scientific, Orlando, FL, USA) equipped with an auto sampler (AI 1310, Thermo Scientific, Orlando, FL, USA), a polyethylene glycol column (length: 60 m, i.d: 0.25 mm, film thickness: 0.25 µm) (TG-WAXMS, Thermo Scientific, Orlando, FL, USA) and a flame ionization detector (FID).The carrier gas was helium at a constant flow rate of 1.5 ml • min −1 .The injection volume was 0.5 µl.The samples were injected with split mode.The injection port temperature was 280 °C.Oven temperature was programmed to increase from 160 °C to 180 °C at a rate 20 °C • min −1 .Air flow was 350 ml • min −1 , and hydrogen flow was 35 ml • min −1 .The temperature of FID detector was 300 °C.Oven run time was 10 min.The concentrations of VFAs expressed as mmol • l −1 were identified using a Xcalibur software programme (Thermo Scientific, Orlando, FL, USA).The percentage of VFA [acetic (A), butyric (B) and propionic (P) acids], and A/P and (A+B)/P ratios were calculated.

Statistical analysis
Firstly, the data were subjected to Levene's test to detect the variance homogeneity.One-way variance analyses (ANOVA) were implemented for homogeneous variances by General Linear Model procedures to test treatment differences.Data was analysed based on the statistical model: Yij = µij +Si + ei where: Yij -general mean common for each parameter under investigation, Si -i th effect of maturity stages of Jerusalem artichoke herbage on the observed parameters, ei -standard error term.The means were separated by Tukey's multiple range test at P < 0.05.
Linear relations among the maturity stage and chemical composition with gas production values and some estimated parameters were determined using Pearson's correlation coefficients (r).

Nutrient composition of Jerusalem artichoke herbage harvested at different maturity stages
The CP content of Jerusalem artichoke herbage was 16.36% at the vegetative stage; otherwise this value decreased to 6.59% with plant maturation (P < 0.001).The ash content was different according to maturity stages: 13.64% at the vegetative, 11.73% at the early flowering, 11.16% at full flowering and 8.9% at early seeding (P = 0.007).The EE content of Jerusalem artichoke herbage ranged from 0.94% to 2.19% (P < 0.001).The aNDFom (P = 0.019), ADFom (P = 0.005) and ADL (P = 0.024) increased with plant maturation and NFC content decreased (P = 0.027).The TCT content (0.47%) at the early seeding stage was higher than (0.35%) at the vegetative stage.The TSP content ranged from 1.66 to 0.43 mg • g −1 DM, and decreased with plant maturation (P < 0.001).
The individual carotenoids composition of herbage was the highest in the full flowering stage (P < 0.001).The β-carotene, α-carotene, γ-carotene, lycopene, zeaxanthin and total carotenoid content in the examined herbage increased from the vegetative stage up to the full flowering stage, and then decreased at the early seeding stage (P < 0.001).The lutein content was higher than the other carotenoids except early flowering maturity stage.In addition, the lowest concentration of carotenoids was determined for β-carotene at the early seeding stage (Table 3).

In vitro digestion parameters of Jerusalem artichoke herbage in horses
In horses the in vitro cumulative gas production of Jerusalem artichoke herbage at vegetative stage .001SD -standard deviation; SEM -standard error of the means; CP -crude protein; aNDFom -neutral detergent fibre assayed with a heat stable amylase and expressed exclusive of residual ash; ADFom -acid detergent fibre expressed exclusive of residual ash; ADL -acid detergent lignin; EE -diethyl ether extract; NFC -non-fibrous carbohydrate; TCT -total condensed tannin; TSP -total saponin; TCar -total carotenoid = α-carotene + β-carotene + γ-carotene + lutein + lycopene + zeaxanthin; a-d -means with different superscripts within the same row are significantly different at P < 0.05 (Tukey's test) was higher than those of other maturity stages (especially at 24 h).Generally, cumulative gas production of the early flowering stage was similar to that of the full flowering stage.The in vitro cumulative gas production at the early seeding stage was lower than those of other stages up to 24 h of incubation (P < 0.001) (Table 4).
The in vitro methane production of Jerusalem artichoke herbage was found in a wide range 0.22-0.33ml • g −1 DM at 24 h for different maturity stages (P = 0.001).The in vitro T-DMd, T-OMd and ME values of Jerusalem artichoke herbage for horses were the highest at the vegetative stage and decreased with plant maturation (P < 0.001).In addition, the in vitro GY 24 (P = 0.116), and PF 24 (P = 0.090) values did not differ between maturity stages.The in vitro MCP value in vegetative stage was higher than those of other stages (P = 0.001; Table 5).DM -dry matter; methane -in vitro methane production as ml • g −1 DM at 24 h (methane production, ml = (GP 24 , ml × methane as percent in GP 24 )/100); GP 24 -volume (ml) of total gas produced by g DM at 24 h (ml • g −1 DM); T-DMd -in vitro true dry matter disappearance (mg) for g DM at 24 h (mg • g −1 DM); T-OMd -in vitro true organic matter disappearance (mg); GY 24 -gas yield is total gas volume (ml) produced for g T-DMd at 24 h (ml • g −1 DM); PF 24 -partial factor is ratio T-DMd to GP 24 at 24 h (mg • ml −1 GP 24 ); ME -metabolizable energy as MJ • kg −1 DM; MCPmicrobial crude protein produced at 24 h (mg • g −1 DM); SEM -standard error of the means; SD -standard deviation; abc -means with different superscripts within the same column for ruminants and horses separately are different at P < 0.05 (Tukey's test) The molar concentration (mmol • l −1 ) of acetic acid (P = 0.019), propionic acid (P = 0.019), butyric acid (P = 0.007) and sum of VFAs (P = 0.015) in the horse fermentation fluid decreased with plant maturation (Table 6).

In vitro digestion parameters of Jerusalem artichoke herbage in ruminants
In ruminants, the in vitro cumulative gas production of Jerusalem artichoke herbage decreased with plant maturation up to 96 h (P < 0.05).For cumulative gas production, the difference between the vegetative and the early seed stages was significant (P < 0.05).
The in vitro ruminal methane production at 24 h at the vegetative and early flowering stages was higher than that of other maturity stages (P = 0.003).The in vitro T-DMd (P = 0.014), T-OMd (P = 0.015) and ME (P = 0.001) values decreased with plant maturation but GY 24 (P = 0.671) and PF 24 (P = 0.690) did not differ among the maturity stages.The in vitro MCP values in early and full flowering stages were higher than those of vegetative and early seeding stages (P = 0.004) (Table 5).

Discussion
Chemical composition of Jerusalem artichoke herbage.According to the obtained results, Jerusalem artichoke herbage can be described as forage with high or moderate protein content at the vegetative stage (NRC, 2001(NRC, , 2007)).In line with the present study, Stauffer et al. (1980) reported that the CP content in Jerusalem artichoke herbage was reduced from 18.0 to 6.2% depending on the progress of maturity stage.Besides, in the previous study it was found that the aerial biomass of Jerusalem artichoke at harvest time (September) contains 2.1-6.1% protein in DM (Gunnarsson et al., 2014).The CP content in lucerne herbage is 22.2% at the vegetative stage, 19.3% at the full flowering stage and 18.7% at the seeding stage according to NRC (2007).In the previous study, in which alterna-tive forage for arid lands was investigated, it was showed that the forage qualities of Atriplex patula and Plantago lanceolata herbages were similar to lucerne herbage; and CP content of these alternative forages at the early flowering stages was 14.64 and 10.74%, respectively (Kara et al., 2016).In addition, CP content of Italian grass was 17.9% at the early vegetative stage and 10.3% at the end of the vegetative stage (NRC, 2007).The fact that the CP content of Jerusalem artichoke herbage at vegetative stage was lower about 60% from that at early seed stage may be related to the high plant cell wall component with plant maturation.
The structural carbohydrate content (aNDFom, ADFom and lignin) of Jerusalem artichoke herbage was increasing along with plant maturation which involves increased lignifications and decreased proportion of leaves to stems (Van Soest et al., 1991),  , 2007).Gunnarsson et al. (2014) determined that cellulose content in aerial parts of Jerusalem artichoke at harvest time for tuber (September) was similar with that obtained in the present study, but lignin content varied from 2.1 to 6.1% in DM.In addition, the aNDFom and ADFom contents of Jerusalem artichoke herbage at the full flowering and early seeding stages were parallel to those of lucerne herbage at the same maturity stages (NRC, 2001(NRC, , 2007;;Kara et al., 2016).In addition, the contents of fibre substances in Jerusalem artichoke herbage obtained in this study are in agreement with the findings of Karsli and Bingöl (2009).The TCT content of Jerusalem artichoke herbage was increasing along with plant maturation, which may be related to increasing aNDFom and ADFom contents.According to current results, the TCT content of Jerusalem artichoke herbage was positively correlated with plant cell wall components (fibrous components).In the present study, ash content increased with plant maturation, which was in relation with results of Gunnarsson et al. (2014).The NFC contents of Jerusalem artichoke herbage were high for all maturity stages.These soluble carbohydrates were at recommended by NRC (2007) concentrations for the forage or diet of dairy and beef cattle.
The saponins are glycosides that occur in plants and are characterized by the ability to produce a soapy lather.They are derivatives of secondary plant metabolism, related to the plant defence system; they are either triterpenoid or steroid glycoside compounds in some saponin-containing plants, mainly legumes.The saponin-rich plants (>4% in DM) can be potentially used as feed additives in diets for ruminants to control ammonia and odour production.This is possible due to the effects of saponins on nitrogen metabolism (Cheeke, 2000).Feeding domestic animals forage with high levels of saponin can lead to toxicity.The saponin toxicity leads to photosensitization followed by liver and kidney degeneration in ruminants as well as gut problems like gastroenteritis and diarrhoea (Wina et al., 2005).The total saponin content of Jerusalem artichoke herbage decreased with plant maturation and was similar to the results of Lima et al. (2013) and was lower than those of saponin-rich plants, e.g., Quillaja saponaria (Rosaceae), Glycyrrhiza species (Leguminosae) and Yucca schidigera (Agavaceae) (Gracindo et al., 2014).
Despite the large variety of carotenoids in plants, there have been found approximately 10 types important for forages; the most important are lutein and β-carotene (Nozière et al., 2006).The lutein content of Jerusalem artichoke in the current study was slightly lower than that indicated for different forage by previous researchers (Calderón et al., 2006).The zeaxanthin content in the present study was similar with the values indicated for natural mountain grassland (in mid-June) (Calderón et al., 2006).In the present study, the β-carotene concentrations of Jerusalem artichoke herbage were satisfactory for vegetative, early flowering and full flowering stages, but its concentration was low at early seeding stage (Calderón et al., 2006).The high total carotenoid content of Jerusalem artichoke herbage at full flowering may be related to the yellow coloured flowers of this plant.In the present study, the lutein and β-carotene concentrations of Jerusalem artichoke herbage were higher (two or three times) than other carotenoids until the end of the full flowering stage.Reynoso et al. (2004) reported that the lutein and β-carotene concentrations of green forages are two-to three-fold higher than the same species in humid versus dry tropical areas.In the current experiment, the low β-carotene content in herbage at the early seeding stage may be associated with plant maturation, rapidly oxidized by light and the changing of leaf:stem ratio (Reynoso et al., 2004;Kane, 2009).
In vitro digestion parameters of Jerusalem artichoke herbage.In ruminants and horses, in vitro cumulative gas produced by Jerusalem artichoke herbage decreased along with plant maturation.The total gas production of Jerusalem artichoke herbage at the vegetative stage was adequate to obtain good quality forage.The total gas (155-220 ml • g −1 DM) produced by Jerusalem artichoke herbage in ruminants was higher than that (110-173 ml • g −1 DM) in horses at 24 h.Frape (2004) stated that caecal microorganisms in horses tend to be less efficient in digestion of hay than ruminal microbes in ruminants.The fibre digestion difference has been attributed to the effects of differences in cellulolytic microbial species (Frape, 2004).However, total gas production of Jerusalem artichoke herbage at the early seeding stage was lower about 30% than that of vegetative stage in ruminants for 24 h incubation.This difference may be due to the in vitro methods used for different animal species and the digestive fluids.Generally, the in vitro gas production values of feed substances are connected with soluble carbohydrates (NFC) content (Kara et al., 2015(Kara et al., , 2016;;Kara, 2016).
In both ruminants and horses, the T-DMd, T-OMd and ME values of Jerusalem artichoke herbage and the molar concentration of organic acids in fermentation fluid were decreasing with plant maturation, and these values may be related with changes in nutrient composition (NFC, CP, lignocellulose).For both animal species these digestion parameters were the highest at the vegetative stage, and the lowest at the early seeding stage.It is known that these in vitro digestion parameters in animal model are negatively affect by the increasing structural carbohydrates content and the rate of structural components (i.e.cellulose, hemicellulose and lignin) of aNDFom in the plant cells (Menke and Steingass, 1988;Kara et al., 2015).Besides, digestion rates of feed substances by different animal species are not the same due to enzymatic, microbial diversity, microorganisms number/count differences of their digestive tracts.This status may affect contents of inoculum used by the in vitro digestion studies.Differences of in vitro digestion rate in horses and ruminants may be due to the reasons mentioned above (Allen, 1997;NRC, 2001;Frape, 2004).In the present study, the structural carbohydrate contents of Jerusalem artichoke increased with maturation; and then these changes of structural carbohydrates could negatively affect digestion parameters.The fibrolytic enzymes being a part of complicated lingocellulose in plant cell wall can reduce the effectiveness of cellulose digestion.In addition, the ADL content was negatively correlated with in vitro gas production which is an important indicator of low digestion value in horses (Kara, 2016).In horse, in vitro gas production was positively correlated with T-DMd, T-OMd and ME values which was in agreement with previous in vitro digestion studies (Elghandour et al., 2015;Kara et al., 2015Kara et al., , 2016;;Kholif et al., 2016).The estimated digestion values and ME values of Jerusalem artichoke herbage were satisfactory for horses forage according to NRC (2007).
In the current study, in vitro methane produced during Jerusalem artichoke herbage digestion in ruminants was at the expected level for forage (Kara et al., 2016).Methane production decreased with plant maturation, and this decrease may be related to the reduction in soluble carbohydrates composition and plant digestibility (Hook et al., 2010;Kara et al., 2015).The in vitro methane production in horses of this herbage harvested at different maturity stages was low or similar with the findings of Kholif et al. (2016) and Kara and Baytok (2017).These values demonstrated that the digestive tract of horses does not produce as much methane as that of ruminants.
The microbial profile of the horse caecum and right ventral colon and the rumen in ruminants are related to the diet composition (Julliand et al., 2001).The VFAs are one of the fermentation endproducts of feedstuffs in rumen and caecum.The amount of forage and concentrate fed also affects the VFA composition in the hindgut (horses) and the rumen (ruminants), with higher acetate concentrations of forage diet and higher propionate concentrations of high starch diet (Julliand et al., 2001;Kara et al., 2016).Even though the concentration of VFAs highly differs among diets, it generally ranges from 60 to 120 mmol • l −1 of rumen fluid (Belanche et al., 2015;Kara et al., 2017).Murray et al. (2010) stated that the molar proportions of VFAs in the in vitro fermentation fluid with equine faecal inoculum changed from 71 to 90 mmol • l −1 due to lucerne:sugar beet pulp ratio in diet.According to the results of the present study, the molar concentrations of the sum of VFAs (90.49-81.19mmol • l −1 vs 91.13-87.47mmol • l −1 ) were similar for in vitro fermentation with faeces (for horses) and rumen inoculum (for ruminants).These results were similar with the in vitro study of Macheboeuf et al. (1997).The molar concentration of acetic, propionic and butyric acids and sum of VFAs in the in vitro fermentation with faeces and rumen inoculum were positively correlated with in vitro digestion values such as gas production, T-DMd and T-OMd.This is an indication that concentrations of organic acids change depending on the digestibility of the plant (Menke and Steingass, 1988;Kara et al., 2016).In the present study, the individual molar concentration of acetic, propionic and butyric acids for the in vitro fermentation with faeces and rumen inoculum were lower with increasing plant maturity stage, which may be associated with low NFC, high lignin and lignocellulose complexes content.

Conclusions
Jerusalem artichoke (Helianthus tuberosus L.) herbage, especially at the vegetative stage, has potential for use as good quality forage in terms of high/moderate nutrient composition and satisfactory digestion values in both horses and ruminants.
Nevertheless, the digestion values and the effect on performance of the aerial parts of Jerusalem artichoke for ruminants and horses should be investigated in more detail in in vivo studies.In addition, it is necessary to further investigate the Jerusalem artichoke during all its maturity stages in order to produce the forage of good quality.

Table 1 .
Maturity stages of Jerusalem artichoke herbage

Table 2 .
Composition of in vitro fermentation medium

Table 3 .
Nutrient composition of Jerusalem artichoke herbage harvested at different maturity stages

Table 5 .
In vitro digestion values of Jerusalem artichoke herbage at different maturity stages in horses and ruminants

Table 4 .
In vitro cumulative total gas production (GP) of Jerusalem artichoke herbage at different maturity stages in horses and ruminants

Table 6 .
Effect of Jerusalem artichoke herbage on organic acid composition in digestion fluid in horses and ruminants in vitro

Table 8 .
Correlation coefficient (r) relationship of chemical composition and some in vitro digestion parameters of Jerusalem artichoke herbage for ruminants

Table 7 .
Correlation coefficient (r) relationship of chemical composition and some in vitro digestion parameters of Jerusalem artichoke herbage for horses Maturation -plant maturation of Jerusalem artichoke herbage; GP 24 , ME, T-DMd, T-OMd -see Table5; B -butyric acid; P -propionic acid; A -acetic acid; aNDFom, ADFom, ADL, NFC, TCT -see Table3; ** -P < 0.01; * -P < 0.05 but NFC content was decreasing.The aNDFom and ADFom content of Jerusalem artichoke herbage at the vegetative stage were similar to the values reported for lucerne herbage at the vegetative stage (NRC