Effects of formic acid and corn flour supplementation of banana pseudostem silages on nutritional quality of silage,growth,digestion,rumen fermentation and cellulolytic bacterial community of Nubian black goats

ZhaNg hao,ChENg Xuan,Mabrouk ELSaBagh,LiN Bo,WaNg hong-rong

1 Laboratory of Metabolic Manipulation of Herbivorous Animal Nutrition,College of Animal Science and Technology,Yangzhou University,Yangzhou 225009,P.R.China

2 Joint International Research Laboratory of Agriculture and Agri-Product Safety,Ministry of Education/Yangzhou University,Yangzhou 225009,P.R.China

3 Department of Animal Production and Technologies,Faculty of Agricultural Sciences and Technologies,Niğde Ömer Halisdemir University,Niğde 51240,Turkey

4 Department of Nutrition and Clinical Nutrition,Faculty of Veterinary Medicine,Kafrelsheikh University,Kafr El-Sheikh 33516,Egypt

5 College of Animal Science,Guangxi University,Nanning 530004,P.R.China

abstract The objective of this study was to evaluate the effects of adding formic acid and corn flour supplementation to banana pseudostem silages on the nutritional quality of these silages,growth,digestion,rumen fermentation and cellulolytic bacterial community of Nubian black goats fed these silages. Banana pseudostem silage was prepared either conventionally without any additives (CON) or mixed with 0.6% formic acid (F),10% corn flour (C),or both (F+C). Four experimental diets containing 40% of the corresponding silages were designed with roughage to concentrate ratio of 50:50 (dry matter (DM) basis). A total of 48 Nubian black castrated goats (body weight (BW),(22.64±1.82) kg;4-mon-old) were randomized into one of the four treatment groups with 12 replicates of one castrated goat per replicate for each treatment in a completely randomized design. Each group was fed on one of the four experimental diets for 40 days. A factorial arrangement of treatments of 2 (formic acid levels:0 and 0.6%)×2 (corn flour:0 and 10%) was adopted. Formic acid supplementation increased (P＜0.05) average daily gain,as well as lactic acid,propionate and butyric acid and water-soluble carbohydrate concentrations,but decreased (P＜0.05) the feed conversion rate,pH value,acetate/propionate ratio,and butyric acid concentration relative to the CON group. Corn flour supplementation increased (P＜0.05) the apparent digestibility of crude protein,neutral detergent fiber,and non-fibrous carbohydrate and Fibrobacter succinogenes,Ruminococcus albus,and Butyrivibrio fibrisolvens populations,but decreased (P＜0.05) the Ruminococcus flavefaciens population relative to the CON group. There were no F×C treatment interactions (P＞0.05) for any of the other indices except for the apparent digestibility of non-fibrous carbohydrate (NFC) (P＜0.05). The results demonstrated that adding 0.6% formic acid and 10% corn flour supplementation to banana pseudostem silages improved the nutritional quality of these silages and enhanced the growth performance of Nubian black castrated goats by improving apparent nutrient digestibility,and regulating ruminal fermentation and bacteria populations.

Keywords:banana pseudostem,silage,goat,nutrient digestion,rumen fermentation,ruminal cellulolytic bacteria

1.introduction

Using the agro-industrial by-products as animal feeds is an effective strategy for lowering the feeding costs and the costs of waste disposal (Benet al.2008;Vastaet al.2008). This strategy further contributes to avoiding competition between animal and human foods as well as stopping the waste of vital resources (Lange and Meyer 2019). As a result,developing sustainable strategies to integrate range vegetation,crop residues,shrubs,and agro-industrial byproducts in small ruminant production systems are of vital importance in the arid zones (Zhanget al.2019).

It is of vital importance to optimize the yield of livestock products which are a key component of food security (Smithet al.2013). Feeding with silage is extensively employed to alleviate the nutritional stress in the dry season (South China) when feeds are in shortage (Siebert and Hunter 1982). Bananas are produced seasonally,so banana stem can be preserved well as silage to be used for ruminant feeding in the entire year (Viswanathanet al.1989;Marie-Magdeleineet al.2010;Wanget al.2016;Barberaet al.2018). Liuet al.(2013) has reported that banana pseudostem could be directly ensiled for 2 mon and can be used for short-term storage without dehydration. Banana pseudostems are suitable for preservation as the direct silage,and the ensiled banana pseudostems stabilized after 50 days of fermentation (Wanget al.2016). However,the intake of banana pseudostem direct silage is limited when provided in ensiled form as an exclusive source of food. Such a phenomenon can be ascribed to the high content of condensed tannins (Hembade and Niras 2014;Oliveiraet al.2014). Furthermore,because of the relatively high moisture content and low crude protein (CP) of banana pseudostem direct silage,to make a higher quality silage,it is necessary to supplement it with suitable substrate(s) (lactic acid bacteria,molasses,wheat bran,and alfalfa) to reduce the moisture content,boost the CP content and ensure aerobic stability (Yanget al.2012;Elahiet al.2019;Mitikuet al.2020).

Corn flour,as a nutrient absorbent,can provide the watersoluble carbohydrate (WSC) and adjust the water content of raw silage materials,which decreases the nutrient loss by microorganisms. Corn flour,as an alternative fermentation (carbohydrate) substrate,is considered suitable for silage stability. Fermented tilapia silage prepared with corn flour attained a pH of less than 4.5 within 7 days of incubation at 30°C,and remained stable for 30 days (Fagbenro and Jauncey 1994). Direct addition of organic acids such as formic acid has been widely used in silage production.Formic acid supplementation rapidly decreased the pH of silage and limited the loss of protein and carbohydrates during fermentation (Luckstadt 2009). Formic acid applied at moderate rates to forages of low ensiling capacity (i.e.,low WSC concentration,high water activity,and high buffering capacity) is particularly effective in inhibiting the activity of undesirable bacteria,such asEnterobacteriaandClostridia(Lorenzo and O’Kiely 2008). Formic acid has been shown to inhibit undesirable microbial growth and dry matter (DM) losses (Goeseret al.2015). Nevertheless,little is known about the effects of formic acid and corn flour supplementation of banana pseudostem silages on the nutritional quality of these silages,growth,digestion,rumen fermentation and the cellulolytic bacterial community of Nubian black goats fed these silages.

Based on the previous studies,it was hypothesized that formic acid and corn flour supplementation of banana pseudostem silages might improve the nutritional quality of these silages and enhance the performance of Nubian black castrated goats fed these silages. Therefore,this study aimed to assess the effects of formic acid and corn flour supplementation of banana pseudostem silages on the nutritional quality of these silages,growth,digestion,rumen fermentation and the cellulolytic bacterial community of Nubian black goats fed these silages.

2.Materials and methods

This work was performed at Guangxi Dibaoli Agriculture and Animal Husbandry Co.Ltd.,Guangxi,China,following the Guidelines for the Care and Use of College of Animal Science and Technology by Yangzhou University. The temperature was maintained at 15.1-26.2°C throughout the study period in an indoor facility with heating radiators,and the mean relative humidity was maintained at 60.8%.

2.1.Banana pseudostem collection and silage production

Banana trees (Musa acuminataL.) were randomly selected from a certain banana farm (Nanning,Guangxi,China) following fruit harvesting. The upper two-thirds of the pseudostems of these plants were collected and cut into pieces of 2-3 cm length. The moisture content of banana pseudostem used for making the silage was 91.3%. Formic acid (98% purity) and corn flour were used to treat the banana pseudostem based on anin vitropilot experiment which indicated that 0.6% formic acid or 10% corn flour (on fresh matter basis) was enough to improve the degradation rates of DM,acid detergent fiber (ADF),and neutral detergent fiber (NDF),and to increase gas production after 24 h fermentation.The chemical compositions of banana pseudostem,corn flour,and their mixture (90:10) before ensiling are shown in Table 1. The banana pseudostems were ensilaged as follows:(i) banana pseudostem conventional silage group (CON) where silage was prepared without any additives,(ii) banana pseudostem+0.6% formic acid silage group (F);undiluted formic acid was sprayed on banana pseudostems before ensiling,(iii) banana pseudostem+10% corn flour silage group (C),and (iv) banana pseudostem+10% corn flour+0.6% formic acid silage group (F+C). In each group,the banana pseudostem was thoroughly mixed. After mixing,25 kg of the mixed material was tightly packed into a plastic bag (28 cm×50 cm) and rewrapped using plastic bag packaging (60 cm×120 cm) to prevent light intervention,followed by sealing with a polyethylene sheet to avoid air circulation and maintain an anaerobic environment. The sealed polyethylene sheets were kept under anaerobic conditions for 45 days. A total of 24 silos (4 treatments×6 replicates per treatment) were sampled for fermentation quality analysis.

2.2.animals and treatments

Four experimental diets were designed to contain 40% of the respective banana pseudostem silages with a final roughage:concentrate ratio of 50:50 (on a DM basis) (Tables 2 and 3).These diets satisfy the nutrient requirements for growth of Nubian black castrated goats at an average daily gain (ADG) of 100 g according to NRC (2007). A total of 48 Nubian black castrated goats (body weight (BW),(22.64±1.82) kg;4-mon-old) were randomly assigned to one of the four experimental groups with 12 replicates of one castrated goat per replicate for each treatment in a completely randomized design. Each group was offered one of the four experimental diets. Animals had free access to a daily concentrate mixture at 07:00 a.m.and 07:00 p.m.for 3 h following roughage feeding. The roughage amount given to the free access group was adjusted every day considering a 10% refusal based on the DM intake (DMI) of the previous day. The offered and residual roughage amounts were recorded every day. Before the beginning of the trial,the goats were adapted to various experimental diets for 10 days,and the feeding trial continued for 40 days. Goats were drenched using ivermectin at 0.2 mg kg-1of BW and were housed within individual pens (3.2 m×0.8 m). All pens were equipped with feeders and automatic water suppliers. The health status of goats was monitored daily before the morning feeding throughout the study period. The BWs of goats were measured at the initial and final experimental days before morning feeding. The ADG and average dry matter intake (ADMI) of goats were determined to calculate the feed conversion ratios (FCR).

Table 1 Chemical composition of banana pseudostem,corn flour,and their mixture (90:10) before ensiling (% of DM)

2.3.Digestion trial

A total of 48 Nubian black castrated goats were raised within individual metabolic cages (1.5 m×0.9 m) and given the corresponding diets according to the previous four treatments to determine nutrient digestibility. Animals were adapted to the experimental diets for five days,after which digestibility trial was initiated on day 25 of the experiment and lasted for seven days. The offered and residual feed amounts,as well as all feces,were weighed,followed by individual homogenization,and a 10% subsample was extracted in the early morning every day for seven days. Then,the collected samples were subject to oven drying for 72 h at 55°C,grinding by a Willey mill (Arthur H.Thomas,Philadelphia,PA,USA) to pass a 1-mm sieve and preservation before use. Afterward,the feeds,orts,and feces samples in the same class were pooled to prepare the composite samples for all animals and all experimental periods.

2.4.Sample extraction and laboratory analysis

Fermentation products of banana pseudostem silages were determined in the cold-water extracts of silage. Wet silage (100 g) samples were subjected to homogenization using 300 mL disinfected distilled water,and preservation overnight at 4°C (Caoet al.2009). The pH value of the filtrate was determined by a glass-electrode pH meter (Horiba D-21;Horiba,Tokyo,Japan). The content of lactic acid in the banana pseudostem silages was measured according to Barker and Summerson’s method (1941).Ammonia nitrogen (NH3-N) was determined by Conway’s method (1962). Volatile fatty acid (VFA) was distilled under steam,followed by qualitative and quantitative measurements using gas chromatography (GC,G-5000A;Hitachi,Tokyo,Japan). Also,WSC was measured according to the method described by McDonald and Henderson (1964). Briefly,0.2 g dried sample was ground using the mill to pass through a 1-mm screen,followed by mixing with sterile distilled water (10 mL) into a glass tube and 30 min of incubation at 100°C. Subsequently,the filtrate was transferred to a volumetric flask,followed by dilution to 100 mL using sterile distilled water. A 1 μL aliquot was collected,sufficiently mixed with 5 mL anthrone-sulfuric acid solution (containing 0.4 g anthrone+100 mL of 88% sulfuric acid) and incubated for 10 min at 100°C. The absorbance of the resultant mixture was then measured at 620 nm,with glucose as the reference substance. The standard curve wasy=116.8x+1.0821,R2=0.9996,in whichyrepresents the filtrate content (μg mL-1),whilexstands for the absorbance value measured at 620 nm. The filtrate was diluted at thex-value of ＞1.0,with the lower limit of detection of 10.48 mg L-1(Liet al.2016).

On day 40 of the feeding trial,20 mL rumen fluid was collected using a flexible gastric tube and a 100 mL syringe for pumping before the morning feeding (06:30 a.m.).Subsequently,the rumen fluid samples were filtered using two-layer cheesecloth,the filtrate pH value was recorded,and then a 5 mL sub-sample was subjected to acidification using sulphuric acid (3 drops,18 mol L-1) followed by storage in the refrigerator until analysis for NH3-N,microbial protein (MCP),and VFA content. The concentration of MCP was measured by purines according to Zinn and Owen’s method (1986) after modification by Makkar and Becker (1999).The purines-to-N ratio in the separated bacteria was used to estimate MCP;meanwhile,the RNA from yeast was utilized as the standard. Afterward,the samples subjected to acidification were stored in the refrigerator,followed by 15 min of centrifugation at 15 000×g at 4°C,and filtering with the Whatman #1 filter paper. The VFA content in the supernatant was determined by gas/liquid chromatography (Harvatineet al.2002).

Feed and feces samples were analyzed according to AOAC (1990) for DM (method number 930.15),CP (method number 990.02),ether extract (EE;method number 920.39),ash content (method number 924.05),as well as calcium (Ca) and phosphorus (P) (method numbers 968.08 and 965.17,respectively). Also,the contents of NDF together with ADF were measured according to the method of Van Soestet al.(1991). Non-fibrous carbohydrate (NFC) was calculated as 100-(NDF+CP+EE+Ash) (Mertens 1997). Total digestible nutrient (TDN) concentrations were calculated as the sum of the digestible fractions using the equation by Weiss (1999):TDN=DCP+2.25×Dlipid+DNFC+DNDF;where DCP,Dlipid,DNFC and DNDF are digestible CP,lipid,NFC and NDF,respectively. Total tannin (TT) was estimated by Folin-Ciocalteu reaction (Makkar 1999). Concentrations of condensed tannin (CT) were estimated according to the butanol-HCl assay (Díazet al.2010). Gross energy (GE) in the dietary ingredients and feces were measured using a bomb calorimeter (C200;IKA Works Inc.,Staufen,Germany). The metabolizable energy (ME) was estimated according to NRC (2007).

2.5.DNa collection and quantitative real-time PCR

DNA from ruminal microbes was collected from the ruminal fluid (1.5 mL) based on Yu and Morrison’s method (2004) using repeated bead beating combined with a column and was stored at -20°C until PCR. TheRuminococcus albus,Fibrobacter succinogenes,Ruminococcus flavefaciensandButyrivibrio fibrisolvenspopulations were obtained through real-time PCR,which was used as part of the entire bacterial 16S rDNA following the formula (Peiet al.2013):

Relative quantification=2-(Cttarget-Cttotalbacteria)

where Ct represents the threshold cycle. The primer sets used for real-time PCR were forward primer 5´-CCC TAAAAGCAGTCTTAGTTCG-3´,reverse primer 5´-CCTCCTTGCGGTTAGAACA-3´ forR.albus(Koike and Kobayashi 2001),forward primer 5´-CGAACG GAGATAATTTGAGTTTACTTAGG-3´,reverse primer 5´-CGGTCTCTGTATGTTATGAGGTATTACC-3´ forR.flavefaciens,forward primer 5´-GTTCGGAATTA CTGGGCGTAAA-3´,reverse primer 5´-CGCCTGCCCCTG AACTATC-3´ forF.succinogenes,and forward primer 5´-CG GCAACGAGCGCAACCC-3´,reverse primer 5´-CCATTG TAGCACGTGTGTAGCC-3´ for total bacteria (Denman and McSweeney 2006),forward primer 5´-TCTGGAAACGGATGGTA-3´,reverse primer 5´-CCTTTAAGACAGGAGTTTACAA-3´ forB.fibrisolvens(Forsteret al.1996).

The MxPro-Mx3005P multiplex quantitative PCR systems (Stratagene,La Jolla,CA) were employed for real-time PCR.The 20 μL reaction mixture was comprised of SYBR Premix TaqTM(10 μL,TaKaRa Biotechnology Co.,Ltd.,Dalian,China),PCR forward primer (0.8 μL,0.2 μmol L-1),PCR reverse primer (0.8 μL,0.2 μmol L-1),ROX feference dye (0.4 μL,50×),dH2O (6.0 μL) as well as template DNA (2 μL). The DNA quantity for each sample was screened in triplicate with an ND-1000 UV spectrophotometer (NanoDrop Technologies,USA) to obtain the averages. The PCR conditions were as follows:at 50°C for 2 min,at 95°C for 2 min for 1 cycle to perform initial denaturation,followed by 95°C for 15 s and 60°C for 1 min for 40 cycles to perform primer annealing and product elongation (Denman and McSweeney 2006;Peiet al.2013).

2.6.Statistical analyses

All data were analyzed as a 2×2 factorial arrangement of treatments by two-way ANOVA using the general linear model procedures of SAS (SAS Institute). The statistical model included the effects of formic acid level,corn flour level,and their interactions. One silage bag or an individual castrated goat is the experimental unit.When there was a significant interaction,post hoc testing was conducted using Bonferroni’s multiple comparison test. AP-value of ≤0.05 was declared as statistically significant.

3.Results

3.1.Chemical composition of banana pseudostem silages

There were no F×C treatment interactions (P＞0.05) on the chemical compositions of the different banana pseudostem silages (Table 2). Supplemental formic acid level or corn flour level affected (P＜0.05) DM,GE,EE,CP,NFC,ash,NDF,ADF,TT and CT concentrations. A 0.6% formic acid supplementation increased (P＜0.05) DM,CP,and NFC concentrations,but decreased (P＜0.05) the ash,NDF,and CT concentrations compared with the CON group. A 10% corn flour supplementation increased (P＜0.05) GE and NFC concentrations,but decreased (P＜0.05) the ADF,TT and CT concentrations compared with the CON group.

Table 3 Ingredient and chemical compositions of the experimental diets (DM basis)

3.2.Fermentation quality of banana pseudostem silages

There were no F×C treatment interactions (P＞0.05) on the fermentation quality of banana pseudostem silages (Table 4).Butyric acid was not detected in the F+C treatment.Supplemental formic acid level or corn flour level affected (P＜0.03) the concentrations of butyric acid,lactic acid,and WSC,and the values of pH and NH3-N/total N,but had no effect (P＞0.05) on acetate or propionate concentrations.A 0.6% formic acid supplementation increased (P＜0.05) lactic acid and WSC concentrations,but decreased (P＜0.05) the values of pH and NH3-N/total N and butyric acid concentration compared with the CON group. A 10% corn flour supplementation increased (P＜0.05) lactic acid concentration,but decreased (P＜0.05) the pH value and butyric acid concentration compared with the CON group.

3.3.apparent nutrient digestibility

There was an F×C treatment interaction (P＜0.05) on the apparent digestibility of NFC (Table 5). The castrated goats in both supplemental formic acid and corn flour group had the highest (P＜0.05) apparent digestibility of NFC relative to the others. Supplemental formic acid level or corn flour level affected (P＜0.04) the apparent digestibility of the DM,CP,NDF,ADF,NFC,EE,as well as digestible energy (DE)/GE,TDN concentration and TDN intake. Compared with the CON group,a 0.6% formic acid supplementation increased (P＜0.05) the apparent digestibility of the DM,NFC,EE,and DE/GE and TDN intake. Compared with the CON group,a 10% corn flour supplementation increased (P＜0.05) the apparent digestibility of the CP,NDF,ADF,and DE/GE and TDN concentration.

3.4.Ruminal fermentation

There were no F×C treatment interactions (P＞0.05) on ruminal fermentation characteristics (Table 6). Supplemental formic acid level or corn flour level affected (P＜0.03) the MCP,total volatile fatty acid (TVFA),propionate,NH3-N and butyric acid concentrations,and the acetate/propionate ratio,but had no effect (P＞0.05) on the PH value or acetate concentration. A 0.6% formic acid supplementation increased (P＜0.05) MCP and butyric acid concentration,but decreased (P＜0.05) the NH3-N concentration compared with the CON group. A 10% corn flour supplementation increased (P＜0.05) propionate and butyric acid concentrations,but decreased (P＜0.05) the acetate/propionate ratio compared with the CON group.

3.5.Ruminal bacteria populations

There were no F×C treatment interactions (P＞0.05) on ruminal bacteria populations (Table 7). Supplemental formic acid level or corn flour level affected (P＜0.03) theF.succinogenes,R.albus,B.fibrisolvens,andR.flavefacienspopulations. A 0.6% formic acid supplementation increased (P＜0.05) theF.succinogenesandB.fibrisolvenspopulations,but decreased (P＜0.05) theR.flavefacienspopulation compared with the CON group. A 10% corn flour supplementation increased (P＜0.05) theR.albuspopulation,but decreased (P＜0.05) theR.flavefacienspopulation compared with the CON group.

3.6.growth performance

There were no F×C treatment interactions (P＞0.05) on the growth performance of Nubian black goats (Table 8). There was no difference (P＞0.05) for IBW among all groups. Supplemental formic acid level or corn flour level affected (P＜0.04) the FBW,ADG,ADMI and FCR. A 0.6% formic acid supplementation increased (P＜0.05) the FBW and ADMI,but decreased (P＜0.05) the FCR relative to the CON group. A 10% corn flour supplementation increased (P＜0.05) the ADG and ADMI relative to the CON group.

4.Discussion

The results from the present study have supported our hypothesis that formic acid and corn flour supplementation of banana pseudostem silages might improve the nutritional quality of these silages and enhance the performance of Nubian black castrated goats fed these silages. These results indicated that 0.6% formic acid and 10% corn flour supplementation improved the nutritional quality of these silages and enhanced the growth performance of Nubian black castrated goats by improving apparent nutrient digestibility,and regulating ruminal fermentation and bacterial populations. Our results provided a scientific experimental basis for the application of formic acid and corn flour supplementation in banana pseudostem silages for goats.

In the present study,the F or C treatments,at the rates of 0.6% formic acid or 10% corn flour addition,respectively,can effectively change the chemical composition of silage material. The F-treated silage had a higher DM content than CON silage,and the C+F-treated silage had a higher CP content than the others. The F-treated silage had a lower ash content than CON silage. A similar lower level of ash content was reported during the formic acid ensiling of shrimp head silage meal (Foxet al.1994).This reduction of ash content in the present study might be due to the deprivation of inorganic constituents within the banana pseudostem used for silage preparation.

In this study,supplemental addition of 10% corn flour and 0.6% formic acid to the banana pseudostem silage increased the contents of the DM,CP,and GE,which might be beneficial for silage microbial fermentation.The NDF and ADF contents were lower but the NFC content was higher in F,C and F+C treatments than those of the CON treatment,respectively,which might be due to a greater degradation of cellulose in the F,C and F+C treatments during ensiling. These changes were marked in silages treated with corn flour because corn flour has a lower fiber content. Also,the fiber levels were lower and the NFC content was higher after ensiling.

A higher lactic acid content and lower pH value (pH＜4.2) have been identified as vital factors to ensure success in preserving silage (Bolsenet al.1996). Meanwhile,the elevated WSC level in herbage contributes to the fermentation of lactic acid,which will result in a lower silage pH (Wanget al.2016). Although lactic acid bacteria count was not a part of this study,higher production of lactic acid is an indication of a dominant population of lactic acid bacteria. The lower pH value of silage in this study might be due to the higher lactic acid production.

The ratio of NH3-N to total N reflects the decomposition degree of proteins and amino acids in silage,and the higher the ratio,the more protein breakdown,indicating poor silage quality (Kunget al.2004). The NH3-N is generated through degradation of the amino acids by the microbes. The well-fermented silage should possess＜10% of NH3-N/total N (McDonaldet al.2002). In this study,NH3-N/total N of each group was less than 10%,showing that the silages of banana pseudostems were well preserved,especially for the F+C silage group. In addition,the NH3-N concentration is consistent with the degree of butyric acid fermentation;i.e.,the more proteolysis,the higher the butyric acid concentration,indicating a poor silage quality (Valenteet al.2003). In our study,butyric acid content was not detected in the F+C group,suggesting that the addition of both 10% corn flour and 0.6% formic acid could improve the banana pseudostem silage quality.

Digestibility has been used as an early indicator of nutrient availability in livestock feed (Yusmadiet al.2008;Trujilloet al.2017;Huet al.2019). Nutrient intake and digestibility can serve as vital indicators for assessing the nutritional significance of the feedstuffs (NRC 2007). Our results indicated that the apparent digestibility of the DM,CP,EE,and NDF in the F,C,and F+C treatments were markedly higher than those in the CON treatment,which showed that the F,C and F+C treatments (especially the F+C treatment) made greater contributions to nutrient digestibility than the CON treatment. Also,the high digestibility of banana pseudostem silages supplied with 10% corn flour and 0.6% formic acid could be ascribed to some decreases in the anti-nutritional factors,such as the TT and CT. It has been indicated that a low concentration of tannin can combine with dietary protein to form by-pass ruminal protein,thus avoiding the excessive degradation of ingested proteins by rumen microorganisms,and this is beneficial for the absorption of protein in the ruminant abomasum and small intestine (Rakhmaniet al.2005). However,a high concentration of tannin can combine with dietary protein to form insoluble complexes in the rumen,which makes dietary protein only minimally utilized by rumen microbes. In the present study,the apparent digestibility of CP was higher,and the concentration of tannin was lower,in the F,C,and F+C treatments than those in the CON treatment,which is consistent with the above-mentioned reports. The digestibility of forage is mainly related to the composition of the cell wall and the access of ruminal microorganisms to the plant tissues. Therefore,the presence of lignin in this fraction may interfere with the use of fibrous carbohydrates (Moore and Jung 2001). The ADF showed similar behavior to that of lignin content,probably because it is composed of cellulose and lignin;however,the lower the lignin content,the higher the ADF digestibility (Carmoet al.2018). This might be an explanation for the finding in this study that the apparent digestibility of the ADF and NDF in the F,C,and F+C treatments were higher than that in the CON treatment.About 85-95% of dietary lipids can be decomposed by microorganisms in the rumen (Bauchartet al.1990). In the present study,the apparent digestibility of EE in the F+C treatment was higher than in the other treatments,which may be due to the higher ruminal microbial activity increasing the lipid decomposition in the F+C treatment.

The rumen pH was within the range of 6.91-7.19 in each diet,and such a range is optimal (6.7±0.5) for the normal cellulolytic bacterial activity (Van Soest 1994),while the pH＞6.0 is optimal for the synthesis of microbial protein (Sniffenet al.1992). Our results recorded a ruminal pH of 6.91-7.19,which is appropriate for ruminal microbial activity,especially for the cellulolytic bacteria (Russell and Wilson 1996). Ruminal ammonia content is affected by the degradation of protein and microbial protein synthesis (Santosoet al.2007). Ammonia is the main source of N for synthesizing microbial proteins,and a lower concentration of ruminal ammonia could be a result of stimulation of rumen microbiota,which directly utilize ammonia to synthesize proteins,so its concentration in the rumen is a case that had to be observed (Satter and Slyter 1974). This could be an explanation for the lower ammonia-N concentration in the rumen,and the greater MCP concentration and ruminal microorganismsF.succinogenes,R.albusandB.fibrisolvenspopulations in the F,C and F+C groups (especially in the F+C group) than in the CON treatment in this study. According to McDonaldet al.(2002),6-21 mmol L-1is the optimal NH3level for synthesizing rumen microbial proteins. The NH3-N content in our study lied within the limits reported previously. The CP apparent digestibility in the F,C,and F+C treatments was elevated with the decrease in NH3-N level,indicating that the F,C,and F+C treatments could improve N utilization compared with the CON treatment.

The rumen VFA profile could be used to describe whether a ration is appropriate to the livestock (Dal Pizzolet al.2017).Most of the energy required by ruminants is supplied by VFA and is directly related to the conditions of fermentation and the control of the ruminal environment,and diets with greater digestibility normally result in a higher VFA concentration in the rumen (Gilaverteet al.2011). Therefore,the higher total VFA concentration in the rumen might be due to the higher apparent nutrient digestibility in the F,C and F+C groups (especially in the F+C group) than in the CON treatment in this study. There was no difference in the concentration of acetate among all groups. This might have occurred due to the active degradation of the microbial population,considering that ruminal pH value was maintained close to neutral in all treatments (Carmoet al.2018). The 0.6% formic acid and 10% corn flour supplementations increased the concentrations of propionate and butyrate in the ruminal fluid in comparison with the other groups. The higher NFC concentration in the F+C diet than the other diets -due to the high ruminal degradation rate and better fermentative quality (Oliveiraet al.2014) -favors the production of propionate,which explains these results. The production of propionate is more energetically efficient and,theoretically,reduces energy loss as methane that is normally associated with acetate production (Riveraet al.2010). Therefore,it is possible to hypothesize that when reducing the molar proportion of acetate:propionate,0.6% formic acid and 10% corn flour supplementation will enable an increase in the glycogenic potential of the diets,enhance the efficiency of the energy metabolism,and decrease the energy lost as methane.

The complex symbiotic microbiota of the rumen (anaerobic rumen fibrolytic bacteria,protozoa,and fungi) is responsible for the breakdown of plant fiber allowing ruminants to utilize plant fiber for nutrition. Bacteria are the most numerous of these microbiota and play a major role in the biological degradation of dietary fiber (Wanapat and Cherdthong 2009). TheR.albus,R.flavefaciens,F.succinogenesandB.fibrisolvensare the dominant cellulolytic bacterial populations in the rumen (Wanapatet al.2014). In the past years,a great number of studies have been performed to examine the adhesion as well as digestion capacities of these bacteria for fibrous plant particles (Mironet al.2001;Wanget al.2016;Kavithaet al.2017). The growth and multiplication of these ruminal cellulolytic bacteria contribute to the secretion of fibrolytic enzymes for the digestion of fibrous materials (Wanget al.2016). Our results indicated thatR.albus,F.succinogenesandB.fibrisolvenspopulations in the F,C and F+C groups (especially in the F+C group) were notably increased compared with the CON treatment,which suggested that supplementing diets 10% corn flour or 0.6% formic acid could stimulate their growth and multiplication. This could be an explanation for the higher apparent digestibility of the ADF and NDF in the F,C,and F+C treatments compared to those in the CON treatment in this study.

Given that the production of banana pseudostem is seasonal,its use as animal feed over the whole year requires adequate preservation and storage. At the same time,a cost-effective feeding approach is needed for profitable goat and sheep entrepreneurs. The potential use of banana or its by-products as part of small ruminant diets have been reported by other authors (Marie-Magdeleineet al.2009,2010;Wanget al.2016). Nevertheless,limited information is available documenting the effect of 0.6% formic acid or 10% corn flour supplementation on goat performance and evaluating the potential of pseudostem as ruminant forage. In this study,we found that 0.6% formic acid and 10% corn flour supplementation induced a greater increase in the DMI and growth performance than the CON treatment,respectively,which were in accord with the results of ruminal fermentability,digestibility trials and ruminal microorganism populations in the goats.

5.Conclusion

Supplementation with 0.6% formic acid and 10% corn flour (especially for the F+C treatment) improved the nutritional quality of these silages and enhanced the growth performance of Nubian black castrated goats fed these silages by improving the apparent nutrient digestibility,and regulating ruminal fermentation and bacteria populations.

acknowledgements

This research was supported by the China Special Fund for Agro-scientific Research in the Public Interest (Investigation on strategies of fattening and high-quality meat productive techniques for herbivores based on resources of unconventional roughage in Southern China) (201303144),the Top Talents Award Plan of Yangzhou University,China (2016,2020),the Cyanine Project of Yangzhou University (2020),the Technology Specialty Fund for Cooperation between Jilin Province and the Chinese Academy of Sciences,2016SYHZ0022),the National Key Research and Development Program of China (2016YFD0700201),the National Natural Science Foundation of China (31902180),and the Natural Science Foundation of Jiangsu Province Research Project,China (BK20170488). The authors thank all the members of the Prof.Wang Hongrong’s laboratory who contributed to sample determination.

Declaration of competing interest

The authors declare that they have no conflict of interest.

Ethical approval

Animal experiments were approved by the Institutional Animal Care and Use Committee of the Yangzhou University Animal Experiments Ethics Committee,China (permit number:SYXK [Su] IACUC 2012-0029). All experimental procedures were performed in accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals approved by the State Council of the People’s Republic of China.