Research Article Received: 5 August 2013
Revised: 12 November 2013
Accepted article published: 19 November 2013
Published online in Wiley Online Library: 21 December 2013
(wileyonlinelibrary.com) DOI 10.1002/jsfa.6485
In vitro evaluation on neutral detergent fiber and cellulose digestion by post-ruminal microorganisms in goats Jinzhen Jiao,a,b Pengpeng Wang,a,b Zhixiong He,a Shaoxun Tang,a Chuanshe Zhou,a Xuefeng Han,a Min Wang,a Duanqin Wu,a,b Jinhe Kanga and Zhiliang Tana∗ Abstract BACKGROUND: Post-ruminal digestion of fiber has received much less attention than its ruminal digestion. Using in vitro incubation techniques, the present study explored whether variations in fiber digestion occurred in different segments of the post-ruminal tract and whether fiber structure could influence its digestibility. A split plot design was conducted with gut segments (jejunum, ileum, cecum and colon) as main plot and substrates (neutral detergent fiber (NDF) and cellulose (CEL)) as subplot. RESULTS: With the same substrate, the final asymptotic gas volume (V F ), gas production at ti (Vti ), digestibility, microbial crude protein (MCP), total bacteria number (TBN), total short-chain fatty acids (TSCFA) and xylanase in incocula from the cecum and colon exceeded (P < 0.01) those in incocula from the jejunum and ileum, while the NH3 -N in the former was less (P < 0.01). For the same gut segment, the digestion of CEL was superior to NDF, as reflected in greater V F , Vti , maximum rate of gas production, digestibility, enzyme activities and SCFA but lower pH and NH3 -N. CONCLUSION: The current results imply that the intestinal contents from the cecum and colon have greater potential to digest fiber than those from the jejunum and ileum, and CEL is more easily digested in the post-ruminal tract than NDF. c 2013 Society of Chemical Industry Keywords: digestion; fiber; goat; in vitro; post-ruminal tract
INTRODUCTION
J Sci Food Agric 2014; 94: 1745–1752
the extent of fiber digestion at different segments of the postruminal digestive tract. It is hypothesized that there are significant variations in fiber digestion at different segments of the postruminal tract, the extent of which is influenced by fiber structure. Therefore the objective of this study was to evaluate the potential of post-ruminal microbes to utilize different fiber sources (neutral detergent fiber (NDF) and cellulose (CEL)) at the jejunum, ileum, cecum and colon using in vitro incubation techniques based on relevant digesta from goats.
∗
Correspondence to: Zhiliang Tan, Key Laboratory for Agro-Ecological Processes in Subtropical Region, Hunan Research Center of Livestock and Poultry Sciences, South-Central Experimental Station of Animal Nutrition and Feed Science in Ministry of Agriculture, Institute of Subtropical Agriculture, The Chinese Academy of Sciences, Changsha, Hunan 410125, China. E-mail: [email protected]
a Key Laboratory for Agro-Ecological Processes in Subtropical Region, Hunan Research Center of Livestock and Poultry Sciences, South-Central Experimental Station of Animal Nutrition and Feed Science in Ministry of Agriculture, Institute of Subtropical Agriculture, The Chinese Academy of Sciences, Changsha, Hunan, 410125, China b University of the Chinese Academy of Sciences, Beijing, 100049, China
www.soci.org
c 2013 Society of Chemical Industry
1745
Ruminants can utilize large quantities of low-quality roughages as energy sources by microbial degradation of fiber in the gastrointestinal tract (GIT). Anaerobic degradation of fiber is a complex process involving several functional groups of microorganisms, especially fibrolytic bacteria (e.g. Ruminococcus albus, Fibrobacter succinogenes and Ruminococcus flavefaciens),1 and results in the production of many end products (e.g. shortchain fatty acids (SCFA), CO2 , CH4 and H2 ).2 While digestion of fiber in the rumen has been intensively studied, its post-ruminal digestion has received much less attention. However, the total quantity of wet digesta at the hindgut of sheep is reported to be 700–1200 g, equivalent to 15–26% of the mass in the rumen (4.5 kg).3 In the post-ruminal tract, total bacterial counts reach 1012 cells g−1 digesta.4 These microbes exert numerous physiological functions, including fermentation of fiber not digested in the forestomach. Tan et al.5 reported that although ruminal fiber digestion was depressed by increased dietary non-structural carbohydrates in sheep, this could be partially compensated by increased microbial fermentation in the hindgut. Although post-ruminal microbes can utilize considerable fiber, limited knowledge is available about
www.soci.org
MATERIALS AND METHODS Animals and experimental design Three adult Liuyang black wether goats (1.5 years of age with an average body weight of 20.1 ± 1.2 kg), a local breed of southern China, were used. The goats were housed individually, had free access to drinking water and were fed an experimental diet (forage/concentrate, 55:45 w/w) formulated to meet 1.4 times metabolizable energy (ME) requirements for maintenance according to the feeding standard of Chinese goats. The diet was composed of 550 g rice straw, 100 g soybean meal, 284 g corn meal, 4 g CaH2 PO4 , 0.2 g CaCO3 , 43 g fat powder, 9 g NaCl and 10 g premix kg−1 dry matter (DM) and contained 91.4 g crude protein, 2.3 g calcium, 1.6 g phosphorus, 414 g NDF and 9.38 MJ ME kg−1 DM. There was a total adaptation period of 3 weeks, with the first week to accustom the goats to the diet and the last 2 weeks with regular feeding, before the first slaughter. Animals were fed the diet as two equal amounts daily at 08:00 and 18:00, and no refusals occurred after the adaptation period. The three goats were slaughtered at 1 week intervals. The digesta collected from the jejunum, ileum, cecum and colon of each goat provided the four inocula. After tying off the jejunum, ileum, cecum and colon with plastic strips, the entire GIT was removed and taken to the laboratory within 1 h under an atmosphere of CO2 to maintain anaerobic conditions. The various digesta were weighed and diluted 1:5 (w/v) with pre-warmed (39 ◦ C), anaerobic, sterile phosphate-buffered saline (pH 7.4). The diluted materials were then homogenized using a hand-mixer for 60 s and strained through four layers of cheesecloth. The resulting liquids were used as inocula. The experimental scheme was as follows: 2 fermented substrates (NDF and CEL) of 0.4 g each × 4 gut segments (jejunum, ileum, cecum and colon) × 4 in vitro incubation time points (12, 24, 48 and 72 h). The three goats were treated as replicates, and each incubation time point had three bottles as repeats. The experiments were approved by the Animal Care Committee, Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha, Hunan, China. Preparation of culture medium, substrates and inocula A semi-defined medium, modified from Lowe et al.,6 in which no carbon source was added, was used for in vitro cultivation studies. The medium was prepared by mixing 76 mL of basal solution, 4 mL of bicarbonate buffer, 1 mL of reducing agent and 1 mL of vitamin/phosphate buffer solution. NDF was extracted from rice straw by the method of Yang and Xie.7 CEL was extracted from rice straw by the method of Sun et al.8 Following overnight feed withdrawal, goats were anaesthetized and slaughtered. Each inoculum was then dispensed into the pre-warmed bottles containing substrate and medium.
1746
In vitro incubation procedures From preliminary studies in vitro (data not shown), an NDF dose of 0.4 g was chosen as appropriate, since it allowed good growth of colonic microbes with suitable gas production (∼50 mL) during 72 h of in vitro incubation. The NDF proportion of the post-ruminal digesta in vivo ranged from 5.5 to 19.0%, equivalent 0.11–0.38 g NDF per 2 g digesta. In our in vitro experiment, 2 g of digesta was collected and diluted with phosphate-buffered saline (pH 7.4) (w/v 1:5), homogenized using a hand-mixer for 60 s and strained through four layers of cheesecloth. Then 10 mL of filtrate (with negligible amount of NDF) was used as inoculum. In practice, the 0.4 g of NDF with 10 mL of inoculum used in vitro was similar to
wileyonlinelibrary.com/jsfa
J Jiao et al.
both the NDF and microbial content found in 2 g of digesta in vivo. For the in vitro incubation, 0.4 g of NDF or CEL was incubated at 39 ◦ C under strictly anaerobic condiitons in 132 mL glass bottles with 30 mL of medium (pH adjusted to 6.8–7) and 10 mL of inoculum. The bottles were placed in a shaking water bath with 50 movements min−1 at 39 ◦ C and gas volumes were recorded every 1 h for 72 h. Bottles were removed from fermentation at 12, 24, 48 and 72 h of incubation, and fluid samples from the bottles were immediately stored at −20 or −80 ◦ C for later analyses. Bottles filled with 10 mL of inoculum (jejunum, ileum, cecum or colon) and 30 mL of medium, but without NDF and CEL, were run simultaneously as controls. The residual solutions after 12, 24, 48 and 72 h of fermentation were filtered into pre-weighed Gooch filter crucibles, dried at 105 ◦ C for 24 h and weighed for the determination of in vitro NDF or CEL digestibility. Calculation of in vitro gas production parameters The in vitro gas curves were fitted using NLREG Version 5.0,9 which employs a combination of the Gauss–Newton and Levenberg–Marquardt methods.10 In general, the logistic–exponential (LE) model suggested by Wang et al.11,12 was fitted without major problems, although initial values were selected based on visual inspection of the plots in order to reach convergence and limit the number of calculations and calculation time. The LE model was V = VF 1 − exp (−kt) / 1 + exp (b − kt) where V is the cumulative gas production (mL) at time point t, V F is the final asymptotic gas volume (mL), k is the rate of gas production and b is a shape parameter. Then FRD0 (initial fractional rate of degradation) = k/[1 + exp(b)], ti (time at which inflexion of the LE model occurred) = b/k, Vti (gas production at ti ) and λ (derived lag time) = ti − Vti /RGti were derived using the LE model. Chemical analysis of in vitro culture fluids After incubation for 12, 24, 48 and 72 h, the bottle contents were filtered through four layers of cheesecloth. The fluids obtained after filtration were used for the following analyses. The pH of culture fluids was measured immediately (pH meter model 2000, Beckman Instruments, Fullerton, CA, USA). The NH3 N concentration was determined by the phenol/hypochlorite method using a UV–visible spectrophotometer at 550 nm.13 SCFA (i.e. acetate, propionate, butyrate, valerate, isobutyrate and isovalerate) concentrations were analyzed using a gas chromatograph (7890A, Agilent, USA) according to the method of Garcia-Gonzalez et al.14 Microbial crude protein (MCP) was determined using the trichloroacetic acid method suggested by Licitra et al.15 The activities of carboxymethylcellulase (CMCase, an endoglucanase) and xylanase were assayed based on procedures described by Rey et al.16 An enzyme activity unit (U) was defined as the amount of enzyme required to release 1 µmol reducing sugars (xylose or glucose equivalent) min−1 mL−1 culture fluid. Quantitative polymerase chain reaction (qPCR) determination of total bacteria number (TBN) Genomic DNA of in vitro culture fluids (before filtration) was extracted from 1 mL aliquots using a QIAamp DNA Stool Minikit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The purified DNA was quantified using NanoDrop ND1000 (NanoDrop Technologies, Wilmington, DE, USA). The
c 2013 Society of Chemical Industry
J Sci Food Agric 2014; 94: 1745–1752
Fiber digestion in post-ruminal tract
www.soci.org
forward primer 5 -CGGCAACGAGCGCAACCC-3 and reverse primer 5 -CCATTGTAGCACGTGTGTAGCC-3 cited by Denman and McSweeney17 were used in qPCR to quantify total bacteria in culture fluids. The purified PCR product amplified with this pair of primers was used to generate plasmid DNA, followed by construction of the standard curve between Ct value and plasmid copies through serial dilution of plasmid copies. The qPCR assays were performed on an ABI 7900HT system (Applied Biosystems, Foster City, CA, USA) in a total volume of 10 µL using SYBR Premix Ex Taq (Takara, Ostu, Japan) following the suggested protocol. The copy numbers of total bacteria were determined by relating the Ct value to the standard curves. The final copy numbers of targeted total bacteria (TBN) mL−1 culture fluid were calculated as described by Zhou et al.18 Statistical analysis The repeated data (data from three bottles) were averaged to be analyzed using the MIXED procedures of SAS (SAS Institute, Cary, NC, USA) according to a split plot design. Gut was included in the main plot and substrate in the subplot. For in vitro gas production parameters, the model consisted of random (goat within (substrate, gut, substrate × gut)) and fixed (substrate, gut, substrate × gut) effects. Digestibility, MCP, TBN, enzyme activities, pH, NH3 -N and SCFA were analyzed using the MIXED procedures of SAS in a REPEATED statement with incubation time as the repeated factor and goat as the subject. The compound symmetry was used as the covariance structure. The model included as random effects goat and substrate, gut and time, plus their interactions within goat. Fixed effects were substrate, gut and time, plus their interactions. Contrast statements included the main effect comparing fermented substrate (NDF vs CEL). Digestibility, MCP, TBN and enzyme activity data were reported at each time point, whereas pH, NH3 -N and SCFA data were reported as mean values of four time points.
RESULTS Gas production of NDF and CEL by digesta from post-ruminal tract The V F and Vti of NDF and CEL were similar in inocula from the cecum and colon but exceeded (P < 0.01) those in inocula from the ileum, while the ti and λ of CEL were greater in inocula from the ileum than in inocula from the cecum and colon (Table 1). Maximum rates of gas production (i.e. the rates at ti ) of NDF and CEL were similar between inocula from the cecum and colon and occurred at about 40 and 30 h of incubation respectively. These rates exceeded those observed in the ileal inocula, which occurred at approximately 48 and 56 h of incubation respectively. Little gas was produced in the jejunal inocula during 72 h of incubation. Furthermore, CEL had higher (P < 0.01) V F and Vti compared with NDF.
J Sci Food Agric 2014; 94: 1745–1752
MCP and TBN of NDF and CEL by digesta from post-ruminal tract For both NDF and CEL, MCP levels in the hindgut inocula were greater (P < 0.01) than those in the small intestinal inocula after 48 and 72 h of incubation (Table 3). Additionally, for both NDF and CEL, TBN levels in the two hindgut inocula were greater (P < 0.01) than those in the two small intestinal inocula (Table 4). When incubated with NDF, after 24, 48 and 72 h of incubation, TBN levels in the small intestinal inocula were less (P < 0.05) than those in the hindgut inocula. When incubated with CEL, after 12, 24 and 48 h of incubation, TBN levels in the hindgut inocula exceeded (P < 0.05) those in the small intestinal inocula. Enzyme activities related to NDF and CEL digestion in post-ruminal tract inocula When incubated with NDF, after 72 h of incubation, the activity of CMCase in the ileal inocula was greater (P < 0.01) than that in the jejunal inocula (Table 5). When incubated with CEL, after 48 h of incubation, the greatest activity of CMCase was present in the cecal inocula and exceeded (P < 0.01) those observed in inocula from the jejunum and ileum; after 72 h of incubation, the greatest activity of CMCase was observed in inocula from the ileum, followed by inocula from the jejunum and cecum, with the colonic inocula being the least active. Furthermore, the activities of CMCase were greater (P < 0.05) for CEL than for NDF for all post-ruminal tract inocula except those from the jejunum. For both NDF and CEL, the activities of xylanase were greater (P < 0.01) in inocula from the cecum and colon than in inocula from the jejunum (Table 5). Only in the cecal inocula was the activity of xylanase less (P = 0.035) for NDF than for CEL. pH, NH3 -N and SCFA from NDF and CEL in post-ruminal tract inocula When incubated with CEL, the pH in the jejunal inocula was greater (P < 0.01) than those in inocula from the cecum and colon (Table 6). Moreover, for the hindgut inocula, NDF had greater (P < 0.05) pH when compared with CEL. Similarly, when incubated with NDF, NH3 -N concentrations in inocula from the ileum, cecum and colon were less (P < 0.01) than that in inocula from the jejunum. When incubated with CEL, the NH3 -N concentration in the ileal inocula was less (P < 0.01) than that in the jejunal inocula but greater (P < 0.05) than those observed in the hindgut inocula. The SCFA concentrations were greater (P < 0.05) in the hindgut inocula than in the small intestinal inocula for both NDF and CEL. In contrast, the acetate/propionate (A/P) ratios in the small intestinal inocula were greater (P < 0.01) than those in the hindgut inocula. Furthermore, in the jejunual inocula, CEL had greater (P < 0.05) butyrate and A/P ratio than NDF, while, in the ileal inocula, NDF had less (P < 0.01) butyrate than CEL. For both hindgut inocula, CEL produced greater (P < 0.05) acetate, propionate and total SCFA (TSCFA) compared with NDF.
DISCUSSION The post-ruminal tract is a very complex system involving interactions between the host and microorganisms and also
c 2013 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1747
Digestibility of NDF and CEL by digesta from post-ruminal tract Compared with the small intestine segments (jejunum and ileum), digestibilities of both NDF and CEL were greater (P < 0.01) in the hindgut segments (cecum and colon) (Table 2). After 48 h of incubation, digestibilities of NDF and CEL were less (P < 0.01) within the small intestine sections than within the hindgut sections. After 72 h of incubation, digestibilities of NDF and CEL in the ileal inocula were 35.4 and 52.2% respectively, greater (P < 0.01) than
those observed in the jejunal inocula but less (P < 0.01) than those observed in inocula from the cecum and colon. Furthermore, CEL had greater digestibility than NDF in inocula from the jejunum (P = 0.031), ileum (P = 0.011), cecum (P = 0.0003) and colon (P = 0.0003).
www.soci.org
J Jiao et al.
Table 1. In vitro gas production parameters of neutral detergent fiber (NDF) and cellulose (CEL) estimated from logistic–exponential (LE) model with inocula from jejunum, ileum, cecum and colon Gut Parametera −1
V F (mL g
Substrate
Jejunumb
NDF CEL NDF CEL NDF CEL NDF CEL NDF CEL
— — — — — — — — — —
)
FRD0 (% h−1 ) ti (h) Vti (mL g−1 ) λ (h)
Ileum 45.7cC 99.3cA 0.18 0.06 48.3 56.2a 22.7cC 49.6cA 31.6 46.3a
P values Cecum
Colon
125.6aC 183.1aA 0.08 0.12 40.4 33.7c 62.3aC 90.6aA 25.0 20.8c
115.8aC 176.1aA 0.09 0.08 41.2 34.6c 57.3aC 87.5aA 24.8 21.9c
Gut
Substrate
Gut × substrate
Revised: 12 November 2013
Accepted article published: 19 November 2013
Published online in Wiley Online Library: 21 December 2013
(wileyonlinelibrary.com) DOI 10.1002/jsfa.6485
In vitro evaluation on neutral detergent fiber and cellulose digestion by post-ruminal microorganisms in goats Jinzhen Jiao,a,b Pengpeng Wang,a,b Zhixiong He,a Shaoxun Tang,a Chuanshe Zhou,a Xuefeng Han,a Min Wang,a Duanqin Wu,a,b Jinhe Kanga and Zhiliang Tana∗ Abstract BACKGROUND: Post-ruminal digestion of fiber has received much less attention than its ruminal digestion. Using in vitro incubation techniques, the present study explored whether variations in fiber digestion occurred in different segments of the post-ruminal tract and whether fiber structure could influence its digestibility. A split plot design was conducted with gut segments (jejunum, ileum, cecum and colon) as main plot and substrates (neutral detergent fiber (NDF) and cellulose (CEL)) as subplot. RESULTS: With the same substrate, the final asymptotic gas volume (V F ), gas production at ti (Vti ), digestibility, microbial crude protein (MCP), total bacteria number (TBN), total short-chain fatty acids (TSCFA) and xylanase in incocula from the cecum and colon exceeded (P < 0.01) those in incocula from the jejunum and ileum, while the NH3 -N in the former was less (P < 0.01). For the same gut segment, the digestion of CEL was superior to NDF, as reflected in greater V F , Vti , maximum rate of gas production, digestibility, enzyme activities and SCFA but lower pH and NH3 -N. CONCLUSION: The current results imply that the intestinal contents from the cecum and colon have greater potential to digest fiber than those from the jejunum and ileum, and CEL is more easily digested in the post-ruminal tract than NDF. c 2013 Society of Chemical Industry Keywords: digestion; fiber; goat; in vitro; post-ruminal tract
INTRODUCTION
J Sci Food Agric 2014; 94: 1745–1752
the extent of fiber digestion at different segments of the postruminal digestive tract. It is hypothesized that there are significant variations in fiber digestion at different segments of the postruminal tract, the extent of which is influenced by fiber structure. Therefore the objective of this study was to evaluate the potential of post-ruminal microbes to utilize different fiber sources (neutral detergent fiber (NDF) and cellulose (CEL)) at the jejunum, ileum, cecum and colon using in vitro incubation techniques based on relevant digesta from goats.
∗
Correspondence to: Zhiliang Tan, Key Laboratory for Agro-Ecological Processes in Subtropical Region, Hunan Research Center of Livestock and Poultry Sciences, South-Central Experimental Station of Animal Nutrition and Feed Science in Ministry of Agriculture, Institute of Subtropical Agriculture, The Chinese Academy of Sciences, Changsha, Hunan 410125, China. E-mail: [email protected]
a Key Laboratory for Agro-Ecological Processes in Subtropical Region, Hunan Research Center of Livestock and Poultry Sciences, South-Central Experimental Station of Animal Nutrition and Feed Science in Ministry of Agriculture, Institute of Subtropical Agriculture, The Chinese Academy of Sciences, Changsha, Hunan, 410125, China b University of the Chinese Academy of Sciences, Beijing, 100049, China
www.soci.org
c 2013 Society of Chemical Industry
1745
Ruminants can utilize large quantities of low-quality roughages as energy sources by microbial degradation of fiber in the gastrointestinal tract (GIT). Anaerobic degradation of fiber is a complex process involving several functional groups of microorganisms, especially fibrolytic bacteria (e.g. Ruminococcus albus, Fibrobacter succinogenes and Ruminococcus flavefaciens),1 and results in the production of many end products (e.g. shortchain fatty acids (SCFA), CO2 , CH4 and H2 ).2 While digestion of fiber in the rumen has been intensively studied, its post-ruminal digestion has received much less attention. However, the total quantity of wet digesta at the hindgut of sheep is reported to be 700–1200 g, equivalent to 15–26% of the mass in the rumen (4.5 kg).3 In the post-ruminal tract, total bacterial counts reach 1012 cells g−1 digesta.4 These microbes exert numerous physiological functions, including fermentation of fiber not digested in the forestomach. Tan et al.5 reported that although ruminal fiber digestion was depressed by increased dietary non-structural carbohydrates in sheep, this could be partially compensated by increased microbial fermentation in the hindgut. Although post-ruminal microbes can utilize considerable fiber, limited knowledge is available about
www.soci.org
MATERIALS AND METHODS Animals and experimental design Three adult Liuyang black wether goats (1.5 years of age with an average body weight of 20.1 ± 1.2 kg), a local breed of southern China, were used. The goats were housed individually, had free access to drinking water and were fed an experimental diet (forage/concentrate, 55:45 w/w) formulated to meet 1.4 times metabolizable energy (ME) requirements for maintenance according to the feeding standard of Chinese goats. The diet was composed of 550 g rice straw, 100 g soybean meal, 284 g corn meal, 4 g CaH2 PO4 , 0.2 g CaCO3 , 43 g fat powder, 9 g NaCl and 10 g premix kg−1 dry matter (DM) and contained 91.4 g crude protein, 2.3 g calcium, 1.6 g phosphorus, 414 g NDF and 9.38 MJ ME kg−1 DM. There was a total adaptation period of 3 weeks, with the first week to accustom the goats to the diet and the last 2 weeks with regular feeding, before the first slaughter. Animals were fed the diet as two equal amounts daily at 08:00 and 18:00, and no refusals occurred after the adaptation period. The three goats were slaughtered at 1 week intervals. The digesta collected from the jejunum, ileum, cecum and colon of each goat provided the four inocula. After tying off the jejunum, ileum, cecum and colon with plastic strips, the entire GIT was removed and taken to the laboratory within 1 h under an atmosphere of CO2 to maintain anaerobic conditions. The various digesta were weighed and diluted 1:5 (w/v) with pre-warmed (39 ◦ C), anaerobic, sterile phosphate-buffered saline (pH 7.4). The diluted materials were then homogenized using a hand-mixer for 60 s and strained through four layers of cheesecloth. The resulting liquids were used as inocula. The experimental scheme was as follows: 2 fermented substrates (NDF and CEL) of 0.4 g each × 4 gut segments (jejunum, ileum, cecum and colon) × 4 in vitro incubation time points (12, 24, 48 and 72 h). The three goats were treated as replicates, and each incubation time point had three bottles as repeats. The experiments were approved by the Animal Care Committee, Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha, Hunan, China. Preparation of culture medium, substrates and inocula A semi-defined medium, modified from Lowe et al.,6 in which no carbon source was added, was used for in vitro cultivation studies. The medium was prepared by mixing 76 mL of basal solution, 4 mL of bicarbonate buffer, 1 mL of reducing agent and 1 mL of vitamin/phosphate buffer solution. NDF was extracted from rice straw by the method of Yang and Xie.7 CEL was extracted from rice straw by the method of Sun et al.8 Following overnight feed withdrawal, goats were anaesthetized and slaughtered. Each inoculum was then dispensed into the pre-warmed bottles containing substrate and medium.
1746
In vitro incubation procedures From preliminary studies in vitro (data not shown), an NDF dose of 0.4 g was chosen as appropriate, since it allowed good growth of colonic microbes with suitable gas production (∼50 mL) during 72 h of in vitro incubation. The NDF proportion of the post-ruminal digesta in vivo ranged from 5.5 to 19.0%, equivalent 0.11–0.38 g NDF per 2 g digesta. In our in vitro experiment, 2 g of digesta was collected and diluted with phosphate-buffered saline (pH 7.4) (w/v 1:5), homogenized using a hand-mixer for 60 s and strained through four layers of cheesecloth. Then 10 mL of filtrate (with negligible amount of NDF) was used as inoculum. In practice, the 0.4 g of NDF with 10 mL of inoculum used in vitro was similar to
wileyonlinelibrary.com/jsfa
J Jiao et al.
both the NDF and microbial content found in 2 g of digesta in vivo. For the in vitro incubation, 0.4 g of NDF or CEL was incubated at 39 ◦ C under strictly anaerobic condiitons in 132 mL glass bottles with 30 mL of medium (pH adjusted to 6.8–7) and 10 mL of inoculum. The bottles were placed in a shaking water bath with 50 movements min−1 at 39 ◦ C and gas volumes were recorded every 1 h for 72 h. Bottles were removed from fermentation at 12, 24, 48 and 72 h of incubation, and fluid samples from the bottles were immediately stored at −20 or −80 ◦ C for later analyses. Bottles filled with 10 mL of inoculum (jejunum, ileum, cecum or colon) and 30 mL of medium, but without NDF and CEL, were run simultaneously as controls. The residual solutions after 12, 24, 48 and 72 h of fermentation were filtered into pre-weighed Gooch filter crucibles, dried at 105 ◦ C for 24 h and weighed for the determination of in vitro NDF or CEL digestibility. Calculation of in vitro gas production parameters The in vitro gas curves were fitted using NLREG Version 5.0,9 which employs a combination of the Gauss–Newton and Levenberg–Marquardt methods.10 In general, the logistic–exponential (LE) model suggested by Wang et al.11,12 was fitted without major problems, although initial values were selected based on visual inspection of the plots in order to reach convergence and limit the number of calculations and calculation time. The LE model was V = VF 1 − exp (−kt) / 1 + exp (b − kt) where V is the cumulative gas production (mL) at time point t, V F is the final asymptotic gas volume (mL), k is the rate of gas production and b is a shape parameter. Then FRD0 (initial fractional rate of degradation) = k/[1 + exp(b)], ti (time at which inflexion of the LE model occurred) = b/k, Vti (gas production at ti ) and λ (derived lag time) = ti − Vti /RGti were derived using the LE model. Chemical analysis of in vitro culture fluids After incubation for 12, 24, 48 and 72 h, the bottle contents were filtered through four layers of cheesecloth. The fluids obtained after filtration were used for the following analyses. The pH of culture fluids was measured immediately (pH meter model 2000, Beckman Instruments, Fullerton, CA, USA). The NH3 N concentration was determined by the phenol/hypochlorite method using a UV–visible spectrophotometer at 550 nm.13 SCFA (i.e. acetate, propionate, butyrate, valerate, isobutyrate and isovalerate) concentrations were analyzed using a gas chromatograph (7890A, Agilent, USA) according to the method of Garcia-Gonzalez et al.14 Microbial crude protein (MCP) was determined using the trichloroacetic acid method suggested by Licitra et al.15 The activities of carboxymethylcellulase (CMCase, an endoglucanase) and xylanase were assayed based on procedures described by Rey et al.16 An enzyme activity unit (U) was defined as the amount of enzyme required to release 1 µmol reducing sugars (xylose or glucose equivalent) min−1 mL−1 culture fluid. Quantitative polymerase chain reaction (qPCR) determination of total bacteria number (TBN) Genomic DNA of in vitro culture fluids (before filtration) was extracted from 1 mL aliquots using a QIAamp DNA Stool Minikit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The purified DNA was quantified using NanoDrop ND1000 (NanoDrop Technologies, Wilmington, DE, USA). The
c 2013 Society of Chemical Industry
J Sci Food Agric 2014; 94: 1745–1752
Fiber digestion in post-ruminal tract
www.soci.org
forward primer 5 -CGGCAACGAGCGCAACCC-3 and reverse primer 5 -CCATTGTAGCACGTGTGTAGCC-3 cited by Denman and McSweeney17 were used in qPCR to quantify total bacteria in culture fluids. The purified PCR product amplified with this pair of primers was used to generate plasmid DNA, followed by construction of the standard curve between Ct value and plasmid copies through serial dilution of plasmid copies. The qPCR assays were performed on an ABI 7900HT system (Applied Biosystems, Foster City, CA, USA) in a total volume of 10 µL using SYBR Premix Ex Taq (Takara, Ostu, Japan) following the suggested protocol. The copy numbers of total bacteria were determined by relating the Ct value to the standard curves. The final copy numbers of targeted total bacteria (TBN) mL−1 culture fluid were calculated as described by Zhou et al.18 Statistical analysis The repeated data (data from three bottles) were averaged to be analyzed using the MIXED procedures of SAS (SAS Institute, Cary, NC, USA) according to a split plot design. Gut was included in the main plot and substrate in the subplot. For in vitro gas production parameters, the model consisted of random (goat within (substrate, gut, substrate × gut)) and fixed (substrate, gut, substrate × gut) effects. Digestibility, MCP, TBN, enzyme activities, pH, NH3 -N and SCFA were analyzed using the MIXED procedures of SAS in a REPEATED statement with incubation time as the repeated factor and goat as the subject. The compound symmetry was used as the covariance structure. The model included as random effects goat and substrate, gut and time, plus their interactions within goat. Fixed effects were substrate, gut and time, plus their interactions. Contrast statements included the main effect comparing fermented substrate (NDF vs CEL). Digestibility, MCP, TBN and enzyme activity data were reported at each time point, whereas pH, NH3 -N and SCFA data were reported as mean values of four time points.
RESULTS Gas production of NDF and CEL by digesta from post-ruminal tract The V F and Vti of NDF and CEL were similar in inocula from the cecum and colon but exceeded (P < 0.01) those in inocula from the ileum, while the ti and λ of CEL were greater in inocula from the ileum than in inocula from the cecum and colon (Table 1). Maximum rates of gas production (i.e. the rates at ti ) of NDF and CEL were similar between inocula from the cecum and colon and occurred at about 40 and 30 h of incubation respectively. These rates exceeded those observed in the ileal inocula, which occurred at approximately 48 and 56 h of incubation respectively. Little gas was produced in the jejunal inocula during 72 h of incubation. Furthermore, CEL had higher (P < 0.01) V F and Vti compared with NDF.
J Sci Food Agric 2014; 94: 1745–1752
MCP and TBN of NDF and CEL by digesta from post-ruminal tract For both NDF and CEL, MCP levels in the hindgut inocula were greater (P < 0.01) than those in the small intestinal inocula after 48 and 72 h of incubation (Table 3). Additionally, for both NDF and CEL, TBN levels in the two hindgut inocula were greater (P < 0.01) than those in the two small intestinal inocula (Table 4). When incubated with NDF, after 24, 48 and 72 h of incubation, TBN levels in the small intestinal inocula were less (P < 0.05) than those in the hindgut inocula. When incubated with CEL, after 12, 24 and 48 h of incubation, TBN levels in the hindgut inocula exceeded (P < 0.05) those in the small intestinal inocula. Enzyme activities related to NDF and CEL digestion in post-ruminal tract inocula When incubated with NDF, after 72 h of incubation, the activity of CMCase in the ileal inocula was greater (P < 0.01) than that in the jejunal inocula (Table 5). When incubated with CEL, after 48 h of incubation, the greatest activity of CMCase was present in the cecal inocula and exceeded (P < 0.01) those observed in inocula from the jejunum and ileum; after 72 h of incubation, the greatest activity of CMCase was observed in inocula from the ileum, followed by inocula from the jejunum and cecum, with the colonic inocula being the least active. Furthermore, the activities of CMCase were greater (P < 0.05) for CEL than for NDF for all post-ruminal tract inocula except those from the jejunum. For both NDF and CEL, the activities of xylanase were greater (P < 0.01) in inocula from the cecum and colon than in inocula from the jejunum (Table 5). Only in the cecal inocula was the activity of xylanase less (P = 0.035) for NDF than for CEL. pH, NH3 -N and SCFA from NDF and CEL in post-ruminal tract inocula When incubated with CEL, the pH in the jejunal inocula was greater (P < 0.01) than those in inocula from the cecum and colon (Table 6). Moreover, for the hindgut inocula, NDF had greater (P < 0.05) pH when compared with CEL. Similarly, when incubated with NDF, NH3 -N concentrations in inocula from the ileum, cecum and colon were less (P < 0.01) than that in inocula from the jejunum. When incubated with CEL, the NH3 -N concentration in the ileal inocula was less (P < 0.01) than that in the jejunal inocula but greater (P < 0.05) than those observed in the hindgut inocula. The SCFA concentrations were greater (P < 0.05) in the hindgut inocula than in the small intestinal inocula for both NDF and CEL. In contrast, the acetate/propionate (A/P) ratios in the small intestinal inocula were greater (P < 0.01) than those in the hindgut inocula. Furthermore, in the jejunual inocula, CEL had greater (P < 0.05) butyrate and A/P ratio than NDF, while, in the ileal inocula, NDF had less (P < 0.01) butyrate than CEL. For both hindgut inocula, CEL produced greater (P < 0.05) acetate, propionate and total SCFA (TSCFA) compared with NDF.
DISCUSSION The post-ruminal tract is a very complex system involving interactions between the host and microorganisms and also
c 2013 Society of Chemical Industry
wileyonlinelibrary.com/jsfa
1747
Digestibility of NDF and CEL by digesta from post-ruminal tract Compared with the small intestine segments (jejunum and ileum), digestibilities of both NDF and CEL were greater (P < 0.01) in the hindgut segments (cecum and colon) (Table 2). After 48 h of incubation, digestibilities of NDF and CEL were less (P < 0.01) within the small intestine sections than within the hindgut sections. After 72 h of incubation, digestibilities of NDF and CEL in the ileal inocula were 35.4 and 52.2% respectively, greater (P < 0.01) than
those observed in the jejunal inocula but less (P < 0.01) than those observed in inocula from the cecum and colon. Furthermore, CEL had greater digestibility than NDF in inocula from the jejunum (P = 0.031), ileum (P = 0.011), cecum (P = 0.0003) and colon (P = 0.0003).
www.soci.org
J Jiao et al.
Table 1. In vitro gas production parameters of neutral detergent fiber (NDF) and cellulose (CEL) estimated from logistic–exponential (LE) model with inocula from jejunum, ileum, cecum and colon Gut Parametera −1
V F (mL g
Substrate
Jejunumb
NDF CEL NDF CEL NDF CEL NDF CEL NDF CEL
— — — — — — — — — —
)
FRD0 (% h−1 ) ti (h) Vti (mL g−1 ) λ (h)
Ileum 45.7cC 99.3cA 0.18 0.06 48.3 56.2a 22.7cC 49.6cA 31.6 46.3a
P values Cecum
Colon
125.6aC 183.1aA 0.08 0.12 40.4 33.7c 62.3aC 90.6aA 25.0 20.8c
115.8aC 176.1aA 0.09 0.08 41.2 34.6c 57.3aC 87.5aA 24.8 21.9c
Gut
Substrate
Gut × substrate
