Effects of Starch Content on Hydrogen and Methane Productions, Rumen Fermentation and Microbial Protein Synthesis During in Vitro Ruminal Culture

Background: Starch has faster rate of rumen fermentation than ber, and always causes a rapid increase in ruminal molecular hydrogen (H 2 ) partial pressure and microbial protein synthesis, which may promote other H 2 sinks to compete H 2 from methanogenesis. The study was designed to investigate the effects of increasing starch content on methane (CH 4 ), hydrogen gas (gH 2 ) production, rumen fermentation, metabolic hydrogen ([H]) production, microbial protein (MCP) synthesis through in vitro ruminal batch incubation. Methods: Seven different treatments was prepared by replacing corn straw with corn grain, and starch content were 72, 185, 297, 410, 525, 634 and 747 g/kg DM. Results: Elevating starch content increased DM degradation (P linear < 0.001), and decreased the CH 4 (P linear and P quadratic < 0.001) and gH 2 (P linear < 0.001) productions relative to DM degraded. Elevating starch content increased VFA concentration (P linear < 0.001), propionate molar percentage (P linear < 0.001; P quadratic = 0.001) and MCP concentration (P linear and P quadratic < 0.001), and decreased acetate molar percentage (P linear < 0.001), acetate to propionate ratio (P linear < 0.001) and estimated net [H] production relative to DM degraded (P linear < 0.001). Elevating starch content decreased molar percentage of [H] utilized for CH 4 (P quadratic = 0.003) and gH 2 (P linear < 0.001) production. Conclusion: Increasing starch content alters rumen fermentation pathway from acetate to propionate production with reduction in eciency of [H] production, promotes H 2 utilization with enhanced MCP synthesis and leads to the reduction in eciency of CH 4 and gH 2 production.


Background
Methane (CH 4 ) is an important greenhouse gas, which receives great attention worldwide for its impact on global climatic change [1]. Ruminant CH 4 emissions accounts for approximately 17% of the global CH 4 emissions [2]. Furthermore, CH 4 emission is also an important energy loss which represents 2-12% of dietary gross energy and strongly associated with e ciency of ruminants production [3,4]. Thus, CH 4 mitigation is bene cial to the environment and animal performance.
Molecular hydrogen (H 2 ) is the precursor of ruminal methanogenesis and mainly produced during the fermentation of carbohydrate to volatile fatty acid (VFA) [5]. Comparing with forage ber, starch has faster rate of rumen fermentation and ATP production [6] and is always accompanied with a rapid increase in ruminal H 2 partial pressure [7]. Other H 2 sinks, such as reductive acetogenesis, biohydrogenation, propionate production and microbial protein (MCP) synthesis [8], can be promoted and serve as H 2 competitors in the rumen, when ruminal H 2 partial pressure is increased [9]. Studies have investigated relationship between fermentation pathways and methanogensis [7,10]. However, few studies have conducted to investigate contribution of H 2 utilization by methanogenesis on the metabolic hydrogen ([H]) generated by VFA production, when different types of carbohydrates was fermented. Loading [MathJax]/jax/output/CommonHTML/fonts/TeX/fontdata.js We hypothesized that increasing starch content could decrease the contribution of H 2 disposal by methanogenesis, so that fermentation pathway might not be only cause of the decreased CH 4 production in starchy diet. In vitro ruminal batch culture was employed, as it is effective method to measure the actual net fermentation products. Increasing starch content was then achieved by replacing corn straw with corn grain. We measured the kinetics of total gas, CH 4 and H 2 gas (gH 2 ) production, fermentation end products, estimated net [H] production and MCP after 48-h in vitro ruminal fermentation.

Materials And Methods
The experiment was approved by the Animal Care Committee, Institute of Subtropical Agriculture, the Chinese Academy of Sciences, Changsha, China.

Experimental design
The corn grain and corn straw (Table 1) were employed to generate different starch contents for the incubated substrates. Seven treatments were ratios of corn grain to corn straw being 6:0, 5:1, 4:2, 3:3, 2:4, 1:5 and 0:6, and grounded to pass 1-mm aperture sieve. The starch content of seven treatments were 72, 185, 297, 410, 525, 634 and 747 g/kg DM. (GC, Agilent 7890 A, Agilent Inc., Palo Alto, California, USA) via a computer-controlled three way solenoid valve, the released gas was automatically vented into a GC for measuring CH 4 and hydrogen gas concentrations. Gas production (GP), CH 4 and gH 2 accumulations were calculated using the equation equation described by Wang et al. [13].
In vitro ruminal fermenation was stopped at 48 h. About 2 mL of liquid without visible particles were collected from each bottle and centrifuged at 15,000 g for 10 min at 4°C. The supernatant (1.5 mL) was acidi ed using 0.15 mL of 25% (w/v) metaphosphoric acid, and stored at -20°C for analysis of VFA and ammonia. The pH was measured immediately with a portable pH meter (Starter 300; Ohaus Instruments Co. Ltd., Shanghai, China). About 8 ml of samples were collected for measuring microbial protein after intense shaking of the bottle to ensure that representative portions of liquid and particle fractions. Solid residues were ltered into pre-weighed Gooch lter crucibles, dried at 105°C to constant weight and weighed to determine degradation of incubated substrates.
Each run had four replicates for each treatment. Two bottles were used for measuring pH and DM degradation, and the other two bottles were used for obtaining samples for measuring fermentation endproduct and microbial protein. Each run was repeated three times, each on different days, so that each treatment was conducted in triplicate.

Sample analyses
The dry matter (DM) content was determined by drying at 105℃ for 24 hours in an oven, and the organic matter (OM) content was determined by ashing at 550℃ for 12 hours in a mu e furnace. Gross energy (GE) was measured using an isothermal automatic calorimeter (5EAC8018; Changsha Kaiyuan Instruments Co. Ltd, Changsha, China). The contents of crude protein (CP) (N × 6.25) in feed samples were determined according to procedures of AOAC [14]. The contents of neutral detergent ber (NDF) and acid detergent ber (ADF) in feed samples were determined according to the methods described by Van Soest et al. [15] and expressed as inclusive of ash. Heat stable α-amylase was added to for NDF analysis.
Volatile fatty acid concentration was measured according to the procedure described by Wang  and centrifugation (150 × g for 10 min), and microbial nitrogen production was measured colorimetrically according to Bradford [19], Using a Coomassie brilliant blue kit (Build a biopharmaceutical research institute, Nanjing, China).

Calculations and statistical analysis
The logistic-exponential model [20] was employed to analyze the kinetics of total gas and CH 4 production by using the Nonlinear Regression Analysis Program (NLREG, version 5.4) [21], and was expressed as follows: Where GPt is the accumulated gas or CH 4 production at time t (ml/g); VF is the nal asymptotic gas or CH 4 production (ml/g); k is the fractional rate of gas or CH 4 production (/h); b is the shape parameter.
The kinetics of gH 2 production (V H2 ) was analyzed using the equations provided by Wang et al. [22], which was expressed as follows: Where V H2t is the accumulated H 2 gas production at time t (ml/g); VF H2 is the nal asymptotic H 2 gas volume (mL/g), b H2 and c H2 are shape parameters of H 2 gas curve without dimension, k H2 is the fractional rate of H 2 gas formation (/h), µ H2 is the fractional rate of H 2 gas re-solution (/h), and lag H2 is discrete lag The data were analyzed using general linear model (GLM) with SPSS 26.0 (Chicago, IL, USA), and are presented as mean and SEM. The analytic model included treatment (n = 7) as xed effect and run (n = 3) as random effect, and were analyzed for linear or quadratic responses to starch content using orthogonal contrasts. Statistical signi cance was considered at P ≤ 0.05 with 0.05 < P ≤ 0.10 considered as a tendency.

Discussion
It's well-known that corn grain and corn straw have different types of carbohydrate components and are rich in starch and ber respectively. Ruminal degradation rate of carbohydrates depends on their monosaccharide and bond composition, molecular size, sugar arrangement at molecular level and physical morphology [23,24]. Starch is mainly formed by α-1,4 glycosidic bond and easily hydrolyzed by enzyme [6,25], while ber is the major component of cell wall, formed by β-1,4 glycosidic bond, and contains "crystals" which formed by the connection of cellulose macromolecules through hydrogen bonds and resistant to acid and enzyme hydrolysis [26]. Comparing with starch, cellulose and hemicellulose are less susceptible to microbial degradation in rumen [5]. In our study, increasing starch content linearly increased substrate degradation, gas production and fractional rate of gas production during in vitro 48-h batch incubation. These results were consistent with previous studies [27,28], which report that starch has greater and faster rumen degradability in comparison with ber. Methane is the end-product during ruminal carbohydrate fermentation [3]. It is not surprising that elevated starch content linearly increased fractional rate of CH 4 production, as starch has faster rate of fermentation than straw ber. The amount of CH 4 produced is related to the degree of substrate degradation and e ciency of CH 4 produced (i.e. CH 4 produced per unit of substrate degraded) [5]. Starch and ber fermentation have different e ciency of CH 4 produced. Readily fermentable feed high in starch leads to lower CH 4 production, while slowly fermentable feed in cellulose and hemicellulose causes higher CH 4 production [29][30][31]. In the present study, the elevated starch content quadratically increased the 48-h and nal asymptotic volume of CH 4 production, but linearly decreased amount of CH 4 produced per unit of DM degraded. Although starch had greater fermentation rate than straw ber, reduced CH 4 production in starchy treatment could be caused by the reduction in e ciency of CH 4 produced.
Hydrogen is produced during carbohydrate fermentation to VFA and mainly consumed by methanogens to produce CH 4 as the end product [32]. The unused H 2 will be evolved from liquid to gas phase in the headspace and vent to air. Normally, the ruminal H 2 partial pressure is very low to facilitate the rumen fermentation [5]. Rooke et al. [33] report that H 2 emitted just accounts for less than 2% of the estimated total H 2 production from fermentation in beef cattle. In our study, both 48-h and nal asymptotic gH 2 production was less than 1 ml/g DM, indicating that most of H 2 produced were utilized by methanogens to produce CH 4 .
Furthermore, starch and ber had different pro le of gH 2 production. For example, elevated starch content increased fractional rate of gH 2 consumption, although fractional rate of gH 2 production was not altered.
Elevated starch content also quadratically decreased the 48-h and nal asymptotic volume of gH 2 production and linearly decreased amount of gH 2 produced per unit of DM degraded. We proposed that the starch exhibited lower e ciency of H 2 production than straw ber, which can be likely to be related to their different pathways of rumen fermentation.
Fermentation of feed rich in starch produces more propionate and butyrate, and less acetate than feed rich in cellulose and hemicellulose [5,34]. In our study, elevated starch content linearly increased propionate molar percentage and reduced acetate molar percentage and acetate to propionate ratio. DMD, dry matter degradation; VF GP , the nal asymptotic volume of total gas production; b, shape parameter of gas production; k GP , the fractional rate of gas production; VF CH4 , the nal asymptotic volume of CH 4 production; k CH4 , the fractional rate of CH 4 production; DDM, degraded dry matter; VF H2 , the nal asymptotic volume of hydrogen gas production; b H2 , shape parameter of hydrogen gas; k H2 , the fractional rate of hydrogen gas production; µ H2 , the fractional rate of hydrogen gas utilization. P NH2 , estimated net [H] production; R NH2 , estimated net [H] production relative to the amount of total VFA produced.