Introduction

Biodiesel production has increased worldwide, as well as the volume of co-products generated, particularly crude glycerin.

Brazilian production of biodiesel was approximately 3.4 billion L in 2014, which generated about 311 million L of crude glycerin (ANP, 2014). There is no feasible alternative to absorb this volume of crude glycerin in the existing Brazilian National Policy for the Production and Use of Biodiesel. Utilizing glycerin as a feed ingredient in ruminant diets could potentially inhibit fat degradation by bacteria and may promote ruminal passage of total lipid content, thereby providing higher proportions of beneficial unsaturated fat for incorporation into beef products (^{Krueger et al., 2010}). Glycerin retention within the rumen is directly associated with its absorption. It can be absorbed from the rumen in significant amounts (Krueger et al., 2010; Omazic et al., 2015). Omazic et al. (2015) reported that glycerin also disappeared via microbial digestion and outflow from the rumen through the omasal orifice. Absorbed glycerin is metabolized in the liver and requires glycerin kinase (Hagopian et al., 2008), an enzyme responsible for channeling glycerin into the triose phosphate step of glycolysis/gluconeogenesis.

Gross energy of glycerin ranges from 3 to 6 Mcal/Kg, depending upon its composition (^{Mendoza et al., 2010}, Meurer et al., 2012; ^{Françozo et al., 2013}; ^{Eiras et al., 2014}). Since glycerin has high-energy content - similar to cornstarch (^{Yang et al., 2012}) - it could potentially be used for animal feed. Therefore, it is necessary to evaluate crude glycerin as an alternative energy feedstuff to replace corn grain in ruminant diets.

The objective of this study was to evaluate the effect of increasing levels of crude glycerin replacing corn in corn-alfalfa hay diets on digestibility, cumulative gas production and ruminal fermentation kinetics using in vitro techniques for true digestibility and cumulative gas production.

Material and methods

Ethical considerations

Animal care procedures throughout the study followed protocols approved by the Ethics Committee for Animal Use (CEUA) at the University of Rio Grande do Sul, Brazil (number 18442/2010).

Experimental design

The dietary treatments consisted in the inclusion of 0, 4, 8, and 12% of crude glycerin replacing ground corn (dry matter basis; DM basis). The experimental diets were composed of 60% alfalfa hay and 40% concentrate (DM basis). One gram (dry matter) of composite sample was added to 250 mL fermentation bottles. Four replicates were used per treatment. Four blank fermentation bottles were also incubated (i.e. no substrate), totaling twenty bottles for the experiment. Bottles were kept in an oven at 39 °C. One hundred mL of a mixture containing culture medium (^{Goering and Van Soest, 1970}), also maintained at 39 °C, and saturated with CO₂ were added to the bottles. The bottles were saturated with CO₂, closed and kept in the incubator until 25 mL of inoculum was added to each bottle. The bottles were then placed in an incubator at 39 °C for 48 hours. The nutritional composition of ingredients and the chemical composition of experimental diets are described in Tables 1 and 2, respectively.

Table 1 Nutritional composition of ingredients. 

*Methanol content lower than 45 mg/L. Dry matter (DM), organic matter (OM), crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF),gross energy (GE), acid detergent lignin (ADL) and glycerin contents.

Table 2 Chemical composition of the experimental diets. 

Dry matter (DM), organic matter (OM), crude protein (CP), neutral detergent fiber (NDF), ether extract (EE), gross energy (GE) contents.

Samples of the incubated feedstuffs (alfalfa hay and ground corn) were analyzed for dry matter (DM; AOAC, 1990; method 934.1 and 930.15), organic matter (OM, AOAC, 1975; method 7.010), and crude protein (CP, AOAC, 1990; method 988.05). Neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL) were determined according to ^{Van Soest et al. (1991}). A specialized laboratory evaluated the nutritional composition of crude glycerin (B.E.T. Laboratories do Brasil S/C Ltda). All analyses were performed in triplicates.

Two Texel sheep (60 ± 1.2 Kg average body weight) fitted with a rumen fistula were used as inoculum donors. Two hours after the animals received the morning meal, ruminal material was collected (1 part of liquid to 1 part of solid). One hundred mL of buffer solution plus 25 mL of rumen fluid were incubated.

In vitro cumulative gas pressure was automatically measured every 10 minutes for 48 hours, using an automatic gas-measuring equipment (ANKOM Gas Production System, Macedon, NY, USA), equipped with a pressure transducer attached to each bottle that transmits by radiofrequency the cumulative gas pressure values to a computer. The bottles were automatically vented every 10 minutes. Four runs were carried out with two replicates per treatment. Gas production values were corrected for the blanks. Gas production data were fitted to the dual-pool logistic model (^{Schofield et al., 1994}). The model parameters (A, B, C, D, E, and F) were estimated by the iterative method of Marquardt using the NLIN procedure of SAS (SAS Institute, Cary, NC, USA; 2002). The partitioning factor (PF) was determined according to Makkar (2004), where: PF = mg of TDOM (truly degraded organic matter)/mL gas produced.

After 48 hours of incubation, fermentation was stopped by putting the bottles in ice. The bottles were then centrifuged, the supernatant was removed, and 50 mL of neutral detergent solution (^{Goering and Van Soest, 1970}) was added to the bottles, which were maintained at 90 °C in an oven during 16 hours to extract the neutral detergent soluble fraction (^{Chai and Udén, 1998}). The residue was filtered in sintered glass crucibles with coarse pore diameter (100 to 160 microns) and placed in an oven at 105 °C for 12 hours, weighed, and burnt in a muffle at 450 °C for 5 hours.

The in vitro organic matter true digestibility (IVOMTD) and in vitro neutral detergent fiber digestibility (IVNDFD) were calculated as the difference between the incubated OM or NDF and the non-digested OM or NDF residue remaining in the crucibles (^{Goering and Van Soest, 1970}).

After removing from the incubator and centrifuging the bottles, and before adding the neutral detergent solution, 20 mL of the supernatant was removed and acidified for ammonia nitrogen (NH₃-N) analysis. NH₃-N concentration was determined by distillation with magnesium oxide (AOAC, 1990; method 990.03).

Statistical analysis

Maximum gas production from the rapidly and slowly degradable fractions of organic matter (^{Schofield et al., 1994}), their respective degradation rates and lag time were estimated by the NLIN procedure. The effects of increasing crude glycerin inclusion levels in the diet on the estimated parameter, in vitro digestibility with 48 hours of incubation, ammonia nitrogen and partitioning factor were analyzed using the GLM procedure. Means were compared by the REG procedure. In vitro cumulative gas production was evaluated using the GLM procedure as repeated measures. All statistical procedures were performed using the SAS package (Statistical Analysis System, version 9.3, SAS Institute, Cary, NC, USA; 2002).

Results

As there was no significant effect (p>0.05) on the parameters evaluated, only the significant effects of glycerin levels (p<0.05) will be shown. Increasing crude glycerin levels to replace corn did not influence in vitro neutral detergent fiber digestibility (IVNDFD) nor in vitro true organic matter digestibility (IVTOMD) after 48 hours of incubation. Mean values for those parameters were 57.8 and 80.4%, respectively (Table 3).

Table 3 Effects of four inclusion levels of crude glycerin on the mean values of several parameters. 

1Ŷ = 48.53 + 5.44x; 2Ŷ = 41.13 - 2.03x; 3Ŷ = 40.75 + 3.87x; 4Ŷ = 9.66 - 0.22x. In vitro neutral detergent fiber digestibility (IVNDFD, %), in vitro true organic matter digestibility (IVTOMD, %), maximum gas production of the rapidly (A, mL) and slowly (D, mL) degradable fractions of organic matter and their respective degradation rates (B and E, %/hours), lag times (C and F, hours), partitioning factor (PF - mg TDOM/mL gas, 24 hours of incubation), and ammonia nitrogen values (NH3-N, mg/dL).

Total cumulative gas production ranged between 82.1 (0% glycerin in the diet) and 123.1 (12% glycerin in the diet) mL/g of incubated OM (Figure 1). There was no effect of glycerin inclusion on cumulative gas production during the first hours of incubation. However, after 14 h of incubation, the inclusion of 12% glycerin yielded higher cumulative gas production (79.9 mL/g

OM) than the treatment with no glycerin (54.8 mL/g OM; p = 0.0613), and this pattern was observed until 30 hours of incubation. After 30 hours of incubation, the 12 and 8% crude glycerin levels produced more gas (118.2 and 107.7 mL/g OM, respectively) compared with the no-glycerin treatment (79.3 mL/g OM; p<0.01), and this trend was maintained until the end of 48 hours of incubation. The standard deviations of gas production values at 16, 30, and 48 hours of incubation were 15.03, 18.83, and 19.65, respectively.

Figure 1 Effect of crude glycerin inclusion level to replace the corn in the diet on the mean values of cumulative gas production (mL/g of OM incubated) during 48 hours of incubation. 

The partitioning factor, which integrates total cumulative gas production with substrate disappearance, is an indicator of fermentative efficiency. The crude glycerin levels used in the present study significantly decreased the PF (Table 3).

Maximum gas production of the rapidly (A) and slowly (D) degradable fractions of organic matter were affected by increasing levels of crude glycerin in replacement of corn (Table 3).

Mean degradation rates of the rapidly (B) and slowly (E) degradable fractions were 0.016 and 0.015 %/h, respectively, and were not affected by crude glycerin inclusion levels. The mean NH₃-N result obtained in the present study was 15.6 mg/dL (Table 3).

Discussion

The digestibility results are consistent with previous studies, which showed that feeding glycerin levels up to 12% of the total ration does not have an effect on nutrient digestibility or animal performance (^{Krueger et al., 2010}; ^{Meral et al., 2015}). On the other hand, ^{Abo El-Nor et al. (2010}) observed that glycerin decreased the in vitro organic matter digestibility and in vitro neutral detergent fiber digestibility in diets. These results suggest that glycerin can modulate digestion in a dose-dependent manner. ^{Drouillard (2012}) describes that the deleterious effects of glycerin on fiber digestion are evident because the inhibitory effects of higher crude glycerin levels on cellulolytic bacteria and fungi activity are clearly evident and may provide a plausible explanation for reduced fiber digestion. In this study, IVOMTD and IVNDFD were not affected by crude glycerin due to its low level in the diet, allowing optimum rumen fermentation, as growth, adhesion and cellulolytic activity were inhibited when glycerin was included at high concentration but not at low concentration (^{Benedeti et al., 2016}).

Total cumulative gas production values are low, but according to ^{Lee et al. (2011}), adding glycerin to alfalfa or corn tended to decrease the in vitro gas production compared to alfalfa or corn substrates when they were incubated without glycerin. ^{Krueger et al. (2010}) also reported a linear increase in gas production when glycerin was added to alfalfa hay at 10, 20 and 40% of the DM in vitro. The differences in the gas production profiles were not expected as corn and glycerin produce the same amount of gas (Lee et al., 2011), and the replacement of corn for glycerin should not cause modification. However, we can hypothesize that alfalfa hay, corn and glycerin combination could affect fermentation, suggesting the occurrence of associative effects. Corn grain has a hydrophobic protein matrix, which interferes negatively on its ruminal degradation rate (^{Larson and Hoffman, 2008}). However, alfalfa hay has a higher percentage of soluble fraction and lower NDF, thus more nonstructural carbohydrate content. These carbohydrates are associated with glycerin, which is rapidly metabolized (between 4 and 6 h), providing greater synchronism with the sources of nitrogen of rapid degradation, and consequently producing more gas.

Partitioning factors indicate fermentative efficiency, and therefore, higher PF values mean greater incorporation of degraded organic matter to the microbial mass, thereby increasing the efficiency of microbial protein synthesis. According to Makkar (2004), optimal PF values range between 2.74 and 4.41 mg TDOM/mL gas produced. The high PF values obtained in the present study indicate that fermentation conditions were good, stimulating production of microbial protein and resulting in high partitioning values.

The higher gas production of the rapidly degradable fraction relative to the slowly degradable one (81.2 vs 28.9 mL/g OM) indicates that glycerin increased the soluble fraction of the diet. This result is consistent with the findings by ^{Getachew et al. (2004}) and ^{Krueger et al. (2010}), who reported that maximum gas production of the rapidly degradable fraction (A) increased linearly as the amount of glycerin in the diet increased.

The B and E degradation rates may be considered as intrinsic characteristics of the feed, and depend on the chemical composition of the roughage, the maturation stage of the plant, and the cell wall structure. The degradation rate can influence the efficiency of microbial synthesis, thereby affecting gas production of each fraction of the incubated feed. According to ^{Caton and Dhuyvetter (1997}), there are usually no effects of dietary energy supplementation on substrate degradation rate, as observed in the results obtained in the present study, probably due to efficiency of energy use.

To be degraded in the rumen, the substrate must undergo an attachment process by rumen bacteria, known as colonization time or lag time, which allows enzymes to reach the substrate. In the present study, the inclusion of increasing glycerin levels did not affect the colonization time of bacteria to the rapidly degradable fraction (C, h) of feed. However, the colonization time to the slowly degradable fraction (F, h) of feed was showing direction of the effect, probably due to a direct relationship between the presence of highly fermentable glycerin and the colonization time.

In this study, the kinetics of gas production from glycerin fermentation were similar to results reported by ^{Lee et al. (2011}) and ^{Krueger et al. (2010}) who reported long colonization times. This indicates that kinetics of in vitro fermentation of glycerin seem to depend on the colonization of rumen microbes. In addition, fermentation of glycerin may be altered by the presence of other feed ingredients (Lee et al., 2011).

Fermentation kinetics of feedstuffs can be determined by the gas production and by buffering of volatile short-chain fatty acids (SCFAs) and depends on the relative ratio of soluble, insoluble but degradable, and non-degradable particles of the feed. In in vitro media, NH₃-N concentration is an indicator of protein degradability because there is no nitrogen absorption or recycling, compared with the in vivo rumen media (NRC, 1985).

The mean NH₃-N results obtained in the present study were above 5 mL/dl, which is, according to ^{Satter and Slyter (1974}), the concentration that maximizes microbial protein synthesis. These results are consistent with previous works showing that feeding glycerin in substitution for corn or barley grain does not affect NH₃-N concentration (Abo El- Nor et al.,2010; ^{Avila et al., 2011}).

In conclusion, inclusion of up to 12% crude glycerin (on dry matter basis) in replacement of corn did not affect diet digestibility. The greater volume of gas produced was observed at the highest inclusion level of glycerin, indicating that alfalfa hay, corn and glycerin combination could change fermentation, suggesting the occurrence of associative effects.

Conflicts of interest

The authors declare they have no conflicts of interest with regard to the work presented in this report.

Notes

To cite this article: Peripolli V, Prates ÊR, Barcellos JOJ, Costa JBG, Lopes RB, Camargo CM. Partial replacement of corn with glycerin: digestibility and ruminal fermentation kinetics by in vitro gas production. Rev Colomb Cienc Pecu 2016; 29:218-225