Nutritional value of coastal Bermuda grass in swine

Nutritional value of
coastal Bermuda grass in swine

I.B. Kim, B. Hansen[1], J. Hansen[2], R. Dvorak[3], and T. van Kempen

Summary

The data obtained suggest that the pigs used in these experiments had difficulty digesting Bermuda grass. This inability to digest Bermuda grass was strong in pigs not adapted to high fiber diets, but was also strong in gestating sows that were fed low levels of Bermuda grass. Higher levels, however, appeared to be digested better. Fibrozyme™ improved the energy digestibility of Bermuda grass in finishing pigs to the extent that it is technically feasible to use Bermuda grass at 14%. Whether this is economically attractive is outside the scope of this experiment.

Introduction

In the South-eastern US, swine waste is typically flushed from buildings. This leads to a rather dilute waste stream, commonly less than 2% dry matter. This waste ends up lagoons; clay-lined ponds designed to maximize degradation of organic matter. During the holding period in these lagoons, the chemical and biological oxygen demand of waste is substantially reduced as organic matter is digested by micro-organisms forming carbon dioxide and methane. Nitrogen is partially converted to nitrogen gas and ammonia. What remains can be divided into two, relatively stable, fractions. Sludge that settles to the bottom of the lagoon and which contains the majority of the phosphorus, and the liquid fraction that contains a substantial portion of the remaining nitrogen.

The sludge fraction is typically only utilized after substantial accumulation has occurred. The liquid fraction, which accumulates much faster, especially in periods of heavy rainfall, has to be utilized frequently. However, as it is low in dry matter it is of relative low fertilization value, which makes long-range transport cost-prohibitive. The liquid fraction is applied to irrigation plots, on which a crops are grown based on their ability to utilize nitrogen, as nitrogen sets the land-application rate. Coastal bermuda grass is often selected for this purpose. Bermuda grass can utilize approximately 200 pounds of nitrogen per acre, and is considered the (practical) crop with the highest ability to utilize nitrogen.

In order for the nutrient cycle to work properly, this Bermuda grass subsequently has to be utilized, and ideally it is fed to livestock such as cattle. However, this route of Bermuda grass utilization is unable to handle the large supply and variable quality of Bermuda grass produced by the intensive swine industry. As an alternative to feeding Bermuda grass to ruminants the possibility to feed Bermuda grass to swine is of interest. However, as Bermuda grass is a rather fibrous feed ingredient, its nutritional value is likely low unless it is fed with fiber degrading enzymes or is fed to animals that have developed a substantial rate of fiber degradation through fermentative action in the large intestines, thus older animals.

The experiments outlined here were designed to evaluate the nutritional value of Bermuda grass in heavy finisher pigs in the presence and absence of a fiber-degrading enzyme and in gestating sows that should have developed substantial fermentative capacity.

Materials and Methods

General. All experiments were approved by the NCSU animal care and use committee. Animal health was monitored twice daily, and in both experiments no health problems other than refusal of feed were noted.

Bermuda grass. The Bermuda grass evaluated was obtained from Eastern North Carolina and was sun-dried coastal Bermuda grass chopped to particles less than 1 cm long. The composition of the Bermuda grass used is provided in Table 1 and was determined by the University of Missouri Chemistry labs, except for energy, which was determined at NCSU.

Table 1. Composition of the Bermuda grass as used in Experiments 1 and 2 in percent on an as-is basis, except where noted.

Proximate parameters	Content	Amino acids	Content
Gross energy (kcal/g DM)	4.1	Lys	.30
Moisture	8.4	Thr	.29
Protein	9.6	Try	.05
Crude fat	1.3	Met	.12
Crude fiber	31.2	SAA	.22
ADF	38.8	Val	.36
NDF	71.4	Ile	.26
Ca	.22	Leu	.52
P	.13

Fibrozyme™. The fiber degrading enzyme cocktail Fibrozyme™ used in Experiment 1 was donated by Alltech Inc. (Nicolasville, KY). Fibrozyme™ contains fermentation extracts and solubles from Aspergillus niger and Trichoderma longibachiatum. As such, it is a crude mixture containing a range of enzymes. Xylanase activity is guaranteed at 100 xylose units/g (one xylose unit is defined as the amount of enzyme that liberates one micromole of xylose per minute under the assay conditions). Fibrozyme™ is stable from pH 3 to 7 and has a maximal activity at pH 6.

Experiment 1. Eight finisher pigs fitted with ileal T-cannulas were used to study the ileal and fecal digestibility of Bermuda grass fed with and without Fibrozyme™. At the start of the experiment, animals weighed, on average, 130 kg and had been fed diets that were relatively low in fiber prior to this experiment. Animals were individually housed in solid wall pens that provided ad libitum access to water.

The experimental diets were formulated in line with industry recommendations for finisher pigs. Bermuda grass was included at 14.5% as determined through least-cost formulation using nutritional values as available from Murphy Family Farms and the Atlas of Nutritional Data on United States and Canadian feeds, National Academy of Sciences, 1971. The composition of the diets is provided in Table 2 (not based on the actual analyzed composition of the Bermuda grass).

Table 2a. Ingredients in the experimental diet used in Experiment 1 (in percent).

	Control	Bermuda	BG+enzyme
Corn	75.4	58.0	57.8
SBM	21.1	19.7	19.7
Fat	0.44	5.00	5.00
Bermuda	-	14.5	14.5
Lys	0.14	0.15	0.15
Met	-	0.04	0.04
Limestone	0.98	0.81	0.81
DiCal	0.73	0.70	0.70
Salt	0.40	0.35	0.35
Vit/Min	0.25	0.25	0.25
Cr₂O₃	0.50	0.50	0.50
Enzyme	-	-	0.20

Table 2b. Calculated composition of the experimental diet used in experiment 1.

	Control	Bermuda	BG+enzyme
GE (Kcal/g)	4.01	4.19	4.19
ME kcal/g	3.28	3.28	3.28
Protein	16.1	15.7	15.7
Fat	3.3	7.3	7.3
Fiber	2.3	6.0	6.0
Ash	3.7	4.2	4.2
Lys	0.92	0.95	0.95
Thr	0.64	0.66	0.66
Trp	0.18	0.19	0.19
Met	0.28	0.32	0.32
SAA	0.55	0.57	0.57

Diets were fed at a rate of 70g/(kg^.75×day) (based on group average weight) and feed was provided twice daily. The experiment was carried out as a Latin Square design with 3 treatments and 3 periods. Each period consisted of a five-day adaptation period to the diet followed by a two-day ileal juice collection period. Feces were collected as grab samples on Day 5 while ileal juice samples were collected for 12-hour periods on Day 6 and 7. The diet, fecal, and ileal juice samples were analyzed for energy, nitrogen, and chromium.

Experiment 2. Thirty gestating sows (average parity 3.6, and 30 to 50 days post-breeding) weighing approximately 200 kg were assigned to one of six treatments based on body weight and parity such that five animals were assigned to each treatment. Animals were housed in standard gestation crates with one empty crate in between each treatment group. Water was available ad libitum. During the course of the experiment, animals were limit-fed with a standard gestation diet with chromium added as a marker for the indigestible fraction (see Table 3) supplemented with either 0, 2, 4, 6, 8, or 10% Bermuda grass. Feeding levels used were 2, 2.04, 2,08, 2.12, 2.16, and 2.20 kg/day for the animals fed 0, 2, 4, 6, 8, and 10% BG, respectively.

Table 3. Composition of the basal diet as used in Experiment 2.

Ingredients	%	Calculated
Corn	83.5	GE (kcal/g)	3.93
SBM	12.7	ME (kcal/g)	3.28
DiCaP	2.0	CP (%)	13.0
Limestone	0.7	Ca (%)	0.80
Salt	0.5	P (%)	0.70
Vit/Min	0.25	Lys (%)	0.60
Cr₂O₃	0.30	Met (%)	0.23
Total	100	Thr (%)	0.48

Animals were fed for a period of eight days after which fecal grab samples were collected on two subsequent days. Both the diet and the fecal grab samples were analyzed for energy, nitrogen, and chromium (as a marker for the indigestible fraction) using the methods as outlined above, based on which energy and protein digestibility was calculated. Results were analyzed statistically using SPSS version 8 (SPSS Inc., Chicago, IL).

Results

Experiment 1. As the Bermuda grass was rather coarse and poorly digested in the ileum, ileal digesta collected were not representative of the material in the ileum, as was apparent from the large variation in the energy to chromium ratio in the collected digesta. Therefore, no data for ileal digestibility of Bermuda grass are reported.

Animals were reluctant to consume the diet containing Bermuda grass; several animals only consumed half the feed provided, while one animal completely refused to eat diets containing Bermuda grass (this animal was deleted from the experiment).

Table 4. Apparent fecal digestibility of Bermuda grass (BG) in Finishing Pigs.

	Control Diet	BG Diet	BG + Enzyme
Energy (%)	79.7±1.1^a	65.0±1.1^c	71.2±1.1^b
Nitrogen (%)	78.5±2.1	74.5±1.9	76.5±2.1

Differences in energy digestion were significant at p<0.01; differences in protein digestion were not statistically different.

A marked decrease in fecal energy digestibility was observed (Table 4) for the diets containing Bermuda grass. Data suggest that Bermuda grass has an effective energy[4] content of zero kcal/g while Fibrozyme™ improved the digestibility of the bermuda grass containing diet such that bermuda grass yielded an effective energy content of 1.7 kcal/g.

Somewhat surprising, no statistical differences in nitrogen digestion were observed. This contrasts the observation that, at the ileal level, typically enzymes improve both energy and protein digestion to a similar extent. Numerically, however, differences in N digestion showed a pattern similar to energy digestion.

Experiment 2. Gestating sows consumed the diets supplemented with Bermuda grass completely. Energy and nitrogen digestibility data as observed in this experiment are summarized in Figure 1. For energy digestibility, the relationship that best described the data was quadratic (r²=.77, with all parameters significant at the p<.01 level, bermuda grass inclusion in %): Energy digestibility = 92.7-1.58xBG+.1xBG² r²=.77

This relationship suggests that at low levels of Bermuda grass inclusion, this Bermuda grass did not contribute any digestible energy. At higher inclusion levels, however, the animals seemed to be able to gain some energy from the Bermuda grass (but still only approximately 1/3rd of the energy contained within the Bermuda grass (or 1.4 kcal/g effective digestible energy) at the 10% level.

Fig. 1. Energy and protein digestibility of a Bermuda grass containing diet as observed in gestating sows. For energy, a quadratic function provided the best fit, while for protein a linear function of Bermuda grass inclusion was observed. See text for details.

For protein digestibility, the following relationship was calculated using linear regression (BG inclusion in %): Protein digestibility = 89.5-.89xBG inclusion r²=.67

This relationship suggests that effectively the protein in Bermuda grass is not digestible; in fact, it decreases apparent digestibility to the extent that a negative apparent digestibility coefficient should be assigned to Bermuda grass.

Discussion

The results obtained in Experiments 1 and 2 suggest that Bermuda grass has a very poor digestibility when fed as part of a complete feed. In the finishing pigs not adapted to high fiber diets, Bermuda grass yielded no ‘effective’ digestible energy[5]. In gestating sows, the ‘effective energy digestibility’ of Bermuda grass was negative at low inclusion rates but it increased when higher levels were fed and at 10% inclusion approximately 1/3^rd of the energy in the Bermuda grass was digested. Whether the non-linear nature of energy digestibility is an artifact of the experiment is not clear. Sows fed high levels of fiber may have adapted more (or faster) to the high fiber content of the diet. Thus, in those animals hind-gut fermentation may have developed to the extent that a portion of the fiber (and other nutrients trapped in the fiber) was digested. This response deserves further research as it may imply that feeding Bermuda grass for a longer period and/or at high enough levels will induce the ability to utilize at least a portion of the nutrients contained in the Bermuda grass.

Fecal protein digestibility with high-fiber diets may be confounded by the fact that fiber stimulates fermentation in the large intestines, which can lead to the accretion of nitrogen in the large intestines. The underlying mechanism is that the microbes in the large intestines are supplied with an energy source in the form of fiber, but for growth nitrogen is required. This nitrogen is ‘borrowed’ from the host, thus reducing urinary nitrogen excretion (Canh, 1998). Whether this phenomenon was relevant in the experiments described can not be concluded from these experiments. In Experiment 1, addition of Bermuda grass to the diet did reduce energy digestibility but no effects on nitrogen digestibility were noted suggesting that nitrogen accretion by microbes was not relevant. The relatively low energy digestibility values also suggest that fermentation activity was rather low as can be expected in animals poorly adapted to high-fiber diets.

In Experiment 2, energy digestibility was much higher, in line with the expectation that in sows fiber degradation is much more developed. In these animals, increasing Bermuda grass led to a decrease in protein digestion that was numerically stronger than the decrease in energy digestibility. This may well be caused by fermentation activity that led to the accretion of nitrogen in the large intestines, thus depressing the apparent digestibility of protein. Fibrozyme™ increased the energy digestibility of the Bermuda grass containing diet such that the effective digestible energy of the Bermuda grass increased from zero kcal/g to 1.7 kcal/g. Protein digestion, however, was not affected statistically (although it showed a similar trend; 29% effective apparent digestibility without Fibrozyme™ and 48% with Fibrozyme™). This is surprising, as, although the enzymes contained in Fibrozyme™ are predominantly carbohydrases, their ability to degrade carbohydrates should lead to the release of protein thus increasing protein digestion. The reason why this did not occur in Experiment 1 is unclear. It may be due to the low protein content of Bermuda grass; thus, the carbohydrases had little protein to release. Alternatively, it may be an artifact of the experiment.

References:

Canh, T.T. 1998. Ammonia emission from excreta of growing-finishing pigs as affected by dietary composition. Ph.D. thesis, Agricultural University Wageningen, The Netherlands.

[4] The effective energy content of Bermuda grass was calculated by presuming that the digestibility of the basal diet was not affected by Bermuda grass inclusion.

[5] Zero effective energy digestibility of Bermuda does not mean that the Bermuda grass per se was not digested (at all). High fiber ingredients have a negative effect on apparent digestibility of other ingredients as nutrients are trapped by the fiber but also as fiber increases endogenous losses.

Proximate parameters

Amino acids

Ingredients

Calculated