Successful dairy farming in the present age is a challenge as today’s dairies face the universal problems of diminishing returns on equity and ever more pressing environmental rules. Solving these problems is not a matter of finding one magical answer, as success lies in understanding and utilizing the knowledge base of multiple disciplines. Nutrition is but one of these disciplines; however as ration expense is usually the largest cost component faced by producers, consideration of any process or product that raises returns is necessary.
Improvements in the engineering of dairy feeding systems and a better understanding of ration digestibility have shown obvious promise (Schurig and Rodel, 1993), as have harvesting equipment advances such as kernel processors. Production increases of 10% (Straub et al., 1996) and improvements in dry matter intake (Schurig and Rodel, 1993) have been reported when kernel processing was employed.
Responses have not always been positive nor have there been consistent improvements in milk protein or fat levels (Miller et al., 1969). The reasons for this variability in response are not entirely understood, but the stage of crop maturity, ruminal starch disappearance and the dynamics of fiber disappearance are reasonable guesses. It must however be remembered that kernel processors are designed primarily to affect the rate of starch availability; and their effect on fiber digestion per se does not appear to be large (Spain, 1998; Johnston et al., 1999). As most corn silages are 40-60% grain, the implications regarding digestion of the fiber portion are significant and should not be ignored.
Exogenous fibrolytic enzymes represent one possible option for improving dietary fiber disappearance rates; and several authors have reported on their successful use (Rode et al., 1999; Kung et al., 2000; Zinn and Salinas, 1999). Most of the research reported to date involved the use of enzymes sprayed on the forage before its incorporation into the final diet.
This paper will deal with the evaluation of a dry commercial enzyme supplement (Fibrozyme, Alltech, Inc.) intended as a feed supplement rather than a for-age treatment. The product was evaluated through measurement of its effects on in vitro neutral detergent fiber (NDF) (IVNDF) digestion of corn silage and the production responses of a commercial dairy herd fed diets balanced to take into account the enzyme effects on silage digestion characteristics using CPM ration formulation.
Measuring NDF digestibility
Measurements of NDF digestibility rates can be achieved using in vivo, in situ, or in vitro techniques. Of the three, the in vivo procedure is undoubtedly the best, but is very costly and time consuming. In situ techniques have been shown to offer an accurate measurement of rumen degradation rates over various outflow rates (Arieli et al., 1996), but the logistics of chemical analysis of the digested samples are time consuming and do not lend themselves to automation.
In vitro analyses, on the other hand, have been greatly simplified with the development of new analytical systems such as Ankom Technologies, and their use for measuring 30 hr IVNDF digestibilities has grown. This type of analysis is not without problems, as it relies on several assumptions about pH and residency times. Time course IVNDF digestibilities are, however, a tool that can be used to screen large numbers of samples in an effective and affordable manner.
Methodology is obviously of critical importance, as the measurement technique must demonstrate accuracy without interfering in fermentation in order to describe feed ingredients in a manner that can be used in rumen models. Grant and Mertens (1992) reported that lag times were an important concern as they can be lengthened by decreased pH levels or by the inclusion of non-structural carbohydrates. Recent work (Piwonka and Firkins, 1996) has shown that neither the lag time nor the extent of NDF digestion is affected when pH is in the 6.2 to 6.8 range, although the overall rate can be diminished. Several studies (Grant and Weidner, 1992; Hiltner and Dehority, 1983) have shown that pH of 6.2-6.8 is the threshold for optimal NDF digestion; hence in vitro analyses should be conducted with inoculum pH in this range.
The in vitro measurement of digestibility has progressed from the early two-stage technique described by Tilley and Terry (1963) to the more automated systems of today. Several authors have reviewed the newer techniques (Garman et al., 1997; Cohen et al., 1997) and have shown that the results obtained from the new systems are reliable and repeatable. One technique reported employed the Ankom Daisy II, with which NDF digestibility can be measured for forages, grains, or total mixed rations.
The system utilizes miniature filter bags immersed in filtered rumen inoculum and buffer. The fermenting chamber holds up to four jars that can contain up to 25 filter bags each. This system offers a great deal of flexibility as far as experimental design is concerned. Analysis of results using the system has shown that similar samples can be successfully fermented within one jar or mixed with other samples within the same jar. As the samples are digested in separate filter bags, sequential ADF and NDF analyses can be performed with a minimum of problems once the fermentation is complete.
Use of in vitro analysis to screen forages
In vitro techniques to measure 30 hr IVNDF disappearance are certainly of value and have been used as one method of ranking corn silage varieties (Thomas et al., 1998). In this work, varieties from three different companies were compared. In vitro true dry matter disappearance (IVTDMD) was shown to differ (P<0.07), as did in vitro NDF disappearance. Further work conducted at the Miner Institute revealed IVNDF disappearance rates for corn silages ranging from 24.87 to 61.56% with a coefficient of variation (CV) of 19.53% for samples originating from farms throughout New England (Allshouse et al., 1998). Multivariate analysis of these samples demonstrated that rainfall and temperature also contributed significantly to the variation in INTDM disappearance (Majewski et al., 1998).
A similar use of in vitro analyses began at Ritchie Feed and Seed in the summer of 1997. The initial thought was to attempt a categorization of forage samples by variety and geographic area. This effort soon proved all but futile, as large intra- and intervarietal differences were observed both within and between regions. Some of this variability was anticipated given the soils of eastern Ontario, but the CV was unexpectedly large.
Upon realizing the degree of variability (Table 1), work began to further quantify the differences among forage samples. Production differences between herds with similar average days in milk fed TMRs based on corn silages with similar 30 hr IVNDF disappearance values led us to believe that the answer perhaps lay in the degree and extent of digestion expressed at various time points. The problem then becomes one of which time point(s) to select, and whether the choice remains constant. Our experience has shown that time point selection is not constant as the effects of maturity at harvest require varied time point selection.
Implications of variations in NDF digestibility
The dairy industry is very focused on dry matter intake. The more dry matter a cow consumes, the more milk and milk components the cow will produce. To maximize dry matter intake, it is important to understand the quality and digestibility of the forage component of the diet. The techniques currently used in North America to predict energy from the acid detergent fiber (ADF) or NDF content are unsatisfactory, as the values give variable production responses.
The basic assumption in these approaches is that all fibers are digested at the same rate and are passed from the rumen at the same rate. These techniques worked initially, but newer data have shown that it is time to move on. Japanese workers have developed the Oa and Ob system (Sniffen and Chalupa, 1998; Zinn and Salinas, 1999), which delineates fast and slow fiber fractions. The Oa fraction is the amount of fiber that disappears in a 4 hr digestion using a fiber digesting enzyme system, while the Ob fraction represents the slower-digesting fraction.
With the release of feeding programs that include predictions of rumen microbioal protein yield, we have been able to see how alterations in the NDF disappearance rate can influence level of milk produced through changes in rumen pH and the extent of rumen microbial protein synthesis. The technique found to be most successful was use of time point analyses to estimate rapid and slow NDF digestion rates. Results from this approach are given in Table 2, which shows how a swing of 10 points in the soluble NDF fraction can have a dramatic effect on milk production and on predicted NEL levels. The data in Table 2 were generated from in vitro analyses using the CPM model and corrected 48 hr NDF levels. The carbohydrate B2 (C:B2) fraction represents the rate of NDF disappearance.
Response to enzymes in ruminant diets
The inclusion of enzymes in monogastric diets is quite common. Their use in ruminant diets has not been as prevalent as the fibrolytic activity of rumen microbes is such that the inclusion of exogenous enzymes in the diet was felt to be unwarranted.
Additional concerns have been that exogenous enzymes cannot survive proteolysis in the rumen and (or) that enzyme activity would be diminished by conditioning systems such as pelleting. Recent research has demonstrated that these concerns may be unwarranted, as positive effects have been noted in both dairy and beef cattle (Beauchemin et al., 1995; 1999). Production responses have ranged from feed efficiency improvements of 10% in beef cattle (Beauchemin et al., 1995) to milk increases of 2.5 kg when lactating cattle were fed a TMR diet treated with a fibrolytic enzyme product (Stokes and Zeng, 1995).
A recent two year study examining effects of a cellulase/xylanase mixture fed to lactating Holsteins demonstrated that the productive improvements in milk yield and 3.5% fatcorrected milk yield were most likely due to improved NDF digestion. These authors reported a significant (P<0.05) increase in 48 hr NDF digestion, but only a numerical increase at 12 hr. Other authors have shown a significant increase in NDF digestion within six hours while the effect was lost at digestion times of 30 or 48 hrs (Johnston and Shivas, 1999) (Table 3). This time difference is important, as it would appear that productive responses do occur if NDF digestion improvements take place within the first 6 hrs after ingestion.
STUDIES WITH FIBROZYME
In vitro response: effects of corn silage variety and planting density
As the practical impact of the various commercial enzyme products would be expected to vary owing to differences in amounts and types of enzyme activities present, generalized conclusions cannot really be drawn. With this in mind, a field trial was undertaken to see if the product Fibrozyme (Alltech Inc., Nicholasville, KY) affected in vitro NDF disappearance rates, and if so in what manner. The study involved four corn silage hybrids planted at three densities of 22K, 27K, and 32K per acre.
Samples were taken following a 90 day fermentation, with NDF disappearance rates being calculated from the results of in vitro analyses. Fibrozyme was added at a rate that would be equivalent to 15 g/head/day. The corn hybrids selected represented a reasonable cross section of those available to today’s producers and included both leafy and grain-type corn plants. NDF levels were determined using equipment produced by Ankom Technologies. Rumen fluid was taken from lactating Holsteins producing in excess of 30 kg/day. The fluid was strained through four layers of cheese cloth and added to a Kansas State buffer system. The samples were fermented for 0, 3, 6, 30 and 48 hrs at a constant temperature of 39.1°C. The hybrids used in the study were grown with three replicates per hybrid for all planting densities and were analyzed with three replicates.
Fibrozyme increased disappearance of NDF (P<0.05) across corn silage hybrids and planting rates within the first 6 hrs of in vitro fermentation (Table 3). This effect is of particular interest as field reports on the success of enzyme supplements have been mixed. Use in the field has no doubt been based on ‘one enzyme fits all corn silage varieties’ strategy; and the fact that there was a significant difference in how different hybrids responded to Fibrozyme offers one reason why field responses to enzymes may vary.
Correlations between Fibrozyme response and plant characteristics or digestion rates were calculated (Table 4, Figures 1,2). Several relationships between ADF and NDF were also examined; as it has been hypothesized that the success of Fibrozyme could be predicted by the numerical difference between NDF and ADF. This hypothesis was not proven as the correlation coefficient between Fibrozyme response and NDF-ADF was very low (Table 4).
The correlation analyses revealed a window of opportunity for use of Fibrozyme based on initial NDF digestion (NDF disappearance within the first 6 hrs of in vitro fermentation). The window lies within the 37 to 42% NDF range as may be seen in Figure 2. The practical importance of this information lies in the fact that Fibrozyme would not appear to alter the NDF disappearance of very high quality corn silage nor will it improve that of poor quality silage. It would however significantly affect the corn silages that fall in between, as shown in Table 1. This includes a wide range of corn silages. The data in Table 1 were drawn from the analyses of corn silage samples grown in similar environmental conditions during the summer of 1998 in the Ottawa Valley.
Application of Fibrozyme and IVDMD data at a commercial dairy
Having shown that a window of opportunity existed for the use of the Fibrozyme product, the next step was to ascertain whether this in vitro screening technique could be applied in ration formulation to gain a response in milk production. A farm with corn silage (1999 crop) within the correct NDF ‘window of opportunity’ was selected and Fibrozyme was added to the diet in a commercial supplement. The silage had fermented for 80 days prior to its being fed, with pH and visual assessments indicating that the silage had fermented well. The diet was balanced using the CPM model with considerations being made for rate and extent of fiber digestion.
The DHIA results for the farm are given in Table 5, with the 1999 corn silage usage beginning two weeks prior to the October test date. Milk yield dropped 2 liters/day once the 1999 corn silage crop was fed. The response to the inclusion of the 1999 corn silage, when the other diet components were left similar, was a loss of milk within five days.
The effect of inclusion of Fibrozyme was a return of milk production to previous levels. These dietary changes fell in line with monthly DHIA testing; and the milk response was recorded with similar days in milk. This was extremely encouraging, as it appeared to demonstrate that diet formulation based on the corn silage parameters determined with in vitro analysis did in fact work. Fortunately, few cows freshened or were dried off during this period. While this was due more to good luck than good experimental management, the production improvements indicated that it would be worthwhile to run a more controlled experiment that would allow removal of ‘cow day’ effects while also determining whether effect of the enzyme was due to changes in dry matter intake.
A second trial was run under more controlled experimental conditions. A free stall dairy with 137 cows was used in a switchback design with two week periods. The majority of the cows on test were in either their first or second lactations, but this was felt to be a positive as these animals would face the greatest challenge as far as intake was concerned (Table 6).
Corn silage samples were collected prior to and during the course of the trial to identify the NDF characteristics; and the average of 10 samples is shown in Figure 3. The NDF fractions of the corn silage fell within the 37 to 42% NDF window of opportunity and the CPM model was used to design the diet used in the trial (Table 7). Fibrozyme was added to the diet at 15 g/ head/day by inclusion in a protein supplement that was pelleted at 165ºF and 137 kPa in order to minimize the possibility of enzyme denaturation. Once the trial began, the diet proportions remained constant save for dry matter corrections. In order to minimize any production variability arising from incompletely fermented corn silage, the diet was based on silage harvested in the fall of 1998 and stored in a sealed bunker since harvest.
As previous field experience had shown that forages with constant rates of NDF disappearance yield more productive milk responses in lactating cattle, efforts were made when formulating the diet to keep the overall disappearance rate of NDF in the total mixed ration (TMR) diet as constant as possible. This was done by matching the NDF rates of the ingredients to the theoretical ideal, and the results are shown in Figure 4. Of interest is that the predicted C:B2 from the CPM model (CPM-Version 1, 1998) was 7.6%/ hr while the calculated rate from our results was 7.2 %/hr; and therefore some degree of confidence can be placed in the concept of designing rations along these lines. A similar exercise was run on the protein side, with the theoretical soluble intake protein (SIP) being 41%, while the calculated amount from the lab analyses was 49%.
The performance results showed that addition of Fibrozyme increased milk yield while its removal from the diet was associated with a decline in milk production (Table 8). The production data demonstrated that in vitro analysis of NDF disappearance rate is a valid and valuable tool when designing rations.
The unequal parity and wide range in days in milk of the cows in the trial made the statistical analysis a challenge. The average milk production change across parities amounted to 1.32 litre/day, with the second lactation cows rising the most with a 1.88 litre/day change. The statistically significant milk response was lost, as expected, during the Fibrozyme withdrawal portion of the trial. Questions could be raised as to the measured effect when the Fibrozyme was reintroduced during the last part of the trial; and one possible answer is that there was a refractory response. If this is the case, it would be of minimal significance on a commercial basis as producers would not add and remove the enzyme on such short notice.
Of particular interest was the numerical increase in milk protein, which approached statistical significance (P<0.06) in the first parity cows. This increase was presumably due to increased microbial protein production in the rumen. One could speculate as to whether this improvement was due to a direct enzymatic effect or changes in rumenal microbial populations, however drawing such a conclusion is far beyond the scope of this field trial.
Economic impact of changes in NDF disappearance
Reviews of the diets fed high producing dairy herds have shown that they usually have a large percentage of highly fermentable NDF, and that these types of ingredients will result in both increased rumen microbial yields (CPM-Version 1, 1998) and improved milk production. The improvements occur for many reasons, but an overall improvement in ruminal health/function would appear to be the most logical explanation. The economic potential of harnessing improvements in fiber digestion can be significant, and in today’s tight market it can be a lifesaver.
As an example of just how large savings in production costs can be, data were collected from a 150 cow free stall herd producing approximately 30 liters per cow per day. The farm has two silos for corn silage. One silo had corn silage stored at 36% DM with a C:B2 rate of 3.07 %/hr and the other silo had corn silage at 32% DM with a C:B2 of 7.0%/hr. Switching from one silo to the other resulted in a 0.75 liter in increase in milk and improved herd health (Dairy Farmers of Ontario, 1999). Formulating diets based on the corn silage with a 7.0% C:B2 fermentation rate allowed the option of either an increase in potential milk production or a savings in feeding costs. As may be seen from Table 9, switching between the two corn silages offered an opportunity to alter how ingredients were used in the final TMR. The economic benefits of these changes were impressive. Savings of $0.68 CDN were seen on a daily basis, while on an annualized basis for 150 milking cows, this difference could contribute $37,250 to the producer’s bottom line.
Assessing the economic benefits of using Fibrozyme can be done using the same type of logic. Unlike most other feed additives, this product has a predictable effect on the disappearance of corn silage NDF. If the NDF is within the window of opportunity, theoretical improvements of 12% are possible, and depending on the formulation of the final TMR this can result in a milk production increase of 1.52 liters. The studies discussed above demonstrated that the theoretical and practical worlds can in fact meet as evidenced by increased average milk production of 1.5 liters/day in a diet formulated using IVNDF disappearance data to characterize silage NDF.
Allshouse, R.D., C.J.Majewski and C.J. Sniffen. 1998. Investigations in forage quality. 1. Variation in forage quality in the northeast. W.H. Miner Agricultural Research Institute.
Arieli, S.J., J. Mabjeesh, Z. Shabi, I. Bruckental, Y. Aharoni, S. Zamwel and H. Tagari. 1996. In situ assessment of degradability of organic matter in the rumen of the dairy cow. J. Dairy Sci. 81:1965-1990.
Beauchemin, K.A., L.M. Rode and V.J.H. Sewalt. 1995. Fibrolytic enzymes increase fiber digestibility and growth in steers fed dry forages. Can. J. Anim. Sci. 75:641-644.
Beauchemin, K.A., W.Z. Yang and L.M. Rode. 1999. Effect of grain source and enzyme activity on site and extent of nutrient digestion in dairy cows. J. Dairy Sci. 82:378-390.
Cohen, M.A., H.E. Maslanka and L. Kung. 1997. An evaluation of automated and manual in vitro methods for estimation of NDF digestion. Conference on Rumen Physiology, Chicago, IL.
CPM-Dairy version 1.0, Copyright 1998 by The Center for Animal Health and Productivity, School of Veterinary Medicine, University of Pennsylvania, Kennett Square PA; The Department of Animal Science, Cornell University, Ithaca NY; and The William H. Miner Agricultural Research Institute, Chazy NY.
Dairy Farmers of Ontario. 1999. Milk shipments, February. Garman, C.L., L.A. Holden and H.A. Kane. 1997. Comparison of in vitro dry matter digestibility of nine feedstuffs using three different methods of analysis. J. Dairy Sci. 80(Suppl. 1):260.
Grant, R.J. and D.R. Mertens. 1992. Development of buffer systems for pH control and evaluations of pH effects on fibre digestion in vitro. J. Dairy Sci. 75:1581-1587.
Grant. R.J. and S.J. Weidner. 1992. Digestion kinetics of fibre: influence of in vitro buffer pH varied within observed physiological range. J. Dairy Sci. 75:1060.
Hiltner, P. and B.A. Dehority. 1983. Effect of soluble carbohydrates on digestion of cellulose by pure cultures of rumen bacteria. Appl. Environ. Microbiol. 46:642.
Johnson, L., J.H. Harrison, C. Hunt, K. Shinners, C.G. Doggett and D. Sapienza. 1999. Nutritive value of corn silage as affected by maturity and mechanical processing: a contemporary review. J. Dairy. Sci. 82:2525- 2944.
Johnston, J.D. and I.C. Shivas. 1999. Formulating the fourth diet. In: Biotechnology in the Feed Industry. Proceedings of the 15th Annual Symposium (T.P. Lyons and K Jacques,eds). Nottingham University Press, UK.
Kung Jr., L., R.J. Treacher, G.A. Nauman, A.M. Smagla, K.M. Endres and M.A. Cohen. 2000. The effect of treating forage with fibrolytic enzymes on its nutritive value and lactation performance of dairy cows. J. Dairy Sci 83:115-122.
Majewski, C.J., R.D. Allshouse and C.J. Sniffen. 1998. Investigations in forage quality. II. Variability in forage quality parameters for corn hybrids. W.H. Miner Agricultural Research Institute.
Miller, C.N., C.E. Polan, R.A. Sandy and J.T. Huber. 1969. Effect of altering physical form of corn silage on utilization by dairy cattle. J. Dairy Sci. 52:1955-1960.
Piwonka, E.J. and J.L. Firkins. 1996. Effect of glucose fermentation on fibre digestion by ruminal microorganisms in vitro. J. Dairy Sci. 79:2196- 2206.
Rode, L.M., W.Z. Yang and K.A. Beauchemin. 1999. Fibrolytic enzyme supplements for dairy cows in early lactation. J. Dairy Sci. 82:2121-2126.
Schurig, M. and G. Rodel. 1993. Power consumption and the effect of corncrakers. ASAE paper 931586. American Society of Agricultural Engineers, St Joseph, MI.
Sniffen, C.J. and W. Chalupa. 1998. Protein and amino acid nutrition of lactating cows – an update. Dairy Professional Program, Cornell 1998.
Spain, Jim. 1998. Profitable forage management strategies. Western Canadian Dairy Seminar, Advances in Dairy Technology. Red Deer, Alberta.
Stokes. M.R. and S. Zeng. 1995. The use of carbohydrate enzymes as feed additives for early lactation cows. 23rd Biennial Conf. Rumen Function, Chicago, Il., 23:35 (Abstr.).
Straub, R.J., R.G. Kroegel, L.D. Satter and T.J. Krause. 1996. Evaluation of a corn silage processor. ASAE paper no. 961033. American Society of Agricultural Engineers, St. Joseph. MI.
Thomas, E.D., C.S. Ballard, C.J.Sniffen, D.S. Tsang, R.D. Allhouse and P. Mandebvu. 1998. Effect of hybrid on corn silage yield, nutrient composition, in vitro digestion, intake by Holstein heifers, intake and milk production by lactating Holstein cows. W.H. Miner Agricultural Research Institute, Chazy, NY. 98-13.
Tilley, J.M.A. and R.A. Terry. 1963. A two-stage technique for the in vitro digestion of forage crops. Journal of British Grassland Society 18:104- 111.
Zinn, R.A. and J. Salinas. 1999. Influence of Fibrozyme on digestive function and growth performance of feedlot steers fed a 78% concentrate growing ration. In: Biotechnology in the Feed Industry. Proceedings of Alltech’s 15th Annual Symposium (T.P. Lyons and K.A. Jacques). Nottingham University Press, Nottingham, UK.