HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ellis, W. C. Articles by Matis, J. H. Search for Related Content PubMed PubMed Citation Articles by Ellis, W. C. Articles by Matis, J. H.

Effects of size of ingestively masticated fragments of plant tissues on kinetics of digestion of NDF

W. C. Ellis^*,1, M. Mahlooji^*, C. E. Lascano^*,2 and J. H. Matis

^* Departments of Animal Science and and Statistics, Texas A&M University, College Station 77843

	Abstract

Top Abstract Introduction Materials and Methods Results and Discussion Implications Literature Cited

 
Ingestively masticated fragments were collected and sized via sieving. Different sizes of esophageal masticate and ruminal digesta fragments, and ground fragments of larger masticated pieces were incubated in vitro, and undigested NDF remaining at intervals of up to 168 h of incubation was determined. The ruminal age-dependent time delay () for onset of digestion of NDF was positively correlated (P < 0.004) with the mean sieve aperture estimated to retain 50% of the fragments between successive sieve apertures (MRA). Degradation rate of potentially degradable NDF (PDF) and level of indigestible NDF were not related (P > 0.10) to MRA of masticated and ground fragments. Estimates of were positively related to MRA, with slopes of bermudagrass < corn silage < ruminal fragments of corn silage. It was concluded that fragment size-, and consequently, ruminal age-dependent onset of PDF degradation of a mixture of different fragment sizes results in an age-dependent rate of degradation of the more rapidly degrading of two subentities of PDF. Models are proposed that assume a before onset of simultaneous degradation of PDF from two pools characterized as having gamma-modeled age-dependency and age-constant rates. The ruminal age-dependent pool seems to be associated with the faster-degrading pool, and its rate parameter increases with range in MRA in the population of fragments. Conceptually, the ruminal age-dependent rate parameter for PDF degradation seems to represent a composite of several effects: 1) effects of the size-dependent ; 2) range in MRA of the population of ingestively masticated fragments; and 3) subentities of PDF that degrade via more rapid age-dependent rates compared with subentities of PDF that degrade via age-constant rates. The estimated fractional rates of ruminative comminution of ingestively masticated fragments (0.060 to 0.075/h) were of a magnitude similar to the mean fractional rates of PDF digestion (0.030 to 0.085/h), which implies that ruminative comminution may be first-limiting to fractional rate of PDF digestion. The in vivo roles of ingestive and ruminative mastication of fragments on PDF degradation must be considered in any kinetic system for estimating PDF digestion in the rumen. These results and others in the literature suggest that the rate of surface area exposure rather than intrinsic chemical attributes of PDF may be first-limiting to degradation rate of PDF in vivo.

Key Words: Digestion • Mastication • Neutral Detergent Fiber • Plant Tissues • Rumen

	Introduction

Top Abstract Introduction Materials and Methods Results and Discussion Implications Literature Cited

 
Ruminal microbial degradation of potentially degradable NDF (PDF) requires physical access and attachment as a prerequisite for its hydrolysis by the fibrolytic bacterial ecology (Weimer, 1996; Morrison and Miron, 2000). Preferred surface sites of PDF (Akin and Amos, 1975) are colonized before erosive degradation of sub-surface PDF can occur (Akin and Barton, 1982; Akin, 1983). It is the array of fragments produced by ingestive mastication that enter the rumen and first become subject to microbial colonization and PDF hydrolysis. Ingestive mastication during prehension of plant parts ruptures indigestible surface barriers (Akin and Barton, 1976, 1982; Akin, 1983) and yields fragments containing a heterogeneous mixture of tissues (Pond et al., 1987). The PDF internal to larger ingestively masticated fragments is limited in its access for microbial colonization because of the dominance of indigestible cutile and vascular bundle tissues (Hannah et al., 1976). Without regard for these in vivo aspects of physical NDF degradation, hydrolysis of PDF is commonly estimated in vitro or in situ from ground samples of forages and modeled as substrate unlimited age-constant rates of PDF degradation. Ruminal age-constant degradation models assume unlimited levels of available PDF degraded at a constant rate when rate is expressed as a fraction of the PDF remaining (i.e., a first-order or age-constant process) as the result of an exponentially diminishing quantity of unlimited PDF being degraded at an age-constant fractional rate. Mechanistic considerations suggest that PDF hydrolysis in vivo must be interactive with the size of the fibrolytic population and size and accessible surface sites for microbial colonization and hydrolysis. We conducted experiments to test the hypothesis that size of ingestively masticated fragments of forage tissues affects the kinetics of NDF digestion from masticated and ground fragments undergoing comminution and hydrolysis.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results and Discussion Implications Literature Cited

 
These experiments were conducted before requirements for institutionally approved animal use protocols were passed (Mahlooji, 1985). However, the animal use protocols used then are virtually the same as those subsequently approved by an institutional animal use committee at Texas A&M University.

Forage and Tissue Fragments

Coastal bermudagrass hay ([Cynodon dactylon {L.} Pers.]; 120 g of CP/kg of DM and 710 g of NDF/kg of DM) or corn silage (Deswysen et al., 1988) were individually fed ad libitum to two esophageally and ruminally cannulated steers for at least 7 d before collection of masticated fragments during two meals as esophageal extrusa and, in the case of corn silage, fragments from the rumen. Sufficient extrusa or ruminal digesta were collected for experimentation from two meals via the open esophageal fistula and the ruminal cannula of the two steers (Ellis et al., 1984). Esophageal extrusa and ruminal digesta from the two animals and two meals were individually mixed and fragments sized using a wet sieving apparatus (Analysette No. 10, Fritsch GmbH, Idar-Oberstein, Germany).

The wet sieving procedure involved a column of successive sieves whose apertures are indicated in Table 1. Each sieve (10 cm diameter x 3 cm deep) was nested and sealed via water-tight O-rings. The uppermost sieve was covered by a transparent dome connected to a water supply, whereas the bottom pan provided for efflux of liquid. Influx rate of water was adjusted to an efflux rate such that the volume of water in each sieve was at least 1/3 full. An adjustable strap provided force to create a watertight seal by each O-ring of individual sieves and to transmit vibrating forces actuated by 250 to 300 vibrations/min. Water flux was continued until the efflux was clear of color and fragments represented a repeatable distribution as established by previous use (three to five per minute). The vibration rate chosen produced a continuous vertical distribution of fragments on the top-most sieve, suggesting that effects of plane-of-sieve and elongated fragments were minimized (Mertens et al., 1984; Vaage et al., 1984). Approximately 5 g of DM was sieved as three to four successive aliquots to minimize overloading the uppermost sieves. Overloading was detected as deviation from an exponential distribution oversize of the retaining sieves’ aperture. Fractions collected on each sieve were then freeze-dried to prevent adhesion of particles as subsequently used.

The larger tissue fragments were further partitioned by plant part/tissue type as indicated in Table 1. Larger fragments (retained on sieve apertures >4,760 µm) of corn silage masticate were separated visually into tissues identified as cob, leaf, pith, and epithelial parts/tissues. Leaf and stem fragments of bermudagrass masticate were separated based on their different aerodynamic properties in a vertical cylinder of air. The velocity of air through the vertical cylinder was adjusted to entrap leaf fragments at an elevation unachievable by stem fragments. This machine was custom-constructed to separate vegetative plant parts from seeds and comprised a vertical column, into which a sufficient rate of air was regulated into the bottom to float and trap leaf fragments in an upper horizon, whereas stem fragments remained in the lower horizons. Sample mass and air flow rate were adjusted and maintained until no fragments remained in the intermediate horizons.

Further Size Reduction

To distinguish between digestive effects of particle size and chemical composition of certain masticated particle sizes, three of the largest fractions were ground to pass a 1- and 0.5-mm sieve in a No. 2 Wiley mill (Thomas Scientific, Swedesboro, NJ) and then subjected to in vitro digestion.

Expression of Distribution of Different Sized Fragments

The procedure described by Pond et al. (1984b) was used to compute the exponential distribution coefficient, k, expressing the accumulative distribution of fragment DM recovered on a series of sieves of successively smaller apertures. The median retaining aperture (MRA) was computed as (0.693/k) + W, where W represented the aperture of the smallest sieve in the series of sieves. Parameters k and W were estimated by regression (PROC NLIN, SAS Inst., Inc., Cary NC) of a cumulative distribution of DM on each sieve (Pond et al., 1984b) expressed as a fraction of the sum of DM recovered on all sieves (Deswysen et al., 1988). The MRA computed from a series of sieve apertures is referred to as serial MRA, and was an estimate of the sieve aperture (µm) that would retain 50% of the total masticate fragments recovered from the series of sieves.

The MRA for the distribution of fragments trapped between discrete pairs of sequential sieves (escaping and retention sieves) was computed assuming an exponential distribution of DM between the two discrete apertures of escaping and retention sieves. This MRA was referred to as a discrete MRA. Discrete MRA was computed as the ln of the aperture of discrete MRA (µm) = ([ln of aperture of escaping sieve, µm] + [ln of aperture of retaining sieve, µm])/2. Calculated discrete MRA is equivalent to geometric mean size (µm; calculated as [escape sieve aperture x retaining sieve aperture, µm]^0.5). The MRA, rather than geometric mean size, was preferred because it could be estimated without assumptions as to smallest and largest sieve size apertures. Discrete MRA was used to differentiate distributions of fragments trapped between different pairs of sequential sieves (e.g., 1,600- and 1,000-µm escaping and retaining apertures, respectively) and may be considered constituent distributions of fragments within the distribution of total masticated fragments (serial MRA).

Chemical Analyses and In Vitro Digestion

Analysis for NDF was as described by Goering and Van Soest (1970), with the exception that NDF OM (NDFOM) was computed as the loss of DM upon ashing the residue insoluble in neutral detergent solvent without amylase and filtered through a medium-porosity sintered glass filter crucible. The in vitro digestion procedure described by Goering and Van Soest (1970) was modified to accommodate a greater number of large fragments required to achieve their representative sampling. Fermentation vessels ranged from 80-mL polyethylene tubes (3 cm x 16 cm) to 500-mL Erlenmeyer flasks, as needed to accommodate a mass of each fraction composed of a minimum of three of the larger sized fragments. Forty-two milliliters of medium and 10 mL of inoculum were used per 0.5 ± 0.2 g of DM of each fragment. The inoculum was collected from two steers receiving the same forage.

Incubations were conducted over 168 h, with individual vessels being removed every 6 h between 0 and 48 h and every 24 h thereafter. Triplicate samples were incubated for each incubation interval for fragments of <1,600 µm, with quintuplet samples incubated for fragment >1,600 µm. Digestion was terminated by addition of 1 mL of 5% (wt/vol) mercuric chloride solution per 52 mL of medium and stored at 4°C. Residues were analyzed chemically by the addition of 100 mL of NDF solvent per 0.5 ± 0.2 g of initial DM followed by reflux and filtration through medium porosity sintered glass filter crucibles. Undigested NDF OM (UNDF_t) was computed as the NDFOM solvent-insoluble OM lost upon ashing (500°C) and expressed as a fraction of the initial NDFOM before digestion.

Incubations were conducted by type of forage masticate (bermudagrass hay or corn silage) and for ruminal fragments (Table 1). Residues from larger size fragments incubated in Erlenmeyer flasks were filtered on an 11 cm of weighed filter paper (Whatman No. 41; Whatman, Ltd., Maidstone, U.K.) in a Buchner funnel. After air-drying, each sample and its filter paper were weighed, and fragments removed and ground in a micro-Wiley mill to pass through a 490-µm screen. Aliquots (0.5 g) of the ground fragments were then analyzed for undigested NDF as described for smaller fragments.

Models

Models used and their comparison and interpretation are presented in a companion article (Ellis et al., 2005). In the present report, a model assuming a gamma distribution of age-constant ruminal degradation rates of PDF and a discrete time delay () before onset of PDF degradation (E(G)) was fitted to profiles of UNDF_t obtained for in vitro incubations of different masticated fragment sizes. Model E(G) estimated shape () and scale (ß) parameters for the distribution of age-constant rates, from which the mean age-constant rate () could be estimated as 1/[( –1) x ß] (Table 2 of Ellis et al., 2005). Simulation studies used an age-constant distribution of rates subsequent to a (Model E) to estimate an age-constant rate (k).

Statistics

Each data set was fitted to each model by PROC NLIN. Fit of various models to the data was evaluated using residual mean square (RMS) computed by adjustments for the number of estimated parameters of each model: RMS = residual sum of squares (RSS) at convergence by PROC NLIN/(number of observations – number of estimated parameters for each specific model). With 20 observations per data set, the number of parameters for the single-pool models, E, was three; the number for the two-pool models, GN/E was five; and the number for the gamma mixture model E(G) was four. Because the number of parameters differed, it was not appropriate to compare RSS. However, the residual degrees of freedom (dfE) were 17, 15, and 16, respectively, for the three classes of models. Hence, when the RMS were calculated (RSS/dfE), the number of parameters in the respective models was taken into account.

The standard nonlinear least squares programs, such as that used by PROC NLIN of SAS, are based on a Taylor series linearization. The assumed parameter values are treated as variables in a multiple linear regression analysis, and the estimated regression coefficients in the linear regression analysis are the estimated changes in the parameter values for the next iteration (Neter et al., 1996). This tie-in with multiple regression is relevant because the RMS criterion is a simple, widely accepted procedure for comparing multiple regression models with differing numbers of variables, and in fact, it is a standard approach in SAS for determining the number of variables in variable selection procedures (Neter et al., 1996). This is particularly true if the differences in the number of parameters are small compared with the number of observations. Due to the linkage between nonlinear regression with multiple parameters and linear regression with multiple variables, the use of RMS in the latter justifies its use in the former. Wilcoxon’s signed rank procedure (Ott, 1977) was used to test the hypothesis of equality of RMS distributions as estimated by each model for various fractions. In cases where the null hypothesis of equal EMS distributions of two models was rejected, the model with a smaller median RMS was considered superior.

Sources of variation in digestion parameters were evaluated by partitioning the variance due to forage type (bermudagrass hay and corn silage) of fragments (masticate, ground masticate and ruminal fragments) via PROC GLM of SAS. Effects of grinding fragments were evaluated by a paired t-test between ground and whole but otherwise identical fragments (t = sample mean difference/[standard deviation of n differences/n]; Ott, 1977). Relationships between postulated dependent variables and independent variables were evaluated by PROC GLM.

	Results and Discussion

Top Abstract Introduction Materials and Methods Results and Discussion Implications Literature Cited

 
Model Selection

Application and interpretation of results from fitting varied individual models to these 29 datasets have been discussed in detail in a companion paper (Ellis et al., 2005). Models with a pool of gamma 3 or gamma 4 degree of age-dependent degradation rates () and a pool of an age-constant rate of NDFOM degradation (k₂) provided best fit (RME and smallest residual mean square error) to all 29 datasets (models G3/E and G4/E, respectively, or, more generally, the GN/E family of models where N = a degree of age-dependence 2). Superior fit by GN/E vs. E models to all datasets could be attributed to their ability to fit the biphasic form of lifetime distributions of UNDF_t observed among individual datasets. Lifetime distributions of UNDF_t for the 29 individual data sets ranged from an exponential distribution of age-constant degradation rates, k, expected for a single PDF entity to a more complex biphasic distribution of lifetimes expected for concurrent degradation rates of PDF1 and PDF2. However, because of insufficient quantity of data during early lifetimes (<12 h), and were estimated with large standard errors by the GN/E models (see Figure 2D of Ellis et al., 2005). Representing a single pool of a continuum of age-constant rates, the gamma distribution of the ruminal age-constant rates model (E(G)) provided flexibility and robustness in fitting PDF degradation lifetimes, although it did not mimic the biphasic distribution of PDF1 and PDF2. Fitting the E(G) model resulted in similar RME but imprecise (large standard errors) model estimates of because of insufficient data during the first 12 h of incubation. Therefore, parameters estimated by fitting the E(G) model to the current data sets were used initially to evaluate sources of variation in NDFOM utilization.

View larger version (14K): [in this window] [in a new window]   Figure 2. Degradation rate of potentially degradable NDF (PDF), undegradable NDF (UDF), and time delay (panels A, B, and C, respectively) as influenced by mean retaining sieve size (MRA) of populations of ingestively masticated fragments of corn silage (black plus sign) or after grinding (white plus sign) large masticated fragments of corn silage to pass a 1-mm sieve.

Effects of forage species (bermudagrass vs. corn silage), form of fragment (masticated, ground masticated, and ruminal fragments), and fragment size (discrete MRA) were all significantly (P < 0.05) related to variations in and undegradable NDF (UDF), but were unrelated (P = 0.25 to 0.57) to . The relative importance of sources of variation in was indicated by F-values of 45.0, 18.6, and 11.8 for fragment form, forage species origin, and discrete MRA, respectively. One form of fragment (ground) was achieved by altering the discrete MRA of six otherwise identical fragments of the same forage (Table 1). Thus, in these cases, effect of form was an effect of changing the distribution of discrete MRA within a population of fragments of otherwise identical chemical composition. A paired t-test for these six fragments (masticated fragments vs. ground fragments from such masticated fragments) indicated significant (P < 0.04) effects of grinding on for bermudagrass and corn silage masticates, on UDF for bermudagrass, and no difference (P = 0.20 to 0.80) on UDF for corn silage or for of either forage species. The effect of grinding on discrete and serial MRA of three ingestively masticated fragments is illustrated in Table 2 and in Figures 1 to 3. It was concluded that fragment size, as represented by discrete MRA, was a major source of variation in for individual populations of fragments produced either via ingestive mastication or a laboratory mill.

View larger version (15K): [in this window] [in a new window]   Figure 1. Degradation rate of potentially degradable NDF (PDF), undegradable NDF (UDF), and time delay (panels A, B, and C, respectively) as influenced by mean retaining sieve size (MRA) of populations of ingestively masticated fragments of bermudagrass hay () or fragments after grinding () large masticated fragments to pass a 1-mm sieve.

View larger version (13K): [in this window] [in a new window]   Figure 3. Degradation rate of potentially degradable NDF (PDF), undegradable NDF (UDF), and time delay (panels A, B, and C, respectively) as influenced by mean retaining sieve size (MRA) of populations of ruminal fragments of corn silage (black plus sign).

The was consistently, positively, and exponentially related to discrete MRA of forage tissue fragments (P < 0.002) regardless of forage species origin (Figures 1C and 2C) or prior fermentation (Figure 3C). In contrast, relationships between discrete MRA and were not significant (P = 0.26 or greater) for any forage tissue, and the relationship between discrete MRA and UDF only approached significance (P = 0.058) for a quadratic relationship (Figure 1B).

The range of fragment sizes resulting from ingestive mastication reflects a wide variation in the content of plant tissues of variable degradation characteristics (Akin, 1982; Pond et al., 1984a). It is therefore somewhat surprising that variations in were so dominantly related to simple fragment size. The quadratic trend for an UDF vs. MRA relationship may simply be a result of the increased physical loss of smaller fragments through the 30- to 50-µm porosity used here (Van Hellen, 1972; Ehle et al., 1983) as discrete MRA is decreased.

Discrete MRA represents an estimate of the mean sieve aperture that will retain half the fragments trapped between a pair of escape and retention sieves assuming an exponential distribution of individual fragment sizes. Because of the exponential distribution of fragments, a wider range in escaping and retaining sieve apertures was used to obtain sufficient quantities of fragments for reliable measurement (Table 2). Thus, estimates of discrete MRA of progressively smaller size fragments contain a progressively wider range in size of fragments with a progressively wider ratio between the smallest size fragments and mean discrete MRA. Alternatively, the aperture of the retaining sieve would represent the discrete MRA of the smallest fragment within the entrapped population. To test for systematic measurement bias associated with different ranges in smallest fragment to discrete MRA, the aperture of the retaining sieve was evaluated and found to yield essentially the same results as discrete MRA. We concluded that the primary effect of size of ingestively masticated fragments was on of fragments and was unrelated to their content of UDF and, consequently, PDF or k.

Biological Basis for vs. Fragment Size Relationship

Hydrolysis of PDF is accomplished by a complex of PDF enzymes associated with the cell walls of fibrolytic species of bacteria, and these enzymes are required for bacterial adhesion to specific surfaces of plant tissues as well as PDF hydrolysis (Weimer, 1997; Morrison and Miron, 2000). Bacterial degradation of PDF is a fragment surface phenomenon determined by spatial distribution of preferred sites of PDF and on the proportion of such surfaces accessible to fibrolytic bacteria (Hannah et al., 1976). Thus, a positive relationship between and size would be expected for fragments homogeneous in their shape and having similar proportions of their surface area accessible to fibrolytic bacteria. Fragmentation of the varied plant tissues comprising the ingested plant parts results in a heterogeneous population of fragment shapes, chemistry, and conceivable distributions of microbially accessible levels of PDF (Pond et al., 1984b; Ellis et al., 1988). In spite of this heterogeneity at the individual fragment level, the current results suggest that size of fragment, as measured by discrete MRA, provides a functional and quantitative relationship to microbial accessible surface area as a mechanism regulating onset of PDF degradation.

Fragment Size-Dependent and Age-Dependent PDF Degradation Rates

The GN/E models assume a before the onset of concurrent degradation of two subentities of PDF, PDF1, and PDF2. Proportions of PDF1 are distinguished by a gamma lifetime distribution (GN) of , whereas PDF2 is distinguished by an age-constant, exponential (E) distribution of degradation rates of k₂, respectively. When fitted to data, it is the faster degradation rate that best conforms to degradation lifetimes of a GN of PDF1. Thus, it can be inferred that PDF1 represents subentities with the more rapid PDF degradation. It also follows that a fragment size-dependent will result in a fragment size-dependent onset of simultaneous degradation of PDF1 and PDF2 and is, in part, causal of and contributes to the ruminal age-dependent rate parameter estimated for a population of different size fragments. The present data was used to simulate lifetime distributions of undegraded PDF (UPDF) expected for hypothetical populations of fragments homogeneous with respect to fragment size having an assumed age-constant rate of PDF degradation of 0.08/h and a fragment size-dependent . The discrete MRA of these populations of bermudagrass fragments of homogeneous size were chosen from data in Table 3, and their corresponding, size-dependent were computed from the regression equation given in Figure 1C.

View larger version (17K): [in this window] [in a new window]   Figure 4. Hypothetical lifetimes of undegraded, potentially degradable NDF (UPDF) from five different size fragments homogeneous in size, having size-dependent time delays of 0, 2, 4, 6 or 8 h and age-constant degradation rates (0.08/h) individually fermented (dashed lines in panel A) or fermented as a mixture (solid line in panel A). Panel B shows expected (solid line) and observed (dashed) lifetimes from fitting simultaneous PDF degradation from an age-G3 age-dependent and age-constant degradation of PDF pools subsequent to a time delay.

Being exponentially distributed, increasing the upper range in fragment size will alter the size-dependent and, consequently, . This was illustrated by comparing estimates of PDF degradation parameters from hypothetical mixtures of different fragment sizes to mimic that observed for bermudagrass and corn silage (Table 3). Results (Table 4) illustrate that increasing the upper range in fragment size from 3 to 6 µm affected the age-dependent rate parameter, , only. Such a sole effect would be expected if fragment size-dependent was causal of the age-dependent increase in onset of PDF degradation, and consequent age-dependent rates of PDF degradation, .

PDF Degradation from Mixtures of Fragments

Results of fitting the G3/E model to data from two other data sets were compared with those for the two hypothetical mixtures of different ranges (Table 4). Deviations of degradation parameters for the hypothetical mixture vs. other data sets may be due to the purposeful oversimplification of five (bermudagrass) or six (corn silage) different exponentially distributed sizes of fragments.

Results from recalculations of the masticate data of Playne et al. (1978) and the data of Carrette-Carreon (2001), obtained on ground samples, suggest that approximately 0.4 (f₁) of the total PDF was degraded via an age-dependent rates process (Table 3). The ruminal in situ data of Carrette-Carreon (2001) involved early and replicated fermentation times between 0 and 16 h so should be most reliable in partitioning lifetime due to and . Relative rates of age-dependent PDF and age-constant degradation rates seem to decrease with expected range in MRA of datasets (i.e., 0.1 to 3 > 0.1 to 30 > masticate > ground forage; Table 3).

Results in Figure 4 and Table 4 indicate that, due to the age-dependent onset of age-dependent degradation of PDF, the conceptual will inevitably be smaller than that estimated by the GN/E. For example, a of 2.2 h was estimated for the hypothetical mixture of five different fragment sizes. In contrast, a of 0 and 2 h was assumed for the two smallest size fragments so the hypothetical must have been 0. However, the hypothetical for the mixture of fragments was not statistically discernable by the nonlinear regression procedure because of the small contribution of the two smallest fractions (0.017 and 0.05 g/g; Table 3).

The hypothetical, age-dependent mean residence times due solely to an age-dependent were 7.2 and 10.8 h; values, which, considering the quality and unintended use of the data, probably did not differ from observed values for ground and masticated forages of 7.7 to 15.4 h. Relatively smaller values for for ground samples support this proposal. We conclude that the observed age-dependent turnover rate, , is primarily accounted for by a size- and consequent age-dependent turnover rate.

Effects of Partial Prior Digestion on

The prior residence time of the ruminal fragments used here is unknown. Presumably, colonizing microbes from prior ruminal in vivo digestions were inactivated via the fragment collection and preparative procedures for in vitro digestion.

Compared with in vitro digestion of ingestively masticated corn silage, resumption of ruminal corn silage fragment digestion in vitro resulted in a larger intercept (6.7 vs. 16.4 h) and a more rapid change in per change in discrete MRA (9.0 vs. 4.8). These observations are consistent with successive colonization by different microbial species with different site preferences for colonization. Akin (1982) observed results suggesting sequential colonization involving epithelial > mesophyll > fragments of vascular bundle sheaths > schlerenchyma and vascular bundle guard cells. Colonization of sclerenchyma and guard cells occurred only after digestion of the more digestible tissues. No microbial attachment was observed for highly lignified tissues such as mestome bundle sheaths. Thus, prior ruminal PDF digestion in vivo seems to have depleted the microbial accessible surface sites preferred by initial colonizing ecologies leaving a different, but wide, nonexponential distribution of microbial accessible surface sites for subsequent in vitro colonization.

Effects of Grinding Masticated Fragments

Few comparisons of ingestively masticated vs. ground fragments were found in the literature. Playne et al. (1978) concluded that ingestive chewing was sufficient to achieve maximal extent of apparent DM digestion from masticated vs. fragments escaping a 1- or 2-mm screen. Problems of distinguishing between forage and microbial DM vs. cell contents and lack of data between 0 and 12 h and subsequent to 72 h limit interpretation of these data.

Noziere and Michalet-Doreau (2000) reported that effects of grinding were especially evident during early time intervals. When fineness of forage grind (escape from mill screens varying from 0.8 to 6 mm) increased, rate of digestion increased and decreased. Robles et al. (1980) observed significant (P < 0.05) positive regressions between in vitro rates of digestion of NDF and geometric mean size (equal to discrete MRA) of ground fragments of alfalfa (319 to 703 µm) but not orchard grass (262 to 650 µm) produced by variations in grinding to escape mill screen sizes of 1 to 8 mm. Emanuele and Staples (1988) observed that different fragmentation patterns on grinding parallel veinated grasses (bermudagrass) vs. cross-veinated legumes (peanut and alfalfa). Fragments of ground bermudagrass fragments escaping 710- vs. 355- or 106-µm sieves had increased (2.1 vs. 0.4 and 0.6 h, respectively), whereas for peanut foliage did not differ due to sieved size. Grinding affected (P < 0.01) rate of digestion of both types of forage and resulted in significant forage x fragment size interactions for both rate and extent of DM digestion. Fragments from cross-veinated leaves, such as peanut foliage, are largely bounded by vascular bundle sheaths along all fractures of the leaf’s surfaces, which may account for their slower unaffected by fragment size (3.2 to 3.6 h). Thus, effects of fragment size on are present within distribution of fragments expectedly having relatively small serial MRA (<300 µM; Table 3) and produced by fragmentation processes other than ingestive mastication (Figure 1A).

Ruminative Mastication, PDF Degradation, and Ruminal Escape

Von Milgen et al. (1991) proposed a sequential, two-compartmental model to describe in situ degradation of PDF, which assumes that not all of PDF was immediately available for degradation but was degraded with age-constant kinetics once it becomes available. Von Milgen et al. (1991) suggested the first compartment functioned as a lag compartment preceding the second compartment functioning as an age-constant, mass-action PDF degradation. Later, Von Milgen and Baumont (1995) proposed an age-dependent function to describe changes in microbial efficiency of PDF degradation as preceding an age-constant degradation rate.

We propose that interactive events involving the repetitive "mixing" forces of ruminal motility sequences and the continual "unmixing" forces of buoyancy of aging fragments are intimately involved in determining the intraruminal kinetics of PDF degradation. We (Ellis et al., 1999, 2000; Waltz et al., 2004) propose a two-phased process regulating ruminal degradation of PDF. Firstly, fermentation-based buoyancy of younger, larger, and more abundant surface sites of PDF for preferred attachment of the fibrolytics (Sutherland, 1986) is the force selectively positioning larger, younger fragments into "lag-fermentation" flow-paths leading to ruminative mastication. As fragments age within ruminal digesta, they become smaller, depleted of surface levels of PDF, less efficient in entrapping gases of fermentation and exposing new surfaces of PDF. Further age-dependent changes in size become smaller as progressively smaller fragments are less efficiently comminuted (Figure 13 of Ellis et al., 1999). Consequently, with further aging, fragments become denser due to increased content of denser UDF and as their probability for escape from the "lag-rumination flow-paths" to "mass-action turnover flow-paths" increases. Escape from the mass-action turnover flow-paths is via mass-action competition of fragments of like specific gravity and affords residence time for degradation of PDF via erosion (Gardner et al., 1999). Further aging of ruminatively masticated fragments is associated with decreased buoyancy of fragments due to limiting microbial accessible PDF and increased specific gravity associated with increasing concentrations of denser UDF.

It is suggested that such a sequential, two-phase degradation progress for PDF accounts of the efficient digestion of PDF (Figure 18 of Ellis et al., 2000). The sequential and combined actions of lag-rumination and mass-action turnover flow-paths provide a mechanism whereby efficient degradation of PDF could occur at the level of individual fragments. Such a mechanism infers that the drive to efficiently degrade PDF is the force constraining ruminal escape of forage residues until their microbial accessible PDF is largely depleted. Thus, exposure rate of PDF via ruminative mastication necessary to constrain fragments in lag-rumination flow-paths seems more important in constraining ruminal escape than fragment size per se. Presumably, the small amount of undigested PDF is associated with microbial inaccessible PDF within the smallest fragments.

Huhtanen and Kukkonen (1995) indicated the need to consider an additional ruminal residence time to that associated with mass action turnover; a pool such as that accounted for by the "lag-rumination" flow-paths. Such considerations raise questions as to what is determining the degradation rate of PDF: efficiency of ruminative comminution or intrinsic chemical attributes of the forage fragment.

Efficiency of Ruminative Mastication

Data obtained from the same corn silage used in this study (Deswysen et al., 1988) and from grazed bermudagrass (Lascano, 1979) are summarized in Table 5. The results in Table 5 indicate an approximately fourfold larger serial MRA for ingestively masticated fragments and ruminal fragments of corn silage vs. bermudagrass, which required longer ruminal residence times (38.2 vs. 25.0 h) to achieve ruminal escape to the feces of fragments of similar serial MRA (359 vs. 316 µm). The mean comminution rate was estimated by assuming that ruminative comminution resulted in a successive series of exponentially distributed fragments over time (Figure 13 in Ellis et al., 2000). Therefore, the difference between the ln of the serial MRA entering and the ln of serial MRA of fragments escaping the rumen divided by the mean ruminal residence time of such fragments should yield an estimate of the ln of the fractional comminution rate. Mean fractional comminution rates of 0.06/h for bermudagrass hay and 0.075/h for corn silage were observed (Table 5).

Deswysen and Ellis (1990) and Deswysen et al. (1993) reported variations (P < 0.05) in patterns of rumination among individual heifers, variations that seemed related to voluntary forage intake and estimated genetic merit of individual heifers. Based on the current results, we suggest that such variation in patterns of rumination among individual animals may be associated with variations in efficiency of rumination comminution among individual animals and to be heritable.

Validity of Non-In Vivo Degradation Data

Non-in vivo methods commonly used to estimate the kinetics of PDF degradation commonly use ground fragments and therefore do not account for effects of ruminal residence time involved in comminution of fragments as a prerequisite for PDF degradation. The potential significance of comminution processes of ruminative mastication was investigated by comparing estimates of rate of PDF for the same forage by in vivo vs. non-in vivo methods.

The ruminal indigestible marker flux and pool size method (ruminal evacuation method; Poppi, 1979) provide estimates of mean ruminal residence time computed as mean ruminal pool size of UDF/mean daily flux of UDF. Computing PDF as NDF – UDF and assuming mass action dilution of the flux of UDF and undegraded PDF, a single age-constant rate of PDF fractional degradation rate can be estimated as the reciprocal of its mean ruminal residence time due to degradation: the sum of + 3/ + 1/k₂ (Ellis et al., 2005). Thus, the mean PDF degradation rate is equivalent to the reciprocal of the sum of + 3/ + 1/k₂, if a G3/E degradation model is assumed.

Of the six experiments summarized in Table 7, five reported rates of PDF degradation consistently greater than for rates of in situ degradation. A reason for the difference in the results of Lund et al. (2002) is not obvious. Other reports indicate that differences in the number of fibrolytic bacteria (Mackie and Meyer, 1986) and levels of fibrolytic enzymes (Huhtanen and Khalie, 1992; Noziere and Michalet-Doreau, 1996; Huhtanen et al., 1998) are lower for in situ vs. in vivo environments. Huhtanen et al. (1998) suggested that the surface area of bag or forage mass per volume of bag was as critical as pore size. Thus, a critical interchange of microbes and substrates may occur for the in situ vs. the in vivo environment as well as exclusion of effects of ruminative comminution.

Grinding such forages to escape a 1-mm screen should yield a distribution of fragments having serial MRA somewhat smaller than that escaping ruminal digesta (316 to 359 µm; Table 5). However, if the buoyancy was large due to fermentation and entrapment of fermentation gases from microbially accessible surface levels of PDF, milled fragments of this size would not escape (Histrov et al., 2003). It was concluded that ruminal in situ methods may underestimate PDF degradation rate and that such underestimation may involve unfettered mobility of in situ confined fragments within the normal and sequential flow-paths of the lag-rumination and mass-action turnover pools.

	Implications

Top Abstract Introduction Materials and Methods Results and Discussion Implications Literature Cited

 
A fragment size-dependent time delay of 8 to 45 h before onset of digestion of potentially digestible fiber of progressively larger ingestively masticated forage fragments seems causal of an age-dependent onset of degradation from the size-variable mixture of masticated fragments. This age-dependent delay must be considered in integrating digestion and passage models. Rate of comminution of fragments by rumination may be first-limiting to degradation rate of potentially digestible fiber and raises questions concerning the interpretation of results from digestion procedures that preclude effects of ruminative mastication, such as in vitro digestion of finely ground samples.

	Footnotes

 
² Current address: CIAT, Division of Tropical Pastures, Cali, Columbia.

¹ Correspondence: 2471 TAMU (phone: 979-845-5063; fax: 979-845-5292; e-mail: w-ellis{at}tamu.edu).

Received for publication December 17, 2003. Accepted for publication March 30, 2005.

	Literature Cited

Top Abstract Introduction Materials and Methods Results and Discussion Implications Literature Cited

 


Aitchison, E. M., M. Gill, J. Francis, and M. S. Dhanoa. 1986. Comparison of methods to describe the kinetics of digestion and passage of fibre in sheep. J. Sci. Food Agric. 37:1065–1072.

Akin, D. E. 1982. Microbial breakdown of feed in the digestive tract. Pages 201–223 in Nutritional Limits to Animal Production from Pastures. J. B. Hacker, ed. Commonwealth Agricultural Bureaux, Slough, U.K.

Akin, D. E., and W. Barton, III. 1983. Rumen microbial attachment and degradation of plant cell walls. Fed. Proc. 42:114–121.[Medline]

Carrette-Carreon, F. O. 2001. Kinetics of digestion and flow of plant tissue residues in ruminants. Ph.D. Dissertation, Texas A & M Univ., College Station.

Carrette-Carreon, F. O., C. E. Cole, J. H. Matis, W. C. Ellis and C. Lowe. 2001. Models for describing kinetics of fiber digestion in the rumen. J. Anim. Sci. 79(Suppl. 1):288. (Abstr.)

Deswysen, A. G., P. Dutilleul, J. P. Godfrin, and W. C. Ellis. 1993. Nycterohemeral eating and ruminating patterns in heifers fed grass or corn silage: Analysis by finite Fourier transform. J. Anim. Sci. 71:2739–2747.[Abstract]

Deswysen, A. G., and W. C. Ellis. 1990. Fragmentation and ruminal escape of particles as related to variations in voluntary intake, chewing behavior and extent of digestion of potentially digestible fiber in heifers. J. Anim. Sci. 68:3871–3879.[Abstract]

Deswysen, A. G., K. R. Pond, E. Rivera-Villareal, and W. C. Ellis. 1988. Effects of time of day and monensin upon the distribution of different size particles within digestive tract sites of heifers fed corn silage. J. Anim. Sci. 67:1773–1783.

Ehle, F. R. 1983. Influence of particle size on determination of fibrous feed components. J. Dairy Sci. 52:653–657.

Ellis, W. C., M. Mahlooji, and J. H. Matis. 2005. Modes for estimating parameters of neutral detergent fiber digestion by ruminal microorganisms. J. Anim. Sci. 83:1591–1601.[Abstract/Free Full Text]

Ellis, W. C., J. H. Matis, T. M. Hill, and M. R. Murphy. 1994. Methodology for estimating digestion and passage kinetics of forages. Pages 682–756 in Forage Quality, Evaluation, and Utilization. G. C. Fahey, ed. Univ. of Nebraska, Lincoln.

Ellis, W. C., D. Poppi and J. H. Matis. 2000. Feed Intake in Ruminants: Kinetic Aspects. Pages 335–363 in Farm Animal Metabolism and Nutrition: Critical Reviews. J. P. F. D’Mello, ed. Commonwealth Agric. Bur. Int., Oxon, U.K.

Ellis, W. C., D. P. Poppi, J. H. Matis, H. Lippke, T. M. Hill, and F. M. Rouquette, Jr. 1999. Dietary-digestive-metabolic interactions determining the nutritive potential of ruminant diets. Pages 423–481 in Proc. 5th Int. Symp. Nutr. Herbivores. H.-J. G. Jung and G. C. Fahey, Jr., ed. Am. Soc. Anim. Sci. Savoy, IL.

Ellis, W. C., M. J. Wylie, and J. H. Matis. 1988. Dietary digestive interactions determining the feeding value of forages and roughages. Pages 177–225 in Feed Science. Vol B-4. E. R. Orskov, ed. Elsevier Science Publishers, Amsterdam, The Netherlands.

Emanuele, S. M., and C. R. Staples. 1988. Effect of forage particle size on in situ digestion kinetics. J. Dairy Sci. 71:1947–1954.[Abstract/Free Full Text]

Gill, M., J. France, R. C. Siddons, and M. S. Dhanoa. 1984. Compartmental models incorporating particle size for the estimation of digestion in the rumen. Pages 142–153 in Techniques in Particle Size Analysis of Feed and Digesta in Ruminants. P. M. Kennedy, ed. Can. Soc. Anim. Sci., Edmonton, Alberta.

Goering, H. K., and P. J. Van Soest. 1970. Forage Fiber Analyses (Appartus, Reagents, Procedures, and Some Applications). Agric. Handbook No.379. ARS-USDA, Washington, DC.

Hannah, W. W., W. G. Monson, and G. W. Burton. 1976. Histological and in vitro digestion study of 1- and 4-week old stems and leaves of high and low quality bermudagrass genotypes. Agron. J. 68:219–222.[Abstract/Free Full Text]

Hristov, A. N., S. Ahvenjarvi, T. A. McAllister, and P. Huhtanen. 2003. Composition and digestive tract retention time of ruminal particles with functional specific gravity greater or less than 1.02. J. Anim. Sci. 81:2639–2648.[Abstract/Free Full Text]

Huhtanen, P., and S. Jaakkola. 1993. The effects of forage preservation method and proportion of concentrate on digestion of cell wall carbohydrates and rumen digesta pool size in cattle. Grass Forage Sci. 48:155–165.

Huhtanen, P., S. Jaakkola, and U. Kukkonen. 1995. Ruminant plant cell wall digestibility estimated from digestion and passage kinetics utilizing mathematical models. Anim. Feed Sci. Technol. 52:159–173.

Huhtanen, P., and H. Khalie. 1992. The effect of sucrose supplements on particle-associated carboxymethyl-cellulose (EC 3.2.1.4) and xylanase (EC 3.2.1.8) activities in cattle given grass-silage based diets. Br. J. Nutr. 67:245–255.[Medline]

Huhtanen, P., and U. Kukkonen. 1995. Comparison of methods, markers, sampling sites and models for estimating digesta kinetics in cattle fed at two levels of intake. Anim. Feed Sci. Technol. 52:141–158.

Huhtanen, P., A. Vanhatalo, and T. Varvikko. 1998. Enzyme activities of rumen particles and feed samples incubated in situ with different types of cloth. Br. J. Nutr. 79:161–168.[Medline]

Lascano, C. E. 1979. Determinants of grazed forage voluntary intake in cattle. Ph.D. Diss., Texas A&M Univ., College Station.

Mackie, R. I., and J. Meyer, 1986. Microbial evaluation of the intraruminal in sacculus technique. Appl. Environ. Microbiol. 51:622–629.[Abstract/Free Full Text]

Mahlooji, M. 1985. Fiber digestion from different size particles produced by mastication. Ph.D. Diss., Texas A&M Univ., College Station.

Morrison, M., and J. Miron. 2000. Adhesion to cellulose by Ruminococcus albus: A combination of cellulosomes and pil-proteins? FEMS Microbiol. Lett. 185:109–115.[Medline]

Noziere, P. and B. Michalet-Doreau. 2000. In Sacco Methods. Pages 233–253 in Farm Animal Metabolism and Nutrition: Critical Reviews. J. P. F. D’Mello, ed. Commonwealth Agric. Bur. Int., Oxon, U.K.

Playne, M. J., W. Khumnualthong, and M. G. Echevarria. 1978. Factors affecting the digestibility of oesophageal fistula samples and hay samples in nylon bags in the rumen of cattle. J. Agric. Sci. (Camb.) 90:193–204.

Poppi, D. P. 1979. Voluntary intake of leaf and stem fractions of tropical forages offered to cattle and sheep. Ph.D. Diss., Queensland Univ., Brisbane, Australia.

Pond, K. R., W. C. Ellis, and D. E. Akin. 1984a. Ingestive mastication and fragmentation of forages. J. Anim. Sci. 58:1567–1575.[Abstract/Free Full Text]

Pond, K. R., W. C. Ellis, C. E. Lascano, and D. E. Akin. 1987. Fragmentation and flow of grazed coastal bermudagrass through the digestive tract of cattle. J. Anim. Sci. 65:609–618.

Pond, K. R., E. A. Tolley, W. C. Ellis, and J. H. Matis. 1984b. A method for describing the weight distribution of particles from sieved forage. Pages 123–134 in Techniques in Particle Size Analysis of Feed and Digesta in Ruminants. P. M. Kennedy, ed. Can. Soc. Anim. Sci. Edmonton, Alberta.

Rinne, M., P. Huhtanen, and S. Jaakkola. 1997. Grass maturity effects on cattle fed silage-based diets. 2. Cell wall digestibility, digestion and passage kinetics. Anim. Feed Sci. Technol. 67:19–35.

Rinne, M., P. Huhtanen, and S. Jaakkola. 2002. Digestive processes of dairy cows fed silages harvested at four stages of grass maturity. J. Anim. Sci. 80:1986–1998.[Abstract/Free Full Text]

Robles, A. Y., R. L. Belyea, F. A. Martz, and M. F. Weiss. 1980. Effect of particle size upon digestible cell wall and rate of in vitro digestion of alfalfa and orchard grass forages. J. Anim. Sci. 51:783–792.

Sutherland, T. M. 1986. Particle separation in the forestomach of sheep. Pages 43–73 in Aspects of Digestive Physiology in Ruminants. A. Dobson, and M. H. Dobson, ed. Cornell Univ. Press, Ithaca, NY.

Vaage, A. S., J. S. Shelford and G. Moseley. 1984. Theoretical basis for the measurement of particle size when sieving elongated feed particles. Pages 76–82 in Techniques in Particle Size Analysis of Feed and Digesta in Ruminants. P. M. Kennedy, ed. Can. Soc. Anim. Sci. Edmonton, Alberta.

Von Milgen, J. and R. Baumont. 1995. Models based on variable fractional digestion rates to describe ruminal in situ digestion. Br. J. Nutr. 73:793–807.[Medline]

Von Milgen, J., M. R. Murphy and L. L. Berger. 1991. A compartmental model to analyze ruminal digestion. J. Dairy Sic. 74:2515–2529.

Van Soest, P. J. 1982. The Nutritional Ecology of Ruminants. O and B Books, Corvallis, OR.

Weimer, P. J., J. M. Lopez-Guisa and A. D. French. 1990. Effect of cellulose fine structure on kinetics of its digestion by mixed ruminal microorganisms in vitro. Appl. Environ. Microbiol. 56:2421–2429.[Abstract/Free Full Text]

Weimer, J. P. 1996. Why don’t ruminal bacterial digest cellulose faster? J. Dairy Sci. 79:1496–1502.[Abstract]

Wylie, M. J., W. C. Ellis, J. H. Matis, E. M. Bailey, W. D. James and D. E. Beever. 2000. The flow of forage particles and solutes through segments of the digestive tracts of cattle. Br. J. Nutr. 83:295–306.[Medline]

This article has been cited by other articles:

				J. L. Foster, A. T. Adesogan, J. N. Carter, A. R. Blount, R. O. Myer, and S. C. Phatak Intake, digestibility, and nitrogen retention by sheep supplemented with warm-season legume haylages or soybean meal J Anim Sci, September 1, 2009; 87(9): 2899 - 2905. [Abstract] [Full Text] [PDF]

				J. C. Burns ASAS Centennial Paper: Utilization of pasture and forages by ruminants: A historical perspective J Anim Sci, December 1, 2008; 86(12): 3647 - 3663. [Abstract] [Full Text] [PDF]

				W. C. Ellis, M. Mahlooji, and J. H. Matis Models for estimating parameters of neutral detergent fiber digestion by ruminal microorganisms J Anim Sci, July 1, 2005; 83(7): 1591 - 1601. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ellis, W. C. Articles by Matis, J. H. Search for Related Content PubMed PubMed Citation Articles by Ellis, W. C. Articles by Matis, J. H.