NUTRITION, FEEDING, AND CALVES Digestion Kinetics of Fiber: Influence of In Vitro Buffer pH Varied Within Observed Physiological Range1 R. J. GRANT2 and S. J. WEIDNER Department of Animal Science University of Nebraska Uncoln 68583-0908
Abbreviation key: AH = alfalfa hay, DH = bromegrass hay, CS com silage, IR indigestible residue, Kd fractional rate constant of digestion.
ABSTRACT
= =
In vitro buffer pH reflective of the diurnal variation in ruminal pH was evaluated for its impact on digestion kinetics of NDF from three forage sources. Alfalfa hay, bromegrass hay, and com silage were incubated in phosphatebicarbonate buffer solution adjusted to pH 6.8, 6.5, 6.2, 6.0, 5.8, or 5.5 using 1 M citric acid. Ash-free NDF was measured after 0,6, 12, 18, 24, 48, 72, and 96 h of fennentation. The experiment was replicated three times; kinetic parameters of fiber digestion were estimated by logarithmic transformation and by linear and nonlinear regression procedures. Lag in NDF digestion increased as pH fell from 6.8 to 6.5 and again when pH decreased below 6.0. Decreasing buffer pH below 6.2 dramatically reduced NDF digestion rate for alfalfa hay and com silage but had no significant effect on NDF digestion rate of bromegrass hay. Lag in NDF digestion increased below pH 6.0 for both alfalfa and com silage but increased only when pH fell below 6.2 for bromegrass. Results suggest that lowered pH exerts its negative effect on NDF digestion between pH 6.2 and 5.8, as evidenced by increased lag and decreased rate, and that critical pH and specific, affected component of digestion varied among forages. (Key words: digestion kinetics, forage fiber, pH, in vitro buffer)
INTRODUCTION
Received September 13, 1991. Accepted November 25, 1991. lPublished willi the approval of the director as Paper Number 9624, 10urnal Series, Nebraska Agricultural Research Division. 2R eprint requests. 1992 1 Dairy Sci 75:1060-1068
=
Ruminal pH, which is depressed as a result of feeding large amounts of readily available carbohydrates, may reduce fiber digestion (2, 14). Stewart (22) demonstrated that fiber digestion was inhibited when pH dropped from 6.9 to 6.0, and cellulolytic populations fell simultaneously from 1()6 to 103/00. Mould et al. (13) postulated two factors that accounted for the depression in fiber digestion when sheep were fed high starch diets. The digestibility depression that was alleviated with bicarbon~ bufferng of ruminal conterns was termed "pH effect"; the residual depression that peIhaps was due to microbial adaptations to starch was termed "carbohydrate effect". In their experiments, pH effect appeared to be three times larger than carbohydrate effect. Low ruminal pH is likely to reduce fiber digestion in high producing dairy cows fed diets containing 50 to 60% concentrates. Robinson et al. (15) fed dairy cows a diet containing 32% starch in the concentrate at 24 kg/d and found average ruminal pH to be 6.1. The diurnal pattern for ruminal pH indicated that it was below 6.2 for 70 to 80% of each day (15). If the ruminal pH of lactating dairy cows is below 6.0 for extended periods, and if low pH is postulated to have a substantial negative impact on fiber digestion, determining the mechanism by which pH alters kinetics of fiber digestion is important. Grant and Mertens (5) showed that decreasing in vitro buffer pH from 6.8 to 5.8 increased digestion lag and decreased rate of NDF digestion. Using continuous culture, Hoover et al. (9) found pH 6.5 to be optimal for fiber and OM digestion, whereas pH of 5.5 or 7.5 signif-
1060
IN VTIRO BUFFER pH AND FIBER DIGESTION
icantly decreased fiber digestion. Data suggest a fairly narrow pH range for optimal fiber digestion in the romen, but threshold pH, below which lag increases and rate decreases, was not defined. Grant and Mertens (6) developed an in vitro buffer system capable of controlling pH between 5.5 and 6.8 throughout 96 h of batch fennentation. Using this system, the specific effect of pH on fiber digestion kinetics can be determined for forage substrates commonly used in dairy cow diets. JlDlg and Varel (10) showed clearly that forage source had a significant impact upon ruminal microbial populations and subsequent in vitro fibrolytic activity of ruminal inoculum. The nature of the forage by pH interaction on fiber digestion kinetics must be determined lDlder carefully controlled, in vitro conditions to lDlderstand how fiber digestion proceeds in vivo when cows consume diets that lower ruminal pH. The objective of this research was to determine the impact of buffer pH on in vitro NDF digestion kinetics of common forage substrates when pH was varied within the diurnal range observed for lactating dairy cows fed diets that depress ruminal pH. MATERIALS AND METHODS
sample Preparation
Forages were dried at 60°C for 48 h and ground through a I-mm screen using a Wiley mill (Arthur H. Thomas Co., Philadelphia, PA). Early bloom, first-cutting alfalfa hay (AH) contained 19% CP and 48% NDF (ashfree, DM basis); mature, first-cutting bromegrass hay (BH) contained 11.5% CP and 68% NDF; and corn silage (CS) at physiological maturity contained 8.0% CP and 54% NDF. A 300-mg sample of each of the three substrates was weighed into 5o-ml polypropylene tubes. In Vitro Procedure
The in vitro system consisted of 5o-ml polyproplyene tubes with gas-release rubber stoppers. The buffer solution was that described by Goering and Van Soest (3) adjusted to pH 6.8, 6.5, 6.2, 6.0, 5.8, or 5.5 with 1 M citric acid as described by Grant and Mertens (6). This buffer system maintained a stable pH throughout 96 h of fermentation with
1061
forage substrates (5, 6). Citrate, in place of distilled water, was added to C02-Saturated buffer solutions at 39°C until the desired pH was obtained. Use of 1 M citric acid to adjust pH of the buffer solution had no effect on NDF digestion in vitro for forage or for forage plus starch substrates (6). Thus, even though citrate can be fermented by ruminal bacteria, use of this acid to adjust pH had no demonstrable confounding effect on our ability to measure NDF digestion kinetics in vitro. Given its tricarboxylic nature, citric acid buffered more strongly within the pH range 5.5 to 6.8 than did phosphoric acid (6, 23). All buffer solutions were reduced (3), warmed to 39°C, and saturated with C~ prior to dispensing. The ruminal fluid inoculum was obtained from a steer fed medium quality AH. At the time of collection, the pH of the ruminal fluid was measured and averaged 6.35 ± .21 for all replicates. Ruminal inoculum of steers fed AH has been shown to degrade fiber fractions of forages more rapidly than inoculum from steers fed BH or switchgrass, as judged by 48-h in vitro cell-wall digestibility (10). Furthermore, inoculum of ruminal microbes selected at pH 5.5 did not digest fiber any better in batch culture at pH 5.0 or 5.5 than inoculum from bacteria selected for pH 6.5 (19). Therefore, inoculum pH of 6.35 should not have had a major impact on fiber digestion kinetics within the pH range in our study. While purging with CO2, a 20% solution of ruminal fluid in buffer of appropriate pH was dispensed in 3o-ml aliquots per tube using a Unispense IT automatic dispensing machine (Wheaton Instruments, Millville, NJ). Tubes were sealed immediately with gas-release stoppers, gently swirled, and allowed to ferment for the appropriate time. The pH of the contents of the tube was measured at each time for each substrate to monitor stability of buffer pH, which did not decrease more than .10 pH unit for any substrate during 96 h of fermentation. Tubes containing AH or BH experienced a decline in pH of no more than .07 unit, whereas pH of CS tubes declined a maximum of .10 unit, presumably because of the presence of rapidly fermented starch. Thus, actual pH for each treatment remained separated sufficiently for the results to be biologically important. Preliminary experiments were to compare this in vitro system with the optimal system Journal of Daily Science Vol. 75, No.4, 1992
1062
GRANT AND WEIDNER
described by Grant and Me~ns (4). Those researchers determined that a system using continuous C~ gassing and reducing agents provided the greatest opportunity to detect differences in substrates related to their intrinsic properties. This system used sodium sulfide nonahydrate and cysteine hydrochloride as reducing agents and used tryptone and microminerals as described by Goering and Van Soest (3) as nutritive additives. In vitro systems that do not maximize digestion kinetics may not detect differences in substrate that are due to constraints of the system. Preliminary fiber digestion rates (per hour) at pH 6.8 for our system versus the one described by Grant and Mertens (4, 6) were the following: for AR, .091, .070; for BH, .058, .051; and for CS, .032, .035. Digestion kinetics determined using this system compared favorably with that obtained using the optimal system described. Fermentation times were 0, 6, 12, 18, 24, 48, 72, and 96 h. Ash-free NDF was measured (3). Modification included use of heat-stable amylase (Termamyl 120L, Novo Laboratories, Inc., Danbury, at filtering for CS. The in vitro experiment was replicated three times.
en
Statistical Analysis
The model for kinetics of NDF digestion was that described by Mertens and Loften (11).
where Y
=
Do Kd L I
= = = =
NDF residue at time t of in vitro fermentation, digestible NDF fraction, fractional rate constant of digestion, discrete lag, and indigestible NDF fraction.
The potentially digestible fraction at each fermentation time was transformed logarithmically, and linear regression was used to estimate digestion parameters. These estimates were used as starting points for iteration by the nonlinear regression procedure of SAS (17). Parameters of the digestion model estimated by nonlinear regression were analyzed using the general linear models procedure of SAS (17); the model included factors for replicate, forage type, pH level, and interactions. Means for which a significant main effect was detected were compared using Student-NewmanKeuls test (17). Regression analysis (17) was used to examine the relationship between buffer pH and fiber digestion kinetics. Unless otherwise stated, statistical significance was at P < .05.
TABLE 1. Comparison of in vitro kinetics of NDF digestion averaged over all pH conditions studied and at pH 6.8 only. ,
Item pH 5.5 to 6.8 Alfalfa bay Bromegrass bay Com silage SE pH 6.83 Alfalfa bay Bromegrass bay Com silage SE
Lag
Estimates of digestion kinetics IR. l Rate (Xci>
(h)
(JIl)
4.55 b 5.4Ob
.043 .049 .045 .004
22.63b 19.(j(f 27.61.97
55.70b 70.7252.59b 2.17
.070-
21.90 19.34 25.19 .85
57.10 70.40 62.05 1.90
6.8~
.77 1.79b 2.6Q&b 3.47-
.25
('Ai)
.051 b .062ab
.005
a,b·"Means within a colmnn with different superscripts differ (P
mn2
< .05).
lIR. = Indigestible NDF residue expressed as a percentage of DM in original sample. 2PED = Potential extent of NDF digestion expressed as a pen:entage of NDF in original sample. 3mcludes three replicates of experima1t as descnDed in Materials lIDd Methods plus one additional replicate of alfalfa bay incubated at pH 6.8.
Journal of Dairy Science Vol. 75. No.4. 1992
1063
IN VI1RO BUFFER pH AND FIBER DIGESTION RESULTS
When averaged over all pH conditions evaluated in this study, CS had a significantly longer lag and higher indigestible residue (IR) than All or BH (fable 1). Bromegrass hay had the least IR. Forage type had no effect on Kd (fable 1). These results are atypical compared with those usually reported for fiber digestion of these forages (11). The longer lags and equivalent rates of fiber digestion reflect inclusion of digestion data for pH 5.5, 5.8, and 6.0. To demonstrate that the in vitro system used in this study actually generated nonnal fiber digestion kinetics at pH 6.8, Table 1 illustrates the effect of forage source upon kinetics of NDF digestion at pH 6.8 only. Alfalfa hay had a significantly shorter lag and faster rate of NDF digestion than BH and CS. Indigestible residue and potential extent of digestion did not differ among forages. These values of NDF digestion at pH 6.8 agree with previously reported kinetic data for these forages (4, 5, 6). We concluded that the in vitro system used in our study provided ample opportunity for detecting differences in forage substrates related to their structural properties and for quantifying the interaction between forage physicochemical properties and pH. Averaged over substrates, lag increased as pH declined (Figure 1). Lag increased significantly as pH fell from 6.8 to 6.5 and again when pH declined from 6.0 to 5.8. Buffer pH had no effect on lag between pH 5.5 and 5.8 or on lag in the range of 6.0 to 6.5. Rate of NDF digestion decreased significantly below pH 6.2 when averaged over all forages (Figure 2). No difference in Kd was significant between pH 6.2 and 6.8. Rate of NDF digestion was not affected by pH below 6.0, although Kd still tended to decline (P < .10). No differences in IR were significant because of pH level (Figure 3). Interestingly, a noticeable but nonsignificant increase in IR occurred between pH 6.2 and 6.0, which paralleled the changes ~ served in Kd (Figure 2). A significant pH by forage interaction was found (P < .012). Table 2 illustrates the response in kinetics of NDF digestion to pH for All, BH, and CS. Lag did not increase significantly for All or CS until pH dropped below 6.0. In contrast, lag increased for BH below pH 6.2. Com silage showed the most pronounced effect of low pH on lag. Rate of NDF digestion for BH showed no response to pH, although Kd decreased nonsignificantly as pH
declined. For All and CS, Kd decreased when pH dropped below 6.2. The lower than expected Kd for All at pH 6.5 was due to a low value for one replicate of the experiment. For all forages, IR increased with decreasing pH, although differences were not significant. To facilitate comparison with reported values, p0tential extent of digestion was calculated. Extent of NDF digestion declined nonsignificantly as pH fell. Table 3 illustrates the potential effect of altering pH on apparent extent of forage fiber digestion. Using equations in Miller and Muntifering (12) and assuming a rate of fiber passage of .OS/h, the apparent extent of digestion for All and CS was lower than for BH as pH declined. The potential apparent extent of digestion decreased below pH 6.2 for all forages. DISCUSSION
The ability of ruminal fibrolytic bacteria to digest plant cell wall is a function of bacterial species, forage species, and the mminal environment in which digestion occurs. Research
10
a
8
--
6
~
4
2
5.5
5.8
6.0
6.2
6.5
6.8
pH Figure 1. Effect of in vitro buffer pH on lag of NDF digestion. Letters (a,b,c) denote significant differences (P
< .05). Journal of Dairy Science Vol. 75. No.4. 1992
1064
GRANT AND WEIDNER 25
.06
.05
--
24
'#.
-
G)
.04
~
'C
.c
~ G)
10 a:
23
'0 G) 'G)
:a
.03
i
22
G)
0)
=e
.02
.5 21
.01 20
5.5
5.5
5.8
6.0
6.2
6.5
6.8
pH
5.8
6.0
6.2
6.5
6.8
pH Figom 3. Effect of in vitro buffer pH on NDF indigestible residue.
Figure 2. Effect of in vitro buffer pH on digestion rate of NDF (KII>. Letters (a,b) denote significant differences (P < .05).
has focused on pH as a primary factor governing fiber digestion (20). Slyter (19) was unable to select for aciduric fibrolytic bacteria after 3 wk in a pH 5.5 continuous fennenter, and he concluded that this sort of microbial adaptation to low pH was unlikely to occur in the rumen as well. Thus, consideration of diurnal fluctuations in ruminal pH between pH 5.5 and 6.8 is critical. Delineation of threshold pH for various forage species with respect to kinetics of NDF digestion was a primary objective of our study. We examined AB, a representative legume, BH, a C3 grass, and CS, a grass with considerable inherent starch to impact NDF digestion potentially. These three forages represent a wide range in plant cell-wall characteristics, are commonly fed to dairy cows, and should therefore be well-suited substrates for quantifying the interaction between in vitro pH and forage NDF digestion kinetics. Detennination of differences in the in vitro digestion Joumal of Dairy Science Vol. 75, No.4, 1992
kinetics of forage fiber types within a physiological range of pH would supply rate and lag infonnation essential for modeling the impact of diurnal pH fluctuations on total ruminal fiber digestion. The existence of a significant pH by forage interaction in our study complicates the interpretation of the main effect means for lag and Kd. Examination of response means, however, allows general conclusions to be drawn concerning pH threshold levels for lag and Kd' In our study, as pH declined from 6.8 to 5.5, lag in NDF digestion increased, and Kd decreased. Although differences in fiber digestion were significant among forages and were most greatly affected by low pH, NDF digestion suffered severely below pH 6.2 to 6.0 for all forages (Table 2; Figures 1 to 3). Using inoculurn from a pH 5.5 fennenter, Slyter (19) found 13% NDF digestion during 24-h periods at pH 5.5, a 45% NDF digestion
1065
IN VITRO BUFFER pH AND FIBER DIGESTION TABLE 2. Effect of forage type and pH on in vitro digestion kinetics of NDF.t Lag
Rate (Kd)
(h)
(/h)
---(%)---
.05~ .045 ab .055a .000b .024b .036b .003 .051 .050 .053 .053 .045 .039 .001 .062a .063a .065 a .029b .031 b .019b
21.94 22.14 21.90 21.99 23.66 24.12 .29 19.34 18.44 18.46 20.33 20.30 20.73 .44
56.49 56.95 56.53 57.59 51.71 54.93 .52
25.19 26.92 26.14 29.21 28.54 29.64 .67
60.56 59.24 54.56 46.15 53.92 41.10 1.69
Forage
pH
Alfalfa hay
6.8 6.5 6.2 6.0 5.8 5.5
1.79b 4.00b 1.92b 3.89b 7.58a 8.10& .38
6.8 6.5 6.2 6.0 5.8 5.5
2.60b 3.47b 4.47b 6.82 a 6.89a 8.12a .26 3.47b 6.32 b 5.90b
SE Bromegrass hay
SE Com silage
6.8 6.5 6.2 6.0 5.8 5.5
4.nb
9.5~ 11.07a
.28
SE
.002
70.75 73.25 71.95 69.83 68.44 70.08 .41
a,"Means within a column and substrate combination differ significantly (P < .05). lAccuracy of fitting data to the model for each pH level as indicated by R 2 was pH 6.8, .945; pH 6.5, .972; pH 6.2, .989; pH 6.0, .986; pH 5.8, .865; pH 5.5, .905. 2IR = Indigestible NDF residue expressed as a percentage of DM in original sample. Potential extent of NDF digestion as a percentage of NDF in original sample. 3pED
=
during 24-h periods at pH 6.5, but only 1% digestion at pH 5.0. In our study, kinetics of fiber digestion were influenced most profoundly for all forages between pH 6.2 and 5.8. At pH 5.5, Kd had become minimal, whereas lag in digestion was approaching 10 h. A pH of 5.5 may be the lower practica1lirnit for fiber digestion in the rumen. Below pH 5.5, little fiber digestion occurs, even if bacteria remain alive (8, 19). Fortunately, diurnal variations in pH are such that it rarely falls below 5.5 for very long (15). Given the linear increase in lag below pH 6.0 observed in our study (R2 = .98) and the linear decrease in rate below pH 6.2 (R2 = .97), little fiber digestion evidently occurs below pH 5.5. These observations are consistent with severely decreased numbers of cellulolytic bacteria in steers fed all concentrate with nnninal pH below 5.3 (21).
Mould et al. (13) described the pH effect on fiber digestion as biphasic: pH reduction from 6.8 to 6.0 resulted in moderate depressions in
fiber digestion, and pH reduction below 6.0 resulted in severe inhibition. Although mechanisms involved at low pH (below 6.0) may concern transmembrane proton flux (16), attachment phenomena may be associated with moderate declines in pH from 6.8 to 6.0. Activity of isolated fibrolytic enzymes remains high in this pH range (7). Also, significant decreases in numbers of cellulolytic bacteria are not associated with small declines in pH within the range of pH 6.0 to 6.8 (19). Our data showed significant increases in lag as pH fell from 6.8 to 6.5. As buffer pH fell from 6.8 to 6.5, lag increased by 2 h, averaged over all forages. Between pH 6.5 and 6.0, lag was constant, averaging 4.6 h. As pH fell below 6.0, however, lag increased markedly to over 8 h (Figure 1). For individual forages, the increases in lag between pH 6.8 and 6.5 were 2.2 h (AH), .9 h (BH), and 2.9 h (CS). Low pH has been associated with prevention of tight attachment of bacteria to plant cell walls (1). Shriver et al. (18), using continuous culture, Journal of Dairy Science Vol. 75, No.4, 1992
1066
GRANT AND WEIDNER
TABLE 3. PoteDtial effect of pH on apparent extent of forage fiber digestion as estimated by in vitto fermentation. l pH Item
6.8
6.5
Alfalfa hay (e-/Cl-)2
.81S& .47oab
K'y(Kd + K~3
e-Kl-- x K'y(Kd + K~4
Bromegrass hay e-/C-}-
+ K~
e-K';- x KdI(Kd + K~
ABD,%
20.Sb
.S7~
.840"
.796&
.5OS
.495
.51S
.~
5.S
.360b
.38Sb
31.1&
6.0
.823& .442b
n.r:/'
AED,S %
Kd/(Kd
6.2
.42oa 30.7&
.4Q9& 29.0.740'.560"
5.5
.682b 32r:/' .22rf
llAc
.708b .476
.340"
23.1 b
SE
.667b .418b
.031 .019
.27~
.020 1.2
.670b .448 .300b
.020 .006
lS.3c
21.0b
.014
1.1
Com silage (e-K';-)
.841&
.725&
Kd/(Kd + K~
.444&
.5S6&
.374& 22.6&
.401&
-KL
e
P
x Kd/(Kd +
ABO, %
K~
23.S&
.414& 22.6&
.788&
.360b .286 13.1b
b
.616b .383b .23Sb
12.l b
.572b .273b .15Sb 6.4c
.026 .028
.027 1.9
a,b,CMeans within a row with different superscripts differ (P < .OS). lCalculations assume a fiber passage rate (K~ of .OS/h. 2Praction of potentially digestible fiber in rumen when digestion begins at end of lag (12). 3orheoretical maximum proportion of fiber disappearance from the rumen that is due to digestion when lag = 0 (12). 4praction of potentially digestible fiber aetua1Iy digested in rumen (12). SAED = Apparent extent of digestion = (potential extent of digestion) x K'y(Kd +
showed that 65% concentrate diets at pH 5.8, 6.2, 6.6, and 7.0 had 43% fewer microbes attached to fiber particles as pH dropped from 6.2 to 5.8. The significant increase in lag as pH dropped from 6.8 to 6.5 in our study may reflect these altered attachment mechanisms. The nature of the interaction between low pH and attachment to fiber particles is unknown, but we hypothesize that this interaction is at least partially responsible for effects of pH on lag that we observed, especially at moderate pH reductions (pH 6.8 to 6.5; Figure 1). Table 2 illustrates the nature of the pH by forage interaction examined in our study. Alfalfa hay and CS clearly differed from BH, the C3 grass forage, in lag and rate of NDF digestion. The pH at which lag increased significantly was .2 pH units lower in our in vitro system for AH and CS than for BH. As the fiber percentage of the substrate increased, the importance of the low pH effect increased relative to the carbohydrate effect discussed by Mould et al. (13). We compared the two substrates in our study in which NDF came from Jonmal of Dairy Science VoL 7S, No.4, 1992
K~
x e-/CpL (12).
forage only (AH, BH). For BH (68% NDF), lag increased significantly between pH 6.2 and 6.0; for AH (48% NDF), lag increased between pH 6.0 and 5.8. There was no significant pH effect upon Kd for BH. The IR increased nonsignificantly below pH 6.2 for BH, but only below pH 6.0 for AH. Therefore, that initial digestion of the higher fiber forage (BH) as measured by lag apparently was more quickly reduced with lower pH. The component of NDF digestion most acutely affected was lag (Table 2). Even more striking, Kd for BH was not significantly affected by decreasing pH. The difference in Kd between pH 6.8 and 5.5 for BH was .012Jh; for AH, .023Jh; and for CS, .043Jh (Table 2). Similar results for CS NDF, showing a combined negative effect of low pH and starch, were reported by Grant and Mertens (5). These dramatic differences among forage fiber sources in NDF digestion kinetics, if found also in vivo, may suggest feeding strategies that effectively mesh fiber availability for microbial degradation with diurnal pH fluctuation.
IN VI1RO BUFFER pH AND FIBER DIGESTION
Terry et al (23) found that the extent of cellulose digestion by microorganisms in vitro (at 96 h) was dependent on media pH. Immature alfalfa had 48-h cellulose digestions of 6, 15, and 15% at pH 5.5, 6.0, and 6.8. At 96 h of in vitro incubation, 6,16, and 15% of cellulose were digested. We observed nonsignificant decreases in potential extent of digestion as pH declined (fable 2). In contrast with much of the reported research, Miller and Muntifering (12) found a decrease in fescue fiber digestion with increasing grain levels in situ, which was related primarily to decreased potential extent of digestion. Lag and rate effects were not significant contributors to their observed decline in fiber digestibility. Mertens and Loften (11) studied similar levels of starch addition with fescue hay in vitro and found a major negative effect on lag. Although our in vitro results showed a nonsignificant increase in IR with pH depressions of 6.2 to 6.0, pH effects on lag and rate were much more significant. Mertens and Loften (11) were unable to elicit large effects of starch addition on fiber Kd in an in vitro system buffered at pH 6.8 only. However, in our study batch culture and buffers within the pH range of 5.5 to 6.8 showed dramatic pH effects on rate and lag, perhaps unique to forage type. Why kinetic analysis of starch effects on fiber digestion, as related to extracellular pH, yields such different results in situ versus in vitro is unknown. Perhaps this difference is due to the fluctuating ruminal pH inherent in the in situ digestion technique compared with the stable pH maintained throughout 96 h of in vitro batch fermentation. Although use of Kd derived from in vitro systems to estimate apparent extent of digestion in vivo necessarily restricts inferences drawn, there is a tremendous potential range possible (Table 3) in ruminal fiber digestion for various forage types under conditions in which pH varies substantially. CONCLUSIONS
Decreasing pH moderately increased fiber digestion lag between pH 6.8 and 6.5 and more severely below pH 6.0. Averaged over all substrates, decreasing pH below 6.2 reduced Kd significantly. However, the effect of low pH on fiber digestion kinetics appeared to be a function of forage fiber source. In particular,
1067
low pH had no significant effect on Kd for BH compared with AH or CS. Common feeding practices result in low ruminal pH in lactating dairy cows. Thus, kinetic data reported here should suggest in vivo studies to help devise feeding strategies that recognize the multifaceted nature of the interaction between low pH and kinetics of forage fiber digestion for specific fiber sources. ACKNOWLEDGMENTS
The authors wish to thank Mary Rieger for her assistance in completing much of the laboratory analysis. REFERENCES 1 Cheng, K. 1., C. S. Stewart. D. Dinsdale, and J. W. Costerton. 1984. Electron microscopy of bacteria involved in the digestion of plant cell walls. Anim. Feed Sci. Teclmol. 10:93. 2 EI-Sbazly, K .• B. A. Dehority, and R R. Johnson. 1961. Effect of starch on the digestion of cellulose in vitro and in vivo by rumen microorganisms. 1. Anim. Sci. 20:268. 3 Goering, H. K., and P. J. Van Soest. 1970. Forage fiber analyses (apparatus, reagents, procedures, and some applications). Agric. Handbook No. 379. ARSUSDA, Washington. DC. 4 Grant, R. 1., and D. R. Mertens. 1990. Effect of in vitro fermentation technique upon kinetics of neutral detergmt fiber digestion. 1. Dairy Sci. 73(Suppl. 1): 129.(Abstr.) S Grant, R 1., and D. R Mertens. 1990. Impact of forage type, corn starch addition and buffer pH upon in vitro digestion kinetics of neutral detergent fiber. 1. Anim. Sci. 68(Suppl. 1):S89.(Abstr.) 6 Grant, R. 1., and D. R. Mertens. 1990. A buffer system for pH control and evaluation of pH effects upon fiber digestion in vitro. Page 109 in US Dairy Forage ReselttCh Center 1989 research snmmaries. ARS-USDA, Madison, WI. 7 Groleau, D., and C. W. Forsberg. 1983. Partial characteri2ation of !he extraee11u1ar carboxymethylcellulase activity produced by the rumen bacterium Bacteroides sucdnogenes. Can. 1. Microbiol. 29:504. 8 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. 9Hoover, W. H., C. R. Kincaid, G. A. Varga, w. V. Thayre, and L. L. 1unkins. Jr. 1984. Effects of solids and liqaid flows on fermentation in continuous cultures. IV. pH and dilution rate. 1. Anim. Sci. 58:692. 101ung, H. G., and V. H. Varel. 1988. Influence of forage type on ruminal bacterial populations and subsequent in vitro fiber digestion. J. Dairy Sci. 71:1526. 11 Mertens, D. R., and J. R. Loften. 1980. 1be effect of starch on forage fiber digestion kinetics in vitro. J. Dairy Sci. 63:1437. 10urnal of Dairy Science Vol. 75, No.4, 1992
1068
GRANT AND WEIDNER
12 Miller, B. G., and R B. Muntifering. 1985. Effect of forage:concentrate on kinetics of forage fiber digestion in vivo. 1. Dairy Sci. 68:40. 13 Mould, F. L., E. R. 0:rsk0v, and S. O. Mann. 1984. Associative effects of mixed feeds. I. Effects of type and level of supplementation and the influence of the rumen fluid pH on ceUulolysis in vivo and dry matter digestion of various roughages. Anim. Feed Sci. Technol. 10:15. 140rs!l:ov, E. R, and C. Fraser. 1975. The effects of processing of barley-based supplements on rumen pH, rate of digestion, and voluntary intake of dried grass in sheep. Br. J. Nutr. 34:493. 15 Robinson, P. H., S. Tamminga, and A. M Van Vuuren. 1986. Influence of declining level of feed intake and varying the proportion of starch in the concentrate on rumen fermentation in dairy cows. Livest. Prod. Sci. 15:173. 16 Russell, J. B. 1987. Effect of extracellular pH on growth and proton motive force of Bacteroides suecinogenes, a cellulolytic rumiDal bacterium. Appl. Enviton. Microbiol. 53:2379. 17 SAS~ User's Guide: Statistics, Version 5 Edition.
Journal of Dairy Science Vol. 75, No.4, 1992
1985. SAS lost, Inc., Cary, NC. 18 ShriVel', B. 1., W. H. Hoover, J. P. Sargent, R J. Crawford, Jr., and W. V. Thayne. 1986. Fel'mentation of a high concentrate diet as affected by ruminal pH and digesta flow. J. Dairy Sci. 69:413. 19 Slyter, L. L. 1986. Ability of pH-selected mixed ruminaI microbial populations to digest fiber at various pHs. Appl. Environ. Microbiol. 52:390. 20 Slyter, L. L., M. P. Bryant, and M J. Wolin. 1966. Effect of pH on population and fermentation in a continuously cultured rumen ecosystem. Appl. MicrobioI. 14:573. 21 Slyter, L. L., R. R Oltjen, D. L. Kern, and F. C. Blank. 1970. Influence of type and level of grain and diethylstilbestrol on the rumen microbial populations of steers fed aU-concentrate diets. J. Anim. Sci. 31: 996. 22 Stewart, C. S. 1977. Factors affecting the cellulolytic activity of rumen contents. Appl. Environ. Microbiol. 33:497. 23 Terry, R A., J.MA. Tilley, and G. E. Outen. 1969. Effect of pH OD cellulose digestion under in vitro conditions. J. Sci. Food Agric. 20:317.
Abbreviation key: AH = alfalfa hay, DH = bromegrass hay, CS com silage, IR indigestible residue, Kd fractional rate constant of digestion.
ABSTRACT
= =
In vitro buffer pH reflective of the diurnal variation in ruminal pH was evaluated for its impact on digestion kinetics of NDF from three forage sources. Alfalfa hay, bromegrass hay, and com silage were incubated in phosphatebicarbonate buffer solution adjusted to pH 6.8, 6.5, 6.2, 6.0, 5.8, or 5.5 using 1 M citric acid. Ash-free NDF was measured after 0,6, 12, 18, 24, 48, 72, and 96 h of fennentation. The experiment was replicated three times; kinetic parameters of fiber digestion were estimated by logarithmic transformation and by linear and nonlinear regression procedures. Lag in NDF digestion increased as pH fell from 6.8 to 6.5 and again when pH decreased below 6.0. Decreasing buffer pH below 6.2 dramatically reduced NDF digestion rate for alfalfa hay and com silage but had no significant effect on NDF digestion rate of bromegrass hay. Lag in NDF digestion increased below pH 6.0 for both alfalfa and com silage but increased only when pH fell below 6.2 for bromegrass. Results suggest that lowered pH exerts its negative effect on NDF digestion between pH 6.2 and 5.8, as evidenced by increased lag and decreased rate, and that critical pH and specific, affected component of digestion varied among forages. (Key words: digestion kinetics, forage fiber, pH, in vitro buffer)
INTRODUCTION
Received September 13, 1991. Accepted November 25, 1991. lPublished willi the approval of the director as Paper Number 9624, 10urnal Series, Nebraska Agricultural Research Division. 2R eprint requests. 1992 1 Dairy Sci 75:1060-1068
=
Ruminal pH, which is depressed as a result of feeding large amounts of readily available carbohydrates, may reduce fiber digestion (2, 14). Stewart (22) demonstrated that fiber digestion was inhibited when pH dropped from 6.9 to 6.0, and cellulolytic populations fell simultaneously from 1()6 to 103/00. Mould et al. (13) postulated two factors that accounted for the depression in fiber digestion when sheep were fed high starch diets. The digestibility depression that was alleviated with bicarbon~ bufferng of ruminal conterns was termed "pH effect"; the residual depression that peIhaps was due to microbial adaptations to starch was termed "carbohydrate effect". In their experiments, pH effect appeared to be three times larger than carbohydrate effect. Low ruminal pH is likely to reduce fiber digestion in high producing dairy cows fed diets containing 50 to 60% concentrates. Robinson et al. (15) fed dairy cows a diet containing 32% starch in the concentrate at 24 kg/d and found average ruminal pH to be 6.1. The diurnal pattern for ruminal pH indicated that it was below 6.2 for 70 to 80% of each day (15). If the ruminal pH of lactating dairy cows is below 6.0 for extended periods, and if low pH is postulated to have a substantial negative impact on fiber digestion, determining the mechanism by which pH alters kinetics of fiber digestion is important. Grant and Mertens (5) showed that decreasing in vitro buffer pH from 6.8 to 5.8 increased digestion lag and decreased rate of NDF digestion. Using continuous culture, Hoover et al. (9) found pH 6.5 to be optimal for fiber and OM digestion, whereas pH of 5.5 or 7.5 signif-
1060
IN VTIRO BUFFER pH AND FIBER DIGESTION
icantly decreased fiber digestion. Data suggest a fairly narrow pH range for optimal fiber digestion in the romen, but threshold pH, below which lag increases and rate decreases, was not defined. Grant and Mertens (6) developed an in vitro buffer system capable of controlling pH between 5.5 and 6.8 throughout 96 h of batch fennentation. Using this system, the specific effect of pH on fiber digestion kinetics can be determined for forage substrates commonly used in dairy cow diets. JlDlg and Varel (10) showed clearly that forage source had a significant impact upon ruminal microbial populations and subsequent in vitro fibrolytic activity of ruminal inoculum. The nature of the forage by pH interaction on fiber digestion kinetics must be determined lDlder carefully controlled, in vitro conditions to lDlderstand how fiber digestion proceeds in vivo when cows consume diets that lower ruminal pH. The objective of this research was to determine the impact of buffer pH on in vitro NDF digestion kinetics of common forage substrates when pH was varied within the diurnal range observed for lactating dairy cows fed diets that depress ruminal pH. MATERIALS AND METHODS
sample Preparation
Forages were dried at 60°C for 48 h and ground through a I-mm screen using a Wiley mill (Arthur H. Thomas Co., Philadelphia, PA). Early bloom, first-cutting alfalfa hay (AH) contained 19% CP and 48% NDF (ashfree, DM basis); mature, first-cutting bromegrass hay (BH) contained 11.5% CP and 68% NDF; and corn silage (CS) at physiological maturity contained 8.0% CP and 54% NDF. A 300-mg sample of each of the three substrates was weighed into 5o-ml polypropylene tubes. In Vitro Procedure
The in vitro system consisted of 5o-ml polyproplyene tubes with gas-release rubber stoppers. The buffer solution was that described by Goering and Van Soest (3) adjusted to pH 6.8, 6.5, 6.2, 6.0, 5.8, or 5.5 with 1 M citric acid as described by Grant and Mertens (6). This buffer system maintained a stable pH throughout 96 h of fermentation with
1061
forage substrates (5, 6). Citrate, in place of distilled water, was added to C02-Saturated buffer solutions at 39°C until the desired pH was obtained. Use of 1 M citric acid to adjust pH of the buffer solution had no effect on NDF digestion in vitro for forage or for forage plus starch substrates (6). Thus, even though citrate can be fermented by ruminal bacteria, use of this acid to adjust pH had no demonstrable confounding effect on our ability to measure NDF digestion kinetics in vitro. Given its tricarboxylic nature, citric acid buffered more strongly within the pH range 5.5 to 6.8 than did phosphoric acid (6, 23). All buffer solutions were reduced (3), warmed to 39°C, and saturated with C~ prior to dispensing. The ruminal fluid inoculum was obtained from a steer fed medium quality AH. At the time of collection, the pH of the ruminal fluid was measured and averaged 6.35 ± .21 for all replicates. Ruminal inoculum of steers fed AH has been shown to degrade fiber fractions of forages more rapidly than inoculum from steers fed BH or switchgrass, as judged by 48-h in vitro cell-wall digestibility (10). Furthermore, inoculum of ruminal microbes selected at pH 5.5 did not digest fiber any better in batch culture at pH 5.0 or 5.5 than inoculum from bacteria selected for pH 6.5 (19). Therefore, inoculum pH of 6.35 should not have had a major impact on fiber digestion kinetics within the pH range in our study. While purging with CO2, a 20% solution of ruminal fluid in buffer of appropriate pH was dispensed in 3o-ml aliquots per tube using a Unispense IT automatic dispensing machine (Wheaton Instruments, Millville, NJ). Tubes were sealed immediately with gas-release stoppers, gently swirled, and allowed to ferment for the appropriate time. The pH of the contents of the tube was measured at each time for each substrate to monitor stability of buffer pH, which did not decrease more than .10 pH unit for any substrate during 96 h of fermentation. Tubes containing AH or BH experienced a decline in pH of no more than .07 unit, whereas pH of CS tubes declined a maximum of .10 unit, presumably because of the presence of rapidly fermented starch. Thus, actual pH for each treatment remained separated sufficiently for the results to be biologically important. Preliminary experiments were to compare this in vitro system with the optimal system Journal of Daily Science Vol. 75, No.4, 1992
1062
GRANT AND WEIDNER
described by Grant and Me~ns (4). Those researchers determined that a system using continuous C~ gassing and reducing agents provided the greatest opportunity to detect differences in substrates related to their intrinsic properties. This system used sodium sulfide nonahydrate and cysteine hydrochloride as reducing agents and used tryptone and microminerals as described by Goering and Van Soest (3) as nutritive additives. In vitro systems that do not maximize digestion kinetics may not detect differences in substrate that are due to constraints of the system. Preliminary fiber digestion rates (per hour) at pH 6.8 for our system versus the one described by Grant and Mertens (4, 6) were the following: for AR, .091, .070; for BH, .058, .051; and for CS, .032, .035. Digestion kinetics determined using this system compared favorably with that obtained using the optimal system described. Fermentation times were 0, 6, 12, 18, 24, 48, 72, and 96 h. Ash-free NDF was measured (3). Modification included use of heat-stable amylase (Termamyl 120L, Novo Laboratories, Inc., Danbury, at filtering for CS. The in vitro experiment was replicated three times.
en
Statistical Analysis
The model for kinetics of NDF digestion was that described by Mertens and Loften (11).
where Y
=
Do Kd L I
= = = =
NDF residue at time t of in vitro fermentation, digestible NDF fraction, fractional rate constant of digestion, discrete lag, and indigestible NDF fraction.
The potentially digestible fraction at each fermentation time was transformed logarithmically, and linear regression was used to estimate digestion parameters. These estimates were used as starting points for iteration by the nonlinear regression procedure of SAS (17). Parameters of the digestion model estimated by nonlinear regression were analyzed using the general linear models procedure of SAS (17); the model included factors for replicate, forage type, pH level, and interactions. Means for which a significant main effect was detected were compared using Student-NewmanKeuls test (17). Regression analysis (17) was used to examine the relationship between buffer pH and fiber digestion kinetics. Unless otherwise stated, statistical significance was at P < .05.
TABLE 1. Comparison of in vitro kinetics of NDF digestion averaged over all pH conditions studied and at pH 6.8 only. ,
Item pH 5.5 to 6.8 Alfalfa bay Bromegrass bay Com silage SE pH 6.83 Alfalfa bay Bromegrass bay Com silage SE
Lag
Estimates of digestion kinetics IR. l Rate (Xci>
(h)
(JIl)
4.55 b 5.4Ob
.043 .049 .045 .004
22.63b 19.(j(f 27.61.97
55.70b 70.7252.59b 2.17
.070-
21.90 19.34 25.19 .85
57.10 70.40 62.05 1.90
6.8~
.77 1.79b 2.6Q&b 3.47-
.25
('Ai)
.051 b .062ab
.005
a,b·"Means within a colmnn with different superscripts differ (P
mn2
< .05).
lIR. = Indigestible NDF residue expressed as a percentage of DM in original sample. 2PED = Potential extent of NDF digestion expressed as a pen:entage of NDF in original sample. 3mcludes three replicates of experima1t as descnDed in Materials lIDd Methods plus one additional replicate of alfalfa bay incubated at pH 6.8.
Journal of Dairy Science Vol. 75. No.4. 1992
1063
IN VI1RO BUFFER pH AND FIBER DIGESTION RESULTS
When averaged over all pH conditions evaluated in this study, CS had a significantly longer lag and higher indigestible residue (IR) than All or BH (fable 1). Bromegrass hay had the least IR. Forage type had no effect on Kd (fable 1). These results are atypical compared with those usually reported for fiber digestion of these forages (11). The longer lags and equivalent rates of fiber digestion reflect inclusion of digestion data for pH 5.5, 5.8, and 6.0. To demonstrate that the in vitro system used in this study actually generated nonnal fiber digestion kinetics at pH 6.8, Table 1 illustrates the effect of forage source upon kinetics of NDF digestion at pH 6.8 only. Alfalfa hay had a significantly shorter lag and faster rate of NDF digestion than BH and CS. Indigestible residue and potential extent of digestion did not differ among forages. These values of NDF digestion at pH 6.8 agree with previously reported kinetic data for these forages (4, 5, 6). We concluded that the in vitro system used in our study provided ample opportunity for detecting differences in forage substrates related to their structural properties and for quantifying the interaction between forage physicochemical properties and pH. Averaged over substrates, lag increased as pH declined (Figure 1). Lag increased significantly as pH fell from 6.8 to 6.5 and again when pH declined from 6.0 to 5.8. Buffer pH had no effect on lag between pH 5.5 and 5.8 or on lag in the range of 6.0 to 6.5. Rate of NDF digestion decreased significantly below pH 6.2 when averaged over all forages (Figure 2). No difference in Kd was significant between pH 6.2 and 6.8. Rate of NDF digestion was not affected by pH below 6.0, although Kd still tended to decline (P < .10). No differences in IR were significant because of pH level (Figure 3). Interestingly, a noticeable but nonsignificant increase in IR occurred between pH 6.2 and 6.0, which paralleled the changes ~ served in Kd (Figure 2). A significant pH by forage interaction was found (P < .012). Table 2 illustrates the response in kinetics of NDF digestion to pH for All, BH, and CS. Lag did not increase significantly for All or CS until pH dropped below 6.0. In contrast, lag increased for BH below pH 6.2. Com silage showed the most pronounced effect of low pH on lag. Rate of NDF digestion for BH showed no response to pH, although Kd decreased nonsignificantly as pH
declined. For All and CS, Kd decreased when pH dropped below 6.2. The lower than expected Kd for All at pH 6.5 was due to a low value for one replicate of the experiment. For all forages, IR increased with decreasing pH, although differences were not significant. To facilitate comparison with reported values, p0tential extent of digestion was calculated. Extent of NDF digestion declined nonsignificantly as pH fell. Table 3 illustrates the potential effect of altering pH on apparent extent of forage fiber digestion. Using equations in Miller and Muntifering (12) and assuming a rate of fiber passage of .OS/h, the apparent extent of digestion for All and CS was lower than for BH as pH declined. The potential apparent extent of digestion decreased below pH 6.2 for all forages. DISCUSSION
The ability of ruminal fibrolytic bacteria to digest plant cell wall is a function of bacterial species, forage species, and the mminal environment in which digestion occurs. Research
10
a
8
--
6
~
4
2
5.5
5.8
6.0
6.2
6.5
6.8
pH Figure 1. Effect of in vitro buffer pH on lag of NDF digestion. Letters (a,b,c) denote significant differences (P
< .05). Journal of Dairy Science Vol. 75. No.4. 1992
1064
GRANT AND WEIDNER 25
.06
.05
--
24
'#.
-
G)
.04
~
'C
.c
~ G)
10 a:
23
'0 G) 'G)
:a
.03
i
22
G)
0)
=e
.02
.5 21
.01 20
5.5
5.5
5.8
6.0
6.2
6.5
6.8
pH
5.8
6.0
6.2
6.5
6.8
pH Figom 3. Effect of in vitro buffer pH on NDF indigestible residue.
Figure 2. Effect of in vitro buffer pH on digestion rate of NDF (KII>. Letters (a,b) denote significant differences (P < .05).
has focused on pH as a primary factor governing fiber digestion (20). Slyter (19) was unable to select for aciduric fibrolytic bacteria after 3 wk in a pH 5.5 continuous fennenter, and he concluded that this sort of microbial adaptation to low pH was unlikely to occur in the rumen as well. Thus, consideration of diurnal fluctuations in ruminal pH between pH 5.5 and 6.8 is critical. Delineation of threshold pH for various forage species with respect to kinetics of NDF digestion was a primary objective of our study. We examined AB, a representative legume, BH, a C3 grass, and CS, a grass with considerable inherent starch to impact NDF digestion potentially. These three forages represent a wide range in plant cell-wall characteristics, are commonly fed to dairy cows, and should therefore be well-suited substrates for quantifying the interaction between in vitro pH and forage NDF digestion kinetics. Detennination of differences in the in vitro digestion Joumal of Dairy Science Vol. 75, No.4, 1992
kinetics of forage fiber types within a physiological range of pH would supply rate and lag infonnation essential for modeling the impact of diurnal pH fluctuations on total ruminal fiber digestion. The existence of a significant pH by forage interaction in our study complicates the interpretation of the main effect means for lag and Kd. Examination of response means, however, allows general conclusions to be drawn concerning pH threshold levels for lag and Kd' In our study, as pH declined from 6.8 to 5.5, lag in NDF digestion increased, and Kd decreased. Although differences in fiber digestion were significant among forages and were most greatly affected by low pH, NDF digestion suffered severely below pH 6.2 to 6.0 for all forages (Table 2; Figures 1 to 3). Using inoculurn from a pH 5.5 fennenter, Slyter (19) found 13% NDF digestion during 24-h periods at pH 5.5, a 45% NDF digestion
1065
IN VITRO BUFFER pH AND FIBER DIGESTION TABLE 2. Effect of forage type and pH on in vitro digestion kinetics of NDF.t Lag
Rate (Kd)
(h)
(/h)
---(%)---
.05~ .045 ab .055a .000b .024b .036b .003 .051 .050 .053 .053 .045 .039 .001 .062a .063a .065 a .029b .031 b .019b
21.94 22.14 21.90 21.99 23.66 24.12 .29 19.34 18.44 18.46 20.33 20.30 20.73 .44
56.49 56.95 56.53 57.59 51.71 54.93 .52
25.19 26.92 26.14 29.21 28.54 29.64 .67
60.56 59.24 54.56 46.15 53.92 41.10 1.69
Forage
pH
Alfalfa hay
6.8 6.5 6.2 6.0 5.8 5.5
1.79b 4.00b 1.92b 3.89b 7.58a 8.10& .38
6.8 6.5 6.2 6.0 5.8 5.5
2.60b 3.47b 4.47b 6.82 a 6.89a 8.12a .26 3.47b 6.32 b 5.90b
SE Bromegrass hay
SE Com silage
6.8 6.5 6.2 6.0 5.8 5.5
4.nb
9.5~ 11.07a
.28
SE
.002
70.75 73.25 71.95 69.83 68.44 70.08 .41
a,"Means within a column and substrate combination differ significantly (P < .05). lAccuracy of fitting data to the model for each pH level as indicated by R 2 was pH 6.8, .945; pH 6.5, .972; pH 6.2, .989; pH 6.0, .986; pH 5.8, .865; pH 5.5, .905. 2IR = Indigestible NDF residue expressed as a percentage of DM in original sample. Potential extent of NDF digestion as a percentage of NDF in original sample. 3pED
=
during 24-h periods at pH 6.5, but only 1% digestion at pH 5.0. In our study, kinetics of fiber digestion were influenced most profoundly for all forages between pH 6.2 and 5.8. At pH 5.5, Kd had become minimal, whereas lag in digestion was approaching 10 h. A pH of 5.5 may be the lower practica1lirnit for fiber digestion in the rumen. Below pH 5.5, little fiber digestion occurs, even if bacteria remain alive (8, 19). Fortunately, diurnal variations in pH are such that it rarely falls below 5.5 for very long (15). Given the linear increase in lag below pH 6.0 observed in our study (R2 = .98) and the linear decrease in rate below pH 6.2 (R2 = .97), little fiber digestion evidently occurs below pH 5.5. These observations are consistent with severely decreased numbers of cellulolytic bacteria in steers fed all concentrate with nnninal pH below 5.3 (21).
Mould et al. (13) described the pH effect on fiber digestion as biphasic: pH reduction from 6.8 to 6.0 resulted in moderate depressions in
fiber digestion, and pH reduction below 6.0 resulted in severe inhibition. Although mechanisms involved at low pH (below 6.0) may concern transmembrane proton flux (16), attachment phenomena may be associated with moderate declines in pH from 6.8 to 6.0. Activity of isolated fibrolytic enzymes remains high in this pH range (7). Also, significant decreases in numbers of cellulolytic bacteria are not associated with small declines in pH within the range of pH 6.0 to 6.8 (19). Our data showed significant increases in lag as pH fell from 6.8 to 6.5. As buffer pH fell from 6.8 to 6.5, lag increased by 2 h, averaged over all forages. Between pH 6.5 and 6.0, lag was constant, averaging 4.6 h. As pH fell below 6.0, however, lag increased markedly to over 8 h (Figure 1). For individual forages, the increases in lag between pH 6.8 and 6.5 were 2.2 h (AH), .9 h (BH), and 2.9 h (CS). Low pH has been associated with prevention of tight attachment of bacteria to plant cell walls (1). Shriver et al. (18), using continuous culture, Journal of Dairy Science Vol. 75, No.4, 1992
1066
GRANT AND WEIDNER
TABLE 3. PoteDtial effect of pH on apparent extent of forage fiber digestion as estimated by in vitto fermentation. l pH Item
6.8
6.5
Alfalfa hay (e-/Cl-)2
.81S& .47oab
K'y(Kd + K~3
e-Kl-- x K'y(Kd + K~4
Bromegrass hay e-/C-}-
+ K~
e-K';- x KdI(Kd + K~
ABD,%
20.Sb
.S7~
.840"
.796&
.5OS
.495
.51S
.~
5.S
.360b
.38Sb
31.1&
6.0
.823& .442b
n.r:/'
AED,S %
Kd/(Kd
6.2
.42oa 30.7&
.4Q9& 29.0.740'.560"
5.5
.682b 32r:/' .22rf
llAc
.708b .476
.340"
23.1 b
SE
.667b .418b
.031 .019
.27~
.020 1.2
.670b .448 .300b
.020 .006
lS.3c
21.0b
.014
1.1
Com silage (e-K';-)
.841&
.725&
Kd/(Kd + K~
.444&
.5S6&
.374& 22.6&
.401&
-KL
e
P
x Kd/(Kd +
ABO, %
K~
23.S&
.414& 22.6&
.788&
.360b .286 13.1b
b
.616b .383b .23Sb
12.l b
.572b .273b .15Sb 6.4c
.026 .028
.027 1.9
a,b,CMeans within a row with different superscripts differ (P < .OS). lCalculations assume a fiber passage rate (K~ of .OS/h. 2Praction of potentially digestible fiber in rumen when digestion begins at end of lag (12). 3orheoretical maximum proportion of fiber disappearance from the rumen that is due to digestion when lag = 0 (12). 4praction of potentially digestible fiber aetua1Iy digested in rumen (12). SAED = Apparent extent of digestion = (potential extent of digestion) x K'y(Kd +
showed that 65% concentrate diets at pH 5.8, 6.2, 6.6, and 7.0 had 43% fewer microbes attached to fiber particles as pH dropped from 6.2 to 5.8. The significant increase in lag as pH dropped from 6.8 to 6.5 in our study may reflect these altered attachment mechanisms. The nature of the interaction between low pH and attachment to fiber particles is unknown, but we hypothesize that this interaction is at least partially responsible for effects of pH on lag that we observed, especially at moderate pH reductions (pH 6.8 to 6.5; Figure 1). Table 2 illustrates the nature of the pH by forage interaction examined in our study. Alfalfa hay and CS clearly differed from BH, the C3 grass forage, in lag and rate of NDF digestion. The pH at which lag increased significantly was .2 pH units lower in our in vitro system for AH and CS than for BH. As the fiber percentage of the substrate increased, the importance of the low pH effect increased relative to the carbohydrate effect discussed by Mould et al. (13). We compared the two substrates in our study in which NDF came from Jonmal of Dairy Science VoL 7S, No.4, 1992
K~
x e-/CpL (12).
forage only (AH, BH). For BH (68% NDF), lag increased significantly between pH 6.2 and 6.0; for AH (48% NDF), lag increased between pH 6.0 and 5.8. There was no significant pH effect upon Kd for BH. The IR increased nonsignificantly below pH 6.2 for BH, but only below pH 6.0 for AH. Therefore, that initial digestion of the higher fiber forage (BH) as measured by lag apparently was more quickly reduced with lower pH. The component of NDF digestion most acutely affected was lag (Table 2). Even more striking, Kd for BH was not significantly affected by decreasing pH. The difference in Kd between pH 6.8 and 5.5 for BH was .012Jh; for AH, .023Jh; and for CS, .043Jh (Table 2). Similar results for CS NDF, showing a combined negative effect of low pH and starch, were reported by Grant and Mertens (5). These dramatic differences among forage fiber sources in NDF digestion kinetics, if found also in vivo, may suggest feeding strategies that effectively mesh fiber availability for microbial degradation with diurnal pH fluctuation.
IN VI1RO BUFFER pH AND FIBER DIGESTION
Terry et al (23) found that the extent of cellulose digestion by microorganisms in vitro (at 96 h) was dependent on media pH. Immature alfalfa had 48-h cellulose digestions of 6, 15, and 15% at pH 5.5, 6.0, and 6.8. At 96 h of in vitro incubation, 6,16, and 15% of cellulose were digested. We observed nonsignificant decreases in potential extent of digestion as pH declined (fable 2). In contrast with much of the reported research, Miller and Muntifering (12) found a decrease in fescue fiber digestion with increasing grain levels in situ, which was related primarily to decreased potential extent of digestion. Lag and rate effects were not significant contributors to their observed decline in fiber digestibility. Mertens and Loften (11) studied similar levels of starch addition with fescue hay in vitro and found a major negative effect on lag. Although our in vitro results showed a nonsignificant increase in IR with pH depressions of 6.2 to 6.0, pH effects on lag and rate were much more significant. Mertens and Loften (11) were unable to elicit large effects of starch addition on fiber Kd in an in vitro system buffered at pH 6.8 only. However, in our study batch culture and buffers within the pH range of 5.5 to 6.8 showed dramatic pH effects on rate and lag, perhaps unique to forage type. Why kinetic analysis of starch effects on fiber digestion, as related to extracellular pH, yields such different results in situ versus in vitro is unknown. Perhaps this difference is due to the fluctuating ruminal pH inherent in the in situ digestion technique compared with the stable pH maintained throughout 96 h of in vitro batch fermentation. Although use of Kd derived from in vitro systems to estimate apparent extent of digestion in vivo necessarily restricts inferences drawn, there is a tremendous potential range possible (Table 3) in ruminal fiber digestion for various forage types under conditions in which pH varies substantially. CONCLUSIONS
Decreasing pH moderately increased fiber digestion lag between pH 6.8 and 6.5 and more severely below pH 6.0. Averaged over all substrates, decreasing pH below 6.2 reduced Kd significantly. However, the effect of low pH on fiber digestion kinetics appeared to be a function of forage fiber source. In particular,
1067
low pH had no significant effect on Kd for BH compared with AH or CS. Common feeding practices result in low ruminal pH in lactating dairy cows. Thus, kinetic data reported here should suggest in vivo studies to help devise feeding strategies that recognize the multifaceted nature of the interaction between low pH and kinetics of forage fiber digestion for specific fiber sources. ACKNOWLEDGMENTS
The authors wish to thank Mary Rieger for her assistance in completing much of the laboratory analysis. REFERENCES 1 Cheng, K. 1., C. S. Stewart. D. Dinsdale, and J. W. Costerton. 1984. Electron microscopy of bacteria involved in the digestion of plant cell walls. Anim. Feed Sci. Teclmol. 10:93. 2 EI-Sbazly, K .• B. A. Dehority, and R R. Johnson. 1961. Effect of starch on the digestion of cellulose in vitro and in vivo by rumen microorganisms. 1. Anim. Sci. 20:268. 3 Goering, H. K., and P. J. Van Soest. 1970. Forage fiber analyses (apparatus, reagents, procedures, and some applications). Agric. Handbook No. 379. ARSUSDA, Washington. DC. 4 Grant, R. 1., and D. R. Mertens. 1990. Effect of in vitro fermentation technique upon kinetics of neutral detergmt fiber digestion. 1. Dairy Sci. 73(Suppl. 1): 129.(Abstr.) S Grant, R 1., and D. R Mertens. 1990. Impact of forage type, corn starch addition and buffer pH upon in vitro digestion kinetics of neutral detergent fiber. 1. Anim. Sci. 68(Suppl. 1):S89.(Abstr.) 6 Grant, R. 1., and D. R. Mertens. 1990. A buffer system for pH control and evaluation of pH effects upon fiber digestion in vitro. Page 109 in US Dairy Forage ReselttCh Center 1989 research snmmaries. ARS-USDA, Madison, WI. 7 Groleau, D., and C. W. Forsberg. 1983. Partial characteri2ation of !he extraee11u1ar carboxymethylcellulase activity produced by the rumen bacterium Bacteroides sucdnogenes. Can. 1. Microbiol. 29:504. 8 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. 9Hoover, W. H., C. R. Kincaid, G. A. Varga, w. V. Thayre, and L. L. 1unkins. Jr. 1984. Effects of solids and liqaid flows on fermentation in continuous cultures. IV. pH and dilution rate. 1. Anim. Sci. 58:692. 101ung, H. G., and V. H. Varel. 1988. Influence of forage type on ruminal bacterial populations and subsequent in vitro fiber digestion. J. Dairy Sci. 71:1526. 11 Mertens, D. R., and J. R. Loften. 1980. 1be effect of starch on forage fiber digestion kinetics in vitro. J. Dairy Sci. 63:1437. 10urnal of Dairy Science Vol. 75, No.4, 1992
1068
GRANT AND WEIDNER
12 Miller, B. G., and R B. Muntifering. 1985. Effect of forage:concentrate on kinetics of forage fiber digestion in vivo. 1. Dairy Sci. 68:40. 13 Mould, F. L., E. R. 0:rsk0v, and S. O. Mann. 1984. Associative effects of mixed feeds. I. Effects of type and level of supplementation and the influence of the rumen fluid pH on ceUulolysis in vivo and dry matter digestion of various roughages. Anim. Feed Sci. Technol. 10:15. 140rs!l:ov, E. R, and C. Fraser. 1975. The effects of processing of barley-based supplements on rumen pH, rate of digestion, and voluntary intake of dried grass in sheep. Br. J. Nutr. 34:493. 15 Robinson, P. H., S. Tamminga, and A. M Van Vuuren. 1986. Influence of declining level of feed intake and varying the proportion of starch in the concentrate on rumen fermentation in dairy cows. Livest. Prod. Sci. 15:173. 16 Russell, J. B. 1987. Effect of extracellular pH on growth and proton motive force of Bacteroides suecinogenes, a cellulolytic rumiDal bacterium. Appl. Enviton. Microbiol. 53:2379. 17 SAS~ User's Guide: Statistics, Version 5 Edition.
Journal of Dairy Science Vol. 75, No.4, 1992
1985. SAS lost, Inc., Cary, NC. 18 ShriVel', B. 1., W. H. Hoover, J. P. Sargent, R J. Crawford, Jr., and W. V. Thayne. 1986. Fel'mentation of a high concentrate diet as affected by ruminal pH and digesta flow. J. Dairy Sci. 69:413. 19 Slyter, L. L. 1986. Ability of pH-selected mixed ruminaI microbial populations to digest fiber at various pHs. Appl. Environ. Microbiol. 52:390. 20 Slyter, L. L., M. P. Bryant, and M J. Wolin. 1966. Effect of pH on population and fermentation in a continuously cultured rumen ecosystem. Appl. MicrobioI. 14:573. 21 Slyter, L. L., R. R Oltjen, D. L. Kern, and F. C. Blank. 1970. Influence of type and level of grain and diethylstilbestrol on the rumen microbial populations of steers fed aU-concentrate diets. J. Anim. Sci. 31: 996. 22 Stewart, C. S. 1977. Factors affecting the cellulolytic activity of rumen contents. Appl. Environ. Microbiol. 33:497. 23 Terry, R A., J.MA. Tilley, and G. E. Outen. 1969. Effect of pH OD cellulose digestion under in vitro conditions. J. Sci. Food Agric. 20:317.
