Color profile: Disabled

Composite Default screen

650

NOTE / NOTE

Attempts to inhibit ruminal methanogenesis by

blocking pyruvate oxidative decarboxylation
E.M. Ungerfeld, S.R. Rust, and R. Burnett

Abstract: The inhibition of pyruvate oxidative decarboxylation as a means of decreasing ruminal methanogenesis
in vitro was studied. In the first experiment, the addition of adenosine and adenine (with and without ribose) to
ruminal batch cultures did not decrease methanogenesis. In the second experiment, the addition of oxythiamin de-
creased methanogenesis by 23%. In the third experiment, three pyruvate derivatives did not inhibit methanogenesis,
although hydroxypyruvate improved organic matter fermentation from 57.8% to 64.2%. The additives did not seem to
inhibit pyruvate oxidative decarboxylation.

Key words: methane, inhibition, rumen, pyruvate decarboxylation, thiamin.

Résumé : Nous avons étudié l’inhibition de la décarboxylation oxydative du pyruvate comme moyen de réduire la
méthanogénèse ruminale in vitro. Dans la première expérience, l’ajout d’adénosine et d’adénine, avec ou sans ribose, à
des cultures ruminales en discontinu n’a pas diminué la méthanogénèse. Dans la seconde expérience, l’ajout d’oxythia-
mine a diminué la méthanogénèse de 23 %. Dans la troisième expérience, trois dérivés du pyruvate n’ont pu inhiber la
méthanogénèse, bien que l’hydroxypyruvate a amélioré la fermentation des matières organiques de 57,8 % à 64,2 %.
Les additifs n’ont vraisemblablement pas inhibé la décarboxylation oxydative du pyruvate.
Mots clés : méthane, inhibition, rumen, décarboxylation du pyruvate, thiamine.

[Traduit par la Rédaction] Ungerfeld et al. 654

Methane production by ruminants is a source of carbon and
energy loss and contributes to global warming. The inhibition
of CH4 production in the rumen would have significant eco-
nomic and environmental benefits (Nagaraja et al. 1997).
Pyruvate oxidative decarboxylation is the first step in the
conversion of pyruvate to acetate and to butyrate in the
rumen and produces CO2, H2, and formate, the precursors
for CH4 formation (Russell and Wallace 1997). As thiamin
pyrophosphate is a cofactor of pyruvate oxidoreductases
(Williams et al. 1990), it was hypothesized that the inhibi-
tion of thiamin utilization by ruminal microorganisms could
block pyruvate oxidative decarboxylation, decreasing the
availability of precursors for methanogenesis and diverting
pyruvate into propionate formation.
Experiment 1: inhibition of thiamin synthesis
All the studies herein presented were approved by the
Michigan State University All-University Committee on An-
imal Use and Care (Animal Use Form 03/02-043-00), in ac-
Received 7 April 2003. Revision received 15 September 2003.
Accepted 19 September 2003. Published on the NRC Research
Press Web site at http://cjm.nrc.ca on 28 November 2003.
E.M. Ungerfeld, S.R. Rust,1 and R. Burnett. Department of
Animal Science, Michigan State University, East Lansing, MI
48824, U.S.A.
1
Corresponding author (e-mail: rust@msu.edu).
Can. J. Microbiol. 49: 650–654 (2003) doi: 10.1139/W03-079
cordance with the Guide for the Care and Use of Laboratory
Animals (National Institute of Health 1996).
Adenine and adenosine impaired thiamin synthesis in Esch-
erichia coli (Iwashima et al. 1968) by lowering the hydroxy-
methylpyrimidine moiety synthesis (Kawasaki et al. 1969).
The effects of adenine and adenosine on CH4 production and
on fermentation by ruminal batch cultures were evaluated.
The effects of adenosine’s ribose moiety alone or with ade-
nine were also studied. Ruminal fluid was collected after the
morning feeding from two nonlactating Holstein cows fed
alfalfa hay, strained through a double cheesecloth and
blended. One part of ruminal fluid was mixed with four parts
of Goering and Van Soest (1975) buffer. Three samples of
the mixture were taken for subsequent determination of vol-
atile fatty acids (VFA) initial contents. Fifty millilitres of the
mixture was anaerobically delivered into 125-mL Wheaton
bottles. Each bottle had 600 mg of ground (0.2 mm screen)
alfalfa hay (1.8% nitrogen, dry matter basis) as substrate,
and adenine (Sigma A 8626; Sigma Chemical Co., St. Louis,
Mo.), ribose (Sigma R 7500), adenine + ribose, or adenosine
(Sigma A 9251) was added as a solid to achieve 10 mmol/L
final concentration for each of the compounds. Control bot-
tles had no added chemical. The bottles were sealed under
an O2-free CO2 atmosphere and incubated in a shaking water
bath at 39 °C for 24 h. Fermentation was terminated by in-
jecting 3 mL of 12 mol/L H2SO4 into each bottle.
Gas composition (CH4 and CO2) was analysed as de-
scribed by Callaway and Martin (1996), with a flame ioniza-
© 2003 NRC Canada

I:\cjm\cjm4910\W03-079.vp
November 21, 2003 10:21:51 AM
Color profile: Disabled
Composite Default screen
Ungerfeld et al.
tion detector gas chromatograph (Gow Mac Instruments Co.,
Bridgewater, N.J.) equipped with a 122 cm × 0.6 cm DC
200 column (150 °C, carrier gas was N2 at 820 kPa). Gas
production was expressed as micromoles at 25 °C and 1 atm
(1 atm = 101.325 kPa). A 5-mL aliquot of the fermentation
medium was centrifuged (26 000g, 4 °C, 30 min). Volatile
fatty acids, lactate, formate, and ethanol were quantified by
differential refractometry with a Waters 712 Wisp HPLC
(Waters Associates Inc., Milford, Mass.) equipped with a
BioRad HPX 87H column (BioRad Laboratories, Hercules,
Calif.). The solvent was 0.005 mol/L H2SO4 at 0.6 mL/min.
Column temperature was 65 °C. Sample injection volume
was 15 µL. Ammonia was analysed as described by Chaney
and Marbach (1962).
Apparently fermented organic matter (FOM) was esti-
mated from VFA production (Marty and Demeyer 1973), but
using isobutyrate instead of caproate. Initial VFA contents
were subtracted, and VFA production expressed in micro-
moles. Hydrogen balance were calculated as done by Unger-
feld et al. (2003).
Four replicates per treatment were used. The experimental
model was as follows: response = overall mean + treat-
ment + residual. Data were tested for homogeneity of vari-
ances (Modified Levene’s test, Neter et al. 1995) and
analysed as a one-way ANOVA. If homogeneity of variances
was rejected (P < 0.05), a Kruskal–Wallis rank sum test
(Neter et al. 1995) was conducted. Planned contrasts of in-
terest were (i) control vs. adenine, (ii) control vs. adenosine,
(iii) adenine vs. adenosine, (iv) ribose vs. adenosine, and
(v) adenine + ribose vs. adenosine. A Bonferroni adjustment
(Neter et al. 1995) for five nonorthogonal comparisons was
used, and significance declared at P < 0.01 (for an experi-
ment-wise type I error probability of 0.05). All other com-
parisons were done by the Scheffé test (Neter et al. 1995).
Adenine, with or without ribose, did not affect CH4 pro-
duction (Table 1). The addition of ribose or adenosine alone
tended to increase (P = 0.03 and P = 0.07, respectively,
Scheffé test) methanogenesis. However, when CH4 output
was related to the amount of organic matter apparently fer-
mented, there were no differences among treatments (not
shown).
The addition of ribose, either pure (control vs. ribose, and
adenine vs. adenine + ribose) or as part of the adenosine
molecule (adenine vs. adenosine), always promoted (P <
0.05) FOM (Table 1). On the contrary, adenine was inhibi-
tory for fermentation. The addition of ribose, alone (control
vs. adenine + ribose, P = 0.69, Scheffé test) or as the ribose
moiety in adenosine (control vs. adenosine, P = 0.58), re-
lieved the inhibition caused by adenine. If, as hypothesized,
adenine caused an inhibition of thiamin synthesis, the reac-
tions of the pentose phosphate pathway catalyzed by trans-
ketolase could have been affected. However, this should not
affect the supply of ribose, as ribose phosphate is a substrate
rather than a product of these reactions (Voet and Voet
1995). Therefore, the relief of adenine’s fermentation inhibi-
tion by ribose would have been due to the use of the latter as
an energy source.
The addition of adenine, with (P = 0.02) or without ribose
(P < 0.01), and adenosine (P < 0.01) increased NH +4 concen-
tration; however, NH +4 was higher (P < 0.01) with adenosine
than with adenine (Table 1). McAllan and Smith (1973)
I:\cjm\cjm4910\W03-079.vp
November 21, 2003 10:21:52 AM
651
found that 62% of added adenine (approximately
1.2 mmol/L) was degraded after 4 h of incubation in ruminal
fluid. Hypoxanthine and xanthine accounted for 33% and
7%, respectively, of the adenine initially present. All the
added adenosine (approximately 1.2 mmol/L) was degraded
within 1 h, and 78% of its initial concentration was recov-
ered as inosine, which was in turn converted to hypoxan-
thine. Therefore, and in agreement with the present
observations, deamination seemed to proceed faster when
the ribose moiety was present.
Hydrogen (H) recovery was low in all treatments. This
may be partly accounted for by H sinks that were not mea-
sured, like sulfate and nitrate reductions (Stewart et al. 1997)
and fatty acids biohydrogenation and synthesis (Czerkawski
1986).
Experiment 2: inhibition of thiamin uptake
and utilization
Thiamin structural analogues can inhibit bacterial growth
(Koser 1968) and hinder thiamin uptake in protozoa (Shi-
geoka et al. 1987). In a second experiment, the effects of
thiamin structural analogues amprolium and oxythiamin on
CH4 production and fermentation by ruminal batch cultures
were evaluated. Ruminal fluid was prepared and incubated
as described above, but collected prior to the morning feed-
ing. Each Wheaton bottle had 200 mg of alfalfa hay. One
millilitre of aqueous solutions of amprolium (Sigma A 0542)
or oxythiamin (Sigma O 4000) was added to achieve final
concentrations of 5 or 10 mmol/L. Controls received 1 mL
of deionized water. Fermentation was terminated by inject-
ing 1 mL of a 10% phenol solution instead of H2SO4, as in
the previous experiment, so that the final pH could be mea-
sured (Digital Benchtop pH meter, Cole-Parmer Instrument
Company, Vernon Hills, Ill.). Total gas production was mea-
sured, and gas composition analysed as before. Dihydrogen
was also analysed (RGD2 Reduction Gas Detector, Trace
Analytical, Menlo Park, Calif.). Analysis of VFA and NH +4
and the calculation of FOM were done as above. The num-
ber of replicates and the statistical model and analyses were
the same as in Experiment 1. Orthogonal polynomial con-
trasts for linear and quadratic trends were tested (Neter et al.
1995).
Methane production was decreased (P < 0.05) by 23% and
8% by oxythiamin and amprolium at 5 mmol/L, respectively
(Table 2). Increasing the concentrations to 10 mmol/L
further decreased methanogenesis by only 2% and 3% for
oxythiamin and amprolium, respectively. The effects of am-
prolium on CH4 production were due to less fermentation, as
CH4 production per micromole of FOM did not change (not
shown). Although oxythiamin addition decreased (P < 0.05)
fermentation, CH4 output per micromole of FOM was de-
creased (P < 0.01) by 17% at 5 and 10 mmol/L oxythiamin.
Oxythiamin and amprolium could have inhibited fermenta-
tion by affecting processes other than pyruvate oxidative
decarboxylation, like the pentose phosphate pathway (Voet
and Voet 1995) or valine synthesis (McDonald et al. 1995).
The catabolism of some amino acids could also be affected
by blocking oxidative decarboxylations (Voet and Voet
1995); this may explain the decrease in NH +4 concentration
observed. Oxythiamin decreased (P < 0.01) acetate and bu-
© 2003 NRC Canada
Color profile: Disabled
Composite Default screen

652 Can. J. Microbiol. Vol. 49, 2003

Table 1. Effects of adenine, adenosine, and ribose on end products of ruminal in vitro fermentation (Experiment 1).
Significant
(P < 0.01)
Adenine + planned
Control Adenine Ribose ribose Adenosine SEM P contrastsa
FOM (%) 52.8 46.6 58.8 51.2 53.3 0.72 <0.01 1, 3, 4
CH4 (µmol) 513 454 668 585 594 29.1 <0.01b 3
CO2 (µmol) 2302 2243 2477 2432 2443 36.0 <0.01b 3
Total VFA (mmol/L) 87.3 86.3 103 100 102 0.47 <0.01 2, 3
Acetate (µmol) 2274 2281 2834 2736 2756 35.4 <0.01 2, 3
Propionate (µmol) 706 633 828 818 842 10.5 <0.01 1, 2, 3
Butyrate (µmol) 223 208 320 296 307 8.80 <0.01 2, 3
Acetate (molar %) 69.1 70.3 68.9 68.8 67.9 0.13 <0.01 1, 2, 3, 4
Propionate (molar %) 21.5 19.5 20.1 20.6 20.8 0.16 <0.01 1
Butyrate (molar %) 6.8 6.4 7.8 7.4 7.6 0.08 <0.01 2, 3
NH +4 (mg/dL) 38.9 52.8 33.9 48.7 62.5 1.67 <0.01 1, 2, 3, 4, 5
H produced (µmol) reducing 6457 6524 8098 7944 8240 57.2 <0.01 2, 3
equivalents pairs
H incorporated (µmol) reducing 4084 3717 5220 4807 4964 61.8 <0.01b 2, 3
equivalents pairs
H recovery (%) 63.2 57.0 64.5 60.5 60.3 0.86 0.05 None
Note: FOM, apparently fermented organic matter; VFA, volatile fatty acids.
a
1, control vs. adenine; 2, control vs. adenosine; 3, adenine vs. adenosine; 4, ribose vs. adenosine; 5, adenine + ribose vs. adenosine.
b
The Kruskal–Wallis rank sum test was performed because of heterogeneity of variances (Levene test on absolute deviations P < 0.05).

Table 2. Effect of thiamin structural analogues on ruminal in vitro fermentation (Experiment 2).
Oxy Amp
Control 5 mmol/L 10 mmol/L 5 mmol/L 10 mmol/L SEM P Relationshipa
FOM (%) 56.7 53.2 51.5 51.4 48.2 1.50 0.02 1, 3
CH4 (µmol) 315 244 239 289 279 5.8 <0.01 1, 2, 3
CO2 (µmol) 646 647 643 674 705 30.7 0.58b None
H2 (µmol) 0.43 0.46 0.49 0.60 0.48 0.07 0.58 None
Total VFA (mmol/L) 48.9 47.7 47.1 46.9 45.8 0.28 0.02b 1, 3
Acetate (µmol) 814 713 697 744 724 20.3 <0.01b 1, 3
Propionate (µmol) 298 353 348 280 251 5.3 <0.01 1, 2, 3
Butyrate (µmol) 74.3 58.8 56.6 61.3 55.6 3.66 <0.01 1, 2, 3
Acetate (molar %) 67.8 62.7 62.8 67.9 69.8 0.12 <0.01 1, 2, 3, 4
Propionate (molar %) 24.9 31.0 31.4 25.7 24.3 0.21 <0.01 1, 2, 4
Butyrate (molar %) 6.2 5.2 5.1 5.6 5.4 0.11 0.02 1, 3
pH 7.3 7.2 7.1 7.2 7.2 0.07 0.18 1
NH +4 (mg/dL) 30.7 27.9 24.8 29.3 28.2 0.60 <0.01 1, 3
H produced (µmol) reducing 2511 2305 2214 2303 2199 32.4 0.02 1, 3
equivalents pairs
H incorporated (µmol) reducing 2024 1870 1820 1891 1789 23.6 0.02 1, 3
equivalents pairs
H recovery (%) 80.7 81.1 82.2 82.6 81.3 1.02 0.95 None
Note: Oxy, oxythiamin; Amp, amprolium; FOM, apparently fermented organic matter; VFA, volatile fatty acids.
a
1, significant (P < 0.05) linear relationship for oxythiamin; 2, significant (P < 0.05) quadratic relationship for oxythiamin; 3, significant (P < 0.05) lin-
ear relationship for amprolium; 4, significant (P < 0.05) quadratic relationship for amprolium.
b
The Kruskal–Wallis rank sum test was done because of heterogeneity of variances (Levene test on absolute deviations P < 0.05).

tyrate production and increased (P < 0.01) propionate. This
would be expected if pyruvate oxidative decarboxylation
was inhibited; however, an increase in propionate production
is a general consequence of inhibiting methanogenesis
(Nagaraja et al. 1997). The changes observed in the VFA
profile may result from the inhibition of methanogenesis
rather than the latter been caused by changes in the fermen-
tation pattern. Dihydrogen did not accumulate, and the elec-
trons spared from methanogenesis were relocated into propi-
onate production.
Amprolium increased (P < 0.05) acetate and decreased
butyrate (P < 0.05) molar percentages (Table 2). Horton and
Stockdale (1979) found that amprolium decreased propio-
nate molar percentage without affecting other VFA. Heit-
mann and Yehya Taka (1970–1971) reported a decrease in
propionate and an increase in butyrate when amprolium was

© 2003 NRC Canada

I:\cjm\cjm4910\W03-079.vp
November 21, 2003 10:21:52 AM
Color profile: Disabled
Composite Default screen

Ungerfeld et al. 653

Table 3. Effect of pyruvate derivatives on end products of in vitro ruminal fermentation (Experiment 3).
Significant
(P < 0.0083)
planned
Control Pyruvate Bromopyruvate Fluoropyruvate Hydroxypyruvate SEM P contrastsa
FOM (%) 57.8 58.8 56.5 55.4 64.2 5.29 0.01b 5, 6
CH4 (µmol) 467 457 423 423 453 31.4 0.20 None
CO2 (µmol) 1041 1054 1019 1090 1080 156 0.60b None
H2 (µmol) 0.4 0.4 0.3 0.8 0.4 0.065 <0.01 4, 6
Total VFA (mmol/L) 51.0 52.9 51.4 50.9 55.2 0.25 0.01b 5, 6
Acetate (µmol) 1244 1329 1260 1249 1336 17.7 0.02b 1, 5, 6
Propionate (µmol) 335 350 334 338 382 5.5 <0.01 5, 6
Butyrate (µmol) 136 135 135 127 194 0.27 0.04b 5, 6
Acetate (molar %) 70.0 70.9 70.2 70.6 67.0 0.09 <0.01 1, 2, 3, 5, 6
Propionate (molar %) 18.9 18.7 18.6 19.1 19.2 0.17 0.69 None
Butyrate (molar %) 7.6 7.2 7.5 7.2 9.8 0.14 <0.01 None
pH 6.8 6.8 6.8 6.8 6.7 0.29 0.51 None
NH +4 (mg/dL) 15.1 16.9 16.2 16.1 15.5 0.23 0.10b None
H produced (µmol) reducing 3532 3728 3584 3523 4038 32.5 0.01b 5, 6
equivalents pairs
H incorporated (µmol) 2903 2886 2738 2719 3101 36.3 0.01 5, 6
reducing equivalents pairs
H recovery (%) 82.2 77.4 76.4 77.3 76.8 0.94 0.20 None
Note: FOM, apparently fermented organic matter; VFA, volatile fatty acids.
a
1, control vs. pyruvate; 2, control vs. average pyruvate derivatives; 3, pyruvate vs. average pyruvate derivatives; 4, bromopyruvate vs. Na-
fluoropyruvate; 5, bromopyruvate vs. hydroxypyruvate; 6, Na-fluoropyruvate vs. hydroxypyruvate.
b
The Kruskal–Wallis rank sum test was performed because of heterogeneity of variances (Levene test on absolute deviations P < 0.05).

added to ruminal continuous cultures. Their results and the
present ones would be contrary to the hypothesis that
amprolium would impair thiamin utilization, resulting in
pyruvate being diverted from acetate and butyrate formation
towards propionate production.
Final pH values slightly above 7 (Table 2) would be a
consequence of the low VFA concentration, due to the low
amount of substrate used, and of the loss of dissolved CO2
after the bottles were opened.
Experiment 3: direct inhibition of pyruvate
oxidoreductases
In a third experiment, the direct inhibition of pyruvate
oxidoreductases through the use of pyruvate derivatives, as
shown by Flournoy and Frey (1989) and Williams et al.
(1990) with purified components reactions, was attempted.
The effects of the pyruvate derivatives bromopyruvate, fluo-
ropyruvate, and hydroxypyruvate on CH4 production and
fermentation were evaluated. Ruminal fluid was collected,
prepared, and incubated as in Experiment 2. Wheaton bottles
contained 300 mg of alfalfa hay each. One millilitre of bro-
mopyruvate (Sigma B 9630), sodium fluoropyruvate (Sigma
F 4004), and hydroxypyruvate (Sigma H 9270) aqueous
solutions were added to Wheaton bottles to achieve final
concentrations of 2 mmol/L. Controls received 1 mL of
deionized water. Sodium pyruvate (Sigma P 2256) at
2 mmol/L was also included as a positive control. Fermenta-
tion was terminated by injecting 1 mL of a 10% phenol solu-
tion. Total gas production and composition, VFA production
and NH +4 concentration, and the calculation of FOM were
done as described above. The additives’ final concentrations
were determined by high-pressure liquid chromatography
(HPLC) along with the concentration of VFA. Substrate ap-
parently fermented was calculated by subtracting the amount
of additive disappeared from FOM.
The number of replicates, the model used, and the statisti-
cal analyses were the same as above. Planned contrasts of
interest were (i) control vs. sodium pyruvate, (ii) control vs.
average of pyruvate derivatives, (iii) sodium pyruvate vs.
average pyruvate derivatives, (iv) bromopyruvate vs. sodium
fluoropyruvate, (v) bromopyruvate vs. hydroxypyruvate,
(vi) sodium fluoropyruvate vs. hydroxypyruvate. A Bon-
ferroni adjustment (Neter et al. 1995) for six nonorthogonal
comparisons was used, so significance was declared at P <
0.0083 (for an experiment-wise type I error probability of
0.05). All other comparisons were done by the Scheffé test.
None of the additives had any effect on CH4 production or
CO2 release (Table 3). Disappearance of all four additives
was complete (not shown). As the VFA molar percentages
were the same with bromopyruvate or fluoropyruvate as with
pyruvate (P > 0.05, Scheffé test; not shown), it is possible
that these derivatives were dehalogenated, converted into
pyruvate, and metabolized to VFA. Hydroxypyruvate, how-
ever, shifted (P < 0.05) the VFA molar proportions towards
butyrate at the expense of acetate, which may indicate that
this additive was also catabolized by an alternative pathway.
Interestingly, hydroxypyruvate increased total FOM (sub-
strate + additive) from 57.8% to 64.2% (P < 0.01; Table 3),
and the alfalfa substrate apparent fermentation from 57.8%
to 63.2% (P = 0.02; not shown).
The pyruvate derivatives did not inhibit CH4 production
and, except for hydroxypyruvate, did not alter the VFA pro-
file. This indicates that they probably did not inhibit pyru-

© 2003 NRC Canada

I:\cjm\cjm4910\W03-079.vp
November 21, 2003 10:21:52 AM
Color profile: Disabled
Composite Default screen
654
vate oxidoreductases, as was hypothesized. As they were
totally metabolized, they must have been taken up by ru-
minal microorganisms. They might have been metabolized
before they could inhibit pyruvate oxidoreductases, or the
structures of ruminal pyruvate oxidoreductases could be dif-
ferent from the ones these compounds were previously re-
ported to inhibit.
In conclusion, oxythiamin was the only additive that
achieved some inhibition of methanogenesis, although it also
decreased fermentation. It seems that pyruvate oxidative
decarboxylation could not be inhibited, either directly or
through the inhibition of thiamin utilization. Because of a
lack of direct measurements on thiamin metabolism, it is not
know if the additives in Experiments 1 and 2 were taken up
by ruminal microorganisms and, if they were, why the intra-
cellular effects hypothesized did not occur. An antimethano-
genic strategy based on the use of oxythiamin would require
the delivery of ruminal-protected thiamin together with oxy-
thiamin to avoid potential toxic effects of the latter on the
host animal. Future work in this line would need basic re-
search on thiamin uptake, synthesis, and utilization and on
the enzymology of pyruvate oxidoreductases of different
ruminal microorganisms.
References
Callaway, T.R., and Martin, S.A. 1996. Effects of organic acid and
monensin treatment on in vitro mixed ruminal microorganism
fermentation of cracked corn. J. Anim. Sci. 74: 1982–1989.
Chaney, A.L., and Marbach, E.P. 1962. Modified reagents for de-
termination of urea and ammonia. Clin. Chem. 8: 130–132.
Czerkawski, J.W. 1986. An introduction to rumen studies. Per-
gamon Press Ltd., Oxford, U.K.
Flournoy, D.S., and Frey, P.A. 1989. Inactivation of the pyruvate
dehydrogenase complex of Escherichia coli by fluoropyruvate.
Biochemistry, 28: 9594–9602.
Goering, H.K., and Van Soest, P.J. 1975. Forage fiber analyses (ap-
paratus, reagents, procedures and some applications). Agricul-
ture Handbook No. 379. Agricultural Research Services (ARS),
United States Department of Agriculture (USDA), Washington,
D.C.
Heitmann, R.N., and Yehya Taka, M.R. 1970–1971. An in vitro
study of the effects of amprolium in the rumen. Mesopotamia J.
Agric. 5, 6: 32–40.
Horton, G.M.H., and Stockdale, P.H.G. 1979. Effects of amprolium
and monensin on oocyst discharge, feed utilization, and ruminal
I:\cjm\cjm4910\W03-079.vp
November 21, 2003 10:21:52 AM
Can. J. Microbiol. Vol. 49, 2003
metabolism of lambs with coccidiosis. Am. J. Vet. Res. 40:
967–970.
Iwashima, A., Kawasaki, T., Nakamura, M., and Nose, Y. 1968.
Effect of amino acids and purine bases on thiamin synthesis by
Escherichia coli. J. Vitaminol. (Kyoto), 14: 203–210.
Kawasaki, T., Iwashima, A., and Nose, Y. 1969. Regulation of thia-
min biosynthesis in Escherichia coli. J. Biochem. 65: 407–416.
Koser, S.A. 1968. Vitamin requirements of bacteria and yeasts.
Charles C. Thomas Publishers, Springfield, Ill., U.S.A.
Marty, R.J., and Demeyer, D.I. 1973. The effect of inhibitors of
methane production on fermentation pattern and stoichiometry
in vitro using rumen contents from sheep given molasses. Br. J.
Nutr. 30: 369–376.
McAllan, A.B., and Smith, R.H. 1973. Degradation of nucleic acid
derivatives by rumen bacteria in vitro. Br. J. Nutr. 29: 467–474.
McDonald, P., Edwards, R.A., Greenhalgh, J.F.D., and Morgan,
C.A. 1995. Animal nutrition. Longman Group Ltd., Harlow,
Essex, U.K.
Nagaraja, T.G., Newbold, C.J., Van Nevel, C.J., and Demeyer, D.I.
1997. Manipulation of ruminal fermentation. In The rumen mi-
crobial ecosystem. Edited by P.N. Hobson and C.S. Stewart.
Blackie Academic & Professional, London. p. 523.
National Institute of Health. 1996. Guide for the care and use of
laboratory animals. National Academy Press, Washington, D.C.
Neter, J., Kutner, M.H., Naschsheim, C.J., and Wasserman, W.
1995. Applied linear statistical models. 4th ed. McGraw–Hill,
Boston, Mass., U.S.A.
Russell, J.B., and Wallace, R.J. 1997. Energy-yielding and energy-
consuming reactions. In The rumen microbial ecosystem. Edited
by P.N. Hobson and C.S. Stewart. Blackie Academic & Profes-
sional, London. pp. 246.
Shigeoka, S., Onishi, T., Maeda, K., Nakano, Y., and Kitaoka, S.
1987. Thiamin uptake in Euglena gracilis. Biochem. Biophys.
Acta, 929: 247–252.
Stewart, C.S., Flint, H.J., and Bryant, M.P. 1997. The rumen bacte-
ria. In The rumen microbial ecosystem. Edited by P.N. Hobson
and C.S. Stewart. Blackie Academic & Professional, London.
pp. 10.
Ungerfeld, E.M., Rust, S.R., and Burnett, R. 2003. Use of some
novel alternative electron sinks to inhibit ruminal methano-
genesis. Reprod. Nutr. Dev. 43: 189–202.
Voet, D., and Voet, J.G. 1995. Biochemistry. John Wiley and Sons,
Inc., New York.
Williams, K.P., Leadlay, P.F., and Lowe, P.N. 1990. Inhibition of
pyruvate:ferredoxin oxidoreductase from Trichomonas vaginalis
by pyruvate and its analogues. Comparison with the pyruvate
decarboxylase component of the pyruvate dehydrogenase com-
plex. Biochem. J. 268: 69–75.
© 2003 NRC Canada