Responses of Rumen Microbes to Excess Carbohydrate
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy
in the Graduate School of The Ohio State University
By
Timothy John Hackmann, B.S. & M.S.
Graduate Program in Ohio State University Nutrition Program
The Ohio State University
2013
Dissertation Committee:
Dr. Jeffrey L. Firkins, Adviser
Dr. Daniel R. Bond
Dr. Thaddeus Ezeji
Dr. Gönül Kaletunç
Dr. William P. Weiss
Copyrighted by
TIMOTHY JOHN HACKMANN
2013
Abstract
Some species of rumen microbes respond to excess carbohydrate by synthesizing
reserve carbohydrate, but others respond by spilling energy (producing heat alone). The
response of mixed cultures of species, however, has not been directly studied, and
elucidating the response was the focus of the present research.
The first aim was to evaluate methods for detecting reserve carbohydrate, in an
attempt to find one that was quantitative. Methods compared were those based on the (i)
anthrone reaction and (ii) hydrolysis with amyloglucosidase.
Compared to the
amyloglucosidase hydrolysis method, the anthrone method detected a larger increase (P =
0.017) in cell carbohydrate when glucose (20 mM) was dosed in cultures. Additionally, it
detected a larger decrease (P = 0.049) in cell carbohydrate after glucose was exhausted.
This result indicated that the anthrone method detected more carbohydrate that functions
as a reserve material. For the anthrone method, recoveries for energy (97.5%), carbon
(100.2%), and cell components (99.8%) were high, indicating carbohydrate was
completely detected.
lower.
For the amyloglucosidase hydrolysis method, recoveries were
The anthrone method appeared to accurately quantify changes in reserve
carbohydrate and shows merit for quantitative studies.
The second aim was to determine if a mixed rumen microbes would respond to
excess carbohydrate by accumulating reserve carbohydrate, spilling energy, or both.
Mixed microbes from the rumen were washed with N-free buffer and dosed with glucose.
ii
Total heat production was measured by calorimetry. Energy spilling was calculated as
heat production not accounted by (i) endogenous metabolism and (ii) synthesis of reserve
carbohydrate. For cells dosed with 5 mM glucose, synthesis of reserve carbohydrate and
endogenous metabolism explained nearly all heat production (93.7%); no spilling was
detected (P = 0.226). For cells dosed with 20 mM glucose, energy spilling was not
detected immediately after dosing, but it became significant (P < 0.05) approximately 30
min after dosing glucose. Energy spilling accounted for as much as 38.7% of heat
production in one incubation. As documented for some pure cultures, mixed microbial
communities from the rumen can respond to large excesses of carbohydrate by spilling
energy.
The third aim was to determine how Gibbs energy ( G )—a thermodynamic
property of reactions—was impacted by the specified physical state of gases. The second
experiment (above) required calculation of thermodynamic properties, but it was found
different authors specify different states for gases for these calculations. Our analysis
indicated the aqueous, not gaseous, state is that used by microbes and should be the one
specified in calculations.
Compilation of literature values showed that microbial
reactions create disequilibrium between aqueous and gaseous concentrations of gases,
and the pattern of disequilibrium suggests the aqueous state is used. The greater the
disequilibrium, the greater G was impacted (up to 60.50 kJ/mol) when changing from
the gaseous to the aqueous state. Because G was so impacted, our results suggest that
aqueous gas concentrations must be measured to accurately estimate  r G ' .
iii
By
establishing appropriate calculations for  r G ' , our results should advance understanding
of microbial processes where energetics play a key role.
iv
Acknowledgments
This dissertation is the work of many:

Dr. Jeffrey Firkins, who made me more than just a better scientist

Katherine Backus, Leanne Diese, and Bethany Keyser, who did most of the
real work in the lab

Dr. Gönül Kaletunç and Jeffrey Rachford, who taught me calorimetry

Drs. Thaddeus Ezeji and MacDonald Wick, who provided insight into thinlayer chromatography

Drs. Daniel R. Bond, William P. Weiss, Zhongtang Yu, and Mark
Morrison, who further shaped the vision of this dissertation

Dr. Jo Ann Van Kessel, whose review of preliminary manuscripts vastly
improved them

Reagen Bluel and the farm staff at Waterman Dairy Farm, who maintained
the cows within tight tolerances

Bong-Joo Lee, Stephanie Metzger, Donna Wyatt, and Josie Plank, who
offered technical assistance

Dr. Joseph Hogan, who helped in untold ways

Staff and graduate students of the Department of Animal Sciences, who
adopted me as one of their own

My parents and friends

Shujin (Yang) Hackmann, who remains by my side

Father God, who continues to orchestrate it all
v
Vita
2003........................................................ Lindbergh High School, St. Louis, MO
2006........................................................ B.S. Biological Sciences, University of
Missouri-Columbia, MO
2008........................................................ M.S. Animal Sciences, B.S. Biological Sciences,
University of Missouri-Columbia, MO
2013........................................................ Ph.D., The Ohio State University, Columbus, OH
Publications
1.
Hackmann, T. J. 2011. A system for predicting energy and protein requirements
of wild ruminants. Zoo Biol 30:165-88.
2.
Hackmann, T. J., J. D. Sampson, and J. N. Spain. 2010. Using ytterbium-labeled
forage to investigate particle flow kinetics across sites in the bovine reticulorumen. Anim
Feed Sci Tech 157:1-12.
3.
Hackmann, T. J., and J. N. Spain. 2010. Invited review: ruminant ecology and
evolution: perspectives useful to ruminant livestock research and production. J Dairy Sci
93:1320-34.
vi
4.
Hackmann, T. J., and J. N. Spain. 2010. A mechanistic model for predicting
intake of forage diets by ruminants. J Anim Sci 88:1108-24.
5.
Hackmann, T. J., J. D. Sampson, and J. N. Spain. 2010. Variability in in situ
ruminal degradation parameters causes imprecision in estimated ruminal digestibility. J
Dairy Sci 93:1074-85.
6.
Williams, J., D. Spiers, L. Thompson-Golden, T. Hackman, M. Ellersieck, L.
Wax, D. Colling, J. Corners, and P. Lancaster. 2009. Effects of Tasco in alleviation of
heat stress in beef cattle. Prof Anim Sci 25:109-117.
7.
Hackmann, T. J., J. D. Sampson, and J. N. Spain. 2008. Comparing relative feed
value with degradation parameters of grass and legume forages. J Anim Sci 86:2344-56.
Fields of Study
Major Field: Ohio State University Nutrition Program
vii
Table of Contents
Abstract ............................................................................................................................... ii
Acknowledgments............................................................................................................... v
Vita..................................................................................................................................... vi
List of Tables .................................................................................................................... xii
List of Figures .................................................................................................................. xiv
Chapter 1: Introduction ...................................................................................................... 1
Chapter 2: Literature Review ............................................................................................. 5
Occurrence of energy spilling...................................................................................... 6
Occurrence and identity of reserve carbohydrate ........................................................ 8
Relative magnitude of responses ................................................................................. 9
Other responses to excess carbohydrate ...................................................................... 9
Occurrence of excess carbohydrate in the rumen ...................................................... 10
Function of energy spilling ........................................................................................ 10
Function of reserve carbohydrate .............................................................................. 12
Detection of reserve carbohydrate ............................................................................. 13
viii
Calculation of spilling ............................................................................................... 14
Heat released by formation of fermentation products ............................................... 16
Physical state and thermodynamic calculations ........................................................ 17
Improving microbial growth efficiency and its prediction ........................................ 18
Conclusion ................................................................................................................. 20
Chapter 3: Evaluation of methods to detect changes in reserve carbohydrate for mixed
rumen microbes ................................................................................................................. 34
Abstract ......................................................................................................................... 34
Introduction ................................................................................................................... 35
Materials and Methods .................................................................................................. 37
Preparation of mixed cultures and sampling. ............................................................ 37
Carbohydrate analysis................................................................................................ 38
Additional analyses.................................................................................................... 41
Energy and carbon recovery. ..................................................................................... 42
Statistics. .................................................................................................................... 43
Results ........................................................................................................................... 43
Characterization of reserve carbohydrate ...................................................................... 43
Detecting reserve carbohydrate ................................................................................. 44
Sensitivity of iodine staining ..................................................................................... 45
ix
Discussion ..................................................................................................................... 46
Chapter 4: Quantifying the responses of mixed rumen microbes to excess carbohydrate 61
Abstract ......................................................................................................................... 61
Introduction ................................................................................................................... 62
Materials and Methods .................................................................................................. 64
Chemical analyses ..................................................................................................... 64
Reaction properties .................................................................................................... 65
Measurement of heat production ............................................................................... 66
Heat production and cellular functions ...................................................................... 68
Energy and carbon recovery ...................................................................................... 69
Sensitivity analysis .................................................................................................... 70
Statistics ..................................................................................................................... 71
Results ........................................................................................................................... 71
Discussion ..................................................................................................................... 76
Appendix 1: Thermodynamic symbols and units ......................................................... 81
Chapter 5: Impact of physical state of gases on calculating Gibbs energy for microbial
reactions ............................................................................................................................ 99
Abstract: ........................................................................................................................ 99
Introduction ................................................................................................................. 100
x
Experimental Procedures............................................................................................. 101
Results ......................................................................................................................... 102
Discussion ................................................................................................................... 103
Differences between  r G' (aq) and  r G' (g) .............................................................. 103
Evidence microbes use gases in the aqueous state. ................................................. 104
Implications for measuring gas concentrations ....................................................... 105
Appendix 1: Demonstration that  r G' (aq) =  r G' (g) under saturation ................... 105
Appendix 2: Derivation of Eq. 2 ................................................................................ 106
Chapter 6: Conclusions .................................................................................................. 113
References ....................................................................................................................... 116
xi
List of Tables
Table 1. Occurrence of energy spilling in microbes. ....................................................... 21
Table 2. Occurrence and characteristics of isolated reserve polysaccharide in rumen
microbes ............................................................................................................................ 24
Table 3. Methods for detecting carbohydrate in rumen microbes. .................................. 26
Table 4. Reserve carbohydrate for cells harvested 30 min after dosing 20 mM final
concentration glucose........................................................................................................ 51
Table 5. Sum of cell components for cells harvested 30 min after dosing 20 mM glucose
(final concentration) .......................................................................................................... 52
Table 6. Properties of reactions relevant to glucose use in rumen microbes (T = 39˚C, pH
= 6.8, I = 0.25 M unless otherwise noted)......................................................................... 84
Table 7. Thermodynamic properties of species and linkages relevant to glucose use in
rumen microbes (T = 25˚C, pH = 0, I = 0) ........................................................................ 85
Table 8. Thermodynamic properties of species and linkages relevant to glucose use in
rumen microbes (T = 39˚C unless otherwise noted, pH = 6.8, I = 0.25) .......................... 87
Table 9. Thermodynamic properties of reactants relevant to glucose use in rumen
microbes (T = 39˚C, pH = 6.8, I = 0.25) ........................................................................... 88
Table 10. Amount of energy spilling following changes in assumed parameter values for
spilling calculations .......................................................................................................... 89
xii
Table 11. Saturation ratios (α) of gases in selected environments of microbes............. 108
Table 12. Difference between ∆rG’(aq) and ∆rG’(g) for selected reactions occurring in
microbial environments. ................................................................................................. 109
xiii
List of Figures
Figure 1. Partitioning of ATP energy towards growth functions, non-growth functions,
and storage. ....................................................................................................................... 31
Figure 2. Bucket model of energy spilling. ....................................................................... 32
Figure 3. Summary of methods for quantifying reserve carbohydrate. ........................... 33
Figure 4. Length of bead-beating and reserve carbohydrate measured from cell
suspensions hydrolyzed with amyloglucosidase. .............................................................. 53
Figure 5. Thin-layer chromatgraphy of cytoplasmic carbohydrate and isolated reserve
polysaccharide................................................................................................................... 54
Figure 6. Absorbance spectrum of isolated reserve polysaccharide in iodine reagent. ... 55
Figure 7. Reserve carbohydrate for washed cells relative to dosing glucose (20 mM final
concentration). .................................................................................................................. 56
Figure 8. Measures relevant to energy and carbon recovery for washed cells relative to
dosing glucose (20 mM final concentration). ................................................................... 58
Figure 9. Absorbance spectra of cell suspensions added to iodine reagent. .................... 60
Figure 10. Response of mixed rumen microbes to glucose dosed at 20 min. .................. 91
Figure 11. Cell components of mixed rumen microbes in response to 20 mM glucose
dosed at 20 min. ................................................................................................................ 93
Figure 12. Response of mixed rumen microbes to starvation. ......................................... 94
xiv
Figure 13. Integrated heat production of mixed rumen microbes in response to glucose
dosed at 20 min. ................................................................................................................ 95
Figure 14. Example illustration of the calculations and overall approach used to quantify
energy-spilling from mixed ruminal microbes from cow 472 dosed with 20 mM of
glucose. ............................................................................................................................. 97
Figure 15. Illustration of aqueous state of hypothetical gas X at saturation, below
saturation, or above saturation. ....................................................................................... 110
Figure 16. Model of CH4(aq) use in Joergensen and Degn ........................................... 112
xv
Chapter 1: Introduction
Owing to ruminants’ ability to digest fiber, ruminant livestock production systems
convert low-quality fibrous and other feedstuffs into highly-nutritious meat and milk. In total,
these systems produce nearly 30 and 100% of the world’s supply of these meat and milk
products (FAO Statistical Databases [http://faostat.fao.org/]).
Although they are a keystone in the world’s food supply, ruminant production
systems also draw public scrutiny by releasing nitrogen (N) into the environment.
Reducing protein fed to ruminants would reduce N release because ruminants excrete
0.11 to 0.14 kg N/kg feed protein [70 to 90% of total N in feed protein; (1)]. In total,
ruminant livestock worldwide excrete 70∙108 kg N/yr (2), including 7.9∙108 kg N/yr for
U.S. dairy cattle alone (3). Nitrogen excreted by ruminants and other livestock leads to
2/3 of anthropogenic output of NO2 (a potent greenhouse gas) and NH3 (which causes
soil acidification, eutrophication, and lower atmospheric visibility) (4-5). Feeding less
protein is otherwise beneficial because it would reduce feed costs to livestock producers.
Feed protein is expensive, with high-quality protein sources such as blood meal and meat
and bone meal more than double the price of low-protein sources, such as corn (6).
Increasing microbial synthesis of amino acids is one strategy to reduce dietary
protein (7). Feed crude protein often has low content of essential amino acids (8), but
microbes growing in the rumen can convert feed crude protein into microbial protein,
1
which is rich with essential amino acids (9). Improving microbial growth efficiency
could thus increase supply of essential amino acids per unit of feed, permit feeding of
lower protein diets, and decrease feed costs and N excretion of ruminant livestock
operations.
In addition to improving growth efficiency, improving prediction of that
efficiency is another strategy to reduce dietary protein. Microbial growth efficiency
ranges widely, but it is poorly predicted by diet formulation systems. For example, one
popular system documented a 4.5-fold range in efficiency, but its prediction equations
could explain only 35% of variation in that efficiency (8). Improving this prediction
would improve prediction of microbial amino acids supplied to the animal, increase
confidence in diet formulation, and potentially eliminate 10% or larger safety margins for
protein in diets (10).
Microbes may growth with low and variable efficiency because they can direct
energy towards (i) futile cycles (energy spilling) and (ii) maintaining their cells
(maintenance) (11). Energy spilling is most commonly a response to excess carbohydrate
(12). Although both spilling and maintenance generate only heat (no growth), spilling
may be considered particularly wasteful because it may not be strictly required for cell
survival, unlike maintenance (11). By wasting carbohydrate otherwise available to
microbial growth, spilling would decrease microbial protein available to ruminants and
require more protein to be fed.
When carbohydrate is in excess, microbes can also respond by directing some
energy towards storage (accumulation of reserve carbohydrate).
2
Unlike energy for
spilling, energy in reserve carbohydrate can later be mobilized for growth (Figure 1) (13),
albeit some ATP energy is spent on synthesizing the reserve carbohydrate and cannot be
recovered.
Directing more energy towards reserve carbohydrate vs. spilling could
improve microbial growth efficiency.
Despite the dualism between spilling and reserve carbohydrate, no study has
examined the relative magnitude of these responses to excess carbohydrate. Studies have
examined spilling and reserve carbohydrate in isolation, such as by using pure cultures
that spill energy but do not accumulate reserve carbohydrate (14). Further, spilling has
been demonstrated for pure cultures of rumen bacteria (15), but the mixed community of
the rumen includes bacteria, protozoa, fungi, and Achaea (16). We can only speculate on
the prevalence of spilling and accumulation of reserve carbohydrate in the rumen.
However, determining their prevalence is a first step towards understanding their impact
on efficiency of microbial growth and feed protein use.
My research aimed to determine how much excess carbohydrate rumen microbes
direct towards synthesis of reserve carbohydrate vs. energy spilling. Before addressing
the central aim, however, I first evaluated methods for detecting reserve carbohydrate, in
an attempt to find one that was quantitative.
I focused on methods based on (i)
hydrolysis with amyloglucosidase and (ii) the anthrone reaction.
I then returned to the central aim of determining the magnitude of reserve
carbohydrate synthesis vs. energy spilling. I washed mixed rumen microbes with N-free
buffer and dosed them with glucose.
I then measured reserve carbohydrate and
3
calculated the amount of cellular heat production accounted by synthesis reserve
carbohydrate vs. spilling (as well as endogenous metabolism).
Experiments described above required calculation of thermodynamic properties.
In the course of these experiments, I found that calculation of one property, transformed
Gibbs energy of reaction (  r G ' ), was sensitive to choice of aqueous vs. gaseous state of
gases. My final aim was to (i) use a compilation of literature values to infer which state
(gaseous, aqueous) microbes indeed use and (ii) determine how much  r G ' is impacted
by choice of state under typical conditions in the literature.
4
Chapter 2: Literature Review
Ruminants can digest fibrous feedstuffs owing to microbes that inhabit their forestomach
(rumen). These microbes (bacteria, protozoa, fungi, archaea) anaerobically degrade feed
to produce ATP (and byproducts such as short-chain fatty acids). With this ATP,
microbes synthesize cellular protein that makes more than 50% of amino acids reaching
the small intestine of their ruminant host (9, 17). A major goal in livestock research is to
predict and improve efficiency with which microbes harness fermentation energy for
growth (production of cells) (18).
Microbes often do not grow with high efficiency because they do not direct all
energy towards growth, but rather direct some energy towards non-growth functions—
e.g., maintenance and energy spilling (Figure 1).
Maintenance functions are those
required for cellular “housekeeping” and include re-synthesis of protein following
turnover (14). Energy spilling [e.g. futile cycling of ions or reserve carbohydrate; (11,
14, 19)] refers to energy dissipated as heat when the ATP generated from catabolism
exceeds ATP consumed for maintenance, growth, and synthesis of reserve carbohydrate.
It can be analogized to water spilling over the brim of an overfilled bucket (Figure 2). It
is commonly a response to excess carbohydrate (12), as can occur when the ruminant is
fed grain.
5
Maintenance and energy spilling alike reduce growth efficiency (c.f., Figure 1),
and reducing each would improve growth efficiency.
Spilling, however, may be a
prominent target to improving efficiency because, unlike maintenance, it may not strictly
be needed for cell survival. Energy spilling has been detected in some pure cultures (15),
but it has been seldom investigated in mixed communities. Thus, the impact of spilling
on growth efficiency in still incompletely understood.
When given excess carbohydrate, microbes can also direct energy towards
storage, such as synthesis of reserve carbohydrate (Figure 1). Unlike with maintenance
or spilling, energy in reserve carbohydrate can later be mobilized for growth (Figure 1,
Figure 2) (13), albeit some ATP energy is spent on synthesizing the reserve carbohydrate
during storage and cannot be recovered (Figure 2).
Directing more energy towards reserve carbohydrate vs. spilling or maintenance
would improve microbial growth efficiency. As a first step towards this goal, we need to
determine how much energy is normally directed towards these functions.
Occurrence of energy spilling. Although maintenance can depress microbial
growth efficiency, only energy spilling can explain very low efficiency during
carbohydrate excess (and other growth-limiting conditions). Russell (15) demonstrated
energy spilling by pulse-dosing rumen bacterial cultures with glucose.
Cultures
fermented excess glucose rapidly, produced very little protein (growth efficiency
approached 0), and dissipated (spilled) energy by producing heat. Energy spilling may be
a major sink for fermentation energy: rumen bacteria may ferment glucose 10-fold faster
when spilling energy (20).
6
Many pure cultures probably spill energy (Table 1).
Spilling has been
demonstrated extensively in a few rumen (Streptococus bovis) and non-rumen
(Escherichia coli, Klebsiella aerogenes) bacteria. Some evidence for spilling exists for
other microbes, including protozoa, fungi, and Archaea.
Spilling can be evidenced by depressed growth efficiency or elevated heat
production in response to excess carbohydrate (Table 1). Carbohydrate excesses have
been generated by pulse dosing glucose or growing cells under limitation of an anabolic
substrate (e.g., N, Mg, P, S, K). Spilling can also be evidenced in response to (i)
ammonia-N replacing amino-N (ii) excess H2 or CO2 (for methanogens), and (iii)
exogenous protonophore.
When measuring changes in growth efficiency and heat
production, one must account for any changes in maintenance, reserve carbohydrate, and
growth (c.f., Figure 1), though this is sometimes not done (Table 1).
The mechanism of spilling is by futile cycles of ions, glycogen, or trehalose
(Table 1). The best-elucidated mechanism is for S. bovis, for which spilling occurs by
futile cycling of protons. This cycling results from growth limitation and a cascade of
biochemical events (16). Specifically, a growth limitation decreases use of ATP for
protein synthesis, increasing fructose 1,6 bisphosphate, decreasing intracellular
phosphate, increasing Gibbs energy of ATP hydrolysis, increasing activity of a protonpumping ATPase, and decreasing membrane resistance to protons. The net result is heat,
with no work done by the protons.
For E. coli, the mechanism of energy spilling is by cycling of K+ or NH4+ or a
combination thereof, depending on the extracellular concentration of K+ and NH4+
7
(Table 1). For many other organisms, glycogen and trehalose cycling may occur (Table
1) and suggest presence of spilling, even though more direct evidence of spilling
(depressed growth efficiency or elevated heat production) has often not been reported.
Some authors have proposed that fructose-6-phoshate/fructose-1,6-bisphosphate (21) and
other substrates (22) can be cycled. However, such cycling is difficult to establish in
microbes under physiological conditions (11, 19).
Many pure cultures have been demonstrated to spill energy (Table 1), but
examples of mixed cultures spilling energy are few.
Van Kessel and Russell (20)
suggested that rumen bacteria spilled energy when grown under ammonia-N limitation,
but they did not measure reserve carbohydrate. Some energy may have in fact been
directed to reserve carbohydrate synthesis, not spilling. Chen et al. (23) induced spilling
by adding a protonophore, but spilling was not measured under more physiological
conditions. Further study is needed to determine if spilling routinely occurs in mixed
cultures, such as those from the rumen, to determine in turn how relevant spilling may be
to livestock operations.
Occurrence and identity of reserve carbohydrate. When carbohydrate is in
excess, many bacteria synthesize large amounts of carbohydrate (12), rather than spilling
some or all energy. This reserve carbohydrate may exceed 50% of cell weight for pure
cultures of both rumen (12) and non-rumen (24) bacteria. Some protozoa (Isotrichidae)
accumulate enough reserve carbohydrate to turn opaque (c.f., Figure 3.6 in ref. (25)]
For rumen microbes, reserve carbohydrate appears to be glycogen [glucan with
(α1→4) and (α1→6) linkages]; when reserve carbohydrate has been isolated and
8
characterized, its identity has always been glycogen (Table 2). Such characterization has
involved determining linkages (usually by enzymatic or gas chromatography
approaches), chain length, molecular weight, iodine staining, and infrared spectrum.
Results of this characterization again are consistent with reserve carbohydrate being
glycogen (Table 2).
Glycogen appears to be ubiquitous across rumen bacteria, fungi, and protozoa
(Table 2). Organisms that accumulate glycogen represent a wide range of niches, such as
sugar-utilizers (mixed Isotricha and Dasytricha sp.), cellulose-utilizers (Fibrobacter
succinogenes), and generalists (Prevotella ruminicola). Glycogen was not detected in
one species of hyper-ammonia bacteria, Eubacterium pyruvativorans (26), but additional
species have not been studied. Glycogen has not been studied in any rumen methanogen.
Trehalose, a reserve carbohydrate of yeast and other organisms (27-29), has not
been found in rumen microbes. However, only in one study (of fungi) did authors
specifically test for its presence (30).
Relative magnitude of responses. Because many rumen microbes can spill
energy or synthesize reserve carbohydrate (Table 1, Table 2), we might expect these
responses to occur concurrently in the rumen. However, studies have examined single
responses only in isolation, such as by using pure cultures that spill energy but do not
accumulate reserve carbohydrate (14). Measuring these responses together is a first step
to understanding their combined impact on efficiency of microbial growth in the rumen.
Other responses to excess carbohydrate. Rumen microbes may respond to
excess carbohydrate in ways other than spilling energy and synthesizing reserve
9
carbohydrate. These other responses include reducing ATP yield by releasing metabolic
intermediates (overflow metabolites), and shifting to catabolic pathways that yield less
ATP (12). Responses are similar for non-rumen microbes (11, 14, 24, 31). These
responses seem minor for rumen microbes, but they might still need to be considered
when otherwise trying to compare energy spilling vs. synthesis of reserve carbohydrate
for mixed microbial communities.
Occurrence of excess carbohydrate in the rumen. Carbohydrate is in greatest
excess in the rumen for grain-fed animals, particularly those transitioning to a high-grain
diet. For these animals, glucose can reach high concentrations [c. 5 mM (32-33)]. For
grain-fed animals, N availability can be low (34), also, and intensify carbohydrate excess.
(34). For animals where N is chiefly in the form of ammonia, carbohydrate excess could
be further intensified (20) because rumen microbes grow far slower with ammonia-N
than amino-N (20, 35). Responses to excess carbohydrate, as discussed above, may be
especially active in animals under these feeding conditions (12).
When animals are fed high-forage diets or adapted to grain, concentrations of
glucose in the rumen are low, rarely exceeding 1 mM (36).
Responses to excess
carbohydrate may therefore be less important than for the grain-fed animal. Even still,
mixed rumen bacteria had accumulated reserve polysaccharide when their hosts were
provided low-quality grass diets (37), and thus responses discussed above may still have
some relevance.
Function of energy spilling. Spilling energy may seem counter-intuitive: why
expend what could be conserved? One simple reason a microbe may spill energy is to
10
deprive competitors of energy, particularly when N (or another anabolic substrate) limits
growth. Having vanquished its competitors, the microbe would have a growth advantage
when N is replenished. However, this argument has been called teleological because it
presumes that microbes can forecast and plot for the future (11).
Spilling may offer a growth advantage for a different reason.
Spilling may
function in keeping a high rate of catabolism needed for growth, even when N (or another
anabolic substrate) limits growth. If N is replenished, microbes spilling energy could reinitiate growth faster than those whose catabolism was stalled (11, 38). This function is
thought to evolved in mixed communities, which can be flooded with nutrients after
being nutrient-starved for long periods (38). However, spilling has not been conclusively
demonstrated in mixed communities (see above).
Rather than offer a growth advantage, spilling may function to prevent toxicity
arising from excess carbohydrate. When glucose is in excess, some microbes produce
methylglyoxal, a toxic metabolite (12).
Spilling could provide an alternate to
methylglyoxal production and prevent toxicity (11).
Specifically, methylglyoxal is
produced because the methylglyoxal shunt does not require ADP, which is scarce during
glucose excess because of activity of ATP-generating pathways. Microbes that spill
energy, however, would not need to produce methylglyoxal because spilling regenerates
ADP from ATP, allowing ATP-generating pathways to continue.
This argument
presumes that methylglyoxal can reach toxic concentrations, but methylglyoxal can be
detoxified to D-lactate by glyoxalase (12). For Prevotella ruminicola, methylglyoxal
11
reached toxic concentrations because detoxification was negligible (39). For other rumen
microbes, such toxicity has not been demonstrated.
Regardless of its function, energy spilling does not seem essential for survival for
the rumen community as a whole, and it is thus particularly wasteful from the perspective
of growth efficiency.
To understand how potentially “wasteful” it is in livestock
operations, further study is needed.
Function of reserve carbohydrate. Reserve carbohydrates conventionally have
been viewed as true storage compounds—first accumulated during growth limitation and
carbohydrate excess, then later mobilized during carbohydrate limitation (27, 40).
According to this view, these carbohydrates are mobilized for survival during nutrient
starvation (27, 40) or for growth when N is replenished (13). Glycogen may otherwise be
mobilized for spore formation and maturation (40). Trehalose may otherwise serve a
protective function during heat or osmotic stress (29).
Another, emerging view for glycogen is that serves as a “carbon capacitor” (41),
modulating the flow of carbon into glycolysis. This view recognizes that glycogen can be
cycled (Table 1) and synthesized during growth (41-42), unlike a strict storage
compound.
It suggests that the net direction of cycling (synthesis or degradation)
modulates the flow of carbon into glycolysis, in order to prevent excesses or shortages of
glycolytic intermediates. Under certain conditions, glycogen may even be synthesized
during growth.
Although ATP is expended for this simultaneous degradation and
synthesis, cycling may still yield a net advantage for growth: cells able to degrade
glycogen (and thus potentially cycle it) grew faster than mutants (41).
12
The carbon capacitor view contradicts the notion that glycogen synthesis and
degradation are tightly regulated (40, 43), at least for model organisms E. coli and
Saccharomyces cerevisiae.
Indeed, some authors have concluded glycogen cycling
probably does not occur for these organisms (11), prohibiting glycogen from serving as a
carbon capacitor. Some authors suggest a small amount of cycling may indeed occur in E.
coli and Saccharomyces cerevisiae, based on mutants deficient in glycogen
phosphorylase (44-45). However, such genetic studies are indirect, and cycling has not
been demonstrated using more direct methods [viz., with isotopic labeling (19)].
If the carbon capacitor view applies to rumen microbes, then we should expect
rumen microbes that accumulate glycogen to cycle it—i.e., spill at least some energy.
This perspective provides even more justification for investigating spilling and synthesis
of reserve carbohydrate concurrently.
Detection of reserve carbohydrate. Reserve carbohydrate must be detected
quantitatively to determine exactly how much excess carbohydrate rumen microbes direct
towards reserve carbohydrate vs. energy spilling. However, studies have used different
methods for detecting carbohydrate in rumen microbes. These studies use divergent
approaches for extracting carbohydrate, hydrolyzing it, and detecting it after hydrolysis
(Table 3). In some studies, the carbohydrate is also isolated (Table 3), again by divergent
approaches.
Despite the diversity of approaches in Table 3, some common methods can be
identified in these studies and those with non-rumen microbes (Figure 3). One method is
determining glycogen and trehalose enzymatically (e.g., by amyloglucosidase or
13
trehalase) (46-48). Another method is determining total hexoses chemically [e.g., by the
anthrone reaction (28, 49)].
Methods can employ cell suspensions or reserve
polysaccharide isolated using ethanol precipitation (49). For enzymatic detection and for
isolation of reserve polysaccharide, the carbohydrate must first be extracted (46-48, 50).
Comparison of methods suggests they detect disparately different values of
carbohydrate (51-52), albeit such comparisons have been made at only one time point
during experiments.
Some authors have suggested their method (amyloglucosidase
hydrolysis) completely detects glycogen because iodine staining was lost after hydrolysis
(47, 50), but the sensitivity of iodine staining was not determined.
Further work is needed to determine which method detects the reserve
carbohydrate completely and quantitatively. Because methods are frequently used to
monitor changes in reserve material, methods should be compared at multiple time points
over an experiment; previous comparisons at single time points (51-52) cannot resolve
such changes. Iodine staining should be evaluated to determine how sensitively it can
detect non-hydrolyzed glycogen. Finally, methods should be compared to determine if
any gives complete recovery of energy, carbon, and cell components—perhaps the most
conclusive way to determine if detection was complete. Once a method is identified, it
can be applied to determining how much carbohydrate rumen microbes direct towards
reserve carbohydrate vs. energy spilling.
Calculation of spilling. As mentioned, spilling is evidenced by elevated heat
production in response to carbohydrate.
Before inferring that spilling has indeed
occurred, one must account for heat production from other functions (maintenance,
14
reserve carbohydrate synthesis, and growth; c.f., Figure 1). How does one calculate heat
production for these functions and thus spilling?
Results from Cook and Russell (53) show the approach for these spilling
calculations. These authors dosed S. bovis with excess glucose, and then measured
change in heat production and cell protein. Lactate was the sole fermentation product.
Heat production from maintenance was assumed a constant value using earlier work (54).
Heat production from growth was calculated by multiplying cell protein, growth yield per
ATP, and heat per lactate, and ATP per lactate as
g cell protein/s x mol ATP/g protein / (1 mol ATP/mol lactate) x -44.1 kJ
1 (1)
heat/mol lactate x -1
No reserve carbohydrate was accumulated by S. bovis, and thus it did not factor into the
calculations. Spilling was calculated as total heat production minus that for maintenance
and growth. The authors expressed their final results in terms of glucose consumption
using a constant conversion factor between heat produced and glucose consumed (-88.2
kJ/mol glucose).
As Figure 18-12 in Cook and Russell (53) shows, S. bovis responded to excess
glucose with elevated glucose consumption (heat production), but only a portion of that
glucose consumption was accounted by spilling.
Without correcting for glucose
consumption accounted by growth and maintenance, spilling would have been
overestimated.
The calculations of Cook and Russell (53) apply only to a simple scenario
(formation of 1 fermentation product, no reserve carbohydrate accumulation). However,
15
calculations could be generalized to complex scenarios likely with mixed cultures. To
accommodate more fermentation products, the calculation would simply need weighted
averages of (i) mol ATP/mol fermentation product and (ii) kJ heat/mol fermentation
product (see eq. 1). Values of mol ATP/mol fermentation product can be found in texts
[e.g., (16)].
Values of kJ heat/mol fermentation product can be calculated using
enthalpies from thermodynamic tables (as described below).
To accommodate reserve carbohydrate accumulation, the calculation of Cook and
Russell (53) would need the additional equation
g reserve carbohydrate/s x mol ATP/g reserve carbohydrate x mol ATP/mol
2 (2)
fermentation product x kJ heat/mol fermentation product
that follows the form of eq. 1 above. Equation 2 ignores heat released from condensation
of glucose to reserve carbohydrate, but it could be included if found to be non-negligible.
Values of g reserve carbohydrate/s could be measured experimentally.
Heat released by formation of fermentation products.
Heat released by
formation of fermentation products (kJ/mol) can be calculated from enthalpies in
thermodynamic tables. The full set of calculations is
 r H '0 
N'
 vi'  f H i'0
3 (3)
i 1
heat released by formation of product i (kJ/mol)   r H '0 / vi'
4 (4)
where  r H '0 is the standard transformed enthalpy of reaction (kJ/mol), N ' is the number
of reactants, v i' is the stoichiometric number of reactant i (dimensionless), and  f H i'0 is
the standard transformed enthalpy of formation of reactant i (kJ/mol).
16
An example calculation is done for glucose (Glc) fermentation to lactate (Lac):
Glucose = 2 Lactate
5 (5)
'0
'0
'
'
By definition, vGlc
 2 . Further,  f H Glc  1264.25 and  f H Lac  685.94 ,
 1 and vLac
which were calculated using the thermodynamic tables and equations in Alberty (55) at
temperature = 39˚C, pH = 7, and ionic strength = 0.25 M.
 r H '0  1  1264.25  2  685.94  107.63
6 (6)
heat released by formation of lactate (kJ/mol)   107.63/2  53.82
7 (7)
This final value, -53.82 kJ/mol, can be deployed in energy spilling calculations (e.g., in
eq. 1). This compares with the value -44.1 kJ heat/mol lactate used by Cook and Russell
(53) and appearing in eq. 1. The source of -44.1 kJ heat/mol lactate was not referenced
by those authors.
Physical state and thermodynamic calculations. The simple equations above
can make thermodynamic calculations appear as a textbook exercise. However, the
experimenter must make key decisions to deploy these equations.
Among other
decisions, one must specify the physical state of reactants.  f H '0 differs across these
states (55) and will impact the calculation of  r H '0 accordingly. Transformed Gibbs
energy of formation (  f G '0 ; kJ/mol), another thermodynamic property, also differs
across states (55) and would impact its counterpart, transformed Gibbs energy of reaction
 r G '0 .
17
Most reactants exist in one state (aqueous solution) (55), and thus specifying state
is a non-issue. However, gases can exist in both the aqueous states (dissolved in liquid)
and gaseous states (in headspace) in most microbial cultures. Because calculation of
energy spilling requires  r H '0 , state of gases should impact energy spilling calculations
in turn; we must investigate exactly how quantatively important the impact is.
We could further investigate how Gibbs energy, also, is impacted by state of
gases. Although the impact of state on  r G '0 is known (55), it is not known for a related
property, transformed Gibbs energy of reaction (  r G '0 ).  r G ' is similar to  r G '0 but
with the concentration of reactants specified (e.g., to physiological conditions), and so
 r G ' is more physiologically relevant than  r G '0 . Because  r G ' is a key property in
microbial energetics (56), we need to determine the impact of specifying state of gases on
its calculation.
Improving microbial growth efficiency and its prediction.
Improving
efficiency of microbial growth is one strategy to reduce protein fed to ruminants (7).
Rumen microbes generally have higher content of essential amino acids than feed crude
protein (8-9), and so less protein has to be fed if appreciable amounts of dietary protein
are converted to microbial protein. As mentioned, directing more energy towards reserve
carbohydrate vs. spilling would improve microbial growth efficiency, and thus study of
these responses could reduce dietary protein.
Improved prediction of microbial growth efficiency is another strategy to reduce
dietary protein. Microbial growth efficiency ranges widely [4.5-fold range in one review
(8)]. Accurate prediction of this efficiency would improve prediction of microbial amino
18
acids supplied to the animal, increase confidence in diet formulation, and potentially
eliminate the 10% or larger safety margins for protein in diets (10).
Approaches for predicting growth efficiency currently exist but are imperfect. For
example, the best equation of the popular NRC (8) system explained only 35% of variation in
efficiency.
Similar to most other empirical systems, the Dairy NRC (8) predicts
efficiency from a simple constant, adjusted for availability of ruminally-degraded N. It
does not account for spilling or reserve carbohydrate, despite their likely impact on growth
efficiency.
There exist more sophisticated, mechanistic approaches to predicting growth
efficiency. Some mechanistic models already account for maintenance, energy spilling and
reserve carbohydrate (57-60). However, many of their parameter values are heuristic (not
derived from data) or simple constants, owing to lack of experimental data. For example,
most models assume storage of reserve carbohydrate is a constant fraction of microbial
biomass (57-60). The application of these models has been restricted mainly to research.
Experiments on energy spilling and reserve carbohydrate could better parameterize these
models, improving their prediction of growth efficiency and broadening their application.
Another limitation of empirical systems is that they often express growth efficiency
in terms of total tract digestible OM (8, 61). Rumen microbes cannot generate ATP for
growth from all OM digested in the total tract. In particular, rumen microbes cannot generate
ATP from OM digested post-ruminally, and they generate little or no ATP from long-chain
fatty acids, short-chain fatty acids, and protein digested in the rumen (61). Mechanistic
models have addresses this limitation by expressing growth efficiency in terms of ATP or
ruminally-fermented carbohydrate (61).
19
Conclusion. Many microbes respond to excess carbohydrate by spilling energy,
which decreases growth efficiency. Others may synthesize reserve carbohydrate, which
can be later used for growth. Directing more energy towards reserve carbohydrate vs.
spilling or maintenance could improve microbial growth efficiency.
Because many rumen microbes can spill energy or synthesize reserve
carbohydrate, we might expect to detect those responses concurrently in the rumen. We
would expect to detect them concurrently, also, if glycogen serves as a carbon capacitor.
However, the relative magnitude of these responses has not been determined. Further,
spilling has seldom been studied in mixed communities—much less those from the
rumen—thus its impact on in rumen becomes difficult to assess. To measure these
responses accurately, we need a quantitative method for detect reserve carbohydrate, but
several methods of unknown accuracy exist.
Overcoming these limitations could
improve our understanding of microbial energy metabolism and translate to improved use
of feed protein of livestock operations.
20
Table 1. Occurrence of energy spilling in microbes.a
Species
Ruminal
Heat production or
isolate
growth yield evidence
Bacteria
Escherichia coli N
Lower growth yield per
ATP under Mg, P, S, K
vs. C-limitation
N
N
Klebsiella aerogenes N
21
Fibrobacter Y
succinogenes
Prevotella ruminicola Y
Selemonas Y
ruminantium
Streptococus bovis Y
Mechanism (type
of cycling)
Notes
Reference
NH3/NH4+/H+
Mechanism applies during
high NH3/NH4 and low K+
(62)
Lower growth yield per
ATP of wild-type vs. K+
transport mutant
ND
K+
Mechanism applies during
low K
(63)
NH3/NH4+/H+/K+
(11)
Lower growth yield per
ATP under NH3, P, S, or
K vs. C-limitation
ND
ND
Mechanism applies during
low NH3/NH4
Heat production rose
rapidly during glucose
excess
Heat production rose
rapidly during glucose
excess
Heat production not
accounted by
maintenance or growth
ND
(64-65)
Glycogen
ND
H+
(42)
Growth and reserve
carbohydrate not
accounted explicitly
Growth and reserve
carbohydrate not
accounted explicitly
(15)
(15)
(53, 6668)
CONTINUED
21
Table 1: CONTINUED
Mixed rumen bacteria Y
Protozoa
Isotricha protostoma Y
Dasytricha Y
ruminantium
Fungi
Saccharomyces N
cerevisiae
22
N
N
Archaea
Methanobacterium N
thermoautotrophicum
Lower growth yield per
hexose under NH3-N vs.
amino-N
Not defined
Reserve carbohydrate or
other cell composition
changes not accounted
explicitly
(20)
ND
ND
Glycogen
Glycogen
(69)
(70)
Lower growth yield per
ATP under glucose
excess
Higher heat production
and lower growth yield
under N vs. glucoselimitation
ND
ND
(71)
Lower growth yield per
CH4 under H2- or CO2excess vs. limitation
ND
Reserve carbohydrate not
accounted explicitly
(72)
Trehalose
Mechanism applies under
heat shock
(73)
ND
Reserve carbohydrate or
other cell composition
changes not accounted
explicitly
(74-75)
CONTINUED
22
Table 1: CONTINUED
Undefined
Mixed microbes in N
activated sludge
Lower growth yield with
protonophore (3,3’,4’5tetrachlorosalicylanilide)
a
N = no, Y = yes, ND = Not determined
23
23
H+
(23)
Table 2. Occurrence and characteristics of isolated reserve polysaccharide in rumen microbesa
Species
Monomer
Linkages
Chain
Molecular
Iodine
Infrared
Reference
length
weight
stainingb spectrumb
(MDa)
Bacteria
Bacteriodes amylogenes Glucose
ND
ND
ND
Y
NA
(76)
Fibrobacter succinogenes ND
ND
ND
ND
Y
Y
(77)
Megasphaera elsdenii Glucose
(α1→4) and
9
>0.2
Y
Y
(78-79)
(α1→6)
Provetella ruminicola Glucose
24
(α1→4) and
(α1→6)
Ruminococcus albus Glucose
ND
Glucose
(α1→4) and
(α1→6)
Selenomonas ruminantium Glucose
ND
Glucose
(α1→4) and
(α1→6)
Mixed streptococci Glucose, trace at least
mannosec
(α1→4)
Fungi
Neocallimastix sp. strain LM-1 Glucose
(α1→4) and
(α1→6)
Protozoa
Mixed Entodinium sp. (chiefly Glucose
(α1→4) and
E. simple and E. nanellum)
(α1→6)
8
>0.2
ND
ND
(80)
ND
ND
ND
>0.2
ND
Y
ND
Y
(81)
(82)
12
23.5
0.5 to >2
20
Y
Y
ND
ND
(83)
(84)
ND
ND
Y
ND
(85)
ND
ND
ND
ND
(30)
22 to 25
ND
Y
Y
(86)
CONTINUED
24
Table 2: CONTINUED
Mixed Entodinium sp. (chiefly Glucose
E. caudatum and E. simplex)
Mixed Isotricha and Glucose
Dasytricha sp.
a
ND = not determined, Y = yes
b
Characteristic of glycogen.
c
Possible trace rhamnose
(α1→4)
19
ND
Y
ND
(87)
(α1→4) and
(α1→6)
23
0.17
Y
ND
(88)
25
25
Table 3. Methods for detecting carbohydrate in rumen microbes.
Method
Species
Bacteria
Clostridium
aminophilum
Clostridium
sticklandii
Fibrobacter
succinogenes
Extraction
Isolation of reserve
polysaccharide
Hydrolysis
Detection of
released
carbohydrate
Reference
26
None
None
70% H2SO4 (100˚C, 7 min)
Anthrone
(89)
None
None
70% H2SO4 (100˚C, 7 min)
Anthrone
(89)
2% SDS
(unspecified
time and
temperature)
None
None
Amyloglucosidase
Glucose oxidase- (42)
peroxidase
None
Anthrone
(77)
None
None
Phenol-sulfuric
acid
(77)
None
None
75% H2SO4 (100˚C, 10
min)
100% H2SO4 (RT for 10
min, 25 to 30˚C for 10 to
20 min)
70% H2SO4 (100˚C, 7 min)
Anthrone
(90)
CONTINUED
26
Table 3: CONTINUED
Megasphaera Lysozyme
elsdenii (37˚C, 2 h),
sonification
(100 W, 90 s),
30% KOH
(100˚C, 3 h)
Lysozyme
(37˚C, 2 h),
sonification
(100 W, 90 s),
30% KOH
(100˚C, 3 h)
None
27
Peptostreptoccus None
anaerobius
Prevetella None
ruminocola
None
None
Ruminococcus 30% NaOH
albus (100˚C, 3 h)
Centrifugation to remove
cell debris; repeated
precipitation with 95%
ethanol; dialysis
70% H2SO4 (100˚C, 7 min)
Anthrone
(78)
Centrifugation to remove
cell debris; repeated
precipitation with 95%
ethanol; dialysis
1 N HCl or 1 N H2SO4
(100˚C, 18 h)
Glucose oxidase
(78)
None
100% HCl (100˚C, 4 h)
(91)
None
70% H2SO4 (100˚C, 7 min)
Phenol-sulfuric
acid
Anthrone
None
70% H2SO4 (100˚C, 7 min)
Anthrone
(15)
None
None
Precipitation with 86%
ethanol; washing with
66% ethanol; filtration to
remove insoluble debris
70% H2SO4 (100˚C, 7 min)
70% H2SO4 (100˚C, 7 min)
10% HCl (100˚C, 3 h)
Anthrone
Anthrone
Copper
reduction
(93)
(80)
(81)
(92)
CONTINUED
27
28
Table 3: CONTINUED
Lysozyme
(37˚C, 2 h),
sonication (100
W, 90 s), 30%
KOH (100˚C, 3
h)
Lysozyme
(37˚C, 2 h),
sonication (100
W, 90 s), 30%
KOH (100˚C, 3
h)
Ruminococcus None
flavefaciens
Selenomonas 30% KOH
ruminantium (100˚C, 3 h)
Centrifugation to remove
cell debris; repeated
precipitation with 95%
ethanol; dialysis
75% H2SO4 (100˚C, 10
min)
Anthrone
(82)
Centrifugation to remove
cell debris; repeated
precipitation with 95%
ethanol; dialysis
1 N HCl or 1 N H2SO4
(100˚C, 18 h)
Glucose oxidase
(82)
None
100% HCl (100˚C, 4 h)
(94)
100% H2SO4 (RT for 10
min, 25 to 30˚C for 10 to
20 min)
None
Centrifugation to remove
cell debris; precipitation
with 72% ethanol;
washing with 60%
ethanol
None
Phenol-sulfuric
acid
Phenol-sulfuric
acid
Anthrone
(83)
None
None
Phenol-sulfuric
acid
(83)
100% H2SO4 (100˚C, 7
min)
100% H2SO4 (RT for 10
min, 25 to 30˚C for 10 to
20 min)
(83)
CONTINUED
28
Table 3: CONTINUED
French pressure
cell (100
kg/cm2)
2 N HCl (100˚C, 3 h)
Glucose oxidase- (86)
peroxidase
None
4% SDS (100˚C, 15
min), DNase and Rnase
treatment; centrifugation
to remove peptidoglycan
None
70% H2SO4 (100˚C, 7 min)
Anthrone
None
None
100% HCl (100˚C, 4 h)
None
α-amylase, glucoamylase
None
Amyloglucosidase
Phenol-sulfuric
(95)
acid
Glucose oxidase- (96)
peroxidase
Glucose oxidase (97)
None
1 N H2SO4 (100˚C, 4 h)
None
None
None
None
None
None
None
None
100% H2SO4 (RT for 10
min, 25 to 30 C for 10 to
20 min)
2 N H2SO4 (110˚C, 2h)
70% H2SO4 (100˚C, 7 min)
100% HCl (100˚C, 4 h)
50% CaCl2
Mixed bacteria (125˚C, 15 min)
H2O (100˚C, 4
h)
None
(15)
29
Ion-exchange
chromotography
Phenol-sulfuric
acid
(97-101)
Neocuproine
Anthrone
Phenol-sulfuric
acid
(102)
(37)
(103)
(102)
CONTINUED
29
Table 3: CONTINUED
Fungi
Neocallimastix 0.25 M Na2CO3, None
sp. strain LM-1 0.5 M HClO4
(unspecified
time and
temperature)
Protozoa
1 N NaOH
None
Dasytricha (100˚C, 5 min)
ruminantium
none
None
30
Isotricha none
prostoma
Mixed bacteria
and protozoa
0.2 N NaOH
(100˚C, 15 min)
Sonication (2
min, 20 kHz)
Sonication (2
min, 20 kHz)
Ethanol and
DMSO (100˚C,
7 min)
None
None
None
None
None
30
75% H2SO4 (100˚C, 10
min)
Anthrone
(30)
100% H2SO4 (room
temperature for 10 min, 25
to 30˚C for 10 to 20 min)
66% H2SO4 (100˚C, 15
min)
66% H2SO4 (100˚C, 15
min)
Phenol-sulfuric
acid
(104)
α-amylase,
amyloglucosidase
72% H2SO4 (RT,
overnight)
72% H2SO4 (RT,
overnight)
α-amylase, pullanase and
β-amylase,
amyloglucosidase
Glucose oxidaseperoxidase
Phenol-sulfuric
acid method
Phenol-sulfuric
acid method
Glucose oxidaseperoxidase
Glucose oxidase- (70)
peroxidase
Glucose oxidase- (69)
peroxidase
(105)
(106)
(106)
(107)
Growth
Fermentation
Feed
Storage
Non-growth
functions
ATPReserve
Cells functions equivalcarbohydrate
ents Mobilization
Energy
spilling
Maintenance
Figure 1. Partitioning of ATP energy towards growth functions, non-growth functions,
and storage. ATP-equivalents can include ATP or ATP-yielding carbon compound (e.g.,
glucose). Modified from (14, 108).
31
Catabolism
Energy spilling
Reserve
carbohydrate
Growth
Maintenance
Figure 2. Bucket model of energy spilling. The large bucket represents the main pool of
ATP-equivalents available to cell functions (maintenance, growth, reserve carbohydrate,
energy spilling). The smaller bucket represents pool of ATP-equivalents in reserve
carbohydrate, which can be stored from and mobilized to the main pool by pumps.
Modified from (16).
32
Microbial cells
-Hydrolysis with 75% H2SO4/
detection of released hexoses
by anthrone reaction
Hexoses in
cell suspensions
-Extraction
(e.g., bead-beating or 50% KOH)
Lysed cells
-Hydrolysis with amyloglucosidase or
trehalase
-Detection of released glucose by
glucose oxidase-peroxidase
Glycogen or
trehalose in cell
suspensions
-Removal of cellular debris
-Precipitation and
washing with ethanol
Isolated reserve
polysaccharide
-Hydrolysis with 75% H2SO4/
detection of released hexoses
by anthrone reaction
Hexoses in isolated
polysaccharide
-Digestion with amyloglucosidase
-Detection of released glucose
by glucose oxidase-peroxidase
Glycogen in isolated
polysaccharide
Figure 3. Summary of methods for quantifying reserve carbohydrate. Shown are most
common methods in Table 3 and in seminal studies with non-rumen microbes (cited in
text).
33
Chapter 3: Evaluation of methods to detect changes in reserve carbohydrate for mixed
rumen microbes
Abstract: Thin-layer chromatography and other analyses revealed that rumen microbes
accumulated large amounts of glycogen. The aim of this study was to identify a method
that would most accurately quantify synthesis and utilization this reserve carbohydrate.
For whole cells, the anthrone reaction detected more (P < 0.001) carbohydrate than did
hydrolysis with amyloglucosidase, even after exhaustive extraction by bead beating (45
min) or KOH digestion (3 h). Less carbohydrate was detected after isolating reserve
polysaccharide by ethanol precipitation. Compared to the amyloglucosidase hydrolysis
method, the anthrone method detected a larger (P = 0.017) increase in cell carbohydrate
when glucose (20 mM) was dosed in cultures. Additionally, it detected a larger (P =
0.049) decrease in cell carbohydrate after glucose was exhausted. This result indicated
that the anthrone method detected more carbohydrate that functions as a reserve material,
which accumulates during energy excess and is utilized for energy during starvation. For
the anthrone method, recoveries for energy (97.5%), carbon (100.2%), and cell
components (99.8%) were high, indicating carbohydrate was completely detected. For
the amyloglucosidase hydrolysis method, recoveries of energy (88.9%), carbon (91.6%),
and cell components (92.8%) were lower. Some authors have inferred from iodine
34
staining that amyloglucosidase hydrolyzes all glycogen in cells. However, iodine did not
stain glycogen remaining after cells were intentionally incompletely extracted.
The
anthrone method appeared to accurately quantify changes in reserve carbohydrate and
shows merit for quantitative studies, whereas the amyloglucosidase hydrolysis method
detected smaller changes and was less consistent with expected carbon and energy
recovery.
Introduction
Rumen microbes are a diverse community that that facilitate digestion of fiber by their
ruminant hosts (17). They seldom encounter rapidly-utilizable, non-fibrous carbohydrate
above those found in immature forages (17). When their ruminant hosts are fed higher
amounts of grain to stimulate productivity, however, carbohydrate concentrations in the
rumen can exceed the capacity for microbial growth (16). Some microbes respond to this
excess carbohydrate by synthesizing reserve carbohydrate (12, 25).
This reserve
carbohydrate may be an important source of carbohydrate digested in the small intestine
of the ruminant (97), and its synthesis may prevent rumen acidosis by diverting
carbohydrate from immediate fermentation after a meal (25).
Reserve carbohydrate may exceed 50% of cell weight for pure cultures of both
rumen (12) and non-rumen (24) bacteria. Because reserve carbohydrate is such a major
sink for excess carbohydrate, completely detecting it is essential for quantitative studies
of carbohydrate metabolism.
Accurate detection of reserve carbohydrate under the
dynamic conditions of the rumen limits the implementation of mechanistic knowledge of
rumen microbial growth to improve ruminant animal productivity and consequently to
35
decrease environmental impact (109). For mechanistic studies on mixed ruminal
microbes, carbon, energy, and cell recoveries would be otherwise incomplete without
absolute quantification of reserve carbohydrate synthesis and depletion; reciprocally,
more accurate quantification of varying reserve carbohydrate pool size limits the ability
to interpret or integrate carbon balance measurements, which might be underrepresented
by as much as 15% (105).
Several methods exist for detecting reserve carbohydrate (Figure 3). Glycogen
and trehalose, which are major components of this carbohydrate, can be detected
enzymatically by hydrolyzing these glucans and measuring released glucose (46-48).
Total hexoses can be detected chemically, such as by the anthrone reaction (28, 49).
Determinations (enzymatic or chemical) can be made using cell suspensions or using
reserve polysaccharide isolated using ethanol precipitation or other fractionation (49).
For enzymatic detection and for isolation of reserve polysaccharide, the carbohydrate
must first be extracted either chemically [KOH digestion; (46-47, 50)] or mechanically
[sonication, bead-beating; (47-48)].
Workers have recognized that methods detect disparate values of reserve
carbohydrate (51-52).
However, few have addressed which method, if any, detects
reserve carbohydrate completely and quantitatively. Some authors have suggested that
their method (amyloglucosidase hydrolysis) completely detects glycogen because iodine
staining was lost after hydrolysis (47, 50). However, authors did not determine energy,
carbon, and cell recoveries in these or other studies to evaluate if detection was
quantitative.
36
Objectives were to evaluate methods of detecting reserve carbohydrate, focusing
on those methods based on (i) hydrolysis with amyloglucosidase and (ii) the anthrone
reaction. We not only determined which method would detect the most carbohydrate, but
we determined which would detect more carbohydrate functioning as reserve material.
We also measured carbon, energy, and cell recoveries to evaluate which method would
measure carbohydrate completely
Materials and Methods
Preparation of mixed cultures and sampling. Rumen fluid was collected from
1 of 5 cannulated Jersey cows fed a lactation diet (50% corn silage, 4.5% alfalfa hay,
21% corn wet milling product [Cargill Corn Milling, Dayton, OH], 9.05% ground corn,
4.64% soybean meal, 1.30% Amino Plus [Ag Processing Inc. Hiawatha, KS], 1.30%
soyhulls, 0.38%, fat, 2.01% vitamin and minerals) ad libitum in two equal meals. At 2.5
h after feeding, rumen contents were strained through 4 layers of cheesecloth. The
strained fluid was diluted 1:1 with N-free buffer [Simplex type, pH = 6.8 (3)] and added
to a separatory funnel. All glassware was pre-warmed to 39˚C and pre-gassed with O2free CO2. Plant particles, which rose to the top, were removed by aspiration after 45 min
of incubation at 39˚C. To prepare mixed rumen microbes, particle-free rumen fluid (40
mL) was centrifuged at 10 000 g for 10 min (washing once with Simplex buffer) on a
pre-warmed rotor (JA-17 rotor, J2-21 centrifuge; Beckman, Brea, CA) under anaerobic
conditions. Cells were anaerobically resuspended in Simplex buffer and transferred to a
culture bottle, which was capped with a butyl rubber stopper and incubated at 39˚C.
Cells were dosed with glucose (20 mM final concentration). Cell pellets were harvested
37
by centrifuging 1 mL culture (10 000 g, 10 min, 4°C), washed once in 0.9% NaCl, and
stored at -20°C. Unless otherwise noted, pellets were harvested at intervals to give 3
points prior to dosing glucose, at least 3 points during glucose excess, and at least 3
points after glucose was exhausted. Cell-free supernatant was prepared by combining
supernatant from cell harvesting and washing.
Based on direct counts (110), recovery of prokaryotes during centrifugation was
high [96.8 (6.0 SEM) %, n = 8 total across 2 cows] and not different from 100% (P =
0.610). Direct counts (111) indicated recovery of the protozoa was likewise high [98.0
(4.4 SEM) %, n = 8 total across 2 cows; P = 0.663). Unpaired t-tests were used in this
comparison and all subsequent ones. In additional preliminary experiments, protozoal
counts were 3.92∙105 (6.58∙104 SEM)/mL cell suspension or 1.05∙108 (1.19∙107 SEM)/g
total microbial protein (n = 4 total across 3 cows). Composition of protozoa was 93.71
(0.84 SEM) % genus Entodinium, 2.44 (0.99 SEM) % genus Isotricha, 1.91 (0.53 SEM)
% genus Dasytricha, 1.07 (0.75 SEM) % genus Epidinium, 0.74 (0.22 SEM) % subfamily
Diplodiinae, and 0.13 (0.13 SEM) % genus Ophryoscolex.
Direct cell counts of
prokaryotes were 4.26∙109 (1.05∙109 SEM)/mL cell suspension or 1.12∙1012 (1.75∙1011
SEM)/g total microbial protein (n = 4 total across 3 cows). Protein was determined as
described below.
Carbohydrate analysis.
Total cytoplasmic carbohydrate was prepared from
pellets by bead-beating. Pellets were resuspended to 1 mL with distilled water and boiled
(15 min) to inactivate glycosidic enzymes. Cytoplasmic contents were extracted from
cells using bead-beating with a procedure based on Yu and Morrison (21); contents were
38
transferred to a 2-mL screw-cap tube containing zirconia beads (0.3 g of 0.1 mm and 0.1
g of 0.5 mm), beaten for 3 min at maximum speed on a Mini-Beadbeater™ (BioSpec
Products, Bartlesville, OK, USA), and incubated on ice for 3 min. Beating/ice incubation
was done 15 times (giving 45 min beating in total). After centrifuging to remove cellular
debris (10 000 g, 10 min, 4°C), protein was precipitated from the supernatant with
trichloroacetic acid (TCA; 5 % w/v final concentration) and removed by centrifugation
(10 000 g, 10 min, 4°C). Samples were neutralized with 2 N NaOH, and TCA was
extracted with 4 volumes of diethyl ether. The aqueous (lower) phase was removed and
evaporated in a vacuum concentrator (to remove residual ether) then resuspended in
distilled water.
Reserve polysaccharide was extracted by bead-beating and precipitated with
ethanol. Cytoplasmic contents were extracted with bead-beating and deproteinized with
TCA as before. Reserve polysaccharide was then precipitated with 2.2 volumes of cold
ethanol, removed by centrifugation (15 000 g, 15 min, 4°C), washed twice in 60% v/v
cold ethanol, resuspended in 1 mL hot distilled water, and then stored at -20°C.
Recovery of beef liver glycogen (Sigma-Aldrich Co, St. Louis, MO) similarly extracted
and precipitated was high (98.5 (3.0 SEM) %, n = 5) and not different (P = 0.643) from
100%.
For comparison, extraction was also done by boiling the pellet in 50% KOH
(120 μL) for 3 h (3, 11).
Thin-layer chromatography (TLC) was used for analysis of total cytoplasmic
carbohydrate and isolated reserve polysaccharide (112).
Samples were prepared as
above, or they were hydrolyzed with Aspergillus niger amyloglucosidase (MegaEnzyme,
39
Ireland Ltd., Bray, Co. Wicklow, Ireland) in 0.2 N sodium acetate (pH = 5.2) (46) after
first evaporating the original solvent (distilled water) in a vacuum concentrator. Activity
of amyloglucosidase was 1.2 U/mL final concentration (46), which we confirmed
completely hydrolyzed sources of isolated glycogen under our conditions (see Results).
As specified by the manufacturer, 1 U of amyloglucosidase releases 1 μmol glucose/min
from soluble starch at 40°C and pH 4.5. Salts were removed with mixed bed resin
(TMD-8; Sigma-Aldrich Co, St. Louis, MO). Samples were evaporated and resuspended
in a small volume of water. Carbohydrate was applied to TLC plates (Si 60 F254, 0.2
mm layer thickness; Merck KGaA, Darmstadt, Germany) in 1-μL volumes. Amount
applied was 3-μg glucose equivalents of each standard to standard lanes, and 3-μg
glucose equivalents total to sample lanes.
propanol:distilled water (2:2:1).
Plates were developed with acetone:1-
Spots were detected by dipping the plate in
H2SO4:ethanol (1:9) and charring at 110°C for 10 min. The detection limit of standards
was 25-ng of glucose equivalents.
Reserve polysaccharide (extracted by bead-beating and precipitated with ethanol)
was analyzed for total hexoses by the anthrone method according to Herbert et al. (28).
Other aliquots of isolated polysaccharide was also analyzed for glycogen by measuring
the glucose released by glucose oxidase-peroxidase (113) after hydrolysis with
amyloglucosidase. Beef liver glycogen from Sigma yielded 96.5 (2.2 SEM) % (n = 9)
total hexoses, which did not differ (P = 0.826) from the manufacturer-specified amount
of glucose (96%). This beef liver glycogen yielded 94.0 (0.8 SEM) % glycogen (n = 9),
which was lower (P = 0.045) than the specified amount of glucose.
40
Cell pellets were analyzed directly for glycogen using enzymatic hydrolysis.
Carbohydrate was extracted by bead-beating for 45 min as described above, except the
pellet was suspended and beaten in sodium acetate buffer. This suspension was analyzed
directly for glycogen without removal of cellular debris or purification. In preliminary
experiments, 30 to 45 min of beating were sufficient to give maximal detection of
glycogen in cell pellets (Figure 4). Recovery of beef liver glycogen extracted and
analyzed similarly was 97.4 (1.3 SEM) % (n = 6) and tended (P = 0.104) to be less than
100%. Recovery of beef liver glycogen spiked in a cell pellet was high (95.0 [4.6 SEM]
%, n = 5) and not different (P = 0.357) from 100%. For comparison, polysaccharide was
also extracted by 50% KOH (neutralizing with 6 N HCl before analysis for glycogen).
Pellets were also analyzed for total hexoses using the anthrone method (28).
The absorbance spectrum of isolated reserve polysaccharide in iodine reagent was
determined according to Krisman (114). The absorption spectrum was corrected for
absorbance of (i) isolated reserve polysaccharide prior to addition of iodine reagent and
(ii) iodine reagent alone. Cell suspensions were similarly analyzed.
Additional analyses. Protein was determined using Pierce BCA Protein Assay
Kit (product #23227; Thermo Scientific, Rockford, IL) after hydrolyzing the pellet in
NaOH (0.2 N final concentration, 100°C, 15 min). DNA was determined using the
diphenylamine method, and RNA was analyzed using the orcinol method after extracting
nucleotides with 0.5 N perchloric acid (20 min, 70°C) (28). Dry matter was determined
by transferring cells to aluminum pans and drying at 105°C overnight.
determined using a Folch extraction (115).
41
Lipid was
Supernatant was analyzed for short-chain fatty acids by gas chromatography.
Cell-free supernatant was prepared for analysis by combining 0.85 mL with 0.15 mL 10
N H3PO4 plus 0.1 mL of 2-ethylbutyric acid (10 mM) internal standard.
The gas
chromatograph was a Hewlett-Packard 5890 A with a 6-ft x 2-mm glass column packed
with GP 15% SP-1220/1% H3PO4 stationary phase on 100/120 Chromosorb W AW
support. The carrier gas was N2 (20 mL/min). Temperature of the oven was initially
113°C and increased 2°C/min over 14 min, temperature of injector was 150°C, and
temperature of detector was 180°C. Separate aliquots of supernatant were analyzed for
D-/L-lactic acid with a kit from R-Biopharm (product code 11112821035, Marshall, MI).
Free glucose was analyzed as above, except that N-ethylmalemide was added to prevent
interference posed by cysteine∙HCl on this method (116); 2 mol N-ethylmalemide per
mol cysteine∙HCl was found adequate to prevent interference. (data not shown).
Energy and carbon recovery.
Energy recovery and carbon recovery were
calculated from measured concentrations of reactants (glucose, glycogen, short-chain
fatty acids) and heat production according Chapter 4. Briefly, the increase in energy and
carbon after dosing glucose (20 mM final concentration) was calculated. For recovery of
100%, the observed increase in energy and carbon after dosing the cells with glucose
must equal the energy and carbon in dosed glucose. To determine actual recovery, the
actual increase (averaged over time) was compared to the expected increase for 100%
recovery.
Heat production (W/L) was measured using microcalorimetry (Thermal Hazard
Technology μRC; Piscataway, NJ) according Chapter 4. Briefly, cell suspension was
42
placed in the calorimeter (set to 39.00°C). Water served as the reference. Glucose (1 M)
was injected using a syringe fitted onto the calorimeter. Heat production was integrated
and corrected for protonation by buffer.
Statistics. Data were analyzed using unpaired t-tests or one-way analysis of
variance with PROC MIXED of SAS (v. 9.1.3, SAS Inst. Inc., Cary, NC). Effects were
discrete and fixed, and correction for multiple comparisons was done using Tukey
adjustment. Local regression [LOCFIT package of R; (117)] was used to fit time-series
data to smooth curves.
A local quadratic model with Gaussian kernel and nearest
neighbor bandwidths between 0.4 to 0.7 was chosen.
In figures, original data are
presented alongside the smooth curves, and smooth curves were used for calculations
(e.g., for recoveries of energy and carbon).
Results
Characterization of reserve carbohydrate.
We characterized cytoplasmic
carbohydrate and isolated reserve polysaccharide of mixed rumen microbes using TLC.
The chromatogram of cytoplasmic carbohydrate showed a dark spot corresponding to
glycogen, with faint spots for glucose and maltose (Figure 5).
Hydrolysis with
amyloglucosidase produced one spot corresponding to glucose, supporting that the
original carbohydrate was primarily glycogen, but it also contained glucose and maltose.
Results were similar for isolated reserve polysaccharide, except glycogen alone was
present in the unhydrolyzed sample.
We further studied the isolated reserve polysaccharide to confirm its presumptive
identity of glycogen. Reserve polysaccharide had a ratio of glycogen (glucose detected
43
after hydrolysis with amyloglucosidase) to total hexoses (carbohydrate detected by the
anthrone method) that was not different from 1 (ratio = 0.994 [0.020 SEM], n = 5, P =
0.779, cow 478). Additionally, the isolated reserve polysaccharide formed a complex
with iodine reagent characteristic of glycogen, with an absorbance peak at 490 to 510 nm
(Figure 6).
Detecting reserve carbohydrate. After characterizing the reserve carbohydrate
as predominantly glycogen, a method was sought to detect it quantitatively.
First,
methods were screened to determine which would detect the most carbohydrate. For
cells harvested 30 min after dosing glucose, the anthrone method detected more
carbohydrate than the amyloglucosidase hydrolysis method (Table 4). It detected more
carbohydrate even though amyloglucosidase hydrolysis was preceded by prolonged beadbeating (45 min) or KOH digestion (3 h).
Bead-beating led to detection of more
carbohydrate than KOH digestion. More carbohydrate was detected when the source was
cell suspensions vs. isolated reserve polysaccharide. Two additional experiments showed
similar results (data not shown).
Next, the method that would detect larger changes in carbohydrate was
determined. Cells were harvested at intervals relative to dosing 20 mM glucose, and
carbohydrate was detected by the anthrone or amyloglucosidase hydrolysis methods on
bead-beaten cell suspensions (Figure 7). Compared to the amyloglucosidase hydrolysis
method, the anthrone method always detected a larger increase in carbohydrate
immediately after dosing and always detected a larger decrease subsequently (Figure 7).
In quantitative terms, the increase in reserve carbohydrate after dosing glucose was 24.2
44
(6.1 SEM) % larger for anthrone vs. amyloglucosidase methods and significantly
different from 0% (P = 0.017; n = 5). The decrease was 37.8 (13.5 SEM) % larger for
anthrone vs. amyloglucosidase methods and again significant (P = 0.049; n = 5). The
increase in carbohydrate occurred during glucose excess, and the decrease began around
glucose exhaustion (Figure 8A and data not shown).
Energy and carbon recoveries were determined for one experiment (that showing
the largest relative difference between the anthrone and amyloglucosidase hydrolysis
methods; cow 486). In this experiment, calculations based on the anthrone method
yielded higher average energy recovery (97.5%) than those based on the
amyloglucosidase hydrolysis method (88.9%) (Figure 8E). The anthrone method yielded
higher carbon recovery (100.2%) than the amyloglucosidase hydrolysis method (91.6%)
(Figure 8F).
In a separate set of experiments, the anthrone method also yielded a higher sum of
cell components (reserve carbohydrate, protein, DNA, RNA, lipid) than the
amyloglucosidase hydrolysis method. For cells harvested 30 min after dosing glucose,
the anthrone method yielded sum approaching 100% (Table 5), whereas the sum was
lower for amyloglucosidase hydrolysis on bead-beaten cell suspensions.
Sensitivity of iodine staining. A multi-part experiment was used to determine
how sensitively iodine stains glycogen (to evaluate if loss of iodine staining can indicate
glycogen was completely extracted and hydrolyzed). In the first part of experiment, cell
suspensions were extracted by bead-beating, but they were not subsequently hydrolyzed
with amyloglucosidase (Figure 9). After adding iodine reagent to these non-hydrolyzed
45
cell suspensions, an absorbance peak corresponding to glycogen was observed (c.f.,
Figure 9). This finding confirmed that iodine can detect glycogen in cell suspensions as
it did in isolated reserve polysaccharide. In the second part of the experiment, cell
suspensions were neither extracted nor hydrolyzed with amyloglucosidase. A peak was
present but low, suggesting iodine did not stain glycogen completely because glycogen
extraction was intentionally made incomplete. In the third part of the experiment, cells
suspensions were both extracted and hydrolyzed with amyloglucosidase. No absorbance
peak was detected, and thus no glycogen remained or else any actually remaining was
below the sensitivity of iodine staining. In the fourth part of the experiment, cell
suspensions were not extracted, but they were hydrolyzed with amyloglucosidase. No
absorbance was detectedeven though non-hydrolyzed glycogen was present (subsequent
bead beating and incubation with amyloglucosidase hydrolyzed an additional 0.685 g/L
glycogen). These findings suggested that non-hydrolyzed glycogen could remain in
appreciable quantities and yet iodine staining would not be sensitive enough to detect it.
Therefore, loss of iodine staining could not confirm that glycogen was completely
extracted and hydrolyzed.
Discussion
This study compared two methods of detecting reserve carbohydrate of mixed rumen
microbes in an attempt to identify one that detects this carbohydrate completely. To
guide this comparison between these two methods, the reserve carbohydrate of these
microbes was characterized using TLC, enzymatic hydrolysis, and iodine staining. Our
finding that isolated reserve carbohydrate of mixed rumen microbes was glycogen agrees
46
with work for rumen protozoa (88, 118) as well as pure cultures of rumen bacteria (78,
80, 83) and fungi (30). Our work further establishes that cytoplasmic carbohydrate was
predominantly glycogen, with trace amounts of maltose and glucose. Trehalose is a
common carbohydrate in yeast and other organisms (27-28), but it was neither found with
thin-layer chromatography in the present study nor previously in rumen fungi (30). Our
work did not characterize glycogen associated with the cell wall (47) with TLC,
enzymatic hydrolysis, and iodine staining because this material cannot be isolated from
cell wall material.
After characterizing reserve carbohydrate as predominantly glycogen, we
screened methods based on the anthrone reaction and amyloglucosidase hydrolysis to
determine which method would detect the most carbohydrate.
Results confirmed
findings with yeast that the anthrone method detects more carbohydrate than the
amyloglucosidase hydrolysis method (51) or similar enzymatic methods (52). Results
also confirmed that less carbohydrate was detected after isolating polysaccharide (47),
probably resulting from losses of glycogen associated with the cell wall (47) or glucose
and maltose. Results verified that mechanical extraction (bead-beating) can be more
complete than chemical extraction using KOH digestion (28).
Although previous studies distinguished which methods detected the most
carbohydrate, they did not determine which method detected the most carbohydrate
functioning as a reserve material—central to the aim of this study. A reserve material
should be able to both accumulate during energy excess and subsequently be utilized for
energy during starvation (13). To establish if a method detects carbohydrate functioning
47
as reserve material, one must therefore monitor changes in detected carbohydrate over
energy excess and subsequent starvation periods.
Previous studies that compared
anthrone and amyloglucosidase hydrolysis did so at single time points only (51-52).
These static comparisons cannot establish which method (anthrone, amyloglucosidase
hydrolysis) detects more reserve material.
To establish which method detects more reserve material, this study compared
anthrone and amyloglucosidase hydrolysis over the course of glucose excess and
starvation.
Compared with the most efficient method of amyloglucosidase hydrolysis
(that for bead-beaten whole cell suspensions), the anthrone method detected more
carbohydrate accumulating during glucose excess and more being utilized for energy
after glucose exhaustion. Based on the previously described criterion, the anthrone
method detected more carbohydrate functioning as reserve material.
Larger values might result from the anthrone vs. amyloglucosidase methods partly
because the anthrone method reacts with non-glycogen hexoses in the cell wall and
exopolysaccharide (28, 119). This cross-reactivity cannot explain the large utilization
(decline) in carbohydrate after glucose exhaustion, however, because exopolysaccharide
and cell wall carbohydrates are not mobilized as reserve material (13, 49, 120).
To further compare methods and determine if they detected carbohydrate
completely, recoveries of cell components, energy, and carbon and were compared. In
one experiment, recovery of cell components for the anthrone method approached 100%,
whereas it was lower for the amyloglucosidase hydrolysis method.
In another
experiment, the anthrone method yielded appreciably higher energy and carbon recovery
48
(8.6%) than did the amyloglucosidase hydrolysis method. Further, recoveries for the
anthrone method approached 100% in this experiment. In a larger set of experiments in
which the anthrone method alone was used, recoveries also approached 100% for that
method (99.9 [5.0 SEM] % for carbon, 97.6 [5.2 SEM] % for energy; n = 8; Chapter 4).
These results indicated that the anthrone method detected changes in reserve
carbohydrate completely, thereby suggesting the amyloglucosidase hydrolysis method
did not.
This study’s findings conflict with the frequent claim that methods based on
amyloglucosidase hydrolysis completely detect glycogen for yeast and bacteria (28, 4748, 50). Authors have based this claim on loss of iodine staining after extraction and
hydrolysis (47, 50). However, results from the current study establish that iodine staining
was insensitive; iodine did not stain glycogen when extraction was intentionally made
incomplete (by omitting bead-beating) even though non-hydrolyzed glycogen remained.
That is, loss of iodine staining does not guarantee complete extraction and hydrolysis in
our study and should not confirm complete quantification of glycogen using
amyloglucosidase hydrolysis in those other reports.
With rumen microbes, the anthrone method detected more carbohydrate and
produced larger changes in reserve carbohydrate than did the amyloglucosidase
hydrolysis method. Based on energy, carbon, and cell recoveries, the anthrone method
appeared to accurately detect the change in reserve carbohydrate. The anthrone method
can cross-react with hexoses in cell wall and exopolysaccharide, but this cross-reactivity
cannot explain the large decline detected by this method once the glucose is exhausted
49
because these hexoses are not mobilized as reserve carbohydrate. Methods based on the
anthrone reaction therefore appear suited for quantitative studies of reserve carbohydrate
metabolism.
50
Table 4. Reserve carbohydrate for cells harvested 30 min after dosing 20 mM final
concentration glucose.
Source of
reserve
carbohydrate
Anthrone
Method
Bead-beating/
amyloglucosidase
50% KOH/
amyloglucosidase
SEM P-value
----------------g glucose equivalents/L---------------Cell
5.43a
4.88b
4.59c
0.07 <0.001
suspension
Isolated
ND
3.40a
2.76b
0.16 0.02
Cells were originally collected from rumen fluid from cow 492, washed in N-free buffer,
and assayed to contain 5.08 g protein/L. Data represent 1 experiment with 5 samples per
method and source of carbohydrate, and each sample was analyzed in triplicate.
a,b,c
Means within rows with different superscripts differ at P-value shown
51
Table 5. Sum of cell components for cells harvested 30 min after dosing 20 mM glucose
(final concentration)
Component
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Protein
DNA
RNA
Lipid
Total excluding carbohydrate
[(1) + (2) + (3) + (4)]
Carbohydrateb
Total [(5) + (6)]
Method
Amyloglucosidase1 Anthrone
% of DM
50.7
3.9
14.1
10.8
79.4
SEM P-value
1.1
0.3
1.3
1.2
2.1
NAa
NA
NA
NA
NA
0.8 <0.001
13.4
20.4
1.6 0.001
92.8
99.8
Amyloglucosidase hydrolysis was performed on bead-beaten cell suspensions. Cells
were originally collected from rumen fluid of cow 478 and washed in N-free buffer. Data
represent 1 experiment with 5 samples per component (10 samples for carbohydrate, with
5 for amyloglucosidase and 5 for anthrone). Each sample was analyzed in triplicate.
a
Not applicable
b
Glucose equivalents∙0.9
52
Figure 4. Length of bead-beating and reserve carbohydrate measured from cell
suspensions hydrolyzed with amyloglucosidase. Cells were harvested 30 min after
dosing 20 mM glucose (final concentration). Cells were originally collected from rumen
fluid of cow 478, washed with N-free buffer, and contained 6.01 g protein/L. Data
represent 1 experiment with 5 samples per time point, and each sample analyzed in
triplicate. Values are means ± SEM.
53
Figure 5. Thin-layer chromatgraphy of cytoplasmic carbohydrate and isolated reserve
polysaccharide. Cells were harvested 30 min after dosing 20 mM glucose (final
concentration). Cells were originally collected from rumen fluid of cow 472 and washed
with N-free buffer. Cytoplasmic carbohydrate was extracted by bead-beating cell
suspensions, and reserve polysaccharide was isolated with ethanol precipitation. Three
micrograms of glucose equivalents were applied to each sample lane, and 3-µg glucose
equivalents of each standard were applied to standard lanes. Data represent 1 experiment
with 1 sample per lane. Data from 2 similar experiments are not shown. Lanes 1 and 6,
standards (G1 = glucose, G2 = maltose, G3 = maltotriose, G7 = maltoheptaose, Gly =
glycogen). Lane 2: cytoplasmic carbohydrate. Lane 3: cytoplasmic carbohydrate
hydrolyzed by amyloglucosidase. Lane 4: isolated reserve polysaccharide. Lane 5:
isolated reserve polysaccharide hydrolyzed by amyloglucosidase.
54
Figure 6. Absorbance spectrum of isolated reserve polysaccharide in iodine reagent.
Cells were harvested 30 min after dosing 20 mM glucose (final concentration). Cells
were originally collected from rumen fluid of cow 472 and washed with N-free buffer.
Polysaccharide was extracted by bead-beating and isolated by ethanol precipitation, and
0.25-g glucose equivalents/L were added to iodine reagent. Absorbance was read at 10nm intervals. Data represent 1 experiment with 5 samples per point, and samples were
analyzed in triplicate. Values are means ± SEM; for clarity, error bars are shown only at
20-nm intervals.
55
Figure 7. Reserve carbohydrate for washed cells relative to dosing glucose (20 mM final
concentration). Cells were originally collected from rumen fluid from cows (A) 486, (B)
491, (C) 492, (D) 490, and (E) 492. Cells were washed with N-free buffer and dosed
with glucose at 20 min. Reserve carbohydrate was determined using the anthrone
reaction or amyloglucosidase hydrolysis method on bead-beaten cell suspensions. Cows
are ordered according the relative difference in carbohydrate detected by anthrone vs.
amyloglucosidase hydrolysis methods, with cow 486 (A) showing the largest difference
and cow 472 (E) the smallest. Each panel represents 1 experiment, with 1 sample per
point, and each sample analyzed in triplicate.
56
Figure 7
57
Figure 8. Measures relevant to energy and carbon recovery for washed cells relative to
dosing glucose (20 mM final concentration). Measures included (A) glucose in media,
(B) cellular protein, (C) fermentation products, including acetate (Ac), methane (CH4),
propionate (Pr), butyrate and isobutryate (But + IBut), valerate + isovalerate (Val + IVal),
and lactate (Lac), (D) integrated heat production, (E) increase in energy after dosing
glucose, and (F) increase in carbon after dosing glucose. Reserve carbohydrate was
determined using the anthrone or amyloglucosidase hydrolysis methods and shown in
Figure 7A. For (E) and (F), lines representing 100% recovery are shown. Protein was
not included in calculation of energy or carbon recoveries but is shown for reference.
Data represent 1 experiment with one sample, and samples were analyzed in triplicate.
Heat production was measured at 1-s intervals and integrated as described in the text.
Increase in energy and carbon after dosing glucose was calculated as described in text.
58
Figure 8
59
Figure 9. Absorbance spectra of cell suspensions added to iodine reagent. Cells were
harvested 30 min after dosing 20 mM glucose (final concentration). Cells were originally
collected from rumen fluid of cow 472 and washed with N-free buffer. In different
experiments, bead-beating (BB) and amyloglucosidase hydrolysis (A) were either
performed (+) or omitted (-). Reserve carbohydrate originally in suspensions (measured
by bead-beating and hydrolysis with amyloglucosidase) was 4.17 g glucose
equivalents/L. After bead beating and amyloglucosidase hydrolysis (if any), suspensions
were diluted 10-fold and added to iodine reagent. Absorbance was read at 10-nm
intervals. Values are means ± SEM (n = 5); for clarity, error bars are shown only at 20nm intervals. Data represent 1 experiment with 5 samples per point, and samples were
analyzed in triplicate.
60
Chapter 4: Quantifying the responses of mixed rumen microbes to excess carbohydrate
Abstract: The aim of this study was to determine if mixed microbial community from
the bovine rumen would respond to excess carbohydrate by accumulating reserve
carbohydrate, energy spilling (dissipating excess ATP energy as heat), or both. Mixed
microbes from the rumen were washed with N-free buffer and dosed with glucose. Total
heat production was measured by calorimetry. Energy spilling was calculated as heat
production not accounted by (i) endogenous metabolism (heat production before dosing
glucose) and (ii) synthesis of reserve carbohydrate (heat from synthesis itself and
reactions yielding ATP for it). For cells dosed with 5 mM glucose, synthesis of reserve
carbohydrate and endogenous metabolism accounted for nearly all heat production
(93.7%); no spilling was detected (P = 0.226). For cells dosed with 20 mM glucose,
energy spilling was not detected immediately after dosing, but it became significant (P <
0.05) by approximately 30 min after dosing glucose. Energy spilling accounted for as
much as 38.7% of heat production in one incubation. Nearly all energy (97.9%) and
carbon (99.9%) in glucose were recovered in reserve carbohydrate, fermentation acids,
CO2, CH4, and heat. This full recovery indicates that products were measured completely
and that spilling was not a methodological artifact. These results should aid future
61
research aiming to mechanistically account for variation in energetic efficiency of mixed
microbial communities.
Introduction
When ruminant livestock are fed grain, rumen microbes often encounter large
excesses of carbohydrate (12). Pure cultures of rumen bacteria can respond to these
carbohydrate excesses with different strategies (Figure 1). Some species use excess
energy and carbohydrate to synthesize reserve carbohydrate and other reserve materials
(12, 24). Other species spill energy, whereby they dissipate excess energy (ATP) through
futile or substrate cycles (12, 14). Such cycles may involve flux of ions through the cell
membrane (14) or simultaneous synthesis and degradation of glycogen (19, 121). Other
responses include reducing ATP yield by releasing metabolic intermediates (overflow
metabolites) and shifting to catabolic pathways that yield less ATP (12). Responses are
similar for non-rumen microbes (11, 14, 24, 31).
Agriculturally, energy spilling may be detrimental because of its potential to
reduce growth of rumen microbes—an important source of protein for ruminant livestock
(9). Environmentally, energy spilling may be detrimental because its net products are
heat and fermentation products, the latter of which may include the greenhouse gas
methane (122). Reserve carbohydrate would be less detrimental because it may be
mobilized later for growth (13) or pass from the rumen without being fermented (97),
albeit ATP energy is spent on its synthesis and cannot be recovered. There thus exist
62
agricultural and environmental incentives to study the magnitude of these responses in
rumen microbes.
Although many studies have examined pure cultures of rumen microbes, few have
examined mixed cultures to determine how they respond to excess carbohydrate. Indeed,
responses such as energy spilling have seldom been investigated in any mixed
community (11, 31), despite the suggestion that spilling evolved to confer competitive
advantage in these communities (38). We know of 2 studies that investigated energy
spilling in mixed communities. Chen et al. (23) induced spilling in activated sludge by
adding an exogenous protonophore, but spilling was not demonstrated under conditions
that are physiologically relevant to typical livestock production. Van Kessel and Russell
(20) suggested that rumen bacteria spilled energy when grown under ammonia-N
limitation, but they did not measure reserve carbohydrate. Some energy may have in fact
been directed to reserve carbohydrate synthesis, not spilling—a possibility that the
current study will investigate and account for.
Further, studies have generally examined only one response to excess
carbohydrate at a time, such as by examining species that spill energy but do not
accumulate reserve carbohydrate (14). They have not examined if multiple responses can
occur simultaneously among species in a mixed community.
The aim of this study was to determine how mixed rumen microbes would
respond to excess carbohydrate (glucose). The hypothesis was that rumen microbes
would direct small amounts of glucose to reserve carbohydrate, but larger amounts would
be progressively directed to spilling. To test this hypothesis, the study deploys a new
63
method to quantify energy directed towards reserve carbohydrate synthesis vs. spilling.
To our knowledge, this study is the first to examine responses to excess carbohydrate in a
mixed community and the relative magnitude of these responses.
Materials and Methods
Mixed rumen microbes were prepared from rumen fluid as described in Chapter
3. Briefly, rumen fluid was collected from 1 of 4 Jersey cows fed a lactation diet. At 2.5
h after feeding, rumen contents were strained through 4 layers of cheesecloth, diluted
with Simplex buffer (3), and flocculated to remove plant particles. Particle-free rumen
fluid (40 mL) was centrifuged at 10 000 g for 10 min (washing once with Simplex
buffer) on a pre-warmed rotor (JA-17 rotor, J2-21 centrifuge; Beckman, Brea, CA) under
anaerobic conditions.
Cells were anaerobically resuspended in Simplex buffer,
transferred to a culture bottle, and incubated at 39˚C. Cells were dosed with glucose (20
mM final concentration). Cell pellets were harvested by centrifuging 1 mL culture (10
000 g, 10 min, 4°C), washed once in 0.9% NaCl, and stored at -20°C. Pellets were
harvested at intervals to give 3 incubation points prior to dosing glucose, at least 3 points
during glucose excess, and at least 3 points after glucose was exhausted. Cell-free
supernatant was prepared by combining supernatant from cell harvesting and washing.
Chemical analyses. Supernatant and pellets were analyzed according to Chapter
3. Briefly, supernatant was analyzed for short-chain fatty acids using gas
chromotography, D-/L-lactate using a kit (product code 11112821035; R-Biopharm,
Marshall, MI), and glucose using glucose oxidase-peroxidase. Pellets were analyzed for
reserve carbohydrate using the anthrone method (28), protein using Pierce BCA Protein
64
Assay Kit (product
#23227; Thermo Scientific, Rockford, IL), DNA using the
diphenylamine method, and RNA using the orcinol method.
Reaction properties. Stoichiometry and thermodynamic properties of reactions
were tabulated (Table 6) for further calculations. Thermodynamic symbols and units are
summarized in Appendix 1. Reaction stoichiometry and change in ATP (  r ATP(l) ;
mol/mol) were from Russell (16).
This stoichiometry, along with measured
concentrations of reactants (glucose, reserve carbohydrate, fermentation acids), was used
to infer concentrations of reactants not measured (CH4, CO2, H2O). The stoichiometric
number ( v ' (l ) ; mol/mol) was calculated also for reaction stoichiometry. For  r ATP(l) of
glucose-utilizing reactions, it was assumed that glucose transport required no net ATP.
Thermodynamic properties for reactions included standard transformed enthalpy
of reaction l (  r H ' 0 (l ) ; kJ/mol), standard transformed Gibbs energy of reaction l (
 r G ' 0 (l ) ; kJ/mol), and change in binding of hydrogen atoms in a biochemical reaction l (
 r N H (l ) ; mol/mol). Properties were calculated at temperature (T) = 39˚C, pH = 6.8 (that
of Simplex buffer), and ionic strength (I) = 0.25 M according to Alberty (55).
To calculate properties of reactions, first, thermodynamic properties of chemical
species at standard conditions (T = 25˚C, pH = 0, and I = 0) were compiled from the
literature (Table 7). Second, properties of species were adjusted to T = 39˚C, pH = 6.8,
and I = 0.25 (Table 8). Third, properties of chemical reactants (made up of species) were
calculated (Table 9), yielding standard transformed Gibbs energy of formation of a
reactant i (  f G'i0 ), standard transformed enthalpy of formation of a reactant i (  f H 'i0 ),
65
and number of hydrogen atoms in reactant i (
N H (i )
). Finally, properties for reactions
(Table 6) were calculated according to
 r H ' 0 (l ) 
N'
 vi'  f H 'i0
8 (1)
i 1
 r G ' 0 (l ) 
N'
 vi'  f G'i0
9 (2)
i 1
 r N H (l ) 
N'
 vi'  f N H (i)
10(3)
i 1
For species, correction of properties to T = 39˚C assumed that heat capacity was
constant over temperature. For reactants that were gases (CH4, CO2, H2), the gaseous
state (g) was assumed.
For reserve carbohydrate, carbohydrate was assumed to be
glycogen [glucan with (α1→4) and (α1→6) linkages] with a chain length of 15 residues
[typical of rumen bacteria and protozoa; (79-80, 83, 87-88, 118)].
The reaction with
isobutyrate was considered identical to that for butyrate because properties for
isobutyrate were not found; reactions for isovalerate and valerate were also considered
identical because properties for isovalerate were not found.
dq t
Measurement of heat production. Rate of heat production ( Vdt , W/L) was
measured using isothermal microcalorimetry (Thermal Hazard Technology μRC;
Piscataway, NJ). This calorimeter measures heat production by power compensation.
According to this principle (123), the measurement cell is heated or cooled to compensate
for heat consumed or produced by the sample within. The power needed to maintain
constant temperature within the cell is related to heat production by a calibration
66
constant. A reference cell is present and corrects for thermal perturbations from the
environment.
Cell suspension (1 mL) was added to a 2-mL autosampler vial with a rubber
septum and placed in the measurement cell. Water (1 mL) was placed in the reference
cell. A glass syringe (250 μL) was filled with 1 M glucose and placed in the syringe
tower of the calorimeter. The calorimeter was set to 39.00°C. Heat production was
recorded at intervals of 1 s.
Rate of heat production was corrected for a baseline obtained with water in the
sample and reference cells before and after each experiment. Calibration was done using
an internal electric heater. This calibration was verified by injecting 0.1 N HCl into 0.01
N NaOH (123), which, when corrected for the apparent heat of injecting 0.1 N HCl,
produced 102.2 (1.4 SEM) % of heat expected (n = 6) (124).
Rate of heat production was integrated to give integrated heat production (kJ/L)
qt

V
t
dqt
0 Vdt dt
11(4)
Here and throughout, integration was done using the rectangle method with 1-s intervals.
Integrated heat production was expressed as an enthalpy change in the calorimetric
experiment (kJ/mol) by dividing by change in glucose concentration.
 r H t (cal) 
qt dcglu cos e,t
/
/ 103
V
dt
12(5)
where c glu cos e,t is concentration of glucose (mol/L) at time t. Here and throughout,
differentiation was done using the finite difference method with 1-s intervals.
67
Heat is released when fermentation acids protonate buffer, and quantities above
require correction for this heat release.  r H t (cal) was corrected for this heat release by
13(6)
r Ht  r H t (cal)  r NH r H (Buff )
where  r H t is the enthalpy of reaction at time t (kJ/mol) and  r H (Buff ) = enthalpy of
dissociation of buffer (kJ/mol H+).  r H (Buff ) (Table 6) was measured by injecting 0.01
N HCl into Simplex buffer at 39˚C.
 r N H was calculated from values in Table 6,
weighting according to the relative rates of individual reactions. Subsequently,
q*
dqt
were corrected for heat release from buffer (giving t
Vdt
V
and
qt
and
V
dqt*
) by backVdt
calculation.
Heat production and cellular functions.
account for total integrated heat production (
qt
):
V
Three functions were assumed to
(i) endogenous metabolism, (ii)
synthesis of reserve carbohydrate, and (iii) energy spilling. Following a similar approach
to Cook and Russell (53), we calculated the amount of total heat production accounted by
each of these 3 functions over the incubation.
For the first function, rate of heat production from endogenous metabolism was
assumed to be equal to the rate measured prior to dosing glucose. This rate was assumed
not to remain constant over the incubation, but to decline by 7.3%/h (as for control
experiments where cells were never dosed with glucose; see Results). Rate of heat
production was then integrated.
68
For the second function, rate of heat production from synthesis of reserve
carbohydrate was the absolute value of multiplying (i) rate of accumulation of reserve
carbohydrate (mol glucose equivalents/s) and (ii) molar heat of reserve carbohydrate
synthesis (kJ/mol). Molar heat of reserve carbohydrate synthesis was calculated as
molar heat of reserve carbohydrate synthesis(kJ/mol)   r H '0 (GlcRc) 
14 (7)
 r ATP(GlcRc) 
(
 r H ' 0 (GlcAc)dc'Ac ,t
v'Va (GlcVa)dt
(

v'Ac (GlcAc)dt
 r H ' 0 (GlcVa)dc'Va ,t

 r ATP(GlcAc) dc'Ac ,t
v'Ac (GlcAc)dt
 r ATP (GlcVa)dc'Va ,t
v'Va (GlcVa)dt
where
c i ,t
'
 r H ' 0 (GlcPr) dcPr
,t
v'Pr (GlcPr) dt
'
 r H ' 0 (GlcLa )dcLa
,t
'
vLa
(GlcLa )dt




'
 r ATP (GlcPr) dcPr
,t
v'Pr (GlcPr) dt
'
 r ATP(GlcLa )dcLa
,t
'
vLa
(GlcLa )dt
'
 r H ' 0 (GlcBu) dcBu
,t
'
vBu
(GlcBu) dt

 r H ' 0 (CO 2 CH 4 )dc'CH4 ,t
v'CH4 (CO 2 CH 4 )dt


'
 r ATP(GlcBu) dcBu
,t
'
vBu
(GlcBu) dt
)/

 r ATP(CO 2 CH 4 )dc'CH4 ,t
v'CH4 (CO 2 CH 4 )dt
)
= concentration of reactant i at time t (M), Glc = glucose, Ac = acetate, Pr =
propionate, Bu = butryrate and isobutryate, La = lactate, and Va = valerate and
isovalerate. Abbreviation for reactions is provided in Table 6; for example, GlcAc refers
to glucose fermentation to acetate. Rate of heat production was then integrated.
For the third function, heat production accounted by energy spilling was
calculated as total integrated heat production (
qt
) minus (i) integrated heat production
V
accounted by endogenous metabolism and (ii) integrated heat production accounted by
reserve carbohydrate synthesis.
Energy and carbon recovery.
Energy recovery and carbon recovery were
calculated from concentrations of reactants (glucose, reserve carbohydrate, fermentation
69
acids, CO2, CH4, H2O) and rate of heat production (see below). First, amounts of energy
and carbon in the culture were calculated:
energy (kJ/L) 
N'

i 1
(ci ,t  f H ' 0 ) 
i
qt*
V
carbon (M )  ci,t N C (i)
where
 f H '0i
15(8)
16(9)
is the standard transformed enthalpy of formation of reactant i (kJ/mol;Table
9), N’ = number of reactants, and
N C (i)
= number of carbon atoms in reactant i (Table
9). Next, increases in energy and carbon after dosing glucose were calculated; energy
and carbon after dosing were subtracted from the average of their respective values
before dosing glucose. The amount of energy and carbon in the glucose dose was
energy in glucose dose (kJ/L)  cglucose, dosing   f H '0glucose
17(10)
carbon in glucose dose (mM)  cglucose, dosing N C (glucose)
18(11)
For recovery of 100%, increase in energy and carbon after dosing glucose equals energy
and carbon in dosed glucose. The actual increase (averaged over time) was compared to
the expected increase for 100% recovery.
Sensitivity analysis. Sensitivity analysis (125) was performed to test if energy
spilling changed appreciably after changing the assumed parameters values used in its
calculation. For any calculation for which some parameter values are assumed and
subject to error, a sensitivity analysis investigates how much those errors impact the
results of that calculation (125). In this study, assumed parameter values were those that
concerned (i) endogenous metabolism as an estimate of maintenance functions, (ii) rate
70
of decline in endogenous metabolism, (iii) ATP requirements for glucose transport, (iv)
ATP yield of fermentation reactions, (v) identity of reserve carbohydrate (constituent
monomers and chain length), (vi) physical state of gases for calculation of
thermodynamic properties, and (vii) pH for calculation of thermodynamic properties.
Assumed values were changed one-at-a-time to alternate values, and spilling recalculated. These alternate values and reason for their choice are given in Table 10.
Statistics. Data were analyzed using unpaired t-tests. Local regression [LOCFIT
package of R; (117)] was used to fit time-series data to smooth curves. The model
chosen was a local quadratic with Gaussian kernel and nearest neighbor bandwidths
between 0.4 to 0.7, which reduced noise in the data without oversmoothing (e.g.,
truncating peaks). Original data have been presented alongside the smooth curves in
figures, and smooth curves were used for calculations (e.g., for heat production accounted
by cellular functions and recoveries of energy and carbon).
First-order rates of
exponential decline were calculated for time-series data as the linear decline after logtransformation.
Results
When rumen microbes were washed with N-free buffer and dosed with glucose, they
responded immediately by consuming glucose, accumulating reserve carbohydrate, and
producing heat (Figure 10). From glucose dosing to glucose exhaustion, cells dosed with
20 mM glucose (final concentration) accumulated 356% more reserve carbohydrate than
did cells dosed 5 mM glucose (final concentration) (SEM = 65%; P = 0.002). In the same
71
interval, cells dosed with 20 mM glucose produced 467% more heat than did cells dosed
5 mM glucose (SEM = 69%; P < 0.001).
At the point of glucose exhaustion, reserve carbohydrate and rate of heat
production began to decline (Figure 10). For cells dosed 5 mM glucose, rate of heat
production at glucose exhaustion was slightly higher than that prior to dosing glucose
(133 [12 SEM] % pre-dosing; P = 0.068; n = 4). It declined to pre-dosing values
relatively quickly (Figure 10I).
For cells dosed 20 mM glucose, the rate of heat
production at glucose exhaustion was much higher than that prior to dosing (202 [20
SEM] %; P = 0.014; n = 4), and it persisted above pre-dosing values long after glucose
was exhausted (Figure 10J). Reserve carbohydrate declined at an increasing rate after
glucose exhaustion continued (Figure 10C,D).
Over the duration of individual experiments, no large changes in protein were
obvious (Figure 10E,F). Across experiments (n = 8), an exponential decline was detected
numerically (1.08 [0.93 SEM] %/h, n = 8), but it was not significant (P = 0.283), and
protein averaged 3.82 (0.43 SEM) g/L. No large changes in DNA or RNA were observed
in a preliminary experiment (Figure 11), and these components were not measured
subsequently.
Lactate initially accumulated (Figure 10G,H), and concentrations peaked at 1.17
(0.15 SEM) mM and 3.99 (0.67 SEM) mM for cells dosed 5 and 20 mM glucose (n = 4
for each). Lactate began to decline before glucose was exhausted (Figure 10G,H), and
concentrations by the end of incubation were low: 0.25 (0.11 SEM) mM and 0.74 (0.32)
72
mM for cells dosed 5 and 20 mM glucose (n = 4 for each). Initial pH (that of Simplex
buffer) was 6.8, and final pH was never below 6.65.
For control incubations in which cells were starved for carbon (washed with Nfree buffer and not dosed with glucose), heat production, reserve carbohydrate, and
protein declined (Figure 12).
Across experiments (n = 4), the decline in reserve
carbohydrate was rapid (-12.8 [0.7 SEM] %/h) and significant (P < 0.001). The decline
for heat production was slow (7.3 [1.0 SEM] %/h; P = 0.006). The decline for protein
was even slower (2.0 [0.5 SEM] %/h; P = 0.031).
For the earlier experiments in which cells were dosed with glucose, we calculated
how much heat production was accounted by 3 cellular functions:
endogenous
metabolism, synthesis of reserve carbohydrate, and energy spilling (Figure 13). Figure
14 illustrates these calculations with example data.
Heat production accounted by
endogenous metabolism was calculated (i) from the rate of heat production measured
prior to dosing glucose and (ii) assuming a decline of 7.3%/h (that measured above for
cells not dosed with glucose; c.f., Fig 2C). Heat production accounted by synthesis of
reserve carbohydrate was calculated by multiplying (i) rate of reserve carbohydrate
accumulation by (ii) molar heat of reserve carbohydrate synthesis. Heat production for
energy spilling was that not accounted by endogenous metabolism or reserve
carbohydrate synthesis.
When cells were dosed with 5 mM glucose, endogenous metabolism and synthesis
of reserve carbohydrate accounted for nearly all of the heat production (Figure 13).
Across experiments (n = 4), these 2 functions accounted for 93.7 [4.1 SEM] %) of the
73
heat production when glucose was dosed. The remaining 6.3% of heat production,
accounted by energy spilling, differed from 0 numerically but not statistically (P =
0.226). When cells were dosed with higher concentration of glucose (20 mM),
endogenous metabolism and synthesis of reserve carbohydrate accounted completely for
heat production, but only initially (Figure 13). Across experiments (n = 4), energy
spilling became detectable (P < 0.05) by 49 min from the start of the incubation and
averaged 21.1 [3.8 SEM] % after dosing glucose. It eventually accounted for as much as
38.7% of heat production in one incubation (cow 472; Figure 13).
At glucose exhaustion, spilling was 4.0 [3.0 SEM] % of total heat production for
cells dosed 5 mM glucose and did not differ (P = 0.271) from 0. It was 21.5 [6.2 SEM]
% for cells dosed 20 mM glucose and differed (P = 0.0404) from 0. At the end of
incubations (which were variable length), spilling was 7.0 [3.1 SEM] % (P = 0.107) for
cells dosed with 5 mM glucose and 28.6 [5.3 SEM] % (P = 0.0124) for cells dosed 20
mM glucose.
Molar heat of reserve carbohydrate synthesis averaged -86.62 (0.75 SEM) kJ/mol
across experiments (n = 8). Related quantities were -196.7 (2.3 SEM) kJ/mol utilized
hexose equivalents (6 mol carbon in products) and 4.32 (0.07 SEM) mol ATP/mol hexose
equivalents (n = 8).
In a sensitivity analysis, assumed parameter values for calculating spilling were
successively changed from their current values to alternate ones. After making this
change, the calculated amount of spilling changed little (<5% for all but one case) (Table
10), even as the alternate values often represented the extreme range of possible values.
74
A central aim of this study was comparing energy spilling for cells dosed 5 vs. 20 mM of
glucose. In the sensitivity analysis, the difference between 20 and 5 mM glucose ranged
narrowly between 13.7 to 15.3% for all but one case (Table 10), indicating that calculated
amount of spilling for 5 mM and 20 mM glucose was impacted similarly by changing
parameter values.
Thus, there is little opportunity for compounding of errors from
parameterization of equations used to calculate energy spilling.
Although energy spilling generally changed little in the sensitivity analysis, one
exception occurred when changing the estimate of maintenance functions.
When
changing from the current to alternate assumed parameter value, the calculated amount of
spilling changed by >5% for cells dosed with 5 mM glucose (Table 10). Further, the
difference between 5 and 20 mM glucose was only 10.6% (Table 10). The current
parameter value was that maintenance functions were equal to endogenous metabolism.
This value averaged 33.7 (2.7 SEM) mW/g protein in our study (average across n = 8
experiments in which glucose was dosed).
The alternate parameter value was that
maintenance functions were equal to energy use of rumen bacteria extrapolated to growth
rate = 0 (not measured in our study). The value of energy use of rumen bacteria
extrapolated to growth rate
= 0 was set equal to 28.0 mW/g protein, which was
calculated from Isaacson (126) [2.6∙10-4 mol glucose/cell DM/h reported by Isaacson
(126) x 196.7 kJ/mol hexose reported in this study x 1 g DM/0.507 g protein reported by
in Chapter 3 x 1 h/3600 s x 106 mJ/kJ].
For experiments in which cells were dosed with glucose (n = 8), carbon recovery
(99.9 [5.0 SEM] %) and energy recovery (97.6 [5.2 SEM] %) were not different from
75
100% (P = 0.658 and 0.985, respectively). Components included in the calculation of
recovery included reserve carbohydrate, fermentation acids, CO2, CH4, and heat. DNA,
RNA, and protein were excluded from the calculation because these components did not
change over the duration of experiments where glucose was dosed (see above). Complete
recovery confirmed that appropriate components were included and that these were
measured completely.
Discussion
When limited for N, some pure cultures of microbes respond to excess carbohydrate by
accumulating reserve carbohydrate (127), whereas others spill energy (14). This study
determined whether mixed cultures of rumen microbes would respond by accumulating
reserve carbohydrate, spilling, or both. Ruminal protozoa rapidly convert consumed feed
into reserve carbohydrate and can potentially and profoundly influence growth efficiency
measurements (105). Based on isotopic flux or precursor/product analyses, both the
gram-positive Clostridium cellulolyticum
(128) and gram-negative Fibrobacter
succinogenes (121) changed catabolite usage and end product formation extensively
when substrate availability or source was modified. Even though that strain of C.
cellulolyticum synthesizes little glycogen relative to F. succinogenes, both reports noted
the important regulatory role of glycogen formation and cycling in those primary
cellulolytic bacteria. The current approach is based on a previous study to improve the
quantification of reserve carbohydrate to measure energy and carbon recovery (Chapter
3); however, the current results extend conditions to providing excess glucose for a novel
method to quantify energy spilling in a mixed community of rumen microbes.
76
In Chapter 3, mixed rumen microbes synthesized large amounts of reserve
carbohydrate when washed in N-free buffer and given excess glucose. This reserve
carbohydrate was identified as glycogen from thin-layer chromatography, enzymatic
analysis, and iodine staining.
Changes in reserve carbohydrate were quantitatively
measured by the anthrone method, as suggested by carbon and energy recovery. That
study did not quantify energy spilling; however, its verification of our assay’s
quantitative measurement of reserve carbohydrate supports the current study’s accuracy
of energy spilling calculation.
The current study determined the extent of energy spilling by measuring heat
production by calorimetry. After first measuring heat production, we calculated the
amount of total heat production accounted by (i) synthesis of reserve carbohydrate and
(ii) endogenous metabolism. Finally, unaccounted heat, if any, was attributed to energy
spilling. Heat production accounted by growth was ignored because cells were washed in
N-free buffer and growth was absent (protein, DNA, and RNA did not change). Cook and
Russell (53) used a similar conceptual approach to calculate how much total glucose
consumption by Streptococcus bovis was accounted by maintenance, growth, and
spilling. These authors measured heat production, also, but they expressed final results in
terms of glucose consumption using a constant conversion factor  r H 0 = -88.2 kJ/mol
glucose. S. bovis formed only 1 product (lactate) in that study. In our study, multiple
fermentation products were formed,  r H '0 was therefore not constant, and results could
not be expressed as directly in terms of glucose consumption.
77
Based our accounting of heat production, energy spilling did occur when the
mixed cell cultures were incubated under large excesses of glucose (20 mM). Spilling
did not occur immediately after the cells were dosed with glucose. By approximately 30
min after dosing, however, a significant proportion of heat production could not be
explained by synthesis of reserve carbohydrate and endogenous metabolism, indicating
energy spilling. Some spilling occurred before the point of glucose exhaustion, and it
continued after glucose was exhausted. Carbon and energy recoveries were complete,
and energy spilling was therefore not an artifact of incompletely recovering reserve
carbohydrate, an overflow metabolite, or other product. When cells were dosed with less
glucose (5 mM), energy spilling was not significantly detected at any point in the
incubations.
To calculate the magnitude of spilling, this study required a number of assumed
parameter values (e.g., estimate of maintenance functions).
How much are our
calculations impacted by changes to these values? When these parameter values were
systematically changed in a sensitivity analysis, the impact on our calculations was small
(Table 10). The impact on calculations was also similar for 5 and 20 mM glucose (Table
10). Note that changes investigated in this sensitivity analysis were often extreme, such
as by assuming glucose transport was 100% ion-driven (instead of 0% as originally
assumed).
For rumen bacteria, only some transport is ion-driven at low glucose
concentrations (1 mM), and none is ion-driven at high concentrations (5 mM) (36).
Because changes in the simulation analysis were often extreme, ranges of values in Table
10 probably overestimate the errors in our calculations.
78
Although the sensitivity analysis generally supported our conclusion that our
calculations were robust, spilling was impacted moderately when changing the estimate
for maintenance functions. Estimating energetic requirements for these functions [e.g.,
establishment of ion gradients or turnover of macromolecules; (11, 14)] remains elusive
(31, 129), and multiple approaches exist. Following the definition of Dawes (130), this
study estimated these requirements from endogenous metabolism prior to dosing glucose
(during starvation and absence of growth). Following Pirt (131), others [e.g., (53)] have
estimated requirements by extrapolating energy use to growth rate = 0 (during fed
conditions and growth). This study’s estimate was 20% larger than an earlier estimate for
rumen bacteria where energy use was extrapolated to growth rate = 0 (Table 10).
Consequently, spilling increased moderately when using this alternate estimate for rumen
bacteria instead of our current estimate.
Some studies have suggested that spilling may occur or be induced in mixed
communities. Van Kessel and Russell (20) reduced growth efficiency of mixed rumen
microbes by replacing amino-N with ammonia-N, and they suggested the reduction
resulted from spilling. However, they did not account for energy used in synthesis of
reserve carbohydrate and could have overestimated spilling. When adding the exogenous
protonophore (3,3’,4’5-tetrachlorosalicylanilide; TCS), Chen et al. (23) reduced growth
efficiency of microbes in activated sludge.
Rate of substrate consumption did not
change, and they inferred that TCS addition induced energy spilling. However, spilling
under more physiological conditions (in absence of the protonophore) was not
79
demonstrated.
Our study confirms that spilling can occur in mixed microbial
communities from the rumen and under simple excess of carbohydrate.
Excesses of carbohydrate generated in this study should approach the upper
bound of excesses encountered in the rumen. Whereas we dosed 5 and 20 mM glucose,
concentrations of glucose in the rumen rarely exceed 1 mM for animals fed high-forage
diets or those adapted to grain (36). Still, concentrations of c. 5 mM glucose can be
reached after feeding unadapted animals large quantities of grain (32, 132). The highest
concentration of glucose reported was 18 mM [after feeding glucose (133)]. The highest
concentration of soluble sugars was 69 mM glucose equivalents [after feeding beet pulp;
(134)]. These concentrations pertain to the bulk fluid, and concentrations may be higher
for microenvironments [e.g., around starch particles to which amylolytic bacteria attach;
(36)]. In our study, carbohydrate excess was reinforced by removing N (cells were
washed with N-free buffer). For animals fed grain, N in the rumen is present but low,
creating carbohydrate excess (34). If N is chiefly in the form of ammonia, carbohydrate
excess could be intensified (20) because rumen microbes grow far slower with ammoniaN than amino-N (20, 35). Energy spilling detected here probably falls in the upper
ranges in the rumen encountered for high-sugar diets and microenvironments, especially
when N is in the form of ammonia.
Energy spilling is not restricted to pure laboratory cultures, but it can also occur
in mixed ruminal communities that include bacteria, archaea, protozoa, and fungi. This
finding supports the suggestion that spilling evolved to confer competitive advantage in
environments, such as the rumen, that are periodically exposed to nutrient excesses (38).
80
Rumen microbes respond to carbohydrate predominantly by synthesis of reserve
carbohydrate, without spilling, under small excesses of carbohydrate. Future work will
identify the microbial groups and biochemical mechanisms responsible for the spilling
observed under large excesses. In particular, glycogen cycling (19, 121) should be a
suspected mechanism because reserve carbohydrate was central in the response to excess
carbohydrate.
Appendix 1: Thermodynamic symbols and units
aq = aqueous state
 r ATP(l) = change in ATP for reaction l (mol/mol)
c i ,t
= concentration of reactant i at time t (M)
C pm ( j )
= standard molar heat capacity of species j (J/mol/K)
C ' 0pm ( j )
= standard transformed molar heat capacity of species j (J/mol/K)
I = ionic strength (M)
g = gaseous state
 f G 0j
= standard Gibbs energy of formation of species j (kJ/mol)
 f G ' 0j
= standard transformed Gibbs energy of formation of species j (kJ/mol)
 f G 'i0 = standard transformed Gibbs energy of formation of reactant i (kJ/mol)
 r G ' 0 (l ) = standard transformed Gibbs energy of reaction l (kJ/mol)
 f H 0j
= standard enthalpy of formation of species j (kJ/mol)
81
 f H ' 0j
= standard transformed enthalpy of formation of species j (kJ/mol)
 f H 'i0 = standard transformed enthalpy of formation of reactant i (kJ/mol)
 r H ' 0 (l ) = standard transformed enthalpy of reaction l (kJ/mol)
 r H t enthalpy of reaction at time t (kJ/mol)
 r H (Buff ) = enthalpy of dissociation of buffer (kJ/mol)
 r H t (cal) = enthalpy change in the calorimetric experiment at time t (kJ/mol)
N’ = number of reactants
N C ( j ) = number of carbon atoms in species j (mol/mol)
N C (i)
= number of carbon atoms in reactant i (mol/mol)
N H ( j ) = number of hydrogen atoms in species j (mol/mol)
N H (i)
= number of hydrogen atoms in reactant i (mol/mol)
 r N H (l ) = change in binding of hydrogen atoms in reaction l (mol/mol)
q = heat flow into a system (kJ)
dq t
Vdt = rate of heat production (W/L)
dq t*
Vdt = rate of heat production corrected for protonation of buffer (W/L)
qt
V = integrated heat production (kJ/L)
q t*
V = integrated heat production corrected for protonation of buffer (kJ/L)
82
rj
= mol fraction of species j in a reactant (mol/mol)
T = temperature (K)
V = volume (L)
vi' (l ) = stoichiometric number of reactant i in reaction l (mol/mol)
zj
= charge number of species
83
Table 6. Properties of reactions relevant to glucose use in rumen microbes (T = 39˚C, pH = 6.8, I = 0.25 M unless otherwise
noted)a,b
Reaction
Abbreviation Stoichiometry
 r ATP
v i'
 r G'0
 r H '0
r NH
mol/mol
Fermentation
Glucose to acetate
GlcAc
84
Glucose to propionate
GlcPr
Glucose to butyratec
GlcBu
Glucose to valerated
GlcVa
Glucose to lactate
CO2 to CH4
GlcLa
CO2CH4
Synthesis of reserve
carbohydrate
GlcRc
Dissociation of buffere
Buff
Glucose + 2 H2O = 2 Acetate + 4
H2(g) + 2 CO2(g)
Glucose + 2 H2(g) = 2 Propionate
+ 2 H2O
Glucose = Butryate + 2 H2(g) + 2
CO2(g)
Glucose + H2(g) = Valerate +
CO2(g)
Glucose = 2 Lactate
CO2(g) + 4 H2(g) = CH4(g) + 2
H2O
Glucose + [reserve carbohydrate]n
residues = [reserve carbohydrate]n+1
residues+ H2O
HBuff = Buff-
a
-------kJ/mol-----
2 -231.83
75.33
-1.99
4
2 -367.14 -333.66
-1.99
4
1 -277.09
-62.86
-0.99
3
1 -342.56 -264.31
-0.99
2
2 -202.76 -107.63
1 -125.02 -252.50
-2.00
-4.00
2
1
75.33
0.00
-2
2.79
-1.00
0
NA -231.83
NA
ND
See text for symbols and source of values.
Values for v i' are for major fermentation product (e.g., acetate in GlcAc reaction)
c
Reaction with isobutyrate was considered identical GlcBu because properties for isobutyrate were not found
d
Reaction with isovalerate was considered identical GlcVa because properties for isovalerate were not found
e
Measured under conditions described in text.
b
84
----mol/mol---
Table 7. Thermodynamic properties of species and linkages relevant to glucose use in rumen microbes (T = 25˚C, pH = 0, I =
0)a
Reactant
Species or linkage
Source C  ( j ) Source
NH ( j)
N C ( j)  G 
zj
 H
f
-
j
f
j
pm
Acetate
CH3COO (aq)
-1
3
2
-369.32
-486.01
(135)
86.19
(135)
Butyrate
CH3COOH(aq)
CH2(CH2)2COO-(aq)
0
-1
4
7
2
4
-396.48
-354.18
-485.76
-538.19
(135)
(135)
169.70
133.05
(135)
(135)
CH2(CH2)2COOH(aq)
0
8
4
-381.62
-535.34
(135)
127.61
(135)
CO2(g)
CO2(aq) + H2O(l)
CO3-2(aq)
0
0
-2
0
2
0
1
1
1
-394.36
-623.16
-527.81
-393.5
-699.64
-677.14
(55)
(55)
(55)
37.11
318.29
-55.00
(124)
(136)
(137)
HCO3-(aq)
-1
1
1
-586.77
-691.99
(55)
-302.00
(137)
H2CO3(aq)
0
2
1
-606.33
-694.91
(55)
69.00
Fructose
Galactose
C6H12O6(aq)
C6H12O6(aq)
0
0
12
12
12
12
-915.51 -1259.38
-908.93 -1255.20
(55)
(55)
369.00
319.00
(137138)
(139)
(139)
Glucose
H+
C6H12O6(aq)
H+(aq)
0
1
12
1
6
0
-915.90 -1262.19
0.00
0.00
(55)
(55)
336.30
0.00
(139)
(124)
H2(g)
H2(aq)
H2O
H2(g)
H2(g)
H2O(l)
0
0
0
2
2
2
0
0
0
0.00
0.00
-237.19
0.00
0.00
-285.83
(55)
(55)
(55)
28.82
28.82
75.29
(124)
(124)
(124)
Lactate
CH3CH(OH)COO-(aq)
-1
5
2
-516.72
-686.64
(55)
211.29
(140)
Mannose
C6H12O6(aq)
0
12
12
-910.00 -1258.66
(55)
342.00
(139)
CH4(g)
CH4(g)
0
4
1
CO2(g)
CO2tot
85
85
-50.72
-74.81
(55)
35.31
(124)
CONTINUED
Table 7: CONTINUED
CH4(aq)
Propionate
Reserve
carbohydrate
Valerate
a
b
CH4(aq)
CH3CH2COO-(aq)
0
-1
4
5
1
3
-34.33
-363.09
-89.04
-513.08
(55)
(135)
0.00
110.88
NAb
(135)
CH3CH2COOH(aq)
0
6
3
-390.99
-512.41
(135)
253.13
(135)
Glucose-glucose (α1→4) linkage
0
0
0
-15.65
4.53
39.50
(141)
Glucose-glucose (α1→6) linkage
0
0
0
-7.06
-5.80
0.00
NA
CH3(CH2)3COO-(aq)
-1
9
5
-345.72
-562.25
(141142)
(143144)
(135)
160.25
(135)
CH3(CH2)3COOH(aq)
0
10
5
-373.38
-559.36
(135)
260.66
(135)
See text for symbols and units
Not available and assumed to be 0
86
86
Table 8. Thermodynamic properties of species and linkages relevant to glucose use in
rumen microbes (T = 39˚C unless otherwise noted, pH = 6.8, I = 0.25)a,b
Reactant
Species or linkage
rj
C ' 0pm ( j )  f G ' 0j
 f H ' 0j
Acetate
CH3COO-(aq)
Butyrate
CO2(g)
CO2tot
75.48
-241.09
-485.93 0.995
CH3COOH(aq)
CH2(CH2)2COO-(aq)
148.29
100.93
-227.46
-57.88
-485.64 0.005
-539.71 0.994
CH2(CH2)2COOH(aq)
CO2(g) + H2O(l)
CO2(aq) + H2O(l)
CO3-2(aq)
84.79
37.11
307.59
-33.59
-44.62
-394.43
-537.31
-524.34
-538.07
-392.98
-696.31
-675.65
-302.00
-541.29
-696.22 0.837
HCO3-(aq)
Fructose
Galactose
Glucose
H2(g)
H2CO3(aq)
C6H12O6(aq)
C6H12O6(aq)
C6H12O6(aq)
H2(g)
58.29
304.77
254.77
272.07
18.12
-519.75 -695.07 0.000
-399.30 -1260.51 1.000
-395.31 -1257.32 1.000
-405.06 -1264.25 1.000
82.45
-0.72 1.000
H2(aq)
H2O
Lactate
H2(aq)
H2O(l)
CH3CH(OH)COO-(aq)
-10.71
64.59
189.88
101.09
-152.49
-303.61
Mannose
C6H12O6(aq)
277.77
-390.75 -1260.02 1.000
CH4(g)
CH4(aq)
Propionate
CH4(g)
CH4(aq)
CH3CH2COO-(aq)
13.90
-21.41
89.47
115.32
133.17
-150.84
-76.57 1.000
-91.30 1.000
-513.79 0.993
CH3CH2COOH(aq)
Glucose-glucose (α1→4)
linkage
Glucose-glucose (α1→6)
linkage
CH3(CH2)3COO-(aq)
221.01
39.50
-138.06
-16.62
-512.24 0.007
5.08 NAc
0.00
-7.12
117.42
34.55
-564.51 0.993
CH3(CH2)3COOH(aq)
207.13
47.52
-561.35 0.007
Reserve
carbohydrate
Valerate
a
See Appendix 1 for symbols and units
0
T = 25˚C for C ' pm ( j )
c
Not applicable
b
0.006
1.000
0.161
0.014
87
-5.33 1.000
-285.90 1.000
-685.94 1.000
-5.80
NA
Table 9. Thermodynamic properties of reactants relevant to glucose use in rumen
microbes (T = 39˚C, pH = 6.8, I = 0.25)a
Reactant
 f G 'i0
 f H 'i0
N H (i )
N C (i)
Acetate
Butyrate
-241.11
-57.90
-485.93
-539.70
3.01
7.01
2.00
4.00
CO2(g)
CO2tot
Fructose
Galactose
Glucose
-394.43 -392.98
-541.45 -696.20
-399.30 -1260.51
-395.31 -1257.32
-405.06 -1264.25
0.00
2.00
12.00
12.00
12.00
1.00
1.00
6.00
6.00
6.00
H2(g)
H2(aq)
H2O
Lactate
Mannose
CH4(g)
83.05
-0.72
101.69
-5.33
-152.49 -285.90
-303.61 -685.94
-390.75 -1260.02
116.52
-76.57
2.00
2.00
2.00
5.00
12.00
4.00
0.00
0.00
0.00
3.00
6.00
1.00
CH4(aq)
Propionate
Reserve carbohydrate
Valerate
a
See Appendix 1 for symbols and units
88
134.37
-91.30
4.00
1.00
-150.86
-268.60
34.53
-513.78
-973.94
-564.49
5.01
10.00
9.01
3.00
6.00
5.00
Table 10. Amount of energy spilling following changes in assumed parameter values for spilling calculations
Assumption
Current
Alternate
Note and reference
for alternate
assumption
All
Maintenance functions estimated
by endogenous metabolisma
Endogenous metabolism declines
by 7.3%/h
0 mol ATP/mol glucose
transported
None (no changes)
By energy use of rumen bacteria
extrapolated to growth rate = 0b
Is constant
See text
Energy spilling
5 mM
20 mM Difference
glucose glucose
(A)
(B)
(B) – (A)
% of heat production
6.3
21.1
14.8
14.7
25.4
10.6
89
Hypothetical
4.5
18.3
13.8
0.33 mol ATP/molc
Cost of ion-driven
transport (16)
2.6
16.4
13.7
4 mol ATP/2 mol propionated
2 mol ATP/2 mole
2.5
16.2
13.7
1 mol ATP/mol CH4f
0.3 mol ATP/molg
Yield of acrylate
pathway (16)
Lowest yield in
(145)
Highest yield in
(145)
Hypothetical
3.5
17.9
14.4
8.0
23.1
15.1
7.8
22.8
15.1
Hypothetical
Hypothetical
Lowest
measurement (80)
Highest
measurement (118)
9.0
8.0
6.1
24.3
23.1
20.8
15.3
15.1
14.7
6.4
21.2
14.8
1.5 mol ATP/molh
Reserve polysaccharide has chain
length of 15 residues and is
polymer of glucosei
Polymer of fructosej
Polymer of galactosek
Polymer of mannosel
8 residuesm
25 residuesn
CONTINUED
89
Table 10: CONTINUED
Thermodynamic properties of gas
reactants assume states in
gaseous state (g)o
Thermodynamic properties of
species assume pH = 6.80q
a
b
Aqueous state (aq)p
Vaporization of
aqueous state can
be slow (146)
Lowest pH
measured in current
study
pH = 6.65r
2.4
16.3
13.9
6.3
21.1
14.8
33.7 mW/g cellular protein (SEM = 2.7; n = 8; see text)
28.0 mW/g cellular protein [see text; adapted from (126)]
c  r ATP
for glucose-utilizing reactions decrease by 0.33; see Table 1 for current values and Equation 7 for application.
d  r ATP(GlcPr )
= 4; see Table 1 for current value and Equation 7 for application.
e  r ATP(GlcPr )
= 2; see Equation 7 for application.
f  r ATP(CO 2 CH 4 )
= 1; see Table 1 for current value and Equation 7 for application.
g  r ATP(CO 2 CH 4 )
h  r ATP(CO 2 CH 4 )
90
0
i  f H 'Rc
=
= 0.3; see Equation 7 for application.
= 1.5; see Equation 7 for application.
-973.94; see Table S3 in the supplemental material for current value and Equation 1 for application.
0
j  f H 'Rc
= -970.21; see Equation 1 for application.
0
k  f H 'Rc
= -967.01; see Equation 1 for application.
0
l  f H 'Rc
= -969.71; see Equation 1 for application.
0
m  f H 'Rc
= -974.47; see Equation 1 for application.
0
n  f H 'Rc
= -973.68; see Equation 1 for application.
See current values in Table S3 in the supplemental material; see Equations 1, 2, and 3 for application.
p
See alternate values in Table S3 in the supplemental material; see Equations 1, 2, and 3 for application.
q
See current value of properties in Table S2 in the supplemental material; see text for application.
r
For brevity, alternate values are not shown.
o
90
Figure 10. Response of mixed rumen microbes to glucose dosed at 20 min. (A,C,E,G,I)
5 mM glucose. (B,D,F,H,J) 20 mM glucose. Data are from cow 472 and represent 1
experiment per concentration of glucose; data for 3 other cows (representing 3 additional
experiments per concentration of glucose) had similar responses (data not shown). (A,B)
Glucose in media. (C,D) Reserve carbohydrate. (E,F) Cell protein. (G,H)
Fermentation products, including acetate (Ac), methane (CH4), propionate (Pr), butyrate
and isobutryate (But + IBut), valerate + isovalerate (Val + IVal), and lactate (Lac). (I,J)
Rate of heat production.
Reserve carbohydrate was expressed in mM monomeric
glucose equivalents to be in same units as glucose in media. Each datum point represents
one sample that was analyzed in triplicate. Heat production was measured at 1-s intervals
as described in the text.
91
Figure 10
92
Figure 11. Cell components of mixed rumen microbes in response to 20 mM glucose
dosed at 20 min. (A) Reserve carbohydrate. (B) Protein. (C) DNA. (D) RNA.
Reserve carbohydrate was expressed in g/L (for comparison to other components) and
mM (for comparison to concentration of glucose dosed), both in monomeric glucose
equivalents. Data represent one experiment, and each datum point represents one sample
that was analyzed in triplicate.
93
Figure 12. Response of mixed rumen microbes to starvation. (A) Reserve carbohydrate.
(B) Cell protein. (C) Rate of heat production. Data are from cow 472 and represent one
experiment; data for 3 other cows (representing 3 additional experiments) had similar
responses (data not shown). Reserve carbohydrate was expressed in mM monomeric
glucose equivalents. Each data point represents one sample that was analyzed in
triplicate. Heat production was measured at 1-s intervals as described in the text.
94
Figure 13. Integrated heat production of mixed rumen microbes in response to glucose
dosed at 20 min. (A,C,E,G) 5 mM glucose. (B,D,F,H) 20 mM glucose. Shown are heat
production accounted by endogenous metabolism, synthesis of reserve carbohydrate, and
energy spilling for cows 472 (A,B), 486 (C,D), 490 (E,F), and 492 (G,H). Each panel
represents 1 experiment. Methods of calculation are described in the text and Figure 14.
95
Figure 13
96
Figure 14. Example illustration of the calculations and overall approach used to quantify
energy-spilling from mixed ruminal microbes from cow 472 dosed with 20 mM of
glucose. Panel A (reserve carbohydrate) also appears as Figure 10D. The units of reserve
carbohydrate concentration are expressed in monomeric glucose equivalents to be in
same units as the glucose dose. From these data, the slope corresponding with each 1-s
time increment was plotted in panel B and designated as the numerical derivative of
reserve carbohydrate with respect to time. Fermentation products over time are presented
in Panel C, which also appears as Figure 10H. The molar heat of reserve carbohydrate
synthesis over time (Panel D) was calculated when individual measured values of endproducts from Panel C and the parameter values in Table 6 were inputted into Equation 7
from the text. The rate of heat production accounted by reserve carbohydrate synthesis is
shown in Panel E. At 1-s time intervals, these points were calculated by multiplying
values illustrated in Panel B by absolute values illustrated in Panel D. The integrated heat
production accounted by reserve carbohydrate synthesis (Panel F) was calculated using
the rectangle method for each 1-s time increment. The rates of total heat production after
glucose dosage and from endogenous metabolism are both shown in Panel G (which also
appears as Figure 10J). Total rate of total heat production is output from the calorimeter.
The rate of endogenous heat production was set equal to the average rate of heat
production prior to dosing (when the two curves are converged) but then plotted to
decline by 7.3%/h, which is the average from corresponding experiments from microbial
populations that were not dosed with glucose (see Results). At each 1-s time increment,
these rates were integrated using the rectangle rule. These integrated heat production
values over time are illustrated in Panel H (which also appears as Figure 10B). The
integrated heat production accounted by reserve carbohydrate synthesis (i.e., the area
from Panel F) was stacked on top of endogenous heat production in Panel G. The
estimate of energy spilling was determined by difference of the integrated total heat
production minus the integrated heat production that was accounted by endogenous
metabolism minus the integrated heat production that was accounted by reserve
carbohydrate synthesis.
97
Figure 14
98
Chapter 5: Impact of physical state of gases on calculating Gibbs energy for microbial
reactions
Abstract: Calculating change in Gibbs energy ( G ) is a convenient way to predict
whether a reaction can proceed, but this calculation is impacted greatly by the conditions
specified for it. When calculating G for reactions, textbooks specify gases (such as O2
and H2) in the gaseous, not aqueous, state. However, our analysis indicated microbes use
gases in the aqueous state. Compilation of literature values shows microbial reactions
usually create disequilibrium between aqueous and gaseous concentrations of gases. The
pattern of disequilibrium suggests the aqueous state is used, suggesting that the aqueous
state should be the one specified in calculating G . To determine the error in specifying
the gaseous state instead, we explored how changing state impacted G under
physiological concentrations of gases (transformed G of reaction,  r G ' ). The greater
the disequilibrium between aqueous and gaseous concentrations, the greater  r G ' was
impacted (up to 60.50 kJ/mol) when changing states. Our results suggest that aqueous
gas concentrations need to be measured to accurately estimate  r G ' . By establishing
appropriate calculations for  r G ' , our results should advance understanding of microbial
processes where energetics play a key role.
99
Introduction
Change in Gibbs energy ( G ) is a central property of energetics. It determines whether
a biochemical reaction can proceed, specifies how much energy it makes potentially
available for cell work (147), and can even empirically predict growth yields (148).
Under specified (e.g., physiological) concentrations of reactants, G is called
transformed G of reaction (  r G ' ; kJ/mol) and calculated using the text-book equation
 r G '   r G '°  RT ln
[C ]vc [ D]vd
[ A]va [ B]vb
19 (1)
where  r G'° is standard transformed G of reaction (kJ/mol) and calculated according
to (55); R is the gas constant (8.314∙10-3 kJ∙mol-1∙K-1); T is temperature (K); [A], [B], [C],
and [D] are concentrations of reactants and products (M); and v is stochiometric
coefficient. For reactants or products in the gaseous state, concentrations are replaced by
partial pressures (bar).
As simple Eq. 1 appears, it can be deployed only after making certain
specifications, such as the physical state of reactants and products.  r G'° often changes
when specifying different states (55), as do reactant and product concentrations (see
below).
For gases such as H2 and O2, specification of physical state is particularly
polemic. Textbooks and reviews generally calculate G with gases specified in the
gaseous state (in the headspace) (56, 147, 149-151). Other authorities argue instead gases
be specified in the aqueous state (dissolved in the liquid) (55, 148, 152-153).
100
Despite the disagreement over state for gases, it is not known how much  r G '
changes when specifying the aqueous instead of the gaseous state.
 r G'° is already
known to change (55), but  r G ' is more biologically relevant than  r G'° because the
former applies under specified (e.g., physiological) concentrations of products and
reactants (Eq. 1). More fundamentally, it has not been resolved which state microbes use
and thus which state is correct for calculating  r G ' .
To answer which state (gaseous, aqueous) should be specified, and how important
the specification is for ∆rG’, we examined studies that report concentrations of both states
in microbial environments. We found these concentrations were often in disequilibrium,
causing ∆rG’ to change unpredictably when changing from the gaseous to the aqueous
state. The pattern of disequilibrium suggested microbes use gases in the aqueous state,
indicating ∆rG’ be calculated with the aqueous state specified.
Experimental Procedures
Thermodynamic calculations were done using Eq. 2 and saturation ratios presented in
Results. For uniformity, we assumed throughout T = 25˚C. The principal quantity from
Eq. 2 that we discuss,  r G' (aq) -  r G' (g) , varies little over normal ranges of
environmental and physiological temperature.
We made no assumptions for ionic
strength, pH, or pMg because any impact they have on  r G ' (55) cancels out in
computing the difference  r G' (aq) -  r G' (g) .
101
Results
Theoretical considerations show that  r G ' should not differ by which state is specified, if
gas concentrations are at saturation (gas in headspace and liquid are in equilibrium;
Appendix 1). However,  r G ' should differ when gases are under-saturated (below
expected concentration at equilibrium) or over-saturated (above expected concentration).
Explicitly, the expected difference is
  vc  vd
 r G' (aq)   r G' (g)  RT ln  c va d vb
 a b




20(2)
where  r G' (aq) is  r G' when gases are specified in the aqueous state;  r G' (g) is  r G'
when gases are specified in the gaseous state; and α is the saturation ratio (154). The
saturation ratio is actual concentration of a gas divided by its concentration expected at
saturation by Henry’s law (155) (Appendix 2).
To provide data for Eq. 2 and determine actual differences between  r G' (aq) and
 r G' (g) , we surveyed microbiology and environmental studies for values of α. This
survey showed that concentrations of gases were often undersaturated (α << 1) or
oversaturated (α >> 1) (Table 11). When we applied these values of α to Eq. 2, we found
 r G' (aq) and  r G' (g) differed for several reactions occurring in microbial environments
(Table 12). As expected, the difference between  r G' (aq) and  r G' (g) was small when
gases approached saturation (α ≈ 1), as for glucose oxidation occurring at surface waters
of the ocean. However, the difference was large when gases were vastly undersaturated
(α << 1) or oversaturated (α >> 1), as for glucose oxidation at surface waters of an
102
Amazon floodplain. Differences between  r G' (aq) and  r G' (g) for other cases were
intermediate between these extremes.
Discussion
Calculating ∆rG’ is a convenient way to determine whether microbial reactions can
proceed (147), and it can also be used to predict growth yields of (148). However,
correct specifications, such as physical state of gases, must be chosen to yield correct
values of ∆rG’. Textbooks and reviews typically specify the state of gases as gaseous
(56, 147, 149-151). Even still, some authorities specify the aqueous state (55, 148, 152153) because, they argue, cells live in an aqueous environment (55, 152-153). This
disagreement over physical state leads to 2 questions. First, does ∆ rG’ differ when
changing from the gaseous to aqueous state? Second, if ΔrG’(g) and ΔrG’(aq) do differ,
which values are the correct ones?
Differences between  r G' (aq) and  r G' (g) .
We found that ∆rG’(g) and
∆rG’(aq) indeed can differ. The difference was only small (as little as -0.5 kJ/mol) when
aqueous concentrations approached saturation. Saturation is the condition when gas in
the liquid is in equilibrium with gas in the headspace (Figure 15A). The difference
between ΔrG’(aq) and ΔrG’(g) could be large (as large as 60 kJ/mol) however, when
aqueous gas concentrations were under- or over-saturated—i..e, concentration of aqueous
gas was lower or higher than expected under equilibrium (Figure 15B,C).
Although papers generally assume aqueous gas concentrations are at saturation (α
= 1) [see (146)], this condition is not guaranteed. Our survey of microbial environments
revealed concentrations were vastly under-saturated (α << 1) or over-saturated (α >> 1)
103
for many studies. This result may be expected for environments like sludge-bed and
upflow blanket reactors, which are poorly-mixed and thus have slow transfer of gas from
liquid to headspace (146). Less expected, undersaturation occurred even for well-mixed
laboratory cultures (156-157). With over- and under-saturation so pervasive, so too will
ΔrG’(aq) often differ from ΔrG’(g).
Previous work indicated that  r G'° changed when changing state of reactants or
products (55), but our work with  r G' is more biologically relevant because  r G' applies
under physiological concentrations of products and reactants (Eq. 1). Strictly speaking,
ΔrG’° applies only when pressure = 1 bar, solute concentrations = 1 M, and pH is
specified (e.g., pH = 7) (55).
Evidence microbes use gases in the aqueous state. Having established that
specification of state impacts  r G ' , we tried to answer which is the state that microbes
use—i.e., which state is correct for calculating  r G ' . Examining saturation data in Table
11 show that microbes use the aqueous state. For example, data from Joergensen and
Degn (156) show M. trichosporium OB3b consumed CH4(aq), not CH4(g), during
methane oxidation. In this study, CH4(aq) was initially undersaturated while M.
trichosporium OB3b was oxidizing methane, but CH4(aq) quickly approached saturation
after oxidation stopped (Table 11). This pattern of saturation suggests cells directly
consumed CH4(aq), depressing its concentration below saturation until oxidation stopped
and transfer of CH4(g) into the liquid could replenish its concentration (Figure 16). For
similar reasons, data from Jensen and Cox (157) and others in Table 11 lead to a similar
104
conclusion that the aqueous, not gaseous state, is used.
Because these data show
microbes use gases in the aqueous state, they support using  r G' (aq) and  r G' (g) .
Implications for measuring gas concentrations. Most experimenters measure
concentrations of gases in gaseous state. Despite the added demands of sample extraction
(158-159) or use of special equipment (electrodes or mass spectrometry), measuring
aqueous concentrations may be needed to avoid errors shown in Table 12. Only when α
≈ 1 does  r G' (aq) ≈  r G' (g) and thus the experimenter could use gaseous concentrations
without error, and no simple method exists for predicting α [see Pauss et al. (146)].
Whereas  r G' is usually calculated with gases specified in the gaseous state, our
results justify specifying the aqueous state instead. Microbes use gases in the aqueous
state, indicated by the pattern of under- and over-saturation of gases in microbial
environments. That same under- and oversaturation can cause  r G' (aq) and  r G' (g) to
differ (up to +60 kJ/mol). Aqueous concentrations of gases may be needed to accurately
estimate  r G' . By establishing appropriate calculations for  r G' , our results should
advance understanding of microbial processes in which energetics play a key role.
Appendix 1: Demonstration that  r G' (aq) =  r G' (g) under saturation
∆rG’ can be calculated as
 r G'   v f G' (products)   v f G' (reactants)
21 (3)
where  f G ' is the transformed G of formation (kJ/mol). The phase transition of gas X
is
105
X (g)  X (aq)
22(4)
Applying Eq. 3 to 4 gives ∆rG’ for the phase transition
 r G'   f G'[ X (aq)]   f G'[ X (g)]
23(5)
For reactions at equilibrium (and constant temperature and pressure), ∆rG’ = 0 by
definition. Therefore, Eq. 5 gives us at phase equilibrium (saturation),
 f G'[ X (aq)]   f G'[ X (g)]
24(6)
 f G ' for gases is thus the same whether they are considered to be in the aqueous or
gaseous state. Consequently,  r G ' should be the same if other reactants and products
(non-gases), too, have the same  f G ' (c.f., Eq. 1). That is, following the definition of
∆rG’(aq) and ∆rG’(g),
 r G' (aq )   r G' (g)
25(7)
Appendix 2: Derivation of Eq. 2
For a given reaction,  r G' (aq ) when gas concentrations are at saturation can be defined
as
 r G' (aq ) sat   r G'°  RT ln
[C ] sat c [ D] sat
v
vd
v
v
[ A] sat a [ B] sat b
For the same reaction,  r G' (aq ) when gas concentrations are those under actual
conditions (not necessarily saturation),
106
26(8)
 r G' (aq ) actual   r G'°  RT ln
v
v
v
v
27(9)
[C ]actual c [ D]actual d
[ A]actual a [ B]actual b
Substituting Eq. 8 into 9 gives
 [C ]actualvc [ D]actualvd [C ] sat vc [ D] sat vd
 r G' (aq ) actual   r G' (aq ) sat  RT ln 
/
va
vb
v
v
[
A
]
[
B
]
[ A] sat a [ B] sat b
actual
actual





28(10)
Because ∆rG’ (aq)sat = ∆rG’(g) at phase equilibrium (Appendix 1),
 r G' (aq) actual
 [C ]actualvc [ D]actualvd [C ] sat vc [ D] sat vd
  r G' (g)  RT ln 
/
va
vb
v
v
[ A] sat a [ B] sat b
 [ A]actual [ B]actual




29(11)
The saturation ratio is

30 (12)
[ X ] actual
[ X ] sat
Combining Eq. 11 and 12
  vc  vd
 r G' (aq) actual   r G' (g)  RT ln  va vb
 



Rearranging and dropping the subscript “actual” from ∆rG’ (aq)actual gives Eq. 2.
107
31 (13)
Table 11. Saturation ratios (α) of gases in selected environments of microbes1
Environment
Further description
Gas α
Surface waters, global ocean
O2
M actual/M
at saturation
1.03
a
CO2 0.996
Source
(160)
Surface waters, South Pacific
N2
1.08
(161162)
(163)
Surface waters, North
Atlantic
CH4 1.10
(164)
H2
6.25
Surface waters, Amazon
floodplain
O2
0.44
Culture, fermentative
digester, glucose
(159)
CO2 25.7
CH4 640
N2-sparging
H2
11
(158)
CO2 1.7
No N2-sparging
H2
3
CO2 1.6
Sludge-bed reactor
Upflow blanket reactor
H2
52c
CO2 1.33c
Culture, Anabaena
variabilis, photosynthetic N2
fixationd
During N2 fixation (light)
N2
0.597d
After N2 fixation stopped
(dark)
During CH4 oxidation
N2
1e
40 min after CH4
oxidation stopped
CH4 0.996
Culture, Methylosinus
trichosporium OB3b, CH4
oxidation
a
H2
38b
Culture, fermentativemethanogenic digester,
sucrose-acetate-yeast extract
(146)
CO2 1.41b
CH4 0.294f
(157)
(156)
Average global ΔpCO2 in Fig. 20d of ref. (161)/average atmospheric pCO2 of 360 µatm in ref.
(162).
b
For second steady-state.
c
For first steady-state.
d
Average for 3 steady-state levels during light periods, with saturated concentrations taken as
those during dark
e
See main text of (157)
f
Average for the 7 steady-state levels prior to O2 removal.
108
Table 12. Difference between ∆rG’(aq) and ∆rG’(g) for selected reactions occurring in microbial environments.a
Reactionb
Stoichiometry
Environment
Glucose oxidation
Methane oxidation
Glucose fermentation
to acetate and H2
Glucose + 6 O2 →
6 CO2 + 6 H2O
CH4 + 2 O2 →
CO2 + 2 H2O
Glucose + 2 H2O →
2 Acetate + 4 H2 + 2 CO2
109
Glucose fermentation
to propionate
Glucose + 2 H2 →
2 Propionate + 2 H2O
Surface waters, global ocean
∆rG’(aq)
- ∆rG’(g)c
kJ/mol
-0.50
Surface waters, Amazon floodplain
Surface waters, Amazon floodplain
60.50
-3.90
Culture, fermentative digester, N2-sparging
26.41
Culture, fermentative digester, no N2-sparging
13.22
Culture, fermentative-methanogenic digester, sludge-bed
reactor
Culture, fermentative-methanogenic digester, upflow blanket
reactor
Culture, fermentative digester, N2-sparging
37.77
Culture, fermentative digester, no N2-sparging
-2.72
Culture, fermentative-methanogenic digester, sludge-bed
reactor
Culture, fermentative-methanogenic digester, upflow blanket
reactor
-9.02
a
40.59
-5.94
-9.79
∆rG’(aq) is ∆rG’ when gases are written in the aqueous state, and ∆rG’(g) is ∆rG’ when gases are written in the gaseous state.
Selected reactions are among those documented to occur by sources cited in Table 11 or otherwise reasonably expected to
occur.
c
∆rG’(aq) - ∆rG’(g) calculated using equation 1 in text and α values in Table 11.
b
109
Figure 15. Illustration of aqueous state of hypothetical gas X at saturation, below
saturation, or above saturation. (A) At saturation. Gas in liquid and headspace are fully
equilibrated.
According to Henry’s law, [X(aq)] is proportional to [X(g)].
Proportionality was chosen to make [X(aq)]/[X(g)] = 1. (B) At 5-fold undersaturation.
The ratio [X(aq)]/[X(g)] equals 0.2, and there is net transfer to liquid (to restore
equilibrium ratio of 1). (C) At 5-fold oversaturation. The ratio of [X(aq)]/[X(g)] equals
5, and there is net transfer to headsapce (to restore equilibrium ratio of 1). Arrows show
transfer of [X(aq)] of [X(g)] across the liquid-gas interface, with size of arrow head
indicating rate of transfer.
110
Figure 15
(A)
At saturation
Headspace
[X(aq)]
1
[X(g)]
Liquid
= X(g)
= X(aq)
(B)
5-fold undersaturation
Headspace
[X(aq)]
 0.2
[X(g)]
Liquid
= X(g)
= X(aq)
(C)
5-fold
Gas oversaturation
Headspace
[X(aq)]
5
[X(g)]
Liquid
= X(g)
= X(aq)
111
(A)
During CH4 oxidation
Headspace
Liquid
Bacterium
= CH4(g)
= CH4(aq)
(B)
After CH4 oxidation
Headspace
Liquid
Bacterium
= CH4(g)
= CH4(aq)
Figure 16. Model of CH4(aq) use in Joergensen and Degn (156). (A) During oxidation.
CH4(aq) is being depleted by oxidation, and concentration of CH4(aq) is thus below
saturation. (B) After oxidation. Transfer of CH4(g) to liquid has replenished CH4(aq)
and concentration of CH4(aq) is at saturation. Arrows show consumption of CH4(aq) by
the bacterium and transfer of CH4(aq) of CH4(g) across the liquid-headspace interface.
For better illustration, the depicted ratio of [CH4(aq)]/[CH4(g)] at saturation is 1 and
higher than actual; it should be 0.0285 (156) owing to the low solubility of CH4(aq).
112
Chapter 6: Conclusions
This research elucidated the response of mixed rumen microbes to excess
carbohydrate. Although the responses of laboratory pure cultures have been well-studied,
responses for mixed rumen microbes had not been. This research documented mixed
rumen microbes responded to large excesses of carbohydrate by spilling energy, although
synthesis of reserve carbohydrate predominated under small energy excesses.
The
anthrone method accurately quantified changes in reserve carbohydrate, and it was thus
identified as suitable for quantitative studies of carbohydrate metabolism. Experiments
above required calculation of thermodynamic properties. It was found that calculation of
one property, transformed Gibbs energy of reaction (  r G ' ), was accurate only when the
aqueous concentrations of gases was used.
To elucidate the response of mixed rumen microbes to excess carbohydrate, we
first needed a method to quantitatively detect reserve carbohydrate. However, several
methods of unknown accuracy have historically been used. As documented in Chapter 3,
the anthrone method detected larger changes in cell carbohydrate than did the
amyloglucosidase hydrolysis method. Specifically, the anthrone method detected a larger
increase in cell carbohydrate when glucose was in excess, and a larger decline after
glucose was exhausted. This result indicated that the anthrone method detected more
carbohydrate that functions as a reserve material (such material accumulates during
113
energy excess and is utilized for energy during starvation).
The anthrone method
appeared to accurately detect these changes in reserve carbohydrate because it gave
energy, carbon, and cell recoveries approaching 100%. Methods based on the anthrone
reaction therefore appear suited for quantitative studies of reserve carbohydrate
metabolism.
By combining measurement of reserve carbohydrate with that of heat production,
we quantified the response of mixed rumen microbes to excess carbohydrate (glucose).
As described in Chapter 4, cells dosed with 5 mM glucose did not spill energy; synthesis
of reserve carbohydrate and endogenous metabolism explained nearly all heat production
(93.7%); no spilling was detected. For cells dosed with 20 mM glucose, energy spilling
was eventually detected (by approximately 30 min after dosing glucose) and accounted
for as much as 38.7% of heat production in one experiment. Our results showed that
energy spilling is not restricted to pure laboratory cultures, but it can also occur in the
mixed community of the rumen under large excesses of carbohydrate. These results
establish that spilling may occur in the rumen and thus is relevant to ruminant production
systems.
Chapter 4 required calculation of certain thermodynamic properties. In the course
of these calculations, we found calculation of a related property,  r G ' , was sensitive to
specifying gases in the aqueous vs. gaseous state. In Chapter 5, we investigated which
state should be specified and how much  r G ' would be impacted by wrongly specifying
state. Compilation of literature values showed that microbial reactions create
disequilibrium between aqueous and gaseous concentrations of gases, and the pattern of
114
disequilibrium suggests the aqueous state is used. This suggested the aqueous state
should be the one specified in calculating  r G ' .
To determine the error in specifying
the gaseous state instead, we explored how changing state impacted  r G ' . The greater
the disequilibrium between aqueous and gaseous concentrations, the greater  r G ' was
impacted (up to 60.50 kJ/mol). Because  r G ' was so impacted, our results suggest the
aqueous concentrations must be measured to accurately estimate  r G ' . These results
establish appropriate calculations where  r G ' is used. Because  r G ' was not used in
energy spilling calculations (Chapter 4), these results are of interest independent from
spilling.
In these production systems, directing more energy towards reserve carbohydrate
vs. spilling would improve microbial growth efficiency. Improved growth efficiency
would in turn improve use of feed protein, to environmental and economic benefit. To
reduce spilling, future work should identify the microbial groups and biochemical
mechanisms responsible for the spilling observed under large excesses of carbohydrate.
In particular, glycogen cycling should be a suspected mechanism because reserve
carbohydrate was central in the response to excess carbohydrate. Future work should
also address if excess energy from protein (not only carbohydrate) can be spilled.
115
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Rotz C. 2004. Management to reduce nitrogen losses in animal production. J
Anim Sci 82:E119.
Sheldrick W, Keith Syers J, Lingard J. 2003. Contribution of livestock excreta
to nutrient balances. Nutrient Cycling in Agroecosystems 66:119-131.
St-Pierre NR, Thraen CS. 1999. Animal grouping strategies, sources of
variation, and economic factors affecting nutrient balance on dairy farms. J Anim
Sci 77 Suppl 2:72-83.
Steinfeld H, Wassenaar T. 2007. The role of livestock production in carbon and
nitrogen cycles. Annu Rev Environ Resour 32:271-294.
Galloway JN, Cowling EB, Seitzinger SP, Socolow RH. 2002. Reactive
nitrogen: too much of a good thing? Ambio 31:60-63.
Anonymous. 2010. Ingredient market. Feedstuffs 82:21.
Satter L, Klopfenstein T, Erickson G. 2002. The role of nutrition in reducing
nutrient output from ruminants. J Anim Sci 80:E143-E156.
NRC. 2001. Nutrient requirements of dairy cattle, 7th revised ed. National
Academies Press, Washington, D.C.
Clark JH, Klusmeyer TH, Cameron MR. 1992. Microbial protein synthesis and
flows of nitrogen fractions to the duodenum of dairy cows. J Dairy Sci 75:23042323.
AFRC. 1993. Energy and protein requirements of ruminants. CABI, Wallingford,
UK.
Russell JB, Cook GM. 1995. Energetics of bacterial growth: balance of anabolic
and catabolic reactions. Microbiol Rev 59:48-62.
Russell JB. 1998. Strategies that ruminal bacteria use to handle excess
carbohydrate. J Anim Sci 76:1955-1963.
Wilkinson J. 1959. The problem of energy-storage compounds in bacteria. Exp
Cell Res 7:111-130.
Russell JB. 2007. The energy spilling reactions of bacteria and other organisms. J
Mol Microbiol Biotechnol 13:1-11.
Russell JB. 1986. Heat production by ruminal bacteria in continuous culture and
its relationship to maintenance energy. J Bacteriol 168:694-701.
Russell J. 2002. Rumen microbiology and its role in ruminant nutrition. James B.
Russell, Ithaca, NY.
116
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Russell JB, Rychlik JL. 2001. Factors that alter rumen microbial ecology.
Science 292:1119-1122.
Firkins JL. 1996. Maximizing microbial protein synthesis in the rumen. J Nutr
126:1347S-1354S.
Portais JC, Delort AM. 2002. Carbohydrate cycling in micro-organisms: what
can (13)C-NMR tell us? FEMS Microbiol Rev 26:375-402.
Van Kessel JS, Russell JB. 1996. The effect of amino nitrogen on the energetics
of ruminal bacteria and its impact on energy spilling. J Dairy Sci 79:1237-1243.
Otto R. 1984. Uncoupling of growth and acid production in Streptococcus
cremoris. Arch Microbiol 140:225-230.
Newsholme E, Challiss R, Crabtree B. 1984. Substrate cycles: their role in
improving sensitivity in metabolic control. Trends Biochem Sci 9:277-280.
Chen GH, Mo HK, Saby S, Yip WK, Liu Y. 2000. Minimization of activated
sludge production by chemically stimulated energy spilling. Water Sci Tech
42:189-200.
Preiss J, Romeo T. 1989. Physiology, biochemistry and genetics of bacterial
glycogen synthesis. Adv Microb Physiol 30:183-238.
Williams A, Coleman G. 1992. The rumen protozoa. Springer-Verlag, NY.
Wallace RJ, McKain N, McEwan NR, Miyagawa E, Chaudhary LC, King
TP, Walker ND, Apajalahti JH, Newbold CJ. 2003. Eubacterium
pyruvativorans sp. nov., a novel non-saccharolytic anaerobe from the rumen that
ferments pyruvate and amino acids, forms caproate and utilizes acetate and
propionate. Int J Syst Evol Microbiol 53:965-970.
Lillie SH, Pringle JR. 1980. Reserve carbohydrate metabolism in Saccharomyces
cerevisiae: responses to nutrient limitation. J Bact 143:1384-1394.
Herbert D, Phipps P, Strange R. 1971. Chemical analysis of microbial cells.
Methods Microbiol 5B:209-344.
Arguelles JC. 2000. Physiological roles of trehalose in bacteria and yeasts: a
comparative analysis. Arch Microbiol 174:217-224.
Phillips MW, Gordon GL. 1989. Growth characteristics on cellobiose of three
different anaerobic fungi isolated from the ovine rumen. Appl Environ Microbiol
55:1695-1702.
Tempest DW, Neijssel OM. 1984. The status of YATP and maintenance energy as
biologically interpretable phenomena. Annu Rev Microbiol 38:459-486.
Mackie R, Gilchrist F, Robberts A, Hannah P, Schwartz H. 1978.
Microbiological and chemical changes in the rumen during the stepwise
adaptation of sheep to high concentrate diets. J Agric Sci 90:241-254.
Ryan RK. 1964. Concentrations of Glucose and Low-Molecular-Weight Acids in
the Rumen of Sheep Changed Gradually from a Hay to a Hay-Plus-Grain Diet.
Am J Vet Res 25:653-659.
NRC. 2000. Nutrient requirements of beef cattle, 7th revised ed. National
Academies Press, Washington, D.C.
Argyle JL, Baldwin RL. 1989. Effects of amino acids and peptides on rumen
microbial growth yields. J Dairy Sci 72:2017-2027.
117
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
Kajikawa H, Amari M, Masaki S. 1997. Glucose transport by mixed ruminal
bacteria from a cow. Appl Environ Microbiol 63:1847-1851.
Van Kessel J, Russell J. 1997. The endogenous polysaccharide utilization rate of
mixed ruminal bacteria and the effect of energy starvation on ruminal
fermentation rates. Journal of Dairy Science 80:2442-2448.
Tempest D. 1978. The biochemical significance of microbial growth yields: a
reassessment. Trends Biochem Sci 3:180-184.
Russell JB. 1993. Glucose toxicity in Prevotella ruminicola: methylglyoxal
accumulation and its effect on membrane physiology. Appl Environ Microbiol
59:2844-2850.
Preiss J. 1984. Bacterial glycogen synthesis and its regulation. Annu Rev
Microbiol 38:419-458.
Belanger AE, Hatfull GF. 1999. Exponential-phase glycogen recycling is
essential for growth of Mycobacterium smegmatis. J Bacteriol 181:6670-6678.
Gaudet G, Forano E, Dauphin G, Delort AM. 1992. Futile cycling of glycogen
in Fibrobacter succinogenes as shown by in situ 1H-NMR and 13C-NMR
investigation. Eur J Biochem 207:155-162.
Wilson WA, Roach PJ, Montero M, Baroja-Fernandez E, Munoz FJ,
Eydallin G, Viale AM, Pozueta-Romero J. 2010. Regulation of glycogen
metabolism in yeast and bacteria. FEMS Microbiol Rev 34:952-985.
Alonso-Casajus N, Dauvillee D, Viale AM, Munoz FJ, Baroja-Fernandez E,
Moran-Zorzano MT, Eydallin G, Ball S, Pozueta-Romero J. 2006. Glycogen
phosphorylase, the product of the glgP Gene, catalyzes glycogen breakdown by
removing glucose units from the nonreducing ends in Escherichia coli. J Bacteriol
188:5266-5272.
Parrou JL, Teste MA, François J. 1997. Effects of various types of stress on the
metabolism of reserve carbohydrates in Saccharomyces cerevisiae: genetic
evidence for a stress-induced recycling of glycogen and trehalose. Microbiology
143:1891-1900.
Parrou JL, Francois J. 1997. A simplified procedure for a rapid and reliable
assay of both glycogen and trehalose in whole yeast cells. Anal Biochem
248:186-188.
Gunja-Smith Z, Patil NB, Smith EE. 1977. Two pools of glycogen in
Saccharomyces. J Bacteriol 130:818-825.
Schulze U, Larsen ME, Villadsen J. 1995. Determination of intracellular
trehalose and glycogen in Saccharomyces cerevisiae. Anal Biochem 228:143-149.
Trevelyan W, Harrison J. 1956. Studies on yeast metabolism. 7. Yeast
carbohydrate fractions. Separation from nucleic acid, analysis, and behaviour
during anaerobic fermentation. Biochem J 63:23.
Becker JU. 1978. A method for glycogen determination in whole yeast cells.
Anal Biochem 86:56-64.
Kaeppeli O, Aeschbach H, Schneider H, Fiechter A. 1975. A comparative
study of carbon energy reserve metabolism of C. tropicalis growing on glucose
and on hydrocarbons. Appl Microbiol Biotech 1:199-211.
118
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
Rothman LB, Cabib E. 1969. Regulation of glycogen synthesis in the intact
yeast cell. Biochemistry 8:3332-3341.
Cook GM, Russell JB. 1994. Energy-spilling reactions of Streptococcus bovis
and resistance of its membrane to proton conductance. Appl Environ Microbiol
60:1942-1948.
Russell JB, Baldwin R. 1979. Comparison of maintenance energy expenditures
and growth yields among several rumen bacteria grown on continuous culture.
Appl Environ Microbiol 37:537-543.
Alberty RA. 2003. Thermodynamics of biochemical reactions. John Wiley &
Sons, Hoboken, NJ, USA.
Thauer RK, Jungermann K, Decker K. 1977. Energy conservation in
chemotrophic anaerobic bacteria. Bacteriol Rev 41:100-180.
Hackmann TJ, Spain JN. 2010. A mechanistic model for predicting intake of
forage diets by ruminants. J Anim Sci 88:1108-1124.
Baldwin RL. 1995. Modeling ruminant digestion and metabolism. Springer.
Russell J, O'connor J, Fox D, Van Soest P, Sniffen C. 1992. A net carbohydrate
and protein system for evaluating cattle diets: I. Ruminal fermentation. Journal of
Animal Science 70:3551-3561.
Dijkstra J, Neal HD, Beever DE, France J. 1992. Simulation of nutrient
digestion, absorption and outflow in the rumen: model description. J Nutr
122:2239-2256.
Dijkstra J, France J, Davies DR. 1998. Different mathematical approaches to
estimating microbial protein supply in ruminants. J Dairy Sci 81:3370-3384.
Buurman ET, Teixeira de Mattos MJ, Neijssel OM. 1991. Futile cycling of
ammonium ions via the high affinity potassium uptake system (Kdp) of
Escherichia coli. Arch Microbiol 155:391-395.
Mulder MM, Teixeira de Mattos MJ, Postma PW, van Dam K. 1986.
Energetic consequences of multiple K+ uptake systems in Escherichia coli.
Biochim Biophys Acta 851:223-228.
Neijssel OM, Tempest DW. 1976. Bioenergetic aspects of aerobic growth of
Klebsiella aerogenes NCTC 418 in carbon-limited and carbon-sufficient
chemostat culture. Arch Microbiol 107:215-221.
Teixeira de Mattos MJ, Tempest DW. 1983. Metabolic and energetic aspects of
the growth of Klebsiella aerogenes NCTC 418 on glucose in anaerobic chemostat
culture. Arch Microbiol 134:80-85.
Russell JB, Strobel HJ. 1990. ATPase-dependent energy spilling by the ruminal
bacterium, Streptococcus bovis. Arch Microbiol 153:378-383.
Bond DR, Russell JB. 1996. A role for fructose 1,6-diphosphate in the ATPasemediated energy-spilling reaction of Streptococcus bovis. Appl Environ Microbiol
62:2095-2099.
Bond DR, Russell JB. 1998. Relationship between intracellular phosphate,
proton motive force, and rate of nongrowth energy dissipation (energy spilling) in
Streptococcus bovis JB1. Appl Environ Microbiol 64:976-981.
119
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
Prins RA, Van Hoven W. 1977. Carbohydrate fermentation by the rumen ciliate
Isotricha prostoma. Protistologica 13:549-556.
Van Hoven W, Prins RA. 1977. Carbohydrate fermentation by the rumen ciliate
Dasytricha ruminantium. Protistologica 13:599-606.
Van Urk H, Mark PR, Scheffers WA, Van Dijken JP. 2004. Metabolic
responses of Saccharomyces cerevisiae CBS 8066 and Candida utilis CBS 621
upon transition from glucose limitation to glucose excess. Yeast 4:283-291.
Larsson C, von Stockar U, Marison I, Gustafsson L. 1993. Growth and
metabolism of Saccharomyces cerevisiae in chemostat cultures under carbon-,
nitrogen-, or carbon- and nitrogen-limiting conditions. J Bacteriol 175:4809-4816.
Hottiger T, Schmutz P, Wiemken A. 1987. Heat-induced accumulation and
futile cycling of trehalose in Saccharomyces cerevisiae. J Bacteriol 169:55185522.
Morgan RM, Pihl TD, Nolling J, Reeve JN. 1997. Hydrogen regulation of
growth, growth yields, and methane gene transcription in Methanobacterium
thermoautotrophicum deltaH. J Bacteriol 179:889-898.
Schönheit P, Moll J, Thauer RK. 1980. Growth parameters (Ks, μmax, Ys) of
Methanobacterium thermoautotrophicum. Arch Microbiol 127:59-65.
Doetsch RN, Howard BH, Mann SO, Oxford AE. 1957. Physiological factors
in the production of an iodophilic polysaccharide from pentose by a sheep rumen
bacterium. J Gen Microbiol 16:157-168.
Stewart CS, Paniagua C, Dinsdale D, Cheng KJ, Garrow SH. 1981. Selective
isolation and characteristics of Bacteriodes succinogenes from the rumen of a
cow. Appl Environ Microbiol 41:504-510.
Cheng KJ, Hironaka R, Roberts D, Costerton J. 1973. Cytoplasmic glycogen
inclusions in cells of anaerobic gram-negative rumen bacteria. Can J Microbiol
19:1501-1506.
Brown RG, Lindberg B, Cheng KJ. 1975. Characterization of a reserve glucan
from Megasphaera elsdenii. Can J Microbiol 21:1657-1659.
Lou J, Dawson KA, Strobel HJ. 1997. Glycogen formation by the ruminal
bacterium Prevotella ruminicola. Appl Environ Microbiol 63:1483-1488.
Hungate RE. 1963. Polysaccharide storage and growth efficiency in
Ruminococcus Albus. J Bacteriol 86:848-854.
Cheng KJ, Brown RG, Costerton JW. 1977. Characterization of a cytoplasmic
reserve glucan from Ruminococcus albus. Appl Environ Microbiol 33:718-724.
Wallace RJ. 1980. Cytoplasmic reserve polysaccharide of Selenomonas
ruminantium. Appl Environ Microbiol 39:630-634.
Wakita M, Hoshino S. 1980. Physicochemical properties of a reserve
polysaccharide from sheep rumen ciliates genus Entodinium. Comp Biochem
Physiol B Comp Biochem 65:571-574.
Hobson P, Mann S. 1955. Some factors affecting the formation of iodophilic
poly-saccharide in group D streptococci from the rumen. J Gen Microbiol 13:420435.
120
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
Kamio Y, Terawaki Y, Nakajima T, Matsuda K. 1981. Structure of glycogen
produced by Selenomonas ruminantium. Agric Biol Chem 45.
Eadie J, Manners D, Stark J. 1963. The molecular structure of a reserve
polysaccharide from Entodinium caudatum. Biochem. J 89:91.
Forsyth G, Hirst E. 1953. Protozoal polysaccharides. Structure of the
polysaccharide produced by the holotrich ciliates present in sheep's rumen. J
Chem Soc:2132-2135.
Chen G, Russell JB. 1989. More monensin-sensitive, ammonia-producing
bacteria from the rumen. Appl Environ Microbiol 55:1052-1057.
Wells J, Russell JB. 1994. The endogenous metabolism of Fibrobacter
succinogenes and its relationship to cellobiose transport, viability and cellulose
digestion. Appl Microbiol Biotechnol 41:471-476.
Mink RW, Hespell RB. 1981. Survival of Megasphaera elsdenii during
starvation. Curr Microbiol 5:51-56.
Chen GJ, Russell JB. 1988. Fermentation of peptides and amino acids by a
monensin-sensitive ruminal Peptostreptococcus. Appl Environ Microbiol
54:2742-2749.
Howlett MR, Mountfort DO, Turner KW, Roberton AM. 1976. Metabolism
and growth yields in Bacteroides ruminicola strain B14. Appl Environ Microbiol
32:274-283.
Wachenheim DE, Hespell RB. 1985. Responses of Ruminococcus flavefaciens, a
ruminal cellulolytic species, to nutrient starvation. Appl Environ Microbiol
50:1361-1367.
Mink RW, Patterson JA, Hespell RB. 1982. Changes in viability, cell
composition, and enzyme levels during starvation of continuously cultured
(ammonia-limited) Selenomonas ruminantium. Appl Environ Microbiol 44:913922.
Obispo NE, Dehority BA. 1999. Feasibility of using total purines as a marker for
ruminal bacteria. J Anim Sci 77:3084-3095.
McAllan AB, Smith RH. 1974. Carbohydrate metabolism in the ruminant.
Bacterial carbohydrates formed in the rumen and their contribution to digesta
entering the duodenum. Br J Nutr 31:77-88.
McAllan AB, Smith RH. 1976. Effect of dietary nitrogen source on carbohydrate
metabolism in the rumen of the young steer. Br J Nutr 36:511-522.
Smith RH, McAllan AB. 1974. Some factors influencing the chemical
composition of mixed rumen bacteria. Br J Nutr 31:27-34.
McAllan A, Smith R. 1977. Some effects of variation in carbohydrate and
nitrogen intakes on the chemical composition of mixed rumen bacteria from
young steers. British Journal of Nutrition 37:55-65.
Merry RJ, McAllan AB. 1983. A comparison of the chemical composition of
mixed bacteria harvested from the liquid and solid fraction of rumen digesta. Br J
Nutr 50:701-709.
Hoogenraad NJ, Hird FJ. 1970. The chemical composition of rumen bacteria
and cell walls from rumen bacteria. Br J Nutr 24:119-127.
121
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
Leedle JA, Bryant MP, Hespell RB. 1982. Diurnal variations in bacterial
numbers and fluid parameters in ruminal contents of animals fed low- or highforage diets. Appl Environ Microbiol 44:402-412.
Williams A, Harfoot C. 1976. Factors affecting the uptake and metabolism of
soluble carbohydrates by the rumen ciliate Dasytricha ruminantium isolated from
ovine rumen contents by filtration. J Gen Microbiol 96:125.
Hall MB. 2011. Isotrichid protozoa influence conversion of glucose to glycogen
and other microbial products. J Dairy Sci 94:4589-4602.
Czerkawski JW. 2006. Chemical composition of microbial matter in the rumen.
Journal of the Science of Food and Agriculture 27:621-632.
Volden H, Mydland LT, Harstad OM. 1999. Chemical composition of
protozoal and bacterial fractions isolated from ruminal contents of dairy cows fed
diets differing in nitrogen supplementation. Acta Agr Scand A Anim Sci 49:235244.
Russell J, Wallace R. 1997. Energy-yielding and energy-consuming reactions, p.
247-282. In P Hobson and C Stewart (ed), The rumen microbial ecosystem, 2nd
ed. Blackie Academic & Professional, New York, NY.
Firkins JL, Hristov AN, Hall MB, Varga GA, St-Pierre NR. 2006. Integration
of ruminal metabolism in dairy cattle. J Dairy Sci 89 Suppl 1:E31-51.
Holdeman L, Moore W (ed.). 1972. Anaerobe laboratory manual, 2nd ed.
Virginia Polytechnic Institute and State University, Blacksburg, VA.
Dehority BA. 1993. Laboratory manual for classification and morphology of
rumen ciliate protozoa. CRC Press, Boca Raton, FL.
Ezeji TC. 2001. Production, purification and characterization of thermostable
amylolytic enzymes from the newly isolated Bacillus thermodenitrificans
HRO10. Ph.D. thesis. University of Rostock, Rostock, DE.
Karkalas J. 1985. An improved enzymic method for the determination of native
and modified starch. J Sci Food Agric 36:1019-1027.
Krisman CR. 1962. A method for the colorimetric estimation of glycogen with
iodine. Anal Biochem 4:17-23.
Folch J, Lees M, Sloane Stanley GH. 1957. A simple method for the isolation
and purification of total lipides from animal tissues. J Biol Chem 226:497-509.
Haugaard N, Cutler J, Ruggieri MR. 1981. Use of N-ethylmaleimide to prevent
interference by sulfhydryl reagents with the glucose oxidase assay for glucose.
Anal Bioch 116:341-343.
Loader C. 1999. Local regression and likelihood. Springer Verlag, New York,
NY.
Wakita M, Hoshino S. 1980. Physicochemical properties of a reserve
polysaccharide from sheep rumen ciliates genus Entodinium. Comp Biochem
Physiol B 65:571-574.
Sutherland IW. 1990. Biotechnology of microbial exopolysaccharides, vol. 9.
Cambridge University Press, Cambridge, UK.
122
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
Souza AM, Sutherland I. 1994. Exopolysaccharide and storage polymer
production in Enterobacter aerogenes type 8 strains. J Appl Microbiol 76:463468.
Forano E, Delort AM, Matulova M. 2008. Carbohydrate metabolism in
Fibrobacter succinogenes: What NMR tells us. Microb Ecol Health Dis 20:94102.
Martin C, Morgavi DP, Doreau M. 2010. Methane mitigation in ruminants:
from microbe to the farm scale. Animal 4:351-365.
Wadsö I, Goldberg RN. 2001. Standards in isothermal microcalorimetry. Pure
Appl Chem 73:1625-1639.
Wagman DD, Evans WH, Parker VB, Schumm RH, Halow I, Bailey SM,
Churney KL, Nuttall RL. 1982. The NBS tables of chemical thermodynamic
properties. selected values for inorganic and C1 and C2 organic substances in SI
units. J Phys Chem Ref Data 11 Suppl 2.
Pannell DJ. 1997. Sensitivity analysis of normative economic models: theoretical
framework and practical strategies. Agr Econ 16:139-152.
Isaacson HR, Hinds FC, Bryant MP, Owens FN. 1975. Efficiency of energy
utilization by mixed rumen bacteria in continuous culture. J Dairy Sci 58:16451659.
Preiss J. 1989. Chemistry and metabolism of intracellular reserves, p. 189-258. In
JS Poindexter and ER Leadbetter (ed), Bacteria in nature, vol. 3. Plenum Press,
New York, NY.
Desvaux M. 2006. Unravelling carbon metabolism in anaerobic cellulolytic
bacteria. Biotechnol Prog 22:1229-1238.
van Bodegom P. 2007. Microbial maintenance: a critical review on its
quantification. Microb Ecol 53:513-523.
Dawes E. 1985. Starvation, survival and energy reserves, p. 43-79. In M Fletcher
and G Floodgate (ed), Bacteria in their natural environments. Society for General
Microbiology, New York, NY.
Pirt SJ. 1965. The maintenance energy of bacteria in growing cultures. Proc R
Soc Lond B Biol Sci 163:224-231.
Ryan RK. 1964. Concentrations of glucose and low-molecular-weight acids in
the rumen of sheep changed gradually from a hay to a hay-plus-grain diet. Am J
Vet Res 25:653-659.
Piwonka EJ, Firkins JL, Hull BL. 1994. Digestion in the rumen and total tract
of forage-based diets with starch or dextrose supplements fed to Holstein heifers.
J Dairy Sci 77:1570-1579.
Clapperton JL, Czerkawski JW. 1969. Methane production and soluble
carbohydrates in the rumen of sheep in relation to the time of feeding and the
effects of short-term intraruminal infusions of unsaturated fatty acids. Br J Nutr
23:813-826.
Shock EL. 1995. Organic acids in hydrothermal solutions: standard molal
thermodynamic properties of carboxylic acids and estimates of dissociation
constants at high temperatures and pressures. Am J Sci 295:496-580.
123
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
Barbero JA, Hepler LG, McCurdy KG, Tremaine PR. 1983. Thermodynamics
of aqueous carbon dioxide and sulfur dioxide: heat capacities, volumes, and the
temperature dependence of ionization. Can J Chem 61:2509-2519.
Larson JW, Zeeb KG, Hepler LG. 1982. Heat capacities and volumes of
dissociation of phosphoric acid (1st, 2nd, and 3rd), bicarbonate ion, and bisulfate
ion in aqueous solution. Can J Chem 60:2141-2150.
Goldberg RN, Kishore N, Lennen RM. 2002. Thermodynamic quantities for the
ionization reactions of buffers. J Phys Chem Ref Data 31:231-370.
Goldberg RN, Tewari YB. 1989. Thermodynamic and transport properties of
carbohydrates and their monophosphates: the pentoses and hexoses. J Phys Chem
Ref Data 18:809.
Wilhoit RC. 1969. Thermodynamic properties of biochemical substances, p.
305–317. In HD Brown (ed), Biochemical microcalorimetry. Academic Press,
New York, NY, USA.
Goldberg RN, Bell D, Tewari YB, McLaughlin MA. 1991. Thermodynamics of
hydrolysis of oligosaccharides. Biophys Chem 40:69-76.
Tewari YB, Lang BE, Decker SR, Goldberg RN. 2008. Thermodynamics of the
hydrolysis reactions of 1, 4-β-D-xylobiose, 1, 4-β-D-xylotriose, D-cellobiose, and
D-maltose. J Chem Thermodynamics 40:1517-1526.
Tewari YB, Goldberg RN. 1989. Thermodynamics of hydrolysis of
disaccharides. Cellobiose, gentiobiose, isomaltose, and maltose. J Biol Chem
264:3966-3971.
Tewari YB, Goldberg RN. 1991. Thermodynamics of hydrolysis of
disaccharides. Lactulose, alpha-D-melibiose, palatinose, D-trehalose, D-turanose
and 3-o-beta-D-galactopyranosyl-D-arabinose. Biophys Chem 40:59-67.
Thauer RK, Kaster AK, Seedorf H, Buckel W, Hedderich R. 2008.
Methanogenic archaea: ecologically relevant differences in energy conservation.
Nat Rev Microbiol 6:579-591.
Pauss A, Andre G, Perrier M, Guiot SR. 1990. Liquid-to-gas mass transfer in
anaerobic processes: inevitable transfer limitations of methane and hydrogen in
the biomethanation process. Appl Environ Microbiol 56:1636-1644.
Harold FM. 1986. The vital force: a study of bioenergetics. W. H. Freeman, New
York, NY, USA.
Roden EE, Jin Q. 2011. Thermodynamics of microbial growth coupled to
metabolism of glucose, ethanol, short-chain organic acids, and hydrogen. Appl
Environ Microbiol 77:1907-1909.
Gottschalk G. 1986. Bacterial metabolism. Springer Verlag, New York, NY,
USA.
White D. 2007. The physiology and biochemistry of prokaryotes, 3rd ed. Oxford
University Press, New York, NY.
Nelson DL, Cox MM. 2005. Lehninger principles of biochemistry, 4th ed, vol. 1.
W.H. Freeman and Company, New York, NY.
Battley EH. 1987. Energetics of microbial growth. Wiley, New York, NY, USA.
124
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
Von Stockar U, Gustafsson L, Larsson C, Marison I, Tissot P, Gnaiger E.
1993. Thermodynamic considerations in constructing energy balances for cellular
growth. Biochim Biophys Acta Bioenergetics 1183:221-240.
Jones S, Evans G, Galvin K. 1999. Bubble nucleation from gas cavities--a
review. Adv Colloid Interface Sci 80:27-50.
Stumm W, Morgan J. 1996. Aquatic chemistry, 3rd ed. Wiley, New York, NY.
Joergensen L, Degn H. 1983. Mass spectrometric measurements of methane and
oxygen utilization by methanotrophic bacteria. FEMS Microbiol Lett 20:331-335.
Jensen BB, Cox RP. 1983. Direct measurements of steady-state kinetics of
cyanobacterial N2 uptake by membrane-leak mass spectrometry and comparisons
between nitrogen fixation and acetylene reduction. Appl Environ Microbiol
45:1331-1337.
Kraemer JT, Bagley DM. 2006. Supersaturation of dissolved H2 and CO2 during
fermentative hydrogen production with N2 sparging. Biotechnol Lett 28:14851491.
Devol A, Richey J, Clark W, King S, Martinelli L. 1988. Methane emissions to
the troposphere from the Amazon floodplain. J Geophysical Res Atmos 93:15831592.
Millero FJ. 2006. Chemical oceanography, 3rd ed. Taylor and Francis, Boca
Raton, FL.
Takahashi T, Sutherland SC, Wanninkhof R, Sweeney C, Feely RA,
Chipman DW, Hales B, Friederich G, Chavez F, Sabine C. 2009.
Climatological mean and decadal change in surface ocean pCO2, and net sea-air
CO2 flux over the global oceans. Deep-Sea Res II 56:554-577.
Takahashi T, Sutherland SC, Sweeney C, Poisson A, Metzl N, Tilbrook B,
Bates N, Wanninkhof R, Feely RA, Sabine C. 2002. Global sea-air CO2 flux
based on climatological surface ocean pCO2, and seasonal biological and
temperature effects. Deep-Sea Res II 49:1601-1622.
Weiss R, Craig H. 1973. Precise shipboard determination of dissolved nitrogen,
oxygen, argon, and total inorganic carbon by gas chromatography. Deep-Sea Res
20:291-303.
Conrad R, Seiler W. 1988. Methane and hydrogen in seawater (Atlantic Ocean).
Deep-Sea Res A 35:1903-1917.
125