Thermodynamic Analysis of
Trinitrotoluene Biodegradation and
Mineralization Pathways
M. D. Shelley,'* R. I.. Autenrieth? J. R. Wild: and 9. E. Dale1**
'Departments of Chemical Engineering, T i v i l Engineering, and
3Biochemistry and Biophysics, Texas A&M University, College Station,
Texas 77843
Received August 3, 1995/Accepted February 8, 1996
Biodegradation of 2,4,6-trinitrotoluene (TNT) proceeds
through several different metabolic pathways. However,
the reaction steps which are considered rate-controlling
have not been fully determined. Glycolysis and other bio-
logical pathways contain biochemical reactions which are
acutely rate-limiting due to enzyme control. These rate-
limiting steps also have large negative Gibbs free energy
changes. Because xenobiotic compounds such as TNT
can be used by biological systems as nitrogen, carbon,
and energy sources, it is likely that their degradation path-
ways also contain acutely rate-limiting steps. Identifica-
tion of these rate-controlling reactions will enhance and
better direct genetic engineering techniques to increase
specific enzyme levels.
This article identifies likely rate-controlling steps (or
sets of steps) in reported TNT biodegradation pathways
by estimating the Gibbs free energy change for each step
and for the overall pathways. The biological standard
Gibbs free energy change of reaction was calculated for
each pathway step using a group contribution method
specifically tailored for biomolecules. The method was
also applied to hypothetical "pathways" constructed to
mineralize TNT using several different microorganisms.
Pathways steps that have large negative Gibbs free en-
ergy changes are postulated t o be potentially rate-control-
ling. The microorganisms which utilize degradation path-
ways with the largest overall (from TNT to citrate)
negatiave Gibbs free energy changes were also deter-
mined. Such microorganisms can extract more energy
from the starting substrate and are thus assumed to have
a competitive advantage over other microorganisms. Re-
sults from this modeling-based research are consistent
with much experimental work available in the literature.
0 1996 John Wiley & Sons, Inc.
Key words: trinitrotoluene TNT biodegredation Gibbs
free energy
INTRODUCTION
The manufacture of pesticides, explosives, plastics, azo
dyes, and pharmaceuticals generates many products and
byproducts that can pose an environmental threat.'
These products are designed for long life and resist
* Current address: Department of Chemical Engineering, Auburn
University, Auburn, AL 36849.
7 Current address: Department of Chemical Engineering,Michigan
State University, East Lansing, MI 48824.
To whom all correspondence should be addressed.
Biotechnology and Bioengineering, Vol. 50, Pp. 198-205 (1996)
0 1996 John Wiley & Sons, Inc.
reaction with common environmental chemicals.30 In
addition, nitro-substituted aromatics, the main constit-
uents of these products, are recalcitrant to natural bio-
degradation largely because they are toxic and muta-
genic toward most biological systems.' Trinitrotoluene
(TNT) is a nitroaromatic explosive that has been exten-
sively produced because of its stability; safe manufactur-
ing methods; low melting point; and low sensitivity to
impact, friction, and high temperature.44TNT produc-
tion and the refurbishment of existing munitions require
large quantities of water which become contaminated
with TNT.7For several decades, these wastewaters were
disposed into open-air lagoons and directly onto the
grounds surrounding production sites, thereby causing
soil and groundwater ~ontamination.~ The overall envi-
ronmental impact of TNT contamination has not been
fully determined; however, the bioamplification of this
pollutant in the food chain and the subsequent threat
to human health is of ~ o n c e r n . ~
Bioremediation uses microorganisms to decompose
toxic substances into various metabolic products. Con-
version of toxic compounds to carbon dioxide, water,
energy, and various metabolites occurs through numer-
ous different pathways via enzymatic reactions. Nitroar-
omatic contaminants such as TNT can be used as nitro-
gen, carbon, and energy sources by microorganisms.
Several bacteria which appear likely to mineralize TNT
have recently been i ~ o l a t e d . ~ ~In~ ~addition,
,'' the bio-
degradation pathways used by these microorganisms
have also been proposed.
Glycolysis and other biological pathways contain bio-
chemical reactions that are acutely rate-limiting due to
enzyme control. Glycolysis has three steps which are
primary regulation points for ATP synthesis. The cell
derives a substantial part of its energy needs as well as
carbon skeletons from these three steps; these are steps
which also have large negative Gibbs free energy
changes.41Product formation in these steps is spontane-
ous; however, acute enzyme control strictly limits the
rate of reaction. Since xenobiotic compounds such as
TNT are used in at least some biological systems as
nitrogen and carbon sources, it is likely that their degra-
dation pathways also contain enzyme-controlled steps.
CCC 0006-3592/96/020198-08
However, the pathway steps which are considered rate-
controlling have not been fully determined. For the pur-
poses of this work, we assume that steps with large
negative Gibbs free energy changes in TNT biodegrada-
tion are potentially rate-controlling, as they are in gly-
colysis.
In addition, Beurskens et al.4have reported that only
those chlorinated benzene biodegradation reactions
which have the largest Gibbs free energy changes are
experimentally observed. Thus, microorganisms appear
to utilize only those enzyme-catalyzed processes that
produce the greatest possible energy return for the cell.
This has been a basic postulate in the cybernetic model-
ing research conducted by Ramkrishna et al.36340 Biodeg-
radation of TNT may provide a greater energy return for
the cell than its more typical energy sources. Therefore,
these microorganisms could acquire a competitive ad-
vantage over other microbes within an environmental
system.
In the present work, a group contribution method
specifically tailored for biomolecules was used as a ther-
modynamic test to determine potentially rate-control-
ling steps within three reductive TNT biodegradation
pathways. The microorganisms which can utilize degra-
dation pathways with the largest overall negative Gibbs
free energy changes were also identified because these
microorganisms may have a competitive advantage over
other bacteria and fungi.
OUTLINE OF TNT BIODEGRADATION PATHWAYS
TNT can be transformed, mineralized, or conjugated
into more complex products by a variety of microorgan-
isms?’Bioremediation of TNT has been found to occur
under both aerobic and anaerobic conditions. Aerobic
degradation of aromatic compounds typically involves
an oxidative attack on the ring structure by electrophilic
oxygenases. Fewer reaction steps and more rapid reac-
tion rates are advantages of aerobic degradation.” How-
ever, the electron-withdrawing nitro groups inherent to
nitroaromatics reduce the electron density of the aro-
matic ring and inhibit oxidative attack by the electro-
philic oxygenases. The nitro groups are also bulky sub-
stituents which sterically hinder oxidative reaction.22
Furthermore, the conjugation of unstable nitroso and
hydroxyamino intermediates into more complex azo or
azoxy compounds is facilitated under aerobic condi-
tions.” Thus, TNT bioremediation research has focused
mainly on reductive or anaerobic environments rather
than oxidative or aerobic environments. Under both
aerobic and anerobic conditions, the initial steps in TNT
biotransformation typically involve reducing the nitro
groups to amino groups.15In most microorganisms, the
nitro group in the para position is reported to undergo
the first reduction.26 Subsequent nitro group reduc-
tion occurs at either ortho position to create diamino-
nitrotoluene isomers; however, complete reduction
SHELLEY ET AL.: TRINITROTOLUENE BIODEGRADATION AND MINERALIZATION PATHWAYS
of all three nitro groups is restricted to anaerobic
environments.”
A microbial consortium represents one approach to
TNT biodegradation. Remediation could also be con-
ducted in distinct stages to exploit the abilities of various
bacteria and fungi under both aerobic and anaerobic
conditions. The present work treats three separate path-
ways for TNT biodegradation which were obtained from
the literature. These pathways are initially anaerobic;
however, aerobic microorganisms could be used later
during biodegradation for ring cleavage and final miner-
alization of TNT intermediates.
TNT Pathway A
Boopathy et a1.6reported that the newly isolated bacte-
rium, Desulfovibrio sp. (B strain), metabolized TNT
with toluene as the main metabolic product. The first
three steps in the proposed pathway, as outlined in Fig-
ure la, involve reducing the nitro groups as TNT to the
corresponding 2,4,6-triaminotol~ene.~ Toluene is ob-
tained in the final step through complete reductive de-
amination of 2,4,6-triaminotol~ene.~
TNT Pathway 6
Duque et al.13 reported the isolation of two Pseudomo-
nus hybrid strains, sp. strain ClSl and sp. clone A, that
also metabolized TNT to toluene. In the proposed path-
way (Fig. lb), the three nitro groups on TNT are re-
moved as nitrite ions.13 Nitro group removal by these
Pseudomonus strains is thought to be analogous to a
similar reaction on picric acid by the bacterium, Rhodo-
coccus erythropolis HL 24-2.29In this reaction, nitro
group removal involves the formation of a hydride-
Meisenheimer complex (Fig. 2).13,29Vorbeck et al.43
have recently reported conclusive evidence that a
hydride-Meisenheimer complex is the initial metabolite
in aerobic TNT bioconversion. The main metabolic
product, toluene, is recovered after all three nitro groups
on TNT are removed.
Toluene Pathways A, B, and C
Several microorganisms can transform toluene, the met-
abolic product of TNT pathways A and B, into TCA
cycle intermediates under both aerobic and anaerobic
conditions. Aerobic catabolism occurs either through
direct ring attack by oxygen-dependent enzymes (tolu-
ene pathway A) or through hydroxylation of the methyl
group (toulene pathway B).39In the first degradation
approach, oxygenases directly attack the aromatic ring
to form 3-methyl catechol. Another oxygen-dependent
enzyme causes ring cleavage of the methyl catechol in-
termediate.” The second degradation approach in-
volves methyl group hydroxylation to form benzoate.
Catechol is created after a subsequent oxygenase reac-
199
 (a) TNT Pathway A

(b) TNT Pathway B

(c) TNT Pathway C

OH OH
Figure 1. TNT biodegradation p a t h ~ a y s . 6 ~ ' ~ ~ ' ~

tion.'* Ring cleavage occurs through the meta-cleavage
pathway using plasmid-encoded en~ymes.3~
Anaerobic degradation of toluene is postulated to
proceed by ring substitution of carbon dioxide to form
2,4,6-Trinitrotoluene
toluate, by ring substitution of water to form p-cresol,
or by hydroxylation of the methyl group to form benzo-
ate.37Dentitrifying bacteria, including several Pseudom-
onus species, have been reported to favor anaerobic
toluene degradation via methyl group hydroxylation
(toluene pathway C ) . ' TAs
~ ~ previously indicated, aero-
bic biodegradation is advantageous; therefore, toluene
bioremediation would likely be conducted under these
conditions. However, at least one anaerobic pathway is
needed for thermodynamic comparison. Thus, toluene
degradation via methyl group hydroxylation will serve
for this comparison. Under anaerobic conditions, ring
cleavage is facilitated by a hydration rea~tion.'~.~' Fur-
ther catabolism of the cleaved ring involves the incorpo-

Hydride-Meisenheimer complex
-.%: N02'
H
ration of three coenzyme A molecules.16

TNT Pathway C
Using indigenous microorganisms from TNT-contami-
nated soil inoculated with either dinoseb-degrading soil
bacteria or anaerobic TNT-degrading methanogenic
bacteria, Funk et al.I5recently reported the biodegrada-
2,4-Dinitrotoluene tion of TNT beyond triaminotoluene. The stepwise re-
duction of the TNT nitro groups yields triaminotoluene
NO2
as previously outlined in TNT pathway A (Fig. la).
Figure 2. Proposed mechanism for TNT pathway B nitro group The proposed anaerobic pathway for triaminotoluene
removal. The hydride-Meisenheimer complex is a key interme- biodegradation (Fig. lc) proceeds through methyl
diate.13.29,43 phloroglucinol and p-~resol.'~ p-Cresol is known to be

200 BIOTECHNOLOGY AND BIOENGINEERING. VOL. 51, NO. 2. JULY 20, 1996
catabolized under both aerobic and anaerobic condi-
tions by a variety of mi~roorganisms.’~
pCresol Pathways A, B, and C
Several microorganisms can transform p-cresol, the
TNT pathway C metabolic product, into TCA cycle
intermediates under both aerobic and anaerobic condi-
tions. As with toluene biodegradation pathways, aerobic
p-cresol catabolism occurs either through direct ring
attack by oxygen-dependent enzymes (p-cresol path-
way A) or through hydroxylation of the methyl group
(p-cresol pathway B).3,”723,25In the first approach, 4-
methyl catechol is formed by monooxygenase attack on
the aromatic ring.23 Ring cleavage and mineralization
occur via the meta-cleavage pathway.’ The second deg-
radation approach involves methyl group hydroxylation
to form 4 - h y d r o x y b e n ~ o a t eProtocatechuate
.~~~~~ is cre-
ated after a subsequent oxygenase reaction and is miner-
alized via the 3-oxoadipate (ortho-cleavage) pathway.21
Anaerobic degradation of p-cresol is postulated to pro-
ceed through methyl group hydroxylation to form 4-
hydroxybenzoate (p-cresol pathway C).8,’4 Reaction
with coenzyme-A creates 4-hdyroxybenzoyl-CoA,
which is subsequently dehydroxylated to yield benzoyl-
CoA, a key intermediate.2n.42Further catabolism in-
volves ring cleavage through a hydration reaction and
the incorporation of three coenzyme A molecules.
METHOD
We postulate that the likely rate-controlling steps (or
sets of steps) in TNT biodegradation and mineralization
pathways outlined above will have large negative Gibbs
free energy changes. The biological standard Gibbs free
energy change is given by the following equation:
AGO1 = -RT In K (1)
where the equilibrium constant, K, represents the activi-
ties of the products divided by the reactants with each
activity raised to the power of its stoichiometric coeffi-
~ i e n t The
. ~ ~biological standard state is defined as a
dilute aqueous solution at pH 7 and 25°C in which the
reacting species have molal concentrations equal to
unity.41 The biological standard Gibbs free energy
change represents the useful energy liberated by such
chemical reactions, which the cell may then capture for
other cell functions. The biological standard Gibbs free
energy change, however, does not predict the kinetics
or rate of the reaction.
The Gibbs free energy of formation for a compound
is the free energy required to create the compound from
its elements. The difference in Gibbs free energies of
formation between the products and reactants of a par-
ticular reaction represents the standard Gibbs free en-
ergy change, AGO, for that reaction. Unfortunately, the
Gibbs free energy of formation at the biological stan-
SHELLEY ET AL.: TRINITROTOLUENE BIODEGRADATION AND MINERALIZATION PATHWAYS
dard state is not known for most biochemical com-
p o u n d ~ . ~This
~”~ is especially true for xenobiotic com-
pounds such as TNT and its metabolic products.
Therefore, a method to estimate the Gibbs free energy
of formation for TNT biodegradation intermediates is
needed.
Group contribution methods have been widely used
to estimate the ideal gas properites of pure compounds.
These methods are based largely on the structures of
the compounds and to a lesser extent on state variables
such as temperature and pre~sure.’~ Mavrovounioti~~~~~~
recently developed a group contribution method that
accurately estimates the Gibbs free energy of formation
for the biological standard state. Use of the group contri-
bution method requires that biochemical compounds
be decomposed into functional groups (e.g., carboxyl
group, hdyroxyl group, etc.). Each functional group is
assigned a value based on the group’s partial contribu-
tion to the total thermodynamic property of the com-
pound. Functional group values were obtained using a
multiple linear regression routine based on biochemical
compounds with known Gibbs free energies of forma-
tion.33,34The biological standard Gibbs free energy
change is affected only by the functional groups which
change during the reaction; therefore, only the product
and reactant functional groups which change are in-
cluded in the free energy calculation. The biological
standard Gibbs free energy change is determined by
subtracting these reactant functional groups from the
product functional groups.
Although the group contribution method could theo-
retically be used for any organic compound in aqueous
solution, the data used to develop the method were
heavily biased toward biochemical compounds and reac-
t i o n ~For
. ~ ~this reason, some of the functional groups
that occur in xenobiotic compounds such as TNT were
not included in the published database. Functional
groups such as the nitro group, -NO2, simply are not
found in natural biological systems. Because the re-
moval or reduction of the nitro groups is a key step in
TNT biodegradation, the unavailability of a nitro group
contribution presented a major obstacle to using this
method. A nitroaromatic Gibbs free energy of forma-
tion was needed to determine the contribution of the
nitro group. A nitroaromatic compound was preferable
because its group contributions would better reflect pos-
sible nitro group and aromatic ring interactions.
Nitrobenzene was selected by default, because it was
the only nitroaromatic compound for which thermody-
namic data were available.32From a Gibbs free energy
of hydration and a Gibbs free energy of formation in
the ideal gas state, the Gibbs free energy of formation
of nitrobenzene in aqueous solution was
The nitro group contribution was then obtained by back-
calculating its value using known contributions for the
aromatic ring. The calculated nitro group contribution
of +6.9 kcal/mol is similar to those for other nitrogen-
20 1
 Table I. Comparison of experimental and calculated free energy changes in glycoly~is.”~

AGO’ (kcal/mol) AGO’ (kcal/mol) Difference (kcaUmo1)

Reaciton step by enzyme (literature) (calculated) (absolute value)

Hexokinase -4.0 -4.4 0.4

Phosphoglucoisomerase +0.4 -2.5 2.9
Phosphofructokinase-1 -3.4 -4.4 1.o
Aldolase +5.7 +4.6 1.1
Triosephosphate isomerase +1.8 +0.2 1.6
Glyceraldehyde 3- +IS +0.1 1.4
phosphate
dehydrogenase
Phosphoglycerate kinase -4.5 -4.8 0.3
Phosphoglyceromutase +1.1 +2.2 1.1
Eno1ase +0.4 -0.8 1.2
Pyruvate kinase -7.5 -8.5 1.o

Average absolute 1.2

difference (kcaUmol)

containing functional groups. Additional nonbiological RESULTS AND DISCUSSION

functional group contributions were either calculated
Using the reported TNT biodegradation pathways, the
or obtained from reliable literature sources.
group contribution method was applied to each step.
The biochemical group contribution method, as de-
From the computed Gibbs free energy changes, a num-
veloped by Mavrovouniotis et al.,3sprovides a relatively ber of patterns emerge. Several pathway steps have
accurate means for estimating the standard Gibbs free rather large negative Gibbs free energy changes (i.e.,
energy change of a reaction. The error for 85%of calcu- -AG“’ BlO kcal/mol). Because steps with large negative
lated Gibbs free energies of formation has been re-
Gibbs free energy changes are assumed to be potentially
ported to be less than t 2 kcaVm01.~~ The error is less
rate-controlling steps, these reactions are of consider-
than t 5 kcal/mol for 95% of calculated Gibbs free ener-
able interest.
gies of formation.33To confirm the accuracy of our ap-
The large negative free energy change reaction steps
proach, we calculated the AGO’ for each step in the
involve primarily the modification or removal of the
glycolytic pathway and compared these values with
nitro groups on TNT as well as the destruction or cleav-
those obtained from the literature. As given in Table I,
age of the aromatic ring (Tables I1 to IV). Experimental
the greatest absolute difference for a glycolytic pathway results indicate that these same steps have proven diffi-
step is 2.9 k c a h o l , while the average absolute differ- cult for microbes to c a t a l y ~ e . ~Duque
~ ’ ~ ~ et
’ ~al.13
~ ~ as
~
ence for all steps is 1.2 kcal/mol. Thus, the biochemical well as others have indicated that “a crucial point in
group contribution method closely approximates the the biodegradation of nitroaromatics is the removal of
actual Gibbs free energy change values for steps in gly- nitro groups from the molecule” and that “biodegrada-
colysis. This method, therefore, appears to be adequate tion is never complete either because part of the product
for attaining the objectives outlined in this article. remains unattacked or because it is converted to par-
As with any group contribution or property estima- tially reduced forms.” In addition, Boopathy et a1.6 re-
tion method, there are inherent limitations. This bio- ported that “the aromatic ring of TNT was not cleaved
chemical group contribution method, however, does
provide corrections to minimize the estimation error Table 11. TNT pathway A results?
caused by amide groups, pure hydrocarbon compounds,
three atom rings, and heteroaromatic rings.34Nonethe-
less, the method does not accurately predict the increaed
free energy changes of reaction attributable to group
interactions or steric hindrances involving multiple sub-
stituents. Likewise, increased Gibbs free energy changes
of reaction due to the stability of symmetric compounds
such as TNT cannot be determined. Additionally, the
free energy change of reaction associated with the re-
arrangement of substituents around an aromatic ring
cannot be ascertained using this procedure. In such
cases, literature sources, if available, must be consulted
to obtain a more accurate thermodynamic estimate.

202 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 51, NO. 2, JULY 20, 1996
Table 111. TNT pathway B r e ~ u l t s . ' ~
[using the bacterium, Desulfovibrio sp. (B strain)]," and
they "did not identify metabolites other than toluene
even after 6 months of incubation." Therefore, the pres-
ent work both predicts and provides supporting evi-
dence that the removal of the nitro groups on TNT
and the cleavage of the aromatic ring are major rate-
controlling steps in TNT biodegradation.
The Gibbs free energy change is a state function;
therefore, the free energy change between two specified
TNT metabolic intermediates will be constant and inde-
pendent of the reaction path. Several TNT biodegrada-
tion reactions involve intermediate steps that are not
included in the pathways described here, because the
mechanisms of these reactions are not well known. The
calculated free energy change for some steps is quite
large; therefore, these reaction "steps" are probably
composed of several intermediate steps which have not
been fully determined. A close examination should be
conducted to determine whether these are indeed one-
step mechanisms with large -AGO' or whether they are
instead composed of several intermediate steps, one or
more of which may have a large -AGO/.
Table IV. TNT pathway C r e s ~ l t s . ' ~
SHELLEY ET AL.: TRINITROTOLUENE BIODEGRADATION AND MINERALIZATION PATHWAYS
An explosive compound such as TNT possesses an
enormous potential free energy; therefore, it is quite
possible that individual TNT biodegradation steps (or
sets of steps) would have relatively large negative Gibbs
free energy changes as compared to glycolytic and TCA
cycle steps. If intermediate steps with large -AGO/ exist,
these reactions represent a potentially rate-limiting pro-
cess through which the reaction must proceed. The use
of genetic engineering techniques to increase specific
enzyme levels or otherwise alter regulation will require
identification of the enzymes catalyzing these possible
intermediate reactions. In any case, the free energy cal-
culations afforded by the group contribution method
may help identify those reaction steps which limit the
overall rate of biodegradation and mineralization. This
analysis assumes that the transport of intermediate me-
tabolites across cell walls and between microorganisms
in a consortium is not rate-limiting.
The overall Gibbs free energy changes for the TNT
biodegradation pathways are given in Table V. Because
a consortium of microorganisms could be used to de-
grade TNT and its intermediates, all possible pathway
combinations were used to generate this table. For accu-
rate pathway comparisons, the thermodynamic data
given in Table V are based on TNT as the starting
substrate and citrate as the common metabolic product.
Although TNT pathways A, B, and C are initially anaer-
obic, these pathways coupled with their respective aero-
bic toluene or p-cresol pathways possess the largest
overall negative free energy changes. Pathway combina-
tions which are completely anaerobic have the smallest
negative free energy changes. Aerobic reactions obvi-
ously release more free energy than their anaerobic
counterparts. Therefore, the aerobic TNT degradation
pathways would be expected to liberate more free en-
ergy than their anaerobic counterparts, and the results
summarized in Table V fulfill this expectation.
Within each initial TNT degradation route, the path-
way combination possessing the largest overall negative
Table V. Overall Gibbs free energy change" for each hypothetical
pathway?
"Overall AG"' from TNT to citrate.
bHypothetical pathways constructed using a microbial consortium.
203
Gibbs free energy change has been shaded in Table V.
Experimental observations indicate that these pathway
combinations, containing toluene pathway B and
p-cresol pathway B, are the routes used most often by
microorganisms or that have enjoyed the most experi-
mental success in TNT bi~degradation.~,' 1,13,21,2s During
the last decade, an enormous amount of research has
been conducted on toluene pathway B and the plasmid-
encoded biodegradation of toluene.39 Furthermore,
Duque et al.13 used a TOL plasmid-encoded Pseudomo-
nus strain to actively mineralize the toluene created via
TNT pathway B. In addition, many of the Pseudomonas
species and other bacteria that grow with aromatic sub-
strates predominately utilize p-cresol pathway B.21
Thus, the pathways shaded in Table V appear to be the
most favorable routes for TNT bioremediation. This
result also confirms the evidence given by Beurskens et
a1.4 that only those xenobiotic degradation reactions
which liberate the most Gibbs free energy are experi-
mentally observed. Microorganisms appear to utilize
only those enzyme-catalyzed processes that produce the
greatest possible energy return for the 53 Biodeg-
radation of TNT could provide a greater energy return
than more typical energy sources such as glucose or
organic acids. Therefore, the microorganisms utilizing
these pathways may acquire a competitive advantage
over other microbes in a microbial consortium.
Because toluene and p-cresol can be readily degraded
by several different microorganisms, a biodegradation
pathway which yields either of these two TNT biodegra-
dation intermediates is desirable.".39 TNT pathway
A possesses the largest overall negative Gibbs free
energy change (AG"' = -350.4 kcal/mol) and thus
appears to be most favorable. However, TNT path-
way C releases a comparable amount of free energy
(AGO/ = -288.1 kcalhol) through rate-controlling steps
that have smaller stepwise free energy changes. This
could explain why more experimental success has been
achieved by the microorganisms utilizing TNT pathway
C.lSLikewise, TNT pathway B appears to be an optimal
biodegradation pathway, because it contains rate-
controlling steps of relatively small magnitude (smaller
-AGO/). However, this pathway has had only modest
success in TNT biodegradation experiments probably
due to its smaller overall -AGO' (providing less available
free energy).13 Duque et al.13 indicated that the two
Pseudomonas hybrid strains, sp. strain ClSl and sp.
clone A, which utilize TNT pathway B can also degrade
TNT using TNT pathway A. In addition, these bacteria
are inclined to degrade TNT to a much greater extent
through TNT pathway A than through TNT pathway
B.I3 This finding further illustrates the predisposition of
microorganisms to target degradation pathways which
release the most free energy.
Therefore, from a thermodynamic perspective, the
smaller -AGO1 rate-controlling steps within TNT path-
way C and p-cresol pathway B should be the focus of
204 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 51, NO. 2, JULY 20, 1996
future TNT biodegradation research as well as genetic
engineering techniques to increase specific enzyme lev-
els or otherwise alter regulation. Elevating the enzyme
levels for rate-controlling steps should help circumvent
strict enzyme control thereby increasing the forward
reaction rate and overall TNT mineralization.
The authors are grateful for the support provided by the
U.S. Army Research Office under the University Research
Initiative, Project 30371-LS.
References
1. Altenschmidt, U., Fuchs, G. 1991. Anaerobic degradation of tolu-
ene in denitrifying Pseudornonas sp.: indication for toluene meth-
ylhydroxylation and benzoyl-CoA as central aromatic intermedi-
ate. Arch. Microbiol. 156: 152-158.
2. Bayly, R.C., Barbour, M. G. 1984. Degradation of aromatic com-
pounds by the meta and gentisate pathways, pp. 253-286. In:
D. T. Gibson (ed.), Microbial degradation of organic compounds,
Marcel Dekker, New York.
3. Bayly, R. C., Dagley, S., Gibson, D. T. 1966.Metabolism of cresols
by species of Pseudomonas. Biochem. J. 101 293-301.
4. Beurskens, J. E. M., Dekker C. G. C., Heuvel, H. V. D., Swart,
M., Wolf, J. D. 1994. Dechlorination of chlorinated benzenes by
an anaerobic microbial consortium that selectively mediates the
thermodynamic most favorable reactions. Environ. Sci. Technol.
28: 701-706.
5. Boopathy, R., Kulpa, C. F., Manning, J., Montemagno, C. D. 1994.
Biotransformation of TNT by co-metabolism with various co-
substrates. Biores. Technol 4 7 205-208.
6. Boopathy, R., Kulpa, C. F., Wilson, M. 1993. Metabolism of TNT
by Desulfovibrio sp. (b strain). Appl. Microbiol. Biotechnol. 39
270-275.
7. Boopathy, R.,Wilson, M., Kulpa, C. F. 1993. Anaerobic removal
of TNT under different electron accepting conditions. Water Envi-
ron. Res. 6 5 271-275.
8. Bossert, I. D., Young, L. Y. 1986. Anaerobic oxidation ofp-cresol
by a denitrifying bacterium. Appl. Environ. Microbiol. 52: 1117-
1122.
9. Bruhn, C., Lenke, H., Knackmuss, H. J. 1987. Nitrosubstituted
aromatic compounds as nitrogen sources for bacteria. Appl. Envi-
ron. Microbiol. 53: 208-210.
10. Cabani, S.,Gianni, P., Mollica, V., Lepori, L. 1981. Group contri-
butions to the thermodynamic properties of non ionic organic
solutes in dilute aqueous solution. J. Sol. Chem. 10:563-591.
11. Dagley, S., Patel, M. D. 1957. Oxidation of p-cresol and related
compounds by a Pseudomonas. Biochem. J. 66: 227-233.
12. Daubert, T. E., Danner, R. P. 1985. Data compilation tables of
properties of pure compounds. 1st edition. AICHE, New York.
13. Duque, E., Haidour, A., Godoy, F., Ramos, J. L. 1993. Construc-
tion of a Pseudornonas hybrid strain that mineralizes TNT.
J. Bacteriol. 175:2278-2283.
14. Evans, W. C., Fuchs, G. 1988. Anaerobic degradation of aromatic
compounds. Ann. Rev. Microbiol. 4 2 289-317.
15. Funk, S. B., Roberts, D. J., Crawford, D. L., Crawford, R. L. 1993.
Initial-phase optimization for bioremediation of munition com-
pound-contaminated soils. Appl. Environ. Microbiol. 59: 2171-
2177.
16. Gallus, C., Schink, B. 1994.Anaerobic degradation of pimelate by
newly isolated denitrifying bacteria. Microbiology 140:409-416.
17. Geissler, J. F., Hanvood, C. S., Gibson, J. 1988. Purification and
properties of benzoate-Coenzyme A ligase, a Rhodopseudornonas
palustris enzyme involved in the anaerobic degradation of benzo-
ate. J. Bacteriol. 170 1709-1714.
18. Gibson D.T., Subramanian V. 1984. Microbial degradation of
aromatic hydrocarbons, pp. 181-252. In: D. T. Gibson (ed.), Mi-
crobial degradation of organic compounds, Marcel Dekker,
New York.
19. Gibson, K. J., Gibson, J. 1992. Potential early intermediates in
aromatic benzoate degradation by Rhodopseudomonas palustris.
Appl. Environ. Microbiol. 5 8 696-698.
20. Glockler, R., Tschech, A,, Fuchs, G. 1989.Reductive dehydroxyla-
tion of 4-hydroxybenzoyl-CoA to benzoyl-CoA in a denitrifying,
phenol-degrading Pseudomonas species. FEBS Lett. 251 237-240.
21. Gottschalk, G. 1976. Bacterial metabolism. 2nd edition. Springer-
Verlag, New York.
22. Higson, F. K. 1992. Microbial degradation of nitroaromatic com-
pounds, pp. 1-19. In: S. L. Neidleman and A. L. Laskin (eds.),
Advances in applied microbiology, vol. 37. Academic Press, San
Diego, CA.
23. Hopper, D. J., Taylor, D. G. 1975. Pathways for the degradation
of m-cresol and p-cresol by Pseudomonas pufida. J. Bacteriol.
122 1-6.
24. Joback, K. G., Reid, R. C. 1987. Estimation of pure-component
properties from group contributions. Chem. Eng. Commun. 57
233-243.
25. Jones, K. H., Trudgill, P. W., Hopper, D. J. 1993. Metabolism of
p-cresol by the fungus Aspergillus fumigatus. Appl. Environ. Mi-
crobiol. 59: 1125-1130.
26. Kaplan, D. L. 1993.Biotechnology and bioremediation for organic
energetic compounds. In: P. Marinkas (ed.), Organic energetic
compounds. Nova Science Publishers, New York.
27. Kaplan, D. L., Kaplan, A. M. 1982. Mutagenicity of 2,4,6-trinitro-
toluene surfactant complexes. Bull. Environm. Contam. Toxicol.
28: 33-38.
28. Koch, J., Eisenreich, W., Bacher, A,, Fuchs, G. 1993. Products of
enzymatic reduction of benzoyl-CoA, a key reaction in anaerobic
aromatic metabolism. Eur. J. Biochem. 211:649-661.
29. Lenke, H., Knackmuss, H.-J. 1992. Initial hydrogenation during
catabolism of picric acid by Rhodococcus erythropolis HL 24-2.
Appl. Environ. Microbiol. 58: 2933-2937.
SHELLEY ET AL.: TRINITROTOLUENE BIODEGRADATION AND MINERALIZATION PATHWAYS
30. Levin, M. A., Gealt, M. A. 1993. Biotreatment of industrial and
hazardous waste. 1st edition. McGraw-Hill, New York.
31. Lovley, D. R., Lonergan, D. J. 1990. Anaerobic oxidation of tolu-
ene, phenol, andp-cresol by the dissimilatory iron-reducing organ-
ism. GS-15. Appl. Environ. Microbiol. 56: 1858-1864.
32. Marsh, K. N. 1994.Personal communication, Director, Thermody-
namics Research Center, Texas A&M University, College Sta-
tion, TX.
33. Mavrovouniotis, M. L. 1990. Group contributions for estimating
standard gibbs energies of formation of biological compounds in
aqueous solution. Biotechnol. Bioeng. 36: 1070-1082.
34. Mavrovouniotis, M. L. 1991. Estimation of standard Gibbs energy
changes of biotransformations. J. Biol. Chem. 266: 14440-14445.
35. Mavrovouniotis, M. L., Bayol, P., Lam, T. K. M., Stephanopoulos,
G. 1988. A group contribution method for the estimation of equi-
librium constants for biochemical reactions. Biochem. Techniq.
2:23-28.
36. Ramkrishna, D., Kompala, D. S., Tsao, G. T. 1987. Are microbes
optimal strategists? Biotechnol. Prog. 3: 121-126.
37. Schoeher, R. J., Seyfried, B., Vasquez, F., Zeyer, J. 1991. Anaero-
bic degradation of toluene by pure cultures of denitrifying bacte-
ria. Arch. Microbiol. 157 7-12.
38. Smith, J. M., Van Ness, H. C. 1987. Introduction to chemical engi-
neering thermodynamics. 4th edition. McGraw-Hill, New York.
39. Smith, M. R. 1990. Biodegradation of aromatic hydrocarbons by
bacteria. Biodegradation 1: 191-206.
40. Straight, J. V., Ramkrishna, D. 1994. Cybernetic modeling and
regulation of metabolic pathways. Biotechnol. Prog. 10 574-587.
41. Stryer, L. 1988. Biochemistry. 3rd edition. W. H. Freeman & Co.,
New York.
42. Tschech, A., Fuchs, G. 1987. Anaerobic degradation of phenol
by pure cultures of newly isolated denitrifying pseudomonads.
Arch. Microbiol. 148:213-217.
43. Vorbeck, C., Lenke, H., Fischer, P., Knackmuss,H. J. 1994.Identi-
fication of a hydride-Meisenheimer complex as a metabolite of
2,4,6-trinitrotoluene by a Mycobacferium strain. J. Bacteriol.
176:931-934.
44. Yinon, J. 1990. Toxicity and metabolism of explosives. 1st edition.
CRC Press, Boca Raton, FL.
205