the pesticide molecule. These two categories of
degradations are also not mutually exclusive and may
coexist in the same environment. At the community
level, primary and ultimate degradations are the functional equivalents of metabolism and cometabolism
described for pure strains. The ecological role of complex microbial communities was later reconsidered by
Bull (1980) who recognized that pure culture studies had
benefits for the elucidation of the metabolic functioning
of microorganisms and transformation pathways. However, they had also many limitations for extrapolation to
natural environments that allow enzyme complementarity and redundancy required to sustain a collaborative
process of degradation. The presence, within a microbial
community of isofunctional enzymes, is likely to ensure
better enzyme efficacy especially under varying environmental conditions. This is also the case for consortia
that drive alternative routes of transformation. At last,
the diversification of microbial hosts resulting from
horizontal gene transfer contributes to stabilize the
genetic determinants of degradation and to maintain
the overall degrading capacity. Nevertheless, at the present state of technical advances, the isolation of
degrading strains or microbial communities requires
the use of liquid culture media. This is a major drawback
because it focuses on culturable microorganisms which
are likely to have limited contribution to the degradation
of persistent compounds with ecological significance.
It may also explain some of the pitfalls and failures
encountered in bioremediation experiments as will be
discussed in the following.
Genetic Basis of Pesticide Biodegradation
The knowledge about genetic processes responsible for the
capacity of soil microorganisms to degrade pesticides
mainly results from advanced studies carried out on
accelerated biodegradation of pesticides. This phenomenon
that relies on the growth and development of the size of
degrading-microbial community diminished pesticide persistence in the environment (for reviews, see Topp et al.
2004; Arbeli and Fuentes 2007). It is well known that
accelerated biodegradation of pesticides can last for several
crop rotations. In several cases, accelerated biodegradation
was shown to diminish the efficacy of pesticide treatment:
this phenomenon is particularly well described for the insecticide carbofuran (Felsot et al. 1982). The stability of
pesticide-degrading capabilities observed in situ is in contradiction with the frequently reported instability of
degrading capabilities of pure microbial strains maintained
under laboratory conditions (Changey et al. 2011). Indeed, it
has been observed by different researchers that pesticidedegrading potential can easily be lost under laboratory
conditions by simply sub-cultivating degrading strains on
agar medium without the selection pressure exerted by the
pesticide. Recently, Changey et al. (2011) suggested that in
absence of the selection pressure, pesticide-degradinggenetic potential represents a genetic burden favoring the
selection of strains having selectively lost this burden by
different genetic mechanisms. Keeping in mind this instability observed on pure strains kept under laboratory
conditions, several hypotheses have been drawn to tentatively explain the stability of pesticide-degrading ability of
soil microbial community observed in situ. Among these
hypotheses, survival of pesticide-degrading populations
and maintenance of the integrity of their pesticidedegrading-genetic potential might result from the presence
in the soil of natural compounds sharing enough structural
homologies with the pesticide to maintain the selective
advantage favorable for their survival. This hypothesis was
verified for 2,4-D biodegradation for which it was found that
both 4-chlorocatechol (intermediary metabolite of 2,4-D)
and homogentisic acid (2,5-dihydroxyphenylacetic acid,
intermediary metabolite of the biosynthesis of aromatic
amino acid) are able to induce the expression of 2,4-D
biodegradation pathway in Arthrobacter sp. (Sandmann
and Loos 1988). Similarly, for atrazine biodegradation,
Udikovic-Kolic and collaborators (2012) suggested that the
prevalence of trzN (responsible for the dechloration of atrazine) might results from its ecological advantages: It is faster
(Vmax 31 mmol min
À1 mg
À1 of protein), it has a higher
affinity for the substrate (Km 25 μM), and it has a higher
range of substrates including natural ones. One could also
hypothesize that under oligotrophic conditions found in the
soil, induction and repression mechanisms do not act with
the same intensity than under pleiotropic* conditions. It is
now well admitted that a lot of the genes coding for pesticide
catabolic enzymes are plasmid-borne. A lot of these plasmids
are conjugative and are part of the so-called mobilome which
can be transferred from a donor to a receptor population by
horizontal gene transfer (HGT) mediated by bacterial conjugation (Frost et al. 2005). The first demonstration of the
plasmidic location of pesticide-degrading genes was done at
the end of the 1970s on Alcaligenes paradoxus able to
degrade 2,4-D, thanks to degrading genes located on the
pJP1 plasmid (Pemberton and Fisher 1977). Thereafter,
several other plasmids harboring 2,4-D-degrading genes
were isolated among which pJP4, isolated from Alcaligenes
eutrophus JMP 134 (Don and Pemberton 1985) later on
renamed Ralstonia eutropha and then Cupriavidus necator
JMP 134, has particularly been studied. All the genes coding
for the catabolic enzymes, transporter, and regulators have
692
J.-C. Bertrand et al.
degradations are also not mutually exclusive and may
coexist in the same environment. At the community
level, primary and ultimate degradations are the functional equivalents of metabolism and cometabolism
described for pure strains. The ecological role of complex microbial communities was later reconsidered by
Bull (1980) who recognized that pure culture studies had
benefits for the elucidation of the metabolic functioning
of microorganisms and transformation pathways. However, they had also many limitations for extrapolation to
natural environments that allow enzyme complementarity and redundancy required to sustain a collaborative
process of degradation. The presence, within a microbial
community of isofunctional enzymes, is likely to ensure
better enzyme efficacy especially under varying environmental conditions. This is also the case for consortia
that drive alternative routes of transformation. At last,
the diversification of microbial hosts resulting from
horizontal gene transfer contributes to stabilize the
genetic determinants of degradation and to maintain
the overall degrading capacity. Nevertheless, at the present state of technical advances, the isolation of
degrading strains or microbial communities requires
the use of liquid culture media. This is a major drawback
because it focuses on culturable microorganisms which
are likely to have limited contribution to the degradation
of persistent compounds with ecological significance.
It may also explain some of the pitfalls and failures
encountered in bioremediation experiments as will be
discussed in the following.
Genetic Basis of Pesticide Biodegradation
The knowledge about genetic processes responsible for the
capacity of soil microorganisms to degrade pesticides
mainly results from advanced studies carried out on
accelerated biodegradation of pesticides. This phenomenon
that relies on the growth and development of the size of
degrading-microbial community diminished pesticide persistence in the environment (for reviews, see Topp et al.
2004; Arbeli and Fuentes 2007). It is well known that
accelerated biodegradation of pesticides can last for several
crop rotations. In several cases, accelerated biodegradation
was shown to diminish the efficacy of pesticide treatment:
this phenomenon is particularly well described for the insecticide carbofuran (Felsot et al. 1982). The stability of
pesticide-degrading capabilities observed in situ is in contradiction with the frequently reported instability of
degrading capabilities of pure microbial strains maintained
under laboratory conditions (Changey et al. 2011). Indeed, it
has been observed by different researchers that pesticidedegrading potential can easily be lost under laboratory
conditions by simply sub-cultivating degrading strains on
agar medium without the selection pressure exerted by the
pesticide. Recently, Changey et al. (2011) suggested that in
absence of the selection pressure, pesticide-degradinggenetic potential represents a genetic burden favoring the
selection of strains having selectively lost this burden by
different genetic mechanisms. Keeping in mind this instability observed on pure strains kept under laboratory
conditions, several hypotheses have been drawn to tentatively explain the stability of pesticide-degrading ability of
soil microbial community observed in situ. Among these
hypotheses, survival of pesticide-degrading populations
and maintenance of the integrity of their pesticidedegrading-genetic potential might result from the presence
in the soil of natural compounds sharing enough structural
homologies with the pesticide to maintain the selective
advantage favorable for their survival. This hypothesis was
verified for 2,4-D biodegradation for which it was found that
both 4-chlorocatechol (intermediary metabolite of 2,4-D)
and homogentisic acid (2,5-dihydroxyphenylacetic acid,
intermediary metabolite of the biosynthesis of aromatic
amino acid) are able to induce the expression of 2,4-D
biodegradation pathway in Arthrobacter sp. (Sandmann
and Loos 1988). Similarly, for atrazine biodegradation,
Udikovic-Kolic and collaborators (2012) suggested that the
prevalence of trzN (responsible for the dechloration of atrazine) might results from its ecological advantages: It is faster
(Vmax 31 mmol min
À1 mg
À1 of protein), it has a higher
affinity for the substrate (Km 25 μM), and it has a higher
range of substrates including natural ones. One could also
hypothesize that under oligotrophic conditions found in the
soil, induction and repression mechanisms do not act with
the same intensity than under pleiotropic* conditions. It is
now well admitted that a lot of the genes coding for pesticide
catabolic enzymes are plasmid-borne. A lot of these plasmids
are conjugative and are part of the so-called mobilome which
can be transferred from a donor to a receptor population by
horizontal gene transfer (HGT) mediated by bacterial conjugation (Frost et al. 2005). The first demonstration of the
plasmidic location of pesticide-degrading genes was done at
the end of the 1970s on Alcaligenes paradoxus able to
degrade 2,4-D, thanks to degrading genes located on the
pJP1 plasmid (Pemberton and Fisher 1977). Thereafter,
several other plasmids harboring 2,4-D-degrading genes
were isolated among which pJP4, isolated from Alcaligenes
eutrophus JMP 134 (Don and Pemberton 1985) later on
renamed Ralstonia eutropha and then Cupriavidus necator
JMP 134, has particularly been studied. All the genes coding
for the catabolic enzymes, transporter, and regulators have
692
J.-C. Bertrand et al.