cells. In addition, the acquired genetic material (1) can be
used as template for DNA repair if the foreign DNA
originated from a representative of the same species and/or
(2) can carry genes (or alleles) conferring a selective advantage (e.g., resistance to antibiotics, new metabolic
capabilities, virulence factors, etc.) to the recipient cell.
The DNA involved in transformation, called transforming
DNA, may have various origins: lysed or decomposed cells,
dislocated viral particles, living cells secreting DNA, etc.
While the transforming DNA is usually of chromosomal
origin, the transfer of plasmids is also possible but involves
additional steps (see below). The rate of transformation in
natural populations is very difficult to measure, because it is
likely linked to the persistence of DNA in the environment.
The half-life of DNA depends on several factors such as
temperature, pH, presence of enzymes degrading DNA, etc.
It can vary from a few minutes to several weeks depending
on the environments.
Transformation was discovered by Griffith (1928;
Downie 1972), who studied in the 1920s the virulence of
Streptococcus pneumoniae (Firmicutes) in mice. He showed
that the injection of S. pneumoniae S1, a virulent strain
carrying capsular antigens of type S1, caused the death of
animals. In contrast, mice survived when the cells were
previously killed by heat. In parallel, Griffith realized
another series of experiments using S. pneumoniae R2, a
non-virulent strain derived from the virulent strain S2, which
had lost its capsular antigens. He showed that the simultaneous injection of killed S1 cells together with living R2
cells caused the death of animals. The examination of dead
mice revealed living S. pneumoniae S1 cells. Based on these
results, Griffith hypothesized that the R2 cells have acquired
S1 capsular antigens from the dead S1 cells. This acquisition
would have restored the virulence of the R2 strain. The
principle of transformation has been correctly formulated
by Avery and colleagues in the 1940s (Avery et al. 1944;
Downie 1972).
Transformation has been put in evidence in prokaryotes
belonging to very different bacterial phyla: Cyanobacteria,
Firmicutes, Deinococcus-Thermus, Chlorobi, etc., and in
Archaea (Lorenz and Wackernagel 1994). It requires that
the recipient cells reach a genetically programmed stage of
development called competence*, which involves 20–50
proteins. To a few exceptions (e.g., Neisseria gonorrhoeae,
Betaproteobacteria), natural competence is a transient state
depending on environmental conditions (e.g., nutrient depletion, high cell density, etc.). Entry and exit into competence,
as well as the proportion of cells entering into competence
within a population, are influenced by environmental and
genetic factors, which vary from one species (or strain or
isolate) to another (Maughan and Redfield 2009). For
instance, in the case of Bacillus subtilis (Firmicutes), 20 %
of cell population on average enters into competence for a
period lasting several hours; in the case of Streptococcus, the
whole population is capable of entering into competence but
remains in this state only for a few minutes. This explains
why it is so difficult to determine whether a species, strain,
or isolate is capable of natural transformation or not. Nevertheless, current data suggest that only 1 % of bacterial
described species are capable of natural transformation. It
is important to note that while many microbial species (e.g.,
Escherichia coli strains, Gammaproteobacteria) are not naturally transformable, artificial methods of transformation
based on the alteration of the permeability of membranes
(e.g., electroporation) have been developed and are routinely
used in the laboratory. It has been proposed that electrical
storms could create comparable physicochemical conditions
in soils and result in transformation by electroporation of
cells present in soil (Ceremonie et al. 2004).
The second step of transformation involves the noncovalent binding of the DNA present in the environment to
DNA-binding proteins (e.g., ComAE proteins in Bacillus
subtilis, Firmicutes) located at the surface of the cell
(Fig. 12.1a). DNA must be double-stranded and present a
free end. In Gram-negative bacteria, the outer membrane is
an additional obstacle to the entry of transforming DNA into
the cells. Secretins (e.g., PilQ in N. gonorrhoeae, a
betaproteobacterium) form a channel and allow the
transforming DNA to cross the outer membrane. Other
mechanisms involve small membrane vesicles, called
transformasomes, protruding on the external surface of the
cell. The transforming DNA is captured by these vesicles
and transported into the periplasmic space. Some
microorganisms are able to filter selectively the incoming
DNA during its translocation. This selective step involves
the recognition of specific 10 base pair sequences called
uptake signal sequences (USS), such as 5
0 -
GCCGTCTGAA-3
0
in
N. gonorrhoeae
or
5
0 -
AAGTGCGGT-3
0 in Haemophilus influenzae, a gammaproteobacterium (Danner et al. 1980; Goodman and Scocca
1988). USS are highly repeated in the genomes of the species
carrying them, whereas according to their relative large size,
the expected occurrence of a given USS in any genome is
low. For instance, 1,900 copies are present in
N. gonorrhoeae and 1,471 in H. influenzae (Smith et al.
1995). Their overrepresentation in these genomes suggests
a positive selection pressure to maintain a high number. It
has been shown that the recipient cells translocate preferentially transforming DNA carrying USS similar to their own
USS. This suggests that USS recognition systems exist. Even
if they have not been identified yet, in N. gonorrhoeae their
activity appears to depend on the pilus-coding genes (Chen
and Dubnau 2004). It is important to note that the existence
of USS seems to be an exception rather than a rule because
many microorganisms are able to translocate efficiently
transforming DNA regardless of its sequence. The entry
446
C. Brochier-Armanet and D. Moreira
used as template for DNA repair if the foreign DNA
originated from a representative of the same species and/or
(2) can carry genes (or alleles) conferring a selective advantage (e.g., resistance to antibiotics, new metabolic
capabilities, virulence factors, etc.) to the recipient cell.
The DNA involved in transformation, called transforming
DNA, may have various origins: lysed or decomposed cells,
dislocated viral particles, living cells secreting DNA, etc.
While the transforming DNA is usually of chromosomal
origin, the transfer of plasmids is also possible but involves
additional steps (see below). The rate of transformation in
natural populations is very difficult to measure, because it is
likely linked to the persistence of DNA in the environment.
The half-life of DNA depends on several factors such as
temperature, pH, presence of enzymes degrading DNA, etc.
It can vary from a few minutes to several weeks depending
on the environments.
Transformation was discovered by Griffith (1928;
Downie 1972), who studied in the 1920s the virulence of
Streptococcus pneumoniae (Firmicutes) in mice. He showed
that the injection of S. pneumoniae S1, a virulent strain
carrying capsular antigens of type S1, caused the death of
animals. In contrast, mice survived when the cells were
previously killed by heat. In parallel, Griffith realized
another series of experiments using S. pneumoniae R2, a
non-virulent strain derived from the virulent strain S2, which
had lost its capsular antigens. He showed that the simultaneous injection of killed S1 cells together with living R2
cells caused the death of animals. The examination of dead
mice revealed living S. pneumoniae S1 cells. Based on these
results, Griffith hypothesized that the R2 cells have acquired
S1 capsular antigens from the dead S1 cells. This acquisition
would have restored the virulence of the R2 strain. The
principle of transformation has been correctly formulated
by Avery and colleagues in the 1940s (Avery et al. 1944;
Downie 1972).
Transformation has been put in evidence in prokaryotes
belonging to very different bacterial phyla: Cyanobacteria,
Firmicutes, Deinococcus-Thermus, Chlorobi, etc., and in
Archaea (Lorenz and Wackernagel 1994). It requires that
the recipient cells reach a genetically programmed stage of
development called competence*, which involves 20–50
proteins. To a few exceptions (e.g., Neisseria gonorrhoeae,
Betaproteobacteria), natural competence is a transient state
depending on environmental conditions (e.g., nutrient depletion, high cell density, etc.). Entry and exit into competence,
as well as the proportion of cells entering into competence
within a population, are influenced by environmental and
genetic factors, which vary from one species (or strain or
isolate) to another (Maughan and Redfield 2009). For
instance, in the case of Bacillus subtilis (Firmicutes), 20 %
of cell population on average enters into competence for a
period lasting several hours; in the case of Streptococcus, the
whole population is capable of entering into competence but
remains in this state only for a few minutes. This explains
why it is so difficult to determine whether a species, strain,
or isolate is capable of natural transformation or not. Nevertheless, current data suggest that only 1 % of bacterial
described species are capable of natural transformation. It
is important to note that while many microbial species (e.g.,
Escherichia coli strains, Gammaproteobacteria) are not naturally transformable, artificial methods of transformation
based on the alteration of the permeability of membranes
(e.g., electroporation) have been developed and are routinely
used in the laboratory. It has been proposed that electrical
storms could create comparable physicochemical conditions
in soils and result in transformation by electroporation of
cells present in soil (Ceremonie et al. 2004).
The second step of transformation involves the noncovalent binding of the DNA present in the environment to
DNA-binding proteins (e.g., ComAE proteins in Bacillus
subtilis, Firmicutes) located at the surface of the cell
(Fig. 12.1a). DNA must be double-stranded and present a
free end. In Gram-negative bacteria, the outer membrane is
an additional obstacle to the entry of transforming DNA into
the cells. Secretins (e.g., PilQ in N. gonorrhoeae, a
betaproteobacterium) form a channel and allow the
transforming DNA to cross the outer membrane. Other
mechanisms involve small membrane vesicles, called
transformasomes, protruding on the external surface of the
cell. The transforming DNA is captured by these vesicles
and transported into the periplasmic space. Some
microorganisms are able to filter selectively the incoming
DNA during its translocation. This selective step involves
the recognition of specific 10 base pair sequences called
uptake signal sequences (USS), such as 5
0 -
GCCGTCTGAA-3
0
in
N. gonorrhoeae
or
5
0 -
AAGTGCGGT-3
0 in Haemophilus influenzae, a gammaproteobacterium (Danner et al. 1980; Goodman and Scocca
1988). USS are highly repeated in the genomes of the species
carrying them, whereas according to their relative large size,
the expected occurrence of a given USS in any genome is
low. For instance, 1,900 copies are present in
N. gonorrhoeae and 1,471 in H. influenzae (Smith et al.
1995). Their overrepresentation in these genomes suggests
a positive selection pressure to maintain a high number. It
has been shown that the recipient cells translocate preferentially transforming DNA carrying USS similar to their own
USS. This suggests that USS recognition systems exist. Even
if they have not been identified yet, in N. gonorrhoeae their
activity appears to depend on the pilus-coding genes (Chen
and Dubnau 2004). It is important to note that the existence
of USS seems to be an exception rather than a rule because
many microorganisms are able to translocate efficiently
transforming DNA regardless of its sequence. The entry
446
C. Brochier-Armanet and D. Moreira