Bacterial Transformation

History

Transformation was first demonstrated in Streptococcus pneumoniae (a.k.a. Pneumonococcus or Diplococcus) by Fred Griffiths in 1928. In 1944, Avery, McCarty and MacLeod proved that DNA was the transforming principle.

[6.1] 6-1

 

"transformation is genetically visible only if it leads to modification of at least one of the characters of the recipient cells."

 

Bacterial strains such as those used by Griffiths and Avery have NATURAL COMPETENCE, i.e. they have the ability to take up DNA from the medium. Natural competence is a genetically programmed physiological state. Natural transformation is distinct from artificial transformation by techniques such as electroporation, protoplast formation, and microprojectiles. In addition, some bacterial strains, such as E. coli, can be made ARTIFICIALLY COMPETENT using CaCl₂ and heat shock treatment.

The process of transformation in both gram negative and gram positive bacteria is broadly similar. In both cases, DNA fragments must first bind to the exterior of the cell. They are then fragmented which may occur concommitantly with uptake into the periplasm in gram negative bacteria. Finally, one DNA strand is taken up into the cytoplasm while the other is degraded. Recombination appears to be quick in gram negative bacteria but somewhat slower in gram positive bacteria.

[Dubnau-1]

 

Natural transformation in gram positive bacteria

It is not known if transformation is a natural phenomenon in all bacteria -- it may be widespread in nature but it can be difficult to observe in the laboratory. As mentioned before, it was discovered first in the gram positive Streptococcus pneumoniae. It has since been observed and characterized in the gram negative Haemophilus influenzae and in the gram positive Bacillus subtilis.

Cells need to be competent to take up DNA from the external milieu. In gram positive bacteria this requires the presence of a DNA-binding protein on the surface of the cell. The presence of this protein is correlated with nutritional shift-down -- i.e. when the cells start to run out of nutrients.

[6.5] 6-6

In Bacillus subtilis and Streptococcus pneumoniae,a complex of 3 - 5 proteins including

• a labile competence factor
  
  
• a specific endonuclease
  
  
• the DNA-binding polypeptides
  
  
• an autolysin to increase cell permeability
  

are required for transformation. The competence complex is exposed by autolysin and can then bind to double-stranded DNA fragments. Only dsDNA can be used to transform cells. If ssDNA is used, no tranformants are observed.

Bound fragments are digested by an endonuclease into fragments of size ~15 Kbp and then an exonuclease degrades one of the two strands as the other enters the cell (it is protected by ssDNA binding proteins from further degradation). The endonucleolytic reaction may be required to position the end of the DNA fragment correctly so that it can be imported into the cell.

[J10.4] [P16.1]

The resulting ssDNA is recombined into the host chromosome by some sort of strand displacement mechanism.

The existence of a ssDNA intermediate within the cell has been inferred from the fact that the cell enters an ECLIPSE period after it has been transformed.

[6.7] 6-8

 

THE ECLIPSE PHASE

In theory, once a cell has taken up DNA -- containing a specific marker -- from the medium then it should be possible to isolate total DNA immediately from the newly transformed cells and then use that DNA in a second transformation -- selecting for the same marker.

However, in naturally transformed B. subtilis cells, this is not possible -- no successful transformants will be found.

They will only be found if one waits a period of time before isolating total DNA to carry out the second transformation.

During transformation of B. subtilis, DNA from the medium is taken up as ssDNA molecules -- but B. subtilis cannot be transformed with ssDNA. So, until the ssDNA is converted into dsDNA as a result of recombination with the host chromosome, it will not be possible to obtain any transformants for the selected marker.

The eclipse period is the time required to convert ssDNA into a stable dsDNA form.

 

Genetics of transformation

Natural tranformation requires the expression of the late competence genes whose products mediate DNA binding and uptake:

ComC is a peptidase which cleaves ComG so that it is no longer an integral membrane protein thereby activating it.

There are seven ComG proteins all of which resemble the protein pilin and which may form a structure that permits access of DNA to ComEA.

ComEA is the receptor that binds DNA for import.

 

from: Inês Chen and Emil C. Gotschlich (2001) ComE, a Competence Protein from Neisseria gonorrhoeae with DNA-Binding Activity Journal of Bacteriology 183 (10): 3160-3168.

ComEC may form an aqueous transport channel through which DNA enters the cell.

ComFA appears to be a helicase which functions in concert with ComEA and ComEC so that it is ssDNA that enters the cell.

[Dubnau-2]

During transformation, dsDNA is thought to bind to the C-terminal domain of ComEA which then delivers the fragment to the active site of the endonuclease (now thought to be NucA). The terminus of the freshly cleaved fragment is then delivered to ComEC and ComFA for transport.

Transcription of all the late competence genes encoding the DNA binding and uptake machinery (comC, comE, comF, comG) as well as genes necessary for recombination (recA, addAB) requires the competence transcription factor, ComK.

 

 

ComK is a transcription activator

ComK positively regulates its own expression. Expression of comK is also subject to a complex regulatory network which includes AbrB, ComA, SinR, and MecAB.

In a set of experiments published in 1998, Leendert W. Hamoen and colleagues have shown that ComK recognizes short A/T-rich sequences arranged in a unique, flexible pattern along the DNA helix. [LINK]

First, they showed first that ComK is sufficient to activate transcription at the comG promoter.
DNase I footprinting analysis of six ComK activated promoters (comC, comG, comE, comF, addAB, and comK) did not give them a clear idea of the ComK recognition sequence.
They determined that ComK did not bend DNA when it binds to it - suggesting that a specific interaction is required.
Gel mobility shift analysis using a mixture of native ComK and a recombinant fusion of ComK with MBP, they showed that four molecules of ComK bind at each promoter.
Hydroxy-radical footprinting analysis of ComK binding to the addAB promoter provided a more detailed look at the binding region and allowed them to conclude that ComK binds to an AT rich sequence AAAAN5TTTT.
All of the ComK activated promoters have two copies of the ComK AT box. each box is recognised by a ComK dimer; full activation requires interaction between two dimers. Deletion analysis has confirmed that box AT boxes are required for activation.
An experiment in which the two boxes were moved away from one another by multiples of 5 bp showed that the distance between the AT boxes is significant. They must both be on the same face of the DNA helix. This result also helps to explain why ComK by itself is a relatively poor activator of its own promoter - the two AT boxes are quite far apart.

The comK promoter is negatively regulated by binding of AbrB and CodY and positively activated by binding of AbrB, SinR, and DegU.

 

• CodY is a GTP-binding protein that senses the intracellular GTP concentration as an indicator of nutritional conditions and regulates the transcription of early-stationary-phase and sporulation genes, allowing the cell to adapt to nutrient limitation. [Manoja Ratnayake-Lecamwasam, Pascale Serror, Ka-Wing Wong, and Abraham L. Sonenshein (2001) Bacillus subtilis CodY represses early-stationary-phase genes by sensing GTP levels. Genes Dev. 15(9); 1093-1103.]
  
  
• DegU helps ComK to bind more tightly to the comK promoter. This is important at low concentrations of ComK as it would be unable to activate its own expression adequately.
  

  	DNase footprinting analysis of the comK promoter in the presence of DegU and ComK - from: Leendert W. Hamoen, Aske F. Van Werkhoven, Gerard Venema, and David Dubnau The pleiotropic response regulator DegU functions as a priming protein in competence development in Bacillus subtilis Proc. Natl. Acad. Sci. USA. 97(16): 9246-9251, 2000.

The following figure summarises the control of ComK expression.

FROM Leendert W. Hamoen, Aske F. Van Werkhoven, Gerard Venema, and David Dubnau The pleiotropic response regulator DegU functions as a priming protein in competence development in Bacillus subtilis Proc. Natl. Acad. Sci. USA. 97(16): 9246-9251, 2000.

 

Regulation of ComK stability

The level of ComK in the cell is controlled directly by the activity of MecA and the ClpCP protease.

ClpC is a chaperone-like protein of the Hsp100 family which assists the protease, ClpP.

MecA binds with ComK and presents it to ClpCP for degradation.

However, this activity is blocked by ComS, a short protein containing 46 aas which is expressed within the srf operon whose gene products function to express surfactin synthetase.

[LINK]

 

ComK then activates transcription of the late competence genes as well as the anti-sigma factor, FlgM.

FlgM binds directly to s²⁸ (or s^D) and prevents it from interacting with the RNA polymerase core complex. s²⁸ is required for transcription of the flagellar operons.

FlgM is an unusual polypeptide because in its native form it has very little secondary structure. When the C-terminal portion of FlgM binds with the C-terminal region of s^D, then it becomes structured.

 

FlgM, as the name suggests, was originally identified as a regulatory gene involved in production of flagellae. Its action to oppose production of flagellae in response to activation of competence makes sense.

ComK also activates expression of recA so that there is a supply of this protein in the cell to promote strand exchange in the homolgous recombination reactions that must occur during transformation. It has recently (Nov 2001) been shown that ComK does this without displacing the LexA repressor. This activation of recA is independent of the SOS response.

When cells are faced with a shortage of nutrients they must adapt. In general, adaptation must be gradual since shortages are gradual and progressive. Cells therefore respond in stages. Induction of flagellar synthesis is a less drastic response than competence or sporulation and permits the cells to swim to new sources of nutrients. If they are not available and more drastic measures are necessary then the cell turns first to competence and then to sporulation as a survival mechanism. Note, however, the discussion below which holds that competence is most likely not a mechanism to aquire nutrients.

 

Competence Factors

Two small oligopeptide competence factors are known that regulate the entire process.

 

• The ComX pheromone is a 10 aa peptide with the sequence IYTNGNWVPS in which the tryptophan residue has been modified.
  
  
• Csf (Competence and Sporulation Factor) is an unmodified 5 aa peptide with the sequence ERGMT.

Both peptides stimulate expression of the srf operon. Their action is mediated through the phosphorylation of ComA.

[6.2] 6-3 ***

ComA~P is a transcription activator; ComA is not. Phosphorylation is catalysed by the ComP kinase. Removal of the phosphate group is catalysed by RapC, an aspartyl phosphate phosphatase.

ComX acts by binding to the external domain of ComP which is an integral membrane protein. This binding activates the cytoplasmic domain which is a histidine kinase. ComA interacts with ComP~P to become phosphorylated.

ComP/ComA form two components of a bacterial signal transduction or sensory system. There are many such systems in bacteria all of which function in very similar ways.

Csf blocks the activity of the RapC phosphatase. This has the effect of increasing the intracellular concentration of ComA~P thereby stimulating comS and ultimately comK expression.

ComQ, ComX and ComP form a species-specific competence recognition system. ComQ produces the correct ComX that is recognised by the host ComP and not by the ComP proteins of other Bacillus species.

 

This complex interplay is summarized in the following figure:

 

• If ComX accumulates in the growth medium and then binds to the external input domain of of the sensor protein ComP, it activates autophosphorylation of ComP which, in turn, phosphorylates ComA which can then activate expression of the srf operon. This operon is not specifically required for competence, however it also contains the comS gene which is. ComS is a small polypeptide (46 aas) which indirectly activates ComK which then activates expression of the late competence genes.
  
  
• ComK is inactivated by MecA/ClpCP. ComS prevents this.
  
  
• MecA is also required for optimal expression of s^D dependent genes.
  
• FlgM is an anti-sigma factor which binds to s^D, which is required for transcription of the flagellar operons, and prevents it from associating with core RNA polymerase.
  
  
• If the competence stimulating factor accumulates in the growth medium, it is transported into the cell via the Spo0K (aka Opp) protein. It blocks the activity of RapC phosphatase thereby increasing the amount of activate ComA in the cell.
  
  
• Expression of the SinR (Sporulation inhibition) protein prevents sporulation and permits competence. It forms a tetramer that functions as a repressor. Binding of SinI to SinR overcomes the repressive activity of SinR. The structure of the SinR monomers is remarkably similar to the structure of the bacteriophage l repressors CI and Cro suggesting an evolutionary relationship between the two adaptive responses - prophage induction and sporulation.
  
  .

Regulation by the ComP/ComA proteins is an example of a two-component regulatory system -- such systems are now quite common in bacteria. DegS and DegU also form a two-component regulatory system. Note that it is the unphoisphorylated form of DegU that regulates comK; but it is the phosphorylated form that acts in the DegS/DegU system.

[Box12-1] [P20-16] [P20-17] [P20-18] [P20-20]

 

Expression of Csf

Csf is encoded by the phrC gene. It is synthesized as the 40 aa precursor protein, PhrC. PhrC contains a signal sequence which suggests that is is secreted via a Sec-dependent pathway. Removal of the signal sequence leaves a peptide of 11-25 amino acids. It is not yet known how the final pentapeptide is produced.

The fact that Csf is formed only after it is secreted from the cell makes sense. If it were formed inside the cell, then the internal regulatory circuits would quickly be swamped by the peptide. Secretion and external processing ensures that only when a high cell density is achieved is there much Csf to enter the cell.

 

The phrC gene is part of an operon that also contains the rapC gene. Expression of this operon is subject to positive and negative regulation by both of its gene products. In both cases, control is indirect and is mediated by ComA~P.

 

Expression of the operon is activated by ComA~P. Once the RapC protein is translated, however, it dephosphorylates ComA~P, thereby negatively regulating its own expression.

However, PhrC is also translated, exported and processed into Csf. As the concentration of Csf increases, it enters the cell, inhibits the activity of RapC, permitting ComA~P to remain phosphorylated, thereby positively regulating its own expression.

 

Csf and Sporulation

Sporulation and the development of competence appear to be alternative strategies pursued by the bacteria in response to nutrient deprivation. Csf plays a role in regulating gene expression in both pathways.

At low concentrations (1-5 nM) Csf stimulates srf (and hence comS) expression by inhibiting the RapC phosphatase.

At higher concentrations (>20 nM) Csf inhibits srf expression and stimulates sporulation. Inhibition of srf expression is hypothesized to occur by means of inhibiting the ComP kinase activity. This would prevent formation of ComA~P which would block activation of all of the downstream genes. Stimulation of sporulation arises because Csf inhibits the activity of the RapB phosphatase which dephosphorylates Spo0F~P, a protein that is an essential component of the phosphorelay that controls sporulation.

As cells enter stationary phase, a second mode of expressing Csf becomes significant. The phrC gene can also be expressed from a s^H dependent promoter located within rapC upstream of phrC. As cells enter stationary phase, the concentration of s^H increases. This boosts the external concentration of Csf and prompts cells to sporulate rather than induce competence.

The key genes for the initiation of the sporulation response are Spo0A, s^H and AbrB. Their expression is required for stage II of sporulation.

 

Spo0A is activated by a response regulatory pathway similar to the ComP/ComA pathway.

[P25.2]

• One of a number of histidine kinases inside the cell is activated and phosphorylated.
  
  
• The phosphorylated Kinase interacts with Spo0F which becomes phosphorylated at an aspartic acid residue.
  
  
• Spo0F~P interacts with Spo0B and transfers the phosphate moiety to an histidine residue.
  
  
• Spo0B~P interacts with Spo0A to phosphorylate it.
  
  Spo0A~P is both an activator and a repressor of transcription. It activates the expression of a number target genes that are essential for sporulation. In particular, it activates expression of s^E in the mother cell and s^F in the forespore. s^F is replaced by s^G later in development, and s^E is replaced by s^K.

 

[See Biochemistry 3107 lecture notes on Bacterial RNA Polymerase for more on the sigma subunit.]

[P25.5]

 

As mentioned above, Csf influences the pathway of Spo0A activation. RapB is a phosphatase which dephosphorylates Spo0F~P. When that happens, the signal is no longer transduced to Spo0A preventing or reducing the sporulation response. By blocking the activity of RapB, Csf helps to ensure that Spo0A is activated.

Spo0A also plays an indirect role in regulating the expression of s^H. It acts as a repressor at the abrB promoter thereby inhibiting the expression of AbrB which, in turn, is a repressor of s^H expression. If AbrB is not continually expressed then its concentration falls and repression is lifted. Recent (Dec 2000) evidence indicates that AbrB is a novel type of DNA binding protein that may conform to its DNA binding sites.

The Kinase/Spo0F/Spo0B/Spo0A system is another example of a bacterial two-component regulatory system although in this case it contains four rather than two components.

 

 

Natural transformation in gram negative bacteria

Natural transformation has been observed in some gram negative bacteria. Transformation in Haemophilus influenzae, the first gram negative bacterium in which natural competence was found, is different at least two important respects from that in B. subtilis.

[6.6] 6-7

 

• DNA uptake is associated with the formation of small membraneous structures, called transformasomes, which protrude outside the cell. The transforming DNA is taken into these vesicles where it is then internalized into the cell. One of the two strands is degraded while the remaining strand may recombine with the host chromosome.
  
  
• Unlike gram positive bacteria, DNA uptake in gram negative bacteria appears to require or involve the recognition of specific sequences. The sequences or some bacteria are as follows:
  

H. influenzae	AAGTGCGGTCA
N. gonorrhoeae	GCCGTCTCAA

Note that 10-11 bp is close to the minimum length sequence necessary for specificity in sequences. Sequences this length or longer are unlikely to occur at random in the typical prokaryotic genome.

[Dubnau-3]

 

Artificial transformation

Artificial transformation has been demonstrated in a number of bacterial species, most notably in E. coli, where it is used routinely for cloning DNA. However, even in E. coli, the process or mechanism is not well understood.

E. coli cells can be made competent for transformation simply by treating them with calcium chloride. But, even this simple treatment is a bit of an art and variables such as the temperature of growth and the density of the culture seem to be important in determining competence.

Why do bacteria have natural competence?

NUTRITION

Uptake of DNA would provide bacteria with a source of nucleotides. However, this seems unlikely to be an efficient method of nutrient uptake. It would seem more efficient to catalyse complete breakdown of DNA in the extracellular environment and then to use normal import channels to recover the nutrients.

REPAIR

Uptake of potentially homologous DNA would allow repair of damage. This seems consistent with the requirement for specific DNA uptake sequences which identify DNA fragments as 'self'. However, there is little evidence that competence systems are induced by damage to DNA.

[6.9] 6-10

DIVERSITY

Uptake of DNA from other strains offers the potential for increasing genetic diversity by uptake and incorporation of non-homologous DNA.

David Dubnau (1999) DNA UPTAKE IN BACTERIA. Annu. Rev. Microbiol. 53:217-244.

The Bacillus subtilis genome sequence website will