A rapid method to construct gene cassette libraries enriched with first gene cassettes and an associated screening method for the clone selection

First Gene Cassettes of Integrons as Targets in Finding Adaptive Genes in Metagenomes
Lionel Huang, Christine Cagnon, Pierre Caumette, and Robert Duran
Applied and Environmental Microbiology, 2009, 75（11）：3823–3825

Here, I will introduce a short publication focusing on a very narrow specific problem. This paper propose a rapid method for the selection of clones carrying an integron first gene cassette that is useful for finding adaptive genes in environmental metagenomic libraries.

Integrons are genetic structures capable of capturing and excising gene cassettes, which usually encode adaptive proteins in different environmental contexts, such as genes for degradation of pollutants, and resistance to antibiotics or heavy metals. Thus, environmental pressures may favor the propagation of cassettes conferring a selective advantage.

Integrons are not a new and hot topic, but there is an increasing interest in finding of adaptive genes associated with integrons among the huge metagenomes. The integration of a new gene cassette, catalyzed by the integrase, occurs by recombination between the attC site and the attI site of the integron. The first gene cassette of an integron is, therefore, the last one integrated. In previous studies, the determinations of gene cassette collection from environmental metagenomes did not target first gene cassettes, since they were performed by PCR methods targeting attC sites. As the first gene cassette is the closest gene to the promoter, its expression level is the highest in the integron. Thus, it is a good target to find new adaptive genes in metagenomes.

The method was developed by using a pure strain isolate, Xanthomonas campestris ATCC 33913T, which carry an integron possessing 23 gene cassettes. One environmental sample, coastal sediment, was used to validate the method. There are two key points in developing this method, one is the construction of first gene cassettes libraries. To amplify the first gene cassettes, a forward primer targeting the intI gene or attI site must be used. Forward primer AJH72 was used for PCR of X. campestris DNA, and primer ICC48 (intB-inverted primer), targeting the class 1 integron intI, was used for PCR of sediment metagenome. A less-degenerated primer ICC21, was designed to target the attC sites from class 1 and 2 integrons. The other key point is the trick for clone selection. Due to the particular structure of the attC site with inverted repeat sequences, the reverse primer was also used in the forward direction. As a result, several amplified fragments were obtained, and the sequence analyses revealed that most of them were gene cassettes other than the first one. Here, the authors develop a triplex PCR method by labeling the forward primer with HEX (6-carboxyhexafluorescein). The fluorescent PCR fragments were selected for further sequencing.

This method was then applied to coastal mud metagenomes, and 23 fluorescent fragments were detected and sequenced. A total of 29 open reading frames (ORF) were characterized as potentially transcribed by an integron promoter. The first-gene cassettes of integrons appear to be good candidates to find gene cassettes, which aid bacteria in effecting a rapid adaptive response. We are now able to reveal integron last gene acquisitions of environmental bacterial communities submitted to stressful conditions. The PCR method combined with the screening method leads to 100% of clones carrying a first gene cassette. Thus, this new method allows the focus to be on spreading first gene cassettes in metagenomes after a specific stress.

In this paper, the authors propose a good idea to construct gene cassette libraries enriched with first gene cassettes and an associated screening method for the clone selection. However, the only one drawback in this paper is that the relationship between the function of ORF and the oil degradation should be discussed further. Since the first gene cassettes was enriched by adding oil into the coastal sediment, it should be served as the environmental pressure for selecting adaptive genes. I have checked the information from EMBL, where the author deposited the sequences, but still no related information in the database.



Hui Li, PhD
University of Idaho

DNA transfer proteins of broad-host-range plasmid can mediate chromosomal DNA transfer

The R1162 mob proteins can promote conjugative transfer from cryptic origins in the bacterial chromosome. Richard Meyer (2009) J. Bacteriol. 191: 1574-1580


It is well-known that plasmids mediate horizontal gene transfer among bacteria and play major roles in the rapid spread of antibiotic resistance. Generally, plasmid-mediated horizontal gene transfer from one bacterial chromosome to another requires the help of transposons as follows; in the first step, a transposon that has captured chromosomal genes moves into the plasmid; in the second step, the plasmid moves into a recipient cell by conjugative transfer; and in last step the transposon on the plasmid jumps into the chromosome of the recipient cell.

This paper showed that the above-described scheme is not the only way that plasmids can participate in the horizontal transfer of chromosomal genes. The author found that a plasmid can directly transfer donor chromosomal DNA into the recipient chromosome.


In a canonical model, conjugative DNA transfer starts with the DNA cleavage at the origin of transfer "oriT", a unique site on the plasmid, which is mediated by a protein called "relaxase". Previously, the author found that relaxase of the broad-host-range plasmid R1162 (also called RSF1010) can initiate DNA transfer at several sequence variants of oriTR1161. The observed promiscuous activity of the relaxase raised the possibility that plasmids can directly mediate the transfer of chromosomal DNA from oriT-like sequences in a chromosome (Jandal S. and Meyer R., 2006).


Draper et al. (2005) showed that relaxase can mediate recombination between two directly repeated oriTs on the transferred DNA in the recipient cell. This nature of the relaxase was used in the experiment to test the hypothesis that relaxase can initiate DNA transfer from cryptic oriT in a chromosome. A plasmid that has the original oriT and a selectable drug-resistance gene marker was artificially integrated into the downstream region of one of the candidate oriT sites in the chromosome of donor strain Pectobacterium atrosepticum to make directly repeated oriT sites. If single-strand DNA is branched out from the candidate oriT and moves into recipient cells, the transferred DNA would become a plasmid that carries a hybrid oriT comprised of the candidate oriT and the original oriT due to the recombination activity of the relaxase. As the author expected, the plasmid that has a part of the donor chromosome and a hybrid oriT was obtained in the recipient E. coli. This result indicates that the initiation of transfer does happen at the candidate oriT in P. atrosepticum.


The next questions are "What length of DNA is it possible to transfer?" and "Is the transferred chromosomal DNA integrated into chromosome in the recipient cell?" To answer these questions, the author used an E. coli strain as donor strains, that has a drug-resistance gene marker at a particular location in the chromosome. In the presence of helper plasmids that express relaxase and pillus proteins, the drug resistance gene marker was transferred and integrated into the chromosome of recipient E. coli. The candidate oriT closest to the resistance gene was 40 kbp away from the resistance gene in the donor chromosome. It suggests that a chromosomal DNA fragment of at least 40 kbp was transferred in the mating process. Surprisingly, even when the closest oriT was eliminated from the donor chromosome, the transfer of the drug resistance maker was observed with almost the same frequency as it was in the presence of the closest candidate oriT. This result suggests that chromosome transfer can be initiated at multiple cryptic oriT sites in the donor chromosome. Given that the second closest candidate oriT is 708 kbp away from the resistance gene, a DNA fragment of at least 708 kbp was indicated to be transferable in this experiment.

The authors estimated that there are 10 candidate oriT sites in the P. atrosepticum chromosome and 8 in the E. coli chromosome, which could be active in the presence of R1162 relaxase. Although it is still not clear how many plasmids have a potential to mobilize fragments of the chromosome, this article clearly showed that there is a novel manner of horizontal gene transfer.


Bacteriophage are also known to mediate the transfer of host's DNA in the manner called "general transduction", where host's DNA are accidentally packed in phage's capsid and are introduced into new host cells. Relaxase-mediated chromosomal DNA transfer resembles phages' general transduction, but different in that the length of transferable DNA is not limited in the relaxase-mediated chromosomal DNA transfer; the size of transferable DNA is limited in general transduction due to the limited size of phages' capsid. Given the size of transferable DNA, it seems that plasmids play much more important roles in bacterial evolution than bacteriophages.



References:

Meyer R. (2009) The R1162 mob proteins can promote conjugative transfer from cryptic origins in the bacterial chromosome. J. Bacteriol. 191: 1574-1580

Jandle S. and Meyer R. (2006) Stringent and relaxed recognition of oriT by related systems for plasmid mobilization; implications for horizontal gene transfer. J. Bacteriol. 188: 499-506

Draper O., César C. E., Machón C., de la Cruz F., and Llosa M. (2005) Site-specific recombinase and integrase activities of a conjugative relaxase in recipient cells. Proc. Natl. Acad. Sci. USA 102: 16385–16390.



H.Yano. University of Idaho

Plasmid Capture by the Bacillus thuringiensis Conjugative Plasmid pXO16

Plasmid Capture by the Bacillus thuringiensis Conjugative Plasmid pXO16
Sophie Timmery, Pauline Modrie, Olivier Minet, and Jacques Mahillon
Journal of Bacteriology, April 2009, p. 2197-2205, Vol. 191, No. 7


In this article authors describe the ability of Bacillus thuringiensis plasmid pXO16 to transfer as well as mobilize three other plasmids. It is very important issue to study plasmid transfer in Gram positive bacteria as this process is not as well known as plasmid transfer in Gram negative bacteria. Gram positive bacteria are very important as they can be as deadly as Bacillus anthracis – the causative agent of anthrax, but also can be opportunistic pathogens associated with food, like Bacillus cereus. The widely used B. thuringiensis plays very important role as the source of insecticidal toxins. Other Gram positive bacteria like Streptococcus sp. and Staphylococcus sp. are also pathogenic for human. The presence and horizontal transfer of mobile genetic elements that can play a role in spreading an antibiotic resistance as well as pathogenic determinants is very important issue.
So this paper gives us some information about pXO16 plasmid transfer and its ability to mobilize other plasmids. I would like to point out some parts of this paper. First, I was intrigued by the title of this paper. The title could suggest that pXO16 can capture other plasmids DNA molecules and incorporate into its own DNA. From the second sentence we can figured out what authors have on mind talking about capture of other plasmids by pXO16, and that it is the mobilization and retromobilization of other plasmids to the host cells by this Bacillus thuringiensis plasmid.
In the introduction section authors presented very briefly the “state of art” in the plasmid transfer and mobilization field. The special interest is put on retromobilization which is very interesting event. Retromobilization occurs when recipient DNA, either plasmid or chromosomal markers are transferred to the donor during conjugation. This reciprocal DNA transfer is very interesting and it was discussed as it is against the unidirectional DNA transfer rule. The schematic representation of retromobilization models on Fig. 1. is very easy and easy understandable. In the results section we have a very nice story about pXO16 plasmid transfer, and mobilization of “mob” plasmids pUB110, and pE194, as well as no mobilizable plasmid pC194. One thing that bothers me is the way to present conjugation efficiency as the transconjugant to recipient ratio. I know that it is the way, but it does not show overall number of donor, recipient and transconjugant cells in the conjugation mixture and I personally do not like it. In the triparental matings with the plasmid free cells as recipients only two plasmids were used, pUB110 and pC194. Generally lower frequencies of plasmid transfer were observed in triparental than in biparental matings. Authors described also the ability of pXO16 to mobilize pUB110 in three different “media” cow, soy and rice milk and showed the influence of media on mobilizable plasmid transfer. It was also shown that plasmid pXO16 can be used to capture the pUB110 plasmid from other bacteria in all 3 kinds of milk. This is quite interesting that food products can be used as media in this kind of experiments and that biological events can occur in such an environment. We should consider this and look on the food quality especially on expiration date in “milk related” products.
Very interesting results are presented in the section describing plasmid transfer kinetics where it is shown that the plasmid transfer occurs only in a short period of time reaching plateau after certain period of time. It is consistent with other results and proves that plasmid transfer is controlled by specific mechanisms in the cell and these functions are not the only plasmid related.
In discussion authors focused on the mode of retrotransfer and showed that retrotransfer in the case of B. thuringiensis pXO16, and plasmids used in the paper support the two step model. It is interesting as we always thought that we need two kinds of cells, donor and recipient for conjugation to occur. And now we have two kinds of cells but both contain the same conjugative plasmid. Mobilizable plasmid can be transfer only if two cells form mating pair. Mobilizable plasmid encode no functions related to mating pair formation so it means that two “donor” cells can form mating pair and maybe reciprocally transfer DNA molecules. That is really exciting…
To summarize I found this paper quite interesting but some weak points, like poorly written materials and methods, as well as graphical representation of results depreciate the quality of this work and should be corrected before publication.

Jarek Krol, PhD
UofI

A new Family of Gram-positive bacterial plasmids

Weaver, K. E., Kwong, S. M., Firth, N. & Francia, M. V. (2009).The RepA_N replicons of Gram-positive bacteria: a family of broadly distributed but narrow host range plasmids. Plasmid 61, 94–109.

Considering the rate at which sequence databases, and therefore our knowledge of existing plasmids, are growing, it is becoming increasingly important to organize and categorize plasmids into relevant groups. Proper organization of known plasmids would allow us to unravel evolutionary histories and make more efficient the process of integrating our current knowledge. In this article, Weaver et. al. propose to create a family of plasmids characterized by RepA_N, a highly conserved domain in the initiator protein.

Unlike most other plasmid classification systems, this family is not based upon the incompatibility of two plasmids due to similar replication machineryc. Indeed, many of the plasmids within the RepA_N family can stably coexist within a single host bacterium (Kwong et. al. 2008). Rather, the determining characteristic of this plasmid is this conserved initiator protein domain. Phylogenies based on RepA_N matched those of each plasmid’s hosts. What’s more, members of this family are narrow host range plasmids but are found in a diverse range of low G+C gram-positive bacteria (Firth et al., 2000). This suggests that RepA_N family plasmids were present in ancestral gram-positive bacteria of low G+C content and then proceeded to diverge with individual hosts at an early split in the host evolution.

When compared to phylogenies based on other protein domains the modular nature of plasmid evolution becomes apparent. Phylogenies based on RepB, for instance, do not match with host phylogenies or those of RepA_N. RepB is just one among many examples of how plasmids can acquire complete, functional units of DNA from various sources throughout their evolutionary histories. The authors site the specific examples of the replication, partition, and conjugative components of RepA_N plasmids as evolving by “shuffling” between various other plasmids that are found within the same host. Again, RepA_N serves as a good classification marker in that it is conserved within each host and matches its host’s phylogeny.

This article touched on several points that bear particular attention. To begin, the authors point out that this sort of a study cannot be effective without a certain volume of raw data in sequence databases, which was not feasible even a few years ago but is now available and growing. Secondly, with such data evolutionary histories of plasmids and their coevolution with their bacterial hosts can be elucidated and that such information is vital to our understanding of why plasmids are distributed as they are today (e.g., how a family of plasmids can be both broadly distributed and only stably transferred to and maintained in a narrow host range). Finally, this study provides several excellent examples of the modular nature of plasmid evolution. Hopefully in the future available information on previously uncharacterized plasmids will continue to grow and will continue to be organized such that more such evolutionary insights may be made in the future.

Citations:

Kwong, S.M., Lim, R., LeBard, R.J., Skurray, R.A., Firth, N., 2008. Analysis of the pSK1 replicon, a prototype from the staphylococcal multiresistance plasmid family. Microbiology 154, 3084–3094.

Firth, N., Apisiridej, S., Berg, T., O’Rourke, B.A., Curnock, S., Dyke, K.G.H., Skurray, R.A., 2000. Replication of staphylococcal multiresistance plasmids. J. Bacteriol. 182, 2170–2178.


Julie Hughes
Graduate Student
Department of Biological Sciences
University of Idaho

Bacterial Toxin–Antitoxin Systems: More Than Selfish Entities?
Laurence Van Melderen and Manuel Saavedra De Bast

Bacterial Toxin-Antitoxin systems are diverse and widespread in the prokaryotic world. They are composed of two components, a toxin that can harm the host and its corresponding antitoxin that is needed by the host to prevent cell death. TA systems that are found on chromosomes are hypothesized to have been acquired by horizontal gene transfer. Some bacteria are known to have around 50 putative TA systems such as Nitrosomonas europeae, and Sinorhizobium meliloti. Others have none or a few TA systems. Plasmid encoded TA systems act as addiction modules and help in maintaining plasmid-containing cells. Thus, while the function of plasmid encoded TA systems is well known, those found on chromosomes are not as well understood. There are several theories on the physiological roles of chromosomal TA systems. Following are some of these proposed models:
1) The programmed cell death model: This model is based on the chromosomally located mazEF system of E. coli [1]. Under conditions of stress such as amino acid starvation, high temperature or presence of antibiotics, transcription of mazEF is affected. This is followed by degradation of MazE (antitoxin) by an ATP-dependent protease and subsequent toxification by previously produced MazF (toxin). This in turn leads to cell death.
2) The growth modulation model: This model is based on the relBE system of E coli [2,3]. The primary difference between this model and the previous one is that this model proposes cell growth inhibition under conditions of amino acid starvation and not cell death.
3) The developmental model: This model was proposed for the toxin gene in Myxococcus xanthus [4] an organism that forms fruiting bodies under nutrient starved conditions. The genome of this organism has a homologue of the mazF toxin gene. During fruiting body formation, MazF protein is produced which induces cell death. In fact, nearly 80% of the cells that undergo fruiting body formation die by lysis. However, MazF has also been found to be essential for fruiting body formation.
4) The stabilization model: The model proposes that TA systems could help stabilize some regions of the genome that are unstable and prone to being lost [5]. Such TA systems are often found on structures called super integrons that carry many essential and non-essential genes. The TA systems stabilize the super integrons as well as unstable plasmids or genomic regions.
5) The anti-addiction model: This model proposes that chromosomal TA systems can benefit their hosts during post seggregational killing [6]. The chromosomal TA system of Erwinia chrysanthemi 3937 was found to prevent post seggregational killing of bacterial cell after loss of plasmid.
The above models show the different ways in which TA systems can confer a selective advantage to their hosts.
Thus TA systems on chromosomes are diverse and have evolved multiple roles from being simple addiction modules to more complex systems involved in cell physiology.
References:
[1] Engelberg-Kulka H, Amitai S, Kolodkin-Gal I, Hazan R. Bacterial programmed cell death and multicellular behavior in bacteria. PLoS Genet. 2006;2:e135. doi:10.1371/journal.pgen.0020135.
[2] Christensen SK, Mikkelsen M, Pedersen K, Gerdes K. RelE, a global inhibitor of translation, is activated during nutritional stress. Proc Natl Acad Sci U S A. 2001;98: 14328–14333.
[3] Christensen SK, Pedersen K, Hansen FG, Gerdes K. Toxin-antitoxin loci as stress-response-elements: ChpAK/MazF and ChpBK cleave translated RNAs and are counteracted by tmRNA. J Mol Biol. 2003; 332:809–819.
[4] Nariya H, Inouye M. MazF, an mRNA interferase, mediates programmed cell death during multicellular Myxococcus development. Cell. 2008;132:55–66.
[5] Rowe-Magnus DA, Guerout AM, Biskri L, Bouige P, Mazel D. Comparative analysis of superintegrons: engineering extensive genetic diversity in the Vibrionaceae. Genome Res. 2003;13:428–442.
[6] Saavedra De Bast M, Mine N, Van Melderen L. Chromosomal toxin-antitoxin systems may act as antiaddiction modules. J Bacteriol. 2008;190:4603–4609.


DIYA SEN
GRADUATE STUDENT
BIOLOGICAL SCIENCES
UNIVERSITY OF IDAHO

A novel gene module protecting bacteria from phage infection

The phage abortive infection system, ToxIN, functions as a protein-RNA toxin-antitoxin pair. Fineran PC, Blower TR, Foulds IJ, Humphreys DP, Lilley KS, and Salmond GP. (2009) Proc Natl Acad Sci U S A. 106:894-899.

Recently it has become easy to determine the complete sequence of 100 kb long plasmids. However, the ensuing annotation process, assigning a function to a DNA sequence, is still a frustrating process for people who are working on sequence analysis; often, the newly sequenced DNA segment does not show homology to well-known genes which makes difficult to judge whether or not there is a gene in the segment.

Let's hope that no homology is a good sign for a big discovery. Here, I introduce a discovery of the novel gene module on Erwinia carotovora plasmid pECA1039, which protect host bacterium from phage infection.

Erwinia carotovora is a plant pathogenic bacterium that causes rot in diverse vegetables. The author's group has been studying virulence mechanisms of E. carotovora (Barnard A.M. and Bowden S.D. et al., 2007). They identified a probable protein coding sequence, named toxN, through the complete sequence analysis of E. carotovora's 5,620-bp cryptic plasmid., whose product shows 31% amino-acid identity to a protein associated with phage abortive infection (Abi) in gram-positive bacteria (Emond E. and Shirley E.D. et al., 1998). Genes related to Abi generally exert a cellular process that shuts down the phage lytic cycle or that kills phage-infected cells to prevent the phage particle production (Chopin M.C. and Chopin A. et al., 2005). The toxN gene on pECA1039 is adjacent to a potential coding sequence toxI that includes five copies of a 36-bp sequence tandem repeat, followed by an inverted repeat sequence. The potential gene product ToxI exhibits no similarity to proteins in databases. Since the Abi system was not so common among gram-negative bacteria, the authors focused the study on the toxI-toxN region and experimentally showed that the toxI-toxN region confers phage-resistance to the host.

The authors have two major subject to be addressed: one is the mechanism of phage resistance, and another is whether or not toxI encodes a protein.

The authors showed that toxN encodes a toxic protein that inhibits the host's growth, which means that the mechanism of phage resistance might be a growth inhibition induced by phage infection. The authors also found that the co-expression of toxI with toxN can counteract the toxic activity of ToxN. But, the hypothetical protein ToxI seemed to not be produced from the toxI gene region according to Western blotting analysis, using the modified toxI gene fused with a sequence coding for hexa-histidine tag.

It is possible that toxI RNA itself has an activity to counteract ToxN. To test this hypothesis, the authors introduced a translational stop codon into the toxI coding sequence and found that the ToxN-counteracting activity was still retained in the mutant toxI region. Furthermore, point mutations that do not change the amino-acid sequence of ToxI but do change the transcript sequence resulted in the loss of ToxN-counteracting activity. These results suggest that the toxI RNA is responsible for the antitoxin function. They also pointed out that similar gene modules that are comprised of the tandem repeat region and the toxN homologous gene are present in diverse Eubacteria and Archea.

Detailed mechanisms of toxin (ToxN) induction and the interaction between the toxI RNA and the ToxN protein are still unclear. However, it is obvious that the authors identified a novel type of functional RNA. The authors' work proved that we can still discover new biological concepts from plasmid sequences.


References:
Fineran PC, Blower TR, Foulds IJ, Humphreys DP, Lilley KS, Salmond GP. (2009) The phage abortive infection system, ToxIN, functions as a protein-RNA toxin-antitoxin pair. Proc Natl Acad Sci U S A. 106:894-899.

Barnard AM, Bowden SD, Burr T, Coulthurst SJ, Monson RE, Salmond GP. (2007) Quorum sensing, virulence and secondary metabolite production in plant soft-rotting bacteria. Philos Trans R Soc Lond B Biol Sci. 362:1165-1183.

Chopin MC, Chopin A, Bidnenko E.　(2005) Phage abortive infection in lactococci: variations on a theme. Curr Opin Microbiol 8:473-479.

Emond E, Dion E, Walker SA, Vedamuthu ER, Kondo JK, Moineau S. (1998) AbiQ, an abortive infection mechanism from Lactococcus lactis. Appl Environ Microbiol. 64:4748-4756.


posted by H. Yano (University of Idaho)

Disrupting antibiotic resistance propagation by inhibiting the conjugative DNA relaxase

Scott A. Lujan, Laura M. Guogas, Heather Ragonese, Steven W. Matson, Matthew R. Redinbo,
PNAS 2007 vol. 104 no. 30 12282-12287

Since mid sixties when so-called “R factors” were found to spread antibiotic resistance between different bacterial strains, lots of effort has been put on finding the simplest and universal method to stop horizontal plasmid transfer. Many papers described the effect of various substances on conjugation. Many substrates have a detrimental effect on plasmid transfer just by influencing the bacterial propagation and general life functions. Sometimes the results could be quite surprising and unexpected, but some ordinary everyday products like green tea (epigallocatechin gallate), coffee (caffeine), and papaya seed macerate can inhibit conjugation and ipso facto spreading of an antibiotic resistance. The most common approach to this kind of study is just to add the material under investigation to the conjugation mixture and evaluate plasmid transfer efficiency over a time. Conjugation is a very complicated, multistep process. The functions necessary for conjugation are encoded by plasmids and do not depend on the host. Many studies have been done to establish the function of particular genes. One of the most important proteins involved in plasmid transfer is a DNA relaxase. The conjugative relaxase initiates DNA transfer with a site- and strand-specific ssDNA nick in the transferred strand (T-strand) at the origin of transfer (oriT), forming a covalent 5′-phosphotyrosine intermediate. The nicked T-strand moves from the donor cell to the recipient cell via an intercellular junction mediated by a type IV secretion system. The relaxase completes DNA transfer by reversing the covalent phosphotyrosine linkage and releasing the T-strand. In the F plasmid, this relaxase is located in the N-terminal domain of a large multifunctional protein, TraI (DNA helicase I). The relaxase active site contains one or several tyrosine residues: F-like relaxases (found in IncF, IncN, IncP9 and IncW plasmids) contain 2 to 5 tyrosines while relaxases of IncQ, IncP, IncI plasmids and Ti plasmid of Agrobacterium sp. possess only one. F-like relaxases encoded by the traI gene shares significant sequence identity with relaxases of many R plasmids (e.g., 98% with R100 TraI); thus, the F plasmid serves as a model system for examining conjugative plasmids and the inhibition of conjugative transfer.
In this study the authors first describe the role that the relaxase enzyme plays in the initiation and termination of DNA conjugation and then use that information to identify potent relaxase-specific inhibitors. This is the first paper which described a bottom-up approach to identify the first small molecule inhibitors of conjugative DNA transfer.
The authors determined the 2.4-Å crystal structure of the 300-residue N-terminal relaxase domain of F plasmid TraI and found it similar to other, previously described relaxase domains. They found that the tyrosine at the active site is responsible for binding the oriT thymidine. Based on electron density they found presence of a divalent cation in the active site and identified it as Mg2+. A survey of magnesium-binding proteins in the Protein Data Bank revealed that the chelation of Mg2+ by neutral amino acid residues is diagnostic of a site that simultaneously binds to multiple phosphate groups. Mutation of the metal-chelating residue histidine-159 to glutamic acid eliminated relaxase activity. These data indicated that the 2+ charge on the bound metal ion is critical to relaxase function. The proposed models for relaxase role in binding DNA strands during conjugation suggested that relaxase binds two phosphate groups. To prove this theory authors used a simple and relatively stable bisphosphonate – imidobisphosphate (PNP) molecule and found that at nanomolar concentration PNP inhibited relaxase activity in vitro. Further studies established that relaxase can be effectively inhibited by substrates where two phosphonate residues are separated by three or fewer atoms and have no additional negative charge at pH 7.4. Five additional inhibitors were found: methylenediphosphonic acid (PCP), iminobis(methylphosphonic acid) (PCNCP), etidronic acid (ETIDRO), clodronic acid (CLODRO), and 1,2-bis(dimethoxyphosphoryl)benzene (PBENP). ETIDRO and CLODRO are bisphosphonates clinically approved as drugs used to treat bone loss by inhibiting farnesyl diphosphate synthase Two other inhibitors identified, PCP and PNP, have been used as radioisotope carriers in humans. The simplest inhibitors, PCP, ETIDRO, and CLODRO, were then characterized further by using a kinetic assay and exhibited purely competitive inhibition, with Kic,app values ranging from 3 to 145 nM. Taken together with the PNP results, these data validate the prediction that F-like conjugative relaxases can accommodate two phosphotyrosine intermediates simultaneously within their active sites. Significantly, these data also establish that bisphosphonates (including clinically approved compounds) potently inhibit the in vitro relaxase activity of F TraI with Ki values in the nanomolar range.
The in vivo tests confirmed the results obtained in vitro. In addition to conjugation inhibition, micromolar concentrations of PNP caused death of plasmid-containing, but not plasmid free cells by blocking relaxase activity.
The presented results show that the clinically approved bisphosphonates etidronate (Didronel) and clodronate (Bonefos) are potently effective at killing F+ cells and preventing conjugative DNA transfer. These particular compounds could also be combined with existing antibiotics to create potent antimicrobial cocktails. Etidronate and clodronate exhibit low absorption and can be administered at high oral doses. According to the authors, extrapolating from the results, approved doses of etidronate and clodronate would be expected kill >90% of plasmid + cells and to stop >80% of conjugative transfer within the gastrointestinal tract. Such results are relatively mild, given the large bacterial populations present in the gastrointestinal tract or at wound sites, but may be enough shift the balance toward success in a variety of recalcitrant clinical infections, especially given the prevalence of conjugative plasmids within multidrug-resistant bacterial strains. The treatment of skin infections, primary sites of nosocomial antibiotic resistance transfer, using topical applications of bisphosphonates may also be effective. In summary, this study establishes conjugative relaxases as a unique antimicrobial target. The results suggest that approved therapeutics could have an immediate impact, alone or in combination with existing antibiotics, in the prevention of resistance propagation during clinical treatment of bacterial infections, thus extending the lifetime of our antibiotic arsenal.
In conclusion this paper shows us a very important thing: basic research on the conjugation process in plasmid transfer that showed the crucial role of relaxase protein led to more detailed application studies that give us a potential weapon to fight bacterial conjugation and the spread of antibiotic resistance in the world of microorganisms.

Additional papers.

Zhao, W.-H. , Z.-Q. Hu, Y. Hara and T. Shimamura 2001 Inhibition by epigallocatechin gallate (EGCg) of conjugative R plasmid transfer in Escherichia coli. J.Infect. Chemotherapy 7: 195-197

Tiagunenko IuV, Glatman LI, Antsiferova NG., 1975. Caffeine as an inhibitor of the conjugation transfer of R-factors. A study of certain aspects of the mechanism of action of caffeine on the conjugation transfer of R-factors], Antibiotiki 20: 253-257.

Leite A.A.M., Nardi R.M.D., Nicoli J.R., Chartone-Souza E. and Nascimento A.M.A., 2005. Carica papaya seed macerate as inhibitor of conjugative R plasmid transfer from Salmonella typhimurium to Escherichia coli in vitro and in the digestive tract of gnotobiotic mice. Gen. Appl. Microbiol. 51: 21-26

Fernandez-Lopez R., Machón C., Longshaw C.M., Martin S., Molin S., Zechner E.L., Espinosa M., Lanka E. and de la Cruz F. 2005. Unsaturated fatty acids are inhibitors of bacterial conjugation. Microbiology 151: 3517–3526


dr Jaroslaw E. Krol
UofI