MULTIDRUG RESISTANT EFFLUX PUMPS IN SOIL BACTERIA

1
MULTIDRUG RESISTANT EFFLUX PUMPS IN SOIL BACTERIA
?
Ashley Wilson Summer USRP Project under the
leadership of Dr. Michael F. Hynes

• ABSTRACT
• In rhizobia there are a number of genes that
  show homology to genes encoding multidrug
  resistant (MDR) efflux pumps found in pathogenic
  bacteria, but the function of these homologs has
  never been explored in rhizobia. This research
  involved the amplification of the homologous
  genes in question using polymerase chain reaction
  (PCR), with the aim of mutating these genes via
  insertion of an antibiotic resistance cassette,
  (within E. coli) and transferring of mutant genes
  back into rhizobia via mating and homologous
  recombination. The mutant rhizobia will then
  have their growth compared to the wildtype on
  media containing various chemicals at different
  concentrations to determine what that particular
  homologous gene confers resistance to. The focus
  of the testing will initially be on compounds
  that the homologous genes present in pathogenic
  bacteria were found to be resistant to (see
  Summary of Origin and Resistance of Investigated
  Efflux Pumps below).
• INTRODUCTION
• Bacteria are susceptible to many compounds that
  are found in their environments, particularly if
  the chemical can move through the bacterial cell
  membrane via osmosis or a membrane protein or
  porin. In order to protect themselves, bacteria
  have developed specialized membrane proteins
  known as MDR efflux pumps, that can expel a
  variety of harmful compounds from the bacterial
  cell by expending energy. There are several
  major families of MDR efflux pumps including
  multidrug and toxic compound extrusion (MATE)
  proteins, small multidrug resistant (SMR)
  proteins, the major facilitator superfamily
  (MFS), resistance nodulation division (RND), and
  the ATP binding cassette (ABC) superfamily1.
• In Rhizobium leguminosarum bv. viciae 38412,
  the rhizobia strain used in this research, there
  are many genes that are predicted to encode MDR
  efflux pumps due to the homology they express to
  MDR efflux pump genes that have been studied in
  pathogenic bacteria. Although the presence or
  chemicals acted upon by these putative MDR efflux
  pumps in rhizobia have not yet been studied, it
  is predicted that these pumps act on different
  chemicals than their homologues in pathogenic
  bacteria as rhizobia bacteria live in soil
  environments that are dramatically different than
  those of many pathogenic bacteria. It seems
  likely that the putative MDR efflux pumps in
  rhizobia could convey resistance to toxins that
  are produced by plants, fungi, animals and other
  bacteria and toxins that are found in soils.
• The focus in this study will be on the genes in
  R. leguminosarum bv. viciae 3841 that show
  homology to the Escherichia coli sugE3 and emrB
  (qacA in Staphylococcus aureus)1,4 genes, Vibrio
  cholerae vcrM 5 gene, Vibrio parahaemolyticus
  norM 6,7 gene, and Pseudomonas aeruginosa mexB
  (acrB in E. coli)1,7 gene and a putative MFS gene
  (mdr3841-1).
• SUMMARY OF ORIGIN RESISTANCE OF INVESTIGATED
  EFFLUX PUMPS
• SugE is a SMR homologue found in E. coli that
  confers resistance to a narrow range of
  quaternary ammonium compounds3.
• EmrB is a multidrug resistance proteins found
  in E. coli, with a similar protein to EmrB, QacA,
  found in S. aureus, a protein belonging to the
  MFS. EmrB in E. coli has been shown to confer
  resistance to carbonylcyanide m-chlorophenylhydraz
  one, 2-chlorophenylhydrazine hydrochloride,
  tetrachlorosalicylanilide, nalidixic acid, and
  phenylmercury acetate4.
• VcrM is a MATE protein from V. cholerae that
  confers resistance to acriflavine,
  6-diamidino-2-phenylindole, Hoechst 33342,
  rhodamine 6G, tetraphenylphosphonium chloride and
  ethidium bromide5.
• NorM is a protein belonging to the MATE family
  and found in V. parahaemolyticus where it makes
  these bacteria resistant to kanamycin,
  norfloxacin, ethidium bromide6, and
  fluoroquinolone7.
• MexB as part of the MexAB OprM pump belongs to
  the RND family and is found in P. aeruginosa7. A
  homologous pump is also found in E. coli where
  MexB is homologous to AcrB in the AcrAB TolC RND
  pump. The MexAB OprM pump and its homologues are
  known to efflux chloramphenicol, lipophillic
  b-lactams, ofloxacin, fluoroquinolones,
  tetracycline, rifampin, novobiocin, fusidic acid,
  ethidium bromide, acriflavine, bile salts, short
  chain fatty acids, SDS, Triton X-100, triclosan,
  glycylcyclines, trimethoprim, and many other
  compounds, although for many of these compounds
  the rate of efflux is not significant to confer
  clinical levels of resistance1,7,8.
• Mdr3841-1 protein shows similarity to a
  putative MFS transporter in P. aeruginosa
  (BLASTX, NCBI) and thus the resistance
  potentially conferred by this putative pump
  remains unknown.

• SUMMARY OF METHODS
• Bioinformatics survey to find genes in R.
  leguminosarum 3841 bv. viciae that are homologous
  to MDR efflux pump genes in pathogenic bacteria
• Growth and isolation of R. leguminosarum 3841 bv.
  viciae DNA
• Primers designed via information gained from the
  bioinformatics survey to be used for PCR
• Amplification of MDR efflux pump genes via PCR
• Cloning of PCR products into pCR2.1-TOPO vector
• Cloning of PCR product into pJQ200SK vector
  (suicide vector)
• Mutation of MDR efflux pump gene inserted in
  vector
• Mating of suicide vector back into R.
  leguminosarum 3841 bv. viciae
• Testing of mutant rhizobia for resistance to
  various compounds versus the wildtype
• TABLE 1 GENES AND PCR PRIMERS

The insert had to be transferred from
pCR2.1-TOPO vector to pJQ200SK vector because
pCR2.1-TOPO is unable to be transferred to
rhizobia, which is necessary in order to create a
R. leguminosarum bv. viciae 3841 with a mutant
MDR efflux pump gene to compare the resistance to
various compounds versus the resistance to the
same compounds by the wildtype R. leguminosarum
bv. viciae 3841. pJQ200SK is a suicide vector in
rhizobia, that is the vector can be transferred
to rhizobia but is unable to replicate.
Infrequently the homologous region in the
pJQ200SK vector (the mutated MDR efflux pump gene
insert) and the homologous region in the wild
type R. leguminosarum bv. viciae 3841 (the
unmutated MDR efflux pump gene) will undergo
homologous recombination producing a mutant
rhizobia. These mutant rhizobia will be screened
for by plating the bacteria on media with the
antibiotic corresponding to the antibiotic
resistance cassette inserted to mutate the MDR
efflux pump gene. Only those rhizobia that
undergo homologous recombination will retain the
cassette that confers antibiotic resistance, and
thus, will survive. Before the homologous
recombination can occur, the pJQ200SK vector with
insert needs to be transferred from E. coli
strain DH5a to S17-113, as S17-1 is able to
conjugate with rhizobia, enabling the transfer of
the pJQ200SK vector. Additionally, before the
pJQ200SK vector is moved to the S17-1 E. coli
strain, the insert had to be mutated via
insertion of an antibiotic resistance cassette,
either kanamycin or tetracycline. These were
obtained from pBSL190 (tetracycline cassette) or
pBSL99 (kanamycin cassette)14. Restriction
enzymes will be used to cut within the MDR efflux
pump gene contained in the pJQ200SK vector as
well as to cut out the antibiotic resistance
cassette from pBSL190 or pBSL99 in order to
ligate the insert and the cassette together. In
the case of the sugE MDR efflux pump gene, BclI
(Fermentas, Burlington, Ontario) restriction
enzyme was to be used. However, E. coli strain
DH5a DAM methylates the restriction site
preventing cutting by the BclI restriction
enzyme. Thus the pJQ200SK vector with sugE
insert had to be transferred to a DAM- E. coli
strain, Gm4815, and after mutation, to
S17-1. Once the mutant rhizobia were obtained
their growth would be compared to the growth of
the wildtype rhizobia on media containing various
types of compounds, beginning with those outlined
in the summary of origin and resistance of
investigated efflux pumps above. RESULTS AND
DISCUSSION All the primers designed using the
R. leguminosarum bv. viciae 3841 sequence (table
1) were successful in amplifying their
corresponding gene with the exception of the
primers for the mdr3841-1 gene. It is predicted
that the mdr3841-1 gene was unable to be
amplified using these primers as the PCR product
is approximately 700 nucleotides longer that the
largest successful amplification by Taq
polymerase. The sugE, emrBpRL11JI, vcrM and
mexBtype2 genes were successfully cloned into the
pCR2.1-TOPO vector. Furthermore, the sugE gene
was successfully moved from pCR2.1-TOPO vector
into pJQ200SK. The sugE gene mutation was
attempted using both the kanamycin and
tetracycline cassettes but was unsuccessful.
Initially, the sugE insert that was replicated
within DH5a E. coli was unable to be cut with the
BclI due to DAM methylation of the restriction
site. However, after transferring the sugE
insert within the pJQ200SK vector to GM48 E.
coli, which is DAM-, the insert was still unable
to be cut by the BclI enzyme for unknown
reasons. As no mutant rhizobia were obtained,
the resistance conveyed by the various putative
MDR efflux pump genes could not be tested. As
shown in table 3, most of the putative MDR efflux
pump genes found in R. leguminosarum bv. viciae
3841 have homologous proteins in many other
species within the rhizobia family. The presence
of these across the rhizobia family suggests that
they many be important for the protection against
toxins commonly found in the rhizobias soil
environment. However, it is interesting to
note the low homology within the rhizobia family
for the vcrM gene, particularly the low homology
to R. etli CFN42 at 39. As R. etli CFN42 and R.
leguminosarum bv. viciae 3841 belong to the same
genus it is expected that the VcrM proteins would
have the highest degree of homology, around 90
as is shown in table 3 for many other proteins.
As the homology is so low both within the
Rhizobium genus and the rhizobia family this
implies that the vcrM gene is found only in R.
leguminosarum bv. viciae 3841 and not in other
closely related bacteria. This suggests that
either the vcrM gene was inherited after the
division of the R. leguminosarum and R. etli
species or that the vcrM gene was lost from the
R. etli lineage. Furthermore, if it is the case
the vcrM gene was inherited after the division of
the R. leguminosarum and R. etli species, the
maintenance of the vcrM gene within R.
leguminosarum suggests that this gene confer some
advantage in the soil environments where R.
leguminosarum bv. viciae 3841 is
found. Additionally for mexBtype2, although
there is significant homology to species within
the rhizobia family there is low similarity to R.
etli CFN 42, implying that the mexBtype2 gene
exists within the rhizobia family but not within
R. etli CFN 42. This implies that the mexBtype2
gene was lost from the R. etli linage, and is
consistent with the R. etli genome being smaller
than the R. leguminosarum genome.
TABLE 3 GENE ORIGINS, HOMOLOGY AND PRESENCE IN
RHIZOBIA
TABLE 2 PCR CONDITIONS
CONCLUSION There are a large number of
potential MDR efflux pump genes in the rhizobia
family with some being specific to certain
species such as vcrM and others conserved among
the rhizobia family such as sugE and emrB. Eight
of the nine putative MDR efflux genes, where
amplification via PCR was attempted, were
successful. The sugE, emrBpRL11JI, vcrM and
mexBtype2 genes were successfully cloned into the
pCR2.1-TOPO vector, with the sugE gene also being
cloned into the pJQ200SK vector. These
constructs are available for future work where
mutation and testing for MDR can occur.
REFERENCES 1- Piddock LJ V. (2006) Clinically
Relevant Chromosomally Encoded Multidrug
Resistance Efflux Pumps in Bacteria. American
Society for Microbiology. 19 382-402. 2- Poole
PS, Blyth A, Reid CJ, Walters K. (1994)
Myo-inositol catabolism and catabolite repression
in Rhizobium leguminosarum. Microbiology-UK 140
2787-2795. 3- Sikora CW, Turner RJ. (2005) SMR
proteins SugE and EmrE bind ligand with similar
affinity and stoichiometry. Biochemical and
Biophysical Research Communications.
335105-111. 4- Lomovskaya O, Lewis K. (1992)
emr, an Escherichia coli locus for multidrug
resistance. Proceedings of the National Academy
of Science, USA. 898938-8942. 5- Huda MN, Chen
J, Morita Y, Kuroda T, Mizushima T, Tsuchuya T.
(2003) Gene cloning and characterization of
VcrM, a Na-coupled multidrug efflux pump, from
Vibrio cholerae non-01. Microbiology and
Immunology. 47419-27. 6- Otsuka M, Yasuda M,
Morita Y, Ostuka C, Tsuchiya T, Omote H, Moriyama
Y. (2005) Identification of Essential Amino Acid
Residues of the NorM Na/Multidrug Antiporter in
Vibrio parahaemolyticus. Journal of
Bacteriology. 1871552-1558. 7- Poole K. (2005)
Efflux mediated antimicrobial resistance. Journal
of Antimicrobolgy and Chemotherapy. 5620-51 8-
Brooun A, Liu S, Lewis K. (2000) A Dose-Response
Study of Antibiotic Resistance in Pseudomonas
aeruginosa Biofilms. Antimicrobial Agents and
Chemotherapy. 44640-646. 9- Beringer J. (1974) R
factor transfer in Rhizobium leguminosarum.
Journal of General Microbiology. 84188-198. 10-
Young JPW, Crossman LC, Johnston AWB, et al.
(2006) The genome of Rhizobium leguminosarum has
core and accessory components. Genome Biology
7R34. 11- Sambrook J, Fritsch EF, Maniatis T.
(1989) Molecular Cloning a Laboratory Manual.
Cold Spring Harbor, NY Cold Spring Harbor
Laboratory. 12- Quandt J, Hynes MF. (1993)
Versatile suicide vectors which allow direct
selection for gene replacement in Gram-negative
bacteria. Gene. 127 15-21. 13- Simon R, Priefer
U, Pühler A. (1983) A broad host range
mobilization system for in vivo genetic
engineering transposon mutagenesis in gram
negative bacteria. Biotechnology 1784-791. 14-
Alexeyev MF, Shokolenko IN, Croughan TP. (1995)
Improved antibiotic-resistance gene cassettes and
omega elements for Escherichia coli vector
construction and in vitro deletion/insertion
mutagenesis. Gene 160 63-67. 15- Marinus MG.
(1973) Location of methylation genes on the
Escherichia coli K-12 genetic map. Molecular and
General Genetics 127 47-55.
DETAILS OF CLONING, TRANSFORMATION AND
MUTATION The cleaned PCR products were cloned
into pCR2.1-TOPO (Invitrogen, Burlington,
Ontario) vector, which has kanamycin and
ampicillin resistance. Plasmid DNA was then
isolated from an overnight culture. To ensure
the desired PCR product was inserted into the
pCR2.1-TOPO vector, the vector was digested
overnight at 37C using EcoRI (Invitrogen) and
visualized on an agarose gel (0.8 w/v). The
insert was then moved from pCR2.1-TOPO vector to
pJQ200SK12 vector via ligation using restriction
enzymes. In the case of the sugE gene, XhoI
(Invitrogen) was used to remove the insert and
digest pJQ200SK. A ligation to pJQ200SK of
emrBpRL11JI, vcrM, and mexBtype2 was attempted
using SpeI and ApaI (Invitrogen) (at 37C ) to
remove the insert from pCR2.1-TOPO vector as well
as digest the pJQ200SK vector.