Introducing specific mutations into the Escherichia coli chromosome using linear transformation

CONTENTS

I. INTRODUCTION AND GENERAL BACKGROUND
1. The bacterium Escherichia coli
2. Urinary tract infections (UTIs)
3. Resistance to antibiotics
3.1 Resistance mutations
3.2 Resistance to fluoroquinolones
3.2.1 Target site mutations
3.2.1.1 Mechanism of DNA gyrase and topoisomerase IV
3.2.2 Resistance caused by increased efflux
3.2.2.1 The mar-locus
3.3 Minimum inhibitory concentration
3.4 Mutation rate versus resistance
4. The system

II. AIM OF THE STUDY

III. MATERIAL AND METHODS
1. Growth media and solutions
1.1 Induction of the plasmid pKD
1.2 Solution for Sequencing
2. Antibiotics
3. Bacterial strains and plasmids
3.1 l plasmid pKD
3.2 Plasmid pCP
4. PCR general
4.1 PCR Beads
4.2 PCR for amplifying tetRA from Tn
4.3 PCR for amplifying tetRA from the plasmid pCP
4.4 Primers for PCR and DNA sequencing
5. Treatment with Dpn1
6. Agarose gel electrophoresis
7. Gel extraction
8. Measurement of DNA concentration
9. Using single-stranded oligonucleotides to introduce
point mutations into a gene
10. Preparing electrocompetent cells with the l Red system induced
10.1 Chromosomal l-Red system
10.2 Plasmid-borne l-Red system
11. Electroporation
12. Recombined chromosome
13. The use of Flp
14. Transformation with heat shock
15. Thermal cycle sequencing
15.1 Sequencing protocol
16. P1 phage preparation and transduction
17. The general scheme of this recombination method

IV. RESULTS
1. Optimising electroporation efficiency
2. Recombining tetracycline resistance into mutS
3. Recombining FRT-tetracycline resistance-FRT into mutS and marR
4. Introducing a point mutation into the genes gyrA and parE
5. Removing tetracycline resistance using Flp
6. Assaying recombinant phenotypes
7. Sequencing recombination junctions

V. DISCUSSION
1. General
2. Electroporation efficiency
3. Dpn1 treatment
4. Recombinants
5. Chromosome versus plasmid
6. Using oligonucleotides
7. Final conclusions

VI. SUMMARY

VII. REFERENCES

VIII. APPENDIX
1. Abbrevations
2. Codon table
3. Curriculum vitae

I. INTRODUCTION AND GENERAL BACKGROUND

1. The bacterium Escherichia coli

In 1885 Theodor Escherich described the gram negative bacterium Escherichia coli (E.coli) (Escherich T., Rev.1989). The gram negative rod belongs to the family of the enterobacteriaceae. It is a natural inhabitant of the human and animal intestine. E.coli can also cause diseases like diarrhea, inflammation of the urinary tract or the gall bladder.

illustration not visible in this excerpt

Figure 1: Electron microscope picture of the bacteria E.coli (from http://commtechlab.msu .edu/sites/dlc-me/zoo/Pf07002.jpg; accessed 11.07.03).

2. Urinary tract infections (UTIs)

Urinary tract infections (UTIs) are one of the most common reasons for antibiotic therapy worldwide (Burman and Olsson-Liljequist, 2001). People with UTIs suffer inflammation of the urinary tract, and frequently the kidneys. Two-thirds of patients with UTIs are women (Canbaz et al. 2002). This is related in part to the shortness of the urethra, which makes colonization of the bladder by bacteria more likely. The elderly and those who undergo genitourinary operations and catheterisation are also frequent sufferers of UTIs (Orenstein and Wong, 1999). The leading causative agent of UTIs is E.coli (60-80 %), usually originating from the patients own faecal flora, followed by Staphylococcus saprophyticus (10 %), Klebsiella sp., other Gram negative bacteria and enterococci (Burman and Olsson-Liljequist, 2001). The antibiotic class most frequently prescribed to treat UTIs in Western Europe and North America is the fluoroquinolones. Fluoroquinolones are synthetic antibiotics derived from nalidixic acid. Resistance to the synthetic fluoroquinolone antibiotics is increasing among the organisms that cause UTIs.

3. Resistance to antibiotics

Antibiotics are low-molecular-weight compounds that can inhibit the growth of, or kill, microorganisms (Salyers and Whitt, 2002). Many antibiotics in clinical use are natural products, or their derivatives, e.g. the penicillins and the macrolides. Because bacteria and fungi have been producing antibiotics, resistance to these antibiotics has also evolved in many target bacteria (Burman and Olsson-Liljequist, 2001).

Antimicrobial resistance is the result of microorganisms changing in ways that reduce or eliminate the effectiveness of drugs, chemicals, or other agents to cure or prevent infections (Fromhttp://www.cdc.gov/drugresistance/miscellaneous/glossary.htmantimicrobialresistance; accessed 14.07.03). Use of antibiotics in medicine and agriculture has imposed a strong selection on bacterial populations. Antibiotics have selected for bacteria with mutations causing resistance, and also for the transfer of genetic material from resistant to sensitive bacteria (Diarmaid Hughes, personal communication).

The mechanisms of antibiotic resistance include (Hughes, 2003):

- Target alteration: The target alters in a way that the affinity of the antibiotic for its target is reduced.
- Increased antibiotic efflux: Mutations that upregulate the expression of trans-membrane efflux pump, can reduce the concentration of antibiotic in the bacterial cell. In many cases, efflux pumps can pump out several different antibiotics causing a multidrug-resistance phenotype.
- Decreased antibiotic uptake: Import of an antibiotic can be inhibited by mutations that downregulate, delete or modify outer-membrane porins.
- Antibiotic inactivation by enzymatic activity: Enzymes that modify, cleave or otherwise inactivate an antibiotic are another common cause of resistance. The most important example clinically is the widespread occurrence of β-lactamases that cleave the β-lactam ring of β-lactam antibiotics.

3.1 Resistance mutations

Strains with clinically relevant levels of resistance (above a defined break-point) typically have several putative resistance mutations. Previous work from this laboratory (Patricia Komp Lindgren and Diarmaid Hughes, personal communication), in which fluoroquinolone resistant strains of E.coli were selected, shows that the successive accumulation of resistance mutations is associated with increasing fitness costs. In contrast, fluoroquinolone resistant clinical isolates of E.coli show no obvious association between the presence of resistance mutations and fitness costs. This difference illustrates a serious problem in trying to interpret the significance of putative resistance mutations. The specific nature of the connection between genotype and phenotype needs to be established by appropriate genetic reconstructions, for example with linear transformation, before definite conclusions can be drawn.

3.2 Resistance to fluoroquinolones

E.coli can become resistant to fluoroquinolones by altering the target enzymes, reducing permeability of the cell to inhibit their entry, or by actively pumping the drug out of the cell. All these resistance mechanisms can play a role in high-level fluoroquinolone resistance however target site mutations appear to be most important (Webber and Piddock, 2001).

3.2.1 Target site mutations

Fluoroquinolone resistant E. coli can have mutations in one or more genes including: gyrA and gyrB (coding for the two subunits of DNA gyrase); parC and parE (coding for the two subunits of topoisomerase IV) (Hughes and Andersson, 2001).

3.2.1.1 Mechanism of DNA gyrase and topoisomerase IV

Topoisomerases are ubiquitous enzymes necessary for controlling the interlinking and twisting of DNA molecules. The DNA gyrase (topoisomerase II) subunit A encoded by the gene gyrA introduces negative supercoils in the DNA. The Topoisomerase IV subunit which is encoded by the gene parE removes DNA positive and negative supercoils (Hooper D.C, 1998). Quinolones inhibit the action of DNA gyrase and topoisomerase IV and kill bacteria by binding to these enzyme-DNA complexes, thereby disrupting DNA replication. Mutations in the genes that encode for DNA gyrase and topoisomerase IV can change the structure of the subunit of the enzyme. These mutations generally occur in a discrete sequence of the bacterial genes, called the quinolone resistance determinant region. A mutation in the quinolone resistance determinant region cause resistance to the antibiotic. Quinolone-resistance mutations in gyrA have all been localized to the amino terminus between amino acid 67 and 106, with the most common sites being alterations at positions 83 and 87. Resistance caused by alterations in gyrA correlates with reduced binding of quinolones to the resistant mutant enzyme DNA complex (Hughes and Anderson, 2001).

illustration not visible in this excerpt

Figure 2: Position of topoisomerase IV and DNA gyrase in the replication fork (Bearden and Danziger, 2001).

3.2.2 Resistance caused by increased efflux

Increased efflux is associated with increased activity of the Acr multidrug efflux pump. Mutations can also occur in marR (coding for a repressor of the marA gene; the marA gene is an activator of the Acr efflux pump); acrR (a repressor of AcrA, the efflux pump protein). It is likely that mutations in other, currently unidentified, genes are also involved in fluoroquinolone resistance in E.coli (Patricia Komp Lindgren and Diarmaid Hughes, personal communication).

3.2.2.1 The mar-locus

The mar -locus consists of two divergently positioned transcriptional units that flank the operator marO. marC is a putative integral inner membrane protein of unknown function that is encoded by one transcriptional unit. The other transcriptional unit comprises marRAB which encodes the 144 amino acid repressor marR, the 127 amino acid transcriptional activator marA, and marB, a small protein of unknown function. Once the repressor marR is inhibited from binding to the operator marO (e.g. by a knock-out mutation of the marR gene) the transcriptional activator marA is transcribed. marA upregulates a total of 47 genes, including AcrA and TolC, both components of the AcrAB efflux pumps, which pump out fluoroquinolones and other structurally unrelated antibiotics (Randall and Woodward, 2002).

illustration not visible in this excerpt

Figure 3: Diagrammatic representation of the mar-locus. The arrows show the direction of transcribtion.

marR was chosen as a target because its inactivation is predicted to increase the efflux of fluoroquinolones, resulting in a resistance phenotype. That means for the laboratory work; if the marR gene is successfully knocked out there should be an increase in the MIC values of antibiotics like nalidixic acid or norfloxacin.

Efflux mutations are increasingly recognised as an important class of mutations involved in resistance in clinical isolates. A problem associated with mutations like marR, is that a single mutation can result in resistance to several antibiotics, so-called multidrug resistance (Diarmaid Hughes, personal communication).

3.3 Minimum inhibitory concentration

MIC is short for minimum inhibitory concentration and is a relative measure of the lowest amount of antibiotic required to inhibit the growth of a bacterium (Burman and Olsson-Liljequist, 2001). The MIC was determined with e-tests (Epsilortest). E-tests (AB Biodisk, Solna, Sweden) consists of plastic strips carrying an antibiotic gradient on the side to be placed onto an inoculated agar plate. After 24 h incubation, the MIC can be read. Testing was performed according to the instructions of the manufacturer. In this study the following antibiotics were chosen; nalidixic acid because it is the first quinolone antibiotic ( Figure 4), norfloxacin as a common fluoroquinolone antibiotic (Figure 5) and chloramphenicol (Figure 6) because active efflux is a main mechanism of chloramphenicol resistance.

illustration not visible in this excerpt

Figure 4: Structure image of nalidixic acid (from http://www.sigmaaldrich.com/cgi-bin/hsrun /Distributed/ HahtShop/HAHTpage/HS_StructureImage; accessed 14.07.03)

illustration not visible in this excerpt

Figure 5: Structure image of norfloxacin (from http://www.sigmaaldrich.com/cgi-bin/hsrun /Distributed/HahtShop/HAHTpage/HS_StructureImage; accessed 14.07.03)

illustration not visible in this excerpt

Figure 6: Structure image of chloramphenicol (from http://www.sigmaaldrich.com/cgi-bin/hsrun /Distributed/HahtShop/HAHTpage/HS_StructureImage; accessed 14.07.03)

3.4 Mutation rate versus resistance

There is a good correlation between an increased general mutation rate and the number of resistance mutations found in clinically resistant isolates (Patricia Komp Lindgren and Diarmaid Hughes, personal communication). All resistant clinical isolates have multiple mutations and the correlation with mutation rate suggests that this might be a factor in the acquisition of the observed mutations. Indeed, a resistant clinical isolate with a high mutation rate has recently been shown to have a deletion of 5’ end of the mutS gene (Patricia Komp Lindgren and Diarmaid Hughes, personal communication). mutS encodes a critical protein for general mismatch repair in E. coli (Bjornson K.P, 2003). Thus, it was decided to target mutS for inactivation, to create a strain pair that could be used in later experiments to measure rate of evolution of resistance in vivo during antibiotic treatment.

[...]