INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

The neutral water-soluble thiols 2-mercaptoethanol, 1-thioglycerol, and dithiothreitol inhibit gram-positive and gram-negative bacteria at millimolar concentrations (19, 21, 24). Although the precise mechanism of growth inhibition is not understood, it is known that several processes are affected. These include the lowering of intracellular concentrations of S-adenosylmethionine (14), inhibition of aerobic respiration (15), and possibly interference with the formation of disulfide bonds of periplasmic and outer membrane proteins (4, 27).

Exposure of aerobically growing Escherichia coli to exogenous thiols also results in a pleiotropic reductive stress response, which includes elevation of riboflavin and porphyrin syntheses (16), blockage of septum formation (18), and changes in the expression of hundreds of genes (17, 20).

In an attempt to look for genes which may be regulating this complex response, we searched for thiol-hypersensitive mutants. The search was based on the rationale that should such a gene(s) exist, its (their) inactivation would likely yield a thiol-hypersensitive phenotype.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Reagents. Amino acids, nucleotides, monosaccharides, coenzyme Q₁₀, and dithiothreitol were purchased from Sigma Chemical Co., St. Louis, Mo. Hexane, petroleum ether, and 1-thioglycerol were bought from Aldrich Chemical Co., Milwaukee, Wis. The Wizard genomic purification kit and the Wizard Minipreps and Wizard PCR Preps DNA purification systems were from Promega Corp., Madison, Wis. Cloned Pfu DNA polymerase was obtained from Stratagene, La Jolla, Calif.

Bacterial strains and plasmids. Strains and plasmids are listed in Table 1.

Growth media. All experiments were performed on aerobically growing cells, at 37°C, in Davis minimal salts medium (12). Glucose (0.3% [wt/vol]) was the carbon source. The media were supplemented for strains THU and IS16 and its derivatives with uracil plus thymine at 20 µg/ml each and with histidine at 40 µg/ml. The growth media for strain IS16B1 and all plasmid-bearing strains were supplemented with 100 µg of ampicillin per ml. dsb mutant strains were supplemented with 25 µg each of leucine, isoleucine, and valine per ml and 2.5 µg of cystine per ml.

Growth inhibition studies. Growth inhibition studies were essentially carried out as described before (15, 21), except that the absorbance was monitored at 540 nm (A₅₄₀).

Alkaline phosphatase assay. Cultures were grown in Davis minimal salts medium until the A₅₄₀ was 0.4 to 0.5. The bacteria were centrifuged and washed twice with low-phosphate-containing Davis medium (Davis medium containing only 10⁴ M K₂HPO₄) at room temperature. The cells were resuspended in low-phosphate Davis medium containing all growth supplements at their initial volumes and were incubated at 37°C with shaking. At 20- or 30-min intervals, 0.1-ml aliquots were collected into 1.9 ml of ice-cold 1 M Tris-HCl buffer (pH 8.0). The enzyme assays were done as described by Brickman and Beckwith (6) except that A₅₄₀ was used instead of A₆₀₀.

Measurement of coenzyme Q₈. Coenzyme Q₈ was routinely extracted by the method of Krivankova and Dadak (23). For its use as a high-performance liquid chromatography (HPLC) marker and for the purpose of quantitation, an extracted sample was further purified by thin-layer chromatography. On a Whatman linear-K silica gel thin-layer plate, with a 70:30 chloroform-light petroleum ether solvent mixture, coenzyme Q₈ had an R_f value of 0.71. This substance was eluted from the plate with methanol, and its concentration was determined by its A (oxidized-reduced)₂₇₅, with 12.7 as the millimolar extinction coefficient (10).

PCR. Genomic DNA was isolated with the Promega Wizard Genomic DNA kit. Cloned Pfu DNA polymerase purchased from Stratagene Cloning Systems was used. The ubiX genes of strains THU and IS16 were isolated by PCR, with the following primers: forward, 5'-ttcgaagcagtgcaacgtcagagcg-3', and reverse, 5'-gaattcaaacagggcaacagcggag-3'. (The 5' end of the forward primer was given a HindIII cutting site, and that of the reverse primer was given an EcoRI cutting site). The primers were synthesized on an Applied Biosystems 394 DNA synthesizer. The products of PCR were purified with the Promega Wizard PCR Preps DNA purification kit. They were sequenced on an Applied Biosystems Model 373 sequencer at the Center for Molecular Biology and Gene Therapy of Loma Linda University.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Isolation of a thiol-hypersensitive derivative of E. coli THU. A culture of strain THU was mutagenized by nitrosoguanidine (1), and the cells were grown on minimal glucose plates containing 25 mM 1-thioglycerol. The smallest colonies were replica plated onto minimal medium- and minimal medium-plus-25 mM-1-thioglycerol-containing dishes. For controls, nonmutagenized THU colonies were replica plated as well. Colonies which were normal size on minimal plates but small on thioglycerol plates were isolated for further study. The isolate exhibiting the greatest hypersensitivity to 1-thioglycerol, named IS16, was used for further studies.

View larger version (13K): [in this window] [in a new window]   FIG. 1.   Inhibition by 1-thioglycerol and dithiothreitol of growth of strains THU, IS16, and IS16B1. The mean growth rates (MGR) of cultures in the presence of the indicated concentrations of thiols were compared with the MGR of untreated cultures.

View larger version (60K): [in this window] [in a new window]   FIG. 2.   Nucleotide sequence of dedF (ubiX) and its flanking regions. The BglII cutting sites are double underlined. Plasmid pPZ2's chromosomal insert sequence begins at the 5' end with G and ends at the 3' end with A. The abbreviated 3' end of purF shows only 69 nucleotides. The single-underlined sequences bracket the portions of the chromosome that were isolated by PCR from IS16 and THU and were inserted into pBR322 to form pHZ1 and pHZ2. Codon 98 in dedF (AGC), printed in boldface, is the site of the mutation in IS16 (AGA).

Measurement of ubiquinone content of strain IS16. Since the complementing plasmid contained only one functional gene, dedF (ubiX), a ubiquinone-biosynthetic gene, the ubiquinone contents of strains THU, IS16, and IS16B1 were measured. The results are shown in Fig. 3. Strain IS16 contained 85% less ubiquinone than did THU, and strain IS16B1, carrying multiple copies of ubiX, had 1.5 times as much ubiquinone.

View larger version (7K): [in this window] [in a new window]   FIG. 3.   Analysis of lipid extracts for ubiquinone by HPLC. Shown are the tracings of the absorption of the effluents at 275 nm. The arrows represent the peaks of ubiquinone Q8. From the areas under the peaks, the ubiquinone contents were calculated. The results are 0.21, 0.03, and 0.31 nmol/mg (dry weight) for strains THU, IS16, and IS16B1, respectively.

Search for the locus of mutation on ubiX in strain IS16. Chromosomal ubiX genes of strain IS16 and strain THU were isolated by PCR, as described in Materials and Methods. Cloned Pfu polymerase was used, because of its considerably greater fidelity than Taq polymerase. The isolated fragments were sequenced from both directions. The ubiX sequence of IS16 differed from the published sequence only at codon 98 (shown in boldface in Fig. 2), where a serine residue (AGC) changed to arginine (AGA). Surprisingly, the ubiX sequence of the parent strain THU was identical to that of strain IS16.

View larger version (14K): [in this window] [in a new window]   FIG. 4.   Inhibition by dithiothreitol of growth of strains AN387 and AN385 (ubiA). MGR, mean growth rate.

View larger version (13K): [in this window] [in a new window]   FIG. 5.   Inhibition of dsb mutant strains by 1-thioglycerol. Closed circles, parent JCB 570; open diamonds, dsbA mutant JCB 571; closed squares, dsbB mutant JCB 789; open inverted triangles, dsbAB mutant JCB 758. The cultures were grown in minimal medium as described in Materials and Methods. MGR, mean growth rate.

View larger version (18K): [in this window] [in a new window]   FIG. 6.   Induction of alkaline phosphatase in strains THU, IS16, IS16B1, AN385, and AN387. Cultures were grown in repressing, high-phosphate-containing medium until A540 was 0.4 to 0.5. Following washing and resuspension in low-phosphate, inducing medium, the enzyme levels were determined at the indicated times. (A) Strains THU (closed circles), IS16 (closed inverted triangles), and IS16B1 (closed squares). (B) Strains AN385 (closed circles) and AN387 (closed inverted triangles).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The low ubiquinone level of strain IS16 was puzzling, since E. coli has two genes, ubiD and ubiX, for the decarboxylation of octaprenyl p-hydroxybenzoate. It required multiple copies of the wild-type ubiX gene to increase this strain's ubiquinone content. The ubiX gene of IS16 had a single deviation from the wild-type sequence, and increasing the number of copies of this variant gene was insufficient to elevate its ubiquinone content to near that of the wild type. It appears that the mutation at codon 98 of the ubiX gene practically inactivated the enzyme. Therefore, serine residue 98 is likely essential for the catalytic role of polyprenyl p-hydroxybenzoate lyase.

Since strain THU, the parent strain of IS16, also carried this mutation, both strains must have been ubiX negative. A reasonable guess is that the mutation in IS16, which resulted in very low ubiquinone levels, had to occur in the ubiD gene. This will have to be confirmed when the ubiD gene is identified.

Ubiquinone deficiency in E. coli is known to give rise to a pleiotropic phenotype of increased resistance to some antibiotics and heat inactivation; the inability to grow on succinate as the sole carbon source; and increased sensitivity to hydrogen peroxide, methyl methanesulfonate, and gamma radiation (8). The results reported here add another phenotypic characteristic to low ubiquinone content, thiol hypersensitivity.

Correlation between low ubiquinone content and thiol hypersensitivity was shown by (i) the loss of thiol hypersensitivity when the ubiquinone level was raised via complementation, (ii) the lessening of thiol hypersensitivity when ubiquinone levels were slightly elevated with multiple copies of the mutated ubiX gene, and (iii) the demonstration of thiol hypersensitivity in a ubiA mutant strain.

Previously published screens for thiol hypersensitivity, with dithiothreitol as the reducing agent, turned up dsbA dsbB, as well as trxA thioredoxin and trxB (thioredoxin reductase), mutants (27). The screen reported here employed 1-thioglycerol, a weaker reducing agent than dithiothreitol. This could be the explanation for finding a thiol-hypersensitive mutant which differed from the previously reported varieties.

A recent report suggests that there is a direct link between the respiratory chain and the DsbA-DsbB disulfide bond-forming system (22). Electrons removed from the periplasmic cysteine residues during disulfide bond formation pass first to the DsbA protein, then to the cytoplasmic membrane-associated DsbB protein, and finally to the respiratory chain. In support of their thesis, the authors showed that a ubiA menA double mutant, when deprived of para-hydroxybenzoate, slowed its growth (presumably because of reduction of ubiquinone content) and accumulated first reduced forms of DsbA and DsbB proteins and then the DsbA-DsbB complex.

This finding could explain how continually low ubiquinone levels such as those seen in strains IS16 and AN385 would reduce the ability of the respiratory chain to accept electrons from the DsbA-DsbB complex. Accumulation of reduced DsbA and DsbB proteins and the DsbA-DsbB complex in turn would result in a Dsb phenotype.

The thiol hypersensitivity of dsb mutants could be the result of accumulation of reduced periplasmic and outer membrane proteins. Dysfunctional (reduced) DsbA proteins would prevent the appropriate folding and integration of outer membrane proteins (28). This scenario gains further credence from an earlier observation that exposure to 1-thioglycerol causes a rapid shift-up in the synthesis of outer membrane proteins OmpA and OmpF (17).

Thiol hypersensitivity of low-ubiquinone-containing cells would be the final outcome of the inability of the respiratory chain to absorb electrons from reduced Dsb proteins in the presence of excess exogenous thiols. Here millimolar concentrations of 1-thioglycerol or dithiothreitol outcompete the cysteine residues of periplasmic proteins for oxidation by DsbA. Growth inhibition would be for the same reasons as in the case of dsb mutants.

It was of interest that multiple copies of the ubiX gene in strain IS16B1 resulted in a 1.5-fold increase in ubiquinone content over that of the parent strain of THU. This implied that, in strain THU, the decarboxylation of octaprenyl p-hydroxybenzoate was rate limiting. However, since strain THU itself is a ubiX mutant, this observation may be applicable only to this strain. Further work will be required to determine whether this reaction is one of the flux-determining steps in ubiquinone biosynthesis in E. coli.