Differentiation of Campylobacter Isolates on the Basis of Sensitivity to Boiling in Water as Measured by PCR-Detectable DNA

Abstract

Differential sensitivity for the release of PCR-detectable genomic DNA upon boiling in water is reported for 45 Campylobacter jejuni and Campylobacter coli strains isolated in Egypt. All of the strains released PCR-detectable DNA when treated with proteinase K and sodium dodecyl sulfate. When DNA was extracted from these strains by boiling in water, nine (20%) of the strains were PCR negative or resistant to boiling, suggesting the presence of boiling-sensitive and boiling-resistant phenotypes.

Campylobacter jejuni and Campylobacter coli are leading bacterial causes of diarrhea in humans (6, 7, 10). These organisms are gram-negative spiral bacteria which possess complex flagella (2). In an attempt to develop a rapid and specific diagnostic method for these pathogens, Oyofo and coworkers (4) developed a PCR assay based on a primer pair designed from a conserved 5′ region of the flaA gene of C. coli VC167. The assay used a boiling strategy to release genomic DNA from the bacteria. The PCR assay was shown to be specific for C. jejuni and C. coli and negative for other Campylobacter spp. and other enteric bacterial pathogens. However, in some instances, C. jejuni and C. coli isolates from Egypt failed to yield a positive PCR result. To further explore this observation, we tested a collection of C. jejuni and C. coli isolates from Egypt for their differential sensitivity to boiling in water as measured by the release of PCR-detectable DNA.

A total of 45 randomly selected C. jejuni and C. coli strains isolated from field studies in Upper Egypt of pediatric diarrhea were used. These strains were confirmed biochemically and morphologically, then were grown on Skirrow’s medium (5) at 42°C for 48 h under an atmosphere of 10% CO₂, 5% O₂, and 85% N₂.

Two methods were used for the preparation of DNA for PCR. The first was the boiling method (9), in which a loopful of bacteria was washed with 1 ml of distilled water (dH₂O) and resuspended in 200 μl of dH₂O. The suspension was boiled for 10 min, placed on ice for 5 min, and then centrifuged for 5 min in a microcentrifuge at 10,000 × g to extract DNA. The second method (8) involved bacterial lysis by treatment with proteinase K and sodium dodecyl sulfate (SDS). The PCR assay was performed as described by Oyofo and coworkers (4) with previously described pg50 and pg3 primers. A Salmonella Vi antigen gene amplification system (1a) was used as a control to test for the presence of inhibitors in Campylobacter reaction mixtures. After amplification, 20 μl of the reaction mixtures was analyzed on a 2% agarose gel containing ethidium bromide, which was run at 100 V for 2 h. The presence of a 450-bp band indicated the successful amplification of the Campylobacter flagellin gene and Salmonella Vi antigen gene, for both of them have the same gene product size.

Forty-five isolates were boiled in water, and the supernatant was subjected to PCR to detect genomic DNA. Nine isolates (four C. jejuni, three C. coli, and two C. jejuni and C. coli) yielded negative PCR results (Table 1). These nine isolates were retested by PCR three times but still yielded negative results.

TABLE 1.

Results of PCR amplification of genomic DNA isolated from C. jejuni and C. coli by the boiling and proteinase K-SDS methods

Campylobacter species	No. tested	No. (%) PCR negative by:
		Boiling	Proteinase K-SDS
C. coli	7	3 (43)	0
C. jejuni	33	4 (12)	0
C. jejuni or C. coli	5	2 (40)	0

Supernatants from two of the PCR-negative Campylobacter isolates did not inhibit the amplification of the targeted S. typhi gene sequence (data not shown). Preparation of the genomic DNA from these PCR-negative strains by the proteinase K-SDS treatment resulted in the detection of the expected 450-bp Campylobacter band in all of the strains tested (Table 1). When the possibility of having mismatches between primer and target sequences of both was explored with purified DNAs from PCR-negative and PCR-positive isolates, no differences in the intensities of PCR-amplified products were seen with 50, 10, 2, or 0.1 ng of DNA sample templates (data not shown). Bands produced from 100 pg were very faint but of equal intensities. However, use of 10 pg of the same DNAs produced no bands. To determine whether the PCR-negative isolates have failed to lyse by boiling, PCR-negative and PCR-positive isolates were boiled, the supernatants were extracted with phenol-chloroform to remove the proteins, and nucleic acids were precipitated. When the optical density was determined, it was found that the PCR-negative isolates gave values that were slightly lower than those obtained from the PCR-positive isolates. To determine the identity of the absorbing material, most of the sample was loaded onto 1% agarose gel and electrophoresed. The PCR-positive samples showed high-molecular-weight genomic DNA at the top of the gel and RNA at the bottom. The PCR-negative samples showed mostly RNA and a very faint smear of low-molecular-weight degraded DNA (Fig. 1).

FIG. 1.

Nucleic acids released by the boiling method. Lanes 1 to 9, PCR-negative strains; lanes 10 to 18, PCR-positive strains.

These results suggest that two phenotypically distinct subgroups of Campylobacter strains exist with respect to their relative capacity to release intact high-molecular-weight DNA upon boiling in water. Whereas upon boiling most isolates readily released genomic DNA, which was detected by PCR, a group of isolates (20%) failed to release intact high-molecular-weight genomic DNA but was able to release RNA and some low-molecular-weight degraded DNA. It appears that these isolates do not completely respond to boiling and instead allow the passage of small molecules (such as RNA and proteins which acquire compact conformations) and hinder the passage of high-molecular-weight molecules, which have an extended large conformation, such as chromosomal DNA.

The variability among these Campylobacter strains in releasing PCR-detectable DNA upon boiling shows that these strains may be divided into two phenotypically distinct subgroups. During the course of this investigation, a similar observation was reported with C. jejuni (3) in the United States, suggesting a global distribution of strains displaying this phenomenon.

It has been established that DNA amplification results with samples prepared by the boiling method may vary significantly among different bacteria. For example, boiling was shown to result in successful DNA amplifications from Leptospira hardjobovis (9) and noncapsulate Haemophilus influenzae (11). In contrast, the same method yielded negative results when applied to Mycobacterium tuberculosis (1), possibly due to differences in cell surface structures. Our data suggest that within the Campylobacter population there is also a subset which does not release PCR-detectable DNA upon boiling in water. Based on these findings, the use of proteinase K-SDS for extraction of DNA is recommended over boiling for a more sensitive and accurate detection of Campylobacter strains.