Survival of Neisseria gonorrhoeae Isolates of Different Auxotypes in Six Commercial Transport Systems

Abstract

We determined the rates of survival of six clinical and two control strains of Neisseria gonorrhoeae in six commercial transport systems, using NCCLS standard M40-A methodology. Differences in strain recovery and system performance were marked. A strain less robust than the recommended ATCC 43069 would provide a more exacting standard.

As direct inoculation of clinical material onto bacteriological media is not always possible, transport systems play an important role in the diagnosis of gonorrhea. In a previous study we demonstrated varied rates of survival of Neisseria gonorrhoeae isolates with different auxotypes in a commercial transport medium (5). An auxotype denotes an absolute requirement for one or more specific growth factors (see below) and may be associated with other strain characteristics. Here we apply the quality control methodology of NCCLS standard M40-A (www.nccls.org) to determine to what extent six commercial transport swab systems preserve different auxotypes of N. gonorrhoeae.

Six clinical isolates of N. gonorrhoeae and two control strains, ATCC 43069 (recommended by the M40-A standard) and NCTC 8375, were studied. The clinical strains had previously been recovered in our laboratory from plates directly inoculated with clinical material; they were auxotyped by the method of Copley and Egglestone (1) and subsequently kept at −70°C. These strains included one that does not require any specific factors and one each that does require proline (P⁻), arginine (A⁻), hypoxanthine (H⁻), proline plus arginine (P⁻A⁻), and arginine plus hypoxanthine (A⁻ H⁻). These were strains considered to have survival curves typical of their auxotype (5). Six widely available swabs from three established manufacturers were evaluated: Copan Diagnostics' M40 Transystem, Medical Wire & Equipment's Transwab, and Starplex Scientific's Starswab II (all were tested with and without charcoal).

Each strain was grown for 18 h on chocolate agar (Oxoid, Basingstoke, United Kingdom) in 5% CO₂ at 37°C, suspended in phosphate-buffered saline (PBS) and adjusted to a 0.5 McFarland standard with a colorimeter. Tenfold dilutions of this suspension to 10⁻⁴ were prepared in PBS, and for each dilution, 50 μl was inoculated onto each of three chocolate agar plates with a spiral plater (Don Whitley, Shipley, United Kingdom).

For each strain, nine swabs of each product were inoculated, using a Gilson pipette, with 100 μl of the 10⁻¹ dilution of the initial inoculum to provide three swabs each for sampling at 0, 24, and 48 h. Each strain was tested against all swabs in a single experiment. Time zero samples were taken within 15 min of swab inoculation. Swabs for the 24- and 48-h sampling were kept at 22.5°C in a chilling incubator. At each sampling time point, bacteria were recovered by vortexing swab tips in 1 ml of PBS for 15 s and then expressing excess fluid. For each washing, serial 10-fold dilutions to 10⁻³ were prepared, and 50 μl of each dilution, including the neat washing, was plated in triplicate (rather than duplicate as recommended by the NCCLS M40-A standard) onto chocolate agar with a spiral plater. After incubation at 37°C in 5% CO₂ for 18 h, the dilution most suitable for counting was selected, and for each strain-product-time combination, the mean of nine counts (three swabs of that product times three plate counts) was taken. From this, the percent recovery of and log₁₀ reduction from the time zero count were calculated.

The initial inocula were between 1.82 × 10⁷ and 1.15 × 10⁸ CFU/ml and therefore within the limits prescribed by the NCCLS M40-A standard; time zero counts were between 2 × 10⁴ and 4.55 × 10⁵ CFU. Results are presented graphically in Fig. 1 and summarized in Table 1, for which we applied the NCCLS M40-A acceptability cutoff for ATCC 43069 at 24 h (3 log₁₀ − 10% reduction from the count at time zero) to all strains and both sampling times.

FIG. 1.

Recovery of eight strains of N. gonorrhoeae from six commercial transport systems. NR, none recovered.

TABLE 1.

Recovery of eight strains of N. gonorrhoeae from six commercial transport systems

Company and product	No. of strains with acceptable level of recovery/total no. tested ata:
	24 h	48 h
Copan Diagnostics
Amies	8b/8	6b/8
Amies + charcoal	8b/8	6b/8
Medical Wire & Equipment
Amies	0/8	0/8
Amies + charcoal	7b/8	0/8
Starplex Scientific
Amies	3b/8	0/8
Amies + charcoal	6b/8	0/8

^aResults are expressed in compliance with the NCCLS M40-A standard. We applied the criteria of the NCCLS for ATCC 43069 (≤3 log₁₀ − a 10% reduction from the count at 0 h to that at 24 h) to samples at both 24 and 48 h.

^bThe number includes ATCC 43069, and therefore the value is equivalent to a “pass” at 24 h.

Essentially, the methods in this study were as recommended by the NCCLS M40-A standard (www.nccls.org) except that inocula were delivered to swabs with a pipette, 48-h samples were taken, and counting was performed in triplicate with a spiral plater. With the exception of Medical Wire & Equipment's Transwab without charcoal, all systems performed satisfactorily with respect to the standard (for ATCC 43069 at 24 h) (Table 1).

In vitro and clinical evaluations of transport medium for N. gonorrhoeae reveal variation in rates of survival of uncharacterized isolates within and between such studies (2-4). As in our previous study (5), auxotypes P⁻ and H⁻ were among the least robust. ATCC 43069 proved to be relatively hardy, and NCTC 8375 (used for quality control purposes in the United Kingdom) was comparable. Although our results confirm the benefits of charcoal, at 24 h Copan Diagnostics' M40 Transystem without charcoal performed better than other manufacturers' charcoal-containing products. Not only were tests using the M40 Transystem with and without charcoal the only ones to preserve all strains at 24 h, both showed good preservation up to 48 h.

As well as risking missed diagnoses, the use of transport media incapable of preserving all strains of gonococci may have epidemiological implications, as the least robust strains—and characteristics associated with them—may be underrepresented in a sampled patient population. Our results reconfirm substantial variation in the rates of survival of different strains of N. gonorrhoeae and in the ability of different commercial systems to preserve them.

The NCCLS M40-A quality control standard provides a much-needed performance standard for manufacturers as well as a means for laboratories to compare and evaluate transport systems. Our results with N. gonorrhoeae suggest that such a means of evaluating performance, using a relatively robust control strain, may not necessarily reflect the in-use value of a product. Our results, expressed not only in compliance with the NCCLS standard (Table 1) but also graphically (Fig. 1), suggest that a less hardy control strain would represent a more exacting standard.