Circulating isolates of Neisseria mucosa do not inhibit the growth of Neisseria gonorrhoeae

: We used agar overlay assays to assess if 24 circulating and historical isolates of Neisseria mucosa could inhibit the growth of 28 circulating and historical isolates of N. gonorrhoeae. We found no evidence of inhibition by N. mucosa (n=24). Positive controls Streptococcus pneumoniae and Escherichia coli demonstrated a strong inhibitory effect against the growth of N. gonorrhoeae .


Introduction
Does Neisseria mucosa inhibit the growth of Neisseria gonorrhoeae, as recently reported by Aho et al. [1]? This is an important question as it may help explain the paradoxical findings of our recently published PReGo Study [2]. This placebocontrolled trial randomized high-risk men who have sex with men to intensive use of Listerine® mouthwash and gargle or placebo to try to reduce the incidence of bacterial STIs in this population, where oral sex plays an important role in STI transmission [3][4][5]. The study found that Listerine increased rather than decreased the incidence of oropharyngeal N. gonorrhoeae. Listerine® had a similar though statistically nonsignificant effect in the OMEGA study [6]. N. mucosa is a healthy core component of the oropharyngeal microbiome and even low concentrations of Listerine® have been shown to be bacteriocidal to Neisseria spp [7]. If the use of the Listerine® mouthwash reduced the prevalence/abundance of N. mucosa and N. mucosa inhibits the growth of N. gonorrhoeae, then Listerine® could increase the susceptibility for N. gonorrhoeae infection [2]. In a similar vein, a randomized controlled trial established that nasal inoculation with N. lactamica reduced the incidence of colonization with N. meningitidis [8]. If the in-vitro anti-gonococcal effect of N. mucosa could be confirmed, N. mucosa might be evaluated as a probiotic to prevent gonococcal infection.
As a first step in evaluating this hypothesis, we tested if our locally circulating isolates of N. mucosa and other commensal Neisseria, including those circulating in the PReGo participants, were able to inhibit N. gonorrhoeae. We found that N. mucosa did not exert this inhibitory effect. The inhibitory effect of a number of bacterial species such as various streptococcal species on N. gonorrhoeae is well established [9][10][11][12][13].
We included a number of these species as positive controls in our experiments.
We also obtained one of the N. mucosa strains used by Aho et al. and confirmed that, in our laboratory, this strain did not inhibit the growth of N. gonorrhoeae. We did, however, note that N. mucosa exerted a repellant effect on the top agar layer in the agar overlay experiments. This created an optical illusion of reduced growth of N.
gonorrhoeae, which may explain the discordant findings between Aho et al., and ourselves.

Inhibitory ("producer") bacterial isolates
We obtained Neisseria isolates from two clinical studies conducted at our centre:

PReGo (Preventing Resistance in Gonorrhoea Study) isolates
This was a single center randomized controlled trial at the Institute of Tropical Medicine in Antwerp, Belgium, between 2019 and 2020 that assessed the efficacy of an antiseptic mouthwash to prevent STIs among 343 MSM using PrEP [2]. A subgroup of 64 participants were selected to participate in a survey of their oropharyngeal microbiomes [2].

ComCom (Commensals in the Community Study) isolates
In June 2020, Institute of Tropical Medicine (ITM) employees were invited to participate in a survey of their oropharyngeal microbiomes [14]. The first 32 eligible employees (male or female) presenting to the study team were included in this survey.

Culture
The details of how the samples were taken and processed has been described elsewhere [2]. In brief, for each individual, an oropharyngeal swab (ESwab TM COPAN Diagnostics Inc., Italy) was taken and inoculated onto blood and Modified Thayer-Martin agar plates using the streak plate technique and incubated at 35-37°C and 5% CO2. Plates were examined after 48 hours and Neisseria-like colonies were selected based on a positive Oxidase test and a Gram stain. Neisseria-like colonies were enriched on blood agar plates and stored in skim milk at -80°C.

Identification
Cultures of Neisseria-like colonies were shipped to Centre Hospitalier Universitaire Saint-Pierre where species were identified using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS).

Agar overlay inhibition testing
We tested all N. mucosa isolates obtained during these two studies (n=14) as well as a random selection of N. meningitidis (n=3) and other commensal Neisseria obtained from these two studies -N. subflava (n=4), N. cinerea (n=2), N. lactamica (n=1), N. oralis (n=3), N. elongata (n=1) and Neisseria spp. (n=1; Table 1). We also included 6 N. mucosa isolates from our ITM historical collection. Five of these were ATCC strains and one was a historical clinical specimen obtained from a patient in 1977. The DSM4631/ATCC 25996 isolate used by Aho et al., was obtained from the DSMZ (https://www.dsmz.de/collection/catalogue/details/culture/DSM-46).

N. gonorrhoeae target strains
Three strains of N. gonorrhoeae were used as target strains for all experiments (WHO-F, WHO-X and MoNg003a clinical isolate obtained from an individual with asymptomatic pharyngeal N. gonorrhoeae infection attending our STI clinic in 2020. In addition, one ATCC strain of N. gonorrhoeae, WHO-W and 23 other circulating strains of N. gonorrhoeae were tested against some of the putative inhibitory bacteria (Table 1).

Agar overlay assay
The details of the agar overlay assay have been described elsewhere [1]. Briefly, all strains used in the experiment were propagated on blood agar plates for 18-24h. The cultures were suspended in 10 µl of phosphate-buffered saline (PBS) containing 10 9 CFU/ml of producer strains. These were spotted onto GC agar and incubated for 24h. 10 ml of melted GCB agar containing 10 6 CFU/ml of a target strain was added to each spotted plate. The plates were then re-incubated for 24 to 48 hours. The diameter of the zone of inhibition surrounding each producer strain was assessed at 24 hours.

Results
None of the commensal Neisseria or N. meningitidis exhibited any activity against N. gonorrhoeae ( Table 1).
The colonies of N. mucosa exhibited a consistently repellant effect on the layer of agar poured over them (Fig. 1). This created 'pitting colonies' or a convex slope between the top of the second layer of agar and the edge of each N. mucosa colony, which created an illusion of reduced N. gonorrhoeae growth around each N. mucosa colony [15]. Closer inspection, however, confirmed that N. gonorrhoeae growth over this convex slope around the N. mucosa colonies was not macroscopically distinguishable from that elsewhere (Fig. 1).
The isolate of Streptococcus pneumoniae demonstrated clear evidence of inhibition against all 9 strains of N. gonorrhoeae (median diameter of inhibition = 21 mm). The inhibitory effect of E. coli was less pronounced. Inhibition was evident in 3 out of 9 N. gonorrhoeae strains testedmedian diameter of inhibition 11mm.

Discussion
Unlike Aho et al., we could find no evidence that N. mucosa or any other commensal Neisseria was able to inhibit the growth of N. gonorrhoeae [1]. This was despite using a large number of clinical and reference strains of N. gonorrhoeae as target strains, and the largest collection of commensal Neisseria tested to date as inhibitory bacteria.
How can these discordant findings be explained? Aho found this inhibitory effect in 3 out of 5 N. mucosa isolates. The isolates were all taken from ATCC collections and the inhibition assays were performed by undergraduates as part of a microbiology course.
No photos were provided of the agar overlay assays showing that N. mucosa inhibited the growth of N. gonorrhoeae. However, one image of N. mucosa inhibiting the growth of N. flavescens was provided.
In our study, we followed an identical agar overlay protocol utilizing a larger panel of isolates of N. mucosa and N. gonorrhoeae. The experiments were performed by a laboratory technician with over 25 years of experience culturing Neisseria species (SA). The plates were examined by this person and two others with extensive experience in culturing Neisseria species (CK and JL). All three concurred that pitting around each colony of N. mucosa created an optical illusion of decreased growth around the colony. Close inspection confirmed that there was no inhibition of growth.
We consider this a parsimonious explanation for the different findings between the two studies. It could be possible that only certain strains of N. mucosa are able to inhibit specific strains of N. gonorrhoeae and that we did not include any of these combinations in our experiments. We did, however, test one of the three isolates of N. We cannot completely exclude the possibility that an unevaluated different experimental condition such as the source of the agar used was responsible for the differences in the results between the two studies.
Our study, unlike that of Aho et al., did include positive controls. These showed clear and consistent evidence of inhibition. Taken together, these findings suggest that N. mucosa is unlikely to have a significant inhibitory effect on the growth of N.
gonorrhoeaeat least in the agar overlay assays evaluated here. More importantly for our current research, we consider it unlikely that a broad range of N. mucosa isolates contains a sufficiently potent compound against our currently circulating strains of N. gonorrhoeae to be able to explain the findings of the PReGo and OMEGA studies.
We concur with Aho et al., that commensal microbes represent a possible source of antimicrobial compounds that could play an important role in reducing the emergence of AMR in N. gonorrhoeae and other bacteria. Based on our findings, we consider it more likely that such anti-gonococcal compounds will be discovered from organisms such as S. pneumoniae than N. mucosa [12,16].