The genus Moraxella comprises catalase and oxidase-positive, nonmotile, non-encapsulated and asaccharolytic Gram-negative coccobacilli or diplobacilli⁵, which show slow growth on blood agar and poor growth on chocolate agar. After 24h, colonies appear circular, small, gray and translucent, turning opaque with increased incubation time². These coccoid or coccobacillary organisms (plump rods), occurring predominantly in pairs and sometimes in short chains, tend to resist decolorization in the Gram stain. Most Moraxella species are susceptible to penicillin and its derivatives, cephalosporins, tetracyclines, quinolones, and aminoglycosides. (-Lactamase production has been rarely reported for Moraxella species other than Moraxella catarrhalis, of which most isolates produce an inducible, cell-associated (-lactamase⁸.

These are low-virulence microorganisms, which are commonly found as commensals of the upper respiratory tract⁴. Occasionally, they can cause infections in vulnerable patients. M. catarrhalis is a leading cause of otitis media in children and acute exacerbations of chronic obstructive pulmonary disease³. Moraxella species other than M. catarrhalis, such as Moraxella lacunata, Moraxella nonliquefaciens and Moraxella osloensis, are mainly associated with ocular infections such as endophthalmitis, keratitis and blepharoconjunctivitis, sometimes with stromal involvement and corneal perforation; although they have also been isolated from invasive infections in humans, including endocarditis, septic arthritis, meningitis, bacteremia, and pericarditis⁷. M. nonliquefaciens and M. osloensis are the two species most frequently isolated, approximately in equal numbers, from nonrespiratory clinical material, especially blood cultures from patients at risk. Moraxella lincolnii is not frequently isolated from clinical samples. Moraxella canis has been isolated from dog bite wounds and from debilitated patients⁸. M. lacunata is the causative agent of 2% of bacterial ocular infections .⁷ Local predisposing factors include wearing contact lenses, previous eye surgery, ocular trauma, and a previous history of Herpes simplex keratitis. Immunocompromised patients, alcoholism, malnutrition, old age, diabetes mellitus and thyroid disease are systemic risk factors⁴.

Due to the very frequent finding of Moraxella spp. in corneal abscesses, various isolates were referred to the Reference Laboratory (RL) to assess the heterogeneity of species within the genus and also the molecular diversity within the species level, to confirm or rule out whether they are part of a clonal complex.

Of a total of 40 Moraxella isolates obtained in the Eye Hospital between January 2018 and November 2019, 17 were randomly selected for an initial study of their identification and clonal diversity. The PFGE technique is widely used for the subtyping of various bacterial species; therefore, we decided to evaluate its functionality in the molecular typing of Moraxella spp. To the best of our knowledge, there are no previous reports that evaluate this technique, nor is it established which restriction enzyme shows a discriminatory power that makes it possible to compare and establish the genetic relationship between isolates of this genus, other than M. catarrhalis ³.

The characteristics of the population affected by these types of microorganisms were analyzed in accordance with the available epidemiological information. The antibiotic susceptibility of all isolates was tested as well.

Seventeen (17) Moraxella spp. isolates obtained from corneal scrapings, which were referred to the RL between January 2018 and November 2019, were analyzed. The median age of patients was 47 years. Thirteen (13) isolates (76.5%) were obtained from male patients. Local risk factors, such as glaucoma (4 patients) and cataracts (1 patient), and systemic risk factors, such as HIV (1 patient), diabetes (3 patients) and arthritis (1 patient), were identified in this study. One or more of these factors were present in nine of the seventeen patients.

The strains were subcultured on blood agar in an atmosphere of 5% CO₂ at 37^oC for 24h. Isolates were observed under a microscope using the Gram staining technique, and both pigment and hemolysis were recorded. The biochemical characterization was carried out using manual phenotypic tests (oxidase, catalase, growth on MacConkey agar, TSI, gelatin hydrolysis, urease, nitrate reduction, DNase and alkalinization with acetate) and by using commercial miniature-version API® 20NE galleries (bioMérieux, France) according to the manufacturer's instructions. Incubation temperature was 37(C and the results of the galleries and biochemical tests were read after 24, 48 and 72h.

Isolates were also identified by mass spectrometry (MALDI-TOF MS Bruker Daltonics, Bruker Taxonomy Database 3.3.1 version+CDC-Microbenet database) using the direct transfer-formic acid method: MALDI-TOF target plates were inoculated into the spots by picking a freshly grown overnight colony and overlaid with 1(l of 70% formic acid (Sigma-Aldrich). Each spot was allowed to dry and subsequently overlaid with 1(l of matrix solution (a cyano-4hydroxycinnamic acid in 50% acetonitrile, 2.5% TFA). The score cut-offs recommended by the manufacturer were used, a score value ≥2,0 indicates species-level identification, a score value between 1,7 and 1,99 indicates genus-level identification, and scores <1,7 indicate no reliable identification. Additionally, the ``10% rule'' was applied, which states that any species scoring >10% below the top-scoring match may be excluded.

Antibiotic susceptibility testing was performed using the disk-diffusion method (the Kirby-Bauer method). The interpretation criterion used is the one established in the CLSI guideline M45-A2 for M. catarrhalis. The culture conditions used were an incubation temperature of 35(C, an atmosphere of 5% CO₂ and Mueller Hinton as the culture medium. The ``susceptible'' category was derived from the extrapolation of the standardized cut-off values for the M. catarrhalis species, which is the only species of the genus with standardized cut-off values¹. The following antimicrobials were tested: amoxicillin-clavulanate, erythromycin, azithromycin, tetracycline, and trimethoprim-sulfamethoxazole.

Figure 1: Band patterns obtained by PFGE of M. nonliquefaciens (lanes 2 and 3) and M. lacunata (lanes 4---9) using Smal. Patternsobserved for Xbal-PFGE can be observed in lanes 10---14 (lanes 10 and 11 M. nonliquefaciens; 12---14 M. lacunata). SalmonellaBraenderup H9812 was used as a molecular weight marker, digested with Xbal.

Patient isolates were analyzed by PFGE to assess genetic relatedness. In brief, chromosomal DNA from the Moraxella isolates was digested with restriction endonucleases SmaI and XbaI to determine their typeability and discriminatory power. Digestion was performed with 30 units of XbaI (10U/(l, Fermentas) for 3h at 37(C and with 50 units of SmaI (10U/(l, Invitrogen) for 3h at 30(C. Salmonella Braenderup H9812 was used as a molecular weight marker, which was digested using XbaI enzyme under the same conditions as the ones previously described. With regard to the running conditions, an initial time of 5s was used and then a final time of 35s, the total run-time was 19.5h. Banding patterns were visually analyzed and interpreted following the criteria described by Tenover et al. (1995) for the typing of bacterial species⁶.

Gram staining showed gram-negative coccobacilli or diplobacilli. Isolates were neither hemolytic nor pigmented. Of the 17 isolates analyzed, 15 turned to be M. lacunata and 2 were classified as M. nonliquefaciens by biochemical tests.

Table 1: Table 1 shows the traditional biochemical tests that allow to identify different species of Moraxella.

Using API® 20NE, the biocode obtained for 15 isolates was 1-0-1-0-0-0-0, 100% concordant with M. lacunata species. For the remaining two isolates, the obtained biocode was 1-0-0-0-0-0-0, 100% concordant with Moraxella spp. Identification by MALDI-TOF MS showed the following results: 4 isolates were identified as Moraxella sp., with score values between 1860 and 1944, 11 as M. lacunata with score values between 2012 and 2443; and 2 isolates were identified as M. nonliquefaciens with score values between 2146 and 2187. Taking into account the 4 isolates identified to the genus level through MALDI-TOF hydrolyzed gelatin, it is worth mentioning that combining basic biochemical tests and MALDI-TOF MS enabled a correct identification at the species level of the Moraxella spp. isolates that were referred to the RL. It should be noted that retrieving M. catarrhalis from corneal abscesses in the Eye Hospital is about 1 out of every 100 Moraxella isolates, which shows a clear predominance of the M. lacunata species in this type of infection.

With regard to antibiotic susceptibility, only 8 of 17 isolates grew on Mueller Hinton. In 8 strains, antibiotic susceptibility testing was carried out in Mueller Hinton medium with 5% horse blood. One of the isolates did not grow on any media tested, therefore, its susceptibility profile could not be determined. All 16 isolates were susceptible to all the antibiotics tested.

Of the two enzymes that were tested, it was only with SmaI that interpretable restriction patterns could be obtained. Therefore, SmaI was selected to test all the strains. Fig. 1 Fig. 1 shows the resolution of the PFGE patterns produced by SmaI and XbaI in a subgroup of the total isolates tested. Of the 15 M. lacunata isolates analyzed, 13 different digestion patterns were obtained. As shown in Fig. 2 Fig. 2, the isolates were distributed into eight clonal types (designated with letters A-H) depending on whether there was a difference of four or more bands among the profiles. Clonal type A was divided into subtypes (A1-A6) since there was just one difference of 1 or 3 bands among these isolates. Subtypes A1 and A2 are comprised of two isolates each, which showed the same banding pattern. The two M. nonliquefaciens isolates, identified as 791 and 403, showed a difference of more than 4 bands from one another.

With regard to antibiotic susceptibility, all isolates were susceptible to the five antimicrobials tested. As for the antibiotic treatment administered to the infected patients, after the ocular sample was collected, an intensive medical treatment was recommended, which included fortified antibiotic eye drops of vancomycin (50mg/ml) and ceftazidime (50mg/ml), and cycloplegic eye drops for pain management. Once the identification and susceptibility results were obtained, vancomycin was removed and the treatment continued with ceftazidime eye drops and, in some cases, with moxifloxacin eye drops as well.

Comparing the results obtained by PFGE with restriction enzymes XbaI and SmaI, it could be observed that the digestion with XbaI produced low molecular weight fragments that could not be adequately resolved in the gel; therefore, this enzyme was considered inappropriate for the analysis of these species by PFGE. The same result was documented for M. catarrhalis when digestion with XbaI was analyzed⁹. Thereby, SmaI, another restriction enzyme, was tested. In most of the literature reviewed, SpeI is used for PFGE of M. catarrhalis, being its drawback that it is a very expensive restriction enzyme, which is not available in our laboratory.

SmaI produced restriction patterns ranging from 11 to 15 bands, which enabled sample comparison and typification. Thirteen (13) different restriction profiles can be observed among the 15 M. lacunata isolates. Subtypes A1 and A2 appear to be genetically indistinguishable, since their PFGE patterns have the same number of bands with the same apparent size. The remaining clonal subtypes represent a single isolate. Due to the genetic diversity obtained, an outbreak could be ruled out and they were, evidently, different isolates. Based on the analysis following Tenover’s criterion⁶, those isolates that belong to the same clonal type were considered closely related isolates.

This study demonstrated that digestion with a single enzyme, SmaI, can confirm or rule out the clonal relationship of M. lacunata and M. nonliquefaciens isolates by PFGE. Moreover, it was concluded that the use of MALDI-TOF MS and biochemical tests, alone or in combination, allows the identification of M. lacunata and M. nonliquefaciens. It is important to mention that 16S rRNA gene sequencing is another useful tool to differentiate Moraxella species such as M. catarrhalis, M. nonliquefaciens, M. lincolnii and M. osloensis. With reference to antibiotic susceptibility, it should be noted that the report should clarify that the interpretation of the categories ``susceptible'', ``resistant'' or ``intermediate'' is based on the extrapolation of M. catarrhalis cut-off points, and that in the future it should be evaluated whether this extrapolation is valid for all Moraxella species. This article highlights the importance of surveillance in the distribution of Moraxella species in corneal abscesses to recognize changes over time, to detect the emergence of clones and, if they do emerge, to reveal if they are associated with some antibiotic resistance in particular.

Funding

This study has been funded by the <ce:grant-sponsor id="gs1" sponsor-id="http://dx.doi.org/10.13039/100003882">National Institute of Infectious Diseases INEI-ANLIS Dr Carlos G Malbrán</ce:grant-sponsor>.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Received: 10-5-2021

Accepted: 14-3-2022