Antimicrobial Susceptibility of Clinical Oral Isolates of Actinomyces spp.

Actinomyces species play an important role in the pathogenesis of oral diseases and infections. Susceptibility testing is not always routinely performed, and one may oversee a shift in resistance patterns. The aim of the study was to analyze the antimicrobial susceptibility of 100 well-identified clinical oral isolates of Actinomyces spp. against eight selected antimicrobial agents using the agar dilution (AD) and E-Test (ET) methods. We observed no to low resistance against penicillin, ampicillin-sulbactam, meropenem, clindamycin, linezolid and tigecycline (0–2% ET, 0% AD) but high levels of resistance to moxifloxacin (93% ET, 87% AD) and daptomycin (83% ET, 95% AD). The essential agreement of the two methods was very good for benzylpenicillin (EA 95%) and meropenem (EA 92%). The ET method was reliable for correctly categorizing susceptibility, in comparison with the reference method agar dilution, except for daptomycin (categorical agreement 87%). Penicillin is still the first-choice antibiotic for therapy of diseases caused by Actinomyces spp.


Introduction
Actinomyces species are Gram-positive, anaerobic bacteria that are colonizing the skin, gastrointestinal and genitourinary tract. They play an important role in the maturation of dental plaque and pathogenesis of periodontitis [1][2][3][4].
Very limited data about the antimicrobial susceptibility of Actinomyces are available. Actinomyces spp. have been described over the years as susceptible to many antibiotics. However, the most recent literature reported increased resistance patterns in Actinomyces spp. and genetically closely related species of the order Actinomycetes, such as Streptomyces, generally being multi-drug-resistant, and newly described resistance mechanisms, such as rifampin-inactivating mechanisms, also have been found [27,28]. Other studies have determined that isolates of A. europaeus and A. urogenitalis showed resistance to piperacillintazobactam, cefriaxone, linezolid and clindamycin, respectively [29][30][31].
Due to their fastidious growth, routine laboratory identification and antimicrobial susceptibility testing can be challenging and thus not usually performed. Currently, there is no standardization for susceptibility testing of Actinomyces spp. and no recommended method. The agar dilution method represents the gold standard for susceptibility testing of anaerobes, although it is time-consuming and laborious. It also requires trained personnel, regular quality control and special equipment that not every routine laboratory provides.
Without periodic susceptibility testing one may oversee a shift in their resistance patterns. Moreover, the interpretation of these data may be hampered by difficulties in the accurate identification of Actinomyces spp. and sometimes the lack of use of standardized susceptibility testing methods.
As antibiotic resistance within the Actinomycetes is emerging, the aim of the study was to analyze the antimicrobial susceptibility of well-identified clinical oral isolates of Actinomyces spp. using agar dilution and the E-Test method.

Bacterial Isolates
A total of 100 clinical oral isolates of Actinomyces spp. were used in this study. They originated from periodontal pockets, supra and subgingival plaque of 20 patients with chronic periodontitis and 15 healthy subjects, as described previously [1]. The initial selection of the clinical strains was determined based on colony and Gram-stain morphology, pigmentation and biochemical methods such as the CAMP test and catalase test.
The species included in this study reflect the diversity of the Actinomyces genus in the oral cavity. Since not all oral Actinomyces species are present in the database of commercial identification kits (Rapid ID 32 A, API Coryne, VITEK 2, ANC ID Card, bioMerieux (Marcy-l'Étoile, France), and VITEK-MS, bioMerieux), the identification based on existing data is challenging. The Identification was performed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI TOF MS) (28 strains) using a new in-house developed identification software under the protocol of T. Borgmann, or by 16s ribosomal RNA (72 strains) [32]. The strain sequences were compared with sequences deposited in the Human Oral Microbiome Database (HOMD) using the program HOMD 16S rRNA Sequence Identification.

Antibiotics
In our study, we tested the minimum inhibitory concentration (MIC) of penicillin (PEN), ampicillin-sulbactam (SAM), meropenem (MRP), clindamycin (CD), daptomycin (DAP), moxifloxacin (MXF), linezolid (LNZ) and tigecycline (TGC). Penicillin is the antibiotic of first choice for clinical infections with Actinomyces spp. Ampicillin/sulbactam combines a β-lactam antibiotic with a β-lactamase inhibitor used against penicillin-resistant bacteria producing β-lactamase. Clindamycin is used against infections with Gram-positive bacteria and is effective against anaerobes. Furthermore, it is an antibiotic of first choice to treat patients with a penicillin allergy. Clindamycin and moxifloxacin are often prescribed in dentistry, and we wanted to determine whether Actinomyces spp. developed resistance. Meropenem is a broad-spectrum antibiotic used against mixed infections that can usually be found in the oral cavity. Daptomycin is a highly effective reserve antibiotic against skin and soft tissue infections caused by Gram-positive bacteria. Tigecycline is a reserve antibiotic with a broad spectrum, including Gram-positive, Gram-negative and multidrug-resistant bacteria in severe infections. Linezolid is also a reserve antibiotic and is used against Gram-positive bacteria. Due to the lack of susceptibility data for Actinomyces spp., we chose to test linezolid in this study.

Inoculum Preparation
The MICs were obtained using the agar dilution method and E-Test methodology. All tests were performed using Brucella blood agar, containing 5% defibrinated sheep blood, which was incubated anaerobically for 48-72 h. Both methods were based on the European Committee for Antimicrobial Susceptibility Testing (EUCAST) standard procedure, which is as follows: The in-house-produced plates had an agar depth of 4 ± 0.5 mm. The plates were used within 3 days for the E-Test method and 8 h for the agar dilution method. We used a sterile cotton swab to pick colonies from an anaerobic overnight culture, then suspended in saline until an even turbidity of 1 McFarland was reached. After inoculation of the agar plates, we incubated them within 15 min in an anerobic chamber. Quality control (QC) was also conducted according to EUCAST recommendations. The cut-off values used for MIC testing were as follows: Benzylpenicillin sensitive (S) ≤ 0.25 mg/L, resistant (R) > 0.5 mg/L; ampicillin/Sulbactam S ≤ 4 mg/L, R > 8; meropenem S ≤ 2 mg/L, R > 8 mg/L; clindamycin S ≤ 4 mg/L, R > 4 mg/L; moxifloxacin PK-PD S ≤ 0.25 mg/L, R > 0.25 mg/L; tigecycline PK-PD S ≤ 0.5 mg/L, R > 0.5 mg/L; and linezolid PK-PD S ≤ 2 mg/L, R > 2 mg/L. For daptomycin there were no clinical breakpoints nor PK-PD breakpoints available at this time, so we postulated R ≥ 4 mg/L for this study.

E-Test (ET) Method
Freshly prepared bacterial suspensions (1 McFarland) as described above were uniformly spread on the surface of Brucella blood agar plates (150 mm in diameter) using a sterile cotton swab. E-Test strip application was carried out in accordance with the manufacturer's guidelines (BioMérieux, Lyon, France). After anaerobic incubation, the MICs were read and visually compared, following the EUCAST guidelines for anaerobic bacteria [33].
A semiautomatic replicator device (A400 Multipoint inoculator, Bachofer GmbH, Germany) was used to inoculate the prepared bacterial suspensions onto the freshly manufactured agar plates. Once inoculated, the agar dilution plates had a final approximate inoculum size of 10 5 CFU/spot.
All the plates were placed into an anaerobic chamber within 15 min after inoculation and then incubated for 48-72 h under anaerobic conditions (Whitley MG 1000, anaerobic workstation, Meintrup Laborgeraete GmbH, Laehden-Holte, Germany). The results were read and visually compared following the European Committee for Antimicrobial Susceptibility Testing (EUCAST) guidelines for anaerobic bacteria [19].

Statistical Analysis
The MIC-Range, MIC 50, MIC 90 , geometric means and percentage of resistance in accordance with the EUCAST breakpoints were determined and compared. In the absence of EUCAST breakpoints for moxifloxacin, tigecycline, linezolid and daptomycin, we used the PK-PD breakpoints to interpret the results. For daptomycin we used an MIC of 4 mg/L as the cut-off value for this study. The essential agreement (EA) between the E-Test and agar dilution methods was determined using the percentage of isolates where the ET MIC was within ±1.0 log 2 dilution of the reference AD MIC value for each agent tested. The categorical agreement (CA) was also calculated, being defined as the percentage of isolates tested that yielded the same categorical interpretation as the reference method. CA discrepancies are subdivided as follows: very major error (VME), major error (ME) and minor error (mE). An adequate susceptibility testing system is expected to have an EA and CA ≥ 90%, and the acceptable error rates are VME ≤ 1.5%, ME < 3% and mE ≤ 10% [21].The descriptive statistics were performed using SPSS (IBM ® SPSS ® Statistics 26.0).

Results
Almost all isolates were susceptible to benzylpenicillin, ampicillin-sulbactam, meropenem, clindamycin, tigecycline and linezolid. In contrast, most of the Actinomyces isolates were resistant to moxifloxacin and showed high MIC values for daptomycin. The results for both methods varied little, showing higher resistance using the E-Test method. Number of isolates, categorical agreement (CA), very major error (VME), major error (ME) and minor error (mE) of the E-Test (ET) MICs compared to the agar dilution (AD) MICs of the selected antimicrobials for 100 isolates of Actinomyces spp. using EUCAST breakpoints or ECOFFS (v10.2) are shown in Table 1. Applying the criteria mentioned above, the E-Test method was reliable for correctly categorizing susceptibility, except for daptomycin, where there are no actual categorizing standards. Moxifloxacin yielded an unacceptably high percentage of errors of categorization. Therefore, the E-Test method did not meet the criteria for a reasonable susceptibility testing system for this antibiotic. n.a n.a n.a n.a Tigecycline n.a n.a n.a n.a Tigecycline PK-PD 99 n.a 0 n.a Values of MIC 50 , MIC 90 , the MIC range and the geometric means for 100 isolates of Actinomyces tested by the E-Test (ET) and agar dilution (AD) methods, as well as the calculated essential agreement between both methods, are shown in Table 2. Benzylpenicillin and meropenem yielded the highest EA with 95% and 92%, respectively. The EA of daptomycin was 81%. We observed major discrepancies in the EA for linezolid (56%), ampicillin-sulbactam, clindamycin (52% each) and moxifloxacin (50%). For linezolid and moxifloxacin, the ET MIC values were two 2-fold dilutions higher than the MIC values obtained by AD. For ampicillin-sulbactam, meropenem, clindamycin and benzylpenicillin the ET MIC values were predominantly a 2-fold dilution lower than those obtained by AD.  Table 3 shows the MIC ranges of representatives of eleven Actinomyces species tested in our study. For linezolid and moxifloxacin, the ET MIC values were two 2-fold dilutions higher than the MIC values obtained by AD. For ampicillin-sulbactam, meropenem, clindamycin and benzylpenicillin the ET MIC values were predominantly a 2-fold dilution lower than those obtained by AD. The poor growth of some isolates sometimes hampered the interpretation of the E-Test MIC. A single isolate of A. odontolyticus showed multi-drug resistance to benzylpenicillin, meropenem, moxifloxacin and daptomycin when using the E-Test method. This resistance pattern was confirmed by the agar dilution method for daptomycin, meropenem and penicillin, but not for moxifloxacin.   Table 4 shows the agar dilution and E-Test MIC range for the eight antimicrobial agents tested against the three most common Actinomyces species in our study. A. oris and A. naeslundii have similar MICs whereas the MICs for isolates of A. odontolyticus are 1 to 2 log2 dilutions higher for meropenem, moxifloxacin, benzylpenicillin, ampicillinsulbactam, linezolid and tigecycline.

Discussion
In our study, we tested 100 clinical isolates of oral Actinomyces spp. against eight antibiotics by the agar dilution and E-Test methods.
Oral Actinomyces susceptibility testing is not usually performed, as this genus is known for its susceptibility to many antibiotics and for a lack of standardized testing methods. While various methods can be applied, the recommended methods for antimicrobial susceptibility testing of anaerobes are the disc diffusion method, broth microdilution method, agar dilution method and gradient tests, such as the E-Test method. The disc diffusion method is a convenient and easily performable technique but recommended for fast-growing anaerobes like Bacteroides fragiles strains. The broth microdilution method is a highly accurate method for testing several antibiotics simultaneously. The disadvantage is the inconsistent growth of anaerobe species, except Bacteroides spp. The agar dilution method, currently representing the gold standard for anaerobic bacteria, is accurate, and a high number of isolates can be tested simultaneously. It is a time-consuming and laborintensive method recommended for specially equipped laboratories. The E-Test method is fast, cost-effective and easy to perform in routine laboratories and especially used for testing a small number of isolates or multiple antibiotics at once. The latter two methods were selected to ensure reliable results.
Numerous studies signaled an increasing resistance of anaerobes in the last years. Thus, assuming that the usually prescribed antibiotics are still highly effective is no longer appropriate. The resistance rates of Actinomyces spp. obtained with the agar dilution method are similar to those from other recently conducted surveys, regarding penicillin, carbapenems, clindamycin, linezolid, daptomycin and moxifloxacin [21][22][23][24][25][26]. We detected one isolate of A. odontolyticus that showed multi-drug resistance (MDR) to benzylpenicillin, meropenem, moxifloxacin and daptomycin. Another isolate of A. gerencseriae was resistant to clindamycin and ampicillin-sulbactam. Though MDR is uncommon among the Actinomyces, this is an increasing problem for other anaerobic bacteria. The literature refers to Bacteroides fragilis isolates with resistance to penicillin, piperacillin-tazobactam, meropenem, clindamycin and metronidazole [35][36][37][38]. Clostridioides difficile showed MDR for moxifloxacin, clindamycin, erythromycin and rifampicin. Furthermore, MDR was observed for Prevotella spp., Finegoldia magna, Veillonella spp. and Cutibacterium acnes [39]. The Actinomycetes order includes beside the genus Actinomyces other important genera, Nocardia and Streptomyces. These genera can develop a remarkable survival ability and can cause life-threatening infections. Reliable identification and susceptibility testing can contribute to reduced mortality [40,41]. Because of the high prevalence of Actinomyces spp. in polymicrobial infections, the emerging resistance patterns of anaerobe bacteria should be observed. Routine susceptibility testing is recommended and using a reliable and convenient method for testing a limited number of isolates is necessary for routine laboratories. However, only a few studies have focused on susceptibility testing of oral Actinomyces spp. Our study underlines the importance of performing susceptibility testing even though there are no guidelines for Actinomyces organisms. To interpret the results, we used for both methods the EUCAST breakpoints for Gram-positive anaerobes, when available, even though it does not specify any method to be used with them. In absence of them, we used the PK-PD (non-species related) breakpoints.
The fastidious character of Actinomyces spp. leads to difficulties in cultivation and identification. Consequently, this can explain some discrepancies between the MIC values obtained using the E-Test and agar dilution methods. The poor and slow growth of some isolates causing diffuse edges of their inhibition zones or the growing of only few colonies makes reading the MICs more difficult and may result in lower MIC values, such as those for ampicillin-sulbactam, meropenem, clindamycin and benzylpenicillin.
Our in vitro data suggest that oral Actinomyces spp. are susceptible to benzylpenicillin, ampicillin-sulbactam, meropenem, clindamycin, linezolid and tigecycline. Our findings are in line with other recently conducted surveys covering oral Actinomyces species [30,31,[42][43][44][45]. In general, there were no differences regarding susceptibility within the different Actinomyces species.
Essentially, the high effectiveness of β-lactam antibiotics, linezolid and tigecycline correspond to other studies [31,46,47], although some studies have found higher resistance rates, from 16-34%, for clindamycin for the Actinomyces spp. tested [48]. The resistance against moxifloxacin was confirmed by other studies [47,49]. Because there are no guidelines or PK-PD breakpoints for daptomycin, we cannot prove the efficacy of this antibiotic. It is noticeable that the MICs for daptomycin where very high in comparison with the other antibiotics used in this study. Our results are similar to those of another study: MIC 50 8 mg/L and MIC 90 32 mg/L [50]. Therefore, it may be assumed that Actinomyces spp. are resistant to daptomycin.
In this study, we showed that the E-Test method could be a promising susceptibility testing alternative to the agar dilution method, especially for ß-lactams for routine laboratory testing.

Conclusions
Although Actinomyces are susceptible to many antibiotics, such as β-lactams, meropenem and reserve antibiotics such as clindamycin linezolid and tigecycline, the empirical use of some antimicrobial agents such as moxifloxacin or daptomycin, especially in polymicrobial infections, could lead to a failed treatment. The constant threat of increased resistance and MDR in anaerobic bacteria needs to be monitored. These data underline the importance of routine susceptibility testing in laboratories using reliable methods, such as the E-Test method, to provide the relevant information to clinicians and to prevent treatment failure.