Genetic diversity and antimicrobial susceptibility of Nocardia species among patients with nocardiosis

The aim of this multicenter study was to determine the genetic diversity and antibiotic susceptibility of clinically isolated Nocardia species. One hundred twenty-seven patients with nocardiosis were randomly selected from 5 provinces of Iran. Molecular diagnosis of Nocardia species was performed using multilocus sequence analysis of gyrase B of the β subunit of DNA topoisomerase (gyrB), and 16S rRNA and subunit A of SecA preproteintranslocase (secA1). Antimicrobial susceptibility testing was performed following the Clinical and Laboratory Standards Institute recommendations. Thirty-five N. cyriacigeorgica, 30 N. asteroides, 26 N. farcinica, 12 N. otitidiscaviarum, and 10 N. abscessus cultures were studied. All isolates were susceptible to linezolid. All isolates of N. cyriacigeorgica, N. asteroides, N. abscessus, and N. otitidiscaviarum were susceptible to trimethoprim-sulfamethoxazole, while 8% of N. farcinica isolates were resistant to this drug. All N. otitidiscaviarum isolates were highly resistant to imipenem, but N. cyriacigeorgica, N. asteroides, N. farcinica, and N. abscessus were only moderate resistant. The susceptibility patterns vary with different species of Nocardia. Resistance to trimethoprim-sulfamethoxazole in Iran is low and this drug should be first line therapy, unless drug susceptibility testing shows resistance. Linezolid also covers Nocardia well and could be a second line agent.

). This study approved by Ethics Committee of Ahvaz Jundishapur University of Medical Sciences (AJUMS), Ahvaz, Iran. Demographic, clinical, and microbiologic data were collected from patients' medical records who signed the informed written consent. Isolates were sent to the Infectious and Tropical Diseases Research Center (AJUMS) for identification and subsequently antimicrobial susceptibility determination. All experimental protocols including sample collection and laboratory methods were approved by scientific committee of Health Research Institute (AJUMS).
A portion of the isolates (32specimens)were identified at the species level by multilocus sequence analysis (MLSA) of 16S rRNA, gyrase B of the ß subunit of DNA topoisomerase (gyrB) and subunit A of SecA preprotein translocase (secA1) as previously described by McTaggart and colleagues 8 to find out the reliability of each marker for identification. The remaining isolates were identified to species level by 16S rRNA analysis because of its acceptable discriminatory power. The 16S rRNA gene was amplified using 27F primer (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1525R (5′-AAGGAGGTGWTCCARCC-3′) and then was sequenced. The sequences were aligned and trimmed in BioNumerics (version 6.0.1) software (Applied Maths, Austin, TX) and were identified to species level. A representative 16S rRNA gene sequence from each of species was deposited in Genbank with KT003507-KT003513 accession numbers.
Broth microdilution testing. The drugs amikacin, amoxicillin-clavulanate, cefepime, cefotaxime, ceftriaxone, ciprofloxacin, clarithromycin, doxycycline, gentamicin, imipenem, linezolid, minocycline, moxifloxacin, tobramycin, trimethoprim-sulfamethoxazole (TMP-SMZ), and vancomycin were selected by testing based on National Committee for Clinical Laboratory Standards (NCCLS) recommendations 9 . Microtiter plates were prepared in-house, using standard twofold dilution of all antimicrobials except ampicillin and amoxicillin-clavulanate in cation-adjusted Mueller-Hinton broth. The plates were stored at − 70 °C and were thawed at room temperature immediately before use. The appropriate dilution of amoxicillin-clavulanate was freshly prepared immediately before use, then aliquoted, and placed in designated microtiter wells. Ten microliters of an inoculum with a turbidity equivalent to that of a 0.5 to 1.0 McFarland standard was dispensed into each well to give a final concentration of 10 4 to 10 5 CFU/mL 9 . The microtiter plates were incubated aerobically at 35 °C and were read after 3 days. Growth was examined daily by visual inspection. The minimum inhibitory concentration (MIC) was defined as the lowest concentration of the drug that inhibited visible growth. MICs at which 50% (MIC50s) and 90% (MIC90s) of isolates are inhibited were determined 9 . MIC50% and MIC90% were selected to provide an interpretation of the clinical significance of concentrations of an antimicrobial that inhibit the growth of an organism or kill it in laboratory systems (in vitro) 10,11 and for defining the starting point for larger preclinical evaluations of novel antimicrobial agents 12 . For TMP-SMZ, the MIC was the 80% inhibition endpoint of growth compared to the control. Susceptible and resistant breakpoints were defined according to the NCCLS recommendations 9 . Quality control of the MICs was performed by the testing of NCCLS recommended reference strains, including Enterococcus faecalis ATCC 29212, Nocardia abscessus DSM 44432, Pseudomonas aeruginosa ATCC 27853, and Staphylococcus aureus ATCC 29213.The control strains were obtained from Iranian Biological Resource Center (IBRC), Tehran, Iran.
N. asteroides. Among the 31 isolates of N. asteroides, all were susceptible to TMP-SMZ and linezolid. Amoxicillin-clavulanic acid, cefepime, ceftriaxone, ciprofloxacin, imipenem, moxifloxacin, and tobramycin had moderate activity, while clarithromycin had poor activity against the clinical isolates of N. asteroides. The MIC90 for both linezolid and TMP-SMZ was 1 (μ g/ml), but, for ceftriaxone this value was 128 (μ g/ml).
N. otitidiscaviarum. All 12 isolates of N. otitidiscaviarum were susceptible to amikacin, linezolid, tobramycin, and TMP-SMZ, whereas there was poor activity to ceftriaxone, doxycycline, and minocycline, and all isolates were resistant to amoxicillin-clavulanic acid, imipenem. The MIC90 for amikacin, linezolid and TMP-SMZ was 0.125 μ g/ml.
One isolate from each species of N. arthritidis, N. kruczakiae, N. nova and N. veteran were studied for drug susceptibility tests. All of them were susceptible to amikacin, amoxicillin-clavulanic acid, ceftriaxone, ciprofloxacin, clarithromycin, imipenem, linezolid, tobramycin and TMP-SMZ.

Discussion
N. asteroides was the most frequently recovered species in our study. It was followed by N. farcinica and N. cyriacigeorgica. This pattern was different between individuals with pulmonary and extrapulmonary nocardiosis, with N. cyriacigeorgica being the most common in extrapulmonary disease. Our study found that extrapulmonary nocardiosis occurs more commonly in younger persons (mean age 38) compared to pulmonary nocardiosis (mean age 56) and in those with rheumatologic disorders taking corticosteroids.
Although cases reports have shown N. cyriacigeorgica 14 , N. asteroides complex 15 and N. nova complex 16 in Iran, but to our knowledge, there is no report of drug susceptibility on clinical isolates of Nocardia from Iran as well as the Middle Eastern countries.
The lungs are the most common organ that Nocardia infects (up to 70%),with N. asteroides complex the species most often isolated from this site 17 .Yamagata and colleagues reported that patients with rheumatologic disorders who took corticosteroid were at higher risk of extrapulmonary nocardiosis 6 . Our study confirmed the higher incidence extrapulmonary nocardiosis in those taking corticosteroids before the Nocardia infection. This information may serve as a warning to clinicians about the risk of corticosteroids and disseminated nocardial infection.
Nocardia species cause a wide variety of diseases and have variable drug susceptibility profiles. Since the 1940s, the sulfonamides have been the treatment of choice for nocardiosis 1,18 . Later, the combination of trimethoprim with sulfamethoxazole became the most commonly recommended treatment for these infections 1 . Other therapies including amikacin, a combination of amikacin and a beta-lactam such as ceftriaxone or imipenem, and a combination with linezolid have also been suggested for therapy of patients with serious disease 1,19 .
Susceptibility testing of Nocardia isolates to the antibiotics showed that N. cyriacigeorgica isolates were generally sensitive to our selected antibiotics. All were susceptible to amikacin, cefepime, gentamicin, linezolid, tobramycin, and TMP-SMZ; and the majority was somewhat less susceptible to cefotaxime, ceftriaxone, doxycycline, imipenem, and minocycline. These findings are consistent with those reported by Glupczynski and colleagues 20 . Further, Larruskain and colleagues noted that N. cyriacigeorgica isolates from Spain were susceptible to amikacin, gentamicin, linezolid, tobramycin, and TMP-SMZ 21 .
Schlaberg and colleagues from the United States reported that N.cyriacigeorgica isolates were susceptible to amikacin, linezolid, tobramycin, and TMP-SMZ, and were resistant to amoxicillin-clavulanic acid, ciprofloxacin, clarithromycin, minocycline, and moxifloxacin 22 . However, our isolates were highly resistant only to amoxicillin-clavulanic acid, ciprofloxacin, and moxifloxacin. Ceftriaxone, imipenem, linezolid, and TMP-SMZ were reported as the most effective antimicrobial agents against N.cyriacigeorgica isolates in Taiwan 23 , which agrees with our results.
Among the 31 isolates of N. asteroides, linezolid and TMP-SMZ were active against all isolates while moderate susceptibility was detected for imipenem, amoxicillin-clavulanic acid, cefepime, ceftriaxone, ciprofloxacin, moxifloxacin, and tobramycin. Clarithromycin had poor activity against clinical isolates of N. asteroides in our study. In the preliminary evaluation of antimicrobial agents against N. asteroides isolates in 1984, the beta-lactams including third-generation cephalosporins were generally reported ineffective, whereas minocycline, doxycycline, and sulfamethoxazole were recommended for therapy 24 . Four years later, Wallace and colleagues showed that the most active parenteral agents against N. asteroides were amikacin, cefotaxime, ceftriaxone, imipenem, minocycline, and sulfonamides 25 . Although Schlaberg and colleagues found that all N. asteroides isolates were susceptible to amikacin, imipenem, linezolid, tobramycin, and TMP-SMZ 22 , we found less susceptibility among N. asteroides isolates in our study.
N. farcinica is more likely to have multidrug resistance and high level resistance to imipenem, ceftriaxone, clarithromycin, tobramycin, and moxifloxacin 21,22 . Although TMP-SMZ has been the drug of choice for the treatment of nocardiosis 7,18,24,25 , we found 8% (2 isolates) of N. farcinica were TMP-SMZ resistant. Larruskain and colleagues in Spain found 16.1% 21 , Uhde and colleagues found 42% 26 from the United States, and Tremblay and colleagues also reported 42% TMP-SMZ resistant strains from Canada 27 . Another study from Spain, reported that 9 of 19N. farcinica isolates (47%) were TMP-SMZ resistant 7 . Furthermore, Lai and colleagues from Taiwan reported a low incidence (9%) similar to ours 23 , and another report from the United States also found only 2% TMP-SMZ resistance 28 and sulfonamide and TMP-SMZ resistance was not seen in South Africa 29 . The similarity between the 2 North American countries and divergence in Europe and Iran suggests there may be geographical differences in N. farcinica drug sensitivity with unknown reasons. We speculate that the difference in drug susceptibility to TMP-SMZ could be related to differences in laboratory methodology and interpretation criteria. More recently, Valdezate and colleagues reported association of high-level sulfonamide resistance and the presence of plasmid-borne integrons carrying sul genes (sul1 and sul2) in SXT-resistant Nocardia strains 30 . These type of integrons, and the corresponding plasmids, are commonly detected in bacteria living in different ecological niches 31 .
In our study, resistance to β -Lactams antibiotics were detected among the isolates, which might be related to a mutational change affecting the inhibitor and active site (s) in the beta-lactamase 32 .
Our Nocardia isolates showed moderate resistance to quinolones. Valdezate and colleagues 30 could not find plasmid-mediated quinolone resistance genes (qnrA, qnrB, qnrC, and qnr) or the gene for the aminoglycoside acetyltransferase for modify ciprofloxacin 33 or efflux pump qepA 34 and or nucleotide changes observed in gyrA 35 . Further, study considering the resistance mechanisms and how antibiotic resistance spreading among Nocardia strains are required.
All N. otitidiscaviarum species were susceptible to TMP-SMZ. Our data was in agreement with those reported by others 22,23 . In contrast with our data, moderate resistance to TMP-SMZ (32%) among N. otitidiscaviarum was reported by Uhde and colleagues 26 .
N. abscessus were susceptible to ceftriaxone, gentamicin, linezolid, tobramycin, and TMP-SMZ in our study. The same susceptibility profile of N. abscessus was reported before 21,22 .
Linezolid, a relatively new class of antibiotics, showed extraordinary in vitro activity against all of the major clinically significant species of Nocardia 19,36 . Our findings are in agreement with reports from different parts of world that clearly demonstrate that linezolid is an effective alternative for the treatment of nocardiosis.
In conclusion, N. asteroides was the most common species isolated from pulmonary nocardiosis and N. cyriacigeorgica was the most frequently recovered species from extrapulmonary nocardial infections. Clinical isolates of Nocardia species in our study had varied drug susceptibility patterns, which were similar to what have been reported from other geographic area, with some exceptions. Importantly, TMP-SMZ resistance was low in the current study. Based on this information, we feel confident recommending TMP-SMZ as the first choice for the treatment of nocardiosis in Iran. Linezolid broadly covers Nocardia and would be a second choice, although the costs are considerably greater. We strongly recommend that drug sensitivity testing is helpful in all patients with serious disease.