INTRUDUCTION: The genus Mycobacterium includes more than 190 species that differ in terms of metabolism, growth rate, epidemiology, pathogenicity, geographical distribution and antimicrobial susceptibility. In addition to species that cause tuberculosis (M. tuberculosis, M. caprae, M. bovis, M. africanum), there are also species that act as opportunistic pathogens capable of causing lymphadenitis and infections of the lungs, skin, soft tissues, joints, tendons, bones and other organs in both animals and humans, and are common called non-tuberculous mycobacteria (NTM). NTM species, also called ecological mycobacteria, have been isolated from water, soil, dust and plants. According to cultural characteristics, they are divided toward the growth rate on the nutrient medium into fastgrowing species in which colonies are visible within seven days (e.g. M. fortuitum, M. chelonae, M. abscessus) and slow-growing species (e.g. M. kansasii, M. marinum, M. avium complex (MAC), M. gordonae) that require a longer incubation time. Growth rate affects sensitivity to antimicrobial drugs as well as clinical signs and pathological changes of the affected organism. The identification of NTM disease is based on bacteriological and biochemical examination and lasts for several weeks. In the end, it often happens that it is not possible to determine the species with certainty. By additional application of commercially available molecular typing kits (Hain, Germany) it is possible to identify about 40 species. Furthermore, NTM are characterized by a very low evolutionary divergence of the genome, and therefore for identification of a particular species it is recommended to use only individual most conserved regions and not the entire genome. One of the characteristics of NTM species is a high level of natural drug resistance, as well as inducible and mutational resistance acquired during suboptimal exposure and drug selection. For this reason, their treatment is much more challenging than treatment of classical tuberculosis, and so far no standardized antimicrobial therapy for NTM has been established. Different types of NTM have different profiles of antimicrobial susceptibility, but there are few publication that indicates those differences. There is currently only one standard for determining antimicrobial resistance for mycobacteria (M24-A2) prescribed by the Clinical and Laboratory Standards Institute. In the territory of the Republic of Croatia, no research has been conducted so far in the form of antimicrobial susceptibility of non-tuberculous mycobacteria isolates in the field of veterinary medicine, and according to our knowledge and available data from the literature, such research has not been conducted in the world. This research is foccused on the determination of different types of NTM present in domestic and wild animals in the Republic of Croatia using molecular methods, as well as determination of antimicrobial resistance of NTM from different sources and geographical units using microdilution method. The results of the research will contribute to the routine identification of pathogens in both animals and humans, and a validated and standardized method will be acceptable and beneficial to health system in Croatia. MATERIAL AND METHODS: In the period from 2012 to 2016, on the presence of Mycobacterium sp. we bacteriologically examined a total of 593 samples of various organs and tissues originating from 153 animals from 14 counties and the City of Zagreb. The research also included a examination of 71 archives isolates of non-tuberculous mycobacteria originating from 64 animals from nine counties and the City of Zagreb. Archival isolates that were identified only up to the level of the genus Mycobacterium by gene multiplication. which encodes a 65 kDa antigen, were stored at a temperature of - 80 ˚C until the beginning of the research. The study covered samples originating from cows, pigs, chickens, ducks, sheep, goats, wild boars and wild deer. Bacteriological examination began with homogenization and decontamination of the material, followed by inoculation on three different nutrient media: Löwenstein-Jensen media supplemented with pyruvate or glycerol and Stonebrink media. The inoculated material were then incubated at 37°C. After 4-7 days of incubation, we controlled the growth of mycobacteria for the first time, a further at one-week intervals. The incubation lasted eight weeks, and if there was no colony growth on the nutrient medium, the incubation was extended to 12 weeks. In case of growth of the colony was transferred onto new nutrient media and stained according to Ziehl - Neelsen in order to determine the presence of acidresistant rods. A total of 155 isolates originating from domestic and wild animals were used for molecular identification. In all isolates, the amplification of 16S rRNA gen region and gen encoding 65 kDa antigen present in all mycobacteria was performed, which is reliable evidence that isolates belong to Mycobacterium sp. Further identification of mycobacterial isolates to the species level was performed by specific hybridization method using "Geno Type® Mycobacterium CM and AS kit", which allows the identification of about 40 species within the genus Mycobacterium. All isolates to which affiliation was confirmed as the M. avium complex, were underwent amplification of the integrated insertion sequence IS901 for determination of subspecies. Isolates in which the affiliation of the species could not be determined by the above mentioned methods, were subjected to sequencing of parts of the 16S rRNA, rpoB, hsp65 genes, and the ITS region. By comparing the resulting sequences with the data in the available databases we identified the species. Antimicrobial susceptibility testing of slowly and rapidly-growing isolates of mycobacterial species was performed by broth microdilution method using Thermo Scientific ™ Sensititre ™ Myco SLOMYCO and RAPMYCO AST Plate commercial kit, using drug concentrations which are obtained from serial, double dilutions. When reading the results, we reported the values of the lowest concentration which interrupts growth (minimum inhibitory concentration, MIC) and interpretation of results as sensitive (S), intermediate (I) and resistant (R). During the test procedure and interpretations of the results we followed the recommendations of the standard published by Clinical and Laboratory Standards Institute – CLSI. The testing was performed on a total of 138 isolates, of which 47 isolates are members of slowly-growing species and 91 isolates are members of rapidlygrowing species. The results were processed by the statistical program Stata 13.1. Descriptive data were presented as values of total number and percentage. Observed differences between the groups we tested by Hi-square test and the Fisher exact test, and the significance of the differences we presented with a P-value where we considered a P-value less than 0.05 statistically significant. The geographical distribution of the obtained antimicrobial susceptibility results was presented by a program QGIS 3.10. RESULTS: Members of the genus Mycobacterium were isolated from a total of 80 animals (52%), namely 29 (36,2%) domestic and 51 (63,7%) wild animals, and we isolated a total of 84 isolates. As earlier mentioned, the study also included 71 archival isolates of nontuberculous mycobacteria originating from 22 (31%) domestic and 49 (69%) wild animals. Out of a total of 155 isolates, 106 (68,4%) were described as members of rapidly growing and 49 (31,6%) as members of slowly growing mycobacteria. Affiliation of those 155 isolates to the genus Mycobacterium was proved by polymerase chain reaction by amplification of the 16S rRNA gene region and the gene encoding the 65 kD antigen present in all mycobacteria. Demonstration of the species of these isolates was continued by the method of specific hybridization using tests "Geno Type® Mycobacterium CM and AS" which successfully typed 61 (39,4%) isolates to the species level. Using this method we proved the following species: M. avium complex (11,6%), M. fortuitum (17,4%), M. gordonae (1,9%), M. celatum (5,2%), M. kansasii (1,9%) and M. intermedium (1,3%). A total of 18 isolates to which affiliation has been confirmed as a M. avium complex (MAC) were subjected to amplification of the integrated insertion sequence IS901 to determine the subspecies. Belonging to the subspecies M. avium subsp. hominissuis were found in 10 isolates and M. avium subsp. avium in eight isolates. In the remaining 94 isolates, we amplified parts of the DNA sequences of the most conserved regions, 16S rRNA, rpoB, hsp65 genes, and intergenic ITS region. By comparing the results obtained by the analysis of these four genes, the affiliation of the species was successfully determined in 72,3% (68/94) of isolates. A total of 14 different species were descibed by sequencing: M. agri, M. arupense, M. chitae, M. elefantis, M. flavescens, M. kumamotonense, M. neoaurum, M. nonchromogenicum, M. peregrinum, M. phocaicum, M. porcinum, M. pulveris, M. triviale and M. vaccae. We described nine isolates as "Most similar to Mycobacterium + species", while for the remaining 17 isolates we were unable to determine the species. Obtained percentage of successful identification up to species level using the 16S rRNA gen was 42.3% (45/94 isolates), hsp65 gen 19.7% (21/94 isolates), rpoB gen 18.9% (20/94 isolates) and ITS region 2.8% (3/94 isolates). Using these methods, we have described a total of 20 different species of non-tuberculous mycobacteria (NTM) and six "Most similar" species. Antimicrobial susceptibility testing was performed by the broth microdilution technique using the VersaTREK kit for rapidly and slowly growing mycobacteria. The study included a total of 47 isolates belonging to nine different slowly growing species and one "Most similar", namely M. avium spp., M. celatum, M. flavescens and "Most similar M. flavescens", M. gordonae, M. intermedium, M. kansasii, M. kumamotonense, M. nonchromogenicum and M. triviale. The tested strains showed the highest percentage of resistance to fluoroquinolones, namely moxifloxacin (77,3%) and ciprofloxacin (65,4%), followed by doxycycline (77%), trimethoprim - sulfamethoxazole (65,4%), linezolid (61,4%) and amikacin (6,8%). A high percentage of resistant strains was also described for rifamycin members rifampin (76,9%), while for rifabutin the percentage was significantly lower (7,7%). The susceptibility of all isolates has only been described to the macrolide clarithromycin. Also, we examined a total of 91 isolates belonging to 11 different rapidly growing species and four "Most similar", namely M. agri, M. arupense, M. chitae and "Most similar M. chitae", M. elephantis, M. fortuitum, M. neoaurum and "Most similar M. neoaurum", M. peregrinum, M. phocaicum, M. porcinum, M. pulveris, M. vaccae and "Most similar to M. vaccae", and "Most similar to M. septicum". The tested strains showed the highest percentage of resistance to the cephalosporin group, namely to cefepime (57,5%) and cefriaxone (47,2%), while to cefoxitin this percentage was significantly lower (3,5%). Resistance was followed by amoxicillin/clavulanic acid (31%), clarithromycin (23%), imipenem (3,5%), and linezolid (1,2%) and trimethoprim/sulfamethoxazole (1,2%). Within the aminoglycoside group, the percentage of tobramycin resistant strains was 14,9%, while all isolates were sensitive to amikacin. Among tetracyclines, 10,4% of resistant isolates were described for doxycycline and 6,9% for minocycline. Among fluoroquinolone group, 1,2% of ciprofloxacin resistant strains were described, while all isolates were susceptible to moxifloxacin. CONCLUSION: During the five-year research period, a total of 155 non-tuberculous mycobacteria isolates originating from domestic and wild animals were collected. From the used molecular methods for the identification to the species level, by the method of specific hybridization using tests "Geno Type® Mycobacterium CM and AS" only 39.4% (61/155) of isolates were described, and by sequencing of the most conserved regions, 16S rRNA, rpoB, hsp65 genes, and intergenic ITS region we described 72.3% (68/94) of isolates. From this we conclude that the method of sequencing and comparison of the most conserved regions of the genome is much more reliable for accurate identification of mycobacterial species. We have described a total of 20 different species of non-tuberculous mycobacteria (NTM) and six "Most similar" species, from which we conclude that NTMs are widespread in both domestic and wild animals, making wild animals possible reservoirs of infection for domestic animals and humans. The most common species in our study were M. fortuitum (17.4%), M. neoaurum (16.1%), M. avium ssp. (11.6%) and M. vaccae (9%), as evidenced by previous studies conducted in the Republic of Croatia describing M. avium ssp. and M. fortuitum as one of the most common species, from which we conclude and confirm the largest presence of the most pathogenic mycobacterial species, namely M. fortuitum and M. avium ssp. in both veterinary and human medicine. Comparing the susceptibility of slowly and rapidly growing NTM isolates, we observed significant differences in resistance to certain antibiotics, and resistance was more pronounced in slowly growers. Among the species of slowly growing mycobacteria, the resistance was equal, while among the rapidly growers it was more pronounced in the species M. fortuitum, M. neoaurum, M. vaccae and M. porcinum. We did not observe significant differences in the incidence of resistant strains isolated from domestic and wild animals, nor differences in different years of observation. Also, the differences in the occurrence of resistant strains originating from different counties were not statistically significant, and no differences in the representation of the described species by counties in the Republic of Croatia were observed. We also described 26 multi-resistant isolates that have shown resistance to five or more antibiotics, and in the case of MAC isolates on three of the four antibiotics tested, of which 15 are members of slow-growing species and 11 are rapidlygrowing species. From the above we conclude that multiple resistance is equally distributed between slow-growing and rapidly-growing species, and a large number of isolates is a worrying fact that indicates the need for further research in understanding the emergence of resistance in the so-called "wild isolates". Isolated strains from both domestic and wild animals showed high percentages of resistance to most antibiotics recommended for therapy in human medicine indicating zoonotic potential with possible challenges in the treatment of NTM infections in humans, and also indicating the possible role of animals as reservoirs of multi-resistant strains of NTM. The adopted and validated method of both identification and antimicrobial susceptibility testing of non-tuberculous mycobacteria will be useful for the health system in the Republic of Croatia, taken into consideration that these methods have not been used so far.