Resistance determinants of emerging pathogens isolated from an intensive care unit as a parameter of population health conditions of the Legal Amazon microregion

Abstract Bacteria responsible for causing infections are common in hospital environments, water, soil, and food products. The infection risk is intensified by the absence of public sanitation, poor quality of life, and food scarcity. These external factors promote the dissemination of pathogens by direct contamination or biofilm formation. In this work, we identified bacterial isolates obtained from intensive care units in the southern region of Tocantins, Brazil. We compared matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) techniques and 16S ribosomal ribonucleic acid (rRNA) molecular analysis; we also performed phenotypic characterization. Fifty-six isolates characterized using morphotinctorial tests were classified as gram-positive (80.4%; n = 45) and gram-negative (19.6%; n = 11) and were resistant to several antibiotic classes; notably, we identified the blaOXA-23 resistance gene in the ILH10 isolate. Microbial identification using MALDI-TOF MS resulted in the identification of Sphingomonas paucimobilis and Bacillus circulans. 16S rRNA sequencing revealed four isolates belonging to the genera Bacillus and Acinetobacter. The similarity was superior to 99% for Acinetobacter schindleri in the Basic Local Alignment Search Tool (BLAST), grouped in the clade superior to 90%. Several strains isolated from intensive care units (ICU) were resistant to various antibiotic classes. These techniques allowed for the identification of several microorganisms of importance in public health, enabling improvements in human infection control and proving the quality of inputs, food, and water.

identification technique for bacteria isolated from biological specimens and can detect various proteins from microbial species (Angeletti, 2017;Tsuchida et al., 2020).
The analysis of microbial diversity in different environments and the multi-drug resistant bacteria profile have essential implications for hospitals stemming from the indiscriminate use of antibiotics.The present work aimed to characterize microbial sources phenotypically in intensive care units (ICUs) and identify isolates using the MALDI-TOF MS and 16S rRNA molecular methods.These techniques are critical for public health because they provide actions to control human infections and benefit the population.

Study location; sample processing; ethical and legal permissions.
The Gurupi Regional Hospital -Tocantins -Brazil (11°43'48" S and 49°04'08" W) is located in the Legal Amazon microregion.It is a referral center for 18 municipalities: Aliança do Tocantins, Araguaçu, Alvorada, Cariri do Tocantins, Crixás do Tocantins, Dueré, Figueirópolis, Formoso do Araguaia, Jau do Tocantins, Lagoa da Confusão, Palmeiropolis, Paranã, Peixe, Sandolândia, São Valério, São Salvador, Sucupira, and Talisman.We used samples from beds of patients hospitalized for more than 48 hours.The collection was performed using sterile swabs passed on the bed surface and transported in Cary Blair medium.Samples were cultured on tryptic soy agar (Kasvi, Italy) and incubated aerobically at 37 °C overnight.Bacterial isolates were phenotypically characterized using manual Gram staining methods and measuring motility properties, growth at 44 °C, citrate utilization, oxidase and urease production, and oxidation/fermentation (Castilho et al., 2017).The Research Ethics Committee of the Federal University of Tocantins approved the work (protocol number 72773417.3.0000.5519).After issuing a substantiated opinion of this study, the protocol was forwarded to the Health Department of the State of Tocantins (SESAU/TO) in compliance with ordinance SES-391/2017.

Phenotypic analysis and resistance classification
We measured antibiotic sensitivity and resistance using the disk diffusion method, following the Clinical and Laboratory Standards Institute guidelines (CLSI, 2017).The isolates were cultured in 100 ml of TSB broth (soy peptone 17 g/l, casein 3 g/l, dextrose 5 g/l, and monobasic potassium phosphate 5 g/l), stirring at 120 rpm at 37 °C for 8 h.Growth was adjusted to 10 8 colony-forming units/ml and seeded on Mueller-Hinton agar (Kasvi, Italy) between 18 and 24 hours of incubation at 37 °C for recording halo

Introduction
Bacterial infections inside and outside hospital environments are the primary causes of illness.The excessive use of antimicrobial agents in humans and domestic animals has generated high rates of bacterial resistance and has become a severe problem worldwide (Bonelli et al., 2014;Moretto et al., 2021).Community and environmental factors contribute to the emergence of multi-drug resistant bacteria, promoted by the lack of adequate public sanitation and the transmission of domestic effluents to water sources.These phenomena contribute to the spread of pathogens, especially in regions with low human development indexes (Gallini et al., 2010;Bartley et al., 2019;Moretto et al., 2021).
The emergence of new resistance mechanisms may arise in vulnerable communities, leading to the dissemination of bacterial genes via conjugative plasmids (Meirelles-Pereira et al., 2002;Cerceo et al., 2016).Among the primary multi-drug resistant bacteria (MDRs), including species in the genera Bacillus, Pseudomonas, Acinetobacter, Klebsiella, and enterobacteria that are commonly found in hospital environments, water, soil, and food products (Logan, 2012;Smet et al., 2014;Ferreira et al., 2019;Arbab et al., 2021;Adwan and Omar, 2021).Food is a potential transmission source between the community and clinical settings (Carvalheira et al., 2021).
Another critical factor is the ability of bacteria to form a structural community of multiple bacterial cells (biofilm) characterized by high nutritional and metabolic versatility and adherence to living and non-living surfaces.These surfaces include glass, plastic, X-ray equipment, benches, beds, fans, and air circulation systems (Bernards et al., 2004;Gaddy et al., 2009).
Conventional methods for diagnosing pathogens involve isolating the organism in culture with subsequent analysis of its phenotypic characteristics (i.e., microscopic and macroscopic morphologies) and biochemical evaluations on various substrates (Stępień-Pyśniak et al., 2017).Other methods used in clinical microbiology laboratories include molecular methods with gene sequencing of 16S rRNA and 23S rRNA.MALDI-TOF MS is a rapid and accurate a 90%.Várias cepas isoladas de ICU foram resistentes a várias classes de antibióticos.Essas técnicas permitiram a identificação de diversos microrganismos de importância em saúde pública, possibilitando melhorias no controle de infecções humanas e comprovando a qualidade dos insumos, alimentos e água.

DNA extraction and sequencing of the 16S rRNA gene
Bacterial isolates were cultured in brain heart broth (Merck, Germany) and incubated overnight at 37 °C with continuous shaking at 120 rpm.DNA extraction was performed using an extraction kit (Invitrogen-PureLink® Genomic DNA Kits).PCR of the rDNA16S region amplified the DNA.The sequence of primers 16SF and 16SR amplify a region of 400 bp (16SF-ACTGCTATCCACCCTCAAAC and 16SR CTGGGTGAAGTTGTAATCGG).The PCR product was purified using exonuclease I and shrimp alkaline phosphatase.The integrity of the genomic DNA was tested by electrophoresis.
The sequencing of the 16S gene was performed on a 3500 sequencer (Thermo Fisher Scientific).Sequencing reactions were set up with 20 ng of DNA, 0.5 µl Big Dye V3.1 (Thermo Fisher Scientific), 2.0 µl of 5X sequencing buffer, and 3.2 µmol of primer.The continuous sequences were compared with reference and outgroup sequences from the National Center for Biotechnology Information using the BLAST system to identify similarities.Phylogenetic analysis of the isolates was performed using the maximum likelihood method and the Tamura-Nei model (Tamura and Nei, 1993) using a MEGA 11 (Tamura et al., 2021).Branch confidence was tested using 500 bootstrap repetitions.

Identification by MALDI-TOF MS
Isolates were seeded in Brain Heart Infusion (Merck, Germany), incubated at 37 ºC for 18 hours, and sent to Instituto Hermes Pardini (Belo Horizonte, Brazil) for microbial identification by automated mass spectrometry equipped with MALDI-TOF MS (BioMérieux, France).Briefly, a fresh colony was spread on a Vitek MS DS target slide, and the preparations were covered with 1 ml of VITEK MS-CHCA matrix solution (a saturated solution of α-cyano-4-hydroxycinnamic acid in 50% acetonitrile and trifluoroacetic acid at 2.5%).After drying, the target plate was loaded into the Vitek MS mass spectrometer and air-dried for 1 to 2 minutes at room temperature.Microbial identification was achieved by obtaining spectra using MALDI-TOF and analyzing the spectra using the VITEK MS database.The spectral peaks were compared with the characteristic pattern of a species, genus, or family of microorganisms, resulting in the identification  (Dubois et al., 2012;Moon et al., 2013;Gonçalves et al., 2019).A percentage probability (99.9%) represents the similarity of specific peaks between the generated spectrum and the database spectra.Lower values were designated as inconclusive

Bacterial identification using 16S rRNA and MALDI-TOF MS
The isolates ILH10, ILH11, ILH13, and ILH2 were identified only in the sequencing of the 16S rRNA gene and belonged to the genera Acinetobacter and Bacillus (Table 3).The ILH10 isolate was subjected to phylogenetic analysis involving 19 nucleotide sequences from the genus Acinetobacter (Figure 2).All ambiguous positions were removed for each pair of sequences (pair deletion option).There were 1593 positions in the final dataset of the ILH10 isolate that showed more significant similarity with 99.3% to Acinetobacter schindleri in BLAST, grouping 90% in the clade with bootstrap higher than 90% (Table 3; Figure 2).The ILH13 isolate was subjected to phylogenetic analysis involving 18 nucleotide sequences from the genus Bacillus (Figure 2).All ambiguous positions were removed for each pair of sequences (pair deletion option).There were 1555 positions in the final dataset of the ILH13 isolate that showed more significant similarity with 100% to Bacillus cereus in BLAST, in the clade with bootstrap higher than 90% (Table 3; Figure 2).The ILH2 isolate was subjected to phylogenetic analysis involving 22 nucleotide sequences from the genus Bacillus (Figure 2).All ambiguous positions were removed for each pair of sequences (pair deletion option).There were 1561 positions in the final dataset of the ILH2 isolate that showed more significant similarity with 99,7% to Bacillus Licheniformis in BLAST, in the clade with bootstrap higher than 80% (Table 3; Figure 2).Only  Emerging pathogens isolated from the hospital environment ILH14 and ILH 16 were identified using MALDI-TOF MS, revealing Sphingomonas paucimobilis and Bacillus circulans (Table 4; Figure 3).ILH10, ILH11, ILH13, and ILH2 were not detected using the technique because the system did not include a reference database for identification (Table 4).

Discussion
The indiscriminate use of antibiotics inevitably leads to antimicrobial resistance, primarily through natural resistance and antibiotic residues in sewage and soils (Moretto et al., 2021).The Gurupi Regional Hospital, the study site, is a referral center for 18 municipalities in the state of Tocantins, which can favor the dissemination of MDRs via dissemination between communities and environments through domestic effluents and food (Meirelles-Pereira et al., 2002;Cerceo et al., 2016).This factor explains the presence of bacteria in the genera Bacillus, Acinetobacter and Sphingomonas, found in Gurupi Regional Hospital, also commonly found in water, soil, and food products (Logan 2012;Smet et al., 2014;Ferreira et al., 2019;Arbab et al., 2021;Carvalheira et al., 2021).
Infections caused by nosocomial pathogens lead to morbidity and mortality in hospitalized patients with the emergence of MDR bacteria (Weber et al., 2010;Otter et al., 2013;Dancer, 2014;Sood and Perl, 2016).Reports have shown that biofilm formation is favorable to the growth of persistent colonization with methicillin-resistant Staphylococcus aureus, Acinetobacter spp., Clostridium difficile, and Pseudomonas aeruginosa (Kramer et al., 2006;Boyce, 2007;Chemaly et al., 2014;Hu et al., 2015).Here, we verified the phenotypic and genotypic characterization of ICU bed surface isolates, showing the presence of MDR bacteria, description of resistance genes, and identification by MALDI-TOF MS and 16S rRNA sequencing.
Limitations regarding the database of protein profiles of isolates make it challenging to use MALDI-TOF MS to identify unconventional bacterial strains of medical interest.Studies have described the failure to identify strains due to the absence of a broad protein spectrum from a reference bank or even the capture of the technique of specific proteins necessary to define the species.(Bizzini et al., 2011;Alizadeh et al., 2021).However, data on its applicability reveal MALDI-TOF MS's importance in quickly typing subspecies of medical importance; the technique contributes to the proper management of antimicrobials and best hospital practices.On the other hand, its limitations highlight the need to increase reference data for reference banks (Croxatto et al., 2012;Angeletti, 2017;Davies et al., 2021).
16S rRNA gene sequencing allowed greater coverage and effectiveness in identifying microorganisms.The technique is used widely for the identification of most bacterial taxa.However, the technique entails substantial performance time and high processing costs (Clarridge III, 2004;Bizzini et al., 2011;Church et al., 2020;Kosecka-Strojek et al., 2019).Our work using the 16S rRNA technique effectively identified the bacterial isolates classified by similarity and construction of the phylogenetic profile.
For ILH-10 (A. schindleri), we observed phenotypes resistant to several antibiotics in different classes.The isolate was positive for OXA-23, which may be associated with a multi-resistance phenotype to beta-lactams.Acinetobacter are opportunistic pathogens, and most are clinically significant, typically causing severe nosocomial infections in immunocompromised hosts (Joly-Guillou, 2005;Ibrahim et al., 2021).Acinetobacter is often associated with the accumulation of different resistance mechanisms to commercially available antibiotics, including carbapenems (Fournier et al., 2006;Périchon et al., 2014;Zander et al., 2014).Species producing class D β-lactamases of the oxa gene type (e.g., A. baumannii, A. johnsonii, A. pittii, A. baylyi, and A. schindleri) have been reported worldwide (Boo et al., 2009;Zhou et al., 2011;Zander et al., 2014;Smet et al., 2014).Furthermore, it is essential to note that resistance genes in non-A.baumannii (A.schindleri) species can cause bacteremia due to carbapenem resistance in immunocompromised patients (Montaña et al., 2017).
S. paucimobilis (ILH14) is an opportunistic pathogen involved in nosocomial infections; it can cause significant problems in clinical settings when resistant to antibiotics.Reports described S. paucimobilis in clinical specimens, blood, urine, endotracheal aspirates, bronchoalveolar lavage, and cerebrospinal fluid of patients with septicemia, meningitis, lower respiratory tract infections, and ventilator-associated pneumonia (Rohilla et al., 2021).Cases of early prosthetic valve infective endocarditis caused by MDR S. paucimobilis have been reported (Saboe et al., 2021).The S. paucimobilis (ILH14) isolated here was identified using the MALDI-TOF MS technique with phenotyping of MDR antibiotics the quinolone, penicillin, aminoglycoside, polymyxin, tetracycline, and oxazolidinone classes.
Bacillus sp.isolates (ILH11, ILH13, ILH2, and ILH16) showed resistance to clinically relevant antibiotics such as cephalosporins, quinolones, and penicillins.This genus is widely distributed in the environment, including soil, dust, air, fomites, water, and hospital environments (Bottone, 2010).The broad diversity of Bacillus allows gene regulation that increases its pathogenicity (Bottone, 2010).Previous studies identified MDR phenotypes in this group (Torkar and Bedenic, 2018;Liu et al., 2018, Mills et al., 2022).Some MDR strains of Bacillus sp obtained in the present study cause food poisoning.They can also cause localized wounds and eye infections; in a hospital environment, they can cause nosocomial bacteremia (Kuroki et al., 2009, Ehling-Schulz et al., 2019) and meningitis in immunocompromised patients (Gaur et al., 2001).Opportunistic infections by Bacillus circulans were identified in immunocompromised patients (Russo et al., 2021).
In summary, our findings suggest that several strains obtained from ICUs showed resistance to several classes of antimicrobials identified by molecular techniques and MALDI-TOF MS; however, the latter technique may be more relevant when associated with antimicrobial resistance phenotyping to assist antimicrobial management.Our findings reinforce the need to investigate resistance genes of the β-lactamase class.Finally, our findings suggest the need for amplification in detecting specific resistance genes.Our findings also contribute to identifying contamination sources responsible for increasing MDR organism levels.Monitoring these sources will improve environmental, food, and public health infection controls.

Figure 2 .
Figure 2. Phylogenetic tree obtained by maximum likelihood of the 16S rRNA gene sequences (400 bp) of the ILH10 isolate with identification of deposit in GenBank ON853562.1 (black arrow).ILH13 isolate with identification of deposit in GenBank ON853564.1 (black arrow).ILH2 isolate with identification of deposit in GenBank ON853570.1 (black arrow).Bootstrap values are indicated at the root of each branch.

Table 1 .
Target genes for PCR amplification, fragment size, and primer sequences.

Table 2 .
Phenotypic characterization of isolates from hospital ICU beds in Tocantins, Brazil.

Table 3 .
Antimicrobial resistance of isolates and 16S rRNA gene sequencing.

Table 4 .
Antimicrobial resistance of isolates and microbial identification by MALDI-TOF MS.