Antimicrobial resistance (AMR) is a natural phenomenon caused by the expression of a variety of resistance genes (Ramamurthy et al., 2022). The development of resistance is exacerbated by the high demand for antimicrobials, their misuse, and unnecessary prescriptions (Belachew et al., 2021). This situation diminishes the effectiveness of antimicrobial treatments, increasing morbidity and mortality rates of infectious diseases, leading to more complex therapeutic approaches, and escalating healthcare costs (Martínez et al., 2017). Currently, AMR is already among the top 10 threats to public health (CDC, 2019). Each year, bacterial infections are responsible for an estimated 7.7 million deaths, with 65% of these deaths being associated with antibiotic-resistant bacterial pathogens (The Lancet, 2024).

The interconnection of human, animal, and environmental health is conceptualized as One Health, an integrated approach in which AMR is relevant across its various dimensions since resistance determinants are as ubiquitous bacteria (Baquero et al., 2019). Within the human body, bacterial communities are part of a diverse ecosystem with other microorganisms and are present in all mucosal surfaces and the skin (Lawton, 2022). It is well-established that these communities (microbiotas) are fundamental for the proper functioning of nearly all human systems (MacAlpine et al., 2023; Pascale et al., 2018; Zhang et al., 2015). However, antimicrobials intended to eliminate pathogens also disrupt commensal microbiota, creating an imbalance condition known as dysbiosis. Altering the composition of a microbial community can increase its susceptibility to colonization by pathogenic bacteria, while also selecting for resistant strains. Antimicrobial resistance genes (ARGs) present in microbial communities, such as the intestinal microbiota, constitute the resistome, which exhibits inter-individual variation and correlates with several factors, including diet, travel, antimicrobial use, and the surrounding environment, such as hospitals (Crits-Christoph et al., 2022). The resistome and the human microbiota serve as a reservoir of ARGs, which can be acquired by pathogenic strains through horizontal gene transfer (HGT) (Zhang et al., 2022). Population-level variations and differences in ARG abundances in commensal microorganisms, such as those occurring during antimicrobial administration, may increase the frequency of successful transmission of these genes to pathogenic strains (Anthony et al., 2022; Crits-Christoph et al., 2022; Smillie et al., 2011).

The western region of Santa Catarina state, located in southern Brazil, is the most significant area for swine and poultry meat production in the country, generating over $2 billion in revenue annually, and is also one of the leading centers of meat production globally (Secretaria de Estado da Agricultura e Pecuária de Santa Catarina, 2023). It is well-documented that antimicrobials are extensively used in animal husbandry, and consequently, this environment is considered a major source of AMR induction (Jacobsen et al., 2023). We have previously shown that farms from this region present a high variety of ARGs in both, farms with open regimens and use of antimicrobials but also in those with closed regimens and no antimicrobial use (Beltrame et al., 2022). Moreover, we detected a colistin resistance gene, mcr-4, in a slurry sample from one of these farms. Unfortunately, few policies currently regulate antimicrobial use in these settings, and existing measures are mostly recommendations for reducing antimicrobial use rather than detailed plans to achieve this reduction.

In this study, we evaluated the gut microbiome and resistome composition of four hospitalized patients who are workers from the western region of Santa Catarina state, an area of intensive animal husbandry, following their admission to the University Hospital, located in the eastern region of the state.

The hospital environment is characterized by the high circulation of individuals, each associated with a unique microbiota, thereby constituting a significant reservoir of ARGs. These factors contribute to a highly complex ecosystem within healthcare facilities, which play a critical role in the ongoing AMR crisis. Thus, the rapid dissemination of resistant microorganisms and ARGs represents an escalating threat to human health, necessitating ongoing efforts to characterize and understand their transmission dynamics.

A total of 14 patients were recruited for our study, of which 4 were selected for further analysis. These 4 patients were specifically chosen because rectal swab samples were collected from them both at the time of admission and upon discharge. The following details pertain to these selected patients:

Patient 04 was an 85-year-old male admitted for a cutaneous infection in the left upper limb. Following a muscle drainage procedure, cultures confirmed the presence of methicillin-resistant Staphylococcus aureus (MRSA). The patient was treated with oxacillin and trimethoprim-sulfamethoxazole (Bactrim) during hospitalization and was subsequently discharged after a 10-day stay, following clinical improvement.

Patient 07 was a 60-year-old male admitted with an infected pressure ulcer in the sacral region. The patient received clindamycin and ceftriaxone during his hospital stay. The admission sample from Patient 07 (7AR) underwent microbiological analysis, revealing four isolates of Escherichia coli and one of Providencia stuartii. The selection of antibiotics for treatment was determined by the attending physician, independently of the data from this study. Subsequently, he was transferred to another hospital to continued care, and was discharged from the University Hospital (UH) after a 7-day admission.

Patient 09 was a 68-year-old male hospitalized for a chemotherapy cycle against blastic plasmacytoid dendritic cell neoplasm. Microbiological cultures identified the presence of Enterococcus faecium, Candida spp., and Pseudomonas aeruginosa resistant to cefepime and meropenem. The patient received multiple antimicrobials during his hospital stay, including trimethoprim-sulfamethoxazole, acyclovir, entecavir, cefepime, meropenem, polymyxin, voriconazole, and vancomycin. The patient died after 21 days of hospitalization.

Patient 13 was a 39-year-old male admitted for a chemotherapy cycle due to recurrent acute myeloid leukemia. He received various antimicrobials during his hospital stay, such as amoxicillin-clavulanate, sulfamethoxazole-trimethoprim, cefepime, meropenem, voriconazole, linezolid, and acyclovir. His hospitalization lasted for a total of 31 days.

The resistome composition analysis revealed a high prevalence of genes associated with resistance to aminoglycosides and tetracyclines across all samples in this study. The genetic element aph(3’)-IIIa, an aminoglycoside phosphotransferase encoded by plasmids in Staphylococcus aureus and Enterococcus spp., was the most prevalent among the resistance genes identified. Notably, this element accounted for 30% of the reads in the admission sample from patient 04.

The MLS_B phenotype, characterized by resistance to macrolides, lincosamides, and streptogramins B, is represented by the erm gene family, with ermB being the most prevalent determinant among the samples analyzed. Notably, samples from patient 07 displayed the highest abundance of this resistance determinant, with 112 reads in the admission sample (7AR) and 198 reads in the discharge sample (7DR). The resistance observed is attributed to target site modification mediated by the erm gene family, which encodes a methylase enzyme that methylates and modifies the antibiotic target, 23S rRNA (Manandhar et al., 2021).

The notable increase in the prevalence of the MLS_B phenotype observed between the admission and discharge samples of patient 07 may be attributed to the use of two antimicrobials during hospitalization: ceftriaxone, a beta-lactam antibiotic, and clindamycin, a lincosamide antibiotic.

Furthermore, we identified eleven bacterial isolates belonging to the order Enterobacterales in patient 07 by classical microbiological analysis, revealing four Escherichia coli isolates and one Providencia stuartii isolate upon admission. At discharge, five additional E. coli isolates and one Klebsiella pneumoniae isolate were identified. Antimicrobial susceptibility testing (AST) indicated that these isolates exhibited resistance to a range of antimicrobials. Notably, eight of the eleven isolates showed resistance to ceftriaxone, encompassing seven E. coli isolates and the Providencia stuartii isolate (data not shown). Additionally, an increase in the prevalence of fluoroquinolone resistance determinants was observed in the 7DR sample compared to 7AR, with the presence of five variants of the qnrB gene family.

The resistome analysis of patient 09’s samples revealed the presence of the mcr-1 gene, which was consistently identified at both admission and discharge. In this case, we believe that it consists of an acquisition outside the hospital and a stable colonization during the hospitalization of this patient. Other longitudinal studies about mcr-1 have already shown its prevalence over the years in patients but also in husbandry environments (Zhong et al., 2018). Notably, the mcr-4 gene was previously detected in a swine production facility located in the same region as the patient (Beltrame et al., 2022). The mcr genes are significant as determinants of colistin resistance transmitted via mobile genetic elements (Gao et al., 2016; Ling et al., 2020). Colistin, a polymyxin antibiotic, is considered a last-resort treatment for infections caused by multidrug-resistant Gram-negative bacteria in humans. Although resistance to colistin was initially believed to arise solely through chromosomal point mutations, the emergence of resistance mediated by mcr genes on plasmids has been documented in bacteria from various sources, including animals, humans, food, farms, and the environment since 2015 (Hussein et al., 2021; Liu et al., 2016). Considering the high complexity of the horizontal transfer of genes in a microbiome context, this finding highlights the importance of new policies to avoid dissemination and further infections of bacteria that are resistant to last-resort antibiotics. It is important to consider also that bacteria tend to accumulate resistance genes through genetic recombination, making colonizing bacteria into reservoirs of genetic mobile elements of pan-resistance. This trend is concerning as it highlights the growing spread of resistance to advanced antimicrobials across different environments.

During infection, the selective pressure exerted by antimicrobials drives microorganisms to acquire mutations in genes targeted by these drugs, leading to the development of resistance. Bacterial strains possessing such mutations often exhibit enhanced proliferation compared to those lacking antimicrobial resistance genes (ARGs) (Jian et al., 2021). For instance, to patient 9 was administered sulfamethoxazole (a sulfonamide) and trimethoprim (a diaminopyrimidine) for 20 days. Notably, resistance determinants dfrA1, dfrA17, and sul2, which are associated with resistance to these antimicrobials, were detected exclusively in the discharge sample. This suggests that the prolonged use of these medications may have contributed to the acquisition of resistance.

Conversely, the tetracycline resistance determinant tetQ, which encodes a ribosomal protection protein and is associated with a conjugative transposon, was detected in seven out of the eight samples analyzed. This gene was particularly prominent in the discharge sample from patient 09 (9DR), where its abundance increased approximately 19-fold compared to the admission sample from the same patient. The association of many tetracycline resistance determinants with mobile genetic elements likely contributes to their widespread distribution across various bacterial genera, thereby explaining their significant abundance (Pavelquesi et al., 2021).

Patient 13 exhibited a greater number of resistance determinants at discharge compared to admission, although there was a decrease in aminoglycoside and beta-lactam resistance determinants. This change is attributed to the diverse antimicrobials administered during the patient’s hospitalization. Over the 31-day stay, the patient received amoxicillin-clavulanate, a penicillin-class antibiotic combined with a beta-lactamase inhibitor that broadens the antimicrobial spectrum; cefepime, a fourth-generation cephalosporin with activity against both Gram-negative and Gram-positive organisms; and meropenem, a carbapenem considered a “last-resort” treatment against infections by Gram-positive and Gram-negative bacteria, due to its resistance to extended-spectrum beta-lactamases (ESBLs) (Huttner et al., 2020; Meurant et al., 2021; Raza et al., 2021). Additionally, the patient also received trimethoprim-sulfamethoxazole and linezolid, with the former belonging to the sulfonamide class, often administered in combination with a diaminopyrimidine to inhibit enzymes in the folic acid metabolic pathway, and the latter being an oxazolidinone, effective in treating Gram-positive pathogens, including methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE) (Roger et al., 2018; Wu et al., 2021).

Despite the small sample size, the findings of this study regarding antimicrobial resistance are consistent with existing literature. A whole-genome study of 335 novel bacterial species from the human microbiota identified MLS_B, aminoglycosides, and tetracyclines as the most prevalent resistance determinants (Khabthani et al., 2022). Similarly, Ma et al. (2016) reported high levels of resistance genes for tetracyclines, multidrug resistance, and aminoglycosides in fecal samples from humans, chickens, and pigs. Additionally, Wang et al. (2020) discovered 330 antimicrobial resistance genes in 18 samples of human, chicken, and pig feces, conferring resistance to 21 classes of antimicrobials, with tetracyclines, MLS_B, and aminoglycosides being among the most common.

Moreover, our results shows that aminoglycoside resistance is strongly related to bacteria from Enterobacteriaceae, Mycobacteriaceae and Neisseriaceae families. Other studies that look for aminoglycoside resistance genes presence in isolates from inpatients also find high prevalence in these families previously cited. A Chinese group, for example, found high prevalence of aminoglycoside-modifying enzyme genes in isolates of Enterobacter cloacae in a teaching hospital (Zhu et al., 2020). Another study suggests treatment alternatives to Klebsiella pneumoniae, Pseudomonas aeruginosa and Acinetobacter baumannii that are co-resistant to aminoglycosides, tetracyclines and other antimicrobials, highlighting the prevalence of these antimicrobials resistance in hospital environments (Karakonstantis et al., 2020).

Despite presenting a higher error rate than other platforms, the low capital investment required to obtain equipment such as MinION (Oxford Nanopore Technologies, Oxford, United Kingdom) contrasts with other companies available on the market, increasing its popularity. In addition, the association of this sequencing technology with refined bioinformatics pipelines highlights the potential of shotgun metagenomic sequencing with MinION (Leggett et al., 2020). Its use has proven to be efficient, user-friendly, and easy to insert into the laboratory routine. With nanopore technology, it is possible to detect antimicrobial resistance genes and their respective host microorganisms (Leggett et al., 2020).

Similar to the impact of clinical antimicrobials, the misuse of disinfectants and insufficient understanding of biosafety principles impose selective pressure, leading to the development of resistant microorganisms (Mc Carlie et al., 2020). Numerous studies have highlighted the correlation between resistance to antimicrobials and disinfectants in various bacteria, a phenomenon attributed to co-selection (Rozman et al., 2021; Staats et al., 2023; Van Dijk et al., 2022). The gene emrD, identified in this study, has been previously reported to encode an efflux pump that facilitates the expulsion of benzalkonium chloride from bacterial cells, thereby contributing to resistance (Nishino and Yamaguchi, 2001). Short et al. (2021) demonstrated that benzalkonium chloride, when used in high concentrations for surface disinfection in healthcare and domestic environments, interferes with the efficacy of aminoglycoside antibiotics in Acinetobacter baumannii and other pathogens in the ESKAPE group, such as Escherichia coli, Enterobacter cloacae, and Klebsiella pneumoniae (Short et al., 2021).

The microbiome composition of the analyzed samples was evaluated, revealing the predominance of the phyla Proteobacteria, Actinobacteria, and Firmicutes. This observation contrasts with the commonly reported dominance of Bacteroidetes and Firmicutes, followed by Proteobacteria. The composition and functionality of the intestinal microbiota can be influenced by various factors, including host genetics, diet, age, and antimicrobial usage (Gomaa, 2020).

Of particular note, the admission and discharge samples from patients 07 and 09 revealed the presence of Lactobacillus fermentum associated with the ermB gene, which confers the MLS_B resistance phenotype. Certain Lactobacillus species, including L. fermentum, are of interest due to their potential use as probiotics (Fernández et al., 2020). Clinical trials have demonstrated that daily intake of a formula containing L. fermentum CECT5716 by six-month-old children significantly reduced the incidence of upper respiratory tract infections, gastrointestinal infections, and overall infections in the subsequent six months (Maldonado-Lobón et al., 2015). Consequently, species within the genus Lactobacillus, including L. fermentum, have been granted Generally Recognized As Safe (GRAS) status by the FDA (Food and Drug Administration, USA) and Qualified Presumption of Safety (QPS) status by the European Food Safety Authority (EFSA) (Fernández et al., 2020).

However, literature indicates that lactobacilli exhibit inherent resistance to vancomycin, aminoglycosides, and most nucleic acid inhibitors (Gueimonde et al., 2013). Resistance to erythromycin and tetracycline is acquired, and thus the associated genetic determinants are often transferable. Acquired erythromycin resistance, encoded by the ermB gene, has been detected in strains of L. fermentum isolated from Chinese fermented foods. Additionally, plasmids potentially involved in transferring antibiotic resistance have been identified in L. fermentum (Anisimova and Yarullina, 2018). A study also demonstrated the successful transfer of the ermB gene from L. fermentum to Enterococcus faecalis (Nawaz et al., 2011). Thus, the detection of a resistance determinant in this non-pathogenic species is concerning, highlighting the need for further investigation into the antibiotic resistance profiles of probiotic cultures at both physiological and molecular levels.

In the admission samples, Escherichia coli and Streptococcus agalactiae emerged as prominent species. E. coli, a Gram-negative bacterium, can exist as both a commensal and pathogenic organism and is known for its broad spectrum of antimicrobial resistance, including resistance to fluoroquinolones (Fuga et al., 2022). Notably, in the United States, where fluoroquinolones are not approved for use in animal production, the resistance rates of E. coli to fluoroquinolones and quinolones are reported to be below 5%. In contrast, the average resistance rate in E. coli exceeds 40% in countries such as Brazil, China, and across the European Union, where the use of fluoroquinolones in animal production is permitted (Roth et al., 2019).

In the discharge samples, Clostridioides difficile, Bacteroides fragilis, and Neisseria meningitidis were notably prevalent. N. meningitidis, the primary causative agent of invasive meningococcal disease (IMD), including septicemia and meningitis, poses a significant health threat due to its potential for rapid progression and epidemic spread, necessitating immediate antibiotic intervention and prophylaxis (Taha and Deghmane, 2022). Although resistance in N. meningitidis is relatively rare, patterns of reduced susceptibility to antibiotics such as ampicillin, penicillin G, ciprofloxacin, levofloxacin, and trimethoprim-sulfamethoxazole have been documented. Furthermore, a strain of N. meningitidis harboring the beta-lactamase gene ROB-1 (bla _ROB-1) has been recently identified in several countries (Shao et al., 2022). The identification of N. meningitidis associated with antimicrobial resistance genes in this study underscores the critical need for ongoing surveillance to monitor and address the emergence of resistant strains.

In our study, the most prevalent bacterial family identified was Mycobacteriaceae, underscoring its significance in the resistome profile observed. Within this family, resistance genes linked to aminoglycosides, peptides, rifamycins, and fluoroquinolones were detected. The presence of these resistance determinants is particularly alarming due to the critical role these antibiotic classes play in the treatment of mycobacterial infections.

Our findings reveal that the primary mechanisms of resistance are associated with mutations in genes encoding essential targets. Notably, these include mutations in 16S rRNA genes related to aminoglycoside resistance, alterations in RNA polymerase subunits associated with rifamycin resistance, and changes in DNA gyrases contributing to fluoroquinolone resistance ( Figure 4 ).

In this study, we analyzed the resistome and microbiome composition of individuals admitted to and hospitalized at the University Hospital in Florianópolis, a coastal city in southern Brazil. Despite the small sample size, a key limitation of the study, the results are particularly significant due to the unique characteristics of this patient cohort. This work was part of a larger project investigating the dynamics of antimicrobial resistance gene (ARG) circulation in Santa Catarina, focusing on the interplay between human healthcare and animal husbandry environments and published in part in 2022 (Beltrame et al., 2022). All recruited patients, although hospitalized in Florianópolis, originated from the Western region of Santa Catarina, a livestock-intensive area located approximately 400–650 km inland. While we initially aimed to recruit a larger cohort, the number of eligible patients from this region was lower than anticipated, further reduced by exclusion criteria and the interruption of the study due to the COVID-19 pandemic. Moreover, our study sought to evaluate the influence of the hospital environment on the resistome and microbiome. The inclusion of admission samples allowed us to account for variations in patient age and enrich our understanding of pre-existing resistome characteristics prior to hospitalization. Firstly, the patients are permanent residents of the western region of Santa Catarina, whose capital is Florianopolis. This region is a leading global producer of swine and poultry meat, which is associated with high levels of antibiotic use in livestock. Although Brazilian legislation mandates stringent controls on antimicrobial use in animal husbandry—including the requirement for veterinary prescriptions, monitoring of antimicrobial resistance, and adherence to safety standards to safeguard public health (MAPA, 2006, 2016)—there is less focus on mitigating environmental contamination from antimicrobials used in these agricultural practices.

Secondly, residents of this region are transported by municipal vehicles for treatment and hospitalization at the University Hospital in Florianopolis, situated at the opposite end of the state. This hospital is renowned for its specialization and high quality, being affiliated with the Federal University of Santa Catarina, ranked as the fifth-best university in Brazil (UFSC, 2024). The transportation and medical treatment provided are free of charge through the Brazilian Unified Health System (SUS).

Lastly, following discharge, patients are transported back to their hometowns using municipal transport services, which, while suitable for inter-municipal travel, are not specifically designed for transporting patients or individuals in recovery.

Thus, the observations presented herein reflect a highly unique situation. This involves the transfer of individuals colonized by various types of multidrug-resistant bacteria, likely acquired in regions with intensive animal production, to a hospital of significant complexity located at the opposite end of the state of Santa Catarina. This hospital, renowned for its advanced medical care, serves numerous patients from diverse cities and municipalities within the state and beyond. These patients receive treatments involving a broad spectrum of antibiotics, potentially influencing the resistome dynamics within this specialized healthcare setting.

All these factors underscore the considerable diversity observed in both the resistome and microbiome compositions. They highlight the significant resistance selection pressures at play, which are proportional to the challenges associated with controlling the emergence of new resistances. These observations should be taken into account in the formulation of policies and the implementation of targeted measures aimed at effectively controlling the development and dissemination of novel multidrug-resistant bacteria.