Phenotypical and molecular characterization of Rhodococcus equi isolated from foals in the Agreste region of Pernambuco - Brazil

Abstract

Equine rhodococcosis is caused by Rhodococcus equi, an intracellular coccobacillus whose main virulence factor is a plasmid that harbors genes encoding proteins from the Vap family, with the vapA gene being the most important in equine isolates. Furthermore, other factors observed in R. equi strains, such as antimicrobial resistance and biofilm production, may represent significant challenges in the treatment of affected animals. The objective of this study was to characterize four isolates of R. equi from foals in the state of Pernambuco, Brazil. All isolates were identified as R. equi through biochemical tests, amplification of the choE gene, and sequencing of 16 S rRNA. PCR analysis revealed that three isolates were positive for the plasmid virulence genes (vapA, -C, -D, -E, -F, -H and traA), although vapD was absent in one of the three isolates. One isolate did not present any virulence genes, possibly due to the loss of the plasmid after repeated passages at 37ºC. In the antimicrobial susceptibility test, all isolates were susceptible to erythromycin, clarithromycin, azithromycin, rifampicin, gentamicin, and doxycycline. However, all isolates were capable of forming biofilms, with moderate biofilm formation in isolates Rhodo1 and Rhodo2, and weak biofilm formation in isolates Rhodo3 and Rhodo4, which may be associated with increased antimicrobial tolerance. This molecular characterization demonstrated, for the first time, the presence of the virulence plasmid in R. equi isolates from foals in Northeast Brazil, as well as their capacity for biofilm formation.

Introduction

The equine rhodococcosis is considered the most common cause of subacute or chronic granulomatous bronchopneumonia in foals under 5 months of age. In recent decades several studies have demonstrated the ability of R. equi to cause disease in humans, especially immunocompromised individuals [1–6]. Several virulence factors are involved in the ability of certain strains to cause disease in humans and animals [7, 8]. However, the central factor responsible for virulence in isolates from sick foals is the presence of an 85-90 kb plasmid that has in its structure a pathogenicity island of 27,536 bp, containing seven virulence-associated protein (vap) genes. The structure is organized in three genes (vapA, vapC and vapD) grouped together, one pair with a gene and a pseudogene (vapE and vapF) and two others located independently (vapG and vapH) [9–12]. The vapB and vapN genes are present on a circular plasmid commonly isolated from pigs and a non-circular plasmid commonly isolated from ruminants, respectively [12–14].

In recent years, several studies have reported the emergence and increase in the prevalence of R. equi isolates resistant to macrolides and rifampicin, these drugs are commonly used, alone or in combination, in the treatment of rhodococcosis in horses [3, 15–17]. This is worrisome due to the risk of spreading antimicrobial resistance between strains and properties.

Biofilm formation is an important virulence factor in several genera of microorganisms. A study carried out with R. equi demonstrated that this pathogen could form biofilms on the surface of polyurethane catheters in hospitalized human patients [18]. After this report, other authors confirmed the formation of biofilm in strains isolated from feces, soil and clinical cases in horses [19–22]. Biofilm formation by R. equi may have important clinical implications, as these studies demonstrated increased antibiotic tolerance in cells present in biofilms by up to ten times the minimum inhibitory concentrations (MICs) observed in free cells [21, 22].

The first report of infection by R. equi in Brazil was made in 1959, in a case of mastitis in a mare [23], and the first outbreak of equine rhodococcosis, with clinical respiratory manifestation in foals occurred only in 1970 [24]. Since then, several studies in the country over the years have reported the isolation of the pathogen in horses, humans, and other species of animals [25–29]. However, most of the scientific works is concentrated in the south and southeast regions of the country, with, to the best of our knowledge, no scientific report to the present date reporting the virulence plasmid characterization of R. equi in Brazil north and northeast regions [30–36].

Therefore, the present work aimed to carry out the first molecular characterization of isolates identified as R. equi from foals that died in the countryside of the state of Pernambuco, Brazil.

Materials and methods

Ethical aspects

The necropsy procedures and sample collection from the animals used in the study were authorized, and the owners of the animals in question agreed to the use of the data generated in the research.

Samples

Samples were collected on a property intended for breeding horses located in the state of Pernambuco, in the Northeast region of Brazil, between the months of March and November 2019. Four foals, aged 1.5 to 3 months that presented clinical symptoms compatible with equine rhodococcosis for 23–30 days, treated with azithromycin and rifampicin, without clinical improvement, resulting in their death. During the necropsy of the animals, the purulent contents of the granulomatous lesions present in the lungs of the animals were aseptically collected. These contents were packaged in sterile test tubes and sent for processing to the Microbiology and Immunology Research Laboratory (LAPEMI) at the Federal University of Agreste de Pernambuco (UFAPE).

Isolation and phenotypic characterization

The isolation and phenotypic characterization of R. equi isolates was carried out as described by Javed et al. [37], with minor modifications. Samples from the foal necropsy were inoculated into 5% sheep blood agar and incubated aerobically at 37 °C for 48 h. Bacterial isolates that presented small, smooth, shiny and non-hemolytic colonies after 24 h of incubation, and subsequently became larger, with a mucoid appearance and salmon-pink color at 48 h, were selected for biochemical tests. Initial confirmation of the isolates as R. equi was made by demonstrating typical cellular morphology in Gram tests. All four R. equi isolates obtained during the study were biochemically characterized using the catalase test and Christein-Atkin-Munch-Peterson (CAMP) test using the Staphylococcus aureus ATCC 25923 strain as indicator strain, urease test and test of sugar fermentation. The strains R. equi ATCC 33701 P+ (vapA positive) and the strains R. equi ATCC 6939 and R. equi ATCC 33701 P- (vapA negative) were used as positive controls in the tests.

Antimicrobial susceptibility test

Antimicrobial susceptibility testing was performed using the disk diffusion method in accordance with the Clinical and Laboratory Standards Institute (CLSI VET01-A4) guidelines [38]. Because the reading was performed at 24 h, the Mueller-Hinton agar was enriched with 5% sheep blood, to improve the growth of R. equi isolates and allow better measurement of inhibition zones at the correct time. The antimicrobials evaluated in this study were erythromycin, clarithromycin, azithromycin, rifampicin, gentamicin and doxycycline (Cecon, SP, Brazil), due to their inhibitory potential on R. equi carried out in in vitro tests [39] and because the combination of a macrolide associated with rifampicin is the most frequent therapy of choice for the treatment of affected animals [40].

To date, there are no interpretative criteria approved by CLSI for susceptibility testing of R. equi strains isolated from animals. Because of this, the criteria adopted by Berghaus et al. [41], were used [Table 1]. The strains S. aureus ATCC 29213 and Enterococcus faecalis ATCC 29212 were used as control strains for the tests, as negatives, and the strains R. equi ATCC 6939 and R. equi ATCC 33701 P + as a positive control.

Table 1.

Interpretive criteria used to classify R. Equi isolates into the described antimicrobials.

Adapted from Berghaus et al. [41]

Antimicrobials a	Breakpoint (mm)
	Susceptible	Intermediary	Resistant
Azithromycin (15mcg)	≥ 18	14–17	≤ 13
Clarithromycin (15mcg)	≥ 18	14–17	≤ 13
Erythromycin (15mcg)	23	14–22	≤ 13
Doxycycline (30 mcg)	≥ 16	13–15	≤ 12
Gentamicin (10mcg)	≥ 16	13–15	≤ 12
Rifampicin (5mcg)	≥ 20	17–19	≤ 16

^a Because there are still no interpretative criteria approved by CLSI for susceptibility testing on R. equi strains, CLSI interpretative criteria for Staphylococcus aureus were used as performed in the reference work [41]

PCR assay

DNA extraction

The isolates were cultivated in Brain Heart Infusion broth (Brain, Heart infusion - BHI, KASVI) for 48 h, then 1 mL of the culture was inserted into a microtest tube and centrifuged at 7000 rpm for 3 min in a microcentrifuge. The extraction of bacterial DNA from the pellet was carried out using the PureLink™ Genomic DNA Mini Kit (Invitrogen, USA) according to the manufacturer’s instructions. The quantity and quality of DNA were determined by spectrophotometry using the GENESYS 10 S Series spectrophotometer (Thermo Fisher Scientific Inc., USA) and by electrophoresis in 1% agarose gel stained with DSView Nucleic Acid Stain (Sinapse Inc.).

PCR

PCR tests were performed to characterize R. equi isolates for the choE genes, vap family genes and traA. The primers used in each PCR reaction and their sources are described in Table 2. The PCR reaction temperatures followed those recommended by previous works [Table 2]. All PCR reactions were performed in the Bio-gener GE9612T-S thermal cycler (Bio-Gener, Hangzhou, China), using the FIREPol™ Master Mix Ready kit to Load (12.5 mM MgCl2) (Solis Biodyne, Tartu, Estonia), with the concentrations recommended by the manufacturer. The R. equi ATCC 33701 P + strain and the R. equi ATCC 6939 and R. equi ATCC 33701 P- strains were used as positive and negative controls for plasmid virulence genes, respectively. The products obtained from each PCR reaction were revealed through electrophoresis in a 1% agarose gel, added with DNA Sybr Safe dye (Invitrogen, USA) and visualized in a UV photo documentation device (Dual Intensity UV Transilluminator Labnet - DyNA).

Table 2.

Gene primers used in the molecular characterization of R. Equi isolates

Gene	Primers 1	Sequence	Size (bp)	Reference
choE	COX-F	GTCAACAACATCGACCAGGCG	959	[67]
	COX-R	CGAGCCGTCCACGACGTACAG
vapA	VAPA-F	ACAAGACGGTTTCTAAGGCG	550	[46]
	VAPA-R	TTGTGCCAGCTACCAGAGCC
vapB	VAPB-F	GAATTCGAAAGCGCAAAGGT	650	[46]
	VAPB-R	TTCCGTGAACATCGTACTGC
vapC	VAPC-F	GGGTCGTCCATCCAAATCGA	700	[46]
	VAPC-R	GGTCAGGCCTATCACCCTTG
vapD	VAPD-F	GGTGGTGCGATGTCAGAATG	400	[46]
	VAPD-R	TGGAACGTCTTGCCCTTCTT
vapE	VAPE-F	ATATGACGACCGTTCACAAG	600	[46]
	VAPE-R	CTCCGATGCCCACCAAACTA
vapF	VAPF-F	AGAATATGCCTGGTATGGGC	350	[46]
	VAPF-R	TCGTCGTATAGCTGCTGCAG
vapH	VAPH-F	AATTCCTATCAAGGACAGC	500	[46]
	VAPH-R	ATACCGATTACGGAGCTCAC
vapN	VAPN-F	GGTACTGCAGGCAACTGCTA	425	[45]
	VAPN-R	GAGCTGCTACTACCGTGGTC
traA	TRAA-F	AGAGTTCATGCGTGACAACG	959	[68]
	TRAA-R	GTCCACAGGTCACCGTTCTT
16 S rRNA	27-F	AGAGTTTGATCCTGGCTCAG	1350	[69]
	1492-R	GGTTACCTTGTTACGACTT

¹ Each primer for each gene was tested in a uniplex PCR

16 S sequencing

The analysis of the 16 S rRNA region was performed by the PCR amplification method using the universal primers 27-F and 1492-R described by Hongoh et al. [42],. [Table 2]. The products resulting from the PCR reaction were then purified with the Illustra GFX PCR DNA and Gel Band Purification Kit (Cytiva, UK) according to the manufacturer’s instructions. The purified DNA samples were sent to the Department of Genetics at the Federal University of Pernambuco (UFPE) for genetic sequencing. Sequence editing and analysis were performed using the Bioedit software (Informer Technologies, Inc.) and consensus sequence alignment using the MEGA package (Molecular Evolutionary Genetics Analysis; Version 11.0. 13).

Biofilm formation test

The assessment of biofilm formation was carried out according to the methodology used by Bujold, Lani and Sanz [20], with modifications. To standardize the initial inoculum of ~ 10 ⁵ CFU/mL, a serial dilution test of the isolates was carried out in BHI broth for 24 h. The procedure was carried out on each of the 4 isolates and control strains in test tubes containing PBS up to a concentration of 10 ^− 5 and inoculated using the spread plate technique on plates containing Mueller-Hinton agar at concentrations 10 ^− 3, 10 ^− 4 and 10 ^− 5 in duplicate. Once the inoculum was standardized, 100 µL of bacterial culture of each isolate at 0 h post- inoculations were added to six wells in parallel to a 96-well plate (Greiner Bio-One, Brazil). The S. aureus ATCC 29213 strain was used as a positive control, being added in the same volume to ten wells. For negative control purposes, sterile BHI broth was added to another ten wells per plate to obtain a blank for performing calculations. For intraspecific control purposes, the strains R. equi ATCC 6939, R. equi ATCC 33701 P + and R. equi ATCC 33701 P- were used in the tests. Three 96-well plates were inoculated identically for the test, and incubated at 37 °C for 24, 48, or 72 h. After the respective incubation period for each plate, the broth was removed by gentle suction and inversion on paper towel. Subsequently, the wells were thermally fixed at 56 °C for 20 min, and then stained with crystal violet (0.1%). Then, washed 3 to 5 times with distilled water (until the water is clear), air-dried and bleached with 70% ethanol. Absorbance was measured at 595 nm using an Asys UVM 340 microplate reader (Biochrom, Cambridge, UK). Additionally, four empty wells per plate were stained with crystal violet in order to confirm that the plastic material did not retain the dye. The tests were carried out in triplicate for each measurement moment.

To interpret and classify the results of the isolates, the criteria established by a previously published study were used, which evaluated biofilm production in Staphylococcus sp [43]. with modifications. From measuring the optical density (OD) of each well, the cutoff OD (ODc) was set at a value of three standard deviations above the mean OD of the negative control. The results of the isolates were interpreted as follows:

(1) non-biofilm producers (OD ≤ ODc);

(2) weak biofilm producers (ODc < OD ≤ 2 × ODc);

(3) moderate biofilm producers (2 × ODc < OD ≤ 4 ×ODc);

(4) strong biofilm producers (4 × ODc < OD).

To achieve greater standardization of the average for each isolate, the highest and lowest measurements of the OD of the wells of each isolate and controls of each plate on each day of measurement were removed from the calculations.

Results

Sample collection

Upon necroscopic examination, all animals showed the same pattern of changes in the thoracic cavity [Fig. 1]. When the cavity was opened by folding the ribs, the disseminated presence of granules in the lung parenchyma was observed, which, when cut, were filled with pus. The lung parenchyma presented atelectatic areas in the dorsal region. No changes were observed in other organs in any of the animals.

Fig. 1.

Foal lung with suspected equine rhodococcosis showing the presence of multiple disseminated to coalescing granular lesions, with purulent content when cut

Biochemical identification and antimicrobial susceptibility test

All isolates showed a Gram-positive coccobacilli morphology. Biochemical tests showed the results: positive catalase, positive urease and positive CAMP test. The results of measuring the halos are shown in Table 3. All isolates were susceptible all the antimicrobials tested according to the reference criteria adopted.

Table 3.

Results in millimeters of the diameters observed in the TSA of R. Equi isolates

Antimicrobials a	Inhibition zone diameter
	Rhodo 1	Rhodo 2	Rhodo 3	Rhodo 4
Azithromycin	30	34	31	31
Clarithromycin	40	44	42	40
Erythromycin	36	38	39	36
Doxycycline	25	29	25	24
Gentamicin	32	32	37	34
Rifampicin	29	33	37	37

^a The diameters of the inhibition zone were observed after 24 h of incubation

Molecular analysis results and 16 S rRNA sequencing

The four isolates studied, as well as the control strains of R. equi, were positive for the choE chromosomal gene due to the presence of the 959 bp amplicon in the agarose gel (1%), thus all isolates were identified as R. equi [Fig. 2].

Fig. 2.

Products of the PCR reaction for the choE gene. All isolates, including the reference strains, tested positive for the chromosomal gene; 1 - Rhodo 1; 2 - Rhodo 2; 3– Rhodo 3; 4– Rhodo 4; 5– R. equi ATCC 6939; 6– R. equi ATCC 33701 P+; 7– R. equi ATCC 33701 P-; 1 Kb DNA Ladder Ready to Load Marker (Solis Biodyne, Tartu, Estonia)

The presence of the virulence plasmid was also observed, based on confirmation of the amplification of the plasmid genes. As for the vap family genes, the Rhodo 1 sample did not show a positive result for any of the proteins nor for the traA gene [Figs. 3 and 4]. The other 3 isolates, as well as the positive control, were positive for the vap family genes, and the traA gene. However, the Rhodo 2 isolate did not demonstrate the presence of the vapD gene [Fig. 3].

Fig. 3.

Results of the PCR reaction for the vapA, vapB, vapC, vapD and vapE genes. Of the four isolates, Rhodo 1 did not demonstrate amplification of any vap family gene, and Rhodo 2 did not demonstrate amplification of vapD; 1 to 8- vapA; 9 to 16- vapB; 17 to 24- vapC; 25 to 32- vapD; 33 to 40- vapE; 1, 9, 17, 25, 33- Rhodo 1; 2, 10, 18, 26, 34– Rhodo 2; 3, 11, 19, 27, 35– Rhodo 3; 4, 12, 20, 28, 36– Rhodo 4; 5, 13, 21, 29, 37– R. equi ATCC 6939; 6, 14, 22, 30, 38– R. equi ATCC 33701 P+; 7, 15, 23, 31, 39– R. equi ATCC 33701 P-; 8, 16, 24, 32, 40– white; 100 bp DNA Ladder Ready to Load marker (Solis Biodyne, Tartu, Estonia)

Fig. 4.

Results of the PCR reaction for the vapF and vapH genes. Except for Rhodo 1, all 3 other isolates presented vap gene products. 1 to 8- vapF; 9 to 16- vapH; 1, 9- Rhodo 1; 2, 10– Rhodo 2; 3, 11– Rhodo 3; 4, 12– Rhodo 4; 5, 13– R. equi ATCC 6939; 6, 14– R. equi ATCC 33701 P+; 7, 15– R. equi 33701 P-; 8, 16– white; 100 bp DNA Ladder Ready to Load marker (Solis Biodyne, Tartu, Estonia)

On the amplification of 16 S rRNA amplification, all isolates presented an amplicon of approximately 1350 bp. Pairwise sequence alignment was performed using the Paired Nucleotide Sequence Alignment tool accessible on the EzBioCloud web page (https://www.ezbiocloud.net/). The 16 S gene sequences of the four isolates were identified as being from the Rhodococcus equi species.

3.5 Biofilm formation

The results of the mean measurements for each strain and controls, as well as the cutoff points and interpretation ranges for the three days of the experiment are shown in Tables 4 and 5, respectively. According to the evaluation criteria, at 24 h the averages of all isolates, except for R. equi 33701 P+, were classified as weak biofilm production.

Table 4.

Average optical densities of isolates and reference strains and cutoff points in each measurement of the biofilm formation assay

Isolates a	Averages
	24 h	48 h	72 h
Rhodo 1	0.087	0.143	0.271
Rhodo 2	0.065	0.073	0.223
Rhodo 3	0.072	0.082	0.092
Rhodo 4	0.091	0.067	0.07
R. equiATCC 6939	0.08	0.074	0.089
R. equi ATCC 33701 P+	0.052	0.061	0.128
R. equi ATCC 33701 P-	0.059	0.063	0.213
S. aureus ATCC 25923	0.082	0.138	0.295
Negative Control	0.049	0.056	0.055

^a The averages were calculated from the values observed in each well. For greater uniformity of values, the highest and lowest values from the wells of each strain were excluded on each measurement day. All clinical isolates were biofilm producers. The highest means for each isolate were observed at 72 h of the assay

Table 5.

Cutoff points and interpretation ranges for each measurement moment of the biofilm formation assay

Rating a	Interpretation ranges
	24 h	48 h	72 h
Cutoff	0.053	0.066	0.069
Non-producer	≤ 0.053	≤ 0.066	≤ 0.069
Weak	0.053 < Odt ≤ 0.106	0.066 < Odt ≤ 0.132	0.069 < Odt ≤ 0.138
Moderate	0.106 < Odt ≤ 0.212	0.132 < Odt ≤ 0.264	0.138 < Odt ≤ 0.276
Strong	Odt > 0.212	Odt > 0.264	Odt > 0.276

^a The cutoff point for each measurement was calculated based on the standard deviation of the OD values of the negative control wells (C-) and then multiplied by three and added to the average C- to obtain it. The values of the classification ranges of the isolates varied according to the cutoff point measured on each day of the test, calculated in accordance with what was described in the methodology

At 48 h, all the isolates showed an increase in the mean OD, and the Rhodo1 isolate started to show moderate biofilm formation, as did the positive control, S. aureus ATCC 25923. However, the R. equi ATCC 33701 P + strains and R. equi ATCC 33701 P-, according to the adopted interpretative criteria, did not show biofilm formation.

At 72 h, the highest average ODs of all isolates were observed. Isolates Rhodo 1, Rhodo 2 and the control strain R. equi ATCC 33701 P- showed moderate biofilm formation, while isolates Rhodo 3 and 4 showed weak formation. The positive control showed high biofilm formation.

Thus, the results indicate that all the isolates were biofilm producers. The assay also demonstrated that there was variation in the formation capacity among the clinical isolates studied, as, at 72 h, two of the four clinical isolates studied proved to be moderate producers and the other two isolates were weak producers.

Discussion

Considered an emerging pathogen [44], Rhodococcus equi has been identified as an etiological agent in several species, including humans [16, 25, 27, 29, 45, 46]. The R. equi strains are classified as virulent or non-virulent based on the presence or absence of the plasmid associated with virulence [9]. To date, three types of plasmids have been identified, pVAPA, pVAPB and pVAPN, the first being identified in equine isolates [14]. Plasmid-free strains have been shown to be unable to survive within macrophages and cause disease in hosts. Furthermore, the pVAPA plasmid demonstrated transfer capacity between virulent and avirulent isolates in vitro [47]. The traA gene, present in the conjugative region of the plasmid, demonstrated a central role in this process, since it’s deletion abolished transmissibility between isolates [48]. Therefore, research into the presence of the plasmid is of great importance due to its pathogenic and zoonotic capacity. Previous work in the northeast region identified R. equi as the etiological agent of clinical pneumonia in foals [34–36], however, none of the studies carried out molecular characterization of the isolates, only being diagnosed through phenotypic tests.

On this study, it was observed two different molecular profiles: vapA, -C, E, -F, -H and traA presented by Rhodo2 and vapA, -C, -D, E, -F, -H and traA, presented in Rhodo3 and − 4. A molecular study of R. equi isolates carried out in southern Brazil found that the 32 virulent isolates found demonstrated six molecular profiles: 100% had the genes vapA, vapD and vapG, 86.6% vapF, 76.6% vapH, 43.3% vapC, 36.6% vapE and no vapB [46]. In another carried out in India, all 28 isolates were positive for vap -A, -C, -D, -E, -F, -G and -H [49]. However, the absence of VapD on an isolate, shown in Rhodo2, wasn’t seen none of these studies.

The main virulence gene in horse strains, vapA encodes a protein that interacts with hosts membrane by causing dysfunction in the organelles endocytic cells leading to the permeabilization of the phagosomal and lysosomal membranes [50] and maintenance of neutral pH allowing the multiplication of R. equi [12]. Although the function of the vapA gene is well known, it is not yet clear whether the remaining five vap genes (-C, -D, -E, -G and -H) play any role in survival or multiplication inside macrophages [51–55]. However, research into these genes may be important since it has been reported that antibodies recognizing VapA do not cross-react with recombinant Vaps C, D or E, which indicates that Vaps can be distinguished immunologically [51]. As for the traA gene, isolates Rhodo 1, 2 and 3 showed positive results.

The virulence plasmid genes were absent on the isolate Rhodo1 as shown by PCR reactions. A work carried out by Takai et al. [56]demonstrated that strains with virulence plasmid positive, when carrying out several passages at 37ºC, there was an attenuation of these strains, which were originally isolated from foal lung infections. The results of this study could be a possible explanation for the absence of virulence plasmid genes in the Rhodo1 sample. Thus, the scientific literature indicates that the R. equi virulence plasmid is possibly lost in the absence of host selection, in addition to the fact that certain factors such as temperature and pH play an important role in preserving the virulence of the isolated strains [13, 14, 56–58].

Regarding antimicrobial resistance, none of the strains isolated in this study were considered resistant to the antimicrobials of the susceptibility test. Most common drugs used in equine rhodococcosis treatment, resistance to macrolides has shown a consistent increase in prevalence since then as indicated by American and European studies [17, 59]. A study carried out in Brazil with clinical samples, collected on equine breeding farms between 1991 and 2013 reported that all isolates were sensitive to azithromycin and clarithromycin; however, 27% (12/44) showed intermediate sensitivity to erythromycin [15].

All R. equi isolates analyzed demonstrated biofilm formation. According to the adopted evaluation criteria, after 24 h of incubation, all strains formed biofilms. At 72 h, Rhodo1 and Rhodo2 presented strong and moderate formations, showing that the isolates presented variated capacities of producing biofilm. The ability to form a biofilm is considered a virulence factor in bacterial pathogens because the cells present in this structure demonstrate increased tolerance to antimicrobial compounds and persistence of infection despite the host’s continued immune response [60–62]. Biofilms formed by R. equi demonstrated the occurrence of this phenomenon. Studies carried out in southern Brazil and Italia observed concentrations as high as ten times the minimum inhibitory concentrations (MICs) for bacteria in planktonic form, was not able to completely destroy R. equi in biofilm [21, 22]. The results of the studies consulted corroborate the possibility that biofilm production plays an important role in the development of a tolerance system to antibiotics in the etiological agent, thus persisting in a quiescent form and, eventually, contributing to the chronicity of the infectious condition. Therefore, when observing the available data, it can be concluded that even though antibiotics are effective in controlling infections, the increased tolerance presented by biofilm-producing bacteria, once this has already been formed, represent a serious challenge to treatment. of the affected animals.

Although R. equi demonstrates growth across a wide temperature range (10 to 40 °C) [63], most studies are based on samples obtained from temperate countries or regions [59, 64, 65]. A recent review analyzing the global distribution of pVAPA plasmids associated with R. equi in equines revealed a lack of data on pVAPA subtypes in tropical countries [66]. Only Brazil, Peru, Venezuela, and Zimbabwe had identified strains regarding plasmid subtypes. However, for studies published in Brazil, research on the presence of plasmids in bacterial samples is concentrated in the southern and southeastern regions, particularly in the states of São Paulo and Rio Grande do Sul. In contrast, the northern and northeastern regions, closer to the equator, have, to date, no studies identifying the presence of virulence plasmids in these areas. From an epidemiological standpoint, although the small sample size in this study does not provide extensive data on the disease in the region, the findings presented here are significantly impactful due to the scarcity of research in tropical climates and the importance of virulent R. equi isolates in equine farming and public health, given their zoonotic risk.

Conclusion

To the best of our knowledge, this is the first study to carry out the molecular characterization of R. equi isolates in the northeast region of Brazil, since the previously published works proceeded only the diagnosis of the etiological agent only through phenotypic and biochemical tests [30–36] In this way, we demonstrated for the first time the presence of R. equi isolates positive for the vapA virulence gene in foals in northeastern Brazil.

Further studies are necessary to characterize the subtypes of the plasmids found, as well to identify other isolates in different areas, in order to enhance our understanding of the genetic diversity and virulence potential of these isolates. Furthermore, the findings from this work can assist other research groups by demonstrating the presence of virulent strains in the northeastern region of Brazil. Also, in other tropical regions given the scarcity of data on rhodococcosis in these localities worldwide. Such information could pave the way for future investigations into the epidemiology of R. equi and its impact on equine health in these areas, ultimately contributing to better management and prevention strategies.

Author contributions

All authors contributed significantly to the research reported in this manuscript. APSG and MM were responsible for the conceptualization of the study. The methodology was developed by MM, APSG and ERS. Validation of results was carried out by MM and FRC, while formal analysis was conducted by APSG, JCSV and GGS. MM and APSG led the investigation, and resources were managed by GFC. Data curation was handled by MM and GFC. The original draft of the manuscript was written by APSG, and subsequent review and editing were performed by MM, FCR, and ERS. GFC was in charge of visualization. Supervision of the project was carried out by MM, with project administration by AOSG and funding acquisition by MM and GFC.

Data availability

The datasets generated during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

None of the authors has any financial or personal relationships that could inappropriately influence or bias the content of the paper.