Antimicrobial resistance and genetic relationships of enterococci from siblings and non-siblings Heliconius erato phyllis caterpillars

Sample collection

The fecal samples used in the present study were collected from fifth instar caterpillars. The caterpillars were sourced from three different populations of Heliconius erato phyllis butterflies and consisted of sibling from the same female. The H. erato phyllis females (HE) were captured with entomological nets in Rio Grande do Sul, South Brazil. The first female (HEAB2) was collected in a forest fragment located in Águas Belas Agronomical Station (30°02′18.1″S; 51°01′23.0″W), the second female (HEV2) from a population in an intense urban area in Viamão (30°09′40.5″S; 50°55′01.5″W) and the third female (HES2) in a forest fragment located in São Francisco de Paula (29°26′34.1″S; 50°36′48.8″W).

Butterflies were kept individually in open-air insectaries with dimensions of 2.3 m × 3 m × 3 m (width, length, height) approximately. Insectaries had many plants for simulation of natural conditions, including P. suberosa, used by females for oviposition. The butterflies were fed daily with a mixture containing water, honey and pollen.

A total of 12 eggs were collected (five from HEAB2, five from HEV2 and two from HES2) with the assistance of a paintbrush. The eggs were transported to the laboratory, and caterpillars were grown individually in cylindrical plastic pots. Immatures were fed exclusively with P. suberosa leaves (Fig. 1). Fecal samples were collected from each caterpillar individually after 48 h of molting to the fifth instar (n = 12), with the aid of a disposable plastic spoon, stored in 1.5 mL microtubes and maintained at −80 °C until processing. The oviposition dates are shown in Table S1.

This study was carried out in accordance with the recommendations of Chico Mendes Institute for Biodiversity Conservation (ICMBio). The protocol was approved by Information Authorization System in Biodiversity (SISBIO) number 33404-1. This study has the Council for the Management of Genetic Patrimony - CGEN - under the Ministry of Environment number A720680.

Isolation and identification of Enterococci

Isolation and identification of enterococci were performed as previously described in Santestevan et al. (2015), with modifications. One milligram of fecal sample was transferred to 10 mL of saline 0.85% and incubated at 37 °C for 24 h. One mL was inoculated in nine mL of Azide Dextrose Broth selective medium (Himedia, Mumbai, India) and incubated for 24 h at 37 °C. Aliquots (one mL) were placed in nine mL of saline 0.85%, and initial samples were further diluted 10-fold. From dilution 10⁻⁵ and 10⁻⁶, 100 µL was inoculated in brain heart infusion (BHI) agar plates (Himedia, Mumbai, India) supplemented with 6.5% NaCl, incubated for 48 h at 37 °C.

Fifteen colonies were randomly selected from each fecal sample. Phenotypic criteria (size/volume, shape, color, gram staining, catalase production and bile aesculin reaction) were used to separate the enterococci group and the non-enterococcal strains. Selected pure colonies were stored at −20 °C in a 10% (w/v) solution of skim milk (Difco, Sparks, MD, USA) and 10% (v/v) glycerol (Neon Comercial Ltda, São Paulo, SP, BR).

Genomic DNA was extracted by physicochemical method as previously described by Depardieu, Perichon & Courvalin (2004). Polymerase chain reaction (PCR) assay was carried out using genus-specific primers targeting the tuf gene, which encodes elongation factor EF-Tu (Ke et al., 1999) (Table 1). Amplifications were performed in a total volume of 25 µL containing: 100 ng of template DNA, 1 X reaction buffer (Ludwig Biotechnology), 0.4 µM of each primer (Ludwig Biotechnology), 1.5 mM MgCl₂, 200 µM of dNTP (Ludwig Biotechnology), 1 U Taq DNA polymerase (Ludwig Biotechnology), and MilliQ water. Amplification was carried out in a conventional thermocycler (Applied Biosystems 2720 Thermal Cycler) according to the following program: initial denaturation at 95 °C for 3 min, followed by 35 cycles of 95 °C for 30 s, 54 °C for 30 s and 72 °C for 1 min, and a final extension at 72 °C for 7 min. The PCR products were visualized using electrophoresis on 1.5% (w/v) agarose gel, stained with SYBR^® Safe DNA Gel, and visualized on a photocumenter. E. faecalis ATCC 29212 was used as positive control.

Characterization of enterococci species

Isolates were screened with the species-specific PCR assay for Enterococcus faecalis, Enterococcus faecium, Enterococcus casseliflavus and Enterococcus mundtii (Cheng et al., 1997; Jackson, Fedorka-Cray & Barrett, 2004; Sedgley et al., 2005). The primers and annealing temperature used are listed in Table 1. Amplifications were prepared as described to tuf gene. PCR amplification was performed in the conventional thermocycler (Applied Biosystems 2720 Thermal Cycler) according to the following program: 94 °C for 5 min followed by 35 cycles of 94 °C for 1 min, appropriate annealing temperature for each species for 1 min, extension at 72 °C for 1 min, and a final extension at 72 °C for 5 min. The DNA fragment amplified by PCR was analyzed in 1.5% (w/v) agarose gels stained with SYBR^® Safe DNA Gel, and visualized on a photocumenter.

Strains that were not identified by PCR reactions were submitted to matrix-assisted laser desorption and ionization time-of-flight technique (MALDI-TOF) applied to Enterococcus sp., according to Sauget et al. (2017).

Isolates classified as Enterococcus sp. were identified by Sanger sequence analysis. The PCR product of 16S rRNA gene, using the 8F and R1522 primers (Coenye et al., 1999) (Table 1), was purified with Illustra™ GFX™ PCR DNA and gel band purification kit (GE Healthcare, Buckinghamshire, UK). To perform Sanger sequencing two additional primers, 519F and 926R, were used (Coenye et al., 1999) (Table 1). Sequencing was performed with the ABI PRISM^® BigDye^® Primer Cycle Sequencing Ready Reaction Kit in an ABI PRISM^® 3100 Genetic Analyzer (Applied Biosystems^®), according to the manufacturer’s protocol. The sequence obtained was compared to nucleotide sequences of reference enterococci strains deposited in GenBank.

Antimicrobial susceptibility testing

Antimicrobial susceptibilities were determined by Kirby-Bauer disk diffusion method recommended by the Clinical and Laboratory Standards Institute (CLSI, 2016). Eleven antibiotics used in clinical and veterinary medicine were tested: ampicillin 10 µg (AMP), vancomycin 30 µg (VAN), erythromycin 15 µg (ERY), tetracycline 30 µg (TET), ciprofloxacin 5 µg (CIP), norfloxacin 10 µg (NOR), nitrofurantoin 300 µg (NIT), rifampicin 5 µg (RIF), chloramphenicol 30 µg (CHL), gentamicin 120 µg (GEN) and streptomycin 300 µg (STR).

Intermediate and resistant strains were considered in a single category and classified as resistant. Strains showing resistance to three or more unrelated antibiotics were considered as multidrug-resistant (MDR) (Schwarz et al., 2010).

Detection of resistance and virulence genes

Erythromycin-resistant strains were tested by PCR for the presence of resistance encoding genes more commonly associated to clinical and environmental enterococci: erm (B), which encodes a ribosomal methylase that mediates MLSB resistance; and msr C, which encodes for a macrolide and streptogtamin B efflux pump. The presence of virulence associated genes gelE (gelatinase enzyme), cylA (activator of cytolysin), ace (accessory colonization factor), esp (associated to biofilm formation) and agg (aggregation substance) was determined by PCR in all enterococcal isolates. The amplifications were performed as described in Prichula et al. (2016). Amplifications were prepared as described to tuf gene. PCR amplification was performed in the conventional thermocycler (Applied Biosystems 2720 Thermal Cycler) according to the following program: 94 °C for 3 min followed by 35 cycles of 94 °C for 1 min, appropriate annealing temperature for each resistance or virulence gene for 1 min, extension at 72 °C for 1 min, followed by final extension at 72 °C for 5 min. The DNA fragment amplified by PCR was analyzed in 1.5% (w/v) agarose gels stained with SYBR^® Safe DNA Gel, and visualized on a photocumenter. The sequences of the primers and annealing temperature are described in Table 1.

Molecular typing of Enterococci by Pulsed-field gel electrophoresis (PFGE)

Eight six enterococci strains isolated from siblings and non-sibling caterpillars were selected for PFGE analysis according to the following criteria: maternal origin (females HEAB2, HEV2 or HES2), hatched larvae, enterococcal species and antimicrobial profile. Chromosomal DNA extraction and electrophoresis conditions were prepared according to Murray et al. (1990) and Saeedi et al. (2002). The restriction enzyme used was Sma I (Invitrogen^®). The electrophoresis was carried out using a clamped homogeneous electric field (CHEF-DRII device; Bio-Rad Laboratories, Richmond, Calif.), with ramped pulse times recommended by Saeedi et al. (2002) at 11 °C. Lambda Ladder PFG Marker (New England Biolabs) was used. The gels were stained with ethidium bromide (0.5 µg/mL for 20 min). The PFGE patterns were interpreted using the program GelCompar II v. 11 6.6, with 1.0% of tolerance, and the percentage of similarity was estimated using the Dice coefficient. The pulsotypes were clustered using the unweighted pair group whit arithmetic averages (UPGMA). A dendrogram was generated to examine the relatedness of PFGE patterns for selected isolates, and cutoff level of 80% applied to this dendrogram (Tenover et al., 1995).

Statistical Analysis

Simpson’s index of diversity (D) was calculated to assess the differentiation of enterococci species among the caterpillars from different maternal origins (Hunter & Gaston, 1988).

Enterococci species present in fecal samples of Heliconius erato phyllis caterpillars

A total of 178 strains of Enterococcus were isolated from fecal samples from fifth-instar caterpillars (Table 2). Enterococcus casseliflavus was the most common species identified (74.15%; n = 132), followed by E. mundtii (21.34%; n = 38) and E. faecalis (1.12%; n = 2). Six strains (3.37%) could not be identified to species level.

Differences in the composition of enterococci were detected between the three groups of caterpillars, as shown in Table 2. The Simpson’s diversity index was different between the three populations, with higher diversity of enterococci species from fecal samples of caterpillars from HES2 (1–D = 0.68), followed by HEV2 (1–D = 0.49) and HEAB2 (1–D = 0.27).

Antimicrobial susceptibility

One hundred and twenty (67.41%) enterococci were resistant to at least one evaluated antimicrobial agent. The frequency of antibiotic-susceptible strains is shown in Table 3. The rifampicin-resistance phenotype was the most commonly observed (56%; n = 100), followed by erythromycin (31%; n = 55). Eight strains (4%) were resistant to norfloxacin and five (3%) to ciprofloxacin. All investigated strains were susceptible to ampicillin, vancomycin, tetracycline, nitrofurantoin, chloramphenicol, gentamicin, and streptomycin.

Single- (SR), double- (DR), and multiple-drug resistance (MDR) were observed in 67% (n = 80), 28% (n = 34), and 5% (n = 6) of strains, respectively, of the 120 resistant strains.

Determinates of resistance and virulence

None of the 55 erythromycin-resistant strains was positive for erm(B) and msrC genes. The presence of virulence genes was evaluated in all strains, and the esp gene was detected in 35.39% (n = 63), ace in 6.74% (n = 12) and gelE in 1.12% (n = 2). No strain was positive for the clyA and agg genes.

Genetic relationships between enterococci isolated from sibling and non-sibling caterpillars in the fifth-instar

Of the 178 strains isolated, 86 (E. casseliflavus, n = 58; E. mundtii, n = 23; E. faecalis, n = 2; and Enterococcus sp., n = 3) were chosen for PFGE (Table S1). From the sibling caterpillars, numbered 6, 7, 10, 11 and 14, of the HEAB2 female were picked E. casseliflavus (n = 32), E. mundtii (n = 8) and E. faecalis (n = 2). All strains were isolated from caterpillars hatched closely in time and fed the same batch of P. suberosa leaves. From the offspring of the HEV2 female (sibling caterpillars 9, 18, 26, 27, and 29) E. casseliflavus (n = 19), E. mundtii (n = 9) and Enterococcus sp. (n = 3) were selected. These isolates were recovered from siblings that hatched at different times and were fed different batches of leaves. From the offspring of the HES2 female (sibling caterpillars 3 and 17) E. casseliflavus (n = 7) and E. mundtii (n = 6) were selected.

The hierarchical relationship between enterococci selected from sibling and non-sibling caterpillars showed 22 patterns (15 patterns and 7 single strain–[singleton]; Fig. 2). Four patterns generated by PFGE indicated a genetic relationship between strains isolated from sibling caterpillars (P7, P8, P9, and P13); 11 were composed of strains isolated from the same caterpillars (P1, P2, P3, P4, P5, P6, P10, P11, P12, P14, and P15). No genetic relationships were observed for strains isolated from non-siblings.

The band patterns for E. casseliflavus (n = 32) isolates from sibling caterpillars (6, 7, 10 11 and 14) of the HEAB2 female showed six PFGE patterns (P5, P7, P8, P9, P11 and P12) and three singletons. Three PFGE patterns (P7, P8, and P9) included 18 of the 32 strains that were isolated from sibling caterpillars 6, 7, 10, and/or 11, with low levels of genetic variability. P5 and P11 each contained two isolates, and P12 with eight isolates showed genetic variation; the remaining three E. casseliflavus isolates were singletons and represented unique PFGE patterns. All E. mundtii isolates from caterpillar 14 were genetically closely related and were clustered into one pattern (P10), as were the two E. faecalis (P3) isolates from caterpillar 6. These results demonstrate that strains may originate from a single lineage.

Seven different PFGE patterns (P4, P6, P13, P15, and three singletons) were obtained from sibling caterpillars 9, 18, 26, 27, and 29 of the HEV2 female. The analysis of the fragment profiles of the E. mundtii (n = 9) and Enterococcus sp. (n = 1) strains isolated from caterpillars 18, 26, 27, and 29 demonstrated a genetic relationship between them (P13). The 19 E. casseliflavus strains showed four PFGE patterns: P4, P6, and P15, each containing six, three, and nine isolates, respectively, and one singleton. These distinct patterns are suggestive of genetic events in these strains.

The seven E. casseliflavus strains from caterpillars 3 and 17 (offspring of the HES2 female) showed two distinct patterns (P1 and P2) with 100% of genetic similarity between them. P1 contained four isolates and P2 had three isolates. Of the six E. mundtii isolates from caterpillar 3, five showed 100% similarity and were clustered in the P14 pattern, suggesting that these strains may be progeny from a single lineage. One strain had distinct and unrelated PFGE by the criteria of Tenover et al. (1995). In addition, most of the patterns were shared by isolates with the same antimicrobial profile.