Introduction

Brucellosis is an infectious disease caused by bacteria of the Brucella genus transmitted from animals to humans [1]. The disease is transmitted from animals to humans via ingestion (unpasteurized milk or dairy products), inhalation, or exposure through the conjunctiva or skin abrasions [1, 2]. Brucellosis predominates in Mediterranean countries, the Middle East, and Latin America [3]. Based on the host preference and pathogenicity, twelve species are recognized within the genus Brucella: B. melitensis, B. abortus, B. suis, B. canis, B. ovis, B. neotomae. B. ceti, B. pinnipediae, B. microti, B. inopinata, B. papionis, and B. vulpis [1, 4]. Brucella melitensis is one of the most virulent species, followed by B. suis and B. abortus [5, 6]. Brucellosis continues to be a reemerging zoonosis worldwide and in Mexico. Although the Mexican government has implemented a campaign for brucellosis control and eradication since 1995, there was an increase in incidence from 2000 to 2011 [7]. The incidence of bovine brucellosis increased from 1 to 15%, and in the case of goats, it increased from 2 to 7% [7]. Mendez et al. reported that in this same period, increases in bovine and goat brucellosis contributed to 15% and 33% of the increase in the incidence of new cases of human brucellosis, respectively [7]. Since brucellosis is a zoonosis, its transmission to humans generally occurs from an animal reservoir [8]. In addition, the ecology of Brucella has been changing, and brucellosis surveillance requires new programs for monitoring unusual hosts and ecosystems. For example, in Egypt during 2010, B. melitensis was identified in Nile catfish (Malapterurus electricus) as a result the dumping of animal waste in the Nile River Delta [9]. These practices are common worldwide; therefore, it is essential to monitor the presence of this pathogen in exposed ecosystems due to the risk of contamination with animal wastes. The gold standard for the diagnosis of brucellosis remains the isolation of Brucella spp. bacteria from clinical, environmental, dairy, and derivative samples. However, PCR-based methods that identify nucleic acid fragments from the bacteria are more useful and practical. Therefore, the objective of this study was to determine the presence of Brucella spp. on tilapia (Oreochromis niloticus) and surface water from a volcanic lake by the application of traditional culture and multiplex PCR methods.

Materials and Methods

Bacteria Strains

In the bacteriological and molecular analysis, the reference strains Brucella suis 1330 (source ATCC 23444) [10] and Brucella abortus 2308, a pathogenic strain from a fetus cow (source Lab collection) [11], were used as controls.

Fish and Water Sampling

The sampling season was during 2016–2017. It started in May 2016 (spring) with 45 surface water (water recollected at ~ 15 cm deep) samples recollected from 15 specific points and 15 tilapia fish swab samples obtained across five different sites on the lake (Supplemental Fig. 1). The same number of samples mentioned above from both surface water and fish specimens were collected in August 2016 (summer), at the beginning of December 2016 (fall), and in February 2017 (winter). The specimens, which were adults of tilapia of commercial size (Oreochromis niloticus), were collected with the help of local fishermen using traditional fishing methods from San Pedro Lake at geographic coordinates of 21° 12′ 33″ north latitude and 104° 43′ 37″ west longitude (Supplemental Fig. 1) and an elevation of 1261 m above sea level within a catchment area of 2.96 km2 belonging to a priority hydrological region called the Crater Lakes of Nayarit, Mexico [12, 13]. All procedures performed in this study using tilapia were in accordance with the ethical standards of CONAPESCA (responsible for fishing in bodies of continental freshwater waters of the federal jurisdiction of the United Mexican States) and based on the Standard Mexican Regulation NOM-060-SAG/PESC-2014 [14]. Fishes were transported individually in sterile plastic bags, and surface water samples were recollected and transported to the laboratory at 4 °C and processed within the next 6 h after being collected, according to the American Public Health Association [15]. The geographical position of the sampling points was determined with a Global Positioning System (GPS, GARMIN® eTrex Legend® H), and HOBO Water Level Data Logger was used to monitor the water temperature (°C) of the lake.

Isolation

A volume of 100 mL of water was filtered in duplicate in a vacuum multiple filtration system (ADVANTEC); the pore size of the filter membrane was 0.22 µm. The first isolation grew in BRUCELLABUAP™ supplemented with Cycloheximide (100 μg/mL), Nystatin (100 U/mL), Nalidixic acid (5 U/mL), and Ethyl violet (0.1% aqueous solution) 1.25 mL/L. The above was modified from Laboratory techniques in brucellosis Manual [16]. Skin swabs were gently rubbed onto the fish surface with a dry sterile cotton-wool swab stick in a manner similar to that used for the filters recovered from each sample and were placed in the culture medium mentioned above, and surface extension was performed on a plate. They were incubated under 5% CO2 conditions at 37 °C, verifying bacterial growth at 48 and 72 h.

Bacteriological Analysis

BRUCELLABUAP™, without dyes or antibiotics, was used to grow colonies, which presented the typical morphology of Brucella genus bacteria. Brucella colonies on clear culture medium are small, transparent, raised, convex, with an entire edge and a smooth, shiny surface. They appear a pale honey color by transmitted light [17]. The bacteria were incubated for 48 to 72 h in 5% CO2 atmosphere at 37 °C, followed by Gram staining of colonies. Gram-negative colonies with coccobacillary morphology and color were subjected to biochemical testing in Simmons Citrate medium (SCM), Triple Sugar Iron Agar (TSI), Methyl Red–Vogues Proskauer (RM-VP), Lysine Iron Agar (LIA), Christensen Urea, Mobility, Indole, Ornithine (MIO), and catalase and oxidase tests. Subsequently, typing was performed with dyes in BRUCELLABUAP™ culture medium with thionine (1:25,000), thionine (1:50,000), safranine (1:10,000), and fuchsin (1:50,000) [18]. For the abovementioned dilutions, stock solutions were prepared at 0.5% for safranine and 0.2% for thionine and fuchsin (mass/volume). Bacterial suspensions were made in saline solution and adjusted to an optical density of 1.4 to 600 nm. The agar plates with the corresponding dilutions of the dyes were inoculated according to Laboratory techniques in brucellosis Manual [16]. As controls, we used the reference strains Brucella abortus strain 2308, and Brucella suis 1330. The plates were incubated at 35 °C with 5% CO2 and monitored at 48 and 72 h.

Bacterial Genomic DNA Isolation and PCR Assay

Genomic DNA extraction and purification were performed with a Quick DNA™ Fungal/Bacterial Miniprep Kit from Zymo Research. DNA integrity was confirmed in 1.5% agarose gel, while the purity was determined by spectrophotometer NanoDrop 2000, Thermo SCIENTIFICMR. The reaction mix for Multiplex PCR (mPCR) assay for the differentiation and identification of presumptive Brucella isolates from water and tilapia skin and reference strains consisted of GoTaq® Green Master Mix (2×) from Promega (Ref M7122) 12.5 μL and DNA template 1.5 μL (250 ng), with a final volume for each reaction of 25 μL with nuclease-free water based on Promega’s specifications. The primers contained in reaction one amplified fragments 774, 587, 450, 272, and 152 bp in size and were used at a concentration of 0.4 μM each. Reaction two contained the primers to amplify fragments 1682, 1071, and 215 bp in size, with a final concentration of 0.4 μM each.

The mPCR was run on a miniPCR™ thermocycler, and the cycle conditions used were one cycle for initial denaturation at 95ºC for 3 min, following by 30 cycles for denaturation at 95ºC for 1 min, annealing of the primers at 50  °C for 45 s and extension at 72  ºC for 3 min, and finally by 1 cycle for final extension at 72ºC for 6 min.

The primers used in this assay are shown in Table 1. The mPCR products were analyzed on 1.3% agarose gel. The amplified products were observed on a MultiDoc-It™ System Imaging System – UVP. The multiplex PCR profiles of the reference strains and sample problems were compared with the Bruce-ladder v2.0 multiplex PCR [19, 20]. In Table 2, we show the reported fragments present in each Brucella species. Additionally, we included a 215-bp fragment that recently has been suggested to be specific to Brucella abortus [21].

Table 1 Primers used in this study
Full size table
Table 2 Bruce-ladder v2.0 multiplex PCR profile
Full size table

Results

Water and fish samples were collected seasonally for one year and were processed to isolate and identify Brucella spp. Seventy-four suspect colonies were isolated by bacteriologic methods and dyes (Table 3). However, of those suspicious colonies, by a multiplex PCR assay. Two colonies isolated from one fish were confirmed to be B. abortus and B. suis and one colony from surface water sample as B. abortus.

Table 3 Colonies isolated and identified by bacteriological analysis
Full size table

The results of the biochemical capacities of the reference strains and Brucella isolates were generally similar and are presented in Table 4.

Table 4 Biochemical capacities for Brucella isolates, Brucella suis, and Brucella abortus 2308
Full size table

The results of the typing by dyes are shown in Table 5. In general, the presumptive Brucella colonies isolated on the BRUCELLABUAP™ by biochemical capacities display the biochemical profile of the Brucella genus. The PCR profile of B. abortus has been previously reported and is distinguished from the rest of the profiles by the absence of the 1071- and 272-bp fragments (Table 2) [31]. The Bruce-ladder v2.0 multiplex PCR of the colonies isolated is shown in Fig. 1. As a positive control, we used a reference strain B. abortus 2308, which has an identical profile to other B. abortus strains reported [19, 20], although in this work, the profile is shown in Fig. 1a and b. To confirm these results, we used specific primers that amplified a 215-bp fragment with Bru ab2 primers, which are not part of the original Bruce-ladder v2.0 multiplex PCR but are specific to B. abortus species, as reported by Alamian et al. (Fig. 1b, lines 2, 3, and 4) [21]. A skin sample from one tilapia fish captured in the fall showed presumptive colonies of Brucella spp., which were later confirmed to be two strain, B. suis and B. abortus afterwards. The multiplex PCR with the DNA from the primary isolate amplified all the fragments reported by the Bruce-ladder v2.0 multiplex PCR. However, the PCR profile did not correspond to any of the Brucella species [19, 20] (data not shown) because the DNA of the primary isolate was a mix of both species identified later. Therefore, single colonies were picked and struck to new BRUCELLABUAP™ to isolate and repeat the Bruce-ladder v2.0 multiplex PCR. The multiplex PCR profile from the isolated colonies showed colonies with specific profiles of B. abortus and B. suis, respectively (Fig. 1). It is worth mentioning that in the reaction mix for panel A (Fig. 1), the primers used were BMEII0987 (152pb), BR0953 (272 pb), BMEI0535–BMEI0536 (450 pb), BMEII0428 (587 pb), and BMEI1426–BMEI1427 (774 pb). In the reaction mix for panel B (Fig. 1), the primers used were Bru ab2 (215 pb), BMEII0843–BMEII0844 (1,071 pb), and BMEI0998–BMEI0997 (1,682 pb).

Table 5 Typing by sensitivity to dyes for bacteria of the genus Brucella
Full size table
Fig. 1
figure 1

Bruce-ladder v2.0 multiplex PCR agarose gel picture used to confirm the Brucella species isolated. a PCR profile showing fragments of 152, 272, 450, 587, and 774. Lane 1, molecular weight marker 100-bp DNA Ladder (PROMEGA), lane 2, Brucella isolate from surface water (B. abortus), lane 3, Brucella isolate from skin fish (B. abortus), lane 4, PCR profile from B. abortus 2308 (positive control). Lanes 5 and 6, Brucella isolate from skin fish (B. suis) and B. suis 1330 (ATCC 23,444) positive control, respectively, which is different from the other profiles above. b PCR profile showing fragments of 215, 1071, and 1682 bp. The 1682- and 1071-bp bands were weakly revealed in panel A) PCR; therefore, they were amplified in a separate PCR with a 215-bp fragment, which is specific for B. abortus. Lane 1, 100-bp molecular weight marker; lane 2, PCR profile of Brucella isolate from surface water (B. abortus); lane 3, Fish skin Brucella isolate (B. abortus); lane 4, PCR profile B. abortus 2308 (positive control); lanes 5, PCR profile isolated from fish skin Brucella (B. suis); and lane 6, B. suis 1330 (ATCC 23,444) positive control

Full size image

Discussion

Brucellosis remains as a major zoonosis worldwide, and according to the World Animal Health Information Database, Mexico had the largest number of reported outbreaks of brucellosis—5514 in 2014 [32]. Nayarit is a state of Mexico that is in the eradication phase of brucellosis. However, brucellosis cases are still presented in the State [33, 34]. San Pedro Lake belongs to a hydrological region called the Crater Lakes of Nayarit, Mexico [13].

One sample from surface water isolated in summer was positive for B. abortus, as confirmed by the Bruce-ladder v2.0 multiplex PCR (Fig. 1) and colonies isolated from fish skin; in fall, the PCR profile showed colonies with specific profiles of B. abortus and B. suis. To the best of our knowledge, this is the first report of two Brucella species isolated from the skin of tilapia (Oreochromis niloticus). Previously, B. melitensis species were isolated from 120 catfish captured from Nile canals and 120 farmed catfish from skin, spleen, liver, and kidney [9]. The authors suggested that Nile catfish are naturally infected by contaminated materials from infected ruminants that reach the water canals and that direct contact with infected catfish may also be a source of human brucellosis [9]. Some of the activities in the Nile River Delta region include agriculture and livestock [35], which are similar to those carried out in San Pedro Lake. At the north of Lake San Pedro, where we isolated the Brucella species, the mouth of the San Pedro stream and the San Pedro Lagunillas town are located and tourist and fishing activities are carried out. In the south of the lake, livestock activities such as grazing cattle and agriculture are carried out around the lake, similar to the activities of the Cuenca Oriental, Mexico. In a previous work, carried out in the crater lakes from the Cuenca Oriental, Mexico, we isolated Brucella spp. from internal organs from charales (Poblana) [36]. It has been reported that Brucella spp. can survive in tap water at – 4 °C for 114 days, 57 days at 8 °C and pH 6.5 in lake water [37], but in the presence of organic matter, such as cattle feces, soil, and lake water at 37 °C, they can survive for more than 2 years [38]. However, in lake water with a temperature of 37 °C and a pH 7.5, the bacterial survival is less than one day [37]. In the case of the volcanic lake of this study, the average water temperatures in summer and fall were 27.03 ± 0.46 °C and 20.04 ± 0.22 °C, respectively. Probably given the water and environmental conditions, it allowed Brucella to survive and to be isolate in this study. Although Brucella has been observed to survive for some time in open environments, the bacterium hardly divides and eventually dies [39]. However, there are no studies about Brucella species survival on fish skin, which could be a source of infection to humans. The isolated strains probably came from breeding animals common in the region (Supplemental Fig. 1), such as pigs or cattle that could be infected with Brucella abortus or B. suis, the latter of which is derived from the breeding and grazing of cattle that takes place on the periphery and the shore of the lake (Supplemental Fig. 2B and C). When the sampling was carried out, it was possible to observe that some cattle entered the lake (Supplemental Fig. 2A), and given the water temperature, the organic matter present in the volcanic lake and the rainfall, which carry animal wastes from pastureland and urban settings, the water may be the source of contamination. In addition, the San Pedro stream passes near cattle and pig raising pens and urban areas before reaching the volcanic lake (Supplemental Fig. 1), where pollutants such as fecal matter from domestic and human livestock can run off relatively easy and quickly into surface waters and contaminate the volcanic lake and the fish that live there.