Bacterial isolates

In total, 5,784 Brucella isolates from confirmed cases of animal and human brucellosis in Italy from 2007 to 2015 were included in the study. Samples were collected from organs and tissues of livestock slaughtered in the framework of the national Brucellosis Eradication Plan or wild animals found dead and submitted for necropsy by competent authorities. Animal welfare during slaughtering procedures was ensured by veterinary services as required by legislation [24].

All strains were isolated by the local Italian Istituti Zooprofilattici Sperimentali (State Veterinary Laboratories) at Brucella spp. level and then sent to the Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise ‘G. Caporale,’ Teramo, Italy [National and World Organisation for Animal Health Manual (OIE) Reference Centre for Brucellosis] for species and biovar typing, according to the rules stated in the Ministerial Order of 14 November 2006. Specimens were transported and delivered in accordance with the World Health Organization (WHO) safety guidelines [25] and the IATA—Infectious Substances Shipping Guidelines—WHO–“Guidance on Regulations for the Transport of Infectious Substances 2015–2016” [26, 27]. The collection of data such as animal species, region of origin, and geographic coordinates was standardized using a form available on the National Brucellosis Reference Centre website (www.izs.it). The Brucella polyvalent and monospecific Brucella A and M antisera were supplied by the Food and Agriculture Organization/WHO Collaborating Centre for Research on Brucellosis (Veterinary Laboratories Agency, Weybridge, UK). All Brucella field isolates were subcultured in Brucella agar base and stored using the Microbank system (Pro-Lab Diagnostics, Toronto, Ontario, Canada) at −80°C.

Isolation procedures

According to the technique described in the OIE Manual of Diagnostic Tests and Vaccines [28, 29], the primary isolation of Brucella was performed by culturing the samples in Brucella broth supplemented with Farrell’s mix of antibiotics [30] and on Brucella agar (Oxoid, Basingstoke, Hampshire, UK) supplemented with 5% horse serum and antibiotics at the following amounts per 1 L of media: bacitracin (25 000 IU), polymyxin B (5000 IU), natamycin (50 mg), nalidixic acid (5 mg), nystatin (100 000 IU), and vancomycin (20 mg). The broth was incubated at 37°C ± 2°C in an atmosphere supplemented with 5% to 10% CO₂ (v/v) for up to 6 weeks. From the broth, two plates per sample were inoculated each week: one plate was incubated in aerobiotic conditions at 37°C ± 2°C and the other in an atmosphere supplemented with 5 to 10% CO₂ (v/v) at 37°C ± 2°C. The plates were observed after 3 days and then daily to identify Brucella-like colonies. The plates were discarded if no specific growth was evident after 7 to 10 days of incubation. Suspected colonies were subcultured onto serum dextrose agar from which subsequent growth was examined microscopically using Gram stain and biochemical (urease, oxidase, and catalase) and motility tests.

Identification methods

Identification were performed with AMOS (abortus, melitensis, ovis, suis) polymerase chain reaction (PCR) (i.e., AMOS-PCR) and PCR-restriction fragment length polymorphism techniques (PCR-RFLP). AMOS-PCR is a multiplex PCR designed to detect four species of Brucella [31]. The PCR Master Mix by Promega (Madison, WI, USA) was used. The assay exploits the polymorphism arising from species-specific localization of the insertion sequence IS711 in the Brucella chromosome. Individual biovars within a species are not differentiated [32]. Amplification was performed for 33 cycles in a thermal cycler (GeneAmp PCR System 9700; PE Applied Biosystems, Waltham, MA, USA) at an annealing temperature of 60°C. Amplicons were checked by fluorescence after electrophoresis in a 1% agarose gel with ethidium bromide.

Three different PCRs were used to amplify three outer membrane protein genes of Brucella: omp2a, omp2b, and omp31 (Fig 1). The amplicons of the omp2a, omp2b, and omp31 genes were digested by endonucleases (Pst I, Hinf I, Taq I, Ava II, and Nco I), and the products of digestion were checked by fluorescence after electrophoresis in 3% agarose gel in the presence of ethidium bromide. The specific biovar pattern was obtained by crossing the results of the single omp restrictions [33, 34, 35, 36].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Brucella AMOS-PCR profiles: In order to identify Brucella species and biovars, for each strain, 4 PCR have been performed: AMOS multiplex PCR, Omp 2a PCR, Omp 2b PCR and Omp 31 PCR.

In this image, this four PCR have been grouped to obtain a good glance and easily identify the Brucella profile. Lane 1 and 22 Marker molecular weights; Lane 2–5: Amos (2)-Omp2a (3)-Omp2b (4)-Omp31 (5) B. abortus biovar 3,5,6,9; Lane 6–9: Amos (6)-Omp2a (7)-Omp2b (8)-Omp31 (9) B. suis biovar 2,3,4,5; Lane 10–13: Amos (10)-Omp2a (11)-Omp2b (12)-Omp31 (13) B. melitensis biovar 1,2,3; Lane 14–17: Amos (14)-Omp2a (15)-Omp2b (16)-Omp31 (17) B. abortus biovar 1,2,4; Lane 18–21: Amos (18)-Omp2a (19)-Omp2b (20)-Omp31 (21) B. ovis.

https://doi.org/10.1371/journal.pone.0213689.g001

For B.abortus and B.suis, to complete the biovars differentiation, addictional tests were performed agglutination with anti-A, anti-M, and anti-R monospecific sera; the production of H₂S; CO₂-dependence; and growth in the presence of basic fuchsin and thionin at a final concentration of 20 μg/ml.

Cattle and water buffalo

Previous studies have isolated of six of the eight known B. abortus biovars in Italy, namely biovars 1, 2, 3, 4, 6, and 7 [46]. However, from 2007 to 2015, only biovars 1, 3, and 6 were isolated. The absence of isolation of biovars 2, 4, and 7 may suggest that these biovars have been eradicated from the country and that these strains may currently be considered exotic. In line with the distribution of the water buffalo population in Italy, most strains were isolated in Campania, the Italian region where 74.2% of the national stock of this species is farmed (Table 2) (286,946 of 386,731 total heads in Italy as of 31 December 2016) (http://statistiche.izs.it/portal/page?_pageid=73,12918&_dad=portal&_schema=PORTAL, accessed on 31 January 2017). Several B. melitensis isolates were also recorded in the present study, both in cattle and water buffalo (Figs 1 and 2). The percentage of B. melitensis isolates among the total number of strains submitted for typing was 9.9% in cattle and 4.1% in water buffalo (Table 2). The number of isolates was not very high, and it dropped significantly compared with previous years. However, B. melitensis can be shed in milk by infected cows, thus constituting a potential hazard for milk and milk product consumers. Infection among farm workers, butchers, and veterinarians may also occur as an occupational disease while handling infected animals or organs after slaughter [47]. Moreover, the occurrence of B. melitensis infection in cattle is of particular concern, given that B. abortus vaccines do not effectively protect against B. melitensis infection. In the past, cattle were the major source of human infection in most countries, and programs to eradicate the disease have been aimed largely at bovine brucellosis. Success has been achieved in northern and eastern European countries, Australia and New Zealand, Japan, Canada, and the US. Cattle are also the source of human brucellosis in most African countries, where large numbers of cattle are maintained and drinking raw milk is a custom. In countries with near universal pasteurization of milk, brucellosis has become an occupational disease and it remains a serious zoonosis for general population in the areas of the world where B. melitensis is endemic in sheep and goats [48].

Sheep and goats

Brucella abortus strains were also isolated in these species, although seldom reported in the past, and the possibility of shedding this strain in milk has been documented at least for sheep [49]. Nevertheless, risk factors including husbandry practices and exposure potential should be evaluated to determine the need to test sheep and goats that may have been exposed to cattle infected with B. abortus. The same risk factor evaluation should be applied to cattle exposed to sheep and goats infected with B. melitensis. Brucella ovis was isolated in sheep in the Piedmont, Abruzzi, Sicily, and Lazio regions (Fig 4). This suggests a widespread presence of the infection in the Italian sheep population; however, the distribution of this strain in sheep populations should be more thoroughly assessed. Actually, reliable information about the distribution of this strain is scarce; this strain has never been actively investigated because is not considered pathogenic for humans. Previous studies in Italy regarding the impact of animal brucellosis on humans have suggested an overlap between the distribution of disease in humans and that in cattle and ovicaprine populations. However, from 1970 to date, B. melitensis has been the pathogen isolated most frequently in human cases, accounting for more than 99% of Brucella spp. isolated from humans. Therefore, the brucellosis problem in Italy seems to be focused more on infection in the ovicaprine than cattle population.

Wild boars and other species

Brucella suis biovar 2 is the main strain responsible for brucellosis in wild boars in Europe [50]. This was confirmed by the results of the present study, in which B. suis biovar 2 was the main strain isolated; this is also in agreement with previous Italians records [51, 52]. However, a small percentage (1.7%) of B. melitensis biovar 3 was isolated from wild boar samples (Table 2) collected during monitoring activities carried out in a regional Park of Piedmont called “La Mandria,” a natural reserve from the 16th century that was used as a hunting reserve of the Savoia court. The occurrence of B. melitensis infection in wild boar in that area might have been a consequence of transmission between the wild boar population and wild ruminants, allowed by strict contact in a closed environment. Brucella suis biovar 2 was the only strain isolated from pigs, as has been reported in the past [53]. However, the presence of this strain in Italian pig farms may be largely underestimated because no specific surveillance plan has been implemented in the country. Most human infections derived from swine are caused by B. suis biovars 1 and 3. These biovars are most prevalent in Latin America, Southern Asia, China, and Oceania. Brucella suis biovar 2 is largely restricted to continental Europe and is maintained in the wild hare populations in an area extending eastward from the Atlantic Coast to the Ural Mountains and southward from the shores of the Baltic Sea to the Mediterranean. The agent is transmitted sporadically to domesticated pigs and these, together with infected wild boars and hares, are a potential source of human infection. However, this biovar seems to possess a low virulence for man, and few verified cases of human brucellosis caused by it have been recorded.

A strain of B. abortus biovar 1 was isolated in a horse located in the Basilicata region; this biovar was cultured by pathologic material collected from the supraspinous bursa. However, other species are farmed in the same environment, particularly cattle, the natural host of B. abortus biovar 1. Sporadic cases of horses infected with B. abortus have been reported. In some cases, infection may remain asymptomatic, but in other cases the infection may be associated with a variety of clinical manifestations, including osteoarthritis and osteomyelitis, abortion, and infertility. Naturally acquired B. abortus infection in horses commonly manifests as chronic bursal enlargement of the neck and withers or navicular bursa, referred to as fistulous withers or poll evil, respectively [54, 55, 56, 57, 58, 59]. Even if it is likely that abortions in mares due to B. abortus may pose a risk for transmission to cattle (and therefore a risk of human infection), documentation of this occurrence is lacking [60].

Infection by B. ceti is common in cetaceans, but only small proportions of infected cetaceans display pathological signs associated with brucellosis, excluding the striped dolphin (Stenella coeruleoalba). This may suggest that many infected cetaceans overcome infection, perhaps remaining as carriers and potential Brucella shedders [61]. The record of seven isolates of B. ceti confirms the presence of the bacterium in the Tyrrhenian and Adriatic Seas. Although Brucella strains from land animals have not been identified in cetaceans, B. ceti strains have been isolated from human cases, stressing the zoonotic potential of this strain. However, despite the few human cases attributed to Brucella isolated from marine mammals, the magnitude of the risk that B. ceti represents for humans remains unknown.

All isolates from human samples were B. melitensis biovar 3, and all were identified in hospitals from the Lombardia, Piedmont, Campania, Emilia Romagna, Lazio, and Puglia regions. This is consistent with previous findings where, in Italy, brucellosis caused by B. melitensis was the most important clinically apparent disease recorded in humans. However, all human cases reported in the present study were notified in hospitals. Actually, because the Italian legislation requires human cases of brucellosis to be reported by the hospitals at which the cases are investigated, the site of disease notification is often different from the site at which the patient acquired the infection. In our cases, strains were received by some of hospitals for supporting the isolation or the typing process, and this may also explain the lower rate of isolation in humans than in animals.