Epidemiological, molecular characterization and risk factors of human brucellosis in Iran

Afshar Etemadi, Rezvan Moniri✉, Mahmood Saffari, Hossein Akbari, Saeed Alamian, Ali Mohammad Behrozikhah

1Department of Microbiology, Faculty of Medicine, Kashan University of Medical Sciences, Kashan, Iran

2Department of Biostatistics and Public Health, Faculty of Health, Kashan University of Medical Sciences, Kashan, Iran; Social Determinants of Health (SDH) Research Center, Kashan, Iran

3Brucellosis Department, Razi Vaccine and Serum Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Karaj, Iran

ABSTRACT Objective: To determine epidemiological, molecular characterization, and potential risk factors of human brucellosis.Methods: This descriptive study was carried out in the clinical setting in Iran between 2017 and 2018. A total of 297 participants enrolled in the study. The sample size was calculated based on the occurrence rate of brucellosis in different areas. Patients were assessed using serological tests and conventional culture methods. Phage and multiplex PCR methods typed all of Brucella isolates. Potential risk factors of disease were determined. Results: A total of 141 of 297 (47.5%) Brucella strains were isolated and all of them were detected as Brucella melitensis biovar 1. Based on serologic titers, high culture positivity was recorded at 1/640 titer (P＜0.006). The risk factors for brucellosis were patients older than 40 years (OR=2.23, 95%CI: 1.4-3.55, P=0.001), animal keeper (OR=7, 95%CI: 1.51-32.41, P=0.005), housewife (OR=8.76, 95%CI: 1.85-41.37, P=0.002), farmer (OR=6.42, 95%CI: 1.21-33.97, P=0.019), and contact with animal (OR=1.31, 95%CI: 0.60-2.85, P=0.005).Conclusions: To the best of our knowledge, this is the first comprehensive report from Iran presenting the detection of Brucella species by the multiplex PCR. Brucella melitensis biovar 1 is still the dominant causative agent in Iran. The consumption of unpasteurized dairy products, living in rural areas, and animal contact were risk factors of brucellosis.

KEYWORDS: Brucellosis; Epidemiology; Molecular detection; Risk factors

1. Introduction

Brucellosis is an important and widespread zoonosis which is transmitted to humans through direct exposure with infected materials or not directly by the eating of contaminated animal products and by inhalation of airborne agents[1]. This disease has high morbidity both for humans and animals, and it is a significant cause of economic loss and a public health risk in many developing countries[2]. Endemic areas for brucellosis comprise countries of the Mediterranean basin, Middle East, Central Asia, China, the Indian subcontinent, sub-Saharan Africa, and parts of Mexico and Central and South America. In Iran cases of human brucellosis are described yearly; most of them are because of Brucella (B.)melitensis. Unpasteurized dairy products, such as fresh cheese are a significant source of infection.Brucellosis is an endemic disease illustrated by a high infection rate in both animals and humans in Iran[3]. Presently, there is limited information existing on the molecular epidemiology of circulating strains. Human infection is related to factors and activities associated with exposure to shedding animals and contaminated food products. Brucellosis transmission to humans is controlled through behavior change and food safety interventions that target milk production. For prevention measures, more information on risk factors is required.The aim of this study is to describe the molecular epidemiology of Brucella strains isolated in high-risk areas in species level and evaluate the risk factors of human brucellosis.

2. Material and methods

2.1. Study areas

This cross-sectional study was performed in the clinical setting at the western, northwestern, northern, and central provinces of Iran from December 2017 to June 2018.

2.2. Sample size

Based on the recent serological-surveillance reports of brucellosis in Iran, provinces of Iran were categorized into five groups, including very high, high, moderate, low, and very low incidence areas[3]. The sample size was determined using a weighted average (weighted mean) formula as the prevalence rate of brucellosis in each province considered as its weight. Regarding the maximum incidence rate of 80-101 per 100 000 persons, minimum sample size at 5% level of precision, 95% confidence level, the calculated sample size was 138. One province selected from each province groups for sampling: Lorestan (very high incidence rate) located around 33̊ 29' 16" N and 48̊ 21' 21" E. Kermanshah (high incidence rate) found approximately 34̊ 18' 51" N and 47̊ 03' 54" E, East Azerbaijan (moderate incidence rate) located around 38̊ 04' N and 46̊ 18' E, Mazandaran (low incidence rate) found about 36̊ 33' 48" N and 53̊ 03' 36" E, and Isfahan (very low incidence rate) located around 32̊ 38' N and 51̊ 39' E. We have chosen 138 patients based on recent serological-surveillance brucellosis in Iran using a weighted average formula, considering the prevalence of brucellosis in each province. It distributed as follows in the five provinces: 105 (Lorestan), 80 (Kermanshah), 64 (Eastern Azerbaijan), 37 (Mazandaran), and 11(Isfahan) (Figure 1).

Figure 1. General map of Iran showing the five provinces included in the study. The selection performed was based on seroprevalence rate of brucellosis. Lorestan: very high (81-100 per 100 000 persons) (n=105); Kermanshah: high (61-80 per 100 000 persons)(n= 80); East Azerbaijan: moderate (41-60 per 100 000 persons)(n=64); Mazandaran: low (21-40 per 100 000 persons)(n=37); Isfahan: very low (1-20 per 100 000 persons)(n=11).

2.3. Research ethics

Written informed consent was obtained from all adult contributors and the parents or legal guardians of minors. Kashan University of Medical Sciences Research Ethics Committee approved this study (IR.Kaums.REC.1395.156).

2.4. Inclusion criteria

The study contributors were patients of two years old and above with clinical doubt of brucellosis described by a stated history of fever and one or more of the following symptoms including chills, night sweats, headache, weight loss, fatigue, myalgia, arthralgia, malaise, weakness and arthritis. The identification of brucellosis was known according to the isolation of Brucella spp. in blood culture or other clinical samples. Noteworthy titers were Wright’s agglutination titer of ≥1/160.

2.5. Questionnaire survey

Designed forms were used to gather data on socio-demographics (e.g., gender, age, education, place of living and job), epidemiology (e.g., ingesting of dairy foods and contacting with animals), as well as the clinical manifestation. The contents of the questionnaire selected based on other studies[3,4].

2.6. Serological tests

The Rose Bengal test for screening and the qualitative assay was used as a recommended procedure[4]. Also, the agglutination tube test was used as a quantitative assay for determining antibody titer. These assays were conducted based on the standard protocol.

2.7. Bacterial culture and phage typing

Ten mL blood samples were inoculated in brain heart infusion broth medium(Merck, Germany). The blood culture bottles were incubated at 37 ℃ for 21 d and a longer time. During the incubation period, samples of each bottle were weekly sub-cultured on selective Brucella Agar plates containing selected antifungal and antibacterial agents (Thermo Scientific™ Oxoid™). The grown strains on selective media with suspected morphological characteristics of Brucella were characterized by traditional culture and phage-based method at species and biovar levels according to the standard pressure. Lysis reaction on strains was determined using Tb phage at two dilutions, routine test dilution (RTD) and RTD×104. Complete lysis was considered a positive reaction after 48-hour incubation at 37 ℃. Subsequently, A and M epitopes and rough phases of colonies were evaluated by A, M, and R mono-specific sera using slide agglutination test. Brucella (B.) abortus biovar 1 strain 544 (ATCC No. 23448), B. melitensis biovar1 strain 16M (ATCC No. 23456), Brucella (B.) suis biovar 1 strain 1330 (ATCC No. 23444), B. abortus Rb51 (vaccine), and B. melitenis Rev.1 (vaccine) reference strains were used as control.

2.8. DNA extraction

Genomic DNA was obtained from the High Pure PCR Template Preparation Kit (Roche Diagnostics GmbHCo, Germany). Briefly, 200 µL of bacterial suspension was centrifuged at 3 000 × g for 5 min, then the supernatant was discarded, and the bacterial pellet was suspended in 200 PBS buffer. After that, 5 µL lysozyme (10 mg/ mL in 10 mM Tris-HCl, pH 8.0) was added and incubated for 15 min at 37 ℃, then, 200 µL binding buffer and 40 µL reconstituted proteinase K was added followed by incubating for 10 min at 70 ℃ .

A total of 100 µL isopropanol was increased and mixed well. After two washing steps, extracted DNA was eluted in TE solution (Tris 10 mM, EDTA 1mM, pH 8.0). The quality and quantity of DNA were measured using NanoDrop Spectrophotometer (NanoDrop®ND-1000 Technologies Inc., USA). It was then stored at -20 ℃ for further analysis.

2.9. Multiplex PCR assay

The multiplex PCR assay was done by nine primer sets, as described previously (Table 1)[5]. Briefly, each PCR reaction consisted of 25 µL 2× QIAGEN Multiplex PCR Master Mix (Qiagen, Hilden, Germany), 5 µL of 10× primer mix (100 pmol/µL), 2 µL template DNA (25 ng/µL) and up to 50 µL RNase-free water.

The reactions were performed in GenAmp Eppendorf Thermal cycler (Germany). The running conditions were: 15 min for enzyme activation at 95 ℃ followed by 35 cycles of 30 s at 94 ℃, 2 min at 60 ℃, and 1 min at 72 ℃, and a final extension step of 5 min at 72 ℃. Amplified products were visualized by electrophoresis using 1.8% agarose gel and stained with ethidium bromide.

Table 1. Primer sets used for differentiation of Brucella strains in this study.

2.10. Statistical analysis

Results from questionnaires and serological tests were stored and analyzed using SPSS version 16. Chi-square and Fisher’s exact tests were figured to calculate the degree of association of the risk factors with brucellosis. P＜0.05 was measured statistically significant. Odd ratios with 95% confidence intervals were calculated for determining the severity of association of all independent variables with culture test results. Multiple logistic regression models were applied to assess the effect of different factors on culture test results. In this model, the backward method was used and P＞0.1 as excluding criteria was considered.

3. Results

An overall of 297 contributors participated in the study. The mean age of the patients was (41.4±17.4) years (range, 2 to 90 years). Of these participants, 164 (55.2%) were males, and 133 (44.8%) were females. A total of 268 out of 297 (90.2%) had contact with sheep and goats, 286 (96.3%) ingesting unpasteurized milk or soft cheese. The prevalence was higher (47.67, 133/297) in rural than urban (44.44%, 8/18) residents, but the difference was not significant.

3.1. Risk factors results

Table 2 presents the risk factors recognized. Older than 40 years (OR: 2.23, 95%CI: 1.40, 3.55), animal keeper (OR: 7, 95%CI: 1.51, 32.41), housewife (OR: 8.76, 95%CI: 1.85, 41.37), farmer (OR: 6.42, 95%CI: 1.21, 33.97), and contact with sheep and goats (OR: 1.31, 95%CI: 0.60, 2.85) increased the risk forbrucellosis. Multiple regression analysis revealed that variables including gender (male), education (under diploma), occupation (animal keepers, housewife), and animal contact were effective on culture test results (Table 3).

3.2. Clinical features results

Common clinical features included fever (83.84%), sweats (83.50%), headache (77.78%), arthralgia (61.25%), backache (85.86%) and myalgia (60.2%). Antibody titers obtained in standard tube agglutination test ranged from 160 IU/mL to 5 120 IU/mL.

3.3. Laboratory finding

Our finding showed that the rate of brucellosis seropositivity was high in male participants (55.21%) is compared with the female (44.78%), as well as within the age group of 40 years old than other age groups. 141 out of 297 (47.5%) blood samples yielded Brucella in culture. Our results showed that the minimum time for detection by culture was three days, and in some cases, it was as long as seventy days. All samples positive cultures were positive by PCR too. All cases had antibodies levels of 160 IU/mL cut-off value in Brucella standard tube agglutination test, 48 (43.2%) patients having antibodies titers of 320 IU/mLwere negative by culture (Table 4). The results of the blood culture and phage typing system revealed that all isolated strains characterized as Brucella melitensis biovar 1. Based on serologic titers, high culture positivity recorded at ≥1/640 titer (P＜0.006). The results of the laboratory test of 297 participants in this study described in Table 4.

Table 2. Potential risk factors for human brucellosis on the basis of univariate analysis.

Table 3. Multiple logistic regression model coefficients of factors affecting the results of culture test.

Table 4. Results of the laboratory test of 297 participants in this study [n(%)].

3.4. Multiplex PCR results

Our findings indicated that all 141 positive cultures, the detected Brucella species were B. melitensis by multiplex PCR. The ability of the multiplex PCR to detect Brucella species and vaccine strains was validated using B. abortus strain 544, B. melitensis strain 16M, B. suis strain 1330, B. abortus strain Rb51 and B. melitensis strain Rev.1 (Figure 2).

Figure 2. Multiplex PCR assay for typing of Brucella isolates. Lanes 1 and 15, mid rang DNA ladder (Invitrogen); Lane 2 Brucella abortus 544; Lane 3 Brucella abortus Rb51; Lane 4 Brucella suis 1330; Lane 5 Brucella melitensis Rev.1; Lane 6 Brucella melitensis 16M; Lane 7-14 isolates of Brucella melitensis.

4. Discussion

Results from our study showed that less than half percent of the patients who had been serologically diagnosed were culture positive for Brucella. Laboratory findings for brucellosis diagnosis interacted with contact history, job, clinical manifestations, and record of past infection[6,7]. Serological techniques, and various molecular approaches are used in identification of brucellosis[8-10]. The interpretation of serological tests is more complex, chiefly in the setting of chronic infections, re-infections, relapses, and in areas that brucellosis is endemic where the brucellosis antibodies have been found in a high percentage of the people[11]. PCR performed on blood or any body tissue samples[12]. However, using PCR-based tests due to the need for standardization of the procedures cannot be supposed usual diagnostic techniques[13]. Identification of the genus and species of Brucella is achievable by several molecular tests[14,15]. Polymerase chain reaction-fragment length polymorphism (PCRRFLP) is considered as an appropriate tool for the detection of Brucella spp. at the species and biovar level in epidemiological studies[16]. Brucella culture positive in other studies differ from 10%-90% according to sample type, phase of infection, previous usage of antibiotics, and by slow-growing character of Brucella[17]. In this survey, the results of culture for brucellosis suspicious cases were 141 out of 297 (47.5%) positive, which is higher than the previous report of 23.4% in Jordan[18] but was lower than the results of 74.1% positive reported from Kuwait[19]. Moreover, the most blood cultures will be positive at the 7th and 21st day[20]. Our results showed the minimum time of detection was three days, and in some cases, it was as long as seventy days. This study showed that 70.3% of patients with negative results in cultures had positive serology 1/160. Our results showed that a major of culture-positive individuals (61.8%) were in the titer group of 1/640. There was a close relation between the serology titer and the positive culture findings (P＜0.006). Identification of species and biotyping of Brucella isolates play role as a helpful epidemiologic screening tool for surveillance studies of outbreaks and importation of Brucella strains in endemic territories[21,22]. All isolates from positive cases belonged to B. melitensis biovar 1, suggesting it as obviously the most common species/biovar implicated in the human disease in Iran. B. abortus biovar 3 is most isolated from Turkey, Egypt, India, Italy, and Africa, but in Iran, it is isolated just from domestic animals[22]. B. melitensis biovar 3 has been reported in the most human brucellosis cases in China[23]. B. melitensis biovar 3 showed the most frequently isolated biotype of Brucella from both animals and humans in Egypt[24]. Similarly, B. melitensis biovar 3 have revealed the causal agent in several epidemiologic studies in Turkey[25,26]. Zoghiet et al. reported B. melitensis biovar 1 as main agent for the disease in Isfahan, Khorasan, Guilan, Khuzestan, Yazd, and Kermanshah provinces in Iran. Still, in Tehran and Azerbaijan, biovars 1, 2, and 3 were the dominant ones[22]. The virulence of B. melitensis and B. suis for humans are more than B. abortus and/or Brucella canis[27-29]. At the present time the host preference and differential virulence between B. melitensis biovars have slight evidences.

Therefore, a supplementary investigation on B. melitensis isolates is needed to elucidate at the biovar level. Several molecular tests are available to clarify the genus and species of Brucella. For example, using 16S rRNA gene sequencing works to identify at genus level, but it is not useful at the species level[30]. Resolution to species level is possible through ribotyping, amplified fragment length polymorphism analysis, DNA sequencing of omp2, and omp25, as well as PCR assays targeting species-specific insertions of IS711 or IS650 elements[31]. The multiplex PCR assays that have been described by several authors can identify Brucella important species, including B. melitensis, B. abortus, and B. suis, simultaneously. Hence, various primer sets targeting bcsp31, omp2b, omp2a, and omp31 have been described to identify the major causative agents of human brucellosis in a single PCR reaction[32-34]. Our data showed that the multiplex PCR approach could discriminate Brucella species and vaccine strains. The multiplex PCR approaches could be used to the eradication in resource-limited laboratories of developing and underdeveloped countries[35]. Depending on the disease phase, clinical manifestations of infections with B. melitensis are variable[36]. Our findings revealed that fever, backache, and sweating were the most prominent symptoms in patients. Fever, arthralgia, and malaise were the most complaints of pediatric brucellosis in Iran[37]. Several risk factors have described for human brucellosis[38,39]. Univariate analysis showed that history of using unpasteurized dairy products, living in rural area, and contact with animals were major risk factors of brucellosis. These findings are agreed with other studies from Iran, Palestine, Lebanon, and Saudi-Arabia[40-42]. We found evidence of a relationship between brucellosis and sheep and goat contact. This is in agreement with other studies that have concerned cattle as asignificant reservoir and it shows a possible role for them in the epidemiology of brucellosis in Iran. Our results that brucellosis is related with animal contact and with using unpasteurized dairy products may suggest that B. melitensis is circulating in Iran.

Performing multiplex PCR only on specimens with Brucella antibody titers ≥160 IU/mL, which determined by standard tube agglutination test, instead of all 297 collected samples due to lack of resources, was one of the restrictions of this study. We recommended that species-level identifications could be done directly from blood samples with a high meticulous diagnostic test through direct multiplex PCR method.

To our information, this is the first inclusive report from Iran presenting the identification of Brucella species by the multiplex PCR. Although the identification of Brucella strains remains a considerable laboratory challenge, particularly in the developing and endemic countries, the multiplex PCR approaches are considered safe and rapid diagnostic methods for Brucella species. B. melitensis biovar 1 was dominant in Lorestan, Kermanshah, Eastern Azerbaijan, Mazandaran, and Isfahan of Iran. The use of unpasteurized dairy foods, living in rural places, and close contact with animals were the significant risk factors of human brucellosis in Iran.

Conflict of interest statement

The authors declared there are no conflicts of interest.

Funding

Kashan University of Medical Sciences supported this research (grant No. 95152).

Authors’contributions

RM, AE, MS conceived and designed the research proposal. AE, AS, BAH performed sampling and experiments. HA performed the data analysis. RM and AE contributed to the final version of the manuscript.