High prevalence of Campylobacter jejuni CC21 and CC257 clonal complexes in children with gastroenteritis in Tehran, Iran

Based on Cochran formula for calculating of sample size, a total of 3000 gastroenteritis cases were examined for suspected cases of sporadic campylobacteriosis. Gastroenteritis was characterized as abdominal pain with ≥3 episodes per day. Children with underlying gastrointestinal disease, physiologic diarrhea or history of antibiotic intake were excluded, which were defined by physicians in hospitals. Suspected cases of bacterial gastroenteritis were subjected to specimen collection, among which 280 cases accompanied WBC (white blood cell) and RBC (red blood cell) shedding in the majority of cases and were considered as suspected cases of campylobacteriosis.

Fecal specimens were collected from children with gastroenteritis, aged 0–5 years, referred to 3 Children’s Medical Center and Hospitals at Tehran, Iran, from May to October 2018. Information on age, clinical symptoms, history of non-pasteurized dairy products consumption, animal contact as well as laboratory results was recorded. Then specimens were transferred from laboratory of hospital to the laboratory of Tarbiat Modares University using Carry-Blair Transport Media (Micro Media-Hungary) and immediately streaked on Brucella agar and modified charcoal-cefoperazone-deoxycholate agar (mCCDA) (Merck-Germany). Plates were incubated at 42 °C for 48 h under microaerophilic condition using Gas Pack C (Merck-Germany). Gram staining, spiral morphology, catalase and oxidase production, nitrate reduction and indoxyl acetate hydrolysis test were used to confirm C. jejuni colonies. Also, hippurate hydrolysis test was used for phenotypic distinguishing of C. jejuni from C. coli from enteritis patients. Eventually, twenty C. jejuni and three C. coli strains were confirmed by Duplex PCR [13] and in the following, twenty C. jejuni was studied.

Twenty-three Campylobacter isolates were identified from 280 stool samples among which 20 and 3 isolates were C. jejuni and C. coli, respectively. All confirmed C. jejuni strains (n = 20) were designated and used for further analyses.

The capsular genotypes of C. jejuni strains were identified using specific primers for most commonly found genotypes HS1, HS2, HS4c, HS19, HS23/36c and HS41 (Table 1) [4, 5].

From 19 different LOS locus classes from A to S, 3 classes (A, B and C) play a key role in the biosynthesis of the sialic acid and are often found in isolates from the stools of patients with GBS [2, 6, 8]. Therefore, sets of primers specific for class A, B, and C classes were used for characterization of LOS identities (Table 2) [6].

Genomic DNA was extracted using a genomic DNA extraction kit (GeNetBio, Korea), according to the manufacturer instructions. PCRs was carried out within a thermal cycler (Eppendorf, Germany) in a final volume of 25 μL containing 1–10 ng DNA template, 2.5 μL 10X PCR buffer, 1 unit of Taq DNA polymerase, 2.0 mM MgCl₂, 0.2 μM of each primer, 0.3 mM each dNTP and sterile deionized water. Amplification conditions was as follow: 95 °C for 5 min, followed by 30 amplification cycles including denaturation at 94 °C for 1 min, annealing at 52 °C for 1 min and extension at 72 °C for 1 min. The reaction was ended with a final extension at 72 °C for 5 min and followed by electrophoresis of amplicons on 1% agarose gel.

Microbial DnaK is a bacterial conserved chaperone protein which has a sequence homology with human peripheral nerve HSP70 and its high level expression is supposed to be related to GBS promotion [14, 15].

RNA extraction was performed on 20 C. jejuni strains, which were previously checked for the presence of capsular types and LOS locus classes using a Favoren Biotec Corp kit (Taiwan). Subsequently, the RNA molecules were treated using the DNase I kit (TaKaRa). A cDNA synthesis kit (Yekta Tajhiz Azma-Iran) was used to generate a single-strand cDNA. The cDNAs were kept at − 20 °C. Quantitative Real Time-PCR was performed using SYBR Green (RealQ Plus Master Mix Green-Denmark) in Qiagen-Rotor-Gene Q with HRM. 16S rRNA gene was used as the internal control. One of the isolates that neither had the selected capsular serotypes nor the LOS locus classes was considered as a reference gene. The PCR reaction mixture consisted of 100 ng to1 mg of cDNA (for dnaK and 16S rRNA genes), 1 mM of each primer (Table 3) [16] and 12.5 mL of SYBR Green I Master Mix. Cycling conditions included an initial denaturing step of 10 min at 95 °C followed by 40 cycles of 15 s at 95 °C and 1 min at 60 °C. 2^-∆∆Ct is a relative quantification method for analyze the relative changes in gene expression from real-time quantitative PCR [16]. SPSS software version 20 was used for the analysis of data.

The MLST method was performed according to Dingle, et al. [17]. Each 25 μL amplification reaction mixture comprised 1–10 ng DNA template, 1X PCR buffer, 1.25 unit of Taq DNA polymerase, 1.5 mM MgCl₂, 1 μM of each primer (Table 4), 0.8 mM each dNTP and sterile deionized water.

The PCR conditions were as follows: denaturation at 94 °C for 2 min; annealing at 50 °C for 1 min, extension at 72 °C for 1 min for 35 cycles. DNA Sequences of each housekeeping gene were submitted to C. jejuni MLST database (http://​pubmlst.​org/​campylobacter) and the related allelic numbers, Sequence Types (ST) and Clonal Complexes (CC) were identified [18]. The Accession Numbers of DNA Sequences of 20 C. jejuni isolates were deposited in GenBank (https://​pubmlst.​org/​campylobacter). Dendrogram was plotted using Interactive Tree Of Life (iTOL) v4 [19].

A circular dendrogram was plotted for comparison of the peer sequences reported worldwide. A total of 72,392 isolates were downloaded from PubMLST website and analyzed [18]. Based on the inclusion and exclusion criteria, a total of 304 isolates were included in dendrogram. The inclusion criteria were: C. jejuni strains from the five recent years (2014–2018), C. jejuni strains from human stool, C. jejuni strains from the sporadic cases and gastroenteritis. The inclusion criteria were based on parameters that well warrant comparisons with our strains. Excluding criteria were unspecified ST or CC as well as repeated ST samples of each country.

The recorded MLST data in PubMLST database were also used to compare STs and CCs between our isolates with those from our neighbor countries. Among Iran neighboring countries only Turkey and Pakistan had recorded data about C. jejuni in PubMLST. Eleven isolates from human stool samples has been reported form these countries which were included for the final analysis. Dendrograms was plotted using Interactive Tree Of Life (iTOL) v4 [19]. Various sequence types were obtained from and plotted using PubMLST database tools [18].

To assess the presence of sialylated LOS classes A, B and C and CPS types in 20 C. jejuni strains isolated from children with gastroenteritis, we used the Cochran’s Q test. Data were analyzed with the Statistical Package for Social Sciences (Version 25.0, SPSS Inc., Chicago, IL, USA); p < 0.05 was considered statistically significant.

From 20 C. jejuni isolates, 17 (85%) expressed one of the selected CPS genotypes under study. CPS types HS23/36c were found in 9 isolates (45%) and appeared as dominant CPS among 20 C. jejuni strains (sig = 0.001; Cochran’s Q test). HS2, HS19, and HS4 were detected in 4 (20%), 2 (10%) and 2 (10%) isolates, respectively. No C. jejuni isolate belonged to HS1 or HS41 genotypes.

Of 20 strains, 16 (80%) expressed sialylated LOS locus class B or C. The LOS class B with 11 isolates (55%) was dominant sialylatd LOS class (sig = 0.003; Cochran’s Q test). No isolate with LOS class A was found among our isolates.

Among C. jejuni strains with LOS class B, the distribution of CPS types was as follows: HS23/36c (n = 6, 54/54%), HS2 (n = 2, 18/18%), HS4 (n = 2, 18/18%) and HS19 (n = 1, 9/9%). Also, the distribution of CPS genotypes among C. jejuni strains with LOS class C was as HS23/36c (n = 2, 40%) and HS2 (n = 2, 40%) (Fig. 1).

The primer efficiency was calculated as 97.51% and 103.54 for dnaK and 16srRNA genes, respectively. The dnaK gene expression was determined by Quantitative Real-Time PCR and according to the 2^-ΔΔCT method. Among isolates that showed one of the LOS classes of A-C or one of six selected capsular genotypes, 18 isolates were classified into three groups including group 1: with identified CPS genotype and sialylated LOS class (B or C) (n = 15), group 2: with identified CPS genotype and without sialylated LOS class (n = 2), group 3: without CPS genotype but with sialylated LOS class (n = 1), and one of the isolates that neither had the selected capsular serotypes nor the LOS locus classes was considered as a reference strain. Due to insufficiency of data, differential expression analysis could not be performed; thus, descriptive statistics was used instead. As a result, dnaK expression level in group 1 was greater than other groups (Table 5). The dnaK expression level was much higher in clinical C. jejuni isolates with one of the CPS genotypes and the LOS classes relevant to GBS patients. Based on the 2^-ΔΔCT method, the graph of the fold change of dnaK gene expression was plotted (Fig. 2).

Based on the MLST analysis of 20 isolates, 6 sequence types and 3 clonal complexes were detected in Iran. Five and 7 isolates were identified as ST-257 and ST-50, respectively. Each of the ST-19 and ST-5326 were detected in three isolates, while ST-1096 and ST-1113, each were only detected in one isolate.

Out of the 3 identified clonal complexes, ST-21 complex dominated in 10 isolates (50%).

Both ST-257 complex and ST-828 complex were found in 5 (25%) and 2 (10%) isolates, respectively.

Distribution of the sequence types and the genetic link between C. jejuni strains isolated from patients with gastroenteritis has been shown in the dendrogram (Fig. 3). The relationship among 20 isolates based on clonal complexes is reflected in minimum spanning tree diagram (Fig. 4) [20].

Totally, 3 CCs (CC828, CC257 and CC21) were identified. The isolates with HS23/36c and HS19 genotypes were found in CC21 (Fig. 5). Majority of isolates in CC21 had LOS class B or LOS class C. In CC257, 4 isolates belonged to HS2 serotype. In this CC, LOS class C and B were observed. Two isolates were assigned to CC828 (Fig. 6).

A phylogenetic analysis revealed that there was no common STs or CCs between Iran and its neighbor countries, Turkey and Pakistan (Fig. 7). This may be due to scarcity of published data from these countries. No data is available from other neighboring countries of Iran. World map related to C. jejuni strains from 1980 to 2018 shows the distribution of ST19 (CC21), ST50 (CC21), and ST257 (CC257) among different continents (Figs. 8, 9 and 10). The frequency of CC21 strains was associated to sporadic cases of human gastroenteritis which was recorded from America (USA and Canada), Europe (UK, The Netherlands and Germany), Asia (Iran, this study) and Australia. Inclusive comparison of CC257 (including ST-257) strains in the same time period showed recorded strains from Asia (Iran, this study), Europe (UK and Spain), America (Chile) and Africa (South Africa).

In a dendrogram constructed by inclusive comparison of MLST results during 2014–2018 a similar picture was depicted for distribution of CC21 and CC257 clonal complexes. Moreover, neighbor-joining results indicated that CC21 and CC353 complexes are the most divers, most frequent and most widely distributed clonal complexes around the world, respectively; although, CC353 was not detected in the current study (Fig. 11) (Table 6).