Antimicrobial resistant Helicobacter fennelliae isolated from non-diarrheal child stool sample in Battambang, Cambodia

Helicobacter fennelliae (H. fennelliae) is a new Campylobacter species originally isolated from asymptomatic, homosexual men with enteritis and proctitis in the past few decades [1]. Like H. cinaedi, this species is classified as enterohepatic Helicobacter that inhabits and causes bacteremia in intestinal and hepatobiliary tracts of various mammal and other species [2]. Additional evidence suggests that H. fennelliae was implicated as a contributing cause of human proctocolitis, gastroenteritis, and bacteremia, particularly in immunocompromised individuals [2, 3]. This Helicobacter species is a fastidious organism that is likely underestimated, and little is known about routes of transmission other than evidence indicates it is a zoonotic infection [2]. As a fastidious organism, molecular genotyping methods are recommended to identify Helicobacter species. Towards that end, the groEL and hsp60 genes encode a 60 kDa chaperonin protein present in virtually all eubacteria, some archaea, and in the plastids and mitochondria of eukaryotes. The utility of this target for bacterial species identification, detection, quantification, phylogenetic analysis, and microbial community profiling was well established [4]. Treatment recommendation guidelines are still not available for enterohepatic Helicobacter species. Various individual and combined antibiotic regimens were successfully used in treating Helicobacter infections; however, there is insufficient information to determine resistance rates of H. fennelliae. The main objective of this report is to describe phenotypic, genotypic, and antimicrobial susceptibility (AST) data from this H. fennelliae isolate from the stool of non-diarrheal child in Cambodia.

A surveillance study to describe diarrhea etiologic agents in children and military personnel in Battambang, Cambodia has been conducted from 2014 until present. Both diarrheal and non-diarrheal stool samples were observed by microscopic examination for the presence of parasites, protozoa, and larvae. Samples were also assessed for the presence of Giardia, Cryptosporidium by enzyme-linked immunosorbent assay (ELISA), and for diarrheagenic E. coli by polymerase chain reaction (PCR) [5]. Enteric pathogens, including Campylobacter species, were isolated and identified by traditional culture methods [6]. The suspected Campylobacter-like colonies were subcultured on blood agar supplemented with 6% sodium formate and fumarate for 48–72 h at 37 °C under microaerobic conditions (10% CO₂ and 5% O₂). The biochemical identifications were included oxidase, catalase, indoxyl hydrolysis, hippurate hydrolysis, nitrate reduction, urease, hydrogen sulfide production, susceptibility to cephalothin and nalidixic acid (30 µg disc) (BD, Spark, USA), oxygen and temperature tolerance test. According to no antimicrobial susceptibility recommendation guidelines, H. fennelliae resistance was determined using the minimal inhibitory concentration (MIC) by E test (Liofilchem, Roseto degli Abruzzi TE, Italy) against azithromycin (AZM), erythromycin (ERY), nalidixic acid (NAL), ciprofloxacin (CIP), levofloxacin (LEV), ceftriaxone (CRO), spectinomycin (SPT), and tetracycline (TET). C. jejuni ATCC 33560 was used as a quality control strain.

Genomic DNA of suspected Campylobacter-like colonies was extracted and subsequently confirmed as belonging to the Campylobacter genus by screening for the 16S rRNA gene [7]. To determine Campylobacter species, the 15 primer sets of cpn60 target gene were used for verified species as described elsewhere [7, 8]. Subsequently, the unknown Campylobacter species beyond 15 primer sets identification were further sequencing analysis by amplifying cpn60 target gene with degenerate primers H729 and H730 [4]. The sequences of degenerate primers were H729: 5′-CGCCAGGGTTTTCCCAGTCACGACGAIIIIGCIGGIGAYGGIACIACIAC-3′ and H730 5′-AGCGGATAACAATTTCACACAGGAYKIYKITCICCRAAI CCIGGIGCYTT-3′. PCR amplification was carried out in a total volume of 50 µL containing 6 µL of genomic DNA template, 2.5 U AmpliTaq Gold^® DNA polymerase (Applied Biosystems, Foster City, Calif.), 5 mM MgCl₂, 100 µM each of the dNTPs and 50 nM each of degenerate primers [4]. The cycling conditions were performed at 94 °C for 5 min, followed by 28 cycles of 1 min at 94 °C, 1 min at 46 °C, 1 min at 72 °C, and a final extension at 72 °C for 10 min. The purified PCR products were additionally differentiate Campylobacter species from Helicobacter and Acrobacter species using primers M13F-pUC (− 40) 5′-GTTTTCCCAGTCACGAC-3′ and M13R (− 20) 5′-GCGGA-TAACAATTTCACACAGG-3′. The result of partial cpn60 sequences (555 bp) was compared with the database in cpnDB (http://​cpndb.​cbr.​nrc.​ca) [4]. The confirmed partial sequence was submitted to the National Center for Biotechnology Information (NCBI) before constructing phylogenetic analysis by BioNumerics software version 7.6 (Applied Maths, Belgium).

A non-diarrheal stool sample of a young child who presented to the hospital with fever and, cough was submitted for laboratory testing. The stool characteristic was loose without mucus, blood, RBCs, or WBCs. No gastrointestinal parasites were detected microscopically or by ELISA. Other enteric bacterial pathogens, including diarrheagenic E. coli, were not identified, except for suspected colonies of a Campylobacter-like organism. The colonies characteristics which were presented after 6 days incubation were thin, flat, film-like colony, with a hypochlorite odor. Biochemical reactions of the colony were positive for oxidase, catalase, and indoxyl acetate hydrolysis. It was susceptible to cephalothin disk but resistant to nalidixic acid disk and could be grown at 42 °C under microaerobic conditions. Culture results indicated that H. fennelliae grows well by supplementing 6% sodium formate and fumarate in blood agar. This is likely due to the fact that formate replaces hydrogen as the electron donor, and fumarate serves as the terminal electron acceptor for hydrogen-required organism growth [9]. Notably, an absence of hydrogen, the low-cost supplemented media, and a long incubation period are suggested to support growth of H. fennelliae.

The MIC results of H. fennelliae and C. jejuni ATCC 33560 were presented in Table 1. Results for this H. fennelliae isolate demonstrated high MICs to macrolides and quinolones, consistent with previous studies [10, 11] and similar to H. cinaedi data [12]. Macrolides, generally considered the drug of choice for Campylobacter treatment [1], may be clinically less effective for Campylobacter-like organism infections such as H. fennelliae and H. cinaedi. Little is known about the antimicrobial resistance mechanisms of H. fennelliae. Mutations of the gyrase and 23S rRNA genes may be responsible for decreased susceptibility to quinolones and macrolides, respectively [10]. However, decreased susceptibility to low MIC macrolide levels were mentioned in a previous study [13]. The H. fennelliae isolate from our study exhibited a high MIC to macrolides, warranting further molecular characterization to explore other resistance mechanisms.

The genotyping confirmation of this H. fennelliae isolate was performed by sequencing the cpn60 gene, and the result was submitted to NCBI under the accession number MG696736. A phylogenetic tree analysis (Fig. 1) divided Helicobacter and Campylobacter strains into seven distinct groups (cut-off of 90%). The MG696736 entry was classified as group IV, which was 97.2% similar to H. fennelliae ATCC 35684, whereas Campylobacter species was classified as group VII, which is distinct from the Helicobacter group VI (cut-off of 90%).

H. fennelliae was suggested as a significant pathogen associated with human gastroenteritis; however, its prevalence and antimicrobial resistant profile might be considerably underestimated due to inadequate isolation and identification methods [14]. To the best of our knowledge, this is the first report of a macrolide and quinolone resistant H. fennelliae identified in a young Cambodian child asymptomatic for intestinal infection. This isolate resembles H. fennelliae, which was previously identified in a boy suffering gastroenteritis and is also isolated from dog specimens [15]. With the introduction of the ‘Cape Town Protocol,’ H. fennelliae may be isolated from stool and blood culture in an H₂-rich microaerophilic atmosphere. Prior evidence indicated that Helicobacter species related to H. fennelliae were isolated from blood of a young child suffering diarrhea symptoms [16]. Nevertheless, the nucleotide sequences of H. fennelliae obtained from blood and stool were not significantly different [17]. Unfortunately blood samples were not available from the child in this study, so that comparison was not achievable. H. fennelliae was predominantly isolated from children who presented with diarrheal symptoms, although stools from asymptomatic diarrheal children with asthma and/or failure to thrive (FTT) were also positive for H. fennelliae [16, 17]. Another possible explanation of this H. fennelliae finding in stool of asymptomatic diarrheal Cambodia child could relate to breastfeeding. Evidence suggests that maternal milk contains a variety of functionally bioactive agents from her innate immune system [18], as well as a mechanism to influence microbial changes in the infant’s gastrointestinal system [19]. As a result of widespread breastfeeding campaigns in the developing world, this may play an important role in the level of asymptomatic carriage within a community [18, 20]. The association between asymptomatic carriage and diarrheal pathogens such as Salmonella, E. coli O157 and Campylobacter was previously reported in outbreaks elsewhere [20]. Identification of an antibiotic resistant H. fennelliae strain from an asymptomatic diarrheal person would probably be transmitted into local communities and environmental contamination. Hence, the public health significance of resistant pathogens in human feces warrants effective monitoring to prevent disease outbreaks.

In conclusion, phenotypic and genotypic assessments confirmed that H. fennelliae was isolated from a non-diarrheal stool sample of a Cambodian child suffering from fever with cough and convulsion. The supplement media, incubation atmosphere, and incubation period utilized permitted culture, isolation, and identification of H. fennelliae. The high MICs values against macrolides (AZM, ERY) and quinolones (NAL, CIP) indicated these are less effective against H. fennelliae. This isolate should be further characterized to better understand associated resistance mechanisms.