Date Published: July 9, 2015
Publisher: Public Library of Science
Author(s): Sabine Dittrich, Weerawat Phuklia, Gareth D. H. Turner, Sayaphet Rattanavong, Vilada Chansamouth, Stephen J. Dumler, David J. P. Ferguson, Daniel H. Paris, Paul N. Newton, David H Walker. http://doi.org/10.1371/journal.pntd.0003908
Abstract: Neorickettsia sennetsu infection is rarely recognized, with less than 100 globally reported patients over the last 50 years. The disease is thought to be contracted by eating raw fish, a staple of many South-East Asian cuisines. In 2009, the first patient with sennetsu was identified in the Lao PDR (Laos), raising the question as to how common this organism and related species are in patients presenting with fever. We investigated the frequency of N. sennetsu infection at hospitals in diverse areas of Laos. Consenting febrile hospital inpatients from central (Vientiane: n = 1,013), northern (Luang Namtha: n = 453) and southern (Salavan: n = 171) Laos were screened by PCR for N. sennetsu, if no previous positive direct diagnostic test was available. A PCR-restriction fragment length polymorphism assay was developed to differentiate between N. sennetsu, Ehrlichia chaffeensis and Anaplasma phagocytophilum. To allow more detailed studies of N. sennetsu, culture was successfully established using a reference strain (ATCC VR-367), identifying a canine-macrophage cell line (DH82) to be most suitable to visually identify infection. After screening, N. sennetsu was identified and sequence confirmed in four (4/1,637; 0.2%) Lao patients. Despite the previously identified high seroprevalence of N. sennetsu antibodies in the Lao population (~17%), acute N. sennetsu infection with sufficient clinical signs to prompt hospitalization appears to be rare. The reservoir, zoonotic cycle and pathogenicity of N. sennetsu remain unclear and require further investigations.
Partial Text: Neorickettsia sennetsu was not reported for 24 years prior to the identification of a Lao patient with sennetsu in 2009 . This pathogen was the first documented cause of glandular fever (infectious mononucleosis) in Japan in 1954, characterized by fever, weakness, anorexia, lymphadenopathy and peripheral blood mononucleosis with atypical lymphocytes . The route of human infection was thought to be the consumption of raw fish (gray mullet, Mugil cephalus) . A recent study in rural, northern Thailand identified two additional acute infections of N. sennetsu in febrile patients by serology .
Prospectively collected samples from consecutive patients were submitted to the Mahosot Hospital Microbiology Laboratory (n = 1,013) from Vientiane (VTE; longitude 102.6119°E, latitude 17.96‟04°N) admitted between January 2010 to December 2011 with suspected typhus to Mahosot Hospital (n = 816), Friendship Hospital (n = 29), Settathirat Hospital (n = 167) and the Military Hospital (n = 1). All patients presented with fever and local physicians suspected typhus. Mahosot Hospital and the four other major hospitals in VTE (1,210 beds total) serve a population of ~900,000 people, including the urban population of Vientiane City and surrounding farming communities of Vientiane Province. Patients from the southern and northern provinces (Salavan Provincial Hospital [SV], longitude 106.2500°E, latitude 15.4300° N; Luang Namtha Provincial Hospital [LNT], longitude 101.4025°E, latitude 20.9606° N) were included from May 2008–December 2010 as part of a large undifferentiated fever study . They were included in the current investigation if they had previously not been assigned a direct (culture/PCR) diagnosis . All samples were screened using a described conventional PCR targeting the 16S rRNA gene, which amplifies N. sennetsu, A. phagocytophilum, E. chaffeensis and related organisms, with some amendments [1,13]. PCR reactions were set up in a total of 25 μL with 1x Platinum UDG-Supermix (Invitrogen) with annealing temperature of 61°C. Positive control-DNA from N. sennetsu, A. phagocytophilum and E. chaffeensis cultures as well as no-template controls, were added in each run. DNA was extracted from 200 μL EDTA buffy coat as described . All positive samples were confirmed using PCRs targeting gltA and omp85, as described . Samples positive for all three targets were processed for DNA sequencing (Macrogen, South Korea) and subsequently analyzed using NCBI-BLAST. To differentiate N. sennetsu, A. phagocytophilum and E. chaffeensis without the need for third party-sequencing, a simple restriction fragment length polymorphism (RFLP) assay on the 16 sRNA amplicon  was developed (Fig 1). Ten microliters of PCR product containing dTTP, was incubated with AluI (NEB, USA; 1 Unit, 37°C, 2h), StyI (NEB, USA; 1 Unit, 37°C, 2h), and BsmF1 (NEB, USA; 1 Unit, 65°C, 2h); and the appropriate reaction buffer. PCR amplicons and RFLP products were visualized on a 4% agarose gel (NuSieve GTG Agarose, Lonza) and RFLP-patterns compared to positive controls (S2 Fig and S1 Table). To further allow characterization and genotyping of patients, N. sennetsu (strain Miyayama) culture was established using the ATCC VR-367 reference strain. It was propagated at biosafety level-3 in an adherent canine macrophage-like cell line (DH82; ATCC CRL-10389), African green monkey kidney cells (VERO; provided by the Australian Rickettsia Reference Laboratory, Geelong, Australia) and mouse fibroblast cell line (L929; provided by the Australian Rickettsia Reference Laboratory, Geelong, Australia). Eagle’s Minimum Essential Medium (EMEM, Invitrogen, UK) was supplemented with 10% fetal bovine serum (FBS; Invitrogen, UK) and 2 mM L-glutamine (Invitrogen, UK) for DH82 cells. RPMI-1640 (Pacific Science, Thailand) was supplemented with 10% FBS (Invitrogen, UK) for VERO and L929 cells and cultures were maintained as described [14,15]. Cultures were observed daily for cytopathic effect (CPE) and a weekly PCR  on supernatant and cells was performed. Confirmation of intracellular growth was obtained by examining cell samples by electron microscopy (EM; Fig 2). Confluent layers of infected cells were trypsinised, washed with 1xPBS and than fixed with 2.5% glutaraldehyde, in 0.1M phosphate buffer. The samples were post-fixed in 2% osmium tetroxide in phosphate buffer and dehydration in ethanol followed by treatment with propylene oxide, prior to embedding in TABB epoxy resin. Thin sections of suitable areas were examined using a Jeol 1200EX electron microscope .
In total, 1,637 febrile patient samples were included from three different geographical locations in Laos, LNT in the north (n = 453), SV in the south (n = 171) and VTE in the center (n = 1,013). The majority of patients were male (56.2%) with a median age of 25 years (range: 1–96). The predominant symptoms of patients at all sites were fever, headache and myalgia. Across the three locations 10.6%-50.7% of patients had lymphadenopathy and 12.2%-28.7% of patients had pharyngitis. Peripheral blood atypical lymphocytes were not determined. All available patient samples were screened and strong positive bands at the correct size were observed in four of 1,642 samples (0.24%) and in addition 75 samples showed weak positive bands (75/1642; 4.6%). Positive and negative controls gave the correct results in all experiments. The strong positive results could be confirmed as N. sennetsu by gltA, omp85 PCR, 16s RNA sequencing as well as the 16 sRNA-RFLP. DNA sequences were deposited on Genbank (Accession Numbers: KM355198-KM355200; LNT554_fragment) (S2 Table and S1 Fig). The remaining weak positive samples were considered negative, as all subsequent confirmatory tests (gltA and ompB-PCR, PCR-16 sRNA RFLP [selected samples only] or sequencing) were negative. The newly established PCR-RFLP assay showed 100% concordance with the sequencing result for N. sennetsu in strongly positive samples (S2 Fig). Due to a lack of A. phagocytophilum and E. chaffeensis positive patient samples, the ability of the PCR-RFLP assay to correctly identify those pathogens in patients rather than positive controls could not be determined.
Large-scale retrospective screening of febrile patients, resulted in the identification of four patients with sennetsu, from three disparate Lao geographical areas, raising the total number of patients recorded in the country to five. All patients had a history of eating raw fish consistent with the suggested route of transmission. Still as eating raw fish is very common in Laos  this does not represent a strong epidemiological link.