Molecular characterization of Clostridium botulinum isolates from foodborne outbreaks in Thailand, 2010

Abstract

Background: Thailand has had several foodborne outbreaks of botulism, one of the biggest being in 2006 when laboratory investigations identified the etiologic agent as Clostridium botulinum type A. Identification of the etiologic agent from outbreak samples is laborious using conventional microbiological methods and the neurotoxin mouse bioassay. Advances in molecular techniques have added enormous information regarding the etiology of outbreaks and characterization of isolates. We applied these methods in three outbreaks of botulism in Thailand in 2010.

Methodology/principal findings: A total of 19 cases were involved (seven each in Lampang and Saraburi and five in Maehongson provinces). The first outbreak in Lampang province in April 2010 was associated with C. botulinum type F, which was detected by conventional methods. Outbreaks in Saraburi and Maehongson provinces occurred in May and December were due to C. botulinum type A1(B) and B that were identified by conventional methods and molecular techniques, respectively. The result of phylogenetic sequence analysis showed that C. botulinum type A1(B) strain Saraburi 2010 was close to strain Iwate 2007. Molecular analysis of the third outbreak in Maehongson province showed C. botulinum type B8, which was different from B1-B7 subtype. The nontoxic component genes of strain Maehongson 2010 revealed that ha33, ha17 and botR genes were close to strain Okra (B1) while ha70 and ntnh genes were close to strain 111 (B2).

Conclusion/significance: This study demonstrates the utility of molecular genotyping of C. botulinum and how it contributes to our understanding the epidemiology and variation of boNT gene. Thus, the recent botulism outbreaks in Thailand were induced by various C. botulinum types.

Figure 1. Multiplex PCR typing of boNT/A and boNT/B genes for Saraburi and Maehongson 2010 outbreak strains.

The PCR pattern of Saraburi 2010 showed positive boNT/A1 amplicon (665 bp, gel A lanes 6–8) and boNT/B1 like amplicon (585 bp, gel B lanes 7–9). The PCR patterns of Maehongson 2010 indicated a boNT/B2-like gene was present as 370 bp amplicon (gel C, lanes 8–11).

Figure 2. Multiplex PCR of ha33 and p47genes for confirmation of boNT/A and boNT/B typing Isolate from Saraburi 2010 strain showed positive ha33 and p47 amplicon (534 bp and 344 bp, lane7), ha 33 of boNT/A1 amplicon (534 bp, lane 1), p47 of boNT/A2 amplicon (344 bp, lane 2).

Figure 3. Phylogenetic analysis of boNT/B nucleotide sequences from various C. botulinum strains.

The phylogenetic tree was generated using the neighbor-joining method in MEGA (v5) software. Bootstrap values (the percentage that each branch would occur after 1,000 bootstrap replicates) and genetic distance (bar) are shown. Clusters corresponding to different phylogenetic groups are labeled according to previous reports (Table 1). The toxin serotypes of the strains are shown on the right, and the Thai isolates from the 2010 outbreaks are highlighted.

Figure 4. PFGE and Southern blot hybridization of undigested bacterial DNA from the different C. botulinum strains and outbreak isolates to identify boNT gene location.

(A and C) undigested PFGE of isolated DNA (B) hybridization using a boNT/A probe (D) hybridization results with the boNT/B probe.