Analysis of a unique Clostridium botulinum strain from the Southern hemisphere producing a novel type E botulinum neurotoxin subtype
© Raphael et al.; licensee BioMed Central Ltd. 2012
Received: 24 August 2012
Accepted: 30 October 2012
Published: 31 October 2012
Clostridium botulinum strains that produce botulinum neurotoxin type E (BoNT/E) are most commonly isolated from botulism cases, marine environments, and animals in regions of high latitude in the Northern hemisphere. A strain of C. botulinum type E (CDC66177) was isolated from soil in Chubut, Argentina. Previous studies showed that the amino acid sequences of BoNT/E produced by various strains differ by < 6% and that the type E neurotoxin gene cluster inserts into the rarA operon.
Genetic and mass spectral analysis demonstrated that the BoNT/E produced by CDC66177 is a novel toxin subtype (E9). Toxin gene sequencing indicated that BoNT/E9 differed by nearly 11% at the amino acid level compared to BoNT/E1. Mass spectrometric analysis of BoNT/E9 revealed that its endopeptidase substrate cleavage site was identical to other BoNT/E subtypes. Further analysis of this strain demonstrated that its 16S rRNA sequence clustered with other Group II C. botulinum (producing BoNT types B, E, and F) strains. Genomic DNA isolated from strain CDC66177 hybridized with fewer probes using a Group II C. botulinum subtyping microarray compared to other type E strains examined. Whole genome shotgun sequencing of strain CDC66177 revealed that while the toxin gene cluster inserted into the rarA operon similar to other type E strains, its overall genome content shared greater similarity with a Group II C. botulinum type B strain (17B).
These results expand our understanding of the global distribution of C. botulinum type E strains and suggest that the type E toxin gene cluster may be able to insert into C. botulinum strains with a more diverse genetic background than previously recognized.
KeywordsBotulism Mass spectrometry Genomics Whole genome sequencing
There are 7 serotypes (types A-G) of botulinum neurotoxins (BoNT) and types A, B, E or F are the most frequent causes of botulism in humans. Strains of Clostridium botulinum producing BoNT/E share similar metabolic characteristics including the inability to digest proteins such as gelatin, casein, or meat. These non-proteolytic strains are psychrophilic with the ability to grow at refrigeration temperatures . In rare cases, strains of Clostridium butyricum have been shown to produce BoNT/E .
Clostridium botulinum type E strains can be isolated from various marine environments and cases of botulism due to BoNT/E typically occur in Canada, Alaska, Northern Europe, and Japan . A total of 56 cases of type E botulism were reported to the Centers for Disease Control and Prevention between 2001–2010 and 87.5% of these cases occurred in Alaska (http://www.cdc.gov/nationalsurveillance/botulism_surveillance.html). Type E botulism has also occurred in the lower 48 states including various outbreaks associated with smoked fish from the Great Lakes [4, 5]. A recent outbreak of botulism in birds and fish in the Great Lakes region was attributed to genetically distinct strains of C. botulinum type E and the organism was also found in lake sediment . A case of infant botulism occurred in Illinois in 2007 although the source of spores in this case could not be determined .
Genetic analysis of 16S rRNA sequences from various C. botulinum strains reveals the presence of distinct phylogenetic groups (I-IV) [8, 9] which correspond to previously recognized metabolic differences. All Group II strains are non-proteolytic and include type E strains and some type B and type F strains. Nucleotide sequencing of various toxin genes has demonstrated the presence of amino acid variation within genes encoding a single toxin serotype and these variants are identified as toxin subtypes [9, 10]. Among type E strains, a total of 8 such subtypes (E1-E8) have been identified . These subtypes differ at the amino acid level by up to 6%.
The genes encoding BoNT/A-G are found in toxin gene clusters that also encode several nontoxic proteins and regulatory proteins. The gene encoding BoNT/E is found within a toxin gene cluster that includes ntnh (nontoxic nonhemagglutinin), p47, and orfX1-3[12, 13]. Hill et al.  demonstrated that the bont/E toxin gene cluster inserted into the rarA operon. The transposon-associated gene, rarA, likely plays a role in this insertion event in which the gene is split into small and large fragments that flank the toxin gene cluster . Remarkably, an intact rarA gene is also located within the toxin gene cluster and the nucleotide sequences of the intact and split genes were shown to differ by phylogenetic analysis. Moreover, the split rarA gene fragments can be pasted together to form a gene with a nucleotide sequence with similarity to the gene found in the Group II C. botulinum type B strain 17B. In another study, the intact and split rarA genes were detected across a panel of 41 type E strains .
In this study, we characterized a previously unreported C. botulinum type E strain isolated in 1995 from soil in Chubut, Argentina. This represents the first report of a type E strain (CDC66177) originating from the Southern hemisphere. We further show evidence that this strain produces a unique type E toxin subtype and that the genetic background of this strain is highly divergent compared to other type E strains.
Results and discussion
Phylogenetic analysis of bont/E in C. botulinumstrains
BLAST analysis of the 16S rRNA nucleotide sequence from strain CDC66177 shared > 99.8% identity with strains Alaska E43 and 17B indicating that the strain clusters with other Group II C. botulinum strains .
Mass spectrometric analysis of BoNT/E produced by strain CDC66177
BoNT/E9 extracted from culture supernatants of strain CDC66177 was subjected to tryptic digestion and the products were analyzed by mass spectrometry to confirm that the toxin's amino acid sequence was indeed unique based on the predicted translation of the DNA sequence. The amino acid sequence of BoNT/E9 was determined with 94.5% coverage (Figure 3B).
DNA microarray analysis of strain CDC66177
A Group II C. botulinum subtyping DNA microarray  was used to evaluate gene content in a panel of 21 Group II strains from the CDC culture collection. Briefly, this array featured 495 probes targeting ~15% of the annotated genes in the C. botulinum type E strain Alaska E43 and 5 additional probes targeting genes present on the bont/B-encoding plasmid (pCLL) in C. botulinum type B strain 17B. Genomic DNA isolated from 15 type E strains (not including CDC66177) hybridized with 90.5% of the probes on this array while DNA isolated from type B strains (N=4) and type F strains (N=2) hybridized with 71.9% and 71.0% of the probes, respectively. Genomic DNA from strain CDC66177 hybridized with 66.8% of the probes present on the array.
Southern hybridization of the split rarAgene in strain CDC66177
Whole genome shotgun sequencing of strain CDC66177
As shown in Figure 6, the regions between orfX3 and the larger split rarA fragment (region I) and between the smaller split rarA fragment and bont/E (region II) contain insertion sequences that are likely involved with transposon-mediated mobility of the toxin gene cluster . It is notable that regions I and II differ in size and nucleotide sequence between strains Alaska E43 and CDC66177. In order to determine if the nucleotide sequences of these regions are strain-specific, we also performed an alignment of these regions with strain Beluga. Interestingly, region I in strain Beluga differed from both CDC66177 and Alaska E43 while region II was identical to that found in Alaska E43. While the mechanism of toxin gene cluster insertion into the rarA operon is unclear, the sequence similarity in region II between strains Beluga and Alaska E43 suggests at least a partial similarity in the origin of the recombination event that results in the insertion of the toxin gene cluster. However, strain CDC66177 lacks similarity to either strain Beluga or Alaska E43 at either region suggesting that the recombination event resulting in the insertion of the toxin gene cluster in strain CDC66177 originated differently compared to strains Beluga or Alaska E43.
Analysis of the genome sequence data explains the unexpected ~1.7 kb band hybridized by the rarA probe in strain CDC66177. The presence of an XbaI site within the toxin gene cluster of both CDC66177 and Alaska E43 and an additional site downstream of the larger rarA fragment in strain CDC66177 yield an ~1.7 kb fragment. Notably the genome sequence of strain 17B also demonstrates the presence of a XbaI site downstream of the intact rarA gene. Similar to other type E toxin gene clusters, strain CDC66177 contains an intact rarA gene that does not hybridize the rarA probe used in our studies. BLAST analysis of this gene demonstrated 98% nucleotide similarity with the gene present in Alaska E43.
Pairwise alignment of toxin gene cluster components
% Nucleotide Identity
Alaska E43/Beluga E
Average nucleotide identity (ANI) of genomic sequences
In a previous study , botulinum toxin-producing clostridia were isolated from 23.5% of soil samples collected in Argentina. The distribution of toxin serotypes reported from the Southern region of Argentina included types A, B, and F. In this study, we characterized a previously unreported C. botulinum type E strain (CDC66177) isolated in 1995 from soil collected in Chubut, Argentina. This region is located at a latitude of approximately 43°S which is located as far from the equator as the Great Lakes are located in the Northern hemisphere. While strain CDC66177 was isolated from soil in proximity to the Atlantic Ocean, it is notable that no cases of type E botulism have been reported in Argentina. This is the first known report of the isolation of this strain and extends the known global distribution of C. botulinum type E.
While the strain CDC66177 produces a novel BoNT/E subtype, the toxin was shown to cleave a peptide substrate in the same location as other BoNT/E subtypes. It remains to be determined if the toxin produced by this strain varies in its neuronal cell receptor compared to other BoNT/E subtypes. Finally, the presence of bont/E in the rarA operon of a strain with genetic similarity to strain 17B raises the intriguing possibility of a bivalent non-proteolytic strain expressing BoNT/E encoded by a chromosomally located gene and BoNT/B encoded by a plasmid (such as pCLL found in 17B).
Bacterial strains used in this study
Bacterial strains used in this study
Fermented seal flipper
Pacific coast, US
Fermented salmon brine
Columbia River, US
DNA extraction, genetic analysis, and DNA microarray
Genomic DNA used in Sanger sequencing and DNA microarrays was extracted using the PureLink Genomic DNA kit (Life Technologies, Grand Island, NY). Neurotoxin and 16S rRNA gene sequences were determined using previously reported primers that amplified overlapping regions [9, 19]. Phylogenetic analysis was performed using CLUSTALX and the resulting phylogenetic tree was rendered using MEGA 5.05 . Comparative analysis among representative BoNT/E subtypes was performed using SimPlot (http://sray.med.som.jhmi.edu/SCRoftware/simplot/) with a 200 amino acid window.
The Group II C. botulinum subtyping microarray was designed as described elsewhere . Briefly, the microarray featured 495 probes representing genes distributed throughout the C. botulinum Alaska E43 genome sequence and 5 additional probes specific for pCLL which encodes the toxin gene cluster in strain 17B. Microarray spotting was performed by ArrayIt (Sunnyvale, CA) or onsite using an Omnigrid Micro microarrayer (Digilab, Holliston, MA). Genomic DNA was labeled with Cy5 random primers and hybridized to the array as previously described . The log of the ratio of the mean fluorescence signal at 635 nm for triplicate probes compared to background fluorescence (locations spotted with buffer alone) was calculated. Log ratios ≥ 1.0 were considered positive and those < 0.5 were considered negative. Log ratios between 0.5 and < 1.0 were considered intermediate likely due to nucleotide sequence variation . Hybridization profiles were converted to binary data by assigning 1 to positive probes and 0 to negative and intermediate probes. Profiles were compared using a UPGMA dendrogram generated with DendroUPGMA (http://genomes.urv.cat/UPGMA/) and selecting the Jaccard coefficient. Microarray data were deposited in the Gene Expression Omnibus with series accession number GSE40271.
Genomic DNA was digested with XbaI for 1 h and run on a 1% TBE agarose gel. Alkaline transfer was performed using the TurboBlotter system (Whatman, Kent, ME). An 874 bp probe corresponding to the large rarA fragment was generated by PCR amplification with primers RarA-F and RarA-R (RarA-F, 5′-GCAAGCACAACTGAAAATCCT-3′; RarA-R, 5′-CTCGGCTTTTGTXCAATTCATTAG-3′) and labeled with the DIG DNA Labeling and Detection kit (Roche, Indianapolis, IN). Hybridization was carried out at 42°C in standard hybridization buffer (5X SSC, 0.1% N-laurylsarcosine, 0.02% SDS, 1% Blocking buffer (from DIG DNA Labeling and Detection kit).
Mass spectrometric analysis
Botulinum neurotoxin in culture supernatant CDC66177 was extracted and tested for light chain protease activity in a manner similar to that previously described , with the exception that 200 μL of culture supernatant was used for this study. Briefly, the neurotoxin was extracted from the culture supernatant using protein G beads coated with antibodies to BoNT/E. Following washing, the beads were then incubated for 4 h at 37°C with a peptide substrate known to be cleaved by BoNT/E in the presence of a reaction buffer. The reaction supernatant was then analyzed by MALDI-TOF mass spectrometry as described previously to determine the location of cleavage of the peptide substrate.
The reaction supernatant was then completely removed from the beads, and the toxin on the beads was digested and analyzed by LC-MS/MS essentially as described previously , with the exception that an Orbitrap Elite was used in place of the fourier transform magnetic trap. Briefly, the beads with toxin attached were digested with trypsin and then chymotrypsin. The resultant peptide mixtures were separated by nano-LC and mass analyzed on an Orbitrap Elite, generating MS/MS of the peptides. The MS/MS data were then searched against a database indexed for only Clostridium spp. for protein identification.
Whole genome sequencing and analysis
Genomic DNA was isolated from strain CDC66177 using the MasterPure kit (Epicenter, Madison, WI) with modifications previously described . This DNA was further purified using a Genomic-tip 100/G column (Qiagen, Valencia, CA). One microgram of genomic DNA was sheared using a Covaris S2 ultrasonicator system to a mean size of 1 Kb. The sheared DNA was used to construct a SMRTbell sequencing library (Pacific Biosciences) according to manufacturer’s instructions. The SMRTbell library was then bound into SMRTbell-DNA polymerase complexes and loaded into zero-mode waveguides (ZMW) on 4 SMRTcells and sequenced using Pacific Biosciences C2 chemistry. This relatively small insert sized library was utilized to promote production of circular concensus reads (CCS) which retain higher accuracy base calls than the longer continuous length reads (CLR). Eight 45 min movies were recorded and processed, yielding ~305 K reads with a mean readlength of 2.9 Kbases and total of 889 Mbases of sequence. CCS reads (140 K reads) were then used to error correct the longer (165 K reads) CLR reads  utilizing the Pacific Biosciences analysis script BLASR and then the combined CCS/corrected CLR fastq format reads were imported into CLC Genomics workbench. Sequence reads were then trimmed of any remaining Pacific Biosciences hairpin adaptor sequences and quality trimmed to a base Q value of 20. The filtered reads were then assembled de novo using the CLC denovo assembler. The 188,898 input reads provided a draft assembly of a 3.85 Mb genome comprised of 119 contigs with an N50 value of 87,742 bases with an average coverage of 28X.
Annotation of the whole genome sequence was performed using RAST . Pairwise alignments of various genes were made with EMBOSS Needle (http://www.ebi.ac.uk/Tools/psa/emboss_needle/nucleotide.html). ANI values were determined using the computer program JSpecies . MLST loci from selected previously reported type E strains were obtained from Genbank . These MLST loci were used to search for the corresponding alleles in the strain 17B genome sequence and the CDC66177 whole genome sequence using BLAST. Concatemers of the alleles for each strain were generated and a multiple sequence alignment was performed using CLUSTALW because the lengths of some alleles in strains 17B and CDC66177 differed due to insertion and/or deletions.
Sanger sequencing was performed in the Genomics Unit within the Division of High Consequence Pathogens and Pathology at CDC. This publication was supported by funds made available from the Centers for Disease Control and Prevention, Office of Public Health Preparedness and Response. ML was supported by an Oak Ridge Institute for Science and Education fellowship. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
- Graham AF, Mason DR, Maxwell FJ, Peck MW: Effect of pH and NaCl on growth from spores of non-proteolytic Clostridium botulinum at chill temperature. Lett Appl Microbiol. 1997, 24: 95-100. 10.1046/j.1472-765X.1997.00348.x.PubMedView ArticleGoogle Scholar
- McCroskey LM, Hateway CL, Fenicia L, Pasolini B, Aureli P: Characterization of an organism that produces type E botulinal toxin but which resembles Clostridium butyicum from the feces of an infant with type E botulism. J Clin Microbiol. 1986, 23: 201-202.PubMedPubMed CentralGoogle Scholar
- Horowitz BZ: Type E botulism. Clin Toxicol. 2010, 48: 880-895. 10.3109/15563650.2010.526943.View ArticleGoogle Scholar
- Kautter DA: Clostridium botulinum in smoked fish. J Food Sci. 1964, 29: 843-849. 10.1111/j.1365-2621.1964.tb00458.x.View ArticleGoogle Scholar
- Whittaker RL, Gilbertson RB, Garrett AS: Botulism, Type E. Ann Intern Med. 1964, 61: 448-454.PubMedView ArticleGoogle Scholar
- Hannett GE, Stone WB, Davis SW, Wroblewski D: Biodiversity of Clostridium botulinum type E associated with a large outbreak of botulism in wildlife from Lake Erie and Lake Ontario. Appl Environ Microbiol. 2011, 77: 1061-1068. 10.1128/AEM.01578-10.PubMedPubMed CentralView ArticleGoogle Scholar
- Lúquez C, Dykes JK, Yu PA, Raphael BH, Maslanka SE: First report worldwide of an infant botulism case due to Clostridium botulinum type E. J Clin Microbiol. 2010, 48: 326-328. 10.1128/JCM.01420-09.PubMedPubMed CentralView ArticleGoogle Scholar
- Collins MD, East AK: Phylogeny and taxonomy of the food-borne pathogen Clostridium botulinum and its neurotoxins. J Appl Microbiol. 1998, 84: 5-17. 10.1046/j.1365-2672.1997.00313.x.PubMedView ArticleGoogle Scholar
- Hill KK, Smith TJ, Helma CH, Ticknor LO, Foley BT, Svensson RT, Brown JL, Johnson EA, Smith LA, Okinaka RT, Jackson PJ, Marks JD: Genetic diversity among botulinum neurotoxin-producing clostridial strains. J Bacteriol. 2007, 89: 818-832.View ArticleGoogle Scholar
- Smith TJ, Lou J, Geren IN, Forsyth CM, Tsai R, Laporte SL, Tepp WH, Bradshaw M, Johnson EA, Smith LA, Marks JD: Sequence variation within botulinum neurotoxin serotypes impacts antibody binding and neutralization. Infect Immun. 2005, 73: 5450-5457. 10.1128/IAI.73.9.5450-5457.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Macdonald TE, Helma CH, Shou Y, Valdez YE, Ticknor LO, Foley BT, Davis SW, Hannett GE, Kelly-Cirino CD, Barash JR, Arnon SS, Lindström M, Korkeala H, Smith LA, Smith TJ, Hill KK: Analysis of Clostridium botulinum serotype E strains by using multilocus sequence typing, amplified fragment length polymorphism, variable-number tandem-repeat analysis, and botulinum neurotoxin gene sequencing. Appl Environ Microbiol. 2011, 77: 8625-8634. 10.1128/AEM.05155-11.PubMedPubMed CentralView ArticleGoogle Scholar
- Chen Y, Korkeala H, Aarnikunnas J, Lindström M: Sequencing the botulinum neurotoxin gene and related genes in Clostridium botulinum type E strains reveals orfx3 and a novel type E neurotoxin subtype. J Bacteriol. 2007, 189: 8643-8650. 10.1128/JB.00784-07.PubMedPubMed CentralView ArticleGoogle Scholar
- Hill KK, Xie G, Foley BT, Smith TJ, Munk AC, Bruce D, Smith LA, Brettin TS, Detter JC: Recombination and insertion events involving the botulinum neurotoxin complex genes in Clostridium botulinum types A, B, E and F and Clostridium butyricum type E strains. BMC Biol. 2009, 7: 66-10.1186/1741-7007-7-66.PubMedPubMed CentralView ArticleGoogle Scholar
- Schiavo G, Matteoli M, Montecucco C: Neurotoxins affecting neuroexocytosis. Physiol Rev. 2000, 80: 717-766.PubMedGoogle Scholar
- Kalb SR, Garcia-Rodriguez C, Lou J, Baudys J, Smith TJ, Marks JD, Smith LA, Pirkle JL, Barr JR: Extraction of BoNT/A,/B,/E, and/F with a single, high affinity monoclonal antibody for detection of botulinum neurotoxin by Endopep-MS. PLoS One. 2010, 5: e12237-10.1371/journal.pone.0012237.PubMedPubMed CentralView ArticleGoogle Scholar
- Raphael BH: Exploring genomic diversity in Clostridium botulinum using DNA microarrays. Botulinum J. 2: 99-108.
- Richter M, Rosselló-Móra R: Shifting the genomic gold standard for the prokaryotic species definition. Proc Natl Acad Sci U S A. 2009, 106: 19126-19131. 10.1073/pnas.0906412106.PubMedPubMed CentralView ArticleGoogle Scholar
- Lúquez C, Bianco MI, de Jong LIT, Sagua MD, Arenas GN, Ciccarelli AS, Fernández RA: Distribution of botulinum toxin-producing clostridia in soils of Argentina. Appl Environ Microbiol. 2005, 71: 4137-4139. 10.1128/AEM.71.7.4137-4139.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Lúquez C, Raphael BH, Maslanka SE: Neurotoxin gene clusters in Clostridium botulinum type Ab strains. Appl Environ Microbiol. 2009, 75: 6094-6101. 10.1128/AEM.01009-09.PubMedPubMed CentralView ArticleGoogle Scholar
- Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S, MEGA5: Molecular Evolutionary Genetics Analysis using Maximum Likelihood, Evolutionary Distance, and Maximum Parsimony Methods. Mol Biol Evol. 2001, 28: 2731-2739.View ArticleGoogle Scholar
- Raphael BH, Joseph LA, McCroskey LM, Lúquez C, Maslanka SE: Detection and differentiation of Clostridium botulinum type A strains using a focused DNA microarray. Mol Cell Probes. 2010, 24: 146-53. 10.1016/j.mcp.2009.12.003.PubMedView ArticleGoogle Scholar
- Kalb SR, Baudys J, Rees JC, Smith TJ, Smith LA, Helma CH, Hill K, Kull S, Kirchner S, Dorner MB, Dorner BG, Pirkle JL, Barr JR: De novo subtype and strain identification of botulinum neurotoxin type B through toxin proteomics. Anal Bioanal Chem. 2012, 403: 215-26. 10.1007/s00216-012-5767-3.PubMedPubMed CentralView ArticleGoogle Scholar
- Raphael BH, Choudoir MJ, Lúquez C, Fernández R, Maslanka SE: Sequence diversity of genes encoding botulinum neurotoxin type F. Appl Environ Microbiol. 2010, 76: 4805-12. 10.1128/AEM.03109-09.PubMedPubMed CentralView ArticleGoogle Scholar
- Bashir A, Klammer AA, Robins WP, Chin CS, Webster D, Paxinos E, Hsu D, Ashby M, Wang S, Peluso P, Sebra R, Sorenson J, Bullard J, Yen J, Valdovino M, Mollova E, Luong K, Lin S, Lamay B, Joshi A, Rowe L, Frace M, Tarr CL, Turnsek M, Davis BM, Kasarskis A, Mekalanos JJ, Waldor MK, Schadt EE: A hybrid approach for the automated finishing of bacterial genomes. Nat Biotechnol. 2012, 30: 701-7. 10.1038/nbt.2288.PubMedPubMed CentralView ArticleGoogle Scholar
- Aziz RK, Bartels D, Best AA, DeJongh M, Disz T, Edwards RA, Formsma K, Gerdes S, Glass EM, Kubal M, Meyer F, Olsen GJ, Olson R, Osterman AL, Overbeek RA, McNeil LK, Paarmann D, Paczian T, Parrello B, Pusch GD, Reich C, Stevens R, Vassieva O, Vonstein V, Wilke A, Zagnitko O: The RAST Server: rapid annotations using subsystems technology. BMC Genomics. 2008, 9: 75-10.1186/1471-2164-9-75.PubMedPubMed CentralView ArticleGoogle Scholar