Cloning Of The Major Toxin Genes In Amanita Exitialis And Studies On The Diversities Of Amanita Toxin Gene Families

The lethal Amanita species contain potent toxic peptide toxins, which are responsible fora large number of mushroom poisoning cases. The studies on Amanita toxin encoding genesand the characterizations of the toxin gene families in these mushrooms remain largelyunknown, but which are of great significance for exploring the rapid detection technology ofthe Amanita toxins, poisoning treatment and discovering new drugs. Thus, to clone moreAmanita toxin genes, clarify the expression features of the major toxin encoding genes, andstudy the diversities of the toxin gene family members are the key scientific problems neededto resolve.The studies were performed as followed: cloning and analyzing the full-lengthcDNA sequences of major toxin genes in the endemic lethal Amanita species in East Asia-Amanita exitialis, characterizing of its expression pattern in different parts and developmentstages of A. exitialis fruitbody, de novo sequencing of the A. exitialis transcriptome, andelucidating the diversities of the toxin gene family members of the6common lethal Amanitaspecies. Main results are listed:1. The full-length cDNA sequences of α-amanitin (AeAMA) and phallacidin (AePHA) inA. exitialis were firstly cloned and analyzed, their full-length cDNA sequences was356bpand321bp, respectively, and the predicted proproteins of AeAMA and AePHA were35and34amino acids respectively. Making a comparison of the cloned, full-length cDNA sequences tothe genomic DNA sequences indicated that both the AeAMA and AePHA have the intron (58nt and56nt, respectively) that interupts the4th from the last codon. The toxin genes of A.extitialis and A. bisporigera shared high DNA sequences and amino acid sequences consensusrates, while except for the toxin region itself and the amino acids immediately upstream, ahigh divergence was found between the Amanita species and Galerina species.2. The stabilities of the5reference genes (60S, GAPDH, NADH, RPB2, and β-actin) inthe different development stages and parts of A. exitialis were evaluated firstly; the β-actinwas selected as the most reliable reference genes for the further study. The expression patternsof α-AMA in selected parts of the fruitbody as well as each developmental stage weresubsequently characterized by using quantitative real-time PCR firstly. Results showed thatthe α-AMA could be expressed in all selected parts and stages of the fruitbody, but the expression levels of α-AMA transcript in these selected parts of the4stages were inequable.Among the selected stages, the whole fruitbodies with a higher expression level in the earlyelongation stage than other3stages; except for the early elongation stage, α-AMA showed thehighest expression level in the pileus of the other3stages; further analysis revealed that theexpression level of α-AMA was closely associated with the development stages, and theexpression of α-AMA was accumulated in the more vigorous growth parts and stages.3. Transcriptome sequencing of A. exitialis was performed using Illumina HiSeq2000technology. A total of25,563,688clean reads (average length of90bp each) were collected,the Q20and GC percentages were93.89%and51.58%, respectively. All high-quality readswere de novo assembled using the Trinity program and were assembled into62,137cDNAcontigs with an average length of481bp and N50length of788bp. A total of39,661unigenes were identified among the assembled contigs, with an average length of662bp andN50length of862bp.27,848unigenes were annotated,27,826coding sequences (CDS) wereidentified,4,269full-length sequences with intact open reading frame (ORF) were found;1,254simple sequence repeat (SSR) were found among1,164unigenes, and the trinucleotiderepeats with the highest rate (48.6%). Functional information, protein sequence similarity,Kyoto Encyclopedia of Genes and Genomes (KEGG), Clusters of Orthologous Groups (COG),and Gene Ontology (GO) information were provided from the unigene annotations. From thisset,11gene sequences encoding the toxins or related cyclic peptides were discovered in thetranscriptome. Four of these sequences matched the peptide toxins α-amanitin, β-amanitin,phallacidin, and amanexitide, while7others matched the unknown peptides. All of the genesencoding peptide toxins were confirmed by reverse transcript PCR in A. exitialis. Thepolymorphism of the toxin gene sequences were discovered, more variant sites were foundbetween the different genera than the sequences of the same genus. The phylogeneticrelationships among the proprotein sequences of the MSDIN family members showed thathigher bootstraps were found among the same kind of toxin sequences.4. Fifty-four gene sequences were obtained from the6lethal Amanita species collectedfrom the Asia and Europe, accounting for70.1%of the known MSDIN family members. Ofthe54gene sequences,20encode α-amanitin,5encode β-amanitin,16encode phallacidin,and13encode peptides of unknown functions. This result further demonstrated that the different Amanita species contain different MSDIN family members. Bayesian analysis ofMSDIN family members showed that10of the13peptides of unknown functions wereclosely related to known phallotoxins, while the remaining3were similar to amatoxins. Byaligning core toxin gene sequences of the α-amanitin and phallacidin, gene polymorphismswere found, there were4and3degenerate sites respectively in the both toxins’ coresequences, all these sites were found at the3rd base of the codon, and the codon usage biaswere calculated at each polymorphic site. Using the PAML4.7a, the dN/dS ratios of theencoding sequences of α-AMA and PHA were calculated, results showed that the α-AMA andPHA should be under the purifying selection. The Bayesian results of α-AMA，PHA, and ITSshowed that the major toxin encoding sequences were conserved at the genus level, the toxinencoding sequences could be used to study the phylogenetic relationships among these toxinproducing species.