| dc.description.abstract | Sulfolobus acidocaldarius DSM 639 is a thermoacidophilic crenarchaeon which can grow in hot springs at 55-85oC and habitat in acidic environments (pH 2-3). The ability to adapt and proliferate in hostile environments compared with other organisms makes it an interesting source for novel extra-cellular proteins. However, the origin of its unusual thermal stability is not fully understood. Via the proteomic approaches, a total of extracellular 184 proteins were identified, representing 24% of the S. acidocaldarius DSM 639 theoretical extra-cellular proteome (exoproteome).Those proteins include 123 proteins from the gel-based system, 23 proteins from the gel-free system. The results demonstrate that the majority of the identified extracellular proteins of S. acidocaldarius DSM 639 were classified as hypothetical proteins (72 unique proteins, 39%), followed by the extracellular enzymes (35 unique proteins, 19%), protein related to ATP binding and transport (25 unique proteins, 14%). Particularly, there are 84 unique proteins (49%) previously not annotated to S. acidocaldarius. These special proteins are identified to 44 archaea respectively. By analyzing the phylogenetic tree with the 44 archaeal species and S. acidocaldarius, there are 24 species (9 hyperthermophiles, 3 thermophiles and 12 mesophiles) belong to Euryarchaeota and 15 species (11 hyperthermophiles and 4 thermophiles) belong to Crenarchaeota, the two major archaeal phyla. Three species (thermophiles) are from the uncultured archaeal phylum Korarchaeota and one species (psychrophiles) is belong to the newly established phylum Thauarchaeota. With the highly homology-conserved archaeal exoproteome between those archaeal species, three rare translation initial codons in prokaryotes, TTG (L), GTT (V) and ACG (T) were found in the 20 proteins from the exoproteome of S. acidocaldarius. This result corresponds to the highlighted phenomenon in recent years, the predicted genes exhibit frequent errors, particularly in start codons. Representing the high-throughput proteomic study provides the annotation correction by combining original proteomic data and comparative genomics. Combination with the previous studies in our team, three species, Aeropyrum pernix K1, Geobacillus kaustophilus ATCC8005 and Thermospora fusca YX, are also predicted by the subcellular localization prediction tools. CELLOv.2.5, PSORTbv.3.0, PRED-SIGNAL, Proteome Analystv.3.0, LocateP, to validate the specificity and recall (Mattews Correlation Coefficient, MCC) and compare with the protein family and domain from Pfam 24.0. It have been found that CELLO.2.5 (Recall: -0.01) and PRED-SIGNAL (Recall: -0.13) are closed to the random predictions (Recall= 0), PSORTbv.3.0 (Recall: 0.45) and Proteome Analyst v.3.0 (Recall: 0.77) are closed to the accurate predictions (Recall= 1). From the results to check PSORTb is the suitable tool for SCL prediction, but still with too much unknown functional results to mislead the extracellular proteins.
| en_US |