Lineage-Specific Domain Fusion in the Evolution of Purine Nucleotide Cyclases in Cyanobacteria


DOI: 10.1007/s00239-008-9127-z

Cite this article as:
Wu, J., Bai, J., Bao, Q. et al. J Mol Evol (2008) 67: 85. doi:10.1007/s00239-008-9127-z


Cyclic nucleotides (both cAMP and cGMP) play extremely important roles in cyanobacteria, such as regulating heterocyst formation, respiration, or gliding. Catalyzing the formation of cAMP and cGMP from ATP and GTP is a group of functionally important enzymes named adenylate cyclases and guanylate cyclases, respectively. To understand their evolutionary patterns, in this study, we presented a systematic analysis of all the cyclases in cyanobacterial genomes. We found that different cyanobacteria had various numbers of cyclases in view of their remarkable diversities in genome size and physiology. Most of these cyclases exhibited distinct domain architectures, which implies the versatile functions of cyanobacterial cyclases. Mapping the whole set of cyclase domain architectures from diverse prokaryotic organisms to their phylogenetic tree and detailed phylogenetic analysis of cyclase catalytic domains revealed that lineage-specific domain recruitment appeared to be the most prevailing pattern contributing to the great variability of cyanobacterial cyclase domain architectures. However, other scenarios, such as gene duplication, also occurred during the evolution of cyanobacterial cyclases. Sequence divergence seemed to contribute to the origin of putative guanylate cyclases which were found only in cyanobacteria. In conclusion, the comprehensive survey of cyclases in cyanobacteria provides novel insight into their potential evolutionary mechanisms and further functional implications.


Cyanobacteria Domain fusion Evolutionary mechanism Purine nucleotide cyclase 

Supplementary material

239_2008_9127_MOESM1_ESM.doc (535 kb)
Purine nucleotide cyclases identified in prokaryotic genomes. (DOC 535 kb)
239_2008_9127_MOESM2_ESM.tif (3.7 mb)
Multiple sequence alignments of the catalytic domains of 51 cyan bacterial cyclases. Substrate specifying residues of the cyclases are marked with asterisks. Metal binding residues of cyclases are marked with circles. Transition state stabilizing residues of cyclases are marked with triangles. Two cyclases (sll1161 and cwat1781), encoded in Synechocystis sp. PCC 6803 and Crocosphaera watsonii WH8501 are found to have either adenylate or guanylate-related substrate specifying residues. (TIF 3,815 kb)
239_2008_9127_MOESM3_ESM.tif (1.6 mb)
An example of cyclase gene neighborhoods in different Prochlorococcus and Synechococcus. Two strains of Synechococcus and one strain of Prochlorococcus are found to contain cyclase, whereas the other strains of Synechococcus and Prochlorococcus may lose this gene. (TIF 1,597 kb)
239_2008_9127_MOESM4_ESM.tif (729 kb)
Functional categories of non-cyclase proteins which share the additional domains with the cyclases from 21 cyanobacterial genomes. The functional categories of non-cyclase proteins are obtained by comparing them to the COG database using BLAST program with an E-value of 10−5. If one can not be assigned to any COG functional category, it was labeled with “No hit”. It’s also notable that some proteins may be classified into multiple categories. (TIF 730 kb)

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  1. 1.Institute of Biomedical Informatics/Zhejiang Provincial Key Laboratory of Medical Genetics, Wenzhou Medical CollegeWenzhouChina
  2. 2.Institute of Oceanology, Chinese Academy of SciencesQingdaoChina
  3. 3.Department of Biochemistry and Molecular BiologyPennsylvania State UniversityUniversity ParkUSA