Chao Y et al. [1] recently reported a novel serpin with antithrombin-like (ATL) activity in Branchiostoma japonicum. In the summary they concluded exaggeratedly, in my opinion, that their study suggested that a primitive coagulation system had already emerged in the protochordate as the amphioxus had a homologue (ortholog?) of vertebrate antithrombin (AT). In this letter I am going to justify their claim after a brief NCBI BLAST, Clustal 2.1 and MrBayes scan of Branchiostomidae and other early emerged deuterostomes. As an introduction to the problem I mention that it has been known for a long time about clotting of colorless lancelet hemolymph and that probably leucocyte-like clumping cells are here responsible for, but no extracellular gelation was observed. [2]. Nothing is known about molecular mechanisms of real working coagulation of amphioxus with any "coagulation factors". Hypothetically, it is possible that there could be some zymogen serine proteases that would activate lancelet "thrombocytes". The other thing is that in hypothesis of simple prevertebrate clotting the most important role is not clotting itself but immobilizing and bonding pathogens by coagulation. We have concluded in our recent paper that in lancelet and other simple deuterostomes are probably present prothrombin-like proteases (PTL), quite related to prothrombin (PT), but multiplied by duplications after divergence of cephalochordates from tree trunk of vertebrate ancestors [3]. They exhibit certain characteristics of thrombin as proteases, but do not have typical thrombin N-terminal domains. I see here a similar situation for B. japonicum antithrombin-like protease. Before going to the main case I would like to mention that when I probationary blasted a fragment of BjATI sequence taken from the article alignment (MAMTYMGARHNTAEQMAAVLHLSEGDVHQAF) the only 100% identical sequence was gb|ABW74215.1| antithrombin-like protein [Branchiostoma belcheri tsingtauense] as also presented in supplementary file with B. belcheri name used in table 2. I have found that according to publication of Zang QJ et al B. belcheri and B. japonicum are separate species [4]. The question to the authors is therefore what species was examined in the study? B. japonicum or B. belcheri? Going back to the main point, I conducted afterwards non-redundant protein NCBI-BLAST search with ABW74215.1 sequence against Branchiostomidae, knowing that B. floridae genome had been known since 2008. Quite interesting results were obtained by my survey (supplementary material). The results returned 5 and 7 different antithrombin-like sequences for B. floridae and B. lanceolatum, respectively (only one for B. belcheri - ABW74215.1, of course 100% identical) . Some of them were more than 50% similar to each other in the same species. I also run and an additional search of sea squirt, acorn worm and sea urchin with B. belcheri sequence. All found sequences were merged with some chosen sequences of vertebrates according to authors’ supplementary material and aligned by ClustalW 2.1 on my computer to generate phylogenetic tree. The optional tree was generated by Mrbayes-3.1.2 from ClustalW nxs output. Both Clustal and MrBayes trees have revealed that all lancelet paralogs are much more closer to each other than to vertebrate AT. This coincides very nicely with a similar situation concerning thrombin homologues of B. floridae, but it cannot be judged that “pritimitive coagulation system had already emerged in the protochordate”. It might be a completely different system in lancelets which could appeared independently from similar building blocks after independent duplications in lancelet ancestors. Moreover, it is not known whether and which of these ATL and PTL copies could take part in it. Only thorough deep research on the biochemistry and molecular biology of blood coagulation of amphioxus this problem would be solved. The last biochemical studies were taken probably in 60’s and maybe 70’s of the last century [5]. Nowadays methods are much more sophisticated.

In summary, there is together a very interesting conclusion from work of Chao Y et al. and mine (apart of the problem of serine protease primitive coagulation of protochordate ancestors of vertebrates). It seems that the ancestors of thrombin could evolve along with its inhibitors, in prevertebrate protochordates as well as in lancelet. I wish I could relate to these issues in my article referring to the paper of Chao Y et al. Unfortunately, my paper was submitted for publication a month before the release of the online version of Chao Y et publication.

References

1. Chao Y, Fan C, Liang Y, Gao B, Zhang S. (2012) A novel serpin with antithrombin-like activity in Branchiostoma japonicum: implications for the presence of a primitive coagulation system. PLoS One. 7:e32392
2. Doolittle RF. (1993) The evolution of vertebrate blood coagulation: a case of Yin and Yang. Thromb Haemost. 70:24-28
3. Zhang QJ, Zhong J, Fang SH, Wang YQ. (2006) Branchiostoma japonicum and B. belcheri are distinct lancelets (Cephalochordata) in Xiamen waters in China. Zoolog Sci. 23:573-579
4. Ponczek MB, Bijak MZ, Nowak PZ.(2012) Evolution of thrombin and other hemostatic proteases by survey of protochordate, hemichordate, and echinoderm genomes. J Mol Evol. 74:319-331
5. Doolittle RF: The comparative biochemistry of blood coagulation. Ph.D. thesis, Harvard University, 1961

Supplementary material:
http://taxus.biol.uni.lod...

Michal B Ponczek

Department of General Biochemistry, Faculty of Biology and Environmental Protection, University of Lodz
Correspondence to Michal B. Ponczek, Department of General Biochemistry, Faculty of Biology and Environmental Protection, University of Lodz, Pomorska 141/143, 90-236 Lodz, Poland; Tel. + 48 42 635 44 82; Fax + 48 42 635 44 84