Evolution of Kallikrein-like enzymes in snakes and lizards

Kallikrein is a serine protease that was first discovered in the pancreas in great amounts and later was found in blood, saliva and urine as well (Raspi 1996). The kallikrein isolated from the venom of snakes and lizard has hydrolytic activity far higher than its orthologs in mammals (Ying et al., 2007). Kallikrein-like venoms are kinin-releasing enzymes. They release the potent vasodepressors bradykinin from kininogen by proteolytic action, and hydrolyze the vasopressor angiotensin II and its precursor angiotension I as well, resulting in the production of hypotension (Komori et al. 1998). Moreover, they increase vascular permeability and degrade the peptide chains of fibrinogen. This altered phenotype is critical to the fitness of the organisms as they become more effective in prey capture, defense and digestion. This report explores the hypothesis that the kallikrein-like venoms do not evolve neutrally, but undergo either positive or negative selection.

Thirteen sequences of kallikrein from venoms of lizard and snakes were chosen and aligned. Five paralogs from one snake species philodryas olfersii and sequences from four different Varanus lizards (monitor lizard) were included in the analysis. A phylogenetic tree was constructed (Figure 1). All the lizard kallikreins form a single clade and they share a common ancestor with the snake kallikreins, which is in agreement with the literature (Fry et al., 2006). When the kallikrein-like orthologs present in the venomous snake philodryas olfersii and venomous lizard Varanus varius were compared, the alignment indicates that the sequences are conserved in over 85% of the region. Moreover, it has already been reported that snake and lizard kallikrein toxins form a monophyletic group to the mammalian tissue (glandular) kallikrein, such as saliva glands (Fry 2005). This surprisingly high identity and the extremely close phylogeny between the orthologs suggest that the driving force of the evolution of kallikrein toxins is the modification of proteins already existed in ancestral salivary tissue rather than gene recruitment events (Fry 2005).

The pairwise alignment was conducted and dn/ds ration was calculated to detect selection forms. The codon-based test of positive selection does not detect any positive selection and the null hypothesis of neutrality cannot be rejected, whereas the codon-based test of purifying selection generates interesting results: no purifying selection was detected in any paralogs of snake philodryas olfersii genus, but all the Varanus lizard kallikrein seem to undergo negative selection. This purifying selection in Varanus lizard kallikrein suggests that mutations in these molecules are likely to be deleterious and therefore they can have severe phenotypic consequences on the survival of the organisms. To further confirm the result, whether the regions with reduced diversity in Varanus lizards are functional domains of kallikrein needs to be analyzed. Then, whether other selective regimes affect the evolution of snake kallikrein alone was investigated. Codon-by-codon analysis of snake kallikrein was carried out. Surprisingly, approximately three quarters of all the regions of the 5 paralogs show an overabundance of nonsynonymous changes, which are the signatures of local positive selection. However, this finding needs to be interpreted with caution because past population size and structure are not taken into account in the analysis. Demographic events can explain this positive selection in snake kallikrein rather than natural selection. Moreover, the codon-by-codon analysis loses power compared to the codon-based test that takes average across the whole sequence.

In order to detect the distinct features of the venom form of kallikrein relative to the other non-venomous proteases, the sequence of kallikrein in Gerrhonotus infernalis, which is a non-venomous alligator lizard, was chosen to compare with the toxins in venomous snakes and lizards. 24 conserved amino acid residues were observed within the venom clade and they are the candidates for functional-important residues. The kallikrein sequence of Gerrhonotus infernalis differs at 7 of the 24 potentially critical amino acids sites (Table 1). These amino acid changes are likely to be associated with the venom property of the protease. The serine to threonine or the opposite change is not so dramatic since the hydroxyl groups are retained. The isoleucine ↔ leucine switch does not change the property of the amino acid much either. However, whether these changes are capable of inducing venomous property is not clear. Proline at position 22, on the other hand, is a secondary structure breaker and therefore can change the three-dimensional conformation of the protein substantially. Nevertheless, this result needs to be analyzed with the knowledge of functional domains of kallikrein.

Another interesting observation is that when a phylogenetic tree was constructed including the kallikrein sequence from non-venomous lizard Gerrhonotus infernalis, it was found that this sequence is more closely related to all the lizard venomous kallikrein than to the whole venom clade (both snakes and lizards). This may indicate that either the common ancestors of all lizard could produce kallikrein-like venom and some of them lost the phenotype during the course of evolution or that the argument of a single early origin of the venom system in lizards and snakes does not hold: instead they evolve independently. In order to draw a solid conclusion, however, more sequences from close relatives of the venomous lizards and snakes need to be included.

Please see the uploaded files for figures and tables
Figure 1: phylogenetic tree of the kallikrein sequences from the venom clade (13 sequences)
Table 1: amino acid changes between Gerrhonotus infernalis and the venom clade (13 sequences)

1. Komori, Y. and T. Nikai. 1998. Chemistry and biochemistry of Kallikrein-like enzyme from snake venoms. TOC 17 (3): 261-277.
2. Fry, B.G., Vidal, N., Norman, J.A., Vonk, F.J., Scheib, H. et al. 2006. Early evolution of the venom system in lizards and snakes. Nature 439 (2): 584-588.
3. Zhang, Y., Ma, Biao, Lin, Y.Q. and Y. Luo. 2007. Long-term and acute toxicity of kallikrein from the venom of Agkistrodon hlays Pallas. J South Med Univ 27 (11): 1756-1758.
4. Fry, B.G. 2005. From genome to “venome”: molecular origin and evolution of the snake venom proteome inferred from phylogenetic analysis of toxin sequences and related body proteins. Genome Res. 15 (3): 403-20.