One-Step Affinity Purification of Leucine-Rich α2-Glycoproteins from Snake Sera and Characterization of Their Phospholipase A2-Inhibitory Activities as β-Type Phospholipase A2 Inhibitors

Snakes contain three types of phospholipase A2 (PLA2)-inhibitory proteins in their blood, PLIα, β, and γ, which protect them from their own venom, PLA2. PLIβ is the snake ortholog of leucine-rich α2 glycoprotein (LRG). Since autologous cytochrome c (Cyt c) serves as an endogenous ligand for LRG, in this study, we purified snake LRGs from various snake serum samples using Cyt c affinity chromatography. All purified snake LRGs were found to be dimers linked by disulfide bonds. Laticauda semifasciata and Naja kaouthia LRGs showed no inhibitory activity against L. semifasciata PLA2 and weak inhibitory activity against Gloydius brevicauda basic PLA2. Elaphe climacophora PLIβ had weaker inhibitory activity against G. brevicauda basic PLA2 than G. brevicauda and Elaphe quadrivirgata PLIs, which are abundant in blood and known to neutralize G. brevicauda basic PLA2. Protobothrops flavoviridis LRG showed no inhibitory activity against basic venom PLA2, PL-X, or G. brevicauda basic PLA2. Binding analysis of P. flavoviridis LRG using surface plasmon resonance showed very strong binding to snake Cyt c, followed by that to horse Cyt c, weak binding to yeast Cyt c, and no binding to P. flavoviridis PL-X or BPI/II. We also deduced the amino acid sequences of L. semifasciata and P. flavoviridis LRG by means of cDNA sequencing and compared them with those of other known sequences of PLIs and LRGs. This study concluded that snake LRG can potentially inhibit basic PLA2, but, whether it actually functions as a PLA2-inhibitory protein, PLIβ, depends on the snake.


Introduction
Snakes possess self-protective mechanisms against venom.Three distinct phospholipase A 2 (PLA 2 )-inhibitory proteins, named PLIα, PLIβ, and PLIγ, have been purified from the serum of the Chinese mamushi, Gloydius brevicauda (renamed Agkistrodon blomhoffii siniticus according to the present taxonomy) [1].To date, PLIα and PLIγ have been reported in many snakes [2][3][4].These inhibitory proteins directly bind to venom PLA 2 s and inhibit their enzymatic activity.PLIαs are generally found in the sera of viperid snakes, and their structure is characterized by the 20 kDa glycosylated subunit with a C-type lectin-like domain [2,5].They specifically inhibit group II acidic PLA 2 s [6] or basic myotoxic PLA 2 s [7].PLIγs can inhibit a broad spectrum of group I, II, and III PLA 2 s [6,8,9], and their primary structures contain two tandem repeats of a cysteine pattern characteristic of the three-finger motifs found in Ly-6-related proteins [10].PLIγ have been identified in the sera of many venomous snakes, including elapid and viperid snakes, as well as non-venomous snakes [2].
PLIβ has nine tandem leucine-rich repeats in its sequence and 33% sequence homology with human leucine-rich α 2 -glycoprotein (LRG) [11].Human LRG is one of the serum proteins of unknown function and has recently accumulated evidence for its contribution to a wide range of human diseases [12].Snake PLIβ selectively inhibits group II basic PLA 2 s and is thought to function as part of a self-defense mechanism against its own venom [1].It was purified from the sera of the viperid snake, G. brevicauda [1] and two non-venomous snakes, Elaphe quadrivirgata [13] and Elaphe climacophora [14], using multi-step purification including ion exchange, gel filtration, and hydrophobic column chromatography.PLIβ transcripts have been found in cDNA libraries from the venom glands and livers of the viperid snakes Lachesis muta [15], Bothrops alternatus, Bothrops jararaca, Bothrops moojeni, Bothrops neuwiedi, and Crotalus durissus terrificus [16].However, PLIβ has not been reported in an elapid snake.Autologous cytochrome c (Cyt c) is the endogenous ligand of LRG and PLIβ [17], and PLIβ can be regarded as the snake LRG with PLA 2 -inhibitory activity.Owing to this, LRGs can be easily purified from serum using a Cyt c affinity column [17,18].
In the present study, we purified snake LRGs from various snake sera using a Cyt c affinity column and investigated whether snake LRGs generally have PLA 2 -inhibitory activity and function as a defense mechanism against snake venom PLA 2 .

Purification of LRGs from Various Snake Sera
Snake PLIβ and human LRG are homologous serum glycoproteins with Cyt c-binding activity, indicating that PLIβ is the snake ortholog of LRG [17].In fact, the PLIβ cDNA sequence was blasted against the blue-ringed sea krait genome using the Ensemble Genome Browser and its synteny was found to be almost identical to that of the human LRG gene, indicating that the PLIβ gene is the same gene as the LRG gene in mammals.Because it has been reported that LRG can be easily purified using Cyt c affinity columns [17,18], we attempted to purify snake LRGs from various snake sera using the same technique.LRGs from the sera of Laticauda semifasciata and Protobothrops flavoviridis that were specifically adsorbed on a Cyt c affinity column and eluted with 0.1 M acetate buffer (pH 4.0) containing 0.5 M NaCl showed almost a single 50 kDa band on the sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) (Figure 1a,b).Similarly, we also purified snake LRGs from the sera of G. brevicauda, E. climacophora, and Naja kaouthia.The molecular weight of the purified snake LRGs was approximately 50 kDa in the SDS-PAGE in the presence of β-mercaptoethanol, but around 100 kDa in its absence, suggesting that these are dimers linked by disulfide bonds (Figure 1c).Since no monomeric bands are detected under non-reducing conditions, this is probably not an artificial dimerization by purification.The N-terminal 10-amino acid sequences of the purified G. brevicauda and E. climacophora LRGs were consistent with the previously determined PLIβ sequences [11,14].The N-terminal amino acid sequences of L. semifasciata, N. kaouthia, and P. flavoviridis LRGs were determined to be Val-Leu-Tyr-X-Pro-Pro-Asp-Pro-Ala, Val-Leu-Tyr-X-Pro-Pro-X-Pro-Ala-Pro-Glu, and Val-Leu-Tyr-X-Pro-Pro-Thr-Pro-Ala-Pro-Glu-Ser-Val-Thr-Glu-Phe-Val-X-Asn-Ser, respectively.The use of Cyt c affinity columns makes it possible to purify PLIs from serum in a single step.Table 1 shows the amount of snake LRG recovered from sera using the Cyt c affinity column.The G. brevicauda serum contained large amounts of LRG (PLIβ), as compared to the other snake sera.[19].G. brevicauda and E. quadrivirgata PLIβ specifically inhibit group II basic PLA 2 s, such as G. brevicauda basic PLA 2 and P. flavoviridis PL-X, but do not inhibit group II acidic and neutral PLA 2 from G. brevicauda venom, group I PLA 2 from Naja atra venom, and group III PLA 2 from honeybee venom [1,13].Elapidae LRGs purified from the sera of L. semifasciata and N. kaouthia did not inhibit the group I PLA 2 , named PLA-I, which was purified from L. semifasciata venom (Figure 2a) but weakly inhibited group II G. brevicauda basic PLA 2 (Figure 2b), with an apparent inhibition constant of 1.03 and 0.35 µM, respectively.Thus, Elapidae LRG does not function as the PLA 2 -inhibitory protein, PLIβ.E. climacophora PLIβ inhibited G. brevicauda basic PLA 2 but was weaker than G. brevicauda and E. quadrivirgata PLIβ, with an apparent inhibition constant of 11.1 nM.P. flavoviridis LRG hardly inhibited G. brevicauda basic PLA 2 and did not inhibit its own venom PLA 2 , PL-X, although G. brevicauda and E. quadrivirgata PLIβ inhibited P. flavoviridis PL-X, with apparent inhibition constants of 26.1 and 66.0 nM, respectively (Figure 2c).Furthermore, P. flavoviridis LRG also did not inhibit G. brevicauda acidic PLA 2 and neutral PLA 2 .This indicated that P. flavoviridis LRG does not function as a PLA 2 -inhibitory protein.Since G. brevicauda PLIβ purified using a Cyt c affinity column in this study showed the same level of PLA 2 inhibitory activity as the previously reported PLIβ purified by multi-step chromatography, it seems that the PLA 2 inhibitory activity of P. flavoviridis LRG was not lost due to the difference in purification methods.

Surface Plasmon Resonance Analyses of the Interaction of P. flavoviridis LRG with Various Cyt c and PLA 2
To identify the endogenous ligand of P. flavoviridis LRG, we immobilized it on a biosensor chip and measured its interactions with Cyt c and PLA 2 .As shown in Figure 3, the sensorgrams indicated that P. flavoviridis LRG had a very high affinity for horse and snake Cyt c, with a binding manner characterized by fast association and slow dissociation rates.The dissociation constant (K d ) was calculated to be 64.5 pM for horse Cyt c, 14.9 pM for L. semifasciata Cyt c, and 1.67 µM for yeast Cyt c.In contrast, P. flavoviridis LRG did not bind to G. brevicauda basic PLA 2 or P. flavoviridis PL-X, as expected based on the results of the PLA 2 -inhibitory activity.Because P. flavoviridis venom contains myotoxic Lys-49-PLA 2 homologues, BPI and BPII, we also examined the binding of P. flavoviridis LRG to BPI/II, but no binding was detected.The species-specificity of P. flavoviridis LRG, which binds more strongly to snake Cyt c than to horse Cyt c, is consistent with a previous study on G. brevicauda PLIβ [17], indicating that the endogenous ligand of LRG is autologous Cyt c rather than its venom PLA 2 s.  [13]); ■, Elaphe climacophora PLIβ; △, Naja kaouthia LRG; ▲, L. semifasciata LRG.

Surface Plasmon Resonance Analyses of the Interaction of P. Flavoviridis LRG with Various Cyt c and PLA2
To identify the endogenous ligand of P. flavoviridis LRG, we immobilized it on a biosensor chip and measured its interactions with Cyt c and PLA2.As shown in Figure 3, the sensorgrams indicated that P. flavoviridis LRG had a very high affinity for horse and snake Cyt c, with a binding manner characterized by fast association and slow dissociation rates.The dissociation constant (Kd) was calculated to be 64.5 pM for horse Cyt c, 14.9 pM for L. semifasciata Cyt c, and 1.67 µM for yeast Cyt c.In contrast, P. flavoviridis LRG did not bind to G. brevicauda basic PLA2 or P. flavoviridis PL-X, as expected based on the results of the PLA2-inhibitory activity.Because P. flavoviridis venom contains myotoxic Lys-49-PLA2 homologues, BPI and BPII, we also examined the binding of P. flavoviridis LRG to BPI/II, but no binding was detected.The species-specificity of P. flavoviridis LRG, which binds more strongly to snake Cyt c than to horse Cyt c, is consistent with a previous study on G. brevicauda PLIβ [17], indicating that the endogenous ligand of LRG is autologous Cyt c rather than its venom PLA2s.S2).LRGcDNA1 and LRGcDNA2 differed from Pf LRG-A by one and four bp, respectively, in terms of the nucleotide sequence, resulting in one and two amino acid substitutions, respectively.The nucleotide sequence of LRGcDNA3 was identical to that of Pf LRG-B, whereas that of LRGcDNA4 differed from that of Pf LRG-B by five bp, resulting in two amino acid substitutions.Thus, LRGcDNA1 and LRGcDNA2 were expected to be variants derived from the Pf LRG-A gene, while LRGcDNA3 and LRGcDNA4 were expected to be variants derived from the Pf LRG-B gene.Since both the Pf LRG-A and Pf LRG-B genes are expressed in the liver, we believe that the purified P. flavoviridis LRG is a mixture of LRG-A and LRG-B, just as E. quadrivirgata PLIβ was a mixture of PLIβ-A and PLIβ-B [13].

cDNA Cloning of P. flavoviridis and L. semifasciata LRGs
The full-length cDNA sequence of L. semifasciata LRG was determined by means of direct sequencing of the 3′ and 5′ rapid amplification of cDNA ends (RACE) products from sea snake liver cDNA (Supplementary Figure S1).L. semifasciata LRG cDNA was 2290 bp (GenBank accession no.LC786336), including 46 bp 5′-untranslated region, 996 bp of the complete coding sequence, and 1145 bp 3′-untranslated region.The predicted open reading frame encodes a protein of 331 amino acids, including a signal peptide consisting of 23 amino acids.Using oligonucleotide primers based on the sequence of G. brevicauda PLIβ, we obtained cDNA clones of LRG (LRGcDNA1-4) with four different sequences from reverse transcripts of P. flavoviridis liver total RNA (GenBank accession no.LC786337-LC786340).A BLAST search of the nucleotide sequences of these four cDNAs against a draft genome sequence of the habu snake [20] retrieved two LRG genes, which we named PfLRG-A and PfLRG-B.The two LRG genes are located on different chromosomal scaffolds, not tandem duplications, and probably resulted from an ancient duplication event.The nucleotide sequences of PfLRG-A and PfLRG-B differed by 62 bp of the

Sequence Comparison and Molecular Phylogenic Tree of Snake LRGs
The deduced amino acid sequences of LRGs from L. semifasciata and P. flavoviridis were compared with those of PLIs from E. climacophore, E. quadrivirgata, and G. brevicauda [11,13,14], and those of sbβPLIs from C. durissus terrificus, B. alternatus, and B. jararaca [16] (Figure 4).The amino acid sequence of L. semifasciata LRG reported here is the first in the Elapidae family and showed 85-86% homology to those of Colubridae LRGs and 72-77% homology to those of Viperidae LRGs.Cysteine residues are conserved in all snake LRGs and based on conformational predictions obtained using AlphaFold2 [21], disulfide bonds were predicted to form between Cys3 and Cys18, Cys261 and Cys287, and Cys263 and Cys306.Cys210, which is specific to Elapidae and Colubridae LRGs, and Cys147 were predicted to occur as free thiol groups, because they are located within the LRR structure.Cys190 and Cys309, which are absent in mammalian LRGs, are likely involved in snake LRG dimer formation.A phylogenetic tree based on the nucleotide sequences encoding mature LRGs was constructed using the neighbor-joining method (Figure 5).Two LRGs, LRG-A and LRG-B, seem to have been generated by gene duplications since the divergence of the two venomous snake families.Both genes are expressed in the liver of P. flavoviridis, but only LRG-B genes are expressed as sbβPLIs in the livers of Latin American Viperidae snakes.Although sbβPLIs from Latin American Viperidae snakes are presumed to inhibit basic PLA 2 from their own venoms, these have never been purified or characterized; therefore, whether they have PLA 2 -inhibitory activity remains questionable.A phylogenetic tree based on the nucleotide sequences encoding mature LRGs was constructed using the neighbor-joining method (Figure 5).Two LRGs, LRG-A and LRG-B, seem to have been generated by gene duplications since the divergence of the two venomous snake families.Both genes are expressed in the liver of P. flavoviridis, but only LRG-B genes are expressed as sbβPLIs in the livers of Latin American Viperidae snakes.Although sbβPLIs from Latin American Viperidae snakes are presumed to inhibit basic PLA2 from their own venoms, these have never been purified or characterized; therefore, whether they have PLA2-inhibitory activity remains questionable.

Discussion
By utilizing the strong binding of LRG to Cyt c, various snake LRGs can be purified from snake sera in one step, using a Cyt c affinity column.Mammalian LRGs are monomers, whereas all snake LRGs purified in the present study were dimers formed by disulfide bonds.Our previous study reported that G. brevicauda and E. quadrivirgata PLIβ is a trimer based on its molecular weight of 160 kDa from gel-filtration and the results of crosslinking experiments; however, the present results lead us to believe that it is a dimer rather than a trimer, since the molecular weight of LRR proteins upon gel-filtration tends to be larger than the actual molecular weight [22] and the 120 kDa band in the SDS-PAGE of the crosslinking experiment is likely not a trimer.In the previous multi-step purification procedure, only 0.1 mg of LRG (PLIβ) was purified per 10 mL of E. climacophora serum [12]; however, in the current one-step procedure using the Cyt c affinity column, the yield was 0.4 mg.Purification by means of one-step chromatography rather than multi-step chromatography is expected to significantly increase the recovery of LRG, thereby allowing for estimation of the amount of LRG in the serum.Table 1 shows the recovery of LRG purified from 10 mL snake sera in the present study.Although 0.16 mg of E. quadrivirgata PLIβ could be purified from 2 mL of serum using a multi-step purification procedure [13], it is expected that several milligrams of PLIβ can be obtained from 10 mL of serum, if purified using the present one-step purification procedure.Unlike E. climacophora, E.

Discussion
By utilizing the strong binding of LRG to Cyt c, various snake LRGs can be purified from snake sera in one step, using a Cyt c affinity column.Mammalian LRGs are monomers, whereas all snake LRGs purified in the present study were dimers formed by disulfide bonds.Our previous study reported that G. brevicauda and E. quadrivirgata PLIβ is a trimer based on its molecular weight of 160 kDa from gel-filtration and the results of crosslinking experiments; however, the present results lead us to believe that it is a dimer rather than a trimer, since the molecular weight of LRR proteins upon gel-filtration tends to be larger than the actual molecular weight [22] and the 120 kDa band in the SDS-PAGE of the crosslinking experiment is likely not a trimer.In the previous multi-step purification procedure, only 0.1 mg of LRG (PLIβ) was purified per 10 mL of E. climacophora serum [12]; however, in the current one-step procedure using the Cyt c affinity column, the yield was 0.4 mg.
Purification by means of one-step chromatography rather than multi-step chromatography is expected to significantly increase the recovery of LRG, thereby allowing for estimation of the amount of LRG in the serum.Table 1 shows the recovery of LRG purified from 10 mL snake sera in the present study.Although 0.16 mg of E. quadrivirgata PLIβ could be purified from 2 mL of serum using a multi-step purification procedure [13], it is expected that several milligrams of PLIβ can be obtained from 10 mL of serum, if purified using the present one-step purification procedure.Unlike E. climacophora, E. quadrivirgata has an ophiophagous habit and often feeds on the Japanese mamushi Gloydius blomhoffii [23].Therefore, E. quadrivirgata PLIβ probably functions as a defense against envenomation.Thus, the higher amounts of G. brevicauda and E. quadrivirgata PLIβ in the serum, compared to the amounts of LRG/PLIβ in the serum of other snakes, are probably due to the protective

Purification of LRGs from Various Snake Sera
LRGs were purified from various snake serum samples using a Cyt c affinity column.Horse heart Cyt c was coupled to a HiTrap NHS-activated HP column (GE Healthcare, Buckinghamshire, UK) to produce a Cyt c affinity column, as described previously [17].Sera from various snakes were loaded into the Cyt c affinity column, equilibrated with 20 mM sodium phosphate buffer (pH 7.4) containing 0.15 M NaCl, and washed with the same buffer.LRG was eluted with 0.1 M sodium acetate buffer (pH 4.0) containing 0.5 M NaCl.SDS-PAGE was analyzed under reducing (5% β-mercaptoethanol) and non-reducing conditions, using a 10% polyacrylamide gel.Protein bands were stained with Quick CBB (Wako Pure Chemical Industries).The N-terminal sequences of the LRGs were determined using a protein sequencer (491HT; Applied Biosystems, Foster City, CA, USA).

Inhibition of PLA 2 Enzymatic Activity
PLA 2 activity was measured fluorometrically using 1-palmitoyl-2-(10-pyrenyldecanoyl)sn-glycero-3-phosphoryl-choline (10-Pyrene-PC; Cayman Chemical, Ann Arbor, MI, USA) as a substrate, as described previously [6] in the presence of various concentrations of LRGs.The apparent inhibition constant (K i app ) was determined by means of non-linear least-squares analysis of relative PLA 2 activity.

Binding Analysis Using Surface Plasmon Resonance
The BIAcore ® X System (GE Healthcare) was utilized to study the interaction of P. flavoviridis LRG with Cyt c and PLA 2 .P. flavoviridis LRG was coupled to a CM5 sensor chip using an amine coupling kit according to the manufacturer's instructions.Binding experiments were performed at 25 • C, using HEPES running buffer (50 mM HEPES buffer containing 0.05% Tween 20, pH 7.5, with NaCl added to obtain an ionic strength of 0.2).Cyt c or PLA 2 samples at different concentrations were perfused over the sensor chip surface to obtain real-time binding data at a flow rate of 10 µL/min.The samples are replaced by running buffer in the dissociation phase.The sensor chip was regenerated by injection of a 10 mM sodium acetate buffer (pH 4.0) containing 0.5 M NaCl.All experiments used repetitive cycles of the same injection and regeneration protocol.The association and dissociation curves were analyzed with BIAevaluation 3.0, using the 1:1 Langmuir binding model with a drifting baseline (global fitting).The apparent K d values were calculated from the association rate and dissociation constants.

cDNA Cloning and Sequence Analysis
Total RNA was isolated from L. semifasciata or P. flavoviridis liver using a RNeasy Mini Kit (Qiagen, Valencia, CA, USA), according to the manufacturer's instructions.A SMARTer™ RACE cDNA Amplification Kit (Clontech Laboratories, Palo Alto, CA, USA) was used to clone the 3 -and 5 -ends of L. semifasciata LRG cDNA.The first-strand 3 -and 5 -RACE-ready cDNA samples were prepared according to the manufacturer's protocol and used as templates for 3 -and 5 -RACE, respectively.The first and nested PCR amplification of L. semifasciata LRG cDNA 3 -and 5 -ends was carried out with the Advantage™ 2 PCR Kit (Clontech Laboratories) and two primers, Lsβ1 (5 -CCCTTCCGAGTGGCCTCTTCCGTA-3 ) and Lsβ2 (5 -CAGGTTAGAAGATTGTCCATGCGGCAGT-3 ), corresponding to sequences 323-346 and 926-899, respectively, of E. climacophora PLIβ cDNA (GenBank accession no.AB462511).The PCR products were subjected to electrophoresis on 1.4% agarose gels and extracted using an Agarose Gel Extraction Kit (Roche Diagnostics, Mannheim, Germany).Total RNA from P. flavoviridis liver was reverse transcribed to cDNA using the PrimeScript RT Reagent Kit (Takara, Kusatsu, Japan).The cDNA fragments of P. flavoviridis LRG were amplified by means of PCR, using the following primer pair: Pf β1 (5 -GGCAGGGTGTCCAGCGTCCTTTACTGCCCACCC-3 ) and Pf β2 (5 -GCCCTCTAGAC TCGAGTTAGCAGGGACAAATTTGGT-3 ), designed based on the cDNA sequence of G. brevicauda PLIβ (GenBank accession no.AB007198).P. flavoviridis LRG cDNA fragments with 15 bp flanking sequences homologous to the vector ends were integrated into the plasmid pcDNA3.1,using the In-Fusion ® HD Cloning Kit (Clontech Laboratories), and transformed into Escherichia coli DH5α Competent Cells (Takara).The PCR products and plasmids were sequenced using the DYEnamic ET Terminator Cycle Sequencing Premix Kit (GE Healthcare) on ABI 310 and 3500 Genetic Analyzers (Applied Biosystems).The cDNA sequences determined in the present study were deposited in the DDBJ/EMBL/GenBank nucleotide sequence databases under accession nos.LC786336 (L.semifasciata LRG) and LC786337-LC786340 (cDNA1-4 of P. flavoviridis LRG).Analysis of the DNA sequence data, alignment of the amino acid sequence, and construction of a phylogenetic tree were performed using GENETYX ver.6 (Genetyx, Tokyo, Japan).

Protein Structure Prediction
Protein structures and complexes were predicted using AlphaFold2 [21] and AlphaFold2multimer [27].Sequence alignments and templates were generated using MMseqs2 and HHsearch.Both multiple sequence alignment and AlphaFold2 predictions were performed using ColabFold [28].

Figure 2 .
Figure 2. Inhibition of the enzymatic activity of Laticauda semifasciata PLA-I (a), Gloydius brevicauda basic PLA 2 (b), and Protobothrops flavoviridis PL-X (c) by snake LRGs (PLIβs).The PLA 2 activity was measured fluorometrically with 10-Pyrene-PC as a substrate, in the presence of various concentrations of LRG. , G. brevicauda PLIβ; •, P. flavoviridis LRG; , Elaphe quadrivirgata PLIβ (data from ref. [13]); , Elaphe climacophora PLIβ; , Naja kaouthia LRG; , L. semifasciata LRG.2.4.cDNA Cloning of P. flavoviridis and L. semifasciata LRGsThe full-length cDNA sequence of L. semifasciata LRG was determined by means of direct sequencing of the 3 and 5 rapid amplification of cDNA ends (RACE) products from sea snake liver cDNA (Supplementary FigureS1).L. semifasciata LRG cDNA was 2290 bp (GenBank accession no.LC786336), including 46 bp 5 -untranslated region, 996 bp of the complete coding sequence, and 1145 bp 3 -untranslated region.The predicted open reading frame encodes a protein of 331 amino acids, including a signal peptide consisting of 23 amino acids.Using oligonucleotide primers based on the sequence of G. brevicauda PLIβ, we obtained cDNA clones of LRG (LRGcDNA1-4) with four different sequences from reverse transcripts of P. flavoviridis liver total RNA (GenBank accession no.LC786337-LC786340).A BLAST search of the nucleotide sequences of these four cDNAs against a draft genome sequence of the habu snake [20] retrieved two LRG genes, which we named Pf LRG-A and Pf LRG-B.The two LRG genes are located on different chromosomal scaffolds, not tandem duplications, and probably resulted from an ancient duplication event.The nucleotide sequences of Pf LRG-A and Pf LRG-B differed by 62 bp of the total length of 930 bp, while the amino acid sequences differed by 10 amino acid residues out of the total 310 amino acids (Supplementary FigureS2).LRGcDNA1 and LRGcDNA2 differed from Pf LRG-A by one and four bp, respectively, in terms of the nucleotide sequence, resulting in

Figure 3 .
Figure 3. Binding curves for the interactions of Cyt c and PLA 2 with the immobilized Protobothrops flavoviridis LRG, measured in real-time using surface plasmon resonance.Sensorgrams for the interaction of immobilized P. flavoviridis LRG with (a) 5 nM horse heart Cyt c, 5 nM Laticauda semifasciata Cyt c, and 200 nM yeast Cyt c, (b) 50 nM Gloydius brevicauda basic PLA 2 , (c) 50 nM P. flavoviridis PL-X, and (d) 20 nM P. flavoviridis BPI/II.

Figure 4 .
Figure 4. Amino acid sequence alignment of snake LRGs and PLIβs.Cysteine residues are shown with black background, in reversed type.N-glycosylation sites are shown in bold type.Residues that differed from those in the other sequences are shown in yellow boxes.

Figure 4 .
Figure 4. Amino acid sequence alignment of snake LRGs and PLIβs.Cysteine residues are shown with black background, in reversed type.N-glycosylation sites are shown in bold type.Residues that differed from those in the other sequences are shown in yellow boxes.

Figure 5 .
Figure 5. Phylogenetic tree of snake LRGs and PLIβs.The tree was constructed from the nucleotide difference value, based on the sequence alignments shown in Figure 4, according to the neighborjoining method.

Figure 5 .
Figure 5. Phylogenetic tree of snake LRGs and PLIβs.The tree was constructed from the nucleotide difference value, based on the sequence alignments shown in Figure 4, according to the neighborjoining method.

Table 1 .
The LRG amounts recovered from 10 mL sera of various snakes, by means of a one-step purification procedure using a Cyt c affinity column.

Snakes LRG Amount (mg/10 mL Serum)
Snake venom PLA 2 s are classified into two groups, I and II.Elapidae venom contains group I PLA 2 s, while Viperidae venom contains group II PLA 2 s