Expression of Bv8 in Pichia pastoris to identify structural features for receptor binding

Expression of Bv8 in Pichia pastoris to identify structural features for receptor binding

Protein Expression and Purification 73 (2010) 10–14 Contents lists available at ScienceDirect Protein Expression and Purification journal homepage: ww...

299KB Sizes 8 Downloads 88 Views

Protein Expression and Purification 73 (2010) 10–14

Contents lists available at ScienceDirect

Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep

Expression of Bv8 in Pichia pastoris to identify structural features for receptor binding Rossella Miele a,*, Roberta Lattanzi b, Maria Carmela Bonaccorsi di Patti a, Alessandro Paiardini a, Lucia Negri b, Donatella Barra a a

Istituto Pasteur-Fondazione Cenci Bolognetti, Dipartimento di Scienze Biochimiche ‘A. Rossi Fanelli’ and CNR Istituto di Biologia e Patologia Molecolari, Sapienza Università di Roma, Piazzale Aldo Moro 5, I-00185 Roma, Italy Dipartimento di Fisiologia Umana e Farmacologia ‘V. Espamer’, Sapienza Università di Roma, Italy

b

a r t i c l e

i n f o

Article history: Received 29 January 2010 and in revised form 19 March 2010 Available online 19 April 2010 Keywords: Bv8 mutants Prokineticin receptors Pichia pastoris AVIT proteins

a b s t r a c t Bv8 is an amphibian peptide belonging to the widely distributed AVIT protein family. The mammalian orthologues of Bv8 were named prokineticin 1 and prokineticin 2. Two G-protein-coupled receptors for Bv8–prokineticins have been identified. The biological activities of Bv8/PK proteins range from angiogenesis and involvement in reproduction and cancer, to neuronal survival and neurogenesis, hypothalamic hormone secretion, circadian rhythm control and immunomodulatory processes. Identifying the structural determinants required for receptor binding of Bv8–PKs is mandatory for the design of PKR antagonists, which may be useful in the treatment and prevention of various disease states. Here we describe a procedure for the production in Pichia pastoris of Bv8 and 3 mutants: W24A-Bv8, in which the tryptophan in position 24 is substituted by alanine, the double mutant M1-W24A-Bv8, that contains an additional methionine at the N-terminus and Bv8-TyrTyr that includes two additional tyrosines at the C-terminus. The results evidence a relevant role of tryptophan 24 in Bv8–PKRs interaction. Ó 2010 Elsevier Inc. All rights reserved.

Introduction A small protein, named Bv8 to indicate its origin from the skin secretion of Bombina variegata and its molecular weight (8 kDa), represents the first amphibian member of the AVIT family [1,2]. Homologues of Bv8 are present in the secretion of other Bombina species, in the venom of the snake black mamba (mamba intestinal toxin, MIT-1), in lizards, and in fishes. The orthologues of Bv8, identified in mammals, were named prokineticin 1 (PK1 or EG-VEGF) and prokineticin 2 (PK2 or mBv8) [3]. They both share high sequence identity (80%) with MIT-1 [1]. Two G-protein-coupled receptors for Bv8–PKs, prokineticin receptor 1 (PKR1)1 and prokineticin receptor 2 (PKR2), encoded within distinct chromosomes in both mouse and human, share about 85% amino acid identity: most of the differences are located at the Nterminus, whereas the sequences are almost identical in the transmembrane domains [4,5]. Given their close homology, it is not surprising that PK1 and PK2 are able to activate both receptors in a relatively non-selective fashion [6]. PKRs have been reported to couple either to Gi or to Gq/o proteins, indicating that PKRs activate multiple intracellular signal transduction pathways [4,7]. * Corresponding author. Fax: +39 06 49917566. E-mail address: [email protected] (R. Miele). 1 Abbreviations used: PKR1, prokineticin receptor 1; PKR2, prokineticin receptor 2; IBS, irritable bowel syndrome; IBD, inflammatory bowel disease. 1046-5928/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2010.04.012

Striking features of members of the AVIT family are the presence of 10 cysteines, engaged in five disulfide bridges [8] and the identical amino-terminal sequence, AVITGA. Mutants in which two cysteines (Cys 18 and 60) were substituted with serine and arginine failed to activate PKR1 and PKR2 [9]. The highly conserved AVITGA motif is required for biological activity: deletions or substitutions in these residues produce antagonist molecules. The N-terminal deletion of alanine and valine, in Bv8 (dAV-Bv8), yields an analog lacking any biological activity but still able to bind the receptors, acting as PKR antagonist in vitro and in vivo [9,10]. The solution structure of MIT demonstrated that the fold shared among members of the AVIT family is similar to the one that had been previously determined for colipase [11]. The colipase fold is found in a wide range of proteins other than colipases, such as the Dickkopf family and protease inhibitors [1]; it consists of two subdomains sharing a very similar topology, linked by disulfide bonds. Each subdomain has a central anti-parallel b-sheet connected by loops and stabilized by disulfide bonds resulting in finger-like structures that may serve as interactive surfaces [12]. Intensive research over the past few years has shown that the biological activities of Bv8/ PK proteins range from angiogenesis and involvement in reproduction and cancer, to neuronal survival and neurogenesis, hypothalamic hormone secretion, circadian rhythm control, and modulation of complex behaviors such as feeding and drinking [13]. The high expression level of human Bv8/PK2 in bone marrow,

R. Miele et al. / Protein Expression and Purification 73 (2010) 10–14

lymphoid organs, and leukocytes suggested an involvement of these peptides in hematopoiesis and in inflammatory and immunomodulatory processes [14,15]. Hence the PKRs are potential targets for novel analgesic drugs that block the nociceptive information before it reaches the brain [16]. In a recent paper by Balboni et al. [17], it has been demonstrated that compound 1 [(2-(5-(4-ethylbenzyl)-1-(4-methoxybenzyl)-1,4,5,6-tetrahydro-4,6-dioxo-1,3,5-triazin-2-ylamino)ethyl) guanidine], a 1,3,5-triazin-4,6-dione derivative acts as a strong PKR1 ligand and as an antagonist of PKR1 and PKR2, behaving as a competitive inhibitor of PKR1-MIT binding [17]. This finding strongly suggests that the AVIT family of proteins and compound 1 adopt a similar mechanism to bind PKR1 and, possibly, PKR2. For this reason, we performed a structural comparison of an energy minimized conformer of compound 1 with the three-dimensional structure of MIT (Fig. 1). This analysis revealed that the former might be able to mimic structural features required for receptor binding of MIT and the other members of the AVIT family. Indeed, the evolutionarily conserved and positively charged N-terminal AVIT sequence can be mimicked by the 1,3,5-triazin-2-ylaminoethyl-guanidine moiety, while the tryptophan residue in position 24 is topologically equivalent to the methoxybenzyl moiety of compound 1. Although the conformer of compound 1 obtained by energy minimization means may not fully represent the bioactive conformer, nevertheless the presence of important structural features shared with MIT (that is, a positively charged moiety and, at a comparable distance, an aromatic ring) led us to suppose that (i) Trp24 could play a fundamental role in the interaction with PKR1/PKR2, (ii) members of the AVIT family could interact with PKR1/PKR2 receptors by orienting the protein region that comprises the AVIT sequence and Trp24.

11

In this study, we describe a procedure that makes use of an eukaryotic expression system, the methylotrophic yeast Pichia pastoris [18], for the production of Bv8 and three mutants: W24A-Bv8, in which the tryptophan in position 24 is substituted by alanine, the double mutant M1-W24A-Bv8, that contains an additional methionine at the N-terminus and Bv8-TyrTyr that includes two additional tyrosines at the C-terminus. Our results evidence a relevant role of tryptophan 24 in Bv8–PKRs interaction. Materials and methods Construction of the expression plasmid cDNA coding for Bv8 from B. variegata was cloned into the P. pastoris integrative vector pPIC9K (Invitrogen). The coding sequence of the signal peptide of the a-factor of Saccharomyces cerevisiae present on the vector was amplified by PCR, using two oligonucleotides: pa1-BamHI and pa2-XhoI, which contain restriction sites for BamHI and XhoI, respectively (Table 1). The resulting fragment, PS BamHI–XhoI, of about 250 bp, was extracted from the gel and digested with XhoI. The sense primer used for Bv8 cloning (Bv8kex2XhoI) was designed according to the N-terminal amino acid sequence of Bv8 [2]. A XhoI restriction site in the 50 -end was inserted to allow inframe cloning into the a-factor secretion signal of pPIC9K. A sequence encoding the KEX2 cleavage site comprising 12 nucleotides was placed ahead of the mature Bv8 cDNA. The antisense primer (Bv8EcoRIdw) was designed based on the C-terminal amino acid sequence of Bv8 and possesses a stop codon and an EcoRI restriction site at the 30 -end (Table 1). The fragment of about 300 bp, named Bv8 XhoI–EcoRI, was digested with XhoI. Finally, the two resulting fragments, PS BamHI–XhoI and Bv8 XhoI–EcoRI, were ligated and used as template for a new PCR amplification using the oligonucleotides pa1-BamHI and Bv8EcoRIdw. The generated fragment, containing the signal peptide of the a-factor of S. cerevisiae upstream the Bv8 coding region, was digested with BamHI and EcoRI and inserted into pPIC9K vector, previously cleaved with the same enzymes in order to cut the signal sequence of the vector. This plasmid (PS-Bv8#13) was transformed into competent Escherichia coli Top10F0 and sequenced. Construction of Bv8 mutants Bv8-TyrTyr, W24A-Bv8 and M1-W24ABv8 The antisense primer (TyrTyr Bv8-dw) was designed based on the C-terminal amino acid sequence of Bv8 and permits the introduction of two tyrosine residues and an EcoRI restriction site in the 30 -end (Table 1). Using as template PS-Bv8#13, the fragment was amplified by PCR with the oligonucleotides pa1-BamHI and TyrTyr Bv8-dw.

Table 1 Sequences of oligonucleotides used in this work. Oligonucleotides Sequence Fig. 1. Structural superposition of an energy minimized conformer of compound 1 (gold) with the three-dimensional structure of MIT (pink). The three-dimensional structure of [(2-(5-(4-ethylbenzyl)-1-(4-methoxybenzyl)-1,4,5,6-tetrahydro-4,6dioxo-1,3,5-triazin-2-ylamino)ethyl)guanidine] (compound 1) was built starting from its 2D structure and energetically minimized by means of the PRODRG server [23]. The structural superposition between the NMR structure of MIT (PDB code: 1IMT) and compound 1 was performed manually, in order to find the best matching between chemically similar moieties. Oxygen and nitrogen atoms are colored in red and blue, respectively. The figure was rendered with PyMol (DeLano Scientific LLC). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

pa1-BamHI pa2-XhoI Bv8kex2XhoI Bv8EcoRIdw TyrTyr Bv8-dw Mut-W-dw Bv8-W-up Bv8-met-up

50 -GCG GAT CCA AAC GAT GAG ATT TCC-30 50 -GCC TCG AGA GAT ACC CCT TCT TC-30 50 -TTA TCT CGA GAA AAG AGC TGT TAT CAC TGG CGC CTG TG-30 50 -ATA GAA TTC TCA TCA AGA ACA CTT AAA TTT TTC TCC-30 50 -ATA GAA TTC TCA TCA GTA GTA AGA ACA CTT AAA TTT TTC TCC-30 50 -CGC GCT AGC GCA GCA GG-30 50 -GCT GCT AGC GCT TCA CGT AAC ATC AG-30 50 -TAT CTC GAG AAA AGA ATG GCT GTT ATC ACT GGC GCC TGT G-30

12

R. Miele et al. / Protein Expression and Purification 73 (2010) 10–14

To obtain the W224A mutant, two different fragments were amplified by PCR using as template PS-Bv8#13: the first was obtained using the oligonucleotides pa1-BamHI and Mut-W-dw. The latter oligonucleotide anneals upstream the tryptophan 24 region and introduces at the 30 -end a NheI restriction site obtained with a silent single base substitution (Table 1). The second fragment was obtained using the oligonucleotides Bv8-W-up and Bv8EcoRIdw. The former anneals to the tryptophan 24 region, and contains three base mutations allowing both the W24A substitution and the insertion of the NheI restriction site (Table 1). The two resulting fragments were digested with NheI and ligated. The product was used as template for a new PCR amplification using the oligonucleotides pa1-BamHI and Bv8EcoRIdw. The generated fragment, containing the signal peptide of the a-factor of S. cerevisiae upstream the Bv8 coding region, was again digested with BamHI and EcoRI and inserted into pPIC9K. This plasmid (Bv8#4B), containing a Bv8 cDNA carrying the W24A substitution, was transformed into competent E. coli Top10F0 and sequenced. For the double mutant cDNA was obtained by ligation of two different fragments. The first PS BamHI–XhoI, coding sequence of the signal peptide of the a-factor of S. cerevisiae was obtained as described. The second fragment was obtained using the oligonucleotides Bv8-met-up and Bv8EcoRIdw (Table 1). The two resulting fragments were digested with XhoI and ligated. The product was again digested with BamHI and EcoRI and inserted into pPIC9K vector. This plasmid (Bv8 mbv#2) was transformed into competent E. coli Top10F0 and sequenced. Expression of Bv8 and Bv8 mutants in P. pastoris Plasmids PS-Bv8#13, Bv8#4B and Bv8 mbv#2 were linearized with SalI, to favor integration at the his4 locus of P. pastoris genome, and transformed into P. pastoris GS115 by electroporation. Selection of His+ transformants was done on minimal selective MD medium (1.34% YNB, 4  105% biotin, 1% dextrose, and 1.5% agar). To confirm the presence of the Bv8 expression cassette in P. pastoris His+ transformants, colonies were screened by PCR using the specific primers, pal-BamHI and Bv8EcoRIdw on purified genomic DNA. Fourteen recombinant P. pastoris clones were analyzed for the amount of secreted Bv8 through small-scale expression trials in 4 mL cultures. The highest Bv8 expressing P. pastoris colony was grown in 5 mL BMG medium, pH 6.0, for approximately 24 h at 30 °C with constant shaking. These cells were cultured further in 600 mL BMG, pH 6.0, for 18 h at 30 °C with constant shaking. The induction of Bv8 synthesis was carried out for 120 h by daily supplementation of 1% methanol. The pH was adjusted to 5.0 every 24 h. Purification of recombinant proteins The crude culture was centrifuged to remove the cells. Supernatant was diluted 1:5 and applied to a CM-Sephadex C-25 (Pharmacia) column in 20 mM BES, pH 7.0. At pH 7, Bv8, with a pI of 8.46, binds to the resin. The elution buffer was 20 mM BES, pH 7.0/0.2 M NaCl. Recombinant Bv8 fractions were pooled and dialyzed against 20 mM Tris–HCl (pH 7.0) buffer. Then, the recombinant protein was purified by reverse-phase HPLC on a Vydac C-18 column. The elution was performed with a linear gradient of acetonitrile (18–45%, v/v) in 0.1% TFA at 1.35% per minute. The flow rate was 5.0 mL/min and the absorbance of the effluent was monitored at 220 nm. The fractions collected were dried under vacuum and dissolved in milli-Q water. Protein concentration was measured integrating the area of two peaks 33–35 min from C-18 reverse column and comparing with a natural Bv8 standard.

SDS–PAGE was carried out on 18% (w/v) polyacrylamide gel in the Laemmli system. Proteins were visualized by silver staining. The molecular mass of purified Bv8 was determined by mass spectrometry. The mass spectrometry analyses were performed in a Voyager-DE STR instrument (Applied Biosystems, Framingham, MA) operating in reflector mode. Receptor binding assay Bv8 and its analogs were evaluated for their ability to displace [125I]MIT binding from membrane preparations of PKR1- and PKR2-transfected CHO cells, as already described [10]. Briefly, membranes (20 lg of proteins for PKR1 and 40 lg of proteins for PKR2) were incubated with 4.0 pM [125I]MIT and 100 ll of graded concentrations of each compound (made in duplicate), in a final volume of 1 mL at 37 °C for 90 min. Membranes were harvested on Whatman GF/B filters pre-soaked in 0.5% polyethyleneimine (Sigma–Aldrich) and transferred to counting vials. Radioactivity was measured in a gamma-counter (Packard, Cobra II auto-gamma). Non-specific binding was determined in the presence of 1 lM Bv8. Displacement curves and IC50 were calculated with the PRISM software (GraphPad Software). Results Expression of Bv8 in P. pastoris As a first approach to investigate the structure–function relationships in the AVIT protein family, we explored the possibility of producing large quantities of Bv8 and suitable mutants, overcoming the difficulties related to the correct formation of disulfide bridges. In fact, the expression of proteins rich in disulfide bonds in prokaryotic cells is often unsuccessful. In the past, the expression of prokineticins in E. coli required a supplementary step of in vitro refolding with a tedious and inefficient procedure [9]. The alternative method using Baculovirus determines low level production of Bv8 [19]. As the integrity of the amino-terminal region is essential for the function of Bv8, its cDNA was inserted, in the proper reading frame, to avoid the inclusion of additional residues between the a-factor signal peptide and the start codon of the protein. The pPIC9K integrative expression vector was modified replacing the original signal peptide proteolytic recognition sequences, containing both KEX2 and STE13 sites, with the one containing only the KEX2 protease site. This strategy was adopted on the basis of previous results showing the inefficacy of STE13 cleavage: in fact, processing of Bv8 expressed in P. pastoris resulted in a secreted protein containing an additional tetrapeptide (EAEA) at its N-terminus. Transformed P. pastoris GS115 colonies presenting the stronger amplification products were grown in 4 mL cultures for small-scale expression trials and the amount of recombinant protein was determined by reversed-phase HPLC. Following this procedure, we selected the highest Bv8-secreting P. pastoris clone for subsequent large-scale protein expression. The expression protocol included a sequential growth of P. pastoris GS115/Bv8 for 24 h in 5 mL BMG, followed by 18 h in 2 L BMG in order to obtain high cell densities. Methanol solution was added once per day to maintain a 1% (v/v) methanol concentration. The optimal time of maximum Bv8 secretion was determined by monitoring the Bv8 content of methanol induced cultures for 144 h. The highest expression peak of recombinant Bv8 was observed at 120 h (Fig. 2). The supernatants of expression cultures were subjected to CMSephadex chromatography. This step was essential to eliminate

13

R. Miele et al. / Protein Expression and Purification 73 (2010) 10–14

M

0h

24h

72h

120h

116 66 45 35

25

18.4

1 Fig. 2. Analysis of recombinant Bv8 fractions. Small-scale expression of Bv8 in P. pastoris GS115 induced with methanol and time-course study analyzed on 18% SDS–PAGE. Lane 1 shows a set of standards with the apparent molecular masses indicated to the left. The arrow indicates the 8-kDa apparent molecular mass of Bv8.

high molecular weight proteins and salts, which appear to clog the reversed-phase HPLC column used for final sample purification. The retention time of recombinant Bv8 was similar to that of the natural protein. The amount of secreted Bv8, as estimated by time-integrated surfaces of the HPLC peak, was found to be approximately 1 mg/L of culture medium. To further confirm the identity of the peptide, a mass spectrometry analysis was performed. The expected molecular mass of oxidized Bv8 is m/z 8341.83. The experimental value for the recombinant protein was found to be m/z 8338.46, suggesting that the thiol status of the recombinant peptide is the correct one. Expression of Bv8 mutants in P. pastoris Analysis with DALI software showed that the structure of AVIT, colipase and Dickkopf proteins shares similar features, typical of proteins lacking extensive secondary structures and stabilized by abundant disulfide bridges. These proteins can be described as an assembly of protruding fingers held together at one end by a network of five disulfide bridges. Recently a crucial role of tryptophan in the second finger loop in Dickkopf 2 was demonstrated [20]. In all members of the AVIT protein family a tryptophan is present in the same site. Therefore, we decided to substitute this tryptophan with alanine, and produce two other mutant forms with the aim to gain information on the structural features for receptor binding. The mutants were expressed as soluble proteins and the yield of each mutant was equal to or higher than that of the wild-type Bv8. Mass spectrometry analysis of the W24A mutant confirmed the correct mass (calculated value m/z 7900.66, experimental value m/z 7897.02). These results supported the advantage claimed for the use of the P. pastoris expression system in order to obtain a large amount of nearly pure secreted recombinant proteins. Receptor binding Binding of the fully purified Bv8 protein and natural secretionderived Bv8 to CHO expressing PKR1 and PKR2 was measured by competition of the unlabeled ligands with 125I-MIT (Fig. 3 and Table 2). The competition curves are coincident indicating identical binding affinity of the recombinant and skin secretion-derived peptide (Fig. 3). Similar concentrations of Bv8 were necessary to

Fig. 3. Ligand-binding properties of Bv8 mutants to PKR1 (upper panel) and PKR2 (lower panel). Binding to PKR1 or PKR2 expressed in CHO cells was assayed by displacement of 125I-MIT-1 by Bv8 mutants. Data are means ± SEM of at least three independent experiments.

displace bound [125I]MIT from PKR1 (IC50 = 1.3 ± 0.8 nM) or PKR2 (IC50 = 1.0 ± 0.5 nM) expressed in CHO cells. The W24A-Bv8 mutant [Ala24Bv8] showed 40 times lowered affinity for both PKR1 (IC50 = 41.1 ± 4.7 nM) and PKR2 (IC50 = 41.7 ± 5.3 nM). The addition of methionine at the N-terminus of Ala24Bv8 generated a peptide, Met-Ala24Bv8, with a 20- and 16-fold decreased affinity for PKR1 and PKR2, respectivley (Fig. 3). This result was in accord with the observation that a mutant of PKs obtained by inserting a Met residue at the N-terminus is a potent antagonist of Bv8. In fact, the mutant maintained the capacity of receptor binding, but determined a reduction of receptor activation [9]. The C-terminal region of the protein has a minor role for peptide–receptor interaction. For prokineticins, the exchangeability of the C-terminal domains between PK1 and PK2 was demonstrated [9]. The addition of two tyrosine residues at the C-terminus of Bv8 generated a compound, Bv8-TyrTyr, with a 10- and 3-fold lower affinity for PKR1 (IC50 = 9.7 ± 1.5 nM) and PKR2 (IC50 = 3.5 ± 0.5 nM), respectively (Fig. 3).

Table 2 IC50 values for binding of Bv8 mutants to PKR1 and PKR2.

PKR1 (nM) PKR2 (nM)

Bv8

Ala24Bv8

Met-Ala24Bv8

Bv8-TyrTyr

1.3 ± 0.8 1 ± 0.5

41.1 ± 4.7 41.7 ± 4.2

27.2 ± 3 16.4 ± 1.5

9.7 ± 1.5 3.5 ± 0.5

14

R. Miele et al. / Protein Expression and Purification 73 (2010) 10–14

Discussion The choice of Trp24 as main target of the mutagenesis study presented in this work was prompted by recent results obtained by Balboni et al. on compound 1 [17; cfr. Introduction]. This compound was assessed to act as a strong PKR1 ligand and as an antagonist of PKR1 and PKR2, behaving as a competitive inhibitor of PKR1-MIT binding [17]. Structural comparison of an energy minimized conformer of compound 1 with the three-dimensional structure of MIT (Fig. 1) revealed that the former might be able to mimic structural features required for receptor binding of MIT and the other members of the AVIT family, suggesting a critical role of Trp24 in the interaction with PKR1/PKR2. In order to test this hypothesis we constructed and expressed three different mutants using the methylotrophic yeast P. pastoris. The increasing popularity of this expression system can be attributed to the ability of P. pastoris to produce foreign proteins at high level, either intracellularly or extracellularly, and the capability of performing many eukaryotic post-translational modifications, such as glycosylation, disulfide bond formation and proteolytic processing [18]. In this study, we have demonstrated that Bv8 can be recombinantly expressed by the yeast P. pastoris and secreted as active peptide into the culture medium, with correct disulfide bond formation. The tryptophan residue in position 24 is conserved in AVIT proteins isolated from many species including invertebrates and vertebrates, implying a role in the function. Because of this topology, the presence of this residue in the hydrophobic surface, could suggest a role in mediating receptor interaction. In this study, we have analyzed the biological activities of Bv8 molecules with the substitution of W24 with alanine. Using membranes from CHO cells expressing PKR1 or PKR2, it could be shown that the affinity of W24A-Bv8 for these receptors was 40 times lower than that of Bv8. Insertion of methionine at the N-terminus determines a reduction of binding to the prokineticin receptors. In contrast insertion of hydrophobic amino acids (TyrTyr) at the C-terminus does not determine a drastic alteration of receptor interaction capacity. In conclusion, these results demonstrate that a highly conserved aromatic residue, a tryptophan at position 24 is essential for AVIT protein binding to PKR1 and PKR2 receptors. Identifying the structural determinants required for receptor binding and hyperalgesic activity of Bv8–PKs is mandatory for the design of PKR antagonists, which may be useful for controlling inflammation and inflammatory pain [21]. Furthermore, PK receptor antagonists could be useful in treating cancer-specific angiogenesis, preventing tumor development/progression [22]. Acknowledgments This work was partially supported by grants from Sapienza Università di Roma. We wish to thank Dr. Alessandra Giorgi for the mass spectrometry analyses. References [1] A. Kaser, M. Winklmayr, G. Lepperdinger, G. Kreil, The AVIT protein family. Secreted cysteine-rich vertebrate proteins with diverse function, EMBO Rep. 4 (2003) 469–473.

[2] C. Mollay, C. Wechselberger, G. Mignogna, L. Negri, P. Melchiorri, D. Barra, G. Kreil, Bv8, a small protein from frog skin and its homologue from snake venom induce hyperalgesia in rats, Eur. J. Pharmacol. 374 (1999) 189–196. [3] M. Li, C.M. Bullock, D.J. Knauer, F.J. Ehlert, Q.Y. Zhou, Identification of two prokineticin cDNAs: recombinant proteins potently contract gastrointestinal smooth muscle, Mol. Pharmacol. 59 (2001) 692–698. [4] D.C. Lin, C.M. Bullock, F.J. Ehlert, J.L. Chen, H. Tian, Q.Y. Zhou, Identification and molecular characterization of two closely related G protein-coupled receptors activated by prokineticins/endocrine gland vascular endothelial growth factor, J. Biol. Chem. 277 (2002) 19276–19280. [5] Y. Masuda, Y. Takatsu, Y. Terao, S. Kumano, Y. Ishibashi, M. Suenaga, M. Abe, S. Fukusumi, T. Watanabe, Y. Shintani, T. Yamada, S. Hinuma, N. Inatomi, T. Ohtaki, H. Onda, M. Fujino, Isolation and identification of EG-VEGF/ prokineticins as cognate ligands for two orphan G-protein-coupled receptors, Biochem. Biophys. Res. Commun. 293 (2002) 396–402. [6] L. Negri, R. Lattanzi, E. Giannini, M. Canestrelli, A. Nicotra, P. Melchiorri, Bv8/ Prokineticins and their Receptors A New Pronociceptive System, Int. Rev. Neurobiol. 85 (2009) 145–157. [7] J. Chen, C. Kuei, S. Sutton, S. Wilson, J. Yu, F. Kamme, C. Mazur, T. Lovenberg, C. Liu, Identification and pharmacological characterization of prokineticin 2 beta as a selective ligand for prokineticin receptor 1, Mol. Pharm. 67 (2005) 2070– 2076. [8] J. Boisbouvier, J.P. Albrand, M. Blackledge, M. Jaquinod, H. Schweitz, M. Lazdunski, D. Marion, A structural homologue of colipase in black mamba venom revealed by NMR floating disulfide bridge analysis, J. Mol. Biol. 283 (1998) 205–219. [9] C.M. Bullock, J.D. Li, Q.Y. Zhou, Structural determinants required for the bioactivities of prokineticins and identification of prokineticin receptor antagonists, Mol. Pharmacol. 65 (2004) 582–588. [10] L. Negri, R. Lattanzi, E. Giannini, M.A. Colucci, G. Mignogna, D. Barra, F. Grohovaz, F. Codazzi, A. Kaiser, G. Kreil, P. Melchiorri, Biological activities of Bv8 analogues, Br. J. Pharmacol. 146 (2005) 625–632. [11] H. Van Tilbeurgh, L. Sarda, R. Verger, C. Cambillau, Structure of the pancreatic lipase-procolipase complex, Nature 359 (1992) 159–162. [12] H. Van Tilbeurgh, S. Bezzine, C. Cambillau, R. Verger, F. Carriere, Colipase: structure and interaction with pancreatic lipase, Biochim. Biophys. Acta 1441 (1999) 173–184. [13] E.S. Ngan, P.K. Tam, Prokineticin-signaling pathway, Int. J. Biochem. Cell Biol. 40 (2008) 1679–1684. [14] J. LeCouter, C. Zlot, M. Tejada, F. Peale, N. Ferrara, Bv8 and endocrine glandderived vascular endothelial growth factor stimulate hematopoiesis and hematopoietic cell mobilization, Proc. Natl. Acad. Sci. USA 101 (2004) 16813–16818. [15] J. Monnier, M. Samson, Cytokine properties of prokineticins, FEBS J. 275 (2008) 4014–4021. [16] L. Negri, R. Lattanzi, E. Giannini, P. Melchiorri, Modulators of pain: Bv8 and prokineticins, Curr. Neuropharmacol. 4 (2006) 207–215. [17] G. Balboni, I. Lazzari, C. Trapella, L. Negri, R. Lattanzi, E. Giannini, A. Nicotra, P. Melchiorri, S. Visentin, C.D. Nuccio, S. Salvadori, Triazine compounds as antagonist at Bv8-prokineticin receptors, J. Med. Chem. 51 (2008) 7635–7639. [18] J.L. Cereghino, J.M. Cregg, Heterologous protein expression in the methylotrophic yeast Pichia pastoris, FEMS Microbiol. Rev. 24 (2000) 45–66. [19] J. LeCouter, J. Kowalski, J. Foster, P. Hass, Z. Zhang, L. Dillard-Telm, G. Frantz, L. Rangell, Identification of an angiogenic mitogen selective for endocrine gland endothelium, Nature 412 (2001) 877–884. [20] L. Chen, K. Wang, Y. Shao, J. Huang, X. Li, J. Shan, D. Wu, J.J. Zheng, Structural insight into the mechanisms of Wnt signaling antagonism by Dkk, J. Biol. Chem. 283 (2008) 23364–23370. [21] F. Shojaei, M. Singh, J.D. Thompson, N. Ferrara, Role of Bv8 in neutrophildependent angiogenesis in a transgenic model of cancer progression, Proc. Natl. Acad. Sci. USA 105 (2008) 2640–2645. [22] E. Giannini, R. Lattanzi, A. Nicotra, A.F. Campese, P. Grazioli, I. Screpanti, G. Balboni, S. Salvadori, P. Sacerdote, L. Negri, The chemokine Bv8/prokineticin 2 is up-regulated in inflammatory granulocytes and modulates inflammatory pain, Proc. Natl. Acad. Sci. USA 106 (2009) 14646–14651. [23] D.M. van Aalten, R. Bywater, J.B. Findlay, M. Hendlich, R.W. Hooft, G. Vriend, PRODRG, a program for generating molecular topologies and unique molecular descriptors from coordinates of small molecules, J. Comput. Aided Mol. Des. 10 (1996) 255–262.