Báo cáo khoa học: Cupiennin 1a, an antimicrobial peptide from the venom of the neotropical wandering spider Cupiennius salei, also inhibits the formation of nitric oxide by neuronal nitric oxide synthase pptx

of the neotropical wandering spider Cupiennius salei,also inhibits the formation of nitric oxide by neuronal nitric oxide synthase Tara L.. The final group comprises peptides with masses

of the neotropical wandering spider Cupiennius salei,

also inhibits the formation of nitric oxide by neuronal

nitric oxide synthase

Tara L Pukala1, Jason R Doyle2, Lyndon E Llewellyn2, Lucia Kuhn-Nentwig3, Margit A Apponyi1, Frances Separovic4and John H Bowie1

1 Department of Chemistry, The University of Adelaide, Australia

2 Institute of Marine Science, Townsville, Queensland, Australia

3 Zoological Institute, University of Bern, Switzerland

4 School of Chemistry, Bio21 Institute, University of Melbourne, Australia

The neotropical wandering spider Cupiennius salei is a

large, nocturnal hunting spider distributed throughout

Central America, located as far north as Veracruz state

in Mexico and extending to Honduras in the south It

is restricted to altitudes ranging from 200 to 1250 m,

and resides in the tropical rain forests of this region

[1] The spider is brown, with small, light spots on the

abdomen and many dark longitudinal stripes,

predom-inantly on the carapace The legs, patella and femurs

are also striped with lighter circles, and the underbody

is red–yellow with thin black vertical stripes under the

abdomen Females can reach up to 35 mm in body

length and have a 100 mm leg span, whereas males are

typically smaller and less brightly coloured [1]

The venom of C salei is a natural insecticide, caus-ing a rapid and dose-dependent paralysis of prey up to

a critical lethal dose [1] Three classes of molecules comprise the venom, and can be categorized on the basis of molecular weights The first group consists of low molecular weight compounds, including ions, free amino acids, amines and polyamines [2] The second group includes mainly proteins with masses between 25 and 27 kDa Among these, a highly active hyaluroni-dase has been reported, which is a spreading factor used to accelerate toxin transport into the tissue [2] Even under extreme test conditions, only very low lev-els of proteolytic enzymes are observable The final group comprises peptides with masses generally in the

Keywords

cupiennin 1a; Cupiennius salei; neuronal

nitric oxide synthase activity;

two-dimensional NMR

Correspondence

J H Bowie, Department of Chemistry, The

University of Adelaide, Adelaide, SA 5005,

Australia

Fax: +61 08 830 34358

Tel: +61 08 830 3567

E-mail: john.bowie@adelaide.edu.au

(Received 22 August 2006, revised 17

Janu-ary 2007, accepted 1 FebruJanu-ary 2007)

doi:10.1111/j.1742-4658.2007.05726.x

(GFGALFKFLAKKVAKTVAKQAAKQGAKYVVNKQ-ME-NH2) is a potent venom component of the spider Cupiennius salei Cupiennin 1a shows multifaceted activity In addition to known antimicro-bial and cytolytic properties, cupiennin 1a inhibits the formation of nitric oxide by neuronal nitric oxide synthase at an IC50 concentration of 1.3 ± 0.3 lm This is the first report of neuronal nitric oxide synthase inhi-bition by a component of a spider venom The mechanism by which cupi-ennin 1a inhibits neuronal nitric oxide synthase involves complexation with the regulatory protein calcium calmodulin This is demonstrated by chem-ical shift changes that occur in the heteronuclear single quantum coherence spectrum of 15N-labelled calcium calmodulin upon addition of cupien-nin 1a The NMR data indicate strong binding within a complex of 1 : 1 stoichiometry

Abbreviations

Ca2+-CaM, calcium calmodulin; HSQC, heteronuclear single quantum coherence; NOS, nitric oxide synthase; nNOS, neuronal nitric oxide synthase.

range 2–8 kDa These include: (a) the neurotoxins

CSTX-1, CSTX-9 and CSTX-13 [3,4]; and (b) the

anti-microbial and cytolytic cupiennins 1a to 1d [5–8] The

sequence of cupiennin 1a is

The cupiennins are membrane-active wide-spectrum

antimicrobials; the most stable structure of

cupien-nin 1a (as determined by two-dimensional NMR

meth-ods in trifluoroethanol⁄ water, 1 : 1 [9]) is the hinged

structure shown in Fig 1 (the hinge occurs at Gly25)

It has been suggested that the role of these

antimicro-bial peptides in the venom of C salei may be two-fold:

(a) the cheliceral claws, which first penetrate the prey,

are heavily exposed to external pathogens, and thus

the antibacterial peptides may be involved in

protec-tion of the venom apparatus against infecprotec-tion; and

(b) the cytolytic activity of these peptides may afford

the neurotoxins better access to their intercellular

tar-gets [7]

After the secondary structure determination of the

strongly basic and hinged peptide cupiennin 1a was

finalized (Fig 1), it was clear that this peptide showed

some structural features in common with certain

amphibian peptides that inhibit the formation of NO

by neuronal nitric oxide synthase (nNOS) [10] These

particular amphibian peptides (e.g the caerins 1 and

splendipherin [11]), are basic and hinged, and inhibit

the operation of nNOS by complexing with the

regula-tory protein calcium calmodulin (Ca2+-CaM)

In this article, we report that cupiennin 1a also

com-plexes with Ca2+-CaM, and is one of the more active

of the known peptide inhibitors of nNOS

Results

Cupiennin 1a was tested for the ability to inhibit nNOS, using an assay that measures the conversion of [3H]arginine to [3H]citrulline by this enzyme [10] Cupi-ennin 1a produces a dose-dependent inhibition of nNOS: the IC50 value and Hill slope [12] are 1.3 ± 0.3 lm (5.1 ± 1.1 lgÆmL)1) and 3.5 ± 1.0, respectively These values are comparable to those of amphibian peptides, which inhibit nNOS by complex-ing with the regulatory protein Ca2+-CaM [10,11]

A 15N heteronuclear single quantum coherence (HSQC) titration was performed to determine whether cupiennin 1a interacts with CaM to inhibit the action

of nNOS Increasing quantities of unlabelled cupien-nin 1a were added to 15N-labelled Ca2+-CaM, and a high-resolution 15N HSQC spectrum was recorded after each addition The chemical shift changes were then tracked by overlaying each of the spectra, as can

be seen in Figs 2 and 3 Chemical shift changes were considered to be significant when they were greater than 0.5 p.p.m in the nitrogen dimension and greater than 0.05 p.p.m in the hydrogen dimension [13] Evidence of complex formation is apparent, with the titration series showing distinct chemical shift changes for a large number of residues throughout the Ca2+ -CaM sequence The chemical shifts do not change as a function of concentration; rather, a second set of peaks appear with distinct chemical shifts after addition of only 0.4 equivalents of cupiennin 1a Peak intensities for the bound and unbound conformers at 0.6 equiva-lents of peptide are comparable, suggesting a 1 : 1

stoi-Fig 1 The lowest calculated potential

energy structure of cupiennin 1a in

d3-trifluoroethanol ⁄ water (1 : 1).

chiometry and providing evidence of slow exchange

binding The peptide was fully bound and gave rise to

a completely new set of protein chemical shifts at a

1 : 1 molar concentration, with continued addition of

peptide to 2 : 1 equivalents having no further effect on

the chemical shifts (data not shown) The peak

intensi-ties for the bound and unbound conformations are

approximately equal upon full saturation, further

indi-cating that the complex is stable on the NMR

time-scale and exists in a slow exchange regime

Selected resonances in the HSQC spectrum were

assigned on the basis of the chemical shifts reported

previously for unbound Ca2+-CaM [14,15] No

attempt was made to assign either unbound resonances

in the NMR spectra of Ca2+-CaM, or of the fully

bound complex, in regions where the density of signals

would result in ambiguity Even so, distinct chemical

shift changes occur for a large number of Ca2+-CaM

resonances, and this can be seen from the selected

labelled peaks shown in Fig 3 (e.g T5, G25, I27, T29,

G61, G98, L116, T117, N137 and A147) This is

con-sistent with a substantial change in Ca2+-CaM

confor-mation following complexation with cupiennin 1a

Discussion

NO is unique among biological signals for its rapid

diffusion, ability to permeate cell membranes, and

intrinsic instability, properties that eliminate the need

for extracellular NO receptors or targeted NO

degra-dation [16,17] NO is produced by three NOSs (in ver-tebrates), which oxidize l-arginine to NO and citrulline, thereby controlling NO distribution and

Fig 2 15N HSQC spectra of CaM in the absence of cupiennin 1a

(red), and with the addition of cupiennin 1a in a 1 : 1 molar ratio

(purple).

Fig 3 Partial overlaid15N HSQC spectra for the titration of Ca2+ -CaM with cupiennin 1a The peptide ⁄ protein ratio is indicated.

concentration The isoforms of NOS are homodimers

with subunits of 130–160 kDa, differing in amino acid

sequence identity, but sharing an overall

three-compo-nent construction, namely: (a) an N-terminal catalytic

oxygenase domain that binds heme,

tetrahydrobiopter-in and l-argtetrahydrobiopter-intetrahydrobiopter-ine; (b) a C-termtetrahydrobiopter-inal reductase domatetrahydrobiopter-in

that binds FMN, FAD and NADPH; and (c) an

inter-vening CaM-binding region that regulates electronic

communication between the oxygenase and reductase

domains [16,17] NOS enzymes are found in most

life-forms [16,17], including bacteria [18–21] and insects

[22–24]

Ca2+-CaM is a dumbbell-shaped 148-residue protein

containing two terminal units, each of which may

con-tain two Ca2+ Ca2+-CaM is required for the

activa-tion of nNOS: it is a regulatory protein that acts as an

electron shuttle and Ca2+ transporter It also alters

the conformation of the reductase domain, allowing

reactions to proceed at the heme site [17] nNOS-active

peptides interfere with communication between

Ca2+-CaM and nNOS, because the complex formed

between the active peptide and Ca2+-CaM has a

dif-ferent shape from that of Ca2+-CaM [10,11,25,26],

and therefore adversely affects binding of CaM to the

Ca2+-CaM-binding domain of nNOS CaM is not

only essential for the operation of nNOS and the other

NOS isoforms, but is also the regulatory protein for a

variety of other enzymes, including kinases [27–29]

Two peptide–CaM binding modes have been

identi-fied by NMR studies [30] These are shown in Fig 4 In

the first, Ca2+-CaM adopts a compact, globular shape,

with the peptide engulfed in a hydrophobic channel

formed by the two terminal domains [25,30–32] The

example shown in Fig 4A is that of the 26-residue

pep-tide fragment of skeletal myosin light chain kinase, with

residues 3–21 encompassed within Ca2+-CaM [25]: this

type of structure is also adopted by Ca2+-CaM when it

binds to the CaM-binding domain of endothelial nitric

oxide synthase (eNOS) [33] The second example is

when the C-terminal lobe of Ca2+-CaM binds part of

the target peptide This is shown in Fig 4B for the

20-residue binding domain of the plasma membrane Ca2+

pump⁄ Ca2+-CaM complex, where the first 12 residues

of the peptide are encompassed by the C-terminal end

of Ca2+-CaM [26]

Previous studies have indicated that binding of

pep-tides to Ca2+-CaM requires the peptide: (a) to adopt

an amphipathic a-helical conformation when binding

to CaM [10,11,25,26]; (b) be positively charged

[10,11,25,26]; and (c) display large hydrophobic

resi-dues in conserved positions, which point to one face in

a presumed helical conformation [30] It has been

pro-posed that the extent of hydrophobic anchoring

deter-A

B

Fig 4 (A) Myosin light-chain kinase ⁄ Ca 2+

-CaM complex [25] (B) Binding domain of the plasma membrane Ca 2+ pump ⁄ Ca 2+ -CaM complex [26].

mines which of the two binding modes (Fig 4A,B) is

adopted by the complex [30]: this is supported by

small-angle X-ray scattering experiments [34]

Cupiennin 1a conforms to all of these prerequisites

It is unstructured in water [6], but adopts a helical

struc-ture in the membrane mimicking the solvent d3

-trifluor-oethanol⁄ water (1 : 1) [9], as shown in Fig 1 The large

number of lysine residues gives the peptide an overall

charge of + 8 As the N-terminal helix of cupiennin 1a

is more amphipathic than the C-terminus, and also has

a greater positive charge, it is reasonable to suppose

that Ca2+-CaM is more likely to bind to this region of

the peptide Furthermore, the hydrophobic face of the

amphipathic N-terminal helix of cupiennin 1a has a

sig-nificant number of long-chain aliphatic or aromatic

resi-dues (L, V and F) available to act as hydrophobic

anchors Given the length of the first helix of

cupien-nin 1a, and the number of hydrophobic anchors

avail-able, it seems likely that the cupiennin 1a⁄ Ca2+-CaM

complex is analogous to the structure shown in Fig 4A

In such a case, some 20 residues of cupiennin 1a could

be situated within the globular Ca2+-CaM

This proposal is consistent with the data shown in

Figs 2 and 3, which indicate that chemical shift

chan-ges occur throughout the CaM sequence, including the

C-terminal and N-terminal domains This means that a

substantial change in conformation occurs for the

regulatory protein upon binding of cupiennin 1a to

Ca2+-CaM, indicating that the complex forms with a

significantly different structure, rather than localized

structural differences at the binding interfaces

Analy-sis of the chemical shift changes for those resonances

that are readily assigned (Fig 3) is also consistent with

those reported for the structure shown in Fig 4A [35]

Conclusion

Cupiennin 1a is a potent venom component of C salei

with multifaceted activity, including antimicrobial

activity and inhibition of nNOS We propose that the

inhibition of nNOS involves the formation of a

glob-ular Ca2+-CaM⁄ cupiennin 1a complex, which prevents

Ca2+-CaM from occupying the CaM-binding domain

of nNOS This will drastically influence numerous

pro-cesses that rely on NO as a neurotransmitter in both

prokaryotic and eukaryotic cells CaM is not only

essential for the operation of nNOS and the other

NOS isoforms, but is also the regulatory protein for a

variety of kinase phosphorylating enzymes and

adeny-late cyclase [28], and is involved in regulation of the

eukaryote cytoskeleton [28] The likelihood is that

cupiennin 1a will interfere with many cellular functions

simultaneously, causing maximum inconvenience and

deterrence to any attacker or pathogen, and also assist with the rapid immobilization of prey

Experimental procedures

nNOS bioactivity testing

nNOS inhibition testing was conducted by the Australian Institute of Marine Science (Townsville, Australia) Inhibi-tion was measured and analyzed by monitoring the conver-sion of [3H]arginine to [3H]citrulline by nNOS, using a method reported previously [10]

15N HSQC titration

15N-labelled Ca2+-CaM was prepared by a method based

on that of Elshorst et al [26] Briefly, CaM was expressed

in Escherichia coli strain BL21(DE3), using the expression vector pET28 (Novagen, Madison, WI, USA) CaM expres-sion was induced by addition of isopropyl thio-b-d-galacto-side (0.1 mm), cells were harvested and lysed by sonication, and CaM was purified from the supernatant using anion exchange and size exclusion chromatography

Samples used for the titration series contained15N-labelled

Ca2+-CaM (3.16 mg, 1.89· 10)7mol), potassium chloride (100 mm), calcium chloride (6.2 mm) and 10% D2O in aque-ous solution at pH 6.3 Sodium azide (0.02%) was added as

a preservative [14] Cupiennin 1a (1.44 mg, 3.79· 10)7mol) was dissolved in water, adjusted to pH 6.3 using sodium hydroxide, and then divided into aliquots such that succes-sive additions would achieve total peptide concentrations of 0.2, 0.4, 0.6, 0.8, 1 and 2 molar equivalents The aliquots were then lyophilized, and the dried peptide portions added to the CaM sample in sequence The pH was readjusted back to 6.3 with the addition of small quantities of hydrochloric acid or sodium hydroxide solutions as required

Spectra were recorded using a Varian (Palo Alto, CA, USA) Inova-600 NMR spectrometer, with a 1H frequency

of 600 MHz and a13C frequency of 150 MHz Experiments were conducted at 25C, and referenced to sodium 3-(tri-methylsilyl)propane-1-sulfonate at 0 p.p.m in 1H, whereas the 15N dimension was centred at 120 p.p.m relative to

NH3as 0 p.p.m The standard gNhsqc pulse sequence from the VNMR library was used, with 256 increments, each comprising 16 transients, acquired over 2048 data points A spectral width of 7197.5 Hz was used in the1H dimension, and a spectral width of 2200 Hz in the15N dimension The resultant spectra were processed using nmrpipe [36], and viewed with sparky software (version 3.111) [37]

Acknowledgements

J H Bowie and F Separovic thank the Australian Research Council for the financial support of this

project T L Pukala and M A Apponyi acknowledge

the award of postgraduate scholarships The pET28

vector containing the calmodulin gene was a generous

gift from Dr Joachim Krebs of the Max Plank

Insti-tute for Biophysical Chemistry, Go¨ttingen, Germany

References

1 Barth FG (2002) A Spider’s World Senses and

Beha-viour.Springer-Verlag, Berlin

2 Kuhn-Nentwig L, Schaller J & Nentwig W (2004)

Biochemistry, toxicology and ecology of the venom

of the spider Cupiennius salei (Ctenidae) Toxicon 43,

543–553

3 Kuhn-Nentwig L, Schaller J & Nentwig W (1994)

Puri-fication of toxic peptides and the amino acid sequence

of CSTX-1 from the multicomponent venom of

Cupien-nius salei(Ctenidae) Toxicon 32, 287–302

4 Wullschleger B, Kuhn-Nentwig L, Tromp J, Kanpfer U,

Schaller J, Schurch S & Nentwig W (2004) CSTG-13, a

highly synergistically acting two-chain neurotoxic

enhancer in the venom of the spider Cupiennius salei

(Ctenidae) Proc Natl Acad Sci USA 101, 11251–11256

5 Haeberli S, Kuhn-Nentwig L, Schaller J & Nentwig W

(2000) Characterisation of antibacterial activity of

pep-tides isolated from the venom of the spider Cupiennius

salei(Ctenidae) Toxicon 38, 373–380

6 Kuhn-Nentwig L, Mu¨ller J, Schaller J, Walz A, Dathe M

& Nentwig W (2002) Cupiennin 1, a new family of highly

basic antimicrobial peptides in the venom of Cupiennius

salei(Ctenidae) J Biol Chem 277, 11208–11216

7 Kuhn-Nentwig (2003) Antimicrobial and cytolytic

peptides of venomous arthropods Cell Mol Life Sci 60,

2651–2668

8 Wullschleger B, Nentwig W & Kuhn-Nentwig L (2005)

Spider venom: enhancement of venom efficacy mediated

by different synergistic strategies in Cupiennius salei

J Expt Biol 208, 2115–2121

9 Pukala TL, Boland M, Gehman JD, Kuhn-Nentwig L,

Separovic F & Bowie JH (2006) Solution structure and

interaction of cupiennin 1a, a spider venom peptide,

with phospholipid bilayers Biochemistry in press

10 Doyle JR, Llewellyn LE, Brinkworth CS, Bowie JH,

Wegener KL, Rozek T, Wabnitz PA, Wallace JC &

Tyler MJ (2002) Amphibian peptides that inhibit nNOS

The isolation of lesueurin from the skin secretion of the

Australian Stony Creek Frog Litoria lesueuri Eur J

Bio-chem 269, 100–109

11 Pukala TL, Bowie JH, Maselli VM, Musgrave IF &

Tyler MJ (2006) Host-defence peptides from the

glandu-lar secretions of amphibians: structure and activity Nat

Prod Rep 23, 368–393

12 Fersht A (1985) Cooperative ligand binding and

allos-teric interactions in enzyme structure and mechanism

In Enzyme Structure and Mechanism, pp 208–225 Free-man WH, New York, NY

13 Schon O, Friedler A, Freund S & Fersht AR (2004) Binding of p53-derived ligands to MDM2 induces a variety of long range conformational changes J Mol Biol 336, 197–202

14 Ikura M, Kay LE & Bax A (1990) A novel approach for sequential assignment of1H,13C and15N spectra of larger proteins: heteronuclear triple-resonance three-dimensional NMR spectroscopy Application to calmo-dulin Biochemistry 29, 4659–4667

15 Torizawa T, Shimizu M, Taoka M, Miyano H & Kainosho M (2004) Efficient production of isotopically labelled proteins by cell free synthesis: a practical proto-col J Biomol NMR 30, 311–325

16 Kerwin JF, Lancaster JR & Feldman PL (1995) Nitric oxide: a new paradigm for secondary messengers J Med Chem 38, 4344–4362

17 Stuehr DJ & Ghosh S (2000) Nitric Oxide Handbook of Experimental Pharmacology(Mayer B, ed.), pp 33–70 Springer-Verlag, Berlin

18 Choi WS, Chang MS, Han JW, Hong SY & Lee HW (1997) Identification of NOS in Staphylococcus aureus Biochem Biophys Res Commun 237, 554–558

19 Morita H, Yoshikawa H, Sakata R, Nagata Y & Tanaka H (1997) Synthesis of nitric oxide from the two equivalent quanidino nitrogens of L-arginine by Lacto-bacillus fermentum J Bacteriol 179, 7812–7815

20 Choi WS, Seo DW, Chang MS, Han JW, Paik WK & Lee HW (1998) Methyl esters of L-arginine and N-nitro-L-arginine induce NOS in Staphylococcus aureus Biochem Biophys Res Commun 246, 431–435

21 Er H, Turkoz Y, Ozerol IH & Uzmez E (1998) Effect of NOS inhibition in experimental Pseudomonas keratitis

in rabbits Eur J Opthalmol 8, 137–141

22 Muller U (1994) Ca2+calmodulin-dependent nitric oxide synthase in Apis mellifera and Drosophila melano-gaster Eur J Neurosci 6, 1362–1370

23 Regulski M & Tully T (1995) Molecular and biochem-ical characterization of dNOS: a Drosophila Ca2+ cal-modulin-dependent nitric oxide synthase

Proc Natl Acad Sci USA 92, 9072–9076

24 D’yakonova VE & Krushinskii AL (2006) Effects of an NOS inhibitor on aggressive and sexual behaviour in male crickets Neurosci Behav Physiol 36, 565–571

25 Ikura M, Clore GM, Gronenborn AM, Zhu G, Klee

CB & Bax A (1992) Solution structure of a calmodulin target peptide complex by multidimensional NMR Science 256, 632–638

26 Elshorst B, Hennig M, Forsterling H, Diener A, Maurer

M, Schulte P, Schwalbe H, Greisinger C, Krebs J, Sch-mid H et al (1999) NMR solution structure of a com-plex of calmodulin with a binding peptide of the Ca2+ pump Biochemistry 38, 12320–12332

27 Klee CB & Vanaman TC (1982) Calmodulin Adv Prot

Chem 35, 213–219

28 Chin D & Means AR (2000) Calmodulin: a prototypical

calcium sensor Cell Biol 10, 322–328

29 Hoedlich KP & Ikura M (2002) Calmodulin in action:

diversity in target recognition and activation

mechan-isms Cell 108, 739–742

30 Vetter SW & Leclerc E (2003) Novel aspects of

calmo-dulin target recognition and activation Eur J Biochem

270, 404–414

31 Roth SM, Schmeider DM, Strobel L, van Berkum M,

Means A & Wand AJ (1991) Structure of the smooth

muscle myosin light-chain kinase calmodulin-binding

domain peptide bound to calmodulin Biochemistry 30,

10078–10084

32 Meador WE, Means AR & Quiocho FA (1992) Target

enzyme recognition by calmodulin: 2.4 A˚ structure of a

calmodulin peptide complex Science 257, 1251–1255

33 Aoyagi M, Arvai AS, Tainer JA & Getzoff ED (2003)

Structural basis for endothelial nitric oxide synthase

binding to calmodulin EMBO J 22, 766–775

34 Kataoka M, Head JF, Vorherr T, Krebs J & Carafoli E (1991) Small-angle X-ray scattering study of calmodulin bound to two peptides corresponding to parts of the calmodulin-binding domain of the plasma membrane Ca-pump Biochemistry 30, 6247–6251

35 Ikura M, Kay LE, Krinks M & Bax A (1991) Triple-resonance multidimensional NMR study of calmodulin complexed with the binding domain of skeletal muscle myosin light-chain kinase: indication of a conforma-tional change in the central helix Biochemistry 30, 5498–5504

36 Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J & Bax A (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes J Biomol NMR 6, 277–293

37 Goddard TD, Kneller DG (2006) SPARKY 3

University of California, San Francisco, CA