Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://vertebrata.bio.msu.ru/Samgina_et_al_2009.pdf Äàòà èçìåíåíèÿ: Mon Feb 8 19:41:58 2010 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 19:30:33 2012 Êîäèðîâêà:

RAPID COMMUNICATIONS IN MASS SPECTROMETRY Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.3994

Mass spectrometric study of peptides secreted by the skin glands of the brown frog Rana arvalis from the Moscow region
T. Yu. Samgina1, K. A. Artemenko2, V. A. Gorshkov1, S. V. Ogourtsov3, R. A. Zubarev and A. T. Lebedev1*
1 2 3

2

Organic Chemistry Depar tment, Moscow State University, Moscow, Russia Depar tment of Cell and Molecular Biology, Uppsala, Sweden Biology Depar tment, Moscow State University, Moscow, Russia

Received 22 December 2008; Revised 13 February 2009; Accepted 20 February 2009

A high-performance liquid chromatography nano-electrospray ionization Fourier transform mass spectrometry (HPLC/nanoESI-FTMS) approach involving recording of collision-activated dissociation (CAD) and electron-capture dissociation (ECD) spectra of an intact sample and two its modifications after performic oxidation and reduction followed by carboxamidomethylation helps to establish peptide profiles in the crude secretion of frog species at mid-throughput level, including de novo sequencing. The proposed derivatization procedures allow increasing of the general sequence coverage in the backbone, providing complementary information and, what is more important, reveal the amino acid sequence in the cystine ring (`rana box'). Thus purely mass spectrometric efficient sequencing becomes possible for longer than usual proteolytic peptides. Seventeen peptides belonging to four known families were identified in the secretion of the European brown frog Rana arvalis inhabiting the Moscow region in Russia. Ranatuerins, considered previously a unique feature of the North American species, as well as a new melittin-related peptide, are worth special mention. The developed approach was previously successfully used for the identification of peptides in the skin secretion of the Caucasian green frog Rana ridibunda. Copyright # 2009 John Wiley & Sons, Ltd.
Several recent papers published by our group deal with mass spectrometric study of the primary structure of peptides generated by the dorsal glands of ranid frogs inhabiting Russia.1­3 Antibacterial peptides secreted by amphibians in response to external impulses possess double scientific interest. Due to their wide spectrum of biological activity they are considered to be a powerful source of future pharmaceuticals,4 while they may also serve as taxonomic markers allowing distinction of the related species.5 In addition, the establishment of the composition of the peptide fraction of the amphibian skin secretion allows phylogenetic connections to be found between various species and families in the order of tailless amphibia Anura rafinesque, required to understand pathways of evolutionary development of hylid and ranid frogs.6,7 It is known that the skin secretion of amphibians is a complex mixture of biologically active compounds including biogenic amines, alkaloids and peptides.8 The latter group of compounds in the case of ranid frogs usually includes the following components: (1) neuropeptides of bombesins,9 ranatensins,9 tachykinins10 and bradykinins11; (2) long bactericidal disulphide-containing peptides belonging to 13 various families: brevinins-1 and -2;12 esculentins-1 and -2;13,14 ranatuerins-1 and -2;15 ranalexins;16,17 palustrins-1, -2 and -3;18 tigerinins;19 japonicins-1 and -2;20 and, finally, (3) temporins ­ the shortest species among the known antimicrobial peptides.21 Direct sequencing of the intact disulphide-containing antimicrobial peptides by means of mass spectrometry (MS) is rarely successful, although the use of negative ion MS allowed the primary structure of the authentic brevinin1E, isolated from the skin secretion of the green frog Rana ridibunda, to be established.22 Unfortunately, in the majority of the known examples this approach does not bring the desired results. Thus, preliminary destruction of the SþS bond is required for complete C-terminal sequencing. There are two general methods of SþS bond modification in peptides: (1) reduction followed by derivatization of the reduced cystine residues and (2) oxidation. A recent review23 summarizes the known agents for reducing disulphide bonds as well as derivatizing agents. The reported variants of oxidation are also listed. The present work aims at establishing the primary structure of the peptides constituting the secretion of the skin glands of the common brown frog Rana arvalis from the Moscow region in Russia by means of mass spectrometry. Rana arvalis inhabits a vast territory from south Sweden and Finland to France, south-east Europe and even Siberia.24
Copyright # 2009 John Wiley & Sons, Ltd.

*Correspondence to: A. T. Lebedev, Organic Chemistry Department, Moscow State University, Moscow 119991, Russia. E-mail: lebedev@org.chem.msu.ru


1242 T. Yu. Samgina et al.

There are no published data on the skin peptides of this species. The present study involved consecutive nanospray electrospray ionization Fourier transform mass spectrometry (nanoESI-FTMS) analysis of the crude secretion of Rana arvalis and two samples of the same crude secretion after modification of the SþS bonds. Reduction with dithiothreitol followed by carboxamidomethylation as well as oxidation with performic acid were used. Since each of these two procedures has some advantages and shortcomings their joint application provides complementary structural information, increasing the reliability of the mass spectrometric sequencing of the disulphide-containing peptides.25 To further increase the reliability electron capture dissociation (ECD) and collision-activated dissociation (CAD) spectra were recorded in each case. The developed approach was previously successfully used for the identification of peptides in the skin secretion of the Caucasian green frog Rana ridibunda.2

detector at 214 nm (Spectra Systems, San Jose, CA, USA). The dry samples were stored at þ268C.

Mass spectrometric sequencing of the peptides
All experiments were performed on a 7-Tesla hybrid mass spectrometer with linear ion trap, quadrupole and ion cyclotron resonance (ICR) cell (LTQ FT; Thermo Fisher Scientific, Bremen, Germany) coupled with an 1100 nanoflow system (Agilent, Santa Clara, CA, USA). The mass spectrometer toggles between a survey MS scan (resolving power 100 000) and consecutive ECD and CAD MS/MS scans (resolving power 25000) of the two most abundant peptides detected eluting at that moment from the nano-LC column. The experimental details are described in Kocher et al.27

Disulphide bond reduction and alkylation
The HPLC-purified and lyophilized crude secretion was redispersed in ammonium buffer (100 mM NH4HCO3, pH 8.0) with 10 mM iodoacetamide and incubated in the dark for 1 h at 378C. Prior to treatment with iodoacetamide, dithiothreitol was added at a concentration of 4 mM while nitrogen was flushed into the sample to provide an inert atmosphere throughout the reaction.28

EXPERIMENTAL Skin secretion
Ten male and female specimens of the Moor frog R. arvalis were caught near Zvenigorog in the Moscow region (Russia). The animals were maintained in captivity under conditions close to natural ones. Secretion from the skin glands was provoked by mild electrical stimulation. The procedure details have been described previously.26 The skin of the animal was moistened with deionized water and then treated with a bipolar platinum electrode connected to a laboratory electrostimulator (ESL-1). The pulse parameters were as follows: voltage, 10 V; pulse duration, 5 ms; pulse frequency, 50 Hz. The skin secretion was washed out with a small amount (up to 25 mL) of deionized water into a container with an equal volume of methanol. The mixture was then centrifuged for 15 min at 3000 rpm, the supernatant was filtered through a Millex-FH membrane (PTFE 0.45 mm; Millipore, Billerica, MA, USA) and concentrated at 358C on a rotary evaporator to a volume of 1 mL. To avoid possible seasonal changes in the peptide profile, `milking' was always performed in the same month of the year.

Disulphide bond oxidation
Performic oxidation of the peptides was performed as described previously.29 Formic acid (Sigma, St. Louis, MO, USA) was mixed with 30% hydrogen peroxide (Sigma) (19:1 v/v) and incubated for 1 h at room temperature, forming performic acid. The peptides were dissolved in formic acid and incubated for 1 h with a triple volume excess of the performic acid at 08C. Finally, the sample was diluted fivefold with Milli-Q water, and the resulting solution was lyophilized.

Edman degradation
Automated Edman sequencing was performed using a standard procedure on a model 470 peptide sequencer (Applied Biosystems, Foster City, CA, USA), equipped with a 900A data-processing module.30 Best results were achieved when peptides were adsorbed from 90% acetonitrile in water on an Immobilon membrane treated with ethanolic bioprene solution.

Peptides synthesis HPLC separation of the skin secretion
High-performance liquid chromatography (HPLC) fractionation was carried out using a C18 reversed-phase analytical ° column (5 mm, 110 A, 150 á 4 mm; BioChimMac, Moscow, Russia), equilibrated with 10% acetonitrile/aqueous 0.1% trifluoroacetic acid (TFA) (Acros, Geel, Belgium). The single injection volume was less than 20 mL to improve the separation. Peptide separation was performed with an HPLC system from ThermoSeparation Products (Piscataway, NJ, USA), equipped with two P2000 gradient pumps (ThermoSystem, San Jose, CA, USA). Peptides were purified using a linear gradient from 10 to 70% (60 min) acetonitrile containing 0.1% TFA. The flow rate was 1.0 mL/min. The absorption was measured using a model UV3000 UV
Copyright # 2009 John Wiley & Sons, Ltd.

Synthetic samples of brevinin-1AVa (molecular weight (MW) 1809 Da) and the C-terminal fragment of ranatuerin2AVa (MW 1902 Da) (85% purity) were prepared by GenScript Corporation (Piscataway, NJ, USA) using L-isomers of amino acids.

Antimicrobial activity tests
Synthetic samples of brevinin-1AVa and the C-terminal fragment of ranatuerin-2AVa were tested for their antimicrobial activity at the Department of Microbiology of the Adelaide Institute of Medicine and Veterinary Science (Adelaide, Australia). The method involved establishing inhibition zones produced by the testing peptides on a thin plate covered by agar in the presence of various microorganisms. The process is discussed in detail in Jorgensen
Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 DOI: 10.1002/rcm


Peptides secreted by the skin glands of R. arvalis
2
Non-peptide material

1243

2250 2000 1750

2250 90

2000 80

1750 70

Intensity (mAU)

1250 1000 750 500 250 0 0

1250 50

16 10 8 9 11 12 13 14 15

1000 40

750 30

1

34
10

5 6
12

7

500 20

17 10
250 0

0

2

4

6

8

14 Time (min)

16

18

20

22

24

26

28

Figure 1. HPLC chromatogram of intact skin secretion of R. arvalis.

et al.31 The peptides were tested towards Gram-positive bacteria: Bacillus cereus, Enterococcus faecalis, Leuconostic lactis, Listeria innocua, Micrococcus luteus, Staphylococcus aureus, Staphylococcus epidermis, Staphylococcus uberis and Gramnegative species: Enterobacter clorae, Escherichia coli, Pasteurella multocida and Pseudomonas aeruginisa.

RESULTS AND DISCUSSION
Figure 1 represents a general view of the HPLC chromatogram of the authentic skin secretion of R. arvalis. The sequences of 17 peptides identified in the secretion of this frog are summarized in Table 1. Four disulphide-containing peptides (1­4), one melittin-related peptide 5, neuropeptide bradykinin 6 with its nine intermediates (7­15) as well as intermediates of two bradykinin modifications (16­17) are listed.

Disulphide-containing peptides in the secretion are presented by brevinin-1 and ranatuerin-2 families. In the opinion of some researchers the array of disulphidecontaining peptides in the skin secretion of ranid frogs may characterize certain species and thus these compounds may be treated as a group of taxonomic markers.5 Conlon et al.,5 summarizing the data on frog peptides available in 2004, concluded that ranatuerins-2 may be treated as identity markers for the North American species, never having been reported in the European species. Nevertheless, two peptides from the ranatuerin-2 group were discovered by us in the secretion of the European frog R. arvalis (see Table 1). The unique characteristic feature of these peptides involves formation of a six-membered C-terminal disulphide ring, via the thiol groups of the side chains of cysteins. Figure 2 represents mass spectra of a peptide with nominal molecular mass of 2824 Da in the intact crude R. arvalis

Table 1. Peptides isolated from the skin secretion of the brown frog R. arvalis from the Moscow region
MW 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1809 1872 2824 2911 2361 1059 806 903 1187 1243 1285 1340 1498 1712 2355 1728 1225 Name Brevinin-1AVa Brevinin-1AVb Ranatuerin-2AVa Ranatuerin-2AVb Melittin-related peptide Bradykinin [des-Arg1,Pro2]Bradykinin [des-Arg1]Bradykinin AR-11 RA-11 RI-11 RP-12 RS-14 RL-16 RD-21 [Hyp3]RL-16 [Glu5]RA-11 Sequence FLPLLAASFACTVTKKC-OH FVPLLVSKLVCVVTKKC-OH GLLDVVKGAAKNLLASALDKLKCKVTGC-OH GLMoDMVKGAAKNLFASALDTLKCKITGC-OH FVGAALKVLANVLPPVISWIKQ-NH2 RPPGFSPFR-OH PGFSPFR-OH PPGFSPFR-OH AGRPPGFSPFR-OH RPPGFSPFRIA-OH RPPGFSPFRII-OH RPPGFSPFRIAP-OH RPPGFSPFRIAPAS-OH RPPGFSPFRIAPASTL-OH RPPGFSPFRIAPASTLKR(DE)D-OH RPHypGFSPFRIAPASTL-OH RPPGESPFRIA-OH Peak 13 14 16 15 17 2 3 4 5 7 1 5 6 9 10 8 5

Copyright # 2009 John Wiley & Sons, Ltd.

Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 DOI: 10.1002/rcm

%CH3CN

1500

1500 60


1244 T. Yu. Samgina et al.

y26

(a)

100 90 80

y25
2543.393

Relative Abundance

70 60 50 40 30 20 10 0 400 600 800 1000 1200 1400

1547.835

1774.965

y15

y17 y21
2102.137 2230.219 2329.308

736.359

y7 b5
480.367 498.278

1434.748

1205.677

849.406 882.602

1974.089

597.340

2084.129

b12

y12
1600 m/z

y18
1800

y20

2000

2212.171 2214.224

2200

2400

2600

2800

(GL)LDVVKGAAKNLLASALDKLK( CKVTGC-OH)
MH+

(b)

100 90 80

x6

Relative Abundance

70 60
742.479

2087.135 2108.275

1207.180 1261.671

1397.786

20 10 0

z

20

800

1000

1200

1400

1600

1800 m/z

2000

2200

2400

2413.353

2528.381 2565.516

849.983 870.541

1480.914

30

c9 c10
941.579

1069.672

1752.059

z12

c15

21

c18

1995.171

c20

c21

z24

z

25

z

27

2600

GLLD(VVK)GAAKNL(LA)S(AL)(DK)L(KCKVTGC-OH)
Figure 2. Mass spectra of the peptide with nominal molecular mass of 2824 Da: (a) CAD spectrum and (b) ECD spectrum. If no cleavages are obser ved between some amino acids those groups are presented in parentheses. Series of y and b ions and z and c ions were used to assign fragment identities.

sample. The spectra were recorded in CAD (Fig. 2(a)) and ECD (Fig. 2(b)) modes. The information provided by the b-y and c-z ion series allows us to obtain the complete sequence, excluding the disulphide C-terminal ring. The spectrum of the quadruply charged molecule is richer than that of triply charged one (data not shown). Unfortunately, higher charge states are not generated in the experiment as they are unstable for a peptide with a molecular mass below 3 kDa.32 To define the C-terminal sequence of the peptide the original crude sample was oxidized with performic acid.
Copyright # 2009 John Wiley & Sons, Ltd.

This procedure results in the transformation of the thiol moieties of the cysteins into sulphonic acid groups (-SO3H ­ groups, denoted as COx below).33 Figure 3 represents CAD and ECD spectra of a peptide with molecular mass 2824 Da after its oxidation. The information obtained in the two experiments is sufficient to establish the complete sequence of this peptide, including the C-terminus. The only difficulty involves the relative positions of two residues (Ala17 and Leu18). Thus, four spectra of this peptide consisting of 28 amino acid residues (Figs. 2 and 3) provide its complete sequence:
Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 DOI: 10.1002/rcm

2754.544 2783.567 2794.609

40

c11

1567.943

50

c7

c16 z

2800

2826.582

2826.592

b6

1850.098

y8

977.514 996.632 1092.566

y11

1363.704

y9

b19

2428.377

y10

1660.920

y13

y14

y16 y19

y22 y23 y24 MH+

2656.475


Peptides secreted by the skin glands of R. arvalis

1245

b

22

(a)

100 90 80
659.519

2219.352

b24
1075.516 2498.428

y9 y10
1190.537 1214.652

Relative Abundance

70 60 50

b21 y14
1532.718 2091.257

1461.690

19

671.962

918.496 947.414

1392.856

1980.759

40 30 20 10 0

1567.148

800

1000

1200

1400

1600

1645.270

1800 m/z

2000

2200

2400

2427.296

2526.366

y15

2328.225

b14

1850.076

y8

2072.076

y22

y23 y24

2600

(GL)LDVV(KGA)(AKNL)LA(SAL)DKLK(CoxK)VTGCox-OH
2924.592 742.484 799.506 870.541 941.580 1028.528 1069.675 1175.530 1183.719

(b)

100 90 80

Relative Abundance

70 60 50 40 30 20 10 0 500 1000 1500 m/z 2000 2500 3000

c12 c16
1567.958

1743.890 1750.972 1857.932 1867.105 1951.445

301.224

2128.100 2185.119 2236.377

1308.266 1359.655

1480.917

416.251

2387.366

12

21

z

14

z

26

GLLD(VVK)GAAKN(LL)AS(AL)D(KL)KCox(KVTG)Cox-OH
Figure 3. Mass spectra of the peptide with nominal molecular mass 2824 Da after performic oxidation: (a) CAD spectrum and (b) ECD spectrum. Cox is used to indicate triply oxidized cysteine. If no cleavages are obser ved between some amino acids those groups are presented in parentheses. Series of y and b ions and z and c ions are used to assign fragment identities.

GLLDVVKGAAKNLLASALDKLKCKVTGC-OH. According to the accepted nomenclature,5 and taking into account the structural similarity approach34 (number of amino acid residues in the chain and six-membered C-terminal cycle), we classified it as a new ranatuerin-2, called ranatuerin2AVa. The residues in positions 7, 11, 20, 22, 24 of the ranatuerin2AVa backbone are considered to be Lys, based on the exact
Copyright # 2009 John Wiley & Sons, Ltd.

mass measurements. The amino acid residues in the second and third positions were assigned as Leu on the basis of the MS3 experiments, carried out on the doubly charged b10 ion. Figure 4 shows the ECD product ion spectrum of the doubly charged b10 ion. ECD spectra allow us to distinguish between the isomeric pairs Leu/Ile, based on the losses of isopropyl or ethyl radicals from the respective z ions.35 Figure 4 demonstrates
Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 DOI: 10.1002/rcm

2958.818

c22

653.944

2511.353

c4

c9 c10

z

c15

17

c19

20

z

2772.605

2881.581

c7

z

z

c23

24

2626.382

c3

c8

c11z

10

z

16

z z

25

c27

MH+

2698.540 2755.565

y13

b

y19

b

26

b

27

2800


1246 T. Yu. Samgina et al.
870.535

100
742.479

90 80

43.05 Da

Relative Abundance

70
799.501

60 50
743.484

43.05 Da

725.453

713.382

744.477

756.439

9

10 0 720

726.448

740

742.168

760

780

m/z

800

820

840

860

Figure 4. Par t of the ECD spectrum of the doubly charged b10 ion of the peptide with nominal molecular mass 2824 Da (without extraction).

w8 and w9 (m/z 713.38 and 826.46) ions, formed due to losses of 43.05 Da from the z8 and z9 ions (m/z 756.44 and 869.52), and thus confirming the presence of two Leu residues in the second and third positions of the ranatuerin-2AVa sequence. The remaining four Leu residues (positions 13, 14, 18 and 21) are defined by the Edman degradation of the ranatuerin2AVa fragment, formed by the proteolytic rupture of the peptide bond at the N-terminus of Lys11 (MW 1901 Da, peak 11 in the chromatogram in Fig. 1). The standard procedure for obtaining skin secretions of R. arvalis by electrostimulation of the dorsal glands26 always leads to the formation of two ranatuerin pieces in the peptide mixture: chromatographic peaks 11 (MW 1901 Da, KNLLASALDKLKCKVTGC-OH) and 12 (MW 1923 Da, KNLFASALDTLKCKITGC-OH). To prevent this process taking place, we slightly changed the procedure ­ washing the secretion with deionized water from the skin into the container already filled with methanol. Thus, the time available for the endoproteolytic reaction of splitting the peptides at the Ala­Lys bond was minimized. Ranatuerin-2AVb (MW 2911 Da) was also identified on the basis of mass spectra and other available information. The assignment of Leu residues in positions 2, 13, 18 and 21 was made by analogy with the related ranatuerin-2AVa. The identified C-terminal sequence completely coincided with that characteristic of ranatuerins-2: CKITGC.34 Thus, the amino acid residue in the cystine ring was assumed to be Ile. A peculiarity of this peptide is the presence of two methionines in its sequence. One of them in position 3 was always detected in the oxidized form ­ Metox3. We did not succeed in detecting the non-oxidized form of ranatuerin-2AVb in the authentic crude serum of R. arvalis. However, the secretion always contains two peptides formed
Copyright # 2009 John Wiley & Sons, Ltd.

by proteolytic cleavage at the N-terminus of Lys in ranatuerin-2AVb: GLMDMVKGAA-OH (MW 991 Da) and KNLFASALDTLKCKITGC-OH (MW 1923 Da). The first one contains both methionines in the non-oxidized form. Therefore, oxidation of the authentic peptide takes place during sample preparation. In addition to the two ranatuerins, two new disulphidecontaining 17-residue peptides with molecular masses 1809 and 1872 Da were detected in the secretion. Their sequences were also established on the basis of the ECD and CAD spectra of the authentic and modified samples. Two types of modification were used: (1) performic oxidation of the SþS bond and (2) reduction of the SþS bond followed by carboxamidomethylation. Distinguishing between Leu/Ile pairs was achieved by ECD spectra. The w13, w14 and w16 ions in the ECD spectrum of the doubly protonated peptide with MW 1809 Da, as well as the w9, w13, w14 ions in the ECD spectrum of the doubly protonated peptide with MW . 1872 Da, were formed as a result of C3H7 (43.055 Da) losses from the corresponding z ions, proving the presence of Leu in the mentioned positions. The sequences of these two peptides established by means of mass spectrometry were confirmed by Edman degradation. They were assigned to the brevinin-1 family and named brevinin-1AVa and brevinin1AVb, respectively. Structurally (the number of the amino acid residues in the chain and the size of the disulphide cycle) both new brevinins are closely related to the 17-residue brevinin-1Ta detected previously in the secretion of the European brown frog R. temporaria: FITLLLRKFICSITKKC.8 However, the sequences (CTVTKKC or CVVTKKC) of their disulphide moieties (excluding the second residue) coincide with these of ranalexines, skin peptides of the North American frogs: CAVTKKC.34
Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 DOI: 10.1002/rcm

869.516

20

771.426 772.458 773.014

872.545

30

z

800.507 801.498

w8

826.466 827.487 828.492

853.511 854.513

40

w

9

8

z

871.540


Peptides secreted by the skin glands of R. arvalis

1247

Biotesting of brevinin-1AVa and the C-terminal fragment, consisting of 18 residues, of ranatuerin-1AVa towards 15 species of Gram-positive and Gram-negative bacteria proved their activity for Leuconostic lactis: their IC50 values are 32.4 and 72.9 mM, respectively. Bradykinin (6) and bradykinin-related peptides (7­17) are also present in the skin secretion of R. arvalis. Bradykinin itself appeared to be the major component of the peptide mixture (Fig. 1). This compound is a powerful neuropeptide present in the skin secretion of various types of ranid frogs. It guarantees the functioning of the defense properties of amphibian skin.34 The secretion also contains [des-Arg1, Pro2]bradykinin (7) and [des-Arg9]bradykinin (8). It is worth mentioning that the latter is also present in the plasma of human blood, being generated from bradykinin by carboxypeptidase-C.36 Quite a number of peptide intermediates having several additional amino acid residues at the C- or N-termini of bradykinin were also detected. These are: AR-11 ­ Nextension; RA-11, RI-11, RP-12, RS-14, RL-16, and RD-21 ­ Cextension. These compounds are considered to form as a result of interaction of preprobradykinin with various endoproteases. The detected substitution of one residue at the terminal extension may be rationalized by expression of more than one gene coding bradykinins and by the fact that preprobradykinins contain several copies of the target peptide.36,37 R. arvalis secretion contains also intermediates of two modifications of bradykinin (16, 17). The first one is [Hyp3]RL-16 with Pro3 substituted for hydroxyproline. The intermediate of the modified bradykinin [Glu5]RA-11 has a substitution of phenylalanine for Glu. The biological activity of hydroxyprolylbradykinin is comparable with that of bradykinin itself, while data on the activity of [Glu5]bradykinin are absent so far.34 Non-modified [Hyp3]bradykinin and [Glu5]bradykinin, formed due to proteolysis of 16 and 17 at the C-terminus of Arg, were not detected. This may be a result of the rapid deactivation of endoproteases by methanol during secretion collection. Since bradykinin is the major component of the skin secretion of the brown frog R. arvalis, it is not surprising that an acidic spacer of preprobradykinin DEDEDEYAGEAKAEDVKR-OH was also discovered. It is known that peptide predecessors are kept in the glands of amphibians in a deactivated form. Stimulation of the dorsal glands results in secretion of the inactive propeptides consisting of an acidic spacer and the peptide itself. Elimination of the spacer requires interaction with endoproteases which are secreted by frogs together with the peptides. The liberated peptide then becomes active.6,7,38 It is worth specially mentioning an interesting peptide 5 (Table 1) identified in the secretion of Rana arvalis. Melittin is the basic active component of the venoms of honeybees Apis mellifera and Apis florae.39 Melittin-related peptides (MRPs) have been reported as components of the secretions of brown frogs: European R.temporaria 21,40 and Japanese R. tagoi 41 and R. sakuraii.42 MRPs as well as authentic melittin are active towards Gram-positive and Gram-negative microorganisms as well as fungi. It should be stressed that their cytolytic activity is lower than that of melittin itself.38,39 The MRP from
Copyright # 2009 John Wiley & Sons, Ltd.

the brown frog R. arvalis is a C-terminal a-amidated peptide, containing 22 amino acid residues: FVGAALKVLANVLPPVISWIKQ-NH2. Its sequence is 50% analogous to melittin (48% and 52% ­ to the MRP from brown Japanese frogs R. tagoi and R. sakuraii and 77% to MRP from European R. temporaria). Leu9, Leu13 and Ile20 are established on the basis of the ECD spectrum of the triply protonated MRP with MW 2361 Da. Leu6 and Ile17 are assigned using an analogous principle and taking into account the sequences of melittin and other known MRPs. Reaction of the original skin secretion sample with performic acid resulted in parallel oxidation of tryptophan residues. The masses of the tryptophan-containing product ions were 15.995 m/z units higher. An earlier publication43 states that hydroxytryptophan (HTRP) formation takes place. Similarly to the bradykinin case an acidic spacer of the predecessor of MRP was detected in the secretion of R. arvalis. Its sequence ­ EEENGGEVKEEE ­ completely coincides with that of the acidic spacer of the prepropeptides of MRPs of the brown Japanese frogs R. tagoi and R. sakuraii.44

CONCLUSIONS
A HPLC/nanoESI-FTMS approach involving recording of CAD and ECD spectra of an intact sample of crude secretion and two samples of the same crude secretion after its performic oxidation and reduction followed by carboxamidomethylation allows us to successfully establish peptide profiles of the skin secretion of frog species at the midthoughput level, including de novo sequencing of rather long peptides containing the `rana box' feature. Seventeen peptides belonging to four known families were identified in the secretion of the European brown frog R. arvalis inhabiting the Moscow region in Russia, with the neuropeptide bradykinin being the major constituent. Seventy-seven precent of the sequence of the newly discovered melittin-related peptide FVGAALKVLANVLPPVISWIKQ-NH2 coincides with that of the MRP identified in the skin secretion of the related species of the European brown frog R. temporaria. The secretion from the Moscow species of R. arvalis contains ranatuerins-2 closely related to those found in the secretion of various North American species. The IC50 values of brevinin-1AVa and the 17-residue Cterminal fragment of ranatuerin-1AVa towards Grampositive bacteria Leuconostic lactis are 32.4 and 72.9 mM, respectively.

Acknowledgements
The authors would like to acknowledge Prof. John H. Bowie (Adelaide University, Australia) for the inspiration, the University of Adelaide for providing Edman data and the smooth muscle and antibiotic test results, and Drs R. Doyle and L. Llewellyn for the nNOS results (from AIMS, Queensland). The authors would like to thank Shakhparonov V. V. for keeping the frogs and biological assistance.
Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 DOI: 10.1002/rcm


1248 T. Yu. Samgina et al.

REFERENCES
1. Artemenko KA, Samgina TY, Lebedev AT, Doyle JR, Llewellyn LE, Bilusich D, Bowie JH. Mass-spektrometria 2007; 4: 79. 2. Samgina TY, Artemenko KA, Gorshkov VA, Ogourtsov SV, Zubarev RA, Lebedev AT. Rapid Commun. Mass Spectrom. 2008; 22: 3517. 3. Samgina TY, Artemenko KA, Gorshkov VA, Lebedev AT. Russian Chem.Bull. 2008; 57: 1061. 4. Clarke BT. Biol. Rev. Camb. Philos. Soc. 1997; 72: 365. 5. Conlon JM, Kolodziejek J, Nowotny N. Biochim. Biophys. Acta 2004; 1696: 1. 6. Apponyi MA, Pukala TL, Brinkworth CS, Maselli VM, Bowie JH, Tyler MJ, Booker GW, Wallace JC, Carver JA, Separovic F, Doyle J, Llewellyn LE. Peptides 2004; 25: 1035. 7. Vanhoye D, Bruston F, Nicolas P, Amiche M. Eur. J. Biochem. 2003; 270: 2068. 8. Simmaco M, Mignogna G, Barra D. Biopolymers 1998; 47: 435. 9. Erspamer V. Ann. N. Y. Acad. Sci. 1988; 547: 3. 10. Severini C, Salvadori S, Guerrini R, Falconieri-Erspamer G, Mignogna G, Erspamer V. Peptides 2000; 21: 1587. 11. Anastasi A, Montecucchi P, Erspamer V, Visser J. Experientia 1977; 33: 857. 12. Morikawa N, Hagiwara K, Nakajima T. Biochem. Biophys. Res. Commun. 1992; 189: 184. 13. Simmaco M, Mignogna G, Barra D, Bossa F. FEBS Lett. 1993; 324: 159. 14. Simmaco M, Mignogna G, Barra D, Bossa F. J. Biol. Chem. 1994; 269: 11956. 15. Goraya J, Knoop FC, Conlon JM. Biochem. Biophys. Res. Commun. 1998; 250: 589. 16. Clark DP, Durell S, Maloy WL, Zasloff M, Ranalexin. J. Biol. Chem. 1994; 269: 10849. 17. Halverson T, Basir YJ, Knoop FC, Conlon JM. Peptides 2000; 21: 469. 18. Basir YJ, Knoop FC, Dulka J, Conlon JM. Biochim. Biophys. Acta 2000; 1543: 95. 19. Sai KP, Jagannadham MV, Vairamani M, Raju NP, Devi AS, Nagaraj R, Sitaram N. J. Biol. Chem. 2001; 276: 2701. 20. Isaacson T, Soto A, Iwamuro S, Knoop FC, Conlon JM. Peptides 2002; 23: 419. 21. Simmaco M, Mignogna G, Canofeni S, Miele R, Mangoni ML, Barra D. Eur. J. Biochem. 1996; 242: 788. 22. Bilusich D, Maselli VM, Brinkworth CS, Samguina T, Lebedev AT, Bowie JH. Rapid Commun. Mass Spectrom. 2005; 19: 3063.

23. Artemenko KA, Samgina TY, Lebedev AT. Mass-spektrometria 2006; 3: 225. 24. Kuzmin SL. Amphibians of the Former USSR. Tovarischestvo nauchnyh izdani KMK: Moscow (Russia), 1999. 25. Samgina TY, Artemenko KA, Gorshkov VA, Poljakov NB, Lebedev AT. J. Am. Soc. Mass Spectrom. 2008; 19: 479. 26. Tyler MJ, Stone DJ, Bowie JH. J. Pharmacol. Toxicol. Methods 1992; 28: 199. 27. Kocher T, Savitski MM, Nielsen ML, Zubarev RA. J. Proteome Res. 2006; 5: 659. 28. Allen G. In Laboratory Techniques in Biochemistry and Molecular Biology, vol. 9, Work TS, Burdon RH (eds). Elsevier: Amsterdam, 1981; 17. 29. Barlet O, Yang CY, Gaskell SJ. J. Am. Soc. Mass Spectrom. 1992; 3: 337. 30. Hunkapiller MW, Hewick RM, Dreyer WJ, Hood LE. Methods Enzymol. 1983; 91: 399. 31. Jorgensen JH, Cleeland R, Craig WA, Doern G, Ferraro MJ, Finegold SM, Hansen SL, Jenkins SG, Novick WJ, Pfaller MA, Preston DA, Reller LB, Swenson JM. Methods for Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically, (3rd edn). epproved standard vol. 13. National Committee for Clinical Laboratory Standards, 1993. 32. Lebedev AT. Mass Spectrometry in Organic Chemistry. BINOM. Laboratory of Knowledge: Moscow, 2003. 33. Hirs CHW. Methods Enzymol. 1967; 11: 197. 34. Pukala TL, Bowie JH, Maselli VM, Musgrave IF, Tyler MJ. Nat. Prod. Rep. 2006; 23: 368. 35. Zubarev RA. Mass Spectrom. Rev. 2003; 22: 57. 36. Conlon JM, Aronsson U. Peptides 1997; 18: 361. 37. Chen T, Orr DF, Bjourson AJ, McClean S, O'Rourke M, Hirst DG, Rao P, Shaw C. Eur. J. Biochem. 2002; 269: 4693. 38. Ganz T. CIBA Found. Symp. 1994; 186: 62. 39. Kreil G. FEBS Lett. 1973; 33: 241. 40. Cevikbas A. Toxicon 1978; 16: 195. 41. Conlon JM, Sonnevend A, Patel M, Camasamudram V, Nowotny N, Zilahi E, Iwamuro S, Nielsen PF, Pal T. Biochem. Biophys. Res. Commun. 2003; 306: 496. 42. Suzuki H, Iwamuro S, Ohnuma A, Coquet L, Leprince J, Jouenne T, Vaudry H, Taylor CK, Abel PW, Conlon JM. Peptides 2007; 28: 505. 43. Finley EL, Dillon J, Crouch RK, Schey KL. Protein Sci. 1998; 7: 2391. 44. Suzuki H, Conlon JM, Iwamuro S. Peptides 2007; 28: 2061.

Copyright # 2009 John Wiley & Sons, Ltd.

Rapid Commun. Mass Spectrom. 2009; 23: 1241­1248 DOI: 10.1002/rcm