Isolation and Characterization of Cyclotides from Brazilian Psychotria: Significance in Plant

Defense and Co-occurrence with Antioxidant Alkaloids

Hélio N. Matsuura,† Aaron G. Poth,‡ Anna C. A. Yendo,† Arthur G. Fett-Neto,† and David J. Craik‡,*

†Center for Biotechnology and Department of Botany, Federal University of Rio Grande do Sul, Porto

Alegre, RS, Brazil

‡Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia.

1

ABSTRACT

Plants from the genus Psychotria include species bearing cyclotides and/or alkaloids. The elucidation of factors affecting the metabolism of these molecules as well as their activities may help to understand their ecological function. In the present study, high concentrations of antioxidant indole alkaloids were found to co-occur with cyclotides in Psychotria leiocarpa and P. brachyceras. The concentrations of the major cyclotides and alkaloids in P. leiocarpa and P. brachyceras were monitored following herbivore- and pathogen-associated challenges, revealing a constitutive, phytoanticipin-like accumulation pattern. Psyleio A, the most abundant cyclotide found in the leaves of P. leiocarpa, and also found in P. brachyceras leaves, exhibited insecticidal activity against Helicoverpa armigera larvae. Addition of ethanol in the vehicle for peptide solubilization in larval feeding trials proved deleterious to insecticidal activity, and resulted in increased rates of larval survival in treatments containing indole alkaloids. This suggests that plant alkaloids ingested by larvae might contribute to herbivore oxidative stress detoxification, corroborating, in a heterologous system with artificial oxidative stress stimulation, the antioxidant efficiency of Psychotria alkaloids previously observed in planta. Overall, the present study reports data for eight novel cyclotides, the identification of P. leiocarpa as a cyclotide-bearing species, and the absence of these peptides in P. umbellata.

2

Rubiaceae is among the plant families able to produce both cyclotides and alkaloids, with important representatives of each class of compounds being found in Oldenlandia affinis1 and Coffea arabica,2 respectively. Psychotria is the largest genus within the Rubiaceae family, distributed throughout tropical and subtropical regions of the world.3-5 Understanding the distribution, role, and possible synergies between indole alkaloids and cyclotides in Rubiaceae plants may be a useful tool to clarify their complex phylogeny and biological targets.

Cyclotides are the most abundant naturally-occurring circular proteins in plants.6-7 To date, the main ecological role ascribed to cyclotides relates to their insecticidal properties,8-9 which are thought to arise by cyclotide-induced rupture of midgut epithelium cells.10 Their interaction with membrane lipids appears to occur in a receptor-independent manner,11 a feature that may prevent insects from acquiring resistance.12-13 Cyclotides contain 28 to 39 amino acids (www.cybase.org.au), and are characterized by their distinct structure: three disulfide bonds form a “cystine knot”14 embedded within the circular peptide backbone, which is made up of six loops between the cysteine residues (Figure 1).

The structure is stabilized by hydrogen-bonding interactions with a conserved glutamic acid in loop 1.15

This structure engenders cyclotides with exceptional stability and has led to them being proposed as potential scaffolds in the design of target-specific drugs.12, 16 Biological effects described for naturally occurring cyclotides include uterotonic, anti-HIV, antimicrobial, insecticidal, and molluscicidal activities.10, 17-20

Figure 1. Cyclopsychotride A from Psychotria vellosiana (formerly P. longipes), the first cyclotide described in the genus Psychotria.

3

The described bioactivities of Psychotria cyclotides include: cytotoxic activity against MCF-7 and MCF-7/ADR breast cancer cell lines for psyle A, C and E (P. leptothyrsa);21 antimicrobial properties and hemolytic activity against human red blood cells for cyclopsychotride A (P. vellosiana);17 and inhibitory activity against prolyl-oligopeptidase by psysol 2 (P. solitudinum).22 To date, 16 cyclotide sequences have been identified across eight Psychotria species; namely, P. vellosiana (cyclopsychotride A),15 P. suterella (PS-1),6 P. leptothyrsa (psyles A-F),23 P. brachiata

(psybra 1), P. deflexa (psydef 1 and 2), P. poeppigiana (psypoe 1), P. suerensis (psysue 1 and 2), and

P. solitudinum (psysol 1 and 2).22, 24 In addition, six cyclotides (caripe 1–6) have been described from

Carapichea ipecacuanha (formerly P. ipecacuanha).24 Other cyclotide-containing species, with sequences yet to be characterized, are: P. brachyceras, P. buchtienii, P. chiriquiensis, P. elata, P. goldmanii, P. mortoniana, P. pilosa, P. prunifolia, P. punctata and P. trichophora.6, 24 Sixty-one

Psychotria species apparently lack cyclotides.6, 23-24

Psychotria species are also rich sources of alkaloids.25 Although alkaloids are often ascribed antiherbivore roles, the primary functions of these metabolites in some plants may in fact be in oxidative stress detoxification and control.26 As defense molecules, alkaloids can be neurotoxic or disrupt cell signaling.27-28 Shoots of the southern Brazilian Psychotria accumulate significant concentrations of indole alkaloids (0.2% to 4.5% dry weight) with strong antioxidant activity, but these metabolites are not herbivore deterrents.29-33 It is possible that in these plants a combined defense strategy is in play, with antioxidant alkaloids, both inducible34-35 and constitutive, acting in plant oxidative stress detoxification29-30 and cyclotides fulfilling the role of herbivore inhibition.

The current study focused on the isolation, identification, and structural characterization of new cyclotides from selected Psychotria species, as well as on understanding their potential role in plant defense against insects. In addition, it was sought to better understand whether cyclotide and indole monoterpene alkaloid expression is integrated in the plant defense framework of two of the most common Psychotria species in the understory of the southern Atlantic Forest,36 P. leiocarpa and P. brachyceras.

4

RESULTS AND DISCUSSION

Novel Cyclotides and a Novel Cyclotide-Producing Plant. Until now, cyclotides have not been found in pantropical Psychotria species, with the only exception being P. punctata, which is thought to be a cyclotide-containing species but for which no sequence has yet been reported.24 Of the four species analyzed in this study (P. brachyceras, P. leiocarpa, P. carthagenensis and P. umbellata), only P. brachyceras and P. leiocarpa were found to be cyclotide-producing plants (Table 1). No evidence for cyclotides in P. umbellata leaves was found, indicating it to be a cyclotide non-producing species, along with P. carthagenensis, which has previously been described as a cyclotide non- producing species.21

In P. leiocarpa, evidence for at least eight cyclotides was found and five complete sequences were obtained through sequencing based on MS/MS data from MALDI-TOF/TOF and static nanospray experiments, all of which were novel (psyleio A-E). Evidence for at least 17 cyclotides was obtained for P. brachyceras (Table 1). In total, seven complete sequences – three novel and unique to P. brachyceras, three shared with P. leiocarpa, and one known sequence (cycloviolacin O17 from Viola odorata) – were obtained. Evidence for all Psychotria cyclotide sequences is presented in the

Supporting Information (Figures S1–S6).

The new cyclotides were named following the nomenclature scheme proposed by Broussalis et al.,37 using the first letters of the genus and the specific epithet of each species. Cyclotides were alphabetically ordered based on yields observed in plant material.

5

Table 1. Cyclotides in Psychotria Species

no. peptide native massa RA massa digested massa (Da) (Da) (Da) Psychotria. leiocarpa 1 psyleio A 2988.06 3336.16 3354.15 2 psyleio B 2948.06 3296.16 3314.15 3 psyleio C 2921.04 3269.16 3287.12 4 psyleio D 2925.02 3273.12 3291.11 5 psyleio E 3026.03 3374.15 3392.14 6 3088.97 3436.09 7 3266.13 3614.24 3632.23 8 3289.10 3637.24 Psychotria. brachyceras 1 psyleio A 2988.85 3336.94 3354.91 2 psyleio B 2948.85 3296.92 3314.92 3 psyleio D 2925.80 3273.89 3291.87 4 psybry A 3289.03 3637.10 3655.07 5 psybry B 3271.01 3619.66 6 3187.01 3535.09 7 2940.82 3288.90 8 cyO17 3153.04 3501.12 9 3227.99 3575.03 3593.98 10 psybry C 3231.99 3579.08 11 3281.14 3629.18 12 3291.02 3639.09 3657.09 13 3299.02 3647.12 14 3303.03 3651.10 15 3307.03 3655.10 16 3320.99 3668.07 17 3345.04 3693.13 a All presented masses correspond to uncharged masses (M0).

Psychotria carthagenensis from southern Brazil lacks cyclotides, as reported for specimens from several locations,6, 23 and is also devoid of alkaloids.38 However, some variations in metabolite profiles have been described for plants originating from other regions. Specimens found in northern

South America were used as a replacement for P. viridis as a source of the psychoactive alkaloid N,N- dimethyltryptamine.

Cyclotide Yields in Plant Leaves. Based on yields of the isolation protocol, P. leiocarpa contained four major cyclotides: psyleio A–D. Psyleio A was present at a concentration of approximately 0.1% dry weight, and was at least twice as abundant as psyleio B (approximately 0.05% dry weight). P. brachyceras expressed three abundant cyclotides: psybry A, psyleio A and psybry B, with psybry B being around ten times less abundant (0.01% dry weight) than psybry A (approximately

0.1% dry weight); psyleio A was found in concentrations near 0.05% dry weight in P. brachyceras leaves. Psyleio B and D were found in P. brachyceras leaves in much lower concentrations.

Representative sequencing and characterization data for the first characterized cyclotide in this study, psyleio A, are shown in Figure 2. 6

Figure 2. Characterization of psyleio A from Psychotria. leiocarpa. (A) MALDI-MS of Psychotria. leiocarpa extract. The putative cyclotide with a mass of 2989.3 Da was reduced and alkylated, and trypsin- or endoproteinase GluC-digested prior to MS/MS analysis. (B) MALDI-TOF/TOF sequencing of the 3355.2 Da precursor from trypsin and endoproteinase GluC digests allowed the characterization of psyleio A based on observed b- and y-ions. All masses are monoisotopic. MS evidence for other cyclotides in Psychotria leiocarpa and Psychotria brachyceras is provided in the Supporting Information (Figures S1–S6).

Insecticidal Activity. Psyleio A had insecticidal activity against Helicoverpa armigera larvae.

Specifically, larvae fed an artificial diet containing psyleio A at 1 µmol g-1 exhibited a slower rate of growth than control larvae (p < 0.01), and 50% of the psyleio A-fed larvae died within 48 h (Figure 3).

For all other treatments, the larval growth rate was equivalent to the control diet, with the fastest growth rate observed between 48 h and 72 h, with no larval mortality (Figure 3). Psyleio A has either five or eight amino acid differences when compared to kalata B2 or kalata B1 (O. affinis), respectively.

A similar difference in composition is true for kalata B1 and kalata B2,39 with both showing insecticidal activity. The mode of action for kalata B1 involves disruption of cell membranes,11 and

7 amino acids implicated in insecticidal activities (L2, P3 and P24 from loops 5 and 6) are also found in the psyleio A structure, as are residues implicated in anthelmintic activity, i.e. G6, E7, T8 and G12, from loops 1 and 2.40

Figure 3. Larval feeding assay of Psychotria cyclotides on Helicoverpa armigera. Relative growth (fold-change from starting mass) after 24, 48, and 72 h was compared between larvae fed with control diet, or a diet containing cyclotides (psyleio A, psybry A) solubilized in water. Geometric forms represent individual larvae. Statistical differences were analyzed using one-way ANOVA, followed by a post-hoc Tukey test: **p < 0.01.

The most abundant cyclotide in P. brachyceras (psybry A, present at approximately 0.1% dry weight in leaves) did not affect the growth rate or survival of H. armigera in larval feeding assays.

However, among the cyclotides shared between P. brachyceras and P. leiocarpa, psyleio A, the most

8 abundant cyclotide in P. leiocarpa and second most abundant in P. brachyceras, showed insecticidal activity against H. armigera. Moreover, psybry A could target different guilds of herbivores. Only single-purified cyclotides were tested but it is possible that the pool of cyclotides reported in both plants (at least six untested cyclotides in P. leiocarpa and 14 in P. brachyceras) could enhance insecticidal activity synergistically. Psychotria alkaloids such as N,β-D-glucopyranosyl vincosamide

(GPV) in P. leiocarpa29 have strong antioxidant properties, possibly helping plants to cope with oxidative stress associated with herbivory and abiotic stress conditions.26 Accordingly, these metabolites might also be acting as scavengers of reactive oxygen species and alleviating the cytotoxic effect of ethanol on the Helicoverpa larvae. While the presence of alkaloids seemed to be beneficial to larvae, reflected by reduced mortality (Figure 4, graphs A and B), statistical analysis did not reveal a statistically significant difference for response between larvae fed alkaloid-spiked and ethanol-only diet in log-rank analysis, and thus should be addressed in future studies. Plant extracts also showed a similar profile (Figure 4, graph C); however, it is important to consider that plants contain a range of compounds, including flavonoids, alkaloids, cyclotides, and tannins, which could potentially be harmful or protective to the larvae.

9

A p s y c h o ll a t in e 1 µ M 1 0 0

b r a c h y c e r i n e 1 µ M

)

%

(

l 7 5

G P V 1 µ M

a

v

i

v r

u 5 0

c o n tr o l

s

e

a v

r 2 5

a l

0

B p s y c h o lla t in e 5 n M 1 0 0

b r a c h y c e r in e 5 n M

)

%

(

l 7 5

G P V 5 n M

a

v

i

v r

u 5 0

c o n tr o l

s

e

a v

r 2 5

a l

0

C P . u m b e l la t a 1 0 0

P . l e i o c a r p a )

% P . b r a c h y c e r a s

(

l 7 5

P . c a r t h a g e n e n s i s

a

v

i

v r

u 5 0

c o n tr o l

s

e

a v

r 2 5

a l

0 0 2 4 4 8 7 2 H o u r s

Figure 4. Kaplan-Meier survival curve for Helicoverpa armigera (n = 4) after 72 h feeding trials in 2% ethanol treatments containing the alkaloids brachycerine, GPV, or psychollatine, at 1 µM (A) or 5 nM (B), or containing Psychotria extracts (C). The data shows the percentage of larvae alive as a function of time from the start of the trial. N.S. via log-rank (Mantel-Cox) analysis.

Response to Wounding and Defense Signaling Molecules Related to Herbivores and

Pathogens in P. brachyceras and P. leiocarpa. The extracts from both Psychotria species examined here are very complex, which made the use of a refined method to monitor specific compounds necessary, the first step being to isolate target compounds from the respective plants. Complementary 10 to a previous study on the induction of cyclotide precursors in Oldenlandia affinis41 no alterations were found in cyclotide levels in either P. leiocarpa or P. brachyceras following mechanical wounding or exposure to herbivore and pathogen-related phytohormones (jasmonate and salicylate) (Figure 5), indicating a phytoanticipin-like (i.e., constitutive accumulation) profile for these compounds, at least for psybry A and psyleio A–D in these two Psychotria species. This is consistent with a previous qPCR study which found no induction of cyclotides genes in Viola baoshanensis following mechanical wounding.42 No alterations in alkaloid content were observed 96 hours post stimulation (Figure 5).

Figure 5. Cyclotides and alkaloid levels in Psychotria leiocarpa and Psychotria brachyceras leaves 96 h after treatment (mechanical wounding, jasmonate and salicylate). The four most abundant cyclotides (psyleio A–D) and major alkaloid (GPV) from Psychotria leiocarpa, and the most abundant cyclotide and alkaloid from Psychotria brachyceras (psybry A and brachycerine, respectively) were monitored. Three cyclotides (psyleio A, B and D) shared between the two species were also monitored in Psychotria brachyceras. Data from the extracted ion chromatograms (XIC) from LC-MS analyses of each sample were normalized using an internal standard (a conotoxin) and internal standard curve. No statistically significant differences were observed in yields between different treatments (one-way ANOVA followed by a post-hoc Tukey test; p < 0.05). Error bars indicate one standard deviation.

11

GPV (P. leiocarpa) and brachycerine (P. brachyceras) levels have been shown previously to be non-responsive and responsive, respectively, to the elicitation treatments used in the present work.

Although GPV was reported to be present at constant levels of 2.5% dry weight in adult plants,29 the level of brachycerine (up to 0.3% dry weight) has been shown previously to increase in response to wounding and jasmonate treatments in P. brachyceras.30, 34 The lack of evidence for induction of brachycerine accumulation by wounding is consistent with earlier work which demonstrated peak induction 48 h following damage and a return of alkaloid concentration to control levels at 96 h. 34

While previous work demonstrated that brachycerine levels accumulate consistently over 6 days to significantly higher levels in P. brachyceras leaf tissue after jasmonic acid application, methyl jasmonate-treated leaves in the present study did not accumulate brachycerine. This observed lack of response at 96 hours therefore may reflect differences in the time course of induction of brachycerine in whole cuttings versus leaf disks, or between jasmonic acid and methyl jasmonate.

The cyclotide yields obtained here through physical isolation or quantified via the “standard addition” method were in agreement, confirming that the concentration of the major cyclotide from P. leiocarpa, psyleio A, and the major cyclotide from P. brachyceras, psybry A, is close to 0.1% dry weight in the leaves of each plant. The yield of the combined pool of naturally occurring cyclotides is close to 0.2% dry weight in each species and may be even more effective in preventing herbivory attack than is seen from assays on the individual cyclotides.

In summary, we present a new cyclotide-containing species, P. leiocarpa, and eight novel cyclotides (Table 2) from two Psychotria species, increasing the total number of cyclotides in the genus to 24. The work presented in this study has established that cyclotides from southern Brazilian

Psychotria species seem to be involved in protection against herbivores, a finding that could explain the high degree of leaf predation observed in the field for the cyclotide-free and alkaloid-free plant P. carthagenensis. In support of this proposal, P. carthagenensis is not only pollinated both by exotic and native bee species, but also interacts with a relatively large spectrum of secondary pollinators,43 which could be a reflection of reduced chemical defenses. A combined strategy, with indole alkaloids acting

12 in a general defense mechanism as antioxidants and cyclotides imparting insecticidal properties, could be advantageous and might help to explain the much lower leaf predation observed in the field for P. leiocarpa and P. brachyceras, in addition to their relative abundance in the understory of Brazilian

Atlantic Forest.36

Table 2. Novel Cyclotides in Psychotria Species (Rubiaceae) and Corresponding Sequences species peptide sequence mass (Da)a Loop 1 2 3 4 5 6 Psychotria psyleio A -G-LPI C GET C -FTGT C --NTPG C S C -TYPI C TRD 3354.22 leiocarpa psyleio B -GDLPI C GET C -FGGT C --NTPG C V C -AWPV C NR 3314.22 psyleio C -GDLPV C GET C -FGGT C --NTPG C V C -AWPV C TR 3287.21 psyleio D -G-LPV C GES C -FGGT C --NTPG C S C -TWPV C TRD 3291.17 psyleio E SVTPIV C GET C -FGGT C --NTPG C S C -SWPI C TK 3392.28 Psychotria psyleio A -G-LPI C GET C -FTGT C --NTPG C S C -TYPI C TRD 3354.22 brachyceras psyleio B -GDLPI C GET C -FGGT C --NTPG C V C -AWPV C NR 3314.22 psyleio D -G-LPV C GES C -FGGT C --NTPG C S C -TWPV C TRD 3291.17 psybry A -GFNP- C GET C IWFPT C --HAPG C T C SIANI C VRN 3655.40 psybry B -GFNP- C GET C WNKPT C --HAPG C T C SIANI C VRN 3637.39 psybry C -GFNP- C GET C QIDQT C --HAPG C T C SIANI C VRN 3596.35 cycloviolacin O17 -G-IP- C GES C -VWIP C ISAAIG C S C -KNKV C YRN 3515.44 Oldenlandia kalata B1 -G-LPV C GET C -VGGT C --NTPG C T C -SWPV C TRN 3256.19 affinis kalata B2 -G-LPV C GET C -FGGT C --NTPG C S C -TWPI C TRD 3319.19 a Masses (M0) of reduced, carboxyamidomethylated and linearized forms of cyclic peptides are shown. b Assignment of isoleucine/leucine was based in sequence homology with known cyclotides (www.cybase.org.au); for cycloviolacin O17 chymotrypsin digests using the same conditions as for trypsin were also conducted to identify the positions of leucine residues.

Psychotria umbellata proved to be devoid of cyclotides, although it displayed some of the highest concentrations of a single indole alkaloid, psychollatine (4% dry weight in inflorescences),44 in addition to three psychollatine-derived alkaloids.45 Although the insecticidal activity of P. umbellata alkaloids is yet to be evaluated, it is interesting to note that this cyclotide-lacking species has a relatively scarce distribution in the understory of Brazilian Atlantic Forest, being found in specific locations and with few individual plants. P. gitingensis was reported to be an alkaloid-free species,46 and if it also proves to be devoid of cyclotides, it could be used to further evaluate the role of these natural products in determining the susceptibility to insects, in parallel with similar studies on P. carthagenensis. Screening other Psychotria species for the co-occurrence of antioxidant indole alkaloids and cyclotides as a putative (and perhaps widespread) defense strategy against herbivores should be pursued in the future.

13

EXPERIMENTAL SECTION

Plant Material. Psychotria spp. were collected between January and February 2015 in Morro

Santana – Federal University of Rio Grande do Sul (UFRGS) in the city of Porto Alegre, RS, Brazil

(30°04’11”S, 51°07’15”W), under harvest authorization Sisbio/ICMBio 32855-1 (authentication code:

58482685). Leaves of P. brachyceras Müll Arg., P. carthagenensis Jacq., P. leiocarpa Cham. and

Schltdl. and P. umbellata Vell. were immediately frozen in liquid nitrogen after harvesting, freeze- dried, and stored at -80 °C until processing. Voucher specimens (7899: P. brachyceras; 7901: P. carthagenensis; 138157: P. leiocarpa; and 98869: P. umbellata) were deposited at the ICN herbarium

(UFRGS). Freeze-dried material was pulverized for analysis and isolation steps (referred to as dried plant material). For wounding and jasmonate and salicylate exposure assays, shoots of P. leiocarpa and

P. brachyceras containing six to eight leaves were harvested and acclimated for five days under controlled conditions [10% Murashige and Skoog nutrient solution47 pH 5.8, 16 h per day of photoperiod at 73 µmol m-2 s-1 of photosynthetically active radiation and 25 ± 3°C] prior to treatment application.

Extraction. Extraction of cyclotides from P. leiocarpa and P. brachyceras leaves was performed based on the protocol described by Mahatmanto et al.,48 with minor modifications. A solution of acetonitrile, water and formic acid (25:24:1) [acetonitrile (Millipore, Billerica, MA, USA); formic acid (Sigma-Aldrich, St. Louis, MO, USA)] was added to dried plant material (1:20; m/v) and extracted for 1 h with agitation via a magnetic stirrer. The extract was filtered through qualitative filter paper, centrifuged at 5500 g for 10 min at room temperature and the supernatant washed with dichloromethane (1:1; v/v); the aqueous layer had most of organic solvent removed by rotary evaporation (below 45 °C), and then filtered through a 0.45 µm membrane before being freeze-dried.

This product was referred to as a “crude extract”.

Reduction, Alkylation and Enzymatic Digestion of Extracts. A plant extract was obtained in each case by dissolving dried plant material (P. brachyceras, P. carthagenensis, P. leiocarpa, and P. umbellata) in 20% acetonitrile (Sigma-Aldrich, St. Louis, MO, USA). Reduction of disulfide bonds

14 was performed by adding 100 mM ammonium bicarbonate buffer (Sigma-Aldrich) (1:1; v/v) (pH 8.0), and 100 mM dithiothreitol (DTT) (Astral, Gymea, NSW, Australia) to samples (1:10; v/v) before incubation at 60 °C for 30 min; alkylation of reduced cysteine residues was performed by addition of

250 mM iodoacetamide (Sigma-Aldrich) to reduced samples (1:10; v/v) and incubation for 20 min at room temperature. Reduced and alkylated samples were enzymatically digested by adding 50 ng µL-1 of either trypsin (Sigma-Aldrich) or endoproteinase GluC (Sigma-Aldrich) or both to samples (1:1; v/v) and incubating for 3 h at 37 °C; the reaction was quenched by adding 1% formic acid (1:4; v/v). All samples were desalted using C18 Zip Tips (Millipore) before MS analysis.

MALDI-TOF/TOF Analysis and Cyclotide Sequencing. MALDI-TOF/TOF analysis was performed on crude native, reduced and alkylated, and digested samples in a 4700 Proteomics Analyzer

(AB Sciex, Framingham, MA, USA) operated in positive reflector mode acquiring 1600–10,000 total shots per spectrum with a laser intensity of 3700. α-Cyano-4-hydroxycinnamic acid (CHCA) (Protea,

St. Louis, MO, USA) 5 mg mL-1 in 50% (v/v) acetonitrile was used as matrix; 0.7 µL of matrix were added to 0.7 µL of sample, and 0.7 µL of the mixture were spotted on the MALDI plate. Spectra were acquired and processed using 4700 Analyzer Software, and manual sequencing was used to identify cyclotides. Additionally, sequencing by nanospray MS/MS on a QStar Pulsar XL (AB Sciex) mass spectrometer was performed as required. Spectra were acquired between m/z 60 and 2000 under a capillary voltage of 900 V for both TOF and product-ion spectra; the collision energy for peptide fragmentation varied between 10 and 50 V, and Analyst 1.1 software was used to acquire and process data.6 Peptides were sequenced based on the N-terminal b-ion and the C-terminal y-ion fragmentation observed in the MS/MS spectra.

Cyclotide Purification. The crude extracts from P. leiocarpa and P. brachyceras were solubilized in 10% acetonitrile (Sigma-Aldrich) and processed using 10 g solid-phase extraction C18 columns (Agilent, St. Clara, CA, USA). Sample elution from the column was done with stepwise solvent elution, with increasing concentrations of acetonitrile [100 mL of each solution, containing 1% formic acid and acetonitrile at 20, 30, 40, 50, 60, 70 or 80% (v/v)]. Fractions containing masses in the

15 cyclotide-compatible range (2800–3900 Da); as identified by LC-MS, were combined (30% to 60% acetonitrile) and freeze-dried. Combined dried fractions were resuspended in 10% acetonitrile and the components separated via HPLC using a preparative C18 column (Jupiter; 250 x 21.2 mm, 300 Å;

Phenomenex, Torrance, CA, USA) with a linear gradient starting with solvent A (0.05% TFA) and ending with solvent B (90% acetonitrile, 0.05% TFA), at 1% min-1 solvent B gradient and a flow rate of 8 mL min-1. Samples with peaks indicating cyclotide masses (checked by LC-MS) were freeze-dried, resuspended in 10% acetonitrile, and purified by semi-preparative HPLC (Jupiter C18 column; 250 x 10 mm, 300 Å, Phenomenex) using a linear gradient starting with solvent A and ending with solvent B, at

0.33% min-1 solvent B gradient and a flow rate of 3 mL min-1.

Insecticidal Assay. Helicoverpa armigera eggs were obtained from The University of

Queensland, where a culture is maintained at 25 °C in a 12 L : 12 D cycle,49 supplemented with wild- caught moths. H. armigera eggs were hatched on wet filter paper and the larvae were grown on an artificial diet until the second instar, where the larval masses ranged from 4 to 11 mg. Before the feeding trials, larvae were starved for 22 h (n = 4). Feeding trials were conducted for 72 h with larvae maintained at 25 °C throughout the experiment, and their weights recorded at 0, 24, 48, and 72 h. Diets containing wheat germ, yeast, soy flour, agar and a set of antibacterial and antifungal agents50-51 were provided. Test diets contained the cyclotides psyleio A, psyleio B, or psybry A, or the alkaloids N,β-D- glucopyranosyl vincosamide (GPV), brachycerine, or psychollatine, at 1 µmol g-1 or 5 nmol g-1; P. leiocarpa, P. brachyceras, P. umbellata and P. carthagenenis leaf extracts were also tested at 1.25 mg g-1. N,β-D-glucopyranosyl vincosamide (GPV), brachycerine and psychollatine were isolated previously from Psychotria leiocarpa Cham. & Schltdl., P. brachyceras Mull. Arg. and P. umbellata

Vell. leaves, respectively (harvest license by Federal Authority Sisbio/ICMBio 32855-1; authentication code: 58482685), following described methods.31-32, 52 High Performance Liquid Chromatography

(HPLC) was used to evaluate compound purity. These alkaloids were of at least 97% purity. The control diet contained no added peptide or alkaloid. Psyleio B, the alkaloids, and the plant extracts were solubilized in 2% undenaturated ethanol due to solubilization issues in water; a 2% undenaturated

16 ethanol control without added peptides or alkaloids was also evaluated in this case.

Wounding Assay. Previously acclimated shoots of P. leiocarpa and P. brachyceras had half of the total leaves on the shoots damaged four to six times using tweezers. Control plants remained intact under the same experimental conditions (i.e., with the same acclimation treatment). Plants remained in the same room for 96 h. Preliminary experiments were conducted to screen for changes in secondary metabolism due to signaling by volatile compounds; no changes were observed, allowing treated and untreated plants to be kept in the same room. After 96 h, samples were harvested, immediately frozen in liquid nitrogen, and stored at -80 °C until processing.

Jasmonate and Salicylate Assay. Leaf disks 1 cm in diameter were prepared from previously acclimated shoots of P. leiocarpa and P. brachyceras. Thirty disks were distributed per petri dish (total of 120 disks distributed in four petri dishes per treatment) containing 10% Murashige and Skoog nutrient solution, pH 5.8, in the absence (control) or presence of jasmonate or salicylate at a final concentration of 400 µM. After 96 h, the samples were harvested, immediately frozen in liquid nitrogen, and stored at -80 °C until processing. The wounding inflicted during the process of making the leaf disks did not affect the general metabolism of P. leiocarpa and P. brachyceras, as previously reported.29-30

Method of Standard Addition for Absolute Quantitation of Indole Alkaloids and

Cyclotides. Levels of the main alkaloid and the most abundant cyclotides from P. brachyceras and P. leiocarpa were evaluated in control and treated samples using the method of standard addition,53 allowing an accurate quantitation of each analyzed compound. Samples from the wounding, jasmonate and salicylate treatment groups were dissolved in 50% acetonitrile, 1% formic acid at a final concentration of 125 µg mL-1, and injected onto a Shimadzu CT-20A UPLC system (Shimadzu, Kyoto,

-1 Japan) with a flow rate of 4 µL min on a Kinetex column (1.7 µm C18 100 Å, 100 x 2.10 mm)

(Phenomenex). UPLC eluent was coupled directly to a 4000 QTRAP LC/MS/MS System (Applied

Biosciences/MDS Sciex, Foster City, CA, USA) with a turbospray ionization source. Five compounds specific to each species were monitored in samples via SIM (single ion monitoring) as follows:

17 cyclotides psyleio A–D and the alkaloid GPV in P. leiocarpa samples, and cyclotides psybry A, psyleio

A, B, D, and the alkaloid brachycerine in P. brachyceras. Source conditions were essentially set as previously described,54 with minor modifications. Standard compounds were added at concentrations of

125, 250, 375 and 500 ng mL-1, and an unrelated peptide, conotoxin cVc1.155 at a final concentration of

1000 ng mL-1 was used as an internal standard. Data were acquired and processed using Analyst QS 2.0 software and compounds quantitated via peak height.

Experimental Procedures and Statistics. All assays were performed in biological quadruplicate, and a technical duplicate was included whenever possible. The results were analyzed using a Student’s t-test or one-way ANOVA, followed by Tukey post-processing, using GraphPad

Prism 6® software. Statistical significance was evaluated at p < 0.01 or p < 0.05.

ASSOCIATED CONTENT

AUTHOR INFORMATION

*Corresponding Author:

Professor David J. Craik

Tel: 61-7-3346 2019

Fax: 61-7-3346 2101

E-mail: [email protected]

Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS

D. Craik is funded by an Australian Research Council Australian Laureate Fellowship (FL150100146).

A. Fett-Neto is a National Council for Scientific and Technological Development (CNPq-Brazil)

Professorial Fellow. A. Fett-Neto’s research and H. Matsuura’s scholarships were supported by grants

18 from CNPq (306079/2013-5) and the Committee for Improvement of Higher Education Personnel

(CAPES) (registration 99999.009943/2014-05). We are grateful to O. Cheneval and Y.-H. Huang (The

University of Queensland) for HPLC technical support, A. Wan Mamat (The University of

Queensland) for assistance with the culture of H. armigera larvae, D. Demartini (Federal University of

Rio Grande do Sul) for initial peptide analysis from Psychotria material, Prof. M. Zalucki for kindly donating H. armigera eggs, and A. Jones and the Institute for Molecular Bioscience Mass

Spectrometry Facility (The University of Queensland) for expertise and access to MS equipment. We thank A. Cooper for editorial assistance.

REFERENCES

(1) Saether, O.; Craik, D. J.; Campbell, I. D.; Sletten, K.; Juul, J.; Norman, D. G., Biochemistry

1995, 34, 4147-4158.

(2) Denoeud, F.; Carretero-Paulet, L.; Dereeper, A.; Droc, G.; Guyot, R.; Pietrella, M.; Zheng, C.;

Alberti, A.; Anthony, F.; Aprea, G.; Aury, J. M.; Bento, P.; Bernard, M.; Bocs, S.; Campa, C.; Cenci,

A.; Combes, M. C.; Crouzillat, D.; Da Silva, C.; Daddiego, L.; De Bellis, F.; Dussert, S.; Garsmeur, O.;

Gayraud, T.; Guignon, V.; Jahn, K.; Jamilloux, V.; Joet, T.; Labadie, K.; Lan, T.; Leclercq, J.;

Lepelley, M.; Leroy, T.; Li, L. T.; Librado, P.; Lopez, L.; Munoz, A.; Noel, B.; Pallavicini, A.;

Perrotta, G.; Poncet, V.; Pot, D.; Priyono; Rigoreau, M.; Rouard, M.; Rozas, J.; Tranchant-Dubreuil,

C.; VanBuren, R.; Zhang, Q.; Andrade, A. C.; Argout, X.; Bertrand, B.; de Kochko, A.; Graziosi, G.;

Henry, R. J.; Jayarama; Ming, R.; Nagai, C.; Rounsley, S.; Sankoff, D.; Giuliano, G.; Albert, V. A.;

Wincker, P.; Lashermes, P., Science 2014, 345, 1181-4.

(3) Nepokroeff, M.; Bremer, B.; Sytsma, K. J., Syst. Bot. 1999, 24, 5-27.

(4) Nepokroeff, M.; Sytsma, K. J.; Wagner, W. L.; Zimmer, E. A., Syst. Biol. 2003, 52, 820-838.

(5) Razafimandimbison, S. G.; Taylor, C. M.; Wikström, N.; Pailler, T.; Khodabandeh, A.; Bremer,

B., Am. J. Bot. 2014, 101, 1102-1126.

19

(6) Gruber, C. W.; Elliott, A. G.; Ireland, D. C.; Delprete, P. G.; Dessein, S.; Göransson, U.; Trabi,

M.; Wang, C. K.; Kinghorn, A. B.; Robbrecht, E.; Craik, D. J., Plant Cell 2008, 20, 2471-2483.

(7) Gerlach, S. L.; Göransson, U.; Kaas, Q.; Craik, D. J.; Mondal, D.; Gruber, C. W., Biopolymers:

Pept. Sci. 2013, 100, 433-437.

(8) Jennings, C.; West, J.; Waine, C.; Craik, D.; Anderson, M., Proc. Natl. Acad. Sci. U. S. A. 2001,

98, 10614-10619.

(9) Poth, A. G.; Colgrave, M. L.; Philip, R.; Kerenga, B.; Daly, N. L.; Anderson, M. A.; Craik, D.

J., ACS Chem. Biol. 2011, 6, 345-355.

(10) Barbeta, B. L.; Marshall, A. T.; Gillon, A. D.; Craik, D. J.; Anderson, M. A., Proc. Natl. Acad.

Sci. U. S. A. 2008, 105, 1221-1225.

(11) Henriques, S. T.; Craik, D. J., ACS Chem. Biol. 2012, 7, 626-636.

(12) Jackson, M. A.; Gilding, E. K., In Advances in Botanical Research: Plant Cyclotides, Craik, D.

J., Ed. Academic Press: London, 2015, Vol. 76, Chapter 10, pp 305-333.

(13) Henriques, S. T.; Huang, Y.-H.; Chaousis, S.; Sani, M.-A.; Poth, A. G.; Separovic, F.; Craik, D.

J., Chem. Biol. 2015, 22, 1087-1097.

(14) Craik, D. J.; Daly, N. L.; Bond, T.; Waine, C., J. Mol. Biol. 1999, 294, 1327-1336.

(15) Witherup, K. M.; Bogusky, M. J.; Anderson, P. S.; Ramjit, H.; Ransom, R. W.; Wood, T.;

Sardana, M., J. Nat. Prod. 1994, 57, 1619-1625.

(16) Thorstholm, L.; Craik, D. J., Drug Discov. Today: Technol. 2012, 9, e13-e21.

(17) Tam, J. P.; Lu, Y. A.; Yang, J. L.; Chiu, K. W., Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 8913-

8918.

(18) Gran, L.; Sandberg, F.; Sletten, K., J. Ethnopharmacol. 2000, 70, 197-203.

(19) Gustafson, K. R.; McKee, T. C.; Bokesch, H. R., Curr. Protein Pept. Sci. 2004, 5, 331-340.

(20) Plan, M. R. R.; Saska, I.; Cagauan, A. G.; Craik, D. J., J. Agric. Food Chem. 2008, 56, 5237-

5241.

20

(21) Gerlach, S. L.; Rathinakumar, R.; Chakravarty, G.; Göransson, U.; Wimley, W. C.; Darwin, S.

P.; Mondal, D., Biopolymers: Pept. Sci. 2010, 94, 617-625.

(22) Hellinger, R.; Koehbach, J.; Puigpinos, A.; Clark, R. J.; Tarrago, T.; Giralt, E.; Gruber, C. W.,

J. Nat. Prod. 2015, 78, 1073-1082.

(23) Gerlach, S. L.; Burman, R.; Bohlin, L.; Mondal, D.; Göransson, U., J. Nat. Prod. 2010, 73,

1207-1213.

(24) Koehbach, J.; Attah, A. F.; Berger, A.; Hellinger, R.; Kutchan, T. M.; Carpenter, E. J.; Rolf, M.;

Sonibare, M. A.; Moody, J. O.; Wong, G. K.-S.; Dessein, S.; Greger, H.; Gruber, C. W., Biopolymers:

Pept. Sci. 2013, 100, 438-452.

(25) Matsuura, H. N.; Porto, D. D.; Fett-Neto, A. G., In Handbook of Natural Products -

Phytochemistry, Botany and Metabolism of Alkaloids, Ramawat, K. G.; Merillon, J.-M., Eds.; Springer:

Heidelberg, 2013, Vol. I, Chapter 5, pp 119-147.

(26) Matsuura, H. N.; Rau, M. R.; Fett-Neto, A. G., Biotechnol. Lett. 2014, 36, 191-200.

(27) Mithöfer, A.; Boland, W., Annu. Rev. Plant Biol. 2012, 63, 431-50.

(28) Zhang, X.; Kuča, K.; Dohnal, V.; Dohnalová, L.; Wu, Q.; Wu, C., J. Appl. Biomed. 2014, 12,

63-77.

(29) Matsuura, H. N.; Fett-Neto, A. G., Nat. Prod. Res. 2013, 27, 402-411.

(30) Porto, D. D.; Matsuura, H. N.; Vargas, L. R.; Henriques, A. T.; Fett-Neto, A. G., Nat. Prod.

Commun. 2014, 9, 629-632.

(31) Kerber, V. A.; Gregianini, T. S.; Paranhos, J. T.; Schwambach, J.; Farias, F.; Fett, J. P.; Fett-

Neto, A. G.; Zuanazzi, J. Â. S.; Quirion, J.-C.; Elizabetsky, E.; Henriques, A. T., J. Nat. Prod. 2001,

64, 677-679.

(32) Henriques, A. T.; Lopes, S. O.; Paranhos, J. T.; Gregianini, T. S.; Von Poser, G. L.; Fett-Neto,

A. G.; Schripsema, J., Phytochemistry 2004, 65, 449-454.

(33) Paranhos, J. T.; Fragoso, V.; Henriques, A. T.; Ferreira, A. G.; Fett-Neto, A. G., Tree Physiol.

2005, 25, 251-255.

21

(34) Gregianini, T. S.; Porto, D. D.; Do Nascimento, N. C.; Fett, J. P.; Henriques, A. T.; Fett-Neto,

A. G., J. Chem. Ecol. 2004, 30, 2023-2036.

(35) Matsuura, H. N.; Fragoso, V.; Paranhos, J. T.; Rau, M. R.; Fett-Neto, A. G., Ind. Crops Prod.

2016, 86, 210-218.

(36) Müller, S. C.; Overbeck, G. E.; Pfadenhauer, J.; Pillar, V. D., Flora 2012, 207, 586-598.

(37) Broussalis, A. M.; Göransson, U.; Coussio, J. D.; Ferraro, G.; Martino, V.; Claeson, P.,

Phytochemistry 2001, 58, 47-51.

(38) Leal, M. B.; Elisabetsky, E., J. Ethnopharmacol. 1996, 54, 37-40.

(39) Gruber, C. W.; Cĕmažar, M.; Anderson, M. A.; Craik, D. J., Toxicon 2007, 49, 561-575.

(40) Colgrave, M. L.; Kotze, A. C.; Huang, Y. H.; O'Grady, J.; Simonsen, S. M.; Craik, D. J.,

Biochemistry 2008, 47, 5581-5589.

(41) Mylne, J. S.; Wang, C. K.; van der Weerden, N. L.; Craik, D. J., Biopolymers: Pept. Sci. 2010,

94, 635-646.

(42) Zhang, J.; Li, J.; Huang, Z.; Yang, B.; Zhang, X.; Li, D.; Craik, D. J.; Baker, A. J. M.; Shu, W.;

Liao, B., J. Plant Physiol. 2015, 178, 17-26.

(43) Mesquita-Neto, J. N.; Silva-Neto, C. M.; Franceschinelli, E. V., Plant Ecol. Evol. 2015, 148,

229-236.

(44) Paranhos, J. T.; Fragoso, V.; da Silveira, V. C.; Henriques, A. T.; Fett-Neto, A. G., Biochem.

Syst. Ecol. 2009, 37, 707-715.

(45) Kerber, V. A.; Passos, C. S.; Klein-Júnior, L. C.; Quirion, J.-C.; Pannecoucke, X.; Salliot-

Maire, I.; Henriques, A. T., Tetrahedron Lett. 2014, 55, 4798-4800.

(46) Tan, M. A.; Eusebio, J. A.; Alejandro, G. J. D., Biochem. Syst. Ecol. 2012, 45, 20-22.

(47) Murashige, T.; Skoog, F., Physiol. Plant. 1962, 15, 473-497.

(48) Mahatmanto, T.; Poth, A. G.; Mylne, J. S.; Craik, D. J., Fitoterapia 2014, 95, 22-33.

(49) Perkins, L. E.; Cribb, B. W.; Hanan, J.; Glaze, E.; Beveridge, C.; Zalucki, M. P., Arthropod

Plant Interact. 2008, 2, 197-207.

22

(50) Teakle, R. E.; Jensen, J. M.; Giles, J. E., J. Invertebr. Pathol. 1985, 46, 166-173.

(51) Poth, A. G.; Colgrave, M. L.; Lyons, R. E.; Daly, N. L.; Craik, D. J., Proc. Natl. Acad. Sci. U.

S. A. 2011, 108, 10127-10132.

(52) Both, F. L.; Kerber, V. A.; Henriques, A. T.; Elisabetsky, E., Pharm. Biol. 2002, 40, 336-341.

(53) Ovesen, R. G.; Göransson, U.; Hansen, S. H.; Nielsen, J.; Hansen, H. C. B., J. Chromatogr. A

2011, 1218, 7964-7970.

(54) Colgrave, M. L.; Poth, A. G.; Kaas, Q.; Craik, D. J., Biopolymers: Pept. Sci. 2010, 94, 592-601.

(55) Clark, R. J.; Jensen, J.; Nevin, S. T.; Callaghan, B. P.; Adams, D. J.; Craik, D. J., Angew. Chem.

Int. Ed. 2010, 49, 6545-6548.

23

Graphical Table of Contents

24