Advertisement

The acyclotide ribe 31 from Rinorea bengalensis has selective cytotoxicity and potent insecticidal properties in Drosophila

  • Tien T. Dang
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
  • Yen-Hua Huang
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
  • Stanislav Ott
    Affiliations
    Program in Neuroscience and Behavioral Disorders, Duke-NUS Medical School, Singapore, Singapore
    Search for articles by this author
  • Peta J. Harvey
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
  • Edward K. Gilding
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
  • Benjamin J. Tombling
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
  • Lai Y. Chan
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
  • Quentin Kaas
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
  • Adam Claridge-Chang
    Affiliations
    Program in Neuroscience and Behavioral Disorders, Duke-NUS Medical School, Singapore, Singapore

    Institute for Molecular and Cell Biology, A∗STAR, Singapore

    Department of Physiology, National University of Singapore, Singapore, Singapore
    Search for articles by this author
  • David J. Craik
    Correspondence
    For correspondence: David J. Craik
    Affiliations
    Institute for Molecular Bioscience, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, The University of Queensland, Brisbane, Queensland, Australia
    Search for articles by this author
Open AccessPublished:August 22, 2022DOI:https://doi.org/10.1016/j.jbc.2022.102413
      Cyclotides and acyclic versions of cyclotides (acyclotides) are peptides involved in plant defense. These peptides contain a cystine knot motif formed by three interlocked disulfide bonds, with the main difference between the two classes being the presence or absence of a cyclic backbone, respectively. The insecticidal activity of cyclotides is well documented, but no study to date explores the insecticidal activity of acyclotides. Here, we present the first in vivo evaluation of the insecticidal activity of acyclotides from Rinorea bengalensis on the vinegar fly Drosophila melanogaster. Of a group of structurally comparable acyclotides, ribe 31 showed the most potent toxicity when fed to D. melanogaster. We screened a range of acyclotides and cyclotides and found their toxicity toward human red blood cells was substantially lower than toward insect cells, highlighting their selectivity and potential for use as bioinsecticides. Our confocal microscopy experiments indicated their cytotoxicity is likely mediated via membrane disruption. Furthermore, our surface plasmon resonance studies suggested ribe 31 preferentially binds to membranes containing phospholipids with phosphatidyl-ethanolamine headgroups. Despite having an acyclic backbone, we determined the three-dimensional NMR solution structure of ribe 31 is similar to that of cyclotides. In summary, our results suggest that, with further optimization, ribe 31 could have applications as an insecticide due to its potent in vivo activity against D. melanogaster. More broadly, this work advances the field by demonstrating that acyclotides are more common than previously thought, have potent insecticidal activity, and have the advantage of potentially being more easily manufactured than cyclotides.

      Keywords

      Abbreviations:

      MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide), PE (phosphatidyl-ethanolamine), POPC (1-palmitoyl-2-oleoyl-glycero-3-phosphocholine), POPE (1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine), RBC (red blood cell), RP-HPLC (reverse-phase HPLC), SPR (surface plasmon resonance)
      Cyclotides have attracted interest as next-generation therapeutics and agricultural agents due to their unique topology and diverse biological activities (
      • Craik D.J.
      • Daly N.L.
      • Bond T.
      • Waine C.
      Plant cyclotides: a unique family of cyclic and knotted proteins that defines the cyclic cystine knot structural motif.
      ,
      • de Veer S.J.
      • Kan M.W.
      • Craik D.J.
      Cyclotides: from structure to function.
      ,
      • Wang C.K.
      • Craik D.J.
      Designing macrocyclic disulfide-rich peptides for biotechnological applications.
      ). These plant-derived peptides comprise a head-to-tail cyclic backbone of 28 to 37 amino acids and three intramolecular disulfide bonds. Together with the cyclic backbone, the three disulfide bonds form a structure known as the cyclic cystine knot (CCK) motif, which bestows cyclotides with exceptional resistance to proteolytic, thermal, or chemical degradation (
      • Craik D.J.
      • Daly N.L.
      • Bond T.
      • Waine C.
      Plant cyclotides: a unique family of cyclic and knotted proteins that defines the cyclic cystine knot structural motif.
      ,
      • Colgrave M.L.
      • Craik D.J.
      Thermal, chemical, and enzymatic stability of the cyclotide kalata B1: the importance of the cyclic cystine knot.
      ).
      Cyclotides are divided into three main subfamilies: Möbius, bracelet, and trypsin inhibitor. The Möbius and bracelet subfamilies are distinguished by the presence or absence of a cis Xaa-Pro bond in loop 5, respectively (
      • Craik D.J.
      • Daly N.L.
      • Bond T.
      • Waine C.
      Plant cyclotides: a unique family of cyclic and knotted proteins that defines the cyclic cystine knot structural motif.
      ). Cyclotides from the trypsin inhibitor family retain the CCK motif but have low sequence similarity and different biological activities to either Möbius or bracelet cyclotides (
      • Hernandez J.F.
      • Gagnon J.
      • Chiche L.
      • Nguyen T.M.
      • Andrieu J.P.
      • Heitz A.
      • et al.
      Squash trypsin inhibitors from Momordica cochinchinensis exhibit an atypical macrocyclic structure.
      ).
      Host defense has been widely considered as the main function of cyclotides in plants (
      • Craik D.J.
      Host-defense activities of cyclotides.
      ,
      • Simonsen S.M.
      • Sando L.
      • Ireland D.C.
      • Colgrave M.L.
      • Bharathi R.
      • Göransson U.
      • et al.
      A continent of plant defense peptide diversity: cyclotides in Australian Hybanthus (Violaceae).
      ). This function has stimulated interest in the use of cyclotides for agricultural applications, which is supported by their wide range of biological activities, including antimicrobial (
      • Nguyen K.N.T.
      • Nguyen G.K.T.
      • Nguyen P.Q.T.
      • Ang K.H.
      • Dedon P.C.
      • Tam J.P.
      Immunostimulating and Gram-negative-specific antibacterial cyclotides from the butterfly pea (Clitoria ternatea).
      ,
      • Porto W.F.
      • Pires A.S.
      • Franco O.L.
      CS-AMPPred: an updated SVM model for antimicrobial activity prediction in cysteine-stabilized peptides.
      ,
      • Pränting M.
      • Lööv C.
      • Burman R.
      • Göransson U.
      • Andersson D.I.
      The cyclotide cycloviolacin O2 from Viola odorata has potent bactericidal activity against Gram-negative bacteria.
      ), antifouling (
      • Göransson U.
      • Sjögren M.
      • Svangård E.
      • Claeson P.
      • Bohlin L.
      Reversible antifouling effect of the cyclotide cycloviolacin O2 against barnacles.
      ), cytotoxic (
      • Dang T.T.
      • Chan L.Y.
      • Huang Y.H.
      • Nguyen L.T.T.
      • Kaas Q.
      • Huynh T.
      • et al.
      Exploring the sequence diversity of cyclotides from Vietnamese viola species.
      ), insecticidal (
      • Pinto M.F.S.
      • Fensterseifer I.C.M.
      • Migliolo L.
      • Sousa D.A.
      • de Capdville G.
      • Arboleda-Valencia J.W.
      • et al.
      Identification and structural characterization of novel cyclotide with activity against an insect pest of sugar cane.
      ), and molluscicidal activities (
      • Plan M.R.R.
      • Saska I.
      • Cagauan A.G.
      • Craik D.J.
      Backbone cyclised peptides from plants show molluscicidal activity against the rice pest Pomacea canaliculata (golden apple snail).
      ). For example, the prototypic Möbius cyclotide kalata B1, discovered from Oldenlandia affinis, causes significantly reduced growth of Helicoverpa punctigera—a major pest of corn and cotton (Fig. 1A) (
      • Jennings C.
      • West J.
      • Waine C.
      • Craik D.
      • Anderson M.
      Biosynthesis and insecticidal properties of plant cyclotides: the cyclic knotted proteins from Oldenlandia affinis.
      ). Similarly, a recent study showcased the activity of the prototypic bracelet cyclotide cycloviolacin O2 (cyO2) against Myzus persicae probing and feeding behavior (
      • Dancewicz K.
      • Slazak B.
      • Kielkiewicz M.
      • Kapusta M.
      • Bohdanowicz J.
      • Gabrys B.
      Behavioral and physiological effects of Viola spp. cyclotides on Myzus persicae (Sulz.).
      ). The bioinsecticide Sero-X (https://innovate-ag.com.au/sero-x/) is produced from an extract of the cyclotide-bearing plant Clitoria ternatea (Fabaceae) and was recently approved for agricultural use against various insect pests including Helicoverpa species, green mirids, green vegetable bugs, and whiteflies. The cyclotide Cter M, isolated from C. ternatea (Fig. 1B) (
      • Poth A.G.
      • Colgrave M.L.
      • Lyons R.E.
      • Daly N.L.
      • Craik D.J.
      Discovery of an unusual biosynthetic origin for circular proteins in legumes.
      ), displays insecticidal activity against the cotton bollworm Helicoverpa armigera, suggesting it may be responsible for the activity of Sero-X. Peptide-based bioinsecticides offer potential advantages over traditional pesticides of lower off-target toxicity, reduced environmental impacts, and more sustainable manufacture.
      Figure thumbnail gr1
      Figure 1Insecticidal activity of kalata B1, Cter M, and ribe 31 against different insects. A, sequence and insecticidal activity of kalata B1 against Helicoverpa punctigera, major pests of corn and cotton. B, sequence and insecticidal activity of Cter M against Helicoverpa armigera larvae (H. armigera), a cotton pest. C, sequence and insecticidal activity of ribe 31 against Drosophila melanogaster (D. melanogaster). Cysteine residues are marked red in yellow circles. Yellow lines connecting two cysteine residues represent disulfide bonds.
      Recently, noncyclic analogs of cyclotides, called acyclotides, have attracted increasing attention (
      • Tammineni R.
      • Gulati P.
      • Kumar S.
      • Mohanty A.
      An overview of acyclotides: past, present and future.
      ,
      • Nguyen G.K.
      • Lian Y.L.
      • Pang E.W.H.
      • Phuong Q.T.N.
      • Tran T.D.
      • Tam J.P.
      Discovery of linear cyclotides in monocot plant Panicum laxum of Poaceae family provides new insights into evolution and distribution of cyclotides in plants.
      ). Acyclotides have a similar sequence to cyclotides and also contain a cystine knot but lack a head-to-tail cyclic backbone. The backbone cyclization is absent due to a stop codon occupying the position that would encode an Asx (Asn or Asp) residue in a typical cyclotide precursor (
      • Dang T.T.
      • Chan L.Y.
      • Tombling B.J.
      • Harvey P.J.
      • Gilding E.K.
      • Craik D.J.
      Planta discovery and chemical synthesis of bracelet cystine knot peptides from Rinorea bengalensis.
      ). This key Asx residue and an adjacent C-terminal tail peptide in the precursor facilitate cyclization in cyclotides by asparaginyl endopeptidase enzymes in plants (
      • Nguyen G.K.
      • Lian Y.L.
      • Pang E.W.H.
      • Phuong Q.T.N.
      • Tran T.D.
      • Tam J.P.
      Discovery of linear cyclotides in monocot plant Panicum laxum of Poaceae family provides new insights into evolution and distribution of cyclotides in plants.
      ,
      • Jackson M.A.
      • Gilding E.K.
      • Shafee T.
      • Harris K.S.
      • Kaas Q.
      • Poon S.
      • et al.
      Molecular basis for the production of cyclic peptides by plant asparaginyl endopeptidases.
      ). Despite their acyclic backbone, acyclotides have comparable stability and structures to cyclotides, primarily due to the stabilizing effect of the cystine knot motif (
      • Dang T.T.
      • Chan L.Y.
      • Tombling B.J.
      • Harvey P.J.
      • Gilding E.K.
      • Craik D.J.
      Planta discovery and chemical synthesis of bracelet cystine knot peptides from Rinorea bengalensis.
      ,
      • Ireland D.C.
      • Colgrave M.L.
      • Nguyencong P.
      • Daly N.L.
      • Craik D.J.
      Discovery and characterization of a linear cyclotide from Viola odorata: implications for the processing of circular proteins.
      ). Apart from local changes at the terminal residues, the NMR structures of two acyclotides, violacin A, and MCoTI-V have shown that linearization has negligible effects upon the structure of the cystine knot core (
      • Ireland D.C.
      • Colgrave M.L.
      • Nguyencong P.
      • Daly N.L.
      • Craik D.J.
      Discovery and characterization of a linear cyclotide from Viola odorata: implications for the processing of circular proteins.
      ,
      • Mylne J.S.
      • Chan L.Y.
      • Chanson A.H.
      • Daly N.L.
      • Schaefer H.
      • Bailey T.L.
      • et al.
      Cyclic peptides arising by evolutionary parallelism via asparaginyl-endopeptidase–mediated biosynthesis.
      ). Examination of species from the five dicot families known to produce cyclotides (namely the Violaceae, Rubiaceae, Solanaceae, Cucurbitaceae, and Fabaceae) suggests that fewer acyclotides than cyclotides are produced by these plants (
      • Wang C.K.
      • Kaas Q.
      • Chiche L.
      • Craik D.J.
      CyBase: a database of cyclic protein sequences and structures, with applications in protein discovery and engineering.
      ,
      • Mulvenna J.P.
      • Wang C.
      • Craik D.J.
      CyBase: a database of cyclic protein sequence and structure.
      ). Some monocots from the Poaceae family express acyclotides but no cyclotide has so far been discovered in this plant family (
      • Nguyen G.K.
      • Lian Y.L.
      • Pang E.W.H.
      • Phuong Q.T.N.
      • Tran T.D.
      • Tam J.P.
      Discovery of linear cyclotides in monocot plant Panicum laxum of Poaceae family provides new insights into evolution and distribution of cyclotides in plants.
      ), despite these plants containing asparaginyl endopeptidase enzymes. Although acyclotides host features common to the bracelet or Möbius cyclotide subfamilies, they are topologically linear. Examples include violacin A (Möbius subfamily) (
      • Ireland D.C.
      • Colgrave M.L.
      • Nguyencong P.
      • Daly N.L.
      • Craik D.J.
      Discovery and characterization of a linear cyclotide from Viola odorata: implications for the processing of circular proteins.
      ), kalata B9 (bracelet subfamily) (
      • Plan M.R.R.
      • Göransson U.
      • Clark R.J.
      • Daly N.L.
      • Colgrave M.L.
      • Craik D.J.
      The cyclotide fingerprint in Oldenlandia affinis: elucidation of chemically modified, linear and novel macrocyclic peptides.
      ), and modi 6 (trypsin inhibitor subfamily) (
      • Du J.
      • Chan L.Y.
      • Poth A.G.
      • Craik D.J.
      Discovery and characterization of cyclic and acyclic trypsin inhibitors from Momordica dioica.
      ). Overall, there are currently around 105 acyclotides reported with about 85% of them being homologous to bracelet subfamily cyclotides. Due to their apparent scarcity in nature compared to cyclotides, the biological activities of acyclotides have been less extensively studied. Specifically, there have been no reported studies assessing the insecticidal activities of acyclotides, highlighting this as an unexplored research field.
      Pesticides are a keystone in the agricultural sector, and the need for developing alternative insecticidal compounds is becoming increasingly apparent due to the negative environmental impacts of many existing pesticides (
      • Hossard L.
      • Philibert A.
      • Bertrand M.
      • Colnenne-David C.
      • Debaeke P.
      • Munier-Jolain N.
      • et al.
      Effects of halving pesticide use on wheat production.
      ). Some widely used conventional insecticides are composed of chemicals that cause environmental damage to surroundings and can pose health risks to the wider population (
      • Braga A.R.C.
      • de Rosso V.V.
      • Harayashiki C.A.Y.
      • Jimenez P.C.
      • Castro Í.B.
      Global health risks from pesticide use in Brazil.
      ). For instance, there have been reports of residual insecticides leading to increased cancer risk (
      • Gill J.P.S.
      • Bedi J.S.
      • Singh R.
      • Fairoze M.N.
      • Hazarika R.A.
      • Gaurav A.
      • et al.
      Pesticide residues in peri-urban bovine milk from India and risk assessment: a multicenter study.
      ) and poisoning (
      • Wang N.
      • Jiang Q.
      • Han L.
      • Zhang H.
      • Zhu B.
      • Liu X.
      Epidemiological characteristics of pesticide poisoning in Jiangsu Province, China, from 2007 to 2016.
      ). Thus, environmentally friendly insecticides, such as Sero-X, are attractive alternatives to harsh chemicals, potentially helping to reduce contamination and damage to ecosystems.
      Drosophila suzukii is a major fruit pest in the North America and Asia (
      • Goodhue R.E.
      • Bolda M.
      • Farnsworth D.
      • Williams J.C.
      • Zalom F.G.
      Spotted wing drosophila infestation of California strawberries and raspberries: Economic analysis of potential revenue losses and control costs.
      ) and is closely related to the vinegar fly Drosophila melanogaster, which is widely used in the laboratory as a model organism for biological studies, that is, toxicity, persistent behavior, aging, cellular morphogenesis, and human disease modeling (
      • Riaz B.
      • Zahoor M.K.
      • Zahoor M.A.
      • Majeed H.N.
      • Javed I.
      • Ahmad A.
      • et al.
      Toxicity, phytochemical composition, and enzyme inhibitory activities of some indigenous weed plant extracts in fruit fly, Drosophila melanogaster.
      ,
      • Tolwinski N.S.
      Introduction: Drosophila-A model system for developmental Biology.
      ), as well as for testing behavior and phenotype effects of insecticides (
      • Scott J.G.
      • Buchon N.
      Drosophila melanogaster as a powerful tool for studying insect toxicology.
      ,
      • Akmoutsou P.
      • Mademtzoglou D.
      • Nakou I.
      • Onoufriadis A.
      • Papadopoulou X.
      • Kounatidis I.
      • et al.
      Evaluation of toxicity and genotoxic effects of spinosad and deltamethrin in Drosophila melanogaster and Bactrocera oleae.
      ,
      • Muñoz E.
      • Lamilla C.
      • Marin J.C.
      • Alarcon J.
      • Cespedes C.L.
      Antifeedant, insect growth regulatory and insecticidal effects of Calceolaria talcana (Calceolariaceae) on Drosophila melanogaster and Spodoptera frugiperda.
      ,
      • Chowanski S.
      • Chudzinska E.
      • Lelario F.
      • Ventrella E.
      • Marciniak P.
      • Miadowicz-Kobielska M.
      • et al.
      Insecticidal properties of Solanum nigrum and Armoracia rusticana extracts on reproduction and development of Drosophila melanogaster.
      ). D. melanogaster has also been used to study insecticide resistance (
      • Ffrench-Constant R.H.
      The molecular genetics of insecticide resistance.
      ,
      • Strycharz J.P.
      • Lao A.
      • Li H.
      • Qiu X.
      • Lee S.H.
      • Sun W.
      • et al.
      Resistance in the highly DDT-resistant 91-R strain of Drosophila melanogaster involves decreased penetration, increased metabolism, and direct excretion.
      ).
      In the current study, a family of acyclotides previously discovered from the stem and leaf of Rinorea bengalensis, named ribe peptides (
      • Dang T.T.
      • Chan L.Y.
      • Tombling B.J.
      • Harvey P.J.
      • Gilding E.K.
      • Craik D.J.
      Planta discovery and chemical synthesis of bracelet cystine knot peptides from Rinorea bengalensis.
      ), were isolated and evaluated for their potential insecticidal activity. The toxicity of these peptides was examined against two insect cell lines in vitro and tested for in vivo insecticidal activity against D. melanogaster (Fig. 1C). Ribe 31 was found to be more toxic than the well-characterized insecticidal cyclotide kalata B1. Structural analysis revealed that all ribe peptides display a similar cystine knot motif and tertiary structure to cyclotides. Ribe 31 was shown to bind preferentially to a model membrane containing phospholipids with phosphatidyl-ethanolamine (PE) headgroups, suggesting that its ability to disrupt membrane is related to its insecticidal activity. Our results suggest that ribe 31 is a useful lead for the development of ecofriendly biocides for crop protection.

      Results

      Isolation of cystine knot peptides from R. bengalensis

      Ribe peptides isolated from stem and leaf tissue extracts of R. bengalensis were purified and separated based on their hydrophobicity as evaluated by their elution times in reverse-phase HPLC (RP-HPLC) (Fig. 2A). Generally, peaks with a later elution time are more hydrophobic. From early to late elution, the order of hydrophobicity was acyclotides ribe 10, 20, 24, and 31, and cyclotide ribe 33. Amino acid sequences of the isolated peptides are shown in Fig. 2B.
      Figure thumbnail gr2
      Figure 2Cystine knot peptides derived from R. bengalensis. A, HPLC traces of stem and leaf extracts of R. bengalensis and the relative elution time of cystine knot peptides discovered from it are shown in top panel. B, the amino acid sequences of acyclotides ribe 10, 20, 24, and 31 and cyclotide ribe 33 are listed in the bottom panel. Yellow lines connecting two cysteine residues represent disulfide bonds.

      In vivo toxicity of ribe peptides to D. melanogaster after oral feeding

      In vitro assays indicated that several purified peptides (ribe 10, 20, 24, 31, 33) and the stem extract from R. bengalensis had cytotoxic activity against insect cells (Table 1). The in vivo insecticidal activity of these peptides for D. melanogaster were evaluated and compared to known insecticidal cyclotides cyO2 and kalata B1. Initial screening involved feeding flies with stem extracts (50 μg), ribe peptides (16 μM), or 10% sucrose (vehicle control) using the capillary feeder (CAFE) assay (Fig. S1, Supporting Information). In this screen, only ribe 31 caused a reduction in lifespan.
      Table 1Cytotoxicity and hemolytic activity of isolated peptides against insect cell lines (Sf9 and High Five) and red blood cells
      PeptideCC50 (μM) ± SEM
      Cytotoxic concentration required to kill 50% of the cells (CC50) with data representing the SEM from triplicates.
      Sf9
      Sf9: a clonal isolate derived from Spodoptera frugiperda Sf21 cells.
      High five
      High Five: an insect cell line originating from the ovarian cells of the cabbage looper, Trichoplusia ni.
      RBC
      N.D.: not determined due to insufficient samples.
      ribe 100.8 ± 0.01.3 ± 0.114.0 ± 0.2
      ribe 203.5 ± 0.18.3 ± 0.1>25
      ribe 240.3 ± 0.00.9 ± 0.111.3 ± 0.1
      ribe 310.7 ± 0.02.6 ± 0.121.4 ± 0.2
      ribe 331.6 ± 0.13.2 ± 0.121.9 ± 0.2
      stem extract0.3 ± 0.00.7 ± 0.1N.D.
      kalata B13.2 ± 0.1N.D.8.4 ± 0.1
      cyO20.2 ± 0.01.0 ± 0.022.0 ± 0.2
      a Cytotoxic concentration required to kill 50% of the cells (CC50) with data representing the SEM from triplicates.
      b Sf9: a clonal isolate derived from Spodoptera frugiperda Sf21 cells.
      c High Five: an insect cell line originating from the ovarian cells of the cabbage looper, Trichoplusia ni.
      d N.D.: not determined due to insufficient samples.
      Subsequently, multiple concentrations of ribe 31 were tested to determine a dose–response relationship. As shown in Fig. 3A, the lifespan of the flies was shorter with increasing peptide concentrations. When ribe 31 was administered at 1.6, 16, and 160 μM, the average fly survival decreased by –200, –252, and –352 h, respectively. This average survival rate was drastically lower than that for control flies (average survival of 458 h). Thus, these results confirm that the administration of ribe 31 shortened the lifespan of D. melanogaster in a dose–response manner, with concentrations as low as 1.6 μM, provoking substantial toxicity. Cyclotides cyO2 and kalata B1 have previously been shown to be effective on M. persicae (
      • Dancewicz K.
      • Slazak B.
      • Kielkiewicz M.
      • Kapusta M.
      • Bohdanowicz J.
      • Gabrys B.
      Behavioral and physiological effects of Viola spp. cyclotides on Myzus persicae (Sulz.).
      ) and to impair the growth and development of H. punctigera, respectively (
      • Jennings C.
      • West J.
      • Waine C.
      • Craik D.
      • Anderson M.
      Biosynthesis and insecticidal properties of plant cyclotides: the cyclic knotted proteins from Oldenlandia affinis.
      ). To extend our understanding of toxicity to the Diptera order of insects, cyO2 and kalata B1 were tested on vinegar flies here, and it was found that lifespan declined with increasing concentrations of cyO2 and kalata B1. However, the toxic effect of cyO2 was not as strong as that of ribe 31, as cyO2 administered at 1.6, 16, and 160 μM reduced average survival by –137, –174 and –310 h, respectively (Fig. 3B). In addition, the effect of kalata B1 on survival was weaker; the average survival reduction was –39, –99, and –140 h when administered at 1.7, 17, and 170 μM, respectively (Fig. 3C). More details on the side-by-side comparison of ribe 31 versus cyO2 or kalata B1 on fly survival are shown in Fig. 2.
      Figure thumbnail gr3
      Figure 3The results of a CAFE assay of ribe 31 (A), cyO2 (B), and kalata B1 (C) on D. melanogaster at a range of concentrations (μM). The lifespan of flies fed a sucrose diet (20% sucrose) or a sucrose plus ribe 31 diet at varying concentrations is shown at the top. Differences in average lifespan of each experimental group compared to the sucrose-only control group (‘sucrose’) are shown at the bottom. Each dot in the scatter plot represents the survival of one fly. Error bars show the 95% CI. A, effect size statistics: ribe 31, 1.6 μM (n = 98) versus sucrose (n = 104) = −200 [95% CI −220, −180] p < 1∗10 to 4; ribe 31, 16 μM (n = 207) versus sucrose (n = 104) = −252 [95% CI −269, −236] p < 1∗10 to 4; ribe 31, 160 μM (n = 209) versus sucrose (n = 104) = −352 [95% CI −365, −337] p < 1∗10 to 4. B, effect size statistics: cyO2, 1.6 μM (n = 105) versus sucrose (n = 104) = −137 [95% CI −171, −103] p < 1∗10 to 4; cyO2, 16 μM (n = 105) versus sucrose (n = 104) = −174 [95% CI −202, −146] p < 1∗10 to 4; cyO2, 160 μM (n = 105) versus sucrose (n = 104) = −310 [95% CI −327, −291] p < 1∗10 to 4. C, effect size statistics: kalata B1, 1.7 μM (n = 99) versus sucrose (n = 102) = −39 [95% CI −76, 0] p= 0.0327; kalata B1, 17 μM (n = 105) versus sucrose (n = 102) = −99 [95% CI −136, −62] p < 1∗10 to 4; kalata B1: 170 μM (n = 207) versus sucrose (n = 102) = −140 [95% CI −170, −106] p < 1∗10 to 4. CI, confidence interval.

      Structural analysis of ribe 10, 20, 24, and 31

      The secondary structure elements of the ribe peptides were evaluated by measuring their secondary Hα NMR chemical shifts (i.e., the difference of measured shifts from random coil values). These shifts were compared to those of cyO2, which also belongs to the bracelet cyclotide subfamily (Fig. 4A) (
      • Göransson U.
      • Herrmann A.
      • Burman R.
      • Haugaard-Jönsson L.M.
      • Rosengren K.J.
      The conserved Glu in the cyclotide cycloviolacin O2 has a key structural role.
      ). Overall, the ribe peptides showed similar secondary shifts to cyO2, indicating that all the peptides have similar secondary structure elements. Slight differences were observed between the ribe peptides in the loops that showed sequence variability. In particular, the different loop lengths and sequences of ribe 20 result in chemical shift differences compared to the other ribe acyclotides, in loops 3, 5, and 6 (Fig. S3, Supporting Information). The amino acid sequence of loop 3 was variable among the studied peptides, and the corresponding Hα shifts were also different, suggesting different conformations.
      Figure thumbnail gr4
      Figure 4Secondary Hα chemical shift comparison between cyO2 and ribe 31 and 33 and the three-dimensional NMR structure of ribe 31. A, the comparison of secondary Hα chemical shifts of ribe 31 (orange), ribe 33 (blue) and cyO2 (lime green). The chemical shifts of cyO2 were obtained from the Biological Magnetic Resonance Data Bank (BMRB, ID: 16073 (
      • Wang C.K.
      • Colgrave M.L.
      • Ireland D.C.
      • Kaas Q.
      • Craik D.J.
      Despite a conserved cystine knot motif, different cyclotides have different membrane binding modes.
      )). All cysteines are highlighted in red text with yellow boxes. The cyclic backbones of ribe 33 and cyO2 are indicated with a thick black line, and disulfide bond connectivities (CysI-IV, CysII-V and CysIII-VI) are shown as thin black lines. B, superposition of 20 conformers representing the 3D NMR structure of acyclotide ribe 31. C, the superimposition of 3D NMR structure of cyclotide cyO2 (lime green) and acyclotide ribe 31 (orange).
      To determine the effect of an acyclic backbone on structure, the three-dimensional structure of ribe 31 was determined based on NMR spectroscopy measurements. A total of 326 distance restraints and 32 dihedral angle restraints was used to calculate preliminary structures with the following imposed disulfide connectivities: Cys4-Cys20, Cys8-Cys22, and Cys13-Cys27. Both proline residues in loops 2 and 6 were determined to adopt a trans conformation based on strong Hδ(i)Pro-Hα(i-1) NOEs. Amide temperature coefficients identified six residue pairs involved in hydrogen bond interactions (Cys4 HN–Cys27 O, Ser21 HN–Tyr28 O, Arg23 HN–Val 26 O, Cys27 HN–Glu6 O, Tyr28 HN–Ser21 O, and Arg29 HN–Ile2 O). The final ensemble of structures (Fig. 4B) is well defined with an average pairwise RMSD across backbone atoms of 0.71 Å. The models were of very good quality as evidenced by an average MolProbity score of 1.8 (Table S2). Despite the linear nature of ribe 31, its backbone structure overlays well with that of cyO2 (RMSD 0.90 Å) with conservation of the three stranded antiparallel β-sheet, hydrogen bonding network across the core, and a short α-helix in loop 3 (Fig. 4C).

      Homology modeling of ribe peptides

      Homology models of all tested peptides are illustrated in Fig. 5A as ribbon (left) and surface (right) representations. Analyzing the surface properties of the ribe peptides from the models visualizes the distribution of the hydrophobic patches that derive from residues in loops 2 and 3. Compared to ribe 10 and 20, ribe 24, 31, and 33 have an additional hydrophobic residue in their loop 3. The models suggest that loop 3 drives the difference in hydrophobicity between the peptides. Ribe 24, 31, and 33 indeed have longer HPLC retention times than the other two ribe peptides (10 and 20, see Fig. 2A). The more hydrophobic ribe peptides 24, 31, and 33 were expected to bind with higher affinity to lipid membranes than ribe 10 and ribe 20, as later tested using surface plasmon resonance (SPR).
      Figure thumbnail gr5
      Figure 5Homology model and binding affinity of ribe peptides toward the model membrane. A, homology models of kalata B1 (pink), ribe 10 (purple), ribe 20 (green), ribe 24 (brown), ribe 31 (orange), and ribe 33 (blue) generated using the SWISS-MODEL web server (https://swissmodel.expasy.org) using the structure of cyO2 (PDB ID: 2KCG) as a template. All structures are represented in surface diagrams using PyMol, with the hydrophobic residues highlighted in corresponding colors. B, binding affinity of ribe peptides toward the model membrane composed of POPC/POPE (80:20) evaluated using surface plasmon resonance. C, sensorgrams of ribe 31 binding to the model membrane at a range of concentrations from 4 μM to 32 μM. PDB, Protein Data Bank; POPC, 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine; POPE, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine.

      Cytotoxicity of ribe peptides against insect cells

      To assess the potency of the ribe peptides as insecticidal agents, their activity against two pest insect cell lines was tested using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay. In experiments comparing selected peptides to an extract, the stem extract from R. bengalensis (5 mg/ml) containing a plethora of acyclotides was shown to be toxic to both insect cell lines (Table 1), suggesting that the plant-expressed peptides are potentially cytotoxic. Isolated ribe peptides had different levels of cytotoxicity toward the two insect cell lines, as shown in Table 1. Ribe 10, ribe 24, and ribe 31 had higher toxicity against Sf9 cells, whereas ribe 20 was the least toxic to both insect cell lines. The grand average of hydropathy (
      • Jaspard E.
      • Macherel D.
      • Hunault G.
      Computational and statistical analyses of amino acid usage and physico-chemical properties of the twelve late embryogenesis abundant protein classes.
      ) score, which is the average Kyte–Doolittle hydropathy score for each residue of an amino acid sequence, was computed for each peptide. According to this score, the two most cytotoxic acyclotides, ribe 31 and 24, are also the most hydrophobic (Table S1, Supporting Information). In general, all the ribe peptides showed greater toxicity to High Five cells than to Sf9 cells. The toxicity of the ribe peptides is similar to that of insecticidal cyclotides cyO2 and kalata B1 (
      • Colgrave M.L.
      • Kotze A.C.
      • Huang Y.H.
      • O'Grady J.
      • Simonsen S.M.
      • Craik D.J.
      Cyclotides: natural, circular plant peptides that possess significant activity against gastrointestinal nematode parasites of sheep.
      ,
      • Colgrave M.L.
      • Kotze A.C.
      • Ireland D.C.
      • Wang C.K.
      • Craik D.J.
      The anthelmintic activity of the cyclotides: natural variants with enhanced activity.
      ). To determine whether the peptides demonstrated toxicity against red blood cells (RBCs), the hemolytic activities of the five ribe peptides were tested (Table 1). The isolated peptides showed 3 to 12-fold lower toxicity toward human RBCs compared to insect cells.

      Confocal analysis

      Sf9 insect cells were treated with ribe 31 (1 μM), and cell death was monitored with time-lapse microscopy in the presence of SYTOX Green over a 4 h period. SYTOX Green is a fluorescent nucleic acid stain that is impermeable to live cells but penetrates cells with compromised plasma membranes. From the time-lapse movie, Sf9 cells treated with ribe 31 became fluorescent over the course of the experiment, indicating that cell death is associated with membrane disintegration. The observed toxicity of ribe 31 agreed with the MTT cytotoxicity assay (see Table 1). Cell membrane binding and disintegration induced by these peptides were further observed using confocal microscopy as shown in Fig. 6. The microscopy images demonstrate that Alexa Fluor 488–labeled ribe 31 (Fig. 6B) mainly binds to the outer membranes of Sf9 cells in a similar way to fluorescently labeled cyO2 (Fig. 6A) at 0.5 μM. Compared to cyO2 and ribe 31, Alexa Fluor 488–labeled kalata B1 binds to the cell membrane to a lesser extent at the same concentration but is internalized into cells more efficiently with no sign of membrane disruption (Fig. 6C). Ribe 31 was able to puncture cell membranes and become internalized. Cell membrane disintegration was observed for insect cells incubated with Alexa Fluor 488–labeled ribe 31 and cyO2 at 0.5 μM as indicated by white arrows.
      Figure thumbnail gr6
      Figure 6The interactions of Alexa Fluor 488–labeled cystine knot peptides (green) with live Sf9 cells were imaged using confocal microscopy. Sf9 insect cells were treated with (A) Alexa Fluor 488–labeled cyO2 (A488-cyO2) at 0.5 μM, (B) Alexa Fluor 488–labeled ribe 31 (A488-ribe 31) at 0.5 μM, and (C) Alexa Fluor 488–labeled kalata B1 (A488-kB1) at 0.5 μM. Images were captured after 30 min of treatment. The cell membrane was labeled with wheat germ agglutinin conjugated with Alexa Fluor 647 (blue).

      Membrane binding properties of ribe peptides

      The main phospholipids of larval and prepupal stages of D. melanogaster are PE and phosphatidylcholine (
      • Jones H.E.
      • Harwood J.L.
      • Bowen I.D.
      • Griffiths G.
      Lipid composition of subcellular membranes from larvae and prepupae of Drosophila melanogaster.
      ). The dominant membrane phospholipid at both stages of development is PE, constituting 50% to 60% of the total phospholipids. To provide insights into the insecticidal mode of action of ribe peptides against D. melanogaster, two model membranes, (i) 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC) and (ii) a mixture of POPC and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE) phospholipids (80:20), were used to study their membrane-binding properties using SPR. All the isolated ribe peptides exhibited strong binding to the lipid bilayer containing POPC/POPE, which agrees with previous studies reporting the preferential binding of cyclotides toward PE phospholipids (see Table 2 and Fig. 5B) (
      • Henriques S.T.
      • Craik D.J.
      Importance of the cell membrane on the mechanism of action of cyclotides.
      ,
      • Henriques S.T.
      • Huang Y.H.
      • Castanho M.A.R.B.
      • Bagatolli L.A.
      • Sonza S.
      • Tachedjian G.
      • et al.
      Phosphatidylethanolamine binding is a conserved feature of cyclotide-membrane interactions.
      ,
      • Henriques S.T.
      • Huang Y.-H.
      • Chaousis S.
      • Wang C.K.
      • Craik D.J.
      Anticancer and toxic properties of cyclotides are dependent on phosphatidylethanolamine phospholipid targeting.
      ). As expected, the ribe peptides did not bind to the POPC model membrane (Fig. S4), highlighting the requirement of PE phospholipids for binding. Based on peptide-to-lipid maximum binding (P/L max, mol/mol), ribe 31 displayed better binding to POPC/POPE lipids (Table 2, Fig. 5B). In contrast, ribe 20 displayed the lowest P/L value (0.30 mol/mol), in agreement with its lower in vitro cytotoxicity (Table 1). In summary, the membrane binding molar ratios of the ribe peptides correlate with their cytotoxicity. The membrane-binding potential of ribe 31 provides a mechanistic explanation for its potent in vivo activity against D. melanogaster.
      Table 2Membrane binding of ribe peptides toward two lipid systems determined by surface plasmon resonance
      PeptidePOPC/POPE (80:20)
      P/L max (mol/mol)
      maximum peptide-to-lipid molar binding ratio.
      KD (μM)
      Peptide concentration to achieve half-maximum binding at equilibrium. The SD of the mean was determined by fitting a nonlinear equation (Hill slope).
      ribe 100.33 ± 0.018.9 ± 0.33
      ribe 200.30 ± 0.0017.3 ± 0.49
      ribe 240.37 ± 0.006.2 ± 0.18
      ribe 310.38 ± 0.036.4 ± 0.59
      ribe 330.33 ± 0.003.2 ± 0.04
      cyO20.33 ± 0.003.0 ± 0.07
      kalata B10.27 ± 0.0010.3 ± 0.12
      The P/L max and KD values for POPC could not be defined (data not shown) as the response curves for binding of isolated peptides to this lipid membrane did not reach a plateau in the tested concentration range (4 μM to 32 μM).
      a maximum peptide-to-lipid molar binding ratio.
      b Peptide concentration to achieve half-maximum binding at equilibrium. The SD of the mean was determined by fitting a nonlinear equation (Hill slope).

      Discussion

      The insecticidal activities and modes of action of cyclotides have been discussed extensively (
      • Jennings C.
      • West J.
      • Waine C.
      • Craik D.
      • Anderson M.
      Biosynthesis and insecticidal properties of plant cyclotides: the cyclic knotted proteins from Oldenlandia affinis.
      ,
      • Broussalis A.M.
      • Clemente S.
      • Ferraro G.E.
      Hybanthus parviflorus (Violaceae): Insecticidal activity of a south American plant.
      ,
      • Gruber C.W.
      • Čemažar M.
      • Anderson M.A.
      • Craik D.J.
      Insecticidal plant cyclotides and related cystine knot toxins.
      ,
      • Jennings C.V.
      • Rosengren K.J.
      • Daly N.L.
      • Plan M.
      • Stevens J.
      • Scanlon M.J.
      • et al.
      Isolation, solution structure, and insecticidal activity of kalata B2, a circular protein with a twist: do möbius strips exist in nature?.
      ). Previous studies have shown that the hydrophobicity of cyclotides contributes to their key biological activities, with their surface-exposed hydrophobic patch inserting into, and disrupting, cell membranes (
      • Henriques S.T.
      • Craik D.J.
      Importance of the cell membrane on the mechanism of action of cyclotides.
      ). For example, the insecticidal activity of cyclotide parigidin-br1 is broadly dependent on its membrane-binding ability for its biological effects. The hydrophobic surface patch interacts with the cell membrane and causes extensive swelling of cells, leading to cell death (
      • Pinto M.F.S.
      • Fensterseifer I.C.M.
      • Migliolo L.
      • Sousa D.A.
      • de Capdville G.
      • Arboleda-Valencia J.W.
      • et al.
      Identification and structural characterization of novel cyclotide with activity against an insect pest of sugar cane.
      ). The membrane lytic activity of cyclotides is linked to both insecticidal and hemolytic activities as reported in an alanine-scanning mutagenesis study of kalata B1 (
      • Simonsen S.M.
      • Sando L.
      • Rosengren K.J.
      • Wang C.K.
      • Colgrave M.L.
      • Daly N.L.
      • et al.
      Alanine scanning mutagenesis of the prototypic cyclotide reveals a cluster of residues essential for bioactivity.
      ). To determine whether the biological activities of acyclotides from R. bengalensis depend on their hydrophobicity, the activities of ribe peptides exhibiting a range of values for overall hydrophobicity were evaluated in a series of in vitro and in vivo assays.
      To provide a baseline to interpret the bioassay data, it was important to determine if the structures of acyclotides are rigid and similar to those of cyclotides, which have been extensively studied. The NMR solution structure of ribe 31 reported here suggests that indeed this acyclotide is rigid in solution and has similar secondary structure elements to cyclotides, specifically a ß-sheet and a series of turns built around the cystine knot core (
      • Craik D.J.
      Host-defense activities of cyclotides.
      ). The finding that acyclotides such as ribe 31 have potent insecticidal activity, yet lack a cyclic backbone, indicates that the cystine knot motif is a more dominant structural influence on activity than the cyclic backbone. This finding is in agreement with previous structural studies on cyclotides and synthetic linear analogs (
      • Colgrave M.L.
      • Craik D.J.
      Thermal, chemical, and enzymatic stability of the cyclotide kalata B1: the importance of the cyclic cystine knot.
      ,
      • Wang C.K.
      • Hu S.H.
      • Martin J.L.
      • Sjogren T.
      • Hajdu J.
      • Bohlin L.
      • et al.
      Combined X-ray and NMR analysis of the stability of the cyclotide cystine knot fold that underpins its insecticidal activity and potential use as a drug scaffold.
      ).
      The insect cytotoxic activity of the ribe peptides was tested in MTT assays, and the interaction with the plasma membrane of insect cells was observed using live cell imaging. In vitro toxicity tests indicated that all ribe peptides were cytotoxic toward Sf9 (Spodoptera frugiperda, a pest of global importance) and High Five insect cells, with only moderate variability between the peptides. Ribe 24 and 31 showed the highest toxicity amongst the acyclotides to Sf9 insect cells, which was expected if using overall high hydrophobicity of a particular ribe peptide to be predictive of insect cytotoxicity. Interestingly, the most late-eluting peptide, ribe 33, displayed lower toxicity than the acyclotides ribe 24 and 31. One potential explanation is that the cyclic backbone of ribe 33 has a significant effect on its elution time, and therefore, the difference in elution times between the cyclotide ribe 33 and the acyclotides ribe 24 and 31 is not a true representation of increased hydrophobic patch density.
      The hemolytic activity of the ribe acyclotides against human RBCs was of similar magnitude to that reported previously for a range of cyclotides, that is, EC50 > 10 μM (
      • Henriques S.T.
      • Craik D.J.
      Importance of the cell membrane on the mechanism of action of cyclotides.
      ). The lower toxicity of the ribe peptides toward human RBCs compared to the two insect cell lines suggests that these peptides have a degree of selectivity and hence, with further optimization, would have potential as insecticides with low mammalian toxicity. Simonsen et al. (
      • Simonsen S.M.
      • Sando L.
      • Rosengren K.J.
      • Wang C.K.
      • Colgrave M.L.
      • Daly N.L.
      • et al.
      Alanine scanning mutagenesis of the prototypic cyclotide reveals a cluster of residues essential for bioactivity.
      ) reported that the hemolytic activity of kalata B1 can be eradicated or significantly diminished by mutation of a single amino acid in the bioactive and hydrophobic surfaces. Interestingly, ribe 10 and 31 share high sequence similarity except for three residues at positions 16, 21, and 23. However, the acyclotide ribe 10 shows higher lysis to RBCs than the cyclotide ribe 31. Overall, the toxicity results highlight that the presence of a cyclic backbone does not necessarily correlate with high cytotoxicity. Rather, the data suggest that hydrophobicity plays a key role in the biological activity of both cyclotides and acyclotides.
      The in vivo toxicity of the ribe peptides further supports the hypothesis that the hydrophobicity of cyclotides, extended here to acyclotides, influences their biological activities (
      • Henriques S.T.
      • Craik D.J.
      Importance of the cell membrane on the mechanism of action of cyclotides.
      ). Ribe 31 is the most toxic and the most hydrophobic acyclotide based on the grand average of hydropathy score. The tested cyclotides and acyclotides bound to and destabilized membrane surfaces containing POPE phospholipids but showed weak permeabilization of membranes lacking POPE. This result suggests that the mechanism of membrane destabilization of acyclotides and cyclotides is related to peptide-lipid recognition and subsequent hydrophobic interactions (
      • Henriques S.T.
      • Craik D.J.
      Importance of the cell membrane on the mechanism of action of cyclotides.
      ,
      • Henriques S.T.
      • Huang Y.H.
      • Castanho M.A.R.B.
      • Bagatolli L.A.
      • Sonza S.
      • Tachedjian G.
      • et al.
      Phosphatidylethanolamine binding is a conserved feature of cyclotide-membrane interactions.
      ,
      • Henriques S.T.
      • Huang Y.-H.
      • Chaousis S.
      • Wang C.K.
      • Craik D.J.
      Anticancer and toxic properties of cyclotides are dependent on phosphatidylethanolamine phospholipid targeting.
      ). Ribe 31 has a large hydrophobic patch capable of establishing strong contacts with insect cell membranes. Thus, it appears likely that the insecticidal activity of ribe 31 is due to its ability to effectively bind to and disrupt the lipid membrane of Drosophila cells. Ribe 10 has only one different residue in loop 3 compared to ribe 31 (Gln versus Val), yet ribe 10 has no noticeable in vivo activity. Therefore, it is likely that Val18 of ribe 31 contributes to the exposed hydrophobic patch in loop 3, which is key for membrane-binding activity.
      Henriques et al. (
      • Henriques S.T.
      • Craik D.J.
      Cyclotide structure and function: the role of membrane binding and permeation.
      ) proposed that low concentrations of Möbius and bracelet cyclotides are internalized by targeting cell membranes through initial interaction of a conserved Glu in loop 1 with PE phospholipids and insertion into the outer leaflet via their hydrophobic patch. At high concentrations, it was proposed that cyclotides enter cells through direct membrane translocation or via endocytosis. Ribe 31 (0.5 μM) followed the internalization trend of cyclotides, in that at a low concentration it binds to the outer cell membrane (Fig. 5). The cell necrosis induced by ribe 31 is characterized by a general swelling of the cell and rapid loss of plasma membrane integrity.
      In summary, this is the first study to explore the in vivo insecticidal activity of acyclotides. The acyclotide ribe 31, of all the peptides tested, showed the greatest potency in D. melanogaster and was more potent than the previously reported insecticidal cyclotides cyO2 and kalata B1, with reduced mammalian cell toxicity. A mechanism in which acyclotide ribe 31 targets insect cell membranes through interaction with PE headgroups, followed by insertion into the membrane hydrophobic core and disruption of the cell membrane is supported by the SPR and confocal imaging studies reported here. The observed binding of acyclotides to PE is consistent with their incorporation of a conserved glutamic acid residue in loop 1, which has been identified as the first point of interaction of cyclotides with PE containing membranes (
      • Göransson U.
      • Herrmann A.
      • Burman R.
      • Haugaard-Jönsson L.M.
      • Rosengren K.J.
      The conserved Glu in the cyclotide cycloviolacin O2 has a key structural role.
      ,
      • Henriques S.T.
      • Craik D.J.
      Cyclotide structure and function: the role of membrane binding and permeation.
      ).
      In conclusion, ribe 31 is a promising bioinsecticide lead. NMR structural analysis and stability investigations revealed that this acyclotide adopts the same secondary structural features and has the same stability enhancing characteristics that have stimulated the use of cyclotides as molecular design scaffolds. Given their relative ease of synthesis compared to cyclotides in that no cyclization step is required, acyclotides are an underexplored scaffold for agricultural applications where large scale manufacture would be required. Although less abundant in nature than cyclotides, the potential applications of acyclotides are similarly promising.

      Experimental procedures

      Extraction and isolation of ribe peptides from the stem and leaf of R. bengalensis

      Ribe peptides were individually isolated from stem and leaf of R. bengalensis for the in vitro and in vivo assays. Plant tissues were ground in liquid nitrogen before extraction with 50% MeCN and 1% formic acid in water as previously described (
      • Dang T.T.
      • Chan L.Y.
      • Huang Y.H.
      • Nguyen L.T.T.
      • Kaas Q.
      • Huynh T.
      • et al.
      Exploring the sequence diversity of cyclotides from Vietnamese viola species.
      ). The extracts were centrifuged before supernatant filtration and lyophilization. Crude extracts were redissolved in 10% (v/v) MeCN and 1% (v/v) formic acid and applied to a solid phase extraction reversed-phase column for preliminary fractionation. Isolation of cyclotides from 40% to 50% MeCN solid phase extraction fractions was carried out using RP-HPLC. For preparative peptide separations, samples were diluted in 20% (v/v) MeCN and 1% (v/v) TFA and filtered through a 0.45 μm RC membrane filter and loaded onto a preparative Phenomenex C18 column. Peptides were further purified using semipreparative or analytical HPLC. Linear 1%, 0.5%, or 0.25% min−1 MeCN gradients were delivered to the column at a flow rate of 8 ml min−1 for preparative, 3 ml min−1 for semipreparative, and 1 ml min−1 for standard bore separations, respectively. A dual wavelength UV detector set to 214 and 280 nm was used to monitor eluent with fractions collected. The mass of ribe peptides was confirmed by an electrospray ionization mass spectrometry. Acyclotides ribe 10, 20, 24, and 31 and cyclotide ribe 33 were purified to >95% purity through repetitive RP-HPLC and their purity was confirmed by LC-MS. CyO2 and kalata B1, isolated from Viola odorata and O. affinis, respectively, were used as control peptides for the biological assays in this study.

      Cell culture

      Insect cells High Five and Sf9 were cultured in ESF 921 (Expression Systems) medium at 27 °C, 5% CO2 in a 75 cm2 flask until 80% confluent (
      • Du J.
      • Chan L.Y.
      • Poth A.G.
      • Craik D.J.
      Discovery and characterization of cyclic and acyclic trypsin inhibitors from Momordica dioica.
      ).

      Cytotoxicity assay

      Ribe peptides and positive controls (cyclotides kalata B1 and cycloviolacin O2) were added in triplicate in concentrations from 0.2 μM to 20 μM. Cells in control wells were incubated with 0.1% Triton X-100. Three ‘media only’ wells, which did not contain cells, were also included for background measurement. Cells were incubated for 24 h at 27 °C and 5% CO2. Ten microliters of MTT (5 mg/ml in PBS) was added to each well to a concentration of 0.5 mg/ml. Plates were then incubated for 3 h at 27 °C, 5% CO2. Supernatants were removed, and the insoluble formazan crystals were resuspended in 100 μl of dimethyl sulfoxide. The plate was shaken at room temperature for 10 min to fully dissolve the formazan crystals, and the absorbance of the solutions was measured at 600 nm.

      Hemolytic assay

      Toxicity toward human RBCs from three donors was assessed using the method previously described (
      • Dang T.T.
      • Chan L.Y.
      • Huang Y.H.
      • Nguyen L.T.T.
      • Kaas Q.
      • Huynh T.
      • et al.
      Exploring the sequence diversity of cyclotides from Vietnamese viola species.
      ). Fresh RBCs were collected according to protocols approved by the Human Research Ethics Unit of The University of Queensland (approval number 2013000582). Peptides were prepared at concentrations from 0.2 to 25 μM. Melittin (0.03–3.3 μM) was used as a positive control. Cyclotides or melittin were added to 0.25% RBC solutions in a 96-well round bottom plate. Control wells were included in each plate, where cells incubated with 0.1% Triton-X were used as a positive control (maximum hemolysis) and cells with water as a vehicle control. The plates were incubated at 37 °C for 1 h and then spun at 1000 rpm for 5 min. The supernatant from each well was transferred into new flat-bottomed plates, and the absorbance was measured at 405 nm. Experiments were done in triplicate.

      In vivo feeding assay

      All CAFE experiments were performed with 4 to 6 day old mated female Canton S D. melanogaster at 23 °C. Flies were starved in vials on agarose overnight prior to the capillary feeding experiments (
      • Diegelmann S.
      • Jansen A.
      • Jois S.
      • Kastenholz K.
      • Velo Escarcena L.
      • Strudthoff N.
      • et al.
      The CApillary FEeder assay measures food intake in Drosophila melanogaster.
      ). Each experiment was done with 15 flies per vial, with 7 vials used for each experimental condition (n = 105 flies total per condition). The ribe peptides, cyO2 and kalata B1, were dissolved in milliQ water and quantified using a UV-visible spectrophotometer based on their absorbance at 280 nm. Peptide stocks were diluted in 10% or 20% sucrose and food dye solution to the desired concentrations and fed to the flies through capillaries. The number of dead flies in each vial was recorded every 1 or 2 h excluding the night period. Mortality was monitored for up to 25 days. The experimenter (SO) was blinded to peptide identities and concentrations, which were unmasked after analysis (by TTD). Data visualization and analysis was performed by using estimation statistics as described previously (
      • Ho J.
      • Tumkaya T.
      • Aryal S.
      • Choi H.
      • Claridge-Chang A.
      Moving beyond P values: Data analysis with estimation graphics.
      ). Briefly, the survival of individual flies was displayed in scatter plots where the average survival of the control population was compared with survival of flies that were fed different peptide concentrations. Differences in survival between the control and test groups were displayed as Δ Survival time with 95% confidence interval.

      Structural analysis by NMR

      Peptides were prepared for NMR analysis by dissolving in 90% H2O/10% D2O (v/v) at a concentration of 1 mM. 1H NMR 2D spectra (TOCSY and NOESY) were recorded at mixing times of 80 and 200 ms, respectively, on a Bruker Avance III 600 MHz spectrometer equipped with a cryoprobe at 298 K. Spectra were referenced internally to 4,4-dimethyl-4-silapentane-1-sulfonic acid. NMR data were processed using TOPSPIN 3.6.1 (Bruker), and assignments were made using CCPNMR (version 2.4.4). Additional spectra were acquired for ribe 31 to determine its three-dimensional structure, including TOCSY spectra at variable temperatures (283–303K) and indirect heteronuclear experiments (1H-15N heteronuclear single quantum coherence [HSQC] and 1H-13C HSQC). Initial structures of ribe 31 were calculated using the program CYANA with NOESY-derived distance restraints, distance restraints to define the CI-CIV, CII-CV, CIII-CVI connectivity, and backbone φ and ψ dihedral angles generated using the program TALOS-N (
      • Shen Y.
      • Bax A.
      Protein backbone and sidechain torsion angles predicted from NMR chemical shifts using artificial neural networks.
      ). Hydrogen bond pairs were added based upon preliminary structures and amide temperature coefficients (
      • Cierpicki T.
      • Otlewski J.
      Amide proton temperature coefficients as hydrogen bond indicators in proteins.
      ). CNS (
      • Brunger A.T.
      Version 1.2 of the crystallography and NMR system.
      ) was used to generate a final set of structures using torsion angle dynamics with refinement and energy minimization in explicit solvent; final structures were assessed for stereochemical quality using MolProbity (
      • Chen V.B.
      • Arendall 3rd, W.B.
      • Headd J.J.
      • Keedy D.A.
      • Immormino R.M.
      • Kapral G.J.
      • et al.
      MolProbity: All-atom structure validation for macromolecular crystallography.
      ).

      Cytotoxicity of peptides visualized using confocal microscopy

      Sf9 insect cells were seeded in 8-well borosilicate plates (Thermo Fisher Scientific) at 10,000 cells/well, the day prior to the assay. Right before the experiment, the medium was replaced with phenol red–free Dulbecco's modified Eagle's medium (DMEM) medium containing SYTOX Green. Live cell imaging started immediately after Sf9 cells were treated with acyclotide ribe 31 at 1 μM. Images were captured using a 20 × 0.45 NA S Plan Fluor ELWD objective and SYTOX Green fluorescence signal was amplified using a 525/30 nm band-pass filter. Images were captured every 60 s for 4 h using a custom built inverted widefield Nikon deconvolution microscopy and were analyzed and exported using FIJI software (https://fiji.sc/) (
      • Du Q.
      • Chan L.Y.
      • Gilding E.K.
      • Henriques S.T.
      • Condon N.D.
      • Ravipati A.S.
      • et al.
      Discovery and mechanistic studies of cytotoxic cyclotides from the medicinal herb Hybanthus enneaspermus.
      ). Cells were incubated at 27 °C with 5% CO2 during the imaging process.

      Peptide–membrane interaction visualized using confocal imaging

      To visualize the interaction of ribe 33 and kalata B1 and cyO2 with insect cells, Sf9 insect cells were seeded in 8-well borosilicate plates (Thermo Fisher Scientific) at 15,000 cells/well, 1 day prior to the assay. Immediately before the experiment, the medium was replaced with phenol red–free DMEM medium containing wheat germ agglutinin Alexa-647 conjugate (1:500, v/v), and cells were treated with Alexa Fluor 488–labeled ribe 31, cyO2, or kalata B1 at 0.5 μM, individually. Images were captured after 1 h treatment on Zeiss 880 LSM confocal equipped with a 63x 1.4 NA Plan Apochromat objective running Zeiss Zen Black software. The microscopy images were analyzed using FIJI software (
      • Du Q.
      • Chan L.Y.
      • Gilding E.K.
      • Henriques S.T.
      • Condon N.D.
      • Ravipati A.S.
      • et al.
      Discovery and mechanistic studies of cytotoxic cyclotides from the medicinal herb Hybanthus enneaspermus.
      ). Cells were incubated at 27 °C with 5% CO2 during the imaging process.

      Lipid binding of ribe peptides evaluated using SPR

      Synthetic lipids POPC and POPE were purchased from Avanti Polar Lipids. Two model membranes were prepared: POPC and POPC/POPE (80:20, mol/mol) using the method described previously (
      • Du Q.
      • Chan L.Y.
      • Gilding E.K.
      • Henriques S.T.
      • Condon N.D.
      • Ravipati A.S.
      • et al.
      Discovery and mechanistic studies of cytotoxic cyclotides from the medicinal herb Hybanthus enneaspermus.
      ). Briefly, small unilamellar vesicles resuspended in Hepes buffer (10 mM Hepes, 150 mM NaCl, pH 7.4, filtered) were obtained using a mini extruder (Avanti) with a 50 nm polycarbonate membrane. Membrane binding of acyclotides and cyclotides toward POPC or POPC/POPE was evaluated using an L1 sensor chip on a Biacore T200 instrument. A suspension of small unilamellar vesicles (0.5 mM) was deposited onto an L1 chip surface at 2 μl/min for 40 min, followed by a short pulse of 10 mM NaOH. The cyclotide solutions were injected over the lipid surface at 5 μl/min for 180 s, and the dissociation process was recorded for 10 min. Dose–response curves were established through the calculated peptide-to-lipid molar ratio to evaluate the binding ability of the tested peptides. KD was calculated by fitting in specific binding with Hill slope using GraphPad Prism software (GraphPad Software Inc) (
      • Du Q.
      • Chan L.Y.
      • Gilding E.K.
      • Henriques S.T.
      • Condon N.D.
      • Ravipati A.S.
      • et al.
      Discovery and mechanistic studies of cytotoxic cyclotides from the medicinal herb Hybanthus enneaspermus.
      ).

      Data availability

      Three-dimensional structural data and chemical shifts of ribe 31 have been deposited (PDB ID 7KPD, BMRB Accession No. 30813).

      Supporting information

      The supporting information is available free of charge (
      • Wang C.K.
      • Colgrave M.L.
      • Ireland D.C.
      • Kaas Q.
      • Craik D.J.
      Despite a conserved cystine knot motif, different cyclotides have different membrane binding modes.
      ).

      Conflict of interest

      The authors declare that they have no conflicts of interest with the contents of this article.

      Acknowledgments

      This work was supported by grants from the Australian Research Council (DP150100443 and DP200101299) and by access to the facilities of the Australian Research Council (ARC) Centre of Excellence for Innovations in Peptide and Protein Science (CE200100012). The work was facilitated using infrastructure provided by the Australian Government through the National Collaborative Research Infrastructure strategy (NCRIS). Microscopy was performed at the Australian Cancer Research Foundation (ACRF)/Institute for Molecular Bioscience Cancer Biology Imaging Facility.

      Author contributions

      A. C. C. and D. J. C. conceptualization; T. T. D., Y. H. H., S. O., P. J. H., E. K. G., and L. Y. C. methodology; Q. K. software; Q. K. validation; T. T. D., Y. H. H., S. O., P. J. H., and E. K. G. formal analysis; T. T. D., Y. H. H., S. O., P. J. H., E. K. G., L. Y. C., A. C. C., and D. J. C. investigation; A. C. C. and D. J. C. resources; T. T. D., Y. H. H., S. O., P. J. H., E. K. G., L. Y. C., B. J. T., and Q. K. writing–original draft; T. T. D., Y. H. H., S. O., P. J. H., E. K. G., L. Y. C., Q. K., A. C. C., and D. J. C. writing–review & editing; T. T. D. and S. O. visualization; Y. H. H., L. Y. C., Q. K., A. C. C., and D. J. C. supervision; L. Y. C. project administration; A. C. C. and D. J. C. funding acquisition.

      Funding and additional information

      L. Y. C was supported by the Advance Queensland Women’s Academic Fund ( WAF-6884942288 ) and D. J. C is an NHMRC Leadership Fellow ( GNT2009564 ). A. C. C. and S. O. were supported by grants from the Singapore Ministry of Education ( MOE2017-T2-1-089 ) and by Duke-NUS Medical School.

      Supporting information

      References

        • Craik D.J.
        • Daly N.L.
        • Bond T.
        • Waine C.
        Plant cyclotides: a unique family of cyclic and knotted proteins that defines the cyclic cystine knot structural motif.
        J. Mol. Biol. 1999; 294: 1327-1336
        • de Veer S.J.
        • Kan M.W.
        • Craik D.J.
        Cyclotides: from structure to function.
        Chem. Rev. 2019; 119: 12375-12421
        • Wang C.K.
        • Craik D.J.
        Designing macrocyclic disulfide-rich peptides for biotechnological applications.
        Nat. Chem. Biol. 2018; 14: 417-427
        • Colgrave M.L.
        • Craik D.J.
        Thermal, chemical, and enzymatic stability of the cyclotide kalata B1: the importance of the cyclic cystine knot.
        Biochemistry. 2004; 43: 5965-5975
        • Hernandez J.F.
        • Gagnon J.
        • Chiche L.
        • Nguyen T.M.
        • Andrieu J.P.
        • Heitz A.
        • et al.
        Squash trypsin inhibitors from Momordica cochinchinensis exhibit an atypical macrocyclic structure.
        Biochemistry. 2000; 39: 5722-5730
        • Craik D.J.
        Host-defense activities of cyclotides.
        Toxins. 2012; 4: 139-156
        • Simonsen S.M.
        • Sando L.
        • Ireland D.C.
        • Colgrave M.L.
        • Bharathi R.
        • Göransson U.
        • et al.
        A continent of plant defense peptide diversity: cyclotides in Australian Hybanthus (Violaceae).
        Plant Cell. 2005; 17: 3176-3189
        • Nguyen K.N.T.
        • Nguyen G.K.T.
        • Nguyen P.Q.T.
        • Ang K.H.
        • Dedon P.C.
        • Tam J.P.
        Immunostimulating and Gram-negative-specific antibacterial cyclotides from the butterfly pea (Clitoria ternatea).
        FEBS J. 2016; 283: 2067-2090
        • Porto W.F.
        • Pires A.S.
        • Franco O.L.
        CS-AMPPred: an updated SVM model for antimicrobial activity prediction in cysteine-stabilized peptides.
        PLoS One. 2012; 7e51444
        • Pränting M.
        • Lööv C.
        • Burman R.
        • Göransson U.
        • Andersson D.I.
        The cyclotide cycloviolacin O2 from Viola odorata has potent bactericidal activity against Gram-negative bacteria.
        J. Antimicrob. Chemother. 2010; 65: 1964-1971
        • Göransson U.
        • Sjögren M.
        • Svangård E.
        • Claeson P.
        • Bohlin L.
        Reversible antifouling effect of the cyclotide cycloviolacin O2 against barnacles.
        J. Nat. Prod. 2004; 67: 1287-1290
        • Dang T.T.
        • Chan L.Y.
        • Huang Y.H.
        • Nguyen L.T.T.
        • Kaas Q.
        • Huynh T.
        • et al.
        Exploring the sequence diversity of cyclotides from Vietnamese viola species.
        J. Nat. Prod. 2020; 83: 1817-1828
        • Pinto M.F.S.
        • Fensterseifer I.C.M.
        • Migliolo L.
        • Sousa D.A.
        • de Capdville G.
        • Arboleda-Valencia J.W.
        • et al.
        Identification and structural characterization of novel cyclotide with activity against an insect pest of sugar cane.
        J. Biol. Chem. 2012; 287: 134-147
        • Plan M.R.R.
        • Saska I.
        • Cagauan A.G.
        • Craik D.J.
        Backbone cyclised peptides from plants show molluscicidal activity against the rice pest Pomacea canaliculata (golden apple snail).
        J. Agric. Food Chem. 2008; 56: 5237-5241
        • Jennings C.
        • West J.
        • Waine C.
        • Craik D.
        • Anderson M.
        Biosynthesis and insecticidal properties of plant cyclotides: the cyclic knotted proteins from Oldenlandia affinis.
        Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 10614-10619
        • Dancewicz K.
        • Slazak B.
        • Kielkiewicz M.
        • Kapusta M.
        • Bohdanowicz J.
        • Gabrys B.
        Behavioral and physiological effects of Viola spp. cyclotides on Myzus persicae (Sulz.).
        J. Insect Physiol. 2020; 122104025
        • Poth A.G.
        • Colgrave M.L.
        • Lyons R.E.
        • Daly N.L.
        • Craik D.J.
        Discovery of an unusual biosynthetic origin for circular proteins in legumes.
        Proc. Natl. Acad. Sci. U. S. A. 2011; 108: 10127-10132
        • Tammineni R.
        • Gulati P.
        • Kumar S.
        • Mohanty A.
        An overview of acyclotides: past, present and future.
        Phytochemistry. 2020; 170112215
        • Nguyen G.K.
        • Lian Y.L.
        • Pang E.W.H.
        • Phuong Q.T.N.
        • Tran T.D.
        • Tam J.P.
        Discovery of linear cyclotides in monocot plant Panicum laxum of Poaceae family provides new insights into evolution and distribution of cyclotides in plants.
        J. Biol. Chem. 2013; 288: 3370-3380
        • Dang T.T.
        • Chan L.Y.
        • Tombling B.J.
        • Harvey P.J.
        • Gilding E.K.
        • Craik D.J.
        Planta discovery and chemical synthesis of bracelet cystine knot peptides from Rinorea bengalensis.
        J. Nat. Prod. 2021; 84: 395-407
        • Jackson M.A.
        • Gilding E.K.
        • Shafee T.
        • Harris K.S.
        • Kaas Q.
        • Poon S.
        • et al.
        Molecular basis for the production of cyclic peptides by plant asparaginyl endopeptidases.
        Nat. Commun. 2018; 9: 2411
        • Ireland D.C.
        • Colgrave M.L.
        • Nguyencong P.
        • Daly N.L.
        • Craik D.J.
        Discovery and characterization of a linear cyclotide from Viola odorata: implications for the processing of circular proteins.
        J. Mol. Biol. 2006; 357: 1522-1535
        • Mylne J.S.
        • Chan L.Y.
        • Chanson A.H.
        • Daly N.L.
        • Schaefer H.
        • Bailey T.L.
        • et al.
        Cyclic peptides arising by evolutionary parallelism via asparaginyl-endopeptidase–mediated biosynthesis.
        Plant Cell. 2012; 24: 2765-2778
        • Wang C.K.
        • Kaas Q.
        • Chiche L.
        • Craik D.J.
        CyBase: a database of cyclic protein sequences and structures, with applications in protein discovery and engineering.
        Nucleic Acids Res. 2008; 36: D206-D210
        • Mulvenna J.P.
        • Wang C.
        • Craik D.J.
        CyBase: a database of cyclic protein sequence and structure.
        Nucleic Acids Res. 2006; 34: D192-D194
        • Plan M.R.R.
        • Göransson U.
        • Clark R.J.
        • Daly N.L.
        • Colgrave M.L.
        • Craik D.J.
        The cyclotide fingerprint in Oldenlandia affinis: elucidation of chemically modified, linear and novel macrocyclic peptides.
        ChemBioChem. 2007; 8: 1001-1011
        • Du J.
        • Chan L.Y.
        • Poth A.G.
        • Craik D.J.
        Discovery and characterization of cyclic and acyclic trypsin inhibitors from Momordica dioica.
        J. Nat. Prod. 2019; 82: 293-300
        • Hossard L.
        • Philibert A.
        • Bertrand M.
        • Colnenne-David C.
        • Debaeke P.
        • Munier-Jolain N.
        • et al.
        Effects of halving pesticide use on wheat production.
        Sci. Rep. 2014; 4: 4405
        • Braga A.R.C.
        • de Rosso V.V.
        • Harayashiki C.A.Y.
        • Jimenez P.C.
        • Castro Í.B.
        Global health risks from pesticide use in Brazil.
        Nat. Food. 2020; 1: 312-314
        • Gill J.P.S.
        • Bedi J.S.
        • Singh R.
        • Fairoze M.N.
        • Hazarika R.A.
        • Gaurav A.
        • et al.
        Pesticide residues in peri-urban bovine milk from India and risk assessment: a multicenter study.
        Sci. Rep. 2020; 10: 8054
        • Wang N.
        • Jiang Q.
        • Han L.
        • Zhang H.
        • Zhu B.
        • Liu X.
        Epidemiological characteristics of pesticide poisoning in Jiangsu Province, China, from 2007 to 2016.
        Sci. Rep. 2019; 9: 8604
        • Goodhue R.E.
        • Bolda M.
        • Farnsworth D.
        • Williams J.C.
        • Zalom F.G.
        Spotted wing drosophila infestation of California strawberries and raspberries: Economic analysis of potential revenue losses and control costs.
        Pest Manag. Sci. 2011; 67: 1396-1402
        • Riaz B.
        • Zahoor M.K.
        • Zahoor M.A.
        • Majeed H.N.
        • Javed I.
        • Ahmad A.
        • et al.
        Toxicity, phytochemical composition, and enzyme inhibitory activities of some indigenous weed plant extracts in fruit fly, Drosophila melanogaster.
        Evid. Based Complement. Alternat. Med. 2018; 20182325659
        • Tolwinski N.S.
        Introduction: Drosophila-A model system for developmental Biology.
        J. Dev. Biol. 2017; 5: 9
        • Scott J.G.
        • Buchon N.
        Drosophila melanogaster as a powerful tool for studying insect toxicology.
        Pestic. Biochem. Physiol. 2019; 161: 95-103
        • Akmoutsou P.
        • Mademtzoglou D.
        • Nakou I.
        • Onoufriadis A.
        • Papadopoulou X.
        • Kounatidis I.
        • et al.
        Evaluation of toxicity and genotoxic effects of spinosad and deltamethrin in Drosophila melanogaster and Bactrocera oleae.
        Pest Manag. Sci. 2011; 67: 1534-1540
        • Muñoz E.
        • Lamilla C.
        • Marin J.C.
        • Alarcon J.
        • Cespedes C.L.
        Antifeedant, insect growth regulatory and insecticidal effects of Calceolaria talcana (Calceolariaceae) on Drosophila melanogaster and Spodoptera frugiperda.
        Ind. Crops Prod. 2013; 42: 137-144
        • Chowanski S.
        • Chudzinska E.
        • Lelario F.
        • Ventrella E.
        • Marciniak P.
        • Miadowicz-Kobielska M.
        • et al.
        Insecticidal properties of Solanum nigrum and Armoracia rusticana extracts on reproduction and development of Drosophila melanogaster.
        Ecotoxicol. Environ. Saf. 2018; 162: 454-463
        • Ffrench-Constant R.H.
        The molecular genetics of insecticide resistance.
        Genetics. 2013; 194: 807-815
        • Strycharz J.P.
        • Lao A.
        • Li H.
        • Qiu X.
        • Lee S.H.
        • Sun W.
        • et al.
        Resistance in the highly DDT-resistant 91-R strain of Drosophila melanogaster involves decreased penetration, increased metabolism, and direct excretion.
        Pestic. Biochem. Physiol. 2013; 107: 207-217
        • Göransson U.
        • Herrmann A.
        • Burman R.
        • Haugaard-Jönsson L.M.
        • Rosengren K.J.
        The conserved Glu in the cyclotide cycloviolacin O2 has a key structural role.
        ChemBioChem. 2009; 10: 2354-2360
        • Jaspard E.
        • Macherel D.
        • Hunault G.
        Computational and statistical analyses of amino acid usage and physico-chemical properties of the twelve late embryogenesis abundant protein classes.
        PLoS One. 2012; 7e36968
        • Colgrave M.L.
        • Kotze A.C.
        • Huang Y.H.
        • O'Grady J.
        • Simonsen S.M.
        • Craik D.J.
        Cyclotides: natural, circular plant peptides that possess significant activity against gastrointestinal nematode parasites of sheep.
        Biochemistry. 2008; 47: 5581-5589
        • Colgrave M.L.
        • Kotze A.C.
        • Ireland D.C.
        • Wang C.K.
        • Craik D.J.
        The anthelmintic activity of the cyclotides: natural variants with enhanced activity.
        ChemBioChem. 2008; 9: 1939-1945
        • Jones H.E.
        • Harwood J.L.
        • Bowen I.D.
        • Griffiths G.
        Lipid composition of subcellular membranes from larvae and prepupae of Drosophila melanogaster.
        Lipids. 1992; 27: 984-987
        • Henriques S.T.
        • Craik D.J.
        Importance of the cell membrane on the mechanism of action of cyclotides.
        ACS Chem. Biol. 2012; 7: 626-636
        • Henriques S.T.
        • Huang Y.H.
        • Castanho M.A.R.B.
        • Bagatolli L.A.
        • Sonza S.
        • Tachedjian G.
        • et al.
        Phosphatidylethanolamine binding is a conserved feature of cyclotide-membrane interactions.
        J. Biol. Chem. 2012; 287: 33629-33643
        • Henriques S.T.
        • Huang Y.-H.
        • Chaousis S.
        • Wang C.K.
        • Craik D.J.
        Anticancer and toxic properties of cyclotides are dependent on phosphatidylethanolamine phospholipid targeting.
        ChemBioChem. 2014; 15: 1956-1965
        • Broussalis A.M.
        • Clemente S.
        • Ferraro G.E.
        Hybanthus parviflorus (Violaceae): Insecticidal activity of a south American plant.
        Crop Prot. 2010; 29: 953-956
        • Gruber C.W.
        • Čemažar M.
        • Anderson M.A.
        • Craik D.J.
        Insecticidal plant cyclotides and related cystine knot toxins.
        Toxicon. 2007; 49: 561-575
        • Jennings C.V.
        • Rosengren K.J.
        • Daly N.L.
        • Plan M.
        • Stevens J.
        • Scanlon M.J.
        • et al.
        Isolation, solution structure, and insecticidal activity of kalata B2, a circular protein with a twist: do möbius strips exist in nature?.
        Biochemistry. 2005; 44: 851-860
        • Simonsen S.M.
        • Sando L.
        • Rosengren K.J.
        • Wang C.K.
        • Colgrave M.L.
        • Daly N.L.
        • et al.
        Alanine scanning mutagenesis of the prototypic cyclotide reveals a cluster of residues essential for bioactivity.
        J. Biol. Chem. 2008; 283: 9805-9813
        • Wang C.K.
        • Hu S.H.
        • Martin J.L.
        • Sjogren T.
        • Hajdu J.
        • Bohlin L.
        • et al.
        Combined X-ray and NMR analysis of the stability of the cyclotide cystine knot fold that underpins its insecticidal activity and potential use as a drug scaffold.
        J. Biol. Chem. 2009; 284: 10672-10683
        • Henriques S.T.
        • Craik D.J.
        Cyclotide structure and function: the role of membrane binding and permeation.
        Biochemistry. 2017; 56: 669-682
        • Diegelmann S.
        • Jansen A.
        • Jois S.
        • Kastenholz K.
        • Velo Escarcena L.
        • Strudthoff N.
        • et al.
        The CApillary FEeder assay measures food intake in Drosophila melanogaster.
        J. Vis. Exp. 2017; 12155024
        • Ho J.
        • Tumkaya T.
        • Aryal S.
        • Choi H.
        • Claridge-Chang A.
        Moving beyond P values: Data analysis with estimation graphics.
        Nat. Methods. 2019; 16: 565-566
        • Shen Y.
        • Bax A.
        Protein backbone and sidechain torsion angles predicted from NMR chemical shifts using artificial neural networks.
        J. Biomol. NMR. 2013; 56: 227-241
        • Cierpicki T.
        • Otlewski J.
        Amide proton temperature coefficients as hydrogen bond indicators in proteins.
        J. Biomol. NMR. 2001; 21: 249-261
        • Brunger A.T.
        Version 1.2 of the crystallography and NMR system.
        Nat. Protoc. 2007; 2: 2728-2733
        • Chen V.B.
        • Arendall 3rd, W.B.
        • Headd J.J.
        • Keedy D.A.
        • Immormino R.M.
        • Kapral G.J.
        • et al.
        MolProbity: All-atom structure validation for macromolecular crystallography.
        Acta Crystallogr. D Biol. Crystallogr. 2010; 66: 12-21
        • Du Q.
        • Chan L.Y.
        • Gilding E.K.
        • Henriques S.T.
        • Condon N.D.
        • Ravipati A.S.
        • et al.
        Discovery and mechanistic studies of cytotoxic cyclotides from the medicinal herb Hybanthus enneaspermus.
        J. Biol. Chem. 2020; 295: 10911-10925
        • Wang C.K.
        • Colgrave M.L.
        • Ireland D.C.
        • Kaas Q.
        • Craik D.J.
        Despite a conserved cystine knot motif, different cyclotides have different membrane binding modes.
        Biophys. J. 2009; 97: 1471-1481