ORIGINAL RESEARCH

published: 26 August 2021

doi: 10.3389/fmicb.2021.720381

Identification and Characterization of

Antimicrobial Peptides From
Butterflies: An Integrated
Bioinformatics and Experimental
Study
Min Wang 1,2,3* † , Ziyue Zhou 3 , Simin Li 4 , Wei Zhu 5,6 and Xianda Hu 7* †
1
Beijing Hospital, National Center of Gerontology, Beijing, China, 2 Institute of Geriatric Medicine, Chinese Academy of
Medical Sciences, Beijing, China, 3 Peking Union Medical College Hospital, Chinese Academy of Medical Sciences
and Peking Union Medical College, Beijing, China, 4 Stomatological Hospital, Southern Medical University, Guangzhou,
China, 5 Institute of Stem Cell and Regenerative Biology, College of Animal Science and Veterinary Medicine, Huazhong
Agricultural University, Wuhan, China, 6 Key Laboratory of Agricultural Animal Genetics, Breeding, and Reproduction
of Ministry of Education, Huazhong Agricultural University, Wuhan, China, 7 Beijing Tibetan Hospital, China Tibetology
Edited by: Research Center, Beijing, China
Marc Maresca,
Aix-Marseille Université, France
Reviewed by:
Butterflies represent one of the largest animal groups on Earth, yet antimicrobial peptides
Séverine Zirah, (AMPs) of this group are less studied in comparison with their moth counterparts. This
Muséum National d’Histoire Naturelle, study employed an integrated bioinformatics approach to survey natural AMPs from
France
Hamza Olleik, publicly available genomic datasets. Numerous AMPs, including cecropins, defensins,
University of Technology Compiegne, and moricins, were identified and subsequently used as templates for the design
France
of a series of synthetic AMPs that mimicked the naturally occurring sequences.
*Correspondence:
Min Wang
Despite differing biological effects among the various sequences, the synthetic AMPs
vivian-08152003@hotmail.com exhibited potent antibacterial and antifungal activities in vitro and in vivo, without
Xianda Hu inducing hemolysis, which implied their therapeutic potential in infectious diseases.
hellocean@hotmail.com
† These authors have contributed
Electron and confocal fluorescence microscopies revealed that the AMPs induced
equally to this work and share first distinct morphological and biophysical changes on microbial cell membranes and nuclei,
authorship suggesting that the antimicrobial effects were related to a mechanism of membrane
penetration and nucleic acid binding by the peptides. In conclusion, this study not
Specialty section:
This article was submitted to only offers insights into butterfly AMPs but also provides a practical strategy for
Antimicrobials, Resistance high-throughput natural AMP discoveries that will have implications for future research
and Chemotherapy,
a section of the journal in this area.
Frontiers in Microbiology
Keywords: antimicrobial peptide, butterfly, bioinformatics, identification, characterization
Received: 04 June 2021
Accepted: 21 July 2021
Published: 26 August 2021
INTRODUCTION
Citation:
Wang M, Zhou Z, Li S, Zhu W and Infectious diseases were a major problem in human history until the development of various
Hu X (2021) Identification efficacious antibiotics that target the causative microorganisms. However, to adapt an aphorism
and Characterization of Antimicrobial
of Friedrich Nietzsche, that which does not kill them makes them stronger, consequently,
Peptides From Butterflies: An
Integrated Bioinformatics
antimicrobial resistance emerged thereafter and has become a significant threat to public health
and Experimental Study. worldwide. Declining numbers of new antibiotic candidates has further worsened concern about
Front. Microbiol. 12:720381. the shortage of antibiotics in the near future, and development of new classes of antimicrobial
doi: 10.3389/fmicb.2021.720381 agents is therefore urgently required (Roope et al., 2019).

Frontiers in Microbiology | www.frontiersin.org 1 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

Antimicrobial peptides (AMPs) are a heterogeneous group of performed using Augustus 3.3.11 (Stanke et al., 2008) trained
small amino acid molecules that are evolutionarily conserved on the Heliconius melpomene genome model (parameters:
in the genome and ubiquitously produced throughout the augustus –softmasking = true –species = heliconius_melpomene1
kingdoms of life to combat microbial infections (Unckless genome.fasta), and complete predicted open reading frames
and Lazzaro, 2016). As effector molecules of the innate host (ORFs) of up to 400 amino acids were retained. The predicted
defense, AMPs are vital to all living organisms, especially simpler encoded amino acid sequences were aligned against AMP
organisms that only depend on innate and humoral immunity to sequences obtained from the RefSeq protein database2 using
prevent the onset of infection (Jasper and Bohmann, 2002). In BLASTp (O’Leary et al., 2016), with the highest scoring
recent decades, AMPs have received considerable attention as a homologies considered as final annotation results. Amino
promising group of molecules due to their unique mechanism acid sequences of the predicted genes were also matched
of rapid physical disruption of microbial membranes, although with the corresponding hidden Markov models (HMMs) of
this dogma is now being challenged, as alternative targets have AMPs retrieved from the Pfam database3 with an e-value
also been reported (Manniello et al., 2021). This mechanism threshold of 10−04 (Mistry et al., 2020). The results of all
is preferable for suppression of the deteriorating resistance these analyses were merged as a putative AMP candidates
problem because microbes are eliminated regardless of antibiotic set. In addition, other available sequences of other AMPs and
sensitivity or resistance, and also AMPs are not prone to inducing AMP-encoding transcripts derived from Papilionoidea species
resistant mutants (Zharkova et al., 2019). Furthermore, other in public databases, as well as AMPs identified from our
desirable properties such as broad-spectrum activity and low next-generation transcriptomic data (GSE142679) of Papilio
host toxicity make AMPs appropriate alternatives to antibiotics clytia and Atrophaneura mencius following lipopolysaccharide
(Hazam et al., 2019). (LPS) challenges and verified by Sanger sequencing, were also
To date, numerous AMPs have been researched as drug included for further analyses. The sequences of mature peptides
candidates, and many are currently in clinical trials, the majority were deduced by alignment with the APD3 database (Wang
of which are naturally derived (Greber and Dawgul, 2017). et al., 2016) and identification of potential cleavage sites in the
As one of the most diverse and abundant orders of animals putative prepropeptides.
on Earth, Lepidoptera, which comprises butterflies and moths,
is a likely major source of natural AMPs (Badapanda and Multiple Sequence Alignment and
Chikara, 2016). Among the order Lepidoptera, Papilionoidea
(butterflies), which comprises typical butterflies (Lycaenidae,
Phylogenetic Analyses
The deduced butterfly AMPs and homologous sequences
Nymphalidae, Papilionidae, Pieridae, and Riodinidae), skippers
available from public databases were subjected to phylogenetic
(Hesperiidae), and moth-butterflies (Hedylidae) (Kawahara and
analyses (Supplementary Table 2). Phylogenetic trees were
Breinholt, 2014), is one of the most speciose superfamilies, but
reconstructed using MEGA X (version 10.1.7) software (Kumar
there are limited studies on AMPs of this superfamily. However,
et al., 2018), based on peptidyl sequence similarities and
most available genomic datasets of lepidopteran species belong
differences aligned by ClustalW2 (version 2.1) program4 (Larkin
to the butterflies, which are characterized by their relatively
et al., 2007, 0) using multiple alignment parameters with
small genome size and relatively low repeat content when
default settings.
compared with other eukaryote genomes (Triant et al., 2018).
Therefore, identification of the transcripts and gene-encoded
peptidyl sequences from butterfly genomes is a feasible approach Peptide Design
to investigate whether high-throughput sequencing data are a To reveal the evolutionarily conserved positions of the AMPs,
potential repository for natural butterfly AMP sequences. entropy calculations were performed, and the residue preference
In this study, an integrative bioinformatics analysis was of putative butterfly-derived natural AMPs was investigated
conducted to discover naturally occurring AMPs from butterfly and visualized using the WebLogo program5 (Crooks et al.,
genomes. Based on the sequence and structural characteristics of 2004). A series of peptide analogs were designed based
these butterfly AMPs, a series of synthetic analogs were designed on the sequence conservation and amino acid frequency at
and synthesized. The antimicrobial activities and mechanisms particular positions. Hydropathic characteristics of the amino
of these synthetic molecules were studied and subsequently acids at evolutionarily variable positions were also considered,
discussed (Supplementary Figure 1). which retained a consistent hydrophilicity or hydrophobicity
to respective positions at the amphipathic α-helices. Estimated
physicochemical characteristics of the newly designed AMPs
were calculated using PepCalc tool6 (Lear and Cobb, 2016).
MATERIALS AND METHODS
1
http://augustus.gobics.de/
In silico Screening of Butterfly 2
http://www.ncbi.nlm.nih.gov/RefSeq/
Antimicrobial Peptides 3
http://pfam.xfam.org/
Complete genome sequences of Papilionoidea species available in 4
http://www.ebi.ac.uk/tools/clustalw2
the Gene Expression Omnibus (GEO) database were downloaded 5
http://weblogo.berkeley.edu/
(Supplementary Table 1). Ab initio gene predictions were 6
http://pepcalc.com

Frontiers in Microbiology | www.frontiersin.org 2 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

Secondary structures were constructed by homology modeling Hemolysis Assay

using the SWISS-MODEL server7 (Waterhouse et al., 2018). Hemolysis activity of the synthesized peptides was tested using
freshly drawn mouse erythrocytes, which were rinsed three times
Peptide Synthesis with phosphate-buffered saline (PBS) and then cultured with
The designed AMPs were synthesized by the standard peptides at concentrations of 4, 8, 16, 32, and 64 µmol·L−1 for 2 h
Fmoc-based solid-phase peptide synthesis (SPPS) methodology (Lin et al., 2013). After centrifugation, 100 µl of the supernatant
(GL Biochem Ltd., Shanghai, China) on respective Wang from each sample was transferred to a microtiter plate, and the
resins and Rink Amide MBHA resin for C-terminal amidated cell-free hemoglobin, representing hemolysis, was detected at
peptides. Intramolecular disulfide bond formation was achieved 550 nm with a microplate reader (Pulang, China).
by air-mediated oxidation. Fluorescence-labeled AMPs were
synthesized using the same protocol with conjugation of Cytotoxicity Assay
fluorescein isothiocyanate (FITC) and an ε-aminocaproic To evaluate the potential cytotoxicity on the normal cells, human
(Acp) linker at the N-terminus. Synthesized peptides were embryonic kidney (HEK293) cells were used, and the cell viability
purified using a high-performance liquid chromatography was assessed by cell counting kit (CCK)-8 assays following
(HPLC) system (LC3000, Beijing Chuangxin Tongheng, Beijing, the manufacturer’s protocol (Dojindo Inc., Kumamoto, Japan)
China), and the identity of the peptides was confirmed by (Wang et al., 2018). Briefly, the cells were seeded in microtiter
mass spectrometry (6125B, Agilent Technologies, Santa Clara, plates at a density of 1 × 104 cells per well for 24 h and
CA, United States). cultured with increasing concentrations of AMPs ranging from
4 to 64 µmol·L−1 and cultured for 24 h. The CCK-8 solution
In vitro Antimicrobial Assay was then added to each well, and the cells were incubated for an
In vitro antimicrobial activities of the synthetic peptides against additional 3 h. The optical density (OD) was measured at 450 nm.
different microorganisms, including the Gram-negative bacteria
Escherichia coli (ATCC 25922), the Gram-positive bacteria Membrane Permeability Assay
Staphylococcus aureus (ATCC 6538P), and the fungus Candida SYTOX Green uptake assays were performed according to the
albicans (ATCC 10231), were determined by microtiter broth instructions of the manufacturer (Thermo Fisher Scientific,
dilution assays (Wiegand et al., 2008). Briefly, the peptides were Waltham, MA, United States) to estimate the membrane
dissolved and diluted with respective culture mediums to a permeabilization effect of the AMPs (Jin et al., 2016). Briefly,
gradient concentration range of 0.01–32 µmol·L−1 and were then 1.0 × 108 CFU·ml−1 microorganisms in logarithmic phase were
applied to mid-logarithmic phase microbes with cell densities incubated with AMPs at concentrations of 4× the corresponding
of approximately 1.0 × 108 CFU·ml−1 and cultured for 24 h. MIC value for 1 h. Microbial cells were then collected, washed
The absorbance of the cultures was evaluated at 600 nm using in Hanks’ balanced salt solution (Solarbio, Beijing, China),
a microplate reader (Pulang, China). The minimum inhibitory and incubated with SYTOX Green Nucleic Acid Stain. The
concentration (MIC) was defined as the lowest concentration of fluorescent intensity, which reflects uptake of SYTOX Green, was
each peptide that exhibited apparent growth suppression on the measured using a fluorescence microplate reader (Thermo Fisher
microbes, while the minimum bactericidal concentration (MBC) Scientific), at wavelengths of 485 and 520 nm for excitation and
was defined as the lowest peptide concentration that resulted in emission, respectively.
no obvious colony growth of the microorganisms.
Electron Microscopy
Antibiofilm Assays Samples for electron microscopy were prepared according to a
Biofilms of the aforementioned microbes were allowed to form previously described method (Hartmann et al., 2010). In general,
in microtiter plates under respective culture conditions in the 1.0 × 108 CFU·ml−1 exponential growth phase cultures of
absence (eradication) or presence (prevention) of AMPs. After E. coli, S. aureus, and C. albicans were incubated with AMPs
incubation, the culture supernatant was discarded, and each at 4× the corresponding MIC value for 1 h. Microbial cells
well was gently rinsed to remove planktonic bacteria. Biofilms were then pelleted by centrifugation, fixed with glutaraldehyde
were fixed with methanol and stained with 0.1% crystal violet solution (Phygene, Fuzhou, China), and stained with osmium
for 20 min; then the dye bound to the cells was resolubilized tetroxide (OsO4 ). Morphological and ultrastructural changes
with ethanol, and the absorbance of the solution was evaluated in the microorganisms were observed and photographed
at 570 nm using a microplate reader (Abouelhassan et al., using scanning electron microscopy (SEM; SU8020, Hitachi,
2017). The minimal biofilm inhibitory concentration (MBIC) Tokyo, Japan) and transmission electron microscopy (TEM;
was defined as the lowest concentration of each peptide that JEM-1200EX, JEOL, Tokyo, Japan).
exhibited apparent biofilm inhibition, while the minimal biofilm
eradication concentration (MBEC) was defined as the lowest Confocal Microscopy
peptide concentration that resulted in biofilm eradication. Based on the electron microscopic findings and to further
investigate the mechanism of the bacteriostatic activity of AMPs
on C. albicans, fluorescence-labeled peptides were employed
7
http://swissmodel.expasy.org to localize potential targets of the synthetic AMPs. Briefly,

Frontiers in Microbiology | www.frontiersin.org 3 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

C. albicans cultures were treated with FITC-labeled AMPs data. Alignment of these sequences with AMPs from RefSeq and
at 4× the corresponding MIC value for 1 h and were Pfam databases resulted in 183 computational translated peptides
then sequentially stained with propidium iodide (PI) and 40 , being annotated as AMPs. Thirty-six of these deduced peptides,
6-diamidino-2-phenylindole (DAPI). Images were acquired with including 28 predicted cecropins, five defensins, and three
a confocal microscope (Ti2-E A1R+, Nikon, Tokyo, Japan). moricins, with high sequence similarities had their presumed
propeptide cleavage sites confirmed manually. In addition,
Circular Dichroism Assay two cecropins and one defensin identified from P. clytia and
Genomic DNA was isolated from C. albicans using a fungal A. mencius transcriptomic data were also included. This set of
genomic DNA extraction kit (Solarbio) following the instructions peptide sequences was merged with available sequences obtained
of the manufacturer. The quantity of the resulting DNA by database query and literature research (four moricins from
was assessed by NanoDrop 2,000 µ-volume spectrophotometer Danaus plexippus, Papilio polytes, Pieris rapae, and Bicyclus
(Thermo Fisher Scientific). An aliquot of the prepared DNA anynana). However, no glycine-rich peptides (such as attacins or
was then mixed with an equivalent volume of synthetic AMPs gloverins) or proline-rich peptides (such as lebocins) were found
at 4× the concentration of respective MIC value. DNA-binding in the butterfly genome datasets. Amino acid sequences of the
activity of the peptides was detected by measuring the circular putative AMPs and theoretical physicochemical properties of the
dichroism (CD) spectra in quartz cuvettes of 1-mm path deduced mature peptides are displayed in Table 1.
length and wavelength ranging from 240 to 330 nm using
a Chirascan spectrometer (Applied Photophysics, Leatherhead, Cecropins
United Kingdom) (Garbett et al., 2007). As the first discovered and one of the most intensely studied
classes of AMPs, cecropins are characterized by their cationic-
In vivo Antimicrobial Assay residue-rich and alpha-helical structures, and their potent and
In vivo antimicrobial activities of the designed AMPs were wide-spectrum antimicrobial activities against different microbes
examined in neutropenic mice following subcutaneous (Bulet et al., 2004). Cecropins consist of a signal peptide and
inoculation of 100 µl of 1.0 × 108 CFU·ml−1 E. coli, S. aureus, propeptide at the N-terminus of the mature peptide, and most
or C. albicans (Lee et al., 2005). At 12 h post-infection, 100 µl of cecropins also contain tryptophan (Trp) at the first or second
AMPs at the respective MIC value was applied by subcutaneous position and a carboxamide modification at the N-terminus of
injection each day. Mice were sacrificed on day 4, and the infected the mature peptide (Yi et al., 2014). Based on the bioinformatics
area was carefully excised and homogenized. The CFU count analysis and data from previous studies, 30 cecropins were
was determined after overnight incubation of the homogenate identified from different Papilionoidea species. These cecropins
aliquots. All animal care and experiments were performed in had an average length of 39.23 ± 2.67 amino acids and an
compliance with institutional and government guidelines and average molecular weight (MW) of 4,290.86 ± 289.78 g·mol−1 .
were approved by the Peking Union Medical College Hospital All deduced sequences were presumed to possess a Trp
Animal Care and Use Committee. residue at the N1 or N2 position, while one-third of the
sequences were predicted to have C-terminal amidation by
the presence of a glycine (Gly) residue at the C1 position.
Statistical Analysis A tetrapeptide motif AGPA that formed the hinge region between
All data represent the means ± standard deviations (SDs) the two α-helices featured consistently in almost all cecropins.
of triplicate determinations in three independent experiments. According to the alignment results, the identified cecropins could
Statistical analyses were performed by t-test or one-way analysis be subdivided into two categories that contained prominent
of variance (ANOVA) with the least significant difference (LSD) sequence differences around the Trp residue (Figures 1A,B).
post hoc test for multiple comparisons using SPSS Statistics 19.0 Phylogenetic classification further suggested that these two
software (IBM Corp., Armonk, NY, United States); p < 0.05 was subclasses of cecropins had diphyletic origin, although the result
considered statistically. also illustrated that the two clusters of peptides and other
lepidopteran cecropins could be derived from a single ancestral
sequence (Figure 2A).
RESULTS
Defensins
In silico Screening of Antimicrobial Defensins represent a class of evolutionarily ancient AMPs
Peptides that are present in nearly all multicellular organisms and play
Available genomic sequencing datasets from 32 Papilionoidea important roles in humoral defense reactions (Machado and
species of six families (one species from Hesperiidae, one from Ottolini, 2015). The predominant characteristic of defensins
Lycaenidae, 20 from Nymphalidae, four from Papilionidae, is the highly conserved motif of cysteine (Cys) residues,
four from Pieridae, and two from Riodinidae) deposited in through which intracellular disulfide bridges are formed to
the GEO database up to January 2019 were downloaded maintain the secondary structure despite differences in the
for analyses. From these sequences, 434,811 genes encoding localization of the Cys residues and the pairing patterns thereof
small peptides (<400 residues) were retrieved by Augustus [14] . In contrast to cecropins, the overall sequence similarity

prediction, despite differences in the integrity of the genomic of the defensins was not apparent, and the phylogenetic

Frontiers in Microbiology | www.frontiersin.org 4 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

TABLE 1 | List of antimicrobial peptides of butterflies.

Source species AMP sequence Physicochemical property of mature peptide

Number of amino acids MW PI

Cecropins
1 Lerema accius MKVFNVFLFVFACILALSTVAAAPEP- 39 4,236.98 11.09
RWNPFKKLERVGQNIRDGIIKAGPAVAVVGQAASIYKGK
2 L. accius MKVFNVFLFVFACILALSTVAAAPEP- 39 4,236.90 10.58
RWNPFKKLEKVGQNIRDGIIKAGPAVAVVGEAANIYKGK
3 Neruda aoede MKFTKVFFFVFACFVALSTVAAAP- 39 4,106.62 10.31
WNPFKELEKAGQRVRDAIISAGPAVQVVGQATSIIKGGN
4 Vanessa tameamea MNFAKIFFFIFACIVLTTVSGAPSP- 32 3,589.27 12.17
KWKLFKRIEKIGRNVRNGLIKAGPAIQVVGQA
5 Pieris rapae MNFGKLFFFVFACVLALSTVSAAP- 39 4,354.09 10.47
KWKIFKKIEHFGQNIRDGLIKAGPALQVVGEAATIYKGK
6 Calycopis cecrops MDFSKILFFIFASLLSLNMVAAAP- 40 4,285.84 9.75
WNPLKELERAGQRVRDAIISAGPAVDVVEKTAAIIKGGQQ
7 Papilio xuthus MKYVTIILFVFIAVVAISYVSAEPIP- 44 4,836.22 4.20
WNPFKELERAGQNIRDAIISAGPAVDVVARAQKIARGEDVDEDD
8 L. accius MKVFNVFLFVFACILALSTVAAAPEP- 45 4,959.65 11.40
RWNPFKKLERVGQNIRDGIIKAGPASRCGGGPSREHIQGKMNLSV
9 P. machaon MHYRNCKTLSGVIGAPEPR- 41 4,400.07 10.53
WNPFKKLEKVGQNIRDGIIKAGPAVEVIGQAASIVKPNQGK
10 P. machaon MACLAALSLTTASP-KWKIFKKIEKVGRNVRDGIIKAGPAVAVVGQAATVAK*-G 37 3,946.72 11.23
11 P. machaon MNFAKILFFVVACFAAFSVTSASP- 38 4,294.08 11.25
KWKLFKKIEKVGRNIRNGIIKAGPAVQVVGQASQIYKQ*-G
12 P. machaon MKYVTIILFVLVAVIAISYVSAEPIP- 44 4,864.27 4.28
WNPFKELERAGQNIRDAIISAGPAVDVVARAQKIARGEDVDEEE
13 Papilio xuthus MNFGKILFFVMACLAALSLTTASP- 37 4,002.78 11.62
RWKIFKKIEKVGRNVRDGIIKAGPAVAVVGQAATVVK*-G
14 Papilio polytes MNFAKILFLVVACFAAFSVTSASP- 40 4,286.10 11.23
KWKIFKKIERVGQNIRDGIIKAGPAVAVVGQAASIIKPGK
15 Danaus plexippus MNYKRIFFSLLSILLISMVASSPAP- 40 4,339.16 10.90
KWKPFKKLEKIGQRVRDGIIKAGPAVQVVGEAAAILKPAQ*-G
16 D. plexippus MNFFRLLFFVALAVMVLSGVSASPSP- 39 4,269.07 11.28
RWKFLKKIEKVGRKVRDGVIKAGPAVGVVGQATSIYKGK
17 D. plexippus MDFSKIFFFVFACFLALSNVSAAPSP- 37 4,033.79 11.23
KWKIFKKIEKVGRNVRDGIIKAGPAVQVVGQATSIAK*-G
18 D. plexippus MKFGKLLFFVFACIMAFSTVSGAPSP- 37 4,094.87 10.97
KWKFFKKIEKVGRNIRDGIIKAGPAVQVLGEAKAIGK
19 P. rapae MNFGELYFLIFACVLALSSVSAAP- 37 4,136.89 10.59
KWKIFKKIEHMGQNIRDGLIKAGPAVQVVGQAATIYK*-G
20 V. tameamea MNFAKIFFFIFACIVLTTVSGAPSP- 37 4,074.89 11.62
KWKLFKRIEKLGQRVRDGIIKAGPAVGVIGQASTIIK*-G
21 P. rapae MNFGKLFLFVFACVLALSSVSAAP- 39 4,323.10 10.69
KWKIFKKIEHMGQNIRDGLIKAGPAVQVVGQAATIYKGK
22 P. rapae MNFGKLFFFVFACVLALSTVSAAP- 39 4,324.08 10.47
KWKIFKKIEHMGQNIRDGLIKAGPAVQVVGEAATIYKGK
23 D. plexippus MKFFNLFTFVFACFMVLGLATAAP- 39 4,178.73 10.16
WNPFKELEKAGQRVRDAIISAAPAVEVVGQASSILKGKN
24 Papilio xuthus MKYVTIILFVFIAVVAISYVSAEPIP- 44 4,850.25 4.24
WNPFKELERAGQNIRDAIISAGPAVDVVARAQKIARGEDVDEDE
25 Papilio xuthus MNFAKILFFVVACFAAFSVTSASP- 38 4,279.11 11.25
KWKLFKKIEKVGRNIRNGIIKAGPAVQVVGQASQIYKL*-G
26 Papilio xuthus MNFNKILSFAFVLFAALSSVIAAPEP- 40 4,284.00 11.58
RWNPFKKLERVGQNIRDGIIKAGPAVAVVGQAASIIKPGK
27 Papilio xuthus MNFNIILCFIIVFFTSLSGVIGAPEP- 42 4,527.26 10.90
KWNPFKKLEKVGQNIRDGIIKAGPAVQVIGQAASIVKPNQGK
28 Papilio xuthus MNFGKILFFVMACLAALSLTTASP- 37 4,074.84 11.14
RWKIFKKIEKVGRNVRDGIIKAGPAVAVVEQAATVVK*-G
29 Papilio clytia MNFAKILFFVVACFAAFSVTSASP- 38 4,308.11 11.16
RWKLFKKIEKVGRNIRDGIIKAGPAVQVVGQASQIYKL*-G
30 Atrophaneura mencius MNFNRIMSFLFVFFVAICAVSGAPEP- 40 4,227.94 11.16
RWNPFKKLEKVGQNIRDGIVKAGPAVGVIGQAASIVKPGK
Defensins
1 Calycopis cecrops MARSYQSMLLLVCISFLVIASAPQNGVAA- 44 4,797.29 8.02
DKLIGSC1 VWGAVNYTSNC2 NAEC3 KRRGYKGGHC4 GSFANVNC5 WC6 ET
(Continued)

Frontiers in Microbiology | www.frontiersin.org 5 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

TABLE 1 | Continued

Source species AMP sequence Physicochemical property of mature peptide

Number of amino acids MW PI

2 L. accius MVKSYRSVLLLVCVTFLVIVSSPRNEVAA- 44 4,826.28 7.63

DKLIGSC1 VWGAVNYTSDC2 NAEC3 KRRGYRGGHC4 GSFANVNC5 WC6 ET
3 Phoebis sennae MVQSYRSMLLLVCVSFLVIVSSPSNSAAA- 44 4,855.36 8.01
DKLIGSC1 VWGAVNYTSDC2 NKEC3 KRRGYKGGHC4 GSFANVNC5 WC6 ET
4 Papilio polytes MKGRVIIFVVLIGLAVVAAAEVEEYNESSLVTRLKRETIMVKPPK- 30 3,330.83 8.01
GC1 VFYEC2 IARC3 RQRGYLSGGYC4 TINGC5 QC6 L*-G
5 Papilio memnon MKGRVIIFVILIGLAVAAAAEVEQSKESSLVSRLKRETIMMKLPK- 30 3,330.83 8.01
GC1 VFYEC2 IARC3 RQRGYLSGGYC4 TINGC5 QC6 L*-G
6 P. clytia MKGHVIFFFVFLIGLTYVTSVALKDIRHPSFLTRLKRETIMVKPPK- 30 3,318.82 8.02
GC1 IFYEC2 IARC3 RQRGHLSGGYC4 TINGC5 QC6 L*-G
Moricins
1 D. plexippus MKLIAVLLVMLCLMSVFDTLEASP- 42 4,543.4 12.26
ARIPIGAIRKGAKAVGKGLRAINIAGTVHDIVEVFKPRKRKH
2 Calycopis cecrops MKIYGLFLIVISILALLVAPNEAKP- 41 4,465.24 11.01
GKIPIGAIKKGAKLVGKGLKALNIASTANDVYHFFHHKRKH
3 Calycopis cecrops MKFFGLFLVVLSLLALLASPSEARP- 41 4,415.17 10.65
GKIPIGAIKKGAEVVGKGLKALNIASTANDVYKFFHHKKKH
4 D. plexippus MKFFSVFVVIVTVLAVFLGTGEARP- 42 4,511.27 12.19
GKIPINAIRKGAKAVGHGLRALNIASTAHDIVSAFKHKKRKH
5 Papilio polytes MDFTKMFVLLFGILAIFMGRCNAKP- 39 4,174.75 11.55
GIPIGAIKKGGQWIRKGFGVLSAAGTAHEVYSHVKNRRN
6 P. rapae MDFRKIFLLIVLSVFAIFGSEARP- 42 4,499.34 10.90
GKIPKAVIKKGAKLVGNGLKALNVASTVHDIYSALHHKKKKH
7 Bicyclus anynana MKFTSLLILILGVFSLFIGASDARP- 41 4,615.47 12.04
KIPINAIRKGARAVGKGLRMINYASTAHDIASMFHKKKRKH
Amino acid sequences of all deduced butterfly antimicrobial peptides are shown above, with larger font size representing mature peptides. Asterisk (*) refers to removal of
glycine and subsequent amidation of the penultimate residue at the C-terminus. Superscript numerals (1–6) in defensin indicate the order of cysteine residues that form
three pairs (1–4, 2–5, and 3–6) of disulfide bridges. AMP, antimicrobial peptide; MW, molecular weight; PI, isoelectric point.

analysis revealed complex relationships among diverse organisms Moricins

(Figure 2B). However, the amino acid sequences of the Moricins are a class of highly positively charged AMPs that are
deduced defensins in the Papilionoidea superfamily were highly broadly effective against diverse microorganisms and feature a
conserved. For example, three transcripts from Calycopis long α-helix secondary structure and a cluster of basic amino
cecrops were predicted to differ by nucleotide sequences and acid residues in the C-terminal region (Hemmi et al., 2002).
corresponding precursor peptides but shared the same mature Moricin peptides, initially isolated from the hemolymph of
peptide sequences, which were also identical to those of a Bombyx mori larva under bacterial challenge and later discovered
defensin from Archaeoprepona demophon (Landon et al., 2004). from several other moth species, are considered to only be
This suggested that certain sequences might be preferable expressed in members of the order Lepidoptera (Xu et al., 2019).
to the host defense system of butterflies. After removal of Although limited moricin sequences of butterfly origin have been
duplicated sequences, six unique defensins were identified reported, the current study identified a total of seven moricins
from the genomic and transcriptomic data (Table 1) and by genomic analysis and database retrieval. These moricin
could be categorized into two subgroups according to the sequences had an average amino acid length of 41.14 ± 1.07,
alignment results and theoretical physicochemical parameters an MW of 4,460.66 ± 140.66 g·mol−1 , and a calculated mean
(Figures 1C,D). Sequences of the first cluster of butterfly isoelectric point of 11.51 ± 0.67, which was greater than that
defensins contained 44 residues and had an average MW of the cecropins (10.32 ± 2.12) and defensins (7.95 ± 0.16).
of 4,826.31 ± 29.04 g·mol−1 , while sequences of the other Despite the limited number of published moricin sequences, the
subgroup were composed of 30 amino acids and had a multiple sequence alignment and phylogenetic analyses revealed
mean MW of 3,326.83 ± 6.93 g·mol−1 . The Cys motifs in significant sequence similarity and a monophyletic origin among
the two subgroups were C-X10 -C-X3 -C-X9 -C-X7 -C-X1 -C and all naturally occurring moricins (Figure 2C). The amphipathic
C-X4 -C-X3 -C-X10 -C-X4 -C-X1 -C, which were within the range segment of the α-helix at the N-terminus of the moricins is
of reported Cys patterns of invertebrate defensins (Seufi et al., critical for the antibacterial activities of these peptides, but the
2011). The distinct Cys motifs and primary sequences of the two function of the positively charged segment at the C-terminus
subgroups of butterfly defensins also reflected the far molecular remains unknown (Hemmi et al., 2002; Oizumi et al., 2005).
distance in the phylogenetic analysis, whereas the high sequence However, amino acid frequency analysis at the relative region
similarity within each respective cluster indicated that a tight in the current study suggested that net positive charge rather
phylogenetic relationship exists between defensins in the same than site-specific preference could control the activities of
subgroup (Figure 2B). moricins (Figure 1E).

Frontiers in Microbiology | www.frontiersin.org 6 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

FIGURE 1 | Sequence alignment and logo analysis of naturally occurring AMPs from butterflies. The sequence differences therein suggested that both cecropins
and defensins could be further subclassed into two categories, respectively. The peptidyl sequence similarities and site-specific preference of (A) cecropins subclass
A; (B) cecropins subclass B; (C) defensins subclass A; (D) defensins subclass B; and (E) moricins were analyzed and displayed as graphics.

FIGURE 2 | Phylogenetic classification and visualization of butterfly AMPs. The analysis was conducted based on the amino acid sequences of cecropins (A),
defensins (B), and moricins (C), among butterfly AMPs and their homologous sequences.

Frontiers in Microbiology | www.frontiersin.org 7 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

Design of Butterfly Antimicrobial Peptide defensins displayed visible antibiofilm activity against C. albicans.
Analogs In contrast, MorA displayed the greatest antibiofilm activities
among all the designed AMPs (Table 4).
Analysis of the sequence conservation and amino acid frequency
at particular positions facilitated the design of a series of peptides
with antimicrobial potential, including two cecropins [cecropin Membrane Permeabilization Assay
A (CecA) and cecropin B (CecB)], two defensins [defensin The membrane disruption effects of the synthesized peptides
A (DefA) and defensin B (DefB)], and one moricin [moricin on the representative microorganisms were analyzed using a
A (MorA)], corresponding to aforementioned butterfly AMP SYTOX Green uptake assay as well as SEM and TEM. SYTOX
classifications. The amino acid length, theoretical MW, and Green, a membrane integrity stain, is unable to enter intact
isoelectric point of all designed peptides were within the range of cell membranes; hence, the fluorescence intensity from this
corresponding parameters analyzed (Table 2 and Supplementary stain is positively related to membrane permeability. SYTOX
Table 3). The homology modeling results displayed classical Green staining indicated significant alteration of cell membrane
secondary structural characteristics of cecropins, defensins, and permeability in the assayed microorganisms exposed to the
moricins, which were in accordance with the corresponding synthetic AMPs (Figure 4). Observations of both Gram-negative
conservative structural features described above (Figure 3). and Gram-positive bacteria by electron microscopy revealed
Following this in silico analysis, the peptides were chemically distinct morphological and biophysical changes induced by all
synthesized for further evaluation (Supplementary Figure 2). of the synthetic AMPs (Figure 5). These observations were
indicative of a typical mechanism of AMPs binding to and
disrupting the cell membranes, resulting in leakage of cell
In vitro Antimicrobial Assay contents. However, images of C. albicans showed only a slightly
In vitro antimicrobial activities of the peptides were examined by changed cell morphology and a relatively dense internal structure,
determination of the MIC and MBC values against representative which indicated that an alternative mechanism might be involved
microorganisms (E. coli, S. aureus, and C. albicans) of public in the antifungal activities of the synthetic AMPs (Figure 6).
concern using microtiter broth dilution assays. Despite different
biological effects among the AMPs, the synthesized cecropins, In vitro Peptide Localization Assay
defensins, and moricins all exhibited potent bactericidal actives To further elucidate the underlying antifungal mechanism of
against both Gram-positive and Gram-negative bacteria, with the designed AMPs, FITC-labeled peptides were synthesized
MIC and MBC values in micromolar and sub-micromolar and used for localization of the potential binding targets of
concentration ranges; cecropins were the most effective agents. In the synthetic AMPs by confocal fluorescence microscopy. The
addition, while cecropins exhibited modest antifungal effects, the synthetic cecropins and moricin targeted the DAPI-stained
synthetic moricin possessed relatively strong fungicidal activities, nucleus of C. albicans (Figure 6). In addition, the FITC
with low micromolar MIC and MBC values (Table 3). No fluorescence showed strong colocalization with the PI signal.
apparent hemolysis was detected at concentrations exceeding the DNA-binding activities of the synthetic AMPs were verified
respective MIC and MBC values (Supplementary Figure 3). by CD spectroscopy, which indicated detectable DNA
structural changes (Supplementary Figure 4). Therefore, it
Antibiofilm Assay was hypothesized that the cationic AMPs exert their antifungal
The biofilm inhibitive and degradative effects of the synthetic activities by binding to the anionic nucleic acids, thus interfering
peptides were quantitated by measuring the MBIC and MBEC with bioprocesses and resulting in cell death.
values against E. coli, S. aureus, and C. albicans, which represent
Gram-negative bacteria, Gram-positive bacteria, and fungi, In vivo Antimicrobial Assay
respectively. All the synthetic peptides prevented the formation Finally, the in vivo antimicrobial activities of the most promising
of biofilms of both Gram-negative and Gram-positive bacteria AMPs against bacteria (CecB) and fungi (MorA) were evaluated
and eliminated pre-formed biofilms of both bacterial species. in a skin infection mouse model. CFU counts for the bacteria
Congruent with the results of the antifungal assays, neither of the and fungus in the presence of the tested AMPs were significantly

TABLE 2 | Amino acid sequences and physicochemical properties of the designed peptides.

AMPs Family Sequence Physicochemical property

Number of residues MW PI

CecA Cecropin RWNPFKKLERVGQNIRDGIIKAGPAVAVVGQAASIAK-NH2 37 3,958.62 11.93

CecB Cecropin KWKIFKKIEKVGRNIRDGIIKAGPAVQVVGQAATIYKGK 39 4,310.14 11.12
DefA Defensin DKLIGSCVWGAVNYTSDCNAECKRRGYKGGHCGSFANVNCWCET 44 4,804.36 7.45
DefB Defensin GCVFYECIARCRQRGYLSGGYCTINGCQCL-NH2 30 3,336.91 8.12
MorA Moricin GKIPIGAIKKGAKAVGKGLKALNIASTAHDIYSFHHKKKKH 41 4,363.17 11.06

AMP, antimicrobial peptide; MW, molecular weight; PI, isoelectric point.

Frontiers in Microbiology | www.frontiersin.org 8 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

FIGURE 3 | Secondary structures of the designed peptide. The three-dimensional shapes of the synthetic molecules, e.g., (A) cecropin A, (B) cecropin B, (C)
denfensin A, (D) denfensin B, and (E) moricin A, were constructed by homology modeling, which demonstrated typical structural features of respective classes of
antimicrobial peptides (AMPs).

TABLE 3 | In vitro antimicrobial activities of the designed peptides.

AMPs Escherichia coli Staphylococcus aureus Candida albicans

MIC MBC MIC MBC MIC MBC

CecA 0.5 2 1 4 16 32
CecB 0.5 1 1 2 8 16
DefA 8 32 1 16 ND ND
DefB 2 16 0.5 16 ND ND
MorA 1 2 2 4 1 2
Cec(Ctrl) 0.5 1 1 2 16 32
Def(Ctrl) 1 32 8 32 ND ND
Mor(Ctrl) 1 2 2 4 2 4

Hyalophora cecropia cecropin A [Cec(Ctrl)], Galleria mellonella defensin [Def(Ctrl)], and Manduca sexta moricin [Mor(Ctrl)] served as respective controls. The results of
antimicrobial activities evaluation are presented as minimal inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) value (µmol·L−1 ) against different
microorganisms, while ND stands for not detected.
AMP, antimicrobial peptide.

less than those of the respective controls. This demonstrated that suppress proliferation of C. albicans in vivo (Figure 7). These
the synthetic cecropin could attenuate growth of both E. coli and findings suggested that CecB and MorA may have therapeutic
S. aureus in the animal model, while the synthetic moricin could potential under different infectious circumstances.

Frontiers in Microbiology | www.frontiersin.org 9 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

TABLE 4 | Antibiofilm activities of the designed peptides.

AMPs Escherichia coli Staphylococcus aureus Candida albicans

MBIC MBEC MBIC MBEC MBIC MBEC

CecA 2 8 2 16 16 32
CecB 2 8 2 16 16 32
DefA 8 32 4 16 ND ND
DefB 8 32 2 16 ND ND
MorA 4 8 4 16 4 16
Cec(Ctrl) 2 8 2 16 16 32
Def(Ctrl) 8 32 4 16 ND ND
Mor(Ctrl) 4 8 4 16 4 16

Hyalophora cecropia cecropin A [Cec(Ctrl)], Galleria mellonella defensin [Def(Ctrl)], and Manduca sexta moricin [Mor(Ctrl)] served as respective controls. The results
of antibiofilm activities evaluation are presented as the minimal biofilm inhibition concentration (MBIC) and the minimal biofilm eradicating concentration (MBEC) value
(µmol·L−1 ) against different microorganisms, while ND stands for not detected.
AMP, antimicrobial peptide.

FIGURE 4 | Evaluation of membrane destruction effects of antimicrobial peptides (AMPs) by SYTOX Green uptake assay. Results demonstrated the synthetic
peptides induced significant membrane permeability changes against different microorganisms.

DISCUSSION of naturally occurring AMPs have been recognized from various

lepidopteran insects (moths and butterflies) insects, although
Numerous AMPs have been identified from natural resources, only a few are derived from Papilionoidea species (butterflies).
and many more have been synthetically created by modification Butterfly species could therefore be a hidden treasure for
and ab initio design based on sequence and structural features natural AMP discoveries and may provide potential templates
of the naturally occurring molecules (Hale and Hancock, 2007). for peptide design. The availability of an increasing number of
Many AMPs are currently in preclinical or clinical development high-throughput sequencing datasets from butterflies facilitated
for the treatment of various infectious diseases, and most of the design of an integrated bioinformatics analysis method for
these AMPs were of natural origin (Greber and Dawgul, 2017). the efficient discovery of naturally occurring AMPs.
Conventionally, naturally occurring AMPs are predominantly To process large-scale genome data, many computational
identified by cDNA cloning and sequencing and/or de novo tools have been developed that can perform sequence assembly
peptide sequencing. Moreover, by taking advantage of precursor and gene prediction and identification. In the context of the
sequence conservation of known AMPs within Pisces and current study and identifying AMPs, protein-encoding portions
Amphibia (fishes and amphibians), a more productive shotgun of transcripts are deducible by ORF analysis, which can be
cloning approach and strategies for database retrieval have accomplished by using software such as ORFfinder. Next, the
been developed to facilitate the identification of homologous signal peptide of propeptides can be predicted by calculation of
sequences (Wang et al., 2012; Liu et al., 2015). However, the potential cleavage sites, which is achieved by using a program
precursor sequence-based methodologies are not applicable for such as SignalP. Furthermore, despite the lack of software
screening butterfly AMPs, since such sequence conservation has available to analyze other post-translational modifications, the
not been revealed among butterfly species. Nevertheless, dozens mature peptide can still be identified by homologous sequence

Frontiers in Microbiology | www.frontiersin.org 10 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

FIGURE 5 | Electron microscopic analysis of bacteria treated with synthetic antimicrobial peptides (AMPs). Representative images of both scanning electron
microscopy (SEM; upper panels) and transmission electron microscopy (TEM; lower panels) showed significant changes on cell morphology induced by AMPs at 4×
of corresponding minimum inhibitory concentration (MIC) value, compared with the intact smooth surfaces of the untreated control group (left panel). The TEM assay
further displayed marked decreases in cell densities that resulted from peptide intervention. These findings suggest membrane permeabilization mechanism being
involved in bactericidal activities of the synthesized butterfly AMPs.

alignment and prediction of cleavage sites (Wang and Hu, AMPs unique to the order Lepidoptera is the lebocins; however,
2017). In general, the N-terminus of a pro-AMP undergoes a no butterfly lebocin sequences were found in the current
two-step proteolytic cleavage by signal peptidase and dipeptidyl study. Phylogenetic analyses revealed a monophyletic origin
aminopeptidase (Boman et al., 1989), while peptides with a Gly within butterfly cecropins and moricins, which is yet another
residue present at the C-terminus are subjected to an amide demonstration of the monophyletic origin of the superfamily
modification by a two-step sequential catalytic reaction with Papilionoidea. The molecular distances among defensins of
peptidylglycine α-amidating monooxygenase (PAM) (Kim and butterfly origin were relatively far, which was expected since it has
Seong, 2001). Based on this bioinformatics analysis method, been proposed that higher variabilities of insect defensins exist
a total of 36 new, naturally occurring AMPs of butterfly among closely related species instead of among distantly related
origin were identified from the published genomic data used insects (Bulet et al., 2004).
in the current study. This was substantially greater than the The molecular design of the synthetic peptides in this study
number of previously reported, naturally occurring butterfly was generally based on the mimicry of natural AMPs. The
AMPs. In addition, the designed peptides, which acted as pronounced sequence and structural features, e.g., α-helical
representatives of the discovered natural AMPs, exhibited structures, C-terminal amidation, disulfide bridges, and the
potent and wide-spectrum antimicrobial effects with typical positively charged C-terminal segment, were included in the
membrane-disruption activities. Therefore, the genome-wide synthetic peptide design (Table 1 and Figure 3). However,
in silico screening is suggested to be an effective and practical due to the possibility of producing the designed AMPs by
method for identification of naturally occurring AMPs. microorganism expression systems in the future, more complex
The findings from this study add to previous discoveries, chemical modification was not considered in this research. It is
and it is our understanding that three classes of AMPs recognized that the positively charged amphipathic α-helix of
have been found from Papilionoidea species; these classes cecropins plays a dominant role in their antimicrobial activities,
are cecropins, defensins, and moricins. Of the three classes, allowing them to partition into the phospholipid bilayers and
cecropins are predominantly expressed by lepidopterans and disrupt microbial membranes (Hancock and Rozek, 2002). The
dipterans and are primarily responsible for the host defense of C-terminal hydrophobic helix of cecropins is believed to have a
lepidopterans. Although α-helical AMPs nominated as cecropins synergistic effect and provide selectivity against Gram-negative
have been identified from other species, the sequences of these bacteria (Lee et al., 2015). Moricins, which possess an N-terminal
molecules exhibit a substantially lower degree of similarity amphipathic segment in the one large α-helix structure, were
to insect cecropins. Defensins have rarely been discovered presumed to share a similar mechanism as that of the cecropins.
from moths or butterflies but are ubiquitously expressed The amphipathicity of the α-helices of the designed peptides
throughout multicellular organisms. In contrast, moricins have was visualized via helical wheel projection (Supplementary
been identified only in Lepidoptera species. Another class of Figure 5). Furthermore, C-terminal amidation is suggested to

Frontiers in Microbiology | www.frontiersin.org 11 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

FIGURE 6 | Microscopic investigation of antifungal capabilities of designed antimicrobial peptides (AMPs). The antifungal effects of cecropins and moricins, which
had been confirmed by microtiter broth dilution assays (Table 3), were analyzed by combination of scanning electron microscopy (SEM; first upper panel),
transmission electron microscopy (TEM; second upper panel), and confocal fluorescence microscopies (lower panels). Although there were no apparent
morphological changes observed by electron microscopies, the confocal microscopy demonstrated the fluorescein isothiocyanate (FITC) (green) labeled peptides
colocalized with DAPI (blue)-stained fungus nucleus, and the FITC fluorophore showed high spatial overlaps with propidium iodide (PI) (red) fluorescence, which
indicates a probable cell death induction mechanism caused by DNA binding.

Frontiers in Microbiology | www.frontiersin.org 12 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

be a common modification of the cecropins and is presumed

to contribute to the broad-spectrum antimicrobial activity of
these peptides (Yi et al., 2014). Among the previously discovered
butterfly cecropins, papiliocin from Papilio xuthus (Kim et al.,
2010) and DAN1 and DAN2 from D. plexippus (Duwadi et al.,
2018) were proposed to have amidated C-termini, although a
non-Gly amino acid as the ultimate residue in DAN2 indicated
a natural, non-amidated terminus. On the contrary, hinnavin I
and hinnavin II (Yoe et al., 2006) from P. rapae were reported
to have carboxyl C-termini, even if the presence of a Gly residue
at the C-terminus of hinnavin II suggested a potential amidated
modification. There were no apparent C-terminal amidation FIGURE 7 | Evaluation of in vivo antimicrobial effects of synthetic antimicrobial
tendencies within the naturally occurring butterfly cecropins peptides (AMPs) on various bacteria and fungi in a skin infection mice model.
The most effective antibacterial (CecB against Escherichia coli and
examined in the current study (Table 1). Moreover, antimicrobial
Staphylococcus aureus) and antifungal (MorA against Candida albicans)
activities of the amidated analog of CecB were not augmented agents in the in vitro assays were examined in this study, and the antimicrobial
when compared with the natural prototype, which suggested activities were expressed as CFU per skin sample. Statistical analyses were
that C-terminal amidation may not be required for cecropins performed by t-test, and asterisks indicate significant difference (∗ p ≤ 0.05,
∗∗ p ≤ 0.01).
derived from butterflies. A C-terminal amidation modification
has also been observed in a group of natural butterfly defensins,
and this modification was relatively rare among other species
(Bruhn et al., 2009). C-terminal amidation could be postulated Gram-negative bacteria (Wang et al., 2017). In accordance with
to neutralize the negative charge and increase resistance to this, both SEM and TEM illustrated that the designed peptides
exopeptidase hydrolyzation. Further studies are required to in the current study could exert their bactericidal activities
elucidate the underlying functions of C-terminal amidation of through a mechanism of membrane binding and disruption.
AMPs. A more conserved character of the defensins is the Furthermore, the synthetic cecropins and moricin, especially
Cys-stabilized αβ (CSαβ) motif signature, which is considered as the latter, also demonstrated antifungal potential through a
the main functional structural basis, despite certain mammalian possible DNA-binding mechanism. The in vivo experiment
defensins lacking α-helices (de Oliveira Dias and Franco, 2015). provided further evidence that the designed peptide CecB
Moreover, the positively charged segment at the C-termini of had important therapeutic functions in bacterial elimination,
moricins is the most distinct feature distinguishing them from while also showing that the synthetic MorA could inhibit
other α-helical AMPs. Although tetra- and penta-peptide motifs fungal growth to some extent (Figure 7). These preliminary
at the positively charged part of moricin have been proposed findings suggest that the synthetic AMPs, primarily CecB and
in other species (Hemmi et al., 2002; Oizumi et al., 2005), the MorA, are promising antimicrobial agents that warrant further
butterfly moricin sequences showed diverse residual propensities investigation. All the synthetic sequences generated in the current
in this region. Combined with the confocal microscopy and study demonstrated potent antimicrobial activities, which further
CD spectroscopy results, which demonstrated that moricins supports the rationality and necessity of selecting sequences and
could penetrate the cell membrane and bind to the cell nucleus structures that have been phylogenetically conserved through
(Figure 6 and Supplementary Figure 3), it could be arguably evolution. The characteristics of natural butterfly AMPs could
hypothesized that the basic amino acid cluster significantly be explored further through more extensive sequence and
increased the net charge of moricins and hence facilitated the structural optimization as well as considering the possibility
interaction with the negatively charged nucleic acids. of peptide expression. Furthermore, the lower antimicrobial
AMPs are active against a broad range of microorganisms, activity of defensins does not mean that these AMPs are less
but selectivity is also evident toward respective microbes. The important; there is accumulating evidence for the existence of
activity of cecropins is thought to be particularly directed against immunoregulatory properties of defensins on host immunity
Gram-negative bacteria, while that of defensins and moricins and resultant therapeutic potential, as well as the occurrence
is mainly targeted toward Gram-positive bacteria (Hara and of functional synergies between a defensin and other AMPs
Yamakawa, 1995). However, the in vitro antimicrobial assay in or antimicrobial agents (Pazgier et al., 2007; Lazzaro et al.,
the current study revealed that all designed peptides, particularly 2020). Examination of these theories was beyond the scope of
cecropins, exerted potent but generally undifferentiated activities the current study.
against Gram-negative and Gram-positive bacteria, with The limited numbers of representative microorganisms
cecropins and moricin exhibiting slightly higher antimicrobial applied in the current study mean the antimicrobial spectrum
preference against Gram-negative bacteria and defensins showing and bioactivities of the AMPs require further evaluation to
a stronger effect on Gram-positive bacteria (Table 3). It was not obtain a more detailed characterization of these molecules.
surprising that the designed moricin exhibited predominant For example, AMPs have been found that can target and
antimicrobial activity against Gram-negative bacteria, since the lyse phosphatidylserine (PS), thereby exposing negatively
high amphipathicity and net charge of this class of AMP are charged tumor cells, and these AMPs are even active against
thought to be positively associated with the activities toward drug-resistant tumor cell variants and may not affect their healthy

Frontiers in Microbiology | www.frontiersin.org 13 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

counterparts (Riedl et al., 2011). In addition, some AMPs were experiment. MW prepared the draft manuscript. XH reviewed
reported to inhibit and degrade biofilm formation by coating the draft manuscript. All authors read and approved the
the bacterium or the biomaterial surface (Segev-Zarko et al., final manuscript.
2015); however, the mechanism of antibiofilm activity was not
investigated in this study. AMPs have also demonstrated antiviral
effects on different viruses, ranging from influenza to human FUNDING
immunodeficiency virus (HIV), through multiple mechanisms
(Vilas Boas et al., 2019). Antiviral effects of the synthetic AMPs This work was funded by the National Natural Science
were not explored in the current study. Despite these limitations, Foundation of China (Nos. 81801635 and 81803848) and
the current study provides valuable information on butterfly China Postdoctoral Science Foundation (No. 2017M610812). The
AMPs and an efficient and practical methodology for AMP funders had no role in study design, data collection and analysis,
screening and design. decision to publish, or preparation of the manuscript.

CONCLUSION SUPPLEMENTARY MATERIAL

This study presented an overview of the landscape of The Supplementary Material for this article can be found
butterfly AMPs and insights into their diverse sequence and online at: https://www.frontiersin.org/articles/10.3389/fmicb.
structural features. Based on these sequences, a series of potent 2021.720381/full#supplementary-material
and wide-spectrum peptides with respective antimicrobial
Supplementary Figure 1 | Schematic illustration of the experimental design.
mechanisms were successfully designed. This study not only
provided a practical strategy for high-throughput natural AMP Supplementary Figure 2 | HPLC profiles and spectrometric data of
the synthetic AMPs.
discoveries but also produced a rational methodology for AMP
design, and this will be of importance for future studies. Supplementary Figure 3 | Evaluation of hemolytic and cytotoxic activities of
synthetic AMPs. The hemolysis effects (left) and cytotoxicity on normal cells (right)
of the synthetic peptides at different concentration were tested on mice
erythrocytes and HEK293 cells, respectively. PBS and Triton-X100 were used as
DATA AVAILABILITY STATEMENT the negative and positive control. No apparent hemolysis or cytotoxicity was
detected among groups.
The datasets presented in this study can be found in
Supplementary Figure 4 | CD spectrum of the designed AMPs interacted with
online repositories. The names of the repository/repositories fungal genomic DNA. The top curve corresponded to free DNA, while the lower
and accession number(s) can be found in the article/ curves were that mixed with respective AMPs. The decreased intensities of
Supplementary Material. genomic DNA molar ellipticity exhibited occurrence conformational alteration of
DNA that induced by AMPs, which suggested a DNA-binding mechanism of the
designed sequences.

ETHICS STATEMENT Supplementary Figure 5 | Helical wheel projection analysis of the designed
peptide. (A) N-terminal amphiphilic helix of cecropin A; (B) C-terminal hydrophobic
The animal study was reviewed and approved by Peking Union helix of cecropin A; (C) N-terminal amphiphilic helix of cecropin B; (D) C-terminal
hydrophobic helix of cecropin B; (E) helical segment of defensin A; (F) helical
Medical College Hospital Animal Care and Use Committee, segment of defensin B; (G) helical segment of moricin A.
Peking Union Medical College Hospital.
Supplementary Table 1 | Papilionoidea species and their high-throughput
sequencing data used in this study.

AUTHOR CONTRIBUTIONS Supplementary Table 2 | The AMP sequences for phylogenetic

analysis in this study.
MW and XH conceived the study. WZ, SL, and XH performed Supplementary Table 3 | Sequence alignment of the synthetic and
the in silico experiment. MW and ZZ carried out the wet lab naturally occurring AMPs.

REFERENCES Boman, H. G., Boman, I. A., Andreu, D., Li, Z. Q., Merrifield, R. B., Schlenstedt, G.,
et al. (1989). Chemical synthesis and enzymic processing of precursor forms of
Abouelhassan, Y., Yang, Q., Yousaf, H., Nguyen, M. T., Rolfe, M., Schultz, G. S., cecropins A and B. J. Biol. Chem. 264, 5852–5860. doi: 10.1016/S0021-9258(18)
et al. (2017). Nitroxoline: a broad-spectrum biofilm-eradicating agent against 83628-6
pathogenic bacteria. Int. J. Antimicrob. Agents 49, 247–251. doi: 10.1016/j. Bruhn, O., Paul, S., Tetens, J., and Thaller, G. (2009). The repertoire of equine
ijantimicag.2016.10.017 intestinal alpha-defensins. BMC Genomics 10:631. doi: 10.1186/1471-2164-10-
Badapanda, C., and Chikara, S. K. (2016). “Lepidopteran antimicrobial peptides 631
(AMPs): overview, regulation, modes of action, and therapeutic potentials of Bulet, P., Stöcklin, R., and Menin, L. (2004). Anti-microbial peptides: from
insect-derived AMPs,” in Short Views on Insect Genomics and Proteomics: Insect invertebrates to vertebrates. Immunol. Rev. 198, 169–184. doi: 10.1111/j.0105-
Proteomics, Vol.2 Entomology in Focus, eds C. Raman, M. R. Goldsmith, and 2896.2004.0124.x
T. A. Agunbiade (Cham: Springer International Publishing), 141–163. doi: Crooks, G. E., Hon, G., Chandonia, J.-M., and Brenner, S. E. (2004). WebLogo: a
10.1007/978-3-319-24244-6_6 sequence logo generator. Genome Res. 14, 1188–1190. doi: 10.1101/gr.849004

Frontiers in Microbiology | www.frontiersin.org 14 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

de Oliveira Dias, R., and Franco, O. L. (2015). Cysteine-stabilized αβ defensins: Lee, P. H. A., Ohtake, T., Zaiou, M., Murakami, M., Rudisill, J. A., Lin, K. H., et al.
from a common fold to antibacterial activity. Peptides 72, 64–72. doi: 10.1016/j. (2005). Expression of an additional cathelicidin antimicrobial peptide protects
peptides.2015.04.017 against bacterial skin infection. Proc Natl Acad Sci U S A 102, 3750–3755.
Duwadi, D., Shrestha, A., Yilma, B., Kozlovski, I., Sa-Eed, M., Dahal, N., doi: 10.1073/pnas.0500268102
et al. (2018). Identification and screening of potent antimicrobial peptides Lin, M.-C., Hui, C.-F., Chen, J.-Y., and Wu, J.-L. (2013). Truncated antimicrobial
in arthropod genomes. Peptides 103, 26–30. doi: 10.1016/j.peptides.2018.0 peptides from marine organisms retain anticancer activity and antibacterial
1.017 activity against multidrug-resistant Staphylococcus aureus. Peptides 44, 139–
Garbett, N. C., Ragazzon, P. A., and Chaires, J. B. (2007). Circular dichroism to 148. doi: 10.1016/j.peptides.2013.04.004
determine binding mode and affinity of ligand–DNA interactions. Nat. Protoc. Liu, H., Lei, M., Du, X., Cui, P., and Zhang, S. (2015). Identification of a novel
2, 3166–3172. doi: 10.1038/nprot.2007.475 antimicrobial peptide from amphioxus Branchiostoma japonicum by in silico
Greber, K. E., and Dawgul, M. (2017). Antimicrobial peptides under and functional analyses. Sci. Rep. 5:18355. doi: 10.1038/srep18355
clinical trials. Curr. Top. Med. Chem. 17, 620–628. doi: 10.2174/ Machado, L. R., and Ottolini, B. (2015). An evolutionary history of defensins: a role
1568026616666160713143331 for copy number variation in maximizing host innate and adaptive immune
Hale, J. D. F., and Hancock, R. E. W. (2007). Alternative mechanisms of action of responses. Front. Immunol. 6:115. doi: 10.3389/fimmu.2015.00115
cationic antimicrobial peptides on bacteria. Expert. Rev. Anti. Infect. Ther. 5, Manniello, M. D., Moretta, A., Salvia, R., Scieuzo, C., Lucchetti, D., Vogel, H.,
951–959. doi: 10.1586/14787210.5.6.951 et al. (2021). Insect antimicrobial peptides: potential weapons to counteract the
Hancock, R. E. W., and Rozek, A. (2002). Role of membranes in the activities antibiotic resistance. Cell Mol. Life Sci. 78, 4259–4282. doi: 10.1007/s00018-021-
of antimicrobial cationic peptides. FEMS Microbiol. Lett. 206, 143–149. doi: 03784-z
10.1111/j.1574-6968.2002.tb11000.x Mistry, J., Chuguransky, S., Williams, L., Qureshi, M., Salazar, G. A., Sonnhammer,
Hara, S., and Yamakawa, M. (1995). Moricin, a novel type of antibacterial peptide E. L. L., et al. (2020). Pfam: the protein families database in 2021. Nucleic Acids
isolated from the silkworm Bombyx Mori. J. Biol. Chem. 270, 29923–29927. Res. 49, D412–D419. doi: 10.1093/nar/gkaa913
doi: 10.1074/jbc.270.50.29923 O’Leary, N. A., Wright, M. W., Brister, J. R., Ciufo, S., Haddad, D., McVeigh, R.,
Hartmann, M., Berditsch, M., Hawecker, J., Ardakani, M. F., Gerthsen, D., and et al. (2016). Reference sequence (RefSeq) database at NCBI: current status,
Ulrich, A. S. (2010). Damage of the bacterial cell envelope by antimicrobial taxonomic expansion, and functional annotation. Nucleic Acids Res. 44, D733–
peptides gramicidin S and PGLa as revealed by transmission and scanning D745. doi: 10.1093/nar/gkv1189
electron microscopy. Antimicrob. Agents Chemother. 54, 3132–3142. doi: 10. Oizumi, Y., Hemmi, H., Minami, M., Asaoka, A., and Yamakawa, M. (2005).
1128/AAC.00124-10 Isolation, gene expression and solution structure of a novel moricin analogue,
Hazam, P. K., Goyal, R., and Ramakrishnan, V. (2019). Peptide based antibacterial peptide from a lepidopteran insect, Spodoptera litura. Biochim.
antimicrobials: design strategies and therapeutic potential. Prog. Biophys. Mol. Biophys. Acta 1752, 83–92. doi: 10.1016/j.bbapap.2005.07.013
Biol. 142, 10–22. doi: 10.1016/j.pbiomolbio.2018.08.006 Pazgier, M., Li, X., Lu, W., and Lubkowski, J. (2007). Human defensins: synthesis
Hemmi, H., Ishibashi, J., Hara, S., and Yamakawa, M. (2002). Solution structure and structural properties. Curr. Pharm. Des. 13, 3096–3118. doi: 10.2174/
of moricin, an antibacterial peptide, isolated from the silkworm Bombyx mori. 138161207782110381
FEBS Lett. 518, 33–38. doi: 10.1016/S0014-5793(02)02637-6 Riedl, S., Rinner, B., Asslaber, M., Schaider, H., Walzer, S., Novak, A., et al.
Jasper, H., and Bohmann, D. (2002). Drosophila innate immunity: a genomic view (2011). In search of a novel target — Phosphatidylserine exposed by non-
of pathogen defense. Mol. Cell 10, 967–969. doi: 10.1016/S1097-2765(02)00752- apoptotic tumor cells and metastases of malignancies with poor treatment
9 efficacy. Biochim. Biophys. Acta (BBA) Biomembr. 1808, 2638–2645. doi: 10.
Jin, L., Bai, X., Luan, N., Yao, H., Zhang, Z., Liu, W., et al. (2016). A designed 1016/j.bbamem.2011.07.026
tryptophan- and lysine/arginine-rich antimicrobial peptide with therapeutic Roope, L. S. J., Smith, R. D., Pouwels, K. B., Buchanan, J., Abel, L., Eibich, P.,
potential for clinical antibiotic-resistant Candida albicans vaginitis. J. Med. et al. (2019). The challenge of antimicrobial resistance: what economics can
Chem. 59, 1791–1799. doi: 10.1021/acs.jmedchem.5b01264 contribute. Science 364:eaau4679. doi: 10.1126/science.aau4679
Kawahara, A. Y., and Breinholt, J. W. (2014). Phylogenomics provides strong Segev-Zarko, L., Saar-Dover, R., Brumfeld, V., Mangoni, M. L., and Shai, Y. (2015).
evidence for relationships of butterflies and moths. Proc. R. Soc. B. Mechanisms of biofilm inhibition and degradation by antimicrobial peptides.
281:20140970. doi: 10.1098/rspb.2014.0970 Biochem. J. 468, 259–270. doi: 10.1042/BJ20141251
Kim, K.-H., and Seong, B. L. (2001). Peptide amidation: production of peptide Seufi, A. M., Hafez, E. E., and Galal, F. H. (2011). Identification, phylogenetic
hormonesin vivo andin vitro. Biotechnol. Bioprocess Eng. 6, 244–251. doi: 10. analysis and expression profile of an anionic insect defensin gene, with
1007/BF02931985 antibacterial activity, from bacterial-challenged cotton leafworm, Spodoptera
Kim, S. R., Hong, M. Y., Park, S. W., Choi, K. H., Yun, E. Y., Goo, T. W., et al. littoralis. BMC Mol. Biol. 12:47. doi: 10.1186/1471-2199-12-47
(2010). Characterization and cDNA cloning of a cecropin-like antimicrobial Stanke, M., Diekhans, M., Baertsch, R., and Haussler, D. (2008). Using native
peptide, papiliocin, from the swallowtail butterfly Papilio xuthus. Mol. Cells 29, and syntenically mapped cDNA alignments to improve de novo gene finding.
419–423. doi: 10.1007/s10059-010-0050-y Bioinformatics 24, 637–644. doi: 10.1093/bioinformatics/btn013
Kumar, S., Stecher, G., Li, M., Knyaz, C., and Tamura, K. (2018). MEGA X: Triant, D. A., Cinel, S. D., and Kawahara, A. Y. (2018). Lepidoptera genomes:
Molecular Evolutionary Genetics Analysis across Computing Platforms. Mol. current knowledge, gaps and future directions. Curr. Opin. Insect. Sci. 25,
Biol. Evol. 35, 1547–1549. doi: 10.1093/molbev/msy096 99–105. doi: 10.1016/j.cois.2017.12.004
Landon, C., Barbault, F., Legrain, M., Menin, L., Guenneugues, M., Schott, V., Unckless, R. L., and Lazzaro, B. P. (2016). The potential for adaptive maintenance
et al. (2004). Lead optimization of antifungal peptides with 3D NMR structures of diversity in insect antimicrobial peptides. Philos. Trans. R. Soc. Lond. B Biol.
analysis. Protein Sci. 13, 703–713. doi: 10.1110/ps.03404404 Sci. 371:20150291. doi: 10.1098/rstb.2015.0291
Larkin, M. A., Blackshields, G., Brown, N. P., Chenna, R., McGettigan, P. A., Vilas Boas, L. C. P., Campos, M. L., Berlanda, R. L. A., de Carvalho Neves, N., and
McWilliam, H., et al. (2007). Clustal W and Clustal X version 2.0. Bioinformatics Franco, O. L. (2019). Antiviral peptides as promising therapeutic drugs. Cell.
23, 2947–2948. doi: 10.1093/bioinformatics/btm404 Mol. Life Sci. 76, 3525–3542. doi: 10.1007/s00018-019-03138-w
Lazzaro, B. P., Zasloff, M., and Rolff, J. (2020). Antimicrobial peptides: application Wang, C.-K., Shih, L.-Y., and Chang, K. Y. (2017). Large-scale analysis of
informed by evolution. Science 368:eaau5480. doi: 10.1126/science.aau5480 antimicrobial activities in relation to amphipathicity and charge reveals novel
Lear, S., and Cobb, S. L. (2016). Pep-Calc.com: a set of web utilities for the characterization of antimicrobial peptides. Molecules 22:E2037. doi: 10.3390/
calculation of peptide and peptoid properties and automatic mass spectral peak molecules22112037
assignment. J. Comput. Aided Mol. Des. 30, 271–277. doi: 10.1007/s10822-016- Wang, G., Li, X., and Wang, Z. (2016). APD3: the antimicrobial peptide database
9902-7 as a tool for research and education. Nucleic Acids Res. 44, D1087–D1093.
Lee, E., Kim, J. K., Jeon, D., Jeong, K. W., Shin, A., and Kim, Y. (2015). Functional doi: 10.1093/nar/gkv1278
roles of aromatic residues and helices of papiliocin in its antimicrobial and Wang, M., and Hu, X. (2017). Antimicrobial peptide repertoire of Thitarodes
anti-inflammatory activities. Sci. Rep. 5:12048. doi: 10.1038/srep12048 armoricanus, a host species of Ophiocordyceps sinensis, predicted based on de

Frontiers in Microbiology | www.frontiersin.org 15 August 2021 | Volume 12 | Article 720381

Wang et al. Butterfly Antimicrobial Peptides

novo transcriptome sequencing and analysis. Infect. Genet. Evol. 54, 238–244. cabbage butterfly, Artogeia rapae. Comp. Biochem. Physiol. Part B 144, 199–205.
doi: 10.1016/j.meegid.2017.07.011 doi: 10.1016/j.cbpb.2006.02.010
Wang, M., Li, Y., and Hu, X. (2018). Chebulinic acid derived from triphala is a Zharkova, M. S., Orlov, D. S., Golubeva, O. Yu, Chakchir, O. B.,
promising antitumour agent in human colorectal carcinoma cell lines. BMC Eliseev, I. E., Grinchuk, T. M., et al. (2019). Application of
Comp. Altern. Med. 18:342. doi: 10.1186/s12906-018-2412-5 antimicrobial peptides of the innate immune system in combination
Wang, M., Wang, L., Chen, T., Walker, B., Zhou, M., Sui, D., et al. (2012). with conventional antibiotics—a novel way to combat antibiotic
Identification and molecular cloning of a novel amphibian Bowman Birk- resistance? Front. Cell Infect. Microbiol. 9:128. doi: 10.3389/fcimb.2019.0
type trypsin inhibitor from the skin of the Hejiang Odorous Frog Odorrana 0128
hejiangensis. Peptides 33, 245–250. doi: 10.1016/j.peptides.2012.01.003
Waterhouse, A., Bertoni, M., Bienert, S., Studer, G., Tauriello, G., Gumienny, R., Conflict of Interest: The authors declare that the research was conducted in the
et al. (2018). SWISS-MODEL: homology modelling of protein structures and absence of any commercial or financial relationships that could be construed as a
complexes. Nucleic Acids Res. 46, W296–W303. doi: 10.1093/nar/gky427 potential conflict of interest.
Wiegand, I., Hilpert, K., and Hancock, R. E. W. (2008). Agar and broth
dilution methods to determine the minimal inhibitory concentration (MIC) of Publisher’s Note: All claims expressed in this article are solely those of the authors
antimicrobial substances. Nat. Protoc. 3, 163–175. doi: 10.1038/nprot.2007.521 and do not necessarily represent those of their affiliated organizations, or those of
Xu, X., Zhong, A., Wang, Y., Lin, B., Li, P., Ju, W., et al. (2019). Molecular the publisher, the editors and the reviewers. Any product that may be evaluated in
identification of a moricin family antimicrobial peptide (px-mor) from this article, or claim that may be made by its manufacturer, is not guaranteed or
plutella xylostella with activities against the opportunistic human pathogen endorsed by the publisher.
Aureobasidium pullulans. Front. Microbiol. 10:2211. doi: 10.3389/fmicb.2019.
02211 Copyright © 2021 Wang, Zhou, Li, Zhu and Hu. This is an open-access article
Yi, H.-Y., Chowdhury, M., Huang, Y.-D., and Yu, X.-Q. (2014). Insect distributed under the terms of the Creative Commons Attribution License (CC BY).
antimicrobial peptides and their applications. Appl. Microbiol. Biotechnol. 98, The use, distribution or reproduction in other forums is permitted, provided the
5807–5822. doi: 10.1007/s00253-014-5792-6 original author(s) and the copyright owner(s) are credited and that the original
Yoe, S. M., Kang, C. S., Han, S. S., and Bang, I. S. (2006). Characterization and publication in this journal is cited, in accordance with accepted academic practice. No
cDNA cloning of hinnavin II, a cecropin family antibacterial peptide from the use, distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Microbiology | www.frontiersin.org 16 August 2021 | Volume 12 | Article 720381