Introduction

Given the increasing burden of bacterial infections and multiple-drug resistant bacteria, there is an urgent need for the development of novel antimicrobials (Tacconelli et al. 2018). In the USA alone, at least two million people acquire antibiotic-resistant infections resulting in 23,000 deaths annually (CDC 2018). The rate of emergence of resistant strains is much higher than the rate of introduction of new antibiotics in the market (CDC 2018). The development of antimicrobials from natural products is of prime importance (Mérillon and Rivière 2018; Harvey et al. 2015). Notably, the majority of commercially available natural products are derived from bacteria, fungi and plants. Nearly 70% of antibiotics are derived from soil dwelling bacteria (Smith 2000) such as actinomycin (from Streptomyces antibioticus), erythromycin (from Streptomyces erythraeus), aminoglycosides (from Streptomyces and Micromonospora) etc. Likewise, the first antibiotic penicillin was isolated from fungus Penicillium notatum (Fleming 1929), cephalosporins from Acremonium species (Newton and Abraham 1955) and ascochital, pestalone, indanonaftol A are antibiotics from various fungal species (Bugni and Ireland 2004; Cueto et al. 2001). Similarly, plant and plant products containing sesquiterpenes, triterpenes, flavonoids, procyanidins are shown to possess broad spectrum antibacterial activity against Gram-positive and Gram-negative bacteria (Ahmad et al. 1994). Of note, Kingdom Animalia represents largest diversity with more than 8 million species (Census of Marine Life). Classes such as fishes, amphibians, reptiles, birds and mammals comprises a huge diversity of terrestrial, marine and aquatic fauna (Science daily 2011). Unlike plants, their exposure to polluted environments and disease causing agents is greater. Therefore, it is thought that their ability to defend against pathogenic microorganisms is relevant to humans and must be explored. For example, cockroaches thrive in polluted environments suggesting their innate ability to produce anti-infective agents (Lee et al. 2011). Also, invertebrates particularly insects are used to treat various illnesses and are common in traditional medicines (Costa-Neto 2005). Insects such as hairy arachnids, Chinese black mountain ant, honey bee and bee products, scorpions, grass hoppers, silk worms, termites etc. are believed to possess various health benefits and are used in the treatment of wound healing, pain, cough, inflammation, fever, gastrointestinal related disorders, reproductive illnesses, pneumonia, hemorrhage, diarrhea etc. (Feng et al. 2009; Srivastava et al. 2009). However, the scientific basis of their medicinal properties remains incompletely understood. Previously, we showed the presence of potent antibacterial molecules in cockroaches against methicillin resistant Staphylococcus aureus (MRSA) and neuropathogenic Escherichia coli K1 (Lee et al. 2011; Ali et al. 2016). Several molecules were identified containing isoquinoline group, chromene derivatives, thiazine groups, imidazoles, pyrrole-containing analogs, sulfonamides, furanones, and flavanones with known antibacterial properties (Ali et al. 2016). Among other species, forest centipede, Scolopendra subspinipes, (also named as Vietnamese or Chinese Red-headed centipede) is commonly used in folk medicine, for its various health benefits in the treatment of wounds, pain, inflammation, sores and tumors (Lee et al. 2017; Bajpai et al. 2017; Ding et al. 2016; Choi et al. 2008). Mainly, distributed in East Asian countries, they are large with the maximum length of 20 cm and feeds primarily on insects, arachnids and small vertebrate animals, and encounter pathogens in their natural habitat (Bush et al. 2001). They must have developed mechanisms to counter infections. Hence, we aim to determine antibacterial activity of S. subspinipes against a panel of Gram-positive and Gram-negative bacteria and to identify biological molecule(s) using liquid chromatography mass spectrometry.

Materials and methods

Bacterial cultures

Eight clinical isolates were tested in this study, among which MRSA (Malaysian Type Culture Collection MTCC 381123), Bacillus cereus (MTCC 131621) and Streptococcus pyogenes (ATCC 49399) were Gram-positive; while, Escherichia coli K1 (MTCC 710859), Pseudomonas aeruginosa (American Type Culture Collection ATCC 10145), Klebsiella pneumonia (ATCC 13883), Salmonella enterica (ATCC 14028) and Serratia marcescens (ATCC 13880) were Gram-negative. All the strains were resistant to two or more antibiotics (Table 1). A 24 h old bacterial broth culture was used for experiments as previously described (Khan et al. 2008).

Table 1 Antibiotic susceptibility profile of bacteria used in this study

Organ lysates of centipede

Wild forest centipedes (S. subspinipes) with approximate length of 18 cm were collected from forest plantation from their natural habitat and kept in a glass cage individually overnight at 30 °C with soil organic matter. 70% ethanol was used to disinfect dissection tools. Centipedes were kept at 4 °C for 15 min. The insect was immobilized by the dissection pins on the anterior and posterior end of the body in a wax tray. The head and legs were removed, and the haemolymph was collected aseptically in ethylenediamine tetraacetic acid (EDTA) containing vacutainer by inserting the sterile pipette tip at the lateral opening of the removed limb (Fig. 1). Digestive system was exposed by the vertical incision made along the midline of the body and the sample was removed aseptically. After collecting the haemolymph and gut, muscle tissue was exposed, a sample of which was aseptically removed and suspended in small volume of sterile distilled water. Protease inhibitors (serine/cysteine/metallo-proteases) were added and the samples were processed at 4 °C and gut and muscle tissue were subjected to ten cycles of freeze-thawing. Homogenization of the samples were performed aseptically with mortar and pestle, followed by sonication and cold centrifugation at 10,000g for 30 min. Next, the lysates were filtered with 0.2 μm pore size sterilized filter to avoid contamination and unwanted residual particles, and the protein concentration was determined by Bio-Rad protein assay kit. Lysates were aliquoted and stored at − 20 °C until further usage.

Fig. 1
figure 1

a Dorsal view of S. subspinipes with intact body segments. b Closer view of upper body of centipede. c Internal organs of the centipede along the body cavity

Antibacterial assay

Antibacterial assays were carried out to determine bactericidal and bacteriostatic activities of haemolymph and tissue lysates of centipede as reported previously (Khan et al. 2008). A 24 h old fresh bacterial culture was adjusted to the absorbance of 0.22 at 595 nm using a spectrophotometer. Approximately 106 bacterial cells suspended in 10 μL of broth, were incubated with 100 µg/mL concentration of organ lysates or 10% haemolymph at 37 °C for 2 h. After incubation, serial dilution of reaction mixture containing bacterial cells was performed followed by plating on nutrient agar plates (Ali et al. 2016; Khan et al. 2008). Bacteria incubated in PBS/broth alone were used as negative control, however, bacteria incubated with 100 μg/mL of gentamicin were used as positive control. Percentage bactericidal/bacteriostatic activity was determined as bacteria surviving relative to the control: 100 − (cfu recovered/original inoculum × 100).

Human keratinocyte cell (HaCaT) cultures

Human keratinized skin cells (Hacat) (CLS:300493) were purchased from CLS Cell Lines Service, Germany. Cells were cultured in cell culture media comprising RPMI-1640, 10% heat-inactivated fetal bovine serum, 2 mM glutamine, 100 U penicillin/mL, 100 μg streptomycin/mL, non-essential amino acids, and vitamins as previously described (Ali et al. 2016; Khan and Siddiqui 2009). Cell cytotoxicity assays were carried out in 96-well plates by inoculating 5 × 105 HaCaT cells per well per mL followed by incubation at 37 °C with 5% CO2 for 48 h. Next, complete monolayer formation was observed microscopically prior to cytotoxicity assays.

Bacterial-mediated host cell cytopathogenecity assays

Centipede haemolymph (10%) was incubated with 106 bacterial cells at 37 °C for 2 h followed by co-incubation with approx. 2 × 106 HaCaT cells at 37 °C in a 5% CO2 incubator for 20 h. Next day, cell suspensions containing metabolites and lactate dehydrogenase enzyme (if present) were collected, centrifuged and subjected to reaction with substrate and dye (present in cytotoxicity detection kit) for 10 min and cytopathogenicity was determined by measuring absorbance of test and control wells at 495 nm. Bacterial-mediated host cell cytopathogenicity were determined and untreated bacteria incubated with human cells were used as controls (Ali et al. 2016; Khan and Siddiqui 2009). Percent cytotoxicity was determined by = (sample value − control value)/(total LDH release − control value) × 100.

Liquid chromatography–mass spectrometry (LC–MS): separation and analysis

Centipede haemolymph was tested for further chemical identity. Haemolymph was subjected for LC–MS analysis on Agilent 1290 infinity liquid chromatograph (Agilent Technologies, Wilmington, DE), coupled with an Agilent 6520 Accurate-Mass quadrupole-time of flight (Q-TOF) mass spectrometer with dual electrospray ionization source (ESI). Reverse-phase high performance liquid chromatography was used for separation of compounds, with an agilent Zorbax Eclipse XDB-C18, Narrow-Bore 2.1 × 150 mm, 3.5-micron column at 25 °C, and equilibrated with solvent A (0.1% formic acid in Milli-Q water) and solvent B (0.1% formic acid in Acetonitrile). 0.5 mL/min flow rate with a linear gradient was used as follows: 5% solvent B for 5 min, 100% solvent B for 20 min, and 100% solvent B for 25 min. The total run time was 30 min. The compounds were ionized using dual ESI + Accurate-Mass Q-TOF mass spectrometer. The ion source parameters were as follows: capillary voltage at 4000 V for positive and 3000 V for negative ion polarity. Flow rate of drying gas was 10 L/min with a fragmentor voltage of 125 V and gas temperature of 300 °C. Pressure of nebulizer gas was set at 45 psi with Quadrupole-TOF detector, while 50% MeOH + 50% Milli-Q water was used as blank after processing each sample.

Identification of compounds through Metlin database

As described, haemolymph was processed for liquid chromatography mass spectrometric analysis, in order to obtain the spectra of chromatograms determining molecular mass of the compounds in crude extract. The mass spectra of the compounds retrieved from HPLC were run against Metlin_AM_PCDL-N-170502.cdb for identification with exact homology through Agilent Mass Hunter software, while keeping in view compensation needed for charges in positive ESI MS as well as electron fragmentations, to ensure searches for the correct parent mass. Novelty determination of the identified compounds was performed on Scifinder software. However, previously reported compounds were subjected to literature search for biological activities.

Results

Centipede lysates exhibit potent antibacterial activity against broad range of bacteria

Centipede’s haemolymph was aspirated and lysates were prepared and tested against Gram-positive and Gram-negative bacteria for determination of antibacterial effects. In particular, haemolymph was remarkably active against bacterial strains tested with more than 90% growth inhibitory activities against MRSA and B. cereus, but more than 50% bacteriostatic activity against E. coli K1, K. pneumonia, S. enterica, S. marcescens and S. pyogenes. Muscle lysates exhibited more than 50% bacteriostatic activity against S. enterica, S. marcescens, P. aeruginosa and S. pyogenes (Fig. 2).

Fig. 2
figure 2figure 2

The crude extracts of red centipede’s haemolymph, gut and muscles were prepared and tested in antibacterial bioassays. For negative control, bacteria incubated with nutrient broth/PBS was used and for positive control bacteria incubated with 100 μg/mL of gentamicin was used. Asterisk represents P < 0.05. P values were obtained using two-sample T test and two-tailed distribution. a Represents 0% growth indicating potent bacteriostatic activity of 10% haemolymph, 100 μg/mL of muscle and gut extracts of red centipede against MRSA. b Represents cidal assay, indicating 50%, 80% and 68% viability of respective extracts against MRSA. c Represents more than 90% bacteriostatic activity of all the three extracts against B. cereus. d Also represents more than 90% bactericidal activity for all three extracts against B. cereus. e Represents 49%, 31% and 63% growth in bacteriostatic assays respectively against K. pneumoniae. f Represents 48%, 53% and 75% viability in bactericidal assays respectively against K. pneumoniae. g Represents 22%, 10% and 49% growth in bacteriostatic assays respectively against S. enterica. h Represents 55% and 78% viability for haemolymh and gut extracts however, muscle extracts was not active in bactericidal assays against S. enterica. i Represents 27%, 31% and 70% growth in bacteriostatic assays respectively against E. coli K1. j Represents 49%, 44% and 73% viability in bactericidal assays respectively against E. coli K1. k Represents nearly 50% bacteriostatic activity of all three extracts against S. marcescens. l Represents no bactericidal activity of centipede’s extracts against S. marcescens. m Represents nearly 83, 81 and 47% growth of centipede’s haemolymph, gut and muscles respectively against P. aeruginosa. n Represents no bactericidal activity of centipede’s extracts against P. aeruginosa. o Represents nearly 33, 19 and 38% growth of centipede’s haemolymph, gut and muscles respectively against S. pyogenes. p Represents 63, 50 and 51% viability of the extracts respectively against S. pyogenes. The results are representative of several experiments performed in duplicates and expressed as the mean ± standard error

Host cell cytopathogenecity assays

To determine the toxic effects of haemolymph treated bacteria against primary human keratinocytes, cytopathogenicity assays were performed. Treated and untreated bacterial cells were incubated at 37 °C for 2 h, followed by co-incubation with HaCaT monolayers at 37 °C in a 5% CO2 incubator for 20 h and lactate dehydrogenase enzyme release (cell lysis marker), was measured using a cytotoxicity detection kit. When treated with 10% haemolymph, B. cereus showed host cell death significantly reduced, from 100% to only 36% (P < 0.05). Similarly, E. coli K1 treated with haemolymph also showed significant reduction in producing host cell damage (P < 0.05). Notably, haemolymph alone produced approximately 25% host cell damage (data not shown). Overall, the treatment of bacterial cells with centipede’s haemolymph reduced bacterial-mediated host cell damage as compared to untreated bacteria (Fig. 3).

Fig. 3
figure 3

The haemolymph of red centipede was aspirated and tested in cytopathogenicity assay against Human Keratinocyte HaCaT monolayers. 10% haemolymph was incubated with 106 bacterial cells for 2 h at 37 °C, followed by the co-incubation with HaCaT monolayers in 5% CO2 incubator at 37 °C for 18–20 h. Untreated bacteria incubated with HaCaT monolayers were used as control. Next day, supernatent containg lactate dehygrogenease enzyme were collected, centrifuged and determined by Rosche cytotocicity detection Kit as per guidelines. Significant reduction was observed in the cytopathogenicity caused by the pre-treated bacteria incubated with haemolymph as compared to untreated bacteria. B. cereus showed up to 36% and E. coli K1 showed up to 64% cytotoxicity to human cells when pre-treated with haemolymph in contrast to untreated bacteria which showed 100% cytopathogenicity to human cells. The results are representative of several experiments performed in duplicates and expressed as the mean ± standard error

Identification of biologically active molecule(s) in centipede haemolymph using liquid chromatography–mass spectrometry

Centipede haemolymph was subjected to LC–MS (Agilent Technologies 6520 Accurate-Mass Q-TOF mass spectrometer with dual ESI source) for qualitative analyses. Figure 4 shows spectra from negative and positive ion polarity. Compounds present in haemolymph were separated in the column on the basis of mass to charge ratio (m/z) and retention time. The data obtained from the LC–MS for haemolymph contained 48 compounds in total, out of which identity of 12 compounds was confirmed. These include, (1) l-Homotyrosine, (2) 8-Acetoxy-4-acoren-3-one, (3) N-Undecylbenzenesulfonic acid, (4) 2-Dodecylbenzenesulfonic acid, (5) 3H-1,2-Dithiole-3-thione, (6) Acetylenedicarboxylate, (7) Albuterol, (8) Tetradecylamine, (9) Curcumenol, (10) 3-Butylidene-7-hydroxyphthalide, (11) Oleoyl Ethanolamide and (12) Docosanedioic acid (Table 2). From remaining 36 compounds, limited information regarding retention time, molecular mass and formula of 23 compounds were determined, whereas for 13 compounds, only molecular mass and retention time were determined (Table 3). The 12 compounds identified from centipede haemolymph were subjected for novelty determination via Scifinder software. Interestingly, all of them were found to possess reported biological activities for their exact and homologous structures.

Fig. 4
figure 4

Scolopendra subspinipes haemolymph was subjected to LC–MS (Agilent 6520 Accurate-Mass Q-TOF mass spectrometer with dual ESI source) for qualitative analyses. The compounds were separated based on m/z ratio and retention time in the column. The data obtained from the LC–MS for haemolymph contained numerous peaks, out of which 12 compounds have been detected with full identity and molecular structure, however, 36 compounds were detected with the limited information of only molecular mass and retention time in the column

Table 2 Compounds identified from the red centipede haemolymph
Table 3 Compounds identified in the haemolymph of red-headed centipede

Discussion

Development of robust antimicrobials from novel sources is the current need to counter drug resistant pathogens (Challinor and Bode 2015; Harvey et al. 2015). Most common sources of antimicrobials are bacteria, fungi, plant and plant products that have been used widely in modern medicine (Abraham et al. 1953; Wagman 1980; Negi et al. 1999). In contrast, discovery of antimicrobials from animal sources is an area explored superficially. This is despite the fact that animals particularly invertebrates such as cockroaches, ants, silk worms, scorpions and tarantulas have been used in traditional medicine for centuries (Costa-Neto 2005). For example, larval therapy is used widely to cure non-healing wounds. This involves, the application of mature blow fly larvae belonging to Sarconesiopsis genus on an open wound, resulting in the secretion of antimicrobial peptides and metabolites (Diaz-Roa et al. 2018). Maggot debridement therapy is effective to cure severe necrotizing fasciitis, caused by more than one type of bacteria such as MRSA, Streptococcus pyogenes, enterococci, E. coli, P. aeruginosa, Clostridium and Bacteroides species (Maya et al. 2014). Maggot debridement therapy is useful in patients suffering from necrotizing fasciitis with an underlying disease who cannot be subjected to surgical procedures such as diabetic patients (Dunn et al. 2002). Other studies showed that application of sterile larvae belonging to genus Lucilia sericata, Protophormia terraenovae, Sarconesiopsis magellanica secretes antimicrobial molecules/peptides such as as p-hydroxybenzoic acid, p-hydroxyphenylacetic acid, dioxopiperazine proline, seraticin, defensins, cecropins, diptericins and proline-rich peptides with potent anti-biofilm and wound healing properties (Nigam et al. 2010; Chernysh et al. 2018). Similarly, arthropods such as wild centipedes have been used in traditional Chinese medicine, often used to treat various illnesses such as seizures, apoplexy, stroke induced hemiplegia, diphtheria, tuberculosis, pyocutaneous disease etc. (Moon et al. 1996; Undheim and King 2011). In Korea, crushed centipede is used to treat back pains, sores and furuncles (Douglas 2014). Recent studies also highlight its broad range of antimicrobial activity against various pathogens. For example, S. subspinipes mutilans exhibited antifungal activity by membrane permeabilization in Candida albicans (Choi et al. 2013). Similarly, antimicrobial activity of the peptide lacrain, isolated from body extract of S. viridicornis showed strong bactericidal activity against Gram-negative bacteria (Chaparro and Da Silva Junior 2016). 3,8-Dihydroxyquinoline also known as jineol, isolated from S. subspinipes mutilans showed antibacterial activity by altering the release of potassium ions from food borne pathogenic strains of E. coli O157:H7 and S. aureus KCTC-1621 (Bajpai et al. 2017). Several other AMPs such as Scolopendrasin I, V, VII are known to possess broad range of antimicrobial activities against drug resistant pathogens (Wenhua et al. 2006; Peng et al. 2010). For the first time, here we determined the antibacterial activity of the haemolymph/organ lysates of red-headed centipede S. subspinipes, with molecular identification of biological components using LC/MS. Our findings suggest that haemolymph and tissue extract of centipede exhibited antibacterial activity against Gram-positive and Gram-negative bacteria. Haemolymph subjected to chemical characterization indicated the identification of 12 compounds with reported biological activities against Gram-positive and Gram-negative bacteria, fungi, viruses and parasites (Pascal et al. 1985; Komorowska-kulik et al. 1998; Niu et al. 2018; Bierer et al. 1998; Baba et al. 2015). For example, compounds 1, 3, 4, 5, 6, 8, 9 and 12 possess antimicrobial activity against a broad range of microorganisms such as S. aureus, P. aeruginosa, P. mirabilis E. coli, H. pylori, Aspergillus species, Candida species, F. oxysporum, C. neoformans, dermatophyte T. rubrum, A. alternata, C. purpureum, P. cactorum, P. infestans, V. inaequalis, B. cinerea, E. graminis, P. recondite, Human Papilloma virus, HIV and parasite Giardia.

Moreover, compounds 1, 4, 5, 6, 7, 9, 10, 11, 12 possess anticancer activity against colon cancer cells, MCF-(breast cancer), NCI-H187 (lung cancer) and KB cells, human gastric cancer cells, HepG2 (Liver carcinoma) cells (Pagano et al. 2017; Wisetsai et al. 2018; Jung et al. 2018; Ali et al. 2001; Bigge et al. 2003; Kuo et al. 2016; Pelcman and Bengtsson 2018; Lee et al. 2016; Hakonarson et al. 2018) (Table 2).

Interestingly, some of the compounds identified also possess antidiabetic, anti-neurodegenerative, antioxidant, antiepileptic and anticancer activities (Bigge et al. 2003; Wisetsai et al. 2018; Gong et al. 2016). Identified compounds contain furan, tyrosine, thione, albuterol, amines, curcumenol and pthalide moieties, potentially responsible for biological activities. Notably, compounds 2, 5, 9, 10 and 12 are phytochemicals with antibacterial, antifungal, anti-inflammatory, anticancer and analgesic properties (Giannini et al. 2004; Gupta et al. 2018; Tran et al. 2018; Kacem et al. 2016; Brinkworth and Bairlie 1992). Biological significance of these compounds are due to their distinct features and structural arrangement of the functional groups. For example, sulfides and disulfides in cpd 5 are active ingredients. Sulphur has its characteristic property and is an essential component in antibiotics such as sulphonamides (Mitchard 1988). Curcumenol cpd 9, containing tetrahydrofuran is an active five-membered oxygen heterocyclic compound, commonly found in natural products, mainly responsible for their antibacterial activity (Keglevich 2015). Phthalides and fatty acids present in cpd 10 and 12 are also well known for their broad spectrum activities such as antiinflammatory, antimicrobial and anticancer activities (Bierer et al. 1995; Gao et al. 2013; Wisetsai et al. 2018). Notably, 36 compounds were not identified in this study. These are also of potential interest and could represent novel antibacterials (Table 3).

In summary, the discovery of natural antibiotic molecules from animals/invertebrates, exposed to the environmental wastes and pollutants in their natural habitat is a fascinating though unexploited area of research. Hence, it is anticipated that the antibiotics from natural sources are minimal or less toxic for biological system as compared to synthetic antibiotic molecules. Further identification, in vivo testing and clinical trials of potentially active metabolites can act as a milestone for the synthesis and development of novel drug leads.