Metabolite profile of marine-derived endophytic Streptomyces sundarbansensis WR1L1S8 by liquid chromatography–mass spectrometry and evaluation of culture conditions on antibacterial activity and mycelial growth

Authors

  • I. Djinni,

    1. Bioorganic Chemistry Laboratory, Department of Physics, University of Trento, Povo, Italy
    2. Laboratory of Applied Microbiology, Faculty of Nature Science and Life, University of Bejaia, Targa Ouzemmour, Bejaia, Algeria
    Search for more papers by this author
  • A. Defant,

    1. Bioorganic Chemistry Laboratory, Department of Physics, University of Trento, Povo, Italy
    Search for more papers by this author
  • M. Kecha,

    1. Laboratory of Applied Microbiology, Faculty of Nature Science and Life, University of Bejaia, Targa Ouzemmour, Bejaia, Algeria
    Search for more papers by this author
  • I. Mancini

    Corresponding author
    1. Bioorganic Chemistry Laboratory, Department of Physics, University of Trento, Povo, Italy
    • Correspondence

      Ines Mancini, Bioorganic Chemistry Laboratory, Department of Physics, University of Trento, via Sommarive 14-38123, Povo, Trento, Italy. E-mail: ines.mancini@unitn.it

    Search for more papers by this author

Abstract

Aims

This study was designed to investigate whether culture conditions (media, seawater concentration and pH) could lead Streptomyces sundarbansensis strain (isolated from marine brown algae Fucus sp. collected from Algerian coastline) to produce bioactive secondary metabolites. The most favourable condition for the production of anti-MRSA compound 1 [2-hydroxy-5-((6-hydroxy-4-oxo-4H-pyran-2-yl)methyl)-2-propylchroman-4-one] was determined.

Methods and Results

The profile of metabolites present in the crude extracts was carried out by HPLC analysis equipped with a diode array detector evaporative light scattering detection (DAD-ELSD) or online coupled to electrospray ionization–mass spectrometry (ESI-MS). Compound 1 was the most abundant secondary metabolite by culturing the strains on starch casein agar (SCA) medium in freshwater or 50% seawater at pH 7 or 9 using agar-state fermentation method.

Conclusions

The study has shown the efficiency of HPLC/ESI-MS technique in the analysis of polyketides produced by the strain under investigation. It was possible to establish the best culture conditions for obtaining the most bioactive compound 1, previously isolated by the same strain.

Significance and Impact of the Study

Marine algae–actinobacteria associations are a particularly promising renewable system for the production of new antibacterial metabolites. Based on the promising bioactivity of the chemically characterized compound 1, the analytical methodology here applied has resulted as an effective approach for establishing its optimized production.

Introduction

Thanks to their ability to produce numerous and diverse bioactive natural products, marine micro-organisms, especially bacteria belonging to the Actinomycetales order, are attracting much more attention as stimulated by the need of novel antibiotics for multiresistant pathogenic strains and of antitumour agents. This takes advantage from the diversity of marine environment which has exerted a driving force on bacteria selection leading to new adaptive strategies and to the synthesis of novel bioactive metabolites (Jensen and Fenical 1996; De Carvalho and Pedro 2010).

In our recent screening for bioactive metabolites from marine-derived actinomycetales, Streptomyces sp. WR1L1S8 strain was selected among twenty-two strains isolated from a marine algae collected in Bejaia coastline, North East of Algeria. This strain produced a new polyketide 1 [2-hydroxy-5-((6-hydroxy-4-oxo-4H-pyran-2-yl)methyl)-2-propylchroman-4-one] as well as three known compounds 2–4 (Fig. 1). All the metabolites are members of the phaeochromycins class, for which we have recently proposed the favoured 2-hydroxy-γ-pyrone structure based on a comparison of experimental and calculated IR spectra (Djinni et al. 2013). They are analogues of the polyketides peculiar for the presence of a propyl side chain. These known metabolites include phaeochromycins B (2), C (3) and E (4), recently isolated from the soil Streptomyces phaeochromogenes LL-P018, and compound 3 reported as an inhibitor of MAPKAP-2 kinase and a potential agent in the treatment for rheumatoid arthritis (Graziani et al. 2005). In our previous work, polyketide 1 has shown the highest antibacterial activity, with a selective inhibition of methicillin resistance Staphylococcus aureues (MRSA) [MIC = 2 μg ml−1, 6 μmol l−1] (Djinni et al. 2013), a promising result for further investigations.

Figure 1.

Molecular structures of polyketides 1–4 isolated from Streptomyces sundarbansensis WR1L1S8 strain.

Herein, we reported (i) the isolation details and the taxonomic description of the strain producing polyketides 1–4, (ii) the bioactivity evaluation of crude extracts depending on culture conditions, with the selection of the most favourable one for obtaining 1, and (iii) the comparison of chemical fingerprints through the analysis of metabolite profiles in crude culture extracts by online HPLC-ESI-MS technique.

Materials and methods

Sampling sites and algae collection

Algal samples were collected in the Mediterranean Sea at Bejaia coastline, Algeria (36°39′4.25″N; 5°25′3.88″E). Several specimens (10) of the coastal brown algae Fucus sp. and green algae Ulva lactuca were collected at a depth of 50 cm, in July 2011. The collected seaweed samples were stored in sterile plastic tubes containing natural seawater, cooled on ice and transported immediately to the laboratory, where they were washed three times with sterile seawater.

Selective isolation of actinomycetales from algal samples

The algal tissue was cut into small pieces of c. 0·5 cm2 each and placed onto agar plates. Liquid tissue portions were also prepared and used for actinobacteria isolation. Samples were aseptically cut into small pieces (±10 mm), then placed in a sterile mortar and thoroughly homogenized with sterile seawater (10 ml). Decimal dilutions were made up 10−4. Samples were spread on selective agar plates in triplicate Petri plates and incubated at 28°C for 3 weeks.

Five different media were used for the isolation of actinomycetales strains: starch casein agar (SCA) medium (Küster and Williams 1964), synthetic medium (SM) (Mincer et al. 2002), modified Gausse medium (Ivantiskaya et al. 1978), modified chitin medium (Hayakawa and Nonomura 1987) and glycine–glycerol medium (Küster 1959). The pH value of the isolation media was adjusted to pH 7·2 ± 0·2. All media contained 50% natural seawater and were supplied with penicillin (50 μg ml−1) and K2Cr2O7 (50 μg ml−1). Colonies of streptomycete-like and nonstreptomycete-like strains growing on the isolation plates were inoculated onto SCA medium plates and incubated at 28°C for 7–10 days.

Selection of strains producing antibacterial metabolites

All the isolates were tested to determine their antibacterial activity by the agar cylinders method (ACM). The agar cylinders (6 mm diameter) were formed, from the growth strain, and shifted to the surface of Mueller-Hinton plates (Bastide et al. 1986) uniformly inoculated with a lawn of bacterial strains (Escherichia coli ATCC 25922, Staphylococcus aureus ATCC 25923, MRSA ATCC 43300, Bacillus subtilis ATCC 6633 and Pseudomonas aeruginosa ATCC 27853) on the agar surface (107 CFU ml−1) then kept at 4°C for 2 h. Activity was assessed by the formation of clearing zones around the agar cylinder after incubation at 37°C.

Taxonomic features of the antibiotic producing strain

Morphological and physiological characteristics

Methods and media described by the International Streptomyces Project (ISP) were used to determine most of the morphological characteristics of the isolate (Shirling and Gottlieb 1966). The morphology of the strain cultured on different media was examined under optic (Leica DM 300) and phase contrast microscopy at various magnifications (Leica EZ4D) as well as under scanning electron microscopy (SEM) (JOEL-JSM-7001F, Field Effect Gun), (Kumar et al. 2011). Physiological characterization of the selected strain was based on the lytic activity towards several organic compounds as described by Gordon et al. (1974) for gelatin, starch, casein and tyrosine hydrolysis; moreover, lipase activity assay was carried out using Tween-80 according to Sierra (1957). Different carbon and nitrogen sources were used according to Pridham and Gottlieb (1948). Resistance to some antibiotics was detected by disc diffusion method (Shirling and Gottlieb 1966). Growth responses to different temperatures, pH, seawater requirement and NaCl concentrations were determined on SCA medium.

Chemotaxonomic analysis

Standard procedures were used to determine the diagnostic isomers of diaminopimelic acid (Stanek and Roberts 1974). Whole-cell sugar composition was performed using HPLC-ELSD technique according to Lechevalier and Lechevalier (1970) for the sample preparation. The concentration of the standards and prepared samples was 1 mg ml−1 in acetonitrile/water (70 : 30). The analysis was performed by HPLC-ELSD Agilent 1100 series technologies using a reversed-phase (RP) C18 (Phenomenex, Torrance, CA, USA; HELIC, Luna 5 μm, 250 × 4·60 mm) column in isocratic mode (acetonitrile/water 90 : 10, flow : 1 ml min−1, detector ELSD at 45°C).

After a total lipid extraction as described by Bligh and Dyer (1959), polar lipids were obtained according to the methods reported by Pasciak et al. (2003), with modification in the volume ratio of methanol, chloroform and water. In detail, the wet mycelium was extracted twice with chloroform–methanol (2 : 1, v/v, 15 ml g−1 wet mass) at room temperature. The mixture was evaporated and then resuspended in chloroform/methanol/water (5 : 4·5 : 4 v/v) to remove nonlipidic material.

Total lipid extracts were analysed by LC-MS technique using a reversed-phase RP-18 column and gradient mobile phase (solvent A: 100% methanol, solvent B: methanol/water 7 : 3 with 28 mmol l−1 aqueous ammonium acetate).

The molecular identification of the strain was obtained by the partial 16S rRNA gene by direct sequencing of PCR amplified 16S rDNA (DSMZ, Braunschweig, Germany). The nucleotide sequence of 16S rRNA gene reported in this article was deposited in GenBank database (accession number KC773843).

Influence of culture conditions on bioactivity and strain growth

Selection of the optimized nutrient medium

Five different culture media [SCA, ISP2, glucose yeast extract agar (GYEA), SM and tryptic soy broth (TSB)] were used in comparative studies to select the optimal nutrient medium for antibiotic production. All media were inoculated with WR1L1S8 spores suspension (107 CFU ml−1) at 28°C for 7 days. Antimicrobial activity was performed by agar cylinder method using E. coli, Staph. aureus and MRSA target bacteria. The medium in which the strain exhibited the maximum antibiotic production expressed in terms of inhibition zone was used as the optimized medium for further study.

Selection of the fermentation process

To define the adequate fermentation method [agar surface fermentation (ASF) and submerged fermentation (SbF)], the actinomycetale strain was initially grown on two Petri dishes containing the selected solid medium inoculated with a spore suspension of about 107 CFU ml−1. After an incubation period of 7 days at 28°C, mycelial mass together with the agar of one Petri dish was subjected to sonication and then macerated overnight in ethyl acetate (100 ml). The resulting mixture was filtered, and the maceration in ethyl acetate was repeated twice. The combined filtrates were evaporated in vacuo to dryness. The crude residue was subjected to tests against E. coli, Staph. aureus and MRSA using well-diffusion assay (100 μl) on Mueller-Hinton medium. Agar cylinders obtained from the second Petri dish were tested against the same target bacteria to confirm the bioactivity.

The selected active strain was precultured in flasks of 50 ml containing 10 ml of broth medium, at 28°C for 3 days. The precultured isolate was transferred to 500-ml Erlenmeyer flasks containing 100 ml of fermentation medium (10%) and incubated on a rotary shaker (107 CFU ml−1). The fermentation broth was centrifuged (10 min, 4°C at 15 000 g) to separate the mycelium from the supernatant. The culture supernatant was secondly concentrated by half. In parallel, 100 ml of cell-free supernatant was extracted with an equal volume of ethyl acetate. The resulting dry extract was prepared and then assayed as described for the ASF. The best fermentation method was selected based on the exhibition of a maximum antibiotic production expressed in terms of inhibition zone.

Solvent for extraction

Streptomyces sp. WR1L1S8 active strain was cultivated, in a second time, on the selected medium (107 CFU ml−1), and an extraction test was conducted as reported above to determine the best extraction solvent. Five solvents with different polarities were used: hexane, dichloromethane, chloroform, benzene (carefully used) and ethyl acetate. However, only the ethyl acetate extract of the actinomycetale culture exhibited considerable antibacterial activity. Therefore, the present study was focused on ethyl acetate extracts.

Effects of seawater, pH value, salinity and temperature

To study the effect of seawater on the growth and the antibiotic production, the strain was cultured in the absence and in the presence of artificial seawater at 50 and 100% on SCA medium. The effect of initial pH was studied by growing the actinomycetale cultures in SCA media at four different pH values (5, 7, 9 and 11). The effect of salinity on mycelial growth was also studied by growing the strain in SCA media made of different NaCl concentrations (0, 1, 2, 4, 6, 8 and 10%) at 28°C. Furthermore, the strain was cultured in SCA media at 4, 10, 37 and 45°C to study the effect of temperature on mycelial growth.

Effects of supplementary carbon and nitrogen sources

With the aim of investigating the effect of the carbon and nitrogen sources on the antibiotic activity and growth of WR1L1S8 strain, the optimal nutrient medium for antibiotic production was employed as an original medium for the following optimization studies. Various sample and complex carbon and nitrogen sources were used individually instead of the corresponding carbon and nitrogen sources in the optimal nutrient medium, while other components were kept constant at original concentration.

Antibacterial assay on the crude extracts and growth measurement

Antibacterial assay of the ethyl acetate crude extracts was performed using a well-diffusion agar. A volume of 100 μl of crude extracts was placed onto each agar well to which the medium was previously inoculated by 10CFU ml−1 with E. coli, Staph. aureus and MRSA. Antibiotic activity was expressed as inhibition zones diameter (mm) and as units of activity per millilitre of the culture extracts, where 1 U was defined as a 1·0-mm annular clearing around the antibiotic agar well (Wang et al. 2011a). The strain growth was determined by the measurement of colony diameter (mm) carried out in a 30-day period of incubation, as well as by dry weights of mycelia.

Metabolite profiles by HPLC-DAD-ELSD and by HPLC-ESI-MS analyses

Ethyl acetate crude extract of cultured WR1L1S8 strain was subjected to online high-performance liquid chromatography (HPLC)-electrospray mass spectrometry (ESI-MS) for metabolite profile analyses as well as molecular weight determination. The extract was dissolved in methanol (1 mg ml−1) and then analysed by a Hewlett–Packard HP1100 HPLC-UV Diode Array detector (DAD), online coupled to an Esquire-Bruker–Daltonics ion trap mass spectrometer using a reversed-phase column on the analytical scale (HP, Hypersil BDS-C18, 250 × 4·00 mm). The mobile phase was applied as the linear gradient. The chromatographic parameters are the following: detection – DAD 210 nm; mobile phase – A acetonitrile, B water; 0–10 min 30% A–70% B, 10–30 min 100% A; and flow rate – 1 ml min−1.

Afterwards, to monitor changes in the metabolites profile, the strain WR1L1S8 was simultaneously cultivated on five media, as well as on SCA medium varying seawater concentration and pH value. Ethyl acetate crude extracts were analysed by HPLC-DAD-ELSD technique, using a reversed-phase column on the analytical scale (HP, Hypersil BDS-C18, 250 × 4·00 mm) in the same conditions adopted previously. ChemStation® (Agilent techonology, Waldbronn, Germany) software was used to integrate the area under the chromatographic peaks, allowing the simultaneous comparison of molecular typing data and fermentation metabolite profiles.

Results

Isolation of actinomycetale strains from marine algae

A total of 22 actinomycetale strains were isolated from algal surface tissue and liquid homogenates portions from two taxonomically different marine algae collected on the Bejaia coastline in Algeria and cultivated on a range of selective media. In an attempt to select the suitable culture conditions for the production of bioactive compounds, all the culture media were supplemented with 50% natural seawater. The highest number of actinomycetales was obtained from Fucus sp. yielding 20 isolates followed by Ulva lactuca which displayed only two isolates. It is noteworthy that the number of endophytic actinomycetales isolated from algal inner tissue (18) is higher than those recovered from the surface tissue (4). Actinomycetales were recognized based on colony and microscopic morphology. Colonies with a tough leathery texture, dry or folded appearance, and branching filaments with or without aerial mycelia that could be observed either unaided by eye or using a microscope were selected and studied.

The macro- and micro-morphology of the obtained isolates showed hard consistency and chalky aspects of colonies, fast-growing strains producing branched substrate mycelia, formed highly developed aerial hyphae. The branched aerial mycelia were straight, of variable length, with smooth surfaces. The micro-morphological observations by optic microscope allowed to note the absence of specific structures such as sporangia, synnemata or sclerotia and nonmotile spores, according to which they could be affiliated within the genus Streptomyces.

Five selective culture media were used to select actinomycetes. Using Fucus sp. marine algae, SCA medium showed the highest recovery giving 14 isolates; three strains were obtained from Gausse and chitin media and two isolates from SM, and no strains from glycine–glycerol medium. Ulva lactuca marine algae allowed the isolation of two strains exclusively recovered on SCA medium.

Antimicrobial activity

The evaluation of the antimicrobial activity observed on the isolates is reported in Table 1. The results indicated that only 59·09% of tested strains produced metabolites inhibiting the growth of bacteria used in the assay. Indeed, only 42·85% of isolated strains from SCA medium produced metabolites against at least one tested bacteria, except the strain WR1L1S8 whose metabolites displayed the broadest spectrum of activity, inhibiting the growth of both the Gram-positive and Gram-negative target micro-organisms, while only two isolates (WR1L1S6 and WR1L1S7) were found to inhibit selectively the Ps. aeruginosa strain growth. The strain WR1S2 obtained from algal tissue portions displayed a large inhibition against the E. coli tested strain. The isolate WVL1S7 exhibited rather a slight inhibition against Gram-positive test bacteria. Otherwise, GR1L1S1 and GR1L1S2 obtained on Gausse medium presented antibacterial activities against Gram-positive and Gram-negative tested organisms, and GR1L1S3 selectively against negative tested bacteria. Strains isolated from SM medium showed weak growth inhibitions, whereas A 66·66% of isolated strains from Chitin medium gave a good growth inhibition against E. coli.

Table 1. Antimicrobial activity expressed as mm of inhibition zone, using agar cylinder method, evaluated for the actinomycete strains from Bejaia coastline marine algae Fucus sp. and Ulva lactuca
StrainsaActivity (mm)
E. coli ATCC 25922Staphylococcus aureus ATCC 25923MRSA ATCC 43300B. subtilis ATCC 6633Pseudomonas aeruginosa ATCC 27853
  1. –, No-inhibition of the growth of the test micro-organism.

  2. a

    W, SCA medium; G, Gausse medium; M, synthetic medium; C, chitin medium; R1, marine brown algae Fucus sp., V, marine green algae Ulva lactuca; L, liquid, strains obtained from liquid tissue portion; S, solid; strains obtained from algal tissue portions.

WR1L1S1
WR1L1S2
WR1L1S3
WR1L1S4
WR1L1S5
WR1L1S612·98 ± 0·85
WR1L1S718·44 ± 1·20
WR1L1S813·73 ± 0·5316·95 ± 0·5817·23 ± 0·7110·04 ± 0·76
WR1S1
WR1S229·37 ± 1·199·23 ± 0·55
WR1S3
WR1S415·3 ± 0·47
GR1L1S115·33 ± 0·8912·72 ± 0·4815·76 ± 0·62
GR1L1S216·47 ± 0·6330·5 ± 1·06
GR1L1S314·07 ± 0·8116·56 ± 0·94
MR1L1S112·1 ± 0·68
MR1L1S211·13 ± 0·5711·74 ± 0·67
CR1L1S114·09 ± 0·88
CR1L1S2
CR1L1S322·45 ± 1·96
WVL1S1
WVL1S711·84 ± 0·5613·20 ± 0·63

The Streptomyces sp. WR1L1S8 strain, which presented a broad and interesting growth inhibition of the tested micro-organisms, was selected for further investigations, including its identification, cultivation and analysis of the profiles for its antibacterial compounds.

Characterization of Streptomyces sp. WR1L1S8 strain

According to morphological and chemotaxonomic studies including the presence of diaminopimelic acid isomers, diagnostic whole-organism sugars and phospholipid analysis, the strain WR1L1S8 was identified as belonging to the genus Streptomyces. The isolate shares a range of chemotaxonomic and morphological markers consistent with its assignment to this genus (Lechevalier et al. 1986) because it was a fast-growing strain producing an extensively branched substrate mycelium, formed a highly developed aerial hyphae and contained ll-diaminopimelic acid, glycine and galactose in whole-organism hydrolysates. The branched aerial mycelia were straight with smooth surfaces and exiguous branched protuberances. At the maturation state, these hyphae fragmented irregularly into long spore chains of about 8·38 μm. The individual spores were oval of c. 0·78 μm, smooth-surfaced and nonmotile; moreover, the strain did not produce sporangia.

Furthermore, it was characterized by the presence of a simple phospholipid pattern as directed by LC-MS analysis, consisting of 91·6% phosphatidylethanolamine, 7·1% phosphatidyl-glycerol, 0·8% phosphatidylcholine and 0·4% sphingomyelin and several uncharacterized polar lipid components.

The analysed partial 16S rRNA gene sequence revealed the evolutionary relationship of the strain WR1L1S8 to a group of Streptomyces species and has the highest similarity with Streptomyces sundarbansensis MS1/7T (DSMZ).

The culture characteristics of strain WR1L1S8 on various agar media were similar. The aerial mycelia were abundant, well developed and similarly brown green. The substrate mycelia varied from brown to various shades of green. No diffusible pigments were observed on agar media. The colony became pale green during sporulation with a chalky aspect. It was tolerant to NaCl concentrations till 8% but did not require the presence of NaCl or seawater for growth. Table 2 depicts physiological and biochemical characteristics of strain WR1L1S8, which used most of carbon sources indicating its broad assimilation capacity. Melanoid pigments on ISP6 and ISP7 media were noticed. Extracellular enzymes produced by the strain were found to hydrolyse starch, cellulose and Tween-80, but not casein and gelatin. The strain showed sensitivity to a variety of antibiotics, but was resistant to ampicillin and clavulanic acid/aztreonam, suggesting that the produced metabolites may be responsible for its resistance to antibiotics (Jain et al. 2012), although it is not always true.

Table 2. Physiological and biochemical characteristics of Streptomyces sp. WR1L1S8
TestsResultsTestsResultsTestsResults
Hydrolysis of purine bases and derivativesAmino acids utilization on ISP9 mediumGrowth at
  1. Note: +, positive test; –, negative test; ND, not determined; + + +: very good development, + +: good development, +: slight development, ±: development of the substrate mycelium (SM) only–: no development.

Adenine+Methionine ± pH 5+++
Guanine+Aspartic acid + pH 9+++
Hypoxanthine+CysteinpH 11+++
  Tyrosine4°C
  Histidine++10°C++
Carbohydrate utilizationAlanine++37°C+++
Adonitol ± Arginine+45°C ±
Starch+++Isoleucine+  
Arabinose±Threonine+++  
Galactose+++Proline+++Growth in the presence of inhibitory compounds
Glucose+++Serine++NaNO3 0·01%
Glycerol+++Glycine+++NaNO3 0·001%+++
Inositol TryptophanPhenol 0·05%+
Lactose+++  Phenol 0·1%
Maltose+++  Sensitivity to antibiotics (μg ml−1) (mm)
Mannose+++Extracellular enzyme activityErythromycin Ery 1541·11
Raffinose GelatineGentamicin Gent 1032·35
Rhamnose ± CaseineVancomycin Van 3018·47
Ribose+++Tween-80+Ampicillin Amp 100
Sorbitol ± TyrosineClavulanic acid/Aztreonam0
Sucrose++    
Trehalose+++    
Xylose++    
Cellulose+    
Fructose+    

Influence of culture conditions

It is known that growth media and growth conditions have a remarkable effect on the production of secondary metabolites and they may be different for different strains (Wagner-Döbler et al. 2002). Therefore, different cultivation parameters have been taken into account, to determine whether the growth conditions of the strain WR1L1S8 influenced its metabolites production and growth. According to the results summarized in Fig. S2,A, SCA was the best for the production of bioactive metabolites, as established by E. coli and MRSA assays, giving values of 14·66 and 16·66 mm for diameter inhibition zones, respectively, result in a good agreement with previous report by Mahyudin (2008). Hence, SCA medium was employed as an original medium for further studies. However, ISP2 medium was the most suitable one for strain growth (Fig. S2,B). These results indicated that the conditions allowing a fast cell growth were different from the ones favouring the best metabolite production, as observed by Audhya and Russell (1974).

Concentrated culture broth showed no activities against the tested organisms, whereas ethyl acetate crude extracts obtained from submerged culture and agar plate fermentation presented relatively interesting activities. The minor amount of antimicrobial metabolites observed in liquid than on solid medium is in line with the study by Wang (1989). In sharp contrast to the ethyl acetate, extracts from agar plates using other solvents (acetone, chloroform, dichloromethane, hexane and benzene) showed lower activities.

The influence of seawater concentration on the production of bioactive metabolites and on growth of the strain WR1L1S8 has been evaluated. By HPLC-ESIMS metabolite profiles showing differences in the compounds production, it was possible to establish that the studied strain was able to grow and to produce antibacterial compounds both in the presence and in the absence of seawater (0–50 and 100%) (Fig. S3). However, this fact did not allow to establish whether the strain is a marine actinomycete requiring seawater for its development, or whether it comes from a freshwater or terrestrial environment.

The growth of the strain was strongly influenced by the presence of high concentrations of NaCl (>8%) in the SCA medium (Fig. S4), although it was able to grow both in the absence and in the presence of NaCl. Nevertheless, the optimum growth was observed on SCA 1% NaCl, with a growth rate estimated at 2·6 mm day−1.

Growth occurred between 10 and 45°C with 28°C as the optimal growth temperature.

No growth was observed at 4°C till 30 days of incubation. It was, however, increased while temperature was increased from 10 to 28°C (Fig. S4), but dropped to the lowest at 45°C.

Streptomyces sp. WR1L1S8 grew under a wide range of pH conditions (from 5 to 11). Maximum growth and an increased production of antimicrobial metabolites were obtained at pH 7 (Fig. S5), suggesting the neutrophilic characteristics of the strain.

Based on SCA medium, the effect on antibiotic activity of various carbon and nitrogen sources was studied, using a final concentration of 10 and 0·3 g l−1 for carbon and nitrogen sources, respectively. Among the tested carbon sources, the strain exhibited a maximum antibiotic activity in starch and glucose (171·73 U ml−1 and 177 U ml−1), respectively, against Staph. aureus, and 127·19 and 116·4 U ml−1 against E. coli (Fig. S6). The bioactivity in lactose and galactose ranged from 114·4 to 119·26 U ml−1 against Staph. aureus and from 102·5 to 111 U ml−1 against E. coli. Among the nitrogen sources, casein (129·66 U ml−1) favoured antibiotics production (Fig. S7). SCA medium using peptone, yeast or malt extracts, casein and KNO3 led to a relatively low growth rate of the strain.

Metabolite profile by HPLC-ESI-MS analysis

The target of this analysis is based on the use of a sensitive and rapid method to estimate the production of secondary metabolites present in fermentation extracts. This request was found in HPLC technique using DAD and ELSD as detectors, as well as the online coupling with ESI MS. In Fig. 2, the chromatographic profile of ethyl acetate crude extract of Streptomyces sp. WR1L1S8, performed by HPLC-DAD analysis at 210 nm, is reported, resulting practically superimposable to the profile monitored at 254 nm as an indication that no other metabolites were present in addition to phaeochromycin-like molecules. Extensive NMR analysis and MS fragmentation experiments on the purified compounds resulted in the identification of the metabolites (Djinni et al. 2013), corresponding to the HPLC peaks, in detail 4 (tR = 2·21 min), 2 (tR = 3·72 min), 1 (tR = 10·61 min) and 3 (tR = 15·56 min).

Figure 2.

HPLC-DAD analysis detected at 254 nm of the crude ethyl acetate extract from Streptomyces sp. WR1L1S8 cultured in SCA medium 50% seawater; (gradient elution: 0–10 min, 30% acetonitrile-water; 10–30 min, 100% acetonitrile 100%). Peak/pure compound assignments: tR = 2·21 min: 4;tR = 3·72 min: 2;tR = 10·61 min: 1;tR = 15·56 min: 3.

In the metabolite profiles, the evaluation was expressed as the relative percentage related to the most intense peak area, taking into account the peaks at 10·61, 6·84, 3·72 and 15·56 min. In particular, the main product observed under the selected SCA medium culture conditions was 1 (tR=10·61 min), which was completely absent for the organism cultured on TSB, ISP2, SM and GYEA media. The data reported in Fig. 3 indicate that the production of 1 was particularly favoured in 50% seawater at pH 7, whereas it was not produced in acidic conditions. On the other hand, the metabolite 3 was highly produced in 100% seawater, more favourably at pH 11. Similar chromatographic profiles were obtained for the presence of metabolites in the crude extracts from the cultures under different pH conditions, both with and without neutralization workup.

Figure 3.

Production of metabolites from WR1L1S8 strain cultured in SCA medium under different concentration of seawater (a) and pH values (b). Plotted data are means ± SD of three replicates per treatment. (image_n/jam12360-gra-0001.png) Compound at t= 3.72 min; (image_n/jam12360-gra-0002.png) Compound at t= 6.84 min; (image_n/jam12360-gra-0003.png) Compound at t= 10.61 min and (image_n/jam12360-gra-0004.png) Compound at t= 15.56 min.

The online HPLC-MS analysis carried out on the crude ethyl acetate extract allowed the determination of molecular mass of the products corresponding to each peak. Coupled to the peak at tR = 2·21 min, the ion at m/z 269 in MS spectrum recorded in positive ion mode corresponded to [M + Na]+ ion of compound 4, as supported by LC-MS injection of the pure phaeochromycin E.

The three compounds eluted in correspondence of the chromatographic peaks at t= 3·72, 6·84 and 10·61 min were isomeric, exhibiting the same molecular mass, as deduced by the ion at m/z 329 corresponding to [M–H] in MS spectrum recorded in negative ion mode and by the presence of the signals at m/z 331 and 353 corresponding to [M + H]+ and [M + Na]+, respectively, in positive ion spectrum (Fig. 4). By injection under the same conditions of compound 2, previously purified and structurally characterized (Djinni et al. 2013), it was possible to establish that the peak at tR = 3·72 min corresponded to phaeochromycin B. Similarly, the peak at tR = 10·6 min was assigned to compound 1.

Figure 4.

Online LC-MS analysis of crude ethyl acetate extract from Streptomyces sp. WR1L1S8: chromatographic profile (at the top); the three chromatographic peaks showing the same MS spectra, reported in negative (in the middle) and in positive ion mode (at the bottom).

The chromatographic peak at tR = 15·56 min was associated with an MS signal at m/z 311 corresponding to [M–H], whereas [M + H]+ and [M + Na]+ ions at m/z 313 and 335, respectively, were observed under positive mode detection (Fig. 5). These data were indicative for a structure lacking of one water molecule in comparison with the previous compounds, as further supported by a higher retention time in reversed stationary phase, in line with a less polar molecule. It was established that this metabolite corresponded to phaeochromycin C (3) by LC-MS analysis of the pure compound and fragmentation experiments (Djinni et al. 2013).

Figure 5.

Online LC-MS analysis of pure metabolite 3 from Streptomyces sp. WR1L1S8: chromatographic profile (at the top), and corresponding MS spectra in negative (in the middle) and in positive ion mode (at the bottom).

The peak at tR = 21·3 min (Fig. 2) corresponded to an intense signal at m/z 621 in ESI–MS spectrum detected in negative ion mode.

Discussion

The isolation of bacteria associated with marine organisms has a restriction due to the fact that <1% can be cultured (Webster et al. 2001). In addition, several methods and media have been recommended for the isolation of actinomycetales from different environments, as well as the use of appropriate isolation media is crucial for improving the recovery of actinobacteria.

All of the isolated strains were found to be affiliated within the genus Streptomyces, in agreement with the reports by several authors. In particular, Fenical and Jensen (2006) reported that the genera belonging to Micromonospora, Rhodococcus and Streptomyces are the most dominant actinobacterial genera in marine environments.

The work here reported is the first chemical screening on endophytic actinomycetales isolated from Fucus sp. algae. The bioactive compound 1 (Djinni et al. 2013) was isolated from the screening of only 22 strain isolates, representing a positive results, due to the fact that the search for novel antibiotics among hundreds of soil actinobacteria typically provides mainly the isolation of already known compounds (Bérdy 2005).

It is noteworthy that this analysis has involved only a subset of the Streptomyces obtained from this algal genus, suggesting a potential for the discovery of other natural products.

As reported in Tables 1 and S1, 100% of the isolated strains from Gausse and SM media showed antibacterial activity, whereas 66·66 and 42·85% obtained from chitin and SCA, respectively, were bioactive. This variation likely reflected the effect of media composition, which is consistent with previous observations (Webster et al. 2001).

The cell wall of strain WR1L1S8 contained ll-diaminopimelic acid and glycine, and whole-cell hydrolysates contained galactose as major sugar, indicating that this strain had a cell wall chemotype IC (Lechevalier and Lechevalier 1970). According to Lechevalier et al. (1977), the phospholipid composition of the strain contained a pattern type II actinomycetes strains which strengthened its belonging to Streptomyces genus.

The physiological properties of the strain WR1L1S8 are shown in Table 2. This strain produced melanoid pigments on both ISP6 and ISP7 media. It was able to degrade adenine, guanine, hypoxanthine and Tween-80, but did not degrade inositol, raffinose, casein or gelatine. Growth occurred at 28 and 37°C at all tested pH values. The strain WR1L1S8 was able to utilize histidine, alanine threonine, proline, serine and glycine as nitrogen sources, but not cystein, tyrosine and tryptophane. Among the tested compounds, the strain utilized practically all the carbon sources.

Based on these evidences, it was possible to establish chemotaxonomic properties typical of the genus Streptomyces.

Regarding the WR1L1S8 strain identification as Streptomyces sp. from 16S rRNA gene sequence, the representative species was recently isolated from sediments of the Sundarbans mangrove forest in India and reported to produce 2-allyloxyphenol (Arumugam et al. 2011). Another strain within the genus Streptomyces identified as Streptomyces phaeochromogenes was isolated by Ritacco and Eveleigh (2008) and reported to produce phaeochromycin metabolites, similarly to the strain here investigated.

The influence on growth and antibiotic production of Streptomyces sp. WR1L1S8 strain by environmental factors such as culture medium, fermentation process, seawater concentration, pH value and incubation period was investigated. This effect was analysed by the chromatographic profile of secondary metabolites present in the crude extracts obtained from different culture conditions. Streptomyces sp. WR1L1S8 strain showed a faster growth on freshwater SCA medium than on 100% seawater SCA and displayed a higher antibacterial activity in 50% seawater SCA medium. It is noteworthy that the best conditions of growth (ISP2, freshwater, pH 11, starch, glucose, casein) were different from the ones able to produce a higher level of bioactive metabolites (SCA, 50% seawater, pH 7, starch, casein), as also observed by Wang et al. (2011b) and Yu and Keller (2005). Indeed, the availability and source of carbon had a substantial effect on the production of antibiotics and morphological development; for example, glucose blocks production of actinorhodin by Streptomyces coelicolor, chloramphenicol by Streptomyces venezuelae, cephamycin by Streptomyces clavuligerus, erythromycin by Saccharopolyspora erythraea and streptomycin by Streptomyces griseus (Van Wezel and McDowall 2011). Furthermore, numerous studies have shown that the source of nitrogen can influence the production of antibiotic molecules, in particular the yield of many, but not all, secondary metabolites was reduced by sources of nitrogen that are favourable to growth. Van Wezel and McDowall (2011) proposed one interpretation of this tendency, based on the fact that supplying a good source of nitrogen, most of the available carbon can be used for growth, generating biomass, so that accordingly less carbon is readily available for secondary metabolism when starvation occurs. No antibiotic activity was detected using TSB agar medium in the WR1L1S8 strain culture, as well as a very weak activity was obtained in submerged fermentation process. However, SCA and solid-state fermentation in the presence of 50% seawater at neutral pH were successful for antibiotic production. These results are in line with reports on differences in solid and liquid media fermentation able to affect the production of metabolites (McDaniel et al. 1994). These evidences could be explained by the fact that the micro-organisms associated to marine surfaces may also require conditions that resemble their native environment in order to produce the maximum amount of bioactive compounds. However, it must be pointed out that these results were based on the single-factor designed experiments, and the optimal conditions for the strain growth and bioactivity might be slightly different when multiple factors are concurrently taken into account.

Metabolite profiles of the strain under investigation in different culture conditions (nutrient culture media, concentration of seawater and pH) showed clear differences related to different treatments, in line with Yu and Keller (2005) who reported how even small changes in the culture medium may not only impact the quantity of certain compounds, but also regulate the general metabolic profile of micro-organisms.

Metabolite profiles here reported have been efficiently acquired by HPLC-ESIMS technique (Xing et al. 2007; Andreo et al. 2012) carried out directly on the crude extracts obtained under various culture conditions. By injecting the pure metabolites previously characterized (Djinni et al. 2013), it was possible to assign the peaks present in the chromatographic profile to compounds 1–4. When the fraction containing 1 eluted in analytical conditions at tR = 10·61 min was injected again, an additional minor peak at t= 6·84 min was observed (whose presence could be also detected in Fig. 5), imputable to an isomeric compound of 1, because showing the same molecular mass as established by data reported in Fig. 4. The MS/MS experiment on [M–H] ion at m/z 329 by direct infusion of the compound eluted at tR = 6·84 min showed the same fragmentation pattern observed for compound 1 (with signals at m/z 311, 285, 267 and 243), different from the one reported for the isomeric compound 2 (at m/z 285, 267 and 257), (Djinni et al. 2013). The ion at m/z 311 corresponding to the loss of a molecule of water could be easily explained for a structure like 1, suggesting that the compound eluted at tR = 6·84 min was probably the tautomeric α-hydroxy-pyrone of the γ-form 1. For the metabolite eluted at tR = 21·3 min, the corresponding ESI-MS spectrum in negative ion mode gave a signal at m/z 621, attributable to the [M–H] ion for a compound showing a sort of dimeric structure if compared to 1. Otherwise, phaeochromycin F having two symmetric hydroxyl-pyrone units has been reported as a new polyketide metabolite from Streptomyces sp. DSS-18 (Li et al. 2008).

In conclusion, these results suggest that marine algae–actinobacteria associations are a particularly promising group from which novel metabolites can be elicited. The isolation of strains with antimicrobial activity indicates that marine seaweeds may represent an ecological niche, which harbours a largely untapped microbial diversity and a yet unexploited potential for new secondary metabolites. The presence of the bioactive compound 1 in the metabolite profile in the crude extracts of the strain under different growth conditions has been efficiently evaluated by the coupled HPLC-DAD/ESI-MS technique.

Acknowledgements

The activity of I.D. in Italy was supported by a PhD grant in the Erasmus Mundus–Averroes project. The authors would like to thank Adriano Sterni, Mario Rossi and Nicola Bazzanella, University of Trento, for their technical support in mass spectrometry, HPLC-ELSD chromatography and SEM analysis, respectively.

Conflict of interest

No conflict of interest declared.

Ancillary