By the technique of differential display of transcripts, in total RNA from a field specimen of S. domuncula 32 fragments were identified that were strongly upregulated, compared to the RNA from a specimen kept in the aquarium. After finding by this technique, the corresponding ESTs were also identified in the sponge database [17]. Then the cDNAs were completed. Blast similarity search of the deduced polypeptides revealed that two of those sequences shared highest sequence relationship with the A. suum ASABF-γ peptide from (BAC00498.1; [23]), and that the sequence from S. domuncula was complete; Figure 1A. Therefore, the sponge sequence was termed ASABF peptide (ASABF_SUBDO), deduced from the SDASABF cDNA. The similarity/identity score of the deduced peptide was 36%/27%. The open reading frame (ORF) of the 467 nt long cDNA ranges from nt_47–49 to nt_{296–293(stop)} and comprises 83 aa contributing to a theoretical size of 9029 Da and an isoelectric point (pI) of 9.28. The peptide possesses one distinct signal sequence at the N-terminal end from aa₁ to aa₁₉ (SignalP 3.0 Server; Technical University of Denmark); Figure 1A. The length of this segment is identical to that predicted for the A. suum ASABF-γ peptide [23]; taking this sequence as reference, the predicted mature region starts at aa₂₀ and ranges to the C-terminus. Hence, the theoretical M_w (molecular weight) is 7026, and the corresponding pI is 9.24. Therefore, the S. domuncula ASABF is a cationic protein, with a pI almost identical to that calculated for the mature ASABF from A. suum with 8.9 [23]. The S. domuncula ASABF sequence shows the characteristic consensus of the nematode CSαβ motif reading C-X(3,18)-C-X(3)-C-X(7,9)-[GS]-X-CX(4-13)-CXC [14]. Like the nematode peptide the mature S. domuncula ASABF sequence includes 8 Cys moieties (aa₂₃, aa₃₀, aa₃₄, aa₃₉, aa₄₄, aa₄₉, aa₅₁, aa₅₄) that allow the formation of disulfide bridges as highlighted in Figure 1A, and as predicted [21]. Secondary structure prediction was performed according to Cole et al. [26]. The characteristic CSαβ (cysteine-stabilized α-helix and β-sheet) structural motif [14] has been predicted according to Chou and Fasman [27]; Figure 1A. The N-terminal α-helix includes the first two Cys residues while the second β-strand comprises the last two Cys residues. No other domain could be identified in the sponge peptide [28].

The complete cDNA (LBASABFr), encoding the L. baicalensis ASABF-related peptide (ASABFr_LUBAI), has a length of 558 nts and an ORF spanning the segment nt_27–29 to nt_{321–323(stop)}, predicting a 98 aa moieties long peptide. The theoretical M_w is 10,665 and the pI was predicted with 7.61. Based on comparisons the signal peptide ranges from aa₁ to aa₂₂, followed by the mature peptide with an M_w of 8,172 and a theoretical pI of 7.86. This only slightly cationic predicted peptide has only 4 Cys residues.

In order to estimate the phylogenetic relationship of the sponge peptides to other metazoan sequences, for the alignment the highest related A. suum, Caenorhabditis remanei, Caenorhabditis briggsae, and the C. elegans sequences of the ASABF family, as well as the ASABF-related sequence from the seahorse Hippocampus kuda have been chosen [29]; additionally the two human defensins (the beta defensin-2 and the neutrophil alpha defensin-3 preproprotein) and the only metazoan theta defensin (from Papio anubis) were selected. The phylogenetic tree was established by using the plant ASABF-related putative defensin AMP1 protein from Arabidopsis thaliana as outgroup. The construction of the tree revealed that the ASABF sequences cluster together, including the sponge peptide (ASABF from S. domuncula) which forms the basis for that branch (Figure 1B). Separated from them are the human and baboon defensins. Interesting is the finding that the L. baicalensis ASABF-related peptide clusters together with the defensins that comprise likewise less Cys moieties in their mature peptide [23].

Recombinant protein was prepared in the yeast Pichia pastoris. This eukaryotic expression system allows post-translational modifications such as disulfide bonding, which is required for proper folding of defensins. Accordingly, the complete ORF of SDASABF, without the putative signal sequence (i.e., aa₂₀ to aa₈₃), was expressed as fusion to a 6xHis tag. Following Ni²⁺-NTA metal-affinity purification (Figure 2Ab), the recombinant protein (8.0 kDa (minus His tag: 7.1 kDa)) was used for immunization of rabbits. Then, specificity of the polyclonal anti-ASABF serum (PoAb-rASABFSUBDO) was assessed on Western blots (Figure 2Ba), showing a band at the expected size of the recombinant protein. In control experiments, 100 μL of the antibodies were adsorbed to 20 μg of the recombinant protein prior to their use. Consequently, no band could be observed on the immunoblots after application of these blocked antibodies (Figure 2Bb).

In a parallel set of experiments, the PoAb was applied to detect S. domuncula ASABF in protein extracts of sponge specimens that had been sampled from the sea (Figure 2Bc), or that had been cultured in aquaria for 12 months (Figure 2Bd). Thus, following SDS-PAGE and immunoblotting of identical protein concentrations, a band was detected with a size corresponding to the processed, mature variant of ASABF, lacking the signal sequence (i.e., 7.0 kDa). The band intensity was considerably stronger in extracts of field specimens than in those of aquarium specimens.

The recombinant ASABF caused strong anti-microbial activity in the in vitro assay system (Table 1). Especially strong was the effect towards Gram-positive bacteria (S. aureus, B. subtilis, and M. luteus) with minimum inhibitory concentrations (MIC) between 2 and 5 μg/mL, while the activity against Gram-negative bacteria (P. aeruginosa and E. coli) was considerably lower with concentrations between 12 and 17 μg/mL. Interestingly, ASABF also caused distinct inhibitory effects on growth of eukaryotic fungi (C. albicans and A. niger) with MIC concentrations between 9 and 13 μg/mL. In order to prove that the effect measured was specific, the microorganisms were exposed to recombinant ASABF that had been adsorbed with antibodies prior to the addition to the cultures; such samples were found to be inhibitory only at MIC above 20 μg/mL.

As a convenient assay to study hemolysis human erythrocytes have been used and exposed to recombinant ASABF. The cells were exposed to the peptide and the percentage of the lysed erythrocytes was determined on the basis of the released hemoglobin. Under the assay conditions described under “Experimental Section” it was seen that the recombinant, non-adsorbed ASABF caused at a concentration of about 1 μg/mL 35% hemolysis. The degree of lysis increased to 76% at a concentration of 10 μg/mL (Figure 3). As a control, the ASABF peptide was adsorbed to anti-ASABF serum (PoAb-rASABFSUBDO), as described under “Experimental Section”. Using this ASABF sample the degree of lysis was very low and amounted to about 10% at the high concentration of 10 μg/mL.

If sponge specimens were kept in aquaria at lower aeration rate or in the presence of Fe⁺⁺ chelator Dip the animals started to develop apoptotic spots/areas on their surface, as described [24]. Those regions were initially (after 1 day) marked by a deeper-red color. At a later stage the apoptotic surface regions became unstructured to meander-like. Previously, those regions had been identified to undergo apoptosis, as had been proven by the “Cell Death Detection ELISA test system” [24]. At day 2 these apoptotic regions were colonized by the gastropod Bittium sp. that had been attracted by quinolinic acid, a metabolite that is formed after induction of the 3-hydroxyanthranilate 3,4-dioxygenase (HAD) gene [24]. At days 2 to 3 the specimens of Bittium sp. (1 to 6 per 5 mm²) grazed on the apoptotic areas (Figure 4A); no gastropods were seen on non-apoptotic surfaces (not shown). As observed before, the gastropods ate (Figure 4B,C) and, by that, removed the apoptotic tissue (Figure 4D). After day 7 the previously apoptotic regions had been removed.

Samples from a control specimen and from an animal kept for 1 day in Dip were analyzed by in situ hybridization. Using the DIG-labeled anti-sense ssDNA (SDASABF) probe the extent of reaction to the tissue structures was low in control sections (Figure 5A,B). In contrast, the reaction of the labeled probe with tissue slices from Dip-exposed animals was intense (Figure 5D,E); at a higher magnification the reaction with cell structures was evident (Figure 5F). One control was performed by hybridizing a sample from the Dip-treated animal with a labeled sense ssDNA probe; there, only background signals were recorded (Figure 5C).

Tissue samples from apoptotic areas on the surface of Dip-treated animals, as well as from control animals were taken and subjected to qRT-PCR analysis. For a quantitative assessment steady-state gene expressions of HAD (marker for quinolinic acid synthesis), the caspase (formation of apoptotic tissue [30]) as well as of ASABF (ASABF synthesis) were chosen. The tissue samples for the determination of the respective transcript levels in apoptotic areas were taken from the rim region of those regions. The determinations were performed at the beginning of the exposure (day 0), as well as after 1 to 9 days. The results showed that the expression level of HAD at the beginning of the experiments was approximately 2.1 × 10⁻³ with respect to the expression of the house-keeping gene tubulin (Table 2). While in the controls the expression level did not significantly change during the 9-day incubation period, a strong increase in the steady-state expression of HAD was measured in the Dip-treated animals. The maximum of expression was seen already at day 2 (27.1 × 10⁻³); during the end of the experiment (after 9 days) the expression level returned to normal (3.1 × 10⁻³). Focusing on ASABF expression, likewise no change of the steady-state expression was observed in the controls (around 1.3 × 10⁻²). In the Dip-treated specimens a strong and significant increase was found already after day 1 (2.8 × 10⁻²), while the maximum was seen 4 days after exposure to Dip (9.8 × 10⁻²). Subsequently, the level of expression decreased. Finally, the expression of the caspase gene likewise increased in the Dip-treated animals from 1.5 × 10⁻⁴ (day 0) to 12.4 × 10⁻⁴ at day 3 (Table 2). In the control tissue no significant changes were measured. A diagrammatic sketch, summarizing the expression levels in apoptotic tissue from the surface of Dip-treated specimens, as well as from controls, is given in Figure 6. There, also the surface texture of a 3-day-old apoptotic area with a grazing gastropod is shown with its uneven appearance in contrast to the smooth/plain surface in the control.

For the expression studies in the above described series of experiments tissue from the rim regions had been selected. To obtain a more detailed view of the expression levels of HAD, ASABF and caspase tissue samples were taken from within the apoptotic area. Such a spatially resolved study was only possible after a more progressed stage of apoptosis, 3 days after Dip exposure. In comparison, tissue samples from apoptotic regions after their ablation through the gastropods were analyzed (Figure 7). The data revealed that at day 3 the increase in the steady-state level of expression of the HAD gene and the ASABF gene was 4.8-fold and 5.9-fold, respectively, while the increase of the expression level for the caspase gene was much higher with 12.2-fold.

The above outlined results indicate that the drop of the HAD expression coincided (after about 4 days after exposure to Dip) with the maximum of the expression of the ASABF gene, encoding the antimicrobial peptide (Figure 6). In addition, it is seen that after about 5 days the gastropods disappeared from the previously apoptotic areas. The reason for the elimination of the gastropods is not quinolinic acid, since this dicarboxylic acid is not toxic for those epibionts. Therefore, it was intriguing to check if it was the ASABF peptide which may have caused a toxic effect. Accordingly it was necessary to test if the recombinant peptide affected the viability of the gastropods.

As outlined in the assay conditions described under “Experimental Section”, the gastropods were exposed for 96 h to ASABF in a serial concentration range. After incubation the mortality in the assays with ASABF were determined to be; at 3 μg/mL: 3.5 ± 2%, 10 μg/mL: 26.4 ± 8% and 30 μg/mL: 75.7 ± 12%. In the controls (untreated animals) 100% of the animals survived. According to the standards the toxicity of the ASABF towards the gastropods should be assessed as “highly toxic” (US EPA 2011); Figure 8.

The sources of the materials used for this study were given previously [43]. In addition, Dip (2,2′-dipyridyl) was purchased from Sigma/Aldrich (St. Louis, MO; USA), and PCR DIG (digoxigenin) Probe Synthesis Kit, anti-DIG AP Fab fragments, as well as disodium 2-chloro-5-(4-methoxyspiro {1,2-dioxetane-3,2′-(5′-chloro)tricyclo[3.3.1.1^3,7]decan}-4-yl)-1-phenyl phosphate (CDP-Star) were obtained from Roche (Mannheim; Germany).

Specimens of the marine sponge S. domuncula (Porifera, Demospongiae, Hadromerida) were sampled in the Northern Adriatic near Rovinj (Croatia), and then kept in aquaria in Mainz (Germany) at a temperature of 17 °C for more than 12 months. Extracts were prepared from tissue samples by treatment with four parts of a 0.4 M Tris-HCl buffer (pH 8.0; supplemented with 50 mM NaCl, 0.5 mM EDTA and 5 mM MgCl₂). After grinding, and stirring of the suspension at 2 °C for 1 h a clear supernatant was obtained after centrifugation (20,000 × g, 30 min, 2 °C); this was collected and contained between 3 and 7 mg of protein per mL. This extract was diluted down to 2 mg/mL and used for the experiments.

The experiments were performed in 5 L (artificial sea water [24]) aquaria, containing 3 sponges each. Apoptosis was induced in sponges by keeping the specimens in the presence of 50 μg/mL of the Fe⁺⁺ chelator Dip [44] for 1 day. Under these conditions the specimens responded with the formation of speckled patches on their surfaces, as described earlier [24]. By application of the “Cell Death Detection ELISA test system” (Roche) apoptosis was proven in these regions [24]. To each aquarium, both controls and Dip (2,2′-dipyridyl) supplemented ones, approximately 100 specimens of the gastropod/mollusc Bittium sp. (Mollusca, Gastropoda, Cerithiidae, Bittium) were added. Already 1 day later the apoptotic patches on the sponge surfaces were covered by 1 to 6 specimens of Bittium sp. per 5 mm² (Dip-treated), while on the surfaces of specimens kept in absence of Dip (controls) no gastropods were seen. As described before, the gastropods ablated the apoptotic tissue within 3 days. Seven days after the beginning of the experiments no signs of apoptotic tissue could be seen. Tissue samples were taken at the beginning of the experiments as well as after 1 day, 2 days, 3 days, 4 days, 5 days, 7/8 days, and 9 days from Dip-treated (speckled-type regions) and from control regions, and analyzed for gene expression.

Bittium sp. (Gastropoda: Cerithiidae) are common grazers, epizoons [45], abundantly seen on the surfaces of S. domuncula apoptotic specimens. The gastropods were collected, transferred to nets (mesh size 44 μm) and could be kept in aquaria, for over 30 days, without losing more than 10% of the specimens. Routinely, the gastropods were used for the experiments 6 days after collection from the apoptotic specimens.

S. domuncula ASABF: For the identification of the cDNAs encoding putative defensin-like molecules the technique of differential display was employed according to Müller et al. [4]. Total RNA was extracted from S. domuncula specimens that had been sampled from the sea or had been cultivated in aquaria for more than one year. Specimens that had been taken from nature showed a higher load of bacteria and, in turn, were more likely to express those polypeptides in a wider array of antibacterial defense molecules and at higher expression levels than aquarium animals [46]. The cloning procedure, given in details before [4] is outlined in brief: 1.8 μg of total RNA of aquarium animals or of field specimens was reverse transcribed using the anchored oligo(dT) primer T₁₁CC and SuperScript III Reverse Transcriptase (Invitrogen, Darmstadt; Germany). Following first-strand cDNA synthesis, the samples were diluted 1:10 with H₂O and, then, subjected to polymerase chain reaction (PCR). For this reaction, Platinum Taq DNA polymerase (Invitrogen) was employed in combination with the IR-labeled arbitrary primer 5′-GTGATCGCAG-3′ and an oligo(dT) primer. After initial denaturation at 94 °C for 2 min, cDNA was amplified during 38 cycles each, at 94 °C for 30 s, 42 °C for 120 s, and 72 °C for 30 s, followed by a final incubation of 10 min at 72 °C. Subsequently, amplicons were size-separated during polyacrylamide gel electrophoresis, using an automated DNA sequencer (LI-COR 4300 DNA Analysis System). Then, cDNAs between 250 and 400 nucleotides (nt) of differentially expressed transcripts were excised using the Odyssey (LI-COR) infrared imaging system and subsequently reamplified by PCR. Finally, the products were sub-cloned in the pGEM-T vector (Promega) and sequenced. Complete cDNAs were obtained by primer walking [47]. Thus, the clone SDASABF was identified (among 32 differentially expressed transcripts found within field animals), comprising 467 nts (excluding the poly(A) tail) with an ORF encoding the putative protein ASABF_SUBDO.

L. baicalensis ASABF: Following the identification of S. domuncula ASABF, >2500 L. baicalensis ESTs [28] were screened for related sequences. One clone was discovered, comprising 558 nts with considerable sequence homology to SDASABF, which was termed LBASABFr. The ORF was found to be complete.

Homology searches were conducted through servers at the European Bioinformatics Institute, Hinxton, United Kingdom and the National Center for Biotechnology Information (NCBI), Bethesda, MD [29]. Multiple alignments were carried out with ClustalW version 1.6 [48]. Phylogenetic trees were constructed on the basis of aa sequence alignments applying the Neighbor-Joining method to distance matrices that were calculated using the Dayhoff PAM matrix model [49,50]. The degree of support for internal branches was further assessed by bootstrapping [51]. The graphical output of the bootstrap figures was produced through the “Treeview” software [52] and GeneDoc [53]. Potential domains, subunits, patterns, and transmembrane regions were predicted searching the Pfam [54] or the SMART database [55].

cDNA of S. domuncula ASABF, coding for the complete mature protein (aa₂₀ to aa₈₃), was cloned as gene of interest into the pPIC3.5 yeast expression vector (Invitrogen) via SnaB I and EcoR I. At the 5′ terminus, an ATG initiation codon was placed within the context of a yeast translation initiation sequence ([56] and according to the manufacturer’s protocol). Additionally, the cDNA contained a 3′ 6xhistidine (His) tag coding sequence (to facilitate metal-affinity chromatography purification of the recombinant protein), followed by a stop codon. After digestion with Sal I, chemically competent Pichia pastoris GS115 cells were transformed with the linearized construct. Subsequently, transformants were selected for their ability to grow on histidine-deficient medium. For recombinant protein expression, transformants were cultured for 84–96 h at 30 °C in appropriate medium with 0.5% (v/v) methanol. Then, cells were harvested, transferred into Breaking Buffer (50 mM sodium phosphate pH 7.4, 1 mM PMSF (phenylmethylsulfonyl fluoride), 1 mM EDTA, 5% glycerol), and homogenized (Precellys; PeqLab, Erlangen; Germany). Finally, the recombinant protein (termed rASABF_SUBDO) was purified through nickel-nitrilotriacetic acid (Ni²⁺-NTA) metal-affinity chromatography and checked through SDS-PAGE.

The purified, recombinant ASABF protein (rASABF_SUBDO) was used for the production of polyclonal antibodies (PoAb), as described [8,57]. Thus, female rabbits (White New Zealand) were immunized thrice with 20 μg/boost. Following collection of the PoAb (termed PoAb-rASABFSUBDO) the titer was determined to be 1:5,000. For control experiments, 100 μL of PoAb were incubated with 20 μg of the antigen during an incubation period of 1 h (4 °C). Only then (adsorbed antibody preparation) these blocked antibodies were applied to immunodetection assays.

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was routinely performed as follows. Samples of 6 μg of protein were mixed with loading buffer (Roti-Load; Carl Roth, Karlsruhe; Germany), boiled for 8 min, and subjected to SDS-PAGE (15% acrylamide and 0.1% SDS) as described [58]. The gels were stained with Coomassie brilliant blue. The protein size standard “Dual Color” (Carl Roth) was used to estimate protein sizes. For Western blot analyses, size-separated proteins were transferred to PVDF-Immobilon membranes. Then, the membranes were blocked at room temperature with blocking solution (Roche; 1% (v/v) in TBS-T buffer; 20 mM Tris-HCl pH 7.6, 137 mM NaCl, 0.1% (v/v) Tween-20). Following consecutive incubations with PoAb-rASABFSUBDO (1:3,000 dilution in TBS-T, supplemented with 0.1% (v/v) Blocking solution), alkaline phosphatase (AP)-conjugated species-specific secondary antibodies (1:4000 dilution), and 4-nitro blue tetrazolium chloride (NBT)/5-bromo-4-chloro-3-indolyl phosphate (BCIP) (Invitrogen), proteins were detected colorimetrically. The “Ultra-Low Range Marker” (Sigma; M3546) was used to determine the apparent size of the peptide.

Quantitative real-time RT-PCR (qRT-PCR) was applied to determine the steady-state expression of the gene encoding ASABF (SDASABF), as recently described in detail [59,60]. Following extraction of tissue RNA was treated with DNAse to eliminate contaminating genomic DNA. First strand cDNA synthesis was performed by SuperScript III Reverse Transcriptase in a reaction mixture containing 5 μg of total RNA, dNTPs, oligo(dT)₁₈, and reverse transcriptase buffer (Invitrogen) at 42 °C for 1 h. After enzyme inactivation (65 °C, 15 min), qPCR experiments were performed in an iCycler (Bio-Rad), using 1/10 serial dilutions in triplicate as described [61]. After appropriate dilution, 2 μL of the reaction mixture were employed as a template for 30 μL qPCR assays. Each reaction contained “Absolute Blue SYBR Green” master mixture (ABgene, Hamburg; Germany) and 5 pmol of each primer. All reactions were run with an initial denaturation at 95 °C for 3 min, followed by 40 cycles each, of 95 °C for 15 s, 62 °C for 30 s, 72 °C for 35 s, and 80 °C for 20 s. The following primers were used for amplification of SDASABF: Fwd: 5′-GCAAAGCTGGCAGAGTTGGAT-3′ (nt₁₁₄ to nt₁₃₄) and Rev: 5′-AGATTTTGAACTTGTTGAAAGG-3′ (nt₂₄₅ to nt₂₂₄); the size of the fragment obtained was 132 bp. For the amplification of the S. domuncula caspase gene SDCASPL (NCBI accession number AM392430.1; [8]) Fwd: 5′-GGACAAACAGATGTACACAGAG-3′ (nt₅₉₇ to nt₆₁₈) and Rev: 5′-GATGTAGTTTCCTTCTGGAGGT-3′ (nt₇₃₂ to nt₇₁₁) were applied, resulting in a fragment of 136 bp. Finally, the expression level of the S. domuncula 3-hydroxyanthranilate 3,4-dioxygenase (HAD; AJ298053.1) was quantified using the primer pair Fwd: 5′-CTCAGGTACTATGTTGATGGTA-3′ (nt₃₆₉ to nt₃₉₀) and Rev: 5′-TGGACTTCCATTCTTGTGTTCC-3′ (nt₃₀₀ to nt₂₇₉); size of the fragment, 132 bp. The expression of the house-keeping gene β-tubulin (AJ550806; [62]) was determined with the following primers, Fwd: 5′-AACCGCTGTTTGCGACATCC-3′ (nt_1,126 to nt_1,145) and Rev: 5′-CAATGCAAGAAAGCCTTTCGCC-3′ (nt₁₂₆₆ to nt₁₂₄₅); fragment size, 141 bp. The threshold position was set to 50.0 RFU above PCR subtracted baseline for all runs. Mean Ct values and efficiencies were calculated by the iCycler software. Expression levels of the respective transcripts were correlated with the one for β-tubulin to assess the relative expression levels (E_Tub ^CtTub/E_ASABF ^CtASABF; E_Tub ^CtTub/E_Casp ^CtCasp; E_Tub ^CtTub/E_HAD ^CtHAD); whereby “E” describes the PCR efficiency and “Ct” represents the threshold cycle.

In situ hybridization assays of S. domuncula tissue slices were performed according to Schröder et al. [62]. Briefly, labeling of ssDNA probes (SDASABF) was carried out with the PCR DIG Probe Synthesis Kit (Roche), using either forward primer 5′-ATGAATGCAAAGTTGTGCACTC-3′ (nt₄₇ to nt₆₈; for the sense probe) or reverse primer 5′-CAAGCGCGTTCTTCATCGAC-3′ (nt₂₉₇ to nt₂₇₈; for the anti-sense probe) in combination with a linearized template. The resulting labeled ssDNA probes spanned nt₄₇ to nt₂₉₇ of the SDASABF cDNA. Then, cryosections (8 μm thickness) were fixed with paraformaldehyde, washed with PBS (10 mM sodium phosphate pH 7.4, in 150 mM NaCl) at room temperature, and incubated with Proteinase K. For rehydration, the sections were hybridized with the anti-sense or sense probe (negative control) overnight at 40°C. After blocking, the sections were incubated for 1 h at 37 °C with anti-DIG Fab fragments (conjugated to alkaline phosphatase) and, then, washed twice. For visualization of the signals, the dye reagents BCIP/NBT (5-bromo-4-chloro-3′-indolyphosphate/nitro-blue tetrazolium) were used in a Tris-buffer (100 mM Tris-HCl pH 9.5, 100 mM NaCl, and 50 mM MgCl₂).

Microbicidal/antibacterial assay was performed as earlier described [63,64] using the Mueller Hinton Medium (DIFCO Invitrogen, Karlsruhe; Germany) at pH 7.0. The following microorganisms were tested: the Gram-positive bacteria Staphylococcus aureus ATCC 25293, Bacillus subtilis ATCC 6633, and Micrococcus luteus ATCC 9341, the Gram-negative bacteria Pseudomonas aeruginosa ATCC 9027, and Escherichia coli ATCC 25922), and the fungi Candida albicans ATCC 10231 and Aspergillus niger ATCC 16404. The minimal inhibitory concentration (MIC) was determined and the values were given in μg/mL.

Human type A erythrocytes were used for the hemolytic assay; they came from healthy individuals. The detailed assay has been described earlier [39,65]. PBS-washed erythrocytes were used to prepare a 10% (v/v) suspension by using an isotonic glucose phosphate buffer (1 mM K-phosphate buffer, pH 7.0; supplemented with 287 mM glucose). Serial dilutions of recombinant non-adsorbed or adsorbed ASABF were prepared to a final volume of 200 μL in 96-well V-bottomed microtiter plates (Nunc/Thermo Electron, Langenselbold; Germany). To reach total hemolysis the samples were subjected to 1% (v/v) Triton X-100. After incubation for 1 h (37 °C) the plates were centrifuged (10 min at 2000 × g at 20 °C) and a 150 μL aliquot from the supernatant was taken and used for absorbance measurements at 450 nm. The percentage of hemolysis was calculated as described [65] ( ( A 450   of the ASABF - treated sample - A 450   of buffer - treated sample ) / ( A 450   of Triton X - 100 - treated   sample - A 450   of buffer - treated sample ) ) × 100 %.

The gastropods, collected from one affected specimen of S. domuncula, were exposed at 17°C to 3 μg/mL, 10 μg/mL, or 30 μg/mL of recombinant ASABF in seawater, or remained without this peptide. The animals, 10 specimens per assay, were kept in parallel in aerated aquaria (volume of 200 mL) for 96 h. Subsequently the gastropods were transferred to seawater without the peptide and culture continued for 96 more hours. Then the living gastropods were counted. The means ± standard deviations of three series of experiments were calculated by using the paired Student’s t-test [66].

For the quantification of protein, the Bradford method ([67]; Roti-Quant solution-Roth) was used.