Distinct signalling pathways promote phagocytosis of bacteria, latex beads and lipopolysaccharide in medfly haemocytes

Abstract

In insects, phagocytosis is an important innate immune response against pathogens and parasites, and several signal transduction pathways regulate this process. The focal adhesion kinase (FAK)/Src and mitogen activated protein kinase (MAPK) pathways are of central importance because their activation upon pathogen challenge regulates phagocytosis via haemocyte secretion and activation of the prophenoloxidase (proPO) cascade. The goal of this study was to explore further the mechanisms underlying the process of phagocytosis. In particular, in this report, we used flow cytometry, RNA interference, enzyme-linked immunosorbent assay, Western blot and immunoprecipitation analysis to demonstrate that (1) phagocytosis of bacteria (both Gram-negative and Gram-positive) is dependent on RGD-binding receptors, FAK/Src and MAPKs, (2) latex bead phagocytosis is RGD-binding-receptor-independent and dependent on FAK/Src and MAPKs, (3) lipopolysaccharide internalization is RGD-binding-receptor-independent and FAK/Src-independent but MAPK-dependent and (4) in unchallenged haemocytes in suspension, FAK, Src and extracellular signal-regulated kinase (ERK) signalling molecules participating in phagocytosis show both a functional and a physical association. Overall, this study has furthered knowledge of FAK/Src and MAPK signalling pathways in insect haemocyte immunity and has demonstrated that distinct signalling pathways regulate the phagocytic activity of biotic and abiotic components in insect haemocytes. Evidently, the basic phagocytic signalling pathways among insects and mammals appear to have remained unchanged during evolution.

Introduction

In higher animals, phagocytosis is a protective mechanism, rather than a mode of feeding. In particular, phagocytosis is a receptor-mediated innate immune response against pathogens and also an effective process to eliminate apoptotic cells. Phagocytosis in mammals is mainly achieved by macrophages, monocytes and neutrophils, whereas in insects it is achieved mainly by plasmatocytes or granulocytes.^1–7 Other haemocytes, such as oenocytoids, may also take up pathogens.⁸ The pathogen uptake by a phagocytic cell depends on specific signals. These can be released either directly from the pathogen or generated by the cell surface receptors, which mediate the pathogen recognition.⁹

The process of phagocytosis begins when immune cells make contact with an invader. This activates several signalling pathways in mammals that regulate the formation of phagosomes and target ingestions through a cytoskeleton-dependent mechanism. The signalling pathways activated by the invader in mammalian phagocytes have been extensively studied and include among others tyrosine kinases, serine kinase [mitogen-activated protein kinases (MAPKs)], small GTPases and lipid signalling pathways. However, the signalling pathways participating in phagocytosis in insect haemocytes have not yet been analysed in detail and therefore, there is little information regarding the molecular mechanisms that activate haemocytes and initiate melanogenesis in response to microbial infections.¹⁰ Recently, it has been demonstrated that medfly haemocytes, key mediators of cell-mediated immunity in insects, respond to Escherichia coli lipopolysaccharide (LPS) by the activation of the three major MAPK subfamilies, namely, extracellular signal-related kinase (ERK), c-jun N-terminal kinase (JNK) and p38 in a Ras/Rho-dependent manner, which is similar to mammalian monocytes/macrophages.^1,3,11 This activation regulates, among other things, the E. coli LPS-dependent release, as well as phagocytosis, both in mammalian cell systems and insect haemocytes.^1,3,12,13 However, only in insect haemocytes is the regulated release in response to E. coli LPS a prerequisite for E. coli LPS uptake.^1–3,14 The functional association of E. coli phagocytosis with the activation of MAPKs appears to be the secretion of prophenoloxidase (proPO)-activating proteinases (PAPs), the enzymes that convert the haemocyte surface proPO to the active phenoloxidase (PO), which catalyses the early steps of the pathway leading to melanin formation.² The activation of haemocyte surface proPO via MAPs activation is an important part of innate immunity and hence the process of phagocytosis.² As a result MAPK signalling pathways appear to control both phagocytosis and melanization because they regulate PAP secretion.

The present study is part of a larger investigation of the signalling pathways in medfly haemocytes and their role in the insect immune response. Data on invertebrate systems are very limited^1,15,16 and in mammalian systems phagocytosis can be either dependent on or independent of focal adhesion kinase (FAK)/Src and MAPK pathways^17–19 so this study aimed to clarify further the role of FAK/Src and MAPK pathways in phagocytosis by primary haemocytes from Ceratitis capitata.

Materials and methods

Materials and antibodies

Polyclonal antibodies against FAK (c-903), p^Y397FAK, ERK and c-Src were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Affinity-purified rabbit polyclonal antibodies to, non-p^Y527Src, p^Y527Src, p^Y416Src, as well as to p^{Thr180–Tyr182}p38, p^{Thr202–Tyr204}p44/42 and p^{Thr183/Tyr185}JNK MAPKs, were purchased from Cell Signalling Technology (Beverly, MA). Horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (IgG-HRP), p^{Ser218/Ser222}MEK-1/2 and the rabbit ABC staining system were purchased from Santa Cruz Biotechnology, (Santa Cruz, CA), while goat anti-mouse IgG-HRP was obtained from BD Transduction Laboratories™ (San Diego, CA), and fluorescein isothiocyanate-conjugated goat anti-rabbit IgG (IgG-FITC) was purchased from Molecular Probes, Inc (Eugene, OR). Antibodies against actin, tubulin and goat anti-mouse-FITC were obtained from Sigma (St Louis, MO). U0126 and PD 98059, MEK1/2 inhibitors were obtained from Cell Signalling Technology (Beverly, MA) and SB 202190 and SB 203580, p38 inhibitors, were obtained from Sigma. SP 600125 JNK inhibitor and PP2 (a potent and selective inhibitor of the Src family of tyrosine kinases) were purchased from Calbiochem (Darmstadt, Germany), donkey anti-rabbit IgGs (Amerlex-M magnetic separation reagent) were purchased from Amersham Life Sciences (Amersham, UK) and RGD^{Arg-Gly-Asp-Ser} and RGE^{Arg-Gly-Glu-Ser} peptides and fibronectin were purchased from Sigma. Other materials were obtained as indicated.

FITC-labelled LPS, E. coli, Staphylococcus aureus and latex beads

FITC-labelled LPS and carboxy-modified latex beads were obtained from Sigma. FITC-labelled E. coli (DH10B) and S. aureus (epidermic clinical isolated) were prepared after incubation of 10⁸ heat-killed bacteria with 1 mg FITC, in 0·5 ml 0·5 m Na₂CO₃/0·5 m NaHCO₃ at pH 9·5 for 30 min in the dark. FITC-conjugated E. coli and S. aureus were rinsed three times with phosphate-buffered saline, resuspended in Grace's medium and stored in aliquots at −20°.

Collection of haemocytes and cell viability test

Ceratitis capitata were reared as described previously.²⁰ Isolation and homogenization of haemocytes from third instar larvae were performed according to Charalambidis et al.²¹ In brief, haemolymph was collected and centrifuged at 200 g for 10 min at 4°. Sedimented haemocytes were washed three times with Ringer's solution (128 mm NaCl, 18 mm CaCl₂, 1·3 mm KCl and 2·3 mm NaHCO₃, pH 7·0). The viability of haemocytes was assessed by exclusion of trypan blue dye (Sigma) under a light microscope.

Protein determination

Proteins were determined according to Bradford²² with a modified solution containing 10% (w/v) Coomassie G250 (Merck, Darmstadt, Germany) in 5% (v/v) ethanol, 10% (v/v) H₃PO₄. Optical density was recorded at 595 nm.

Co-immunoprecipitation

Haemocytes were lysed in lysis buffer [50 mm Tris–HCl pH 7·4, 150 mm NaCl, 5 mm ethylenediaminetetraacetic acid (EDTA), 1% Triton X-100, 1 mm sodium orthovanadate, 5 mm NaF, 1 mm phenylmethylsulphonyl fluoride, 10 μg/ml leupeptin and 10 units/ml aprotinin] at 4°. Insoluble material was removed by centrifugation (16 000 g for 15 min at 4°) and supernatant was collected. For immunoprecipitation, 400 μg crude extract protein was incubated with 2 μg anti-FAK, anti-Src and anti-ERK polyclonal antibodies for 2 hr at 4° and then for an additional 1 hr at 25° with an Amerlex-M secondary antibody reagent (Amersham Life Science). The immune complexes were washed four times with TBS (10 mm Tris–HCl, pH 7·5, 100 mm NaCl). Proteins were eluted from the beads by boiling the samples for 3 min in 50 μl electrophoresis sample buffer. Immunoprecipitated proteins were analysed on 10% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and immunoblotted with anti-FAK, anti-Src and anti-ERK polyclonal antibodies.

SDS–PAGE and immunoblot analysis

SDS–PAGE was performed on 10% acrylamide and 0·10% bisacrylamide slab gels, according to Laemmli.²³ Samples with equal amounts of protein were analysed and electroblotted on to Immobilon P polyvinylidene fluoride membranes (Millipore Corp., Billerica, MA). Membranes were incubated in SuperBlock™ blocking buffer (Pierce, Rockford, IL) for 1 hr at 25°. Subsequently, membranes were incubated overnight at 4° with primary antibody diluted 1 : 1000 in TBS, 10% (v/v) SuperBlock™ blocking buffer and 0·05% (v/v) Tween-20. Membranes were washed and incubated with HRP-linked secondary antibody (Transduction Laboratories, Lexington, KY; Cell Signalling Technology; Santa Cruz Biotechnology) for 1 hr at 25°. Immunoreactive proteins were visualized on X-ray film by enhanced chemiluminescence methodology (Amersham). Stripping of membranes was performed according to the manufacturer's instructions (Amersham). Prestained protein markers, broad range, were used to indicate the size of the protein bands (Cell Signalling Technology).

Flow cytometry analysis

Larval haemocytes (5 × 10⁵ cells) were incubated in 100 μl Grace's insect medium containing either LPS-FITC (10 μg/ml) or E. coli-FITC (10 bacteria per haemocyte) or S. aureus (10 bacteria per haemocyte) latex beads-FITC (10 beads per haemocyte) for 30 min at 25° in the presence or absence of Src inhibitor PP2 or FAK dsRNA and paxillin dsRNA (specifically deplete FAK and paxillin expression, respectively) or peptides RGD and RGE or fibronectin (FN). Internalized FITC-conjugated LPS/E. coli/S. aureus/latex beads were measured by quenching surface-exposed FITC-LPS/E. coli/S. aureus/latex beads with or without trypan blue 4% in Ringer's solution. Approximately 20 000 cells from each sample were analysed by flow cytometry using a Coulter EPICS-XL-MCL cytometer (Coulter, Miami, FL), and the data were processed using the xl-2 software. The percentage decrease or increase of phagocytosis was calculated, assuming that the phagocytosis in the cells that were incubated with only LPS/E. coli/S. aureus/latex beads was 100%.

Fluorescence microscopy

Isolated haemocytes were suspended in Grace's medium and allowed to attach to glass slides for 10 min at 25°. The slides were washed with Ringer's solution to remove non-adherent haemocytes. The resulting monolayers were pre-permeabilized with 0·1% Triton X-100 for 10 seconds, washed with TBS and then fixed in a cold 100% methanol chamber at − 20° for 30 min. Slides were then rinsed with TBS and permeabilized with 0·1% Triton X-100 for 1 min at 25°. Haemocytes were treated afterwards with a protein-blocking agent for 10 min at 25° to reduce non-specific binding. Following saturation, slides were incubated with anti-tubulin, pJNK or pp38 or pERK or p^Y397FAK or non-p^Y527Src (diluted 1 : 100 in TBS) for 1 hr at 37° in a humid atmosphere. Following antibody treatment, the slides were washed with TBS and further incubated for 1 hr at 37°, in a humid atmosphere in the shelter of the light, with a secondary anti-rabbit antibody coupled to a fluorochrome (FITC) diluted 1 : 100 in TBS. Haemocyte monolayers were then washed with TBS and mounted with an aqueous mounting medium (Sigma) and observed under a fluorescence microscope or confocal fluorescence microscope to examine the distribution of DNA, tubulin, pJNK, pp38, pERK, non-p^Y527Src and p^Y397FAK in E. coli challenged haemocytes.

Rt-pcr

Reverse transcription polymerase chain reaction (RT-PCR) was carried out using the Qiagen® One Step RT-PCR Kit (Qiagen, Mississauga, Canada), according to the supplier's instructions. Total RNA was isolated from adult Drosophila flies as described in Holmes and Bonner²⁴ and the resulting FAK cDNA (1011 base pairs), after the reverse transcription, was amplified using the following primers that include a 5′-T7 RNA polymerase-binding site (TAATACGACTCACTATAGGGAGACCAC): forward FAK primer (5′-TAATACGACTCACTATAGGGAGACCACAGTCGACCACCTAACCGCGCAGATGACG-3′) and reverse FAK primer (5′-TAATACGACTCACTATAGGGAGACCACTGACTACATGACGATGTGAGAATGCG-3′). The thermal cycling conditions were as follows: 50° for 30 min; 95° for 15 min; followed by 30 cycles of 94° for 30 seconds; 57° for 1 min; and 72° for 1 min; with a final incubation at 72° for 10 min. The PCR products (1011 base pairs) were electrophoresed on 1% agarose gel in TAE buffer (Tris–HCl 0·4 m, EDTA–Na₂–salt 0·01 m, and acetic acid 0·2 m) and visualized by staining with ethidium bromide (0·5 μg/ml).

cDNA cloning

The purified FAK cDNA (1011 base pairs) was cloned into a pCRII plasmid vector, as described in Invitrogen's TA Cloning Kit, and then it was sent for sequencing analysis to confirm that the insert was the FAK cDNA.

PCR amplification

As a template for the PCRs we used the cloned FAK cDNA. Taq polymerase (New England BioLabs, Ipswich, MA) was used according to the manufacturer's recommended protocol with the same forward (5′-TAATACGACTCACTATAGGGAGACCACAGTCGACCACCTAACCGCGCAGATGACG-3′) and reverse FAK primers (5′-TAATACGACTCACTATAGGGAGACCACTGACTACATGACGATGTGAGAATGCG-3′), which include the 5′-T7 RNA polymerase-binding site (TAATACGACTCACTATAGGGAGACCAC), as described before. Thermocycle conditions were as follows: 95° for 2 min; and 25 cycles of 94° for 1 min, 65° for 45 min, 72° for 1 min; and a final extension step of 72° for 10 min. PCR products (1011 base pairs) were photographed after agarose gel electrophoresis.

dsRNA production

The purified PCR products (1011 base pairs) were used as templates to produce double-stranded RNA (dsRNA) by using a MEGAscript RNAi kit (Ambion, Inc.), following the manufacturer's instructions for the synthesis and purification of the dsRNA. The dsRNA products were precipitated with LiCl and resuspended in miliQ-H₂O. RNA concentration was measured as absorbance at 260 nm and in 1% agarose gels; dsRNAs were stored at − 70°.

RNA interference (RNAi) in medfly haemocytes

Suspended medfly haemocytes (5 × 10⁵/100 μl) were incubated in Grace's medium with FAK dsRNA for up to 6 hr and were then either lysed or used for flow cytometry. Longer term incubations were not possible because the viability of the haemocytes decreased considerably.

Enzyme-linked immunosorbent assay (ELISA)

Haemocytes (5 × 10⁵/100 μl) were either incubated for 5·30 hr in Grace's medium followed by 30 min treatment with E. coli or incubated in the presence of FAK dsRNA for up to 6 hr, but for the last 30 min E. coli was added. Haemocytes incubated in plain medium were used as a control. Cells were then lysed in lysis buffer. Insoluble material was removed by centrifugation (16 000 g for 15 min at 4°), supernatant was collected and total protein was determined. Samples were diluted with 150 mm Na₂CO₃-buffered solution, pH 9·0, to a final concentration of 50 mm Na₂CO₃ and 2–4 μg protein/ml. Aliquots of 100 μl from each sample were added to the respective wells of eight-well strips (Costar, Corning, NY) and left overnight at 4°. Strips were washed four times with TBS containing 0·05% (v/v) Tween-20. To avoid non-specific interactions, 250 μl SuperBlock™ (Pierce) was added to all wells, and the strips were allowed to incubate for 2 hr at 37°. The antigens bound on to the wells were detected with anti-pERK, anti-pp38, anti-pJNK, anti-pMEK, anti-p^Tyr416Src and anti-p^Tyr527Src polyclonal antibodies (1 : 1000) in TBS 10% (v/v) SuperBlock™ and 0·05% (v/v) Tween-20, for 1 hr at 37°. Wells were washed and incubated with 100 μl goat anti-rabbit IgG antibody or goat anti-mouse IgG antibodies labelled with horseradish peroxidase for 1 hr at 37°. After washing, peroxidase activity was determined using tetramethylbenzidine (TMB) as a substrate. The enzyme reaction was stopped with 50 μl 49.04 gl⁻¹ H₂SO₄, and colour/optical densities were measured at 450 nm with an ELISA reader.

Results

Uptake of biotic and abiotic components by hemocytes

Although well documented in mammalian systems, it remains unclear how insect haemocytes engulf pathogens. As reported earlier, medfly haemocyte suspensions from the wandering larval stage avidly engulf FITC-labelled bacteria (E. coli), polystyrene latex beads and LPS.¹ In this report more direct and quantitative approaches were used to elucidate the internalization of FITC-labelled Gram-negative (E. coli), FITC-labeled Gram-positive (S. aureus) bacteria, FITC-labelled polysterene latex beads and LPS-FITC by medfly haemocytes. Flow cytometry analysis with and without trypan blue quenching demonstrated the engulfment of the above mentioned biotic and abiotic components (Fig. 1a). In particular, these data show that after trypan blue quenching, about 30% and 70% of the total haemocytes treated, had engulfed FITC-labelled bacteria or LPS, respectively, and an additional 7% and 30%, respectively, appeared to be adhered to the exterior surface of haemocytes (Fig. 1a). The adhesion is probably a transient stage between recognition and engulfment. By contrast, latex beads appeared to be internalized immediately (Fig. 1a). A delay in the internalization of bacteria by haemocytes was also observed when the experiments were performed at 4°, instead of 25° (data not shown). Furthermore, kinetic studies of FITC-labelled E. coli and FITC-labelled S. aureus clearly show a rapid engulfment of bacteria by medfly haemocytes, indicating that the bacterial transient adhesion stage is very short (Fig. 1b,c). Confocal microscopy confirmed the uptake of E. coli by medfly haemocytes (Fig. 1d). These observations led us to postulate that different groups of invaders require different pattern recognition receptors for internalization.

Figure 1.

Bacteria and cell-surface LPS adhered to and were engulfed by medfly haemocytes. (a) Suspended haemocytes (5 × 10⁵ cells/100 ml medium) from wandering larval stage were incubated for 30 min with either FITC-labelled E. coli (i, v) or FITC-labelled S. aureus (ii, vi) or FITC-labelled LPS (iii, vii) or FITC-labelled latex beads (iv, viii) and the uptake of these components by haemocytes was analysed using flow cytometry, either after trypan blue quenching (v–viii), or not (i–iv). (b) and (c) Kinetic analysis of E. coli and S. aureus with flow cytometry. (d) Identification of E. coli phagocytosis by confocal microscopy. Haemocytes were fixed and stained with antibodies against mouse tubulin and propidium iodide. MIF, medium intense fluorescence.

FAK dsRNA and phagocytosis of bacteria, latex beads and LPS

The functional significance of FAK and other mammalian signalling molecules for phagocytosis remains undefined in many instances, and, evidently in insect haemocyte phagocytosis. As reported earlier⁴ using osmotic loading of anti FAK into medfly haemocytes, we concluded that the FAK signalling pathway might play a role in the uptake of bacteria by medfly haemocytes. These results encouraged us to test directly whether FAK is a good candidate for the phagocytosis process. To explore the functional role of FAK in the uptake of Gram-negative bacteria (E. coli), Gram-positive bacteria (S. aureus), abiotic components (latex beads) or small molecules (LPS) as well, dsRNA corresponding to Drosophila FAK was transfected into medfly haemocytes of wandering stage larvae as described previously to specifically deplete FAK expression.²⁵ As reported in this study, dsRNA corresponding to Drosophila FAK works in medfly haemocytes bringing about 55% diminution of FAK expression.²⁵ The depletion was only partial, evidently, because of the short-term incubations, as the viability of haemocytes decreases in long-term incubations.

To understand FAK silencing in the phagocytosis process, medfly haemocytes were incubated for 6 hr with dsRNA corresponding to FAK or to paxillin, as control, in the presence of bacteria, or latex beads or LPS and the level of phagocytosis was monitored by flow cytometry (Fig. 2). The results clearly show a decrease of about 22–24% in the uptake of both bacteria (Fig. 2a,b) and 28% in the uptake of the latex beads (Fig. 2d) in the presence of FAK dsRNA, whereas FAK dsRNA had no effect on the level of LPS phagocytosis (Fig. 2c). The expression of FAK is therefore an important determinant of bacteria and latex bead phagocytosis in the medfly haemocytes. On the other hand, it appears that FAK, although activated by LPS (see below), is not involved in the uptake of LPS by haemocytes.

Figure 2.

FAK regulates phagocytosis of E. coli/S. aureus/latex beads in medfly haemocytes. Suspended haemocytes (5 × 10⁵ cells/100 ml medium) from wandering larval stage were incubated with either FITC-labelled E. coli (a–c), FITC-labelled S. aureus (d–f), FITC-labelled LPS (g–i), or FITC-labelled latex beads (j–l), after preincubation for 6 hr at 25° with 8 μg FAK dsRNA (b, e, h, k). After trypan blue quenching, the haemocytes were analysed with flow cytometry. Control experiments, after trypan blue quenching, were performed in haemocyte suspensions in plain medium (a, d, g, j) or with 8 μg paxillin dsRNA (c, f, i, l). MIF, medium intense fluorescence.

Functional association of FAK with bacteria, latex beads and LPS

FAK is a non-receptor tyrosine kinase that is involved in the control of cell–extracellular interactions. To understand the functional association of FAK with bacteria and abiotic components during phagocytosis in our system, suspended haemocytes in Grace's insect medium were treated with E. coli, S. aureus, latex beads, LPS, FN, RGD or RGE and the haemocyte lysates were immunoblotted with anti-FAK Y397. The results in Fig. 3(a) clearly show that pathogens, latex beads, LPS, FN and RGD phosphorylate FAK at Y397 considerably. Tyrosine 397 (Y397) is the sole autophosphorylation site of FAK.²⁶ Increased levels of phosphorylated Y397 thus correlate well with FAK activation. Immunofluorescence labelling confirmed the phosphorylation of FAK at Y397 in haemocytes challenged with E. coli or S. aureus (Fig. 3b). Control experiments did not cause significant FAK phosphorylation (Fig. 3b).

Figure 3.

Pathogens, latex beads, LPS, fibronectin (FN) and RGD phosphorylate FAK at Y397. (a) Haemocyte suspensions stimulated with E. coli (lane 1), S. aureus (lane 2), LPS (lane 3), latex beads (lane 4), FN (lane 5), and RGD (lane 6). A haemocyte suspension that was stimulated with RGE (lane 7) was used as negative control. Haemocytes incubated in Grace's medium (lanes 8 and 9) were used to show any minor activation because of the isolation procedure. Cell lysates were subjected to immunoblotting analysis with anti-p^Y397FAK. Actin was used as a loading control. (b) Phosphorylation of FAK at Y397 was also detected immunocytochemically. Suspended haemocytes were attached to glass slides and were at first incubated with anti-p^Y397FAK (i, iii, iv), followed by a secondary anti-rabbit antibody coupled to a fluorochrome (FITC), as described in the Materials and methods. Samples that were incubated only with secondary rabbit antibody FITC (ii) were used to detect any non-specific binding sites of secondary antibody. Haemocytes incubated in plain medium were used as a control (i).

Next, we wanted to explore whether the functional association of FAK with extracellular bacteria, latex beads or LPS, takes place through RGD-binding receptors and/or other unknown receptors. Flow cytometry analysis demonstrated that the engulfment of FITC-labelled E. coli and S. aureus was blocked when haemocytes were preincubated with RGD or FN (Fig. 4a,b). Preincubation of haemocytes with RGE did not affect the internalization of bacteria (Fig. 4a,b). On the other hand, the engulfment of LPS by haemocytes was not affected in the presence of RGD or FN (Fig. 4c) and the engulfment of latex beads by haemocytes was increased rather than decreased when haemocytes were preincubated with RGD or FN (Fig. 4d). The above data strongly support RGD recognizing receptor-mediated signalling events via FAK for bacteria phagocytosis, and evidently, the presence of different pattern recognition receptors for latex beads and LPS.

Figure 4.

RGD-binding receptors regulate phagocytosis of E. coli and S. aureus in medfly haemocytes. Suspended haemocytes (5 × 10⁵ cells/100 ml medium) from wandering larval stage were incubated for 30 min with either FITC-labelled E. coli (a–d) or FITC-labelled S. aureus (e–h) or FITC-labelled LPS (i–l) or FITC-labelled latex beads (m–p), after preincubation for 30 min at 25° with 2 mm RGD (b, f, j, n) or 0·01γ/μl FN (c, g, k, o). Control samples were preincubated in plain medium (a, e, i, m) or in 2 mm RGE (d, h, l, p). After trypan blue quenching, haemocytes were analysed with flow cytometry.

FAK downstream targets and phagocytosis

FAK phosphorylation at Y397 creates a high-affinity binding site recognized by the SH-2 domain of the Src family. The recruitment and activation of Src via the formation of a bipartite kinase complex result in activation of the Ras/Raf/MAPK pathway.²⁷ Initially, to explore FAK downstream targets in our system, suspended haemocytes in Grace's insect medium supplemented with PP2 (5 μm), a specific inhibitor of Src, before challenge with E. coli or S. aureus, or latex beads, blocks the uptake of bacteria and latex beads, strongly supporting the involvement of Src in the phagocytosis process (Fig. 5a). On the other hand, the inhibitor of Src repeatedly showed induction, rather than blockage, of LPS uptake (Fig. 5a), indicating that it acts as a negative regulator, as is the case for p38 inhibitor, in Drosophila, which induces genes responsible for antibacterial peptides.²⁸ Immunofluorescence labelling confirmed the phosphorylation of Src in haemocytes challenged with E. coli or S. aureus (Fig. 5b). Control experiments did not cause significant Src phosphorylation (Fig. 5b). Consequently, FAK signals through Src for the uptake of bacteria and latex beads. On the other hand, it is further confirmed that the signalling pathway FAK/Src is not involved in the internalization of LPS.

Figure 5.

Src is involved in the phagocytosis process of bacteria and latex beads. (a) Suspended haemocytes (5 × 10⁵ cells/100 ml medium) from wandering larval stage were stimulated with FITC-labelled E. coli (i, ii), FITC-labelled S. aureus (iii, iv), FITC-labelled LPS (v, vi), FITC-labelled latex beads (vii, viii), after preincubation for 30 min at 25° in plain medium (i, iii, v, vii) or in 2 μl PP2 (ii, iv, vi, viii). After trypan blue quenching, haemocytes were analysed with flow cytometry. (b) The involvement of Src in phagocytosis was detected immunocytochemically. Suspended haemocytes were attached on glass slides and were at first incubated with non-p⁵²⁷Src (i, iii, iv), followed by a secondary anti-rabbit antibody coupled to a fluorochrome (FITC), as described in the Materials and methods. Samples that were incubated only with secondary anti-rabbit antibody FITC (ii) were used to detect any non-specific binding sites of secondary antibody. Haemocytes incubated in plain medium were used as a control (i). MIF, medium intense fluorescence.

Next we explored whether haemocytes engulf E. coli, S. aureus, LPS or latex beads by signalling via activation of MAPKs. To explore this hypothesis, haemocytes suspended in Grace's insect medium were supplemented with E. coli, S. aureus, latex beads, LPS, FN or RGD, and haemocyte lysates were subjected to immunoblotting analysis. Figure 6(a,b) clearly shows that the above treatment caused considerable phosphorylation of MAPKs. Immunofluorescence labelling confirmed the phosphorylation of MAPKs in haemocytes challenged with E. coli. Control experiments did not cause significant MAPK phosphorylation (Fig. 6c).

Figure 6.

RGD-binding receptors phosphorylate MAP kinases to the same extent as pathogens, LPS and latex beads. (a) Haemocyte suspensions were stimulated with E. coli (lane 2), S. aureus (lane 3), latex beads (lane 4), LPS (lane 5), FN (lane 6), and RGD (lane 7). Haemocytes incubated in Grace's medium (lane 1) were used to show any minor activation caused by the isolation procedure. Cell lysates were subjected to immunoblotting analysis with anti-pJNK. Membrane was stripped and reprobed with anti-pp38. Actin was used as a loading control. (b) Haemocyte suspensions were stimulated with E. coli (lane 2), S. aureus (lane 3), latex beads (lane 4), LPS (lane 5), FN (lane 6), and RGD (lane 7). Haemocytes incubated in Grace's medium (lane 1) were used to show any minor activation caused by the isolation procedure. Cell lysates were subjected to immunoblotting analysis with anti-pERK. Actin was used as a loading control. (c) Phosphorylation of MAP kinases was detected immunocytochemically. Suspended haemocytes were attached on glass slides and were at first incubated with anti-pp38 (iii), anti-pJNK (iv), anti-pERK (v), followed by a secondary anti-rabbit antibody coupled to a fluorochrome (FITC). Samples that were incubated only with secondary rabbit antibody FITC (ii) were used to detect any non-specific binding sites of secondary antibody. Haemocytes incubated in plain medium were used as a control (i).

To elucidate whether Src and MAPKs signalling molecules act upstream or downstream of FAK in our system, we analysed the effect of FAK dsRNA, on the phosphorylation of Src and MAPKs. For this purpose haemocytes in suspension were separated into three equal parts. The first part was incubated for 6 hr in Grace's medium (control). The second was incubated for 5½ hr in Grace's medium followed by 30 min treatment with E. coli, in Grace's medium and the third was incubated for 6 h with FAK dsRNA, but for the last 30 min E. coli was added. Concerning the phosphorylation of Src, we observed that FAK expression silencing in response to FAK dsRNA treatment of haemocytes, decreased the phosphorylation of active p^Y416Src, but did not affect the phosphorylation of inactive p^Y527Src (Fig. 7). This behaviour was expected because it is well known that c-Src protein tyrosine kinase activation is the result of, among other things, the overexpression of an integrin-associated signalling protein such as FAK, which interacts with and enhances c-Src PTK activity. On the other hand, the inactive c-Src was not in association with FAK. Concerning the phosphorylation of MAPKs, Fig. 7 shows that FAK expression silencing in response to FAK dsRNA treatment in E. coli-challenged haemocytes reduced the phosphorylation of ERK, MEK, JNK and p38 MAPKs. These gene-silencing experiments convincingly demonstrated that MAPK activation was vital for efficient endo/phagocytosis processes in insect haemocytes, as is the case with mammalian counterparts.

Figure 7.

Effect of FAK silencing on the phosphorylation of ERK, p38, JNK, MEK and Src. Suspended haemocytes from wandering larval stage were either incubated for 5½ hr in Grace followed by 30 min treatment with E. coli (E. coli) or incubated in the presence of FAK dsRNA for up to 6 hr, but for the last 30 min E. coli was added (FAK dsRNA + E. coli). Haemocytes incubated in plain medium were used as a control (control). Afterwards, the haemocytes were lysed and the lysates were plated in a 96-well assay plate and analysed by ELISA to detect the status of ERK, p38, JNK, MEK and Src phosphorylation using polyclonal antibodies against pERK (a), pp38 (b), pJNK (c), pMEK (d), p^Y416Src (e), and p^Y527Src (f), respectively. Each well contained 4 μg/ml haemocyte lysate.

The use of specific inhibitors for Src and MAPKs and phosphorylated site-specific antibody for FAK, phosphorylated on tyrosine 397, in Western blots also confirmed that Src interacts with FAK (see also Fig. 7), and that MAPKs act downstream of FAK (Fig. 8). Upon stimulation the c-Src-SH2 domain binds to the motif surrounding a major autophosphorylation site (Y397) on FAK. Haemocytes of the wandering stage larvae suspended in Grace's insect medium were supplemented with either PD 098059 (8 μm) or U0126 (8 μm) for MEK1/2, SB203580 (0·4 μm) or SB202190 (10 μm) for p38, SP600125 (4 μm) for JNK or PP2 (5 μm) for Src, to block MAPKs and Src. Western blot analysis showed that PP2 blocked FAK phosphorylation, whereas PD 098059, UO126, SB203580, SB202190 and SP600125 inhibitors had no effect on FAK phosphorylation (Fig. 8). These results confirm that Src interacts with FAK and ERK; JNK and p38 are downstream targets that are activated by this FAK/Src complex.

Figure 8.

Src interacts with FAK and MAP kinases act downstream of FAK. Suspended haemocytes were stimulated with E. coli (lanes 2–9) after preincubation for 30 min at 25° with Src inhibitor, 5 μm PP2 (lane 3) or with kinase inhibitors, 0·4 μm SB 203580 (lane 4) and 10 μm SB 202190 (lane 5) for p38, 4 μm SP 600125 (lanes 6 and 7) for JNK and 10 μl U0126 (lane 8) for MEK1/2 and 4 μm PD 098059 (lane 7) for ERK. Haemocytes incubated in Grace's medium (lane 1) were used to show any minor activation as a result of the isolation procedure. Cell lysates were subjected to immunoblotting analysis with anti-pFAK Y397. Actin was used as a loading control.

Physical association of FAK with Src and ERK

It is widely accepted that the signalling molecules participating in a pathway are usually in a direct physical association. To explore whether any physical association exists between the signalling molecules in our system, we analysed the association of FAK with Src and ERK in medfly haemocytes. Mammalian FAK, Src and ERK are closely related to their insect homologues. Drosophila FAK, Src and ERK are 33%, 51·3% and 80% identical in amino acid sequence to their mammalian counterparts.^29–31 Furthermore, the amino acid sequence surrounding the phosphorylation sites in these molecules in humans are also closely related or identical to insect homologues as retrieved from Fly base and NCBI HomoloGene base.

For this purpose, coimmunoprecipitation, reciprocal coimmunoprecipitation and Western blot analysis were performed. Figure 9 shows that when FAK was immunoprecipitated from medfly haemocytes using anti-FAK antibody, coimmunoprecipitation of Src and ERK was observed, in Western blot analysis. Furthermore, when Src or ERK were immunoprecipitated using anti-Src or anti-ERK, FAK was also coimmunoprecipitated. Finally, Src antibody precipitated ERK and vice versa (Fig. 9). These results strongly support a physical association between FAK with Src and ERK.

Figure 9.

Co-immunoprecipitation and reciprocal co-immunoprecipitation of FAK, Src and ERK signalling molecules. Suspended haemocytes from wandering larval stages were isolated and then lysed. FAK, Src or ERK were immunoprecipitated from the lysates. When FAK was immunoprecipitated, using FAK polyclonal antibody, the precipitates were then resolved on 10% SDS–PAGE and blotted with polyclonal antibodies against Src and ERK. When Src or ERK were immunoprecipitated using polyclonal antibodies against Src and ERK, the precipitates were then resolved on 10% SDS–PAGE and blotted with FAK polyclonal antibody.

Discussion

Insect haemocytes responsible for cellular defence responses recognize a variety of foreign biotic and abiotic targets, and respond by activating several intracellular signalling pathways via haemocyte surface receptors including integrins. Key targets of these intracellular signalling pathways in response to pathogen infection appear to be the regulated release of bioactive molecules and phagocytosis. A documented signalling pathway involved is FAK/Src/MAPKs, and the secreted bioactive molecules are PAPs.^1–3 The PAPs are essential bioactive molecules, because they activate cell-free or cell-surface-associated PO. The activated proPO cascade participates in many aspects of insect innate immunity (phagocytosis, melanotic encapsulation and wound healing), in contrast to mammalian innate immunity. The mode of action of the proPO cascade in the phagocytosis and wound healing processes is not known. In the process of melanotic encapsulation, however, it is suggested that the toxic quinones or hydroxyquinones produced via the proPO cascade function as killing agents.³² These studies aimed to clarify further the role of FAK, Src and MAPKs in the process of phagocytosis of bacteria, latex beads and LPS by primary insect haemocytes.

An initial step for E. coli or S. aureus internalization by medfly haemocytes is their binding to an RGD-binding receptor, as judged by inhibition of E. coli or S. aureus uptake in the presence of RGD peptides or fibronectin (Fig. 4a,b). Interestingly, latex beads and LPS are not recognized by an RGD-binding receptor (Fig. 4c,d). These results indicate differentiation of pathogens from abiotic components by pattern recognition receptors. RGD-binding peptides also inhibit phagocytosis in several other insect species including the mosquito Aedes gambiae,³³Drosophila melanogaster and the moth Pseudoslusia includens.³⁴ RGD-containing peptides inhibit integrin-mediated binding to RGD-containing extracellular proteins or fibrinogen in both vertebrates and invertebrates.^35,36 A haemocyte-specific integrin was also found to be required for haemocytic encapsulation in the tobacco hornworm Manduca sexta.³⁷· Therefore, it appears that the molecular mechanisms of adhesion, phagocytosis and encapsulation share at least haemocyte surface receptors.

A key question that immediately arises is what signalling molecules link haemocyte surface receptors to E. coli, S. aureus, latex beads or LPS engulfment with the process of phagocytosis. FAK is considered to be a component of central importance because it is known to associate with multiple cell surface receptors (integrins, growth factor receptors and G-protein-linked receptors) and signalling proteins through which it can modulate the activity of several intracellular signalling pathways.^26,38 FAK is a family of non-receptor and non-membrane-associated tyrosine kinases, which have been implicated in controlling several cellular functions, including cell spreading, migration, apoptosis and cell survival.^25,26 The present data show that E. coli, S. aureus, latex beads, LPS, FN or RGD peptides phosphorylate FAK at Y397 rapidly and considerably (Fig. 3), implying a central role(s) for haemocyte FAK in cell signalling pathways. FAK possesses several tyrosine-phosphorylated sites, which play an important role in its function. FAK autophosphorylation on Tyr³⁹⁷ creates binding sites for the SH2 domain of the Src members, which also promotes phosphorylation of FAK on Tyr⁹²⁵.²⁶ This phosphorylated site on FAK recruits and promotes Grb2 adaptor protein binding to FAK and stimulates the activation of the Ras/MAP pathway.

Previous experiments with antibody-mediated inhibition of FAK resulted in both blockage of FAK phosphorylation and phagocytosis, indicating that the phosphorylation of FAK is a prerequisite for bacteria phagocytosis by insect haemocytes.⁴ In this report by knockdown experiments we further confirmed the involvement of FAK in phagocytosis (Fig. 2). FAK dsRNA was transfected into medfly haemocytes, followed by E. coli, S. aureus, latex beads or LPS infection. Flow cytometry analysis demonstrated a considerable inhibition of the phagocytosis of bacteria and latex beads in response to FAK dsRNA. LPS uptake was not affected by FAK dsRNA (Fig. 2). Consequently, the activation of FAK is a prerequisite for the engulfment of bacteria and latex beads by insect haemocytes. In mammalian systems the results in support of these mechanisms are rather controversial. For example, it has been reported that FAK promotes phagocytosis of integrin-bound photoreceptors³⁹ and is involved in the uptake of Yersinia.⁴⁰ On the other hand, it has been reported that extracellular pressure stimulates macrophage phagocytosis by inhibiting a pathway involving FAK.⁴¹

Given the fact that FAK is phosphorylated at Y397 by bacteria, latex beads and LPS (Fig. 3), there is considerable interest in the pathways by which FAK signals in each case. It is well appreciated that FAK uses multiple downstream effectors to mediate its diverse biological actions; therefore, we focused our interest on the potential roles of Src, which in mammalian systems binds and activates FAK and MAPKs that in mammalian systems form an important set of pathways regulated by integrins and FAK/Src complex.

Previous experiments with osmotic-loaded anti-Src antibodies in medfly haemocytes indicated that this specific inhibition of Src kinases is able to inhibit both FAK phosphorylation and phagocytosis.⁴ In this report, ELISA showed that preincubation of haemocytes with FAK dsRNA, blocks the phosphorylation of Src at Y416 (Fig. 7). Furthermore, flow cytometry analysis demonstrated that PP2, a specific Src inhibitor, blocks bacteria and latex bead phagocytosis, but not LPS uptake, supporting the participation of Src family kinases in the process of bacteria and latex bead phagocytosis (Fig. 5). Consequently, the signalling for the engulfment of pathogens or latex beads, in medfly haemocytes is transmitted from RGD-binding receptors or other unknown receptors via the FAK/Src complex to downstream target substrates (Fig. 10). On the other hand, the FAK/Src complex is not involved in the internalization of LPS.

Figure 10.

Proposed signalling model for the engulfment of pathogens or latex beads or LPS by medfly haemocytes.

In normal mammalian cells, FAK and c-Src association and activation occur in a transient fashion, and both events are tightly regulated. In our system the physical association of bacteria, latex beads and LPS with RGD-binding receptors and/or other unknown haemocyte receptors leads to the activation of FAK and Src for bacteria and latex beads, but not for LPS. Therefore, there is a functional association of ligands via haemocyte receptors with FAK and Src. In this study, immunoprecipitation and coimmunoprecipitation experiments (Fig. 9), demonstrated that FAK and Src are also physically associated in unchallenged haemocytes. Therefore, in medfly haemocytes FAK and Src are both functionally and physically associated.

In mammalian systems, downstream targets that are activated by the FAK/Src complex include ERK, JNK and p38 signalling molecules. The MAPK family is a well-characterized intracellular evolutionarily conserved phosphorylation cascade and is implicated in the regulation of differentiation control, cell proliferation, development, inflammatory response, apoptosis and phagocytosis.^1,42–44 Three major subfamilies have been characterized, the ERKs, the JNKs, also known as stress-activated protein kinases (SAPKs), and the p38 MAP. Previous data as well as silencing gene expression experiments (Figs 6 and 7) convincingly demonstrated that activation of MAPKs is vital for efficient endo/phagocytosis in insect haemocytes, as is the case for their mammalian counterparts.^1,3 Activation of ERK is also required for phagocytosis by the freshwater snail Lymnaea stagnalis¹⁶ and activation of JNK is required for haemocyte uptake in the marine mussel Mytilus.¹⁵ The activation of MAPKs has previously been reported in haemocytes from medfly when challenged with E. coli, latex or LPS^1,3 and in mammalian macrophages following challenge with bacteria^45,46 with similar temporal, transient patterns of phosphorylation. Since macrophages and haemocytes are functionally similar⁴⁷ the comparable responses of the MAPK pathways to bacterial challenge between these cells suggest that MAPK signalling in innate immunity may have been conserved via evolution.

Another interesting finding in this study, as demonstrated by immunoprecipitation and coimmunoprecipitation experiments, was that ERK is not only functionally associated, but also physically associated with the FAK/Src complex (Fig. 9). This finding raises another, as yet unsolved, question of how haemocytes discriminate the pathways promoted by bacteria or latex beads from that of LPS. As stated above, bacteria or latex beads activate MAPKs through the FAK/Src complex, whereas LPS does not.

Taken together, the above results provide strong evidence that (1) bacteria bound at the haemocyte surface activate RGD-binding receptor, FAK, Src and MAPKs sequentially, promoting their phagocytosis by haemocytes, (2) latex beads bound at the haemocyte surface activate an unknown receptor and then FAK, Src and MAPKs sequentially, and (3) LPS bound at the haemocyte surface activate initially an unknown receptor, and then through an undefined signalling pathway, activate MAPKs (Fig. 10). FAK, which is activated by LPS, is not involved in the internalization of LPS. Our efforts have now been focused to discriminate in detail the differences in the signalling pathways concerning the uptake of biotic, abiotic and small molecules by insect haemocytes and to compare the signalling in innate immunity, between mammals and insects.

Abbreviations

ELISA

enzyme-linked immunoabsorbent assays

ERK

extracellular signal-regulated kinase

FACS

fluorescence-activated cell sorter

FAK

focal adhesion kinase

FITC

fluorescein isothiocyanate

FN

fibronectin

HRP

horseradish peroxidase

IgG

immunoglobulin G

JNK

c-jun N-terminal kinase

MAPK

mitogen-activated protein kinase

MEK

MAPK/ERK kinases

PAP

prophenoloxidase-activating proteinases

PI

propidium iodide

PI-3K

phosphatidylinositide 3′-kinase

PO

phenoloxidase

proPO

prophenoloxidase

RNAi

RNA interference

RT-PCR

reverse transcription–polymerase chain reaction

SH2 domain

Src homology domain