- Email: [email protected]
Preliminary characterization of hemolymph coagulation in Anopheles gambiae larvae
ARTICLE IN PRESS
Developmental & Comparative Immunology
Developmental and Comparative Immunology 31 (2007) 879–888
www.elsevier.com/locate/devcompimm
Preliminary characterization of hemolymph coagulation in Anopheles gambiae larvae Bogos Agianiana,b, Christine Leschc, Olga Losevac,d, Mitchell S. Dushaye,f, a
European Molecular Biology Laboratory, Meyerhofstrasse 1, 69117 Heidelberg, Germany Department of Molecular Biology and Genetics, Democritus University of Thrace, Dimitras 19, 68100 Alexandroupolis, Greece c Department of Molecular Biology and Functional Genomics, Stockholm University, 106 91 Stockholm, Sweden d Department of Genetics, Microbiology, and Toxicology, Stockholm University, 106 91 Stockholm, Sweden e Department of Life Sciences, So¨derto¨rns ho¨gskola, 141 89 Huddinge, Sweden f Department of Comparative Physiology, Uppsala University, Norbyva¨gen 18A, 752 36 Uppsala, Sweden
b
Received 30 November 2006; received in revised form 8 December 2006; accepted 18 December 2006 Available online 25 January 2007
Abstract Hemolymph coagulation is a first response to injury, impeding infection, and ending bleeding. Little is known about its molecular basis in insects, but clotting factors have been identified in the fruit fly Drosophila melanogaster. Here, we have begun to study coagulation in the aquatic larvae of the malaria vector mosquito Anopheles gambiae using methods developed for Drosophila. A delicate clot was seen by light microscopy, and pullout and proteomic analysis identified phenoloxidase and apolipophorin-I as major candidate clotting factors. Electron microscopic analysis confirmed clot formation and revealed it contains fine molecular sheets, most likely a result of lipophorin assembly. Phenoloxidase appears to be more critical in clot formation in Anopheles than in Drosophila. The Anopheles larval clot thus differs in formation, structure, and composition from the clot in Drosophila, confirming the need to study coagulation in different insect species to learn more about its evolution and adaptation to different lifestyles. r 2007 Published by Elsevier Ltd. Keywords: Anopheles gambiae; Drosophila; Coagulation; Lipophorin; Phenoloxidase; Insect immunity
1. Introduction Coagulation is one of the first responses to injury in insects as in other animals, and the clot prevents infection as well as stops bleeding and contributes to wound healing (reviewed in [1–3]). While much is known about coagulation in Limulus [4], and studies Corresponding author. Department of Comparative Physiol-
ogy, Uppsala University, Norbyva¨gen 18A, 752 36 Uppsala, Sweden. Tel.: +46 018 471 2805; fax: +46 018 471 6425. E-mail address: [email protected] (M.S. Dushay). 0145-305X/$ - see front matter r 2007 Published by Elsevier Ltd. doi:10.1016/j.dci.2006.12.006
have been done in other arthropods such as shrimp [5] and crayfish [6], the molecular mechanisms of coagulation in insects are not well understood. Studies have begun in Drosophila melanogaster larvae, where the clot has been described [7], and methods have been developed to isolate and identify candidate clotting factors [8,9]. Remarkably, few of these clotting factors have homologs in other insects [8]. This suggests that insect coagulation may have evolved through co-option of genes serving other functions [3,10]. The lack of conservation makes it important to study coagulation in different insects,
ARTICLE IN PRESS 880
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
to learn how the clot is formed in different species, and by comparison to learn how evolution acted on this important defense process. The mosquito Anopheles gambiae is the subject of intense interest because of its role as malaria vector; its humoral immune response has been studied [11–13], and genome published [14]. Although it is a dipteran insect like Drosophila, the two species are separated by 250 million years of evolution, and mosquito larvae are aquatic, so both time and different environmental constraints have acted on coagulation in these species. Coagulation in Anopheles is also of interest because some of its mechanisms may be shared with other immune reactions, including those against malaria parasites, and for the possibility that coagulation could provide a new target for pest control strategies. We have studied coagulation in Anopheles larvae by applying techniques developed to study clotting in Drosophila. To our surprise, we could not draw out strands from coagulating mosquito larval hemolymph as with Drosophila [7], suggesting that the mosquito clot is much more delicate. We used more sensitive techniques to demonstrate the formation of a clot in Anopheles larvae, and we used pullout [8] to identify phenoloxidase (PO) and apolipophorin-I as candidate clotting factors in these mosquitoes. Electron microscopy of negatively stained clot preparations revealed the Anopheles clot is at least partly composed of paracrystalline sheets. Taken together, our data further suggest that these sheets are produced by the assembly of lipophorin particles in a process dependent on PO. Further work will be required to identify additional clotting factors, demonstrate how clot formation is activated, and eluciate the role of coagulation in immune defense in mosquito larvae.
humid chamber 30 min RT. The grids were washed twice in 10 ml drops of 0.1 PBS, 2 in 10 ml 0.5% IGEPAL (Sigma) in 0.1 PBS, and twice in 0.1 PBS. The clot was then stained with 10 ml 0.1 PBS containing 10 mg/ml lectin (either HPL-FITC or PNA-FITC) for at least 20 min, and visualized in a Zeiss Axioplan 2 microscope. 2.3. Bead aggregation Bead aggregation was performed essentially as described in [8]. Tosylactivated Dynabeads (M-280, Dynal) were washed and blocked overnight with 0.2 M Tris (pH 8.5) according to manufacturer’s instructions, then washed and resuspended in 0.1 PBS pH 8.0. For bead aggregation experiments, both Drosophila and Anopheles, 5 larvae were bled onto 10 ml of beads. 2.4. Pullout
2. Materials and methods
Pullout was performed as described in [8]. Briefly, fourth instar Anopheles larvae were rinsed in dH2O and opened with fine antimagnetic scissors under the surface of 50 ml beads held in a watch glass. The larvae were swirled in the buffer during bleeding to maximize contact of clotting factors with the beads. The solution and the beads were then transferred to a microcentrifuge tube, and the watch glass was washed with an additional 100 ml of buffer, which was added to the tube. Tubes were put in a magnetic holder to attract the beads to the side of the tube, and the solution was removed by pipette. Beads were washed (i) 3 in 0.1 PBS, (ii) 3 in 0.1 PBS, 0.5% IGEPAL (Nonidet 40) detergent, and Roche protease inhibitor, and (iii) 3 in 0.1 PBS. The beads were then resuspended in 16 ml 0.1 PBS+5 ml gel loading buffer, heated to 64 1C 10 min, and then stored at 20 1C.
2.1. Mosquito larvae
2.5. Transmission electron microscopy (TEM)
A. gambiae G3 larvae were raised in standard conditions [15]. Fourth instar larvae were used in all experiments because they are large and close in size to third instar Drosophila larvae (Fig. 1A).
Five third instar Drosophila, or 10 fourth instar Anopheles larvae were bled into 10 ml 0.1 PBS (with, or without PTU) on top of glow-discharged carbon-coated grids and incubated for 20 min at RT in humid chambers. Grids were then washed twice in 0.1 PBS, and stained in 1% uranylacetate in 0.1 PBS for 1 min. The samples were examined in a Philips CM120 Biotwin TEM operating at 100 kV. Images were captured using a retractable CCD camera.
2.2. Light microscopy of clot preparations Five Drosophila or Anopheles larvae were bled by ripping open with fine forceps into 10 ml 0.1 PBS pH 8.0 over a copper SEM grid and incubated in a
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
881
Fig. 1. Anopheles fourth instar and Drosophila third instar larvae: (A) Fourth instar Anopheles gambiae larva above, and third instar Drosophila melanogaster larva below, shown against a millimeter scale to indicate the comparable size of the two species. (B) Melanization after wounding. Larvae of the two species are shown after wounding with a fine tungsten needle to demonstrate that like Drosophila, Anopheles larvae are able to heal wounds, which are subsequently melanized. Melanized healed wounds are indicated by arrows.
2.6. MALDI-TOF mass spectrometry analysis and protein identification Proteomic analysis was done essentially as described in [9]; coomassie-stained protein bands were excised from gels using sterile stainless-steel blades. The gel pieces were destained with 50% acetonitrile in 25 mM ammonium bicarbonate and finally dehydrated with 100% acetonitrile. Proteins were in-gel digested with 12.5 ng/ml trypsin (Promega V511A) in 50 mM ammonium bicarbonate at 37 1C overnight. The resultant peptides were extracted with 50% acetonitrile/5% trifluoroacetic acid and dried in a vacuum centrifuge. The recovered peptides were purified and concentrated
on a C18ZipTips (Millipore) column according to manufacturer’s instructions. Mass spectra (MS) were recorded in positive reflection mode using an Applied Biosystems MALDI-TOF Voyager-DE STR mass spectrometer equipped with a delayed ion extraction technology. a-cyano-4-hydroxycinnamic acid was used as the matrix. External calibration was performed using the Sequazyme Peptide Mass Standard kit with Angiotensin I (1296.6853 Da) and ACTH clips 1–17 (2093.0867 Da), 18–39 (2465.1989 Da), 7–38 (3657.9294 Da) (PerSeptive Biosystems, USA) and for internal calibration auto digestion peaks of bovine trypsin were used. The peptide mass profiles produced by MS were analyzed by using the
ARTICLE IN PRESS 882
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
programs Mascot (http://www.matrixsciences.com) and MS-Fit (http://prospector.ucsf.edu). The monoisotopic peptide masses were compared with the theoretical masses derived from the NCBInr database for Metazoan animals. Search parameters included allowed mass accuracy of 0.1 Da, more than four peptide mass hits required for a protein match, consideration of one missed enzymatic cleavage, pI range 3.0–10.0, and molecular mass range 1–400 kDa. For lipophorin (ENSANGP00000018348, theoretical MW/pI 360 kDa/7.4), the Mascot search resulted in 47 matched mass values, which represents 20% sequence coverage. The MS-Fit search matched 46 peptides covering 20.8% of the protein. For prophenoloxidase 3 (ENSANGP0000002437, theoretical MW/pI 78.6 kDa/6.8), the Mascot search resulted in 23 matched mass values with sequence coverage of 46%. The MS-Fit search gave 23 matched peptides covering 46.2% of the protein. 3. Results 3.1. The Anopheles clot is finer than the Drosophila clot Fourth instar A. gambiae larvae are comparable in size to third instar Drosophila larvae (Fig. 1A), so we used these for our experiments for purely
practical reasons. Hemolymph from Anopheles larvae differed from Drosophila hemolymph by containing large amounts of what appeared to be lipid droplets. Some, or all of these may have come from fat body damage. In addition, Anopheles hemolymph appeared to melanize more quickly and to a greater extent than Drosophila hemolymph, although we did not measure this quantitatively. Anopheles larvae stopped bleeding and formed a melanized scar/scab after injury with a fine needle (Fig. 1B), like Drosophila [16]. This clearly suggested, coagulation occurred, and that it would be possible to study the Anopheles larval clot using the draw out, bead aggregation, and pullout techniques developed in Drosophila [7,8]. To our surprise, we were unable to draw coagulating strands out of pooled Anopheles hemolymph, as we could easily do for Drosophila (see also [7]). Furthermore, Anopheles hemolymph did not aggregate dynabeads, an assay that is indicative of clotting in Drosophila [8] (Fig. 2A–C). Because the size or weight of the dynabeads might hinder their aggregation by a finer clot, we tried bleeding hemolymph onto finer India Ink particles. Again, despite trying different incubation times, buffer pH, and salt composition, we did not observe particle aggregation (data not shown). This suggested that the Anopheles clot was substantially different and more delicate than
Fig. 2. Bead aggregation and detergent washed clot on SEM grids stained with FITC-coupled lectins: The well diameter for (A–C) was 8 mm, while the SEM grids in (D–F) had gaps of 23 mm. (A) Drop of untreated dynabeads. (B) Drop of dynabeads incubated with Anopheles larval hemolymph. Note lack of bead aggregation. (C) Drop of dynabeads incubated with Drosophila wild-type larval hemolymph to show bead aggregation. (D) Clot from Drosophila larval hemolymph stained with PNA-FITC for comparison to E and F to show how a different lectin stains clots from the different species. The staining of the Drosophila clot with PNA has been described [7,47]. 66 magnification. (F) Clot from Anopheles gambiae washed with detergent and stained with HPL-FITC to show Anopheles clot material can span gaps in an SEM grid. 66 magnification. Arrowheads indicate staining clot material. (E) Different view of Anopheles gambiae clot at higher, 100 magnification, prepared in the same way as Fig. 2E. Arrowheads indicate staining clot material.
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
Drosophila’s, and that alternative methods would be required to study coagulation in the mosquito. 3.2. Microscopic evidence of an Anopheles clot To see the Anopheles hemolymph clot, we bled larvae onto copper SEM grids with square holes 23 mm wide and looked for gap-spanning material. Extremely fine, wisp-like material spanning the corners of the SEM grids was barely detectable in phase contrast microscopy, in contrast to the bulkier Drosophila clot formed under similar conditions (data not shown). The Drosophila clot is more easily visualized when stained with FITCconjugated peanut agglutinin (PNA) [7], and this clearly labeled the Drosophila clot which adhered to SEM grids through detergent washes (Fig. 2D). However, this lectin did not decorate our Anopheles preparations. Instead, we found that the Anopheles clot stained with FITC-conjugated Helix pomatia lectin (HPL) (arrowheads, Fig. 2E and F). Despite the difference in lectin binding, the occurrence of these fine strands and their persistence through detergent washes strongly indicates a discrete clot. PNA and HPL bind to different sugars (galactose vs. terminal N-acetyl-a-D-galactosamine, respectively), and their differential labeling of Drosophila and Anopheles clots suggests different glycan composition in the two species, either of glycosylated protein clotting factors or constituent lipopolysaccharides. To investigate the role of PO in Anopheles coagulation, we bled larvae onto copper grids in the presence of PTU, an inhibitor of PO. In contrast to the results described above, clot material was very rarely seen on SEM grids in the presence of PTU (data not shown). This suggested that PO could be important for clot formation in Anopheles, in contrast to Drosophila, where PTU does not inhibit clot formation (data not shown), and where it has been demonstrated genetically that PO is not necessary for primary soft clot formation [7]. 3.3. Identifying candidate clotting factors Having demonstrated clot formation in Anopheles, we set out to identify candidate clotting proteins. The pullout method of isolating proteins on dynabeads has been used to identify clotting factors in Drosophila [8]. However, this assay was developed based on the ability of Drosophila hemolymph to aggregate beads. Because Anopheles
883
hemolymph did not aggregate beads, we first performed control experiments to demonstrate that clotting factors adhere to beads even in the absence of bead aggregation. For this, we made use of Drosophila hemolectin (hml) mutants [17], whose hemolymph does not allow coagulating strands to be drawn or bead aggregation ([8], and Lesch and Dushay, unpublished observations). Nonetheless, when we performed pullouts on hemolymph from hml larvae, we obtained protein patterns similar to wild-type Drosophila, except for the absence of Hml protein (Fig. 3A). Thus, under our conditions, Drosophila clotting factors bind to beads even in the absence of bead aggregation. We then isolated Anopheles candidate clotting proteins using the same pullout assay. Pullouts of hemolymph from fourth instar Anopheles larvae showed four prominent bands, a single band around 260 kDa, a doublet centered at 65 kDa and a lower band at 55 kDa (see arrows in Fig. 3B). The bands were excised and analyzed proteomically (as described in Section 2, where sequence coverage and other details are also provided). The tryptic peptide analysis identified the 260 kDa band as lipophorin. The lipophorin band matched the expected MW for apolipophorin-I (apoLp-I), the heavy component of lipophorin. Consistent with this, lipophorin has been identified as a clotting factor in Drosophila [8,9], and in several other insect species [2,18–21]. Tryptic peptide analysis of the 65 and 55 kDa bands identified them as prophenoloxidase 3 (PPO3) with theoretical MW 78.6 kDa. Our finding two bands of smaller molecular weights suggests these were most likely products of specific proteolytic processing, a result of activation of the PPO cascade [23] and production of PO. Similarly sized PO polypeptides, including doublets most likely representing heteromerization, have been detected in the hemolymph of A. gambiae [24] and Aedes aegypti [25]. Our identification of PPO3 is consistent with the high expression of this isoform in late instar larvae [22], However, nine PPO genes have been annotated in the Anopheles genome ([22] and www.ensembl.org/Anopheles_gambiae) with high sequence identity (470%) between their products, so we do not exclude the possibility that other PO isoforms also participate in the clot. 3.4. TEM on clot preparations The extremely thin and fragile nature of the Anopheles clot suggested by the absence of bead
ARTICLE IN PRESS 884
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
aggregation or strand formation and our light microscopy observations led us to examine clot preparations by TEM. Negatively stained preparations of Anopheles and Drosophila hemolymph were made as described in Section 2, including Anopheles samples incubated with PO-inhibitor PTU to test whether PO has an effect on clot formation. When Anopheles hemolymph was incubated in the presence of PTU, an abundance of approximately spherical particles about 10 nm in diameter were observed distributed almost uniformly on the grids (Fig. 4A and B). These particles have notable resemblance to insect lipophorin purified from hemolymph, both in shape and size [26], which allows us to postulate that they represent lipophorin molecules. Remarkably, when we bled Anopheles larvae and incubated hemolymph under identical conditions but in the absence of PTU—allowing PO activity, we observed regular, paracrystalline sheets (Fig. 4C and D, and arrowhead in Fig. 4D), which appeared to be formed from laterally stacked fibers (diameter ca. 10 nm) that are beaded along their long axis (Fig. 4C and D). We also observed sporadic fiber structures of much larger diameter (ca. 50 nm), which stain uniformly along their length (Fig. 4E and F). The molecular nature of these thicker fibers remains unidentified. The same preparations consistently showed a markedly lower number of free particles like the ones shown in Fig. 4A and B. The sheets were never observed in the presence of PTU in several repetitions of the same assays with different larvae, however, they were always observed in the absence of PTU. TEM of similarly prepared Drosophila hemolymph showed a coarser, thicker clot, in which no regular structure was visible (data not shown). Fig. 3. Pullout of Drosophila and Anopheles hemolymph: (A) Drosophila proteins binding to beads analyzed by PAGE on 5% gels. While loss of Hml protein does not allow strand draw out and abolishes bead aggregation, the proteins adhering to the beads show a similar pattern to wild-type hemolymph, except for the absence of Hml (indicated by an arrow). Lane 1: pullout from wild-type hemolymph. Lanes 2 and 3: independent pullouts from Hml mutant larval hemolymph. (B) Pullout from Anopheles gambiae analyzed by PAGE on 10% gel. Apolipophorin (arrow) and phenoloxidase bands (arrowheads) are indicated. Lane 1: pullout from 10 fourth instar larvae. Lane 2: Whole hemolymph from two larvae. We note that the hexamerin bands in Lane 2 appear less abundant in comparison to other published gel patterns of Anopheles larval hemolymph c.f. [48]. This is probably a result of procedural differences (our samples were not diluted and quick frozen at 80 1C).
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
885
Fig. 4. TEM of negatively stained Anopheles hemolymph with and without PTU: (A–B) Different views of preparations with PTU. Scale bars 100 nm. (C–F) Different TEM views of negatively stained Anopheles hemolymph preparations in the absence of PTU. When PTU is omitted, phenoloxidase quickly melanizes hemolymph. Now, it appears that the putative lipophorin molecules associate into a paracrystalline matrix, indicated by an arrowhead in (D). Scale bars: (C) 50 nm, (D) 200 nm, (E) 20 nm, (F) 50 nm.
These observations, in combination with the pullout results (see Section 3.3) lead us to suggest that during Anopheles coagulation, PO activity causes lipophorin particles to coalesce into the sheet structures we observe. A protein crosslinking action of PO is well documented [23], and PO was proposed to crosslink lipophorin in Bombyx mori hemolymph, although this was not linked to coagulation [27]. It has been noted that lipophorin has a tendency to precipitate in low ionic strength conditions [28], but the ordered structures we see (e.g. Fig. 4C and D) are inconsistent with random aggregation. We are not able to distinguish whether PO remains attached to the sheets, or if it takes part as a structural component of the clot.
4. Discussion To our knowledge, this is the first study of coagulation in larvae of the mosquito A. gambiae. We have demonstrated clot formation and shown that the mosquito clot is much more delicate than the clot found in Drosophila. The Anopheles clot stained with HPL (GalNAc), rather than PNA (Gal) which stains the Drosophila clot (Fig. 2), suggesting that Anopheles clotting factors are glycosylated differently than in Drosophila. The functional significance of this finding remains to be shown. Coagulating Anopheles hemolymph is not viscous, does not allow strands to be drawn out, and
ARTICLE IN PRESS 886
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
does not aggregate beads. However, candidate clotting factors can still be isolated by pullout, as demonstrated by control experiments in Drosophila (Fig. 3A). The main mosquito clotting factors we identified were lipophorin and PO (Fig. 3B). While both lipophorin and PO were identified in Drosophila pullouts [8], we show here that they are more prominent, and appear to play a greater role in Anopheles coagulation. A vitellogenin-like protein is the plasma coagulogen in crayfish [6], and lipophorin forms detergent-insoluble precipitates in Galleria [29] and Ephestia [30], and has been identified as a clotting factor in many insect species including Drosophila (references cited in Section 3). An immune role for lipophorin in mosquitoes is supported by the results of Cheon et al. [31] who found lipophorin is upregulated in Aedes aegypti following immune challenge in a Toll/Reldependent way, and by reports showing that the lipophorin precursor gene is essential for the development of Plasmodium parasites in A. gambiae [32]. PO is not required for primary soft clot formation in Drosophila [7]. In contrast, we have demonstrated that PO activity is required for coagulation in Anopheles. Fluorescence light microscopy showed that HPL-staining clot material adhered to SEM grids in the absence, but not in the presence of the PO inhibitor PTU, in contrast to Drosophila, where PTU does not have this effect. PO is an important enzyme for immune reactions in mosquitoes [33,34], its activity is critical for protection against parasite infection [35], and it has long been held to be critical for coagulation in different insect species [2,36,37]. In order to observe details of the fine clot detected by light microscopy and to assess a potential role of PO, we investigated larval hemolymph by TEM. In the presence of PTU, there was an abundance of free 10 nm particles, and these coalesced into paracrystalline arrays in the absence of PTU (Fig. 4). Based on our assumption that these particles represent lipophorin, as suggested by their size and abundance, this suggests that PO acts to crosslink lipophorin during coagulation. Although a spontaneous tendency of lipophorin to aggregate has been previously reported [28], our results show a controlled and specific structural organization inconsistent with random aggregation (Fig. 4C and D). Our proteomic analysis of pullout bands identified only the high molecular mass component of lipophorin, apoLP-1. However, we cannot conclude whether apoLP-II is also part of the clot, since a
smaller apoLP-II band in Fig. 3B would be very faint, assuming a 1:1 ratio to apoLP-1. Apolipophorins I and II are usually tightly associated and dissociate only under strongly chaotic conditions [38–40], and further work will be necessary to resolve this issue. One of the purposes of our study of coagulation in Anopheles using techniques developed in Drosophila was to see how clots differ in different insect species, since few of the clotting factors identified in Drosophila [8,9] had homologs in other species whose genomes have been sequenced, including Anopheles. Using the pullout technique, we identified lipophorin and PO as Anopheles clotting factors. On the one hand, both of these proteins have been shown to be involved in coagulation in Drosophila and other insect and arthropod species. On the other hand, Anopheles seems unique among these insects in having such a delicate clot. It may be that a delicate clot, like the fine sheet structures we observed, is sufficient for aquatic mosquito larvae with low hemolymph pressure, which do not require as strong a clot as terrestrial Drosophila larvae with higher hemolymph pressure. Future experiments should address whether knocking down PO abolishes coagulation and causes injured larvae to bleed to death, and what role the clot plays in preventing infection of mosquito larvae that live in water teeming with microbes. Our results also have implications for Drosophila coagulation. We note the gross similarity of Anopheles hemolymph to hml mutant Drosophila, which despite lacking hemolectin, a major clotting factor, nonetheless stop bleeding and survive injury (Lesch and Dushay, unpublished observations). It remains to be shown whether residual hemostatic mechanisms at work in hml mutant Drosophila larvae have similarities to clotting in Anopheles larvae. If so, it would be reasonable to expect that studies of coagulation in these two insect species will complement each other. Finally, coagulation in Anopheles may share components or processes with other immune responses. A primary defensive response to malaria parasite infection is the formation of a melanotic capsule. Similarities have been observed between coagulation and encapsulation in other insect species [41–43]. We note that encapsulation in mosquitoes may be different from other insects in apparently not requiring direct participation of hemocytes [44–46], but our results show that coagulation in mosquitoes also appears different
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
from other insects. This leaves open the possibility of shared mechanisms between the two immune responses, which will be resolved by further studies in this important insect. Acknowledgments The authors acknowledge Professor Fotis Kafatos and Dr. Kristin Michel for providing facilities, encouragement, and assistance; and Dolores Doherty for cheerful technical support on mosquito colonies. The authors thank Dr. Uli Theopold for advice, and reviewers for critical evaluation of the manuscript. BA is grateful to Prof. Fotis Kafatos for financial support and guidance. C.L. and M.S.D were supported by the Swedish National Research Council and M.S.D. was also supported by a shortterm EMBO fellowship, So¨derto¨rns ho¨gskola, and Uppsala University.
References [1] Gre´goire C. Hemolymph coagulation. In: Rockstein M, editor. Physiology of insecta, vol. 5. New York: Academic Press; 1974. p. 309–60. [2] Bohn H. Hemolymph clotting in insects. In: Brehe´lin M, editor. Immunity in invertebrates. Berlin: Springer; 1986. p. 188–207. [3] Theopold U, Schmidt O, So¨derha¨ll K, Dushay MS. Coagulation in arthropods: defence, wound closure and healing. Trends Immunol 2004;25:289–94. [4] Osaki T, Kawabata S. Structure and function of coagulogen, a clottable protein in horseshoe crabs. Cell Mol Life Sci 2004;61:1257–65. [5] Sritunyalucksana K, So¨derha¨ll K. The proPO and clotting system in crustaceans. Aquaculture 2000;191:53–69. [6] Hall M, Wang R, Antwerpen Rv, Sottrup-Jensen L, So¨derha¨ll K. The crayfish plasma clotting protein: a vitellogenin-related protein responsible for clot formation in crustacean blood. Proc Natl Acad Sci, USA 1999;96: 1965–70. [7] Bidla G, Lindgren MPM, Theopold U, Dushay MS. Hemolymph coagulation in insect larvae and the role of phenoloxidase in Drosophila. Dev Comp Immunol 2005;29: 669–79. [8] Scherfer C, Karlsson C, Loseva O, Bidla G, Goto A, Havemann J, et al. Isolation and characterization of hemolymph clotting factors in Drosophila melanogaster by a novel pull-out assay. Curr Biol 2004;14:625–9. [9] Karlsson C, Korayem AM, Scherfer C, Dushay MS, Theopold U. Proteomic analysis of the Drosophila larval hemolymph clot. J Biol Chem 2004;279:52033–41. [10] Ganfornina MD, Sanchez D. Generation of evolutionary novelty by functional shift. Bioessays 1999;21:432–9. [11] Levashina EA. Immune responses in Anopheles gambiae. Insect Biochem Mol Biol 2004;34:673–8.
887
[12] Osta MA, Christophides GK, Vlachou D, Kafatos FC. Innate immunity in the malaria vector Anopheles gambiae: comparative and functional genomics. J Exp Biol 2004;207: 2551–63. [13] Christophides GK, Vlachou D, Kafatos FC. Comparative and functional genomics of the innate immune system in the malaria vector Anopheles gambiae. Immunol Rev 2004;198: 127–48. [14] Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, et al. The genome sequence of the malaria mosquito Anopheles gambiae. Science 2002;298:129–49. [15] Dimopoulos G, Richman A, Mu¨ller H-M, Kafatos FC. Molecular immune responses of the mosquito Anopheles gambiae to bacteria and malaria parasites. Proc Natl Acad Sci, USA 1997;94:11508–13. [16] Galko MJ, Krasnow MA. Cellular and genetic analysis of wound healing in Drosophila larvae. PLoS Biol 2004;2:E239. [17] Goto A, Kadowaki T, Kitagawa Y. Drosophila hemolectin gene is expressed in embryonic and larval hemocytes and its knock down causes bleeding defects. Dev Biol 2003;264: 582–91. [18] Gellissen G. Lipophorin as the plasma coagulen in Locusta migratoria. Naturwissenschaften 1983;70:45–6. [19] Barwig B. Isolation and characterization of plasma coagulen (PC) of the cockroach Leucophaea madera (Blatteria). J Comp Physiol B 1985;155:135–43. [20] Duvic B, Brehe´lin M. Two major proteins from locust plasma are involved in coagulation and are specifically precipitated by laminarin, a beta-1,3-glucan. Insect Biochem Mol Biol 1998;28:959–67. [21] Li D, Scherfer C, Korayem AM, Zhao Z, Schmidt O, Theopold U. Insect hemolymph clotting: evidence for interaction between the coagulation system and the prophenoloxidase activating cascade. Insect Biochem Mol Biol 2002;32:919–28. [22] Mu¨ller HM, Dimopoulos G, Blass C, Kafatos FC. A hemocyte-like cell line established from the malaria vector Anopheles gambiae expresses six prophenoloxidase genes. J Biol Chem 1999;274:11727–35. [23] Cerenius L, So¨derha¨ll K. The prophenoloxidase-activating system in invertebrates. Immunol Rev 2004;198:116–26. [24] Jiang H, Wang Y, Korochkina SE, Benes H, Kanost MR. Molecular cloning of cDNAs for two pro-phenol oxidase subunits from the malaria vector, Anopheles gambiae. Insect Biochem Mol Biol 1997;27:693–9. [25] Li JS, Ruyl Kim S, Christensen BM, Li J. Purification and primary structural characterization of prophenoloxidases from Aedes aegypti larvae. Insect Biochem Mol Biol 2005; 35:1269–83. [26] Ryan RO, Kay CM, Oikawa K, Liu H, Bradley R, Scraba DG. Effect of particle lipid content on the structure of insect lipophorins. J Lipid Res 1992;33:55–63. [27] Ashida C, Yoshida Y. Limited proteolysis of prophenoloxidase during activation by microbial products in insect plasma and effect of phenoloxidase on electrophoretic mobilities of plasma proteins. Insect Biochem 1988;18:11–9. [28] Chino H, Hirayama Y, Kyomoto Y, Downer RGH, Takahashi K. Spontaneous aggregation of locust liphophorin during hemolymph collection. Insect Biochem 1987;17:89–97. [29] Ma G, Hay D, Li D, Asgari S, Schmidt O. Recognition and inactivation of LPS by lipophorin particles. Dev Comp Immunol 2006;30:619–26.
ARTICLE IN PRESS 888
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
[30] Rahman MM, Ma G, Roberts HL, Schmidt O. Cell-free immune reactions in insects. J Insect Physiol 2006;52:754–62. [31] Cheon HM, Shin SW, Bian G, Park JH, Raikhel AS. Regulation of lipid metabolism genes, lipid carrier protein Lipophorin, and its receptor during immune challenge in the mosquito Aedes aegypti. J Biol Chem 2006;281:8426–35. [32] Vlachou D, Schlegelmilch T, Christophides GK, Kafatos FC. Functional genomic analysis of midgut epithelial responses in Anopheles during Plasmodium invasion. Curr Biol 2005;15:1185–95. [33] Christophides GK, Zdobnov E, Barillas-Mury C, Birney E, Blandin S, Blass C, et al. Immunity-related genes and gene families in Anopheles gambiae. Science 2002;298:159–65. [34] Paskewitz SM, Shi L. The hemolymph proteome of Anopheles gambiae. Insect Biochem Mol Biol 2005;35: 815–24. [35] Volz J, Osta MA, Kafatos FC, Muller HM. The roles of two clip domain serine proteases in innate immune responses of the malaria vector Anopheles gambiae. J Biol Chem 2005; 280:40161–8. [36] Ratcliffe NA, Rowley AF. A comparative synopsis of the structure and function of the blood cells of insects and other invertebrates. Dev Comp Immunol 1979;3:189–221. [37] Rowley AF, Ratcliffe NA. The granular cells of Galleria mellonella during clotting and phagocytic reactions in vitro. Tissue Cell 1976;8:437–46. [38] Kawooya JK, Keim PS, Ryan RO, Shapiro JP, Samaraweera P, Law JH. Insect apolipophorin III. Purification and properties. J Biol Chem 1984;259:10733–7. [39] Shapiro JP, Keim PS, Law JH. Structural studies on lipophorin, an insect lipoprotein. J Biol Chem 1984;259: 3680–5.
[40] Kawooya JK, Wells MA, Law JH. A strategy for solubilizing delipidated apolipoprotein with lysophosphatidylcholine and reconstitution with phosphatidylcholine. Biochemistry 1989;28:6658–67. [41] Schmit AR, Ratcliffe NA. The encapsulation of foreign tissue implants in Galleria mellonella larvae. J Insect Physiol 1977;23:175–84. [42] Rowley AF, Ratcliffe NA. A histological study of wound healing and hemocyte function in the wax-moth Galleria mellonella. J Morphol 1978;157:181–200. [43] Pech LL, Strand MR. Granular cells are required for encapsulation of foreign targets by insect haemocytes. J Cell Sci 1996;109:2053–60. [44] Hall DW. Mosquito hemocytes: a review. Dev Comp Immunol 1983;7:1–12. [45] Hernandez-Martinez S, Lanz H, Rodriguez MH, GonzalezCeron L, Tsutsumi V. Cellular-mediated reactions to foreign organisms inoculated into the hemocoel of Anopheles albimanus (Diptera: Culicidae). J Med Entomol 2002;39: 61–9. [46] Kumar S, Christophides GK, Cantera R, Charles B, Han YS, Meister S, et al. The role of reactive oxygen species on Plasmodium melanotic encapsulation in Anopheles gambiae. Proc Natl Acad Sci, USA 2003;100:14139–44. [47] Korayem A, Fabbri M, Takahashi K, Scherfer C, Lindgren M, Schmidt O, et al. A glue-like Drosophila salivary protein is expressed in hemocytes: evidence for a function in hemolymph coagulation. Insect Biochem Mol Biol 2004;34: 1297–304. [48] Korochkina SE, Gordadze AV, York JL, Benes H. Mosquito hexamerins: characterization during larval development. Insect Mol Biol 1997;6:11–21.
Developmental & Comparative Immunology
Developmental and Comparative Immunology 31 (2007) 879–888
www.elsevier.com/locate/devcompimm
Preliminary characterization of hemolymph coagulation in Anopheles gambiae larvae Bogos Agianiana,b, Christine Leschc, Olga Losevac,d, Mitchell S. Dushaye,f, a
European Molecular Biology Laboratory, Meyerhofstrasse 1, 69117 Heidelberg, Germany Department of Molecular Biology and Genetics, Democritus University of Thrace, Dimitras 19, 68100 Alexandroupolis, Greece c Department of Molecular Biology and Functional Genomics, Stockholm University, 106 91 Stockholm, Sweden d Department of Genetics, Microbiology, and Toxicology, Stockholm University, 106 91 Stockholm, Sweden e Department of Life Sciences, So¨derto¨rns ho¨gskola, 141 89 Huddinge, Sweden f Department of Comparative Physiology, Uppsala University, Norbyva¨gen 18A, 752 36 Uppsala, Sweden
b
Received 30 November 2006; received in revised form 8 December 2006; accepted 18 December 2006 Available online 25 January 2007
Abstract Hemolymph coagulation is a first response to injury, impeding infection, and ending bleeding. Little is known about its molecular basis in insects, but clotting factors have been identified in the fruit fly Drosophila melanogaster. Here, we have begun to study coagulation in the aquatic larvae of the malaria vector mosquito Anopheles gambiae using methods developed for Drosophila. A delicate clot was seen by light microscopy, and pullout and proteomic analysis identified phenoloxidase and apolipophorin-I as major candidate clotting factors. Electron microscopic analysis confirmed clot formation and revealed it contains fine molecular sheets, most likely a result of lipophorin assembly. Phenoloxidase appears to be more critical in clot formation in Anopheles than in Drosophila. The Anopheles larval clot thus differs in formation, structure, and composition from the clot in Drosophila, confirming the need to study coagulation in different insect species to learn more about its evolution and adaptation to different lifestyles. r 2007 Published by Elsevier Ltd. Keywords: Anopheles gambiae; Drosophila; Coagulation; Lipophorin; Phenoloxidase; Insect immunity
1. Introduction Coagulation is one of the first responses to injury in insects as in other animals, and the clot prevents infection as well as stops bleeding and contributes to wound healing (reviewed in [1–3]). While much is known about coagulation in Limulus [4], and studies Corresponding author. Department of Comparative Physiol-
ogy, Uppsala University, Norbyva¨gen 18A, 752 36 Uppsala, Sweden. Tel.: +46 018 471 2805; fax: +46 018 471 6425. E-mail address: [email protected] (M.S. Dushay). 0145-305X/$ - see front matter r 2007 Published by Elsevier Ltd. doi:10.1016/j.dci.2006.12.006
have been done in other arthropods such as shrimp [5] and crayfish [6], the molecular mechanisms of coagulation in insects are not well understood. Studies have begun in Drosophila melanogaster larvae, where the clot has been described [7], and methods have been developed to isolate and identify candidate clotting factors [8,9]. Remarkably, few of these clotting factors have homologs in other insects [8]. This suggests that insect coagulation may have evolved through co-option of genes serving other functions [3,10]. The lack of conservation makes it important to study coagulation in different insects,
ARTICLE IN PRESS 880
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
to learn how the clot is formed in different species, and by comparison to learn how evolution acted on this important defense process. The mosquito Anopheles gambiae is the subject of intense interest because of its role as malaria vector; its humoral immune response has been studied [11–13], and genome published [14]. Although it is a dipteran insect like Drosophila, the two species are separated by 250 million years of evolution, and mosquito larvae are aquatic, so both time and different environmental constraints have acted on coagulation in these species. Coagulation in Anopheles is also of interest because some of its mechanisms may be shared with other immune reactions, including those against malaria parasites, and for the possibility that coagulation could provide a new target for pest control strategies. We have studied coagulation in Anopheles larvae by applying techniques developed to study clotting in Drosophila. To our surprise, we could not draw out strands from coagulating mosquito larval hemolymph as with Drosophila [7], suggesting that the mosquito clot is much more delicate. We used more sensitive techniques to demonstrate the formation of a clot in Anopheles larvae, and we used pullout [8] to identify phenoloxidase (PO) and apolipophorin-I as candidate clotting factors in these mosquitoes. Electron microscopy of negatively stained clot preparations revealed the Anopheles clot is at least partly composed of paracrystalline sheets. Taken together, our data further suggest that these sheets are produced by the assembly of lipophorin particles in a process dependent on PO. Further work will be required to identify additional clotting factors, demonstrate how clot formation is activated, and eluciate the role of coagulation in immune defense in mosquito larvae.
humid chamber 30 min RT. The grids were washed twice in 10 ml drops of 0.1 PBS, 2 in 10 ml 0.5% IGEPAL (Sigma) in 0.1 PBS, and twice in 0.1 PBS. The clot was then stained with 10 ml 0.1 PBS containing 10 mg/ml lectin (either HPL-FITC or PNA-FITC) for at least 20 min, and visualized in a Zeiss Axioplan 2 microscope. 2.3. Bead aggregation Bead aggregation was performed essentially as described in [8]. Tosylactivated Dynabeads (M-280, Dynal) were washed and blocked overnight with 0.2 M Tris (pH 8.5) according to manufacturer’s instructions, then washed and resuspended in 0.1 PBS pH 8.0. For bead aggregation experiments, both Drosophila and Anopheles, 5 larvae were bled onto 10 ml of beads. 2.4. Pullout
2. Materials and methods
Pullout was performed as described in [8]. Briefly, fourth instar Anopheles larvae were rinsed in dH2O and opened with fine antimagnetic scissors under the surface of 50 ml beads held in a watch glass. The larvae were swirled in the buffer during bleeding to maximize contact of clotting factors with the beads. The solution and the beads were then transferred to a microcentrifuge tube, and the watch glass was washed with an additional 100 ml of buffer, which was added to the tube. Tubes were put in a magnetic holder to attract the beads to the side of the tube, and the solution was removed by pipette. Beads were washed (i) 3 in 0.1 PBS, (ii) 3 in 0.1 PBS, 0.5% IGEPAL (Nonidet 40) detergent, and Roche protease inhibitor, and (iii) 3 in 0.1 PBS. The beads were then resuspended in 16 ml 0.1 PBS+5 ml gel loading buffer, heated to 64 1C 10 min, and then stored at 20 1C.
2.1. Mosquito larvae
2.5. Transmission electron microscopy (TEM)
A. gambiae G3 larvae were raised in standard conditions [15]. Fourth instar larvae were used in all experiments because they are large and close in size to third instar Drosophila larvae (Fig. 1A).
Five third instar Drosophila, or 10 fourth instar Anopheles larvae were bled into 10 ml 0.1 PBS (with, or without PTU) on top of glow-discharged carbon-coated grids and incubated for 20 min at RT in humid chambers. Grids were then washed twice in 0.1 PBS, and stained in 1% uranylacetate in 0.1 PBS for 1 min. The samples were examined in a Philips CM120 Biotwin TEM operating at 100 kV. Images were captured using a retractable CCD camera.
2.2. Light microscopy of clot preparations Five Drosophila or Anopheles larvae were bled by ripping open with fine forceps into 10 ml 0.1 PBS pH 8.0 over a copper SEM grid and incubated in a
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
881
Fig. 1. Anopheles fourth instar and Drosophila third instar larvae: (A) Fourth instar Anopheles gambiae larva above, and third instar Drosophila melanogaster larva below, shown against a millimeter scale to indicate the comparable size of the two species. (B) Melanization after wounding. Larvae of the two species are shown after wounding with a fine tungsten needle to demonstrate that like Drosophila, Anopheles larvae are able to heal wounds, which are subsequently melanized. Melanized healed wounds are indicated by arrows.
2.6. MALDI-TOF mass spectrometry analysis and protein identification Proteomic analysis was done essentially as described in [9]; coomassie-stained protein bands were excised from gels using sterile stainless-steel blades. The gel pieces were destained with 50% acetonitrile in 25 mM ammonium bicarbonate and finally dehydrated with 100% acetonitrile. Proteins were in-gel digested with 12.5 ng/ml trypsin (Promega V511A) in 50 mM ammonium bicarbonate at 37 1C overnight. The resultant peptides were extracted with 50% acetonitrile/5% trifluoroacetic acid and dried in a vacuum centrifuge. The recovered peptides were purified and concentrated
on a C18ZipTips (Millipore) column according to manufacturer’s instructions. Mass spectra (MS) were recorded in positive reflection mode using an Applied Biosystems MALDI-TOF Voyager-DE STR mass spectrometer equipped with a delayed ion extraction technology. a-cyano-4-hydroxycinnamic acid was used as the matrix. External calibration was performed using the Sequazyme Peptide Mass Standard kit with Angiotensin I (1296.6853 Da) and ACTH clips 1–17 (2093.0867 Da), 18–39 (2465.1989 Da), 7–38 (3657.9294 Da) (PerSeptive Biosystems, USA) and for internal calibration auto digestion peaks of bovine trypsin were used. The peptide mass profiles produced by MS were analyzed by using the
ARTICLE IN PRESS 882
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
programs Mascot (http://www.matrixsciences.com) and MS-Fit (http://prospector.ucsf.edu). The monoisotopic peptide masses were compared with the theoretical masses derived from the NCBInr database for Metazoan animals. Search parameters included allowed mass accuracy of 0.1 Da, more than four peptide mass hits required for a protein match, consideration of one missed enzymatic cleavage, pI range 3.0–10.0, and molecular mass range 1–400 kDa. For lipophorin (ENSANGP00000018348, theoretical MW/pI 360 kDa/7.4), the Mascot search resulted in 47 matched mass values, which represents 20% sequence coverage. The MS-Fit search matched 46 peptides covering 20.8% of the protein. For prophenoloxidase 3 (ENSANGP0000002437, theoretical MW/pI 78.6 kDa/6.8), the Mascot search resulted in 23 matched mass values with sequence coverage of 46%. The MS-Fit search gave 23 matched peptides covering 46.2% of the protein. 3. Results 3.1. The Anopheles clot is finer than the Drosophila clot Fourth instar A. gambiae larvae are comparable in size to third instar Drosophila larvae (Fig. 1A), so we used these for our experiments for purely
practical reasons. Hemolymph from Anopheles larvae differed from Drosophila hemolymph by containing large amounts of what appeared to be lipid droplets. Some, or all of these may have come from fat body damage. In addition, Anopheles hemolymph appeared to melanize more quickly and to a greater extent than Drosophila hemolymph, although we did not measure this quantitatively. Anopheles larvae stopped bleeding and formed a melanized scar/scab after injury with a fine needle (Fig. 1B), like Drosophila [16]. This clearly suggested, coagulation occurred, and that it would be possible to study the Anopheles larval clot using the draw out, bead aggregation, and pullout techniques developed in Drosophila [7,8]. To our surprise, we were unable to draw coagulating strands out of pooled Anopheles hemolymph, as we could easily do for Drosophila (see also [7]). Furthermore, Anopheles hemolymph did not aggregate dynabeads, an assay that is indicative of clotting in Drosophila [8] (Fig. 2A–C). Because the size or weight of the dynabeads might hinder their aggregation by a finer clot, we tried bleeding hemolymph onto finer India Ink particles. Again, despite trying different incubation times, buffer pH, and salt composition, we did not observe particle aggregation (data not shown). This suggested that the Anopheles clot was substantially different and more delicate than
Fig. 2. Bead aggregation and detergent washed clot on SEM grids stained with FITC-coupled lectins: The well diameter for (A–C) was 8 mm, while the SEM grids in (D–F) had gaps of 23 mm. (A) Drop of untreated dynabeads. (B) Drop of dynabeads incubated with Anopheles larval hemolymph. Note lack of bead aggregation. (C) Drop of dynabeads incubated with Drosophila wild-type larval hemolymph to show bead aggregation. (D) Clot from Drosophila larval hemolymph stained with PNA-FITC for comparison to E and F to show how a different lectin stains clots from the different species. The staining of the Drosophila clot with PNA has been described [7,47]. 66 magnification. (F) Clot from Anopheles gambiae washed with detergent and stained with HPL-FITC to show Anopheles clot material can span gaps in an SEM grid. 66 magnification. Arrowheads indicate staining clot material. (E) Different view of Anopheles gambiae clot at higher, 100 magnification, prepared in the same way as Fig. 2E. Arrowheads indicate staining clot material.
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
Drosophila’s, and that alternative methods would be required to study coagulation in the mosquito. 3.2. Microscopic evidence of an Anopheles clot To see the Anopheles hemolymph clot, we bled larvae onto copper SEM grids with square holes 23 mm wide and looked for gap-spanning material. Extremely fine, wisp-like material spanning the corners of the SEM grids was barely detectable in phase contrast microscopy, in contrast to the bulkier Drosophila clot formed under similar conditions (data not shown). The Drosophila clot is more easily visualized when stained with FITCconjugated peanut agglutinin (PNA) [7], and this clearly labeled the Drosophila clot which adhered to SEM grids through detergent washes (Fig. 2D). However, this lectin did not decorate our Anopheles preparations. Instead, we found that the Anopheles clot stained with FITC-conjugated Helix pomatia lectin (HPL) (arrowheads, Fig. 2E and F). Despite the difference in lectin binding, the occurrence of these fine strands and their persistence through detergent washes strongly indicates a discrete clot. PNA and HPL bind to different sugars (galactose vs. terminal N-acetyl-a-D-galactosamine, respectively), and their differential labeling of Drosophila and Anopheles clots suggests different glycan composition in the two species, either of glycosylated protein clotting factors or constituent lipopolysaccharides. To investigate the role of PO in Anopheles coagulation, we bled larvae onto copper grids in the presence of PTU, an inhibitor of PO. In contrast to the results described above, clot material was very rarely seen on SEM grids in the presence of PTU (data not shown). This suggested that PO could be important for clot formation in Anopheles, in contrast to Drosophila, where PTU does not inhibit clot formation (data not shown), and where it has been demonstrated genetically that PO is not necessary for primary soft clot formation [7]. 3.3. Identifying candidate clotting factors Having demonstrated clot formation in Anopheles, we set out to identify candidate clotting proteins. The pullout method of isolating proteins on dynabeads has been used to identify clotting factors in Drosophila [8]. However, this assay was developed based on the ability of Drosophila hemolymph to aggregate beads. Because Anopheles
883
hemolymph did not aggregate beads, we first performed control experiments to demonstrate that clotting factors adhere to beads even in the absence of bead aggregation. For this, we made use of Drosophila hemolectin (hml) mutants [17], whose hemolymph does not allow coagulating strands to be drawn or bead aggregation ([8], and Lesch and Dushay, unpublished observations). Nonetheless, when we performed pullouts on hemolymph from hml larvae, we obtained protein patterns similar to wild-type Drosophila, except for the absence of Hml protein (Fig. 3A). Thus, under our conditions, Drosophila clotting factors bind to beads even in the absence of bead aggregation. We then isolated Anopheles candidate clotting proteins using the same pullout assay. Pullouts of hemolymph from fourth instar Anopheles larvae showed four prominent bands, a single band around 260 kDa, a doublet centered at 65 kDa and a lower band at 55 kDa (see arrows in Fig. 3B). The bands were excised and analyzed proteomically (as described in Section 2, where sequence coverage and other details are also provided). The tryptic peptide analysis identified the 260 kDa band as lipophorin. The lipophorin band matched the expected MW for apolipophorin-I (apoLp-I), the heavy component of lipophorin. Consistent with this, lipophorin has been identified as a clotting factor in Drosophila [8,9], and in several other insect species [2,18–21]. Tryptic peptide analysis of the 65 and 55 kDa bands identified them as prophenoloxidase 3 (PPO3) with theoretical MW 78.6 kDa. Our finding two bands of smaller molecular weights suggests these were most likely products of specific proteolytic processing, a result of activation of the PPO cascade [23] and production of PO. Similarly sized PO polypeptides, including doublets most likely representing heteromerization, have been detected in the hemolymph of A. gambiae [24] and Aedes aegypti [25]. Our identification of PPO3 is consistent with the high expression of this isoform in late instar larvae [22], However, nine PPO genes have been annotated in the Anopheles genome ([22] and www.ensembl.org/Anopheles_gambiae) with high sequence identity (470%) between their products, so we do not exclude the possibility that other PO isoforms also participate in the clot. 3.4. TEM on clot preparations The extremely thin and fragile nature of the Anopheles clot suggested by the absence of bead
ARTICLE IN PRESS 884
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
aggregation or strand formation and our light microscopy observations led us to examine clot preparations by TEM. Negatively stained preparations of Anopheles and Drosophila hemolymph were made as described in Section 2, including Anopheles samples incubated with PO-inhibitor PTU to test whether PO has an effect on clot formation. When Anopheles hemolymph was incubated in the presence of PTU, an abundance of approximately spherical particles about 10 nm in diameter were observed distributed almost uniformly on the grids (Fig. 4A and B). These particles have notable resemblance to insect lipophorin purified from hemolymph, both in shape and size [26], which allows us to postulate that they represent lipophorin molecules. Remarkably, when we bled Anopheles larvae and incubated hemolymph under identical conditions but in the absence of PTU—allowing PO activity, we observed regular, paracrystalline sheets (Fig. 4C and D, and arrowhead in Fig. 4D), which appeared to be formed from laterally stacked fibers (diameter ca. 10 nm) that are beaded along their long axis (Fig. 4C and D). We also observed sporadic fiber structures of much larger diameter (ca. 50 nm), which stain uniformly along their length (Fig. 4E and F). The molecular nature of these thicker fibers remains unidentified. The same preparations consistently showed a markedly lower number of free particles like the ones shown in Fig. 4A and B. The sheets were never observed in the presence of PTU in several repetitions of the same assays with different larvae, however, they were always observed in the absence of PTU. TEM of similarly prepared Drosophila hemolymph showed a coarser, thicker clot, in which no regular structure was visible (data not shown). Fig. 3. Pullout of Drosophila and Anopheles hemolymph: (A) Drosophila proteins binding to beads analyzed by PAGE on 5% gels. While loss of Hml protein does not allow strand draw out and abolishes bead aggregation, the proteins adhering to the beads show a similar pattern to wild-type hemolymph, except for the absence of Hml (indicated by an arrow). Lane 1: pullout from wild-type hemolymph. Lanes 2 and 3: independent pullouts from Hml mutant larval hemolymph. (B) Pullout from Anopheles gambiae analyzed by PAGE on 10% gel. Apolipophorin (arrow) and phenoloxidase bands (arrowheads) are indicated. Lane 1: pullout from 10 fourth instar larvae. Lane 2: Whole hemolymph from two larvae. We note that the hexamerin bands in Lane 2 appear less abundant in comparison to other published gel patterns of Anopheles larval hemolymph c.f. [48]. This is probably a result of procedural differences (our samples were not diluted and quick frozen at 80 1C).
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
885
Fig. 4. TEM of negatively stained Anopheles hemolymph with and without PTU: (A–B) Different views of preparations with PTU. Scale bars 100 nm. (C–F) Different TEM views of negatively stained Anopheles hemolymph preparations in the absence of PTU. When PTU is omitted, phenoloxidase quickly melanizes hemolymph. Now, it appears that the putative lipophorin molecules associate into a paracrystalline matrix, indicated by an arrowhead in (D). Scale bars: (C) 50 nm, (D) 200 nm, (E) 20 nm, (F) 50 nm.
These observations, in combination with the pullout results (see Section 3.3) lead us to suggest that during Anopheles coagulation, PO activity causes lipophorin particles to coalesce into the sheet structures we observe. A protein crosslinking action of PO is well documented [23], and PO was proposed to crosslink lipophorin in Bombyx mori hemolymph, although this was not linked to coagulation [27]. It has been noted that lipophorin has a tendency to precipitate in low ionic strength conditions [28], but the ordered structures we see (e.g. Fig. 4C and D) are inconsistent with random aggregation. We are not able to distinguish whether PO remains attached to the sheets, or if it takes part as a structural component of the clot.
4. Discussion To our knowledge, this is the first study of coagulation in larvae of the mosquito A. gambiae. We have demonstrated clot formation and shown that the mosquito clot is much more delicate than the clot found in Drosophila. The Anopheles clot stained with HPL (GalNAc), rather than PNA (Gal) which stains the Drosophila clot (Fig. 2), suggesting that Anopheles clotting factors are glycosylated differently than in Drosophila. The functional significance of this finding remains to be shown. Coagulating Anopheles hemolymph is not viscous, does not allow strands to be drawn out, and
ARTICLE IN PRESS 886
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
does not aggregate beads. However, candidate clotting factors can still be isolated by pullout, as demonstrated by control experiments in Drosophila (Fig. 3A). The main mosquito clotting factors we identified were lipophorin and PO (Fig. 3B). While both lipophorin and PO were identified in Drosophila pullouts [8], we show here that they are more prominent, and appear to play a greater role in Anopheles coagulation. A vitellogenin-like protein is the plasma coagulogen in crayfish [6], and lipophorin forms detergent-insoluble precipitates in Galleria [29] and Ephestia [30], and has been identified as a clotting factor in many insect species including Drosophila (references cited in Section 3). An immune role for lipophorin in mosquitoes is supported by the results of Cheon et al. [31] who found lipophorin is upregulated in Aedes aegypti following immune challenge in a Toll/Reldependent way, and by reports showing that the lipophorin precursor gene is essential for the development of Plasmodium parasites in A. gambiae [32]. PO is not required for primary soft clot formation in Drosophila [7]. In contrast, we have demonstrated that PO activity is required for coagulation in Anopheles. Fluorescence light microscopy showed that HPL-staining clot material adhered to SEM grids in the absence, but not in the presence of the PO inhibitor PTU, in contrast to Drosophila, where PTU does not have this effect. PO is an important enzyme for immune reactions in mosquitoes [33,34], its activity is critical for protection against parasite infection [35], and it has long been held to be critical for coagulation in different insect species [2,36,37]. In order to observe details of the fine clot detected by light microscopy and to assess a potential role of PO, we investigated larval hemolymph by TEM. In the presence of PTU, there was an abundance of free 10 nm particles, and these coalesced into paracrystalline arrays in the absence of PTU (Fig. 4). Based on our assumption that these particles represent lipophorin, as suggested by their size and abundance, this suggests that PO acts to crosslink lipophorin during coagulation. Although a spontaneous tendency of lipophorin to aggregate has been previously reported [28], our results show a controlled and specific structural organization inconsistent with random aggregation (Fig. 4C and D). Our proteomic analysis of pullout bands identified only the high molecular mass component of lipophorin, apoLP-1. However, we cannot conclude whether apoLP-II is also part of the clot, since a
smaller apoLP-II band in Fig. 3B would be very faint, assuming a 1:1 ratio to apoLP-1. Apolipophorins I and II are usually tightly associated and dissociate only under strongly chaotic conditions [38–40], and further work will be necessary to resolve this issue. One of the purposes of our study of coagulation in Anopheles using techniques developed in Drosophila was to see how clots differ in different insect species, since few of the clotting factors identified in Drosophila [8,9] had homologs in other species whose genomes have been sequenced, including Anopheles. Using the pullout technique, we identified lipophorin and PO as Anopheles clotting factors. On the one hand, both of these proteins have been shown to be involved in coagulation in Drosophila and other insect and arthropod species. On the other hand, Anopheles seems unique among these insects in having such a delicate clot. It may be that a delicate clot, like the fine sheet structures we observed, is sufficient for aquatic mosquito larvae with low hemolymph pressure, which do not require as strong a clot as terrestrial Drosophila larvae with higher hemolymph pressure. Future experiments should address whether knocking down PO abolishes coagulation and causes injured larvae to bleed to death, and what role the clot plays in preventing infection of mosquito larvae that live in water teeming with microbes. Our results also have implications for Drosophila coagulation. We note the gross similarity of Anopheles hemolymph to hml mutant Drosophila, which despite lacking hemolectin, a major clotting factor, nonetheless stop bleeding and survive injury (Lesch and Dushay, unpublished observations). It remains to be shown whether residual hemostatic mechanisms at work in hml mutant Drosophila larvae have similarities to clotting in Anopheles larvae. If so, it would be reasonable to expect that studies of coagulation in these two insect species will complement each other. Finally, coagulation in Anopheles may share components or processes with other immune responses. A primary defensive response to malaria parasite infection is the formation of a melanotic capsule. Similarities have been observed between coagulation and encapsulation in other insect species [41–43]. We note that encapsulation in mosquitoes may be different from other insects in apparently not requiring direct participation of hemocytes [44–46], but our results show that coagulation in mosquitoes also appears different
ARTICLE IN PRESS B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
from other insects. This leaves open the possibility of shared mechanisms between the two immune responses, which will be resolved by further studies in this important insect. Acknowledgments The authors acknowledge Professor Fotis Kafatos and Dr. Kristin Michel for providing facilities, encouragement, and assistance; and Dolores Doherty for cheerful technical support on mosquito colonies. The authors thank Dr. Uli Theopold for advice, and reviewers for critical evaluation of the manuscript. BA is grateful to Prof. Fotis Kafatos for financial support and guidance. C.L. and M.S.D were supported by the Swedish National Research Council and M.S.D. was also supported by a shortterm EMBO fellowship, So¨derto¨rns ho¨gskola, and Uppsala University.
References [1] Gre´goire C. Hemolymph coagulation. In: Rockstein M, editor. Physiology of insecta, vol. 5. New York: Academic Press; 1974. p. 309–60. [2] Bohn H. Hemolymph clotting in insects. In: Brehe´lin M, editor. Immunity in invertebrates. Berlin: Springer; 1986. p. 188–207. [3] Theopold U, Schmidt O, So¨derha¨ll K, Dushay MS. Coagulation in arthropods: defence, wound closure and healing. Trends Immunol 2004;25:289–94. [4] Osaki T, Kawabata S. Structure and function of coagulogen, a clottable protein in horseshoe crabs. Cell Mol Life Sci 2004;61:1257–65. [5] Sritunyalucksana K, So¨derha¨ll K. The proPO and clotting system in crustaceans. Aquaculture 2000;191:53–69. [6] Hall M, Wang R, Antwerpen Rv, Sottrup-Jensen L, So¨derha¨ll K. The crayfish plasma clotting protein: a vitellogenin-related protein responsible for clot formation in crustacean blood. Proc Natl Acad Sci, USA 1999;96: 1965–70. [7] Bidla G, Lindgren MPM, Theopold U, Dushay MS. Hemolymph coagulation in insect larvae and the role of phenoloxidase in Drosophila. Dev Comp Immunol 2005;29: 669–79. [8] Scherfer C, Karlsson C, Loseva O, Bidla G, Goto A, Havemann J, et al. Isolation and characterization of hemolymph clotting factors in Drosophila melanogaster by a novel pull-out assay. Curr Biol 2004;14:625–9. [9] Karlsson C, Korayem AM, Scherfer C, Dushay MS, Theopold U. Proteomic analysis of the Drosophila larval hemolymph clot. J Biol Chem 2004;279:52033–41. [10] Ganfornina MD, Sanchez D. Generation of evolutionary novelty by functional shift. Bioessays 1999;21:432–9. [11] Levashina EA. Immune responses in Anopheles gambiae. Insect Biochem Mol Biol 2004;34:673–8.
887
[12] Osta MA, Christophides GK, Vlachou D, Kafatos FC. Innate immunity in the malaria vector Anopheles gambiae: comparative and functional genomics. J Exp Biol 2004;207: 2551–63. [13] Christophides GK, Vlachou D, Kafatos FC. Comparative and functional genomics of the innate immune system in the malaria vector Anopheles gambiae. Immunol Rev 2004;198: 127–48. [14] Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, et al. The genome sequence of the malaria mosquito Anopheles gambiae. Science 2002;298:129–49. [15] Dimopoulos G, Richman A, Mu¨ller H-M, Kafatos FC. Molecular immune responses of the mosquito Anopheles gambiae to bacteria and malaria parasites. Proc Natl Acad Sci, USA 1997;94:11508–13. [16] Galko MJ, Krasnow MA. Cellular and genetic analysis of wound healing in Drosophila larvae. PLoS Biol 2004;2:E239. [17] Goto A, Kadowaki T, Kitagawa Y. Drosophila hemolectin gene is expressed in embryonic and larval hemocytes and its knock down causes bleeding defects. Dev Biol 2003;264: 582–91. [18] Gellissen G. Lipophorin as the plasma coagulen in Locusta migratoria. Naturwissenschaften 1983;70:45–6. [19] Barwig B. Isolation and characterization of plasma coagulen (PC) of the cockroach Leucophaea madera (Blatteria). J Comp Physiol B 1985;155:135–43. [20] Duvic B, Brehe´lin M. Two major proteins from locust plasma are involved in coagulation and are specifically precipitated by laminarin, a beta-1,3-glucan. Insect Biochem Mol Biol 1998;28:959–67. [21] Li D, Scherfer C, Korayem AM, Zhao Z, Schmidt O, Theopold U. Insect hemolymph clotting: evidence for interaction between the coagulation system and the prophenoloxidase activating cascade. Insect Biochem Mol Biol 2002;32:919–28. [22] Mu¨ller HM, Dimopoulos G, Blass C, Kafatos FC. A hemocyte-like cell line established from the malaria vector Anopheles gambiae expresses six prophenoloxidase genes. J Biol Chem 1999;274:11727–35. [23] Cerenius L, So¨derha¨ll K. The prophenoloxidase-activating system in invertebrates. Immunol Rev 2004;198:116–26. [24] Jiang H, Wang Y, Korochkina SE, Benes H, Kanost MR. Molecular cloning of cDNAs for two pro-phenol oxidase subunits from the malaria vector, Anopheles gambiae. Insect Biochem Mol Biol 1997;27:693–9. [25] Li JS, Ruyl Kim S, Christensen BM, Li J. Purification and primary structural characterization of prophenoloxidases from Aedes aegypti larvae. Insect Biochem Mol Biol 2005; 35:1269–83. [26] Ryan RO, Kay CM, Oikawa K, Liu H, Bradley R, Scraba DG. Effect of particle lipid content on the structure of insect lipophorins. J Lipid Res 1992;33:55–63. [27] Ashida C, Yoshida Y. Limited proteolysis of prophenoloxidase during activation by microbial products in insect plasma and effect of phenoloxidase on electrophoretic mobilities of plasma proteins. Insect Biochem 1988;18:11–9. [28] Chino H, Hirayama Y, Kyomoto Y, Downer RGH, Takahashi K. Spontaneous aggregation of locust liphophorin during hemolymph collection. Insect Biochem 1987;17:89–97. [29] Ma G, Hay D, Li D, Asgari S, Schmidt O. Recognition and inactivation of LPS by lipophorin particles. Dev Comp Immunol 2006;30:619–26.
ARTICLE IN PRESS 888
B. Agianian et al. / Developmental and Comparative Immunology 31 (2007) 879–888
[30] Rahman MM, Ma G, Roberts HL, Schmidt O. Cell-free immune reactions in insects. J Insect Physiol 2006;52:754–62. [31] Cheon HM, Shin SW, Bian G, Park JH, Raikhel AS. Regulation of lipid metabolism genes, lipid carrier protein Lipophorin, and its receptor during immune challenge in the mosquito Aedes aegypti. J Biol Chem 2006;281:8426–35. [32] Vlachou D, Schlegelmilch T, Christophides GK, Kafatos FC. Functional genomic analysis of midgut epithelial responses in Anopheles during Plasmodium invasion. Curr Biol 2005;15:1185–95. [33] Christophides GK, Zdobnov E, Barillas-Mury C, Birney E, Blandin S, Blass C, et al. Immunity-related genes and gene families in Anopheles gambiae. Science 2002;298:159–65. [34] Paskewitz SM, Shi L. The hemolymph proteome of Anopheles gambiae. Insect Biochem Mol Biol 2005;35: 815–24. [35] Volz J, Osta MA, Kafatos FC, Muller HM. The roles of two clip domain serine proteases in innate immune responses of the malaria vector Anopheles gambiae. J Biol Chem 2005; 280:40161–8. [36] Ratcliffe NA, Rowley AF. A comparative synopsis of the structure and function of the blood cells of insects and other invertebrates. Dev Comp Immunol 1979;3:189–221. [37] Rowley AF, Ratcliffe NA. The granular cells of Galleria mellonella during clotting and phagocytic reactions in vitro. Tissue Cell 1976;8:437–46. [38] Kawooya JK, Keim PS, Ryan RO, Shapiro JP, Samaraweera P, Law JH. Insect apolipophorin III. Purification and properties. J Biol Chem 1984;259:10733–7. [39] Shapiro JP, Keim PS, Law JH. Structural studies on lipophorin, an insect lipoprotein. J Biol Chem 1984;259: 3680–5.
[40] Kawooya JK, Wells MA, Law JH. A strategy for solubilizing delipidated apolipoprotein with lysophosphatidylcholine and reconstitution with phosphatidylcholine. Biochemistry 1989;28:6658–67. [41] Schmit AR, Ratcliffe NA. The encapsulation of foreign tissue implants in Galleria mellonella larvae. J Insect Physiol 1977;23:175–84. [42] Rowley AF, Ratcliffe NA. A histological study of wound healing and hemocyte function in the wax-moth Galleria mellonella. J Morphol 1978;157:181–200. [43] Pech LL, Strand MR. Granular cells are required for encapsulation of foreign targets by insect haemocytes. J Cell Sci 1996;109:2053–60. [44] Hall DW. Mosquito hemocytes: a review. Dev Comp Immunol 1983;7:1–12. [45] Hernandez-Martinez S, Lanz H, Rodriguez MH, GonzalezCeron L, Tsutsumi V. Cellular-mediated reactions to foreign organisms inoculated into the hemocoel of Anopheles albimanus (Diptera: Culicidae). J Med Entomol 2002;39: 61–9. [46] Kumar S, Christophides GK, Cantera R, Charles B, Han YS, Meister S, et al. The role of reactive oxygen species on Plasmodium melanotic encapsulation in Anopheles gambiae. Proc Natl Acad Sci, USA 2003;100:14139–44. [47] Korayem A, Fabbri M, Takahashi K, Scherfer C, Lindgren M, Schmidt O, et al. A glue-like Drosophila salivary protein is expressed in hemocytes: evidence for a function in hemolymph coagulation. Insect Biochem Mol Biol 2004;34: 1297–304. [48] Korochkina SE, Gordadze AV, York JL, Benes H. Mosquito hexamerins: characterization during larval development. Insect Mol Biol 1997;6:11–21.