- Email: [email protected]
Eicosanoids mediate microaggregation reactions to bacterial challenge in isolated insect hemocyte preparations
Journal of Insect Physiology 47 (2001) 1409–1417 www.elsevier.com/locate/jinsphys
Eicosanoids mediate microaggregation reactions to bacterial challenge in isolated insect hemocyte preparations J.S. Miller a, D.W. Stanley b
b,*
a Department of Biological Sciences, Northern Illinois University, De Kalb, IL 60115, USA Insect Biochemical Physiology Laboratory, 311 Plant Industry Building, University of Nebraska, Lincoln, NE 68583-0816, USA
Received 3 April 2001; accepted 31 July 2001
Abstract Nodule formation is the quantitatively predominant insect cellular defense reaction to bacterial challenges, responsible for clearing the largest proportion of infecting bacteria from circulation. It has been suggested that eicosanoids mediate several steps in the nodulation process, including formation of hemocyte microaggregates, an early step in the process. While fat body and hemocytes are competent to biosynthesize eicosanoids, the source of the nodulation-mediating eicosanoids remains unclear. To investigate this issue, we studied hemocyte microaggregation reactions to bacterial challenge in vitro. Hemocyte suspensions from the tobacco hornworm, Manduca sexta, were treated with the phospholipase A2 inhibitor, dexamethasone, then challenged with the bacterium Serratia marcescens. Preparations treated with dexamethasone yielded fewer hemocyte microaggregations than untreated, control preparations. Furthermore, the influence of dexamethasone was reversed by amending experimental (dexamethasone-treated) preparations with the eicosanoid biosynthesis precursor, arachidonic acid. Palmitic acid, which is not a substrate for eicosanoid biosynthesis, did not reverse the influence of dexamethasone on the microaggregation reaction. The influence of dexamethasone was also reversed by adding filtered media from challenged hemocyte preparations to dexamethasone-treated preparations. Finally, most hemocyte preparations treated with selected eicosanoid biosynthesis inhibitors formed fewer hemocyte microaggregations than control preparations. The 5- and 12-lipoxygenase inhibitor, esculetin, did not influence the formation of hemocyte microaggregations in this system. These results are consistent with similar investigations performed in vivo, and we infer that hemocytes are responsible for forming and secreting eicosanoids, which subsequently initiate nodulation by mediating hemocyte microaggregation. 2001 Elsevier Science Ltd. All rights reserved. Keywords: In vitro; Eicosanoids; Insect immunity; Hemocyte microaggregation; Manduca sexta; Serratia marcescens
1. Introduction The innate immune systems in insects and other invertebrates are categorized into hemocytic and humoral immune responses. Hemocytic defense reactions to bacterial infections include phagocytosis and nodule formation (Gupta 1986, 1991). These responses occur within minutes of infection and feature direct cellular interactions between circulating hemocytes and the bacterial cells (Gupta 1986, 1991). While the nodulation reaction is well documented, much less is known about the signaling mechanisms responsible for mediating
* Corresponding author. Tel.: +1-402-472-8710; fax: +1-402-4724687. E-mail address: [email protected] (D.W. Stanley).
nodulation. Because many mammalian cellular defense reactions are mediated by eicosanoids, Stanley-Samuelson et al. (1991) suggested that possibly insect cellular immune reactions also are mediated by eicosanoids. Eicosanoids are oxygenated metabolites of certain C20 polyunsaturated fatty acids. The biological significance, structures and biosynthetic pathways of these compounds are outlined in several reviews (Stanley-Samuelson, 1994a; Stanley and Howard, 1998; Howard and Stanley, 1999; Stanley, 2000). Nodulation is the predominant insect cellular immune reaction, responsible for clearing as much as 90% of infecting bacteria from hemolymph circulation within minutes of infection (Horohov and Dunn, 1982), and we hypothesized that nodulation is a specific cellular immune reaction mediated by eicosanoids. We previously tested this hypothesis in a series of simple
0022-1910/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 0 1 ) 0 0 1 3 1 - 7
1410
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
experiments (Miller et al., 1994). Groups of tobacco hornworms, Manduca sexta, were treated separately with a range of eicosanoid biosynthesis inhibitors, including the phospholipase A2 inhibitor dexamethasone, then artificially infected with the bacterium, Serratia marcescens. Cellular immune reactions to the artificial infections were assessed by counting the number of hemocytic microaggregates (an early phase of nodulation) under phase contrast optics and by counting the number of melanized, mature nodules within the hemocoels using a dissecting microscope. The ability of experimental hornworms to form microaggregates and nodules was severely impaired by the inhibitors. These findings supported the idea that eicosanoids mediate nodule formation in tobacco hornworms (Miller et al., 1994). This idea has been tested in several hemimetabolous and holometabolous insect species, the results of which support the broader hypothesis that eicosanoids mediate insect nodulation reactions to bacterial infections (Miller et al. 1996, 1999; Jurenka et al., 1997; Tunaz et al., 1999; Stanley, 2000). Mandato et al. (1997) identified other eicosanoid-dependent cellular immune reactions, suggesting that eicosanoids mediate certain discrete steps in nodule formation including cell spreading and prophenoloxidase activation, as well as phagocytosis in the wax moth, Galleria mellonella. Morishima et al. (1997) also suggested that eicosanoids mediate expression of two silkworm, Bombyx mori, fat body genes for anti-bacterial proteins, cecropin and lysozyme. The emergent view is that eicosanoids act in several discrete aspects of insect immunity. The idea that eicosanoids mediate various cellular defense reactions to bacterial challenge opens questions on cell–cell signaling in immunity. On a paracrine model, the eicosanoids involved in mediating insect cellular immune reactions could originate in the hemocytes, then influence the actions of other hemocytes. Alternatively, other tissues, particularly fat body, could be responsible for releasing eicosanoids which influence hemocytes. However, there is no direct evidence on the point. Both fat body and hemocytes (the major insect immune tissues) from tobacco hornworms are competent to biosynthesize eicosanoids (Stanley-Samuelson and Ogg, 1994; Gadelhak et al., 1995). Also, Jurenka et al. (1999) reported that bacterial infections simulated increased hemolymph prostaglandin (and other eicosanoid) titers in true armyworms, but again, there is no evidence on the origin of the eicosanoids. We hypothesized that hemocytes are the major source of eicosanoids which influence hemocyte defense reactions to bacterial infection. In this paper we report the outcomes of in vitro experiments which indicate that eicosanoids mediate microaggregation reactions to bacterial infections in isolated hemocyte preparations.
2. Materials and methods 2.1. Organisms Eggs of the tobacco hornworm, M. sexta, were purchased from Carolina Biological Supply (Burlington, NC). Larvae were reared on standard culture medium in individual cups under semi-sterile conditions developed by Dunn and Drake (1983). Cultures of a non-pigmented strain of S. marcescens and nutrient broth (Difco) were purchased from Carolina Biological Supply (Burlington, NC). Bacteria were grown in 50 ml of nutrient broth in an environmental shaker at 37°C and 100 rpm. Bacteria were used in mid-logarithmic or stationary phase at a dose of 2.5–7.0×107 colony forming units (cfu)/ml (Stanley-Samuelson et al., 1991). 2.2. Reagents Grace’s insect medium (Sigma Chemical Co., St. Louis, MO) was used in the preparation of hemolymph suspensions. Eicosanoid biosynthesis inhibitors, including the phospholipase A2 inhibitor, dexamethasone [(11β,16α)9-fluoro-11,17,21-trihydroxy-16-methylpregna-1,4-diene3,20-dione], indomethacin [1-(p-chlorobenzoyl)-5methoxy-2-methyl-3-indolyl-acetic acid], ibuprofen [2(4-isobutylphenyl)propionic acid], phenidone [1-phenyl3-pyrazolidinone], esculetin [6,7-dihydroxycoumarin] and the fatty acids, arachidonic acid [5,8,11,14-eicosatetraenoic acid] and palmitic acid [hexadecanoic acid], were purchased from Sigma Chemical Company (St. Louis, MO). 2.3. Hemolymph collection and preparation Second and third day fifth-instar hornworms were anesthetized by chilling on ice for 15 min, then surface sterilized by swabbing their exteriors with 95% ethanol. Hemolymph was collected by the pericardial puncture procedure described by Horohov and Dunn (1982). Briefly, a 20-gauge sterile, siliconized needle was inserted anteriorly at the thoracic abdominal junction such that the needle penetrates into the pericardial sinus. Freely dripping hemolymph was collected into chilled, sterile polypropylene 1.5 ml centrifuge tubes containing 500 µl of cold Grace’s insect medium. Approximately 500 µl of hemolymph was collected in this manner, gently mixed by inverting the test tube several times, and kept on ice for immediate use in each experiment. 2.4. In vitro assay for hemocyte microaggregate formation A sterile 96-well (250 µl/well), flat bottom, polystyrene, microtiter plate was used for each experiment
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
(Becton Dickenson, Lincoln, NJ). Each well was preloaded with 80 µl cold Grace’s insect medium amended with 2 µl ethanol or a selected pharmaceutical and/or fatty acid, dissolved in 2 µl ethanol. Then 20 µl of hemolymph suspension (approximately 6.4×105 cells/well) was added to each well. The hemocyte preparations were then challenged by adding 5.0 µl of bacterial suspension (approximately 104 cfu). The total volume of each well was adjusted to 150 µl with Grace’s insect medium. The plate was then incubated for selected time periods, as specified in Section 3, at 30°C in an environmental shaker at 100 rpm. After the selected incubation period, 20 µl of the hemocyte preparation was examined in a Bright-Line hemacytometer (AO Instruments Co., Buffalo, NY). After a 3-min settling period, a cover slip was applied and the number of hemocyte microaggregates (defined as a cluster of nine or more cells) in each sample was determined by direct counting using a light microscope equipped with phase contrast optics. Microaggregates were counted in grids which contained 1 µl of hemolymph. Number of microaggregates were normalized to microaggregates/ml hemolymph by multiplying the recorded count times 1000 times the dilution factor. Test preparations were treated with either the phospholipase A2 inhibitor, dexamethasone (final concentration 0.088 mM), one of the cyclooxygenase inhibitors (indomethacin [final concentration 0.098 mM] or ibuprofen [final concentration 0.172 mM]), the dual cyclooxygenase and lipoxygenase inhibitor phenidone (final concentration 0.218 mM), or the 5- and 12-lipoxygenase inhibitor, esculetin (final concentration 0.198 mM), dissolved in ethanol. In some experiments, test preparations were also treated with arachidonic acid or palmitic acid dissolved in ethanol. All pharmaceutical products were administered at dosages of 0.53 µg in 2 µl of ethanol/well. Fatty acids were administered at dosages of 0.2 µg in 2 µl ethanol/well. 2.5. Assessing cytotoxicity of treatments
Potential cytotoxicity of experimental treatments was assessed by Trypan Blue dye exclusion. Hemocyte preparations were set up in microtiter plates as just described, then exposed to experimental treatments without bacterial challenge at 30°C. For positive controls, hemocytes were incubated for 60 min without treatment. For negative controls, hemocytes were incubated in the presence of 75% ethanol for 60 min. Experimental treatments included incubation in the presence of 2.5% ethanol, esculetin, dexamethasone, ibuprofen, phenidone or indomethacin. After 60 min incubation periods, potential cytotoxicity was assessed by counting stained (killed) and living, dye-excluding cells.
1411
2.6. Background control experiments Control preparations were treated with 5 µl of nutrient broth, or 2 µl of 95% ethanol (drug vehicle). The total volume for each well was adjusted to 150 µl and the plate was allowed to incubate for 60 min. The number of microaggregates was then assessed as described above. To assess the effects of Grace’s insect medium on hemocyte microaggregation, wells were set up with 20 µl of hemolymph suspension. The plate was incubated for 60 min and the number of microaggregates assessed. To determine the effects of the orbital shaker on microaggregation, two groups of hemolymph suspensions were created. One group was incubated for 60 min in an orbital shaker at 100 rpm and the other incubated for 60 min without shaking. The number of microaggregates was then assessed. 2.7. Time course of hemocyte microaggregation: influence of dexamethasone Hemocyte preparations were divided into two groups. One group was treated with 2 µl of ethanol as a control. The other group was treated with 0.53 µg of dexamethasone in 2 µl of ethanol, and then 5 µl of the S. marcescens bacterial suspension was introduced to both groups. Plates were incubated for 15, 30, 60, and 120 min post inoculation (PI). At each time point, the number of microaggregates was assessed. 2.8. Fatty acid rescue experiments Individual preparations in two groups were treated with either 2 µl of ethanol as a control group, or 0.53 µg of dexamethasone dissolved in 2 µl of ethanol, then inoculated with 5 µl of bacterial suspension. Immediately after inoculation, the dexamethasone-treated preparations were divided into two sub-groups. Preparations in the first sub-group were treated with 0.2 µg of arachidonic acid in 2 µl of ethanol. Preparations in the second sub-group were treated with 0.2 µg of palmitic acid in 2 µl of ethanol. The total volume for each preparation was adjusted and the plate was allowed to incubate for 60 min. The number of aggregates was then assessed. 2.9. Influence of other eicosanoid biosynthesis inhibitors on hemocyte microaggregation Hemocyte preparations were set up as described, then divided into six groups. One group was treated with 2 µl of ethanol as a control. The other groups were treated with 0.53 µg of the phospholipase A2 inhibitor dexamethasone, or one of the cyclooxygenase inhibitors indomethacin or ibuprofen, or the dual cyclooxygenase-lipoxygenase inhibitor phenidone, or the 5- and 12-lipoxygenase inhibitor esculetin dissolved in ethanol. Then 5 µl of the
1412
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
bacterial suspension was introduced into each preparation. The total volume for each well was adjusted and the plate was allowed to incubate for 60 min. The number of microaggregates was then assessed.
2.10. The influence of conditioned media on hemocyte aggregation
To prepare condition media, hemolymph was collected as previously described in sterile polypropylene 1.5 ml centrifuge test tubes containing 500 µl cold Grace’s insect medium. One ml of hemolymph suspension was then challenged by adding 100 µl of standard bacterial suspension. The preparation was gently mixed by inverting the test tube several times and then incubated for 1 h at 30°C in an environmental shaker at 100 rpm. After the incubation period, the preparation was filtered through a 0.2 µm filter (Millipore). The resulting filtrate, referred to as conditioned media, was then placed on ice for immediate use. To determine the influence of conditioned media on microaggregation, 10 µl of dexamethasone solution (10 mM in ethanol) was added to sterile 1.5 ml microcentrifuge tubes containing either 500 µl of unconditioned cold Grace’s insect medium or 500 µl of cold conditioned media. To each microcentrifuge test tube, 500 µl of hemolymph was added and gently mixed by inverting the test tube several times. The hemolymph suspension prepared with unconditioned Grace’s insect medium was then challenged with a standard bacterial suspension. The hemolymph suspension prepared with conditioned media was not challenged with bacteria. The mixtures were then incubated for 1 h at 30°C in an environmental shaker at 100 rpm. After the incubation period, 20 µl of each preparation was applied to a Bright-Line hemacytometer (AO Instruments Co., Buffalo, NY). After a 3-min settling period, a cover slip was applied and the number of microaggregates for each sample was determined by direct counting using a light microscope equipped with phase contrast optics. Again, numbers of microaggregates were normalized to microaggregates/ml hemolymph.
2.11. Statistical analyses
Significant treatment effects were identified by oneway analysis of variance in the General Linear Models procedure at P⬍0.05. Where appropriate, significant differences among treatment means were determined by protected least significant differences (LSD) test (SAS Institute, Inc.).
3. Results 3.1. Control experiments Because the formation of microaggregates might be caused by the preparation and agitation steps, we examined samples taken from wells that were incubated without shaking as well as those incubated at 100 rpm. In all cases, the mean number of microaggregates was less than 1.0 microaggregate/preparation (data not shown). These results indicate that the environmental shaker is not a contributing factor in the formation of microaggregates. The results of the control experiments are displayed in Table 1. We recorded about 5.1×103 microaggregates/ml hemolymph in wells containing only hemolymph suspension and none in wells set up with hemolymph suspension treated with nutrient broth or with ethanol. We registered 2.4×104 microaggregates/ml hemolymph in preparations challenged with a standard dose of bacterial suspension. To examine the effects of ethanol (the drug vehicle) on microaggregation, standard hemolymph preparations were incubated 60 min with 2 µl of 95% ethanol. The ethanol treatments did not stimulate microaggregation. Standard hemolymph preparations treated with 2 µl of 95% ethanol and bacterial suspension produced about 2.6×104 microaggregates/ml hemolymph, indicating that the drug vehicle did not significantly influence microaggregation in experimental preparations. The results of our cytotoxicity assessments are presented in Table 2. We recorded approximately 94% viability in positive control preparations, and 46% viability in negative control preparations treated with 75% ethanol. All other treatments produced a range of 82–89% hemocyte viability, all statistically different from both positive and negative controls. 3.2. Time course of hemocyte microaggregation: influence of dexamethasone The time course of microaggregation in two groups of hemocyte preparations is displayed in Fig. 1. Table 1 Outcomes of background control experiments. Hemocyte preparations were as described in Section 2. The preparations were then given the indicated treatments. The values represent the mean number±SEM of microaggregates assessed at 60 min Treatments
Number of hemocyte microaggregates±SEM (×103)
(n)
None EtOH Nutrient broth Bacteria EtOH+bacteria
5.1±1.95 0.0±0.0 0.0±0.0 23.7±1.17 25.7±2.24
(7) (7) (5) (8) (8)
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
Table 2 Outcomes of cytotoxicity assessments. Hemocyte preparations and treatments were as described in Section 2. The values represent proportions, as percent total hemocytes (±1 SEM), of viable hemocytes recorded after 60 min incubations Treatments None EtOH (2.5%, drug vehicle) EtOH (75%, negative control) Phenidone Esculetin Dexamethasone Ibuprofen Indomethacin
Proportions of viable hemocytes
(n)
93.9±0.74
(16)
89.2±0.79
(25)
45.7±1.54
(10)
86.8±0.16 84.3±1.57 84.2±2.27 82.5±0.66 81.9±0.81
(10) (10) (10) (10) (10)
1413
3.3. Fatty acid rescue experiments Dexamethasone-treated hemocyte preparations produced fewer microaggregates than ethanol-treated control wells. On the model that dexamethasone inhibits eicosanoid biosynthesis through its inhibitory effect on phospholipase A2 (responsible for releasing arachidonic acid from cellular phospholipids and the first step in eicosanoid biosynthesis [Stanley, 2000]), we reasoned that supplementing dexamethasone-treated preparations with eicosanoid precursor polyunsaturated fatty acids would reverse the influence of dexamethasone on microaggregate formation. Fig. 2 shows that arachidonic acid supplementation reversed the effects of dexamethasone on microaggregate formation. Ethanol-treated control preparations produced about 2.7×104 microaggregates/ml hemolymph, compared to about 1.6×104 microaggregates/ml hemolymph in dexamethasone-treated preparations. We recorded approximately 3.0×104 microaggregates/ml hemolymph in dexamethasone-treated preparations amended with arachidonic acid, on par with the control wells. The rescue effects were specific to eicosanoid precursor polyunsaturated fatty acids, because amending dexamethasonetreated preparations with palmitic acid, a saturated fatty acid which is not a substrate for eicosanoid biosynthesis, did not restore microaggregate formation to control levels (LSD, P⬍0.05).
Fig. 1. Time course of microaggregation in tobacco hornworm, M. sexta, hemocyte preparations in response to challenge with bacteria, S. marcescens. Test preparations were treated with dexamethasone, then challenged with bacteria. Control preparations were treated with the drug vehicle, ethanol, then similarly challenged. At the indicated times PI, microaggregation was assessed by counting samples under phase contrast optics. Each point indicates the mean number of microaggregates in each hemocyte preparation (n=4–8), and the error bars represent 1 SEM.
Dexamethasone-treated preparations yielded about 7.5×103 microaggregates/ml hemolymph at 15 min PI, which increased to about 1.1×104 at 30 min. Longer incubations did not yield further increases in microaggregates. Compared to the experimental treatments, the ethanol-treated control preparations produced more microaggregates at each time point, from about 1.5×104 microaggregates/ml hemolymph at 15 min to a high of about 3.8×104 microaggregates/ml hemolymph at 30 min PI.
Fig. 2. Arachidonic acid reversed the effect of dexamethasone on microaggregate formation. Tobacco hornworm hemocyte preparations were treated with ethanol (EtOH) or dexamethasone (DEX), then challenged with bacteria, S. marcescens. Immediately after bacterial challenge, dexamethasone-treated test preparations were treated with 0.2 µg of arachidonic acid (DEX+AA). To control the influence of the third injection, individuals in another set of controls were treated with palmitic acid (DEX+PA). At 1 h PI, microaggregation was assessed. The height of histogram bars represented the mean number of microaggregates/ml hemolymph (n=10), and the error bars represent 1 SEM. Histogram bars with the same fill pattern are not significantly different from each other (LSD, P⬍0.05).
1414
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
3.4. Influence of other eicosanoid biosynthesis inhibitors on hemocyte microaggregation To dissect the possible roles of cyclooxygenase and lipoxygenase pathways in microaggregate formation, hemocyte preparations were treated with either standard doses of the cyclooxygenase inhibitors indomethacin or ibuprofen, the dual cyclooxygenase-lipoxygenase inhibitor phenidone, or the 5- and 12-lipoxygenase inhibitor esculetin and then inoculated with a standard dose of bacterial suspension. All test preparations, except the esculetin-treated samples, yielded significantly reduced microaggregation compared to ethanol-treated control wells (Fig. 3). Esculetin treatments did not influence microaggregation. Of the remaining inhibitors, we determined there were no significant differences among the effects of individual inhibitors on microaggregation. 3.5. The influence of conditioned media on hemocyte microaggregation The results of experiments with conditioned media are displayed in Table 3. Untreated hemolymph yielded approximately 2.5×103 microaggregates/ml, while hemolymph samples challenged with bacteria produced 8.5×103 microaggregates/ml. We registered significantly reduced (LSD; P⬍0.05) numbers of microaggregates, approximately 4.8×103/ml, in samples which were first treated with dexamethasone, then challenged with bac-
Fig. 3. The influence of individual eicosanoid biosynthesis inhibitors on tobacco hornworm microaggregation in response to bacterial challenge. Test preparations were first treated with either dexamethasone (DEX), or indomethacin (INDO), phenidone (PHEN), ibuprofen (IBU) or esculetin (ESC). Control preparations were treated with ethanol (EtOH). Then all preparations were challenged with bacteria. At 1 h PI, microaggregation was assessed. The height of histogram bars represented the mean number of microaggregates/ml hemolymph (n=5, except for EtOH and DEX, where n=10), and the error bars represent 1 SEM. Histogram bars with the same fill pattern are not significantly different from each other (LSD, P⬍0.05).
Table 3 The influence of conditioned media on hemocyte microaggregation reactions. Tobacco hornworm hemocyte preparations were exposed to the indicated treatments. After 60 min incubations, microaggregation was assessed by counting samples under phase contrast optics. The values represent the mean number±SEM of microaggregates. Means with the same superscript letter are not statistically different (LSD, P⬍0.05) (*dex=dexamethasone) Treatment Untreated hemolymph Hemolymph+bacteria Hemolymph+dex* +bacteria Hemolymph+conditioned media Hemolymph+dex+ conditioned media
Number of hemocyte microaggregates±SEM (×103)
(n)
2.50A±0.60 8.46B±1.28
(6) (12)
4.75A±1.24
(2)
10.04B±0.76
(13)
8.88B±0.94
(4)
teria. Unchallenged hemolymph samples which were treated with conditioned media produced about 1.0×104 microaggregates/ml, similar to the results just described for hemolymph samples challenged with bacteria. Finally, adding conditioned media to hemocyte preparations which were treated with dexamethasone resulted in approximately 8.8×103 microaggregates/ml.
4. Discussion In this paper, we report on microaggregation reactions to bacterial challenge by hemocyte preparations isolated from tobacco hornworms, M. sexta. The outcomes of our experiments support the hypothesis that hemocytes biosynthesize and secrete eicosanoids which mediate microaggregation reactions to bacterial challenge. Four lines of evidence strongly support the hypothesis. One, the number of microaggregates in untreated controls increased over the time course of the experiments, and the microaggregation reaction was significantly attenuated in test preparations treated with the phospholipase A2 inhibitor, dexamethasone. Two, the influence of dexamethasone was reversed by amending experimental hemocyte preparations with the eicosanoid precursor, arachidonic acid. Three, inhibition of eicosanoid biosynthesis with a range of pharmaceutical cyclooxygenase probes reduced microaggregation reactions in test preparations compared to untreated controls. Finally, adding conditioned media to hemocyte preparations stimulated microaggregation reactions. These results are similar to our earlier results of analogous experiments performed on whole insects. For example, in our previous work with tobacco hornworms, we recorded about 8.37×106 microaggregates/ml hemolymph at 1 h post infection in control hornworms, which was reduced by about four-fold in dexamethasone-
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
treated experimental larvae. Similarly, in our timecourse experiments, ethanol-treated control hornworms produced significantly more nodules at each time point than the dexamethasone-treated experimental hornworms (Miller et al., 1994). Hence, the results of our experiments with isolated hemocytes add weight to the idea that eicosanoids mediate insect microaggregation and nodulation reactions to bacterial infections. The information reported in this paper helps generate a novel insight into the roles of eicosanoids in insect cellular immune reactions. The observation of microaggregation reactions to bacterial challenge in isolated hemocyte preparations indicates that the eicosanoids involved in signaling microaggregate formation are produced by the hemocytes, and not by another tissue, such as fat body. However, this observation does not reveal whether the eicosanoids act in an intracellular or intercellular mechanism of action. In an intracellular mechanism, eicosanoids formed in reaction to bacterial challenge would act within the cell, possibly mediating biosynthesis and secretion of another signal moiety which is secreted to interact with other hemocytes. In an extracellular mechanism, eicosanoids would act in a paracrine mode. We designed experiments with conditioned media to distinguish between these possibilities. In our first experiments, untreated hemocytes (not challenged with bacteria) were exposed to conditioned media, and these preparations yielded number of microaggregates which were statistically similar to preparations which had been challenged with bacteria. We interpreted this result to indicate that one or more biochemical factors in the conditioned media were responsible for mediating microaggregate formation. In our second experiments, hemocyte preparations were treated with dexamethasone, then exposed to conditioned media. Again, we recorded number of microaggregates which were similar to preparations which had been challenged with bacteria. The biochemical factors in the conditioned media can stimulate microaggregation in hemocyte preparations in which eicosanoid biosynthesize was inhibited. We infer that isolated hemocytes are able to biosynthesize and secrete eicosanoids, and the secreted eicosanoids mediate microaggregation reactions. Hemocytes can be quite reactive to events, such as exposure to wound sites. We conducted background experiments to control the possibility of adventitious formation of hemocyte microaggregates due to our manipulations (Table 1). In the first place, hemolymph samples were withdrawn using the pericardial puncture first described by Horohov and Dunn (1982). Hemocyte defense reactions are generally not activated by this procedure, and our control experiments indicated that the inoculation and incubation treatments did not stimulate microaggregation. We recorded more than 2.3×104 microaggregates/ml hemolymph in hemocyte preparations challenged with bacteria or bacteria plus etha-
1415
nol, and no microaggregates in preparations treated with ethanol (the pharmaceutical vehicle) or with nutrient broth (the bacterial challenge vehicle). Moreover, our data show that the incubation, ethanol and nutrient broth treatments did not impair the ability of hemocytes to react to the presence of bacteria. In the same vein, the hemocytes were not damaged in obvious ways by the pharmaceutical assessments, as shown by our cytotoxicity assessments. Our assessments indicate the percentages of viable hornworm hemocytes are similar to percentages observed in studies with primary cultures of murine macrophages (R.L. Pardy, personal communication). Hence, the microaggregation reactions we recorded were physiological reactions to experimental bacterial challenge and not due to adventitious cellular actions. Our time course experiments illustrate the rapid onset of cellular reactions, specifically microaggregation, to bacterial challenge. We recorded a steep increase in microaggregation during the first 30 min PI, which did not further increase or possibly decreased over the following 90 min of our experiments. This would be consistent with the kinetics of nodule formation, thought to proceed through formation of microaggregates to development of larger nodules. Because nodules are formed by adhesion of numerous microaggregates, it is not surprising to record little change or decreases in microaggregates in later stages of the nodulation process. Our data show that microaggregation was inhibited in hemocyte preparations treated with inhibitors of cyclooxygenase, but not in preparations treated with esculetin, a specific inhibitor of 5- and 12-lipoxygenase. This indicates that in in vitro experiments, prostaglandins, but not lipoxygenase products, are key mediators of microaggregation. This finding runs contrary to our whole-insect experiments, in which esculetin treatments inhibited in vivo nodulation reactions to bacterial infection in several insect species, including tobacco hornworms (Miller et al., 1994), larvae of the beetle, Zophobas atratus (Miller et al., 1996), adult crickets, Gryllis assimilis (Miller et al., 1999) and adult 17-year periodical cicadas, Magicicada semptemdem (Tunaz et al., 1999). More recent experiments with lipopolysaccharide (LPS) purified from the bacterium S. maracescens also indicate that eicosanoids mediate insect microaggregation reactions to purified LPS challenge (Bedick et al., 2000). In this latter work we reported that esculetin treatments also attenuated cellular reactions to LPS. Nodulation is thought to involve an unknown number of specific cellular actions, some of which may depend on lipoxygenase products while others are mediated by cyclooxygenase products. If any of the cellular actions in the overall process of nodulation were somehow prevented, the inhibition would be recorded as reduced nodulation. It now remains unclear why esculetin did not influence the microaggregation
1416
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
process in isolated hemocyte preparations. We interpret the results in terms of down-stream events, with the suggestion that esculetin influences one or more cellular actions in the post-microaggregation phase of the overall nodulation process. Our results with esculetin open an alternative, intriguing possibility that other tissues, particularly hemopoietic organs, may be involved in one or more aspects of the nodulation process. Circulating hemocyte populations often decline during an infection cycle (Lackie, 1988), which we also recorded in our nodulation experiments with tobacco hornworms (Miller et al., 1994). On the idea that release of sessile hemocytes from hemopoietic organs may contribute to the microaggregation and nodulation processes, our data showing that esculetin did not influence microaggregation reactions in isolated hemocytes may be understood if one or more lipoxygenase products act in hemopoietic organs. If this is so, then we would expect esculetin or other lipoxygenase inhibitors to influence nodulation in intact insects, but not necessarily in isolated hemocytes. Comparison to our earlier data with tobacco hornworms supports this view (Miller et al., 1994). We recorded approximately 8.37×106 microaggregates/ml hemolymph from whole insect preparations, compared to 103–104 microaggregates/ml hemolymph registered in the in vitro work reported here. The lower numbers of microaggregates probably reflect the lower numbers of hemocytes in each preparation, and the absence of sessile hemocytes (possibly in the hemopoietic organs) which could be recruited into the microaggregation process. These and other speculative ideas will motivate future experimental work on nodulation. There are difficulties associated with inferring physiological signaling pathways from results of experiments with pharmaceutical probes (Stanley-Samuelson, 1994b). We have discussed these in detail with respect to the role of eicosanoids in insect cellular immunity, along with discussion of the biological significance of other signaling moieties in insect immunity elsewhere (Miller et al., 1996; Stanley, 2000), and will not reiterate the points here. Such objections notwithstanding, the results of experiments with pharmaceutical probes, when interpreted critically, are valid. The significance of our work with isolated hemocyte preparations lies in gaining a more detailed understanding of the signaling pathways in cellular immune reactions to bacterial infections. In work at the organismal level, we recorded sharply reduced microaggregation and nodulation reactions to bacterial challenge following experimental treatments with pharmaceutical inhibitors of eicosanoid biosynthesis (Miller et al. 1994, 1996; Tunaz et al., 1999; Bedick et al., 2000). These findings, however, did not provide guidance with respect to the source of the eicosanoids involved in mediating cellular defense reactions. The results of our experiments with
isolated hemocyte preparations show that hemocytes biosynthesize and secrete eicosanoids which in turn influence the cellular actions of other hemocytes. Of course, the data presented in this paper do not exclude the possibility that eicosanoids from other tissue sources, particularly fat body, can influence hemocyte actions.
Acknowledgements We thank Dr Ralph Howard and Dr Virginia Naples for a careful reading and commentary on a draft of this manuscript. This is paper number 13,360 of the Nebraska Agricultural Research Division. This work was supported by the Agricultural Research Division, UNL (Project NEB-17-054) and by the Department of Biological Sciences, Northern Illinois University.
References Bedick, J.C., Pardy, R.L., Howard, R.W., Stanley, D.W., 2000. Insect cellular reactions to the lipopolysaccharide component of the bacterium Serratia marcescens are mediated by eicosanoids. Journal of Insect Physiology 46, 1481–1487. Dunn, P.E., Drake, D.R., 1983. Fate of bacteria injected into naive and immunized larvae of the tobacco hornworm, Manduca sexta. Journal of Invertebrate Pathology 41, 77–85. Gadelhak, G.G., Pedibhotla, V.K., Stanley-Samuelson, D.W., 1995. Eicosanoid biosynthesis by hemocytes from the tobacco hornworm, Manduca sexta. Insect Biochemistry and Molecular Biology 25, 743–749. Gupta, A.P., 1986. Arthropod immunocytes: identification, structure, functions, and analogies to the functions of vertebrate B- and Tlymphocytes. In: Gupta, A. (Ed.), Hemolytic and Humoral Immunity in Arthropods. Wiley, New York. Gupta, A.P., 1991. Immunology of Insects and Other Arthropods. In: Gupta, A.P. (Ed.). CRC Press, Boca Raton, FL. Horohov, D.W., Dunn, P.E., 1982. Changes in the circulating hemocyte population of Manduca sexta larvae following injection of bacteria. Journal of Invertebrate Pathology 40, 327. Howard, R.W., Stanley, D.W., 1999. The tie that binds: eicosanoids in invertebrate biology. Annals of the Entomological Society of America 92, 880–890. Jurenka, R.A., Miller, J.S., Pedibhotla, V.K., Rana, R.L., StanleySamuelson, D.W., 1997. Eicosanoids mediate microaggregation and nodulation responses to bacterial infections in black cutworms, Agrotis ipsilon, and true armyworms, Pseudaletia unipuncta. Journal of Insect Physiology 43, 125–133. Jurenka, R.A., Pedibhotla, V.K., Stanley, D.W., 1999. Prostaglandin production in response to a bacterial infection in true armyworm larvae. Archives of Insect Biochemistry and Physiology 41, 225– 232. Lackie, A.M., 1988. Haemocyte behavior. Advances in Insect Physiology 21, 85–178. Mandato, C.A., Diehl-Jones, W.L., Moore, S.J., Downer, R.G.H., 1997. The effects of eicosanoid biosynthesis inhibitors on prophenoloxidase activation, phagocytosis and cell spreading in Galleria mellonella. Journal of Insect Physiology 43, 1–8. Miller, J.S., Nguyen, T., Stanley-Samuelson, D.W., 1994. Eicosanoids mediate insect nodulation responses to bacterial infections. Proceedings of the National Academy of Sciences of the United States of America 91, 12418–12422.
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
Miller, J.S., Howard, R.W., Nguyen, T., Nguyen, A., Rosario, R.M.T., Stanley-Samuelson, D.W., 1996. Eicosanoids mediate nodulation responses to bacterial infections in larvae of the tenebrionid beetle, Zophobas atratus. Journal of Insect Physiology 42, 3–12. Miller, J.S., Howard, R.W., Rana, R.L., Tunaz, H., Stanley, D.W., 1999. Eicosanoids mediate nodulation reactions to bacterial infections in adults of the cricket, Gryllus assimilis. Journal of Insect Physiology 45, 75–83. Morishima, I., Yamano, Y., Knoue, K., Matsuo, N., 1997. Eicosanoids mediate induction of immune genes in the fat body of the silkworm, Bombyx mori. FEBS Letters 419, 83–86. Stanley, D.W., 2000. Eicosanoids in Invertebrate Signal Transduction Systems. Princeton University Press, Princeton, NJ. Stanley, D.W., Howard, R.W., 1998. The biology of prostaglandins and related eicosanoids in invertebrates: cellular, organismal and ecological actions. American Zoologist 38, 369–381. Stanley-Samuelson, D.W., 1994a. Prostaglandins and related eicosanoids in insects. Advances in Insect Physiology 24, 115–212.
1417
Stanley-Samuelson, D.W., 1994b. Assessing the physiological significance of prostaglandins and other eicosanoids in insect physiology. Journal of Insect Physiology 40, 3–11. Stanley-Samuelson, D.W., Ogg, C.L., 1994. Prostaglandin biosynthesis by fat body from the tobacco hornworm, Manduca sexta. Insect Biochemistry and Molecular Biology 24, 481–491. Stanley-Samuelson, D.W., Jensen, E., Nickerson, K.W., Tiebel, K., Ogg, C.L., Howard, R.W., 1991. Insect immune response to bacterial infection is mediated by eicosanoids. Proceedings of the National Academy of Sciences of the United States of America 88, 1064–1068. Tunaz, H., Bedick, J.C., Miller, J.S., Hoback, W.W., Rana, R.L., Stanley, D.W., 1999. Eicosanoids mediate nodulation reactions to bacterial infections in adults of two 17-year periodical cicadas, Magicicada septendecim and M. cassini. Journal of Insect Physiology 45, 923–931.
Eicosanoids mediate microaggregation reactions to bacterial challenge in isolated insect hemocyte preparations J.S. Miller a, D.W. Stanley b
b,*
a Department of Biological Sciences, Northern Illinois University, De Kalb, IL 60115, USA Insect Biochemical Physiology Laboratory, 311 Plant Industry Building, University of Nebraska, Lincoln, NE 68583-0816, USA
Received 3 April 2001; accepted 31 July 2001
Abstract Nodule formation is the quantitatively predominant insect cellular defense reaction to bacterial challenges, responsible for clearing the largest proportion of infecting bacteria from circulation. It has been suggested that eicosanoids mediate several steps in the nodulation process, including formation of hemocyte microaggregates, an early step in the process. While fat body and hemocytes are competent to biosynthesize eicosanoids, the source of the nodulation-mediating eicosanoids remains unclear. To investigate this issue, we studied hemocyte microaggregation reactions to bacterial challenge in vitro. Hemocyte suspensions from the tobacco hornworm, Manduca sexta, were treated with the phospholipase A2 inhibitor, dexamethasone, then challenged with the bacterium Serratia marcescens. Preparations treated with dexamethasone yielded fewer hemocyte microaggregations than untreated, control preparations. Furthermore, the influence of dexamethasone was reversed by amending experimental (dexamethasone-treated) preparations with the eicosanoid biosynthesis precursor, arachidonic acid. Palmitic acid, which is not a substrate for eicosanoid biosynthesis, did not reverse the influence of dexamethasone on the microaggregation reaction. The influence of dexamethasone was also reversed by adding filtered media from challenged hemocyte preparations to dexamethasone-treated preparations. Finally, most hemocyte preparations treated with selected eicosanoid biosynthesis inhibitors formed fewer hemocyte microaggregations than control preparations. The 5- and 12-lipoxygenase inhibitor, esculetin, did not influence the formation of hemocyte microaggregations in this system. These results are consistent with similar investigations performed in vivo, and we infer that hemocytes are responsible for forming and secreting eicosanoids, which subsequently initiate nodulation by mediating hemocyte microaggregation. 2001 Elsevier Science Ltd. All rights reserved. Keywords: In vitro; Eicosanoids; Insect immunity; Hemocyte microaggregation; Manduca sexta; Serratia marcescens
1. Introduction The innate immune systems in insects and other invertebrates are categorized into hemocytic and humoral immune responses. Hemocytic defense reactions to bacterial infections include phagocytosis and nodule formation (Gupta 1986, 1991). These responses occur within minutes of infection and feature direct cellular interactions between circulating hemocytes and the bacterial cells (Gupta 1986, 1991). While the nodulation reaction is well documented, much less is known about the signaling mechanisms responsible for mediating
* Corresponding author. Tel.: +1-402-472-8710; fax: +1-402-4724687. E-mail address: [email protected] (D.W. Stanley).
nodulation. Because many mammalian cellular defense reactions are mediated by eicosanoids, Stanley-Samuelson et al. (1991) suggested that possibly insect cellular immune reactions also are mediated by eicosanoids. Eicosanoids are oxygenated metabolites of certain C20 polyunsaturated fatty acids. The biological significance, structures and biosynthetic pathways of these compounds are outlined in several reviews (Stanley-Samuelson, 1994a; Stanley and Howard, 1998; Howard and Stanley, 1999; Stanley, 2000). Nodulation is the predominant insect cellular immune reaction, responsible for clearing as much as 90% of infecting bacteria from hemolymph circulation within minutes of infection (Horohov and Dunn, 1982), and we hypothesized that nodulation is a specific cellular immune reaction mediated by eicosanoids. We previously tested this hypothesis in a series of simple
0022-1910/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 1 9 1 0 ( 0 1 ) 0 0 1 3 1 - 7
1410
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
experiments (Miller et al., 1994). Groups of tobacco hornworms, Manduca sexta, were treated separately with a range of eicosanoid biosynthesis inhibitors, including the phospholipase A2 inhibitor dexamethasone, then artificially infected with the bacterium, Serratia marcescens. Cellular immune reactions to the artificial infections were assessed by counting the number of hemocytic microaggregates (an early phase of nodulation) under phase contrast optics and by counting the number of melanized, mature nodules within the hemocoels using a dissecting microscope. The ability of experimental hornworms to form microaggregates and nodules was severely impaired by the inhibitors. These findings supported the idea that eicosanoids mediate nodule formation in tobacco hornworms (Miller et al., 1994). This idea has been tested in several hemimetabolous and holometabolous insect species, the results of which support the broader hypothesis that eicosanoids mediate insect nodulation reactions to bacterial infections (Miller et al. 1996, 1999; Jurenka et al., 1997; Tunaz et al., 1999; Stanley, 2000). Mandato et al. (1997) identified other eicosanoid-dependent cellular immune reactions, suggesting that eicosanoids mediate certain discrete steps in nodule formation including cell spreading and prophenoloxidase activation, as well as phagocytosis in the wax moth, Galleria mellonella. Morishima et al. (1997) also suggested that eicosanoids mediate expression of two silkworm, Bombyx mori, fat body genes for anti-bacterial proteins, cecropin and lysozyme. The emergent view is that eicosanoids act in several discrete aspects of insect immunity. The idea that eicosanoids mediate various cellular defense reactions to bacterial challenge opens questions on cell–cell signaling in immunity. On a paracrine model, the eicosanoids involved in mediating insect cellular immune reactions could originate in the hemocytes, then influence the actions of other hemocytes. Alternatively, other tissues, particularly fat body, could be responsible for releasing eicosanoids which influence hemocytes. However, there is no direct evidence on the point. Both fat body and hemocytes (the major insect immune tissues) from tobacco hornworms are competent to biosynthesize eicosanoids (Stanley-Samuelson and Ogg, 1994; Gadelhak et al., 1995). Also, Jurenka et al. (1999) reported that bacterial infections simulated increased hemolymph prostaglandin (and other eicosanoid) titers in true armyworms, but again, there is no evidence on the origin of the eicosanoids. We hypothesized that hemocytes are the major source of eicosanoids which influence hemocyte defense reactions to bacterial infection. In this paper we report the outcomes of in vitro experiments which indicate that eicosanoids mediate microaggregation reactions to bacterial infections in isolated hemocyte preparations.
2. Materials and methods 2.1. Organisms Eggs of the tobacco hornworm, M. sexta, were purchased from Carolina Biological Supply (Burlington, NC). Larvae were reared on standard culture medium in individual cups under semi-sterile conditions developed by Dunn and Drake (1983). Cultures of a non-pigmented strain of S. marcescens and nutrient broth (Difco) were purchased from Carolina Biological Supply (Burlington, NC). Bacteria were grown in 50 ml of nutrient broth in an environmental shaker at 37°C and 100 rpm. Bacteria were used in mid-logarithmic or stationary phase at a dose of 2.5–7.0×107 colony forming units (cfu)/ml (Stanley-Samuelson et al., 1991). 2.2. Reagents Grace’s insect medium (Sigma Chemical Co., St. Louis, MO) was used in the preparation of hemolymph suspensions. Eicosanoid biosynthesis inhibitors, including the phospholipase A2 inhibitor, dexamethasone [(11β,16α)9-fluoro-11,17,21-trihydroxy-16-methylpregna-1,4-diene3,20-dione], indomethacin [1-(p-chlorobenzoyl)-5methoxy-2-methyl-3-indolyl-acetic acid], ibuprofen [2(4-isobutylphenyl)propionic acid], phenidone [1-phenyl3-pyrazolidinone], esculetin [6,7-dihydroxycoumarin] and the fatty acids, arachidonic acid [5,8,11,14-eicosatetraenoic acid] and palmitic acid [hexadecanoic acid], were purchased from Sigma Chemical Company (St. Louis, MO). 2.3. Hemolymph collection and preparation Second and third day fifth-instar hornworms were anesthetized by chilling on ice for 15 min, then surface sterilized by swabbing their exteriors with 95% ethanol. Hemolymph was collected by the pericardial puncture procedure described by Horohov and Dunn (1982). Briefly, a 20-gauge sterile, siliconized needle was inserted anteriorly at the thoracic abdominal junction such that the needle penetrates into the pericardial sinus. Freely dripping hemolymph was collected into chilled, sterile polypropylene 1.5 ml centrifuge tubes containing 500 µl of cold Grace’s insect medium. Approximately 500 µl of hemolymph was collected in this manner, gently mixed by inverting the test tube several times, and kept on ice for immediate use in each experiment. 2.4. In vitro assay for hemocyte microaggregate formation A sterile 96-well (250 µl/well), flat bottom, polystyrene, microtiter plate was used for each experiment
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
(Becton Dickenson, Lincoln, NJ). Each well was preloaded with 80 µl cold Grace’s insect medium amended with 2 µl ethanol or a selected pharmaceutical and/or fatty acid, dissolved in 2 µl ethanol. Then 20 µl of hemolymph suspension (approximately 6.4×105 cells/well) was added to each well. The hemocyte preparations were then challenged by adding 5.0 µl of bacterial suspension (approximately 104 cfu). The total volume of each well was adjusted to 150 µl with Grace’s insect medium. The plate was then incubated for selected time periods, as specified in Section 3, at 30°C in an environmental shaker at 100 rpm. After the selected incubation period, 20 µl of the hemocyte preparation was examined in a Bright-Line hemacytometer (AO Instruments Co., Buffalo, NY). After a 3-min settling period, a cover slip was applied and the number of hemocyte microaggregates (defined as a cluster of nine or more cells) in each sample was determined by direct counting using a light microscope equipped with phase contrast optics. Microaggregates were counted in grids which contained 1 µl of hemolymph. Number of microaggregates were normalized to microaggregates/ml hemolymph by multiplying the recorded count times 1000 times the dilution factor. Test preparations were treated with either the phospholipase A2 inhibitor, dexamethasone (final concentration 0.088 mM), one of the cyclooxygenase inhibitors (indomethacin [final concentration 0.098 mM] or ibuprofen [final concentration 0.172 mM]), the dual cyclooxygenase and lipoxygenase inhibitor phenidone (final concentration 0.218 mM), or the 5- and 12-lipoxygenase inhibitor, esculetin (final concentration 0.198 mM), dissolved in ethanol. In some experiments, test preparations were also treated with arachidonic acid or palmitic acid dissolved in ethanol. All pharmaceutical products were administered at dosages of 0.53 µg in 2 µl of ethanol/well. Fatty acids were administered at dosages of 0.2 µg in 2 µl ethanol/well. 2.5. Assessing cytotoxicity of treatments
Potential cytotoxicity of experimental treatments was assessed by Trypan Blue dye exclusion. Hemocyte preparations were set up in microtiter plates as just described, then exposed to experimental treatments without bacterial challenge at 30°C. For positive controls, hemocytes were incubated for 60 min without treatment. For negative controls, hemocytes were incubated in the presence of 75% ethanol for 60 min. Experimental treatments included incubation in the presence of 2.5% ethanol, esculetin, dexamethasone, ibuprofen, phenidone or indomethacin. After 60 min incubation periods, potential cytotoxicity was assessed by counting stained (killed) and living, dye-excluding cells.
1411
2.6. Background control experiments Control preparations were treated with 5 µl of nutrient broth, or 2 µl of 95% ethanol (drug vehicle). The total volume for each well was adjusted to 150 µl and the plate was allowed to incubate for 60 min. The number of microaggregates was then assessed as described above. To assess the effects of Grace’s insect medium on hemocyte microaggregation, wells were set up with 20 µl of hemolymph suspension. The plate was incubated for 60 min and the number of microaggregates assessed. To determine the effects of the orbital shaker on microaggregation, two groups of hemolymph suspensions were created. One group was incubated for 60 min in an orbital shaker at 100 rpm and the other incubated for 60 min without shaking. The number of microaggregates was then assessed. 2.7. Time course of hemocyte microaggregation: influence of dexamethasone Hemocyte preparations were divided into two groups. One group was treated with 2 µl of ethanol as a control. The other group was treated with 0.53 µg of dexamethasone in 2 µl of ethanol, and then 5 µl of the S. marcescens bacterial suspension was introduced to both groups. Plates were incubated for 15, 30, 60, and 120 min post inoculation (PI). At each time point, the number of microaggregates was assessed. 2.8. Fatty acid rescue experiments Individual preparations in two groups were treated with either 2 µl of ethanol as a control group, or 0.53 µg of dexamethasone dissolved in 2 µl of ethanol, then inoculated with 5 µl of bacterial suspension. Immediately after inoculation, the dexamethasone-treated preparations were divided into two sub-groups. Preparations in the first sub-group were treated with 0.2 µg of arachidonic acid in 2 µl of ethanol. Preparations in the second sub-group were treated with 0.2 µg of palmitic acid in 2 µl of ethanol. The total volume for each preparation was adjusted and the plate was allowed to incubate for 60 min. The number of aggregates was then assessed. 2.9. Influence of other eicosanoid biosynthesis inhibitors on hemocyte microaggregation Hemocyte preparations were set up as described, then divided into six groups. One group was treated with 2 µl of ethanol as a control. The other groups were treated with 0.53 µg of the phospholipase A2 inhibitor dexamethasone, or one of the cyclooxygenase inhibitors indomethacin or ibuprofen, or the dual cyclooxygenase-lipoxygenase inhibitor phenidone, or the 5- and 12-lipoxygenase inhibitor esculetin dissolved in ethanol. Then 5 µl of the
1412
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
bacterial suspension was introduced into each preparation. The total volume for each well was adjusted and the plate was allowed to incubate for 60 min. The number of microaggregates was then assessed.
2.10. The influence of conditioned media on hemocyte aggregation
To prepare condition media, hemolymph was collected as previously described in sterile polypropylene 1.5 ml centrifuge test tubes containing 500 µl cold Grace’s insect medium. One ml of hemolymph suspension was then challenged by adding 100 µl of standard bacterial suspension. The preparation was gently mixed by inverting the test tube several times and then incubated for 1 h at 30°C in an environmental shaker at 100 rpm. After the incubation period, the preparation was filtered through a 0.2 µm filter (Millipore). The resulting filtrate, referred to as conditioned media, was then placed on ice for immediate use. To determine the influence of conditioned media on microaggregation, 10 µl of dexamethasone solution (10 mM in ethanol) was added to sterile 1.5 ml microcentrifuge tubes containing either 500 µl of unconditioned cold Grace’s insect medium or 500 µl of cold conditioned media. To each microcentrifuge test tube, 500 µl of hemolymph was added and gently mixed by inverting the test tube several times. The hemolymph suspension prepared with unconditioned Grace’s insect medium was then challenged with a standard bacterial suspension. The hemolymph suspension prepared with conditioned media was not challenged with bacteria. The mixtures were then incubated for 1 h at 30°C in an environmental shaker at 100 rpm. After the incubation period, 20 µl of each preparation was applied to a Bright-Line hemacytometer (AO Instruments Co., Buffalo, NY). After a 3-min settling period, a cover slip was applied and the number of microaggregates for each sample was determined by direct counting using a light microscope equipped with phase contrast optics. Again, numbers of microaggregates were normalized to microaggregates/ml hemolymph.
2.11. Statistical analyses
Significant treatment effects were identified by oneway analysis of variance in the General Linear Models procedure at P⬍0.05. Where appropriate, significant differences among treatment means were determined by protected least significant differences (LSD) test (SAS Institute, Inc.).
3. Results 3.1. Control experiments Because the formation of microaggregates might be caused by the preparation and agitation steps, we examined samples taken from wells that were incubated without shaking as well as those incubated at 100 rpm. In all cases, the mean number of microaggregates was less than 1.0 microaggregate/preparation (data not shown). These results indicate that the environmental shaker is not a contributing factor in the formation of microaggregates. The results of the control experiments are displayed in Table 1. We recorded about 5.1×103 microaggregates/ml hemolymph in wells containing only hemolymph suspension and none in wells set up with hemolymph suspension treated with nutrient broth or with ethanol. We registered 2.4×104 microaggregates/ml hemolymph in preparations challenged with a standard dose of bacterial suspension. To examine the effects of ethanol (the drug vehicle) on microaggregation, standard hemolymph preparations were incubated 60 min with 2 µl of 95% ethanol. The ethanol treatments did not stimulate microaggregation. Standard hemolymph preparations treated with 2 µl of 95% ethanol and bacterial suspension produced about 2.6×104 microaggregates/ml hemolymph, indicating that the drug vehicle did not significantly influence microaggregation in experimental preparations. The results of our cytotoxicity assessments are presented in Table 2. We recorded approximately 94% viability in positive control preparations, and 46% viability in negative control preparations treated with 75% ethanol. All other treatments produced a range of 82–89% hemocyte viability, all statistically different from both positive and negative controls. 3.2. Time course of hemocyte microaggregation: influence of dexamethasone The time course of microaggregation in two groups of hemocyte preparations is displayed in Fig. 1. Table 1 Outcomes of background control experiments. Hemocyte preparations were as described in Section 2. The preparations were then given the indicated treatments. The values represent the mean number±SEM of microaggregates assessed at 60 min Treatments
Number of hemocyte microaggregates±SEM (×103)
(n)
None EtOH Nutrient broth Bacteria EtOH+bacteria
5.1±1.95 0.0±0.0 0.0±0.0 23.7±1.17 25.7±2.24
(7) (7) (5) (8) (8)
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
Table 2 Outcomes of cytotoxicity assessments. Hemocyte preparations and treatments were as described in Section 2. The values represent proportions, as percent total hemocytes (±1 SEM), of viable hemocytes recorded after 60 min incubations Treatments None EtOH (2.5%, drug vehicle) EtOH (75%, negative control) Phenidone Esculetin Dexamethasone Ibuprofen Indomethacin
Proportions of viable hemocytes
(n)
93.9±0.74
(16)
89.2±0.79
(25)
45.7±1.54
(10)
86.8±0.16 84.3±1.57 84.2±2.27 82.5±0.66 81.9±0.81
(10) (10) (10) (10) (10)
1413
3.3. Fatty acid rescue experiments Dexamethasone-treated hemocyte preparations produced fewer microaggregates than ethanol-treated control wells. On the model that dexamethasone inhibits eicosanoid biosynthesis through its inhibitory effect on phospholipase A2 (responsible for releasing arachidonic acid from cellular phospholipids and the first step in eicosanoid biosynthesis [Stanley, 2000]), we reasoned that supplementing dexamethasone-treated preparations with eicosanoid precursor polyunsaturated fatty acids would reverse the influence of dexamethasone on microaggregate formation. Fig. 2 shows that arachidonic acid supplementation reversed the effects of dexamethasone on microaggregate formation. Ethanol-treated control preparations produced about 2.7×104 microaggregates/ml hemolymph, compared to about 1.6×104 microaggregates/ml hemolymph in dexamethasone-treated preparations. We recorded approximately 3.0×104 microaggregates/ml hemolymph in dexamethasone-treated preparations amended with arachidonic acid, on par with the control wells. The rescue effects were specific to eicosanoid precursor polyunsaturated fatty acids, because amending dexamethasonetreated preparations with palmitic acid, a saturated fatty acid which is not a substrate for eicosanoid biosynthesis, did not restore microaggregate formation to control levels (LSD, P⬍0.05).
Fig. 1. Time course of microaggregation in tobacco hornworm, M. sexta, hemocyte preparations in response to challenge with bacteria, S. marcescens. Test preparations were treated with dexamethasone, then challenged with bacteria. Control preparations were treated with the drug vehicle, ethanol, then similarly challenged. At the indicated times PI, microaggregation was assessed by counting samples under phase contrast optics. Each point indicates the mean number of microaggregates in each hemocyte preparation (n=4–8), and the error bars represent 1 SEM.
Dexamethasone-treated preparations yielded about 7.5×103 microaggregates/ml hemolymph at 15 min PI, which increased to about 1.1×104 at 30 min. Longer incubations did not yield further increases in microaggregates. Compared to the experimental treatments, the ethanol-treated control preparations produced more microaggregates at each time point, from about 1.5×104 microaggregates/ml hemolymph at 15 min to a high of about 3.8×104 microaggregates/ml hemolymph at 30 min PI.
Fig. 2. Arachidonic acid reversed the effect of dexamethasone on microaggregate formation. Tobacco hornworm hemocyte preparations were treated with ethanol (EtOH) or dexamethasone (DEX), then challenged with bacteria, S. marcescens. Immediately after bacterial challenge, dexamethasone-treated test preparations were treated with 0.2 µg of arachidonic acid (DEX+AA). To control the influence of the third injection, individuals in another set of controls were treated with palmitic acid (DEX+PA). At 1 h PI, microaggregation was assessed. The height of histogram bars represented the mean number of microaggregates/ml hemolymph (n=10), and the error bars represent 1 SEM. Histogram bars with the same fill pattern are not significantly different from each other (LSD, P⬍0.05).
1414
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
3.4. Influence of other eicosanoid biosynthesis inhibitors on hemocyte microaggregation To dissect the possible roles of cyclooxygenase and lipoxygenase pathways in microaggregate formation, hemocyte preparations were treated with either standard doses of the cyclooxygenase inhibitors indomethacin or ibuprofen, the dual cyclooxygenase-lipoxygenase inhibitor phenidone, or the 5- and 12-lipoxygenase inhibitor esculetin and then inoculated with a standard dose of bacterial suspension. All test preparations, except the esculetin-treated samples, yielded significantly reduced microaggregation compared to ethanol-treated control wells (Fig. 3). Esculetin treatments did not influence microaggregation. Of the remaining inhibitors, we determined there were no significant differences among the effects of individual inhibitors on microaggregation. 3.5. The influence of conditioned media on hemocyte microaggregation The results of experiments with conditioned media are displayed in Table 3. Untreated hemolymph yielded approximately 2.5×103 microaggregates/ml, while hemolymph samples challenged with bacteria produced 8.5×103 microaggregates/ml. We registered significantly reduced (LSD; P⬍0.05) numbers of microaggregates, approximately 4.8×103/ml, in samples which were first treated with dexamethasone, then challenged with bac-
Fig. 3. The influence of individual eicosanoid biosynthesis inhibitors on tobacco hornworm microaggregation in response to bacterial challenge. Test preparations were first treated with either dexamethasone (DEX), or indomethacin (INDO), phenidone (PHEN), ibuprofen (IBU) or esculetin (ESC). Control preparations were treated with ethanol (EtOH). Then all preparations were challenged with bacteria. At 1 h PI, microaggregation was assessed. The height of histogram bars represented the mean number of microaggregates/ml hemolymph (n=5, except for EtOH and DEX, where n=10), and the error bars represent 1 SEM. Histogram bars with the same fill pattern are not significantly different from each other (LSD, P⬍0.05).
Table 3 The influence of conditioned media on hemocyte microaggregation reactions. Tobacco hornworm hemocyte preparations were exposed to the indicated treatments. After 60 min incubations, microaggregation was assessed by counting samples under phase contrast optics. The values represent the mean number±SEM of microaggregates. Means with the same superscript letter are not statistically different (LSD, P⬍0.05) (*dex=dexamethasone) Treatment Untreated hemolymph Hemolymph+bacteria Hemolymph+dex* +bacteria Hemolymph+conditioned media Hemolymph+dex+ conditioned media
Number of hemocyte microaggregates±SEM (×103)
(n)
2.50A±0.60 8.46B±1.28
(6) (12)
4.75A±1.24
(2)
10.04B±0.76
(13)
8.88B±0.94
(4)
teria. Unchallenged hemolymph samples which were treated with conditioned media produced about 1.0×104 microaggregates/ml, similar to the results just described for hemolymph samples challenged with bacteria. Finally, adding conditioned media to hemocyte preparations which were treated with dexamethasone resulted in approximately 8.8×103 microaggregates/ml.
4. Discussion In this paper, we report on microaggregation reactions to bacterial challenge by hemocyte preparations isolated from tobacco hornworms, M. sexta. The outcomes of our experiments support the hypothesis that hemocytes biosynthesize and secrete eicosanoids which mediate microaggregation reactions to bacterial challenge. Four lines of evidence strongly support the hypothesis. One, the number of microaggregates in untreated controls increased over the time course of the experiments, and the microaggregation reaction was significantly attenuated in test preparations treated with the phospholipase A2 inhibitor, dexamethasone. Two, the influence of dexamethasone was reversed by amending experimental hemocyte preparations with the eicosanoid precursor, arachidonic acid. Three, inhibition of eicosanoid biosynthesis with a range of pharmaceutical cyclooxygenase probes reduced microaggregation reactions in test preparations compared to untreated controls. Finally, adding conditioned media to hemocyte preparations stimulated microaggregation reactions. These results are similar to our earlier results of analogous experiments performed on whole insects. For example, in our previous work with tobacco hornworms, we recorded about 8.37×106 microaggregates/ml hemolymph at 1 h post infection in control hornworms, which was reduced by about four-fold in dexamethasone-
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
treated experimental larvae. Similarly, in our timecourse experiments, ethanol-treated control hornworms produced significantly more nodules at each time point than the dexamethasone-treated experimental hornworms (Miller et al., 1994). Hence, the results of our experiments with isolated hemocytes add weight to the idea that eicosanoids mediate insect microaggregation and nodulation reactions to bacterial infections. The information reported in this paper helps generate a novel insight into the roles of eicosanoids in insect cellular immune reactions. The observation of microaggregation reactions to bacterial challenge in isolated hemocyte preparations indicates that the eicosanoids involved in signaling microaggregate formation are produced by the hemocytes, and not by another tissue, such as fat body. However, this observation does not reveal whether the eicosanoids act in an intracellular or intercellular mechanism of action. In an intracellular mechanism, eicosanoids formed in reaction to bacterial challenge would act within the cell, possibly mediating biosynthesis and secretion of another signal moiety which is secreted to interact with other hemocytes. In an extracellular mechanism, eicosanoids would act in a paracrine mode. We designed experiments with conditioned media to distinguish between these possibilities. In our first experiments, untreated hemocytes (not challenged with bacteria) were exposed to conditioned media, and these preparations yielded number of microaggregates which were statistically similar to preparations which had been challenged with bacteria. We interpreted this result to indicate that one or more biochemical factors in the conditioned media were responsible for mediating microaggregate formation. In our second experiments, hemocyte preparations were treated with dexamethasone, then exposed to conditioned media. Again, we recorded number of microaggregates which were similar to preparations which had been challenged with bacteria. The biochemical factors in the conditioned media can stimulate microaggregation in hemocyte preparations in which eicosanoid biosynthesize was inhibited. We infer that isolated hemocytes are able to biosynthesize and secrete eicosanoids, and the secreted eicosanoids mediate microaggregation reactions. Hemocytes can be quite reactive to events, such as exposure to wound sites. We conducted background experiments to control the possibility of adventitious formation of hemocyte microaggregates due to our manipulations (Table 1). In the first place, hemolymph samples were withdrawn using the pericardial puncture first described by Horohov and Dunn (1982). Hemocyte defense reactions are generally not activated by this procedure, and our control experiments indicated that the inoculation and incubation treatments did not stimulate microaggregation. We recorded more than 2.3×104 microaggregates/ml hemolymph in hemocyte preparations challenged with bacteria or bacteria plus etha-
1415
nol, and no microaggregates in preparations treated with ethanol (the pharmaceutical vehicle) or with nutrient broth (the bacterial challenge vehicle). Moreover, our data show that the incubation, ethanol and nutrient broth treatments did not impair the ability of hemocytes to react to the presence of bacteria. In the same vein, the hemocytes were not damaged in obvious ways by the pharmaceutical assessments, as shown by our cytotoxicity assessments. Our assessments indicate the percentages of viable hornworm hemocytes are similar to percentages observed in studies with primary cultures of murine macrophages (R.L. Pardy, personal communication). Hence, the microaggregation reactions we recorded were physiological reactions to experimental bacterial challenge and not due to adventitious cellular actions. Our time course experiments illustrate the rapid onset of cellular reactions, specifically microaggregation, to bacterial challenge. We recorded a steep increase in microaggregation during the first 30 min PI, which did not further increase or possibly decreased over the following 90 min of our experiments. This would be consistent with the kinetics of nodule formation, thought to proceed through formation of microaggregates to development of larger nodules. Because nodules are formed by adhesion of numerous microaggregates, it is not surprising to record little change or decreases in microaggregates in later stages of the nodulation process. Our data show that microaggregation was inhibited in hemocyte preparations treated with inhibitors of cyclooxygenase, but not in preparations treated with esculetin, a specific inhibitor of 5- and 12-lipoxygenase. This indicates that in in vitro experiments, prostaglandins, but not lipoxygenase products, are key mediators of microaggregation. This finding runs contrary to our whole-insect experiments, in which esculetin treatments inhibited in vivo nodulation reactions to bacterial infection in several insect species, including tobacco hornworms (Miller et al., 1994), larvae of the beetle, Zophobas atratus (Miller et al., 1996), adult crickets, Gryllis assimilis (Miller et al., 1999) and adult 17-year periodical cicadas, Magicicada semptemdem (Tunaz et al., 1999). More recent experiments with lipopolysaccharide (LPS) purified from the bacterium S. maracescens also indicate that eicosanoids mediate insect microaggregation reactions to purified LPS challenge (Bedick et al., 2000). In this latter work we reported that esculetin treatments also attenuated cellular reactions to LPS. Nodulation is thought to involve an unknown number of specific cellular actions, some of which may depend on lipoxygenase products while others are mediated by cyclooxygenase products. If any of the cellular actions in the overall process of nodulation were somehow prevented, the inhibition would be recorded as reduced nodulation. It now remains unclear why esculetin did not influence the microaggregation
1416
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
process in isolated hemocyte preparations. We interpret the results in terms of down-stream events, with the suggestion that esculetin influences one or more cellular actions in the post-microaggregation phase of the overall nodulation process. Our results with esculetin open an alternative, intriguing possibility that other tissues, particularly hemopoietic organs, may be involved in one or more aspects of the nodulation process. Circulating hemocyte populations often decline during an infection cycle (Lackie, 1988), which we also recorded in our nodulation experiments with tobacco hornworms (Miller et al., 1994). On the idea that release of sessile hemocytes from hemopoietic organs may contribute to the microaggregation and nodulation processes, our data showing that esculetin did not influence microaggregation reactions in isolated hemocytes may be understood if one or more lipoxygenase products act in hemopoietic organs. If this is so, then we would expect esculetin or other lipoxygenase inhibitors to influence nodulation in intact insects, but not necessarily in isolated hemocytes. Comparison to our earlier data with tobacco hornworms supports this view (Miller et al., 1994). We recorded approximately 8.37×106 microaggregates/ml hemolymph from whole insect preparations, compared to 103–104 microaggregates/ml hemolymph registered in the in vitro work reported here. The lower numbers of microaggregates probably reflect the lower numbers of hemocytes in each preparation, and the absence of sessile hemocytes (possibly in the hemopoietic organs) which could be recruited into the microaggregation process. These and other speculative ideas will motivate future experimental work on nodulation. There are difficulties associated with inferring physiological signaling pathways from results of experiments with pharmaceutical probes (Stanley-Samuelson, 1994b). We have discussed these in detail with respect to the role of eicosanoids in insect cellular immunity, along with discussion of the biological significance of other signaling moieties in insect immunity elsewhere (Miller et al., 1996; Stanley, 2000), and will not reiterate the points here. Such objections notwithstanding, the results of experiments with pharmaceutical probes, when interpreted critically, are valid. The significance of our work with isolated hemocyte preparations lies in gaining a more detailed understanding of the signaling pathways in cellular immune reactions to bacterial infections. In work at the organismal level, we recorded sharply reduced microaggregation and nodulation reactions to bacterial challenge following experimental treatments with pharmaceutical inhibitors of eicosanoid biosynthesis (Miller et al. 1994, 1996; Tunaz et al., 1999; Bedick et al., 2000). These findings, however, did not provide guidance with respect to the source of the eicosanoids involved in mediating cellular defense reactions. The results of our experiments with
isolated hemocyte preparations show that hemocytes biosynthesize and secrete eicosanoids which in turn influence the cellular actions of other hemocytes. Of course, the data presented in this paper do not exclude the possibility that eicosanoids from other tissue sources, particularly fat body, can influence hemocyte actions.
Acknowledgements We thank Dr Ralph Howard and Dr Virginia Naples for a careful reading and commentary on a draft of this manuscript. This is paper number 13,360 of the Nebraska Agricultural Research Division. This work was supported by the Agricultural Research Division, UNL (Project NEB-17-054) and by the Department of Biological Sciences, Northern Illinois University.
References Bedick, J.C., Pardy, R.L., Howard, R.W., Stanley, D.W., 2000. Insect cellular reactions to the lipopolysaccharide component of the bacterium Serratia marcescens are mediated by eicosanoids. Journal of Insect Physiology 46, 1481–1487. Dunn, P.E., Drake, D.R., 1983. Fate of bacteria injected into naive and immunized larvae of the tobacco hornworm, Manduca sexta. Journal of Invertebrate Pathology 41, 77–85. Gadelhak, G.G., Pedibhotla, V.K., Stanley-Samuelson, D.W., 1995. Eicosanoid biosynthesis by hemocytes from the tobacco hornworm, Manduca sexta. Insect Biochemistry and Molecular Biology 25, 743–749. Gupta, A.P., 1986. Arthropod immunocytes: identification, structure, functions, and analogies to the functions of vertebrate B- and Tlymphocytes. In: Gupta, A. (Ed.), Hemolytic and Humoral Immunity in Arthropods. Wiley, New York. Gupta, A.P., 1991. Immunology of Insects and Other Arthropods. In: Gupta, A.P. (Ed.). CRC Press, Boca Raton, FL. Horohov, D.W., Dunn, P.E., 1982. Changes in the circulating hemocyte population of Manduca sexta larvae following injection of bacteria. Journal of Invertebrate Pathology 40, 327. Howard, R.W., Stanley, D.W., 1999. The tie that binds: eicosanoids in invertebrate biology. Annals of the Entomological Society of America 92, 880–890. Jurenka, R.A., Miller, J.S., Pedibhotla, V.K., Rana, R.L., StanleySamuelson, D.W., 1997. Eicosanoids mediate microaggregation and nodulation responses to bacterial infections in black cutworms, Agrotis ipsilon, and true armyworms, Pseudaletia unipuncta. Journal of Insect Physiology 43, 125–133. Jurenka, R.A., Pedibhotla, V.K., Stanley, D.W., 1999. Prostaglandin production in response to a bacterial infection in true armyworm larvae. Archives of Insect Biochemistry and Physiology 41, 225– 232. Lackie, A.M., 1988. Haemocyte behavior. Advances in Insect Physiology 21, 85–178. Mandato, C.A., Diehl-Jones, W.L., Moore, S.J., Downer, R.G.H., 1997. The effects of eicosanoid biosynthesis inhibitors on prophenoloxidase activation, phagocytosis and cell spreading in Galleria mellonella. Journal of Insect Physiology 43, 1–8. Miller, J.S., Nguyen, T., Stanley-Samuelson, D.W., 1994. Eicosanoids mediate insect nodulation responses to bacterial infections. Proceedings of the National Academy of Sciences of the United States of America 91, 12418–12422.
J.S. Miller, D.W. Stanley / Journal of Insect Physiology 47 (2001) 1409–1417
Miller, J.S., Howard, R.W., Nguyen, T., Nguyen, A., Rosario, R.M.T., Stanley-Samuelson, D.W., 1996. Eicosanoids mediate nodulation responses to bacterial infections in larvae of the tenebrionid beetle, Zophobas atratus. Journal of Insect Physiology 42, 3–12. Miller, J.S., Howard, R.W., Rana, R.L., Tunaz, H., Stanley, D.W., 1999. Eicosanoids mediate nodulation reactions to bacterial infections in adults of the cricket, Gryllus assimilis. Journal of Insect Physiology 45, 75–83. Morishima, I., Yamano, Y., Knoue, K., Matsuo, N., 1997. Eicosanoids mediate induction of immune genes in the fat body of the silkworm, Bombyx mori. FEBS Letters 419, 83–86. Stanley, D.W., 2000. Eicosanoids in Invertebrate Signal Transduction Systems. Princeton University Press, Princeton, NJ. Stanley, D.W., Howard, R.W., 1998. The biology of prostaglandins and related eicosanoids in invertebrates: cellular, organismal and ecological actions. American Zoologist 38, 369–381. Stanley-Samuelson, D.W., 1994a. Prostaglandins and related eicosanoids in insects. Advances in Insect Physiology 24, 115–212.
1417
Stanley-Samuelson, D.W., 1994b. Assessing the physiological significance of prostaglandins and other eicosanoids in insect physiology. Journal of Insect Physiology 40, 3–11. Stanley-Samuelson, D.W., Ogg, C.L., 1994. Prostaglandin biosynthesis by fat body from the tobacco hornworm, Manduca sexta. Insect Biochemistry and Molecular Biology 24, 481–491. Stanley-Samuelson, D.W., Jensen, E., Nickerson, K.W., Tiebel, K., Ogg, C.L., Howard, R.W., 1991. Insect immune response to bacterial infection is mediated by eicosanoids. Proceedings of the National Academy of Sciences of the United States of America 88, 1064–1068. Tunaz, H., Bedick, J.C., Miller, J.S., Hoback, W.W., Rana, R.L., Stanley, D.W., 1999. Eicosanoids mediate nodulation reactions to bacterial infections in adults of two 17-year periodical cicadas, Magicicada septendecim and M. cassini. Journal of Insect Physiology 45, 923–931.