Humoral encapsulation of the fungus Erynia radicans (Entomophthorales) by the potato leafhopper, Empoasca fabae (Homoptera: Cicadellidae)

JOURNAL

OF INVERTEBRATE

PATHOLOGY

52,4%56

(1988)

Humoral Encapsulation of the Fungus Etynia radicans (Entomophthorales) by the Potato Leafhopper, E’mpoasca fabae (Homoptera: Cicadellidae) T. M. BUTT,* S. P. WRAIGHT,~ S. GALAINI-WRAIGHT,$ D. W. ROBERTS,~ AND R. S. SOPER*

R. A.

HUMBER,*

*U.S. Department of Agriculture, Agricultural Research Service, Plant Protection Research Unit, Boyce Thompson Institute, and tBoyce Thompson Institute, Tower Road, Ithaca, New York 14853; and #Department of Entomology, Cornell University, Ithaca, New York 14853 Received October 5, 1987; accepted November 30, 1987 The immune response of the potato leafhopper, Empoasca fabae, to fungal infection is humoral encapsulation. Noncellular, melanotic capsules were formed around invading hyphae of Erynia radicans, Metarhizium anisopliae, and Hirsutella guyana. Melanization of E. radicans hyphae was observed in the cuticle, epidermis, and hemolymph of infected leafhoppers, in some cases within 4-6 hr post-inoculation. The encapsulation response was nearly always ineffective in preventing fungal infection. Rarely, hyphae in the hemolymph were completely encased in thick melanotic capsules. Thin sections of melanized hyphae showed the melanin to be a granular, electron-opaque substance. A few hyphae were observed coated with a fdamentous, electron-opaque substance which also may be melanin. 0 1988AC&& FTCSS, IN. KEY WORDS: Empoasca fabae; Erynia radicans; Metarhizium anisopliae; Hirsutella guyana; melanization; encapsulation; melanin:

INTRODUCTION

mycelium of E. radicans strain ARSEF’ Two types of encapsulation are recog- 1590, prepared according to McCabe and nized as insect immune reactions to fungal Soper (1985), were spread evenly onto wainfection. The first, cellular encapsulation, ter agar in Petri dishes. The dishes were generally involves both hemocytes and then closed and incubated at room tempermelanin formation and is widespread and ature until prodigious sporulation (ca. 12 well documented (Ratcliffe and Rowley, hr). Fifth-instar nymphs of E. fabae from a 1979). The second, humor-al encapsulation, laboratory colony were anesthetized with entails formation of a melanotic capsule CO2 and collected in groups of 15. Each of around the parasite without direct partici- the groups of anesthetized leafhoppers was pation of hemocytes and, previously, has exposed to a different sporulating culture only been reported in dipterans (Gotz and for 7 min, during which time the culture was Vey, 1974; Travland, 1979; Sweeney et al., continuously rotated. The inoculated leaf1983; G&z, 1986). This paper describes hu- hoppers were held at 20°C on cowpea fomoral encapsulation of the entomophthora- liage under moisture-saturated conditions with a 15hr photoperiod. Inoculations ceous fungus Erynia (Zoophthora) radicans by the potato leafhopper, Empoasca fabae were made at the beginning of the photo(Homoptera: Cicadellidae). This is the first phase. Observations were made at 2-hr indescription of humoral encapsulation in a tervals during the first 14 hr post-inocunondipterous insect and the first record of lation and then at various times until death encapsulation of an entomophthoraceous of the nymphs, usually within 48 hr. Specimen preparation (slide mounting and stainfungus. MATERIALS Inoculation

’ U.S. Department of Agriculture, Agriculture Research Service Collection of Entomopathogenic Fungi, Ithaca, New York.

AND METHODS

of E. fabae. Particles of dry 49

0022-2011188 $1.50 Copy&M 0 l!%S by Academic Press, Inc. All rights of rcpraductioll in any form reserved.

50

BUTT ET AL. a)

b)

FIG. 1. Simplified diagram summarizing our interpretation of some of the permutations of Eryniu radicans melanization in the leafhopperEmpoasca fabae. (a) Melanizationof the cuticle in juxtapositionto appressorium; (b) melanization of the pegin the integument;(c) bothpegandinvasionhypha areencapsulated; (d) only the invasionhyphais melanized;(e)melanizationin the cuticleandhemocoel, whereonly part of the invasionhyphais melanized.A, appressorium; B, balloonportion of hypha; C, cuticle; E, epidermis;F, fdamentousportion of hypha; PP, penetrationpeg; S, spore. Blackenedareasrepresentmelanin.

RESULTS ing) precluded repeated observations of the same insects. Melanized and nonmelanized General Observations hyphae dissected from infected insects were also examined. Although this investiMelanization of E. rudicans occurred in gation focused on E. rudicans, some leaf- the cuticle (Figs. 1, 2), epidermis, and hoppers infected with the entomogenous hemocoel of E. fubae (Figs. 1, 3-lo), but hyphomycete fungi Hirsutella guyana (AR- not necessarily in all three at the same time. SEF 881) and Metarhizium anisopliae (AR- Our interpretation of some of the permutaSEF 550) were also examined to determine tions observed are summarized in Figure 1. whether melanization was independent of The melanin was reddish-brown. No lysed the type of challenging fungus. or intact hemocytes were observed at any Fluorescence and electron microscopy. time close to or in contact with fungal eleSpecimens were placed under a coverslip ments. This indicates an encapsulation reon a microscope slide in a drop of 0.5% sponse similar to the humor-al response deaqueous (w/v) Uvitex BOPT (CIBA-Geigy’) scribed by G&z and Vey (1974). Hyphae of H. guyana and M. anisopliae and examined with an Olympus HB2 photomicroscope with an epifluorescence at- were also melanized without formation of tachment; the combination of exciter, chro- hemocytic capsules, showing that the rematic beam splitter, and barrier filters was sponse was not limited to E. radicans. 390-440, 460, and 475 nm, respectively. Specimens were examined using bright- Observations on the encapsulation of Erynia radicans field, epitluorescent , and combined brightfield and epifluorescent illumination. ImMelunotic lesions. Melanotic lesions are ages were recorded on Kodak Technical defined here as wounds on or in the cuticle, Pan or Tri-X Pan film. often marking the site of fungal penetration Specimens were prepared for transmis- or attempted penetration. These lesions sion electron microscopy (TEM) according were found in sclerites and intersegmental to Butt and Beckett (1984) and examined in membranes. Occasionally the cuticle a Zeiss EM-10 at 60 kV. touching a fungal hypha was melanized, but ’ Mention of companynamesand/or proprietary the lesions were usually located at penetramicrosproductnamesdoesnot constituteendorsement by the tion sites (Fii. l-4). Fhroreme U.S. Departmentof Agriculture. copy permitted detection of minute lesions

HORMONAL

ENCAPSULATION

which otherwise would not have been observed with brightfield or phase-contrast optics. Melanotic collar. Soon after E. radicans breached the cuticle, a melanotic collar in some instances formed around the penetration peg (Figs. 3, 4). Most collars were between 0.7 and 1.2 pm thick; however, thicknesses of up to 3 pm were recorded. The melanin usually remained localized in

OF

Etynia radicans

51

or beneath the cuticle but occasionally extended the length of the peg. The pegs varied in length from 1.4 to 14 Frn but rarely exceeded 1.2 pm in diameter. A few pegs measured approximately 2.4 km in diameter. Growth of E. radicans was rarely blocked by melanotic capsules. Most pegs appeared either to breach the collar or to grow faster than the melanin could be syn-

FIG. 2. Fluorescence micrograph showing melanization of the cuticle (arrow) around the appressorium. The appressorium is formed at the end of the primary spore, not a germ tube. x880. FIG. 3. Fluorescence micrograph showing a melanotic collar around a penetration peg (arrows). The peg arises from the bulbous tip of an elongate appressorium. x890. FIG. 4. Combined brightfield and fluorescence micrograph of a partially melanized peg and an invasion hypha encased in a thick sheath of melanin. The spore and nonmelanized portions of the peg fluoresce intensely. Note the cracks in the melanin sheath. x890. FIG. 5. Combined brightfield and fluorescence micrograph of two melanized hyphae in the first abdominal segment. One melanized hypha is still attached to the brightly fluorescing collapsed germ tube/elongate appressorium on the cuticular surface. x420. FIG. 6. Melanized hypha still anchored to the cuticle (out of the plane of focus). Note the hyphal expansion (“ballooning”) prior to growth as a filament. x890.

52

BUTT

thesized and deposited. A few infections were observed in which it appeared that the initial infection peg had been arrested by a thick, complete melanotic capsule, and in response the fungus had produced a second peg from the same appressorium. In such casesthe second peg had successfully penetrated into the hemocoel. Some melanized pegs were highly contorted, which may represent an attempt by the fungus to escape the host response. Mefunization of hyphae. Once the fungus breached the cuticle, it either grew in the epidermis (Fig. 7), between the cuticle and epidermis (Fig. 8), or entered the hemocoel where it “ballooned” (i.e., expanded dramatically), and then continued growth as a filament (Figs. 1, 5, 6, 9). A few hyphae were completely encapsulated (Fig. 4), but

ET AL.

most continued to grow in spite of being melanized (Figs. 5, 6, 9). In the leafhopper host, E. rudicans hyphae grew primarily as unbranched or sparsely branched filaments. The cell wall of these hyphae varied in thickness (compare Figs. 11-13) and structure. Generally, melanin was deposited as a continuous layer of uniform thickness over the area of hyphal surface affected by the encapsulation reaction (Figs. 5, 6, 8). Nonmelanized hyphae measured 4.7-5.2 p,rn in diameter while melanized hyphae reached 9.5 km. Thinner depositions of melanin were observed near the growing apices of hyphae (Fig. 9). It was not unusual to find in a group of fungal cells only one or two melanized hyphae, the rest apparently unaffected.

FIG. 7. A transmission electron micrograph of the fungus (F) growing in the epidermis (E). ~8,600. FIG. 8. A melanized hypha growing between the cuticle (C) and epidermis (E). x 13,600.

HORMONAL

ENCAPSULATION

OF Erynin ru&cuns

FIG. 9. Section through a growing hyphal tip with melanin deposition. Note how the melanin tapers in thickness near the tip of the hypha. x8600. FIG. 10. Thin section of a hypha which is partially coated with a layer of filamentous melanin-like material (arrows). X 13,300. FIG. 11. Details of part of a hypha with an extremely thin cell wall. Note the faint granular staining of the cell wall. ~54,600. FIG. 12. Filaments of melanin-like material deposited on the surface of a fungal cell wall. x 33,900. FIG. 13. Granular melanin coating the surface of a fungal cell wall, X27,300.

53

54

BUTT

Encapsulation was observed in most regions of the host body, including the legs and wing pads. Fewer melanized hyphae were seen in the head and thorax than in the abdomen probably because there were more hyphal penetrations in the abdomen (unpubl.). The number of encapsulated hyphal strands and the total amount of melanization increased with time. Observations on melanin sheath. In thin sections, two types of melanin-like substances were found in the hemocoel. The usual type was a coarsely granular, electron-opaque substance (Figs. 8, 9, 13); less often, a filamentous, electron-opaque coating was observed (Figs. 10, 12). It is unlikely that the filamentous form is an artifact since the distribution of this type of melanin was not random (i.e., it was only found coating fungal hyphae). The granular melanin was deposited on the surface of fungal hyphae as a thin layer that gradually increased in thickness as granules were incorporated into the capsule. This resulted in an amorphous, electron-opaque layer of melanin encapsulating the fungus. Septa and other cell inclusions could be resolved in hyphae which had a relatively thin melanin sheath (i.e., less than 3 pm). Where the sheath thickness exceeded 3 urn, longitudinal and circumferential cracks sometimes appeared (Fig. 4). These cracks could be resolved inside living insects. The thick crust could be fragmented by maceration with a needle, suggesting that the melanin sheath was extremely brittle. In such sheaths, striations were also observed, usually extending along the length of the hypha. The melanin sheath darkened as it increased in thickness, but the color could be partially bleached after 1 hr incubation with 30% hydrogen peroxide.

-Cl

AL. A-

larvae by Beauveria bassiana (see G&z and Vey, 1974), of mosquito Culiseta inornata larvae by Coelomomyces psorophorae (see Travland, 1979), of various mosquito larvae by Culicinomyces clavisporus (see Sweeney et al., 1983), and the crayfish Astacus astacus by Aphanomyces astaci (see Sdderhall et al., 1979). In the latter case, hemocytes are associated with melanized hyphae in the hemocoel but melanization of the cuticle appears to be humoral. However, in all the other systems, encapsulation of the invading fungi is humoral. Goetz et al. (1977) carried out a survey of defense responses in insects from several different orders and found that humor-al encapsulation occurred in insects with few circulating blood cells. This study shows that the defense reaction of E.fabae, which lacks any readily identifiable circulating hemocytes, is humoral, thus supporting the findings of Goetz et al. However, humoral encapsulation is not the defense response of all homopteran insects with few hemocytes. Butt (1983) found no evidence of humoral encapsulation of live entomophthoralean fungi infecting aphids. This included Erynia radicans in the pea aphid, Acyrmus luridus

thosiphon

pisum.

Melanotic

Lesions

The various permutations of melanization reported here (Fig. 1) suggest that the enzymes responsible for melanin synthesis, the phenoloxidases, are present in the integument and hemolymph, and that these may function independently. The presence of phenoloxidase in insect cuticle (Aoki and Yanase, 1970; Anderson, 1985), hemolymph (Ashida et al., 1983), and hemocytes (Leonard et al., 1985) is well documented. In the cuticle, phenoloxidases are responsiDISCUSSION ble for tanning and sclerotization, a related Host reactions involving melanization process, which makes the cuticle stiffer, and encapsulation were observed during less soluble, more resistant to degradation, penetration of the integument and in the and darker (Anderson, 1985). Activation of hemocoel. Similar reactions have been de- these enzymes is elicited by certain sugars, scribed for the infection of midge Chironothe p-1,3-glucans, found in fungal cell wails

HORMONAL

ENCAPSULATION

OF

Erynia radicans

55

(S&lerh&ll and Unestam, 1979; Siiderhall, 1982). This may explain why entomophthoralean fungi which form protoplasts within their respective hosts (Tyrrell, 1977; Butt et al., 1981) do not elicit an immune response (Dunphy and Nolan, 1982). Cuticular phenols are proably exploited as substrates by phenoloxidase, although transport of phenols from the epidermal cells is known to occur (Anderson, 1985). The presence of melanized hyphae between the epidermis and cuticle suggeststhat phenoloxidase and substrate(s) are present in this region, or are translocated from the epidermis. Many investigators have reported melanotic lesions on insect cuticle at sites of fungal penetration (e.g., Brobyn and Wilding, 1977; Butt, 1983, 1987). It is not clear whether the reddish-brown lesions represent a wound repair response or a defense response. In either case, melanization of the cuticle results in both chemical and structural changes in the cuticle. It is generally accepted that entomopathogenic fungi penetrate host cuticle using a combination of enzymes and mechanical force (e.g., Brobyn and Wilding, 1977). Melanization, like sclerotization/tanning, may alter the mechanical properties of the cuticle, making it both more rigid (Anderson, 1985) and resistant to enzymatic degradation. Melanin is known to inhibit microbial enzymes (Kuo and Alexander, 1967)and is fungistatic (Soderhall and Ajaxon, 1982). Melanotic lesions, however, as shown in the present study, did not prevent penetration of the leafhopper cuticle by E. rudi-

the fungi M. anisopliae and Aspergillus niger. Sweeney et al. (1983) reported humoral encapsulation of the invading hyphae of C. clavisporus but found that this did not prevent successful colonization of the mosquito host. It appears from these observations that humoral encapsulation is generally ineffective against fast growing, pathogenic fungi. It might, however, be effective against nongrowing or slow-growing organisms such as eggs or larvae of parasitoids. In cellular encapsulation, the cells of the inner layer usually lyse, releasing melanin or its precursors (Ratcliffe and Rowley, 1979; Butt, unpubl.), thereby, as in the case of humoral encapsulation, ensheathing the foreign organism in melanin. The melanin sheath acts as a physicochemical constraint. Not only is it fungistatic (Siiderhall and Ajaxon, 1982) but it also imposes a physical constraint as suggestedby the contorted appearance of some of the hyphae of E. radicans. The stress lines and cracks observed in relatively thick melanin sheaths suggest that the fungus is exerting some force in its attempt to break through the capsule. This study, however, provides evidence that the fungus is not always able to break through a thick capsule.

cans.

S. 0. 1985. Sclerotization and tanning of the cuticle. In “Comprehensive Insect Physiology, Biochemistry and Pharmacology” (G. A. Kerkut, and L. I. Gilbert, Eds.), Vol. 12, pp. 59-74. Pergamon Press, Oxford. AOKI, J., AND YANASE, K. 1970. Phenol oxidase activity in the integument of the silkworm Bombyx mori infected with Beauveria bassiana and Spicaria fumoso-rosea. .I. Znvertebr. Pathol., 16, 45M.

Melanization

ACKNOWLEDGMENTS This research was supported in part by the Bean/ Cowpea Collaborative Research Support Program of the U.S. Agency for International Development (USAIDZBIFAD Grant AIDIDSAN-X11-G-0261).

REFERENCES ANDERSON,

of Invading

Hyphae

This study shows that the humoral response of E. fabae is usually inadequate for containing growth of the pathogen E. radicans. Similarly, Gotz and Vey (1974) found that growth of the hyphomycete pathogen B. bassiana competed well with the speed of capsule formation and often escaped encapsulation. These workers found that humoral encapsulation was effective against

ASHIDA,

M.,

ISHIZAKI,

Y., AND IWAHANA,

H. 1983.

Activation of prophenoloxidase by bacterial cell walls or B-1,3glucans in plasma of the silkworm, Bombyx mori. Biochem. Biophys. Res. Commun., 113, 562468.

56

BUTT

P. J., AND WILDING, N. 1977. Invasive and developmental processes of Entomophthora species infecting aphids. Trans. Brit. Mycol. Sot., 69, 349-

BROBYN,

366.

Bum, T. M. 1983. “Studies on the Invasive and Developmental Processes of Five Genera of the Entomophthorales with Specific Reference to Erynia neoaphidis on the Pea Aphid, Acyrthosiphon pisum Harris.” Ph.D. thesis. University of Bristol, Bristol. Burr, T. M. 1987. A fluorescence microscopy method for the rapid localization of fimgal spores and penetration sites on insect cuticle. J. Znvertebr. Pathol., 50,72-74.

BUTT, T. M., AND BECKETT, A. 1984. Ultrastructure and behaviour of the spindle pole body of the aphidpathogenic fungus Erynia neoaphidis. Protoplasma, 120, 61-71.

BUTT, T. M., BECKER, A., AND WILDING, N. 1981. Protoplasts in the in vivo life cycle of Erynia neoaphidis. .I. Gen. Microbial., 127, 417-421. DUNPHY, G. B., AND NOLAN, R. A. 1982. Cellular immune responses of spruce budworm larvae to Entomophthora egressa protoplasts and other test particles. J. Znvertebr. Pathol., 39, 81-92. GOETZ, P., ROETTGEN, I., AND LINGG, W. 1977. Encapsulement humoral en tant que reaction de defense chez les Dip&es. Ann. Parasitol., 52, 9597.

G~Tz, P. 1986. Encapsulation in arthropods. In “Immunity in Invertebrates” (M. Brehelin, Ed.), pp. 153-170. Springer-Verlag, Berlin. G~Tz, P., AND VEY, A. 1974. Humoral encapsulation in Diptera (Insecta): Defence reactions of Chironomus larvae against fungi. Parasitology, 68, 193-205. Kuo, M.-J., AND ALEXANDER, M. 1967. Inhibition of lysis of fungi by melanins. J. Bacterial., 94,624-629. LEONARD, C., S~DERH~~LL, K., AND RATCLIFFE, N. A. 1985. Studies on prophenoloxidase and pro-

ET AL.

tease activity of Blaberus sect Biochem.,

craniifer

haemocytes. Zn-

6, 803-810.

MCCABE, D., AND SOPER, R. S. 1985. “Preparation of an Entomopathogenic Fungal Insect Control Agent.” U.S. Patent 4,530,834. NYHLEN, L., AND UNESTAM, T. 1980. Wound reactions and Aphanomyces astaci growth in crayfish cuticle. J. Znvertebr. Pathol., 36, 187-197. RATCLIFFE, N. A., AND ROWLEY, A. F. 1979. Role of hemocytes in defense against biological agents. In “Insect Hemocytes” (A. P. Gupta, Ed.), pp. 331414. Cambridge Univ. Press, Cambridge. S~DERH~~LL, K. 1982. Prophenoloxidase activating system and melanization: A recognition mechanism of arthropods? A review. Dev. Camp. Zmmunol., 6, 601-611. S~DERH~~LL, K., AND AIAXON, R. 1982. Effect of quinones and melanin on mycelial growth of Aphanomyces spp. and extracellular protease of Aphanomyces astaci, a parasite on crayfish. J. Znvertebr. Pathol., 39, 105-109. S~DERHALL, K., HALL, L., UNESTAM, T., AND NYHLEN, L. 1979. Attachment of phenoloxidase to fungal cell walls in arthropod immunity. J. Znvertebr.

Pathol.,

34, 285-294.

SODERHKLL, K., AND UNESTAM, T. 1979. Activation of serum prophenoloxidase in arthropod immunity: The specificity of cell wall glucan activation by purified fungal glycoproteins of crayfish phenoloxidase. Canad.

J. Microbial.,

25, 406-414.

SWEENEY, A. W., INMAN, A. O., BLAND, C. E., AND WRIGHT, R. G. 1983. The tine structure of Culicinomyces clavisporus invading mosquito larvae. J. Znvertebr.

Pathol.,

42, 224-243.

TRAVLAND, L. B. 1979. Initiation of infection of mosquito larvae (Culiseta inornata) by Coelomomyces psorophorae.

J. Znvertebr.

Pathol.,

33, 95-105.

TYRRELL, D. 1977. Occurrence of protoplasts in the natural life cycle of Entomophthora egressa. Exp. Mycol., 1, 259-263.