INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6M x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
University Microfilms International A Bell & Howell Information Com pany 300 North Z eeb Road Ann Arbor Ml 48106-1346 USA 313 761-4700 800 521-0600 Order Number 9326670
Physiological and behavioral interactions between the ectoparasitoidNasonia vitripennis (Walker) (Hymenoptera: Pteromalidae) and its pupal hostSarcophaga buliata P arker (Diptera: Sarcophagidae)
Rivers, David Bradley, Ph.D.
The Ohio State University, 1993
UMI 300 N. Zeeb Rd. Ann Aibor, Ml 48106 PHYSIOLOGICAL AND BEHAVIORAL INTERACTIONS BETWEEN THE
ECTOPARASITOID NASONIA VITRIPENNIS (WALKER) (HYMENOPTERA:
PTEROMALIDAE) AND ITS PUPAL HOST SARCOPHAGA BULLATA
PARKER (DIPTERA: SARCOPHAGIDAE)
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
David Bradley Rivers, B.S.
*****
The Ohio State University
1993
Dissertation Committee:
D.L. Denlinger Approved by
W.F. Hink
R.W. Hall
^ m a m i - D.C. Smith Advisor ^ Department of Entomology To Dr. Harold L. Zimmack, who taught me about life. I also dedicate this dissertation to my family for their
enduring support and encouragement.
ii ACKNOWLEDGMENTS
I express my sincere appreciation to Dr. David L.
Denlinger for his guidance and encouragement throughout
this research. I also thank the other members of my
advisory committee, Drs. w. Fred Hink, Richard W. Hall,
and D. Courtney Smith, for their insight and comments.
Gratitude is extended to Jay Yoder, Brian Almond, Mei-
Ling Zhang, Sherry Johnson, Karl Joplin, and George Yocum
for many helpful comments throughout this study. I am
also indebted to George Keeney for his insight and
friendship, to Foster Purrington for his many helpful
suggestions, and to the graduate students in the
Department of Entomology for their friendship and
support, in particular Dave Kallander, Bob Hancock, Diana
Sammataro, Amadou Ba, and Jane Zumwalt. To my wife,
Cheryl, I offer sincere thanks for your unshakable faith
in me and for the love and encouragement you provided
throughout my endeavors. I also thank my daughter,
Megan, for giving me inspiration. VITA
September 17, 1966 ...... Born - Kokomo, Indiana
1989 ...... B.S., Ball State University, Muncie, Indiana
1989-1990 ...... Research Assistant, Department of Biochemistry, The Ohio State University, Columbus, Ohio
1990-Present ...... Teaching Assistant, Department of Entomology, The Ohio State Univeristy, Columbus, Ohio
PUBLICATIONS
Rivers, D. B. , C. N. Vann, and H. L. Zimmack. 1988. Specificity of the Bacillus thurinaiensis delta- endotoxins toward the larvae of Ostrinia nubilalis. Proceedings of the Indiana Academy of Science 98:21-24.
Milne, R., A. Ge, D. Rivers, and D. Dean. 1990. Specificity of insecticidal crystal proteins: Implications for Industrial Standardization. In: Analytical Chemistry of Bacillus thurinaiensis. L. Hickle and W. Fitch (Eds), American Chemical Society, pp. 22-35.
Rivers, D. B., C. N. Vann, H. L. Zimmack, and D. H. Dean. 1991. Mosquitocidal Activity of Bacillus laterosporus. Journal of Invertebrate Pathology 58:444-447.
iv Ge, A., D. B. Rivers, R. Milne and D. H. Dean. 1991. Functional domains of Bacillus thurinoiensis insecticidal crystal proteins. Journal of Biological Chemistry 266(27):17954-17958.
Rivers, D. B., W. F. Hink, and D. L. Denlinger. 1993. Toxicity of the venom from Nasonia vitripennis (Hymenoptera: Pteromalidae) toward fly hosts, nontarget insects, different developmental stages, and cultured insect cells. Toxicon (In press).
FIELDS OF STUDY
Major Field: Entomology (Insect Physiology)
v TABLE OF CONTENTS
DEDICATION ...... ii
ACKNOWLEDGMENTS ...... iii
VITA ...... iv
LIST OF TABLES ...... viii
LIST OF FIGURES ...... ix
INTRODUCTION ...... 1
CHAPTER PAGE
I. HOST QUALITY AND ITS IMPACT ON FECUNDITY( DEVELOPMENTAL TIME, AND SEX RATIO IN NASONIA VITRIPENNIS. A GREGARIOUS, ECTOPARASITOID OF THE FLESH FLY, SARCOPHAGA B U L L A T A ...... 10
Introduction ...... 10 Materials and Methods ...... 11 R e s u l t s ...... 17 D i s c u s s i o n ...... 36
II. DEVELOPMENTAL FATE OF THE FLESH FLY, SARCOPHAGA BULLATA. ENVENOMATED BY THE PUPAL ECTOPARASITOID, NASONIA VITRIPENNIS ...... 44
Introduction ...... 44 Materials and Methods ...... 4 6 R e s u l t s ...... 51 D i s c u s s i o n ...... 65
III. REDIRECTION OF METABOLISM IN THE FLESH FLY, SARCOPHAGA BULLATA. ENVENOMATED BY THE ECTOPARASITOID NASONIA VITRIPENNIS: A CONTRAST BETWEEN NONDIAPAUSING AND DIAPAUSING HOSTS . . 70
Introduction ...... 7 0 Materials and Methods ...... 73 R e s u l t s ...... 78 vi D i s c u s s i o n ...... 90
IV. TOXICITY OF THE VENOM FROM NASONIA VITRIPENNIS (HYMENOPTERA: PTEROMALIDAE) TOWARD FLY HOSTS, NONTARGET INSECTS, DIFFERENT DEVELOPMENTAL STAGES, AND CULTURED INSECT CELLS ...... 97
Introduction...... 97 Materials and Methods ...... 98 R e s u l t s ...... 103 D i s c u s s i o n ...... 115
SUMMARY ...... 120
LIST OF REFERENCES ...... 125
vii LIST OF TABLES
TABLE PAGE
1. Fecundity of Jl. vitripennis on restricted body regions of its pupal host, S. bullata...... 59
2. Injection of 20-hydroxyecdysone into pupae of S. bullata after envenomation ...... 60
3. Hemolymph amino acid composition in pharate adults of S. bullata ...... 85
4. Activity of the venom from Nasonia vitripennis toward natural and nontarget insect species. . . 110
5. Response of Trichoolusia ni (TN-368) and Sarcophaaa perearina (NIH SaPe4) cell lines to the venom of Nasonia vitripennis ...... 112 LIST OF FIGURES
FIGURE PAGE
1. (A) Mean + SEM number of M. vitripennis adults produced per host on 4 species of fly pupae (£. bullata. £. crassipalpis. £. abnormis. and M- domestical (F« 15.68; df« 3,139; P < 0.01). (B) Mean + SEM time (days) required for 1*. vitripennis to develop from egg (deposition) to adult (emergence from host puparia) at 25°C, L15:D9 on pupae from each species of host (F= 117.1; df= 3,45; P < 0.01). 36-64 pupae of each fly host were used for each experiment. Columns preceded by the same letter do not differ significantly at P > 0.05 (ANOVA) . . . 28
2. The relationship between the number of adults of N . vitripennis produced and pupal weight (mg) of the host, S . bullata. The regress ion i s significant (y= 25*94 + 0.23x, P < 0.05, r2= 0.73, SEjg® 0.18) up to a host weight of 140 mg. Increases in host weight beyond 14 0 mg did not result in an increase in the number of parasitoids produced......
3. (A) Mean + SEM number of eggs deposited (-•- host-fed and unfed wasps) and adults of N. vitripennis produced (-o- by host-fed and -□- unfed wasps) by host-fed and unfed wasps. (B) Adult eclosion (%) of JJ. vitripennis from envenomated (-•- host-fed and -G- unfed wasps) and unenvenomated (-•- host-fed, -o- unfed), nondiapausing puparia of S. bullata. (C) Mean ± SEM number of adult parasitoids produced on hosts of £. bullata killed by a 1 h exposure to —7 0°C. (D) Adult eclosion (%) of N. vitripennis resulting from egg transplantion to unenvenomated, dead hosts. (E) Oviposition and progeny production -o- by £. vitripennis on diapausing pupae of £. bullata. (F) Adult eclosion (%) of N. vitripennis resulting from egg transfer to unenvenomated, diapausing hosts. These experiments were replicated twice using ix 2 3-46 puparia in each replicate for each age and physiological state of the host ...... 32
4. (A) The proportion of males produced by host-fed -•- and unfed -o- JI. vitripennis females in 24 h on nondiapausing puparia of S. bullata. (B) Proportion of males produced on dead puparia and unenvenomated, dead puparia -o-, (C) Proportion of males produced on diapausing puparia -o- of £. bullata and unenvenomated, diapausing puparia Each experiment was replicated twice using 23-46 puparia in each replicate...... 34
5. The developmental fate of £* bullata hosts envenomated by 1$. vitripennis on different days after pupariation...... 61
6. The developmental fate of four different ages of S. bullata hosts envenomated in the anterior (A) or posterior (P) half of the body by N. vitripennis...... 63
7. Changes in the rate of oxygen consumption and metabolites in nondiapausing pharate adults (5 days after pupariation) (left panels) of s. bullata and diapausing pupae (20 days in diapause) (right panels)...... 86
8. Mean + SEM total body alanine content (umoles/g) in nondiapausing pharate adults of S. bullata envenomated by H. vitripennis. All envenomated hosts contained significantly more alanine than any of the unenvenomated flies (P <0.01, One-way ANOVA). Each experiment was replicated twice using 5 pharate adults per replicate...... 88
9. Cytotoxicity of Nasonia vitripennis venom toward TN-3 68 and NIH SaPe4 cells...... 113
X INTRODUCTION
A parasitic lifestyle may be the most efficient means for an insect species to proliferate and develop.
Parasitic relationships may reduce the energy input
required by the parasite to obtain nutriment since the host may serve as the sole food source (an infinite resource if the host still feeds during parasitism). In addition, the parasite may utilize host hormones and metabolites by synchronizing its development with that of the host's. Such a relationship ultimately would lead to a reduction in parasite biosynthesis of metabolic intermediates since these factors are obtainable from the host. Additionally, the parasite may benefit from the physical aspects of the host.
The host integument may serve as a physical barrier to prevent water loss from the developing parasite, or the body of the host can be used to protect the parasite from harsh environmental conditions. Overwintering parasites may benefit from living in a diapausing host by using host-produced cryoprotectants to inhibit ice formation in their own bodies, or the host body may
1 2
simply prevent formation of ice crystals in the parasite
by providing a barrier to contact nucleation. Despite
this seemingly one-sided relationship, the host may
possess an arsenal of defenses to counteract a parasitic
attack. However, a "specialized” group of insect
parasites, referred to as parasitoids, has overcome the
problems associated with host defenses by ensuring that
the host can not pass on to its offspring the genetic
information coding for host resistance.
Insect parasitoids exert a profound effect on the
development, physiology, behavior, and morphology of the
host (Fisher, 1971; Beckage, 1990a; Lawrence, 1986;
Strand, 1984; Shaw, 1981; Moore, 1989), but unlike a parasitic relationship, this type of interaction always
results in death of the host. These host alterations
(including death) have often been attributed to larval feeding on or in the host, but numerous studies indicate that the female parasitoid is responsible for the onset of host manipulation (Beckage & Templeton, 1986; Vinson &
Barras, 1970; Shaw, 1981, and Jones et- a l ., 1986). In fact, the ability to alter host development is regulated by factor(s) (calyx fluid, venom, polydnavirsus, etc.) injected by parasitic wasps during oviposition (Wani et al., 1990; Dover et al., 1987; Strand and Dover, 1991;
Krell, 1991). 3
Insect parasitoids can live either within the host
(endoparasitic) or on the host (ectoparasitic), and as a consequence, these parasitoids have evolved unique attributes to cope with each lifestyle. Such parasitoid adaptations are more fully understood in the endoparasitoids, but this is simply because these insects have received the most attention. The reasons for this attention stem from observations showing that several intriguing interactions occur between endoparasitoids and their hosts: these parasitoids can suppress the immune response of the host, they rely on the host endocrine status for their own development, and endoparasitoids can redirect the intermediary metabolism of the host insect for the benefit of the wasp's progeny.
The egg of an endoparasitic wasp is deposited into the host hemolymph, subjecting the parasitoid to the full force of the host insect's immune response.
Endoparasitoids do not, however, leave their offspring without protection. Females of several species of wasps inject virus-containing calyx fluid and venom into the host with the egg during oviposition (Wani et al., 1990;
Krell, 1991; Strand and Dover, 1991), and the two appear to work synergistically to suppress hemocyte encapsulation of the parasitoid's egg (Stolz and Guzo,
1986; Beckage, 1990b; Whitfield, 1990). The viral 4
material can also induce mimicry of host antigens by
adhering to the egg surface so that the egg is not
recognized as a foreign particle (Schmidt and Schuchmann
Feddersen, 1989) . In addition, the injection of wasp
venom appears to inhibit host immune responses commonly
observed during pathogenic invasion (Beckage et al.,
1989). The egg itself releases a group of cells
(teratocytes) from the enveloping membrane that are
thought to offer protection for the parasitoid, although
there appears to be little agreement as to how this is
achieved (Dahlman, 1990). The endoparasitic insect has
developed the ability to not only survive in this hostile
environment, but it has also found ways to capitalize on
living in the host hemolymph.
The developing endoparasitoid is bathed by the hemolymph of the host and thus must be exposed to a constantly changing hormonal milieu. Since many of the parasitoids can develop in more than one stage of the host, most reports have ruled out any influence of host hormones and believed the parasitoids developed autonomously (reviewed by Beckage, 1985). However, recent observations suggest that many parasitoids demonstrate stage-specific development and that they show a dependence upon the host's endocrine status (Beckage,
1985; Lawrence, 1986; Brown and Reed-Larsen, 1991; Strand et al., 1991). As a consequence of the parasitoid's
reliance on host-derived hormones, the host typically
enters an arrested or retarded development. This
manipulation of the host may also result in alterations
in host metabolism.
Endoparasitic insects have been shown to redirect
the intermediary metabolism of their hosts following
parasitism. This redirection is generally thought to
enhance parasitoid growth (Vinson and Iwantsch, 1980a;
Thompson, 1986b), but evidence demonstrating a
nutritional role of host metabolites in developing
endoparasitoids has not be conclusive (Ferkovich and
Dillard, 1986; Thompson, 1986b). However, the successful development (and eclosion) of these parasitoids in metabolically altered hosts suggests that changes in host metabolism do not harm the intruder and that these
insects have probably developed mechanism(s) to alter the host's nutritional status for the benefit of the endoparasitoid's offspring (Vinson and Iwantsch, 1980a).
The ability to regulate host development is thought to be absent in ectoparasitoids. This conclusion has been reached because the venoms from almost all ectoparasitoids that have been studied are paralytic and either immediately kill the host or completely suppress host development (Beard, 1963; Piek and Spanjier, 1986). 6
In addition, since these insects feed externally, their development is considered to be independent of the host's endocrinological and nutritional status. However, recent evidence (Uetmatsu et al., 1987; Coudron et al., 1990) suggests that ectoparasitoids from at least one family
(Eulophidae) can regulate the development of their hosts.
Since few ectoparasitoids have been examined, it seems likely that other species of ectoparasitic insects will be found that alter and regulate host development in ways comparable to endoparasitoids. One such ectoparasitoid that merits study is Nasonia vitripennis (Walker)
(Hymenoptera: Pteromalidae).
N. vitripennis attacks pupae and pharate adults from several families of Cyclorrhapha (Darling and Werren,
1990). During host attack, the female parasitoid examines the host puparium for a brief period of time and then inserts the ovipositor through the host puparium.
Prior to oviposition, the wasp injects venom into the host (Dawei and Dingxi, 1987) , which appears to kill the host within 24 h (Beard, 1964; Ratcliffe and King, 1967).
The parasitoid's eggs are then deposited on the surface of the host's body and they start to hatch 24-36 h later.
Larval development (3 instars) is completed in 5-7 days at 25°C, and the pupae require an additional 4 days before adult eclosion ends the duration of the wasps's lifecycle within the host puparium. 7
N. vitripennis is a gregarious parasitoid and has evolved a mechanism to regulate the sex ratio of its offspring. The mechanism involves the regulation of egg
fertilization so that only a limited number of males are produced. All of the males are brachypterous, thereby preventing them from leaving the site of emergence. This allows the sons to mate with their siblings, which means the wasp adheres to local mate competition (Werren, 1980,
1984a). Many other factors may contribute to the overall sex ratio of a brood, but superparasitism has been the dominant attribute examined in this wasp (Werren, 1980,
1984a; King and Skinner, 1991b). Due to the extensive use of N. vitripennis in sex ratio theory investigations, it would be expected that the biology of this wasp is fully understood. But, the literature is devoid of any thorough works relevant to host effects on clutch size and sex ratio, and no information is available about this parasitoid's potential to alter or regulate host development. Failure to recognize the importance of the latter may explain why this wasp has achieved only limited success in biological control programs aimed at flies (Legner, 1967; Rutz and Scoles, 1989).
The majority of investigations examining the biology of N. vitripennis have employed the house fly, Musca domestica L. (Diptera; Muscidae), as the host. This fly 8
is not the preferred host of |*. vitripennis (Ohgushi,
19 59; Werren, 1983; Darling and Werren, 1990) and is
seldom encountered by this wasp in nature (Rutz and
Scoles, 1989; Axtell and Rutz, 1986). Therefore, the
reports describing the biology or attributes of host
quality for |f. vitripennis must be viewed with caution.
In this study, the flesh fly, Sarcoohaaa bullata Parker
(Diptera: Sarcophagidae), was selected as the host
because it is a natural host of this wasp, the large size
of the pupa (and pharate adults) ensures that parasitoid
nutrition is not compromised, and the physiology of this
fly has been well studied. In addition, observations by
Denlinger (unpublished) have indicated that parasitized
pupae of £». bullata appear diapause-like in morphology,
and this suggests that death may not be the immediate host response following parasitoid attack.
The major objective of this study was to examine the physiological and behavioral interactions between N. vitripennis and its host S. bullata. This was accomplished by focusing on four specific aims. The
first goal was to identify the factors that defined host range for li* vitripennis by evaluating the impact of different fly species and host quality on fecundity, development and sex ratio in this wasp. The second objective was to determine the developmental fate of venom-injected hosts and evaluate the ecdysteroid 9
dependency of this parasitoid. The third aim of this
study was to examine the metabolic reserves and
respiratory metabolism in hosts attacked by vitripennis. The final objective was to evaluate the
role of the wasp's venom in parasitism by examining its
impact on host development and comparing the effects of
the venom on nontarget insect species. CHAPTER I
HOST QUALITY AND ITS IMPACT ON FECUNDITY, DEVELOPMENTAL
TIME, AND SEX RATIO IN NASONIA VITRIPENNIS. A
GREGARIOUS, ECTOPARASITOID OF THE FLESH FLY, SARCOPHAGA
BULLATA
Introduction
The ectoparasitoid Nasonia vitripennis (Walker)
(Hymenoptera: Pteromalidae) must locate a fly puparium
that is buried in soil and then evaluate its suitability
as a host. This is one of the most critical tasks a
female parasitoid must perform (Takagi, 1985). Failure
to recognize differences in host quality can result in a brood being deposited on a host unable to support the parasitoid's clutch, a wasted investment by the wasp that would decrease fitness (Waage, 1986).
N. vitripennis is capable of utilizing pupae from several families of cyclorrhapha (Smith, 1969; Smith and
Pimentel, 1969; Oghushi, 1959; Cornell and Pimentel,
1978; Wladimirov and Smirnov, 1934), but, reports of success (production of reproductive progeny) have been variable. The reasons for this variability are not understood, nor are the factors that govern its host
10 11
range (Smith, 1969). Can host size, the variable
commonly thought to be synonymous with host quality,
account for individual differences in host suitability?
Or, do the qualitative differences among host species
affect their suitability for N. vitripennis? Possibly N.
vitripennis can not discriminate between different hosts
and differential mortality of the progeny may account for
the variability.
Here, we determine if N. vitripennis is influenced
by inter- and intraspecific differences in host quality.
We compare developmental time, sex ratio, fecundity, and
incidence of diapause of this wasp when presented
different species of hosts. In addition, we examine the
effect of different physiological stages of the flesh
fly, Sarcophaaa bullata Parker (Diptera: Sarcophagidae), on this parasitoid.
Materials and Methods
Parasitoid and host rearing: JJ* vitripennis females were collected in Columbus, Ohio from pupae of Calliphora spp.
(Diptera: Calliphoridae) and maintained as a laboratory culture for 2 years on S. bullata. For all experiments, unless otherwise indicated, 25 females of £J. vitripennis
(3-7 days after eclosion from host puparia) were placed in plastic Petri dishes (15 x 100 mm) with 50 nondiapausing pupae of S. bullata (5-6 days after 12
pupariation) and a 1:1 (v/v) honey-water solution for 24
h prior to each treatment. N. vitripennis treated in
this manner are referred to as host-fed. The wasps were
maintained at 25°C, with a daily light-dark regime of
L 1 5 :D 9 .
Laboratory colonies of two North American species
of flesh flies, S. bullata and S. crassipalpis Macquart
(Diptera: Sarcophagidae), and a Panamanian species,
Peckia abnormis (Enderlein) (Diptera: Sarcophagidae),
were maintained as previously described (Denlinger,
1972a). Nondiapausing pupae were produced by rearing
flies at 25°C, L15:D9 throughout development. To produce
diapausing pupae, adults were held at 25°C, L12:D12 and
larvae at 20°C, L12:D12. Musca domestica L. (Diptera:
Muscidae), Drosophila melanoqaster Meigen (Diptera:
Drosophilidae), and Phorus spp. (Diptera: Phoridae) were
obtained from The Ohio State University Insectary, and
Rhagoletis pomonella (Walsh) (Diptera: Tephritidae) was
provided by Dr. D. C. Smith (The Ohio State University);
all were maintained at 25°C, L15:D9.
Pupal diapause in S. bullata was confirmed using the
criteria described by Fraenkel and Hsiao (1968).
Induction of larval diapause in £. vitripennis was
determined using the criteria of Schneiderman and
Horowitz (1958). 13
Exposure of parasitoid to host
Nondiapausing pupae (4 days after pupariation) of
S. bullata. S. crassioalpis. P. abnormis. R. pomonella.
I), melanoaaster. Phorus spp. , or J5* domestica were exposed singly to a single host-fed female of N. vitripennis for 24 h. After exposure, the wasps were removed and the pupae were kept separately in plastic 1 oz. cups (Dixie) and maintained at 2 5°C, L15:D9. The developmental time (from egg to adult emergence from the host puparium), sex ratio, incidence of diapause, and number of progeny produced by JJ- vitripennis were observed for each host. Hosts used in all experiments were treated as above unless otherwise indicated.
Host categories
The influences of host size (length of the puparium) and weight on oviposition decisions and/or development were assessed for N. vitripennis by utilizing s. bullata.
Puparial length was measured with an ocular micrometer and samples were divided into the following categories:
9.0-9.9, 10.0-10.9, 11.0-11.9 and 12.0-12.9 mm. These pupae were weighed and subdivided into the following groups: 51-60, 61-70, 71-80, 81-90, 91-100, 101-110, 111-
120, 121-130, 141-150 mg.
To test whether the curvature of the host puparium is important, distorted puparial shapes were produced by 14
forcing larvae of S. bullata to pupariate inside 1 ml
disposable pipette tips (Eppendorf). Under these
conditions, the larva can not form its characteristic
barrel-shaped puparium, and the resulting puparium is
greatly elongated.
The effect of host age on oviposition and
development was determined by exposing S. bullata puparia
of different ages (1-11 days after pupariation) to wasps.
To facilitate counting eggs deposited by N. vitrioennis.
puparia were wrapped singly in aluminum foil, so that
only the anterior third of the puparium was exposed.
Preliminary experiments revealed that the presence of
aluminum foil does not interfere with the wasp's
oviposition behavior or number of eggs deposited. Each
host was then exposed singly to individual females of JJ.
vitripennis for 24 h. After removing the wasp, the
anterior cap of the puparium was removed, exposing the
eggs laid on the surface of the host's head. The number
of eggs laid on hosts of each age was recorded. Progeny
production and development time of J4. vitripennis was
assessed by using puparia of S. bullata of different ages
as described above, with the exception that the puparia were not wrapped in aluminum foil. The host age experiments were repeated using females of N. vitripennis that had not previously host-fed. 15
Preliminary experiments indicated that very young puparia (<2 days after pupariation) do not yield adult parasitoids. To determine whether oviposition was attempted, puparia of S. bullata were exposed to single unfed female wasps for 4 h at 6 h intervals during the two days following pupariation. Hosts were observed visually and then dissected to search for £1. vitripennis eggs laid either on or in the host.
Evidence for host feeding on these young puparia was also sought by examining the ovarioles of the wasps used in the above experiment. The ovarioles of each wasp were examined 48 h after being presented hosts. Typically, host feeding by N. vitripennis results in a doubling of the number of mature eggs present in the ovarioles 2-3 days later (Edwards, 1954b).
Pupae and pharate adults of S. bullata were killed by freezing to determine if a host must be alive for successful oviposition and development by JJ. vitripennis.
Hosts (1-11 days after pupariation) were killed by exposure to -70°C (Revco) for 1 h. After freezing, hosts were placed at 2 5°C for 2 h before exposure to JJ. vitripennis.
The impact of the host's sex on oviposition and development of fi. vitripennis was observed by exposing male and female pupae of S. bullata to individual females of JJ* vitripennis. Host sex was determined by removing 16
the posterior portion of each puparium 4 days after
pupariation to observe the size of the footpads on the
last pair of legs (male footpads are 2-3 times larger
than the footpads of the female).
The impact of the host's pupal diapause on
oviposition and development of vitripennis was
determined using diapausing pupae of £. bullata.
Diapausing pupae were offered as hosts at 10 day
intervals throughout the 80 days of diapause.
Transfer of eaas to unenvenomated hosts
To test the importance of host envenomation on the development of l{. vitripennis progeny, unenvenomated hosts representing each of the host categories were used.
Hosts were wrapped in aluminum foil and exposed to N. vitripennis females as described above for the experiments evaluating the effects of host age. The anterior cap of each puparium was opened and 15 eggs were transferred to an unenvenomated host of the same age and physiological state (nondiapausing, diapausing, and dead hosts). As controls, eggs were placed on envenomated, nondiapausing hosts to ensure that damage to the eggs did not occur as a consequence of the transfer. The experiment was repeated using eggs from unfed females placed on nondiapausing hosts. 17
Oviposition site preference
To determine if iJ* vitripennis females show a
preference for a particular body region of its pupal
host, pupae of S. bullata (4 days after pupariation) were
exposed to females of fi. vitripennis for 5 h. After
exposure, each puparium was removed from the pupa and the
body region used for oviposition was recorded (head,
thorax or abdomen). For the few cases in which the wasp
oviposited more than once, more than one body region was
recorded for a single host.
Statistical analysis
Data were analyzed using one and two-way analysis of variance, t-tests, alpha= 0.05, Student-Newman-Keuls multiple comparison tests at the 5% level, and linear regression analysis according to Sokal and Rohlf (1969).
Data analysis were performed using CoStat statistical software (CoHort).
Results
Host species
Far fewer parasitoids were produced on pupae of M. domestica than on the three sarcophagid species (Fig. 1A;
F= 15.68; df= 3,139; P < 0.01), development time was longer (Fig. IB; F= 117.1; df= 3,45; P < 0.01), and a higher proportion of males was produced (mean ± SEM = 18
0.41 ± 0.10, on M. domestica. 0.17 + 0.03, on S. bullata.
0.20 + 0.03, on £. crassipalpis. and 0.15 + 0.02, on P.
abnormis. F= 9.46; df= 3,124; P < 0.01).
The diapause incidences in {£. vitripennis reared on
the 4 species at 25°C, L15:D9 were consistently low and
not significantly different (mean + SEM = 0.02 ± 0.01; F=
4.71; df= 3,136; P > 0.05).
Puparia (4 days after pupariation) representing
three additional fly species (R. pomenella. D. melanoqaster. and Phorus spp.) were rejected as hosts by
N. vitripennis. No eggs or adult parasitoids were recovered from these puparia.
Host size and shape
Two of the rejected species (R. pomenella, weight=
13-19 mg and Phorus spp., 11-22 mg) were similar in weight to M. domestica (18-27 mg), yet only M. domestica was acceptable for progeny production by N. vitripennis.
Pupae of J4. domestica are much smaller than pupae of S. bullata (120-132 mg), s. crassipalpis (117-128 mg), and
P. abnormis (144-161 mg). Pupae of S. bullata were used to test the effect of size alone. Between weight extremes of 51 and 140 mg, there was a positive relationship between weight of the host and the number of parasitoids produced (Fig. 2; y= 25.94 + 0.2 3x; P < 0.05; r2= 0.73; SEj3= 0.18). Weight increases of the host 19
beyond 140 mg do not increase the number of parasitoids
produced (Fig. 2).
To test whether weight or length of the host was
more important in influencing the number of parasitoids
produced, hosts were categorized based on weight and
length of the puparium. In the first experiment, host
weight was constant (120-130 mg) but length was either 10
or 14 mm. There was no difference in the number of eggs
deposited or in the number of adult parasitoids emerging
from 10 mm puparia (38.7 + 0.1 eggs/host) and 14 mm
puparia (40.2 + 0.1 eggs/host) (t= 0.23; df= 64;
P > 0.05). In the second experiment, host puparia of the
same length (10 mm) were placed in 4 different weight
categories. The number of parasitoids increased as host
weight increased (F= 0.98; df= 3,128; P < 0.01): mean +
SEM = 23.9 ± 0.1 parasitoids for hosts weighing 51-60
mg, 34.3 + 0.1 parasitoids on 71-80 mg hosts, 39.9 +0.1
parasitoids on 101-110 mg hosts, and 45.1 + 0.1 parasitoids on 121-130 mg hosts.
When hosts of equal weight but with distorted puparia were compared (81-90 mg), there was no
significant difference between the number of parasitoids produced on normally-shaped puparia (mean ± SEM = 31.3 +
0.2 parasitoid/host) and puparia that were elongated by
40% (32.9 + 0.2 parasitoid/host) (t= 0.49; df= 116; P >
0.05). 20
For the range of host weights tested in Fig. 2, no
significant differences were observed in the time
required for parasitoid development (mean + SEM = 13.7 +
0.2 days; F= 2.05; df= 10,79; P > 0.05), proportion of
males (0.20 + 0.02? F= 1.02; df= 10,221; P > 0.05), and
the incidence of diapause in the parasitoid (1.8 + 0.3%;
F= 2.45; df= 10,221; P > 0.05).
Host aae
No eggs were recovered from hosts that had not yet
pupated (<2 days after pupariation), but eggs were
deposited on all other intrapuparial stages of the host,
even on pharate adults that had begun extrication
behavior. Though fewer eggs were laid and fewer adults
were produced on older hosts (8-11 days after
pupariation), these values were not significantly
different from the number of eggs laid (F= 2.72; df=
10,203; P > 0.05) and adult parasitoids produced (F=
1.04; df= 10,545; P > 0.05) by N* vitripennis on younger
hosts (<8 day old puparia, Fig. 3A).
From these results it was not possible to determine whether females attempted to oviposit on very young puparia (<2 days after pupariation) and chose to leave without ovipositing or if the females laid eggs that never developed. Visual observation indicated that no oviposition occurred, and no eggs were detected on hosts 21
until 48 h after pupariation. By that time, the host had
pupated, thus providing space between the puparium
(integument of the third instar larva) and the pupal
cuticle.
The time required for H- vitripennis progeny to
develop was significantly shorter on 2-4 day old puparia
(mean + SEM = 13.2 + 0.4 days) than on puparia >4 days
old (mean + SEM - 14.1 + 0.2 days; F= 16.47; df= 9,422; P
< 0 .01).
For the host ages tested in Fig. 3A, no significant
differences were observed in the incidences of diapause
(mean + SEM = 0.3 ± 0.1; F= 1.67; df= 9,224; P > 0.05) or
in the proportions of males (Fig. 4A; F= 2.71; df= 9,3 51;
P > 0.05).
Host feeding
Unfed females of N. vitripennis laid significantly
fewer eggs than host-fed wasps (Fig. 3A; F= 8.45; df=
9,210; P < 0.01), and consequently fewer adult parasitoids were produced (Fig. 3A; F= 8.32; df= 9, 104;
P < 0.01). Likewise, progeny required a longer time to develop from egg deposition to adult emergence from the host puparium when the female was denied a feeding opportunity (mean + SEM = 14.1 + 0.4 days for unfed wasp progeny and 12.7 + 0.1 for host-fed; F= 16.4; df= 9, 251;
P < 0.01). 22
The proportion of males produced by unfed wasps did
not differ significantly from that of host-fed wasps
(Fig. 4A; F= 1.23; df= 9,147; P > 0.05), nor did the
incidence of diapause in progeny of unfed and host-fed
wasps (mean + SEM = 0.8 + 0.1; F= 2.98; df= 9,192; P >
0.05) .
Evidence of host feeding (visual evidence and
evidence of stimulated egg maturation) was not observed
until N. vitripennis was presented hosts that had
pupariated 48-54 h earlier. The number of mature eggs
observed in the ovarioles of host-fed wasps (mean + SEM =
37.5 + 0.2) was significantly greater than in unfed wasps
(mean ± SEM = 19.0 + 0.5, t= 4.7, df= 92, P < 0.01).
Dead hosts
Hosts that were killed by exposure to -70°C were
less suitable for N. vitripennis than living hosts (Fig.
3). This was evident both in the decreased number of parasitoids produced (Fig. 3C; F= 12.4; df= 10,381; P <
0.01) and in the 2-3 fold increase in the proportion of male progeny (Fig. 4B; F= 17.1; df= 10,344; P < 0.01).
No significant differences were observed in the interval required by N. vitripennis to develop from egg deposition to adult eclosion reared on the 2 hosts (mean
+ SEM = 12.8 ± 6 days; F= 1.77; df= 10,273; P > 0.05) nor 23
in the incidence of diapause (mean + SEM = 1.2 + 0.6%; F=
2.43; df= 10,330; P > 0.05).
Age of the dead host was not important. There were
no significant age-related differences observed in the
number of parasitoids produced (Fig. 3C; F= 2.12; df=
9,268; P > 0.05), developmental time required (mean + SEM
= 13.3 + 0.4 days; F= 1.18; df= 9,300; P > 0.05),
proportion of males produced (Fig. 4B; F= 3.72; df=
9,266; P > 0.05), or in the incidence of diapause (mean ±
SEM = 0.9 + 0.2%; F= 2.90; df= 9,269; P > 0.05). No
parasitoids developed on hosts that had been dead longer
than 48 h.
Sex of the host
Sex of 4 day-old pupae of S. bullata did not result
in any significant differences in parasitoid developmental time (mean + SEM = 12.2 + 0.2 days on male and 12.2 + 0.2 on female hosts, t= 0.0, df= 8, P > 0.05), progeny production (mean ± SEM = 40.1 ± 3.2 parasitoids on male and 4 0.5 ± 3.3 on female pupae, t= 0.10, df= 76,
P > 0.05) or in the proportion of males produced (mean +
SEM = 0.29 + 0.04 on male and 0.21 + 0.03 female hosts, t= 1.60, df= 86, P > 0.05). No diapausing larvae were produced on either sex of pupa in this experiment. 24
Host diapause
Diapausing pupae of bullata were less suitable
for N. vitripennis than nondiapausing hosts (Fig. 3).
Fewer eggs were laid on diapausing pupae (Fig. 3E; F=
8.93; df= 8,492; P < 0.05), fewer adult parasitoids were
produced (Fig. 3E; F= 13.9; df= 8,464; P < 0.01), and a
higher proportion of males was produced (Fig. 4C; F=
18.7; df= 8,558; P < 0.01).
No significant differences were observed in the
developmental time for N. vitripennis reared on
diapausing and nondiapausing hosts (mean ± SEM = 13.7 +
0.3 days; F= 2.53; df— 8,228; P > 0.05). Likewise, the
incidence of diapause did not differ significantly with
the 2 hosts (mean + SEM = 0.4 + 0.1; F= 1.75; df= 8,330;
P > 0.05).
For the wide range in age of diapausing hosts that were tested, no significant differences were observed in the number of eggs deposited per host (mean + SEM= 32.4 ±
0.8 eggs; F= 2.41; df= 7,384; P > 0.05), in the number of adult parasitoids produced (mean + SEM = 29.7 + 0.5 parasitoids; F= 3.48; df= 7,325; P > 0.05), in the developmental time (mean ± SEM = 13.3 + 0.2 days; F=
1.12; df= 7,299; P > 0.05), in the incidence of diapause
(mean + SEM = 1.2 ± 0.1%; F= 1.20; df= 7,274; P > 0.05), 25
or in the proportion of males produced (mean + SEM = 0.2 7
+ 0.04; F= 3.12; df= 7,392; P > 0.05).
Unenvenomated hosts
Fewer adult wasps eclosed from unenvenomated puparia
than from envenomated hosts (Fig. 3B; F= 12 3.2; df=
10,266; P < 0.01), the time required by N. vitripennis to
develop was significantly longer (mean + SEM = 15.3 + 0.2
days on unenvenomated hosts and 13.2 + 0.1 on
envenomated; F= 318.2; df= 10,215; P < 0.01), and no
parasitoids developed on unenvenomated hosts older than 5
days after pupariation (Fig. 3 B ) .
The proportion of males produced (Fig. 4A; F= 3.67;
df= 10,188; P > 0.05) and the incidence of diapause (mean
+ SEM = 0.3 + 0.1%; F= 0.78; df= 10,2 02; P > 0.05) in N. vitripennis did not differ significantly on unenvenomated and envenomated hosts.
When the experiment was repeated using eggs from unfed parasitoids, the same trends as described for host- fed wasp progeny were observed. Fewer parasitoids eclosed on unenvenomated hosts (Fig. 3B; F= 17.8; df=
10,194; p < 0.01), developmental time was longer (mean +
SEM = 15.7 + 0.2 days on unenvenomated hosts and 14.3 +
0.3 on envenomated; F= 23.4; df= 10,257; P < 0.01), and no wasps developed on puparia older than 5 days after pupariation. 26
The proportion of males produced on unenvenomated
and envenomated hosts did not differ significantly (Fig.
4A; F= 1.27; df= 10,201; P > 0.05), and no diapausing
larvae of N. vitripennis were produced.
To test if envenomation exerts its effect on
parasitoid development simply by preventing host
eclosion, eggs from I*. vitripennis were also transferred
to 2 hosts (dead and diapausing) that could not eclose.
Far fewer parasitoids emerged from unenvenomated, dead
(Fig. 3D; F= 15.7; df= 9;226; P < 0.01) and diapausing
hosts (Fig. 3F; F= 7.82; df= 8,265; P < 0.01) than from
envenomated hosts (Fig. 3 B ) . Both dead (Fig. 4B; F=
8.76; df= 9,205; P < 0.01) and diapausing hosts that were
unenvenomated (Fig. 4C; F= 7.30; df= 8,270; P < 0.01)
produced a higher proportion of males than envenomated
hosts (Fig. 4 A ) .
The time needed for parasitoid development did not differ significantly among unenvenomated, dead (mean +
SEM = 13.2 + 0.3 days; F= 1.58; df= 9,210; P > 0.05} or diapausing hosts (mean + SEM = 12.8 + 0.2 days; F= 2.18; df= 8,178; P > 0.05) and envenomated hosts (mean ± SEM =
13.2 + 0.1 days). In addition, the incidence of diapause in N. vitripennis did not differ significantly in unenvenomated, diapausing hosts (mean + SEM = 1.3 + 0.7%;
F= 1.55; df= 8,281; P > 0.05), unenvenomated, dead hosts 27
(mean ± SEM = 1.2 ± 0.1%; F= 2.40; df= 9,195; P > 0.05) and envenomated hosts (mean ± SEM = 0.8 + 0.1%).
Oviposition site preference
The thorax and abdomen were preferred oviposition sites when 4 day-old puparia of £. bullata were exposed to wasps. The pupal thorax was utilized for oviposition
56.4% of the time, a value which did not differ significantly from the abdomen (43.6%), but did differ from the head region (10.9%; F= 219.1; df= 2,53; P <
0.01). No preference was detected for the dorsal or ventral surfaces of the host (47.3% of ovipositions on dorsal surface and 52.7% on ventral, t= 1.24, df= 53, P >
0.05). Fig. 1. (A) Mean ± SEM number of 1{. vitripennis adults
produced per host on 4 species of fly pupae (£. bullata.
£. crassipalpis. £. abnormis. and tf. domestica) (F*
15.68; df* 3,139; P < 0.01). (B) Mean + SEM time (days)
required for £. vitripennis to develop from egg
(deposition) to adult (emergence from host puparia) at
25°C, L15:D9 on pupae from each species of host (F*
117.1; df- 3,45; P < 0.01). 36-64 pupae of each fly host were used for each experiment. Columns preceded by the same letter do not differ significantly at P > 0.05
(ANOVA).
28 vo ro No. ol adults (mean ± SEM) ± (mean adults ol No. Development lime (days) lime Development
F i g . Fig. 2. The relationship between the number of adults of
&. vitripennis produced and pupal weight (mg) of the host, £. bullata. The regression is significant (y-
25.94 + 0.23X, P < 0.05, r2- 0.73, SEfa- 0.18) up to a host weight of 140 mg. Increases in host weight beyond
14 0 mg did not result in an increase in the number of parasitoids produced.
30 No. of adults / host m II* *r •■■o ft-100 Weigh! of (mg) has! Weigh! i. 2 Fig. 30 11 3 11 4 MUM 11 1*0 1*1 M U M 140 131 130 131 0 I3 M 1 31 Fig. 3. (A) Mean ± SEM number of eggs deposited (-e-
host-fed and unfed wasps) and adults of £.
vitripennis produced (-o- by host-fed and -0- unfed
wasps) by host-fed and unfed wasps. (B) Adult eclosion
(%) of Ji- vitripennis from envenomated (-•- host-fed and
-Q- unfed wasps) and unenvenomated (-e- host-fed/ -o-
unfed), nondiapausing puparia of £. bullata. (C) Mean +
SEM number of adult parasitoids produced on hosts of £.
bullata killed by a 1 h exposure to -70°c.
(D) Adult eclosion (%) of {J. vitripennis resulting from
egg transplantion to unenvenomated, dead hosts.
(E) Oviposition -e- and progeny production - o - by vitripennis on diapausing pupae of £. bullata. (F)
Adult eclosion (%) of Jf* vitripennis resulting from egg transfer to unenvenomated, diapausing hosts. These experiments were replicated twice using 2 3-46 puparia in each replicate for each age and physiological state of the host.
32 33
i *
f * I « * m
Fig. 3 Fig. 4. (A) The proportion of males produced by host-fed
-e- and unfed -o- vitripennis females in 24 h on
nondiapausing puparia of £. bullata. (B) Proportion of males produced on dead puparia and unenvenomated, dead puparia -o-, (C) Proportion of males produced on diapausing puparia -o- of £. bullata and unenvenomated, diapausing puparia . Each experiment was replicated twice using 2 3-46 puparia in each replicate.
34 Proportion itm ol Im {t SEM) O o e o
i a H- (Q I ? a-
fr
u ut 36 Discussion
Host suitability for N. vitripennis is dependent
upon the species and physiological state of the host.
The impact of host species on this parasitoid is evident
in the higher production of progeny and the more rapid
developmental time by N. vitripennis on sarcophagid pupae
(S. bullata. S. crassipalpis. and £. abnormis) than on M. domestica (Fig. 1A & B), and by the rejection of puparia
of R. pomonella. Phorus spp., and D. melanoqaster as hosts. Parasitoid production did not differ significantly among the three sarcophagid species, thus suggesting that these hosts are equally suitable for N. vitripennis.
Can host size differences alone account for differences in host species suitability? Pupae of R. pomonella and Phorus spp. are nearly the same weight as pupae of M. domestica. yet, eggs and parasitoids were recovered only from puparia of H. domestica. In addition, by extrapolation of the regression equation in
Fig. 2, S. bullata pupae similar in weight to JJ. domestica would be expected to produce twice as many parasitoid adults as observed in &. domestica. Thus, variability in the success of M- vitripennis on different species of hosts is not due entirely to host size.
M- vitripennis can detect size differences in the same species of host and regulate the number of eggs laid accordingly. More progeny were produced as the weight of 37
the host increased up to a maximum host weight of 140 mg
(Fig. 2). Beyond 140 mg, there is no increase in the
number of parasitoids produced, and this at least
partially accounts for the fact that heavy pupae of P.
abnormis do not yield more parasitoids than the somewhat
smaller sarcophagid species (Fig. 1A). These results
confirm the observations on Ji. vitripennis by Edwards
(1954a) and Wylie (1967) that "large” hosts produce more offspring than "small,” and are consistent with the
findings for several other parasitic wasps (Purrington and Uleman, 1972; Heinz and Parrela, 1989; Takagi, 1986;
Bai et al., 1992; Klomp and Teerink, 1962; Salt, 1961,
Hardy et al., 1992).
Puparial length of £. bullata was not a cue used by this wasp to estimate host quality. We observed that puparia of the same weight, but of different lengths, received the same number of eggs and produced an equal number of progeny of N. vitripennis. In addition, the curvature of the host puparium was not a factor influencing this wasp's oviposition decision. This was demonstrated by comparing oviposition on elongated
(prevented from contracting longitudinally) and normally contracted puparia of equal weight. No difference was observed in the number of eggs laid or adults that developed on either type of pupa. These results are compatible with the suggestion by King and Skinner 38
(1991b) that sex ratio and clutch size of N. vitripennis
are independent of the physical attributes associated
with the host puparium.
This study was not designed to determine how N.
vitripennis distinguishes between hosts of differing
sizes, yet, our observations suggest that the ovipositor
of the wasp may be utilized in making these assessments.
Curvature and length of the host puparium did not
influence the number of eggs deposited by vitripennis.
but weight of the host was important. Host weight could
be assessed by several methods, by estimating the
exposed surface area of the host puparium or by cues
detected by insertion of the ovipositor through the
puparium. If surface area of the host were important,
length rather than weight would be the feature likely to
be measured by the wasp. Since surface area measurements
do not appear to be important (Wylie, 1967), we suspect
that weight cues are detected by information impinging on
the ovipositor.
If insertion of the ovipositor through the host puparium is required for host discrimination, must the
ovipositor also penetrate the pupal cuticle? Saunders et al. (1970) have reported that "large" pupae of S. argvrostoma f=barbata) have a thicker puparium than
smaller pupae, and they suggest that N. vitripennis can distinguish between hosts based on puparium thickness. 39
This argument, however, can not explain the ability of N.
vitripennis to distinguish between S. bullata puparia of
different ages or dead hosts (Fig. 3A & C). After
pupation, puparium thickness does not change. Though the
puparium thickness may contribute to this decision, it is
clear from these results that information is also derived
from the pupa itself.
After pupation has occurred, all ages of pupae and
pharate adults are suitable hosts, even late pharate
adults that have already initiated extrication behavior.
Before pupation (during the 2-day interval between
puparium formation and pupation) the hosts were not
acceptable for feeding or oviposition. In contrast,
puparia of M. domestica are reported to become less
suitable for N. vitripennis oviposition and progeny production as they get older (Wylie, 1962, 1963; Chabora and Pimentel, 1966). Edwards (1954a) observed that very young puparia (<24 h old) of M. domestica are suitable only for host feeding and not oviposition because of a lack of "space" between the puparium and pupa to lay the eggs. In our experiments with sarcophagid hosts, the presence of this space was essential for both oviposition and feeding; neither occurred earlier than 48-54 h after pupariation of the host.
vitripennis can discriminate between different developmental stages of the same host. Nondiapausing 40
pupae received the largest number of eggs and produced
more offspring than other S. bullata puparia (Fig. 3).
Diapausing and dead puparia yielded only about half the
number of parasitoid adults as nondiapausing hosts (Fig.
3C & E). Edwards (1954a) reported that this wasp does
not utilize dead pupae, but in our experiments dead hosts
were used. This conflict is probably due to the method
used to kill the hosts or how soon the hosts were used
after death. N. vitripennis did not accept S. bullata
puparia that had been dead for more than 48 h.
Eggs from vitripennis transferred to
unenvenomated, nondiapausing puparia did not develop to
adults if the host was older than 5 days after
pupariation, yet, every age of host was suitable for
progeny development if the hosts were envenomated (Fig.
3B). Similarly, unenvenomated diapausing and dead puparia produced adult wasps on every age of host (Fig.
3D & F), but the percent eclosion of I*. vitripennis was
15-50% less than from envenomated, nondiapausing hosts of comparable age (Fig. 3B). Thus, venom injection greatly enhances successful development of parasitoid offspring on suitable hosts, and in nature, females never oviposit without first injecting venom.
Host feeding has a great impact on fecundity of the wasp. Host-fed females deposited twice as many eggs as unfed wasps (Fig. 3A). The lower egg deposition of unfed 41
females could be due to egg resorption, which is known to begin in unfed N. vitripennis as early as 2 days after adult emergence (Edwards, 1954b; P. King, 1962, 1963;
Flanders, 1935) . However, a higher proportion of males would be expected from unfed wasps than from host-fed wasps if resorption were the cause (King, 1962), but this was not observed (Fig. 4A). The more plausable explanation is that eggs mature faster after the female has fed (Edwards, 1954b).
Host species and physiological stage of the host had a great influence on the time required by N. vitripennis to complete development. At 25°C, parasitoid offspring reared on young (2-4 days after pupariation) sarcophagid hosts required 12-13 days to develop from egg deposition to adult emergence (from the host puparium).
Developmental time was lengthened on older puparia (>4 days after pupariation) of S. bullata. dead, diapausing or unevenomated hosts, or on puparia of M. domestica.
Changes in the host's physiology following envenomation may favor the development of this parasitoid's progeny on some hosts, but not others. Such changes have been demonstrated for a number of hosts used by endoparasitoids (reviewed by Vinson and Iwantsch, 1980b), and our preliminary results (unpublished) suggest that a number of important biochemical changes occur in hosts of
E* vitripennis following envenomation. 42
Charnov (1982) suggests that a small host is more
advantageous for the development of a male parasitoid
than a female. Numerous investigations support this
contention (Van Den Assem et al., 1984; Charnov, 1979;
Purrington and Uleman, 1972; B. King, 1991a; Werren,
1984b; Clausen, 1939). But, sex ratio of N. vitripennis
was not altered by host weight, length, age, or sex.
Females of this wasp did, however, produce a
significantly higher proportion of males on diapausing
and dead pupae (Fig. 4B & C), suggesting that these hosts
are more suitable for production of male parasitoids.
Our results using dead hosts contrast with the findings
of King and Skinner (199lb) who reported that use of dead
hosts altered clutch size but not the sex ratio of M . vitripennis.
The incidence of diapause in larvae of £* vitripennis was not altered by our manipulations. Under the conditions of this study, fewer than 4% of the progeny of N . vitripennis entered diapause, regardless
of the host, and thus diapause incidence did not reveal differences in host quality. The use of more stongly diapause-inducing conditions (Schneiderroan and Horowitz,
1958; Saunders, 1965) may reveal differences that were not detected in this study.
Our results provide insight into two important aspects of biological control efforts using $1. 43 vitripennis. First, maximum propagation of this wasp can be achieved by using large (120-140 mg), young, nondiapausing hosts from a sarcophagid species. Second, our results suggest tactics for parasitoid release.
Current efforts using fi. vitripennis are directed toward controlling muscoid flies (Axtell and Rutz, 1986; Rutz and Scoles, 1989). An inoculative release of this wasp would likely prove ineffective since our results show that this wasp produces few adult parasitoids on the museid M* domestica. and of those progeny produced, a high proportion are males. Periodic introductions of female-biased populations of N. vitripennis may be a more effective release strategy for biological control of flies. Such release practices would help increase the number of females in natural populations that are available to parasitize muscoid puparia. CHAPTER II
DEVELOPMENTAL PATE OF THE FLESH FLY, BARCOPHAGA BULLATA.
ENVENOMATED BY THE PUPAL ECOTPARASITOID, NASONIA
VITRIPENNIS
introduction
Maximium utilization of a host is critical for
survival of a parasitic insect, especially if the host
attacked is a finite resource (i.e. a nonfeeding h o s t ) .
Endoparasitic insects have evolved the capacity to
regulate their host's nutritional and/or endocrinological
status (Vinson and Iwantsch, 1980; Beckage, 1985), thus
enabling these parasitoids to overcome the limitations
posed by nonfeeding hosts. In contrast, ectoparasitoids
are generally thought to lack the ability to regulate host
development. This conclusion has been reached for two
reasons: i) venoms from almost all the ectoparasitoids
that have been examined are paralytic and either
immediately kill the host or completely suppress host development (Beard, 1963; Piek and Spanjier, 1986), and
ii) although relatively few ectoparasitic insects have 44 45
been studied, only representatives from one family,
Eulophidae, appear capable of regulating the physiology
of their host (Shaw, 1981; Uetmatsu and Sakanoshita,
1987; Coudron et al., 1990).
The gregarious, ectoparasitoid Nasonia vitripennis
(Walker) (Hymenoptera: Pteromalidae) attacks pupae from
several families of the higher Diptera (Darling and
Werren, 1990) and injects venom at the time of oviposition (Beard, 1964; Dawei and Dingxi, 1987). Hosts have been reported to die immediately (within 24 h) following envenomation (Beard, 1964; Ratcliffe and King,
1967), thus host regulation by this wasp seemed unlikely.
But, our recent observations (Rivers et al., 1993) suggest that the venom of N. vitripennis is nonparalytic and development of envenomated hosts is either arrested or delayed. The importance of host arrestment to the parasitoid is not known.
The goal of this study was to determine the developmental fate of pupae and pharate adults of the flesh fly, Sarcophaaa bullata Parker (Sarcophagidae), that were envenomated by JJ* vitripennis at different stages of development. We separately evaluated the roles of envenomation and feeding by the parasitoid larvae in eliciting the host's response. The parasitoid was experimentally offered select body regions of intact hosts and portions isolated by ligation to evaluate 46 potential differences in responses of the parasitoid and host. In addition, we tested whether 2 0-hydroxyecdysone could "rescue" hosts from the effects of envenomation- induced arrest and permit resumption of development.
Materials and Methods
Parasitoid and host rearing
$J. vitripennis females were collected in Columbus,
Ohio from pupae of Calliphora spp. (Diptera:
Calliphoridae) and maintained as a laboratory culture for
3 years on S. bullata. For all experiments, 25 mated females of vitripennis (3-7 days after emergence from host puparia) were placed in plastic Petri dishes (15 x
100 mm) with 50 nondiapausing pupae of S. bullata (5-6 days after pupariation) and a 1:1 (v/v) honey-water solution for 24 h prior to each treatment. The wasps were maintained at 25°C, with a daily light-dark regime of L 1 5 :D 9 .
Laboratory colonies of the flesh fly, S . bullata. were maintained as previously described (Denlinger,
1972). To generate nondiapausing pupae, the flies were held at 25°C, L15:D9 throughout development. Diapausing pupae were produced by rearing adults at 25°C, L12:D12 and larvae at 20°C, L12:D12. Pupal diapause in S. 47 bullata was confirmed using the criteria described by
Fraenkel & Hsiao (1968).
Exposure of host to parasitoid
Unless otherwise indicated, puparia of £. bullata were wrapped in aluminum foil so that only the anterior third of each puparium was exposed to the wasps. This restricted the site of parasitoid deposition and permitted easy removal of the parasitoid eggs from the host. Each host was exposed to a single female of N. vitripennis for 6 h in a plastic 1 oz. cup (Dixie).
After host exposure, the adult wasps were discarded, the operculum of each puparium was removed, and the parasitoid#s eggs were removed from the surface of the host integument. Each pupa was then kept separately in a plastic 1 oz. cup and maintained at 25°C, L15:D9. The numbers of hosts that entered developmental arrest, deposited eye pigment, formed body bristles, eclosed, and died following envenomation were recorded.
Transfer of ecras to unenvenomated hosts
To determine if the disruption of host development was due to envenomation or to the presence of parasitoid larvae, development of N. vitripennis was monitored on unenvenomated and envenomated pupae. Hosts were wrapped in aluminum foil and exposed to E* vitripennis females 48
for 24 h. The anterior cap of each puparium was opened,
and 15 eggs were transferred to an unenvenomated or
envenomated host of the same age and physiological state
(diapausing or nondiapausing pupae). Pupae were maintained at 25°C, L15:D9 and examined 2-3 days later
for the development of eye pigment or body bristles (in the presence of parasitoid larvae, the host can be completely consumed within 5-7 days after oviposition by the w a s p ) .
Site of envenomation
To determine if the developmental fate of envenomated hosts is influenced by the body region attacked, different regions of £. bullata were exposed to
N. vitripennis. Puparia (4-7 days after pupariation) of
S. bullata were partially wrapped in aluminum foil so that either the anterior or posterior half of each puparium was accessible to Jf* vitripennis. Preliminary experiments demonstrated that aluminum foil does not interfere with the wasp's ability to oviposit on the unwrapped portion of the puparium or influence the number of eggs deposited.
Host ligations
To determine if parasitoid development is likely to be dependent on a certain region of the host's body, 49
diapausing (20 days in pupal diapause) and nondiapausing
(4-7 days after pupariation) fly hosts were ligated with
cotton thread to produce either head-thorax or thorax-
abdomen components. The turgor of the pupal body
precludes the use of both the anterior and posterior
components following ligation, thus only one preparation
could be obtained from one individual. The half to be
sacrificed was first punctured with an insect pin before
the 1igature was pulled tight. Each ligated host was
exposed to a single female parasitoid for 6 h, and then
each host was kept separately in a plastic l oz. cup and
held at 25°C, L15:D9. The number of adult progeny
produced was recorded for each ligated host.
Hormone treatments
To determine if the arrest of host development
following envenomation was due to an ecdysteroid deficiency, envenomated S. bullata were injected with 20- hydroxyecdysone (Sigma). After removal of the parasitoid's eggs from the host's head, each envenomated pupa (4 days after pupariation) was pricked with an insect pin (# 3) and 1-2 ul of hemolymph was removed on filter paper to provide space for injection. Using finely drawn glass capillaries, 1 ul of a 20- hydroxyecdysone solution (0.5-5 ug in 10% ethanol) was injected into the dorsal surface of the host's thorax. 50
Pupae injected with 10% ethanol served as controls.
After injections, pupae were placed on filter paper in plastic Petri dishes and maintained at 25°C, L15:D9.
Pupae were examined daily for eye pigment deposition, body bristle formation, eclosion, or death. Envenomated, diapausing pupae of £. bullata (20 days in diapause) were also injected with 20-hydroxyecdysone as described above.
In a second set of experiments, nondiapausing pupae
(4 days after pupariation) were ligated 60 min after envenomation and isolated abdomens were then injected with a 1 ul solution of 20-hydroxyecdysone (0.5-5 ug in
10% ethanol) or 10% ethanol (controls).
The juvenile hormone analog, methoprene (ZR-515,
Zoecon), was topically applied to diapausing (20 days in diapause) and nondiapausing (4 days after pupariation) pupae either before or after envenomation. Pupae received a 1 ul topical application of a methoprene solution (1-20 ug in acetone) on the dorsal surface of the thorax. 51 Results
Developmental response of nondiapausing hosts to envenomation
Under natural conditions, all hosts used for feeding or oviposition by li. vitripennis are first injected with venom. Parasitoid larvae typically consume the host within 5-7 days after oviposition by the wasp, thus the parasitoid eggs were removed from the host to separate the effects of envenomation from the effects of larval
feeding.
Very young hosts (<2 days after pupariation) were not parasitized and thus >80% of the flies eclosed 11-12 days after pupariation. All other ages of hosts were attacked at a high rate, as indicated by an envenomation incidence exceeding 80% (Fig. 5).
Envenomated hosts either died immediately or entered a developmental arrest. Most young hosts (2 days after pupariation) and late pharate adults (>9 days after pupariation) died within 48 h following envenomation
(Fig. 5). In contrast, <15% (F= 29.6; df= 10,23; P <
0.01) of envenomated hosts of all other ages were dead by
12 days post-envenomation (Fig. 5). Host death was intially determined by comparing oxygen consumption rates
(Scholander respirometer) of envenomated hosts with pupae killed by freezing (2 h exposure to -70°C, unpublished).
The decrease in oxygen consumption rates correlated with 52
visual changes in appearance of the host (necrosis and
desiccation), and thereafter the visual appearance of the
host was used as the criterion for death.
For nondiapausing hosts of an intermediate age (3-9
days after pupariation), the dominant response was an
arrest or retardation of development. Among young,
nondiapausing hosts that were still in the true pupal
stage (3 days after pupariation), development was
consistently arrested by envenomation and the hosts
retained their pupa-like morphology until death.
Similarly, older pharate adults (8-11 days after
pupariation) did not continue to develop following
envenomation, and this was true even of pharate adults
that had begun extrication behavior (expansion of the
ptilinum). Among hosts of intermediate age (4-7 days
after pupariation), pharate adult development proceeded
for several days, but it progressed very slowly.
Unenvenomated hosts (4-6 days after pupariation) deposited eye pigment within 1-3 days, but pigment deposition in envenomated hosts was delayed by 4-6 days
(F= 9.87; df= 5,231; P < 0.01). Bristle formation was also delayed by 5-8 days in these envenomated hosts (F=
7.62; df= 5,274; P < 0.01). Development was not only retarded in envenomated hosts but was also less fully expressed. For example, the compound eyes of envenomated pharate adults remained light orange/pink and never 53
developed the deep red color observed in unenvenomated
pharate adults. Though some hosts envenomated 4-7 days
after pupariation were capable of depositing eye pigment
and forming body bristles, none of these hosts were able
to complete pharate adult development.
The duration of developmental arrest was dependent
on host age; >75% of hosts envenomated 3-9 days after
pupariation were in delayed development 12 days later,
while <30% (F= 48.2; df= 10,728; P < 0.01) of very young
hosts (<3 days after pupariation) or late pharate adults
(>9 days after pupariation) remained in developmental arrest by the 12th day. Hosts not in developmental
arrest by the 12th day (after envenomation) had either died or were not envenomated and eclosed 11-12 days after pupariation.
Developmental response of diapausing hosts to envenomation
Like nondiapausing hosts, a high percentage (>86%, n= 87) of diapausing pupae (20 days in pupal diapause) were envenomated when exposed to H* vitripennis.
Diapausing pupae did not succumb to the pathology associated with envenomation as quickly as nondiapausing hosts: 92 + 5.3% (mean + SEM, n= 87) of the hosts were still alive 40 days after envenomation (F= 58.0; df=
11,25; P < 0.01), but none of the diapausing pupae that 54 were envenomated succeeded in terminating diapause and
intiating pharate adult development.
Impact of parasitoid larvae on host development
The number of nondiapausing, envenomated hosts
(envenomated 5 days after pupariation) that developed bristles in the presence of parasitoid larvae did not differ significantly from envenomated hosts that did not have a parasitoid load (mean + SEM = 20.5 + 1.4% with larvae and 27.2 + 2.5% without larvae; F= 2.13; df= 1,5;
P > 0.05). Likewise, bristle development of unenvenomated hosts was not significantly different in the presence or absence of parasitoid larvae (mean + SEM
= 98.3 + 0.9% with larvae and 100% without larvae; F=
0.71; df= 1,5; P > 0.05), however, far fewer envenomated hosts formed body bristles than unenvenomated hosts (F=
14.76; df= 3,8; P < 0.01). No bristle development was detected for any of the diapausing hosts, with or without parasitoid larvae present.
Site of envenomation
Nondiapausing hosts of S. bullata (4-7 days after pupariation) differed in their response to envenomation depending on whether the anterior or posterior half of the body was exposed to !?• vitripennis. Hosts envenomated in the anterior half of the body were less likely to deposit eye pigment than hosts attacked in the
posterior region (Fig. 6). Likewise, the duration of
host arrestment (prior to death) was much shorter in
hosts envenomated in the anterior region of the body
(mean + SEM = 4.7 + 1.3 days) than in posteriorly-
envenomated hosts (15.1 + 0.3 days; F= 99.0, df= 1,310; P
< 0.01), and the incidence of necrosis (darkening of the
host integument that was not due entirely to
melanization) noted on the 5th day after envenomation was
much higher in anteriorly-envenomated hosts (mean ± SEM =
53.6 + 2.1%) than in hosts envenomated posteriorly (30.8
± 4.5%, F= 37.1; df= 1,5; P < 0.01). Eventually, all
envenomated hosts became necrotic before death.
Hosts (5-7 days after pupariation) that were
posteriorly-envenomated did not differ significantly from
unenvenomated hosts in the number of individuals depositing eye pigment (Fig. 6) or in the time required
for eye pigment deposition (mean + SEM = 2.6 + 0.5 days
for posteriorly-envenomated hosts and 2.2 + 0.3 days for controls; F= 2.02; df= 7,291; P > 0.05). But, far fewer hosts of the same age that were envenomated anteriorly deposited eye pigment (Fig. 6), and the onset of eye pigment deposition was significantly delayed in relation to controls (mean + SEM = 8.6 + 0.4 days for anteriorly- envenomated hosts, F= 67.4; df= 7,301; P < 0.01). 56
Age-related differences were observed in the host's response to envenomation in different body regions. Far fewer 4-5 day (after pupariation) hosts envenomated anteriorly deposited eye pigment or formed body bristles than did the older hosts (Fig. 6; F= 18.6; df= 3,240; p <
0.01). In addition, 4-day (after pupariation) hosts envenomated anteriorly required a longer time to deposit eye pigment (mean + SEM = 16.1 + 3.3 days) than older hosts (5-6 days after pupariation, 5.1 + 0.8 days; F=
96.8; df= 2,306; P < 0.01). When hosts were posteriorly- envenomated, fewer 4-day hosts deposited eye pigment than the older hosts (Fig. 6), and the effects of envenomation on eye pigment deposition and bristle formation decreased with increasing ages of hosts (5-7 days after pupariation).
Progeny production on ligated hosts
On intact nondiapausing hosts, there was no significant difference in the number of progeny produced when oviposition was restricted to the head-thorax or the thorax-abdomen (Table 1). When diapausing hosts were used, progeny production was lower, but again there was no difference as a result of body region used for oviposition. In contrast, H. vitripennis produced far fewer adult wasps on head-thorax preparations isolated by ligation than on thorax-abdomen preparations (Table 1). 57
Though the difference was greater for nondiapausing
hosts, the effect was also evident in diapausing hosts.
Parasitoid production on isolated thorax-abdomen
components did not differ significantly from production
on unligated hosts parasitized in either the head-thorax
or thorax-abdomen (Table 1).
Response of envenomated hosts to 20-hvdroxvecdvsone
Injection of 20-hydroxyecdysone into envenomated pupae did not cause the hosts to resume normal development (Table 2). All intact, unenvenomated nondiapausing hosts injected with either 2 0- hydroxyecdysone (0.5-5 ug in 10% ethanol) or 10% ethanol developed bristles, but only 24-50% of the envenomated hosts injected with the hormone or 10% ethanol formed bristles, and there was no evidence of a dose response
(Table 2). Among the diapausing hosts, the unenvenomated pupae responded in a dose-dependent manner to 20- hydroxyecdysone injection by the termination of diapause and formation of body bristles, but envenomated diapausing pupae failed to respond (Table 2).
A similar response was observed with isolated thorax-abdomens (isolated 60 min after envenomation).
None of the envenomated thorax-abdomen preparations from nondiapausing or diapausing pupae responded to 20- hydroxyecdysone by resuming development (Table 2). In a parallel set of experiments, 1-20 ug of the juvenile hormone analog methoprene was applied topically in 1 ul of acetone to all the above treatment groups. As with 20-hydroxyecdysone, there was no evidence that methoprene could rescue any of the envenomated hosts, isolated abdomens, or isolated heads from developmental arrest (data not shown). 59
Table 1. Fecundity of H* vitripennis on restricted
body regions of its pupal host, £. bullata.
Body rsgion Boat darvslopaaneal itusbsr at parasitoid
ttpesad sutua n progany/bost (X + SEM) Body intacc, rastrictad oviposition
baad-tborax nondiapausing 90 15.2 ± 0.7* Diapausing • 1 7.3 ± 0.1b tborajt-abdoaan Mondiapaus ing 94 17.7 ± 1.9S Diapausing 72 C.S * 0.9b Xsolatad body ragion hsad-chorax nondiapausing 92 3.2 + 0.3C Diapausing 75 a.5 * 0.5be tiio ras-abdossn Nondiapausing S2 20.0 * I.2a Diapausing <9 9.2 - 2.3b
ovifeaaisiaa feyx« s b c i m u i fed• M 4i(!,Mnt fefeCy r a f i w * at i n t a c t i M U fey n w * i a t Ska poyiriw (rn aw kalf •( feaCy Mt «a *• itlliM. x. xi££lMaaia *1U m tvuapt •viHtttiM n a feaat aaay rtfln H M U f ut M«riaB. :h ;«m ta*d- ckaraa 5t ’jianx-iMNM »w » >n,»rK u a aiallar fefeiniar, tlu aka n r u i u M raalk ■vifeaan anly aa aaa kalf at *ja u n M a y . Saatt u f i r u u c waa l a p U M U i Stw at aaa wttt U-IO Matt ia«a tar ••«=* rapitaaca. ralaaa U a w l a n fallaaa« fey ska aaaa larxar 4a aaa tfiftar tiautiauiUy (raw aaoa ackar it t » a.01 [One-nay w o n , lakol • a *aai;, :> £. bullata after envenomation by £. vitripennis. % Boat dovelapaant to tha briatla stags (lata pharata adult) Ooaa li and la pa using Diapausing (ug/pupa) Control mvanoaitad Control fnveimaatert Response of tha whola body 0 100* 33.3c Od Od 0.9 •9.9a 3S.7C 33.«c Od 1.0 looa 31.3c 90.7b 0d 9.0 looa 34. 0C 100a Od HaspoAsa of isolatad thorax-abdoaana 0 10OS 47.0b Od Od 0.9 100a 90.3b 14. 3C Od 1.0 lOOa 39.7b 39.4C od 9.0 100a 44.3b 09. 9b Od 10'hydrexyNdywM (20-8X) vaa praparad In 10% athanal and in)acted into U a dorsal surfscs of taa thoru of anvaneaacad pupaa (* daya aftar pupariation). Controls vara pupaa that had not boon anvaneaacad, but did racaivs an injaction of 20-KZ. In tha ligation axpsrioants, pupaa wara ligatad «0 sin aftar anvanoaatien so that tha postarior half survived. Kxperlsants wara rap lies tad 1 tines with is pupaa us ad par replicate. Percentages wara norsalisad by arcsine transformation bafora being analytad by AHOVA (Jokal and Rohlf, 1***) • Values in tha i i m row or coluon followad by tha aaaa 1 attar do not diffsr significantly at P > 0.09 (Student-Nevaan-Keul’* nulitpla coaparisona tasts, sokal and Rohlf, 19*9). Fig. 5. The developmental fate of £. bullata hosts envenomated by M. vitripennis on different days after pupariation. Stages include prepupae (1 day after pupariation), pupae (2-4 days after pupariation) and pharate adults (5-11 days after pupariation). Each point represents the mean ± SEM % response (death, developmental arrest, or unenvenomated) of £. bullata 48 h after exposure to £. vitripennis. Host responses on the 12th day after exposure to vitripennis did not differ significantly from those observed at 48 h (F* 2.03; df- 21,1067; P > 0.05). Each experiment was replicated three times using 26-43 hosts in each replicate. Hosts were maintained at 25°C, L15:D9 following envenomation. 61 Fata ol hoti imaao : sam %) too 20 0 ■ 40 •o o i ao g ol hosi {daysAgo alter pupuiiairon) i. 5 Fig. unenvanomaied Fig. 6. The developmental fate of four different ages of £. bullata hosts envenomated in the anterior (A) or posterior (P) half of the body by £. vitripennis. Stages of the host include pupae (4 days after pupariation) and pharate adults (5-7 days after pupariation). The progression of pharate adult development in envenomated hosts was assessed by observing the deposition of eye pigment and formation of body bristles. All unenvenomated hosts (4-7 days after pupariation) deposited eye pigment and developed bristles. The deposition of eye pigment always preceded bristle formation, but bristles never developed on a host if eye pigment had not first been deposited. Each experiment was replicated three times using 15-24 hosts in each replicate. Hosts were maintained at 25°C, L15:D9 following envenomation. 63 Developmentalfate of host (%) M Devaiopmentai Arrest B 3 Eye Pigment Deposition Deposition Pigment Eye 3 B Arrest Devaiopmentai M e g A of host (days after pupariation) i. 6 Fig. D C Brittle Formation Brittle 64 65 Discussion All flesh fly hosts used by H* vitripennis for host feeding or oviposition were envenomated. In fact, host envenomation appears to be essential for the successful development of the parasitoid's offspring (Rivers and Denlinger, 1993). If the hosts were either very young (2 days after pupariation) or near the end of pharate adult development (>9 days after pupariation), they died quickly (within 48 h) in response to envenomation, but hosts of intermediate ages (3-9 days after pupariation) entered a developmental arrest in response to envenomation. The halt in host development appears to be due to an arrestment factor(s) injected by the female wasp and is not due to the presence of parasitoid larvae. The arrestment factor is most likely the venom itself since the effects of natural envenomation are identical to those produced by injection of isolated venom into pupae of S. bullata (Rivers et al., 1993). Host age influenced the response of nondiapausing hosts to envenomation. Hosts that were still true pupae when envenomated (2-3 days after pupariation) remained pupa-like in morphology and never initiated pharate adult development. Morphologically, such pupae appeared to be •4 in a "diapause-like" state. And, when diapausing pupae were envenomated by 1{* vitripennis. the hosts remained 66 locked into diapause. Developmentally-suppressed diapausing hosts appeared to be "preserved" and did not show signs of deterioration (e.g. desiccation and necrosis) for up to 40 days after envenomation. Unlike pupae, pharate adults that were envenomated showed some progression in development (eye pigment deposition and body bristles formation), but the onset of these events was delayed by several days. None of the hosts were capable of completing pharate adult development, and they remained developmentally arrested until death, a characteristic shared with other hosts parasitized by parasitic Hymenoptera (Shaw, 1981; Coudron et al., 1990; Pennacchio et al., 1992; Dover et al., 1987; Lawrence, 1988; Webb and Dahlman, 1985; Beckage and Templeton, 1986). The fact that development was arrested in response to envenomation suggested the possibility of an ecdysteroid deficiency. Pupal diapause in flesh flies is the consequence of such a deficiency (Zdarek and Denlinger, 1975), and several days of a sustained high ecdysteroid titer are needed for the completion of pharate adult development (Denlinger, 1981). But, if this were the cause of the host arrestment observed in response to envenomation, we should be able to elicit development in envenomated hosts with an injection of 2 0- hydroxyecdysone. Such attempts failed and we conclude 67 that the suppression of development is not caused by an ecdysteroid deficiency. The body region used for injection of venom by 2*. vitripennis had a great impact on host arrestment. Far fewer hosts envenomated anteriorly (head-thorax) deposited eye pigment or formed body bristles, the time required for both developmental events to occur was much longer, and the duration of developmental arrest (prior to death) was shorter in these hosts than in hosts envenomated in the thorax-abdomen. These results suggest that the arrestment factor(s) is more deleterious to the host when injected in the anterior half of the body. Thus, if host arrestment is essential for the development of parasitoid progeny, N. vitripennis would be expected to oviposit more frequently in the thorax or abdomen of the host than in the head region, and we have observed that this is indeed the case (Rivers and Denlinger, 1993). The region of the host body that was available to the parasitoid influenced the number of wasp progeny produced. 1J- vitripennis produced far fewer adult parasitoids on isolated head-thorax preparations than on isolated thorax-abdomen preparations or on unligated hosts (Table 1). These observations suggest that optimal development of the parasitoid is dependent on some aspect of the posterior end of the host and that the receptor(s) 68 for the arrestment factor is not in the anterior half of the host. In addition, our results show that far fewer hosts envenomated anteriorly progressed pharate adult development, and the onset of necrosis and death was more rapid in these hosts than in flesh flies attacked in the posterior region of the host body. In larvae of Trichoplusia ni parasitized by the ectoparasitoid Euplectrus plathvpenae (Hymenoptera: Eulophidae), the arrestment factor appears to affect epidermal tissue in the thorax and abdomen (Coudron et al., 1990). A similar response may be operating in the flies envenomated by N. vitripennis. The importance of host age on suppression of development suggests that this may be a factor affecting host suitability for JJ. vitripennis. In fact, the ages of hosts most likely to enter developmental arrest following envenomation are identical to the host ages that are most suited for adult parasitoid production (Rivers and Denlinger, 1993). These results also suggest that differences in the ability to enter developmental arrest may help explain why some fly species are more suited than others for progeny production by this wasp (Ohgushi, 1959; Cornell and Pimentel, 1978). Variation in response among different host species may be attributed to differences in the duration of the intrapuparial stages. For example, the interval from puparium formation until adult eclosion is 14 days in S. bullata at 25°C, but the same developmental sequence is completed in 5 days in Musca domestica. a host less suited for parasitoid production (Rivers and Denlinger, 1993). Thus, fly species that develop very quickly (a short intrapuparial life) may be less vulnerable to developmental suppression following envenomation and consequently fewer wasp offspring would be produced. CHAPTER III REDIRECTION OF METABOLISM IN THE FLESH FLY, BARCOPHAGA BPLLATA. FOLLOWING ENVENOMATION BY THE ECTOPARASITOID NASONIA VITRIPENNIS t A CONTRAST BETWEEN NONDIAPAUSING AND DIAPADSING HOSTS Introduction Insect parasitoids do much more than simply feed on the body of their hosts. These "specialized" parasites are adapted to oviposit and develop in specific host stages, and thus they often require specific endocrinological and nutritional factors from the host. In fact, the evolution of host-parasitoid relationships favors regulation of host development for the benefit of the parasitoid's offspring (Vinson and Iwantsch, 1980). host development is regulated by factor(s) (calyx fluid, venom, polydnavirsus, etc.) injected by parasitic wasps during oviposition (Dover et al., 1987; Wani et al., 1990; Strand and Dover, 1991; Krell, 1991). Parasitoid manipulation of a host can arrest or suppress host development. In many cases, this halt in development appears to be due to changes in the hormonal status of the host (Baronio and Sehnal, 1980; Beckage, 70 71 1985; Lawrence, 1986; Webb and Dahlman, 1986; Brown and Reed-Larsen, 1991) and these endocrinological disturbances facilitate utilization of the host by the parasitoid (Beckage, 1985; Brown et al., 1990). Host arrestment, as well as parasitoid development, may also be dependent upon alterations of the host's metabolic reserves. A diverse range of parasites (protozoan, nematode, bacterial, and viral) alter the intermediary metabolism of their insect hosts (Stubblefield et al., 1966; Lewis et al., 1971; Rutherford and Webster, 1978; Womersley and Platzer, 1984; Dunn, 1986; Miranpuri et al., 1992). Similar parasitoid-induced changes in carbohydrate, protein, and lipid levels have been observed (Barlow, 1962; Barras et al., 1970; Dahlman, 1970, 1975; Thompson and Binder, 1984; Thompson et al., 1990), but most previous work has been done on endoparasitoids where it is not possible to separate the effects of the feeding parasitoid from the effects of the arrestment factor(s). The gregarious, ectoparasitoid Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae) can induce a developmental arrest in pupae and pharate adults of Sarcophaaa bullata Parker (Diptera: Sarcophagidae) by injecting venom prior to feeding or oviposition (Rivers and Denlinger, 1993). Arrestment of the host is essential for parasitoid development. The cause of 72 arrestment is unclear, but the halt in development can not be countered with an injection of exogenous ecdysteroid (Rivers and Denlinger, 1993). Possible alterations in host metabolism could account for the arrest of host development and/or the success of parasitoid development. Since the parasitoid's eggs can be removed from the pupal integument without disturbing the host's physiology, the effects of envenomation on the metabolic reserves of £. bullata can be separated from the pathology associated with larval feeding by vitripennis. Here, we examine the rate of oxygen consumption in developmentally arrested pharate adults of £. bullata and compare these values with rates observed in unevenomated flies. We monitor the effects of envenomation on levels of carbohydrate, protein, lipid, amino acids, pyruvate, and oxaloacetate during host arrestment. In addition, we determine if envenomation-induced alterations in metabolic reserves are apparent in hosts that are already in an environmentally-induced developmental arrest, diapause. 73 Materials and Methods Parasitoid and host rearing vitripennis females were collected in Columbus, Ohio from pupae of Calliphora spp. and maintained as a laboratory culture for 3 years on bullata. For each experiment, 25-30 mated females of vitripennis (3-7 days after emergence from the host puparium) were placed in plastic Petri dishes (15 x 100 mm) with 50 pupae of S. bullata and a 1:1 honey-water solution for 24 h prior to each treatment. The parasitoids were held at 25°C, with a daily light:dark cycle of L15:D9. Laboratory colonies of §. bullata were maintained as previously described (Denlinger, 1972a). To generate nondiapausing pupae, the flies were reared at 25°C, L15:D9 throughout development. Diapause-destined pupae were produced by rearing adults at 25°C, L12:D12, and larvae at 20°C, L12:D12. Pupal diapause was confirmed using the criteria described by Fraenkel and Hsiao (1968). Exposure of parasitoid to host Puparia of £5. bullata were wrapped in aluminum foil so that only the anterior third of the puparium was exposed to H. vitripennis. This restricted the site of oviposition and allowed easy removal of the parasitoids 74 from the host. Each host was exposed to a single female of N. vitripennis for 6 h in a plastic l oz. cup (Dixie). After host exposure, the adult wasps were discarded, the operculum from each puparium was removed, and the parasitoid's eggs were removed from the surface of the host integument. Nondiapausing pharate adults used in these experiments were envenomated 5 days after pupariation (young pharate adults), and diapausing pupae had been in diapause for 20 days. Metabolic responses were determined in hosts held at 2 5°C, L15:D9 following envenomat ion. Oxvaen consumption determination Oxygen consumption was determined in diapausing pupae and nondiapausing pharate adults of S. bullata using a Scholander volumetric respirometer as previously described (Denlinger et al., 1972b). Five flies were placed in each respirometer vessel, and the consumption of 02 at 2 5°C was recorded for several days following envenomation. To determine if envenomation-induced 02 consumption rates were similar to 02 consumption associated with injury metabolism, 02 consumption was monitored in nondiapausing pharate adults that were pricked with an insect pin (#3) through the dorsal surface of the head (a 75 (a single insertion of the pin). o2 consumption rates were also determined for pharate adults that were killed by a 1 h exposure to -70°C. Glvcoaen and trehalose isolation Glycogen content was determined as described by Van Handel (1965). An individual pupa or pharate adult was removed from the puparium and homogenized (Virtis 23 homogenizer) in 0.2 ml sodium sulfate (2% w/v) for 2-3 min at 25°C. The homogenate was mixed with 1 ml methanol, centrifuged at 2486 x g for 2 min, and the supernatant was decanted into a clean centrifuge tube (13 x 100 mm). The pellet was resuspended in 1 ml distilled h 20, mixed with 1 ml methanol, and then centrifuged at 2486 x g for 2 min. The supernatant was combined with the first fraction and saved for trehalose determination. The pellet was resuspended in 1 ml H2<3 and the glycogen content determined using anthrone reagent. The trehalose content in each fly host was determined as previously described (Van Handel, 1985a). Protein determination For protein determination, individual hosts (removed from the puparium) were homogenized in 1 ml 25 mM phosphate buffer (pH 7.4), centrifuged at 10,000 x g (Sorvall) for 10 min at 4°C, and the supernatant decanted 76 into a clean test tube. The protein content in each sample was determined as decribed previously (Bradford, 1976), using bovine serum albumin (Sigma) as the standard. Lipid extractions Total body lipid was determined using a modified version of the chloroform-methanol procedure described by Bligh and Dyer (1959). Each host (removed from the puparium) was homogenized for 2 min with a mixture of 1 ml chloroform and 2 ml methanol. An additional 1 ml chloroform was added and the mixture homogenized for 1 min. After the addition of 1 ml H20, 1 ml chloroform and 2 ml methanol were added, the mixture was homogenized for 2 min, and an additional 1 ml chloroform was added. The phases were allowed to separate for 30 min, and then the chloroform layer was transferred to a clean test tube that was placed in a heating block. The solvent was evaporated by heating the tube at 90-110°C for several minutes. The extracted lipids were converted to sulfonic acid derivatives by resuspension in 0.4 ml 12 M sulfuric acid, followed by incubation at 90-92°C for 10 min. The lipid content was determined using vanillin reagent as described previously (Van Handel, 1985b). Sesame oil (Sigma) was used as the standard. 77 Keto acid (ionized) isolation Individual pupae and pharate adults (removed from the puparium) were homogenized in 1 ml 25 mM phosphate buffer (pH 7.4). The homogenates were centrifuged at 2486 x g for 3 min and the supernatant decanted into a clean test tube. Samples were deproteinized by mixing each supernatant with 0.5 ml perchloric acid (6% v/v), the protein was then pelleted by centrifugation (10,000 x g for 10 min at 4°C), and the supernatant adjusted to pH 3.5 (pyruvate) or 6.0 (oxaloacetate) using 5 M potassium carbonate. Pyruvate was determined using the reverse enzymatic reaction described by Bucher et al. (1963) for lactate dehydrogenase (EC 1.1.1.27) (Sigma). The determination of oxaloacetate levels was performed in a similar manner as above, with the exception that malate dehydrogenase (EC 1.1.1.37) (Sigma) was used as the catalyst (Hohorst and Reim, 1963). Amino acid analysis Hemolymph from nondiapausing hosts was collected by centrifuging pharate adults in microcentrifuge tubes at 5000 x g for 1-2 min. This procedure forced the fat body and hemocytes from the fly's head into the thorax and abdomen without disturbing the turgor of the host's body. Hosts were then pricked in the head with an insect pin and relatively cell-free hemolymph was collected in a 1.5 78 ml microcentrifuge tube. Hemolymph from 3 pharate adults was pooled, centrifuged at 10,000 x g for 10 min at 4°C, and the supernatant deproteinized with trichloroacetic acetic acid (6% w/v) as described above for the keto acid analysis. Amino acid concentrations were determined using a Waters Picotag Amino Acid Analysis System (Biochemical Instrumentation Facility, The Ohio State University). Preliminary experiments revealed that alanine was the dominant amino acid in hemolymph from envenomated pharate adults. Thus, alanine levels were monitored for several days following envenomation. Whole body homogenates were prepared from pharate adults as described previously for the keto acid analysis and the pH adjusted to 3.5 using 5 M potassium carbonate. Alanine concentrations were determined using the reverse enzymatic reaction described by Williamson (1974) for alanine dehydrogenase (EC 1.4.1.1) (Sigma). Results Rate of Oxvaen consumption As expected (Denlinger et al., 1972b), the rate of C>2 consumption in unenvenomated pharate adults increased as pharate adult development progressed (Fig. 7 A ) . In contrast, the rate of 02 consumption in unenvenomated 79 diapausing pupae was 1/10-1/15 of the lowest rate observed in pharate adults, and the rates of 0 2 consumption remained constant throughout the 3 5 days of this study. For this experiment, 02 consumption was monitored for groups of 5 diapausing pupae, thus the cycles of 0 2 consumption that are evident in individual pupae (Denlinger et al., 1972b; Slama and Denlinger, 1992) were not observed. Following envenomation, the rate of 0 2 consumption of pharate adults decreased rapidly in the first 12 h and thereafter remained constant (35-50 ul g- 1h“1) until just before death (Fig. 7A). In contrast, the 02 consumption rates in diapausing pupae were already low and envenomation did not significantly lower the rate (F= 2.06; df= 12,98; P > 0.05) until many days later when death occurred (Fig. 7 A ) . The consumption of oxygen in pharate adults wounded with an insect pin did not differ from the o2 consumption rates in unenvenomated hosts (Fig. 7A), thus suggesting that the wound made by envenomation is not responsible for the change in 0 2 consumption. 0 2 consumption in flies killed by a 1 hour exposure to -70°C dropped sharply within 2 h after exposure, and 0 2 consumption in these pharate adults was not detectable 2 days later. This drop in the 0 2 consumption rates to an undetectable level paralleled changes in morphological appearance; 80 pharate adults killed by freezing were severely desiccated and the entire body darkened due to necrosis. Clearly the response of the envenomated flies was quite different than the responses observed in flies that were either mechanically wounded or killed. Glvcogen At most time points, the glycogen content in unenvenomated flies (diapausing and nondiapausing) did not differ significantly from the levels extracted from those that were envenomated (Fig. 7B; F= 1.84; df= 30,320; P > 0.05). One exception was an elevation in glycogen 2 h after envenomation in pharate adults, but 24 h later the glycogen content in these pharate adults returned to the levels observed in unenvenomated flies. In all flies examined, the levels of extractable glycogen steadily decreased over the following 15 days or, in unevenomated flies, until adult eclosion (Fig. 7 B ) . Trehalose The levels of trehalose in unenvenomated pharate adults steadily declined as pharate adult development progressed (Fig. 7C). In contrast, trehalose levels dropped very rapidly within 2 h following envenomation, remained constant for several days, and then slowly declined until the 15th day after envenomation (Fig. 7C). 81 The content of trehalose fluctuated very little in diapausing pupae and was not significantly altered by envenomation (Fig. 7C; F= 2.78; df= 20,295; P > 0.05). Protein Total body protein concentration was 20-25 mg/g higher in diapausing pupae than in pharate adults (Fig. 7D; F= 132.61; df= 13,191; P < 0.01). In pharate adults that were unenvenomated, protein levels remained unchanged with the progression of pharate adult development. Following envenomation, protein concentrations in pharate adults intially remained the same as in their unenvenomated counterparts, but by the 10th day after adult parasitoid attack, the protein content of pharate adults gradually increased to levels approaching the higher levels observed in pupal diapause (Fig. 7D). Protein content of diapausing hosts was not altered by envenomation. bipi<3 Lipid levels in unenvenomated diapausing pupae were consistently 2-3 times higher than observed in nondiapausing pharate adults (Fig. 7E; F= 39.9; df= 13,394; P < 0.01), and, in the absence of envenomation, the lipid content of both groups of flies remained constant throughout the experimental period. When pharate adults were envenomated, the lipid content remained constant during the first 3 days, but then rose sharply on the 4th day following parasitoid attack and remained elevated (6-7 times the pre-envenomation levels and 3-4 times the level observed in diapausing hosts) until 10 days post-envenomation. No differences were observed in diapausing hosts. Pyruvate The quantities of pyruvate isolated from unenvenomated pharate adults increased sharply early in pharate adult develoacid analysis. Amino aciconstant until adult eclosion (Fig. 7F). Following envenomation, pyruvate content in pharate adults was elevated for the first 3 days and then declined rapidly by the 10th day (Fig. 7F). Pyruvate levels in diapausing pupae were significantly lower than in pharate adults (F= 16.0; df= 13,209; P < 0.01) and remained at 0.6-0.7 umoles/g for the 15 days of this experiment (Fig. 7F). After envenomation, pyruvate levels in diapausing pupae were much lower than in unenvenomated diapausing hosts (Fig. 7F; F= 53.3; df= 20,372; P < 0 .0 1 ), with the exception of a brief elevation 3 days after envenomation. 83 Oxaloacetate Oxaloacetate concentrations steadily increased throughout development in unenvenomated pharate adults (Fig. 7F). In envenomated hosts, the oxaloacetate content stabilized around 1 .8-2.0 umoles/g in pharate adults and then slowly declined by 10 days after envenomation (Fig. 7 F ) . This keto acid was not detectable in diapausing pupae that were envenomated or unenvenomated. Amino acids Using pooled hemolymph from 3 pharate adults, 17 amino acids were detected in unenvenomated nondiapausing hosts, and, among these, alanine, glutamate, proline, and glycine were present in the highest concentrations (Table 3). Hemolymph collected from pharate adults 2 days after envenomation showed a similar amino acid profile, but the concentrations of each amino acid were elevated. The most pronounced difference was a 3 fold increase in alanine. Alanine represented 16.9% of the total amino acid hemolymph content of unenvenomated hosts, compared to 2 3.9% in envenomated pharate adults. These preliminary amino acid profiles prompted further experiments evaluating alanine levels in envenomated h o s t s . 84 Total body alanine levels increased slightly at the onset of pharate adult development in unenvenomated hosts and remained constant throughout the remainder of development (Fig. 8 ). In contrast, the alanine content of envenomated hosts doubled within 2 h following parasitoid attack and continued to rise (Fig. 8 ). Twenty days after envenomation, alanine still remained high (mean + SEM = 24.3 ± 0.5 umoles/g). 85 TABLE 3. Hemolymph amino acid composition in pharate adults of £. bullata Hemolymph amino acid concentration (mM) Amino Acid Unenvenomated Envenomated alanine 12. 13 26.27 arginine 6.72 7.74 aspartate 0 . 12 0.71 cysteine 0.05 0 . 08 glutamate 9.47 11.60 glycine 8.91 10.55 histidine 7.36 9.89 isoleucine 1.21 2.49 leucine 2 .02 3 . 30 lysine 2.48 4. 12 methionine 0.53 1.07 phenylalanine 0.76 1.36 proline 8 . 18 11. 53 serine 2 .80 3 .92 threonine 1.29 2.89 tyrosine 4.40 5.35 valine 3.29 7. 14 Total 71.72 110.01 Amino acid concentrations were determined in unenvenomated and envenomated (48 h after envenomation) pharate adults of £. bullata. For controls or envenomated hosts, hemolymph from three pharate adults was centrifuged (10,000 x g, 4°C) to remove hemocytes, deproteinized with 6% TCA (w/v), and amino acid concentrations determined using Waters Picotag Amino Acids Analysis System. Fig. 7 Changes in rate of oxygen consumption and metabolites in nondiapausing pharate adults of £. bullata (5 days after pupariation) (left panels) and diapausing pupae (20 days in diapause) (right panels) in response to envenomation by H. vitripennis. (A) Mean + SEM rate of oxygen consumption (ul 0 2/g/h) (B) Mean ± SEM total body glycogen content (mg/g). (C) Mean + SEM total body trehalose content (mg/g). (D) Mean ± SEM total body protein levels (mg/g). (E) Mean + SEM total body lipid levels (mg/g). (F) Mean + SEM total body pyruvate and oxaloacetate levels (umoles/g). Oxaloacetate was not detectable in any of the diapausing pupae assayed. Each metabolite experiment was replicated three times using 5 diapausing pupae or nondiapausing pharate adults in each replicate. 86 0> (|dO,rB/hf) ill I « ill I i ®hf6oe«i ®hf6oe«i <*"»/g) 0*yg«i oonmmpdon » ____ •f TiaMoi* (mg/g) «i PioMn ( Fig. Fig. 8 . Mean + SEM total body alanine content (umoles/g) in nondiapausing pharate adults of 3 . bullata envenomated by t£. vitripennis. All envenomated hosts contained significantly more alanine than any of the unenvenomated flies (P <0.01, One-way ANOVA). Each experiment was replicated twice using 5 pharate adults per replicate. 88 (timolMfQ) 7.3 + Deys efter envenemeden efter Deys Fig. Fig. 8 9 8 90 Discussion H. vitripennis drills through the host's puparium, envenomates the host and then deposits its eggs on the surface of the pupal (or pharate adult) integument. Prior to egg hatch (24-36 h following envenomation), a series of changes in the host's intermediary metabolism take place that may facilitate parasitoid development. Several earlier studies indicated that the hosts died immediately in response to envenomation (Roubaud, 1917; Beard, 1964; Ratcliffe and King, 1967), and if that were so, a redirection of the host's intermediary metabolism would not be expected. But, our recent evidence (Rivers and Denlinger, 1993) demonstrates that only very young and very old hosts (S. bullata) are killed immediately by N. vitripennis. In hosts of intermediate age, development is arrested, and the host metabolic changes we observed in this study are consistent with the idea that the fly's metabolism is redirected and independent of host decay. Several lines of evidence indicate that envenomation causes a redirection of metabolism: developmental arrest in host pupae and pharate adults, oxygen consumption rates are suppressed, and concentrations of certain metabolites are altered. All of the results reported in this paper can be attributed to envenomation rather than larval feeding because the 91 eggs of II. vitripennis were systematically removed from the host before hatching. In an unenvenomated fly, the rate of 02 consumption continued to increase throughout pharate adult development. In contrast, 02 consumption immediately dropped following envenomation of pharate adults to levels that were more similar to the rates observed in diapausing pupae. This decline in oxygen consumption is similar to the respiratory suppression observed in other hosts used by endoparasitic Hymenoptera (Edwards and Sernka, 1969; Jones and Lewis, 1971; Dahlman and Herald, 1971). Though 0 2 consumption was suppressed, the host remained alive for many days following envenomation. Death (and the final drop in 02 consumption) did not occur until approximately 16 d after envenomation. These results thus suggest that envenomation by M* vitripennis induces a low "diapause-like” respiratory metabolism in pharate adults and confirms our previous observations that venom injection does not immediately kill £. bullata (Rivers et al., 1993). The increased consumption of oxygen in developing pharate adults is consistent with the elevation in keto acids (pyruvate and oxaloacetate) observed in unenvenomated nondiapausing hosts. Pyruvate was also initially elevated in envenomated pharate adults but then began to slowly decline around the 4th day after 92 envenomation and remained at these lower levels until host death. The fact that oxaloacetate did not increase concurrently with pyruvate suggests that host metabolism was not directed toward the citric acid cycle. The elevation of pyruvate and the decline of oxaloacetate in envenomated pharate adults is suggestive of an active gluconeogenic pathway operating in nondiapausing hosts, but a shift in host metabolism toward carbohydrate accumulation was not observed. Alterations in tissue carbohydrate levels have been demonstrated.for a number of host-parasite associations (Dahlman, 1970; Von Brand, 1979; Thompson, 1983, 1986). In most cases, host carbohydrate content decreases and this has been attributed to consumption by the developing parasites. But, the decline of glycogen and trehalose in envenomated pupae and pharate adults of S. bullata can not be due to parasitoid feeding because the parasitoid's eggs were removed from the host, and furthermore, the carbohydrate levels declined even before the time required for hatching. There was a brief glycogen increase 2 h after envenomation, but levels of glycogen then declined throughout the remainder of this study. Most likely the intial glycogen elevation was a stress-induced response in S- bullata. due to the insertion of the wasp's ovipositor, injection of venom, and/or adult parasitoid 93 feeding on the host hemolymph. This would be consistent with the rapid decline in trehalose levels following envenomation and also similar to other stress-induced responses reported for insects (Jankovic-Hladni, 1991). The increased host pyruvate content and the decreased carbohydrate levels in envenomated pharate adults suggest that the rates of lipogenesis may have increased in nondiapausing hosts. This is consistent with the elevation of total body lipid levels extracted from envenomated pharate adults. The rise in lipid content is synchronized with the expected timing of the last larval moult in JJ* vitripennis and the beginning of the heaviest feeding by the developing parasitoids (Rivers and Denlinger, unpublished). This suggests that the elevation of lipid plays a role in the nutrition of the developing larvae of l£. vitripennis. A dependence on host lipid metabolism has been demonstrated for several other parasitoids (Bracken and Barlow, 1967; Thompson and Barlow, 1976; Espelie and Brown, 1990). The elevation in hemolymph concentrations of amino acids that was noted following envenomation suggests that protein hydrolysis increased as a result of envenomation, yet total body protein levels in these hosts also gradually increased following parasitoid attack. We did not separately measure the hemolymph protein 94 concentrations, thus we can not be certain if hemolymph protein levels were affected by envenomation. Alanine, the dominant amino acid in the host hemolymph, increased 3-fold immediately following envenomation. Alanine continued to increase in nondiapausing hosts for several days after envenomation, even after 0 2 consumption rates were no longer detectable. This elevation appeared to be independent of pyruvate concentration, thus indicating that alanine did not accumulate as a result of pyruvate transamination. A redirection of host metabolism was not as apparent in diapausing hosts. The rate of 0 2 consumption in diapausing pupae is much lower (Denlinger et al., 1972b) than the values observed for developing pharate adults and remained constant throughout the study. In envenomated pupae, the rate of 0 2 consumption was not altered until just before death, which occurred about 60 d later. The onset of death in diapausing hosts was much later than in pharate adults, suggesting that the higher rate of metabolism in nondiapausing hosts may accelerate the action of the venom, possibly more target sites are present in pharate adults, or the receptor(s) may have a higher affinity for the venom in nondiapausing hosts. The low rate of 02 consumption in diapausing pupae was indicative of low keto acid concentrations. The 95 pyruvate content of unenvenomated diapausing pupae was much lower than the levels observed in pharate adults and remained constant throughout the study. In envenomated hosts, pyruvate levels did increase, but the onset and duration of this elevation did not coincide with this event in pharate adults, oxaloacetate was not detectable in any of the diapausing pupae, which suggests that, as in pharate adults, host metabolism is not redirected toward the citric acid cycle. The carbohydrate content (glycogen and trehalose) slowly declined in diapausing pupae throughout the study and was not altered by envenomation. These low levels are similar to the carbohydrate levels noted for diapausing pupae of £. crassipalois (Adedokun and Denlinger, 1985) and are consistent with their suggestion that flesh flies rely on lipid metabolism instead of glycogen at the onset of diapause. Lipid levels were consistently 2-3 times higher in unenvenomated diapausing pupae than in developing pharate adults throughout this study. A similar result was noted for £. crassipalpis (Adedokun and Denlinger, 1985). This high lipid content in diapausing pupae was not altered by envenomation, but in pharate adults, lipid levels were highly elevated by envenomation and rose to levels that even exceeded diapause values by 3-4 times. If host lipid metabolism is essential for parasitoid development, 96 then the lower lipid content of envenomated diapausing pupae may explain why fewer adult parasitoids are produced on diapausing hosts than on nondiapausing hosts (Rivers and Denlinger, 1993). Whether the developmental suppression is the result of an altered host metabolism or if host arrestment precedes the metabolic disturbances remains unknown. Yet, the fact that intermediary metabolism of the fly host is altered due to envenomation and that these changes appear to be synchronized with specific developmental events in the life of the larval parasitoid suggest that JJ* vitripennis can redirect host metabolism for the benefit of the wasp progeny and that host arrestment may simply be the consequence of altering host metabolic reserves. Quite likely, the only changes in the host's intermediary metabolism that are relevant to parasitoid development are those that occur within a few days of envenomation. If parasitoid larvae are permitted to hatch on the host, the fly host is normally consumed within 5-7 days after oviposition. CHAPTER IV TOXICITY OF THE VENOM FROM MASONIA VITRIPENNIS TOWARD FLY HOSTS, NONTARGET INSECTS, DIFFERENT DEVELOPMENTAL STAGES, AND CULTURED INSECT CELLS Introduction The parasitic wasp, Nasonia vitripennis (Hymenoptera: Pteromalidae) is a widely distributed ectoparasitoid that attacks pupae from several families of the higher Diptera. Most commonly parasitized are flies in the families Sarcophagidae and Calliphoridae (Darling and Werren, 1990). Though genetics (Whiting, 1967), ecology (Merwe, 1943; Pimentel et al. , 1962), photoperiodism (Schneiderman and Horowitz, 1958; Saunders, 1965, 1975) and sex ratio theory (Werren, 1984a; Orzack and Parker, 1986; King and Skinner, 1991b) have been well studied in this species, very little is known about the venom produced by this wasp. Ratcliffe and King (1967, 1969) presented detailed studies on the morphology and ultrastructure of the Nasonia venom system, but no attempt has been made to characterize the composition or activity of the venom. This leaves unanswered several important questions about Nasonia 97 98 venom. Is the venom actually specific for flies or does it have a broad spectrum of activity? Does it have a paralytic action or does it elict its effect by some other mechanism? And, can this venom be used in the development of a selective bioinsecticide? This question is raised because several insect-specific toxins derived from scorpion and other arthropod venoms have potential for construction of highly selective biopesticides (Zlotkin, 1983; Stewart et al., 1991; Quistad et al., 1988, 1992; Tomalski et al., 1991). The goal of this study is to examine the biological activity of the venom from vitripennis. We compare the activity of this venom against different developmental stages of flies and moths, and toward insects representing ten different orders. In addition, we evaluate the effect of the venom on cultured insect cells and suggest that this novel approach can be used to quickly screen activity. Materials and Methods Insect rearing fi. vitripennis was collected from pupae of Calliphora spp. in Columbus, Ohio and maintained as a laboratory culture for 2 years. In the laboratory, wasps were reared on pupae of the flesh fly, Sarcophaaa bullata. and maintained at 25°C with a daily 15:9 light:dark cycle. Female wasps were allowed to host feed on flesh fly pupae for 24 hours after adult emergence and were then frozen and stored at -20°C until dissection of the venom gland. Laboratory colonies of Phaenicia sericata and £. bullata were reared on beef liver as larvae and fed sugar and water as adults. The flies were maintained at 2 5°c with a daily 15:9 light:dark cycle. Adults of Aphis nerii and Protoca11iphora spp. were field collected in Columbus, Ohio from milkweed plants and bluebird boxes, respectively. Trichoplusia ni and Heliothis virescens were provided by the U.S.D.A. Western Cotton Research Laboratory (Tuscon, Arizona) and maintained on a wheat germ-pinto bean artificial diet. Lvmantria dispar was provided by the U.S.D.A. Forest Service (Delaware, Ohio) and reared on a wheat germ artificial diet. All other insects were obtained from the Ohio State University Insectary. After injection of venom, all insects were kept at 25°C with a light:dark cycle of 15:9. Isolation of venom Venom gland reservoirs from host-fed H* vitripennis females were dissected into phosphate buffered saline (PBS) [10 mM sodium phosphate (dibasic), 0.9% (w/v) N a C l , 15% sucrose, pH 8.0] (Jones and Leluk, 1990). This was accomplished by gently separating the ovipositor from the 1 0 0 abdomen with fine forceps, pulling with it the venom gland and reservoir. The contents of the venom gland and reservoir are equally active (Ratcliffe and King, 1967), therefore to reduce dissection time only the reservoir was used for bioassays. Venom reservoirs were frozen and stored at -70°C. Isolated venom reservoirs were concentrated by centrifugation (Sorvall RC-5B) at 7,000 x g for 10 min, 4°C. Pellets were transferred to a Pyrex ground glass tissue grinder and homogenized (30 strokes) in 100 uls PBS in an ice bath. The homogenized material was transferred to a microcentrifuge tube (1.5 ml) and centrifuged at 12,000 x g for 10 min at 4°C. The pellet was discarded and the supernatant containing the venom was transferred to a clean microcentrifuge tube and stored frozen at ~70°C. A venom reservoir equivalent was defined as the supernatant from 1 homogenized venom reservoir in 1 ul PBS. Bioassavs Venom activity was assayed in different life stages of £. bullata. a natural host of li* vitripennis. Venom was injected into the hemocoel of third instar "wandering” larvae (pre-red spiracle stage) (110-120 mg) and adult females (82-94 mg) using a finely drawn glass capillary. Pupae (110-142 mg) were injected by first 10 1 removing the anterior cap of the puparium so that the head and most of the thorax were exposed. An insect pin was used to bleed off 1-2 uls of hemolymph from the head of each pupa, thus providing space for injection of venom into the hemocoel and to prevent excessive bleeding from the site of injection. Injections were made on the dorsal surface of the thorax, just posterior to the head. L D 5 0 's were calculated for each stage of g. bullata using probit analysis (Finney, 1971). Venom (0.05-10.0 VRE) was also injected into the hemocoel of the house fly, Musca domestica (adult 7-12 mg); greenbottle fly, £. sericata (adult 2 8-35 mg); vinegar fly, Drosophila melanoaaster (adult 0.5-0.7 mg); bird blow fly, Protocalliphora spp. (adult 21-45) ; milkweed bug, Oncopeltus fasciatus (adult 47-55 mg); yellow mealworm, Tenebrio molitor (pupa 140-153 mg, adult 44-71 mg); dermestid beetle, Dermestes spp. (adult 22-29 mg); German cockroach, Blattella germanica (adult 49-60 mg); American cockroach, Periolaneta americana (adult 721-780 mg) ; Indianmeal moth, Plodia interpunctella (larva 12-22 mg, pupa 12-2 0 , adult 10-21 mg); gypsy'moth, i. disoar (larva 140-155 mg, pupa 172-195, adult 110-135 mg); tobacco budworm, H. virescens (larva 141-153 mg, pupa 188-201 mg, adult 122-127 mg); cabbage looper , T. ni (larva 117-129 mg, pupa 145-162 mg, adult 118-129 mg); milkweed aphid, &. nerii (adult 0.05-0.2 mg); workers of 102 the honey bee, Apis mellifera (adult 121-126 mg) ; vitripennis (adult 0.03-0.2 mg); termite, Reticulitermes flavipes (adult 0 .2-0 .4 mg); ring-legged earwig, Euborellia annuljpeg (adult 30-45 mg); and walking stick, Diapheromera femorata (adult 600-671 mg). Adult females were used for venom injections of each insect except £. flavipes. £. annuljpes, and Q. femorata, for which both sexes were used. Mortality was scored at 24 hr intervals for five days. LDso's were calculated by probit analysis for VRE/insect and VRE/gram of insect. Cell Assays Two cell lines were used; TN-368 cells derived from adult ovaries of the cabbage looper, Trichoplusia ni (Hink, 1970), and NIH SaPe4 cells derived from embryonic tissue of the flesh fly, Sarcophaaa peregrina (Takahashi et al., 1980). Cells were pipetted into wells of microtiter plates (Falcon) to give a final concentration of 2 x 103 cells/well in a volume of 100 ul medium (TNM- FH). Venom in PBS was added to each well using a 50 ul Hamilton syringe. Cells were incubated at 27°c for 24 h before scoring mortality. Cell viability was determined by vital staining using Trypan Blue. 10 ul of a 0 .1% (w/v) stock solution of Trypan Blue was added to each well (100 ul) and incubated for 5-10 min before observation. The final concentration of Trypan Blue 103 (approx. 0 .01%) in each well was sufficient to stain all dead cells but sufficiently dilute to permit observation without first having to pipette off the dye. The LC5Q (concentration required to induce 50% mortality) was determined using the method described by Finney (1971). 10 wells were evaluated for each venom concentration and replicated twice. The time required to kill 50% of the cells (LT50) was determined in the same manner as described above for the LC5q. Cell viability was assayed at the following time intervals after the introduction of a LCgg dose of venom: 0, 0.1, 0.2, 0.3, 0.7, l, 1.3, 1.7, 2, 2.3, 5, and 24 h. 10 wells were evaluated for each time interval and replicated three times. Results Venom activity in flies Pupae of cyclorrhaphous Diptera are the natural hosts of N. vitripennis. Within this group of flies, pupae from Calliphoridae and Sarcophagidae are the preferred hosts for oviposition (Darling and Werren, 1990), although muscids are utilized by Nasonia under certain conditions (Rutz and Scoles, 1989). Protocalliphora pupae (Calliphoridae) can serve as hosts (Darling and Werren, 1990), but this does not occur frequently due to the habitat (bird's nests) and feeding 104 behavior (parasitic on fledglings) of this fly. Pupae of Drosophila have not been reported as hosts for Nasonia (Thompson, 1958), probably because this fly is not associated with carrion and the extremely small size of the pupae may make them less suitable for oviposition. Injection of venom into adult representatives from these four families (Table 4) resulted in venom sensitivity similar to oviposition preference (S. bullata > £. sericata > Protocal1iphora > M. domestica = .Q. melanogasterl. No signs of muscular contractions, spasmotic activity or paralysis were evident in any stage of S. bullata following venom injection. Larvae injected with sublethal doses of venom (<5.3 VRE/g) were able to crawl normally and pupariate, but they formed aberrant puparia similar to those described by Zdarek and Fraenkel (1987). Injection of a lethal concentration of venom (>8.0 VRE/g, approx. = LDgo) resulted in the larvae becoming flaccid within several hours. This observation is similar to the flaccid- extended paralysis observed in Sarcophaaa aravrostoma (= faculatal larvae in response to an injection of scorpion venom (Zlotkin, 1983), yet, no other symptoms of this extended paralysis were observed. Adult flies showed no inhibition of movement and were able to display normal flight behavior for 2-3 days after venom injection (<6.0 105 VRE/g). Just prior to death, the adults lost their ability to stand. Paralysis was more difficult to evaluate in pupae because fly pupae are already immobile. The first apparent effect of Nasonia venom in larvae of £. bullata was the appearance of blackening at the site of injection, a response that was evident within 5-10 min. Mortality was observed after 1-2 days at lethal concentrations (>6.0 VRE/g). Pupariation was delayed 3-4 days in larvae exposed to a sublethal dose of venom (0.8-4.5 VRE/g). The first symptoms of envenomation in adults was darkening of the integument at the site of injection 1-2 days after injection (>l.o VRE/g). Pupae (nondiapause and diapause) responded much more slowly to venom injection than larvae or adults. Neither nondiapausing nor diapausing pupae displayed darkening of the integument for several days after envenomation and death did not ensue until much later (12-25 d for nondiapausing pupae, 25-60 d for diapausing pupae). The progression of pharate adult development was retarded by injecting nondiapausing pupae with venom (3.0-8.5 VRE/g). In controls injected with saline, eye pigment was deposited 1-2 days after injection, but in venom injected pupae (<8.5 VRE/g) deposition of pigment was delayed 4-5 days. Near the site of venom injection, the compound eyes remained only partially (1/2 to 3/4) filled with pigment, and black bristle formation was 106 reduced or absent. In addition, the compound eyes of envenomated pharate adults remained light orange/pink and never developed the deep red color observed in controls. None of the diapausing pupae that were envenomated succeeded in intiating pharate adult development. A crude venom preparation from vitripennis was equally toxic to all stages of £. bullata. The LDgo's were nearly identical for third instar larvae, nondiapausing pupae, diapausing pupae, and adult females (Table 4), and the 95% confidence intervals overlapped when activity was based on an individual response or on weight. The response in adult females of p. sericata. Protocalliphora. M- domestica. and p. melanogaster was similar to that of g. bullata adults. All individuals remained capable of flight for 1-2 days after envenomation (<4.5, <6.2, <10.5, and <10.9 VRE/g, respectively). p. sericata adults were equally as sensitive to envenomation as S. bullata. but adults of Protocal1iphora. g. melanogaster. and M* domestica were 2-4 times more tolerant of the venom (Table 4). Venom activity in different stages of nontaroet insects As in flies, paralysis was not detected in any of the nontarget insects injected with venom, with the possible exceptions of J. ni and £. molitor pupae. Injection of venom into 1-2 d old pupae of T. ni resulted 107 in a dose-dependent loss of abdominal mobility followed by a darkening of the integument. Neither the larval nor adult stages of X. ni displayed signs of paralysis, so it seems unlikely that the effects observed in the pupae were due to a neurotoxic component in the venom. Cessation of abdominal movement was also observed in T. molitor pupae after venom injection. Unlike X* ni pupae, however, darkening of the pupal integument in X* molitor was associated with blackening of the entire body, including the hemolymph. All other insects injected with venom continued to feed and walk normally for several hours following venom injection and showed no evidence of paralysis. Larvae of p. interounctella and T. ni exposed to sublethal doses of venom (<6.0 and <4.2 VRE/g, respectively) were able to pupate normally and did not show the type of delay observed in S. bullata larvae. Larvae of p. interounctella and X* ni showed signs of blackening at the site of injection within 5-10 min, but mortality was not evident until 24-3 6 h later. Pupae of X- Hi were the most sensitive (LD5Q = 0.58 VRE/g) of any of the stages or species injected with Nasonia venom (Table 4), including the natural host of this wasp, £. bullata pupae (LD50 = 5.46 VRE/g). Suprisingly, pupae of X* ni were almost 10 times more sensitive than larvae and adults (7.23 and 7.48 VRE/g, respectively). Likewise, pupae of p. interounctella and 108 X- molitor were more susceptible than larvae or adults (Table 4). The contrast in sensitivities of different life stages was most pronounced between adults and pupae of E. interounctella. Pupae of this species were at least 80 times more susceptible to Nasonia venom than the adults (LD50 = 10.6 and >880 VRE/g, respectively). Interestingly, adults of vitripennis were not sensitive to their own venom (LD5Q = >53 3 VRE/g), although another hymenopteran, A* mellifera. was very susceptible (4.62 VRE/g). The hemimetabolous insects and larvae, pupae and adults of 1,. dispar were extremely tolerant of venom injection and thus no LD50's could be calculated (Table 4). Effect of venom on insect cells Bioassays using T. ni (TN-368) and S. perearina (NIH SaPe4) cells proved effective for evaluating the venom response. The LC5Q of Nasonia venom for both cell types was nearly identical (Table 5) and is consistent with the in vivo bioassays using adults of X* Qi and various stages of £• bullata (Table 4). with both cell types, the venom caused a rapid rounding up of the cells followed by slight swelling of the plasma membrane and eventual death. Using Trypan Blue staining as a measure of cell viability, 36% of TN-368 and 37.5% of NIH SaPe4 cells exposed to 0.01 VRE/ul were nonviable 10 min after venom exposure (Fig. 9). However, there were differences in responses of the cell lines in that it took much longer (24 h) to kill 100% of TN-368 cells versus 100 min for NIH SaPe4 cells. Also, the calculated LT50, using LD99 doses, were 27 min for TN-368 and 16 min for NIH S a P e 4 . 110 Table 4. Activity of the venom from Nasonia vitrinenni« toward natural and nontarget species LD50 Insect Stags n VKS/lnaect V W / g r m a Slopa Diptara lareonhiat bttllatn * larva 20 1.04 5.20 1.07 ± 0.40 (0.71,2.40) (4.37,0.50) 3.29 £ 0.57 nendiapauss 45 0.44 5.40 1.00 £ 0.37 F«P« (0.32,0.75) (3.41,13.2) 1.45 £ 0.30 diapause 15 0.50 5.54 1.01 * 0.52 pupa (0.40,1.04) (3.70,9.71) 1.01 £ 0.52 adult 45 0.50 5.70 3.12 £ 1.15 (0.24,1.43) (4.54,7.00) 3.05 ± 0.43 adult 25 0.10 4.49 2.00 £ 0.43 (0.10,0.24) (4.75,0.14*) 3.39 £ 0.07 PEotocallichors sp. adult 25 0.10 0.47 2.09 £ 0.57 (0.11,0.35) (4.50,10.3) 4.27 £ 0.05 adult 45 0.12 17.2 1.32 ± 0.40 (0.04,0.32) (14.0,20.5) 3.17 £ 0.40 adult IS 0.000 15.7 2.04 ± 0.90 (0.004,0.023) (0.27,45.0) 2.44 * 0.90 Lepidoptara larva 20 0.15 14.5 1.07 * 0.49 (0.05,0.31) (4.52,23.3) 3.04 ± 1.13 pupa 12 0.12 10.4 1.42 £ 0.42 (0.03,0.31) (2.47,20.4) 1.42 £ 0.42 adult 15 >2.0 >000 ------——— TiVMntrl1 d i n a r larva 30 >0.0 >53.3 pupa 15 >4.0 >21.0 adult 10 >4.0 >27.1 Trlchoaluala ai larva 15 0.93 7.23 1.54 £ 0.40 (0.54,2.05) (4.31,15.4) 1.04 £ 0.43 pupa 12 0.09 0.50 2.40 £ 0.00 (0.04,0.15) (0.27,0.90) 2.39 £ 0.40 adult 10 0.93 7.40 2.33 £ 0.70 (0.47,1.01) (3.01,14.4) 2.34 £ 0.70 Haliuthi* virasesns larva IS >4.0 >27.4 ------pupa 12 1.24 4.44 2.01 £ 0.71 (0.49,3.37) (3.41,17.4) 2.01 £ 0.71 adult 10 0.41 5.00 3.40 £ 0.09 (0.33,0.95) (2.72,7.77) 3.40 £ 0.09 I l l Tabla 4 Continued. Ryuanoptara HMOnla vltglnannla adult 7 >0.10 >533 ABA* ■elllgera adult 12 0.51 4. <2 3.07 ± 0.77 (0.35,0.13) (2.07,0.00) 3.07 ± 0.77 Colaoptera T«nabrla talttar pupa 10 0.57 3.54 3.59 ± 1.10 (0.33,0.50} (2.25,0.71) 3.50 + 1.09 adult 15 0.03 00.7 2.92 ± 1.35 (4.00,S .09) (07.7,157. 9) 3.15 + 0.98 Cuuatii op. adult 15 0.71 29.9 2.01 ± 0.55 (0.45,1.14) (10.0,47.9) 2.01 ± 0.55 Haalptara Qneopaltua faaeiatua adult IS >4.00 >00.0 Houoptara *Bhl« naril adult 12 >0.25 >1200 Oersaptara EubgrtlUt annul lnaa adult IS >1.0 >23.4 PtiasBatodaa □iapharaaara fauorata adult 10 >3.0 >4.70 Orthoptara Blattalla aaraaniea adult 12 5.31 102.1 2.09 ± 0.50 (3.37,11.9) (50.0,321.0) 1.00 * 0.67 Farlplanata aaarigana adult 15 >0.0 >0.30 Xsoptara ROtXgUllf gaoo glavipaa adult 12 >0.25 >950 Table 5. Response of Trichoplusia ni (TN-368) and Sarcophaaa perearina (NIH SaPe4) cell lines to the venom of Nasonia vitripennis TN-368 NIH SaPe4 Origin of adult cell line ovary embryo LC50 (VRE/ul) 0.0014 0.0010 (0.0005,0.0038) (0.0004,0.0028) Slope 1.76 + 0.43 2.03 + 0.47 LT50 (minutes) 27.4 16.2 (15.7,47.0) (14.5,17.8) Slope 1.25 ± 0.01 2.01 + 0.12 LC50's for T* Hi and £. perearina cell lines were determined by exposing cells to six concentrations of venom and incubating for 24 hrs before observing mortality. Trypan Blue dye exclusion staining was used to determine cell viability. LT50's were calculated in the same manner as for LC^q's utilizing 12 time intervals and replicated twice. Fig. 9. Cytotoxicity of Nasonia vitripennis venom toward TN-368 and NIH SaPe4 cells. Each point on the curve was generated using a venom dosage (0.01 VRE/ul) determined in preliminary tests to yield a LCgg* Cell viability was assessed using Trypan Blue dye exclusion staining. Ten wells were assayed for each time interval, with each well containing 2 x 103 cells/100 ul culture medium. Cell viability was determined from three replicates and the standard deviation is given for each point on the curve. 113 i "K. i. -4- // -- 1--- //I i 5 Tim* itlir Addition or Vonom |hn) 115 Discussion Crude Nasonia venom displays high activity for insects from 4 of the 10 orders tested. Though fly pupae are the normal host, larvae and adults of the flesh fly, S. bullata. were equally sensitive. Adults of the calliphorid, £. sericata. and the sarcophagid, S. bullata. showed similar sensitivities (6.5 and 5.7 VRE/g, respectively) , while the muscid, Jl. domestica. and the drosophilid, B* melanoaaster. were approximately 3 times more tolerant of Nasonia venom. This observation is consistent with the female oviposition preference of N. vitripennis: pupae of sarcophagids and calliphorids are selected over muscids (Ohgushi, 1959). Factitious hosts from the Hemimetabola demonstrated little sensitivity to Nasonia venom. However, pupae of all the holometabolous insects injected with venom were highly susceptible and sensitivity was nearly identical, or even higher in some cases, to that observed in £. bullata. Pupae of T. ni were almost 10 fold more susceptible to Nasonia venom than any of the other insects assayed. This sensitivity observed in nontarget insect pupae suggests that the venom may be affecting some process critical for the successful metamorphosis of holometabolous insects. Most Hymenoptera venoms are not selective for target insects. Venoms from species in several genera of social Hymenoptera are capable of paralyzing insects from many 116 different orders and display virtually no specificity (Quistad et al., 1988; and Piek and Spanjer, 1986). This lack of specificity is presumably due to the indiscriminate cell lysis that is commonly associated with paralytic, defensive venoms (Quistad et al., 1992), or to nonspecific binding of neurotoxic components in the venoms. Venom from Euolectrus olathvpenae (family Eulophidae) appears to demonstrate some selectivity by arresting molting in Lepidoptera larvae (44 species) and representatives of Hemiptera, Coleoptera, and Neuroptera (Coudron and Puttier, 1988), yet, insects from Diptera and Orthoptera were not affected. Microbracon hebetor (family Braconidae) venom is also specific for Lepidoptera larvae, although it is capable of paralyzing insects from several other orders (Drenth, 1974) . To our knowledge, no previous attempt has been made to evaluate the effects of hymenopteran venoms on different life stages of the same insect species. Thus, the high activity we observed with Nasonia venom for a single life stage (pupae from 3 orders of insects: Diptera, Lepidoptera, and Coleoptera) may not be unusual, but insufficient comparative data is available for other Hymenoptera. Blackening of the integument was evident in all insects injected with venom but was not observed in control insects injected with saline. This darkening 117 may have been due, in part, to melanization as a general wound repair response (Lai-Fook, 1966) and/or coagulation of the hemolymph, followed by hardening and tanning by cuticular enzymes (St. Leger, 1991) . The response was rapid, and in larvae of S. bullata darkening of the integument at the site of injection appeared within 10 min. However, mortality was not observed for 1-2 days post-injection (> 6.0 VRE/g), and pupariation was delayed 3-4 days in those larvae exposed to a sublethal dose of venom. Adult insects injected with venom blackened at the site of injection in 1-2 days, but darkening of the entire body was not evident until just before death. Pupae of S. bullata displayed no darkening of the integument for several days following envenomation, a result that is consistent with the slowness of pupal response to venom injection. Pharate adult development in S. bullata was arrested by injecting pupae with venom, but the arrest was incomplete. Some pigmentation was deposited in the compound eyes and black bristles eventually formed on the body, although the appearance of these developmental landmarks was delayed 4-5 days. The response of the pharate adult may be localized: only partial pigment deposition occurred in the compound eyes near the site of injection, and black bristle formation was reduced or absent in that local area. In addition, the eye pigment 118 of envenomated pharate adults remained light orange/pink and never developed the deep red color observed in controls. Though some eye pigment deposition occurred following venom injection, the venom somehow altered this pr o c e s s . Most ectoparasitic wasps paralyze their hosts prior to oviposition (Beard, 1963; Piek and Spanjer, 1986), but most species that have been examined attack mobile hosts such as Lepidoptera larvae. Thus, it is highly advantageous for the host to be paralyzed for some time in order for the ectoparasite to oviposit without disturbance. A paralytic action seems unnecessary for the venoms of parasitoids that utilize an egg or pupal host because these stages already are immobile. Pleurotropis passei does apparently paralyze beetle pupae (Taylor, 1937), but unlike fly pupae, many pupae of Coleoptera are capable of some movement and therefore may require paralysis to facilitate oviposition. None of the insects injected with Nasonia venom showed any evidence of paralysis. The exceptions may be pupae of T. ni and T. molitor. both of which lost abdominal mobility following venom injection. But, this does not conclusively demonstrate a paralytic effect since it was not possible to determine if these pupae had simply died. In most cases, the evidence is clear that Nasonia venom exerts its effects by a nonparalytic mechanism. 119 When tested in vitro against cultured insect cells, venom from Nasonia was cytopathic without lysis of cells. Initial effects of the venom were detected within 10 min for both cell types by vital staining, yet, 100% mortality was not observed until 24 h later for TN-3 68 cells. Complete cell death occurred much faster in NIH SaPe4 cells (100 min). The assessment of venom activity using cultured insect cells presents a novel method for the study of insect venoms. Effects of the venom could be detected within 10 min and were more consistent and reliable than in vivo bioassays. This method is also less labor intensive. We found that 5-10 trays (80 wells/tray) of insect cells could be assayed for activity in the time required to inject one insect species with 4-5 doses of v e n o m . The venom from H- vitripennis may have considerable potential in the development of future biopesticides. It demonstrates activity toward insects from 4 different orders with several fly species displaying high sensitivity to envenomation. The fact that it also appears to be nonparalytic suggests that the active components may be an attractive alternative to the paralyzing toxins of other arthropods. SUMMARY I. Host quality and its impact on fecundity, developmental time, and sex ratio in Nasonia vitripennis. a gregarious, ectoparasitoid of the flesh fly, Sarcophaoa bullata. 1. Host species greatly influenced the fecundity and developmental time of vitripennis: this wasp produced more progeny and developed more rapidly on several species of sarcophagid pupae than on M. domestica. and vitripennis rejected puparia of R. pomonella. melanoaaster. and Phorus spp. 2* There was a positive relationship between weight of the host and the number of wasp offspring produced. Weight increases beyond 14 0 mg did not increase progeny production. 3. N. vitripennis prefered nondiapausing hosts (3-9 days after pupariation) for oviposition and development over diapausing, dead, unenvenomated, and older hosts. 120 121 4. Envenomation of a host was essential for the successful development of E- vitripennis. 5. The sex ratio of N. vitripennis was female-biased when the clutch developed on nondiapausing hosts, but shifted toward 1:1 when the wasp utilized dead or diapausing hosts, or pupae of M. domestica. II. Developmental fate of the flesh fly, Sarcophaqa bullata, envenomated by the pupal ectoparasitoid, Nasonia vitripennis. 1. All hosts used for feeding or oviposition were envenomated. 2. Envenomated hosts either died immediately or entered a developmental arrest. Most young (2 days after pupariation) or late pharate adults (>9 days after pupariation) died within 48 h. The dominant response of intermediate aged hosts (3-9 days after pupariation) was an arrest or retardation of development. Diapausing pupae appeared "preserved" for up to 4 0 days following envenomation. 3. Pharate adult development did occur in some nondiapausing hosts (4-7 days after pupariation) following envenomation, but it proceeded very 122 slowly, and none of these hosts completed pharate adult development. 4. Host developmental arrest was independent of the presence of feeding parasitoid larvae and appeared to be due to the injection of venom by the ovipositing female wasp. 5. The developmental arrest could not be countered with 20-hydroxyecdysone injections. 6 . The onset and duration of the developmental arrest was dependent on host age and the site of envenomation. The injection of venom into true pupae (4 days after pupariation) or into the anterior half of the host body was more deliterious than injection into older hosts (5-7 days after pupariation) or the posterior body region. 7. N. vitripennis produced far more progeny on thorax-abdomen preparations isolated by ligation than on isolated head-thorax preparations. III. Redirection of metabolism in the flesh fly, Sarcophaoa bullata. following envenomation by the ectoparasitoid Nasonia vitripennis: A contrast between nondiapausing and diapausing hosts. 1. Following envenomation of pharate adults, the rate of oxygen consumption dropped sharply and 123 the total body content of oxaloacetate, trehalose, and glycogen steadily declined. 2. Hemolymph amino acid concentrations, most notably alanine, increased 2-3 fold following envenomation. 3* Pyruvate levels intially increased following envenomation but declined rapidly 3-4 days later. 4. The decline of this keto acid occurred concurrently with an increase in total body lipid (an increase of 6-7 times the pre- envenomation levels), suggesting that the rate of host lipogenesis had increased following envenomation. 5. Envenomation of diapausing pupae did not alter the rate of 0 2 consumption nor levels of trehalose, glycogen, alanine, lipid, or protein. Pyruvate briefly increased, but the onset and duration of this elevation did not coincide with the changes observed in pharate adults that were envenomated. IV. Toxicity of the venom from Nasonia vitripennis toward fly hosts, nontarget insects, different developmental stages, and cultured insect cells. 1. Isolated venom from JJ. vitripennis injected into different fly species showed selectivity similar to oviposition preference (S. bullata > P. sericata > Protoca11iohora > M. domestica = D. melanogasterl. The venom was equally toxic to all life stages (larvae, nondiapausing and diapausing pupae, and adults) of S. bullata that were injected. The pupal stage was the most susceptible to venom in the nontarget insects. In fact, pupae of T. ni were 10 times more sensitive to venom injection than the preferred host, S. bullata. All of the hemimetabolous insects screened were tolerant of the venom. Paralysis was not detected in any of the flies or nontarget insects injected with venom. Bioassays using X* ni (TN-368) and £. perearina (NIH SaPe4) cells demonstrated nearly identical LC5 0 's. With both cell types, the venom caused a rapid rounding up of the cells, followed by slight swelling of the plasma membrane and eventual death. Death occurred much more quickly in the NIH SaPe4 cells (100% mortality occurred in 100 min) than in the TN-368 cells (24 h). LIST OF REFERENCES Axtell R.C. and Rutz D.A. (1986) Role of parasites and predators as biological fly control agents in poultry production facilities. In Biological control of muscoid flies. (Eds Patterson R.S. and Rutz D.A.), Miscellaneous Publication 61, pp. 80-100. Entomological Society of America, College Park, Md. Bai B., Luck R.L., Forster L., Stephens B. and Janssen J.A.M. (1992) The effect of host size on quality attributes of the egg parasitoid, Trichoaramma prestiosum. Entomol. exp. appl. 64, 37-48. Barlow J.S. (1962) An effect of parasitism on hemolymph electropherograms. J. Insect Phvsiol. 36, 274-2 75. Baronio P. and Sehnal F. (1980) Dependence of the parasitoid Gonia cinerascens on the hormones of its lepidopterous hosts. J. Insect Phvsiol. 26, 619-626. Barras D.J., Joiner R.L., and Vinson S.B. (1970) Neutral lipid composition of the tobacco budworm, Heliothis virescens (Fab.), as affected by its habitual parasite, Cardiochiles niariceps viereck. Comp. Biochem. Phvsiol. 36, 775-783. Beard R.L. (1963) Insect toxins and venoms. A n n . R e v . Entomol. 8 , 1-18. Beard R.L. (1964) Pathogenic stinging of house-fly pupae by Nasonia vitripennis (Walker). J. Invertebr. Pathol. 6 , 1-4. Beckage N.E. (1985) Endocrine interactions between endoparasitic insects and their hosts. Annu. Rev. Entomol. 30, 371-413. 125 126 Beckage N.E., Metcalf J.S., Nesbit D.J., Schleifer K.W., Zetlan S.T. and de Buron L. (1990a) Host hemolymph monophenoloxidase activity in parasitized Manduca sexta larvae and evidence for inhibition of wasp polydnavirus. Insect; Biochem. 20, 285-294. Beckage N.E., Nesbit D.J., Nielsen B.D., Spence K.D. and Barman M.A. (1989) Alterations of hemolymph polypeptides in Manduca sexta: A two-dimensional electrophoretic analysis and comparison with major bacteria-induced proteins. Arch. Insect Biochem. Phvsiol. 10, 29-45. Beckage N.E. (1990b) Parasitic effects on host development In New Directions in Biological Control: Alternatives for Suppressing Agricultural Pests and Diseases. pp. 497-515. Alan R. Liss, Inc., New York. Beckage N.E. and Templeton T.J. (1986) Physiological effects of parasitism by Apanteles conqreaatus in terminal-stage tobacco hornworm larvae. J. Insect Phvsiol. 32, 299-314. Bligh E.G. and Dyer W.J. (1959) A rapid method of lipid extraction and purification. Can. J. Biochem. Phvsiol. 37, 912-917. Bracken G.K. and Barlow J.S. (1967) Fatty acid composition of Exeristes comstockii reared on different foods. Can. J. Zool. 45, 57-61. Bradford M.M. (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein dye binding. Analvt. Biochem. 72, 248-254. Brown J.J. and Reed-Larsen D. (1991) Ecdysteroids and insect host/parasitoid interactions. Biol. Control 1, 136-143. Brown J.J., Reed-Larsen D. and Ahl J. (1990) Physiological relationship between a diapausing endoparasitoid fAscooaster guadridentata^ and its dormant host fCvdia pomonella^. Arch. Insect Biochem. Phvsiol. 13, 229-238. Bucher T., Czok R., Lamprecht W. and Latzko E. (1963) Methods of enzymatic analysis (Edited by Bergmeyer H.U.), pp. 324-327. Academic Press, New York. 127 Chabora P.C. and Pimentel D. (1966) Effect of host (Musca domestica Linnaeus) age on the pteromalid parasite Nasonia vitripennis (Walker). Can. Ent. 98, 1226-1231. charnov E.L. (1979) The genetical evolution of patterns of sexuality: Darwinian fitness. &n. Nat. 113, 465-480. Charnov E.L. (1982). The theory of sex allocation, Princeton University Press, Princeton. Clausen C.P. (1939) The effect of host size upon the sex ratio of hymenopterous parasites and its relation to methods of rearing and colonization. J* N. entomol. S o c . 47, 1-9. Cornell H. and Pimentel D. (1978) Switching in the parasitoid Nasonia vitripennis and its effect on host competition. Ecol. 59, 297-303. Coudron T . A . , Kelly T.J. and Puttier B. (1990) Developmental responses of Trichoplusia ni (Lepidoptera: Noctuidae) to parasitism by the ectoparasite Euplectrus plathvpenae (Hymenoptera: Eulophidae). Arch. Insect Biochem. Phvsiol. 13, 83-94. Coudron T.A. and Puttier B. (1988) Response of natural and factitious hosts to the ectoparasite Euplectrus plathvpenae (Hymenoptera: Eulophidae). Ann. Entomol. Soc. Am. 81, 931-937. Dahlman D.L. (1970) Trehalose levels in parasitized and non-parasitized tobacco hornworm, Manduca sexta. larvae. Ann. Entomol. Soc. Am. 63, 615-617. Dahlman D.L. (1975) Trehalose and glucose levels in hemolymph of diet-reared, tobacco leaf-reared and parasitized tobacco hornworm larvae. Comp. Biochem. P h v s i o l . 50A, 165-167. Dahlman D.L. (1990) Evaluation of teratocyte function: an overview. Arch. Insect Biochem. Phvsiol. 13, 159-166. Dahlman D.L. and Herald F. (1971) Effects of the parasite, Apanteles conareaatus. on respiration of tobacco hornworm, Manduca sexta. larvae. Comp. Biochem. P h v s i o l . 40A, 871-880. 128 Darling D.C. and Werren J.H. (1990) Biosystematics of Nasonia (Hymenoptera: Pteromalidae): two new species reared from bird's nests in North America. Ann. Entomol. Soc. Am. 83, 352-370. Dawei H. and Dingxi L. (1987) Studies on the behavior of Nasonia vitripennis (Walker). Sinozooloaia 5, 177-182. Denlinger D.L. (1972a) Induction and termination of pupal diapause in Sarcophaaa (Diptera: Sarcophagidae). Biol. B u l l . 142, 11-24. Denlinger D.L. (1981) Hormonal and metabolic aspects of pupal diapause in Diptera. Ent. Gen. 7, 245-259. Denlinger D.L., Willis J.H. and Fraenkel G. (1972b) Rates and cycles of oxygen consumption during pupal diapause in Sarcophaga flesh flies. J. Insect Phvsiol. 18, 871-875. Dover B.A., Davies D.H., Strand M.R., Gray R.S., Keeley L.L. and Vinson S.B. (1987) Ecdysteroid-titre reduction and developmental arrest of last-instar Heliothis virescens larvae by calyx fluid from the parasitoid Campoletis sonorensis. J. Insect Phvsiol. 33, 333-338. Drenth D. (1974) Susceptibility of different species of insects to an extract of the venom gland of the wasp Microbracon hebetor (Say). Toxicon 12, 189-192. Dunn P.E. (1986) Biochemical aspects of insect immunology. Annu. Rev. Entomol. 31, 321-339. Edwards J.S. and Sernka T.J. (1969) On the action of Bracon venom. Toxicon 6 , 303-305. Edwards R.L. (1954a) The host-finding and oviposition behaviour of Mormoniella vitripennis (Walker) (Hymenoptera: Pteromalidae), a parasite of muscoid flies. Behaviour 7, 88-112. Edwards R.L. (1954b) The effect of diet on egg maturation and resorption in Mormoniella vitripennis (Hymenoptera: Pteromalidae). Quart. <1. Micr. Sci. 95, 459-468. 129 Espelie K. and Brown J.J. (1990) Cuticular hydrocarbons of species which interact on four trophic levels: apple, Malus pumila: codling moth, Cydia pomonella L. ? a hymenopteran parasitoid, Ascogaster quadridentata Wesmael; and a hyperparasite, Perilampus fulvicornis Ashmead. Comp. Biochem. Phvsiol. 95B, 131-136. Ferkovich S.M. and Dillard C. (1986) A study of the radiolabeled host proteins and protein synthesis during development of the eggs of the endoparasitoid, Microolitis croceipes (Cresson) (Braconidae). Insect Biochem. 16, 337-345. Finney D.J. (1971) Probit Analysis (3rd edition). Cambridge Univesity Press. Fisher R.C. (1971) Aspects of the physiology of endoparasitic Hymenoptera. Biol. Rev. 46, 243-278. Flanders S.E. (1935) An apparent correlation between the feeding habits of certain pteromalids and the condition of their ovarian follicles. Ann. Entomol. Soc. Am. 28, 438-444 . Fraenkel G. and Hsiao C. (1968) Morphological and endocrinological aspects of pupal diapause in the flesh fly, Sarcophaqa aravrostoma (Robineau-Desvoidy). J. Insect Phvsiol. 14, 707-718. Hardy I.C.W., Griffiths N.T. and Godfray H.C.J. (1992) Clutch size in a parasitoid wasp: a manipulation experiment. J. Anim. Ecol. 61, 121-129. Heinz K.M. and Parrella M.P. (1989) Attack behavior and host size selection by Diglyphus beqini on Liriomvza trifolii in chrysanthemum. Entomol. exp. appl. 53, 147-156. Hink W.F. (1970) Established insect cell line from the cabbage looper, Tricholousia ni. Nature. Lond. 226, 466-467. Hohorst H-J. and Reim M. (1963) Methods of enzvmatic analysis (Edited by Bergmeyer H.U.), pp. 33 5-3 39. Academic Press, New York. Jankovic-Hladni M.I. (1991) Hormones and metabolism in insect stress. In Hormones and metabolism in insect stress (Eds Ivanovic J. and Jankovic-Hladni M.), pp. 5- 23. CRC Press, Boca Raton. 130 Jones D. , Jones G. , Rudnicka M. , Click A. , Reck- Malleczewen and Ivaya M- (1986) Pseudoparasitism of host Trichoplusia ni by ChSlQTms spp. as a new model system for parasite regulation of host physiology. J. Insect Phvsiol. 32, 315-328. Jones D. and Leluk J. (1990) Venom proteins of the endoparasitic wasp Che1onus near curvimaculatus: characterization of the major components. Arch. Insect B i o chem. Phvsiol. 13, 95-106. Jones R.L. and Lewis W.J. (1971) Physiology of the host- parasite relationship between Heliothis zea and Microplitis croceipes. J. Insect Physiol. 17, 921-927. King B.H. (1991a) A field study of host size effects on sex ratio of the parasitoid wasp Spalanoia cameroni. Am. Midi. Nat. 125, 10-17. King B.H. and skinner S.W. (1991b) Proximal mechanisms of the sex ratio and clutch size responses of the wasp Nasonia vitripennis to parasitized hosts. Anim. Behav. 42: 23-32. King P.E. (1962) The effect of resorbing eggs upon the sex ratio of the offspring in Nasonia vitripennis (Hymenoptera: Pteromalidae). J. exp. Biol. 39, 161-165. King P.E. (1963 The rate of egg resorption in Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae) deprived of hosts. Proc. R. Ent. Soc. Lond. A 38, 98-100. Klomp H.B. and Teerink J. (1962) Host selection and number of eggs per oviposition in the egg parasite Trichoaramma embrvophaaum Htg. Nature 195, 1020-1021. Krell P.J. (1991) The polydnaviruses: multipartite DNA viruses from parasitic Hymenoptera. In Viruses of invertebrates (Ed Kurstak E.), pp. 141-177. Marcel Dekker Inc., New York. Lai-Fook J. (1966) The repair of wounds in the integument of insects. J. Insect Phvsiol. 12, 195-226. Lawrence P.O. (1986) Host-parasite hormonal interactions: an overview, £. Insect Phvsiol. 32, 295-298. 131 Lawrence P.O. (1988) Ecdysteroid titers and integument changes in superparasitized puparia of Anastrepha suspensa (Diptera: Tephritidae). J. Insect Phvsiol. 34, 603-608. Legner E.F. (1967) The status of Nasonia vitripennis as a natural parasite of the house fly, Musca domestica. Can. E n t . 99, 308-309. Lewis L.C., Mutchmor J.A. and Lynch R.E. (1971) Effect of Perezia pvraustae on oxygen consumption by the European corn borer, Ostrinia nubilalis. j. Insect Physiol. 17, 2457-2468. Miranpuri G.S., Erlandson M.A., Gillespie J.P. and Khachatourians G.G. (1992) Changes in hemolymph of the migratory grasshopper, Melanoolus sanauinipes. infected with an entomopoxvirus. J. Invertebr. Pathol. 60, 274-282. Moore S.D. (1989) Regulation of host diapause by an insect parasitoid. Ecol. Entomol. 14, 9 3-98. Ohgushi R. (1959) The longevity, egg laying capacity, and the number of hosts parasitized by one female of Mormoniella vitripennis (Walker). (Hymenoptera: Pteromalidae) reared in three species of fly puparia. Insect Ecol. 8 , 46-57. Orzack S.H. and Parker E.D. Jr. (1986) Sex-ratio control in a parasitic wasp, Nasonia vitripennis. I. Genetic variation in facultative sex-ratio adjustment. Evolution 40, 331-340. Pennacchio F., Vinson S.B. and Trembly E. (1992) Host regulation effects on Heliothis virescens (F.) larvae induced by teratocytes of Cardiochiles niariceps Viereck (Lepidoptera, Noctuidae-Hymenoptera, Braconidae). Arch. Insect Biochem. Phvsiol. 19, 177-192. Pimentel D., Nagel w.p. and Madden J.L. (1962) Space-time structure of the environment and the survival of parasite-host systems. &m. Naturalist 97, 141-167. Purrington F.F. and Uleman J.S. (1972) Brood size of the parasite wasp Hvssopus thymus (Hymenoptera: Eulophidae): functional correlation with the mass of a cryptic host. Ann. Entomol. Soc. Am. 65, 280-281. 132 Quistad G.B., Skinner W.S and Schooley D.A. (1988) Venoms of social Hymenoptera-toxicity to the lepidopteran, Manduca sexta. Insect Biochem. 18, 511-514. Quistad G.B., Dennis P.A. and Skinner W.S. (1992) Insecticidal activity of spider (Araneae), centipede (Chilopoda), scorpion (Scorpionida), and snake (Serpentes) venoms. J . Econ. Entomol. 8 5 , 33-3 9. Ratcliffe N.A. and King P.E. (1967) The venom system of Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae). Proc. R. ent. Soc. Lond. (A). 42, 49-61. Ratcliffe N.A. and King P.E. (1969) Morphological, ultrastructure, histochemical, and electrophoretic studies on the venom system of Nasonia vitripennis walker (Hymenoptera: Pteromalidae). Morph. 127, 177-204. Rivers D.B., Hink W.F. and Denlinger D.L. (1993) Toxicity of the venom from Nasonia vitripennis (Hymenoptera: Pteromalidae) toward fly hosts, nontarget insects, different developmental stages, and cultured insect cells. Toxicon (In press). Roubaud E. (1917) Observations biologiques sur Nasonia brevicornis Ashm. Chalcidide parasite des pupes de muscides. Bull. sci. France et Bela. 11, 21-32. Rutherford T.A. and Webster J.M. (1978) Some effects of Mermis niarescens on the hemolymph of Schistocerca areaaria. Can. £. Zool. 56, 339-347. Rutz D.A. and Scoles G.A, (1989) Occurrence and seasonal abundance of parasitoids attacking muscoid flies (Diptera: Muscidae) in caged-layer poultry facilities in New York. Environ. Entomol. 18, 51-55. Salt G. (1961) Competition among insects parasitoids. SvrnP. S o c . E x p . B i o l . 15, 96-119. Saunders D.S. (1965) Larval diapause of maternal origin: Induction of diapause in Nasonia vitripennis (Walk.) (Hymenoptera: Pteromalidae). J. exp. Biol. 42, 495-508. Saunders D.S. (1975) Spectral sensitivity and intensity thresholds in Nasonia photoperiodic clock. Nature 253, 732-734. 133 Saunders D.S., Sutton D. and Jarvis R.A. (1970) The effect of host species on diapause induction in Nasonia vitripennis. J. Insect FhYgiPl• 405-416. Schmidt o . and Schuchmann-Feddersen I. (1 9 8 9 ) Role of virus-like particles in parasitoid-host interaction of insects. In Subcellular Biochemistry (Ed Harris J.R.), Vol. 1 5 , pp. 9 1 - 1 1 9 . Plenum Press, New York. Schneiderman H.A. and Horowitz J. (1958) The induction and termination of facultative diapause in the chalcid wasps Mormoniella vitripennis (Walker) and Tritneptis kluaii (Ratzeburg). J. exp. Biol. 35, 520-551. Shaw M.R. (1981) Delayed inhibition of host development by the nonparalyzing venoms of parasitic wasps. £. Invertebr. Pathol. 37, 215-221. Slama K. and Denlinger D.L. (1992) Infradian cycles of oxygen consumption in diapausing pupae of the flesh fly, Sarcophaaa crassipalois. monitored by a scanning microrespirographic method. Arch. Insect Biochem. PhVSiol. 20, 135-143. Smith G.J. (1969) Host selection and oviposition behavior of Nasonia vitripennis (Hymenoptera: Pteromalidae) on two host species. Can. Ent. 101, 533-538. Smith G.J.C. and Pimentel D. (1969) The effect of two host species on the longevity and fertility of Nasonia vitripennis (Walk.). Ann. Entomol. Soc. Am. 62, 305-308. Sokal R.R. and Rohlf F.J. (1969) Biometry. w. H. Freeman and Company, San Francisco. St. Leger R.J. (1991) Integument as a barrier to microbial infections. In Physiology of the insect epidermis (Eds Binnington K. and Retnakaran A.), pp. 284- 3 06. CSIRO, Melbourne. Stewart L.M.D., Hirst M., Ferber M.L., Merryweather A.T., Cayley P.J. and Possee R.D. (1991) Construction of an improved baculovirus insecticide containing an insect- specific toxin gene. Nature 352, 85-89. 134 stolz D.B. and Guzo D. (1986) Apparent hemocytic transformations associated with parasitoid-induced inhibition of immunity in Melacosoma disstria larvae. J. Insect Phvsiol. 32, 377-388. Strand M.R. (1984) The physiological interactions of parasitoids with their hosts and their influence on reproductive strategies. In Insect Parasitoids (Eds Waage J. and Greathead D.), pp. 97-136. Academic Press, L ondon. Strand M.R., Baehrecke E.H. and Wong E.A. (1991) The role of host endocrine factors in the development of polyembryonic parasitoids. Biol. Control 1 , 144-152. Strand M.R. and Dover B.A. (1990) Developmental disruption of Pseudoolusia includens and Heliothis virescens larvae by the calyx fluid and venom of Microolitis demolitor. Insect Bjoghem- Phvsiol. 18, 131-145. Stubblefield R.D., Bennet G.A., Shotwell O.L., Hall H.H. and Jackson R.w. (1966) Organic acids in the hemolymph of healthy and diseased Pooillia iaponica (Newman) larvae. J. Insect Phvsiol. 12, 949-956. Takagi M. (1985) The reproductive strategy of the gregarious parasitoid, Pteromalus puparum (Hymenoptera: Pteromalidae) 1. optimal number of eggs in a single host. Oecoloqia (Berlin) 6 8 , 1-6. Takagi M. (1986) The reproductive strategy of the gregarious parasitoid, Pteromalus puoarum (Hymenoptera: Pteromalidae) 2. host size discrimination and regulation of the number and sex ratio of progeny in a single host. Oecoloqia (Berlin) 70, 321-325. Takahashi M., Mitsuhashi, J. and Ohtaki T. (1980) Establishment of a cell line from embryonic tissues of the fleshfly, Sarcophaaa perearina (Insecta: Diptera). Develop.. Growth and Differ. 22, 11-19. Taylor T.H.C. (1937) The biological control of an insect in Fiii. The Imperical Institute of Entomology, London. Thompson S.N. (1983) Biochemical and physiological effects of metazoan endoparasites on their host species. Comp. Biochem. Phvsiol. 74B, 183-211. 135 Thompson S.N. (1986a) Effects of the insect parasite Hvposter exioruae (Viereck) on the carbohydrate metabolism of its host, Trichoplusia ni (Hubner). J. insect Phvsiol. 32, 287-293. Thompson S.N. (1986b) The metabolism of insect parasites (parasitoids): an overview. £. Insect Phvsiol. 32, 421-423. Thompson S.N. and Barlow J.S. (1976) Regulation of lipid metabolism in the insect parasite Exeristes roborator (Fabricius). J. Parasitol. 62, 303-306. Thompson S.N. and Binder B.F. (1984) Altered carbohydrate levels and gluconeogenic enzyme activity in Trichoplusia ni parasitized by the insect parasite, Hvposter exiauae. J- Parasitol. 70, 644-651. Thompson S.N., Lee R. W.-K. and Beckage N.E. (1990) Metabolism of parasitized Manduca sexta examined by nuclear magnetic resonance. Arch. Insect Biochem. Phvsiol. 13, 127-143. Thompson w.R. (1958) A catalogue of the parasites and predators of insect pests: Part 5 Hosts of the Hymenoptera. Lepidoptera. Mid gtrepsjptera, pp. 583-624. Commonwealth Institute of Biological Control, Ottawa. Tomalski M.D. and Miller L.K. (1991) Insect paralysis by baculovirus-mediated expression of a mite neurotoxin gene. Nature 352, 82-84. Uematsu H. and Sakanoshita A. (1987) Effects of venom from an external parasitoid, Euplectrus kuwanae (Hymenoptera: Eulophidae) on larval ecdysis of Aravroaramma albostriata (Lepidoptera: Noctuidae). Ap p . Ent. Zool. 22, 139-145. Van Den Assem J., Putters F.A. and Prins T.C. (1984) Host quality effects on sex ratio of the parasitic wasp Anisopteromalus calandrae (Chalcidoidea, Pteromalidae). Neth. J. Zool. 34, 33-62. Van Der Merve J.S. (1943) Investigations on the biology and ecology of Mormoniella vitripennis (Walker). J. Ent. S££. £2 - A f E j s a 6 , 49-64. Van Handel E. (1965) Estimation of glycogen in small amounts of tissue. Analvt. Biochem. 11, 256-265. 136 Van Handel E. (1985a) Rapid determination of glycogen and sugars in mosquitoes. J. Mosa. Control Assoc. 1, 299-301. Van Handel E. (1985b) Rapid determination of total lipids in mosquitoes. &in. Mosa. Control Assoc. 1, 302-304. Vinson S.B. and Barras D.J. (1970) Effects of the parasitoid, Cardiochiles niariceps, on the growth, development, and tissues of Heliothis virescens. j. Insect Phvsiol. 16, 1329-1338. Vinson S.B. and Iwantsch G.F. (1980a) Host regulation by insect parasitoids. Quart. Rev. Biol. 55, 143-165. Vinson S.B. and Iwantsch G.F. (1980b) Host suitability for insect parasitoids. Annu. Rev. Entomol. 25, 397-419. Von Brand T. (1979) Pathophysiology of the host. In Biochemistry and physiology of endoparasites. pp. 321- 3 90. Elsevier/North Holland Biomedical Press, Amsterdam. Waage J.K. (1986) Family planning in parasitoids: adaptive patterns of progeny and sex allocation. In Insect Parasitoids (Eds Waage J. and Greathead D.), pp. 63-96. Academic Press, London. Wani M., Yagi S. and Tanaka T. (1990) Synergistic effect of venom, calyx and teratocytes of Apanteles karivai on the inhibition of larval pupal ecdysis of the host, Pseudaletia separata. Entomol. exp. appl. 57, 101-104. Webb B.A. and Dahlman D.L. (1985) Developmental pathology of Heliothis virescens larvae parasitized by Microplitis croceipes: parasite-mediated host developmental arrest. Arch. Insect Biochem. Phvsiol. 2, 131-143. Webb B.A. and Dahlman D.L. (1986) Ecdysteroid influence on the development of the host Heliothis virescens and its endoparasite Microolitis croceipes. J. Insect Phvsiol. 32, 339-345. Werren J.H. (1980) Sex ratio adaptations to local mate competition in a parasitic wasp. Science 208, 1157-1159. Werren J.H. (1983) Sex ratio evolution under local mate competition in a parasitic wasp. Evol. 37, 116-124. 137 Werren J.H. (1984a) A model for the sex ratio selection in parasitic wasps: local mate competition and host quality effects. Neth. Zool. 34, 81-96. Werren J.H. (1984b) Brood size and sex ratio regulation in the parasitic wasp Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae). Neth. £. Zool. 34, 123-143. Whitfield J.B. (1990) Parasitoids, polydnaviruses and endosymbioses. Parasitol. Today 6, 381-384. Whiting A.R. (1967) The biology of the parasitic wasp Mormoniella vitripennis (= Nasonia brevicornis) (Walker). Quart. Rev. Biol. 42, 333-406. Williamson D.H. (1974) L-Alanine determination with alanine dehydrogenase. In Methods of enzvmatic analysis (Ed Bergmeyer H.U.), Vol. 4, pp. 1679-1682. Academic Press Inc., New York. Wladimirov M. and Smirnov E. (1934) Uber das Verhalten der Schlupfwespe Mormoniella vitripennis Wlk zu verschiedenen Fliegenarten. Womersley C. and Platzer E.G. (1984) A comparison of tricarboxylic acid cycle intermediates in the haemolymph of healthy Culex pjpiens. Aedes taeniorhvnchus. and Anopheles ouadrimaculatus larvae and larvae parasitised by Romanomermis culicivorax. Insect Biochem. 14, 401-406. Wylie H.G. (1962) An effect of host age on female longevity and fecundity in Nasonia vitripennis (Walk.) (Hymenoptera: Pteromalidae). Can. Ent. 94, 990-993. Wylie H.G. (1963) Some effects of host age on parasitism by Nasonia vitripennis (Walk.) (Hymenoptera: Pteromalidae). C a n . E n t . 95, 881-886. Wylie H.G. (1967) Some effects of host size on Nasonia vitripennis and Muscidifurax raptor (Hymenoptera: Pteromalidae). Can. Ent. 99, 742-748. Zdarek J. and Denlinger D.L. (1975) Action of ecdysoids, juvenoids, and non-hormonal agents on termination of pupal diapause in the flesh fly. 3_. Insect Phvsiol. 21, 1193-1204. 138 Zdarek J. and Fraenkel G. (1987) Pupariation in flies: a tool for monitoring effects of drugs, venoms, and other neurotoxic compounds. Arch. Insect Biochem. Phvsiol. a , 29-46. Zlotkin E. (1983) Insect selective toxins derived from scorpion venoms: an approach to insect neuropharmacology. Insect Biochem. 13, 219-236.