24 Rivers et al.

Archives of Insect Biochemistry and Physiology 61:24–41 (2006)

Characterization and Biochemical Analyses of Venom

From the Ectoparasitic Wasp Nasonia vitripennis
(Walker) (Hymenoptera: Pteromalidae)

David B. Rivers,1* Fevzi Uckan,2 and Ekrem Ergin2

During parasitism, the ectoparasitic wasp Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae) induces a developmental
arrest in host pupae that is sustained until the fly is either consumed by developing larvae or the onset of death. Bioassays
using fluids collected from the female reproductive system (calyx, alkaline gland, acid gland, and venom reservoir) indicated
that the venom gland and venom reservoir are the sources of the arrestant and inducer(s) of death. Infrared spectroscopic
analyses revealed that crude venom is acidic and composed of amines, peptides, and proteins, which apparently are not
glycosylated. Reversed phase high performance liquid chromatography (HPLC) and sodium dodecyl polyacrylamide gel electro-
phoresis (SDS-PAGE) confirmed the proteinaceous nature of venom and that it is composed mostly of mid to high molecular
weight proteins in the range of 13 to 200.5 kilodaltons (kDa). Ammonium sulfate precipitation and centrifugal size exclusion
membranes were used to isolate venom proteins. SDS-PAGE protein profiles of the isolated venom fractions displaying biologi-
cal activity suggest that multiple proteins contribute to arresting host development and eliciting death. Additionally, HPLC
fractionation coupled with use of several internal standards implied that two of the low molecular weight proteins were
apamin and histamine. However, in vitro assays using BTI-TN-5B1-4 cells contradict the presence of these agents. Arch. Insect
Biochem. Physiol. 61:24–41, 2006. © 2005 Wiley-Liss, Inc.

KEYWORDS: wasp venom; infrared spectroscopy; HPLC; SDS-PAGE; size exclusion

INTRODUCTION mental arrest (Beard, 1963; Coudron, 1991; Piek

et al., 1982; Rivers and Denlinger, 1994a; Rivers et
Venoms produced by parasitic Hymenoptera al., 2002a; Strand and Pech, 1995; Whitfield and
possess unique regulatory agents that functionally Asgari, 2003). In certain endoparasitic species,
aid in subduing an insect host. These venom com- other maternally derived secretions (e.g., virus-like
ponents are known to facilitate host invasion particles, polydnaviruses, or ovarian proteins) are
through induction of paralysis, suppression of host also needed for successful parasitism (Beckage,
immune responses, stimulation or augmentation 1998; Webb and Luckhart, 1994; Vinson, 1990).
of the activity of other parasitoid factors (e.g., en- Regardless of the source of regulatory material, only
dosymbiotic viruses), and/or initiation of develop- a rudimentary understanding of the mode of ac-

1
Department of Biology, Loyola College, Baltimore, Maryland
2
Department of Biology, Faculty of Science-Literature, Balikesir University, Balikesir, Turkey
Contract grant sponsor: USDA-NRICGP; Contract grant number: 2001-1005; Contract grant sponsor: Loyola College, Baltimore, Maryland.
Abbreviations used: HPLC = high performance liquid chromatography; I.R. = infrared spectroscopy; kDa = kilodalton; LC99 = lethal concentration to kill
100% of population; LD = light-dark; MWCO = molecular weight cutoff; SDS-PAGE = sodium dodecyl sulfate polyacrylamide gel electrophoresis; TFA =
trifluoroacetic acid; VRE = venom reservoir equivalent.
*Correspondence to: Dr. David Rivers, Department of Biology, Loyola College in Maryland, 4501 North Charles Street, Baltimore, MD 21210.
E-mail: [email protected]
Received 24 February 2005; Accepted 21 July 2005

© 2005 Wiley-Liss, Inc. Archives of Insect Biochemistry and Physiology

DOI: 10.1002/arch.20094
Published online in Wiley InterScience (www.interscience.wiley.com)
 Characterization of Venom From N. vitripennis 25

tion or the molecular target sites is available. Con- teins are thought to have an immunosuppressive
sequently, precise pathways triggered by parasitic role (Asgari et al., 2003a,b; Parkinson et al., 2001,
wasp venoms remain unclear. 2003) or even induce host castration (Digilio et
The availability of specific venom proteins re- al., 2000).
sponsible for disruption of host development and Nasonia vitripennis (Walker) (Hymenoptera:
physiology as well as inducers of death would Pteromalidae), like other species of parasitoids, has
greatly enhance efforts aimed at understanding the been the subject of several studies focused on de-
mechanisms of action of regulatory proteins found ciphering how the wasp functions in the parasitic
within venoms. Unfortunately, only a limited num- relationship (reviewed by Rivers et al., 1999a). This
ber of studies have attempted to characterize the intensively studied wasp has a well-developed
composition and biochemical properties of para- venom system, which produces a proteinaceous
sitic wasp venoms (Quicke, 1997). Quistad et al venom in the acid gland and stores the venom in
(1994) identified several paralytic toxins in the active form within a single reservoir (Ratcliffe and
venom from the ectoparasitoid Habrobracon (= King, 1967, 1969). Functionally, venom inhibits
Bracon) hebetor, with at least two of the isolated host cellular immune responses (Rivers et al,
proteins demonstrating high insecticidal activity 2002a), depresses respiratory metabolism within
toward 6 species of lepidopteran larvae. Two mid- 6–8 h (Rivers and Denlinger, 1994b), stimulates
range molecular weight proteins (33 and 52 kDa) increases in lipid levels within hemolymph and fat
have been isolated and sequenced from the endo- body (Rivers and Denlinger, 1995; Rivers et al.,
parasitoid Chelonus near circumaculatus (Jones et al., 1998), and ultimately induces death. In muscoid
1992), and though the 52-kDa protein shares a flies, N. vitripennis venom can retard adult fly mo-
high degree of homology with an insect chitinase bility within 1–2 h when injected artificially, and
(Krishnan et al., 1994), no functional role in para- the flies will succumb to death in less than 24 h
sitism has been established. By contrast, a 66-kDa (Beard, 1964; Rivers et al., 1993). The venom has
protein was isolated from the ectoparasitic wasp been shown to be highly active toward several life
Euplectrus comstockii. When injected into host lar- stages (wandering larvae, pupae, pharate adults,
vae, it was capable of inducing developmental ar- adults) of flies from at least 4 families (Muscidae,
rest that resembled natural parasitism (Coudron Drosophilidae, Calliphoridae, and Sarcophagidae)
and Brandt, 1996). Molecular and biochemical (Rivers et al., 1993). Additionally, in vitro assays
analyses of two wasp species (Pimpla hypochondriaca using cultured insect cells have revealed venom
and P. turionella) from the genus Pimpla suggest stimulation of G-protein dependent signal trans-
that these endoparasitoid venoms possess a large duction pathways that promote mobilization of
and diverse number of enzymes, including laccase, intracellular calcium from mitochondria and en-
serine protease, reprolysin-like metalloprotease, doplasmic reticulum, elevations in cAMP, and cell
phospholipases, and phenoloxidases (Parkinson et death via an oncotic lytic mechanism (Rivers et
al., 2001, 2002a,b, 2003; Uçkan et al., 2004). al., 2002b, 2005).
Uçkan et al. (2004) also observed the presence of The diversity of biological functions ascribed to
noradrenaline, apamin, and melittin in venom this wasp venom argues that multiple venom pro-
from P. turionella, consistent with the paralytic ac- teins or factors are involved in subduing the fly
tion of the venom in multiple life stages of lepi- host, a feature that would be consistent with most
dopteran hosts (Kansu and U—ur, 1984). Similarly, parasitoid venoms that have been at least partially
an aspartylglucosaminidase-like protein has been purified (Digilio et al., 2000; Doury et al., 1997;
identified in the venom of Asobara tabida (Moreau Jones et al., 1992; Parkinson et al., 2002a). How-
et al., 2004), with a speculative role in host pa- ever, despite the accumulation of evidence that has
ralysis through induction of aspartate-dependent shed light on how venom from N. vitripennis op-
excitatory pathways. Other identified venom pro- erates in vivo and in vitro, scarce information ex-

January 2006
26 Rivers et al.

ists on the chemical composition of the venom, vented glass jar (1 liter) with 1–2 ml tap water.
or on how many different proteins may be involved Larvae were held under these conditions for 3 days
in manipulating the host condition. This lack of at 25°C with frequent (3–5 times/day) water
knowledge has contributed to the precise target tis- changes. This “wet” treatment temporarily inhib-
sues in the host remaining obscure, as have key its the release of ecdysteroids until the larvae are
aspects of the cellular pathways associated with placed in dry conditions, thereby synchronizing
host biochemical changes, developmental disrup- pupariation (Ohtaki, 1966).
tion, and death. A necessary next step to address
these issues is to characterize and identify venom Cell Culture
components responsible for the observed host al-
terations. This study was aimed at biochemical BTI-TN-5B1-4 cells (embryos from T. ni) (Davies
analyses of venom components through high per- et al., 1993) (also called High Five™) were pur-
formance liquid chromatography, infrared spectros- chased from Invitrogen (San Diego, CA) and grown
copy, and protein gel electrophoresis (SDS-PAGE). in TC-100 (Sigma Chemical Co., St. Louis, MO)
We also attempted to isolate active venom proteins containing 10% fetal bovine serum (FBS) (Sigma)
by using ammonium sulfate precipitation and size at 27°C.
exclusion centrifugal membranes. Isolated venom
fractions were then further analyzed electrophoreti- Isolation of Crude Wasp Venom and Other
cally and through in vivo and in vitro biological Reproductive Tissues
assays.
Unless otherwise indicated, crude venom from
MATERIALS AND METHODS N. vitripennis was isolated from host-fed females
in phosphate-isolation buffer [10 mM sodium
Insect Rearing phosphate (pH 8.0), 0.9% (w/v) NaCl, 15% (w/
v) sucrose, 1 mM ethylenediamineteraacetic acid,
A laboratory colony of N. vitripennis was main- and 1 mM phenylmethylsulfonyl fluoride] (Rivers
tained on pupae and pharate adults of the flesh et al., 1993) and stored frozen at –70°C. Venom
fly, Sarcophaga bullata, as described previously (Riv- activity was confirmed in vitro and in vivo with
ers and Denlinger, 1994a). Adults and larvae were BTI-TN-5B1-4 cells and young pharate adults (5
reared under a 15:9 h light-dark cycle at 25°C. days after pupariation at 25°C) of S. bullata, re-
Twenty to thirty females (3–7 days after emergence spectively, as described previously (Rivers et al.,
from host puparia) were placed in a plastic con- 1993).
tainer (15 × 100 mm) with 30–50 nondiapausing Fluids from calyx tissue, alkaline glands, venom
pupae (4 days after pupariation at 25°C) of S. glands (acid gland), and venom reservoirs were col-
bullata and a 50% (v/v) honey solution. After 24 lected essentially as described above for crude
h, the adult wasps were removed and parasitized venom isolation. Fine forceps were used to pull
pupae maintained at 25°C, LD 15:9 h. Under these the ovipositor and sheath from host-fed adult fe-
conditions, N. vitripennis develops from egg to males covered in phosphate-isolation buffer. This
adult (eclosion) in 12 days. technique allowed removal of ovaries, oviducts, and
A colony of S. bullata was reared according to accessory glands along with the ovipositor. Specific
Denlinger (1972). Larvae and adults were fed beef glands or tissues were dissected free from the re-
liver throughout development at 25°C with a light- maining tissue using iris scissors and forceps, the
dark cycle of LD 15:9 h. To synchronize fly devel- identity of each tissue was confirmed using the cri-
opment for assessing host age, third instar larvae teria of Ratcliffe and King (1967), and then the
that had begun to wander from food (but prior to tissue was processed as described above. Extract
crop emptying) were collected and placed in a from each tissue was tested for biological activity

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 27

as described below for young pharate adult injec- was maintained at 30°C until complete evapora-
tions and cell venom assays. tion. The dried material was homogeneously
ground with potassium bromide before infrared
Protein Determination spectroscopic analyses were performed at room
temperature using a Perkin-Elmer Spectrum BX-II
Total protein in crude venom was determined infrared spectrometer (Perkin-Elmer, Beaconsfield
colorimetrically at 562 nm using a micro-BCA Pro- Buks, England).
tein Assay kit (no. 23235, Pierce, Rockford, IL). Bo-
vine serum albumin (Sigma) served as the standard. High Performance Liquid Chromatography

Effect of pH on Venom Activity High performance liquid chromatography

(HPLC) was used to separate and identify venom
To determine the optimal pH for maintenance components. Reverse-phase liquid chromatography
of venom activity (lethality), the pH of crude was performed by loading 20 µl of a crude venom
venom was adjusted using buffers of different buff- solution (150 lyopholized venom reservoirs recon-
ering capacities. Crude venom was isolated as de- stituted in 150 µl isolation buffer) onto an AceIII
scribed above, with the exception that venom C18 reverse phase column (12.5 cm by 4.0 mm i.d.;
reservoirs were homogenized in buffers of differ- MAC-MOD Analytical, Chadds, Ford, PA) as de-
ent pH [citrate-phosphate (pH 5–6.8), sodium scribed by Uçkan et al. (2004). The mobile phases
phosphate (pH 7.0–8.0), and Tris-HCl (pH 8.2– were (A) 0.1% trifluoroacetic acid (TFA) in aceto-
9) (Stoll and Blanchard, 1990)], but otherwise the nitrile:water (80:20) and (B) 0.1% TFA in water.
same composition as phosphate isolation buffer. Crude venom was fractioned by linear gradient (5–
The ability of venom in each buffer to elicit death 80%) using mobile phase A at 40 min. The flow-
toward cultured cells and adult flies was assessed rate was maintained at 1.0 ml/min, and the elution
as described. monitored at 280 nm.

Psoralen Treatment Venom Injections

To determine whether endosymbiotic viruses or Young pharate adults (5 days after pupariation
other microorganisms may be harbored in wasp at 25°C) of S. bullata were injected with venom lat-
venom, and thus responsible for venom biologi- eral to the dorsal midline of the thorax, just behind
cal activity, crude venom was incubated with the neck. To prepare pharate adults for injection,
psoralen, a DNA cross-linking agent. The venom- the operculum of each puparium was removed to
psoralen mixture was then placed in a UV-cross- expose the head-neck region, and then pricked with
link oven for 55 min (Stratelink). Under these an insect pin to bleed-off approximately 1–2 µl of
conditions, psoralen has been shown to cross-link hemolymph. The latter was performed to tempo-
viral DNA, thereby preventing viral replication and rarily alleviate the high hemocoelic pressure result-
inhibiting all activity (Strand and Noda, 1991). ing from longitudinal contraction of the body and
shrinkage of the cuticle during puparium forma-
Infrared Spectroscopy tion (Zdarek et al., 1979). Injections were accom-
plished by means of finely drawn glass capillaries
Infrared spectroscopic analysis was performed (Rivers et al., 2004). Mortality was assessed at 24-
essentially as described previously (Uçkan et al., h intervals for 30 days (Rivers et al., 1993). Induc-
2004). In brief, a 50-µl sample of lyopholized tion of developmental arrest was determined using
crude venom, reconstituted in phosphate isolation the criteria of Rivers and Denlinger (1994a).
buffer (final protein concentration = 1.42 µg/µl), In parallel experiments, 48-h-old imagoes (after

January 2006
28 Rivers et al.

eclosion at 25°C) were injected (1 µl/fly) with crude cold acetone, a reagent that promotes protein pre-
venom lateral to the dorsal midline of the abdo- cipitation. The solution was mixed by gentle vor-
men using finely pulled capillaries. Adults injected texing for 30 sec, and then placed at –20°C for
with phosphate isolation buffer served as controls. 24 h to facilitate protein precipitation. The mix-
ture was then centrifuged at 10,000 rpm for 30 min
Venom Assays at 4°C. The protein precipitate that formed was
washed twice in phosphate isolation buffer (pH
BTI-TN-5B1-4 cells were counted with a hema- 8.0) by centrifugation at 7,000 rpm for 10 min
cytometer and seeded (2 × 103 cells/well) into 96- (4°C), followed by re-suspending the pellet in 100
well plates (Falcon) with 100 µl TC-100 containing µl (the original volume of crude venom) phosphate
10% FBS. Cells were grown at 27°C for 2–3 days. isolation buffer. Venom activity was assessed in vivo
Confluent monolayers were washed with PBS (pH and in vitro as described above.
7.4) by removing spent culture media, adding 100 The above method coupled with protein elec-
µl PBS, and then gently rocking the plate for 10–20 trophoresis confirmed that the active components
sec before discarding the saline. After the wash, 100 in crude venom that trigger developmental arrest
µl TC-100 with 10% FBS was added, and wasp and death are proteins. Therefore, in an attempt
venom (0.0003–0.01 VRE/µl) was pipetted into each to isolate these active proteins, ammonium sulfate
well. Cell viability was assessed with trypan blue precipitation was performed. Ammonium sulfate
dye exclusion staining (final concentration was was added to crude venom solutions (1 VRE/µl)
0.04%) as previously described (Rivers et al., 1993). as described by Englard and Seifter (1990) stepwise
HPLC analyses suggested that both apamin and until 100% saturation. The solutions were gently
histamine were present in crude venom. To confirm stirred for 30 min at 4°C, and then centrifuged at
the presence of these agents, venom assays were 12,000g for 10 min at room temperature. The pre-
performed with pure apamin (Sigma) and anti-his- cipitate was washed twice (centrifugation at 7,000
tamine antibodies (Sigma), either alone or in com- rpm for 5 min at room temperature) with phos-
bination with a LC99 dose of crude venom from N. phate isolation buffer, and the final pellet re-sus-
vitripennis (Rivers et al., 1993). Similarly, phenol- pended in 200 µl phosphate isolation buffer. To
oxidases have been identified as major venom com- facilitate de-salting, these solutions were then
ponents of some wasp species (Dani et al., 2003; loaded onto a 10,000 molecular cut-off filter
Parkinson et al., 2001), as well as components of (Amicon, Bedford, MA) and centrifuged at 3,000
parasitoid larval secretions, including N. vitripennis rpm for 30 min at room temperature. The retentate
(Gerling and Legner, 1968; Thompson, 1986; Whit- was removed and adjusted to 50 µl with phosphate
ing, 1967). Consequently, parallel experiments were isolation buffer. The ammonium sulfate superna-
conducted with anti-phenoloxidase antibodies (Re- tants from above were loaded into dialysis mem-
search Diagnostics, Flanders, NJ) either alone or si- branes (10,000 MWCO, Pierce, Rockford, IL) and
multaneously with crude venom. Cell viability was dialyzed with several exchanges of 10 mM PBS (pH
assessed with trypan blue dye exclusion staining (fi- 8.0) overnight at 4°C. Following dialysis, the
nal concentration was 0.04%) as previously de- samples were concentrated using 10,000 MWCO
scribed (Rivers et al., 1993). centrifuge filters (Amicon) as detailed above.
Venom fractions were analyzed by biological as-
Protein Precipitation says and SDS-PAGE.

The proteinaceous nature of crude venom was Size Exclusion Separation of Crude Venom
examined by mixing a crude venom solution (1
VRE/µl) prepared either from isolated venom Size exclusion membranes (centrifuge filters)
glands or venom reservoirs with 3 volumes of ice- were used to estimate the molecular size of venom

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 29

components responsible for host developmental ar- RESULTS

rest and death. Crude venom samples (30–50 ml,
Source of Biological Activity
1.51 µg protein/µl) in phosphate isolation buffer
were loaded onto low-protein binding (cellulose
Upon encountering a fly host in nature, females
acetate) centrifuge membrane filters of different
of N. vitripennis inject secretions into pupae and
molecular weight cut-off exclusion limits [10,000
pharate adults via the ovipositor that evoke a de-
to 100,000 MWCO (Spectrum, Gardena, CA or velopmental arrest and eventual host death (Riv-
Viviscience, Hannover, Germany)]. Filters were cen- ers and Denlinger, 1994a). The source of the
trifuged at 12,000g (4°C) for 30–45 min, or until arrestant and inducer of death appears to be the
the volume had been reduced to approximately 10 venom gland and venom reservoir as evidenced by
µl. Retentates were transferred to a microcentrifuge injection of material from either tissue into young
tube (0.5 ml) and the volume adjusted to 25 µl pharate adults of S. bullata inducing a developmen-
with phosphate isolation buffer. Retentates and fil- tal arrest of the fly reminiscent of natural parasit-
trates were then tested for biological activity in vivo ism that eventually culminated in death (Table 1).
and in vitro as described above, and also subjected Similarly, incubation of BTI-TN-5B1-4 cells with
to SDS-PAGE (below). fluids from venom glands and reservoirs resulted
in cell death (Table 1). Prior to death, the plasma
Protein Electrophoresis and nuclear membranes swelled, which was fol-
lowed by cell lysis, a series of events previously
Sodium dodecyl sulfate polyacrylamide gel elec- shown for crude wasp venom (Rivers et al., 1999b).
trophoresis (SDS-PAGE) was performed by the By contrast, saline, calyx fluid, and fluids from al-
modified Laemmli method described by Garfin kaline glands displayed no biological activity in
(1990) using Tris-HCl pre-cast gels [10% and lin- vivo or in vitro (Table 1).
ear (4–15%)] (BioRad, Hercules, CA) in a Mini
Protean III apparatus (BioRad). After electrophore- Psoralen and Acetone Treatments
sis, gels were placed in fixative [40% methanol (w/
v), 10% trichloroacetic acid (w/v)] for 1 h at room Incubation of psoralen with crude venom, fol-
temperature on a Labnet agitator (Orbit P4), fol- lowed by UV cross-linking did not alter wasp
lowed by Coomassie blue staining (Garfin, 1990). venom’s ability to trigger developmental arrest or
Gel images were captured using a Versadoc image its toxicity toward young pharate adults of S. bullata
analysis system (BioRad) connected to a Macintosh (Table 1).
G-4 computer (Apple) equipped with Quantity When venom (collected from either venom
One 1-D analysis software (V. 4.4, BioRad). Mo- glands or venom reservoirs) was incubated with
lecular weights of protein bands were estimated ice-cold acetone, a precipitate formed. Following
with reference to molecular weight markers (Ka- centrifugation and pellet washes with PBS, the re-
leidoscope standards, BioRad). suspended protein pellet was injected into fly hosts.
This fly treatment induced host arrestment and
Statistical Analyses death consistent with natural parasitism (Table 1).
In contrast, the supernatant resulting from acetone
Means were compared using one- and two-way precipitation displayed no biological activity to-
analysis of variance (ANOVA) and Student New- ward pharate adults of S. bullata.
man-Keul’s multiple comparisons tests using Stat- In parallel experiments, the supernatant and re-
View statistical software (v. 5.01, α = 0.05). suspended pellets were incubated in vitro with
Percentage data was arcsine transformed prior to confluent monolayers of T. ni cells and yielded
analysis. similar results to those observed in vivo: the

January 2006
30 Rivers et al.

TABLE 1. Sources of Insecticidal Activity From Tissues and Fluids of Nasonia vitripennis*
% Response
In vivo In vitro
Source N Developmental arrest Death N Death
a a
Saline (isolation buffer) 45 0 0 9,781 0a
Alkaline gland 45 0a 0a 8,643 0a
Venom gland 45 95.9 ± 2.7b 100b 10,175 100b
Venom reservoir 45 98.7 ± 3.0b 100b 7,421 100b
Calyx fluid 45 0a 0a 7,704 0a
Acetone precipitate
Venom gland 45 78.1 ± 4.6c 87.4 ± 2.2c 10,305 86.3 ± 1.8c
Venom reservoir 45 75.4 ± 6.8c 83.6 ± 1.8c 8,895 88.6 ± 3.2c
Psoralen treatment
Venom gland 45 97.8 ± 4.2b 100b 9,547 100b
Venom reservoir 45 96.0 ± 2.9b 100b 9,036 100b
*Fluids were collected by dissecting tissues from host-fed adult females in isolation buffer [10 mM so-
dium phosphate (pH 8.0) containing 15% (w/v) sucrose, 0.9% (w/v) NaCl, and 1 mM EDTA]. In vivo
assays were performed by injecting samples with finely drawn glass capillaries into the dorsal surface of
adults of the flesh fly, S. bullata. In vitro assays used confluent monolayers of BTI-TN-5B1-4 cells grown in
96-well plates at 27°C.

acetone supernatant was not lethal toward BTI-TN- examined using solutions with differing buffer ca-
5B1-4 cells but the re-suspended precipitate dis- pacities. Toxicity of crude venom toward adult flies
played high cytotoxicity (Table 1). and cultured insect cells was highest at alkaline pH
values at or near pH 8.0 (Fig. 1). As the pH became
Effect of pH on Venom Activity more alkaline than 8.2 or more acidic than 7.4, the
toxicity of venom in vivo declined steadily. In con-
The stability of the arrestment and lethal activ- trast, in vitro lethality decreased much more slowly
ity of crude wasp venom at different pH levels was with increasing or decreasing buffer pH (Fig. 1).

Fig. 1. Effect of pH on insecticidal activ-

ity of crude venom from N. vitripennis. A
LC99 or LD99, respectively, dose of crude
venom was either incubated with con-
fluent monolayers of BTI-TN-5B1-4 cells
at 27°C, or artificially injected into adults
of the flesh fly, S. bullata. Mortality was
assessed 24 h later.

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 31

Total Protein Determination TABLE 2. Analysis of Characteristic Infrared Absorption Bands

Associated With Crude Venom From N. vitripennis*
Frequency (cm–1) Possible assignment
Total protein in crude venom extracted from
venom glands and venom reservoirs was deter- 3,430 N–H stretching, secondary amines
2,930 Asymmetric C–H stretching of CH2
mined colorimetrically using bovine serum albu- 2,362 CO2
min as standards. The mean protein content of 1,645, 1,513 C = O stretching, amides
1,456 C–N stretching
venom reservoirs (X ± SEM = 1.48 ± 0.3 µg, n = 8 1,418 C–O–H in plane bending, carboxylic acids
samples containing 450–500 venom reservoirs in 1,340 O–H bending
1,058 P–O–H bending
phosphate isolation buffer) was significantly higher
996, 928 P–O stretching
(t = 25.7, df = 7, P < 0.001) than that of venom
*Interpretation of infrared absorption bands characteristic of crude venom as well as
glands (0.78 ± 0.2 µg, n = 8). However, this may band assignments are based on 3 replicates of i.r. spectroscopic analysis using 50-
simply reflect the difference in tissue size, and µl lyopholized samples of crude venom reconstituted in isolation buffer (Fig. 1).
hence fluid volume, between glands and reservoirs.

Infrared Spectroscopy
expected for proteinaceous venoms (Stuart, 1997).
Infrared spectral analyses were performed to char- The observed band at 2,930 cm–1 is characteristic of
acterize the peptide and protein composition of alkanes, while the vibration at 1,418 cm–1 reflects
wasp venom. The resulting infrared spectrum of the acidic (carboxylic) nature of venom (Leonard,
crude venom from N. vitripennis is shown in Figure 1972). Absorption bands at 1,058, 996, and 928
2 with interpretation of characteristic absorption cm–1 suggest that some venom components, possi-
bands in Table 2. Absorption bands at 3,430, 1,645, bly enzymes, contain phosphorous. The noticeable
and 1,513 cm–1 are consistent with the presence of absence of bands at 3,600 (OH peak), 2,900 (C-H
secondary amine and amide groups, as would be stretching), and 1,700 (C=O stretching) cm–1 (Fig.

Fig. 2. Infrared radiation spectroscopic analysis of crude in isolation buffer (pH 8.0) were used for i.r. analysis.
venom from N. vitripennis. Fifty microliters of crude venom

January 2006
32 Rivers et al.

2) indicates that the contents of venom reservoirs (Fig. 4). Comparisons made to the retention times
lack a carbohydrate moiety (Fritz and Schenk, 1979; of several internal standards suggest that apamin
Stuart, 1997). (Fig. 3, t = 9.90 min, peak 4) and histamine (Fig.
3, t = 25.38 min, peak 7) are major components
HPLC Analysis of crude venom. However, by comparison to crude
venom alone, confluent monolayers of BTI-TN-
Crude venom isolated from venom reservoirs 5B1-4 cells displayed very little sensitivity to pure
was fractioned using a reversed phase C18 column. apamin, and this toxin did not augment venom
Based on the absorption peaks monitored at 280 activity when co-incubated with T. ni cells (Table
nm, venom appears to be a complex mixture of 3). Similarly, incubation of crude venom with
peptides and proteins (Fig. 3), which was further anti-histamine or anti-phenoloxidase did not
supported by SDS-PAGE profiles of crude venom lower the toxicity of venom toward cultured cells,

Fig. 3. Fractionation of crude venom (150 venom reser- an injection volume of 20 µl. Absorbance was monitored
voir equivalents of lyophilized venom dissolved in 150 µl at 280 nm for (A) crude venom and (B) standards (1)
isolation buffer) from N. vitripennis by reversed phase HPLC noradrenaline, 0.7 mg/ml; (2) dopamine, 10.3 mg/ml; (3)
(Acelll C18 column, 12.5 cm × 4.0 mm). Eluent A: 0.1% serotonin, 11.2 mg/ml; (4) apamin, 1 µg/µl; (5) phospho-
TFA in acetonitrile: water (80:20); eluent B: 0.1% TFA in lipase B, 0.5 µg/µl; (6) phospholipase A2, 1 µg/µl; (7) his-
water. Fractionation was performed using a linear gradient tamine, 22.8 mg/ml; and (8) melittin, 1 µg/µl.
of 5–80% A at 40 min with a flow rate of 1.0 ml/min and

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 33

Salt Precipitation

Ammonium sulfate precipitation was used as a

means to differentially isolate the venom proteins
responsible for developmental arrest and death.
Ammonium sulfate was added stepwise to crude
venom (in phosphate isolation buffer) until 100%
saturation was achieved, and then each fraction or
cut was assayed for the ability to induce develop-
mental arrest and death in young pharate adults
of S. bullata, and also trigger cell death in vitro.
Under these conditions, three salt fractions (5, 80,
and 100%) displayed biological activity toward S.
bullata in terms of halting fly development and
evoking death, with the 80% cut most closely re-
sembling crude venom (Table 4). Likewise, when
the fractions was assayed in vitro using T. ni cells,
Fig. 4. SDS-PAGE analysis of crude venom from N. all 3 fractions were cytotoxic, but the 80% cut was
vitripennis using 10% Tris-HCl polyacrylamide gels (Bio- most lethal toward the cultured cells (Table 4).
Rad). Lane 1: Molecular weight standards (10 µl), mysosin SDS-PAGE analysis of the salt fractions indicated
(199.6 kDa), β-galactosidase (129.8 kDa), bovine serum that the 5% cut had a nearly identical protein pro-
albumin (82.5 kDa), and carbonic anhydrase (40.8 kDa); file as crude venom (Fig. 5, Lanes 2–3). Protein
Lane 2: Crude venom (30 µg) in isolation buffer; Lane 3: bands with estimated molecular weights ranging
Crude venom (30 µg) in isolation buffer lacking sucrose. from 13 to 146 kilodaltons (kDa) were detected
Proteins were visualized using 0.1% (w/v) Coomassie bril- in both crude venom and the 5% fraction (Fig. 5).
liant blue G-250 staining (Garfin, 1990). However, two high molecular weight proteins
(188.1 and 200.5 kDa) observed in crude venom
even at antibody levels (1:10) that should have were absent from the 5% cut, and the 200.5-kDa
band was also not present when crude venom had
saturated binding sites on any venom proteins
been prepared in phosphate isolation buffer mi-
(Table 3).
nus sucrose (Fig. 4).
The protein profile of the 80% ammonium sul-
fate fraction revealed partial purification of 4 pro-
TABLE 3. In Vitro Evaluation of Apamin, Histamine, and Phenoloxidase teins with apparent molecular masses of 125,
Activity in Crude Venom From N. vitripennis* 116.9, 100.3, and 67.7 kDa (Fig. 5). Two of these
Treatment N Cell death (%) bands (67.7 and 100.3 kDa) were observed in the
Untreated 9,708 1.2 ± 0.8a
100% cut, although the biological activity of this
Saline (isolation buffer) 11,417 2.9 ± 1.3a fraction was significantly less than that of the 80%
Venom 10,046 100b cut in terms of inducing developmental arrest (F =
Apamin (10 µg) 8,643 1.7 ± 0.9a
Venom + apamin 12,479 100b
132.5, df = 8, 300, P < 0.001) and triggering death
Anti-histamine (1:10) 11,580 2.6 ± 0.5a in vivo (F = 99.2, df = 8, 300, P < 0.001) and in
Venom + anti-histamine 10,989 100b vitro (F = 92.4, df = 8, 300, P < 0.001) (Table 4).
Anti-phenoloxidase (1:10) 9,056 3.8 ± 2.1a
Venom + anti-phenoloxidase 10,047 100b
*The possible presence of apamin, histamine, and phenoloxidase in crude venom Membrane Separations
was assessed in vitro by evaluating the insecticidal activity in the presence of
apamin, anti-histamine, and anti-phenoloxidase. Assays used confluent monolayers
of BTI-TN-5B1-4 cells and mortality was assessed at 24 and 48 h post-incubation
Fractionation of crude venom proteins was also
using vital staining. attempted using centrifuge filters containing mem-

January 2006
34 Rivers et al.

TABLE 4. Insecticidal Activity in Venom Fractions Separated by Ammonium Sulfate

Precipitation*
% Response
In vivo In vitro
Fraction N Developmental arrest Death N Death
Saline 30 0a 0a 8,976 0a
Crude venom 30 95.1 ± 6.1b 100b 7,760 100b
Salt fraction
5% AS 30 54.2 ± 5.0c 80.4 ± 3.7c 10,056 91.4 ± 4.5b
15% AS 30 0a 0a 9,004 1.5 ± 0.8c
25% AS 30 0a 0a 11,307 0a
40% AS 30 0a 0a 8,925 1.1 ± 0.4c
65% AS 30 0a 0a 8,458 0.8 ± 0.6c
80% AS 30 93.2 ± 2.7b 100b 9,350 100b
90% AS 30 0a 0a 10,336 2.3 ± 1.0c
100% AS 30 63.4 ± 3.8c 56.7 ± 4.8d 11,974 64.8 ± 6.7d
*Salt was added stepwise to crude venom from N. vitripennis until 100% saturation, and then the precipi-
tate that formed with each fraction was assayed for biological activity. In vivo activity was assessed by
injecting 1 µl of test sample into the dorsal surface of adult S. bullata, whereas in vitro activity was mea-
sured by incubating BTI-TN-5B1-4 cells with 1 µl of each salt fraction. Mortality was observed 24 h later.

Fig. 5. SDS-PAGE analysis of ammonium sulfate fraction- venom (30 µg) in isolation buffer; Lane 3: 5% ammonium
ation of crude venom from N. vitripennis using 4–15% Tris- sulfate (AS) cut (precipitate); Lane 4: 15% AS cut; Lane 5:
HCl polyacrylamide gels (BioRad). Lane 1: Molecular 25% AS cut; Lane 6: 40% AS cut; Lane 7: 65% AS cut;
weight standards (10 µl), mysosin (199.6 kDa), β-galactosi- Lane 8: 80% AS cut; Lane 9: 90% AS cut; Lane 10: 100%
dase (129.8 kDa), bovine serum albumin (82.5 kDa), car- AS cut. Proteins were visualized using 0.1% (w/v) Coom-
bonic anhydrase (40.8 kDa), soybean trypsin inhibitor (31.5 massie brilliant blue G-250 staining (Garfin, 1990).
kDa), and lysozyme aprotinin (17.0 kDa); Lane 2: Crude

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 35

branes with specific size exclusion limits (10,000 tein profiles as crude venom (Fig. 6, only the 50-
to 100,000 MWCO). Retentates of all centrifuge and 75-K MWCO membranes are shown as repre-
membranes with MWCOs between 10,000–75,000 sentations). The protein profile of the 50- and 75-
were highly toxic when injected into adult flies or K MWCO filtrate indicated partial purification of
exposed to confluent monolayers of BTI-TN-5B1- at least 3 proteins with apparent molecular masses
4 cells (Table 5). Only the 100-K MWCO retentate of 125.7, 114.6, and 67.1 kDa (Fig. 6). This pro-
failed to display any lethal activity. By contrast, in file is nearly identical to that of the 80% ammo-
vitro and in vivo toxicity were observed with fil- nium sulfate fraction with the exception of a
trates from membrane filters with MWCOs rang- protein band of 100.3 kDa in the 80% cut. The
ing from 30,000–100,000 (Table 5). latter protein may be particularly important for
When samples were assayed for the ability to venom activity as filtrates from the 50- and 75-K
induce developmental arrest in young pharate MWCO membranes showed much lower biologi-
adults of S. bullata, retentates from all filters ex- cal activity than the 80% ammonium sulfate frac-
cept the 100K MWCO membrane elicited a halt in tion (Tables 4 and 5).
fly development comparable to crude venom
(Table 5). In contrast, only filtrates derived from DISCUSSION
75- and 100-K MWCO membranes were capable
of stimulating developmental arrest (Table 5). As During natural parasitism, N. vitripennis induces
a control, phosphate isolation buffer alone was a developmental arrest in fly hosts that lasts until
loaded onto each type of membrane filter, and the fly is consumed by parasitoid larvae (Rivers and
retentates and filtrates assayed. In all cases, no bio- Denlinger, 1994a). Developmental arrest as well
logical activity was observed. as death result from envenomation, and in the ab-
SDS-PAGE analyses of the membrane fractions sence of feeding wasp larvae, the halt in fly devel-
revealed that retentates from all membranes be- opment is sustained for an extended period of time
tween 10–75-K MWCO had nearly identical pro- (30–60 days at 25°C) (Rivers and Denlinger, 1995).

TABLE 5. Insecticidal Activity in Venom Fractions Separated by MWCO Membranes*

% Response
In vivo In vitro
Fraction N Developmental arrest Death N Death
a a
Saline (sugar-free 30 0 0 11,987 0a
isolation buffer)
Crude venom 30 97.1 ± 4.2b 100b 10,983 100b
MWCO
10 K retentate 30 94.3 ± 6.7b 100b 12,348 100b
Filtrate 30 0a 0a 11,149 0a
30 K retentate 30 95.9 ± 3.4b 100b 10,806 100a
Filtrate 30 0a 56.1 ± 3.6c 8,359 71.9 ± 4.8c
50 K retentate 30 95.4 ± 5.5b 100b 12,007 100b
Filtrate 30 0a 47.8 ± 7.16c 9,939 65.9 ± 5.0c
75 K retentate 30 97.0 ± 3.7b 100b 10,035 100b
Filtrate 30 94.3 ± 4.2b 100b 11,972 100b
100 K retentate 30 0a 0a 8,365 0a
Filtrate 30 98.3 ± 6.8b 100b 12,894 100b
*Retentates and filtrates were assayed for insecticidal activity in vivo by injecting 1 µl of test sample into
adult S. bullata, or examined in vitro by incubating BTI-TN-5B1-4 cells with 1 µl of each venom fraction.
Mortality was observed 24 h later.

January 2006
36 Rivers et al.

fluids from venom glands versus reservoirs, sug-

gesting that venom is modified in the venom res-
ervoir. Our study did not find any evidence to
support the idea of venom modification after pro-
duction in the acid gland. In fact, SDS-PAGE pro-
tein profiles were identical for crude venom
prepared from venom glands and venom reser-
voirs. When venom was extracted from wasp tis-
sues in buffer lacking a protein-stabilizing agent
such as sucrose (at high concentrations), or enzyme
inhibitors (Rivers, unpublished observations) of
buffer pH was less than optima, biological activ-
ity of venom decreased. Based on SDS-PAGE
Fig. 6. SDS-PAGE analysis of MWCO (molecular weight analyses of crude venom prepared in sucrose-free
cut-off) membrane fractionation of crude venom from N. buffer, the protein composition of venom was
vitripennis using 4–15% Tris-HCl polyacrylamide gels likely altered by these treatments. Venom analy-
(BioRad). Lane 1: Molecular weight standards (10 µl), ses performed by Ratcliffe and King (1967) did
mysosin (199.6 kDa), β-galactosidase (129.8 kDa), bovine not report controlling such factors when compar-
serum albumin (82.5 kDa), carbonic anhydrase (40.8 ing the contents of venom reservoirs and glands.
kDa), and soybean trypsin inhibitor (31.5 kDa); Lane 2:
It is also important to note that these early inves-
75 K (MWCO) retentate; Lane 3: 50 K retentate; Lane 4:
tigations (Ratcliffe and King, 1967, 1969) failed
75 K filtrate; Lane 5: 50 K filtrate. Proteins were visual-
to detect developmental arrest as a feature of en-
ized using 0.1% (w/v) Coommassie brilliant blue G-250
staining (Garfin, 1990). venomation, largely because non-natural host de-
velopmental stages (adult) were used in the
biological assays. Previous characterization of the
Here, we show that the developmental arrestant insecticidal properties of venom glands and res-
and the agent(s) that induce host death are present ervoirs (Beard, 1964; Ratcliffe and King, 1967;
in the venom gland and venom reservoir but not Tiegs, 1922) must also be interpreted cautiously
in other tissues associated with the female repro- as the method (i.e., dipping an insect pin in
ductive system. Psoralen treatment of crude venom venom prior to pricking the body) of venom in-
preparations indicated that biological activity was jection was not necessarily quantifiable or con-
associated with maternally derived secretions in the sistently reproducible.
gland and reservoir and not attributed to products Biochemical and electrophoretic analyses of
from symbiotic organisms (endosymbiotic viruses) crude venom indicate that venom is predominantly
as shown for some endoparasitic species (Quicke, composed of mid to high molecular weight pro-
1997; Stolz and Whitfield, 1992) or ectoparasitoids teins that are acidic in nature. This range of pro-
harboring microsporidia (Geden, 1996). These ob- tein sizes as well as complexity of the protein
servations are consistent with those of Ratcliffe and profile in terms of total number of proteins is con-
King (1967), who argued that contents of the sistent with most parasitic wasp venoms that have
venom gland (acid gland) alone are responsible been examined (Digilio et al., 2000; Leluk et al.,
for induction of death in adults of M. domestica, 1989; Nakamatsu and Tanaka, 2003; Parkinson et
and that the venom reservoir stores venom in ac- al., 2002a; Uçkan et al., 2004). By contrast, low
tive form (Ratcliffe and King, 1969). molecular weight proteins and peptides are typi-
Despite this agreement, previous electrophor- cal of venoms of the aculeate Hymenoptera (Leluk
etic analysis of the wasp venom (Ratcliffe and et al., 1989; Schmidt, 1982), as are proteins abun-
King, 1967) revealed unique protein profiles for dant in neutral and basic amino acids (Piek and

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 37

Spanjier, 1986). Infrared spectral analyses suggest venom yielding biological activity (induction of
that venom proteins do not possess a carbohydrate developmental arrest and death) all contained a
moiety, a feature shared with venom from P. protein with an estimated molecular weight of 67–
turionella (Uçkan et al., 2004) and many social hy- 70 kDa. It seems highly improbable, however, that
menopterans (Leluk et al., 1989) but unique from a single protein in venom from N. vitripennis is
other parasitic wasp venoms that appear to be rich responsible for inducing both host arrestment and
in glycosylated proteins (Leluk et al., 1989; Piek death. Indeed, though a protein band correspond-
and Spanjier, 1986). The infrared spectrum also ing to approximately 67 kDa was observed in the
revealed the presence of phosphorus-containing 100% ammonium sulfate fraction, this salt cut in-
structures, typical of enzymes, in the venom. duced developmental arrest in significantly fewer
Though the presence of enzymatic activity in N. pharate adults of S. bullata than the 80% fraction
vitripennis venom has yet to be confirmed, venoms that possessed two additional high molecular
from several endoparasitic species and social Hy- weight proteins (114.6 and 125.7 kDa). A similar
menoptera have been shown to contain multiple reduction in toxicity, as well as ability to arrest fly
types of enzymes (Moreau et al., 2004; Parkinson development, was observed with venom fractions
et al., 2001; 2002a,b; 2003; Piek and Spanjier, (filtrates from the 50- and 75-K MWCO) isolated
1986; Schmidt, 1982; Uçkan et al., 2004). Enzyme by centrifugal membranes with correspondingly
activity, however, does not likely account for the similar but not identical protein profiles to the
ability of venom from N. vitripennis to arrest host 80% salt cut: a 100.3-kDa protein was detected in
development or evoke death as crude venom prepa- the salt fraction in addition to 3 proteins present
rations contained the protease inhibitor PMSF as in the membrane fractions. Collectively, these ob-
well as the chelating agent EDTA, and high con- servations argue that multiple proteins are required
centrations of anti-phenoloxidase antibodies had for full venom activity. This certainly is consistent
no influence on venom activity. with the complex nature of parasitic wasp venoms
Venom stability at alkaline pH is in agreement that have been characterized and the milieu of host
with wasp toxins identified in venom from the effects that they evoke (Quistad et al., 1994;
ectoparasitoids H. hebetor (Quistad et al., 1994) Parkinson et al., 2002a,b, 2003).
and E. comstockii (Coudron and Brandt, 1996). H. Some lethal activity was also detected in filtrate
hebetor produces a potent venom with at least two collected from the 30-K MWCO membranes. How-
highly paralytic protein toxins (Brh-I and Brh-V, ever, this toxicity was significantly less by compari-
estimated molecular masses 71–73 kDa) (Quistad son to filtrates from membranes with molecular
et al., 1994) and one smaller, less insecticidal toxin cutoffs higher than 75 K and fly development was
(ca. 20–40 kDa). All three proteins appear to be not arrested by these samples. Surprisingly, no pro-
glycosylated. A 66-kDa venom protein has been tein bands were observed by SDS-PAGE that corre-
isolated from E. comstockii that does not induce sponded to the filter size exclusion limits. Failure
paralysis but does arrest host development by in- to detect proteins smaller than 30 kDa may reflect
hibiting larval–larval ecdysis (Coudron and Brandt, protein levels below the detection limits of the
1996). This arrestment protein is thought to be Coomassie staining (Garfin, 1990). It is also pos-
comprised of at least two subunits, suggesting that sible that the actual lethal protein is larger than
the native protein is possibly a dimer composed the size cutoff limit of the membranes used since,
of two 31–33-kDa proteins (Coudron and Brandt, like dialysis membranes, the listed size represents
1996). Size and activity estimates for venom pro- a median pore size, with both larger and smaller
teins from these two parasitoids are in agreement pores also present on the membranes (Pohl, 1990).
with SDS-PAGE profiles for partially purified Consequently, reduced venom activity may be at-
venom proteins from N. vitripennis: ammonium tributed to a small amount of the protein passing
sulfate and MWCO membrane fractions of crude through the membrane. SDS-PAGE analyses sup-

January 2006
38 Rivers et al.

port the latter scenario for both the 30- and 50-K wasp inhibits melanization of the host hemolymph. In-
MWCO membranes. sect Biochem Mol Biol 33:1017–1024.
HPLC separations of crude venom coupled with Asgari S, Zareie R, Zhang G, Schmidt O. 2003b. Isolation and
the use of internal standards suggested that low characterization of a novel venom protein from an
molecular peptides or proteins exist in venom in endoparasitoid, Cotesia rubecula (Hym.: Braconidae). Arch
the form of apamin and histamine. It is not sur- Insect Biochem Physiol 53:92–100.
prising, then, that neither were detected electro-
phoretically since their low molecular masses Beard RL. 1963. Insect toxins and venoms. Annu Rev Entomol
8:1–18.
would allow them to migrate much more rapidly
than other venom proteins in 10–12% acrylamide Beard RL. 1964. Pathogenic stinging of house-fly pupae by
gels (Garfin, 1990). Though these paralytic agents Nasonia vitripennis (Walker). J Invertebr Pathol 6:1–4.
are common in venoms from social Hymenoptera
Beckage NE. 1998.. Modulation of immune responses to para-
(Piek and Spanjier, 1986) and the endoparasitoid
sitoids by polydnaviruses. Parasitology 116:S57–S64.
P. turionella (Uçkan et al., 2004), biological assays
using cultured cells indicate that either apamin or Blum MS. 1981. Proteinaceous venoms. In: Blum M, editor.
histamine do not affect the cells tested or neither Chemical defenses of arthropods. New York: Academic
was present in active form within venom from N. Press. p 288–302.
vitripennis. This finding is not unexpected consid-
ering that apamin and histamine are typical of Coudron TA. 1991. Host-regulating factors associated with
parasitic Hymenoptera. In: Hedin PA, editor. Naturally oc-
paralytic venoms used primarily for defense (Blum,
curring pest bioregulators. ACS Symp Ser 449:41–65.
1981; Schmidt, 1982). In contrast, venom from N.
vitripennis is non-paralytic (Rivers and Denlinger, Coudron TA, Brandt SL. 1996. Characteristics of a develop-
1994a), and is used to manipulate a non-mobile mental arrestant in the venom of ectoparasitoid wasp
host (pupa) to maximize progeny production (Riv- Euplectrus comstockii. Toxicon 34:1431–1441.
ers et al., 1998). Clearly, additional studies are
needed to further characterize and isolate venom Dani MP, Richards EH, Isaac RE, Edwards JP. 2003. Antibac-
proteins. The stability of venom at alkaline pH sug- terial and proteolytic activity in venom from the endopara-
sitic wasp Pimpla hypochondriaca (Hymenoptera: Ichneu-
gests that anion-exchange chromatography is a vi-
monidae). J Insect Physiol 49:945–954.
able next step to purify individual venom proteins
(Quistand et al., 1994). Davies TR, Wickham TJ, McKenna KA. 1993. Comparative
recombinant protein production of eight insect cell lines.
ACKNOWLEDGMENTS In Vitro Cell Dev Biol Anim 29:388.

Denlinger DL. 1972.. Induction and termination of pupal dia-

The authors thank Kenneth Bujold and Timothy
pause in Sarcophaga (Diptera: Sarcophagidae). Biol Bull
Crawley for providing assistance in isolation of crude
142:11–24.
wasp venom. We also express thanks to Drs. Brain
Barr, Neena Din, and Terry Bird (Loyola College) Digilio MC, Isidoro N, Tremblay E, Pennacchio F. 2000. Host
for many conversations concerning protein purifi- castration by Aphidius ervi venom proteins. J Insect Physiol
cation techniques. This work was supported in part 46:1041–1050.
by USDA-NRICGP Seed grant 2001-1005 (D.B.R.)
and by a Loyola College Faculty Development Grant Doury G, Bigot Y, Perguet G. 1997. Physiological and bio-
chemical analysis of factors in the female venom gland
(D.B.R.)
and larval salivary secretions of the ectoparasitoid wasp
Eupelmus orientalis. J Insect Physiol 43:69–81.
LITERATURE CITED
Englard S, Seifter S. 1990. Precipitation techniques. In:
Asgari S, Zhang G, Zareie R, Schmidt O. 2003a. A serine pro- Deutscher MP, editor. Guide to protein Purification. Lon-
teinase homolog venom protein from an endoparasitoid don: Academic Press. p 285–300.

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 39

Fritz JS, Schenk GH. 1979. Electronic absorption spectra, fluo- Ohtaki T. 1966. On the delayed pupation of the flesh fly,
rescence and infrared spectroscopy. In: Fritz JS, Schnek GH, Sarcophaga peregrina Robineau-Desvoiday. Jpn J Med Soc
editors. Quantitative analytical chemistry. Boston: Allyn Biol 19:97–104.
and Bacon. p 430–460.
Parkinson NM, Smith I, Weaver R, Edwards JP. 2001. A new
Garfin DE. 1990. One-dimensional gel electrophoresis. In: form of arthropod phenoloxidase is abundant in venom
Deutscher MP, editor. Guide to protein Purification. Lon- of the parasitoid wasp Pimpla hypochondriaca. Insect
don: Academic Press. p 425–440. Biochem Mol Biol 31:57–63.

Geden CJ. 1996. Nosema disease of the house fly parasitoid Parkinson NM, Conyers C, Smith I. 2002a. A venom protein
Muscidifurax raptor and its impact on fly IPM programs. from the endoparasitoid wasp Pimpla hypochondriaca is
Proceedings of IOBC conference Technology Transfer in similar to snake venom reprolysin-type metalloproteases.
Biological Control: From Research to Practice, Montpellier, J Invertebr Pathol 79:129–131.
FR Bull IOBC 19:208.
Parkinson NM, Richards EH, Conyers C, Smith I, Edwards JP.
Gerling D, Legner EF. 1968. Developmental history and re- 2002b. Analysis of venom constituents from the parasi-
production of Spalangia cameroni, parasite of synathropic toid wasp Pimpla hypochondriaca and cloning of cDNA en-
flies. Ann Entomol Soc Am 61:1436–1443. coding a venom protein. Insect Biochem Mol Biol 32:
729–735.
Jones D, Sawicki G, Wozniak M. 1992. Sequence, structure,
and expression of a wasp venom protein with a negatively Parkinson NM, Conyers CM, Keen JN, MacNicoll AD, Smith
charged signal peptide and a novel repeating internal struc- I, Weaver RJ. 2003. cDNAs encoding large venom proteins
ture. J Biol Chem 267:14871–14878. from the parasitoid wasp Pimpla hypochondriaca identi-
fied by random sequence analysis. Comp Biochem Physiol
Kansu IA, U—ur A. 1984. Pimpla turionellae (L.) (Hym,-Ichneu-
134C:513–520.
monidae) ile konukçusu bazý lepidopter pupalarý arasýn-
daki biyolojik iliÕkiler üzerine araÕtýrmalar. (Investigations Piek T, Spanjier W. 1986.. Chemistry and pharmacology of
on the biological relationships between Pimpla turionella (L.) solitary wasp venoms. In: Piek T, editor. Venoms of the
(Hym.-Ichneumonidae) and some lepidopterous pupae Hymenoptera. London: Academic Press. p 161–307.
hosts J. Nat. Sci.) Doga Bilim Dergisi 8:160–173.
Piek T, Veenedaal RL, Mantel P. 1982. The pharmacology of
Krishnan A, Nair PN, Jones D. 1994. Isolation, cloning, and Microbracon venom. Comp Biochem Physiol 72C:303–309.
characterization of new chitinase stored in active form in
chitin-lined venom reservoir. J Biol Chem 269:20971–20976. Pohl T. 1990. Concentration of proteins and removal of sol-
utes. In: Deutscher MP, editor. Guide to protein purifica-
Leluk J, Schmidt J, Jones D. 1989. Comparative studies on tion. London: Academic Press. p 68–82.
the protein composition of hymnenopteran venom reser-
voirs. Toxicon 27:105–114. Quicke DLJ. 1997. Parasitic wasps. London: Chapman and
Hall.
Leonard GJ. 1972. The isolation of a toxic factor from sawfly
(Lophyrotoma interrupta klug) larvae. Toxicon 10:597–603. Quistad GB, Nguyen Q, Bernasconi P, Leisy DJ. 1994. Purifi-
cation and characterization of insecticidal toxins from
Moreau SJM, Cherqui A, Doury G, Dubois F, Fourdrain Y, Sabat- venom glands of the parasitic wasp, Bracon hebetor. Insect
ier L, Bulet P, Saarela J, Prevost G, Giordanengo P. 2004.
Biochem Mol Biol 24:955–961.
Identification of an apartylglucosaminidase-like protein in
the venom of the parasitic wasp Asobara tabida (Hymenop- Ratcliffe NA, King PE. 1967. The venom system of Nasonia
tera: Braconidae). Insect Biochem Mol Biol 34:485–492. vitripennis (Walker) (Hymenoptera: Pteromalidae). Proc R
Ent Soc Lond 42:49–61.
Nakamatsu Y, Tanaka T. 2003. Venom of ectoparasitoid,
Euplectrus sp. Near plathypenae (Hymenoptera: Eulo- Ratcliffe NA, King PE. 1969. Morphological, ultrastructural,
phidae) regulates the physiological state of Pseudaletia histochemical, and electrophoretic studies on the venom
separata (Lepidoptera: Nocuitdae) host as a food resource. system of Nasonia vitripennis Walker (Hymenoptera:
J Insect Physiol 49:149–159. Pteromalidae). J Morph 127:177–204.

January 2006
40 Rivers et al.

Rivers DB, Denlinger DL. 1994a. Developmental fate of the Rivers DB, Zdarek J, Denlinger DL. 2004. Disruption of
flesh fly, Sarcophaga bullata (Diptera: Sarcophagidae), pupariation and eclosion behavior in the flesh fly, Sarco-
envenomated by the pupal ectoparasitoid, Nasonia vitri- phaga bullata Parker (Diptera: Sarcophagidae), by venom
pennis (Hymenoptera: Pteromalidae). J Insect Physiol from the ectoparasitic wasp Nasonia vitripennis (Walker)
40:121–127. (Hymenoptera:Pteromalidae). Arch Insect Biochem Physiol
57:78–91.
Rivers DB, Denlinger DL. 1994b. Developmental fate of the
flesh fly, Sarcophaga bullata (Diptera: Sarcophagidae), Rivers DB, Crawley T, Bauser H. 2005. Localization of intrac-
envenomated by the pupal ectoparasitoid, Nasonia vitri- ellular calcium release in cells injured by venom from the
pennis (Hymenoptera: Pteromalidae). J Insect Physiol ectoparasitoid Nasonia vitripennis (Walker) (Hymenoptera:
40:121–127. Pteromalidae) and dependence of calcium mobilization
on G-protein activation. J Insect Physiol 51:
Rivers DB, Denlinger DL. 1995. Venom-induced alterations
149–160.
in fly lipid metabolism and its impact on larval develop-
ment of the ectoparasitoid Nasonia vitripennis (Walker) Schmidt JO. 1982. Biochemistry of insect venoms. Annu Rev
(Hymenoptera: Pteromalidae). J Invertebr Pathol 66: Entomol 27:339–368.
104–110.
Stoll VS, Blanchard JS. 1990. Buffers: Principles and practice.
Rivers DB, Hink WF, Denlinger DL. 1993. Toxicity of the In: Deutscher MP, editor. Guide to protein purification.
venom from Nasonia vitripennis (Hymenoptera: Ptero- London: Academic Press. p 24–37.
malidae) toward fly hosts, nontarget insects, different de-
velopmental stages, and cultured insect cells. Toxicon Stoltz DB, Whitfield J. 1992. Viruses and virus-like entities
31:755–765. in the parasitic Hymenoptera. J Hymenoptera Res 1:125–
139.
Rivers DB, Pagnotta MA, Huntington ER. 1998. Reproduc-
tive strategies of 3 species of ectoparasitic wasps are modu- Strand MR, Noda T. 1991. Alterations in the hemocytes of
lated by the response of the fly host, Sarcophaga bullata Pseudoplusia includens after parasitism by Microplitis demol-
(Diptera: Sarcophagidae) to parasitism. Ann Entomol. Soc itor. J Insect Physiol 37:839–850.
Am 91:458–465.
Strand MR, Pech LL. 1995. Immunological basis for compat-
Rivers DB, Ruggiero L, Yoder JA. 1999a. Venom from Nasonia ibility in parasitoid-host relationships. Annu Rev Entomol
vitripennis: a model for understanding the roles of venom 40:31–56.
during parasitism by ectoparasitoids. Trends Entomol 2:
1–17. Stuart B. 1997. Biological applications of infrared spectros-
copy. New York: John Wiley and Sons.
Rivers DB, Genco M, Sanchez RA. 1999b. In vitro analysis of
venom from the wasp Nasonia vitripennis: susceptibility of Thompson SN. 1986. Nutrition and in vitro culture of insect
different cell lines and venom-induced changes in plasma parasitoids. Annu Rev Entomol 31:197–219.
membrane permeability. In Vitro Cell Dev BiolAnim
35:102–110. Tiegs OW. 1922. Researches on insect metamorphosis, Parts
I and II. Trans R Soc S Aust 46:319–527.
Rivers DB, Rocco MM, Frayha AR. 2002a. Venom from the
ectoparasitic wasp Nasonia vitripennis increases Na+ influx Uçkan F, Sinan S, SavaÕçý S, Ergin E. 2004. Determination of
and activates phospholipase C and phospholipase A2 de- venom components from the endoparasitoid wasp Pimpla
pendent signal transduction pathways in cultured insect turionellae L. (Hymenoptera: Ichneumonidae). Ann Ento-
cells. Toxicon 40:9–21. mol Soc Am 97:775–780.

Rivers DB, Ruggiero L, Hayes M. 2002b. The ectoparasitic wasp Vinson SB. 1990. How parasitoids deal with the immune sys-
Nasonia vitripennis (Walker) (Hymenoptera: Pteromalidae) tem of their host: Overview. Arch Insect Biochem Physiol
differentially affects cells mediating the immune response 13:3–27.
of its flesh fly host, Sarcophaga bullata Parker (Diptera:
Sarcophagidae). J Insect Physiol 48:1053–1064. Webb BA, Luckhart S. 1994. Evidence for an early immuno-

Archives of Insect Biochemistry and Physiology

 Characterization of Venom From N. vitripennis 41

suppressive role for related Campoletis sonorensis venom and Whiting AR. 1967. The biology of the parasitic wasp Mor-
ovarian proteins in Heliothis virescens. Arch Insect Biochem moniella vitripennis (= Nasonia brevicornis) (Walker). Q Rev
Physiol 26:147–163. Biol 42:333–406.

Whitfield JB, Asgari S. 2003. Virus or not? Phylogenetics of Zdarek J, Slama K, Fraenkel G. 1979. Changes in internal
polydnaviruses and their wasp carriers. J Insect Physiol pressure during puparium formation in flies. J Exp Zool
49:397–405. 207:187–196.

January 2006