EPA/600/A-92/028
VERTEBRATE TOXICOLOGY OF THE SOLUBiLIZED PARASPORAL CRYSTALLINE
PROTEINS OF BACILLUS THURiNGIENSIS SUBSP. ISRAELENSIS
R. M. Roe, V. L. Kallapur,
W. C. Dauterrnan and F. W. Edens
Departments of Entomology,
Toxicology, and Poultry Science
North Carolina State University
Raleigh, NC 27695, U.S.A.
M. E. Mayes, G. A. Held and C. Y. Kawanishi
U.S. Environmental Protection Agency
HERL (MD-67)
Research Triangle Park, NC 27711, U.S.A.
A. R. Alford
Department of Entomology
University of Maine
Orono, ME 04469, U.S.A.
C. W. Clifford
Division of Natural Science and Mathematics
Northeastern State University
Tahlequah, OK 74464, U.S.A.

-------
INTRODUCTION
Within the sporangium of the bacterium. Bacillus thurinaiensis (Bt), is synthesized
a parasporal, proteinaceous crystalline inclusion (1,2) that has found widespread use
as a biological control agent (3). Bt is classified into varieties (also called subspecies
and serovars) based on the antigenic reactivity of the flagellae (H-antigen) found in
young, motile cells (4). As of 1982, cultures representing 28 serovars and 768
different isolates were collected and catalogued (5). The parasporal crystalline
proteins (PCP) from these serotypes when ingested have varying degrees of
selective toxicity against more than 182 species of insects (3). Research has
concentrated on Bt serovar 3a3b, kurstaki (Btk), because of its larvicidal activity
against major agricultural pests in the insect order, Lepidoptera (3). The serotype Bt
serovar 14, israelensis (El), differs from Bik by being toxic to members of the insect
order, Diptera (6,7), with little known toxicity to Lepidoptera. B|i is of special interest
because of its use in the control of mosquitoes and biting flies.
The anticipated expanded use of JB1 in insect control through applications in
biotechnology, has stimulated interest in the toxicology of Bi poisoning. Most of the
vertebrate toxicological research has focused on Bit The crystals of BJi consist of
several'protein components of differing molecular weights. The genes of five
polypeptides that compose the Bti parasporal crystal have been cloned. Cry IVA, cry
IVB, cry IVC, cry IVD and cytA genes encode polypeptides of 134,127.8, 77.8, 72.4
and 27.4 K in molecular weight, respectively (8). The cry IVA and B gene products
are proteolytically converted to toxic fragments that are in the 58 to 70 K molecular
range. Cry IVD protein is proteolytically converted to a fragment of about 30 K. Other
polypeptides are also observed as a component of purified crystal but are believed
to be degradation products of the major proteins or cellular contaminants (9). The
role of the major proteins that make up the parasporal crystal in vertebrate toxicity
has been the focus of our research.

-------
TOXICITY OF INJECTED SOLUBILIZED PARASPORAL CRYSTALLINE PROTEIN
FROM BT ISRAELENSIS
The alkaline solubllized parasporal crystalline protein (SPCP) of M was injected
into the hemocoel of insects from 6 orders and intraperitoneal^ into mice, rats and
quail (Table 1). In these studies,
Table 1. Injected toxicity of SPCP.	
Animal	24b LD50 (irg/Kg)
Insect* (intrahemocoel injections):
Adults
Aedes aeovpti (yellowfever mosquito)	11.6+2.2
Husca domesticus (housefly)	10.9+2.2
Qncooeltus fasciatus (milkweed bug)	27.7+7.0
Periolaneta americana (American cockroach)	4.42+0.36
Larvae
Trichoolusia ni. (cabbage looper)	3.71+0.32
Heliothis zea (bol Iworm)	73.6+3.0
Tenebrio molitor (yellow mealworm) >100
Vertebrates (intraperitoneal injections):
Swiss-Webster mice	1.31+0.23*
2.33B"
CD rats	K95 (1.78-2.12)®
Japanese quail	22.7 (21.7-24.1)"3
aFrom Roe et al. (10). Values are the mean + 1 standard deviation.
bFrom Hayes et al. (11).
cFrom Hayes, Kallapur, Held, Dauteraian, Roe and Kawanishi (unpublished).
Value reported is the mean ± the 95% confidence interval (a = 0.05).
From Kallapur, Hayes, Edens, Held, Dauterman, Kawanishi and Roe
(unpublished). Value reported is the mean ± the 95% confidence interval
(a = 0.05).
M was toxic to all of the animals tested except the yellow mealworm. Mice, rats,
cabbage loopers, and cockroaches were the most sensitive to SPCP injection with

-------
IDSO's of 1.31,1.95, 3.71 and 4.42 mg/kg, respectively. Among the insects tested,
the injected toxicity was not limited to the Diptera (mosquitoes and flies) but was also
observed in the Hemiptera, Orthoptera and Lepidoptera. Insect toxicity varied
significantly within a single insect family (the Noctuidae) with the LD50 ranging from
3.7 mg/Kg for the cabbage looper to 73.6 mg/Kg for the bollworm. In the vertebrates
tested, the lowest LD50 was 1.31 mg/Kg for mice compared to the highest of 22.7
mg/Kg for Japanese quail. In control experiments, B|| SPCP demonstrated typical,
oral mosquitocida! activity and was not toxic when fed to Lepidoptera; this was
consistent with previous descriptions of the toxic action of M (12). These results
demonstrate that the solubilized parasporaS crystalline protein of By, is highly toxic
and non-specific when introduced by injection into insects and vertebrates.
VERTEBRATE TOXICITY OF fill SPCP USING DIFFERENT ROUTES OF
INTRODUCTION
The toxicity of fit SPCP introduced by different routes was investigated in the rat,
mouse and Japanese quail (Table 2).
Table 2. Toxicity of SPCP bv different routes of introduction.	
Animal	Route of Introduction Dose (mg/Kg) 24h % Mortality (n)
CD rata
Intraperitoneal
9
100
(7)

Subcutaneous
9
0
(12)

Intravenous
21
0
(6)

Intratracheal
10
0
(6)

Gavage (oral)
9
0
(6)
Swiss-Webster




mouseb
Intraperitoneal
1.4
100
(5)


1.1
40
(5)

Gavage (oral)
>30
0
(10)
Japanese quail0
Intraperitoneal
30
70
(10)

Subcutaneous
100
0
(10)

Intravenous
100
0
(8)

Intranasal
40
0
(5)
aFrom Hayes, Kallapur, Held, Dauterman, Roe and Kawanishi (unpublished).
bFrom Roe et a].. (10).
cFrom Kallapur, Mayes, Edens, Held, Dauterman, Kawanishi and Roe
(unpublished).

-------
The only route resulting in death was intraperitoneal injection causing 100% mortality
at 9 mg/Kg in rats, 100% mortality at 1.4 mg/Kg and 40% mortality at 1.1 mg/Kg in
mice, and 70% mortality at 30 mg/Kg in quail. SPCP was not lethal by subcutaneous,
intravenous, intratracheal, and intranasal injection or by gavage. However,
subcutaneous injections in rats resulted in localized necrosis relative to dose.
In general fit has proven to be a very safe insecticide. The route of intoxication
responsible for its insecticida! activity in field applications is ingestion. The PCP of
Btk ,upon entering the gut of lepidopteran larvae, is activated by the alkaline
conditions and proteolytic activity of the digestive system (13-15). The gut epithelial
cells swell, vacuoles form, and the cells separate from the basement membrane and
each other ultimately disrupting the gut-hemocoel barrier (16-19). Incapacitation and
death occurs soon afterward. Similar observations have been made in mosquito
larvae fed Bt|.
In contrast to insects, no oral toxicity of SPCP in vertebrates was found (Table 2).
The fact that the only route of introduction of Bt| SPCP that was lethal to vertebrates
was intraperitoneal injection and no deaths were noted by other routes, suggests that
the target is the peritonea! cavity and its closely associated organs, that interactions
occur in this region necessary for toxic action elsewhere, and/or that sequestration or
metabolism at other sites of introduction prevent access to the site of action. These
possible explanations will be discussed in more detail later.
NEUROTOXIC, MYOTOXIC AND CYTOLYTIC ACTIVITY OF BE SPCP
The injection of SPCP into insects produced a number of obvious neuromuscular
effects almost immediately. These included cardiac arrest, paralysis of the body
region near the site of injection, abnormal crawling behavior, and eventual total
paralysis within 1 h. Pharmacological studies of ventral nerve cord function in the
cabbage looper (Fig. 1) showed that M SPCP acted as a nerve poison. Prior to

-------
treatment, ventral nerve cord electrical activity was minima! and a clear response as
PRE-TREATMENT t	POST-TREATMENT	
—1—1—1
0 rnin	j ' 7-60 rnirs j . 24 h
Figure 1. Ventral nerve cord response to injection of 13 mg/Kg of BJi SPCP into the
hemocoe! of the cabbage looper. In the neuro-physiofogical preparation,
the head, thorax and gut were removed from the larva. Tungsten
electrodes were placed into the hemocoe! along side abdominal ganglion
VIII, the ventral nerve cord and the abdominal wall. Injections of SPCP
were made into the second pair of abdominal prolegs. Mechanical
sensory stimulation with a glass probe was applied at the ana! proleg
(indicated as an "S" on the electrical trace of ventral nerve cord activity).
Data taken from Roe §t a|. (10).
indicated by increased electrical activity was noted after stimulation of the insect
abdominal prolegs with a glass rod. From 7 to"60 min after the injection of 12 mg/Kg
of SPCP, spontaneous-high frequency discharges were recorded in the ventral nerve
cord. The insect at this time was completely paralyzed. Hyperexcitability was
followed by reduced background activity and sensitivity to sensory stimulation.
Similar findings were made in isolated ventral nerve cord and peripheral nerve
preparations from the crayfish, Procambarus clarki. treated directly with SPCP (Roe
and Grossfeld, unpublished). Neurotoxic and myotoxic activity in insects was also
reported by other researchers (20,21).

-------
The behayiora! responses of mice to SPCP injection were not as dramatic as that of
insects. Mice at 0 to 1 h after injection appeared to be in a stupor and manifested
reduced alertness, exploratory behavior, and responsiveness to stimuli. They were
also slow in righting themselves due in part to a loss of hind leg control. Some of the
dead animals exhibited a constriction at their waist
The cytolytic activity of M SPCP has been well documented. Fig. 2 shows the
relationship between SPCP concentration and the percent hemolysis
ICO
60-
6tJ -
CO
55
>
o
£
in
z
SHEEPRSC
20 -
0
40
10
20
30
SPCP CONCENTRATION (jig/ml)
Figure 2. The in vitro hemolytic activity of M SPCP (from Kallapur, Mayes, Edens,
Held, Dauterman, Kawanishi and Roe, unpublished).
of quail and sheep red blood cells in vitro. It is interesting that this hemolytic activity

-------
is inhibited by preincubation of SPCP with serum. Hemolytic activity has been shown
for red blood cells from a number of animals (22) and cytolytic activity for both insect
and mammalian cells in culture (23-27). Using the appearance of cytosolic lactate
dehydrogenase (LDH) in insect hemolymph after injection of SPCP as a marker for
cytolytic activity, Roe et Si- (10) reported that SPCP was a potent in vivo cytotoxin in
the cabbage looper. However, in vivo cytotoxic activity was not affected by
temperature as was the toxicity and onset of hyperexcitability. These results
suggested a cause and effect relationship between hyperexcitability and toxicity but
not cytotoxicity as measured by LDH levels. Despite the obvious hemolytic activity of
Bti SPCP, intraperitoneal injections into mice which produced death in 35 min had no
effect on the red blood cell concentration which averaged 11 X 106 ce!ls/|il at 0, 20
and 35 min after injection (Roe and Clifford, unpublished). Similar findings were
reported by Mayes Ma!- (11 )•
TOXIC ACTIVITY OF THE COMPONENTS OF THE Eli PARASPORAL CRYSTAL
Davidson and Yamamoto (28) found that a 25K molecular weight (MW)
polypeptide, possibly a fragment of the 28K cytA protein was insecticidal, cytolytic,
and lethal to mice. This 25K MW fragment was purified by Sephacryl S-200 gel
permeation and DEAE-cellulose chromatography from alkaline dissolved crystal.
Similar cytolytic and/or larvicidal properties for 24 to 28K proteins were reported by
other researchers (29-32). Contrary to these findings, Hurley Mai- (33), Cheung and
Hammock (34), Held Mai- (35), and Visser it aj- (9) found that 25 to 28K proteins
were cytolytic but possessed either minimal or no mosquitocidal activity. The
purification of Hurley Mfil (33) was by gel permeation chromatography on a Bio-Gel
P-150 column, Cheung and Hammock (34) by DEAE-Sephacel chromatography,
Held Mai- (35) by immunoaffinity chromatography, and Visser Mai- (9) by sucrose
gradient ultracentrifugation. Delecluse §| si, (36) most recently found that disruption
of the cytA gene eliminates the hemolytic activity of the parasporal crystal.

-------
Ultrastructura! studies have revealed that the paraspora! body of Bti contains
three major inclusion types. Ibarra and Federici (32) isolated the Type 2 inclusion
using the pressure of centrifugation to dislodge the inclusion and a NaBr gradient for
purification. The Type 2 inclusion which consisted almost exclusively of a 65K
protein was only slightly toxic to mosquitoes. Hurley eial- (33, purification previously
described) and Kim ei ai. (37) found that 65K and 67K proteins, respectively, were
mosquitocidal. Kim gisi. (37) purified the 67K protein by Sspharose CL-4B gel
filtration and DEAE-cellulose chromatography from alkaline dissolved PCP. Visser et
ai- (9) reported mosquitocidal activity and no hemolytic activity for isolated 230 and
130k proteins from Bii. The toxicity of crystalline proteins of Bii derived from
recombinant E. coli and Bacillus was recently reviewed by Hofte and Whiteley (8)
and Federici §i si. (38). There is evidence that deletion of the cytA gene has no effect
on the mosquitocidal activity of the parasporal body (36) and others report that
crylVA, B, C and D mosquitocidal activity is enhanced by the presence of cytA
(39,40).
In our investigations to assign different toxic activities to crylV and cytA genes, a
28K protein from BJi (MW determined by SDS-PAGE) was purified and tested for its
mosquitocidal, hemolytic, neurotoxic and mouse toxic activities. For the neurotoxicity
studies, the 28K protein was purified by DEAE-Sephace! chromatography (41) while
for the other tests, the 28K protein was isolated by monoclonal antibody affinity
chromatography (11). Table 3 shows that the 28K protein had minimal mosquitocida!
Table 3. The mosquitocidal and hemolytic activity of the components
	of Bti SPCPa.			
Fraction
Aedes aeavoti larval
LC50 (/ig/ml)b
Protein conc. at 5055.
hemolysis )
Whole SPCP
0.72a
9.3
28K
23.1b
6.1
SPCP less the 28K


component
ai-._ 	c it
0.81a
no hemolysis at 19.1
aFrom ref. 11.-
bMeans followed by different letters are significantly different by the
Tukey's procedure (a = 0.05).

-------
activity as compared to both whole solubilized crystal and SPCP without the 28K
component. In contrast, the 28K protein had similar hemolytic activity to that of the
whole solubilized crystal while the remainder of the crystalline proteins had no effect
on hemolysis. The toxic activity in injected mice could also be attributed to the
cytolytic 28 K component of the solubilized crystal as illustrated in Fig. 3.
too
80 -
60 -
>-
<
J—
££
i
*
40 -
20 -
12
10
8
6
4
2
DOSE (mg/kg)
Figure 3, The mouse toxic activity of whole Bti SPCP and the 28K component.
Toxins were injected intraperitoneally. The SPCP less the 28K
component demonstrated no toxic activity in these studies. Taken from
Ref. 11.
The 28K component had comparable activity to that of the whole crystal while SPCP
less the 28K protein demonstrated no toxicity in the mouse. In separate studies
(10,41), it was discovered that a 25K fragment of the 28K component was also
neurotoxic to ventral nerve cord preparations of the cabbage looper causing nerve
death in a similar fashion to that of the cytolytic protein, phospholipase A2. Therefore,

-------
it appears from these studies, that the 28K component has multiple activities that
include cytolytic, neurotoxic, and injected mouse toxic activity. However, the 28K
protein has low mosquitocida! activity.
MODE OF ACTION OF INJECTED fill SPCP IN VERTEBRATES
Although the mode of action of M SPCP when introduced into vertebrates by
intraperitoneal injection is unknown, a number of common physiological responses
have been measured both in mice and quail. These include a reduction in heart rate
and a decrease in body temperature (Fig. 4). In addition to these effects, peripheral
vasodilation is apparent in mice indicated by extreme reddening of the ears, feet and
tail. From histopathological studies, the only organs affected by BJi SPCP
intraperitoneal injection in mice and rats were the liver and jejunum (11). The
jejunum was characterized by hemorrhaging in the lamina propria, especially at the
tips of villi and this hemorrhaging was accompanied by epithelial necrosis and cell
sloughing. The liver exhibited centrilobular congestion. One explanation for these
histopathological responses to M SPCP is reduced vascular perfusion possibly
resulting from hypotension. The resulting hypoxia would cause cell death.
Peripheral circulatory system dilation, decreasing heart rate and temperature, and
hypoxia are symptoms consistent with septic shock. The apparent lack of red blood
cell hemolysis in vivo in mice dosed with SPCP (discussed earlier) suggests that
systemic, general cytolytic activity is not a major contributing toxic factor and that the
septic shock-like response is initiated by interactions in the region of the peritonea!
cavity. Recall that no other routes of introduction into rats and quai! other than by
intraperitoneal injection were lethal. The role of cytotoxicity on tissue necrosis in the
liver and jejunum and its contribution to the septic shock-like symptoms associated
with death is not clear. The localized necrosis due to subcutaneous injections
appears to be due to cytolytic activity. Additional references on the vertebrate
toxicology of B| are cited (42-47).

-------
700
Figure 4,
-£>—-c—2
CONTROL (MICE)
SFCP (MICE)
CONTROL (QUASI)
SFCP (QUAIL)
50
—j—
10©
i
150
200
250
30 -
O
25 -
20 -
3
t-
<
CC
Hi.
a.
£
H
15 -
CONTROL (MICE)
SFCP (MICE)
CONTROL (QUAIL)
SFCP (QUAIL)
10 -
50
100
150
25G
0
200
TIMS (M'N) AFTER IP INJECTION
The effect of intraperitoneal injection of Eli SPCP on heart rate and body
temperature in mice (11) and Japanese quail (Kaliapur, Mayes, Edens,
Held, Dauterman, Kawanishi and Roe, unpublished). The SPCP dose
was 10 mg/Kg for mice and 30 mg/Kg for quail.

-------
CONCLUSIONS
The solubilized parasporal crystalline protein of Bacillus thurinaiensis israelensis
has broad-spectrum toxic activity when injected into the hemocoel of insects and into
the peritoneal cavity of vertebrates. The mouse LD50 was as low as 1.3 mg of whole
SPCP/Kg body weight. SPCP was not lethal when introduced into mice, rats, and
quail by other routes including intratracheal, intranasal, intravenous, subcutaneous,
and oral administration, demonstrating that Bti is a relatively safe insecticide.
Subcutaneous injection does cause localized necrosis. The 28K cytA component of
the parasporal crystal is responsible for neurotoxic, hemolytic and mouse toxic
activity but has low oral mosquito toxicity. Intraperitoneal injections of SPCP in mice
and quail reduced the heart rate and body temperature, produced peripheral
vasodilation in mice, and caused jejunal hemorrhaging and liver centrilobular
congestion in both mice and rats. These responses are consistent with symptoms
associated with endotoxin-induced septic shock.
ACKNOWLEDGEMENTS
This research was supported by the NC State Agricultural Research Service, a
Biomedical Research Support Grant (RR7071), the US Public Health Service (ES-
07046 and ES-GQ044), and the US Environmental Protection Agency. This
document has been reviewed in accordance with US Environmental Protection
Agency policy and approved for publication. The mention of trade names or
commercial products does not constitute endorsement or recommendations for use.
We would like to especially acknowledge Dr. J. P. Woodring in the Department of
Physiology at Louisiana State University for his advice on experimental approaches.

-------
REFERENCES
1.	H.J. Somerville, Insect toxin in spores and protein crystals of Bacillus
thurinoiensis. Trends Biochem. Sci. 3:108-110 (1978).
2.	L. A. Bulla, Jr., D. B. Bechtel, K. J. Kramer, Y. I. Shethna, A. I. Aronson and P. C.
Fitz-James, Ultrastructure, physiology, and biochemistry of Bacillus
thurinaisnsis. Q. R. C. Crit. Rev. Microbiol. 8:147-204 (1980).
3.	P. Luthy, J.-L. Cordier and H.-M. Fischer. Bacillus thurinaiensis as a
bacterial insecticide: basic considerations and application, in:
"Microbial and Viral Pesticides", E. Kurstak, ed., Marcel Dekker, New
York (1982).
4.	H. de Barjac and A. Bonnefoi, Essai de classification biochemique et
serologique de 24 souches de Bacillus du type fi. thurinoiensis.
Entomophaaa 7:5-31 (1962).
5.	H. T. Dulmage, C. C. Beegle, H. de Barjac, D. Reich, G. Donaldson, J.
Krywienczyk and G. Donaldson, Bacillus thurinoiensis cultures available
from the U.S. Department of Agriculture, USDA. ARS. ARM-S-30/Qct
(1982).
6.	L. J. Goldberg and J. Margalit, A bacterial spore demonstrating rapid larvicida!
activity against Anooheles seraentii. Uranotaenia unouiculata. Culex
univitattus. Aedes aeovoti and Culex oioiens. Moso. News. 37:355-358
(1977).
7.	H. de Barjac, Une nouvelle variete de Bacillus thurinaiensis tres toxique pour
les moustiques: Bacillus thurinoiensis var. israelensis serotype 14, Q. JR.
Acad. Ssi- (Paris) 2880:797-800 (1978).
8.	H. Hofte and H. R. Whiteley, Insecticidal crystal proteins of Bacillus
thurinaiensis. Microbiol. Rev. 53:242-255 (1989).
9.	B. Visser, M. van Workum, A. Dullemans and C. Waaiwijk, The mosquitocidal
activity of Bacillus thurinaiensis var. israelensis is associated with Mr 230
000 and 130 000 crystal proteins. FEMS Microbiol. Lett. 30:211-214 (1988).

-------
10.	R. M. Roe, P. Y. K. Cheung, B. D. Hammock, D. Buster and A. R. Alford,
Endotoxin of Bacillus thurinaiensis var. israelensis. broad- spectrum toxicity
and neural response elicited in mice and insects, in: "Bioregulators for Pest
Control, Symposium Series 276", P. Hedin, ed., American Chemical
Society, Washington, D.C. (1985).
11.	M. E. Mayes, G. A. Held, C. Lau, J. C. Seely, R. M. Roe, W. C. Dauterman and C.
Y. Kawanishi, Characterization of the mammalian toxicity of the crystal
polypeptides of Bacillus thurinaiensis subsp. israelensis. Fund. Add!. Tox.
13:310-322 (1989).
12.	M. J. Klowden, G. A. Held and L. A. Bulla, Jr., Toxicity of Bacillus thurinaiensis
subsp. israelensis to adult Aedes aeavoti mosquitoes, Aool. Environ.
Microbiol. 46:312-315(1983).
13.	L. A. Bulla, Jr., K. J. Kramer, D. J. Cox, B. L. Jones, L. I. Davidson and G. L.
Lookhart, Purification and characterization of the entomocidal protoxin of
Bacillus thurinaiensis. J. Biol. Chem. 256:3000-3004 (1981).
14.	K. W. Nickerson, Structure and function of the Bacillus thurinaiensis protein
crystal, Biotechnol. Bioena. 22:1305-1333 (1980).
15.	D. J. Tyrell, L. A. Bulla, Jr., R. E. Andrews, Jr., K. J. Kramer, L. I. Davidson and P.
Nordin, Comparative biochemistry of entomocidal parasporal crystals of
selected Bacillus thurinaiensis strains, Bacterid. 145:1052-1062 (1981).
16.	Y. Endo and J. Nishiitsutsuji-Uwo, Mode of action of Bacillus thurinaiensis
endotoxin: histopathological changes in the silkworm midgut, J. Invertebr.
Pathol. 36:90-103 (1980).
17.	J. Percy and P. G. Fast. Bacillus thurinaiensis crystal toxin: ultrastructurai
studies of its effect on silkworm midgut cells, J. Invertebr. Pathol. 41:8S-98
(1983).
18.	G. R. Sutter and E. S. Raun, Histopathology of European-coin-borer larvae
treated with Bacillus thurinaiensis. J. Invertebr. Pathol. 9:90-103 (1967).
19.	H. R. Ebersold, P. Luethy and M. Mueller, Changes in the fine structure of the
gut epithelium of Pieris brassicae induced by the endotoxin of Bacillus
thuringiensis, Bull. Soc. Ent. Suisse 50:269-276 (1977).

-------
20.	G. J. P. Singh and S. S. Gill, Myotoxic and neurotoxic activity of Bacillus
thurinaiensis var. israelensis crystal toxin, Pest. Biochem. Physiol. 24:408-
414 (1985).
21.	G. J. P. Singh, L. P. Schouest, Jr. and S. S. Gill, Action of Bacillus thurinaiensis
subsp. israelensis endotoxin on the ultrastructure of the house fly larva
neuromuscular system in vitro, J. Invertebr. Pathol. 47:155-168 (1988).
22.	S. A. Weinstein, A. W. Bernheimer and J. D. Oppenheim, Isolation of a
hemolysin from a spore-crystal mixture of Bacillus thurinaiensis israelensis
(serotype H-14), Toxicon 28:733-746 (1988).
23.	N. Sandler, R. Zomper, A. Keynan and J. Margalit, Bacillus thurinaiensis var.
israelensis crystal hemolysis as a possible basis for an assay of larva!
toxicity, Aool. Microbiol. BiotechnoL 23:47-53 (1985).
24.	W. E. Thomas and D. J, Eiiar, Bacillus thurinaiensis var israelensis crystal
endotoxin: effects on insect and mammalian cells in vitro and in vivo. J. Cell.
M- 60:181-197 (1983).
25.	D. W, Murphy, S. S. Sohi and P. G. Fast, Bacillus thurinaiensis enzyme-
digested endotoxin; effect on cultured insect cells, Science 194:954-958
(1976).
26.	D. E. Johnson, Toxicity of Bacillus thurinaiensis entomocidal protein toward
cultured insect tissue, Invertebr. Pathol. 38:94-101 (1981).
27.	J. Nishiitsutsuji and Y. Endo, Mode of action of Bacillus thurinaiensis
endotoxin: effect on TN-368 ceils, J. Invertebr. Pathol. 34:267-275 (1979).
28.	E. W. Davidson and T. Yamamoto, Isolation and assay of the toxic component
from the crystals of Bacillus thurinaiensis var israelensis. Curr. Microbiol.
11:171-174 (1984).
29.	J. L. Armstrong, G. F, Rohrmann and G. S. Bsaudreau, Delta endotoxin of
Bacillus thurinaiensis subso israelensis. J. Bacterid. 161:39-46 (1985).
30.	E. S. Ward, D. J. Ellar and J. A. Todd, Cloning and expression in Escherichia
coli of the insecticidal endotoxin gene of Bacillus thurinaiensis var.
israelensis. FEBS Lett. 175:377-382 (1984).

-------
31.	T. Yamamoto, T. lizuka and J. N. Aronson, Mosquitocida! protein of Bacillus
thurinaiensis subsp. israelensis: identification and partial isolation of the
protein. Curr. Microbiol. 9:279-284(1983).
32.	J. E. Ibarra and B. A. Federici, Isolation of a relatively nontoxic 65-kiiodalton
protein inclusion from the parasporal body of Bacillus thurinaiensis subso.
israelensis. J. Bacteriol. 165:527-533 (1988).
33.	J. M. Hurley, S. G. Lee, R. E. Andrews, Jr., M. J. Klowden and L. A. Bulla, Jr.,
Separation of the cytolytic and mosquitocida! proteins of Bacillus
thurinaiensis subsp. israelensis. Biochem. Bioohv. Res. Comm. 128:981 -
965 (1985).
34.	P. Y, K. Cheung and B. D. Hammock, Separation of three biologically distinct
activities from the parasporal crystal of Bacillus thurinaiensis var.
israelensis. Curr. Microbiol. 12:121-126 (1985).
35.	G. A. Held, Y.-S. Huang and C. W. Kawanishi, Effect of removal of the cytolytic
factor of Bacillus thurinaiensis subsp. israelensis on mosquito toxicity,
Biochem. Bioohvs. Res. Commun. 141:937-941 (1988).
36.	A. Delecluse, J.-F. Charles, A. Klier and G. Rapoport, Deletion by in vivo
recombination shows that the 28-kiIodalton cytolytic polypeptide from
Bacillus thurinaiensis subso. israelensis is not essential for mosquitocida!
activity. J. Bacteriol. 173:3374-3381 (1991).
37.	K.-H. Kim, M. Ohba and K. Aizawa, Purification of the toxic protein from Bacillus
thurinaiensis serotype 10 isolate demonstrating a preferential larvicidal
activity to the mosquito, J. Invertebr. Pathol. 44:214-219 (1984).
38.	B. A. Federici, P. Luthy and J. E. Ibarra, Parasporal body of Bacillus
thurinaiensis israelensis. Structure, protein composition and toxicity, in:
"Bacterial Control of Mosquitoes and Blackflies: Biochemistry, Genetics and
Applications of Bacillus thurinaiensis and Bacillus sohaericus". H. de Barjac
and S. Sutherland, eds, Rutgers University Press, New Brunswick, N.J.
(1990).

-------
39.	D. Wu and F. N. Chang, Synergism in mosquitocidal activity of 26 and 65 kDa
proteins from Bacillus thurinaiensis subsp. israelensis crystal, FEBS Lett.
190:232-236 (1985).
40.	A. Delecluse, C. Bourgouin, A. Kiier and G. Rapoport, Specificity of action on
mosquito larvae of Bacillus thurinaiensis israelensis toxins encoded by two
different genes, Mol. Gen. Genet. 214:42-47 (1988).
41.	P. Y. K. Cheung, D. Buster, B. D. Hammock, R. M. Roe and A. R. Alford, Bacillus
thurinaiensis var. israelensis endotoxin: evidence of neurotoxic action,
Pestic. Biochem. Phvsiol. 27:42-49(1987).
42.	R. E. Warren, D. Rubenstein, D. J. Eilar, J. M. Kramer and R. J. Gilbert, Bacillus
thurinaiensis var israelensis: protoxin activation and safety, Lancet
(8378):678-679 (1984).
43.	P. Y. K. Cheung, R. M. Roe, B. D. Hammock, C. L. Judson and M. A. Montague,
The apparent in vivo neuromuscular effects of the endotoxin of Bacillus
thurinaiensis var. israelensis in mice and insects of four orders, Pestic.
Biochem. Phvsiol. 23:85-94(1985).
44.	J. P. Siegel, J. A, Shadduck and J. Szabo, Safety of the entomopathogen
Bacillus thurinoiensis var, israelensis for mammals, J.. Econ. Entomol.
80:717-723 (1987).
45.	W. M. Hadley, S. W. Burchiel, T, D. McDowell, J. P. Thilsted, C. M. Hibbs, J. A.
Wharton, P. W. Day, M. B. Friedman and R. E. Stoll, Five-month oral (diet)
toxicity/infectivity study of Bacillus thurinaiensis insecticides in sheep, Fund.
Aool. Tax. 8:236-242(1987).
46.	D. G. L. Innss and J. F. Bendell, The effects on small-mammal populations of
aerial applications of Bacillus thurinaiensis. fenitrothion, and Matacil® used
against jack pine budworm in Ontario, Can. j|. Zool. 67:1318-1323 (1989).
47.	J. P. Siege! and J. A. Shadduck, Mammalian safety of Bacillus thurinaiensis
•israelensis. in: "Bacteria! Control of Mosquitoes & Black Flies. Biochemistry,
Genetics & Applications of Bacillus thurinoiensis israelensis and Bacillus
sohaericus". H. de Barjac and D. J. Sutherland, eds., Rutgers University
Press, New Brunswick, NJ (1990).

-------
EPA/600/A-92/028
STmis«9^rfTns
Vertebrate Toxicology of the Solubilized Proteins of
Bacillus thurinqiensis subso. israelensis
i» wSa
RoSj yfL< Kallapur, W.G. Dauterman,
F.W. Edens, H.E. Mayes, G.A. Held, C.Y. Kawanishi,
A. P.	anrf r u rui^d,
i, PzmQmMm® ©aajisuaaTiea nasas assaaa
US EPA (HERL/ERC/DTO/PTB) RTP, NC 27711
NCSU - Raleigh, NC 27695
University of Mains - Orono, Maine 04469
NSU Oklahoma - Tahlequah, Oklahoma ' 74464
il'Ss'a^^^yg'esi^aaaat jsi»	m®'.
ia.	Asaasv «a»g ass amxus
US m (HERL/ESC/DTD/PTB) MB-67
Research Triangle Park, NC 27711
! etmAma&figm ce#§
'id
T«W,
19. T??3 e# *1*9*7 &&Z MSJ53 CCV39SD
i&i^essas^lWS i&abicv east	
^PR/^oc//c-
ia $y*m,g mmr&m mm?3s
n.
This review summarizes the studies done with the mammalian toxic
Bacillus thuringiensls subsp. isrmlsnsis {Bti) 28 kDa cytA protein. The
data is relevant to hazard identification studies with bacteria! pesticides.
The data shews that cytA produces lethal physiological changes in
diverse mammalian species when administered intraperitoneal^ and a
dose-dependent localized necrosis by the subcutaneous route. Challenge
by other routes have no effects. The cytA protein is a minor component of
the insscticldal activity of the Bti paraspora! crystal. Insertion of the
cytA gene by genetic engineering methods into microbial species that
have the potential to invade traumatized tissues or organs could result in
detrimental human health effects. -
*a? cane# ms eseyati»7 asm**
ttMoaswoiass/gn'sa gaaaa nsana
§. essaysw&s&?

RELEASE TO PUBLIC
TTTRu»RW3D2Srn
UNCLASSIFIED
'a47isiu»5i?* et*ai7i
UNCLASSIFIED

fi?£ pM USSm 1 tftafe 4mm	mmfWSs IS «33Sw**S

-------