Two commercial preparations of the b exotoxin of Bacillus thuringiensis influence the mortality of caged adult honey bees, Apis mellifera (Hymenoptera: Apidae)


Vandenberg, J.D.; Shimanuki, H.

Environmental Entomology 15: 6-9

1986


Two commercial preparations of thuringiensin, the β exotoxin produced by some strains of Bacillus thuringiensis Berliner, were tested for their effect on honey bees (Apis mellifera L.). ABC-6162 (ABG) and SAN410SC72 (SAN) were diluted in 50% sucrose. When bees were fed the toxin solutions throughout their life, only the most dilute preparation of ABG had no significant effect on LT„ of replicate cages of bees when compared with untreated controls. All other dilutions of ABG and SAN significantly reduced LT30. A single feeding per bee of a 5 × 10-4 dilution resulted in no significant reduction in LT. for ABG. Higher doses and all doses of SAN caused a significant reduction in LTA. When dilutions in sucrose solutions were sprinkled on bees in cages, only the highest dose of ABG significantly reduced LTA. Lower doses of ABG and all doses of SAN had no significant impact on LTSO. Based on low spray volumes and consequent high thuringiensin concentrations, it is possible but unlikely that foraging honey bees could be exposed to doses sufficient to reduce longevity. However, since sprays are water-based, it is unlikely that honey bees would be sufficiently attracted to it to consume a lethal dose. Furthermore, some spray volumes are large enough, and concentrations low enough, that no toxic effect is expected.

Two
Commercial
Preparations
of
the
,9
Exotoxin
of
Bacillus
thuringiensis
Influence
the
Mortality
of
Caged
Adult
Honey
Bees,
Apis
mellifera
(Hymenoptera:
Apidae)
1
JOHN
D.
VANDENBERG
AND
H.
SHIMANUKI
Bioenvironmental
Bee
Laboratory,
United
States
Department
of
Agriculture,
Agricultural
Research
Service,
Beltsville
Agricultural
Research
Center,
Beltsville,
Maryland
20705
Environ.
Entomol.
15:
166-169
(1986)
ABSTRACT
Two
commercial
preparations
of
thuringiensin,
the
/3
exotoxin
produced
by
some
strains
of
Bacillus
thuringiensis
Berliner,
were
tested
for
their
effect
on
honey
bees
(Apis
mellifera
L.).
ABC
-6162
(ABG)
and
SAN410SC72
(SAN)
were
diluted
in
50%
sucrose.
When
bees
were
fed
the
toxin
solutions
throughout
their
life,
only
the
most
dilute
prepa-
ration
of
ABG
had
no
significant
effect
on
LT„
of
replicate
cages
of
bees
when
compared
with
untreated
controls.
All
other
dilutions
of
ABG
and
SAN
significantly
reduced
LT
30
.
A
single
feeding
per
bee
of
a
5
x
10
-4
dilution
resulted
in
no
significant
reduction
in
LT.
for
ABG.
Higher
doses
and
all
doses
of
SAN
caused
a
significant
reduction
in
LT
A
.
When
dilutions
in
sucrose
solutions
were
sprinkled
on
bees
in
cages,
only
the
highest
dose
of
ABG
significantly
reduced
LT
A
.
Lower
doses
of
ABG
and
all
doses
of
SAN
had
no
significant
impact
on
LT
SO
.
Based
on
low
spray
volumes
and
consequent
high
thuringiensin
concentra-
tions,
it
is
possible
but
unlikely
that
foraging
honey
bees
could
be
exposed
to
doses
sufficient
to
reduce
longevity.
However,
since
sprays
are
water
-based,
it
is
unlikely
that
honey
bees
would
be
sufficiently
attracted
to
it
to
consume
a
lethal
dose.
Furthermore,
some
spray
volumes
are
large
enough,
and
concentrations
low
enough,
that
no
toxic
effect
is
expected.
THURINGIENSIN
IS
the
heat
-stable
exotoxin
pro-
duced
by
some
strains
of
Bacillus
thuringiensis
Berliner
(for
reviews
see
Lecadet
and
DeBarjac
1981,
Sebesta
et
al.
1981).
It
is
toxic
to
many
species
of
insects
including
honey
bees,
Apis
mellifera
L.
(Krieg
and
Herfs
1963,
Cantwell
et
al.
1964,
1966,
Krieg
and
Langenbruch
1981).
Recently,
two
commercial
fi
rms
have
prepared
formulations
of
thuringiensin
for
possible
use
as
control
agents
against
coleopterous,
lepidopterous,
and
acarine
crop
pests.
We
obtained
samples
of
two
formulations
to
test
their
effect
on
adult
honey
bees.
We
prepared
dilutions
of
the
two
products
in
sucrose
solution
and
exposed
caged
adult
honey
bees
to
them
using
three
different
methods. This
paper
describes
the
results
of
those
experiments.
Materials
and
Methods
Newly
emerged
bees
were
obtained
by
incu-
bating
frames
of
sealed
brood
in
cages
at
34°C
and
by
collecting
the
emerging
adults
over
24
h.
All
bees
for
each
test
were
anesthetized
briefly
with
CO
2
in
an
ice-cream
carton
(0.5
liter).
Groups
of
ca.
50
bees
were
placed
into
the
cartons
and
cov-
ered
with
a
plastic
petri
dish
lid
(100
mm
diam).
Two
holes
in
the
lid
held
inverted
gauze
-covered
vials;
one
held
distilled
water,
the
other
held
50%
'
This
article
reports
the
results
of
research
only.
Mention
of
a
proprietary
product
does
not
constitute
an
endorsement
or
a
rec-
ommendation
for
its
use
by
USDA.
(wt/wt)
sucrose
(SS).
Both
groups
were
fed
ad
lib.
throughout
the
experiment
except
for
the
admin-
istration
of
the
products
as
stated
below.
Three
replicate
cages
were
set
up
for
each
product
and
dose.
Dead
bees
in
each
cage
were
counted
daily
and
living
bees
at
day
15
of
the
experiment
were
also
counted.
Two
commercial
preparations
were
donated
for
this
experiment:
ABG-6162
(ABG;
Abbott)
and
SAN410SC72
(SAN;
Zoecon).
Concentrations
of
thuringiensin,
the
active
ingredient,
are
1.8%
for
ABG
and
1.0%
for
SAN.
All
dilutions
of
thurin-
giensin
products
were
made
with
SS.
In
the
first
step,
equal
weights
of
SS
and
toxin
product
were
combined
to
give
the
5
x
10
-'
dilutions.
All
sub-
sequent
dilutions
were
prepared
vol/vol
with
SS.
Three
methods
were
used
to
expose
the
bees
to
thuringiensin.
In
method
1
(tests
started
on
30
May
and
14
June
1984),
the
bees
in
cages
were
given
water
and
toxin
diluted
in
SS
throughout
the
study
and
the
controls
received
SS
and
water.
In
method
2
(test
started
on
26
July
1984),
each
cage
of
bees
was
given
2.0
ml
of
treated
SS
and
water
until
all
the
toxin
had
been
consumed
(within
24
h).
Sub-
sequently,
all
bees
in
the
cages
were
given
SS
and
water.
In
method
3
(test
started
on
12
September
1984),
each
group
of
bees
was
sprinkled
with
2.0
ml
of
treated
SS
at
the
time
of
their
introduction
into
the
cage.
The
cages
were
shaken
gently
to
distribute
the
SS
over
the
bees.
The
bees
were
then
given
SS
and
water.
For
analysis,
cumulative
mortality
data
were
166
February
1986
VANDENBERG
AND
SHIMANUKI:
THURINCIENSIN
EFFECT
ON
HONEY
BEES
167
Table
1.
Effect
of
thuringiensin
products
in
sucrose
solutions
fed
to
caged
adult
honey
bees
Product
Dilution°
nb
LT
50
(days)
Differ
-
ence°
ABC
6
11.6
-
5
x
10
-6
3
6.6
NS
5
x
10
-4
6
4.6
s
5x
10
-5
3
4.0
S
5
x
10
-3
6
2.5
s
5
x
10
-2
3
2.1
s
5
x
10
-1
3
1.0
S
SAN
6
25.5
-
5
x
10
-4
6
4.6
S
5
x
10
-3
6
3.4
S
5
x
10
4
3
3.2
s
5
x
10
-5
3
2.7
S
5
x
10
-6
3
2.5
S
5x
10
-
i
3
1.2
S
°
Concentration
of
product
in
50%
(wt/wt)
sucrose.
h
Number
of
replicate
cages;
ca.
50
bees
each.
"
Results
of
Dunnet's
test
(P
<
0.05)
for
difference
from
control:
S,
significant;
NS,
not
significant.
subjected
to
angular
transformation
(Sokal
and
Rohlf
1969)
and
LT.'s
were
estimated
by
least
squares
regression
of
transformed
mortality
on
time
in
days
for
each
dose/method/product
combina-
tion
(SAS
1982).
The
LT
50
was
then
used
as
the
response
variables
in
analyses
of
variance
(ANOVA)
for
the
effects
of
dose
for
each
product
and
meth-
od
(SAS
1982).
Dunnet's
test
for
comparing
treat-
ment
means
to
a
control
(Crew
1977)
was
used
to
test
the
significance
of
the
LT
n
's.
Results
Regression
of
transformed
mortality
on
time
was
significant
for
all
(135)
dose/method/product
combinations
(P
<
0.05).
ANOVA
of
LT
50
for
each
method
and
product
are
displayed
in
Tables
1-3.
For
method
1
(Table
1),
constant
exposure
to
thu-
ringiensin
in
SS,
only
the
most
dilute
solution
(5
x
10
-6
)
of
ABG
had
no
significant
impact
on
LT50
when
compared
with
controls.
All
other
doses
of
ABG
and
all
doses
of
SAN
resulted
in
reduced
LT.'s
(For
ABG,
F6.23
=
3.45,
P
<
0.01;
for
SAN,
F6.,=
4.47,
P
<
0.004).
A
single
feeding
of
thuringiensin
product
dilu-
tions
in
SS,
method
2
(Table
2),
resulted
in
no
significant
reduction
in
LT
50
at
dilutions
of
5
x
10
-5
and
5
x
10
-4
for
ABG.
More
concentrated
solutions
significantly
reduced
LT.
=
23.15,
P
<
0.0001).
All
dilutions
of
SAN
(5
x
10
-2
to
5
x
10
-5
)
resulted
in
a
significant
reduction
in
LT,,,
=
5.39,
P
<
0.01).
Sprinkling
SS
containing
thuringiensin
on
caged
bees,
method
3
(Table
3),
reduced
LT
50
only
for
the
5
x
10
-3
dilution
of
ABG
(F
3
,
8
=
5.38,
P
<
0,03).
More
dilute
ABG
preparations
(5
x
10
-'
and
5
x
10
-5
)
and
all
three
dilutions
of
SAN
caused
no
significant
reduction
in
LT50
compared
with
controls
(F
3
,
8
=
3.66,
P
>
0.06).
For
ABG,
no
difference
between
methods
2
and
Table
2.
Effect
of
a
single
feeding
of
thuringiensin
in
sucrose
solutions
to
caged
adult
honey
bees
Product
Dilution"
LTso
(days)
Differ-
enceb
ABG
5
x
10
-
s
10.5
NS
0
7.9
5
x
10
-4
7.0
NS
5
x
10
-3
3.9
S
5x
10
-2
2.8
S
SAN
0
13.1
5
x
10
-4
6.4
S
5x
10
-5
4.3
S
5x
10
-3
4.1
S
5
x
10
-2
2.3
S
a
Concentration
of
product
in
50%
(wt/wt)
sucrose.
b
Results
of
Dunnet's
test
(P
<
0.05)
for
difference
from
control:
S,
significant;
NS,
not
significant.
3
was
detected
for
equal
dilutions.
For
SAN,
all
dilutions
tested
(5
x
10
-2
to
5
x
10
-5
)
by
method
2
reduced
the
LT50
but
those
tested
by
method
3
(5
x
10
-3
to
5
x
10
-5
did
not).
Discussion
Method
1
served
as
a
range
-finding
experiment
to
determine
an
approximate
lethal
-dose
range.
Methods
2
and
3
were
designed
to
simulate
fi
eld
conditions.
We
diluted
the
products
with
SS
to
increase
the
likelihood
that
bees
would
consume
the
dose.
Each
cage
of
ca.
50
bees
was
given
2.0
ml
of
SS
containing
diluted
thuringiensin
prod-
ucts.
Thus,
each
bee
could
ingest
ca.
0.04
ml
(0.05
g
at
30°C
for
50%
sucrose
[wt/wt]).
This
falls
with-
in
the
range
of
nectar
loads
observed
by
Park
(1922).
Method
2
resulted
in
the
bees
consuming
the
entire
dose,
but
in
method
3
the
bees
may
not
have
consumed
all
of
the
solutions.
There
is
a
possibility
that
foraging
bees
would
consume
these
products
in
the
field.
If
1,250
ml
of
the
product
were diluted
to
947
liters
in
water
for
application
over
1
ha
(a
dilution
of
1.25
x
10
-3
and
the
equivalent
of
1
pint
product
per
100
gal-
lons
spray
per
acre)
and
coverage
is
uniform,
then
one
foraging
bee
(ca.
1
cm
2
)
would
be
hit
by
9.5
x
Table
3.
Effect
of
sprinkling
sucrose
solutions
con-
taining
thuringiensin
on
caged
adult
honey
bees
Product
Dilution°
LT50
Differ-
enceb
ABG
5
x
10
-5
4.3
NS
0
4.1
5
x
10
-4
3.1
NS
5
x
10
-3
1.5
S
SAN
5
x
10
-5
6.0
NS
0
5.2
5
x
10
-4
3.9
NS
5
x
10
-3
3.8
NS
a
Concentration
of
product
in
50%
(wt/wt)
sucrose.
b
Results
of
Dunnet's
test
(P
<
0.05)
for
difference
from
control:
S,
significant;
NS,
not
significant.
168
ENVIRONMENTAL
ENTOMOLOGY
Vol.
15,
no.
1
10
-5
ml
spray.
If
the
bee
consumed
0.04
ml
(the
equivalent
of
one
nectar
load)
of
this
spray
(as
in
method
2,
Table
2),
it
might
be
sufficient
to
short-
en
its
life
span,
but
this
volume
would
be
spread
over
ca.
420
cm
2
.
When
dilutions
of
5
x
10
-3
and
5
x
10
-4
in
SS
were
sprinkled
on
bees
(ca.
0.04
ml
per
bee,
method
3,
Table
3),
only
the
highest
con-
centration
of
ABG
significantly
reduced
LT..
For
solanaceous
fi
eld
crops
the
concentration
range
for
SAN
is
4.2
x
10
-2
to
8.3
x
10
-2
.
Although
a
dose
of
0.04
ml
at
this
concentration
may
be
high
enough
to
cause
reduced
longevity,
the
volume
of
spray
by
which
bees
would
be
hit
is
only
ca.
3
x
10
-6
ml
and
a
dose
of
0.04
ml
would
be
spread
over
10
4
cm
2
.
Furthermore,
our
tests
used
SS
as
the
diluent;
field
spray
mixtures
would
use
water
and
would
not
be
as
attractive
to
bees.
Our
data
show
different
LT.'s
for
cages
of
con-
trol
bees
in
our
experiments.
Although
we
used
newly
emerged
bees
to
obtain
a
more
uniform
re-
sponse,
these
bees
were
collected
at
four
different
times
between
late
May
and
early
September.
Free
and
Spencer
-Booth
(1959)
demonstrated
that
bees
emerging
as
adults
early
in
the
season
have
a
greater
life
expectancy
than
bees
emerging
in
late
summer.
Nevertheless,
we
consider
it
likely
that
the
impact
of
methods
2
and
3
(tests
begun
26
July
and
12
September
1984,
respectively)
would
be
the
same
on
bees
emerging
in
spring.
Krieg
and
Herfs
(1963)
diluted
autoclaved
cul-
ture
supernatants
of
four
B.
thuringiensis
varieties
in
SS
and
fed
this
to
adult
honey
bees
in
cages.
Their
design
was
similar
to
our
method
1
in
that
the
bees
received
the
toxin
preparation
constantly
in
SS.
Each
bee
received
ca.
0.03
ml
of
the
culture
supernatant.
Mortality
after
9
days
from
B.
thu-
ringiensis
var.
sotto
and
var.
euxoae
was
not
sig-
nificantly
different
from
that
of
controls
(2-15%),
while
mortality
from
var.
dendrolimus
and
var.
thuringiensis
averaged
60
and
100%,
respectively.
Those
authors
concluded
that
toxic
effects
of
these
preparations
on
honey
bees
would
be
expected
only
after
a
long
period
of
application.
We
have
found,
however,
that
a
single
feeding
of
diluted
thurin-
giensin
preparations
can
reduce
honey
bee
longev-
ity
(Table
2).
Cantwell
et
al.
(1964)
fed
autoclaved,
freeze-
dried
aqueous
extracts
of
B.
t.
thuringiensis
spore/
crystal
preparations
to
honey
bee
larvae
and
adults.
This
exotoxin
preparation
caused
increased
mor-
tality
in
adults
and
larvae
when
fed
at
increasing
doses
between
0.31
and
10.00
mg
per
bee.
Those
workers
did
not
compare
mortality
of
treated
bees
with
that
of
controls.
For
our
studies,
using
an
estimated
range
of
1.0-1.8%
active
ingredient
in
the
products,
methods
2
and
3
resulted
in
a
dose
range
of
2.0-3.6
x
10
-2
to
2.0-3.6
x
10
-5
mg
per
bee.
Since
doses
as
low
as
2.0-3.6
x
10
-3
mg
per
bee
caused
a
significant
reduction
of
LT
50
for
both
products
(Table
2),
it
seems
likely
that
the
prep-
aration
used
by
Cantwell
et
al.
(1964)
had
a
low
concentration
of
active
exotoxin.
In
a
later
study,
Cantwell
et
al.
(1966)
observed
mortality
of
a
small
colony
of
ca.
10,000
bees
within
2
weeks
after
a
feeding
of
2
g
exotoxin
preparation
in
200
ml
of
sucrose
solution.
A
similar
concentration
of
either
product
tested
in
this
study
when
fed
directly
to
a
colony
would
probably
also
cause
its
death.
In
conclusion,
while
some
concentrations
of
the
products
tested
in
our
study
did
cause
a
significant
reduction
in
LT.'s,
the
likelihood
of
exposure
of
foraging
honey
bees
to
toxic
doses
is
low.
Upon
contact
with
the
bee,
the
water
-based
spray
would
evaporate
or
run
off.
What
remained
near
the
bee
would
probably
not
be
ingested
because
of
its
lack
of
attractiveness.
Contamination
of
nectar
and
pollen
by
the
spray
is
another
possible
route
of
ingestion
by
honey
bees.
The
range
of
possible
toxin
concentrations
for
ABG
for
orchard
use
is
1.25
x
10
-3
to
3.12
x
10
-
'.
This
overlaps
the
highest
dose
in
our
tests
(5
x
10
-
')
that
caused
no
significant
reduction
in
LT..
Di-
lution
of
the
spray
by
nectar
or
pollen
would
fur-
ther
reduce
the
effective
dose
for
foraging
bees.
Acknowledgment
We
thank
R.
J.
Cibulsky
and
T.
R.
Shieh
for
their
donation
of
the
products
and
some
information
used
in
this
study.
We
acknowledge
G.
E.
Cantwell
for
advice
on
design
and
L.
W.
Douglass
for
advice
on
statistical
analysis.
E. L.
Atkins,
R.
J.
Argauer,
G.
E.
Cantwell,
and
W.
T.
Wilson
made
helpful
suggestions
on
the
manu-
script
and
we
are
grateful
to
them.
Finally,
we
recog-
nize
the
valuable
technical
assistance
of
A.
M.
Lemanski.
References
Cited
Cantwell,
G.
E.,
D.
A.
Knox,
and
A.
S.
Michael.
1964.
Mortality
of
honey
bees,
Apis
mellifera
Linnaeus,
fed
exotoxin
of
Bacillus
thuringiensis
var.
thurin-
giensis
Berliner.
J.
Insect
Pathol.
6:
532-536.
Cantwell,
G.
E.,
D.
A.
Knox,
T.
Lehnert,
and
A.
S.
Michael.
1966.
Mortality
of
the
honey
bee,
Apis
mellifera,
in
colonies
treated
with
certain
biological
insecticides.
J.
Insect
Pathol.
8:
228-233.
Crew,
V.
1977.
Comparisons
among
treatment
means
in
an
analysis
of
variance.
U.S.
Dep.
Agric.
Tech.
Bull.
H-6.
Free,
J.
B.,
and
Y.
Spencer
-Booth.
1959.
The
lon-
gevity
of
worker
honey
bees.
Proc.
R.
Entomol.
Soc.
London
(A)
34:
141-150.
Krieg,
A.,
and
W.
Herfs.
1963.
Uber
die
Wirkung
von
Bacillus
thuringiensis
auf
Bienen.
Entomol.
Exp.
Appl.
6:
1-9.
Krieg,
A.,
and
G.
A.
Langenbruch.
1981.
Suscepti-
bility
of
arthropod
species
to
Bacillus
thuringiensis,
pp.
837-896.
In
H.
D.
Burges
[ed.],
Microbial
control
of
pests
and
plant
diseases,
1970-1980.
Academic,
London.
Lecadet,
M.
-M.,
and
H.
DeBarjac.
1981.
Bacillus
thuringiensis
beta-exotoxin,
pp.
293-321.
In
E.
W.
Davidson
[ed.],
Pathogenesis
of
invertebrate
micro-
bial
diseases.
Allanheld
Osmun,
Totowa,
N.J.
Park,
0.
W.
1922.
Time
and
labor
factors
in
honey
and
pollen
gathering.
Am.
Bee
J.
62:
254-255.
February
1986
VANDENBERG
AND
SHIMANUKI:
THURINGIENSIN
EFFECT
ON
HONEY
BEES
169
SAS.
1982.
Regression
procedure,
pp.
39-83, and
Microbial
control
of
pests
and
plant
diseases,
1970
general
linear
models
procedure,
pp.
139-199.
In
1980.
Academic,
London.
Statistical
analysis
system
user's
guide:
statistics.
SAS
Sokal,
R.
R.,
and
F.
J.
Rohlf.
1969.
Biometry.
Free
-
Institute,
Cary,
N.C.
man,
San
Francisco.
Sebesta,
K.,
J.
Farkas,
K.
Horska,
and
J.
Vankova.
1981.
Thuringiensin,
the
beta-exotoxin
of
Bacillus
Received
for
publication
8
April
1985;
accepted
25
thuringiensts,
pp.
249-281.
In
H.
D.
Burges
led.],
October
1985.