26
DATA
EVALUATION
REPORT
6/
17/
2001
Reviewed
by:
Chris
A.
Wozniak,
Ph.
D.,
Biologist,
OPP
/
BPPD
/
s/
Secondary
Reviewer:
John
L.
Kough,
Ph.
D.,
Biologist,
OPP
/
BPPD
/
s/
MRID
No.:
452422­
05
Active
Ingredients:
Bacillus
thuringiensis
strain
PS149B1
(
NRRL
B­
21553)
binary
insect
control
proteins
as
expressed
in
maize
Product
Name:
PS149B1
insect
control
proteins
as
expressed
in
maize
ID
No.:
068467­
EUP­
G,
068467­
EUP­
L,
029964­
EUP­
R,
022964­
EUP­
G
Submission
No.:
S590323,
S590337
Chemical
No.:
006430
Bacillus
thuringiensis
PS149B1
proteins
DP
Barcode:
D271578
Sponsor:
Dow
Agrosciences
LLC
Authors:
S.
J.
Stelman
Testing
Facility:
Dow
Agrosciences
LLC,
5501
Oberlin
Drive,
San
Diego,
CA
92025.
Study
Titles:
Comparison
of
the
amino
acid
sequence
of
the
Bacillus
thuringiensis
strain
PS149B1
13.6
kDa
and
43.8
kDa
insecticidal
crystal
proteins
to
known
protein
allergens.
Study
Date:
October
13,
2000
Study
No.:
GH­
C
5140
Conclusion:
Based
upon
the
search
requirements
that
at
least
8
contiguous
amino
acids
are
identical
to
a
known
allergenic
sequence,
PS149B1
13.6
and
43.8
kDa
proteins
are
unrelated
to
any
known
allergens
in
that
they
do
not
share
any
linear
epitopes
with
known
protein
allergens.

Classification:
Acceptable.

Good
Laboratory
Practices:
This
study
was
not
conducted
in
compliance
with
the
Certification
of
Good
Laboratory
Practices
as
outlined
in
40
CFR
160.

Purpose:
To
assess
the
potential
homology
between
the
PS149B1
proteins
and
known
allergens
as
present
in
available
databases.

METHODS
General
­
All
sequence
analysis
programs
used
int
his
study
were
part
of
version
10.1
of
the
GCG
(
Genetics
Computer
Group),
Inc.,
Madison,
WI)
package
run
on
a
Silicon
Graphics
Origin
200
computer
under
the
Irix
(
Silicon
Graphics,
Inc.)
6.5
operating
system.

Database
construction
­
The
GCG
DATASET
was
used
to
generate
four
allergen
databases
as
follows:
a
dataset
generated
by
searching
for
`
allergen'
against
the
PIR
(
release
65,
Swiss­
Prot
(
release
39),
TrEMBL
(
release
14)
and
GenPept
(
release
118)
databases
using
GCG
STRINGSEARCH
and
LOOKUP
programs;
three
other
databases
based
upon
published,
curated
protein
databases
of
known
or
putative
allergens
­
Gendel,
S.,
1998.
Sequence
databases
for
assessing
the
potential
allergenicity
of
proteins
used
in
transgenic
foods.
Advances
in
Food
and
27
Nutrition
Research
42:
63­
92
(
see
also
http://
www/
iit.
edu/~
sgendel/
fa.
htm);
Metcalfe,
D.,
et
al.,
1996.
Assessment
of
the
allergenic
potential
of
foods
derived
from
genetically
engineered
crop
plants.
Critical
Reviews
in
Food
Science
and
Nutrition
36(
S):
S165­
S186;
Bairoch,
A.,
1999.
Nomenclature
and
index
of
allergen
sequences.
SWISS­
PROT
Protein
Sequence
Databank.
Release
38.0,
July
1999
and
updates
to
September
1999
(
see
also
http://
www.
ex[
asy.
ch/
cgibin
lists?
allergen.
txt)

To
test
the
sensitivity
of
the
databases
and
search
engines
to
detect
as
little
as
8
contiguous
amino
acids,
a
positive
control
of
8
aa
was
hypothetically
inserted
in
to
the
sequence
data
of
the
13.6
and
43.8
kDa
proteins
from
a
randomly
chosen
sequence
of
the
public
database
(
TrEMBL
accession
Q45630;
Bacillus
sp.
bacteriolytic
enzyme).
These
modified
sequences
were
then
run
through
the
database
screen
for
homology
comparison
for
each
of
the
databases.
A
sequence
of
8
amino
acid
residues
is
considered
the
minimum
for
an
immunological
recognition.

Sequence
comparison
­
Pattern
matching
­
A
UNIX
Bourne
shell
script
was
used
to
parse
the
sequences
of
the
13.6
and
43.8
kDa
PS149B1
proteins
into
all
possible
eight
letter
scripts.
A
data
file
was
generated
for
input
into
the
GCG
FINDPATTERNS
program.

GCG
FINDPATTERNS
was
used
to
compare
each
of
the
four
allergen
databases
with
the
generated
pattern
file
allowing
zero
mismatched
characters.
The
Swiss­
Prot,
TrEMBL,
PIR
and
GenPept
databases
were
then
compared
with
the
generated
pattern
file,
allowing
for
zero
mismatches.
Any
hits
containing
a
match
to
a
known
allergen
sequence
were
compared
back
to
the
allergen
databases,
then
evaluated
for
any
obvious
sequence
relationship.
This
method
would
allow
for
matching
between
any
new
entry
into
the
database
which
contained
an
8
residue
match,
but
which
lacked
the
designation
as
an
allergen.

Sequence
comparison
­
Local
alignments
­
Local
amino
acid
alignment
comparisons
were
performed
with
GCG
FASTA
program
with
the
wordsize
set
at
2
and
both
the
blosum50
default
and
the
constructed
identity
matrices
employed
as
recommended
by
Gendel,
1998a).
Gap
creation
penalty
default
value
of
12
and
gap
extension
penalty
of
2
were
used.
Each
of
the
PS149B1
proteins
was
compared
against
all
four
databases.

RESULTS
Pattern
Matching
­
Analysis
of
the
13.6
kDa
or
the
43.8
kDa
protein
from
B.
thuringiensis
PS149B1
failed
to
show
any
identity
of
8
amino
acid
residues
or
more
between
these
sequences
and
any
of
the
known
allergens
in
the
databases.
The
positive
control
protein
sequence
was
identified
in
the
searches.

The
full
protein
database
search
revealed
5
sequences
with
homology
of
8
amino
acids
with
either
the
13.6
or
43.8
kDa
PS149B1
proteins.
However,
none
of
these
were
from
allergenic
proteins
as
indicated
in
the
database.

Local
alignments
­
Using
the
constructed
identity
matrix
or
the
blosum50
matrix,
FASTA
alignments
did
not
find
any
immunologically
significant
sequence
homology
with
known
28
allergens.

Conclusion
Based
upon
the
search
requirements
that
at
least
8
contiguous
amino
acids
are
identical
to
a
known
allergenic
sequence,
PS149B1
13.6
and
43.8
kDa
proteins
are
unrelated
to
any
known
allergens
in
that
they
do
not
share
any
linear
epitopes
with
known
protein
allergens.
An
attachment
to
this
report
contains
the
amino
acid
sequence.
