Page
1
of
34
July
7,
2005
MEMORANDUM
SUBJECT:
Transmittal
of
Minutes
of
the
FIFRA
Scientific
Advisory
Panel
Meeting
Held
May
3­
4,
2005:
Scientific
Issues
Associated
With
TSCA
Inventory
Nomenclature
for
Enzymes
and
Proteins
TO:
Charles
M.
Auer,
Director
Office
of
Pollution
Prevention
and
Toxics
FROM:
Paul
I.
Lewis,
Ph.
D.
Designated
Federal
Official
FIFRA
Scientific
Advisory
Panel
Office
of
Science
Coordination
and
Policy
THRU:
Larry
C.
Dorsey,
Executive
Secretary
FIFRA
Scientific
Advisory
Panel
Office
of
Science
Coordination
and
Policy
Clifford
J.
Gabriel,
Ph.
D.
Director
Office
of
Science
Coordination
and
Policy
Please
find
attached
the
minutes
of
the
FIFRA
Scientific
Advisory
Panel
open
meeting
held
in
Arlington,
Virginia
from
May
3­
4,
2005.
These
meeting
minutes
address
a
set
of
scientific
issues
being
considered
by
the
U.
S.
Environmental
Protection
Agency
regarding
TSCA
inventory
nomenclature
for
enzymes
and
proteins.

Attachment
Page
2
of
34
cc:

Susan
Hazen
Margaret
Schneider
Mary
Ellen
Weber
Neil
Patel
Oscar
Hernandez
James
Jones
Anne
Lindsay
Janet
Andersen
Debbie
Edwards
Steven
Bradbury
William
Diamond
Richard
Keigwin
Arnold
Layne
Tina
Levine
Lois
Rossi
Frank
Sanders
William
Jordan
Douglas
Parsons
Karen
Chu
Anthony
Britten
Enesta
Jones
Vanessa
Vu
(
SAB)
Greg
Fritz
Mark
Segal
Henry
Lau
FIFRA
SAP
Members
Steven
G.
Heeringa,
Ph.
D.
Janice
Elaine
Chambers,
Ph.
D.,
D.
A.
B.
T.
Stuart
Handwerger,
M.
D.
Gary
Isom,
Ph.
D.

FQPA
Science
Review
Board
Members
Ralph
Bradshaw,
Ph.
D.
Richard
Cammack,
Ph.
D.,
Sc.
D.
Louis
B.
Hersh,
Ph.
D.
Micah
Krichevsky,
Ph.
D.
William
D.
Nes,
Ph.
D.
Robert
S.
Phillips,
Ph.
D.
Joseph
Spradlin,
Ph.
D.
Page
3
of
34
SAP
Report
No.
2005­
03
MEETING
MINUTES
May
3­
4,
2005
FIFRA
Scientific
Advisory
Panel
Meeting,
held
at
the
Holiday
Inn­
Rosslyn
at
Key
Bridge
A
Set
of
Scientific
Issues
Being
Considered
by
the
Environmental
Protection
Agency
Regarding:

TSCA
Inventory
Nomenclature
for
Enzymes
and
Proteins
Page
4
of
34
NOTICE
These
meeting
minutes
have
been
written
as
part
of
the
activities
of
the
Federal
Insecticide,
Fungicide,
and
Rodenticide
Act
(
FIFRA),
Scientific
Advisory
Panel
(
SAP).
This
report
has
not
been
reviewed
for
approval
by
the
United
States
Environmental
Protection
Agency
(
Agency)
and,
hence,
the
contents
of
this
report
do
not
necessarily
represent
the
views
and
policies
of
the
Agency,
nor
of
other
agencies
in
the
Executive
Branch
of
the
Federal
government,
nor
does
mention
of
trade
names
or
commercial
products
constitute
a
recommendation
for
use.

The
FIFRA
SAP
was
established
under
the
provisions
of
FIFRA,
as
amended
by
the
Food
Quality
Protection
Act
(
FQPA)
of
1996,
to
provide
advice,
information,
and
recommendations
to
the
Agency
Administrator
on
pesticides
and
pesticide­
related
issues
regarding
the
impact
of
regulatory
actions
on
health
and
the
environment.
The
Panel
serves
as
the
primary
scientific
peer
review
mechanism
of
the
EPA,
Office
of
Pesticide
Programs
(
OPP)
and
is
structured
to
provide
balanced
expert
assessment
of
pesticide
and
pesticide­
related
matters
facing
the
Agency.
Food
Quality
Protection
Act
Science
Review
Board
members
serve
the
FIFRA
SAP
on
an
ad
hoc
basis
to
assist
in
reviews
conducted
by
the
FIFRA
SAP.
Further
information
about
FIFRA
SAP
reports
and
activities
can
be
obtained
from
its
website
at
http://
www.
epa.
gov/
scipoly/
sap/
or
the
OPP
Docket
at
(
703)
305­
5805.
Interested
persons
are
invited
to
contact
Paul
Lewis,
Designated
Federal
Official,
via
e­
mail
at
lewis.
paul@
epa.
gov.

In
preparing
these
meeting
minutes,
the
Panel
carefully
considered
all
information
provided
and
presented
by
the
Agency
presenters,
as
well
as
information
presented
by
public
commenters.
This
document
addresses
the
information
provided
and
presented
within
the
structure
of
the
charge
by
the
Agency.
Page
5
of
34
TABLE
OF
CONTENTS
Page
Participants
...................................................................................................................
7
Public
Commenters
.......................................................................................................
8
Introduction
..................................................................................................................
8
Summary
of
Panel
Discussion
and
Recommendations
...................................................
8
Panel
Deliberations
and
Response
to
the
Charge..........................................................
11
References
..................................................................................................................
33
Page
6
of
34
SAP
Report
No.
2005­
03
MEETING
MINUTES:
FIFRA
Scientific
Advisory
Panel
Meeting,
May
3­
4,
2005,
held
at
the
Holiday
Inn­
Rosslyn
at
Key
Bridge,
Arlington,
Virginia
A
Set
of
Scientific
Issues
Being
Considered
by
the
Environmental
Protection
Agency
Regarding:

TSCA
Inventory
Nomenclature
For
Enzymes
And
Proteins
Paul
I.
Lewis,
Ph.
D.
Steven
G.
Heeringa,
Ph.
D.
Designated
Federal
Official
FIFRA
SAP
Session
Chair
FIFRA
Scientific
Advisory
Panel
FIFRA
Scientific
Advisory
Panel
Date:
July
7,
2005
Date:
July
7,
2005
Page
7
of
34
Federal
Insecticide,
Fungicide,
and
Rodenticide
Act
Scientific
Advisory
Panel
Meeting
May
3­
4,
2005
TSCA
Inventory
Nomenclature
for
Enzymes
and
Proteins
PARTICIPANTS
FIFRA
SAP
Session
Chair
Steven
G.
Heeringa,
Ph.
D.,
Research
Scientist
and
Director
for
Statistical
Design,
Institute
for
Social
Research,
University
of
Michigan,
Ann
Arbor,
MI
Designated
Federal
Official
Paul
I.
Lewis,
Ph.
D.,
FIFRA
Scientific
Advisory
Panel,
Office
of
Science
Coordination
and
Policy,
U.
S.
Environmental
Protection
Agency,
Washington,
D.
C.

FIFRA
SAP
Members
Janice
Elaine
Chambers,
Ph.
D.,
D.
A.
B.
T.,
William
L.
Giles
Distinguished
Professor
and
Director,
Center
for
Environmental
Health
Sciences,
Mississippi
State
University,
Mississippi
State,
MS
Stuart
Handwerger,
M.
D.,
Director,
Division
of
Endocrinology,
Cincinnati
Children's
Hospital
Medical
Center,
University
of
Cincinnati,
Cincinnati
OH
Gary
Isom,
Ph.
D.,
Professor
of
Toxicology,
School
of
Pharmacy
&
Pharmacal
Sciences
Purdue
University,
West
Lafayette,
IN
FQPA
Science
Review
Board
Members
Ralph
A.
Bradshaw,
Ph.
D.,
Professor,
Department
of
Physiology
and
Biophysics,
School
of
Medicine,
University
of
California
at
Irvine,
Irvine,
CA
Richard
Cammack,
Ph.
D.,
Sc.
D.,
Professor
of
Plant
Biochemistry,
Department
of
Life
Sciences
King's
College
London,
University
of
London,
London,
United
Kingdom
Louis
B.
Hersh,
Ph.
D.,
George
Schwert
Professor
and
Chair,
Department
of
Molecular
and
Cellular
Biochemistry,
College
of
Medicine,
University
of
Kentucky,
Lexington,
KY
Micah
Krichevsky,
Ph.
D.,
Chairman,
Bionomics
International,
Wheaton,
MD
William
D.
Nes,
Ph.
D.,
Program
Director,
Molecular
and
Cellular
Biosciences
Division,
National
Science
Foundation,
Arlington,
VA
Robert
S.
Phillips,
Ph.
D.,
Professor
of
Chemistry,
Biochemistry
&
Molecular
Biology,
Page
8
of
34
University
of
Georgia,
Athens,
GA
Joseph
Spradlin,
Ph.
D.,
Adjunct
Professor,
University
of
Arkansas,
Hot
Springs,
AR
PUBLIC
COMMENTERS
Oral
statements
were
made
by:
Ms.
Alice
Caddow
of
Genencor
International
and
John
Carroll,
Ph.
D.,
of
Novozymes
on
behalf
of
the
Enzyme
Technical
Association
Dan
Robertson,
Ph.
D.,
on
behalf
of
Diversa
Corporation
Mr.
Brent
Erickson
on
behalf
of
the
Biotechnology
Industry
Organization
Ms.
Martha
Marrapese
on
behalf
of
Keller
and
Heckman,
LLP
Written
statements
were
provided
by:
Biotechnology
Industry
Association
Diversa
Corporation
Enzyme
Technical
Association
INTRODUCTION
The
Federal
Insecticide,
Fungicide,
and
Rodenticide
Act
(
FIFRA),
Scientific
Advisory
Panel
(
SAP)
has
completed
its
review
of
a
set
of
scientific
issues
being
considered
by
the
Agency
pertaining
to
its
review
of
the
Toxics
Substance
Control
Act
(
TSCA)
inventory
nomenclature
for
enzymes
and
proteins.
Advance
notice
of
the
meeting
was
published
in
the
Federal
Register
on
March
9,
2005
(
docket
number
OPP­
2005­
0060).
The
review
was
conducted
in
an
open
Panel
meeting
held
in
Arlington,
Virginia,
May
3­
4,
2005.
The
meeting
was
chaired
by
Steven
G.
Heeringa,
Ph.
D.
Paul
Lewis,
Ph.
D.,
served
as
the
Designated
Federal
Official.
Clifford
Gabriel,
Ph.
D.,
Director,
Office
of
Science
Coordination
and
Policy,
EPA,
welcomed
the
Panel
to
the
meeting.
Mr.
Neil
Patel,
Associate
Director,
Economics,
Exposure
and
Technology
Division,
Office
of
Pollution
Prevention
and
Toxics,
EPA,
provided
introductory
remarks.
Greg
Fritz,
Ph.
D.,
Chemist,
Economics,
Exposure,
and
Technology
Division,
Office
of
Pollution
Prevention
and
Toxics,
EPA,
provided
a
background
on
the
TSCA
inventory
system.
Mark
Segal,
Ph.
D.,
Microbiologist,
Risk
Assessment
Division,
Office
of
Pollution
Prevention
and
Toxics,
EPA,
reviewed
the
Agency's
proposed
approach
for
enzyme
identification
on
the
TSCA
inventory.

SUMMARY
OF
PANEL
DISCUSSION
AND
RECOMMENDATIONS
The
FIFRA
SAP
considered
each
of
the
four
data
elements
(
function,
sequence,
source,
and
processing)
the
Agency
is
proposing
to
employ
for
a
comprehensive
listing
and
for
distinguishing
among
enzymes
on
the
TSCA
Inventory.
The
Panel
concluded
that
each
of
these
four
elements
has
merit
for
cataloging
and
distinguishing
among
enzymes,
and
that
the
Agency's
proposed
nomenclature
system
is
useful
and
should
be
retained.
However,
each
of
the
four
elements
do
not
necessarily
carry
equal
weight,
and
some
elements
may
not
be
known
and
likely
Page
9
of
34
are
difficult
to
obtain.
The
Panel
believed
that
many,
if
not
the
majority
of
its
responses
were
most
relevant
when
applied
to
purified
or
isolated
enzyme
preparations,
which
as
noted
above
is
often
not
the
case
with
commercial
enzymes.
The
application
of
the
four
elements
to
impure
enzyme
preparations
could
be
useful,
but
in
many
cases
the
information
available
is
incomplete.

In
considering
the
four
elements,
the
Panel
concluded
that
the
identification
of
enzyme
function
is
of
prime
importance.
However,
function,
by
itself
may
be
too
broad
a
criteria.
Sequences
can
differ,
yet
activity
remains
the
same.
Source,
although
less
useful,
should
not
be
ignored.
Processing
information
can
be
of
importance
in
certain
cases.

The
Panel,
in
preparing
and
reporting
its
deliberations,
noted
that
commercial
enzyme
preparations
generally
lack
homogeneity
and
indeed
in
most
cases,
there
is
no
attempt
by
the
manufacturer
to
purify
the
enzymatic
activity.
Since
classifications
using
descriptors
based
on
sequence,
source
or
processing
largely
presuppose
some
degree
of
purity,
their
indiscriminate
inclusion
in
any
classification
scheme
would
not
serve
either
the
agency
or
the
commercial
purveyors
well.
However,
the
Panel
did
concur
that
improvements
in
identification
and
classification
were
needed
and
reported
that
in
many
instances
information
from
one
or
more
of
the
proposed
categories,
when
added
to
a
functional
description,
could
materially
aid
in
developing
and
maintaining
an
improved
classification
scheme
without
overly
burdening
either
industry
or
the
government
for
collecting
and
categorizing
such
information.

Use
of
the
functionally
based
International
Union
of
Biochemistry
and
Molecular
Biology
(
IUBMB)
nomenclature
system
is
recommended.
The
additional
knowledge
of
reaction
conditions
and
non­
catalytic
functions
of
specific
enzymes
was
believed
to
be
of
limited
use.
On
the
other
hand,
knowledge
of
the
binding
specificity
and
catalytic
mechanism
were
thought
to
be
useful.
Although
some
enzymes
are
multifunctional,
these
are
relatively
few
in
number.
In
the
absence
of
post­
translational
modifications,
knowledge
of
the
amino
acid
sequence
of
an
enzyme
provides
an
exact
chemical
description
of
the
molecule.
However,
large
variations
in
the
sequences
of
enzymes
catalyzing
the
identical
reaction
makes
sequence
not
as
useful
a
criterion
for
describing
function.

The
Agency
addressed
the
issue
of
the
expected
amount
of
variation
in
an
enzyme
amino
acid
sequence
from
various
sources.
When
an
enzyme
is
cloned
from
its
gene,
random
errors
can
be
introduced;
however
these
are
readily
detected
by
DNA
sequencing.
Such
variants
could
be
altered
in
their
stability,
specificity,
or
catalytic
efficiency.
During
production
of
an
enzyme,
variation
in
amino
acid
sequence
can
occur
as
a
result
of
transcriptional
as
well
as
translational
errors.
However,
these
occur
at
a
very
low
frequency
and
the
Panel
concluded
they
would
not
represent
a
significant
fraction
of
the
enzyme
preparation.
The
Panel
knew
of
no
case
in
which
an
enzyme
used
in
commerce
or
research
changed
its
amino
acid
sequence
over
time.

The
Panel
addressed
the
Agency's
question
as
to
the
level
of
maximum
permissible
overall
amino
acid
sequence
variations
that
could
be
determined
when
identifying
a
specific
enzyme.
The
Panel
concluded
that
one
cannot
predict
the
differences
that
might
occur
among
enzyme
variants.
Page
10
of
34
In
some
cases
a
single
amino
acid
change
can
drastically
alter
the
enzyme
while
in
other
cases
large
sequence
variants
catalyze
the
same
reaction.
Changes
at
the
active
site
or
at
specific
functional
motifs
have
a
much
greater
effect
than
changes
at
other
regions.
The
same
can
be
said
for
deletions
and/
or
excisions;
the
site
at
which
they
occur
determines
to
what
extent
they
will
change
enzyme
activity.
It
would
be
difficult
for
an
enzyme
manufacturer
to
determine
the
location
of
the
active
site
or
other
specific
regions,
as
this
requires
techniques
generally
not
used
by
enzyme
manufacturers.

The
Agency
inquired
as
to
the
efficacy
of
existing
sequencing
technologies.
The
Panel
pointed
out
that
both
protein
and
nucleotide
sequencing
are
reliable
ways
of
measuring
amino
acid
sequence,
but
due
to
rapidity
and
cost
most
protein
sequences
are
determined
from
their
nucleotide
sequence.
In
conjunction
with
amino
acid
sequence,
knowledge
of
post­
translational
modifications
is
important
as
these
can
affect
both
structure
and
function.

Regarding
knowledge
of
the
source
of
an
enzyme,
the
Panel
concluded
that
knowing
the
original
source
as
well
as
the
production
source
will
be
of
limited
value
in
differentiating
enzymes.
If
the
original
source
was
used
as
an
identification
element
to
discriminate
among
enzymes,
the
lowest
taxonomic
level
available
should
be
utilized.
However,
it
was
noted
that
scientific
nomenclature
codes
only
extend
down
to
the
subspecies
level.
Knowledge
of
the
tissue
or
organ
source
will
be
of
value
in
a
relatively
limited
number
of
cases.
Similarly,
for
enzymes
only
characterized
by
activity,
the
chemical,
geographic,
and/
or
environmental
condition
from
which
source
organisms
were
isolated
may
be
a
useful
descriptor.
This
is
not
the
case
when
manipulating
the
enzyme's
original
source
prior
to
gene
transfer
or
manipulating
the
production
source.

In
reference
to
processing,
the
Panel
concluded
that
processing
techniques
generally
will
not
affect
the
structure
of
an
enzyme.
However
there
are
documented
cases
when
this
is
not
the
case.
The
use
of
detergents,
and
autolysis
by
endogenous
or
exogenously
added
proteases,
could
change
the
structure
of
a
protein
and
its
activity.
For
example,
the
milk
activity
referred
to
as
xanthine
oxidase,
is
actually
xanthine
dehydrogenase,
and
becomes
an
oxidase
after
treatment
with
thiol
reagents
or
proteases.
This
is
an
example
where
a
simple
processing
step
gives
you
an
enzyme
with
different
protein
structure
and
catalytic
function.
The
Panel
noted
that
although
there
are
a
host
of
processing
techniques
utilized
in
the
research
laboratory,
most
of
these
are
not
utilized
for
the
majority
of
enzymes
on
the
TCSA
inventory.
Changes
in
processing
techniques
are
not
likely
to
occur
with
industrial
enzymes,
and
any
such
changes
would
not
be
expected
to
significantly
alter
an
enzyme's
chemical
structure
or
its
properties.

Other
factors
that
might
be
beneficial
to
describe
an
enzyme
include
the
minimum
structural
requirement
for
a
substrate,
the
mechanism
of
catalysis,
the
presence
or
absence
of
regulatory
sites,
the
assay
used
to
define
an
enzyme's
activity,
and
the
identification
of
enzymes
with
multiple
non­
identical
subunits.
Page
11
of
34
PANEL
DELIBERATIONS
AND
RESPONSE
TO
THE
CHARGE
The
specific
issues
to
be
addressed
by
the
Panel
are
keyed
to
the
Agency's
background
documents,
references
and
Agency's
charge
questions.

EPA
is
proposing
the
use
of
four
data
elements
(
function,
sequence,
source,
and
processing)
for
comprehensively
listing
and
distinguishing
among
enzymes
on
the
TSCA
Inventory.
The
following
questions
are
intended
to
help
the
Agency
make
a
final
decision
on
how
enzymes
will
be
listed
on
the
Inventory
in
the
future.

General
Comment
The
Panel
addressed
the
class
of
proteins
with
a
catalytic
function,
namely
enzymes,
rather
than
proteins
in
general.
However,
many
of
the
answers
can
be
applied
to
proteins
as
a
group.

Function
The
FUNCTION
of
an
enzyme
refers
to
its
catalytic
activity.
Internationally­
accepted
nomenclature
conventions
of
the
Nomenclature
Committee
of
the
International
Union
of
Biochemistry
and
Molecular
Biology
(
NC­
IUBMB)
describe
and
categorize
enzymes
based
on
their
function.
The
NC­
IUBMB
assigns
enzymes
an
Enzyme
Committee
(
EC)
code
number
based
on
the
specific
reaction(
s)
catalyzed
by
the
enzyme,
the
nature
of
the
bond
involved,
and
the
substrate
acted
upon.
EPA
intends
to
incorporate
function
into
TSCA
Inventory
enzyme
listings
by
using
these
EC
codes
and
the
systematic
name
for
the
specific
catalytic
activity.
In
the
questions
below,
please
identify
the
scientific
merit
for
using
function
information
to
differentiate
among
enzymes
and
identify
what
level
of
detail
regarding
function
would
be
scientifically
appropriate
for
this
purpose.

1.
While
the
Agency
recognizes
the
practical,
historical
advantages
of
using
function
to
describe
enzymes,
in
the
context
of
the
Agency's
need
for
unique
and
unambiguous
naming,
what
is
the
scientific
rationale
for
identifying
an
enzyme
based
on
the
chemical
reaction(
s)
it
catalyzes?

The
Panel
qualified
its
response
indicating
that
this
question
can
best
be
addressed
with
regard
to
isolated
enzymes
(
not
necessarily
pure).
The
Panel
concluded
that
in
terms
of
function,
the
Agency's
proposed
nomenclature
system
is
useful
and
should
be
retained.
Identification
of
an
enzyme
based
on
the
chemical
reaction(
s)
it
catalyzes
permits
a
useful
description
of
an
enzyme.
The
chemical
reaction
catalyzed
by
an
enzyme
is
the
most
essential
piece
of
information
needed
in
any
nomenclature
system
for
enzymes.
The
function
of
an
enzyme
is
more
important,
from
a
practical
point
of
view,
in
naming
an
enzyme
than
any
of
its
structural
features.
The
Enzyme
Comission
(
EC)
list,
maintained
by
the
nomenclature
committee
of
the
International
Union
of
Biochemistry
and
Molecular
Biology
(
NC­
IUBMB;
IUBMB
and
IUPAC/
IUBMB
Joint
Commission
on
Biochemical
Nomenclature,
http://
www.
chem.
qmw.
ac.
uk/
iubmb/
enzyme)
defines
enzymes
based
on
the
reaction
catalysed.
Such
a
system
represents
a
clear
descriptor
of
the
Page
12
of
34
enzyme.
In
many
cases,
the
IUBMB
nomenclature
system,
which
is
based
on
function
(
reaction
catalyzed),
further
classifies
enzymes
based
on
mechanism
(
e.
g.
zinc
metalloproteases)
and
provides
insights
into
functional
groups
involved
in
catalysis.
Grouping
of
enzymes
based
on
their
catalytic
mechanism
frequently
permits
insights
into
structural
features
common
to
the
group.
It
also
provides
a
starting
point
for
defining
physiological
function.
If
one
wanted
to
use
an
enzyme
for
some
specific
purpose,
reference
to
the
grouping
based
on
the
reaction
it
catalyzes
would
be
an
important
starting
point.
The
nomenclature
used
by
the
IUBMB
provides
a
reference
point
where
different
enzymes
catalyzing
the
same
reaction
(
e.
g.
proteases)
might
be
used
for
the
same
purposes,
or
provides
a
way
to
recognize
differences
in
the
reactions
catalyzed
by
similar
enzymes,
i.
e.
proteases
with
different
specificities.

While
the
Panel
did
recognize
limitations
with
the
IUBMB
nomenclature
system,
the
Panel
was
generally
in
support
of
the
IUBMB
approach.
The
IUBMB­
EC
system
is
usually
based
on
measurements
of
enzyme
functions
in
an
aqueous
system
reacting
with
a
particular
substrate.
In
some
cases,
this
may
not
be
relevant
to
industrial
uses.
There
are
other
complexities
that
make
the
IUBMB
nomenclature
of
categorizing
enzymes
based
on
the
reaction
they
catalyze
awkward.
For
reasons
of
consistency,
the
system
describes
the
reaction
catalysed
by
a
formalism
that
does
not
necessarily
reflect
the
thermodynamically
favored
direction
of
the
reaction
nor
the
direction
of
the
reaction
in
metabolism.
Categorizing
an
enzyme
based
on
the
chemical
reaction(
s)
it
catalyzes
doesn't
define
specificity
in
a
precise
manner
since
enzymes
may
react
with
multiple
substrates.
Furthermore,
the
enzyme
name
does
not
always
reflect
its
commercial
use
(
e.
g.
xylose
isomerase
is
used
for
catalyzing
fructose
formation).

For
historical
reasons,
the
depth
of
coverage
of
enzymes
with
broad
and
narrow
specificities
is
not
consistent
throughout
the
nomenclature
list.
The
peptidases
(
proteases)
are
listed
in
great
detail,
including
in
some
cases
structure
and
species,
whereas
other
enzymes
are
listed
only
as
broad
classes.
For
example,
the
restriction
endonucleases,
are
not
described
in
the
IUBMB
system
of
naming
enzymes
by
the
reaction
they
catalyze,
but
rather
reference
is
made
to
lists
in
other
databases.
Another
potential
problem
is
that
some
activities
have
two
or
more
EC
numbers
depending
on
whether
the
enzymes
have
broad
or
narrow
specificity.
It
is
estimated
that
about
20%
of
the
enzyme
names
in
the
literature
are
incorrect.

The
IUBMB
number
tells
us
basic
functionality.
More
information
such
as
specificity
would
enhance
the
value
of
the
classification.
Overall,
there
is
more
of
an
interest
in
function
than
structure.
However,
without
the
IUBMB
number,
or
other
description
of
the
reaction
catalyzed,
no
useful
information
is
available.
Bioinformatics
databases
treat
the
IUBMB
number
as
an
identifier
and
by
compiling
all
the
data,
get
an
optimum
description
or
definition
of
an
enzyme.

The
one
property
that
all
enzymes
have
in
common
other
than
being
proteins
is
that
they
catalyze
at
least
one
chemical
reaction.
Clearly
no
list
of
enzymes
should
omit
this
parameter.
However,
it
needs
to
be
recognized
that
there
can
be
ambiguity
involved,
particularly
with
enzymes
whose
activity
is
very
general
such
as
a
non­
specific
protease.
It
is
also
clear
that
none
of
the
databases
or
classification
systems
is
complete,
although
all
have
some
value.
Page
13
of
34
2.
How
precise
is
the
IUBMB
EC
categorizing
system
for
describing
enzyme
function?
For
example,
in
addition
to
the
EC
function
category
to
which
an
enzyme
belongs,
what
additional
information
about
enzyme
structure
and/
or
chemical
properties,
if
any,
would
be
gained
by
a
more
detailed
functional
description
that
included
a.
enzyme
reaction
conditions
(
e.
g.,
pH
range,
reaction
temperature
range)?

The
Panel
concluded
that
at
this
time
little
would
be
gained
by
including
enzyme
reaction
conditions.
The
properties
listed,
including
pH
range,
etc.
are
standard
properties
that
are
generated
during
a
typical
characterization
of
the
protein
in
order
to
establish
its
activity
assay.
These
properties
can
vary
for
a
given
enzyme
or
amongst
enzyme
systems,
depending
on
its
source.
The
Panel
concluded
that
knowledge
of
additional
information
about
enzyme
structure
and/
or
chemical
properties
in
order
to
provide
a
more
detailed
functional
description
would
be
of
limited
use.
The
difficulty
of
comparing
enzyme
preparations
at
different
times
and
in
different
places
has
been
recognized.
Hicks
and
Kettner
(
2003)
described
efforts
to
establish
internationally
agreed
Experimental
Standard
Conditions
of
Enzyme
Characterizations
for
the
measurement
(
assay)
of
enzyme
activity
in
an
initiative
led
by
the
Beilstein
Institute.

b.
non­
catalytic
enzyme
functions
that
are
not
represented
by
EC
codes
(
e.
g.,
binding
properties)?

The
issue
of
non­
catalytic
enzyme
functions
does
not
appear
to
be
important
to
the
inventory
issue.
Additional
information
such
as
allosteric
modulators
(
effectors)
can
be
a
useful
indicator
of
protein
function.
Interacting
proteins
can
also
be
an
important
indicator
of
effector
action.

Information
about
the
binding
properties
of
the
active
site
is
important
in
predicting
what
other
substrates
an
enzyme
will
use
besides
the
originally
described
substrate.
The
binding
properties
are
also
needed
to
predict
the
"
reversion"
products
formed
when
the
catalyzed
reaction
is
at
or
near
thermodynamic
equilibrium.
In
hydrolyses,
the
size
distribution
of
polymer
fragments
depends
heavily
on
the
structure
of
the
binding
site.

c.
other
additional
information
about
function
that
could
be
used
to
differentiate
enzymes
(
please
specify
what
would
be
of
value)?

Other
useful
information
about
an
enzyme's
function
would
be
its
catalytic
mechanism.
Enzymes
that
use
the
same
substrate
and
produce
the
same
products
may
do
so
by
totally
different
catalytic
mechanisms
(
e.
g.
chymotrypsin
and
pepsin).

3.
The
Agency
is
trying
to
gauge
the
probable
comprehensiveness
of
enzyme
catalytic
function
descriptions
for
subsequent
enzyme
reporting.
a.
How
common
are
multifunctional
enzymes?

Multifunctional
enzymes
represent
a
relatively
small
percentage
of
the
total
known
Page
14
of
34
enzymes.
Less
than
1%
of
total
enzymes
are
known
to
be
multifunctional
(
including
orthologous
enzymes
in
different
organisms).
A
key
word
search
for
"
multifunctional
enzyme"
in
the
National
Center
for
Biotechnology
Information
(
NCBI)
protein
data
base
resulted
in
1954
hits,
in
Swiss­
Prot
1666
hits
of
180652
entries
(
0.9%)
and
in
TrEMBL
4011
hits
out
of
1689375
entries
(
0.2%).

Some
multifunctional
enzymes
are
multienzyme
complexes
with
separate
chains
for
each
activity.
Other
multifunctional
enzymes
are
single
chains
with
multiple
catalytic
sites.
In
some
cases,
a
multienzyme
complex
in
one
organism
may
be
a
single
chain
in
another
organism.
Some
enzymes
are
promiscuous,
and
can
catalyze
multiple
activities
in
a
single
site.
It
is
important
to
keep
in
mind
the
distinction
between
enzymes
with
broad
specificity
(
e.
g.
cytochromes
P450)
and
truly
multifunctional
enzymes
(
such
as
ribulose
bisphosphate
carboxylase/
oxygenase).
The
simplest
way
to
make
that
distinction
is
by
catalytic
activity.
Enzymes
may
have
unrelated
structural
or
regulatory
functions,
in
addition
to
catalytic
activity.
These
can
include
protein
machines
(
e.
g.,
proteasomes)
that
contain
both
catalytic
and
non­
catalytic
components.
It
may
also
be
useful
to
distinguish
multisubstrate
enzymes
and
truly
multifunctional
enzymes.

b.
How
frequently
are
new
catalytic
functions
for
existing
enzymes
discovered?

The
discovery
of
new
catalytic
functions
for
enzymes
is
an
infrequent
event.
Based
on
the
low
abundance
of
multifunctional
enzymes
(
less
than
1%
of
total),
the
reporting
of
a
new
catalytic
function
for
an
enzyme
would
be
expected
to
be
an
infrequent
event.
However
a
search
of
PubMed
for
2005
publications
shows
three
new
activities
described
for
previously
identified
proteins.

In
the
past,
the
discovery
of
multifunctional
enzymes
has
often
been
related
to
the
elucidation
of
metabolic
pathways.
Enzymes
that
catalyze
sequential
steps
in
metabolism
may
be
fused
into
a
single
chain.
If
the
activities
are
not
related,
then
the
discovery
of
multifunctional
enzymes
is
usually
due
to
accident.
Researchers
investigating
different
activities
may
find
that
after
purification
and
sequence
determination,
both
enzyme
activities
are
due
to
the
same
protein.

c.
How
good
are
existing
models
to
assess
the
likelihood
that
an
enzyme
may
have
several
catalytic
functions?

There
are
no
general
models
to
discover
new
activities
in
known
enzymes.
Sequence
alignment
of
a
known
enzyme
with
genome
databases
may
show
homology
to
another
enzyme
sequence,
which
may
help
to
indicate
another
activity.
If
a
new
sequence
shows
homology
to
two
different
enzyme
sequences,
that
may
provide
an
indication
that
there
are
two
activities
in
a
single
enzyme.
However,
homology
is
not
a
reliable
predictor
of
a
particular
activity.

d.
What
information
is
required
to
utilize
such
models?

There
is
no
particular
piece
of
information
that
can
be
used
to
assess
the
likelihood
that
an
Page
15
of
34
enzyme
may
have
several
catalytic
functions.
Proteomics
has
opened
the
field
for
finding
interacting
proteins,
and
thus
it
may
be
possible
to
make
predictions
of
activity
based
on
proteinprotein
interactions.
Promiscuous
enzymes
and
those
with
additional
regulatory
or
structural
functions
are
particularly
difficult
cases
for
the
prediction
of
novel
activities
or
functions
from
sequence
data.
In
an
impure
mixture,
such
as
those
used
in
an
industrial
setting,
it
will
be
impossible
to
determine
if
two
different
enzyme
activities
are
properties
of
a
multifunctional
enzyme
or
two
different
enzymes.

Sequence
The
AMINO
ACID
SEQUENCE
of
an
enzyme
is
known
as
its
primary
structure.
It
is
a
systematic
representation
of
the
linear
sequence
of
amino
acids
that
are
connected
via
amide
bonds
to
form
a
polypeptide.
In
the
questions
below,
please
consider
what
scientific
support
there
is
for
using
sequence
information
to
differentiate
among
enzymes
and
what
level
of
detail
would
be
scientifically
appropriate
for
this
purpose.

4.
What
information
about
an
enzyme
could
be
gained
by
identifying
it
based
on
its
amino
acid
sequence?

The
amino
acid
sequence
of
an
enzyme
defines
its
covalent
structure
(
minus
any
co/
post
translational
modifications
and
prosthetic
groups).
This
is
analogous
to
the
structure
of
any
other
organic
molecule
and
can
be
determined
very
accurately.
Not
withstanding
genetic
variation,
this
can
be
viewed
as
the
`
fingerprint'
of
this
molecule.
Clearly,
an
enzyme
could
be
uniquely
identified
by
its
sequence.
There
is
a
good
reason
that
most
biotechnology
patents
on
matter
are
written
on
sequences
of
either
the
protein
or
the
DNA
encoding
it.
Sequence
information
also
often
provides
structural
and
functional
information
of
possible
interest
in
the
commercial
application
of
enzymes.
Since
the
enzyme
sequence,
if
properly
folded,
will
virtually
always
lead
to
the
structure
that
produces
its
function(
s),
in
this
case
catalysis,
it
is
a
more
absolute
descriptor
than
any
other
information
available.

It
should
be
noted
however
that
there
are
sequence
variations
of
the
same
enzymes
with
the
same
catalytic
function
within
an
organism
and
among
organisms.
Thus
one
might
accumulate
a
long
list
of
sequences
for
the
same
functional
enzyme
that
might
vary
in
a
few
amino
acids
or
might
vary
by
a
significant
number
of
amino
acids.
As
a
means
to
address
this,
deciding
when
enzyme
structural
relationships
can
be
ascertained
from
sequence
identity
and
sequence
similarity
calculations
is
helpful.
These
do
not
necessarily
define
function,
although
they
may
define
a
common
reaction
mechanism
used
by
a
group
of
related
enzymes
performing
different
functions.
In
this
regard,
phylogenetic
analyses
can
be
a
useful
adjunct
in
using
this
information
as
a
descriptor.

Thus
although
sequence
information
will
clearly
define
an
enzyme
chemically,
the
large
variations
expected
among
enzymes
catalyzing
the
identical
reaction,
and
thus
having
the
same
function,
makes
this
potentially
a
daunting
task.
This
is
probably
not
such
a
problem
with
genetically
engineered
enzymes,
as
the
changes
made
will
have
been
defined.
On
the
other
hand
Page
16
of
34
any
listings
of
enzymes
based
entirely
on
exact
sequence
would
be
large
and
cumbersome
if
one
considers
both
natural
and
engineered
mutations.

5.
The
Agency
is
trying
to
assess
the
expected
amount
of
variation
in
an
enzyme
amino
acid
sequence
due
to
various
causes
in
spite
of
current
quality
control
standards.

When
an
enzyme
is
cloned
from
one
source
for
expression
in
the
same
or
a
heterologous
source
(
i.
e.
a
gene
from
one
organism
expressed
in
another),
the
preparation
of
the
gene
can
introduce
random
errors
based
on
the
fidelity
of
the
system
used
to
copy
it.
This
can
lead
to
random
point
mutations
in
the
enzyme,
which
can
vary
widely
in
their
resultant
effect.
On
the
one
hand
these
mutations
may
have
no
effect
or
they
can
lead
to
an
enzyme
with
reduced
or
no
activity,
or
an
enzyme
with
a
lower
stability.
However,
in
the
absence
of
DNA
sequencing,
the
errors
can
go
undetected.

When
an
enzyme
is
produced
in
batches
it
is
usually
generated
from
its
cDNA
with
an
appropriate
expression
vector.
In
this
case,
there
are
then
two
general
sources
of
errors
that
can
lead
to
a
variation
in
an
enzyme
amino
acid
sequence;
nucleotide
changes
(
mutations)
in
the
gene
that
arise
from
transcription
errors
and
misincorporation
of
amino
acids
into
the
enzyme
through
translational
errors.

a.
How
much
and
what
type
of
variation
(
including
substitutions,
deletions,
and
additions)
can
be
expected
in
the
amino
acid
sequence
of
an
enzyme
produced
in
multiple
batches
that
will
arise
due
to
unintended
differences
in
production
conditions?
Estimate
a
percentage,
number
of
residues,
or
other
quantifiable
measure
of
variation.

As
noted
above,
variations
in
the
amino
acid
sequence
of
an
enzyme
produced
in
multiple
batches
can
arise
from
mutations
introduced
during
transcription
and
through
translation
errors.
By
far
the
most
common
change
will
be
amino
acid
substitutions.
Deletions
and
additions
would
more
likely
lead
to
unstable
proteins.
The
more
replications
of
a
cloned
gene,
the
greater
the
likelihood
of
a
mutation.
Since
the
mutation
rates
are
low,
if
the
enzyme
were
produced
in
multiple
batches,
one
would
expect
a
relatively
low
number
of
amino
acid
sequence
changes
to
arise,
and
these
would
be
different
in
each
batch
of
enzyme.
It
is
difficult
to
quantitate
the
number
of
expected
changes
since
they
are
dependent
upon
a
number
of
variables
including
the
enzyme
itself,
the
expression
system
used,
and
the
growth
conditions.
When
multiple
batches
are
used,
the
stable
variations
will
continue
to
accumulate
from
batch
to
batch.
Manufacturers
will
avoid
long­
term
deterioration
by
re­
starting
the
fermentation
from
stock
cultures.

b.
How
much
and
what
type
of
variation
(
including
substitutions,
deletions,
and
additions)
can
be
expected
in
the
amino
acid
sequence
of
an
enzyme
within
a
given
sample
of
a
single
production
batch
due
to
individual­
level
variation
in
an
enzyme­
producing
population?
Estimate
a
percentage,
number
of
residues,
or
other
quantifiable
measure
of
variation.

The
same
type
of
variation
observed
in
multiple
batches
(
including
substitutions,
deletions,
and
additions)
will
be
expected
in
the
amino
acid
sequence
of
an
enzyme
within
a
single
production
batch.
However
since
these
will
not
be
cumulative
they
will
be
present
at
a
much
lower
frequency
relative
to
an
enzyme
produced
in
multiple
batches.
Page
17
of
34
The
error
rate
of
transcription
is
generally
low
but
significant
(
about
1
in
107).
The
more
replications
of
a
cloned
gene,
the
greater
the
likelihood
that
it
will
become
mutated.

c.
How
much
and
what
type
of
variation
(
including
substitutions,
deletions,
and
additions)
can
be
expected
in
the
amino
acid
sequence
of
an
enzyme
across
multiple
samples
collected
over
time
(
e.
g.,
in
microbial
cultures
stored
for
extended
periods)
due
to
changes
in
an
enzyme­
producing
population?
Estimate
a
percentage,
number
of
residues,
or
other
quantifiable
measure
of
variation.

i.
Over
what
time
scale
will
such
variation
arise?
That
is,
is
there
a
predictable
relationship
between
the
amount
of
variation
and
the
length
of
time
in
culture?

Differences
in
protein
sequence
can
arise
from
mutations
in
the
gene
or
from
translational
errors.
Mutations
in
the
gene
causing
differences
in
protein
sequence
arise
at
the
mutation
frequency
of
the
system
used
to
express
it.
For
E.
coli
there
are
between
10­
6
and
10­
7
mutations
per
gene
per
generation
and
assuming
any
one
of
three
mutations
could
affect
a
given
amino
acid,
this
would
turn
out
to
be
a
maximal
value
of
3x10­
6
to
3x
10­
7
amino
acid
changes
per
generation.
However,
due
to
the
fact
that
many
base
changes
do
not
affect
amino
acid
coding,
this
number
will
be
significantly
smaller.
If
any
of
these
random
mutations
caused
instability,
the
protein
would
likely
be
cleared
from
the
cell
by
proteolysis.
Thus,
if
the
culture
were
not
under
continuous
culture
for
many
generations,
one
would
expect
a
negligible
number
of
amino
acid
sequence
changes
to
arise
from
mutations,
and
these
would
be
different
in
each
batch
of
enzyme.

Translational
errors
occur
at
approximately
10
times
the
frequency
as
transcriptional
changes
(
Rosenberger
1994a).
During
recombinant
protein
synthesis,
translational
errors
will
occur
at
specific
sites
in
the
protein
with
each
translational
error
occurring
at
an
error
rate
of
2x10­
3
to
2x10­
4
(
Rosenberger
1994a)
and
thus
representing
a
small
but
significant
percentage
of
the
total.
Rosenberger
(
Scorer
et
al.
1991;
Rosenberger
1994b)
measured
the
error
rate
in
synthetic
recombinant
mouse
epidermal
growth
factor
(
EGF)
produced
in
E.
coli
by
measuring
the
phenylalanine
content
in
this
non­
phenylalanine
containing
protein.
It
was
found
that
phenylalanine
was
mis­
incorporated
into
1.1%
of
all
the
amino
acids
present
and
2.6%
of
those
amino
acids
whose
codons
differed
by
a
single
base.

However,
the
translational
error
rate
can
be
protein
dependent.
Weickert
and
Apostol
(
1998)
measured
isoleucine
incorporation
into
coexpressed
di­
alpha­
globin
and
beta­
globin
expressed
in
E.
coli.
They
found
 
0.2
mol
of
isoleucine
per
mol
of
hemoglobin
which
corresponds
to
a
translation
error
rate
of
 
0.001.
They
concluded
that
this
is
not
different
from
typical
translation
error
rates
found
for
other
E.
coli
proteins.
Two
different
expression
systems
that
resulted
in
accumulation
of
globin
proteins
to
levels
equivalent
to
~
20%
of
the
level
of
E.
coli
soluble
proteins
also
resulted
in
equivalent
translational
fidelity.

In
yet
another
study,
Kane
et
al.
(
1992)
found
that
about
2%
of
recombinant
bovine
placental
lactogen
(
bPL)
produced
from
E.
coli
using
rare
arginine
codons
exhibited
an
altered
Page
18
of
34
trypsin
digestion
pattern.
This
was
not
the
case
when
a
preferred
codon
was
used.
They
proposed
a
model
in
which
translational
pausing
occurred
at
the
arginine
residues
encoded
by
an
AGG
codon
because
the
corresponding
arginyl­
tRNA
species
is
reduced
by
the
high
level
of
bPL
synthesis,
and
a
translational
hop
occurs
from
the
leucine
residue
85
TTG
codon
to
the
leucine
residue
87
TTG
codon.
Thus
misincorporation
of
amino
acids
may
depend
on
the
protein
itself
as
well
as
codon
usage
and
the
system
utilized.

Misincorporation
of
amino
acids
occurs
at
an
error
rate
~
10
times
that
of
stop­
codon
readthrough
errors
and
frameshift
errors,
the
latter
being
estimated
to
occur
at
about
1/
2
the
frequency
as
stop­
codon
read­
through
errors
(
Rosenberger
994a).
The
above
numbers
represent
single
batch
data,
and
thus
in
multiple
batches
the
number
of
errors
would
increase.

The
use
of
multiple
samples
collected
over
time
will
reduce
the
accumulation
of
mutations
in
the
culture.
Due
to
the
variables
noted
above
there
is
not
an
easily
quantifiable
relationship
between
the
amount
of
variation
and
the
length
of
time
in
culture.

ii.
What
kinds
of
changes
might
occur
to
an
enzyme
preparation
if
naturally
occurring
variants
become
the
dominant
component
(
e.
g.,
changes
in
rates
of
activity,
reactions
catalyzed,
substrate
range,
response
to
environmental
conditions)?

If
naturally
occurring
variants
became
the
dominant
component,
any
one
of
a
number
of
changes
could
occur
that
could
include
changes
in
the
rate
in
which
the
enzyme
catalyzes
the
reaction
(
usually
slower),
subtle
changes
in
the
enzyme's
substrate
specificity,
possible
changes
in
the
enzyme's
response
to
environmental
conditions,
and
changes
in
stability.
However
it
would
be
very
unlikely
that
the
reaction
mechanism
would
change.

The
production
of
commercial
industrial
grade
enzymes
is
usually
if
not
always
started
from
a
frozen
stock
culture
and
rarely
or
never
from
a
sample
from
a
previous
batch.
This
is
done
to
ensure
that
the
enzymes
produced
from
batch
to
batch
are
the
same
and
that
any
variations
that
may
occur
during
a
production
run
are
never
passed
on
to
future
production
runs.
Very
large
numbers
of
frozen
cultures
are
produced
from
one
highly
controlled
fermentation
and
then
used
in
production
over
several
years.
This
minimizes
the
number
of
replications
and
their
cumulative
mutations.
Therefore,
the
amino
acid
sequence
can
remain
unchanged
for
decades.
Finally,
manufacturers
will
avoid
long­
term
deterioration
by
re­
starting
the
fermentation
from
stock
cultures.

iii.
Have
any
enzymes
in
commerce
or
research
been
known
to
change
in
amino
acid
sequence
over
time?
Have
any
been
known
to
remain
unchanged
in
amino
acid
sequence
for
a
year/
decade
or
longer?

In
the
drug
field,
it
is
well
established
that
biologics
do
not
change
significantly
over
Page
19
of
34
periods
of
at
least
ten
years.
The
Panel
could
not
think
of
any
specific
examples
of
enzymes
in
commerce
or
research
that
are
known
to
have
changed
in
amino
acid
sequence
over
time.
However,
the
Panel
pointed
out
that
the
whole
process
of
evolution
leads
to
changes
in
amino
acid
sequence
over
time.

6.
EPA
is
trying
to
judge
whether
a
scientifically
appropriate
level
of
maximum
permissible
overall
amino
acid
sequence
variation
could
be
determined
when
identifying
a
specific
enzyme.

a.
What
types
of
differences
may
exist
among
enzyme
variants
that
differ
by
a
single
amino
acid
change?
What
types
of
differences
may
exist
among
enzyme
variants
that
differ
in
amino
acid
composition
by
0.5%?
1%?
10%?
etc.?

The
Panel
could
not
specify
the
differences
among
enzyme
variants
that
vary
by
a
single
amino
acid
change,
because
the
consequences
of
such
changes
vary
enormously.
In
some
cases,
a
single
amino
acid
change,
e.
g.
in
the
active
site
can
virtually
eliminate
activity.
On
the
other
hand,
two
proteins
having
only
20%
homology
can
have
a
closely
similar
protein
folding
and
activity.
Usually
a
single
or
few
amino
acid
differences
have
only
a
minor
effect
on
activity
or
specificity;
the
principal
effect
is
on
enzyme
stability.

b.
How
much
does
the
region
of
the
enzyme
in
which
the
variation
occurs
matter?
For
example,
how
important
are
changes
in
the
amino
acid
sequence
of
the
active
site
versus
the
rest
of
the
molecule?
Are
there
other
regions
of
the
enzyme
that
are
considered
important,
i.
e.,
where
sequence
is
generally
conserved?

The
region
of
an
enzyme
in
which
variation
occurs
is
very
important.
Changes
at
surface
residues
generally
will
have
little
effect
while
changes
in
the
active
site
will
have
dramatic
effects.
It
is
noted
in
the
literature
(
Carter
and
Wells,
1988)
that
mutation
of
the
active­
site
serine
of
subtilisin,
for
example,
causes
activity
to
decrease
by
a
factor
of
106.
The
activity
is
still
greater
however
than
the
uncatalysed
reaction.
Amino
acid
changes
in
the
hydrophobic
core
of
a
protein
are
more
likely
to
affect
stability
than
those
on
the
surface
of
the
molecule.

In
addition
to
the
active
site,
there
are
other
regions
of
the
enzyme
that
are
considered
important
and
where
amino
acid
changes
will
likely
have
a
dramatic
effect.
Certain
arrangements
of
amino
acids,
known
as
motifs,
can
be
recognized
in
the
amino­
acid
sequences
of
enzymes,
which
in
their
folded
state,
have
specific
properties
such
as
the
HXXXH
motif
as
a
zinc
binding
motif.
There
are
many
such
motifs
for
binding
specific
substrates
and
for
binding
cofactors
or
effectors.

c.
How
important
are
deletions
and/
or
excisions
in
determining
differences
between
enzymes?

As
noted
by
the
Panel
in
response
to
question
6a
above,
the
site
of
deletions
and/
or
excisions
is
critical.
On
the
one
hand,
a
small
deletion
may
have
a
dramatic
effect
on
activity,
while
a
long
deletion
may
leave
some
activity
intact,
with
decreased
stability.
Page
20
of
34
d.
How
easy
would
it
be
for
a
typical
enzyme
manufacturer
to
determine
the
location
of
the
active
site
or
other
specific
regions
mentioned
in
6b?

There
are
relatively
few
simple
and
inexpensive
ways
for
an
enzyme
manufacturer
to
determine
the
location
of
the
active
site
or
other
specific
regions
of
a
protein
other
than
through
DNA
sequencing
coupled
with
recognition
of
conserved
motifs,
There
are
however
a
number
of
techniques
commonly
used
by
researchers.
For
example,
changes
in
molecular
weight
caused
by
a
mutation
can
be
readily
detected
using
electrospray
ionization
(
ESI)
mass
spectrometry.
However
this
instrumentation
would
likely
not
be
available
to
an
enzyme
manufacturer.
The
ideal
way
to
identify
the
active
site
is
the
determination
of
the
structure
of
the
enzyme,
containing
its
substrate
by
crystallography.
Academic
researchers
and
the
pharmaceutical
industry
are
increasingly
using
this.
Another
way
to
identify
the
active
site
is
to
use
affinity
labeled
or
"
suicide"
substrates,
if
they
are
available.
The
labeled
active­
site
amino
acids
can
then
be
identified
after
peptide
hydrolysis.
However,
many
manufacturers
of
enzyme
products
do
not
sell
pure
proteins
and
they
are
unlikely
to
have
ready
access
to
the
equipment
used
to
locate
the
active
site
or
other
specific
regions
of
an
enzyme.
Determining
the
active
site
of
an
enzyme
activity
contained
within
an
impure
protein
mixture
typically
is
not
feasible.

7.
EPA
wants
to
assess
the
efficacy
of
existing
sequencing
technologies.

a.
How
accurate
and
reproducible
are
readily
available
amino
acid
sequencing
techniques
and
instrumentation?

Both
protein
and
nucleotide
sequencing
are
reliable
ways
of
determining
the
amino
acid
sequence
of
an
enzyme.
Most
protein
sequences
are
deduced
from
nucleotide
sequences
as
it
is
fast,
cheap
and
very
reliable.
The
error
rate
is
about
1
in
1000
bases
or
less,
and
it
is
common
to
obtain
>
700
bases
in
a
single
sequencing
run.
Nucleotide
sequencing
is
used
routinely
to
obtain
enzyme
sequences
after
any
sort
of
manipulation,
such
as
moving
to
a
new
vector,
mutagenesis,
etc.

The
expense
in
determining
the
amino
sequence
of
an
enzyme
through
amino
acid
sequencing
is
illustrated
by
the
following
fee
structure
for
protein
sequencing
at
the
University
of
California,
San
Diego
(
http://
proteinsequencer.
ucsd.
edu/)
for
users
outside
the
UC
System
working
at
a
'
For­
Profit'
organization:

a.
$
46/
amino
acid
residue
for
amino
acids
1­
10
(
minimum
charge
is
5
residues)

b.
After
10
amino
acids,
residues
11
and
beyond
are
charged
half­
price.

The
National
Center
for
Biotechnology
Information
(
NCBI)
also
provides
information
describing
a
typical
sequencing
process,
a
complex
and
highly
specialized
process.

b.
How
accurate
and
reproducible
are
readily
available
nucleotide
sequencing
techniques
and
instrumentation?
Page
21
of
34
The
accuracy
of
results
and
reproducibility
(
actually,
the
precision)
of
nucleotide
sequencing
techniques
and
instrumentation
is
largely
dependent
on
the
repetition
rate
of
determinations.
The
base
error
rate
is
in
the
range
of
1%
for
a
single
determination
and
decreases
by
repetition
and
reading
the
sequence
from
alternate
(
both)
ends.

c.
Does
the
accuracy
of
the
result
depend
on
the
choice
of
method?

The
accuracy
depends
on
a
variety
of
factors:
sample
preparation
and
storage,
method,
skill
of
operator,
etc.
However,
there
is
no
absolute
measure
of
accuracy
since
some
small
residual
of
uncertainty
remains.

d.
How
rapidly
are
sequencing
techniques
improving
or
new
techniques
being
developed?

History
suggests
that
improvements
and
new
techniques
develop
at
an
ever
increasing
rate.

e.
How
reliably
can
one
predict
the
amino
acid
sequence
of
the
final
gene
product
based
on
the
nucleotide
sequence?

Since
nucleotide
sequencing
is
considered
highly
reliable,
one
can
accurately
predict
the
amino
acid
sequence
of
the
final
gene
product
based
on
the
nucleotide
sequence.
The
reliability
of
the
amino
acid
sequence
is
thus
dependent
on
the
nucleotide
sequence.
Modern
laboratory
techniques
now
allow
the
determination
of
protein
sequences
to
be
performed
almost
exclusively
by
decoding
of
nucleotide
sequences.

8.
What
additional
information
would
be
gained,
if
any,
by
a
more
detailed
structural
description
that
included
in
addition
to
amino
acid
sequence:

While
both
sequence
and
function
are
the
most
useful
properties
for
identifying
an
enzyme,
additional
information
can
be
gained
by
recognizing
other
properties
or
characteristics
of
enzymes.
Information
such
as
glycosylation
sites,
coenzymes
and
cofactors
would
be
helpful.
A
description
of
such
factors
are
described
below.

a.
glycosylation
sites
(
and
the
composition
of
these
carbohydrate
moieties),

Glycosylation
is
important
in
some
cases,
such
as
in
some
eukaryotic
extracellular
proteins.
Glycosylation
usually
has
little
effect
on
enzyme
activity,
although
it
can
affect
stability.
Expression
in
bacteria
or
yeast
will
usually
not
do
this
correctly,
so
eukaryotic
expression
systems
such
as
Baculovirus
in
insect
cells
are
employed.
A
more
important
consequence
for
the
health
of
workers
is
that
glycosylation
can
change
the
immunogenicity
and
allergic
reactions
due
to
a
given
enzyme.
Background
on
glycosylation
would
refine
the
description
of
the
enzyme
if
it
included
information
on
the
sites
of
glycosylation
and
the
glycosyl
groups
involved.
Page
22
of
34
b.
coenzymes
(
prosthetic
groups),

In
response
to
both
parts
b
and
c
(
see
below),
it
should
be
noted
that
the
terminology
of
"
cofactors"
and
"
coenzymes"
is
confusing,
and
it
is
better
to
use
more
precise
terms.
There
are
"
cosubstrates"
which
participate
in
the
reaction
by
binding
to
the
enzyme
and
then
dissociate
from
it
as
a
product
and
can
be
recycled
by
other
enzymes.
Prosthetic
groups,
which
are
bound
to
the
enzyme,
remain
unchanged
at
the
end
of
the
reaction
cycle.
It
is
possible
that
modified
prosthetic
groups
may
be
incorporated,
which
would
change
the
specificity
of
the
enzyme.
This
might
be
deliberate,
or
an
accidental
consequence
of
the
host
organism
used
for
expression.

c.
cofactors
Cofactors
remain
bound
to
the
enzyme
and
are
required
for
catalysis.
In
heterologous
expression
systems,
there
is
always
the
possibility
that
cofactors
may
not
be
properly
or
completely
inserted,
yielding
an
enzyme
of
lower
activity.
Another
possibility
is
that
the
host
organism
attempts
to
insert
an
inappropriate
cofactor
or
metal
ion.
An
example
is
the
insertion
by
E.
coli
of
iron­
sulfur
clusters
containing
zinc
instead
of
iron
(
Archer
et
al.
1994)
.

d.
other
post­
translational
modifications
to
residues
of
the
amino
acid
chain?

In
a
few
instances
post­
translational
modifications
of
a
protein
may
be
required
for
its
activity.
For
example
the
conversion
of
a
seryl
to
a
pyruvoyl
residue
is
required
in
some
enzymes
that
use
the
pyruvoyl
residue
in
their
active
site.
Regarding
other
post­
translational
modifications
to
residues
of
the
polypeptide
chain,
there
will
be
variations,
depending
on
whether
the
protein
was
designed
in
the
original
organism
to
be
exported
or
secreted
from
the
cell.
Certain
cellular
systems
such
as
the
twin­
arginine
export
system
in
bacteria
can
either
transport
the
protein
through
the
membrane,
or
embed
the
protein
in
it,
depending
on
the
protein
sequence.
Such
amino
acid
modifications
can,
in
some
cases,
significantly
influence
the
activity
and/
or
stability
of
an
enzyme
preparation.

Source
The
SOURCE
of
an
enzyme
refers
to
(
1)
the
organism
from
which
the
gene
encoding
the
enzyme
was
derived,
i.
e.,
the
original
source
and
(
2)
the
organism
or
manufacturing
platform
(
e.
g.,
tissue
culture)
in
which
the
enzyme
is
produced,
i.
e.,
the
production
source.
In
the
questions
below,
please
consider
what
scientific
support
there
is
for
using
source
information
to
differentiate
among
enzymes
and
what
level
of
detail
would
be
scientifically
appropriate
for
this
purpose.

9.
What
information
about
an
enzyme's
structure
could
be
gained
by
knowing
a.
the
original
source
of
the
enzyme?

Knowing
the
original
source
of
the
enzyme
is
particularly
important
where
the
enzyme
preparation
is
not
a
purified
enzyme,
but
a
crude
extract,
such
as
pancreatic
juice,
which
contain
a
mixture
of
enzymes,
is
often
of
undefined
sequence.
Page
23
of
34
For
individual
enzymes,
the
relevance
of
original
source
depends
on
the
uniqueness
of
the
enzyme
to
that
organism.
A
greater
understanding
of
enzyme
structure
via
knowledge
of
the
original
source
of
the
enzyme
is
dependent
upon
knowledge
of
the
structure
of
the
enzyme
produced
by
a
given
source
organism
and
the
uniqueness
of
the
enzyme
to
that
organism.
In
most
cases,
an
enzyme
structure
will
be
largely
conserved
among
similar
organisms
in
spite
of
peripheral
mutations.
Thus,
source
will
have
limited
value
in
differentiating
enzymes
unless
the
structure
of
the
enzyme
from
each
source
is
known,
and
it
can
be
established
that
the
enzymes
from
different
sources
are
structurally
related
or
different.

b.
the
production
source
of
the
enzyme?

In
theory,
producing
an
enzyme
with
a
biological
system
other
than
the
original
organism
is
done
to
manufacture
the
product
as
an
exact
copy
of
the
original.
If
the
manufacturing
organism
changes
the
enzyme
in
some
way,
the
change
may
or
may
not
be
significant.
In
general,
comparing
enzymes
from
different
sources
requires
some
criteria
to
define
the
decision
plane
of
what
and
when
is
a
difference
significant
for
regulatory
purposes.

10.
If
original
source
information
were
used
as
an
identification
element
to
discriminate
among
enzymes,
what
level
of
taxonomic
specificity
(
e.
g.,
family,
genus,
species,
subspecies,
population,
biovar,
culture
line)
would
be
most
scientifically
appropriate
to
use
for
each
of
the
following
categories?
What
if
production
source
information
were
used?
(
Note:
EPA
recognizes
that
taxonomic
revisions
may
change
the
names
of
particular
organisms
and
can
utilize
mechanisms
for
normalizing
organism
nomenclature,
but
that
consideration
does
not
need
to
be
addressed
by
the
panel.)

a.
plants
b.
animals
c.
fungi
d.
bacteria
e.
other
micro­
organisms
If
original
source
information
were
used
as
an
identification
element
to
discriminate
among
enzymes,
using
the
lowest
taxonomic
level
available
would
be
the
most
desirable.
Differences
among
enzymes
from
different
parts
of
a
source
organism
may
occur
in
some
cases.
However,
such
differences
are
unlikely
to
be
a
general
phenomenon
of
a
specific
organ
or
tissue.
One
cannot
assume
that
all
or
even
a
preponderance
of
the
enzymes
from
one
tissue
or
organ
differ
from
that
of
another
tissue
or
organ,
although
they
may.
Furthermore,
two
enzymes
having
the
same
catalytic
action
may
well
exist
in
the
same
tissue
or
organ.
For
example
enzymes
may
differ
at
different
stages
of
development
(
juvenile
vs.
adult
beta­
galactosidase).
Hence,
this
issue
must
be
addressed
in
terms
of
the
formal
and
informal
nomenclature
structures.
Not
only
do
organism
names
change,
so
do
their
taxonomic
ranks
due
to
these
name
changes.
Nomenclature
codes
as
regulated
by
the
scientific
community
only
extend
down
to
the
subspecies
level.
Even
in
the
official
codes,
there
is
no
central
adjudication
of
the
proper
use
of
a
taxonomic
level
or
its
Page
24
of
34
application
to
a
specific
biological
entity
(
exception:
virology).
There
is
no
universally
accepted
structure
below
subspecies.
In
addition,
there
are
common
practices
that
vary
among
biological
groupings.
The
current
status
of
Codes
of
Nomenclature
and
some
of
their
attributes
is
contained
in
Table
1.

11.
How
could
source
be
described
if
taxonomic
names
were
inappropriate
because
either
the
original
or
production
source
were
artificial?
Examples
of
such
new
technologies
could
include
enzymes
produced/
developed
through
gene
splicing
or
ex
vivo
chemical
synthesis.

As
noted
previously
by
the
Panel,
the
original
source
of
the
enzyme
is
an
important
descriptor.
When
the
production
source
is
artificial,
i.
e.
in
vitro
translation,
the
system
used
to
produce
the
enzyme
will
be
a
defined
one
and
thus
can
be
identified.
To
date
the
synthesis
of
a
truly
artificial
enzyme
by
ex
vivo
chemical
synthesis
is
rare,
and
is
not
likely
to
be
used
commercially.
Enzymes
produced/
developed
through
gene
splicing
are
derived
from
defined
sources
and
usually
contain
the
backbone
from
a
specific
enzyme
that
is
then
modified.
Such
enzymes
may
be
composed
of
parts
of
more
than
a
single
original
source
enzyme,
but
these
can
be
defined.

Generally
the
structure
of
an
engineered
protein
will
be
related
to
a
naturally­
occurring
protein,
even
after
extensive
modification.
The
components
of
a
chimeric
enzyme
(
i.
e.
one
that
is
engineered
from
components
of
other
enzymes)
can
be
described
in
terms
of
its
various
sources,
and
the
modifications
made
to
it.
In
addition,
information
on
the
vector
to
which
the
protein
is
introduced
is
helpful.
In
some
cases
it
could
be
important
to
know
the
source,
location
and
population
of
the
enzyme.
However
as
noted
previously
by
the
Panel,
modifications
to
the
sequence
can
have
drastic
effects
on
the
activity
and
specificity
of
the
enzyme.

12.
What
information
about
an
enzyme's
structure
could
be
gained
by
additional
details
about
source
including:

a.
the
particular
tissue
or
organ
of
a
given
source
organism
from
which
they
were
derived
(
e.
g.,
swine
pancreatic
tissue
vs.
swine
salivary
glands)?

Differences
among
enzymes
from
different
parts
of
a
source
organism
may
occur
and
should
be
taken
into
account.
They
are
termed
iso­
enzymes
(
or
just
isozymes),
and
arise
from
a
variety
of
different
mechanisms,
including
duplicated
genes,
introns,
and
alternative
gene
splicing.
The
distribution
of
isoenzymes
is
a
feature
of
the
specific
organ
or
tissue.
Although
one
cannot
assume
that
all
or
even
a
preponderance
of
the
enzymes
from
one
tissue
or
organ
differ
from
that
of
another,
there
are
known
differences
among
enzymes
from
different
parts
of
a
source
organism.
Those
cases
need
to
be
defined.

b.
the
chemical,
geographic,
and/
or
environmental
conditions
from
which
source
organisms
were
isolated
(
e.
g.,
soil,
water,
feces,
etc.)?
Page
25
of
34
For
unknown
enzymes,
only
characterized
by
an
activity,
the
chemical,
geographic,
and/
or
environmental
conditions
may
be
a
useful
descriptor.
Some
structural
information
may
well
be
associated
with
place
of
isolation.
High
temperature
environments
may
select
for
thermostable
enzymes.
Hypersaline
environments
are
sources
of
halophilic
bacteria,
which
accommodate
their
environment
by
using
high
concentrations
of
ions
to
maintain
internal
pressures.
Their
enzymes
are
in
some
cases
highly
stable,
though
they
may
require
high
salt
concentrations
for
stability
outside
of
the
cell.

c.
manipulations
of
the
enzyme's
original
source
prior
to
gene
transfer
(
e.
g.,
through
rDNA
technology,
radiation
treatment,
altered
rearing
conditions,
etc.)?

It
is
unlikely
that
manipulation
of
the
enzyme's
original
source
would
produce
a
stable
transmissible
change.
Even
if
such
manipulation
had
an
affect
on
enzyme
structure,
the
specific
nature
of
the
change
would
be
unpredictable
in
the
general
sense
and
would
have
to
be
determined
empirically.

d.
manipulations
of
an
enzyme

'
s
production
source
prior
to
and/
or
following
gene
transfer?

The
conditions
noted
in
response
to
part
c
above
are
applicable
to
part
d.

e.
other
relevant
aspects
of
source
that
are
not
mentioned
(
please
specify
what
would
be
of
value).

While
the
conditions
of
growth
of
the
production
organism
will
affect
the
characteristics
of
the
product,
such
as
restriction
or
supplementation
of
nutrients,
trace
elements
or
vitamins,
there
are
no
other
specific
aspects
of
source
outside
those
mentioned
in
sections
12a
through
12b
that
would
add
additional
information
about
the
structure
of
an
enzyme.

Processing
The
PROCESSING
of
an
enzyme
refers
to
procedures
used
to
isolate
the
enzyme
from
the
production
organism
or
manufacturing
platform,
procedures
used
to
purify
the
enzyme,
and/
or
any
chemical
reactions
to
which
the
enzyme
is
subjected
to
produce
the
final
enzyme
product.
In
the
questions
below,
please
consider
what
scientific
support
there
is
for
using
certain
processing
information
to
differentiate
among
enzymes
and
identify
the
level
of
detail
that
would
be
scientifically
appropriate
for
this
purpose.

13.
What
information
about
an
enzyme's
structure
could
be
gained
by
knowing
which
of
certain
processing
techniques
were
used
in
its
production?

Processing
techniques
used
in
the
production
of
an
enzyme
will
generally
not
affect
its
structure.
The
methods
used
for
purifying
enzymes
depend
on
various
general
properties
of
the
protein.
As
an
example,
ion
exchange
chromatography
separates
on
the
basis
of
charge
and
gel
Page
26
of
34
filtration,
size
and
possibly
shape.
These
are
the
commonly
used
techniques
that
are
not
specifically
applied
to
a
particular
enzyme.
Affinity
chromatography,
which
is
now
the
preferred
approach,
can
be
directed
at
catalytic
site
organization
but
one
usually
needs
to
have
some
information
ahead
of
time
to
make
this
effective.
In
most
cases,
since
there
is
no
significant
purification
for
most
commercial
industrial
enzymes,
it
seems
unlikely
that
any
useful
information
about
the
enzyme's
structure
will
be
gained
from
a
consideration
of
its
processing.
Factors
in
production
that
have
a
bearing
on
the
structure
of
the
enzyme
could
include
treatments
that
affect
the
protein's
oligomeric
state,
such
as
cross­
linking,
and
immobilization
to
a
solid
support.
Knowledge
of
processing
techniques
could
be
useful
in
an
exclusionary
sense.
Thus,
if
the
process
does
not
yield
a
pure,
single
protein,
then
it
will
be
immediately
obvious
that
many
informational
details
delineated
by
the
Panel
will
be
unattainable.

14.
EPA
anticipates
that
certain
processing
techniques
may
be
so
routine
and/
or
chemically
inconsequential
that
their
reporting
would
be
unnecessary,
while
other
processing
techniques
would
have
significant
effects
on
the
chemical
structure
and/
or
properties
of
an
enzyme.
The
Agency
is
trying
to
assess
how
practical
it
would
be
to
create
a
list
of
processing
techniques
that
need
not
be
included
as
part
of
enzyme
identity.

a.
What
processing
techniques
are
used
in
the
isolation
and
purification
of
enzymes?

There
are
a
host
of
methods
for
obtaining
a
purified
protein
(
enzyme)
but
few
if
any
of
these
seemed
to
be
used
in
making
TCSA
inventory
enzymes.
These
utilize
various
properties
of
proteins
such
as
charge
and
size.
There
are
also
various
forms
of
affinity
interactions
(
substrate
mimics,
antibodies,
tags
etc).
These
are
indeed
relatively
routine
in
their
application
and
a
list
could
be
created.
However,
new
affinity
purification
methods
are
changing
regularly
and
since
they
are
not
thought
to
be
generally
employed
for
the
commercial
enzyme
preparations
to
be
included
in
the
TCSA
inventory,
there
seems
to
be
little
advantage
in
doing
so.

b.
Which
processing
techniques
could
change
the
chemical
structure
of
the
enzyme?
Which
could
change
chemical
properties
that
would
indicate
an
underlying
structural
change?

By
and
large,
most
methods
for
isolation
of
enzymes
are
designed
to
NOT
alter
proteins
and
ones
that
do
so
are
usually
not
used.
However,
there
are
some
well­
established
methods
that
are
known
to
cause
changes.
Exposure
to
heat,
pH
extremes,
and
oxygen
that
can
routinely
change
structural
features
(
such
as
the
oxidation
of
methionine
and
the
deamidation
of
side
chain
amides).
The
recent
advances
in
mass
spectrometry
analysis
(
MALDI,
ESI)
have
revealed
that
this
occurs
to
a
much
larger
extent
than
was
previously
realized,
but
it
is
unlikely
this
analysis
would
be
applied
to
manufactured
enzymes.
Many
of
these
changes
are
also
time­
dependent
and
occur
simply
as
a
function
of
time.
Thus,
they
cannot
be
strictly
tied
to
the
processing
step,
although
there
are
certainly
conditions
(
such
as
noted
above)
that
will
accelerate
these
changes.
Importantly,
in
considering
any
structural
alterations
from
the
`
native'
state,
one
should
only
Page
27
of
34
consider
covalent
modifications
and
not
transient
changes
such
as
proton
association/
disassociation.

A
commonly­
used
step
particularly
for
crude
enzyme
preparations
is
autolysis;
the
cells
containing
the
enzyme
are
allowed
to
digest
for
some
time.
This
process
depends
on
proteases
and
lipases;
sometimes
these
are
derived
from
the
cell,
sometimes
added
to
the
preparation.
They
help
to
release
insoluble
proteins
such
as
those
that
were
anchored
to
the
membrane,
but
the
product
is
often
cleaved
in
several
places,
and
may
have
sections
removed.
Another
method
that
is
used
to
isolate
membrane­
bound
proteins
is
the
use
of
detergents.

It
is
also
important
to
consider,
particularly
in
mixtures
of
enzymes
that
characterize
commercial
preparations,
that
purification
steps
can
separate
cofactors
or
other
modulaters,
which
would
in
turn
affect
activity.

c.
Describe
the
chemical
or
structural
changes
expected
to
occur
from
the
use
of
the
processing
techniques
identified
in
14(
b).

The
structure
of
the
protein
is
determined
by
means
such
as
crystallography,
on
purified
preparations.
Although
processing
may
lead
to
modifications
of
amino
acids,
the
central
structure
should
remain
the
same.
There
is
a
difference
between
the
applications
to
purified
enzymes
as
opposed
to
unpurified
enzymes.
The
types
of
covalent
modifications
that
might
be
encountered
were
described
above.

d.
Which
processing
techniques
would
not
be
expected
to
cause
any
structural
changes
to
the
enzyme?
Which
would
not
be
expected
to
cause
any
chemical
property
changes?

Most
methods
involving
chromatography
or
electrophoresis
should
not
affect
enzyme
structure.

15.
EPA
is
trying
to
anticipate
whether
inclusion
of
processing
in
enzyme
identity
will
increase
in
importance
as
a
result
of
future
advances
in
enzyme
production.

a.
What
new
processing
techniques
are
being
developed?

Since
industrial
enzymes
are
relatively
inexpensive
products,
new
techniques
are
slow
to
be
implemented.
Established
processing
techniques
that
have
been
used
for
decades
have
been
refined
to
the
point
that
their
efficiencies
are
very
high.
There
is
little
room
in
the
production
cost
of
an
industrial
enzyme
to
absorb
the
cost
of
implementing
a
new
processing
technique
in
order
to
improve
the
profit
margin
in
the
final
product.
New
processing
techniques
that
are
being
developed
are
aimed
at
creating
new
products
that
will
support
a
premium
price
in
the
market
place.
One
example
of
these
new
technologies
would
be
to
change
the
"
format"
of
an
enzyme
from
a
water­
soluble
enzyme
to
an
immobilized
solid
state
enzyme.
Another
example
would
be
techniques
to
selectively
remove
an
undesired
enzyme
activity
that
prevents
the
product
from
Page
28
of
34
being
used
in
a
particular
industrial
application.
One
method
of
purification
of
enzymes
is
via
the
use
of
affinity
chromatography.
This
process
is
expensive
and
often
requires
modification
of
the
gene
coding
for
the
enzyme.
When
affinity
tags
are
added
to
an
enzyme
structure,
they
may
have
to
be
removed
in
another
processing
step
in
which
case,
because
of
the
cost
involved,
only
those
enzyme
products
that
can
support
premium
pricing
will
be
candidates
for
purity
enhancement
using
affinity
techniques.

b.
How
might
these
techniques
change
an
enzyme's
chemical
structure
or
properties?

Most
of
the
processing
techniques
do
not
significantly
change
an
enzyme's
chemical
structure
or
its
properties.
There
are
immobilization
techniques
that
change
the
solubility
of
an
enzyme
but
not
its
catalytic
function.
Except
for
solubility,
immobilization
techniques
do
not
change
the
chemical
structure
or
properties
of
an
enzyme.
Affinity
purification
techniques
that
do
not
require
enzyme
modification
will
not
alter
the
enzyme's
chemical
structure
or
function.
At
this
time,
modification
in
a
gene
coding
for
an
enzyme
for
the
purpose
of
purification
is
far
too
expensive
for
use
in
the
production
of
industrial
enzymes,
though
it
may
be
used
in
the
future.

c.
How
frequently
are
new
processing
techniques
for
enzymes
adopted?

The
adaptation
of
new
processing
techniques
cost
a
lot
of
money
and
can
only
be
cost
effective
for
new
products
that
can
support
a
premium
price.
The
majority
of
new
industrial
enzymes
are
processed
via
the
well­
established
techniques
that
have
been
in
place
for
decades.

Other/
General
Questions:

16.
Aside
from
function,
sequence,
source,
and
processing,
are
any
other
data
elements
crucial
for
enzyme
identification?

The
Panel
considered
assay
conditions
as
a
possible
data
element
for
enzyme
identification,
but
was
unable
to
determine
whether
assay
conditions
itself
should
be
considered
as
a
part
of
the
function
data
element.
Other
conditions
that
could
be
considered
have
been
referred
to
by
the
Panel
in
response
to
previous
questions.

The
minimum
structural
requirement
for
a
substrate
of
an
enzyme
reflects
the
binding
site
requirements.
This
information
would
allow
one
to
predict
other
materials
that
would
serve
as
a
substrate
besides
the
one
used
to
describe
the
function
in
the
nomenclature.

The
mechanism
of
catalysis
used
by
an
enzyme
adds
additional
information
about
an
enzyme.
There
are
several
enzymes
that
act
on
the
same
substrate
and
produce
the
same
products
but
do
so
using
different
mechanisms.

A
number
of
enzymes
contain
allosteric
sites
that
bind
modulators
to
enhance
or
control
activity.
The
presence
or
absence
of
these
sites
could
be
used
to
identify
different
enzymes
Page
29
of
34
catalyzing
the
same
reaction.
Salts
(
both
cations
and
anions)
serve
as
modulators
for
a
number
of
enzymes.
There
are
many
examples
of
enzymes
having
identical
catalytic
function,
but
differ
as
to
the
ion
that
modulates
the
activity.

Many
hydrolytic
enzymes
(
proteases,
lipases,
amylases,
etc)
display
different
levels
of
activity
based
on
the
substrate
used
in
their
assay.
Therefore
when
reporting
enzyme
activity
level,
it
is
important
to
identify
the
substrate
used
in
the
assay
as
well
as
the
enzyme
assay
conditions.

Industrial
enzymes
can
be
very
impure
products
and
often
contain
many
enzyme
activities
along
with
the
marketed
enzyme.
An
identification
element
could
be
constructed
around
the
degree
of
purification
from
the
mixture
produced
during
the
fermentation
or
isolation
process.
Several
industrial
enzymes
are
often
produced
from
a
common
fermentation
or
isolation.
They
often
differ
by
the
removal
of
selected
enzymes
that
prevent
their
use
in
an
industrial
application.

17.
Are
there
any
special
considerations
that
should
be
taken
into
account
when
identifying
enzymes
with
multiple,
non­
identical
subunits?
For
example,
a.
when
only
one
subunit
is
modified?

It
is
important
to
specify
the
nature
of
the
whole
enzyme
complex
as
well
as
the
individual
subunits.
When
a
subunit
of
a
multienzyme
complex
is
removed,
it
may
show
much
lower
activity.
In
addition,
the
reaction
specificity
may
be
altered
or
relaxed.
In
addition,
if
a
subunit
fails
to
be
inserted,
the
remaining
protein
often
fails
to
assemble
correctly.
This
can
lead
to
loss
or
modification
of
structure.
Activity
may
be
decreased
and
specificity
altered.
Therefore,
it
is
important
to
specify
the
nature
of
the
whole
complex
as
well
as
the
individual
subunits.

b.
when
a
modified
enzyme
is
a
component
of
an
enzyme
complex?

The
answer
for
part
17a
above
applies,
whether
the
catalytic
entity
comprises
one
subunit
or
has
multiple
subunits.
The
way
in
which
enzymes
containing
non­
identical
subunits
are
currently
described
in
the
EC
list
is
that
there
is
an
EC
number
for
the
whole
enzyme
and
EC
numbers
for
the
individual
catalytic
entities.
This
is
not
always
done
however,
because
of
the
problem
of
defining
the
primary
EC
class
when
different
classes
of
reaction
are
involved.
The
catalytic
entity
in
some
complexes
is
not
even
an
individual
subunit;
it
may
be
a
domain
of
an
extended
polypeptide.
For
example
many
of
the
polyketide
synthases
have
a
single
polypeptide
chain
comprising
multiple
catalytic
sites
that
work
sequentially
to
assemble
a
molecule
such
as
an
antibiotic.
These
catalytic
domains
behave
essentially
like
subunits,
but
since
they
are
a
single
polypeptide
they
are
more
stable.

c.
when
a
multi­
functional,
multi­
component
enzyme
performs
a
sequence
of
reactions?

In
a
case
such
as
the
polyketide
synthases
noted
above,
the
catalytic
sites
can
be
identified
separately.
Page
30
of
34
d.
when
an
enzyme
has
another
non­
catalytic
function,
e.
g.,
a
binding
site?

If
relevant,
this
could
be
described
as
another
function.

e.
under
any
other
circumstances?

The
ratio
of
subunits
may
be
relevant,
such
as
when
a
complex
contains
too
many
or
(
more
commonly)
too
few
of
a
subunit.
If
a
subunit
fails
to
be
inserted,
the
consequent
failure
to
assemble
correctly
can
lead
to
loss
or
modification
of
structure;
activity
may
be
decreased
and
specificity
altered.

As
presented
previously,
enzymes
have
been
shown
to
contain
more
than
one
catalytic
activity.
It
is
often
the
case
in
multifunctional
enzymes
that
the
various
functions
are
carried
out
by
different
subunits
of
the
molecule.
Taking
into
account
the
differences
in
subunit
structure
in
enzymes
made
up
of
non­
identical
subunits
would
add
little
if
anything
to
their
identity.
It
is
sufficient
to
only
identify
the
catalytic
activities
present
within
the
enzyme.

Any
of
the
four
conditions
above
(
a­
d)
potentially
can
differentiate
among
multiple
subunit
enzymes
and
indeed
between
multiple
subunit
enzymes
and
single
unit
enzymes
having
similar
activity
properties.
Thus,
the
knowledge
of
the
existence
of
such
conditions
enhances
the
ability
to
detect
"
new"
enzyme
products.

18.
Although
EPA
believes
that
all
four
identification
elements
are
critical
for
enzyme
identification
for
TSCA
purposes,
the
Agency
is
trying
to
judge
their
relative
importance.

a.
Do
any
data
elements
warrant
greater
emphasis
than
others
because
differences
in
those
data
element(
s)
reflect
more
significant
differences
in
an
enzyme's
physical
and/
or
chemical
properties
than
the
others
do?

The
identification
of
enzyme
function
is
of
prime
importance,
and
sets
the
context
for
all
else.
However,
function,
by
itself
may
be
too
broad
for
definitive
consideration
if
novelty
is
the
objective.
The
existing
registries
have
to
include
other
information.
Simple
examples
include
differentiation
of
isoenzymes,
names
of
source
organisms
in
some
cases,
etc.
For
reasons
given
previously,
in
some
cases
the
same
enzyme
activity
commands
different
identifiers
in
the
lists.
The
sequence,
if
available,
may
be
helpful
if
compared
sequences
are
the
"
same"
(
as
determined
by
similarity
overall
and/
or
of
reactive
site),
then
novelty
is
excluded.
The
situation
is
more
problematic
as
similarity
decreases
but
the
activity
remains
the
same.
Source
is
less
helpful
as
it
is
not
consistent
in
its
applicability.
Still,
source
should
not
be
ignored
as
it
can
give
a
general
context
for
an
enzyme,
particularly
for
crude
enzyme
containing
extracts.
However,
the
practicality
of
obtaining
reliable
data
from
the
applicant
may
inhibit
utility.
For
example,
a
statement
that
the
enzyme
was
obtained
from
mouse
spleen
raises
the
question
as
the
common
name
"
mouse"
is
defined
variously
as
covering
one
genus
of
rodent
or
two
depending
on
the
dictionary
consulted.
Page
31
of
34
Processing
can
be
of
importance.
The
first
processing
consideration
is
the
purity
of
the
preparation.
A
pure
enzyme
protein
can
be
evaluated
in
ways
that
are
denied
for
a
mixture
of
substances.

b.
If
data
for
sequence,
source,
and
processing
were
the
same
for
two
enzymes
(
at
the
level
of
detail
you
have
determined
to
be
appropriate
in
the
questions
above),
what
additional
information
about
chemical
structure
and/
or
properties
would
be
provided
by
distinguishing
the
enzymes
based
on
function?

The
implication
of
the
sequence
being
the
same
for
two
enzymes
is
based
on
the
assumption
that
the
protein
preparations
being
pure
to
enable
determination
of
the
two
enzymes'
sequences.
If
the
sequence
were
truly
identical
and
without
modification,
then
one
would
expect
everything
else
to
be
the
same.
There
may
be
cases
in
which
multiple
conformations
exist,
but
knowing
the
source
and
processing
for
identical
sequences
and
the
identical
activities
of
the
proteins
is
unlikely
to
define
one
or
the
other
enzyme
as
new.
The
question
becomes
moot
in
the
common
situation
where
enzyme
preparations
are
impure,
rendering
sequence
determination
impractical.
In
such
situations,
as
much
information
as
practical
from
all
the
categories
proposed
by
the
Agency
will
aid
in
determination
of
relative
sameness
or
difference.
Without
a
determination
of
sequence
identity,
only
relative
similarity
can
be
determined.
Thus,
the
more
kinds
of
information
gathered,
the
greater
the
possibility
to
determine
whether
two
enzymes
with
the
same
activity
are
the
same
or
different.

c.
If
data
for
function,
sequence,
and
processing
were
the
same
for
two
enzymes
(
at
the
level
of
detail
you
have
determined
to
be
appropriate
in
the
questions
above),
what
additional
information
about
chemical
structure
and/
or
properties
would
be
provided
by
distinguishing
the
enzymes
based
on
(
1)
original
source
and
(
2)
production
source?

If
the
sequence
were
truly
identical
and
without
modification,
then
one
would
not
gain
much
new
information
based
on
the
original
source
or
the
production
source.
If
on
the
other
hand
the
sequence
was
not
totally
defined,
information
on
the
production
source
could
be
useful.
Limited
useful
information
would
be
derived
by
deducing
the
characteristics
of
the
protein
from
(
1)
original
source
and
(
2)
production
source.
The
same
enzyme
may
differ
in
source
and
still
be
the
same
enzyme,
or
the
same
source
may
yield
two
enzymes
with
the
same
activity.

d.
If
data
for
function,
sequence,
and
source
were
the
same
for
two
enzymes
(
at
the
level
of
detail
you
have
determined
to
be
appropriate
in
the
questions
above),
what
additional
information
about
chemical
structure
and/
or
properties
would
be
provided
by
distinguishing
the
enzymes
based
on
processing?

The
conditions
noted
in
response
to
parts
18.
b
and
18.
c
above
are
applicable
to
part
18.
d.
Page
32
of
34
REFERENCES
Archer
VEV,
Breton
J,
Sanchezgarcia
I,
Osada
H,
Forster
A,
Thomson
AJ
and
Rabbitts
TH.
1994.
Cysteine­
Rich
Lim
Domains
Of
Lim­
Homeodomain
And
Lim­
Only
Proteins
Contain
Zinc
But
Not
Iron.
Proceedings
of
the
National
Academy
of
Sciences
of
the
United
States
of
America.
91:
316­
320.

Carter
P
and
Ells
JA.
1988.
Dissecting
The
Catalytic
Triad
Of
A
Serine
Protease.
Nature,
332:
564­
568.

Hicks
MG
and
Kettner
C.
2003.
Experimental
Standard
Conditions
of
Enzyme
Characterizations.
pp.
203­
213,
Beilstein
Institute,
Frankfurt,
ISBN
3­
8325­
0727­
2
http://
www.
beilsteininstitut
de/
escec2003/
proceedings/

Kane
JF,
Violand
BN,
Curran
DF,
Staten
NR,
Duffin
KL,
and
Bogosian
G.
1992.
Novel
In­
Frame
Two
Codon
Translational
Hop
During
Synthesis
Of
Bovine
Placental
Lactogen
In
A
Recombinant
Strain
Of
Escherichia
coli.
Nucleic
Acids
Res.
Dec
25;
20(
24):
6707­
12.

Rosenberger,
RF
1994a.
Translation
Errors
During
Recombinant
Protein
Synthesis
In
Genetic
Stability
And
Recombinant
Protein
Consistency.
Brown
F.
and
Lubiniecki
As
(
eds)
Dev.
Biol.
Stand.
Basel,
Karger,
1994,
vol
83,
pp
21­
26.

Rosenberger
RF.
1994b.
Translational
Errors
During
Recombinant
Protein
Synthesis.
Dev.
Biol.
Stand.
83,
21­
26.

Scorer
CA,
Carrier
MJ,
and
RF
Rosenberger.
1991.
Nucleic
Acids
Res.
19:
3511
 
3516.

Weickert
MJ
and
Apostol
I.
1998.
High­
Fidelity
Translation
of
Recombinant
Human
Hemoglobin
in
Escherichia
coli
Appl
Environ
Microbiol.
64,
1589
 
1593.
Page
33
of
34
Table
1.
Status
of
Nomenclature
Codes
Org.
acronyms
Orgs.
Overseeing/
man.
Code
Code
Contact
information
Nname
for
mat.
Physical
nature
of
deposit
Requirement
ICSU
International
Council
for
Science
www.
icsu.
org
IUBS
International
Union
of
Biological
Sciences
www.
iubs.
org
ICZN
International
Commission
on
Zoological
Nomenclature
International
Code
of
Zoological
Nomenclature
www.
iczn.
org
Type
specimen
Nature
varies
with
organism
­
deposition
in
a
recognized
institution
not
required
by
the
Code,
most
authors
do,
making
specimens
available.
Now
must
be
designated
and
clearly
identified
for
any
species
described
after
2000
IAPT
International
Association
for
Plant
Taxonomy
International
Code
of
Botanical
Nomenclature
www.
botanik.
univie.
ac.
at/

iapt/
Type
single
plant,
parts
of
one
or
several
plants,
or
of
multiple
small
plants
­

usually
mounted
on
a
single
herbarium
sheet
or
in
an
equivalent
preparation,

such
as
a
box,
packet,
jar
or
microscope
slide
either
a
single
specimen
conserved
in
one
herbarium
or
other
collection
or
institution,

or
an
illustration
for
any
species
described
after
1990
ICNCP
International
Commission
for
the
Nomenclature
of
Cultivated
Plants
International
Code
of
Nomenclature
for
Cultivated
Plants
www.
ishs.
org/
sci/
icracpco
.
htm
IUMS
International
Union
of
Microbiological
Societies
www.
iums.
org
ICSP
International
Committee
on
Systematics
of
Prokaryotes
International
Code
of
Nomenclature
of
Bacteria
www.
the­
icsp.
org
Type
culture
living
culture
Publication
in
the
International
Journal
of
Systematic
and
Evolutionary
Microbiology
and
deposited
in
two
recognized
culture
collections
in
two
different
countries
ICTV
International
Committee
on
the
Taxonomy
of
Viruses
International
Code
of
Virus
Classification
and
Nomenclature
www.
ncbi.
nlm.
nih.
gov/
IC
TV/
rules.
html
Physical
type
not
used
Nomenclatural
type
description
serves
this
function
Accepted
description
maintained
by
ICTV
The
Internet
contact
information
links
to
the
most
current
versions
of
the
Codes,
decisions
on
nomenclature,
information
on
the
history
and
administration
of
the
Codes
and
other
relevant
information.
The
official
printed
versions
of
the
Codes
are
necessarily
out
of
date
as
of
the
publication
in
that
they
obviously
cannot
contain
changes
post
printing.
Page
34
of
34
