Download The size, operation, and technical capabilities of protein and nucleic

Special
Feature
The
size,
operation,
protein
and technical
and nucleic
Hughes
Medicine,
New
53705,
Medical
Haven,
USA;
tThe
Institute,
Department
Connecticut
Rockefeller
06510,
of Molecular
USA;
University,
New
tUniversity
York, New
Molecular Biology Institute,
UCLA School of Medicine,
Great Valley, Pennsylvania 19355, USA; and #Departnt
Ca4fornia
95616,
of 40 protein
and
nucleic
acid
chemistry
facil-
ities has provided
data about the capabilities
of core
facilities and the cost of the services they provide. Approximately
43% of the $158,000 average annual operating
budget
for
from
service
charges.
a typical
university
After
facility
correcting
out
an acid
a protein.
cleaved
tide
A 25-residue
for
costs
$2078,
$874.
synthesized
of
an
for
average
analyses,
syntheses,
A
The
facility
and
of
protein
are
in a protein
1.2 nmol
to
runs,
typically
digest and then
residues
in one
65
can
amino
three
requires
required
the
an
150 pmol
to first
acid
Depend-
200 pmol
accurate
first
be
peptide
syntheses.
obtain
and
pep-
per month
used, from 85 to nearly
to
on
same
output
sequencing
To sequence
of protein
the
work
to
analysis
oligonucleotide
total
required
it was
$65
can be synthesized
16 oligonucleotide
composition.
acid
corresponds
acid
various
facility
sequencing
25-residue
$258.
the
facilities,
amino
peptide
the approach
on
and
whereas
15 amino
ing
acid
hydrolysis
is derived
for
degrees of subsidization
of the different
found that it costs a typical
university
carry
amino
15 amino
acids
compared
with
carry
out
a tryptic
isolate and sequence
the first 15
of the resulting
tryptic
peptides.
-WILLIAMS,
K. R.; NIECE,
R. L.; ATHERTON,
D.;
FOWLER, A. V.; KUTNY, R.; SMITH, A. J. The size, operation, and technical
capabilities
of protein
and nucleic
acid core facilities.
FASEBJ
2: 3124-3130; 1988.
Key
Words:
quencing
.
core facility
peptide
synthesis
.
biopolymer
facility
protein
se-
amino acid analysis
RECENT
ADVANCES IN THE BIGFECHNOWGY
of protein
and nucleic acid sequencing
and synthesis have fostered
the development
of core facilities whose primary
function is to make these techniques
available
to a wide
3124
DONNA
Biophysics
ATHERTON,1
and Biochemistry,
of Wisconsin
York 10021,
AUDREE
Biotechnology
USA;
V. FOWLER,4
Yale University
Center, Madison,
SDepartment
of Biological
School of
Wisconsin
Chemistry
and
Los Angeles, Ca4fornia 90024,
USA; “Eastman Pharmaceuticals,
of Biological Chemistry,
University of Ca4fornia, Davis,
USA
ABSTRACT
A survey
of
acid core facilities’
KENNETH
R. WILLIAMS*z
RONALD
L. NIECE,t
RUSTY KUTNY,II AND ALAN J. SMITH”
*Howard
capabilities
range of investigators,
not only specialists
in these areas
but also cellular and molecular
biologists
in general (1).
An investigator
with little or no training
in protein
sequencing
can now use a relatively
simple technique,
sodium
dodecyl
sulfate (SDS) gel electrophoresis
followed by electroblotting
onto polyvinylidene
difluoride
membranes
(2), to prepare
a protein sample that can be
submitted
to a core facility for NH2-terminal
sequence
analysis.
Based on the resulting
sequence,
the facility
can make a corresponding
synthetic
peptide
as well as
an oligonucleotide.
The synthetic
peptide may be used
for structure/function
studies and as an antigen to elicit
antibodies
that can be used for a variety
of purposes,
including
the screening
of cDNA libraries.
The oligonucleotide
probe can be used to screen cDNA libraries
or to confirm
clones isolated
with antibodies.
Because much of the needed equipment
is expensive
and requires
considerable
expertise
to operate
continuously at peak efficiency, core facilities staffed by professional personnel
represent
the most economical
and
effective means of bringing
these technologies
to bear
on problems
related to biochemical
research.
To effectively utilize a core facility, prospective
users must follow appropriate
sample
preparation
procedures
and
have realistic
expectations
concerning
the capabilities
of these facilities.
These capabilities
relate both to the
quantity
of protein
that is needed
as well as to the cost
and amount
of time required
to complete
the requested
analyses.
To meet this need, we have compiled
a survey
of 40 protein
and nucleic acid core facilities.
Although
most of the facilities that participated
in the survey are
located
in universities,
responses
were received
from
two government
and five industrial
core facilities.
The
‘A preliminary
account of this study was presented at the symposium Core Research Resource Facilities: Practical Aspects For Users sponsored by the Education Affairs Committee
of the American Society
for Biochemistry
and Molecular Biology at the 72nd Annual Meeting of the Federation of American Societies for Experimental
Biology, Las Vegas, Nevada, May 2, 1988.
2To whom correspondence
should be addressed.
0892-6638/88/0002-3
124/$01 .50.
#{174}
FASEB
TABLE
1. Personnel
and equipment
in core facilities
Range
Number
of personnel
with
advanced
Meana
Median5
Mode’
degrees”
B.A./B.S.
0-4
1.5
(±
MS.
0-2
0.48
(± 0.68)
Ph.D.
0-3
0.75
(±
0.87)
0-3
1.1
(±
0.77)
1.0
1
0-10
0-3
0-2
0-2
3.2
1.5
0.73
0.48
(±
(±
(±
(±
1.9)
0.59)
0.63)
0.59)
3.0
1.0
1.0
0.0
2
1
1
0
Number
1.1)
1.3
2
0
0
0.55
1
of instruments
Amino acid analyzers
HPLC’
Protein sequencers
Peptide synthesizers
DNA synthesizers
aArithmetic
average
(±
so).
6The
median
is the
middle
number
in a series
an odd
containing
number
of items
and
the
number
midway
between
the two middle numbers
in an even number
of items.
In the first example,
there were as many
facilities
with more as there were those that had
less than 1.3 personnel
with B.A./B.S.
degrees.
‘The mode is the value most frequently
given among
the facilities surveyed.
dValues
are number
of personnel
x effort.
‘Values
include
on-line
and off-line PTH amino
acid HPLC systems but not HPLC
systems
dedicated
solely to the analysis
of free amino
acids or their phenylthiocarbamyl
derivatives;
these are included
under
amino
acid analyzers.
following
report
summarizes
size, financial
support,
and
facilities.
EXPERIMENTAL
the data
capabilities
concerning
of these
the
core
PROCEDURES
On May 21, 1987, a three-page
survey form was sent to
146 protein
and/or nucleic acid chemists
who had indicated an interest
in being included
on a mailing
list of
core facilities that was compiled
by one of the authors
(R. L. N.) in conjunction
with a Research
Resource
Satellite Meeting
(3). Forty surveys were returned
within
a 2-month
period from directors
of protein
and nucleic
acid chemistry
facilities.
Preliminary
results from these
surveys were presented
at the Satellite
Meeting
on Research Resource
Facilities
held on August 9, 1987, immediately
before the First Symposium
of the Protein
Society
in San Diego,
California.
Except
when specified otherwise,
the results
presented
in this report
refer to the entire group of 40 facilities,
33 of which
were located in universities.
Because a few respondents
did not answer all questions
on the survey, there is some
variation
(as noted throughout
the text) in the sample
size.3
RESULTS
Description
of core
facilities
Based on the data shown in Table 1, an average
core
facility can be described
as follows. It is staffed by the
equivalent
of (about) three full-time
personnel.
Usually
there are one or two full-time
technical
specialists
and,
in most instances,
there is at least a 50% commitment
on the part of one individual
with a Ph.D. Of the facilities surveyed
(34 of 40 responses),
85% are directed
by
someone
who has earned a Ph.D. compared
with three
facilities
that are directed
by an individual
with either
an M.S. or a B.A./B.S.
degree. As evidenced
by the fact
that 24 of the 38 directors
that responded
to this question are principal
investigators
with National
Institutes
SURVEY
OF PROTEIN
AND
NUCLEIC
ACID
CORE
FACILITIES
of Health,
National
Science Foundation,
or private research grants,
at least 60% of the directors
carry on
their own individual
research
programs
as well as oversee a core facility. On the average,
each staff member
of a core facility
is responsible
for operating
two or
three instruments,
which may include
an amino
acid
analyzer,
a protein/peptide
sequencer,
a peptide
synthesizer,
or the three high-performance
liquid chromatography
(HPLC)
systems
typically
found
in a core
facility (Table 1). Fewer than half of the facilities
surveyed contain
an oligonucleotide
synthesizer
(Table 1),
and only one facility was found that had an automated
DNA sequencer.
As for the manufacturers
of this equipment,
nearly
80% of the protein
sequencers
and oligonucleotide
synthesizers used in the facilities surveyed were made by Applied Biosystems
(Foster City, Calif.) (Fig. 1). Likewise,
this manufacturer
accounted
for 60% of the solid-phase
peptide
synthesizers
contained
within
this group
of
facilities
(Fig. 1). In contrast,
the leading
supplier
of
HPLC
systems
and
amino
acid
analyzers
was
Waters
(Milford,
Mass.),
with a 38% market
share in both
categories.
Although
most of the Waters HPLC
systems in this
group of facilities (excluding
those dedicated
to amino
acid analysis)
are being used for peptide
or protein
separations,
the 26% Applied Biosystems
HPLC market
share results almost entirely
from the Model
120 online phenylthiohydantoin
(PTH)
amino acid analyzer.
Only 2 of the 33 Applied
Biosystems
HPLC
units in
these facilities
were the Model
130 peptide/protein
HPLC
systems;
all remaining
31 units were dedicated
to on-line analysis of PTH amino acids. In addition
to
Waters and Applied
Biosystems,
these 40 core facilities
housed HPLC
systems manufactured
by 12 other companies (Fig. 1).
3We request
facilities
report
may
directors
based to
be included
is
of protein
who did not receive
forward
in any
a copy
their names
subsequent
and/or
of the
to K.
nucleic
survey
Williams,
acid
on
chemistry
which
so that
this
they
surveys.
3125
ties responding,
24 of a total of 31, are supported
in part
by user fees based
on a preestablished
schedule
of
charges
for the various
kinds of analyses
offered.
Of the seven university
facilities that did not charge
for performing
analyses,
three were supported
by the
Howard
Hughes
Medical
Institute,
two by U.S. Public
Health
Service
program
projects
and individual
research grants,
and in two instances
it was not possible
to determine
the basis of financial
support from the survey. At the other end of the spectrum,
there were only
six university
facilities that were entirely
supported
by
user fees, and in every case these self-supporting
facilities had an annual
operating
budget
of at least
$225,000.
In fact, the average annual operating
budget
for these six self-supporting
facilities
was $293,000,
Manufacturer
Figure
1.
Major
manufacturers
of
equipment
in
protein
and
nucleic acid chemistry facilities. The total number of instruments
in the 40 facilities surveyed in each of these categories was 59 protein/peptide
sequencers,
42 amino acid analyzers,
29 peptide synthesizers,
19 oligonucleotide
synthesizers,
and 125 HPLC systems
(including
on-line and off-line instruments
dedicated
of phenylthiohydantoin
amino
acids).
Abbreviations:
plied
Biosystems;
Beck.,
Beckman
Instruments;
Packard
Corporation
(Palo Alto, Calif.).
to the analysis
A. B., ApHP,
Hewlett-
Figure 2 shows that no consensus
yet exists regarding
the best approach
for obtaining
accurate
amino
acid
analyses.
As shown, 42% of the amino acid analyzers
in these facilities rely on ion exchange
separation
of the
free amino acids, followed by postcolumn
derivatization
with ninhydrin,
o-phthalaldehyde,
or fluorescamine.
In comparison,
54% of the amino acid analyzers
used
by the core facilities
rely on precolumn
derivatization
with phenylisothiocyanate,
followed
by reverse-phase
HPLC
separation
of the resulting
phenyithiocarbamyl
(PTC) amino acids. Based on all 40 facilities surveyed,
Waters is the manufacturer
most often chosen when a
PTC amino acid analyzer
is purchased.
Of the 23 PTC
amino acid analyzers
within this group of facilities,
14
are based on Waters HPLC
systems.
All five facilities
surveyed
that have multiple
amino
acid analyzers
use more than one method,
which underscores
the fact that there are advantages
and disadvantages
inherent
in both ion exchange
and reversephase approaches.
Each of these five facilities has both
an anion exchange
and reverse-phase
instrument.
50
C/)
w 40
N
>-
-J
z
30
-J
I-
0
20
I-
l0
0
z
<
0
-.-
0
I0.
-
0
0
Ll
LL
Financial
support
of core
W
facilities
2. Amino acid analysis techniques
used in protein chemistry
facilities. The total number of amino acid analyzers
in use in the
40 facilities surveyed was 42. Abbreviations correspond to the following approaches to amino acid analysis: PTC/RP,
phenylisothioFigure
Another
measure of the size of core facilities is the annual
operating
budget.
As shown in Table 2, the average
operating
budget (which includes
all salaries,
supplies,
equipment
repair as well as all other costs except those
for capital equipment
acquisition)
is $158,000 for the 30
university
facilities that responded
to this question on the
survey. This figure compares
with $275,000
for the five
company
facilities and $153,000 for the two government
core facilities that responded.
Most of the university
fadii-
3126
Vol.
2
Dec. 1988
cyanate
derivatization
of the free amino
acids followed by reversephase
HPLC
separation;
IE/NIN,
ion exchange
separation
with
postcolumn
ninhydrin
detection;
IE/OPA,
ion exchange
separation
followed by postcolumn
detection
with o-phthalaldehyde;
FMOC/RP,
precolumn
derivatization
with 9-fluorenylmethyl
chloroformate
followed by reverse-phase
HPLC;
IE/FLUOR.,
ion exchange
separation followed
by postcolumn
detection
with fluorescamine.
The FASEBJournal
WILLIAMS
ET AL.
TABLE
2. Financial
of university core facilities
support
Number
Total
(in
operating
budget
thousands
of dollars)
Percentage
of income
Percentage
of expenses
derived
from
covered
user
by user
fees
equipment
“Arithmetic
Range
30
$ 10-480
31
0-100
32
32
0-100
0-100
31
(±
average
so). See Table
I for definitions
$158.0
0-80
of median
and
Median
Mode
$113
$250
(±
120)
43.0
(±
37)
40
0
28.0
69.0
(±
(±
42)
43)
0
100
0
100
(±
14)
0
0
3.5
mode.
compared
with only $129,000 for those university
facilities that are not completely
self-supporting.
As expected, the self-supporting
facilities had large personnel
staffs, with an average of nearly five full-time
members
compared
with fewer than three full-time
members
in
a non-self-supporting
facility. Averaging
the data from
the 33 university
facilities
that provided
it, the typical
facility
recovers
about
40% of its operating
budget
through
the assessment
of user fees, which most often
are used to purchase
supplies (Table 2). In many cases
the university
subsidizes
the core facility by covering an
average
of 72% of the salaries of the facility personnel
(Table 2).
Among
those university
facilities
that operate
on a
fee-for-service
basis, there is a large range in the percentage
of total income derived
from user fees, from 5
to 100%, which provides
a reasonable
explanation
for
the equally
large range in service charges
assessed
in
different
university
facilities
for seemingly
identical
services.
As shown in Table
3, there is a 10-fold or
greater
range in the charges
assessed by different
university
facilities
for amino
acid analysis,
amino
acid
sequencing,
and peptide
synthesis.
There is a range of
threefold
in charges assessed for oligonucleotide
synthesis. If the varying extent of the percentages
of operating
costs covered by the fees at different
university
facilities
is taken into account,
the range narrows
considerably,
and it is then possible to calculate
the actual total cost
of each analysis.
Hence,
we estimate
that it costs a
university
facility $65 to hydrolyze
and do an amino
acid analysis
on one sample
and $874 to sequence
25
residues
in a protein
or peptide.
The most expensive
TABLE
Mean”
fees
Salaries
Supplies
Capital
of responses
service
offered
by most facilities
is peptide
synthesis
and, based on the data shown in Table 3, it costs $2078
to synthesize
and cleave but not purify
a 25-residue
synthetic
peptide.
In comparison,
a 25-residue
oligonucleotide
can be synthesized
and cleaved for $258.
In addition
to the various degrees of subsidization
of
the core facility by the university
or another
source,
there are several other factors that can contribute
to the
relatively
wide range of fees assessed
by different
university facilities for the same service (Table 3). One of
these
is the portion
of time spent
on nonincomeproducing
operations
such as maintenance,
calibration,
optimization,
and technical
development.
Although
the
mean value for 37 facilities was 19.7% (with an SD of
10.9%),
the amount
of time devoted
to nonincomeproducing
operations
ranged from as little as 5% to as
much as 60% of total instrument
time. Another
factor
that must be taken into account
is that in 11 of the 26
facilities that responded
to this question,
samples
submitted
by the director
of the facility
are run at no
charge.
The average
percentage
of instrument
time
devoted to these kinds of samples is 31.5% (SD 28.3%),
so it is clear that a facility that does not charge
for
analyses
performed
by its own director
might have to
charge other users according
to a somewhat
higher fee
schedule
than a facility
that bills all analyses
at the
same rate structure.
Capabilities
of core
Figure 3 displays
the core facilities
facilities
the five services most often offered by
surveyed.
Amino
acid analysis
and
3. Service charges in university facilities”
Number of
responses
Service
Hydrolysis/amino
acid
analysis
Amino-acid
sequencing
(25 residues)
Peptide synthesis/cleavage
(25 residues)
Oligonucleotide
Average
b
Mean
Range
Median
Mode
expenses
% of operating
co vered by fee
Average
nonsubsidized
26
$1 0-110
$39
(±
19)
$40
$40
60
$65
26
125-1250
542
(±
268)
590
650
62
874
19
270-2850
1330
(±
593)
1250
1500
64
13
98-302
(±
72)
300
76
fee’
2078
synthesis
(25 nucleotides)
“Rates for services
of median and mode.
from the host as opposed
‘Obtained
by dividing
SURVEY OF PROTEIN AND
NUCLEIC
196
to an outside
the mean
institution,
charge
ACID CORE FACILITIES
in U.S.
by the fraction
185
dollars.
of operating
bArithmetic
average (± SD). See Table
expenses covered by the fee.
258
I for definitions
3127
a
a
0
a,
to
0’
C
0
>
0
0
a
a
U-
Figure
3. Availability
in 40 core
facilities.
of protein
and
nucleic
acid chemistry
services
protein
sequencing
are offered by nearly 90% of these
facilities.
Peptide
synthesis,
DNA synthesis,
and peptide isolation
are offered by somewhat
fewer facilities,
that is, about 60% of those surveyed.
These figures are
somewhat
misleading
in that there are very few facilities that offer all of the services
shown in Fig. 3; only
5 of the 40 facilities
surveyed,
or about
12%, offer all
five
services.
The tremendous
range in the size of core facilities is
similarly reflected by the range in the number of analyses
that are carried out per month by each facility offering
the respective
services (Table 4). Hence facilities were
surveyed
that routinely
carry out as few as 2 or as many
as 260 hydrolyses
and amino acid analyses
per month.
The average number
of amino acid analyses carried out
per month
in these facilities
was 74. Similarly,
the average
number
of samples
subjected
to amino
acid
sequencing
was 17 per month, and the average number
of synthetic
peptides
and oligonucleotides
that were made in the
facilities
surveyed
were 6.1 and 41 per month,
respectively.
In addition
to the number
of analyses
that can be
completed
per month,
another
important
question
relating to the capabilities
of core facilities
relates to the
amount
of protein
required
for each analysis.
Although
Table 5 indicates
that there was a 200-fold range in the
responses
noted to a question
concerning
the amount
of
TABLE
4. Number
of services” carried out per month
Service
Hydrolysis/amino
acid analysis
Amino-acid
sequencing
Peptide
synthesis/cleavage”
Oligonucleotide
synthesis’
protein
required
to sequence
the first 15 residues,
nearly 60% of the facilities (19 of the 32 that responded)
agreed that from 50 to 100 pmol is the amount
generally required
to obtain
the data on an Applied
Biosystems instrument.
The average value was 150 pmol,
and only five facilities
replied
that this data could
generally
be acquired
from less than 50 pmol of protein. Although
only four laboratories
surveyed
are
equipped
with Beckman
Instruments
(Fullerton,
Calif.)
spinning
cup sequencers,
it appears
from the limited
data that this approach
requires
at least threefold
larger
amounts
of protein
than the Applied
Biosystems
gas
phase or pulsed liquid phase instruments.
For the purposes of this survey, the latter two instruments
were
grouped
together.
If, instead
of being given a sample ready for direct
NH2-terminal
sequence
analysis,
a facility must first
carry out a tryptic digest and isolate the resulting
peptides by HPLC,
then an average of 1.2 nmol (or an eightfold larger amount)
of protein
is required
(Table 5).
Although
there was a range of 80-fold in the responses
received
to this question
(from 0.05 to 4.0 nmol of protein), 63% of the facilities surveyed
(10 of 16) that are
equipped
with Applied
Biosystems
sequencers
replied
that from 0.2 to 1.0 nmol of protein
would be required
for the initial
tryptic
digest to ensure
that a typical
15-residue
tryptic
peptide
that might
result could be
HPLC
purified
and then completely
sequenced.
Although
only two facilities
were surveyed
that offer
HPLC
peptide
isolation
and are equipped
with Beckman Instruments
spinning
cup sequencers
(Table 5), it
appears from this limited data that this approach
would
also require
more than 200 pmol of protein.
Overall it was reported
that approximately
21% of all
samples
(with a SD of 14%) subjected
to amino
acid
sequencing
fail
to yield
any
usable
data.
Reasons
for
this seemingly
low failure
rate would include
an insufficient
amount
of protein
or peptide,
blocked NH2terminus,
as well as insufficient
purity and instrument
failures.
Based on the data in Table 6, an average of about
190 pmol of protein
(i.e., about
5 tg of a 25,000dalton
protein)
is required
to obtain
an amino
acid
composition
by using
ion exchange
separation
that
is accurate
to within
± 10%. In this regard, PTC amino
acid analysis
appears
to be about
twice as sensitive
since the same data could be obtained
by this approach
in core facilities
Number of
facilities’
Range
35
36
22
16
2-260
3-40
1-15
3-120
Mean’
74.0
17.0
6.1
41.0
(±
(±
(±
(±
76)
9.4)
3.9)
32)
Median
Mode
41.0
15.0
5.5
30.0
10
15
10
30
“Number
of samples
subjected
to sequencing
and the number
of individual
peptides
and oligonucleotides
synthesized
are given rather than number
of cycles completed.
In the case of amino acid sequencing,
a few responses
were in terms of cycles completed,
which were converted
to number
of samples,
based on the assumption
that an average
sequencing
run extends
for 20 cycles.
of facilities
(of the 40 that responded
to the survey) offering
the indicated
service.
‘Arithmetic
average
(± so); see Table 1 for definitions
of median
and mode.
dDoes not necessarily
include
purification
and
characterization.
‘Excludes
purification
and assumes
smallest
scale of synthesis
available.
3128
Vol.
2
Dec.
1988
The FASEB Journal
WILLIAMS
ET AL.
TABLE
5. Amount of protein required to obtain 15
amino
acid residues of sequence
Am ount
Number of
facilities surveyed
Range
of protein,
nmol
Mean”
Median
Mode
From the amino
terminus
Applied
Biosystems
sequencer
Beckman
Instruments
sequencer
32
4
0.005-1000
0.05-1
0.15
0.51
(±
(±
0.20)
0.34)
0.075
0.50
0.050
0.50
From an internal
tryptic
peptideb
Applied
Biosystems
sequencer
Beckman
Instruments
sequencer
16
2
0.05-4.0
0.2-10
1.2
5.1
(±
(±
1.6)
4.9)
0.40
1.0
“Arithmetic
average (± SD); see Table
as a service, what would be the least amount
to completely
sequence a typical 15-residue
DISCUSSION
Although
the development
of microchemical
core facilities is rapidly
bringing
state-of-the-art
techniques
for
synthesis
and structural
analysis
of proteins
and genes
within reach of an ever-increasing
range of biomedical
researchers,
it is essential
that everyone
concerned
have
realistic
expectations
regarding
the support
and capabilities of these facilities. With an average staff of nearly
three
full-time
personnel
and seven complex
instrument systems,
it is clear that to establish
a full-service
facility requires
a substantial
commitment
in terms of
space, personnel,
and financial
backing.
If the facility
is expected
to be self-supporting
and therefore
derive
100% of its operating
expenses
from user fees, then
(based on this survey) it would need to be considerably
larger than the present
average.
Only 6 of the 31 university
facilities
that responded
are completely
selfsupporting,
and they contain
an average
of 10 instrument systems
(where
the amino
acid sequencer
and
are each
considered
TABLE 6. Amount of protein required
separate
systems)
to obtain an amino
and
acid composition
the equivalent
of five full-time
personnel.
The average
annual operating
budget for these self-supporting
facilities was nearly $300,000.
More typical is the “average”
facility that derives about 43% of its $158,000
annual
operating
budget
from user fees and the other 57%
from other sources. The total work output from such an
average
facility can be calculated
from the data contained in Table 4 as the product
of the number
of facilities offering
a particular
service
and the number
of
services
carried
out divided
by 40 (the total number
of
facilities
that responded
to the survey).
The resulting
average
corresponds
to 65 amino
acid analyses,
15
amino
acid sequencing
runs, three peptide
syntheses,
and 16 oligonucleotide
syntheses
per month.
These average data regarding
a typical facility should
be taken as only a first approximation,
since no effort
was made to closely define the scope of each service,
which can vary widely and thus affect the apparent
productivity
as measured
by our survey. For example,
protein sequence
analysis
may or may not include
a final
purification
step, quantitation
by amino acid analysis,
assessment
of purity by SDS polyacrylamide
gel electrophoresis,
or data interpretation,
which ranges from
providing
a called sequence
as well as individual
cycle
yield, background
correction,
carryover
and repetitive
yields to providing
only the raw data sheets.
Peptide
synthesis
may or may not include
ninhydrin
coupling
yields at each step, HPLC
purification,
an analytical
HPLC
profile,
amino
acid analysis,
fast-atom
bom-
that is accurate
to within
± 10%
Amount
Number of
facilities surveyed
Ion
14
9
4
1
30-500
40-500
75-500
-
30
Reverse-phase
Phenylisothiocyanate
22
21
1
4-500
4-500
85
89
chloroformate
-
of protein,
Mean”
Range
exchange
Ninhydrinb
o-Phthalaldehyde
Fluorescamine
Fluorenylmethyl
-
1 for definitions of median and mode.
5The question asked was, “If your facility offers peptide isolation
of a tryptic digest of a 25,000-dalton
protein you would need to subject to HPLC so that you would be able
peptide?”
with about 90 pmol of protein.
Although
an ion exchange/fluorescamine
detection or a fluorenylmethyl
chloroformate
(FMOC),
precolumn
derivatization/reversephase
separation
approach
might
provide
increased
sensitivity
(Table
6), not enough
core facilities
that
responded
to our survey are using these techniques
to
warrant
any firm conclusions.
PTH-HPLC
5.1
179
187
197
8
(±
(±
(±
Median
162)
156)
176)
-
(±
(±
pmol
90
100
107
-
Mode
-
-
108)
40
100
109)
40
100
-
-
-
“Arithmetic
average (± SD): See Table 1 for definitions of median and mode.
bData include three facilities equipped
with Beckman Models
120C or l2lM and four facilities equipped with Beckman Model 6300 analyzers.
The mean amount of protein required for Model 6300 was 260 pmol
compared
with 173 pmol for the older Model 120C or 121 M.
SURVEY OF PROTEIN AND
NUCLEIC
ACID CORE FACILITIES
3129
(FAB) mass spectrum,
and/or
amino
acid
analysis of the final product.
The extent of involvement
of the facility in each of its services obviously
can significantly
affect the productivity
as measured
by
the number
of analyses
completed
per staff member.
The cost of performing
these services ranges from $65
for the amino acid analysis
to $2078 for synthesizing
a
25-residue
peptide.
To sequence
this same
peptide
would cost $874, and to synthesize
a 25-residue
oligonucleotide
would cost $258. Before any comparison
can
be made between
charges assessed by two different
university facilities,
it is essential
that the charges are first
corrected
for the fraction of the actual cost of the analysis that the charge covers. Obviously,
a core facility that
recovers
100% of its operating
expenses
from user fees
would have to charge more for any given service than
a facility that is 60% subsidized
and therefore
has to
recover only 40% of the cost of the analysis
from the
user.
Finally, of crucial concern
to the prospective
user of
the facility is the amount
of material
required
for the
proposed
analysis.
Although
it is certainly
possible
to
use background
subtraction
and PTC amino acid analysis to obtain an amino acid composition
on as little as
600 fmol (about
40 ng) of a single standard
protein
such as bovine serum albumin
(4), it is clear from the
data in this survey that with real samples
in a high
throughput
facility, at least 50 times this amount,
or 85
pmol (i.e., 2 tg of a 25,000-dalton
protein),
is required
to routinely
obtain
an amino acid composition
that is
accurate
to within
± 10%. It should
be noted that a
proportional
amount
of material
would also be required
to obtain a comparable
analysis of a peptide.
Similarly,
although
it is possible to determine
the sequence
of the
first 22 amino acids in myoglobin
by using only 5 pmol
of protein
(5), a well-run
university
facility working
with real samples,
which may have been prepared
by
nonprotein
chemists
from an immense
variety of purification and sample preparation
procedures,
would require at least 30-fold this amount,
or 150 pmol of protein, to routinely
obtain the same amount
of sequence.
If, as is frequently
the case with eukaryotic
proteins
(6),
the NH2-terminus
is blocked,
then more than a nanomole of protein
on the average
would be required
to
carry out a tryptic digest and then isolate and sequence
15 amino acids in one of the resulting
peptides.
bardment
sequence
3130
Vol.
2
Dec.
1988
The results of this survey clearly demonstrate
that instances in which analyses
of extremely
high sensitivity
have been carried
out are relatively
isolated;
they are
not representative
of what can be expected
when a typical
unknown
protein sample prepared
by the investigator
is
submitted
to a well-run
core facility,
whether
it is
located
in a university,
government,
or industrial
laboratory.
Most of this discussion
has centered
on a hypothetical
average
or typical facility that is equally
proficient
at
carrying
out all services.
Actually,
many facilities tend
to specialize
to some extent,
and thus expend
more
energy on and are better equipped
to carry out one or
more of the services
offered.
Hence,
just because
a
particular
facility may perform
above average
in one
kind of analysis
does not necessarily
mean that it will
perform
better than average on all types of analyses and
services.
In the future we can expect to see the creation
of new
microchemical
facilities
and the growth of established
facilities.
Automated
DNA sequencers
are now available; mass spectrometry
has an ever-expanding
role in
protein
chemistry;
and sequence
analyses
will be performed
at even higher
sensitivities.
These
increased
services
at escalating
levels of sophistication
will increase demands
for more technical
personnel.
All past
indications
are that the efficiency, quality, and development of services will improve
with increased
administrative commitment
and greater
user education.
REFERENCES
I.
M.; KENT, S.; CARUTHERS,
M.; DREYER, W.; FIRcA,J.;
GIFFIN, C.; HORVATI-I, S.; HUNKAPILLAR,
T.; TEMPST, P.; HOOD,
L.
A microchemical
facility for the analysis and synthesis of genes and proHUNKAPILLAR,
teins.
Nature (London) 310: 105-111; 1984.
P. Sequence from picomole
quantities
troblotted onto polyvinylidene
difluoride
membranes.
262: 10035-10038;
1987.
3. Nic,
R.; ATHERTON,
D.; FOWLER, A.; KUTNY,
2.
MATSUDAIRA,
of proteins
J. Biol.
elecC/len.
R.; SMITH, A.
Research
Resource
Facility Satellite
Meeting.
Walsh,
K. A., ed. Protein
sequence analysis. Clifton,
N.J.: Humana;
1986: 317.
4. STONE, K.; WILLIAMS,
K. High-performance
liquid chromatographic
peptide
mapping
and amino
acid analysis
in the sub-nanomole
range.
J. Chromatogr. 359: 203-212; 1986.
5. HUNKAPILLAR,
M.; HooD, L. Protein sequence analysis: automated
microsequencing.
Science 219: 650-659;
1983.
6. WOLD, F. Acetylated
N-terminals
in proteins-a
perennial
enigma.
Trends Bioc/lem. Sci. 9: 256-257;
The FASEB Journal
1984.
WILLIAMS
ET AL.