To take a slightly different tack and, perhaps, to "pour some more oil on
the fire", I would like to raise a discussion on the utility of the pepti=
de
mass fingerprinting technique in the light of Ken Williams data below.
In his description of his MS-Fit routine (http://prospector.ucsf.edu/html=
ucsf/instruct/fitman.htm), Karl Clauser
ends his discussion on the improvement of the mass accuracy of modern
MALDI mass spectrometers by suggesting that "proteins can now be confiden=
tly
identified by peptide mass fingerprinting using masses alone with MS-Fit.
Identification certainty is primarily a function of the level of mass acc=
uracy." I have=20
continued to wonder what this certainty level actually was with modern
instruments.
Ken Williams has, certainly for the first time I can recall seeing, quote=
d
some real life figures relating to this kind of protein identification. T=
hese
are, as Ken explains, genuine user supplied samples and he, no doubt with
the greatest experimental care, has been able to identify 65% of a set of=
150 proteins
by peptide mass searching. I take it Ken that this refers to proteins tha=
t
could be unequivocally identified by only the information on the protein
as it was presented to the laboratory and the peptide mass list with no a=
dditional=20
structural information? I also take it that these identifications were=20
confirmed with some supplementary structural techniques like Edman, MS/MS
or PSD?=20
Does this mean that there is, in fact, an overall 35% uncertainty with th=
e=20
peptide mass fingerprinting technique and, if this is so, shouldn't this
stastic bother us given the requirement of our collaborators/customers to
unequivocally identify their proteins?=20
Don't get me wrong here, I think a list of masses from a tryptic digest i=
s
a fantastic thing to have to give you that warm "I know I have got the ri=
ght=20
protein" feeling after getting some sequence information from Edman or
MS/MS.=20
It just bothers me that peptide mass fingerprinting is being promoted as =
a=20
primary technique of protein identification where I believe that Ken=20
Williams data (and my personal experience with a much smaller set of prot=
eins)
suggests that it isn't . I think this type of data is enormously valuable=
and goes
hand in hand with the biological information as well as the physicochemic=
al
data (like pI and MW from a 2D gel) to give another level of confidence t=
o a
protein identification but if there is anyone out there planning to set =
up a
protein identification facility based on peptide mass fingerprinting alon=
e without
the support of Edman or MS/MS, I would suggest they consider Ken's number=
s=20
carefully. I certainly won't report a protein identification based on a=20
technique that only identifies two out of three proteins correctly.
If there is anyone else out there with a good set of data on real
unknowns, it would be great to hear about them, I look forward to the day=
when this
type of fingerprinting will achieve accuracies in the high 90's%, until t=
hen, I
think it important to interpret peptide mass fingerprinting data with due
statistical caution.
In view of Karl Clausers comments on mass accuracy at the beginning of
this message, I think we need to discuss and promote techniques to drive
our mass accuracies up in order to evaluate, with the help of large
unknown sample sets, the real confidence limits to this technique. Lookin=
g
forward to continuing these discussions in San Diego.
Regards to all....Ken Mitchelhill=20
On Fri, 6 Mar 1998, Kenneth Williams wrote:
> Axel,
>=20
> Thank you so much for posting your detailed comments (copy below) abou=
t in
> gel/PVDF digests and please by all means continue to throw all the "oil=
in
> the fire" that you like - it is the only way to shed some light on the
> problems.
> =09
> We agree that 50 pmol is a rather large amount - that frankly we often=
do
> not see. The minimum amount that we believe is generally recommended (=
by
> laboratories that specialize in this area) for conventional Edman inter=
nal
> sequencing is 5-10 pmol with less being needed for MS-based protein
> identification. However, in setting up our studies we have to take int=
o
> account the varying degree of expertise of our members - for instance, =
we
> were delighted that several participants in our current study used our
> sample to try in gel digests for the first time. So, we tried to strik=
e a
> balance by including a 10 pmol minor component (along with the 50 pmol
> major protein) in our current study. We hope you will find the results=
of
> this study, which will be presented at the ABRF Meeting in a few weeks,=
of
> interest.
>=20
> With regard to sample clean-up following in gel digests, in our opini=
on,
> it is not generally needed when there is sufficient protein to permit
> MALDI-MS to be carried out on <10% aliquots of in gel digests of singl=
e,
> Coomassie blue stained gel bands that are carried out as described at t=
he
> references that are given below. Approximately 150 in gel digests have
> recently been directly subjected to MALDI-MS under these conditions
> following a protocol:
> =20
> http://info.med.yale.edu/wmkeck/geldig3.htm)=20
>=20
> that is very similar to the representative protocol on the ABRF Interna=
l
> Sequence Research Group Web Site:
>=20
> http://www.medstv.unimelb.edu.au/ABRF/ResearchCommittees/intprotseqresc=
omm.h
> tml
>=20
> In each case the MALDI-MS has been carried out on a microliter or so of=
the
> digest and the remaining gel/digest then frozen (as is) while the MALDI=
-MS
> is carried out and the database searching completed. Of these 150
> user-submitted, "unknown" samples, 65% have been identified by peptide
> mass searching via the ProFound algorithm. Please note these samples w=
ere
> not "extracted" and that the fraction of proteins identified increases =
to
> above 90% in the case of an organism like yeast whose genome has been
> sequenced.=20
>=20
> Sample clean-up or repeated cycles of vacuum centrifugation (to lower =
the
> ammonium bicarbonate level) may well be needed at extremely low (<pmol)
> levels and when larger fractions of the digest are subjected to MALDI a=
nd
> other types of MS. =20
>=20
> A number of laboratories have developed methods to clean-up low pmol
> levels of sample digests using reverse-phase material either self-packe=
d or
> in a small column. A few comments on these approaches from members of =
our
> committee are listed below.=20
>=20
> One of our members (Ulf) notes sample clean-up takes less than 15 minu=
tes
> (for a beginner) on a one-time use "pico-column", made from an Eppendor=
f
> "Geloader" tip, into which is packed about 1 =B5l of Poros coarse parti=
cles.
> After loading, washing and eluting the peptide mix in a few microliters=
,
> MALDI-MS can then be carried out "undisturbed". At this level the
> non-desalted material does not provide a useable MALDI spectrum. =20
>=20
> Another of our members (Roland) adds, "We only use the cleanup for the
> very lowest level of samples, probably on the order 50 ng or less. The=
n we
> commit approx 25-33% of the sample to clean up, reserving the rest for
> nanoES. We also find we need to use the columns as concentrators when =
we
> had to use large digestion volumes because the spot was from a prep gel.
> One extra trick which I picked up from Peter Roepstroff's lab is to cri=
mp
> the pipet tip by bending it in half before loading the Poros . This
> effectively traps the beads and you no longer have to take pains to kee=
p it
> from going dry. This tremendously simplifies the elution step."
>=20
> Scott adds, we have carried out a detailed comparison (Electrophoresis
> (18, 369-381, 1997), of in-gel and on-membrane digests using standard
> proteins and the results obtained indicated the sample clean-up using a
> reverse-phase method (e.g. BioTechniques 22:244-250, 1997) is required =
for
> successful MALDI-MS peptide mass-mapping (or PSD-MALDI-MS) of low to
> sub-pmol levels of in-gel digested samples.
>=20
> Other relevant references are:
> =20
> Lui et al. (1996) Anal Biochem. 241, 156-166.
> Kussmann et. al. (1997) J. Mass Spectrom. 32, 593-601.
> Shevchenko et al., (1996) Anal. Chem. 68, 850-858
> Otto et al, (1996) Electrophoresis 17, 1643-1650
>=20
> Your remaining points about the need to improve methodologies, the ver=
y
> few laboratories that carry out sequencing of less than pmol amounts of
> novel proteins and what happens to the "missing" 80% of an average dige=
st
> are extremely well taken and map out some very challenging areas for fu=
ture
> studies.
>=20
> Thanks again for taking the time to post your comments,
>=20
> Roland Annan, Ulf Hellman, Scott Patterson, Kathy Stone, Kristine Swid=
erek
> & Ken Williams
>=20
> _________________
>=20
> (Copy of 3/4 posting by Axel Ducret)
>=20
> Ken,
>=20
> I do not want to throw oil in the fire, but the problem with this study
> lies precisely in the protein amounts that have been analyzed. With 50 =
pmol
> on hand, and having good sequencers and/or a good LC/MS system (or, of
> course, a MALDI-Tof), I am quite confident to get a number of relevant
> amino acid sequences, independently if I get the protein in a gel, on a
> PVDF membrane or on a nitrocellulose membrane because, even if the reco=
very
> yields are low, there is still sufficiently material available to perfo=
rm
> the analyses (20% of 50 pmol is still 10 pmol, a lot for modern MS
> instruments or sequencers). Now, what I would like to see is the same s=
tudy
> done with only 5 or 1 pmol of protein available, because now matrixes b=
egin
> to substantially interfere with the analysis. Personally (and, please, =
do
> not see this comment as an attack on your preferred method; this is jus=
t a
> matter of personal preference, as Ken mentioned), I have not managed to=
get
> substantial good results with PVDF membranes. Although I used to prefe=
r
> nitrocellulose membranes over in-gel digest, I also had to realize that
> recovery yields substantially decrease once the low pmol is reached. I=
do
> not like present in-gel digest protocol because you have to clean up th=
e
> digests from a lot of garbage before applying to a MALDI target or a
> capillary column. And clean-up, at least in my eyes, means losses. So, =
what
> I am heading to is the following. I do not think that the current
> bottleneck is the instrument sensitivity (although this could be discus=
sed
> for Edman sequencers). Most of modern MS instruments have currently
> sensitivity going to the low fmol and heading to the amol. However, the=
re
> are not many publications dealing with the sequencing of novel protein
> below the pmol. So where is the problem? Consistently, one of the probl=
em
> that I have observed here in my lab and at others is sample handling. W=
e
> have to realize that methods for protein digestion have not changed
> dramatically over the last 5 years. Yes, of course, we have miniaturize=
d
> the equipment, tried to worked with smaller volumes, concentrated the
> proteins in smaller areas. However, there is no true consensus on how i=
t
> should be done. The ABRF could take a lead here on conducting a study o=
n
> how to manipulate very low amounts of proteins in a real world fashion,
> that is a protein, let's say, that have been isolated from a biological
> tissue using conventional purification technique and contaminated with =
the
> usual cocktail of salt and detergents that typically dooms my own analy=
sis.
> Because if we have currently techniques to retrieve 20% of the availabl=
e
> protein amount, we got to find something to tap in these remaining 80%!
>=20
>=20
********************
Ken. I. Mitchelhill
John Holt Protein Structure Laboratory
St. Vincent's Institute for Medical Research
41 Victoria Parade
Fitzroy 3065 Victoria
Australia
Telephone: 61-3-9288 2480
Fax: 61-3-9416 2676
Email: ken@ariel.ucs.unimelb.edu.au
Webmaster for the Association of Biomolecular Resource Facilities
http://www.medstv.unimelb.edu.au/abrf.html
*********************