Iozzo Renato V. Proteoglycan Protocols
Подождите немного. Документ загружается.
Cartilage and Smooth Muscle Cell Proteoglycans 65
References
1. Doege, K. J. (1995) Aggrecan, in Guidebook to the Extracellular Matrix and Adhesion
Proteins (Kreis, T. and Vale, R., eds.) Oxford University Press, Oxford, New York, Tokyo,
pp. 17–18.
2 Zimmerman D. R. (1995) Versican, in Guidebook to the Extracellular Matrix and Adhe-
sion Proteins. (Kreis, T. and Vale, R., eds.), Oxford University Press, Oxford, New York,
Tokyo, pp. 100–101.
3 Mow, V. C. and Mak, A. F. (1986) Lubrication of diarthrodial joints, in Handbook of Bioengi-
neering (Skalak, R. and Chien, S., eds.), McGraw-Hill, New York, NY, pp. 5.1–5.34.
4. Berenson, G. S., Radhakrishnamurthy, B., Srinivasan, S. R., Vijayagopal, P., Dalferes, E. R.,
and Sharma, C. (1984) Recent advances in molecular pathology: carbohydrate-protein mac-
romolecules and arterial integrity—a role in atherogenesis. Exp. Mol. Pathol. 41, 267–287.
5. Hardingham, T. E. and Muir, H. (1972) The specific interaction of hyaluronic acid with
cartilage proteoglycans. Biochim. Biophys. Acta 279, 401–405.
6. Hascall, V. C. and Heinegård, D. (1974) Aggregation of cartilage proteoglycans. The role
of hyaluronic acid. J. Biol. Chem. 249, 4232–4241.
7. Caterson, B. and Baker, J. (1978) The interaction of link proteins with proteoglycan mono-
mers in the absence of hyaluronic acid. Biochem. Biophys. Res. Commun. 80, 496–503.
8. Kongtawelert, P. and Ghosh, P. (1990) A method for the quantitation of hyaluronan (hyalu-
ronic acid) in biological fluids using a labeled avidin-biotin technique. Anal. Biochem.
185, 313–318.
9. Fosang, A. J., Hey, N. J., Carney, S. L., and Hardingham, T. E. (1990) An Elisa based assay
for hyaluronan using biotinylated proteoglycan G1 domain (HA binding region). Matrix
10, 306–313.
10. Bray, B. A., Hsu, W., and Turino, G. M. (1994) Lung hyaluronan as assayed with a
biotinylated hyaluronan-binding protein. Exp. Lung Res. 20, 317–330.
11. Asari, A., Miyauchi, S., Kuriyama, S., Machida, A., Kohno, K., and Uchiyama, Y. (1994)
Localization of hyaluronic acid in human articular cartilage. J. Histochem. Cytochem. 42,
513–522.
12. Asari, A., Miyauchi, S., Takahashi, T., Kohno, K., and Uchiyama, Y. (1992) Localization of
hyaluronic acid, chondroitin sulfate and CD44 in rabbit cornea. Arch. Histol. Cytol. 55, 503.
13. Edelstam, A. A. B., Lundkvist, O. E., Wells, A. F., and Laurent, T. C. (1991) Localization of
hyaluronan in regions of the human reproductive tract. J. Histochem. Cytochem. 39,1131–1135.
14. Hellström, S., Laurent, C., and Yoon, Y.-J. (1994) Distribution of hyaluronan in the middle
and inner ear. A light microscopical study in the rat using a hyaluronan binding protein as a
specific probe. J. OtorhinoParyngal. Relat. Spec.ORL 56, 253–256.
15. Ripellino, J. A., Bailo, M., Margolis, R. U., and Margolis, R. K. (1988) Light and electron
microscopic studies on the localization of hyaluronic acid in developing rat cerrebellum.
J. Cell. Biol. 106, 845.
16. Ripellino, J. A., Klinger, M. M., Margolis, R. U., and Margolis, R. K. (1985) The hyaluronic
acid binding region as a specific probe for the localization of hyaluronic acid in tissue sec-
tions. Application on chick embryo and rat brain. J. Histochem. Cytochem. 33, 1060.
17. Rosenbaum, D., Peric, S., Holecek, M., and Ward, H. E. (1997) Hyaluronan in radiation
induced lung disease in the rat. Radiat. Res. 147, 585–591.
18. Lee, H. G. and Cowman, M. K. (1994) An agarose gel electrophoretic method for analysis
of hyaluronan molecular weight distribution. Anal. Biochem. 219, 278–287.
19. Pouyani T. and Prestwich, G. D. (1994) Biotinylated hyaluronic acid: a new tool for prob-
ing hyaluronate-receptor interactions. Bioconjugate Chem. 5, 370–372.
66 Melrose
20. Melrose, J., Numata, Y., and Ghosh, P. (1996) Biotinylated hyaluronan: a versatile and
highly sensitive probe capable of detecting nanogram levels of hyaluronan binding proteins
(hyaladherins) by a novel affinity detection procedure. Electrophoresis 17, 205–212.
21. Melrose, J., Little, B., and Ghosh, P. (1998) Detection of aggregatable proteoglycan popu-
lations by affinity blotting using biotinylated hyaluronan. Anal. Biochem. 256, 149–157.
22. Melrose, J., Whitelock, J., Xu, Q., and Ghosh, P. (1998) Pathogenesis of abdominal aortic
aneurysms: possible role of differential production of proteoglycans by smooth muscle
cells. J. Vasc. Surg. 28, 676–686.
23. Allington, W. B., Cordry, A. L., McCullough, G. A., Mitchell, D. E., and Nelson, J. W.
(1978) Electrophoretic concentration of macromolecules. Anal. Biochem. 85, 188–196.
24. Rosenfeld, S. A., Ross, O. H., Corman, J. I., Pratta, M. A., Blessington, D. L., Feeser, W.
S., and Freimark, B. D. (1994). Production of human matrix metalloproteinase-3
(stromelysin) in Escherichia coli. Gene 139, 281–286.
25. Hughes, C. E., Caterson, B., Fosang, A. J., Roughley, P. J., and Mort, J. S. (1995) Mono-
clonal antibodies that specifically recognise neoepitope sequences generated by aggrecanase
and matrix metalloproteinase cleavage of aggrecan: application to catabolism in-situ and
in-vivo. Biochem. J. 305, 799–804.
26. Caterson, B., Christner, J. E., Baker, J. R., and Couchman, J. R. (1985) Production and
characterization of monoclonal antibodies directed against connective tissue pro-
teoglycans. Fed Proc. Fed. Am. Soc. Exp. Biol. 44, 386–393.
27. Caterson, B., Calabro, T., and Hampton, A. Monoclonal antibodies as probes for elucidating
proteoglycan structure and function in Biology of the Extracellular Matrix (Wright, T. and
Mecham eds, R., eds.),Academic Press, New York, NY,1987.
28. Caterson, B., Christner, J. E., and Baker, J. R. (1983) Identification of a monoclonal anti-
body that specifically recognises corneal and skeletal keratan sulfate. J. Biol. Chem. 258,
8848–8854.
29. Heinegård D., Bjorne-Persson A., Cöster L., Franzén S., Gardell A., Malmström M.,
Paulsson, M., Sandfalk, R., and Vogel, K. (1985) The core proteins of large and small
interstitial proteoglycans from various connective tissues form distinct subgroups.
Biochem J. 230:181–194.
30. Bitter, T. and Muir, H. M. (1962) A modified uronic acid carbazole reaction. Anal.
Biochem. 4, 330–334.
31. Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H., Provenzano,
M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Klenk, D. C. (1985) Measurement
of protein using bicinchoninic acid. Anal. Biochem. 150, 76–85.
32. Tengblad, A. (1979) Affinity chromatography on immobilised hyaluronate and its appli-
cations to the isolation of hyaluronate binding proteins from cartilage. Biochim. Biophys.
Acta 578, 281–289.
33. Farndale R. W., Buttle, D. J., and Barrett, A. J. (1986) Improved quantitation and dis-
crimination of sulfated glycosaminoglycans by use of dimethylmethylene blue. Biochim.
Biophys Acta 883, 173–177.
34. Whitelock, J., Mitchell, S., and Underwood, P. A. (1997) The effect of human endothelial cell
derived proteoglycans on human smooth muscle cell growth. Cell Biol. Int. 21, 181–189.
35. Towbin, H., Staehelin, T, and Gordon, J. (1979) Electrophoretic transfer of proteins from
polyacrylamide gels to nitrocellulose sheets. Procedure and some applications. Proc. Natl.
Acad. Sci. (USA ) 76, 4350–4354.
36. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature (Lon.) 227, 680–685.
Preparation for Amino Acid Sequence Analysis 67
67
From:
Methods in Molecular Biology, Vol. 171: Proteoglycan Protocols
Edited by: R. V. Iozzo © Humana Press Inc., Totowa, NJ
8
Preparation of Proteoglycans
for N-Terminal and Internal Amino Acid Sequence Analysis
Peter J. Neame
1. Introduction
Amino acid sequence analysis of proteoglycans is performed using many of the
same methods that are used for conventional proteins and glycoproteins, with some
specific modifications that result from the glycosaminoglycans that are attached to
the protein core. Amino acid sequence analysis of proteoglycans is more challenging
than for conventional proteins for two reasons. First, because proteoglycans are large
molecules, they have the same problems that are inherent in sequence analysis of
larger proteins. The higher molecular weight provides for relatively high levels of
nonspecific cleavage of the protein during Edman chemistry, resulting in a rapidly
increasing background. Second, the presence of glycosaminoglycans, which tend to
bind water, can exacerbate the problem of nonspecific hydrolysis of peptide bonds to
such an extent that it may be impossible to obtain an N-terminal on a proteoglycan
that is over 70% glycosaminoglycan. In an ideal situation, it is difficult to determine
more than 10 amino acids from the N-terminal of an intact proteoglycan.
On the other hand, the glycosaminoglycan chains provide a unique handle that
enables proteoglycans or glycosaminoglycan-containing peptides to be separated
from other proteins or peptides by a combination of size-exclusion chromatography and
anion-exchange chromatography. By judicious use of glycosaminoglycan-specific
endoglycosidases, it is possible to separate different families of proteoglycans from
each other.
Classically, proteins have been best identified or characterized by amino acid
sequencing (Edman degradation). The chemistry has changed little since its original
conception (1), although the analytical tools available have improved enormously. In
general, proteins and proteoglycans are most easily identified or characterized by
obtaining internal peptide sequence, rather than an N-terminal sequence. This elimi-
nates any problems caused by blocked N-termini and allows the investigator to make
68 Neame
multiple attempts to obtain unequivocal data. Direct analysis of the N-terminal is best
achieved from a sample that is Western blotted onto polyvinyl difluoridene (PVDF) to
reduce the risk that the N-terminal may derive from a lower-molecular-weight
contaminant that, while minor on a mass basis, represents a significant contaminant
on a molar basis.
Recently, it has become possible to reliably analyze peptides derived from pro-
teolytic digestion by mass spectrometry. The resulting list of fragment masses can
then be compared to a database of fragment masses and proteins identified with a high
degree of confidence. This is both cheaper and quicker than Edman degradation. How-
ever, a major limitation at present is that the protein must be present in the databases to
be identified, or it must be so similar in sequence to a protein that is in the databases
that most peptides have identical masses. In practice, this means >98% sequence iden-
tity and a restriction to human and murine molecules, as the number of known proteins
in these species begins to approach the complete repertoire found in the genome. Pro-
teins that have extensive posttranslational modification cannot readily be identified in
this way.
Mass spectrometry has also made inroads into direct amino acid sequence analysis;
individual peptides can be separated and partially degraded on the mass spectrometer,
and the fragments that are released analyzed to determine changes of mass. This meth-
odology does not allow differentiation of leucine or isoleucine, but in the hands of a
skilled operator can provide considerable useful data. Extensive novel amino acid
sequence is still difficult to determine by mass spectrometry, but it is a valuable
adjunct to classical methods (2).
It is unlikely that individual investigators will have direct access to mass spectrom-
eters or automated sequencers, so the methodologies described here are designed to
enable the investigator to prepare samples that are suitable for mass spectrometry or
Edman degradation performed by a core facility. As this chapter is designed to help
individuals working with proteoglycans, it will be assumed that glycosaminoglycan
chains will be left intact. The majority of core laboratories are not familiar with exten-
sively glycosylated proteins and may need some education on the unique properties
of these molecules.
To obtain good yields of protein, transfers from one container to another should be
avoided as much as possible, particularly in the later stages of purification. Losses are
generally a result of surface adsorption; it is easy to lose as much as 5 µg protein/
container. If 20 µg of protein is present at the start, there will not be much left after
three transfers from one container to another.
The methods described here have proved to be generally useful for structural analy-
sis of proteoglycans (and several other proteins) in the author’s lab. There are a num-
ber of more detailed descriptions of some of the techniques mentioned; the reader is
referred to publications from the Protein/DNA Technology Center at The Rockefeller
University (3) and the Howard Hughes Biopolymer/Keck Foundation Biotechnology
Resource Laboratory at Yale (4,5).
This overview consists of three parts: (1) preparation of proteoglycans for N-termi-
nal analysis, (2) proteolytic digestion of proteoglycans, and (3) separation of peptides
Preparation for Amino Acid Sequence Analysis 69
by column-based two-dimensional chromatography. The example of two-dimensional
chromatography described here is size-exclusion chromatography followed by
reversed-phase chromatography. However, the first dimension could be ion-exchange
chromatography or affinity chromatography (6). Reversed-phase high-performance
liquid chromatography (HPLC) is the final preparation method of choice for Edman
degradation, as the solvents are completely volatile.
2. Materials
2.1. Sample Preparation
1. Reduction and alkylation buffer 10× concentrate: 1 M Tris-HCl, 0.1 M EDTA pH 8.5.
2. Saturated guanidine HCl (this can be purchased as an 8 M solution in 100-mL bottles
from Sigma or Pierce).
3. Dithiothreitol; keep refrigerated and preferably desiccated and under nitrogen.
4. Chromatography columns (anion-exchange, Mono-Q; large-pore size-exclusion, Superose
6, Sephacryl 400-HR, or Sepharose 4B).
5. It may be useful to perform size-exclusion chromatography in chaotropic conditions (4 M
guanidine HCl or 6 M urea). Use high-quality reagents for this—Sigma Ultra grade, for
example—and filter through a 0.2-µm filter.
6. Dialysis membrane with a molecular-weight cutoff of 12,000 or desalting columns (PD-10,
Amersham Pharmacia Biotech).
7. Iodoacetic acid or iodoacetamide should be used fresh and kept in the dark. Both reagents
should be essentially colorless crystalline salts. Free iodine released by these reagents will
derivatize tyrosine and can be detected by smell and color.
8. Savant Speedvac.
9. Waterbath/heating block capable of reaching 54º C.
10. Eppendorf tubes (avoid glass and polystyrene containers for dilute solutions of proteins).
11. For Western blotting, a high-capacity form of Immobilon, such as Immobilon-P (Millipore)
should be used.
2.2. Enzymes
2.2.1. Glycosidases
Depending on the particular problem being addressed, it may be useful to remove
glycosaminoglycan chains and/or N-linked and O-linked oligosaccharides. Note
that carrier protein (usually bovine serum albumin) is often added to these enzymes,
so removing glycosaminoglycan chains before fragmentation by proteases may cost
more and be more trouble than it is worth. However, if N-terminal analysis from a
blotted sample, or proteolytic mapping using in-gel digestion are desired, then
removal of glycosaminoglycan chains may be useful (see Note 1). Detailed descrip-
tions are inappropriate here, as they will depend on the particular problem being
addressed.
1. Heparatinase, chondroitinase, keratanase (protease free), available from Seikagaku America.
Note that all of these enzymes will leave “stubs,” the undigested remnant of the glycosami-
noglycan chains, attached to the protein core, and so, upon analysis, gaps will appear in the
amino acid sequence.
70 Neame
2. N-glycosidase F (removes N-linked oligosaccharides to leave aspartate in the pro-
tein), O-glycosidase (cleaves between N-acetyl galactosamine and serine/threonine), both
available from Roche Molecular Biochemicals.
2.2.2. Proteases (see Note 2)
1. Sequencing-grade trypsin (Boehringer Mannheim or Calbiochem): dissolve to obtain a
concentration of 1 mg/mL in 0.1% trifluoroacetic acid. This can be stored for up to a week
at 4ºC and may be stored longer if frozen. Trypsin cleaves at the C-terminal of lysine and
arginine. If the following amino acid is proline, cleavage does not usually occur.
2. Endoprotease Lys-C (Boehringer Mannheim): dissolve in water to obtain a concentration of
1 mg/ml. Use immediately. Endoprotease Lys-C has the same characteristics as trypsin, but
does not cleave at arginine.
3. Proteolytic digestion buffer for both enzymes: 50 mM Tris-HCl, pH 8.0.
2.3. Postdigestion Size-Exclusion Separation
This is the first stage of a two-dimensional separation of digestion products. The
objective is to separate peptides by size into those that are easy to resolve by reversed-
phase chromatography (< 2.5 kDa, <20 amino acids) and those that are difficult to
separate (typically glycopeptides or peptides with attached glycosaminoglycan chains).
1. A high-resolution, low-column-volume chromatography system is strongly recommended
to avoid losses. The Amersham Pharmacia Biotech FPLC system is particularly suitable,
but other HPLC systems can also be used. A UV monitor set to 280 nm is required.
2. Columns: Superose 6 or 12 (Amersham Pharmacia Biotech) are ideal, but other columns
with similar porosity could be used.
3. Running buffer: 4 M guanidine HCl in 20 mM phosphate, pH 6.5 (see Note 3).
2.4. HPLC
2.4.1. Hardware
It is assumed that an HPLC is available (note that while a fast protein liquid chro-
matography (FPLC) can separate simple mixtures of peptides by reversed-phase
HPLC, it usually has too high a dead volume for high-resolution applications); if not,
most protein analysis core facilities will perform peptide separations. HPLC eluant
should be monitored at 220 nm; a lower wavelength will give greater sensitivity but at
the cost of increased baseline drift. A low-volume flow cell (< 5 µl internal volume) is
necessary for use of 2.1-mm-internal diameter-columns. While most HPLCs have inte-
grators attached, a chart recorder is quite adequate. It is important that there be some
way of monitoring the absorbance in real time, either via a display on the monitor, a
chart recorder, or an integrator display that updates rapidly.
2.4.2. Column
Highest sensitivity will be obtained with a column with an internal diameter of 2.1
mm or less. Flow rates in the range of 200–250 µL/min are suitable. Column length is
not particularly important, although longer columns (25 cm) will give better resolu-
tion. Optimal column life will be obtained with C18 (octadecylsilane, or ODS col-
umns). A pore size of 300 Å is necessary to separate molecules of the size of peptides.
Preparation for Amino Acid Sequence Analysis 71
As a general rule, the smaller the particle size, the better is the separation. Vydac
columns with 5-µm particles give good separations, but other column manufacturers
also make suitable columns. The column should be stored in 50% methanol in water.
2.4.3. Solvents
Solvent A: 1.54 g of trifluoroacetic acid (this must be UV-transparent) in 1 L of
HPLC-grade water (0.1% TFA v/v). It is not possible to measure TFA accurately by
pipetting. Caution: trifluoroacetic acid is corrosive and causes severe burns on contact
with skin.
Solvent B: 1.31 g of trifluoroacetic acid in 300 mL of HPLC-grade water and 700
mL of HPLC-grade UV-transparent acetonitrile (0.85% TFA v/v in 70% aqueous
acetonitrile v/v). Caution: acetonitrile may cause liver damage with chronic exposure.
Solvents will remain usable for 2–3 wk.
3. Methods
3.1. Sample Preparation
3.1.1. Preparation for Direct N-Terminal Analysis
Direct N-terminal analysis is not a particularly helpful method for identification of
a proteoglycan. The hydrated glycosaminoglycan chain(s) generally result in high
backgrounds, due to nonspecific peptide bond cleavage. However, it is the only
method suitable for defining the N-terminus and has proved to be valuable for identi-
fying, for example, the N-termini of the mature forms of the leucine-rich proteoglycans
decorin and biglycan (7) (not obvious from their cDNA sequences) and for identifying
processing sites in aggrecan (8).
Methods for proteoglycan purification that are described in other chapters of this
volume should be followed. To get clear sequence data, it is essential to pay attention
to the following.
1. Water quality. Very high quality water is essential for all of the later stages of purification.
2. Urea. Some grades of urea can be contaminated with cyanates. If present, cyanates will
block the N-terminal by forming a carbamylate. Urea can be purified by passing it down an
ion-exchange column.
3. Sample concentration. The sample must be in a small volume, and therefore at a high
concentration. Typical amounts that are appropriate for application to a sequencer are in the
region of 10 pmol protein in a volume not greater than 100 µl. For a proteoglycan with a
molecular weight of 250,000 (250 µg = 1 nmol), this corresponds to a concentration of
25 µg/mL. The highest concentration of proteoglycan will be found at the apex of peaks
when they are eluted from chromatography columns. This will also be likely to be the region
where they are at their highest purity. As a general rule, pooling fractions will reduce
concentration and add contaminants to the sample and should be avoided.
4. Sample quantitation. By far the best method for defining the amount of a protein is amino
acid analysis. Proteoglycans may also be quantitated by the dimethylmethylene blue
(DMMB) assay (9) if the percentage of glycosaminoglycan is known. Quantitation of
proteoglycans is somewhat empirical if dry weight cannot be obtained or the size of the core
protein is unknown.
72 Neame
5. Sample contaminants. Because Edman degradation relies on repetitive transitions from
basic conditions (during coupling of phenylisothiocyanate to primary amines on the pro-
tein) to highly acidic condition (during cleavage of the derivatized N-terminal amino
acid), it will not work in the presence of buffers. For this reason, reversed-phase HPLC is
the method of choice for final isolation. Unfortunately, it is inappropriate for intact
proteoglycans. Methods that are suitable for intact proteoglycans are Western blotting
onto a polyvinyldifluoridene membrane (such as Immobilon, sold by Millipore) or exten-
sive dialysis against water. Buffer exchange into water on a Sephadex G25 column (for
example, Amersham Pharmacia Biotech PD 10 columns) is an option, but there is the risk
of the proteoglycan precipitating in the column. Quantitate the proteoglycan after desalting.
A method of identifying the location of proteoglycan on Western blots that we have
found useful for N-terminal analysis is to perform a blot onto Immobilon P using
conventional methods. A thin ribbon of membrane is then removed from the center of
the lane(s) on the membrane and is blocked with casein and developed using a
proteoglycan-specific antibody. This will then precisely identify the location of the
proteoglycan on the remainder of the membrane. The unstained proteoglycan-con-
taining region can then be cut out and used directly for sequencing.
3.1.2. Reduction and Alkylation
The objective is to convert cystine (and cysteine) into a stable, nonoxidizable
derivative. Reduction and alkylation also enables proteins to be digested more com-
pletely by opening globular domains and removes the possibility that peptides held
together by disulfide bonds might be sequenced (which will confusingly give two or
more N-termini).
If the proteoglycan does not have globular domains, this stage can be eliminated.
Minimal losses will occur if the proteoglycan is exchanged into digestion buffer
using a desalting column as described above. It is possible to digest samples on blots
or in gels. See Subheading 3.1.1. and Note 1.
1. Dissolve proteoglycan in the smallest volume possible of water or low-concentration
(<20 mM) buffer.
2. Add 1 volume of 8 M guanidine-HCl (if the proteoglycan has been collected from a size-
exclusion column run in chaotrope, then it may already be in guanidine HCl).
3. Add 0.2 volume reduction and alkylation buffer to give a final concentration of 100 mM
Tris-HCl, pH 8.5.
4. Make up a fresh 100-mM solution of dithiothreitol in water (15.3 mg/mL) and add
enough to the proteoglycan solution to bring the concentration to 2 mM (1/50th volume
from step 3).
5. Briefly degas on a Savant Speedvac, if available. If not, minimize the amount of headspace
by choosing a small container. To degas, start the Speedvac spinning, and once it has come
up to speed, apply a vacuum for 5 min. Release the vacuum and, when air ceases to bleed
into the chamber, stop the Speedvac.
6. Gently blow nitrogen into the headspace of the tube to remove the majority of the air.
Incubate under nitrogen for 45 min at 54ºC.
7. Cool to room temperature and add freshly dissolved iodoacetic acid or iodoacetamide to
bring the final concentration to 4.4 mM. It is most convenient to do this by making a 50 ×
concentrate (40.9 mg/mL)
Preparation for Amino Acid Sequence Analysis 73
8. Incubate for 20 min in the dark.
9. Desalt into digestion buffer using a G25 column or a PD10 column. Desalting by dialysis is
also possible (although not recommended), and should be performed in the dark. When
using Amersham Pharmacia Biotech’s desalting columns, use the following protocol:
a. Equilibrate with 5 column volumes in the buffer of choice (for example, digestion buffer).
b. Add sample in <0.2 column volumes. Allow to run into column. For a PD-10 column,
up to 1.5 mL can be effectively desalted.
c. Wash sample into column with equilibration buffer until (sample volume + buffer
volume) = 0.4 column volumes. For a PD-10 column, this will be 2.5 mL.
d. Elute protein with 2 X sample volumes of buffer.
It is useful to collect 0.5-mL fractions and measure the absorbance of the eluate at 280 nm;
a very rough approximation is that a 1-mg/mL solution will have an absorbance of 1 at 280
nm. However, this value depends on the numbers of tyrosine and tryptophan residues in the
protein and varies considerably.
3.2. Digestion
It is important to maintain an enzyme:substrate ratio high enough to ensure that
complete digestion is obtained but low enough to ensure that background from the
enzyme is immaterial. For high levels of substrate, a ratio of approximately 1:40 is
appropriate. For low levels of substrate, 1:25 will be more useful. It is useful to per-
form a blank digest without substrate that can be used to obtain a baseline.
1. Prepare sample (your sample will either be in solution, will be on a PVDF membrane, or will
be in a gel slice). For proteoglycans, a sample in solution will be assumed. Methods for diges-
tion of blotted proteins or samples in gels have been published elsewhere in this series (3,4).
2. Dry weight is the best method for substrate quantitation but is not recommended, as it can
be difficult to redissolve the sample. Absorbance at 280 nm will give a very rough
approximation; assume that 1 mg/mL solution will have an absorbance of 1.
3. Add enzyme at a molar ratio between 1:25 and 1:40. For example, assume that the
molecular weight of the enzyme is 25,000 (approximately correct for both endoprotease
Lys-C and trypsin); 1 nmol would be 25 µg. If the proteoglycan substrate has a molecular
weight of 100,000, then 1 nmol would be 100 µg of core protein and 1 µg of enzyme
would be appropriate. Avoid digesting less than 10 µg; losses will likely become exces-
sive. A recommended minimum is 50 µg.
3.3. Postdigestion Size-Exclusion Separation
For most proteins, peptides that derive from proteolytic digestion can be separated
by direct application of the digest to a reversed-phase HPLC column (see Subheading
3.4.). Proteoglycans, by virtue of the fact that they have large glycosaminoglycan
chains attached to them, require an additional step. Glycosaminoglycan-containing
peptides and proteins can cause deterioration of reversed-phase columns, but their
most annoying characteristic is that they elute as extremely broad peaks that will give
a background to many of the other peaks. It is therefore worthwhile to insert a size-
exclusion chromatography step into the protocol. This has the added advantage that
the position of the glycosaminoglycan chain can readily be deduced by identifying the
glycosaminoglycan-containing peptide(s) by Edman degradation. A further advantage
is that peptides are separated into pools that have similar molecular weights and
are more easily characterized. A good example of this can be found in the characteriza-
74 Neame
tion of a glycosaminoglycan-containing peptide from epiphycan, in which the peptide
elutes as several broad peaks spread out along the entire acetonitrile gradient (10).
The following protocol is appropriate for a Superose 12 or Superose 6 column on
an Amersham Pharmacia Biotech FPLC system. It is useful to keep 10% of the digest
for analysis on a reversed phase column (see Subheading 3.4.).
1. Equilibrate column in 4 M guanidine-containing buffer at a flow rate of 0.5 ml/min.
2. Set UV monitor to 280 nm.
3. The chart recorder or integrator should be set up so that full-scale deflection corresponds to
an absorbance of 0.02.
3. Load sample (typically 0.5 mL for a Superose 6 column) into loop.
4. Inject.
5. Collect fractions (0.5 mL). The void volume will appear at approximately 7 mL and the
VT will be at approximately 18 mL.
6. The UV trace is only a guide; do not pool peaks based on what you can see, as you will
only be detecting peptides that contain tryptophan and tyrosine.
7. Fractions can be concentrated up to twofold on a Speedvac and injected directly onto a
reversed-phase column (see Note 4).
8. Fractions containing glycosaminoglycan-substituted peptides can be detected using the
DMMB assay (9) (see Note 5).
Because the column is running in guanidine-HCl, proteins and peptides will have
approximately twice the hydrodynamic volume that they would have under “native”
conditions. The exclusion molecular weight for a Superose 6 column is thus approxi-
mately 1,000,000. It will separate peptides as small as 1 kDa from those of 2 kDa.
3.4. HPLC
It will be necessary to calculate an approximate lag time between the flow cell and
the exit point of tubing where you will be collecting peaks. This will be (60 × (internal
volume of the tube coming out of the flow cell + half the volume of the flow cell) / flow
rate) seconds (see Subheading 2.4.).
1. Prepare solvents (If helium sparging is used to remove dissolved gases, perform this before
addition of trifluoroacetic acid. In practice, helium sparging is not generally necessary.
2. Purge pumps.
3. If sample pH is > 7, acidify sample(s) by addition of glacial acetic acid (0.05 volume). This
is necessary to avoid damaging silica-based columns.
4. Equilibrate column in solvent A and run a blank chromatogram. This would also be a useful
time to run approximately 10% of the original digest to determine where peaks are likely to be.
5. Load digested proteoglycan into sample loop (100–200 µL).
6. Inject and start a gradient of 0–100% solvent B in 45–90 min.
7. Collect peaks into Eppendorf tubes by hand. Note that there is likely to be a very large void
volume peak. There may occasionally be peptides here, but this is rare. Most peptides will
not elute before 8 min.
When collecting peaks, start collecting after the UV absorbence has left the baseline and
add the lag time that you have previously calculated. Stop collecting at the point where the
UV trace is approximately halfway back to the baseline, plus the lag time. There are frac-
tion collectors that can do some of this process automatically, but it is much more reliable
to do it by hand and label the tubes based on the peak elution time (see Fig. 1).