Idea Transcript
SEDIMENTATION, DIFFUSION, AND MOLECULAR WEIGHT OF A MUCOPROTEIN FROM HUMAN PLASMA* BY
EMIL
L. SMITH,
DOUGLAS M, BROWN, HENRY RICHARD J. WINZLER
E. WEIMER,
AND
(From the Laboratory for the Study of Hereditary and Metabolic Disorders, and the Departments of Biological Chemistry and Medicine, University of Utah College of Medicine, Salt Lake City, and the Department of Biochemistry and Nutrition, University of Southern California School of Medicine, Los Angeles)
(Received for publication,
February
27; 1959)
Behavior in Ultracentrifuge Sedimentation and diffusion studies were carried out with an electrophoretically homogeneous mucoprotein, having the chemical composition previously reported (4), isolated by fractionation with ammonium sulfate from pooled, normal, human plasma. The sedimentation studies were made in the Spinco’ electrically driven ultracentrifuge with the controls and procedures previously described (5). All of the runs were performed at 59,780 r.p.m., equivalent to centrifugal fields of approximately 240,000 X g and 300,000 X g at the meniscusand base. In Fig. 1, Philpot-Svensson diagrams of the sedimenting mucoprotein are shown as observed in representative runs at different pH values. It is apparent that the substance behaves as a molecularly monodispersesystem under these conditions. In fact, at no time was any evidence found for the presence of other sedimenting boundaries. Individual determinations of the sedimentation constants are shown in Fig. 2. The values are given in Svedberg units (S = 1 X lo-la), * This investigation was aided by grants from the United States Public Health Service and from the Committeeon Growth of the National ResearchCouncil acting for the American Cancer Society. 1 Specialized Instrument Corporation, Belmont, California. 569
Downloaded from http://www.jbc.org/ by guest on May 6, 2019
Investigations by some of us (l-3) have been concerned with the demonstration by electrophoresis of mucoproteins in normal and pathological sera. It was found that at least three electrophoretic components were present in crude mucoprotein fractions. More recently, the isolation in electrophoretically homogeneous form of the major mucoprotein component of normal human plasma has been described, and chemical and electrophoretic characterization of this substance was presented (4). In this communication, further studies of the physical properties of this mucoprotein will be reported.
570
HUMAN
PLASMA
MUCOPROTEIN
and these are corrected for the density and viscosity of the medium (6). The temperature of each run was taken as the average for the time when photographs were made, and the results were then corrected to 20”. It is quite clear that the protein concentration has a significant effect on However, this effect s~O,~, indicating a somewhat asymmetrical molecule
A bJl 4A B
4 D
J
LJM
of a mucoprotein from huFIG. 1. Sedimentation behavior .i o the ultracentrifuge man plasma. The arrow shows the direct,ion of radial migration. A, results at pH 4.1 (0.1 M acetate + 0.1 M sodium chloride) ; first picture 36 minutes after attaining full speed, subseqrent exposures at 32 minute intervals; protein concentration 0.92 per cent. B, the solvent was 0.15 M sodium chloride (pH 5.4) ; protein concentration 1.0 per cent; exposures 30 minutes after reaching full speed and at 16 minute intervals. C, the solvent was 0.1 M Verona1 + 0.4 M sodium chloride at pH 8.4; protein concentration 1.43 per cent; exposures 58 minutes after reaching full speed and at 32 minute intervals. D, results in 0.1 M glycine buffer at pH 9.5; protein concentration 1.14 per cent; exposures 38 minutes after reaching full speed and at 32 minute intervals. E, in 0.1 M glycine buffer at pH 10.5; protein concentration 0.50 per cent; exposures 32 minutes after reaching full speed and at 16 minute intervals.
seemsto depend on the nature .of the solvent. There appears to be some influence of ionic strength at the higher protein concentrations. A single determination (not plotted in Fig. 2) in 0.1 M Verona1 at pH 8.4 gave a value of s20,W= 2.68 S at a protein concentration of 1.5 per cent. At an ionic strength of 0.5, 820,~= 2.94 S. This effect of ionic strength was not observed at protein concentrations below 1 per cent.
Downloaded from http://www.jbc.org/ by guest on May 6, 2019
C
SMITH,
BROWN,
WEIMER,
AND
WINZLER
571
Extrapolated values for sero protein concentration gave ~20.~ = 3.00 S at pH 4.1, 3.16 S at pH 5.6, 3.06 S at pH 8.4, and 3.20 S at pH 9.5. The two determinations at pH 10.5 gave a somewhat higher value, but no significance can be attributed to these measurements at the present time,
pH
10.5
cz 2.4 z 3.2 5 3.0 5 2.8 f 3.2 g 3.0 2 2.8 3.0 2.8 2.6 2.4
0
0.5 Protein
1.0
1.5
Concentration
20
in per cent
of the sedimentation constant (s~.,J in Svedberg units as a funcFIG. 2. Variation tion of the protein concentration as determined at different pH values. The measurements at pH 4.1 were made in an acetate buffer which contained 0.1 M acetate and 0.1 M sodium chloride. Runs were made in 0.1 M Verona1 buffer at pH 8.4 (o), and in the same buffer supplemented with 0.4 M sodium chloride (W). The solvents at pH 9.5 and 10.5 were 0.1 M glycine buffers.
The average of the extrapolated 3.11 s.
values at the four lowest pH values is
Diflusion
Studies
The electrophoresis cell was used for the measurement of the diffusion constants from photographs taken by the Schlieren scanning method by
Downloaded from http://www.jbc.org/ by guest on May 6, 2019
gL.0
572
HUMAN
PLASMA
MUCOPROTEIN
TABLE
Di$usion Protein
concentration per
cent
1
Constants 0s Plasma Mucoprotein %S’,o
x
10’
sp. cm. per sec.
0.98
2.895
0.98 0.39 0.39
2.82 2.79 2.79
~zo,w
x
10’
sp. cm. )W sec.
5.40 5.27 5.21 5.21
Partial Spec$c Volume These measurementswere performed in 10 cc. pycnometers with a preparation of the mucoprotein that was about 95 per cent homogeneouselectrophoretically at pH 4.0. Values of the partial specific volume ranging from 0.671 for a protein concentration of 1.275 per cent to 0.678 for a concentration of 6.380 per cent were obtained by the formula 110 given in Svedberg and Pedersen (6). The average value of 0.675 for the apparent partial specific volume (V) of this mucoprotein is somewhat lower than that previously found for many proteins. However, it is very close to the value of 0.685 found by Fredericq and Deutsch (8) for ovomucoid, a protein which contains proportions of carbohydrate somewhat similar to those of the plasma mucoprotein (4). Molecular Weijht and Frictional Coeficient Using the usual formula for the molecular weight (6), where M = RT.s/D(l - VP), and substituting the values found for s, D, and V, we obtain 44,100.
Downloaded from http://www.jbc.org/ by guest on May 6, 2019
the method described by Longsworth (7). The results were computed by the formula D = A2/(4ntH2) where A is *the area under the curve, H the maximum height, t the time in seconds, and D the diffusion constant in sq. cm. per second. Two runs were performed in Verona1 buffer of 0.1 ionic strength at pH 8.5. The measurements were made at 1.5” and corrected for the difference in viscosity and temperature in the usual manner (7). Each run was performed in duplicate, both halves of the cell being used separately. Six or seven photographs were taken at intervals of 8 to 70 hours and the values averaged. The data are given in Table I. The average diffusion constant found for the mucoprotein is 5.27 f 0.06 X lo-’ sq. cm. per second. It is apparent that, at the two values studied, the protein concentrat,ion has no detectable effect on the diffusion const~ant.
SMITH,
BROWN,
WEIMER,
AND
573
WINZLER
The frictional coefficient, f/jo, was calculated to be 1.78 from the sedimentation and diffusion constants (6). The maximal value of the axial ratio is about 15 for an elongated ellipsoid, or 20 for a flattened ellipsoid. viscosity
studies
TABLE Viscosity Concentration
Studier
i
II
of Plasma
Mucoprotein
Relative viscosity, n,
- -. -
H
grit. per loo cc.. 6.38 3.86 3.00 2.18 0 __-.
1.605 1.318 1.237 1.167 1.000 __._.~
0.4731 0.2761 0.2127 0.1544
I
O
0.0742 0.0715 0.0709 0.0708 (0.069)
increment, Y, calculated from the intrinsic viscosity Ho and the partial specific volume V, by the equation v = 100 Ho/VI, was found to be 10.2. This corresponds to an axial ratio of about 8 for an elongated ellipsoid or approximately 13 for a flattened ellipsoid. The agreement between the axial ratios obtained from viscosity studies and those obtained from diifusion and sedimentation is not particularly good. If the calculations are made on the basis of a hydrated protein containing 0.2 gm. of water per gm. of protein, the discrepancy becomes somewhat less significant, especially for a flattened ellipsoid, the values being 10 from viscosity measurements and 15 from diffusion measurements (flfO = 1.63). DISCUSSION
It has previously been pointed out (4) that this mucoprotein does not appear to bear any easily recognized relationship to the well known protein constituents of human plasma. The high content of carbohydrate, the low isoelectric point (pH l.S), the amino acid composition, and the
Downloaded from http://www.jbc.org/ by guest on May 6, 2019
Viscosity measurements were carried out on a preparation of the mucoprotein which was 90 per cent homogeneous by electrophoresis at pH 2.7, by means of an Ostwald viscosimeter mounted in a water bath at 25”. The results of this study are recorded in Table II. The values were extrapolated by means of the function H = (In n/&/C, H being plotted versus 6. The intrinsic viscosity, Ho, was taken as the extrapolated value of H at C = 0. Here n is the viscosity of the protein solution, no the viscosity of the solvent (double distilled water), and C is the mucoprotein concentration in gm. per 100 cc. of solution. The value of the viscosity
574
HUMAN
PLASMA
MUCOPROTEIN
presence of labile sulfate have all served to distinguish this protein from other plasma proteins. The present studies support the earlier conclusions.2 The sedimentation and diffusion constants of the mucoprotein are quit,e different from those found for other constituents of human plasma (9). The molecular weight of 44,100 establishes this protein as one of the smallest proteins present in human plasma. Of considerable interest is the low partial specific volume (0.675) of the mucoprotein. It is suggested that this low value arises from the high carbohydrate content, since similar results have been found with ovomucoid (8). It is of some interest to compare the molecular weight of the mucoprotein as determined by physical measurements with that computed from the amino acid analysis of Weimer, Mehl, and Winzler (4). Six of the TABLE III Weight of Mucoprotein from Amino Acid Content Amino acid
Content y$ytm . --
.
-
%z
Mol.
wt.
-__ per ccn;
Cystine........................................
Glycine........................................ Histidine....................................... Methionine..................................... Tryptophan.................................... Tyrosine....................................... __-Average......................................
0.59
0.80 1.28 0.64 1.22 1.95 ~~--
* This low value suggests that some cystine is destroyed
40,700 9,380 12,106 23,300 16,700 9,290
1 5 4 2 3 5
40,700* 46,900 48,400 46,600 50,100 46,450 46,509
during hydrolysis.
amino acids are present in small enough amounts to permit such caiculations. The data in Table III are calculated from the amino acid figures given for the anhydrous substance; these lead to an average molecular weight of 46,500 which is fairly satisfactory. As already pointed out (4), methionine and cystine do not account for all of the sulfur, 1.02 per cent, which is present in the mucoprotein. Cal2 It was hoped that further characterization of themucoprotein could be achieved by immunochemical methods. Preliminary experiments have been made by Dr. B. V. Jager of the University of Utah. Attempts were made to stimulate antibody production by repeated injection into rabbits and chickens of alum-precipitated preparations of the mucoprotein. This was tried at 50 to 100 mg. of protein per kilo of body weight. Thus far, none of these efforts has been successful. However, these negative experiments do serve to distinguish the poorly antigenic mucoprotein from other human plasma proteins which are highly antigenic. Antisera to human yglobulin and to human albumin gave no precipitates with the mucoproteirr..
Downloaded from http://www.jbc.org/ by guest on May 6, 2019
Molecular
SMITH,
BROWN,
WEIMER,
AND
WINZLER
575
culation gives 14 or 15 sulfur atoms, while only 4 can be present in the cysteine and methionine residues. It is apparent that 10 or 11 sulfur at,oms are present in another form, probably as labile sulfate esters. SUMMARY
Examination of a purified mucoprotein from human plasma showed that it behaved as a homogeneous substance in the ultracentrifuge over the range of pH 4.1 to 10.5. The average sedimentation constant (~20,~) is 3.11 S, as found by extrapolation to zero protein concentration. The average diffusion constant is 5.27 X lo-’ sq. cm. per second. The apparent partial specific volume is 0.675. The intrinsic viscosity is 0.069. The computed molecular weight is 44,100, and the frictional coefficient is 1.78. Downloaded from http://www.jbc.org/ by guest on May 6, 2019
BIBLIOGRAPHY
1. Winaler, R. J., Devor, A. W., Mehl, J. W., and Smyth, I. M., J. Clin. Invest., 27, 609 (1948). 2. Mehl, J. W., Humphrey, J., and Winzler, R. J., Proc. Sot. Exp. Biol. and Med., 72, 106 (1949). 3. Mehl, J. W., Golden, F., and Winzler, R. J., Proc. Sot. Exp. Biol. nnd Med., 72, 110 (1949). 4. Weimer, H. E., Mehl, J. W., and Winzler, R. J., J. Biol. Chem., 186, 561 (1950). 5. Smith, E. L., Brown, D. M., Fishman, J. B., and Wilhelmi, A. E., J. Biol. Chem., 177, 305 (1949). 6. Svedberg, T., and Pedersen, K. O., The ultracentrifuge, Oxford (1940). 7. Longsworth, L. G., Ann. New York Acad. SC., 41, 267 (1941). 8. Fredericq, E., and Deutsch, H. F., J. Biol. Chem., 181, 499 (1949). 9. Oncley, J. L., Scatchard, G., and Brown, A., J. Phys. and Colloid Chem., 61, 184 (1947).
SEDIMENTATION, DIFFUSION, AND MOLECULAR WEIGHT OF A MUCOPROTEIN FROM HUMAN PLASMA Emil L. Smith, Douglas M. Brown, Henry E. Weimer and Richard J. Winzler J. Biol. Chem. 1950, 185:569-575.
Alerts: • When this article is cited • When a correction for this article is posted Click here to choose from all of JBC's e-mail alerts This article cites 0 references, 0 of which can be accessed free at http://www.jbc.org/content/185/2/569.citation.full.h tml#ref-list-1
Downloaded from http://www.jbc.org/ by guest on May 6, 2019
Access the most updated version of this article at http://www.jbc.org/content/185/2/569.citation