Serum protein concentration and portal pressure determine the ascitic fluid protein concentration in patients with chronic liver disease.
The ascitic fluid and serum concentrations of albumin and globulin were measured simultaneously with transhepatic portal pressure determination in 56 patients with chronic liver disease to determine whether (1) portal pressure correlated with (S-Asc)A and (2) the majority of variation in ascitic fluid protein concentration between patients was related to fluid balance from serum to ascites. The mean ascitic fluid albumin concentration was 1.04 +/- 0.73 gm/dl; globulin concentration 1.31 +/- 0.80 gm/dl; and ascitic fluid total protein concentration 2.35 +/- 1.49 gm/dl. The mean serum albumin concentration was 2.58 +/- .57 gm/dl; globulin concentration 3.91 +/- .86 gm/dl; and total protein concentration 6.49 +/- 1.30 gm/dl. The (S-Asc)A was 1.54 +/- .45 gm/dl. The mean PPIVC was 14.5 +/- 4.3 mm Hg. The (S-Asc)A correlated directly with PPIVC (r = 0.73; p less than 0.0001). The ascitic fluid protein correlated with three variables that did not correlate with each other: serum albumin (r = 0.67; p less than 0.0001), serum globulin (r = 0.44; p less than 0.001), and PPIVC (r = -0.48; p less than 0.0005). The sum of the squared correlation coefficients with these latter uncorrelated variables equaled 0.87 and partial correlation coefficient analyses demonstrated an increase in the correlation of the ascitic fluid protein with the serum albumin concentration when corrected for serum globulin and (S-Asc)A (r = 0.97; p less than 0.0001) or PPIVC (r = 0.90; p less than 0.0001). Thus most of the variation in ascitic fluid protein between patients in this study could be related to serum protein concentrations and PPIVC or (S-Asc)A. Furthermore, multivariate discriminant analysis of patients with an ascitic fluid protein less than or equal to 2.5 vs. greater than 2.5 gm/dl indicated that the majority of differences between the two groups could be attributed to differences in serum albumin and serum globulin in combination with the (S-Asc)A (canonical correlation = 0.808) or PPIVC (canonical correlation = 0.806). These factors could correctly identify the low or high ascitic fluid protein groups in 96% and 93% of patients, respectively. In conclusion: (1) the (S-Asc)A is associated with the degree of portal pressure elevation and (2) the majority of variation in ascitic fluid protein concentration between patients with chronic liver disease is associated with differences in portal pressure and serum protein concentrations.
Effect of protein concentration on the determination of digoxin in serum by fluorescence polarization immunoassay
WH Porter, VM Haver and BA Bush
Determination of digoxin by fluorescence polarization immunoassay (FPIA) with the Abbott "TDx" is significantly influenced by the concentration of total serum protein. Each 10 g/L increase in serum protein results in an 8% decrease in measured digoxin. Studies with [3H]digoxin confirmed that digoxin binds to the protein pellet during the trichloroacetic acid precipitation step before the immunoassay. Serum protein, or equal concentrations of albumin or gamma-globulin, exert an equivalent effect on the apparent digoxin value. Because the total protein concentration of the assay calibrators is low (50 g/L) compared with its reference interval in serum (60-80 g/L), results by FPIA may be expected to be low by an average of 16% (range, 8-24%). Digoxin results by FPIA will be most nearly accurate when the calibrators include a total protein concentration of about 70 g/L. Patients' specimens with abnormally high or low protein content will give falsely high or low results for digoxin.
Serum proteins as health and nutrition indicators
The protein composition of blood responds over the time period of days to weeks to changes in diet as well as inflammatory and infectious disease. Plasma can be similarly analyzed.
Total serum protein is the sum of the individual fractions measured separately (discussed below) as well as the collection of a diversity of proteins found in relatively low concentration. The total serum protein concentration is used in a correlation analysis with albumin, IgG and acute phase proteins. If serum protein concentration is elevated but uncorrelated with an specific fraction then it idicates that the individual is dehydrated from water stress. Typically, total serum protein concentration is correlated with increased concentrations of albumin, IgG and/or acute phase proteins.
[Measured in 2 ul of serum using Coomassie R-250 dye binding and spectrophotometry]
Albumin is the most abundant protein in serum and is synthesized by the liver to maintain homostatic levels when protein intake is sufficient. Decreased albumin concentration, which is typically correlated with decreased total protein concentration, indicates protein malnutrition. Depending on the composition of the diet, decreased albumin may indicate energy malnutrition whereby proteins are catabolized for gluconeogenesis.
[Measured in 3 ul of serum by bromcresol-green dye binding and spectrophotometry]
Immunoglobulin G (IgG) is the second most abundant protein in plasma and is synthesized by B-lymphocytes in response to the sum of immunogenic stimulation. The levels of IgG are decreased by (protein) malnutrition and elevated by infectious disease.
Acute phase proteins (APP) are an ensemble of plasma proteins synthesized by the liver in response to tissue damage and inflammation associated with traumatic and/or infectious disease. Transferrin (Tf) and fibrinogen (Fb) are the most abundant acute proteins that can be accurately measured. There are additional proteins that become detectable when transferrin and fibrinogen levels are increased. Increased IgG and acute phase proteins indicate an infection with concurrent tissue damage. Increased IgG without increased fibrinogen indicates immunological response to infection without concomitant decrease in tissue function.
[IgG and APP are measured in 5 ul of serum using SDS-PAGE]
An example of SDS-PAGE of serum proteins from waved albatrosses (Phoebastria irrorata) Nazca boobies (Sula dactylatra) and common terns (Sterna hirundo). IgG concontration standards (2-10 ug) are used to construct a standard curve after densitometric scanning of the gel.
-----------------------------------------------------------------------------------------------
INFLUENCE OF PROTEIN CONCENTRATION UPON
ELECTROPHORETIC MOBILITY OF
SERUM PROTEINS*
BY RICHARD W. LIPPMANt AND JAY BANOVITZI
(From the Institute for Medical Research, Cedars of Lebanon Hospital, Los Angeles,
and the Gates and Crellin Laboratories of Chemistry,5 California
Institute of Technology, Pasadena, California)
(Received for publication, June 24, 1952)
Previous investigators have reported a change in distribution of serum
protein fractions, as measured by the areas subtending the conventional
electrophoresis curves, as a result of a change in ionic strength of the buffer
(1) or as a result of a change in total protein concentration (1, 2). A
similar effect of total protein concentration on the electrophoretic analysis
of ovalbumin has been shown by Cann (3). Although the effects of buffer
ionic strength, buffer species, and pH on mobility have been well known,
little attention has been given to the effect of protein concentration and
species on the electrophoretic mobility of proteins. Luetscher (4) suggested
that mobility might increase with protein concentration, but this
comment was not amplified. Longsworth and MacInnes (5), in a discussion
of electrophoresis patterns, presented a table which showed a decrease
in the mobility of ovalbumin with increased protein concentration.
Alexander and Johnson (6) refer to an unpublished qualitative observation
(Johnson and Shooter, 1946) that protein mobility decreased with an
increase in protein concentration. In a recent, paper on another subject,
Alberty and Marvin (7) presented a table which showed a 10 per cent decrease
in the mobility of bovine serum albumin at pH 7.0 as the protein
concentration increased from 2 to 8 per cent.
Our attention was called to this subject in the course of an experiment
with rats which had been given nephrotoxic globulin to produce experimental
nephritis. Analysis of sera from these animals suggested an apparent
relation between the mobility of serum albumin and severity of
* This work was supported in part by a grant (H-132-C3) from the National Heart
Institute of the National Institutes of Health, and in part by a grant from the Research
Trust Fund of Los Angeles. An abstract of the work was presented at the
April 15, 1952 meeting of the American Society of Biological Chemists, in New York
(13). The authors deeply appreciate the aid and suggestions of Professor Dan H.
Campbell and Dr. Jerome Vinograd.
t Fellow of the John Simon Guggenheim Memorial Foundation.
$ Present address, Department of Zoology, University of Wisconsin, Madison,
Wisconsin.
5 Contribution No. 1703.
451
Downloaded from www.jbc.org by on January 29, 2008
452 SERUM PROTEINS
the disease. However, in tabulating the results it was found that, the
sera had been handled in a routine manner and diluted to the same extent
before electrophoresis was performed. Since the concentration of total
serum protein falls with the severity of nephrotoxic globulin nephritis in
the rat, the determinations of electrophoretic mobility inadvertently had
been performed in specimens that were progressively more dilute. It,
seemed necessary to know the effect of protein concentration and species
upon the measurements of mobility before any conclusions could be
drawn, and the present experiment was undertaken to secure this information.
Methods and Materials
The purified bovine serum albumin and y-globulin1 uxx homogeneous
by electrophoresis in Verona1 buffer at pH 8.6. Normal serum was
pooled from about thirty rats of the Slonaker-Addis strain, taken from
stock, lightly anesthetized with ether, and exsanguinated from the abdominal
aorta. The production and biological characteristics of nephrotoxic
globulin (NTG) nephritis in the rat have been described in detail
(8, 9). Pooled sera from groups classified for severity of the nephritis in
previous experiments were used. All sera were rapidly frozen and kept
in the frozen state until thawed for dilution, dialysis, and determination
of the electrophoretic mobility.
The sera were dialyzed without dilution or were diluted wit)h Verona1
buffer, ionic strength 0.1, pH 8.6, before dialysis. The bovine proteins
were dialyzed after dilution with the same buffer. Dialysis against the
buffer was performed in cellophane sausage casing at 3”, either for 24
hours with gent,le stirring or for at least 3 days without stirring of thr
buffer. At the conclusion of dialysis, a portion of the solution was analyzed
for total protein concentration by a slight modification of the biuret
method of Kingsley (10). hnother portion was used to determine
the electrical resistivity of the solution by means of a Wheatstone bridge.
Electrophoretic analysis was performed, as indicated in the respclri i1.c
tables, either with open electrode vessels in a Perkin-Elmer electrophorcsis
apparatus, or with one of the electrode vessels closed in an instrument
described by Swingle (11). The protein mobilitics were calculated from
mcasuremellts of t Iic tlcscending boundary only, ncacortlillg to thr rccommendations
of Longsworth and MacInnes (5), 1)~ using t.he first murnelrt
of the gradient curve. The measured resistivity of the dialyzed protein
buffer solution was used for the mobility calculations since this differs
significantly from the resistivity of the buffer solution alone.
1 Bovine proteins were purch:mxl from t hc Arnlour Laboratories, Chicago, Illinois.
Downloaded from www.jbc.org by on January 29, 2008
R. W. LIPPMdN .4ND J. BANOVITZ 453
RESULTS .4ND DISCUSSION
The mobility of bovine serum albumin in the isolated system decreased
17 per cent as the protein concsentration increased from 0.61 to 5.95 percent,
as shown in Fig. I. \Vhc~l the mobility of 1 per cent bovine serum albumin
was determined in the presence of increasing concentrations of bovine
serum y-globulin, there was an equivalent decrease of about 14 per cent in
the mobility of the albumin as the total protein concentration rose from
1.32 per cent to 3.50 per cent, also shown in Fig. 1. In the same expcrimcnts
the mobility of the r-globulin deereased ahout, 7 per cent (SW
I % 8.S.A. + 8.G.G.
0 2.0 4.0 6.0
TOTAL PROTEIN %
FIG. 1. Relation of albumin mobility to total protein concentration of the sample
used for electrophorctic analysis. B. S. A., bovine serum albumin; B. G. G.,
bovine -y-globulin; R. S. A., rat albumin in rat strum.
Table I). Although the direction of change in albumin mobility is the
same with increasing albumin or r-globulin concentrations, Fig. 1 shows
that increasing r-globulin concentrations produce a greater decrease in
the albumin mobility than an equal weight of albumin alone. If the
curve represent.ing mobilit,y of 1 per cent albumin in the presence of -y-globulin
is extrapolated, it should join the curve for albumin alone at the 1 per
ceut level, and iuspcction of Fig. 1 indkatcs an approximation of this
expected result.
Results of experiments with normal rat serum resembled those obtained
with bovine serum albumiu. The mobility of rat serum albumin in whole
normal rat serum decreased as the total protein concentration increased.
In the open electrode system, albumin mobility diminished by 29 per
cent as the total protein concentration rose from 0.72 to 5.36 per cent
Downloaded from www.jbc.org by on January 29, 2008
454 SERUM PRCYTEINS
(Table II). I% the closed electrode system, albumin mobility diminished
by 23 per cent as the total protein concentration rose from 0.53 to 4.82 per
cent (Table III). Since the change in mobility in the open electrode
TABLE I
Bovine Serum Proteins in Closed Electrophorelic System
Mobility, sq. cm. volt-l sec.-l X 10-S
Per cent total protein
Albumin I -y-Globulin
Albumin only
__._--
0.61 7.93 ,
1.46 7.48
3.90 6.67
5.95 6.60
1% albumin + r-globulin
1.32 7.36 1.71
1.73 7.05 1.61
2.24 6.91 1.62
3.50 6.36 1.58
TABLE II
Normal Rat Serum in Open Electrophoretic System
Per cent total protein
0.72
1.09
1.75
2.15
2.68
3.61
4.50
5.36
Albumin
6.92
6.50
6.29
5.78
5.82
5.11
4.97
4.94
Mobility, sq. cm. volt- sec.-’ X 10-*
m-Globulin w-Globulin &Globulin
6.40 4.91 3.5,
5.80 4.5s 3.41
5.6, 4.30 3.28
4.7s 3.4, 2.38
4.96 3.9* 2.62
4.3a* 2.16
4.00 I 2.90 1.9s
4.02* 2.00
7-Globulin
2.7s
2.6,
2.42
1.6s
1.8~
1.52
I.29
1.46
* (Ye- + a*-Globulin, as the pattern did not resolve the peaks.
system is of similar magnitude to that in the closed system, the concentration
effect on mobility cannot be due to anomalies that may be encountered
in the open system. In these experiments, the relative proportion
of serum protein components remained constant, except for the apparently
small percentile changes in the areas subtending the electrophoresis
curves, previously commented upon, which would be produced by
Downloaded from www.jbc.org by on January 29, 2008
R. W. LIPPMAN AND J. BANOVITZ 455
dilution of serum. The other protein components, W-, QZ.-, /3-, and yglobulins,
showed an equivalent decrease in mobility as the total protein
concentration rose.
The albumin mobilities derived from analysis of nephritic rat sera were
plotted against the actual protein concentration of the sample used, to-
TABLE III
Normal Rat Serum in Closed Electrophoretic System
Per cent total protein
0.53
1.23
3.36
4.82
-
Mobility, sq. cm. volt-1 sec.-’ X 10-L
Albumin 1 a&lobulin ax-Globulin ,T-Globulin y-Globulin
6.92 5.8~ 4.8~ 3.21 1.61
6.53 5.5* 4.5, 2.8, 1.5s
5.72 4.81 3.7, 2.3, 1.26
5.37 4.40 3.50 2.2s 1.12
__I ~- I I I -
u) 7.2
t
‘0 b
X
T -0’ l o 0
z .o
T 3 6.0-- OCP
=L 0
00 l
4%
0 2.0 4.0 6.0
TOTAL PROTEIN %
Fro. 2. Relation of electrophoretic mobility of serum albumin to total protein
concentration of the sample used. l , values for normal rat serum albumin; 0,
values for serum albumin of rats with nephrotoxic globulin nephritis.
gether with data obtained from various dilutions of normal rat serum.
As can be seen in Fig. 2, the values obtained in nephritic sera do not differ
significantly from the values obtained in normal sera when the total protein
concentration of the sample used is taken into consideration.
From these results it is apparent that both the concentration of total
protein and the protein species have an important effect upon the determination
of electrophoretic mobility of bovine and rat serum proteins.
Downloaded from www.jbc.org by on January 29, 2008
456 SERUM PROTEINS
Without recognition of these factors, t.he interpret’ation of data has doubtful
significance, as in the statement that, in the rat the urinary albumin
after administrat,ion of renin has a somewhat greater mobility than normal
serum albumin (12). In ol’der to compare mobilities, in addition to the
usual factors of buffer strength, species, and pH, the protein concentration
must be held constant and the protein species must be known. It may
be suggested from theoretical considerations that the protein mobility
at infinite dilution should be constant and i~~depentlent of other protein
species present.
An explanation of the dcpendencc of protein mobility on total protein
concentration and protein species cannot be offered on the basis of OUI
experiments. Such factors as viscositjy (5), reduction of net charge through
protein-protein interactions, total ionic strength of the protein solution, as
well as protein-solvent binding, will have to bc c~onsiderctl in order to ac.
count for the concent,ration effect on mobility.
It was found that the c~lcotrophoretic: mobility on bovine serum albumin
is diminished as the total protein concentration of the sample is increased.
The magnitude of this effect is variable, depending upon the protein xpeties
present. Thus, in the presence of -y-globulin of bovine serum the
mobility of bovine serum albumin is less than in the presence of an equal
weight of protein which is all albumin. The albumin of nephritic rats has
the same mobility as the serum albumin of normal rats when determinations
are compared at the same total protein concentration. Unless these
effects of concentration and species are considered, protein mobilities
cannot be compared.
l3IBLIOGR.WNT
1. Perlmann, G. E., and Kaufman, D., J. Am. Chem. Sot., 67, 638 (1945).
2. .J:lmrson, E., Arch. Biochem., 16, 389 (1947).
3. Cann, J. R., J. Am. Chem. Sot., 71, 907 (1949).
4. l,uetscher, J. A., Jr., J. Clin. Invest., 19, 313 (1940).
5. I,ongsworth, L. G., and MacInnes, D. A., J. Am. Chem. Sot., 62, 705 (1940).
G. Alexander, A. E., and Johnson, P., Colloid science, New York, 1, 326 (1949).
7. Alberty, R. A., and Marvin, H. H., Jr., J. Am. Chem. Sot., 73, 3220 (1951).
8. Lippman, R.. W., Marti, H. U., and Campbell, D. H., Arch. Path., 63, 1 (1952).
9. T,ippnmn, It. W., Marti, H. U., and Jacobs, E. E., Arch. Path., 64. lG9 (1952).
10. Kingsley, G. R., J. Biol. Chem., 131, 197 (1939).
11. Swingle, S. &I., Rev. Scient. Instruments, 16, 128 (1947).
12. Sellers, A. L., Roberts, S., Rask, I., Smith, S., 3rd, Marmorston, J., and Goodman,
H. C., J. Exp. Med., 95, 465 (1952).
13. Lippman, R. W., and Banovitx, J., Pederntion Prnc., 11. 250 (1952).
Downloaded from www.jbc.org by on January 29, 2008
