CLINICAL CHEMISTRY, Vol. 19. No.2, (1973) 228-234
Polyacrylamide-Gel Disc-Electrophoresis as a Screening Procedure
for Serum Lipoprotein Abnormalities1
Herbert K. Naito,2 Mitsuo Wada, L. A. Ehrhart, and Lena A. Lewis
A polyacrylamide-gel
disc-electrophoresis
described
that is suitable
for screening
system
is
sera with a
wide range of lipid concentrations,
and the results
agree with analytical ultracentrifugation
data. The Iipoprotein
pattern so obtained
was compared
with
that obtained
by other recognized
electrophoretic
procedures.
Lipoprotein
determinations
were done
by polyacrylamide
gel electrophoresis,
for 125 patients, to determine
its merit as a clinical screening
method for lipoprotein abnormalities
as compared to
the paper-electrophoretic
method. The lipid dye and
riboflavin concentration
of the gels used give a background that makes it easy to detect chylomicra
in
the
sample
gel
and
allows
for densitometric
quantitation
of the lipoprotein fractions.
is reliable, precise, and accurate.
semi-
The method
Additional Keyphrases: diagnostic aid #{149}Iipoproteinemlas #{149}
comparison with other techniques of electrophoresis and with ultracentrifugation
.
a/
lipo protein ratios
PAGE,3
a recent
addition
to the many techniques
used for electrophoresis,
was first described
by Ornstein (1) and Davis (2), for use in protein electrophonesis. Later, other investigators
modified the method
for the study of LP (3-6). These previously reported
techniques
require an unusually
great amount
of
lipid dye in the sample gel, which makes it very difficult to see whether or not chylomicra
are present
(7).
Likewise,
these methods
are not suitable
for
semiquantitative
estimation of the LP bands because
of the high background
of the sample gel and, to
some extent,
the background
of the spacer gel. Wada
and Mise (8) studied
PAGE and succeeded
in eliminating
most of these shortcomings.
The concentraFrom the Division
of Research
and the Division
of Laboratory
Medicine,
Cleveland Clinic Foundation. Cleveland, Ohio 44106.
‘Presented,
in part, at the 24th National
Meeting,
American
Association
of Clinical
Chemists.
Workshop
on Hyperlipoproteinemia, August 22-23, 1972.
2Research
Fellow of the American
Heart Association.
Northeast
Ohio Chapter,
Inc.
aNonstandard
abbreviations
used:
PAGE,
polyacrylamide-gel
disc-electrophoresis(
-tic); LP, lipoproteins;
Tris, tris( hydroxymmethyl )aminomethane;
TEMED,
N,N,N,’N’-tetramethylethylenediamine; Ris. N,N’-methylene
bisacrylamide;
LDL, low-density
lipoproteins;
HDL, high-density
lipoproteins;VLDL, very-low-densitylipoproteins;
d. relative density.
Received Nov. 17, 1972;accepted Nov. 22, 1972.
228
CLINICAL CHEMISTRY. Vol. 19. No. 2, 1973
tions of lipid dye and dye solvent (ethylene
glyco!) in
the sample
gel and of riboflavin
in the sample
and
spacer gels were modified
to yield resolved
LP patterns that were suitable
for evaluation
by visual in-
spection
and
appeared
possibly
to be suitable
for
densitometric
scanning.
Little
information
is available
on accuracy
and
precision
of PAGE,
its applicability
for screening
sera with a wide range of lipid values, or its potential
use for semiquantitative
estimation
of LP fractions.
These are some of the reasons
why PAGE
is little
used
in clinical
chemistry
for routine
lipoprotein
screening.
This study was designed
to (a) develop
a PAGE
system
that
would
give LP patterns
suitable
for
densitometric
scanning
over a wide range of serum
lipid values, (b) compare
the LP profile so obtained
with other analytical
methods
to determine
the accuracy of this system, and (c) determine
its precision
(reproducibility
and repeatability).
Materials and Methods
Stock
Solutions4
Solution
A. Dissolve 36.6 g of Tris and 0.46 ml of
TEMED
in 48 ml of dilute HC1 (1 mol/liter),
dilute
to 98 ml with distilled
water, adjust the pH to 8.9
with 1 molar HC1, and dilute to 100 ml with distilled
water.
Solution
B. Dissolve 5.98 g of Tris and 2.3 ml of
TEMED
in 48 ml of 1 molar HC1, dilute to 98 ml
with distilled
water,
adjust
the pH to 6.6 with 1
molar HC1, and dilute to 100 ml with distilled
water.
Solution
C. Dissolve 9.6 g of acrylamide
(electrophoresis
grade) and 0.252 g of Bis in distilled
water
and dilute to 100 ml.
Solution
D. Dissolve 10.0 g of acrylamide
and 2.5 g
of BIS in distilled
water and dilute to 100 ml.
Solution
E. Dissolve
4.0 mg of riboflavin
in 2000
ml of distilled water.
Solution
F. Dissolve 15.38 mg of riboflavin
in 1000
ml of distilled water.
4A11 reagents were obtained from Eastman Organic Chemical
Co., Rochester,
N.Y. 14650. Preparation
and storage of solutions
are discussed
by Davis (2).
Solution
G. Dissolve 0.14 g of ammonium
persulfate in 100 ml of distilled water.
Solution
H. Dissolve 10.0 mg of Sudan Black B
dye in 10.0 ml of ethylene glycol. Prepare this solution
two weeks
before
use;
this
dye solution
must
be
aged. Minimize exposure to light.
Reservoir Buffer. Dissolve 4.0 g of Tris and 19.2 g
of glycine (ammonia-free)
in distilled water and dilute to 1000 ml (pH, 8.4).
Working
Solutions
The
diately
system
following
Working
Solutions
are made immebefore use. Unless
otherwise
specified,
this
was used for PAGE.
Separating
Gel, 36 g/liter. Mix 1.0 part of Solution
A, 3.0 parts of Solution
C, and 4.0 parts of Solution
Spacer Gel, 25 g/liter.
Mix 1.0 part of Solution
B,
2.0 parts of Solution D, and 5.0 parts of Solution E.
Sample
Gel, 33.3 g/liter.
Mix 1.00 part of Solution
B, 2.00 parts of Solution D, 0.65 part of Solution F,
and 2.35 parts of Working
Sudan
Working
Sudan Black B Solution
1.2 parts of Stock Sudan
parts of ethylene glycol.
Black
Black
B Solution.
is made by mixing
B Solution and 8.8
Procedure
1. Place
0.9 ml of Separating
Gel in each tube,5
and allow the
overlay
with 0.2 ml of distilled
water,
gel to polymerize
(about 30 mm).
2. Remove the water-overlay
and add 0.1 ml of
Spacer Gel to each tubes overlay with 0.2 ml of distilled water,
and allow the gel to photopolymerize
under 15-watt fluorescent
lamp for about 25 mm.
3. Remove
the water-overlay
and add 0.2 ml of
Sample
Gel to each tube, deliver 20 .tl of serum sample, cover the top of each tube with Parafilm,
and
invert the rack containing
the gel tubes about 12
times to completely
mix serum and sample gel. Overlay with Reservoir
Buffer and allow the gel to
photopolymerize
for about 45 mm.
A tuberculin
syringe is used for precise delivery
of
the amount
of the Gels, and for the overlay process a
2.5-mI syringe
with 26-gauge
needle
is used. If the
Sudan
Black
B dye preciptates
after mixing
the
Sample
Gel components,
it indicates
unclean
glassware, and the Sample
Gel must be prepared
again.
Although
it has been suggested
that, after polymerization,
bands form more discretely
if the Separating
Gels are first stored under refrigeration
for 24 to 48 h
before use (4, 5), we recommend
immediate
use with
the described
system.
even overnight, results
band.
4. Electrophorese
Storing the Separating
Gels,
in poor resolution of the a-LP
the
samples
(3 mA
per
tube,
constant
current)
35 mm or until the fast1 cm from the anodal
is about
5. Scan the electrophoresed
samples
as soon as
possible. Delay in scanning results in decreasing resolution and possible fading of the bands.
We used the “Quick-Disc”
electrophoretic
cell
(Model 1200; Canalco, Inc., Rockville, Md. 20852)
and the Heathkit
Variable
Power Supply
(Cat. No.
IPW-17;
Heath
Co., Benton
Harbor,
Mich. 49022).
Gel tubes were made from glass tubing (i.d., 5.0 mm;
length,
75.0 mm). The PAGE patterns
were scanned
with the “Analytrol”
with a scanning
plate (Model
R-102, Cat. No. 324220; Beckman
Instruments,
Inc.,
Fullerton,
Calif. 92634) and with use of a B-5 cam and
600-nm
filters. The scanning
plate was modified
by
cutting
a 78 X 7 mm slit to accommodate
the gel
tubes.
Serum triglycerides
and total cholesterol
were determined
(9) in all samples studied.
Analytical
Variables
Optimum
separating-gel
concentration.
Using stock
A, C, E, and distilled water as described by
Davis (2), we made 12 various concentrations
of separating
gel (ranging
from 21 to 70 g of acrylamide
per liter).
Twenty
microliters
of the same serum
sample
(triglycerides,
185 mg/dl,
cholesterol,
160
mg/dl)
was added to each sample gel. In addition to
visual
inspection,
we scanned
the electrophoretic
solutions
patterns
to determine
which gel concentration
gave
the clearest LP resolution and was best adapted for
scanning.
Characterization
microliters
strips,
and,
of the
of serum
after
bands.
Fifty
to a pair of paper
lipoprotein
was applied
overnight
electrophoresis
in a Dur-
rum cell, one strip was oven-dried for 30 mm, followed
by Oil Red 0 staining (10). The other strip was immediately
removed from the cell and 1-cm pieces
were cut, starting from the point of application.
The
lipoproteins
were eluted from the pieces with NaC1
solution (9 g/liter), and the eluates were electrophoresed
on the
acrylamide
PAGE
system
separating-gel.
with
use of the
The same
serum
36 g/liter
sample
was also used for serial preparation
of the VLDL (d,
< 1.006), LDL (d, 1.006-1.063),
and HDL (d, 1.063-
1.21) fractions
by the preparative
ultracentrifugal
technique
(11).
The LP fractions
were dialyzed
against saline (9 g/liter),
and an aliquot (10 sl) of
each fraction was electrophoresed
on the polyacrylamide gel support
medium
described
above (see
“Procedure”).
The identity of LP bands on the polyacrylamide
gel was further confirmed by slicing the bands soon
after electrophoresis
and imbedding the gels in agarose (15 g/liter)
5Before use, clean tubes should be soaked in “Tween 20” solution
(1 ml/100
ml of distilled
water)
for several
hours, removed,
and
allowed to air-dry in upright
position.
for about
est moving band (a-LP)
end of the gel tube.
prepared
in barbital
buffer
(pH
8.6,
0.1 ionic strength) for immunodiffusion
against rabbit
antisera
to human
a-LP, 13-LP, albumin,
and to
whole serum.
CLINICAL
CHEMISTRY,
Vol. 19, No.2,
1973
229
Comparison with Other Analytical Methods
Aliquots of the same serum samples were electrophoresed on paper (10), agarose (12), cellulose acetate (13), and in polyacrylamide
gel (with use of acrylamide separating
gel at a concentration
of 36 g/
liter). The cellulose acetate procedure was modified
by electrophoresing
the samples on “Sepraphore
ifi”
membrane
(Gelman
Instrument
Co., Ann Arbor,
Mich. 48106) in a “Microzone”
cell (Beckman)
as
described
by Fletcher and Styliou (14). The membrane was stained and cleared as described elsewhere
(13). The patterns
on all four support media were
scanned to determine
the relative percentage
distribution of the LP moieties, from which the a/3 LP
ratios (a-LP/pre-3
+ 3-LP)
were obtained.
The
same serum samples were also prepared at d 1.21 for
study in the analytical
ultracentrifuge
(10) and the
a/3 LP ratios (HDL/VLDL
+ LDL) were calculated. The ratios obtained by the various methods were
compared.
Reproducibility
test. To test reproducibility,
we
electrophoresed
samples of the same serum in the
first 11 tubes. In the 12th tube a reference serum, lipoprotein content of which was known, was run as a
check on the electrophoretic
run.
Repeatability
test. Aliquots of reference serumprepared from whole serum by removing VLDL by
ultracentrifugation
and stored frozen (15) (triglycerides, 65 mg/dl,
cholesterol,
125 mg/dl)-were
electrophoresed
each day for 12 days to determine
the reliability of the method.
Hyper- and hypolipoproteinemic
sera. Sera from
125 patients were selected without conscious bias for
lipoprotein
determinations,
and electrophoresed
by
the PAGE system to determine its merit as a clinical
screening method for LP abnormalities
as compared
to the paper-electrophoretic
method.
Hyper- and
hypolipoproteinemic
sera were included. The study
was a double-blind
one, in which the electrophoretic
patterns obtained by paper and by PAGE were compared. All patterns by both techniques
were visually
inspected
and classified.
The PAGE patterns
were
also scanned, for more precise evaluation
of the LP
fractions.
The integrated
peak counts of each LP
fraction from the densitometric
scans were compared
to the scores obtained
by visual inspection
of the
paper strips (10). Selected hyper- and hypolipoproteinemic sera were analytically
ultracentrifuged
for
further confirmation
of dyslipoproteinemia.
Results and Discussion
Evaluation of the Method
Optimum
separating-gel
concentration.
These results are shown in Figure 1. If the LP fractions were
to be visually inspected, the optimum concentration
for separating
gel was between 28.0 and 40.7 g/liter,
in agreement with a previous report (8). For both visual and densitometric
evaluation
of the entire LP
230
CLINICAL CHEMISTRY, Vol. 19, No.2,
1973
-
____
PRE-$-LP
_-
a-LP
0 C100Cl
o0.
Ifl00000
PERCENT
Fig. 1. Effect of different separating-gel concentrations
(in percent) on separation and resolution of the serum Iipoprotein bands
E
EEE
222
0
APPL.PT.
$ -LP
-LI
PRE-$LP
S.-’
r’
------------
PAPER
DISC
Fig. 2. Identification of the resolved lipoprotein bands.
W.S., whole serum. Paper-electrophoretic
pattern on left,
for comparison. Scale is in centimeters
spectra (chylomicra
to a-LP), the 36 g/liter gel was
most satisfactory.
For scanning
purposes,
this gel
concentration
allowed better separation
of 3- from
pre-3-LP
than the 37.5 g/liter gel previously
reported (3, 6) and also allowed the pre-3 LP to migrate several millimeters
into the separating
gel
rather than lodging close to the beginning of the separating
gel (16). Separating
gels containing
more
than 40.7 g of acrylamide
per liter caused the a-LP
fraction to separate into four to seven bands, which
may represent different HDL moieties (17, 18).
Characterization
of the lipoprotein bands in acrylamide separating-gel
(36 g/liter). In PAGE the eluates from the paper electrophoretograms
corresponding to fi-, pre-fl, and a-LP fractions migrated
in the same position as the LDL, VLDL, and HDL
moieties isolated by the preparative
ultracentrifugal
technique
(Figure 2). In polyacrylamide
gel, 3-LP
(on paper) and LDL (by ultracentrifugation)
migrate
ahead of the corresponding
pre-fl LP and VLDL
fractions; in contrast, on paper support-medium
the
pre-3 LP and VLDL fractions migrate ahead of the
-LP and LDL fractions. This difference is attributable to the molecular-sieving
effect of the gel, which
separates
macromolecules
primarily
according
to
molecular size rather than by charge. The fast-migrating bands in the 0-1, 1-2, 2-3, and 3-4 cm frac-
tions seen in the photograph
(Figure 2) of the gels
after electrophoresis
were not lipoprotein
bands, but
albumin
bands (yellowish-orange
tinge), from the
bovine albumin of the albuminated
buffer extracted
from the paper strip. The eluate from the 4-5 cm
piece demonstrated
the presence of -LP by PAGE
(Figure 2). This was most likely due to a slightly
slower rate of migration of this band on the paper
strip that was used for LP elution studies than on
the strip stained with Oil Red 0.
Immunodiffusion
(Figure
3) indicates
that the
a-LP band of polyacrylamide
gel reacted against
anti-whole serum, anti-a-LP,
and anti-albumin
antisera with no cross-reaction
against anti-/3-LP antisera. Only the precipitin
line in the position of a-LP
against anti-whole
serum and the line against antia-LP antisera were Oil Red 0 stainable.
Likewise,
the imbedded 3-LP band reacted against anti-whole
serum and anti-3-LP
antisera, but not against antia-LP antisera.
These two precipitin
lines were also
stainable
with Oil Red 0. These data not only further confirm the identification
of the respective LP
bands, but also indicate complete separation
of aand fl-LP.
Comparison
with other analytical
methods.
The
a/3 LP ratios of the same serum samples electrophoresed on different support media are given in Table
1. These data indicate that PAGE is as reliable as
the other three electrophoretic
methods for LP phenotyping
(1).
The similar ratios obtained
by the
PAGE
and
analytical
ultracentrifugal
methods
suggest that PAGE is an accurate
method for determining LP profile.
Reproducibility.
The electrophoresed
samples were
scanned to determine
the relative percent distribution of the pre-,
i3-, and a-LP fractions. The relatively low standard
deviation
of values (1.3, 1.3,
1.3%, respectively)
for each of the three lipoprotein
fractions,
calculated
from replicate
analyses
(11
tubes)
of single
serum
samples,
suggest
that
the
method is precise.
Repeatability.
The a/f3 LP ratios of the reference
serum electrophoresed
daily for 12 days had a mean
Table 1. Comparison of a/fl-Lipoprotein
Analytical
Bela-Lipoprotein
of 0.87 ± 0.04 (SD). This represents
a ±4.6% deviation, which for screening purposes and semi-quantitative studies can be considered a reliable technique.
Clinical
Hyper-
Evaluation
screen sera having
glycerides,
30-2250
double-blind
study
indicates
that
it is easy
while
to
by visual inspection
or width of the band
Hyperlipidenlic5
HypoIipidemic
± 0.04
± 0.14
± 0.03
± 0.03
0.46
0.38
0.56
0.53
PAGE
0.56
Agarose electrophoresis
Paper electrophoresis
Cellulose acetate electrophoresis
± 0.03
0.54 ± 0.03
0.47
0.22
± 0.06
0.41 ± 0.06
101
(64-120)
standard
standard
standard
(range).
(trimg/
Ratio, as Determined by Different Analytical Methods
0.28
0.28
0.33
“Means
±
Means ±
Means ±
d Mean and
can be used to
range of lipid values
cholesterol,
25-620
difficult
to diagnose
correctly
alone. Variations
in sharpness
Normal”
mg/dl!
a wide
mg/dl;
discern differences
in the intensities
of a particular
LP band in sera of patients with accentuated
hyperor hypolipoproteinemic
sera, subtle differences
are
0.60 ± 0.07
Serum cholesterol,
that PAGE
dl). PAGE often reveals additional
bands (such as
double pre-/3-LP and double 3-LP) not found by the
paper-electrophoretic
method. About 32% of the normal sera and 56% of the hyperlipidemic
sera that we
screened showed a “midband”
between the pre-fland 13-LP bands by the PAGE method. Analytical
ultracentrifugal
studies are being done to determine
whether or not this band represents
the Sr 12-20
fraction. The clinical significance
of the double pre-LP, double [3-LP, or midband is not yet known.
The LP profile and intensity
of each band obtained by PAGE method was evaluated visually and
scored in the same procedure
as paper (10). The
Methods
mg/did
sera. Our findings
and hypolipoproteinemic
for the 125 sera indicate
ultracentrifuge
Serum triglyceride,
Alpha-Lipoprotein
Fig. 3. Immunodiffusion
of PAGE (a- or (3-LP) band imbedded in agarose,
15 g/liter,
(center well) against 1,
anti-whole
sera antisera; 2, anti-/3-LP antisera; 3, anti-aLp antisera; and 4, arti-aIbumin
antisera
0.26 ± 0.06
387
(117-2250)
196
260
(180-240)
(155-620)
±
±
±
±
0.01
0.01
0.02
0.03
0.40 ± 0.03
50
(30-85)
60
(25-90)
error for 14 different patients’ sera.
error for 12 different patients’ sera.
error for 5 different patients’ sera.
CLINICAL CHEMISTRY, Vol. 19. No.2.1973
231
introduce
additional
complications
in interpreting
the intensity of the band. Evaluation
of the gel patterns by densitometric
tracings
(integrated
peak
counts) more accurately
reflected the intensity
of
each band than did visual inspection.
Visual scoring
of the band intensities
of the gels agreed 74% of the
time with the scores obtained by paper electrophoresis. The correlation
coefficients
(20), r, comparing
the integrated peak counts and paper electrophoresis
scores for the chylomicra, pre-/3-, 13-, and a-LP bands
were 0.84, 0.96, 0.99, 0.90, respectively,
which
implies good agreement
between the two methods of
determining
band intensities,
as evaluated
by integrated peak count on polyacrylamide
gel and visual
scoring of paper electrophoresis.
Paper-electrophoretic
patterns
of hyperand
hypolipoproteinemic
sera are shown in Figures 4 and
5. PAGE patterns of the same sera are shown in Figures 6 and 7, and their densitometric
scans of the
same sera are shown in Figures 8 to 12. Although
paper electrophoresis
can demonstrate
the various
dyslipoproteinemic
phenotypes,
PAGE can further
NOIMAL
TYPE I
TYPT II..
TYPSIl-b
TYPI III
TYPI IV
0
-
4
.fl
-->
UI
z
‘J
UI
UI
UI
UI
UI
0.
0.
>-).-)i-
0-
0.
)-
O
>-
0.
)-
-
a-LP
-
-.
_
.._
a....#{149} _-.-.
Fig. 6. PAGE patterns
in hyperlipoproteinemia.
represents
healthy
adult male, age 30.
Figure 4
“Normal”
Same sera as in
HYPO-$ -LIPOPROTEINEMIA
TYPI V
PRE-6-LP
.LP
-
i-
CHYLOMICRA
PRE-$-LP
-LP
APPL.PT.
(CHYLOMICIOHt
-
i-
-__
$-LP
-
.
-.9.
P..1-LP
-
‘j”
#{149}.LP
-
a-LP
TO
CHOLIST1IOL
60
110
345
320
110
950
7
210
192
1010
190
(n’./dI)
AOl
30
22
57
37
27
57
57
SIX
N
P
F
N
N
N
N
Fig. 4. Paper-electrophoretic
emia [WHO nomenclature,
healthy male, age 30
patterns in hyperlipoprotein(14)].
“Normal”
Hypo
Norinol
APPL. PT.$-LP
PRE-p-LP
represents
-
N
Fig.
7.
teinemic
PAGE patterns of
sera (107). Same
P
normal (N) and hypo-13-lipoprosera as in Figure 5
_i
IACSO1OUND
--
--
1::-HI
A
-
107
-ffl
.-‘-.‘..
./‘__\
/-,-.--.---___,
:7,.,,:
S...pI.
0.1
Sp.c.
0.1
S.paratln.
-
To
NOtNAL
1S5..,/dIh
CHOIESTIIOL
160
L
g/
ITI
TT
I..
a-LP
0.1
-
I
I-
-
-
_4
-
B’ ITT
CHOLESTEROL
(mg
30
90
TO
:Pr.$.Lp
/dl)
Fig.
8.
A
=
electrophoretic
Fig. 5. Paper-electrophoretic
hypo-/3-lipoproteinemic sera
232
CLINICAL
CHEMISTRY,
$-LP
a-LP
50
90
patterns
Vol. 19, No.2,
of
normal
and
resents
Figures
1973
healthy
4-7)
Background
scan
pattern
of
male,
age
of
“normal”
30
blank
gels;
serum.
(same
normal
B
=
Scan
of
“Normal”
serum
repas
in
‘-tYNuiH’ETT1
TO
3000 mg/a
CHOLESTEROL 110 .19/dI
i--H
--
L
4---
-.
ii-
i-
i.
4 ..,4
-
tfHrrHT
A
r
H
-.
1
CHYLONICRA
11
P.E-LP
11
]
B-LP
;i
±TYPEfla
TO
‘CHOLESTE1OL
-f-
B
tH
-
TG
-
H
/
1050
--
.4.
.,__
- -
-i--i---
-
-.
-
B
H
/
:
g-LP
-
j
pattern of Type I phepattern
of Type lI-a
_.-t
Fig. 9. A, Scan of electrophoretic
notype; B, Scan of electrophoretic
phenotype
TYPE Il-b
:
TO
320 raq/dt
, CHOLESTEROL 300 rag/al
-
20
.--
:
-.
‘jCKOtES!IROI.590 a/
JtTh
-
. -
11
I
___
__
‘Fif
‘f
--
1
-
-j,
:
:
-
-;
t4
-
±-
1_I1ITLj__fT
95 mg/dl
345 mg/dl
‘I
-
-
,,
.-LP
:
‘I-
:ri
H4--
L4.
-,
H -R -----p-’
4t
-1----t
f
TO
--
- --
-
,
,
-
-
.
-:
-
.-
-
-.
...
__.
.
-
.
-
--
U
-‘
of electrophoretic
B, Scan
pattern
of electrophoretic
of Type IV
of Type V
pattern
Eii
ifNaiJLii
:
-.
A, Scan
L
phenotype
___
-
Fig. 11.
phenotype;
-
-E
-
L.
,CNOI.ESTEROI. 190
A
A
-
_.
.._.
L...
-
-.
......
-.
.-
T
i’1
1 /
Pn.sIp
-.
‘-
-
-
‘--
--
-.
.-i----
-
---4-
-----‘-
1U0rag/dI
250 aig/dI
-
-
-----H
TT’T
-
‘j
---
H
phenotype;
B
-
---
---
r
-
_
,--
-
. -
of electrophoretic
B, Scan of electrophoretic
Scan
6), the chylomicra
--
U..
-_
L
_
.
-
:
2.
-
.-_
. -.
_
---
-
--
-.
_____________
2
-!
1-il-r
pattern of Type
pattern
of Type
s-I.,
Il-b
III
clarify the phenotypic
patterns by clearly separating
the LP fractions into discrete bands, which can be
seen and can be evaluated
by densitometric
scanning. Although the patterns shown exemplify each of
the classes of dyslipoproteinemias,
they may not represent the only type of patterns in each class. For example, in Type V hyperlipoproteinemic
serum, three
chylomicron
bands were found in the spacer gel (Figure 6), but we have observed the chylomicra
represented as a single band at the interface of the sample
and spacer gels in some Type V samples and as a
single band in the spacer gel in others. In the Type I
(Figure
rag/al
I:
00
I%SIt
phenotype
case
B-UP0Pt0T
lA
30 rag/al
_______
P-LP
I
A,
TO
‘CHOLESTEROt $0
Li__
10.
‘U
-.
‘H
‘H---H---P-/1L--Lt:
PI’I#{216}-LP
‘INP0
-
__
Fig.
.-LP
‘
H
rTYpEt
T0
‘CHOLESTEROL
$-l.P
--.----...
remain
diffusely
sep-
arated in the sample gel. Whether this is true of all
Type I sera is not yet clear.
These results indicate that our PAGE system can
-
.-LP
I
- .,._..
._._L_
Fig. 12. A, Scan of electrophoretic
pattern of normal
serum; B, Scan of electrophoretic pattern of hypo-$-Iipoproteinemic serum
be used as a valuable analytical
method to determine lipoprotein
distribution
in serum (or plasma)
samples with a wide range of lipid concentrations,
and the results agree with those obtained by analytical ultracentrifugation
and by paper electrophoresis.
If LP interacts with any of the components
used in
the system, it is not evident and does not interfere
with the accuracy of the method. The method is precise, reliable, and accurate;
and can be used in the
clinical laboratory
as an effective screening method
for abnormal serum lipoproteins.
It can also be used
in addition to other routine LP screening procedures,
when more precise identification
of the serum LP
components
is needed, i.e., for presumptive
diagnosis
of phenotypes,
such as that of Type III (19). Our use
of the optimum
concentration
of Sudan
Black
CLINICAL CHEMISTRY. Vol. 19. No.2.1973
B dye
233
in the sample gel and of a decreased concentration
of
riboflavin in the sample and spacer gels diminished
the background
considerably,
which facilitated
visual assessment
of chylomicra.
Furthermore,
because
background was minimized, PAGE patterns could be
scanned with a densitometer.
The integrated
peak
count of the scans provided a more accurate interpretation
of the gel patterns
and the intensities
of
each lipoprotein moiety.
To provide precise quantitative
estimation
of the
lipoprotein
fractions,
investigation
is now in progress
to determine
the optimum
size for serum samples
and for lipid dye concentration
for this PAGE system. An in-depth investigation
of PAGE patterns of
various LP phenotypes is being made.
This study was supported, in part,
the National
Heart and Lung Institute,
by Grant No. HL-6835
NIH, USPHS.
from
References
I. Ornstein,
L., Disc electrophoresis.
Ann. N.Y. Acad. Sci. 121, 321 (1964).
2. Davis, B., Disc electrophoresis.
human serum proteins. Ann. N.Y.
I. Background
and
theory.
II. Method
and application
Acad. Sci. 121, 404(1964).
to
3. Frings, C. S., Foster.
L. B., and Cohen. P. S., Electrophoretic
separation
of serum
lipoproteins
in polyacrylamide
gel. (‘tin.
(‘hem. 17, 111 (1971).
4. Moran,
R. F., Castelli,
of beta lipoprotein
(LDL)
of disc electrophoretograms.
5. Canalco
Diagnostic
6. Narayan.
electrophoresis
(1965).
“Quick
Products
W. P., and Moran, M. V., Quantitation
cholesterol
by densitometric
evaluation
(‘tin. (‘hem. 18, 217 (1972).
Disc” Electrophoresis
of Lipoproteins,
Group, Rockville.
Md. (1969).
K. A., Narayan,
S., and Kummerow,
F. A., Disk
of human
serum
lipoproteins.
Nature
205, 246
7. Wolfman,
L.. and Sachs, B. A., Disc electrophoresis
teins: A reinterpretation
of the chylomicron
band.
16,620(1970).
234
Canalco
CLINICAL CHEMISTRY, Vol.
19, No. 2, 1973
of lipopro(‘tin. (‘hem.
8. Wada, M., and Mise, J., An investigation
of serum lipoprotein
disc electrophoresis.
I. Behavior
of serum lipoproteins
in disc type
acrylamide
gel electrophoresis.
Jap. Circ.J. 36, 121 (1972).
9. AutoAnalyzer
Method
File, Cholesterol
ide N-78, Technicon
Instruments
Corp.,
(1965).
N-24a, and TriglycerTarrytown,
N.Y. 10591
10. Lewis, L. A., Screening
for serum lipoprotein
abnormalities:
Comparison of ultracentrifugal,
paper
and thin-layer
starch-gel
electrophoresistechniques. Lipids 4,60(1969).
11. Lindgren,
F. T., Nichols,
A. V., Freeman,
N., Wills, R. D.,
Wing, L., and Gullberg,
J. E., Analysis
of low-density
lipoprotein
by preparative
ultracentrifugation
and refractometry.
J. Lipid
Res. 5, 68(1964).
12. Pfizer Electrophoresis
tein Phenotyping,
Pfizer
(1972).
Operating
Diagnostic
Manual:
V. Serum LipoproDivision,
New York, N.Y.
13. Chin, H. P., and Blankenhorn,
D. H., Separation
and quantitative analysis
of serum lipoproteins
by means of electrophoresis
on cellulose acetate.
Clin. Chim. Acta 20, 305(1968).
14. Fletcher,
M. J., and Styliou,
M. H., A simple method
arating
serum lipoproteins
by electrophoresis
on cellulose
Clin. Chem. 16,362(1970).
-
for sepacetate.
15. Hulley, S. B., Cook, S. G., Wilson, W. S., Nichaman,
M. Z.,
Hatch,
F. T., and Lindgren,
F. T., Quantitation
of serum lipoproteins by electrophoresis
on agarose
gel: Standardization
in lipoprotein
concentration
units (mg/100
ml) by comparison
with analytical ultracentrifugation.
J. Lipid Res. 12,420(1971).
16. Wada,
M., and Mise, J., An investigation
tein disc electrophoresis.
II. A simple method
poprotein disc electrophoresis
and identification
poproteins.
Jap. Circ.J. 36,335(1972).
17. Utermann,
G., Disc-electrophoretic
high density lipoproteins.
(‘tin. Chim.
IS. Dangerfield,
ma lipoproteins
Acta 30,273(1970).
W. G., and Pratt,
by polyacrylamide
of serum lipoproof human
serum liof the resolved
Ii-
patterns
of human
Acta 36, 521 (1972).
serum
J. J., An investigation
of plaselectrophoresis.
Clin. Chim.
19. Beaumont, J. L., Carlson, L. A., Cooper, G. R., Fejfar, Z.,
Fredrickson,
D. S., and Strasser,
T., Classification
of hyperlipidaemias
and hyperlipoproteinaemias.
Bull.
W. H. 0. 43, 891
(1970).
20. Steel, R. G. D., and
of Statistics,
McGraw-Hill,
Torrie, J. H., Principles
and Procedures
New York, N.Y., 1960, pp 75, 183.