[CANCER RESEARCH 35, 2928 2934, November 1975]
The Isolation of Carcinoembryonic Antigen from Tumor Tissue at
Neutral pH
Robert J. Carrico and Magdalena Usategui-Gomez
Ames Research Laboratory, Ames Company Division of Miles Laboratories, Inc., Elkhart, Indiana 46514
SUMMARY
Carcinoembryonic antigen (CEA) was purified from
tumor tissue under mild conditions at neutral pH by a procedure that utilized affinity chromatography on concanvalin
A. Further purification by gel filtration provided C E A in 10
to 20% yield and 10% purity. Antibody to this preparation
was rendered specific for C E A by adsorption on a column of
normal liver proteins bound to Sepharose. On reaction by
immunodiffusion against a crude tumor extract, the adsorbed antibody produced two precipitin lines, of which one
was relatively weak. These two precipitin lines fused completely with the two respective lines produced by antibody to
perchloric acid-treated CEA. The major antigen found in
crude tumor extracts and in C E A preparations purified at
neutral pH was nearly undetectable in perchloric acid
extracts of t u m o r homogenates. Further investigations
showed that 60 to 70% of the CEA in crude tumor extracts
and in preparations isolated at netural pH is destroyed
a n d / o r becomes insoluble under acidic conditions.
INTRODUCTION
C E A 1 is a tumor-associated antigen first identified in
adenocarcinoma of the colon by Gold and Freedman (11).
Later, it was also found in adenocarcinomas of the pancreas
and liver (12). As more sensitive immunological techniques
were developed, relatively low levels were detected in
normal tissues (2, 21, 22).
Originally, C E A was isolated from tumor homogenates
by extraction with M perchloric acid which precipitates most
of the tissue proteins while a few, including CEA, remain in
solution (17).The soluble proteins were fractionated further
by gel filtration and preparative electrophoresis. All purification procedures developed later used perchloric acid
extraction (7, 16, 18). CEA is generally considered to be
stable in this acid because it is soluble and its immunological properties are not destroyed (17).
CEA appears to be a mixture of structurally related
glycoproteins (6). Isoelectric focusing separated a CEA
preparation into several components with the major ones
being isoelectric between pH 1 and 3 (6). These components
had different amino acid and sugar compositions.
Antiserum raised to CEA preparations usually contains
XThe abbreviations used are: CEA, carcinocmbryonicantigen: Con A,
concanavalin A.
Received March 31, 1975; accepted July 17, 1975.
2928
antibody to antigens found in extracts of noncancerous
tissues. The antiserum is typically rendered monospecific
for CEA by adsorption with proteins from normal tissues
(11, 12, 15, 25). Some adsorbed antisera gave 2 precipitin
lines by immunodiffusion with perchloric acid extracts of
tumor tissue (11, 15, 22). Further fractionation of the CEA
preparations by gel filtration and electrophoresis failed to
separate the 2 antigens (15).
Strong acid can hydrolyze glycosidic bonds in the carbohydrate moiety of CEA and can denature the polypeptide
portion. Purification of CEA without acid conditions could
yield higher quality antigen and, consequently, better antiserum. Binding of C E A to Con A has been reported (4, 5).
We have found that affinity chromatography on immobilized Con A is a useful step for isolation of CEA. Some
properties of C E A isolated at neutral pH are compared with
the antigen prepared by extraction with perchloric acid.
MATERIALS AND METHODS
Purification of CEA. Liver metastases of carcinoma of
the colon were dissected from normal tissue, and 150 g were
homogenized with 450 ml distilled water for 10 rain at 0 ~ in
a Sorvall Omnimixer (Ivan Sorvall Co., Newton, Conn.).
The homogenate was treated sonically for 10 rain at 0 ~ with
a Branson Model W140D sonic oscillator (Branson Sonic
Power Co., Plainview, N. Y.) operated at 70% of full power.
Cell fragments were separated by centrifugation at 9000 •
g for 20 min. The supernatant was dialyzed against
Tris-acetate buffer, pH 6.9, F / 2 0.02, for 40 hr at 5 ~.
Centrifugation at 9000 • g removed some particulate material, and the turbid supernatant was passed into a 5- x 35-cm
column of DEAE-cellulose previously equilibrated with the
Tris-acetate buffer. The chromatogram was developed at
room temperature with a linear salt gradient generated with
2 liters, Tris-acetate buffer, pH 6.9, P / 2 0.02, and 2 liters of
this buffer containing 0.5 M NaC1. Fractions (22 ml) were
collected, and those containing more than 2 ug CEA per ml,
measured by radioimmunoassay, were pooled and passed at
a flow rate of 50 m l / h r through a 2.5- x 29-cm column of
Con A bound to Sepharose (Pharmacia Fine Chemicals,
Piscataway, N. J.); 8 mg Con A bound per ml Sepharose.
Then the column was washed with 0.2 M sodium citrate
buffer, pH 6.5, containing M NaC1, mM MgC12, mM MnC12,
and mM CaCI~. When the absorbance at 280 nm of the
effluent decreased to 0.1, 120 ml of 0.1 M a-methyl-D-mannoside in the citrate buffer was passed into the column.
Elution with sugar was repeated 20 and 44 hr later. The
CANCER RESEARCH VOL. 35
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.
Isolation o f C E A at N e u t r a l p H
sugar eluates were combined and concentrated to 5 ml by
pressure dialysis.
The Con A column was regenerated by washing with 2
bed volumes of the sodium citrate buffer each day for 7 to
10 days. With this treatment, 1 column was used for 12
CEA preparations without noticeable change in binding
efficiency.
The concentrated sugar eluate was chromatographed at
7 ~ on a 2.5- x 90-cm column of Sepharose 4B equilibrated
with sodium acetate-acetic acid buffer, pH 6.5, I ' / 2 0.05,
containing 0.15 M NaCI. Fractions (15 ml) with more than 5
~g CEA per ml were pooled. The protein was concentrated
to 5 ml by pressure dialysis and fractionated further by gel
filtration on 2.5- x 90-cm column of Sephadex G-200
equilibrated with the sodium acetate-acetic acid buffer.
Again, 15-ml fractions were collected, and assays for CEA
revealed 2 components, as described in "Results." The main
peak and shoulder were collected into separate pools. They
were concentrated to 2 to 5 ml each and stored at - 15 ~ until
needed.
Isolation of CEA by Extraction with Perchloric Acid.
Tumor tissue homogenates were extracted with perchloric
acid by the method of Krupey et al. (18). After dialysis
against distilled water, the acid-soluble materials were
dialyzed into Tris-acetate buffer, pH 6.9, I ' / 2 = 0.02. Then
the protein was concentrated by pressure dialysis and used
for immunodiffusion measurements.
Crude Extracts of Normal and Tumor Tissues. Normal
liver and colon were autopsy specimens without known
neoplastic disease. The tissues were homogenized as described above for tumor tissue. Particulate material was
separated by centrifugation and the supernatants were
dialyzed against distilled water. The dialysate was concentrated by pressure dialysis.
Crude extracts of tumor tissue were prepared by the same
procedure for use in immunodiffusion studies.
Binding of Protein to Sepharose. The method of Axen and
Ernbach (1) was used to bind proteins to cyanogen bromideactivated Sepharose. Sepharose-4B, 20 ml packed, was
washed with 250 ml distilled water on a scintered glass
funnel and transferred to a beaker with 40 ml water. The pH
was adjusted to 11 with 5 N N a O H , and 7 g cyanogen
bromide dissolved in 30 ml water were added. The pH was
maintained near 11 by the addition of N a O H , and the
temperature was kept between 20 ~ and 25 ~ by the addition
of ice. When the reaction had proceeded for 12 min, the
activated Sepharose was washed with 250 ml cold (0 ~ 0.2 M
sodium citrate buffer, pH 6.5, on a scintered glass funnel
and was transferred into a flask with 25 ml cold sodium
citrate. The crude extract from 10 g normal liver, prepared
as described above, was added and the suspension was
stirred at 7 ~ overnight. Then the protein-Sepharose conjugate was washed with 250 ml sodium citrate buffer, and
unreacted sites were blocked by stirring for 2 hr with 40 ml 1
M ethanolamine in the sodium citrate buffer adjusted to final
pH 8.0. The conjugated Sepharose was washed with 250 ml
phosphate-buffered saline (0.04 sodium phosphate, pH
7.4, containing 0.12 M NaC1), and poured into a column
for use as an immunoadsorbent.
Preparation of Antiserum. For the initial immunization,
20 to 40 ug of purified C E A in complete Freund's adjuvant
were injected s.c. into goats, near the footpads. Booster
injections in incomplete Freund's adjuvant were administered s.c. in the back at 3-week intervals. Three months after
the initial injection, the antisera had developed sufficiently
high titers for use in immunodiffusion measurements.
Antiserum to CEA purified by perchloric acid extraction
was kindly provided by Dr. C. W. Todd (7).
Adsorption of Antiserum. Antiserum to CEA purified at
neutral pH was rendered monospecific for C E A by adsorption on a column of normal liver proteins bound to
Sepharose. The immunoadsorbent described above was
equilibrated with phosphate-buffered saline, pH 7.4, at
room temperature, and 2 ml antiserum were washed
through at a flow rate such that unbound proteins eluted in
10 to 15 hr. The absorbance at 280 nm of the effluent was
monitored, and serum proteins that eluted were concentrated to 2 ml by pressure dialysis.
lmmunodiffusion Measurements. A suspension of 1%
Nobel agar (Difco Laboratories, Detroit, Mich.) in 1 M
glycine, 0.8 mM sodium azide, and 0.15 M NaC1 was
adjusted to pH 7.0 to 7.5 with N a O H and heated at 95 ~ for
5 min. This suspension was poured into 1.5-mm thick layers
on glass plates. Immunodiffusion was conducted for 24 to
48 hr in a humid chamber at room temperature.
Immunoelectrophoresis was conducted in 1% agarose by
the method of Scheidegger (24).
Radioimmunoassay. CEA and antibody to CEA for the
radioimmunoassay were provided by Dr. C. W. Todd (7).
The C E A was radiolabeled with 125I by the method of Egan
et al. (10). Assays were conducted in phosphate-buffered
saline, pH 7.4, containing 1% bovine serum albumin.
Antibody and antigens were incubated at 5 ~ in a 0.5-ml
reaction overnight, and then rabbit anti-goat lgG was
added. This 2nd antibody was allowed to react at room
temperature for 2 hr. The antigen-antibody precipitate was
collected by centrifugation and washed twice with 1 ml
phosphate-buffered saline containing 1% bovine serum
albumin. The washed precipitate was dissolved in 1 ml N
N a O H for counting 125I decay.
Protein Assay. The method of Lowry et al. (20) was used
to measure total protein during purification of CEA. Bovine
serum albumin was used as the standard.
Electrophoresis, Isoelectric focusing was conducted in a
110-ml column (LKB Productor, Stockholm, Sweden) with
1% (w/v) ampholytes. The protein was included in the dense
solution used to develop the sucrose gradient. The column
was maintained at l0 ~ and a potential of 500 V was applied
for 60 hr. Then the absorbance at 280 nm of the column
effluent was monitored and 2-ml fractions were collected.
The pH of the fractions was measured and they were
assayed for C E A by the radioimmunoassay method.
Disc electrophoresis was conducted in 7% polyacrylamide
gels at pH 8.9 by the method described by Davis (8).
RESULTS
Extraction of Tumor Tissue. C E A is associated with the
tumor cell surface (13) and is partially released by homoge-
NOVEMBER 1975
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.
2929
R . J. C a r r i c o a n d M . U s a t e g u i - G o m e z
nization in distilled w a t e r (Table 1). To determine the most
effective conditions for extraction, aliquots of h o m o g e n a t e
were treated by the m e t h o d s listed in Table 1. P a r t i c u l a t e
m a t t e r was rein oved by centrifugation, and the s u p e r n a t a n t s
were dialyzed against Tris-acetate buffer, pH 6.9, F / 2 0.02,
for 40 hr and then assayed for C E A . The results shown in
Table 1 indicate that the highest yield of C E A was obtained
by sonic disruption, and the lowest by perchloric acid
extraction. The detergents, sodium dodecyl sulfate and
Tween 20, gave essentially the same yield as water alone.
D E A E - C e l l u l o s e C h r o m a t o g r a p h y . C h r o m a t o g r a p h y of
dialyzed t u m o r extracts on D E A E - c e l l u l o s e r e m o v e d protein and lipid m a t e r i a l that interfered with the subsequent
affinity c h r o m a t o g r a p h y on Con A. S o m e heterogeneity of
the C E A is evident from the elution profile in C h a r t 1. Since
the c o m p o n e n t s were poorly resolved, fractions containing
m o r e t h a n 2 ~tg C E A / m l were collected into 1 pool. The
recoveries of C E A varied from 25 to 50% (Table 2).
Affinity Chromatography on Con A. The immobilized
C o n A bound over 90% of the C E A pooled f r o m the
D E A E - c e l l u l o s e c h r o m a t o g r a m , while a large portion of the
protein impurities passed directly through the column
( C h a r t 2). M o s t of the protein bound to Con A was eluted
i m m e d i a t e l y with a - m e t h y l - D - m a n n o s i d e ; however, after
the column stood overnight, additional protein was recovered. Affinity c h r o m a t o g r a p h y increased the purity of
the C E A 8-fold (Table 2).
Gel Filtration. The C E A eluted from Con A was separated from higher-molecular-weight c o n t a m i n a n t s by chrom a t o g r a p h y on Sepharose-4B ( C h a r t 3A). The antigen
eluted as an a s y m m e t r i c a l peak between 75 and 100% of the
column bed volume. F u r t h e r gel filtration on S e p h a d e x
G-200 s e p a r a t e d C E A from some of the lower-molecularweight proteins ( C h a r t 3B). A b o u t 70% of the antigen eluted
as a s y m m e t r i c a l peak between 35 and 50% of the column
bed volume. The r e m a i n i n g 30% a p p e a r e d as a shoulder on
the m a i n peak ( C h a r t 3B). A p p r o x i m a t e l y 10% of the protein in the m a i n peak was C E A , and this represented 10 to
20% of the antigen originally extracted from the t u m o r
tissue (Table 2).
Table 1
Yields o f CEA obtained from a tumor homogenate by various
extraction methods
Extraction metho&
Distilled water
1 MNaCl
1% Tween 20
1% Sodium dodecyl sulfate
1-Butanol ~
1 MHC10,
Sonic disruption c
Yield
(#g CEA)
14
l0
17
13
i1
6.3
23
a Aliquots of a tumor homogenate (2 ml) in distilled water were mixed
with 2 ml of the appropriate solutions to give the designated final
compositions. These mixtures were stirred at room temperature for 30
min and then were processed as described in "Results."
The homogenate was stirred with an equal volume of 1-butanol for 30
min. The aqueous phase was taken for dialysis and CEA analysis.
c Sonic disruption was conducted as described in "Materials and
Methods."
2930
E-
i.,r
15
g
1.0
(~
~ta
Q.8
0.6
o O.4
,~
k l \ ' ~ - e - " ~ ~ ~-o - - ~
_
_
53_
0,2
400
800
12oo
16OO
2000
EFFLUENT
VOLUME
2400
2800
3200
(roll
Chart 1. DEAE-cellulose chromatography of a tumor tissue extract.
The proteins extracted from 150 g of tumor tissue were dialyzed against,
Tris acetate buffer, pH 6.9, F/20.02, and passed into a 5- x 35-cm column
of DEAE-cellulose previously equilibrated with the Tris-acetate buffer.
The chromatogram was developed with a linear salt gradient as described
in "Materials and Methods." The absorbance at 280 (solid line) of the
column effluent was monitored and 22-ml fractions were collected; broken
line, CEA levels measured by radioimmunoassay.
Table 2
Yield and purity of CEA during purification
CEA was purified from 150 g of liver metastases, as described in
"Materials and Methods." The amount of CEA obtained at each purification step was measured by radioimmunoassay, and total protein was
estimated by the method of Lowry et al. (20). Tissue extract refers to the
materials that did not sediment at 9000 • g following dialysis against
Tris-acetate buffer, pH 6.9, F/2 0.02.
Total
CEA
(rag)
Tissue extract
DEAE-cellulose chromatography
Affinity chromatography on Con A
Sepharose chromatography
Sephadex chromatography
Peak 1
Peak 2
10.3
2.9
2.6
2.1
1.3
0.6
Total
protein
(rag)
5900
1600
170
108
11.5
19
Recovery
of CEA
(%)
100
28
25
20
13
6
Isoelectric Focusing. F u r t h e r purification of C E A was
a t t e m p t e d by isoelectric focusing with p H 3 to 6 ampholytes. The protein r e m a i n e d in solution during the run,
and C E A levels m e a s u r e d in the column effluent revealed
an elution profile very similar to that reported by Coligan
et al. (6).
M u c h different results were obtained when isoelectric
focusing was conducted with pH 2.5 to 4.0 ampholytes.
Between 30 a n d 40% of the C E A was recovered in solution
in the column effluent. M o s t of this was in fractions with p H
above 5.5. F r a c t i o n s with p H 3.5 to 4.5 contained precipitated protein which was separated by centrifugation and
extracted with 0.25 M Tris-HC1 buffer, pH 8.6. C E A
recovered in the extract represented 10% of the original
antigen. By mass balance, a p p r o x i m a t e l y 50% of the
original C E A precipitated during isoelectric focusing and
would not dissolve at p H 8.6.
Solubility of CEA under Acidic Conditions. The solubility
of C E A at acid p H was examined further by dialyzing a
purified p r e p a r a t i o n containing 85 ~tg C E A per ml against
sodium acetate-acetic acid buffer, p H 4.3, 1'/2 0.001, for 23
C A N C E R R E S E A R C H VOL. 35
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.
I s o l a t i o n o f C E A at N e u t r a l p H
hr at 7 ~ Precipitate that formed was separated by c e n t r i f u g a t i o n , a n d the s u p e r n a t a n t c o n t a i n e d 30 # g C E A per ml
w h i c h was 35% o f the o r i g i n a l m a t e r i a l . O n l y 3% o f the
o r i g i n a l a n t i g e n was extracted from the precipitate with 0.1
M Tris buffer, p H 8.0. T h e r e f o r e , p r e c i p i t a t i o n o f C E A in
i s o e l e c t r i c - f o c u s i n g e x p e r i m e n t s a p p e a r s to be a n effect o f
p H and is not dependent on the a m p h o l y t e s or sucrose.
Immunological Studies. Immunodiffusion patterns obtained with u n a d s o r b e d a n t i s e r u m to purified C E A s h o w e d
a n t i b o d y a c t i v i t y to several p r o t e i n s in e x t r a c t s of n o r m a l
liver a n d c o l o n (Fig. 1). A d s o r p t i o n o f the a n t i s e r u m on a
c o l u m n of n o r m a l liver p r o t e i n s b o u n d to Sepharose
removed the a n t i b o d i e s to n o r m a l tissue a n t i g e n s (Fig. 1).
T h i s d e m o n s t r a t e s t h a t a c o m p l e x m i x t u r e of tissue p r o t e i n s
c a n be i m m o b i l i z e d a n d used effectively as an i m m u n o a d sorbent.
T h e a d s o r b e d a n t i b o d y r e a c t e d with t u m o r e x t r a c t s to
p r o d u c e the s t r o n g p r e c i p i t i n line seen in Fig. 1 a n d also a
w e a k line w h i c h is evident in Fig. 2. R e a c t i o n s o f i d e n t i t y
were o b t a i n e d w h e n the a d s o r b e d a n t i b o d y was c o m p a r e d
with the a n t i b o d y to p e r c h l o r i c a c i d - t r e a t e d C E A (Fig. 2).
A l t h o u g h this c o n c l u s i o n is not d e m o n s t r a t e d clearly in Fig.
2 for the w e a k p r e c i p i t i n lines, they a p p e a r e d to fuse
c o m p l e t e l y w h e n the i m m u n o d i f f u s i o n plate was viewed
directly.
Further Characterization of the Purified CEA. O n e x a m i n a t i o n by i m m u n o e l e c t r o p h o r e s i s , the purified C E A developed at least 3 precipitin lines w i t h the u n a d s o r b e d antib o d y . W i t h the a d s o r b e d a n t i b o d y , 1 p r e c i p i t i n line developed in t h e / 3 - a n d 3,-globulin regions.
Disc e l e c t r o p h o r e s i s was c o n d u c t e d in 7% a c r y l a m i d e
gels. A f t e r s t a i n i n g with C o o m a s s i e blue, 2 m a j o r b a n d s
were l o c a t e d n e a r the origin of the s e p a r a t i n g gel, a n d these
were also seen in gels t r e a t e d with a p e r i o d i c a c i d - S c h i f f
~ 0.8
o
co
40
A
LU 0.6
30
.~
C
D
~
-
I0
5
200
_...a__--l---~---t--o---T---e
400
600
800
I000
1200
E F F L U E N T V O L U M E (rnl)
A
- 40
1.8:"
§
1.2-
[
,o
30
I
i!
0.8
i
0.(
--'
]
~
- 20
-I0
IX)
o.z
I00
/
200
300
"
400
iaJ
Z
m
0
1.6
u)
1.4
m
"~
1.2
m
B
,
3
1.0
50
0.6
40
/
0.8
]
30
,,U ",, V
o,
I00
200
\
300
EFFLUENT VOLUME (ml)
Chart 3. A, gel filtration of partially purified CEA on Sepharose-4B.
The proteins eluted by sugar from the Con A column were concentrated to
5 to 8 ml and chromatographed on a 2.5- x 90-cm column of Sepharose-4B
equilibrated with 0.05 M sodium acetate-acetic acid buffer, pH 6.5,
containing 0.15 M NaC1. The absorbance at 280 nm of the effluent was
monitored (solid line) and 15-ml fractions were collected. These were
assayed for CEA (broken line). B, gel filtration of CEA on Sephadex
G-200. The proteins pooled from the Sepharose-4B column were concentrated to 4 to 6 ml and chromatographed on a 2.5- x 90-cm column of
Sephadex G-200 equilibrated with 0.05 Msodium acetate-acetic acid buffer,
pH 6.5, containing 0.15 M NaC1. Solid Hne, absorbance at 280 nm of the
column effluent; broken line, the CEA level measured in 15-ml fractions.
3
i
0.2
z.o-
-,' ~ , - ,.
1400
1600
Chart 2. Affinity chromatography of CEA on Con A bound to Sepharose. The pooled fractions from the DEAE-cellulose chromatography
were passed through a 2.5- x 29-cm column of Con A bound to Sepharose.
The absorbance at 280 nm (solid line) of the effluent was monitored and
20-ml fractions were collected. At A, the column was washed with the citrate buffer described in "Materials and Methods"; at B, the eluting buffer
was made 0.1 M with c~-methyl-D-mannoside; at C, the column was allowed to stand with the sugar for 20 hr and then elution was continued.
The waiting period and further elution were repeated at D. Results of
assays for CEA are shown by the broken line.
s t a i n for o x i d i z a b l e sugars (26). S e v e r a l m i n o r b a n d s t h a t
m i g r a t e d a h e a d of the m a j o r ones were also visible with
the C o o m a s s i e blue stain. A n u n s t a i n e d gel was cut in h a l f
a l o n g the l o n g i t u d i n a l axis a n d p l a c e d on a n a g a r plate
p a r a l l e l to a strip of filter p a p e r s a t u r a t e d with the a d s o r b e d
a n t i b o d y to C E A . T h e p r e c i p i t i n line t h a t d e v e l o p e d in the
a g a r i n d i c a t e d t h a t d u r i n g e l e c t r o p h o r e s i s the a n t i g e n had
m i g r a t e d to the region o f the s t a i n e d b a n d s .
The immunological and electrophoretic measurements
i n d i c a t e d t h a t our purified p r e p a r a t i o n s c o n t a i n several
c o m p o n e n t s in a d d i t i o n to C E A . T h e m a j o r c o m p o n e n t s
a p p e a r to be g l y c o p r o t e i n s b e c a u s e t h e y were s t a i n e d by
b o t h C o o m a s s i e blue a n d the p e r i o d i c a c i d - S c h i f f reagent.
T h e p r e p a r a t i o n s h a v e an optical a b s o r p t i o n m a x i m u m at
N O V E M B E R 1975
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.
2931
R. J. Carrico and M.
Usategui-Gomez
278 nm which is further evidence for the presence of protein.
Detailed chemical characterization of the C E A isolated by the procedure presented here must await further
purification.
DISCUSSION
C E A is usually purified from tissue homogenates b y
procedures that use perchloric acid extraction (7, 16).
Although this glycoprotein is considered to be stable in acid,
we have found that 70% of the C E A in tumor homogenates
is lost in M perchloric acid (Table 1). Loss of the antigen
from solution was also a problem under relatively mild
conditions at neutral pH. For example, only 25 to 50% of
the C E A in tumor extracts was recovered during chromatography on DEAE-cellulose at pH 6.9 (Table 2). Furthermore, in preliminary investigations we found that some
purification methods, such as a m m o n i u m sulfate fractionation, gave very low recoveries.
Chu et al. (4, 5) have reported that 60% of the 125I-CEA in
their preparation was bound by Con A and the binding was
reversed by mannose. We used affinity chromatography to
purify CEA from crude tumor extracts and found that 90%
of the CEA was bound by immobilized Con A (Chart 2;
Table 2). Further fractionation of the glycoproteins by gel
filtration provided CEA that was about 10% pure. This
procedure was used successfully to isolate several preparations of CEA from 3 specimens of liver metastases.
Isoelectric focusing has been used to purify C E A and
separate different chemical species (6, 9, 23). When we
attempted to purify our preparation by this method with pH
3 t o 6 ampholytes, a major component of the C E A was
found in fractions with pH 1 to 3. Similar elution profiles
were reported by Coligan et al. (6) and are not very
satisfactory because the pH gradient was very steep in the
region of interest. When we used pH 2.5 to 4.0 ampholytes
in order to cover the low pH range more effectively,
precipitate formed in the region with pH 3.5 to 4.5, and the
soluble CEA, representing 30 to 40% of that applied to the
column, was mostly located above pH 5.5. Apparently, the
C E A precipitated in the acidic region, floated up the sucrose
gradient, and dissolved in the strongly alkaline region.
Possibly, previous workers did not observe precipitation of
C E A during isoelectric focusing because their preparations,
which had been exposed to perchloric acid, contained only
acid-soluble material.
Antiserum raised with the purified CEA was rendered
monospecific for CEA by adsorption with normal liver
proteins bound to Sepharose (Fig. 1). Adsorption was
conducted with immobilized rather than soluble proteins
because the latter method produces soluble antigen-antibody complexes that remain in the antibody preparation (3).
These complexes can interfere with subsequent immunological measurements, especially in radioimmunoassays where
the complexed antigens can exchange with radiolabeled
antigens (3). In some cases, such exchange could explain
observations of cross-reacting antigens when extracts of
normal tissues were examined (19, 23).
Both the adsorbed antibody to CEA purified at neutral
2932
Fig. I. Immunodiffusion patterns obtained with antiserum before and
after adsorption with insolubilized normal liver proteins, a, center well,
unadsorbed antiserum to purified CEA. Wells 1 and 4, crude extracts of
normal liver; Well 2, purified CEA; Well 3, crude extract of tumor tissue;
Well 5, buffer: Well 6, crude extract of normal colon: b, center well,
adsorbed antibody to CEA; Well I, crude extract of tumor tissue; Wells 2
and 3, crude extracts of normal liver and colon, respectively.The precipitin
patterns were developed for 36 hr at room temperature.
pH and the antibody to perchloric acid-treated CEA
produced 2 precipitin lines with crude extracts of tumor
tissue (Fig. 2). Occurrence of 2 immunochemically distinct
antigens has been reported earlier (15, 22, 25). The relative
levels of the 2 antigens in the CEA purified at neutral pH
were very similar to those in the crude extract (Fig. 2). The
CANCER RESEARCH VOL. 35
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.
Isolation of CEA
m a j o r antigen, however, was nearly undetectable in the
perchloric acid extract. Evidently, this antigen is nearly
insoluble, possibly as a result of d e n a t u r a t i o n , in M perchloric acid. The low solubility accounts, at least in part, for
the low recovery of C E A during acid extraction of t u m o r
h o m o g e n a t e s (Table 1).
Precipitin lines produced by antibody to C E A isolated at
neutral pH fused completely with those produced by
antibody to C E A purified by perchloric acid extraction (Fig.
2). The antigen observed in the perchloric acid extract
a p p e a r e d to be identical to the corresponding antigen
isolated at neutral p H (Fig. 2). Results of previous studies
have provided similar evidence that the i m m u n o l o g i c a l
properties of C E A are not altered by perchloric acid (2, 12),
but a detailed c o m p a r i s o n of both antibodies and antigens
has not been reported.
In their initial studies, Gold and F r e e d m a n (1 1) identified
C E A on the basis of its i m m u n o l o g i c a l properties. Later,
the antigen was purified by extraction with perchloric acid
and its solubility in this m e d i u m was used as a further
criterion for identification (16-18). Our experimental results
indicate that the immunological criterion is m o r e satisfactory because t u m o r extracts contain material(s) which is lost
during exposure to perchloric acid but which reacts with
antibody to C E A purified by extraction with this acid.
Several properties of the C E A isolated at neutral p H are
similar to those reported for the antigen isolated by acid
extraction. F o r instance, C E A isolated at neural pH
a p p e a r s to contain c a r b o h y d r a t e since it binds specifically to
Fig. 2. Immunodiffusion patterns obtained with adsorbed antibody to
CEA and antibody to perchloric acid-treated CEA. Wells 1 and 3 contained a perchloric acid extract of tumor tissue; Well 2, CEA purified at
neutral pH; Well 4, antibody to CEA purified by perchloric acid extraction; Well 5, adsorbed antibody to CEA isolated at neutral pH;
Well 6, crude extract of tumor tissue. One ml of the solutions used
in Wells 1, 2, and 3 contained the proteins isolated from about 20 g of
tumor tissue. The precipitin lines were developed at room temperature
for 42 hr.
at N e u t r a l p H
C o n A ( C h a r t 2). The antigen elutes from gel filtration
columns at a p p r o x i m a t e l y the s a m e position as r e p o r t e d
previously for p r e p a r a t i o n s purified by extraction with
perchloric acid ( C h a r t 3; Refs. 7 a n d 18). Finally, the
mobility on i m m u n o e l e c t r o p h o r e s i s was similar to that
reported for C E A by other investigators (7, 16, 18).
The clinical utility of C E A for detection of neoplastic
disease depends on a c c u r a t e quantitative m e a s u r e m e n t of
the antigen in blood. S o m e assays are conducted with
perchloric acid extracts while others are carried out without
special t r e a t m e n t of the serum or p l a s m a (10, 14, 25). The
low solubility of C E A in acid would be expected to decrease
the levels m e a s u r e d following acid extraction. F u r t h e r m o r e ,
perchloric acid a p p e a r s to extract s o m e species of C E A
m o r e effectively than others. These two points m u s t be
t a k e n into consideration in c o m p a r i n g results of clinical
m e a s u r e m e n t s carried out by different methods.
ACKNOWLEDGMENTS
The authors are grateful to Dr. C. W. Todd and Dr. M. L. Egan, who
generously provided CEA and antibody to CEA for the radioimmunoassay.
REFERENCES
1. Axen, R., and Ernback, S. Chemical Fixation of Enzymes to
Cyanogen Halide Activated Polysaccharide Carriers. European J.
Biochem. 18." 351-360, 1971.
2. Burtin, P., Chavanel, G., and yon Kleist, S. Demonstration in Normal
Human Plasma of an Antigen that Cross Reacts with Carcinoembryonic Antigen of Digestive Tract Tumors. J. Natl. Cancer Inst., 49.
1727-1728, 1972.
3. Carrico, R. J., Beagle, R., Usategui-Gomez, M., and Dienhardt, F. A
Radioimmunoassay Specific for the Antigenic Determinants Exposed
by Pepsin Digestion of Human lmmunoglobulin G. Immunochemistry, I1: 573-579, 1974.
4. Chu, T. M. Reaction between Carcinoembryonic Antigen (CEA) and
Concanavalin A. Federation Proc., 32." 986, 1973.
5. Chu, T. M., Holyoke, E. D., and Murphy, G. P. The Reaction between
Carcinoembryonic Antigen and Concanavalin A. Cancer Res., 34.
121-124, 1974.
6. Coligan, J. E., Henkart, P. A., Todd, C. W., and Terry, W. D.
Heterogeneity of the Carcinoembryonic Antigen. Immunochemistry
10: 591-599, 1973.
7. Coligan, J. E., Lautenschleger, J. T., Egan, M. L., and Todd, C. W.
Isolation and Characterization of Carcinoembryonic Antigen. Immunochemistry, 9: 377-386, 1972.
8. Davis, B. Disc Electrophoresis. II. Methods and Application to
Human Serum Proteins. Ann. N. Y. Acad. Sci., 121." 404-427, 1964.
9. Dyce, B. J., and Haverback, B. J. Free and Bound Carcinoembryonic
Antigen in Neoplasms and in Normal Adult and Fetal Tissue.
Immunochemistry, 11: 423-440, 1974.
10. Egan, M. L., Lautenschleger, J. T., Coligan, J. E., and Todd, C. W.
Radioimmune Assay of Carcinoembryonic Antigens. Immunochemistry, 9: 289-299, 1972.
11. Gold, P., and Freedman, S. O. Demonstration of Tumor-Specific
Antigens in Human Colonic Carcinomata by Immunological Tolerance and Absorption Techniques. J. Exptl. Med., 121: 439-462, 1965.
12. Gold, P., and Freedman, S. O. Specific Carcinoembryonic Antigens of
the Human Digestive System. J. Exptl. Med., 122: 467--481, 1965.
NOVEMBER 1975
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.
2933
R. J. Carrico and M. Usategui-Gomez
13. Gold, P., Gold, M., and Freedman, S. O. Cellular Location of
Carcinoembryonic Antigens of the Human Digestive System. Cancer
Res., 28." 1331-1334, 1968.
14. Hansen, H. J., Lance, K. P., and Krupey, J. Demonstration of an Ion
Sensitive Antigenic Site on Carcinoembryonic Antigen: Assay Using
Zirconyl Phosphate. Clin. Res., 19: 143, 1971.
15. Kleinman, M. S, Harwell, L., and Turner, M. D. Studies of Colonic
Carcinoma Antigens. Gut, 12: 1-10, 1971.
16. Krupey, J., Gold, P., and Freedman, S. O. Purification and Characterization of Carcinoembryonic Antigens of the Human Digestive
System. Nature, 215: 67-68, 1967.
17. Krupey, J., Gold, P., and Freedman, S. O. Physicochemical Studies of
the Carcinoembryonic Antigens of the Human Digestive System. J.
Exptl. Med., 128: 387-398, 1968.
18. Krupey, J., Wilson, T., Freedman, S. O., and Gold, P. The Preparation of Purified Carcinoembryonic Antigen of the Human Digestive
System from Large Quantities of Tumor Tissues. Immunochemistry,
9: 617-622, 1972.
19. Kupchik, H. Z., and Zamcheck, N. Carcinoembryonic Antigen(s) in
Liver Disease. Gastroenterology, 63: 95-101, 1972.
20. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.
2934
21.
22.
23.
24.
25.
26.
Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem.,
193." 265-275, 1951.
Mach, J-P., and Pusztaszeri, G. Carcinoembryonic Antigen (CEA):
Demonstration of a Partial Identity between CEA and a Normal
Glycoprotein. Immunochemistry, 9: 1031-1034, 1972.
Martin, F., and Martin, M. S. Demonstration of Antigens Related to
Colonic Cancer in the Human Digestive System. Intern. J. Cancer, 6:
352-360, 1970.
Rule, A. H., and Goleski-Reilly, C. Carcinoembryonic Antigen
(CEA): Separation of CEA/Reacting Molecules from Tumor, Fetal
Gut, Meconium, and Normal Colon. Immunol. Commun., 2: 213-226,
1973.
Scheidegger, J. J, Une micro-mathode de l'immuno~lectrophor~se.
Intern. Arch. Allergy Appl. Immunol., 7:103-110, 1955.
Thomson, D. M. P., Krupey, J., Freedman, S. O., and Gold, P. The
Radioimmunoassay of Circulating Carcinoembryonic Antigen of the
Human Digestive System. Proc. Natl. Acad. Sci. U. S., 64: 161-162,
1969.
Zacharis, R, M., Zell, T. E., Morrison, J. H., and Woodlock, J. J.
Glycoprotein Staining following Electrophoresis on Acrylamide Gels.
Anal. Biochem., 30: 148-152, 1969.
CANCER
RESEARCH
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.
VOL. 35
The Isolation of Carcinoembryonic Antigen from Tumor
Tissue at Neutral pH
Robert J. Carrico and Magdalena Usategui-Gomez
Cancer Res 1975;35:2928-2934.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/35/11/2928
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1975 American Association for Cancer
Research.