-
Abnormal Plasma Components in C3H Mice Bearing Spontaneous
Tumors*
ELIZABETH ESHELMAN MILLER AND PETER BERNFELD
(Bio-Researeh Institute, Cambridge 41, Mass.)
SUMMARY
The plasma proteins of C5tI mice bearing spontaneous mammary
adenocarcinoma were fractionated by zone electrophoresis on
cornstarch gel stiffened by additional amylose, and rabbits were
immunized with purified protein fractions. Precipitin re- actions
by the agar gel diffusion technic clearly demonstrated the presence
of an anomalous protein in the plasma of tumor-bearing mice which
is absent in normal control animals. Differences in concentration
of certain proteins, in particular of a-globulins, between the
plasma of normal and of tumor-bearing animals were also
evident.
In 34 out of 37 instances, tumor-bearing animals were correctly
identified by the observation of an extra precipitin line, whereas
6.5 per cent of the observations were doubtful and only 1.5 per
cent were incorrect. With another rabbit antiserum, producing a
weaker extra line, 74 per cent correct, 14 per cent doubtful, and
1~ per cent incorrect observations were obtained. The significance
of these data lies in the potentiality of this procedure as a
diagnostic tool.
Plasma mucoproteins belonging to the group of a-globulins have
been found in increased amounts in patients and experimental
animals with neo- plastic disease (1, 9, 1~-14). The first paper of
this series (1) has demonstrated statistically significant
increases of a-globulin in Sarcoma 180-bearing C57BL/6 mice.
Preliminary studies with rabbit antisera to a-globulin fractions
from Sarcoma 1- bearing A/Jax mice have indicated the presence of
certain antigens in the plasma from tumor-bearing animals absent in
normal animals of the same strain (10).
Since spontaneous tumors more closely re- semble the conditions
found in human disease, ex- periments were initiated with the
intention of searching for anomalous proteins in the plasma of CJH
mice bearing spontaneous mammary gland adenocarcinoma.
Electrophoretic studies of whole plasma in this host-tumor system
had revealed no increases in the a-globulin components (6). 1
Con-
* Supported in part by research grant C-885~ from the Na- tional
Cancer Institute, National Institutes of Health, U.S. Public Health
Service, and by a gift from Maurice Gordon, Boston,
Massachusetts.
1 p. Bernfeld, unpublished data.
Received for publication January 15, 1960.
sequently, immunological technics were selected as a tool for
the present study, because their sen- sitivity is considered to be
far superior to tha t of the electrophoretic methods. In the
expectation of increasing the antigenicity of small or minute
amounts of anomalous plasma proteins, purified plasma fractions
have been used for the prepara- tion of antisera instead of whole
plasma, and alteration of biological activity during the purifica-
tion was reduced to a minimum by the use of mild fractionation
procedures, such as zone electro- phoresis.
The present paper describes changes occurring in the plasma of
CSH mice bearing spontaneous tumors and reports on the reliability
with which individual plasma from tumor-bearing mice could be
distinguished from normal plasma by the detec- tion of an antigenic
component absent in normal controls.
MATERIALS AND M E T H O D S
Plasma.--The plasma was obtained from inbred CSH mice purchased
from the R. B. Jackson Memorial Laboratories, Bar Harbor, Maine.
Blood was drawn from retired female breeders bearing spontaneous
mammary gland carcinomas
1149
Research. on June 13, 2021. © 1960 American Association for
Cancercancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
1150 Cancer Research Vol. 90, September 1960
ranging in size from 1 to 8 gan. and, for the con- trols, from
normal males which had been mated to the females, as well as from
young C3H males and females, 5-6 weeks old. Blood was collected by
severing the brachial plexus and drawing the blood into a pipette
wetted with saturated sodium citrate.
Fractionation of plasma proteins.--The method for fractionation
of plasma by zone clectro- phoresis on a cornstarch gel stiffened
by additional amylose has previously been described (2). In the
present experiments a commercial preparation of dried amylose
called "Superlose ''2 was used, and a concentration of 3 per cent
Superlose was found to give a gel of suitable consistency for
fractiona- tion.
Superlose was suspended in 1.5 times its weight of ethyl alcohol
and put into solution in distilled water by autoclaving the mixture
for 1 hour at 17 lbs. per sq. in. pressure, whereby most of the
ethanol evaporated. The hot solution was diluted with enough warm
water and veronal citrate buffer of pH 8.6, 0.1 ionic strength, to
give a final concen- tration of 0.05 ionic strength. Filter cel and
dry cornstarch were immediately added to prepare the gel. After
electrophoresis of the plasma in the gel for approximately 60 hours
in a field of about 1 volt/era, the gel was cut into 16 segracnts,
�89 or 1 era. wide.
The protein was recovered from each segment by squeezing the
liquid out of the gel through Whatman No. 1 filter paper on
centrifugal filters with Buchner funnels No. 0, fitted into
aluminum adapters and bottles for ~50-ml. trunnion cups of a
refrigerated International centrifuge. Thirty minutes of
centrifugation at 3,000 r.p.m, yielded a clear filtrate and an
almost dry starch cake on the filter.
Characterization of the protein fractions.--The eluate from each
segment was analyzed by paper electrophoresis for uniformity and
mobility of protein at pH 8.6 and for total protein according to
the method of Ix)wry (8). Eluates which thus were found to contain
components of the same mobilities were combined, This resulted in
the col- lection of eight fractions, representing the major
electrophoretically distinguishable plasma pro- teins, or mixtures
thereof, i.e., (a) albumin, (b) albumin with al-globulin, (c) al-
and a2-globulins, (d) a2-globulin, (e) a3 -and ~-globulins, (f) /~-
globulin, (g) fl- and 7-globulins, and (h) ~,- globulin.
Production of rabbit antisera.--Rabbits weigh- ing 4-6 pounds
were given a single injection of
2 Manufactured by Stein, Hall & Co., Inc., New York.
1 rag. total protein per pound of body weight mixed with
Freund's adjuvant, as described by Cohn (3). Depending on their
availability, each protein fraction was injected into one to four
rabbits. Blood was collected from the marginal ear vein of the
rabbit 6 weeks after injection.
Since the isolated fractions contained soluble starch and buffer
salts from the gel media, the maximal amount of these substances
expected to be found in any one fraction was injected into control
animals in the absence of protein, and neither toxicity nor
antigenicity was found.
Precipitin tests in agar.--The Ouchterlony agar plate diffusion
method (11) as modified by Korn- gold (7) was used for the
precipitin tests. Rabbit antiserum to the isolated protein
fractions was placed in the center cup of the agar plate and
various antigen solutions in four surrounding cups. Eight different
antigen concentrations covering a 1:512 range of dilution were used
for each antigen. On each agar plate a set of two different
antigens was opposed to one another at two corresponding dilutions.
This permitted the direct comparison between the two antigens of a
set, as well as the detection of extra lines in either one or the
other. I)recipitin lines were observed after incubation at room
temperature for ~ weeks. To eliminate sub- jective errors, this
procedure was carried out by two investigators, neither of whom had
knowledge of the identities or distribution of the individual
antigens on the plate.
The antiserum from each of the twenty inl- munized rabbits was
thus tested against sets of antigens consisting of pooled plasma
from tumor- bearing animals on one side of the agar plate and of
pooled plasma from normal old littermates on the other side. With
the use of those two rabbit anti- sera showing the most marked
difference between tumor-bearing females and male littermates, the
suitability of using plasma from normal old males as controls was
ascertained in a series of experi- ments in which individual plasma
from fifteen normal old male C3H mice was compared with in-
dividual plasma from each of ~ young male and female C3H mice of
5-6 weeks of age. To test the frequency of the occurrence of extra
antigens in plasma from tumor-bearing animals, as well as the
absence of these extra antigens in normal animals, plasma from
individual tumor-bearing mice was opposed to plasma from individual
nor- real animals on each of 66 agar plates. The antigen dilutions
were 1:8 and 1:32, which had been found in preliminary experiments
to be within the range of optimum concentration, and the antisera
were again those that produced extra precipitin lines with plasma
from tumor-bearing animals,
Research. on June 13, 2021. © 1960 American Association for
Cancercancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
~[ILI~ER AND BEt~:xFELD--..lbnornlal Plasma ("Oml~ouenls in Mice
1151
i.e., sera from rabbits immunized with an a._,- globulin
fraction and a fraction containing a~- anti ao.-globulins,
respectively. Eleven plates of the same experiment contained
mixtures of plasma from a normal mouse with tha t of a
tumor-bearing mouse on both sides. Tile existence of these plates
was unknown to the investigators reading the plates anti, thus,
considerably increases the sig- nificance of their
observations.
In some instances, isolated protein fractions have been used as
antigens.
Experiments in which plasma from individual nornlal young C'HI
males and females was tested against plasma of the old males, used
as control animals, showed tha t there were no significant
antigenic differences between the plasmas of these different types
of normal CSH mice.
Dist inct differences were observed, however, when plasma from
tumor-bearing and from nornml mice were compared with each other.
These differ- ences include marker variations in tile distant.es of
the precipitin lines from the center cup for
T A B L E 1
I)iFFI.:RENCFS IN ANTIGEN C()N( 'ENTRATI(JN FROM THE DISTAN(
'ES
BFTVCEFN I )RE( ' I I ' IT IN LINES AND (_'.ENTER CUP
]{ABllI T I M M V -
NIZ~:D W I T H :
Albumin
Albumin+ a l-t.;lobulin
c,.,-Globulin
~'- (; lobulin
3,-Globulin
.AtN TI ( ; EN
]) l ] ,U TION
(IN A ( ; A I ~
1 : It56 1:51~2
1 :64
1 :4
n o
1:3~
L I N E
IL
14
b C
R
b I t
b
a
b C
I t
b C
d
] ) I ~ T A N C E IN M M . I-'I~OM
( : E N T E R (-:UI' TO LINI ' : [4*
Tumor Normal plasma plasma.
1`2,4 11. (; 14.1 14.1
20 .8 1 7 8 1~.5 11 .5 11 .0 10 .7
17 .7 18 .7 15 .0 15 .0 15.1~ 16 .~ 1~.5 1"2.5
2"2.5 ~`2.5 20 .0 "20.0 14 .3 14 .3
~1 .5 ~1 .0 19 .0 19.0 16 .8 17."2 11.9 11.9
I ) I . F I - ' E HEN (lG
I ~ E T W E E N I)l,~l -
T A N ( ' E IN " / l i e
o sLmrsl"
- - 0 . 8
- - ~ 5 - 1 o - -0 3
-4-1.0
+1 .o
- 0 . 5
+0.4
* M e a s u r e d on 1.7-fold p h o t o g r a p h i c en la
rgemen t s .
t Pos i t i ve sign deno tes increased concen t r a t i on of
the c o m p o n e n t in the tu- m o r - b e a r i n g an imal
.
RESULTS
Eighteen out of the twenty rabbits immunized with protein
fractions gave good ant ibody reac- tions. Precipitating antibodies
were obtained to all eight fractions, but the antigenic complexity,
as observed by the Ouchterlony technic, varied greatly among the
different fractions. Albumin was the simplest fraction
antigenically, giving only two lines of precipitation, while the
7-globu- lin fraction was very complex, exhibiting at least six
distinct lines. When albumin or ~-globulin fractions were applied
on the agar as antigen against antisera from rabbits immunized with
ao_- globulin, lines of precipit:ation were obtainr with
~-globulin, bu t not with albumin.
normal and tumor antigens, as seen from the data in Table 1,
indicating considerable differences in the concentrat ions of the
components in normal and pathological sera. The most significant
differ- ence appears to be the fact tha t an t i seraf rom rabbits
imnmnizcd with certain a-globulin frac- tions from tumor-bearing
mice produced an adtti- tional precipitation line with plasma from
tumor- bearing mice which was absent when normal mouse plasma was
used as the antigen.
In every instance in which this extra line was observed its
position on the agar plates was found to be characteristic for each
of the two protein fractions used for immunization. A strong extra
line in addition to a single normal line was pro-
Research. on June 13, 2021. © 1960 American Association for
Cancercancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
115~ Cancer Research Vol. ~0, Sep t embe r 1960
duced by the antiserum of the rabbit immunized with a~- and
a2-globulins, as seen in :Figure 1. A much fainter extra line, in
addition to two or three strong normal lines, was observed with
antiserum from the rabbit immunized with a2-globulin.
By the presence of these lines, plasma from tumor-bearing
animals could be distinguished from that of normal animals with a
high degree of certainty. Table ~ shows that the high percentage of
correct readings, in which the observers identi- fied most of the
tumor-bearing animals, is opposed by only 1.5 and 1~ per cent wrong
negatives, and by no wrong positives. The statistical significance
of these results is considerably increased by the presence in this
series of experiments of a certain number of plates, unknown to the
observers, on which the antigen was the same on both sides.
which is absent in the plasma of normal mice of the same strain.
The present data have shown that such an anomalous plasma
component, a protein with the eleetrophoretie mobility of an
a-globulin, is present in animals bearing a spontaneous tumor,
while it had earlier been found also in mice with transplantable
tumors (10). The possibility that normal tissue antigens from the
host, or unspecific antibacterial antigens introduced by tumor
graft- ing, may be involved in this phenomenon is great- ly
reduced, however, in inbred animals bearing spontaneous tumors. The
use of old male litter- mates as normal control animals appears
justi- fiable, since no difference in the antigenic response was
found between old males, young males, and young females. If young
CgH mice had been used as controls, the possibility of age
differences would
TABLE
OCCURRENCE OF EXTRA PRECIPITIN LINES WITH PLASMA FROM
TUMOR-BEARING ~IICE
FRACTION USED
FOR IMMUNIZA-
TION OF RAnBIT
al-&a2-globulin al-&a~-globulin ~2-globulin
a.~-globulin
DISTRIBUTION" OF
ANTIGENS ON
AGAR PL.K TE $
T v s . ~ "
T + N vs. T + N T v s . N T + N vs. T + N
No. PLATES
O n c o r -
r e c t s i d e
37 34, 34 o 0, 0
~9 ~0, ~3 9 0, 0
EXTRA LINES OBSERVED t
Dou]~tful on ec rreet None
si, le
~, 3 1, 0 0 ,0 ~,~ 4, 4 5, 0, 0 9, 9
(In
si, te
0 ,0 0, 0 0, 0 0, 0
Av. RATING OF BOTH OBSERVERS
(PER CENT)
Correct Doubtful readings readings
9~2 6.5 100
74 14 lOO
Wrong readings
1.5
lC2
* T = Plasma from tumor-bearing animals. N = Normal plasma. T +
N = Mixed tumor and normal plasma. The two figures represent
individual readings by two independent observers.
These plates were correctly identified (lines and 4). I t can
also be seen from this table that the blind readings made by two
independent investigators agreed very closely.
DISCUSSION
The demonstration of an extra preeipitin line on agar diffusion
plates when certain rabbit antisera are brought into contact with
plasma from tumor- bearing mice appears to indicate tha t the
plasma of the tumor-bearing animals contains an antigen
not have been excluded. Whether the anomalous plasma component
is a unique feature in all tumor-bearing mice independent of mouse
strain and tumor type, or may occur in certain other conditions
totally unrelated to neoplastic growth, is not known at the present
time.
Eleetrophoretie analyses of the plasma proteins from C3H mice
bearing a spontaneous adeno- carcinoma did not reveal any increase
in a-globulin (6). 1 The detection of an anomalous plasma com-
ponent in this system by immunological methods
FIG. 1.--Preeipitin pat tern on an agar diffusion plate. Cen-
ter cup: rabbit antiserum to a protein fraction from the plasma of
tumor-bearing mice, containing al- and a2-globulins. Cups on upper
right and lower right: plasma from an individual C3H mouse bearing
a sp ontaneous adenoeareinoma, at the dilutions of 1:8 and 1:8~,
respectively. Cups on upper left and lower left: plasma from an
individual normal CSH mouse (male litter-mate), a t the same
dilutions.
Research. on June 13, 2021. © 1960 American Association for
Cancercancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
~[ILLER AND BERNFELD . . . . Abnormal Plasma Components in Mice
115'~
indicates, therefore, that the anomaly nmst be present at
concentrations below the limit of per- ceptibility of the
electrophoretic technic. I t may be concluded that an anomalous
plasma compo- nent, detectable by immunologic procedures, may be
present in a host-tumor system, although electrophoretic plasma
protein analyses have been found to be normal.
I t also appears evident from the present data (Table 1) that a
certain component in the a2- globulin fraction is increased during
the growth of a spontaneous mouse tumor, a fact which had
previously been shown in mice with transplantable tumors (1, 9).
Increased Inucoproteins belonging to the group of a-globulins have
repeatedly been reported in patients with malignancies (1~-14). The
increase of a-globulin in tumor-bearing A/Jax mice, which was
observed by moving boundary electrophoresis (10), was due no doubt
to both an increase of a normal plasma component and to the
appcarancc of small amounts of an anomalous protein.
In addition, the present results clearly show that a certain
constituent of 3,-globulin is in- creased, while another
7-globulin, at least two al-globulin components, and albumin arc
de- creased in the tumor-bearing animals. The de- crease in albumin
in tumor-bearing mice has been described (1) and is commonly
observed in patients with canccr.
Increases in antigenic components in whole plasma of
tumor-bcaring animals were observed by Darcy (4, 5). However, no at
tempt was made by this worker to fractionate the plasma and to
locate specifically the proteins responsible for the antigenic
reactions.
Sincc the use of whole plasma from tumor-bear- ing hosts as
antigen was found to produce highly complex antigenic patterns, 3
we believc that the results described in the present paper were due
in large part to thc fact that purified plasma frac- tions served
as antigens instead of whole plasma. Thus, the anomalous protein
was libcrated from the bulk of competing antigens and became a
major component of the material used for im- munization. I t
appcars possible that its anti- genicity was thereby increased.
Although the technic of purification uscd in this work may not
yicld pure protein fractions, this mild method nevertheless appears
logical and suitable for the purpose of concentrating anomalous
plasma pro- teins from tumor-bearing mice into partially puri- fied
fractions suitable for immunological studies.
P. Bernfeld and E. E. Miller, unpublished results.
The fact that the anomalous component could be detected with a
very high degree of accuracy in individual tumor-bearing animals
and not at all in normal controls makes this immunological pro-
cedure a model experiment of diagnostic tests. Its application to
humans, i.e., the combination of the immunological analysis with
electrophoretic, or possibly chemical purification of plasma
proteins, may become a significant approach to diagnostic
procedures in cancer patients.
A C K N O W L E D G M E N T S
We arc indebted to Dr. Max Goldfrank of Stein, IIall & Co.
for the generous supply of "Superlose" for our investiga- tion.
R E F E R E N C E S
1. BEm'~F~LD, P., and HOMBrJRGER, ~'. The Influence of Tu- inor
Growth on the Plasma Proteins in Mice. Cancer Re- search,
15:359-64, 1955.
~2. BEltNFELD, P., and NlSSr;LBAb'M, J. S. A Modified Method for
Protein Separation by Zone Electrophoresis on a Starch Gel. J.
Biol. Chem., 220:851-60, 1956.
3. Con~', M. In: A. C. COUCORAN (cd.), Methods in Medical
Research, 5:o~71-83. The Year Book Publishers, Inc., Chicago,
195~.
-~. I)AUCY, D. A. Immunological Discrimination between the Blood
of Normal and Tumor-bearing Rats. Nature, 176: 643-44, 1955.
5. Immunological I)emonstrat ion of a Substance in Rat" Blood
Associated with Tissue Growth. Brit. J. Cancer, 11:137-47.
1957.
6. ,JOItNSO.~, R. M.; ALBERT, S.; and PI_~'KUS, II. Serum Pro-
teins in Mice Bearing Induced and Spontaneous Mam- mary Gland
Carcinomas. Cancer Research, 14:830-36, 1954.
7. KOI~N(:;OLD, L. The I)istribution and Immunochcmical
Properties of Human Tissue and Tumor Antigens. Ann. New York Acad.
Sc., 69:681-97, 1957-58.
8. LowRy, 0. H.; ROSEBUO~'aH, N. J.; FARR, A. L.; ~nd RANDALI~,
n . J. Protein Measurement with the Foliu Phenol Reagent. J. Biol.
Chem., 193:265-75, 1951.
9. NISSELBAUM, ,I. S., and BERNFELD, P. The Properties of Two
(;lycoproteins Isolated from the Plasma of Normal and Tumor-bearing
Mice. J. Am. Chem., Soc. 78:687-89, 1956.
10. . Immunological Studies on Anomalous Plasma Proteins from
Tumor-bearing Mice. Proc. Am. Assoc. Cancer Research, 2 : ~235,
1957.
11. O~-CItTV:aLO~CY, O. Antigen-antibody Reactions in Gels. IV.
Types of Reactions in Coordinated Systems of Diffusion. Acta
Pathol. & Microbiol. Scandinav., 32:~231-40, 1958.
1~. PETEItMANN, ~[. L., and HOGNESS, K. R. Elcctrophoretic
Studies on the Plasma Proteins of Patients with Neoplastic Disease.
II. An Acid Protein Present in the Plasma. Cancer, 1:104-8,
I948.
1,~{. SEIIIERT, F. B.; SEIBEH.T, M. V.; ATNO, A. J.; and ChirP-
nELL, H. J. Variation in Protein and Polysaccharide Con- tent of
Scra in the Chronic Discuses, Tuberculosis, Sar- coidosis, and
Carcinoma. J. Clin. Investigation, 26:90-10-2, 1947.
14. WINZLP;H, R. J.; DOyEn, A. W.; MEIIL, J. W.; and SMYTIL i.
M. Studies on the Mucoprotcins of I Iuman Plasma. I. l
)etermination and Isolation. J. Clin. Investigation, 27: 609-16,
1948.
Research. on June 13, 2021. © 1960 American Association for
Cancercancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
g~
Research. on June 13, 2021. © 1960 American Association for
Cancercancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/
-
1960;20:1149. Cancer Res Elizabeth Eshelman Miller and Peter
Bernfeld Spontaneous TumorsAbnormal Plasma Components in C3H Mice
Bearing
Updated version
http://cancerres.aacrjournals.org/content/20/8/1149
Access the most recent version of this article at:
E-mail alerts related to this article or journal.Sign up to
receive free email-alerts
Subscriptions
Reprints and
[email protected] at
To order reprints of this article or to subscribe to the
journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take
you to the Copyright Clearance Center's (CCC)
.http://cancerres.aacrjournals.org/content/20/8/1149To request
permission to re-use all or part of this article, use this link
Research. on June 13, 2021. © 1960 American Association for
Cancercancerres.aacrjournals.org Downloaded from
http://cancerres.aacrjournals.org/content/20/8/1149http://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/20/8/1149http://cancerres.aacrjournals.org/