Journal o f Immunological Methods, 32 (1980) 297--304
297
© Elsevier/North-Holland Biomedical Press
HIGH F R E Q U E N C I E S O F A N T I G E N - S P E C I F I C H Y B R I D O M A S :
DEPENDENCE ON IMMUNIZATION PARAMETERS AND PREDICTION
BY S P L E E N C E L L A N A L Y S I S
CHRISTIAN STAHLI, THEOPHIL STAEHELIN, VINCENZO MIGGIANO,
JORG SCHMIDT and PAUL HJkRING
Pharma Research Department, F. Hoffrnann-La Roche and Cie. A.G., CH-4002 Basel,
Switzerland
(Received 17 September 1979, accepted 14 October 1979)
Hybridomas producing antibodies against soluble antigens have in most cases been
difficult to establish. After fusion of myeloma cells with spleen cells obtained from mice
immunized with a soluble protein, hybridomas secreting specific antibodies have been
observed to occur very rarely among non-specific hybridomas. We found that the
frequency of specific hybridomas correlates directly with the increase over background of
the frequency of blast and/or plasma cells in the spleen (measured by cell size analysis)
after antigenic stimulation. High yields of specific hybridomas were obtained simply by
following a novel immunization technique consisting of several conventional preimmunization courses followed by 4 very high doses of antigen in saline on each of the last
4 days before fusion.
INTRODUCTION
Since t h e i n t r o d u c t i o n o f cell fusion f o r t h e p r o d u c t i o n of antigen-specific
m o n o c l o n a l a n t i b o d i e s b y KShler a n d Milstein ( 1 9 7 5 ) , a large n u m b e r o f
h y b r i d o m a s have b e e n c r e a t e d in m a n y l a b o r a t o r i e s . T h e m e t h o d is e n t i r e l y
successful f o r p a r t i c u l a t e antigens, i.e., s u r f a c e antigens o f cells a n d viruses.
F o r i n s t a n c e , w i t h s h e e p red b l o o d cells (SRBC), h u m a n l y m p h o c y t e s , or
l y m p h o i d cell lines as antigens, even a f t e r a p r i m a r y i m m u n i z a t i o n , usually
b e t w e e n 5% a n d 20% o f h y b r i d o m a clones secrete a n t i b o d y d i r e c t e d against
t h e i m m u n i z i n g antigens, a p p r o x . 3 0 - - 5 0 % secrete n o a n t i b o d y a n d t h e
r e m a i n d e r secrete a n t i b o d y of u n k n o w n s p e c i f i c i t y (specific e f f i c i e n c y =
n u m b e r o f antigen-specific h y b r i d o m a s p e r t o t a l n u m b e r o f h y b r i d o m a s =
5 - - 2 0 % ; G. K 5 h l e r , pers. c o m m . ; M.H. Schreier, pers. c o m m . ; H. F o r s t e r
a n d B. T a k a c s , pers. c o m m . ) . F o r soluble antigens, h o w e v e r , t h e m e t h o d has
n o t p r o v e n as effective. Specific efficiencies < < 1 % have b e e n f o u n d in
fusion e x p e r i m e n t s w i t h the e n z y m e lactic d e h y d r o g e n a s e (K. R a j e w s k i ,
pers. c o m m . ) , w i t h c a r c i n o e m b r y o n i c antigen (CEA, R. Acolla, pers.
c o m m . ) , and in o u r initial e x p e r i m e n t s w i t h h u m a n c h o r i o n i c g o n a d o t r o p i n
( h C G ) a n d w i t h C E A , even t h o u g h several i m m u n i z a t i o n s in m o n t h l y inter-
298
vals before fusion had resulted in high serum antibody titers.
Antibody-secreting hybridomas are the result of fusion between plasma
and/or blast cells and m y e l o m a cells (Anderson and Melchers, 1978; KShler
and Shulman, 1978). Under the assumption that specific efficiency thus
depends on the size of specific, antigen-stimulated clones (blasts and plasma
cells) relative to the pool size of unrelated B-blast and plasma cells, an
immunization scheme was designed to maximize the number of specific
B-blasts at the time of fusion. As a result, the fraction of specific hybridomas
increased by 1--2 orders of magnitude over previous experiments: up to 40%
of all hybridomas were antigen-specific. Furthermore, we established
electronic cell size analysis as a quick m e t h o d to estimate extent of spleen
cell stimulation and predict the specific efficiency of the fusion.
We also compared the efficiency of fusion using an antibody-secreting and
a non-secreting myeloma.
MATERIALS AND METHODS
Mice
BALBc/J female mice were obtained from the Roche Institut ffir Biologisch-Medizinische Forschung, Fiillinsdorf, BL, Switzerland.
Immunization
Mice were immunized as indicated in Table 1. In the first immunization
(age: 6 weeks) hCG was injected subcutaneously in complete Freund's
adjuvant (Difco) supplemented with 109 IU of Bordetella pertussis, in the
second and third immunizations (age: 22 and 27 weeks, respectively)
intraperitoneally (i.p.) in incomplete Freund's adjuvant (Difco); in immunizations 4--8 (age: 11--12 months) hCG was injected in saline either i.p. or in
equal parts intravenously (i.v.) and i.p. as indicated in Table 1.
Myeloma lines
Two m y e l o m a lines were obtained from G. KShler: P3/X63-Ag8 (KShler
and Milstein, 1975) and Sp2/01-Ag (G. KShler, pers. comm.; Sp2/01-Ag does
n o t produce m y e l o m a globulin chains).
Cell fusion and plating
3 X 107 spleen cells and 3 X 107 myeloma cells were washed twice in
RPMI-1640 and pelleted together in a conical Falcon-2070 tube; to the
drained pellet 100 pl of 50% polyethylene glycol (PEG-4000) in RPMI-1640
were added at room temperature; after 60 sec 5 ml RPMI-1640 were slowly
added (initially dropwise) with gentle shaking to suspend the PEG; after a
resting period of approx. 15 min the cell clumps were slightly disrupted by
repeated gentle pipetting; the cells were seeded into 2 Costar Tissue-CultureClusters-3524 (24 wells of 3 ml each); each well contained 3 X 10 s spleen
cells, 3 X 10 s m y e l o m a cells and l 0 s mouse macrophages (peritoneal exudate
b
c
d
e
0.9%
0.2%
Sp0
46
3.2
10
0.24
7.5%
1.1%
Sp0
40
1.8
25
0.73
41%
10
10
10
1700
1
400 a
---130
50
50
10
2600
1
200
400
400
400
80
6.0%
H-10
412
H-9
407
0%
Sp0
46
3.2
1
0.02
0.6%
0.4%
10
10
10
1200
1
50
---210
H-11
414
413
0.5%
3.2%
10
10
10
1100
-200
400
400
400
81
Sp0
46
3.2
21
0.58
18%
H-12
X63
45
2.8
11
0.27
9.2%
H-13
Sp0
39
1.7
11
0.26
15%
H-14
0.5%
2.2%
0.5
5
50
450
1
200
400
400
400
62
460
X63
47
3.8
1
0.02
0.5%
H-15
F r a c t i o n o f cells in channels 51-99 or 71-99 as per c e n t o f t o t a l l y m p h o c y t e s (channels 14-99).
B a c k g r o u n d s are 1.6% and 0.3%, respectively (Fig. 3).
Out o f a total o f 48 wells.
F r o m t h e f r a c t i o n o f wells w i t h g r o w t h a n d t h e f r a c t i o n o f positive wells ( o u t o f 48) the m e a n n u m b e r s o f clones and o f positive
clones per well, respectively, w e r e calculated by the use o f Poisson statistics.
a i.p. + i.v.
Fusion
Mouse
Immunizations:
1
2
3
Titer (As78/min)
4 (day - - 7; i.p.)
5 ( d a y - - 4; i.p.)
6 (day - - 3; i.p. and i.v.)
7 (day - - 2; i.p. and i.v.)
8 (day - - 1 : i.p.)
Titer (day 0)
Size 51-99 cells b
above b a c k g r o u n d c
Size 71-99 cells b
above b a c k g r o u n d c
Myeloma
Wells with g r o w t h d
Clones per well e
Positive wells d
Positive clones per well e
Specific e f f i c i e n c y e
I m m u n i z i n g doses are all pg o f hCG.
I M M U N I Z A T I O N S C H E D U L E S AND T H E I R E F F E C T S ON S E R U M T I T E R , ON F R E Q U E N C Y OF S T I M U L A T E D S P L E E N C E L L S
AND ON F R E Q U E N C Y OF SPECIFIC H Y B R I D O M A S
TABLE 1
t~
¢.O
¢O
300
cells) in 2 ml of selection medium (RPMI-1640 supplemented with 15% fetal
calf serum and HAT). Media were changed twice a week starting with day 3.
Assay for positive wells
The solid-phase radioimmunoassay in Microtiter plates coated with the
antigen hCG is described in detail elsewhere (Miggiano et al., in preparation).
In brief, 50 pl hCG (1 pg/ml) in phosphate-buffered saline (PBS) was added
to each well and incubated for 2 h, followed by 200 pl of 3% bovine serum
albumin (BSA) for 2 h; the cups were then washed with PBS (4X) and incubated with 50 pl of test sample overnight; after washing with PBS (4X),
incubation with 50 pl of iodinated second antibody for 4 h (rabbit antimouse Ig at about 100,000 dpm per 50 pl) and final washing with PBS (4X)
they were counted; backgrounds were measured in wells coated with BSA
only.
Cell-size analysis
Since spleen cells swell upon standing, they were analyzed immediately
for size (with a Coulter-Counter attached to a Channelizer) in Isotone (all
supplied by Coulter Electronics) at low cell concentrations (500--1000
cells/ml, to eliminate coincidences); the cell suspension was repeatedly
pipetted through a thin orifice (tip of pipette) to disrupt possible clumps of
cells; the settings were 1 for 'amplification', 'aperture-current' and 'basechannel-threshold' and 100 for 'window-width'; the cells are scaled into 100
channels (0--99); channel number divided by amplification setting and
current setting is proportional to cell volume and will be referred to as 'size'.
Determination of antibody titers
Anti-hCG titers were determined by a spectrometric latex-agglutination
m e t h o d developed by H. Gallati (pets. comm.). Latex beads covered with
hCG aggregate in the presence of antibody. The aggregation can be followed
as an increase of absorption at 578 nm and its rate is directly proportional to
antibody concentration. 2 ml Tris-HC1 (pH 7.4, 0.2 M) + 50 gl hCG-latex (14
mg/ml, obtained from Pregnex ® test, Roche, Basle, Switzerland) + antibody
are mixed in a 3 ml cuvette and the rise in absorbance at 578 nm is observed.
The initial rate (AAsTs/At) is determined and its value is multiplied by the
factor of dilution of the antibody in 2 ml.
RESULTS
A series of 7 fusion experiments (H-9--H-15) was carried out with spleen
cells of 5 mice which were immunized according to different protocols
(Table 1). After 3 immunizations with hCG over a period of 6 months
according to similar (standard) procedure, the mice were finally boosted
according to different procedures. These consisted of varying antigen doses
given 2--5 times within the week before fusion (Table 1).
'301
RBC S.L.
I
B.C.
I
1
C
C
(..)
-
C
20
40
60
80
cell-size (channel number)
Fig. 1. Cell-size analysis o f spleen cells f r o m an u n i m m u n i z e d m o u s e (a a n d b) a n d of
m i t o g e n - s t i m u l a t e d spleen cells (c). Full-scales are: 1 0 , 0 0 0 cells (a a n d c) a n d 4 0 0 cells (b)
p e r c h a n n e l . M i t o g e n s t i m u l a t i o n : 106 spleen cells were i n c u b a t e d for 4 days in 1 ml
R P M I - 1 6 4 0 m e d i u m c o n t a i n i n g 50 ~g l i p o p o l y s a c c h a r i d e a n d 5 X 10 -s M 2 - m e r c a p t o e t h a n o l ; t h e curve (c) was c o p i e d point-wise f r o m a figure w i t h a d i f f e r e n t scale of t h e
x-axis. Red b l o o d cells (RBC), small l y m p h o c y t e s (S.L.) a n d blast cells (B.C.) are indic a t e d b y arrows.
Before fusion cell-size analysis was carried out with the spleen cells of
these animals (Table 1). The Channelizer scales cells into 100 channels
according to their volume. Channel number will be referred to as 'size' (see
Materials and Methods). A spleen cell preparation of an unimmunized mouse
shows 2 peaks, one.representing RBC (size 5), the other representing small
lymphocytes (size 20; Fig. 1,a). A shoulder of cells larger than small lymphocytes can be seen at higher vertical magnification (lower full scale;
Fig. 1,b), which presumably represent background plasma cells and macrophages. B-blast cells form a peak of size 70 as seen in the profile of lipopolysaccharide (LPS)-stimulated mouse spleen cells (Fig. 1,c). Microscopic
observation of the LPS-culture showed small lymphocytes together with cells
of up to 2--3 times their diameter. In analyzing spleen cells of m a n y immunized mice, no peaks were observed at larger sizes (up to size 400), and
the fraction of cells of sizes > 9 9 never exceeded 0.1%. Presumably sizes
10--15 to 30--40 thus represent small lymphocytes, sizes 20--30 to 40--60
plasma cells and sizes 40--60 to > 100 blast cells.
Spleen cells of the 5 mice were fused with the m y e l o m a lines P3/X63-Ag8
and Sp-0/2-Ag8 (referred to as X63 and Sp0). The cells from each fusion
were distributed into 48 wells at 3 × l 0 s spleen cells per well. Shortly before
cultures became confluent, supernatant was tested by solid-phase radioimmunoassay in microtiter trays. Each supernatant was measured in 2 wells
302
9-
(a)
(b)
8-
~7~55
5 4-
L,
21-
lb
~b
~o
10
40
20
30
40
well number
Fig. 2. Screening o f supernatants from h y b r i d o m a s o f e x p e r i m e n t s H-12 (a) and H-13 (b).
Binding of radiolabel is expressed as fraction o f radiolabel b o u n d w i t h a 1 : 1 0 0 0 diluted
m o u s e anti-hCG reference serum used as standard in each assay. Background binding (to
B S A ) was not subtracted (for illustration). X mark wells w i t h no growth. The backgrounds of each e x p e r i m e n t were analyzed statistically. For all 7 e x p e r i m e n t s the
standard deviations were b e l o w 30% of the mean of backgrounds. Thus, wells with radioactivity > 2x m e a n background were c o u n t e d as positive.
against hCG and in 2 wells against BSA (background). This background
measurement is necessary to identify hybridomas which make antibodies
that appear to bind specifically to the trays. Fig. 2 shows the results of the
screening of 2 fusions (H-12 and H-13). From the number of wells containing hybridomas the mean number of hybridomas per well was calculated.
After a similar determination of the mean number of positive hybridomas,
the specific efficiency (ratio of the 2 means) was determined (Table 1).
Supernatants of each well of the experiment H-9 were collected and pooled
over a period of a few weeks and then retested in the same way (results not
s/9
(a)
.4-
c
.~- 3 .2-
0
i
o13
C-'~m~11 i)115 i
1
i
i
2
3
fraction
4
(b)
119
.?
zm/lO ° 1 3
J
r
5 6
7
m
J
8 °/o
of size-51-99-cefls
05
1.0
15 °/o
fraction of size-71-99-cells
Fig. 3. Plot of specific e f f i c i e n c y vs. fraction of large spleen cells of sizes 51-99 (a) or
7 1 - 9 9 (b). Points are labeled w i t h the numbers o f the respective fusion e x p e r i m e n t s (see
Table 1); fusions w i t h x 6 3 (H-13 and H-15) are represented by open circles. The bars (c)
span the ranges o f cell sizes of 4 u n i m m u n i z e d control mice.
303
shown). All b u t 5 wells (90%) were still positive. Thus the possibility
could be excluded that the hCG-specific antibody detected in the first
screening had been produced by specific spleen cell foci (surviving clumps
of plasma cells producing anti-hCG antibody, which originated from one
stimulated lymphocyte). For the mice used in experiments H-9 to H-15 and
for 4 unimmunized control mice specific efficiency was plotted against the
fractions of spleen cells of 'sizes' 51-99 and 71-99 (Fig. 3). The intercepts of
the straight lines through these points give background values of 1.6% and
0.3% for the 2 cell-size ranges, in agreement with the background values
obtained from the control mice. Not shown is a series of experiments
(H-I--H-4), which differed from those listed in Table 1 only in the final
immunization. These animals received 3 or 4 days before fusion 50 or 100 #g
hCG in incomplete Freund's adjuvant. In these experiments among over
1000 clones recovered no positive clone was found (specific efficiency
<0.1%).
DISCUSSION
As specific efficiency depends on the fraction of freshly stimulated large
lymphocytes (Fig. 3), an immunization schedule optimal for fusion experiments has to generate the largest possible number of specific blast and/or
plasma cells. Maintaining high concentrations of circulating antigen up to
the day of fusion proved to be successful in forcing continued blast cell
proliferation. This was evident from the greatly increased numbers of splenic
blasts and specific hybridomas obtained (fusions H-9, H-12, H-14) as compared to those after a single high dose immunization (fusion H-10). Primary
stimulation of large numbers of weakly cross-reacting virgin clones as an
alternative explanation is incompatible with our finding of 1 IgM and 24 IgG
secreting cultures in fusion H-9. In recent experiments with other soluble
antigens, among them CEA, human IgG, and eukaryotic protein synthesis
initiation factor eIF-3, we found our immunization method equally
successful as with hCG, and the relation between frequencies of large spleen
cells and specific efficiency was confirmed. Thus, cell-size analysis is now
routinely used to test a spleen considered for fusion.
There seems to be a lower limit of circulating antigen concentration
required for the generation of enough specific blast cells in the spleen to find
at least one antigen-specific hybridoma. By extrapolation from fusions H-10
and H-11 the minimal single dose of soluble hCG to reach this concentration
is a b o u t 10 pg. Final immunization with 50--100 pg hCG in
incomplete Freund's adjuvant 3--4 days before fusion is insufficient to build
up this minimal concentration (fusions H-I--H-4), although this regime is
adequate to obtain high serum antibody titers.
Supernatants of hybridomas with the 'silent' Sp0 as fusion partner gave
much higher signals in our solid-phase RIA screening test than those with
X63 (Fig. 2). While the former secrete one sole variety of spleen cell-specific
304
a n t i b o d y , the latter secrete a mixture of 'scrambled' antibodies resulting
f r o m r a n d o m association of spleen cells and X63 specific heavy and light
chains. The 10 different antibodies pr oduced by such a h y b r i d o m a have
different affinities for their antigen in binding with 1 or 2 arms. Our
screening test depends on the cont i nuous association of the a n t i b o d y to hCG
on the plate during the 4 h incubation period with the developing antibody.
Studies on the dissociation rate constant k_l of single combining sites of
these antibodies have shown shorter half-lives of antigen-antibody interaction (unpublished results). Thus, only antibodies attached to 2 hCG molecules are e x p e c t e d to give a signal in the screening test. This prediction is
verified by the results of screening tests of the two pairs of fusions using the
same spleen cells but different myelomas (H-12/H-13 and H-14/H-15).
Within each pair of fusions the hybr i dom a s should essentially produce the
same set of antibodies differing only in the occurrence of scrambling. In
b o th pairs o f fusions the mean signals (mean of cpm a-hCG, minus mean of
cpm a-BSA) of h ybr i dom as producing scrambled antibodies were 16--17% of
those producing h o m o g e n e o u s antibodies.
ACKNOWLEDGEMENTS
The technical assistance of R ut h Fessler and Dagnija T h o r n t o n and the
typing by Susan Hafner are gratefully acknowledged. We t hank Dr. B. Glatthaar for generously supplying us with purified hCG, and Dr. S. Fazekas de
St. Groth and Dr. C. Steinberg for helpful discussions.
REFERENCES
Anderson, J. and F. Melchers, 1978, Curr. Top. Microbiol. Immunol. 81,130..
K5hler, G. and C. Milstein, 1975, Nature 256,495.
KShler, G. and M. Shulman, 1978, Curr. Top. Microbiol. Immunol. 81,143.