Cytotechnology 9: 131-139, 1992. 9 1992 Khm,er Academic Publishers. Printed in the Netherlands.

Enhancing monoclonal antibodies and hybridoma cell lines Sally S. Seaver Hygeia Sciences, 330 Nevada St., Newton, MA 02160, USA Received 30 March 1992: accepted in revised lbrm 6 August 1992

Key words: class switch, hybridoma, monoclonal antibody, secretion levels, stability, subcloning

Abstract We are interested in determining the range o f variants present in a cell population that can actually be isolated. We have used subcloning and sublining to search for variants with increased antibody stability, increased cell line stability to freezing and defrosting, increased cell population viability, increased antibody production and the ability to ~ o w in simpler media. This paper presents the case histories of several different hybridoma cell lines which required some property changed before they became production ready clones. We found that switching the class o f an antibody from IgG3 to IgGi did increase its stability, decrease its tendency to aggregate and allowed it to be used in a commercial diagnostic kit. We could isolate subclones that produced twice the level of antibody with a frequency of 1-3%. It was straight forward to isolate clones that were stable to freezing and defrosting or grew in a simpler media. We were not successful in increasing the maximum viability of a cell line. In conclusion, we have found that any population of hybridoma cells has natural variants with significantly enhanced properties that can be isolated.

Introduction It is obvious when characterizing any parameter in a population of cells by fluorescent activated cell sorting (FACS) that there is a wide range of responses clustered around a mean (Heath et al., 1990; Lee and Palsson, 1990). But there are always points significantly different than the bulk of the population. Do these points simply represent some error in the measurements or does the cell population contain variants with a wide range of properties? If so, how easily is it to select a variant with a specific property? To examine these questions we have looked for hybridoma clones that have enhanced properties, usually during routine subcloning of our cell

lines. For instance, we have subcloned cell lines in simpler media to try to select hybridoma lines that can be stably propagated in this medium and continue to produce at least as much antibody. Although there are notable exceptions (Heath et al., 1990; Tharakan et al., 1986), all too often monoclonal antibody production decreases when cell lines are grown in simpler medium (Tharakan et al., 1986). However, if an enhanced property occurs infrequently (less than 1 in 104 ) and can not be selected by traditional techniques that kill unwanted or unchanged cells, the cell o f interest may be isolated by sublining. In sublining populations of ceils are screened for the presence of a cell that has the desired characteristic. The popu-

132 lation o f cells that has a "desired" cell is then subdivided and the smaller populations are screened for the "desired" cell. Eventually, the population of cells is small enough that the "desired" cell can be isolated by subcloning. Sublining requires a specific assay that can detect a cell with the desirable trait amongst many cells lacking this trait. We used sublining to determine if changing the subclass of an antibody would increase its stability and improve its performance in a commercial diagnostic kit. W e used an isotype specific assay to isolate a subclone that had switched its class. W e could not directly select or screen for such traits as increased rates or levels of antibody secretion, increased cell line stability to freezing and defrosting and increased maximum cell viability. For these we simply screened the subclones generated for the desired trait. We expected to successfully select some enhanced variants since hybridoma lines have been routinely subcloned to "stabilize" antibody production. This paper presents some of our initial efforts in finding enhanced variants.

Methods Cell lines and culture media All hybridoma lines were the result of fusions of spleen cells from Balb/c or A mice with either SP2/0-Ag- 14 or P3-X63-Ag8.653 myeloma cells according to the technique of Kohler and Milstein (1975). Cells were cultured in commercially available media (Mediatech) with 1 - 1 0 % Fetal Bovine Serum. Liquid medium was purchased or made without l-glutamine added. The serum and 1glutamine were added to the medium just before it was used. Unless noted no additional additive were used. Cell viability was determined by trypan blue staining. Subcloning was done by limiting dilution at 0.3-0.5 cells/well in the desired medium the presence of Origen (Igen Inc., Rockville, MD) or freshly prepared spleen cells, macrophages or thymocytes. There was no difference in the num-

ber of antigen positive clones generated with cells subcloned on a feeder layer or on Origen. Macrophages were preferable for normal subcloning. T h y m o c y t e s were used in the class switch subcloning where the initial assay detected IgG1 production. Antibody levels in the culture supernatant were determined in a suicide culture. For this cells were diluted at least 1:5 to 1 x 105 ml -I in fresh medium and were cultured in flasks or plates for 5 - 7 days until the cell viability was less than 20%.

Assays Antibody levels were quantified in antigen coated plate ELISA assays. In the typical assay 100 gl of antigen per well was coated at 4~ overnight onto "high" binding microtiter plates (Nunc or Dynatech) at 0.1-0.5 gg ml -l. All washes were done using phosphate or TRIS buffered saline with 0.1% Tween. The plates were washed three times. If blocking was desired, 0.1% BSA in PBS/ Tween was incubated in the wells at 37~ for 1 hr followed by three washes. 50 gl of cell supernatant was incubated in the wells for 1 hr. The three washes were repeated. 50 tal o f goat anti-mouse conjugated to horseradish peroxidase ( Z y m e d or Jackson ImmunoResearch) was incubated in the wells for 1 hr. The plate was washed five times. 100 ~tl o f tetramethylbenezidine, hydrogen peroxide solution was added and the plate was read after 3 - 1 5 min. The standard curve was generated using the same purified monoclonal antibody. At least three different dilutions o f sample were used for quantification. When a standard deviation is reported, samples were run in triplicate at least three different dilutions in two separate determinations. Gel electrophoresis of purified antibody or ascites fluid was carried out in 1% agarose gels at pH 7.1 in 50 m M Hepes, 10 m M EDTA. The proteins were stained with Coomasie Blue. The stability of the antibody to heat (class switch experiment) was determined by incubating cell culture supematants at 55~ overnight. No special buffer was used to stabilize the antibody.

133 Antibody levels were determined as described above.

Results/discussion Stabilizing a m o n o c l o n a l antibody

It is well known that immunoglobulin isotype can affect antibody function (Kaminski et al., 1986; Lubeck et al., 1988; Steplewski et al., 1985) and that lymphocytes and hybridoma cell lines spontaneously switch their class/isotype with a frequency of 1:105 to 1:106 (Esser and Radbruch, 1990; Gritzmacher, 1989). We have used class switching to select monoclonal antibodies that were better suited for diagnostic assays. Below is an example. We had generated a murine IgG3 monoclonal antibody with an exquisite specificity and sensitivity for an infectious agent. However, our attempts to develop a commercial kit were plagued by the instability of the antibody upon storage and a low level of nonspecific aggregation which increased the rate of false positive results. Since stability to heat and cold and a tendency to aggregate are characteristic of IgG 3 antibodies, we decided to switch the class of the monoclonal antibody to an IgGl to determine if kit performance could be improved. Class switched variants have been isolated using FACS or sublining. In the former technique fluorescent anti-isotype antibody is used to label the surface of t h e lymphocyte which is then separated from the bulk population using a cell sorter (Beyreuther et al., 1981; Parham et al., 1983). In the latter technique cells are sublined at progressively lower numbers per well until the cells secreting antibody of the desired isotype can be subcloned (Muller and Rajewsky, 1983; Spira et al., 1984). We used this latter technique. An IgGi specific ELISA assay which could detect one IgGi secreting cell in the presence of 1000 IgG 3 secreting cells was generated by carefully screening lots of commercially available reagents until background levels were less than 0.020 O.D. By day 5 the signal generated by 1000 IgG3 cells

with one IgGi cell was greater than 0.15 O.D. In the actual sublining procedure about two million hybridoma cells were screened for class switch variants. Hybridoma cells were aliquoted into twenty 96 well plates at 1000 cells/well. Twelve out of 1920 wells gave a positive signal in the IgG1 specific assay. Only half of these reassayed positive. When these six wells were then sublined at 50 or 100 cells per well, four of the original 12 wells had multiple wells that were positive in the IgGj specific assay. Several positive wells from each subline were then subcloned. The antibody laden supematants from the subclones were screened in an antigen coated plate ELISA which was developed with an anti-IgGt conjugated antibody. All four sublines resulted in antigen positive, IgGl positive subclones. Antibodies from each subclone were then isotyped using two different sets of commercially available reagents, neither of which were used in the sublining and subcloning. Subclones from two of the sublines secreted IgG 3 antibody that now cross reacted with the specific anti-IgG I reagents (results not shown). The other two sublines resuited in subclones that secreted IgGl or the now "cross reacting" IgG3 antibody (e.g., clone 12/11/ 4) (Fig. 1). Agarose gel electrophoresis confirmed the results of the isotyping assays (Fig. 1). Antibodies from subcloned sublines that isotyped as IgG3 migrated identically to the IgG 3 parent (e.g., 12/11/4). Antibodies from subcloned sublines that isotyped as an IgGz migrated differently from the IgG3 parent. Antibody from the subclone 12/11/8 which isotyped as both an IgGl and IgG3 contained two bands, one which migrated like the parent IgG3 antibody and one which migrated like the class switched IgGl antibody. Two further assays were done to qualify a class switched clone for use in the commercial kit. First antibody secretion levels in culture were quantified and compared to the parent clone (Table 1). Many of the subclones had decreased levels of antibody expression. This is not unexpected; we find a similar phenomena during any subcloning. Second, an estimate of the stability of the antibody to heat was obtained after subjecting antibody laden culture supernatants to 55~ over-

134 n i g h t ( T a b l e 1). T h e p a r e n t IgG3 w a s c o m p l e t e l y i n a c t i v a t e d b y 5 5 ~ o v e r n i g h t . I n t e r e s t i n g l y antib o d y f r o m o n e s u b l i n e d I g G 3 s u b c l o n e , 12/1 1/4, s u r v i v e d the h e a t t r e a t m e n t . N o t all the c l a s s s w i t c h e d IgG1 a n t i b o d i e s w e r e h e a t s t a b l e (e.g., 12/11/11). B a s e d on t h e s e r e s u l t s the s u b c l o n e 12/9/2 w a s s e l e c t e d for i n c o r p o r a t i o n into the i m m u n o a s s a y kit. T h i s c l a s s s w i t c h e d a n t i b o d y h a d the s a m e s e n s i t i v i t y , s p e c i f i c i t y a n d a p p a r e n t a f f i n i t y as the o r i g i n a l IgG3 a n t i b o d y (results not s h o w n ) . H o w e v e r , the kits m a d e w i t h this a n t i b o d y no l o n g e r had high levels of nonspecific back~ound and were stable upon storage. Preliminary work with the 12/11/7 a n t i b o d y i n d i c a t e d that it b e h a v e d s i m i l a r l y to the 12/9/2 a n t i b o d y . W e h a v e c l a s s s w i t c h e d o t h e r I g M a n d IgG3 c l o n e s . T h e frequency of generating a stable, secreting IgGl c l o n e is a b o u t 1 : 10 6 cells.

Increased levels o f antibody secretion

Fig. I. Agarose gel electrophoresis of antibody. Purified monoclonal antibody (top four lanes) or diluted ascitic fluid (botton seven lanes) was run on 1% agarose gels at pH 7.1. The top three bands are three different IgG I monoclonal antibody standards. The class switched subclones are denoted by the three number designation. The subclass denoted on the fight hand side was determined separately in two different subclass assays (results not shown).

O u r first a t t e m p t s to i s o l a t e a s u b c i o n e w i t h an e n h a n c e d p r o p e r t y f o r w h i c h w e c o u l d not s e l e c t i n v o l v e d s i m p l y l o o k i n g for a s u b c l o n e w i t h an increased level of antibody secretion. The murine h y b r i d o m a , w h i c h h a d an S P 2 / 0 - A g - 1 4 m y e l o m a fusion partner, had already been subcloned twice. It h a d b e e n in c u l t u r e i n t e r m i t t e n t l y for s e v e r a l

Table 1. Antibody secretion levels and stability at 55~ Subclone

Class

Antibody (,tag ml-I) In culture a

After overnight @ 55~

Parent

3

70

0

12/9/1

3

32

0

(0%) b (0%)

12/9/2 12/I 1/4 12/1 I/6 12/11/7 12/11/8 12/11/11 12/12/3 6/15/4

1 3 3 I 1/3 I I 1

85 22 29 75 76 71 45 41

50 15 0 40 32 0.5 28 13

(59%) (68%) (0%) (53%) (42%) (1%) (62%) (32%)

a Antibody activity in 5 day culture supematants was determined in an ELISA assay as described in Methods. All cultures were started at 1 x 105 cells m1-1. bBased on the antibody activity in untreated supernatants.

135

Table 2. Antibody secretion levels for a parent hybridoma and its subclones

Suicide culturea (lag m1-1) Ascites fluidb (mg m1-1)

Parent

Subclone #70

Subclone #78

6.3 + .4 1.6 - 1.8

16 + .9 (254%) 3.4 - 4.2 (200 - 247%)

8.6 (137%)

aCells were diluted to 1 • 105 m1-1 in fresh medium and were cultured for 5-7 days until the viability was less than 20%. Antibody in the supematant was quantified in an antigen based ELISA assay as described in Methods. bThe range of antibody levels shown is from different taps and different injections.

years with no indication of a decrease in antibody production. H o w e v e r , it was a poor producer in both m o u s e ascites and in tissue culture (Table 2). Since there was no way to select for higher production levels, subcloning at 0.5 cells/well was done on several plates. Out of the 92 subclones obtained, 90% of which secreted specific antibody, the five subclones with the highest signals on the E L I S A screening assay were chosen for further study. Antibody production levels were determined in the s u p e m a t a n t s of a "suicide" culture of each subclone. (In a suicide culture the subclone is diluted to 105 cells per ml and cultured for 6 - 7 days until the cell viability is below 20%). T w o of the subclones had significantly higher levels of antibody production than the parent (Table 2). Subclone 70 had a 2.5 fold increase in antibody levels. The growth profile, including the doubling time and m a x i m u m cell density and viability, was similar to the original parent clone (data not shown). This greater than two-fold increase in antibody titers was also found in mice ascitic fluid (Table 2).

Increasing antibody secretion levels in medium

a

simpler

This SP2/0-Ag-14 based h y b r i d o m a had been subcloned once and been frozen for several years. It required that hypoxanthine (H), thymidine (T) and 10% FBS be in the medium. Although the line was very healthy and had a doubling time of less than 24 hr, the m a x i m u m cell viability was only 6 5 - 7 5 % . Attempts to increase this n u m b e r by changing the basal m e d i u m or specific nutrients were unsuccessful (data not shown). The

goals of this subcloning were to increase the level of antibody secretion, grow the ceils in a simpler m e d i u m and increase the m a x i m u m cell viability. The cell line was subcloned in the original m e d i u m containing 10% FBS and H T as well as in m e d i u m containing 5% FBS and HT. Thirty eight and thirty one subclones, both 97% positive in the antigen based E L I S A , were generated in the 10% and 5% FBS m e d i u m respectively. Suicide cultures were done in both the presence and absence of H T with seven subclones that grew in 5% FBS. Table 3 shows the antibody levels in suicide cultures grown in the absence of HT. Only one subclone, 2F10, showed a reduced level of antibody in the culture done without HT. Three subclones had increased levels of antibody by 50%; one subclone, 2 G I 0 , had double the antibody levels. All the cultures showed similar growth profiles. The m a x i m u m cell viability ranged f r o m 6 6 81%. The subclone 2G10 had a m a x i m u m cell viability of 7 8 - 8 2 % . This is higher than the parent clone. Whether this is significantly higher is not clear. N o n e of the subclones had a maxim u m cell viability greater than 90%. H o w e v e r , in this single subcloning we simultaneously enhanced two properties of the subclone 2G10; it secreted twice the level of antibody in a m e d i u m with half the level of serum and without HT.

Minimizing serum requirement while increasing antibody secretion The success of the previous e x a m p l e made us wonder how much we could reduce the serum in the m e d i u m in a single subcloning and still maintain or increase antibody secretion levels. The

136 Table 3. Antibody secretion levels in 50% less serum

FBS %

Suicide culture a (gg m1-1 ) Parent Subclone IA6 1E7 2F7 2F10 2GI0 3E4 3GI0

%

37

10

43 57 44 46 74 56 58

116% 154% 119% 124% 200% 151% 157%

5 5 5 5 5 5 5

apereentage figures are based on antibody levels in parent clone.

n e x t s u b c l o n i n g i n v o l v e d an S P 2 / 0 - A g - 1 4 b a s e d h y b r i d o m a w i t h a s i m i l a r h i s t o r y as the p r e v i o u s example. The parent had been subcloned once a n d ~ e w in m e d i u m with 10% F B S . Its m a x i m u m cell v i a b i l i t y w a s a b o u t 8 0 % . A t t e m p t s to i n c r e a s e this n u m b e r b y c h a n g i n g the b a s a l m e dium or adding nutrients was unsuccessful. T h e cell line w a s s u b c l o n e d in m e d i u m c o n t a i n i n g 5, 2 o r 1% F B S . T h i r t y one, 36 a n d 42 s u b c l o n e s , 100%, 9 7 % a n d 9 5 % p o s i t i v e in the a n t i g e n b a s e d E L I S A , w e r e g e n e r a t e d in the 5, 2 a n d 1% F B S c o n t a i n i n g m e d i u m , r e s p e c t i v e l y . F o u r o f the s u b c l o n e s s e l e c t e d in 2 % F B S a n d 7 o f the s u b c l o n e s s e l e c t e d in 1% F B S w e r e s a v e d . D u r i n g e x p a n s i o n o n e o f the s u b c l o n e s s e l e c t e d in 1% F B S d i e d . S u i c i d e c u l t u r e s o f the six r e m a i n i n g s u b c l o n e s g r o w i n g in 1% F B S i n d i c a t e d that

the c u l t u r e s d i e d 1 - 2 d a y s b e f o r e the p a r e n t c l o n e and that t h e i r m a x i m u m v i a b i l i t y w a s o n l y 5 6 6 6 % . A s a r e s u l t the a m o u n t o f a n t i b o d y s e c r e t e d w a s o n l y 1 8 - 5 6 % o f that s e c r e t e d b y the p a r e n t line (results not s h o w n ) . T h e m a x i m u m cell v i a b i l i t y o f the s u b c l o n e s i n c r e a s e d w i t h an i n c r e a s e in the s e r u m l e v e l u s e d for s e l e c t i o n . T h e v i a b i l i t y o f the c l o n e s s e l e c t e d a n d g r o w n in 2 % s e r u m w a s in the l o w 70s a n d t h o s e s e l e c t e d a n d ~ o w n in 5 % s e r u m w a s 8 0 85%. S u i c i d e c u l t u r e s w e r e d o n e for t h r e e o f the c l o n e s that w e r e o r i g i n a l l y s e l e c t e d in 1% F B S u s i n g m e d i u m w i t h e i t h e r 2 or 4 % F B S . I n c r e a s ing the s e r u m l e v e l i m m e d i a t e l y i n c r e a s e d the cell v i a b i l i t y , p r o l o n g e d the s u i c i d e c u l t u r e a n d inc r e a s e d the a n t i b o d y l e v e l s ( T a b l e 4). T h e 9 B 8 subclone has since been extensively propagated

Table 4. Effects of culturing subclones selected in 1% FBS in 2% or 4% FBS a

Clone

Parent Subclone 9B8 9C5 9F10

% FBS

Antibody

Max. cell viability

p.g ml-I

% of parent

10

39

100%

80%

2 4 2 4 2 4

92 74 83 67 40 45

236% 190% 213% 172% 103% 115%

80% 78% 79% 84% 66% 77%

aAntibody levels in suicide culture of 7 days.

137

Table 5. Stability of a 653 hybridomato freezinga FBS %

Parent Subclones 4C8 5B 11 7B2b 7El0b

Poe-freeze

Post freeze Day 0

Day 1

Final

10%

72%

42%

14%

72%

5% 5% 5% 2% 4%

70% 71% 87% 75% 82%

71% 54% 82% 56% 62%

48% 25% 68% 52% 60%

73% 68% 80% 57% 83%

aExpoessed as the percentage of viable cells as defined by trypan blue exclusion. bOriginally selected in 2% FBS.

in 2% FBS, maintained a viability in the low 80s and produced twice as much antibody as the parent clone.

Stability to freezing and defrosting Our major concern with this P3-X63-Ag8.653 based hybridoma, which had been subcloned three times, was its poor stability to freezing and defrosting (Table 5). The viability of the population usually dropped below 25% the first few days after defrosting. It was several weeks after defrosting before it could be diluted to 1 • 105 cells/ml without dying or having several days of no growth. This poor recovery was not due to the method use for freezing and defrosting the cells. Other lines frozen and defrosted at the same time using the same procedures did not have this problem (e.g., Table 5). We also wanted to reduce

the serum requirements of this cell line from 10% FBS. The hybridoma was subcloned in medium containing half (5% FBS) or 20% o f the serum (2% FBS). Even though many antigen positive subclones were generated in both 5% (55 clones, 96% positive) and 2% FBS (29 clones, 100% positive), most subclones selected in 2% FBS died. Two were rescued by culturing them at 4% FBS. The master subclone was selected after determining the subclones' stability to freezing (Table 5) and antibody production in suicide culture (Table 6). Each of the four subclones is illustrious of the range of variants found after a subcloning. The 4C8 subclone initially had high levels of secretion and adequate stability after defrosting. However, its m a x i m u m cell viability was never over 75% and the more it was cultured, the slower it gTew.

Table 6. Antibody secretion levels of a 653 hybridoma FBS %

Antibody ~g m1-1

Parent Subclone 4C8 5B 11 7B2a 7El0a

10

70

5 5 5 2 4

63 84 31 58 66

aOriginally selected in 2% FBS.

% of parent

90% 120% 44% 82% 94%

138 The subclone 5B 11 was a hardy subclone, proliferated rapidly and produced the most antibody, but its viability upon defrosting mirrored that of the parent clone. The 7B2 subclone which had been selected in the 2% FBS m e d i u m only bec a m e a hardy, rapidly proliferating subclone when the serum was increased to 4%. It was stable to freezing but its antibody secretion level was less than 50% of the parent clone. The 7 E l 0 subclone which was also selected in 2% FBS could be grown in 2% FBS but its performance was marginal. When grown in 4% FBS, 7 E l 0 b e c a m e the choice for production. Its viability was higher than even the parent clone, its antibody production was similar to the parent clone, and its stability after freezing was normal. It grew with a doubling time of less than a day and could be routinely cultured at 105 cells ml -l.

Conclusion We have found that switching the class of a monoclonal antibody can impact its properties. By switching the antibody from an IgG3 to an IgGl we were able to find IgGi subclones that were stable to heat or cold and no longer aggregated. This allowed us to obtain the required stability (which affects storage conditions and expiration dating) and specificity (no aggregation reduces the n u m b e r of false positive results) to launch our c o m m e r c i a l kit. The most critical step in class switching is ensuring that the subclass specific reagents do riot bind the parent antibody. W e were quite surprised to find the high frequency ( 1 - 3 % ) of obtaining a subclone that secreted twice the level of antibody. This high frequency was not due to a high level of nonsecreting cells in our original cell populations. In every subcioning cited greater than 90% of the subclones secreted antigen specific antibody. W e have not had the occasion to repeat the subcloning of any of these lines to determine if the antibody production levels can be increased further. It was also straight forward to generate these higher secreting subclones while reducing the level of serum at least 50%. H o w e v e r , there are

limits that are definitely cell line specific as to how low serum levels can be reduced and still obtain a stable subclone with the same level of antibody production. Subcloning a h y b r i d o m a line in simpler m e d i u m has b e c o m e the preferred way for our laboratory to adapt a clone to that medium. Stepwise reduction in the serum level is not only slower but often results in a decreased level of antibody production and a cell line that does not retain the ability to grow in simpler m e d i u m after freezing and defrosting. We have not tried subcloning directly into serum free medium because it has no production advantage over low serum for our diagnostic reagents. We were equally surprised to find that subcloning could readily improve a cell line's rate of recovery upon defrosting but not its overall viability. We have no explanation for these differences. H o w e v e r , our results indicate that a wide variety o f cell and product characteristics can be selected during routine subcloning if one sirnply takes the time to assay for the desired enhancement.

Acknowledgements I would like to a c k n o w l e d g e the following people who contributed to the work presented in this paper: Marianne Gesner, Chris Huang, Michelle Negrotti, Monica Strernpko and Philip W o n g at Hygeia Sciences and Bruce Brown and Tracey Walker at Hazelton.

References Beyreuther K, Bovens J, Dildrop R. Dorff H, Geske T, Liesegang B. Muller C, Neuberger MS, Radbruch A, Rajewsky K, Sablitzky F, Schreier PH, and Zaiss S (1981) Isolation and characterization of class switch variants of myeloma and hybridoma cells. In: Janeway C, Sercarz EE, and Wigzell, H (eds.), Immunoglobulin ldiotypes. ICN-UCLA Syrup. Mol. Cell Biology, Vol. XX (pp. 229-245) Academic Press, New York. Esser C and Radbruch A (1990) lmmunoglobulin class switching: molecular and cellular analysis. Annu. Rev. Immunol. 8: 717-735. Gritzmacher C (1989) Molecular aspects of heavy-chain class switching. Crit. Rev. lmmunol. 9: 173-200.

139 Heath C, Dilwith R and Belfort G (1990) Methods for increasing monoclonal antibody production in suspension and entrapped cell cultures: biochemical and flow cytometric analysis as a function of medium serum content. J. Biotech. 15: 71-90. Kaminski MS, Kitamura K, Maloney DG, Campbell MJ and Levy R (1986) Importance of antibody isotype in monoclonal anti-idiotype therapy of a murine B cell lymphoma. A study of hybridoma class switch variants. J. Immunol. 136:11231130. Kohler G & Milstein C (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256: 495-497. Lee GM and Palsson BO (1990) Immobilization can improve the stability of hybridoma antibody productivity in serum-free media. Biotech. Bioeng. 36: 1049-1055. Lubeck MD, Kimoto Y, Steplewski Z and Koprowski H (1988) Killing of human tumor cell lines by human monocytes and routine monoclonal antibodies. Cell, Immunol. 111: 107117. Muller CE, and Rajewsky K, (1983) Isolation of immunoglobin class switch variants from hybridoma lines secreting antiidiotope antibodies by sequential sublining. J. Immunol. 131: 877-881.

Parham P, Kipps TJ, Ward FE & Herzenberg LA (1983) Isolation of heavy chain class switch variants of monoclonal anti-DCI hyhridoma cell line: effective conversion of noncytotoxic lgG 1 antibodies to cytotoxic IgG 2 antibodies. Human lmmunol. 8: 141-151. Spira G, Bargellesi A, Teillaud JL and Scharff MD (1984) The identification of monoclonal class switch variants by sib selection and an ELISA assay. J. Immunol. Methods 74: 307-315. Steplewski Z, Spira G, Blaszczyk M, Lubeck MD, Radbruch A, lllges H, Herlyn D, Rajewsky K and Scharff M (1985) Isolation and characterization of anti-monosialoganglioside monoclonal antibody 19-9 class-switch variants. Proc. Natl. Acad. Sci. 82: 8653-8657. Tharakan JP, Lucas, A & Chau PC (1986) Hybridoma growth and antibody secretion in serum-supplemented and low protein serum-free media. J. Immunol. Methods 94: 225-235.

Address Jbr offprints: Sally S. Seaver, Ph.D, Hygeia Sciences, 330 Nevada St., Newton, MA 02160, USA. Current address: Sally Seaver, Ph.D., Seaver & Assoc., 174 Hawthorne Ln., Concord, MA 01742, USA. Tel.: 508-3699000; Fax: 508-369-9000.