journal of
biotechnology
Journal of Biotechnology 39 (1995) 165-173 ELSEVIER
The effect of fatty acids on hybridoma cell growth and antibody
productivity in serum-free cultures
M. Butler *, N. Huzel
Department of Microbiology, University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2
Received 19 November 1994; accepted 13 January 1995
Abstract
A murine B-lymphocyte hybridoma (CC9ClO) was adapted for growth in a serum-free medium. Supplementation
of the medium with k-unsaturated fatty acids (lo-50 PM) improved the cell yield in the order oleic/linoleic
> linoleic > oleic. Initial supplementation with the fatty acids also caused a significant increase (58%) in the
volumetric Mab titre. Continued growth of the cells in the fatty acid supplemented media over five culture passages
resulted in a gradual deterioration of the Mab yield concomitant with the appearance of lipid inclusions in the
cytosol. The higher Mab yield could be restored by a limited period of growth of the lipid-loaded cells in fatty
acid-free medium. These effects were independent of growth rate. This suggests that the optimal intracellular lipid
content is finely balanced between a reduced and an overloaded state. Specific glucose and glutamine utilisation
rates were unaffected by the presence of fatty acids. Also, the optimal glucose and glutamine concentrations for
growth were independent of the fatty acids.
Keywords: Cell culture; Hybridoma; Antibody; Fatty acid; Serum free
1. Introduction
There are numerous chemically defined
serum-free formulations that have been devel-
oped for the in vitro growth and productivity of
mammalian cells and particularly for antibody-
secreting hybridomas (Kawamoto et al., 1983;
Cleveland et al., 1983; Kovar and Franek, 1986;
Tharakan et al., 1986; Jager et al., 1988; Mu-
rakami, 1989). Several of these formulations con-
tain unsaturated fatty acids which are required
* Corresponding author.
Elsevier Science B.V.
SSDI 0168-1656(95)00017-8
for optimal cell growth. The fatty acids are nor-
mally bound to bovine serum albumin (BSA)
which acts as a carrier molecule and to substitute
for the role of a serum supplement (Ashbrook et
al., 1975).
Growth enhancement has been shown to be
more likely with c&unsaturated fatty acids rather
than saturated fatty acids (Rosenthal, 1987;
Barnes and Sato, 19801, although in some cases
combinations of saturated and unsaturated fatty
acids have been used (Spieker-Polet and Polet,
1981). The optimal fatty acid concentration for
growth enhancement of hybridomas is between
15 and 40 PM and higher concentrations may
166 M. Butler, N. Huzel/Journal of Biotechnology 39 (1995) 165-I 73
cause growth inhibition (Kovar and Franek, 1986).
Incorporation of fatty acids into phospholipids
is required in the synthesis of plasma membrane
structure. However, cells appear to be unable to
regulate fatty acid uptake and thus may become
overloaded with an intracellular excess often
recognised by the appearance of cytoplasmic lipid
droplets (Schneeberger et al., 1971). Because of
this, it is important to characterise the effect of
fatty acid addition to cultures on the long-term
growth of cells. In this report we examine the
effect of two unsaturated fatty acids (oleic and
linoleic acids) on the growth and monoclonal
antibody production of a murine hybridoma over
several culture passages.
2. Materials and methods
Cell line
The murine B-lymphocyte hybridoma
(CC9ClO), which secretes a monoclonal antibody
(a)
(IgG,) against bovine insulin, was obtained from
the American Type Culture Collection (ATCC
No. HB123). The cells were shown to be my-
coplasma-free by routine testing in an indepen-
dent laboratory (Rh Pharmaceuticals, Winnipeg).
Cuttut-e
The cells were grown in serum-free medium
consisting of DMEM/ Ham’s F12 in the ratio of
1:l v/v (Gibco). This basal medium was supple-
mented with 10 wg ml-’ bovine insulin, 10 pg
ml -’ bovine transferrin, 0.1% (w/v) Pluronic F-
68, 100 /.LM phosphoethanolamine, 10 FM
ethanolamine and 10 nM sodium selenite. The
cultures were also supplemented, where indi-
cated, with fatty acid-free bovine serum albumin
(0.1 mg ml-‘) complexed with a specific concen-
tration of oleic or linoleic acid. A BSA-fatty acid
concentrate ( X 1001 was formed by mixing 1 ml of
BSA (10 mg ml-‘) with varying volumes (l-10
~1) of fatty acid (200 mg ml-‘) in ethanol. The
concentrate was mixed for 30 min at room tem-
8;
(W
I .
OLL- / I 1._- 1
0 20 40 60 80 100
Oleic acid @M)
0 A_~~_ 1 .~_1._ 1 ---2L
0 20 40 60 80 100
Lmoleic acid (uM)
Fig. 1. The effect of oleic or linoleic acid on cell yield. Cells (CCYClO) were inoculated at 0.5 X lo5 ml-’ into 2 ml of culture media
supplemented with BSA-bound oleic acid (a) or linoleic acid (b) at O-100 PM. Daily viable cell counts were determined during the
culture period. Cell yields for single cultures are shown after 4 d growth. Control cultures established with and without BSA
showed no significant differences in cell yield.
M. Butler, N. Huzel /Journal of Biotechnology 39 (1995) 165-173 167
perature prior to addition to the cultures. Cul-
tures were established in T-flasks or in 24-well
plates and grown under a 10% CO, non-humidi-
fied atmosphere. All chemicals were purchased
from Sigma Chemical Co., unless otherwise
stated.
Sample analysis
Monoclonal antibody concentrations were de-
termined by HPLC using an affinity column (Pro-
AnaMabs from Hyclone). Glucose concentrations
were determined with a specific analyzer (Model
YSI 27, Yellow Springs Instrument Co.). Glu-
tamine was measured using an enzymatic assay
(Lund, 1985). Cell concentrations were deter-
mined by a Coulter counter.
Specific productivities
(AC//N dt) were calculated as the cumulative
concentration change divided by the viability in-
dex (integral of cell concentration and culture
time).
3. Results
3.1. Concentration-dependent effect of fatty acid
addition
The effect of fatty acid addition on the growth
of the CC9ClO cells was initially determined from
2 ml cultures incubated in 24-well plates. The cell
yield was determined after 4 d for cultures con-
taining O-100 PM oleic or linoleic acid (Fig. 1).
Control cultures were established with and with-
out supplemented BSA. The presence of BSA
had no effect on cell yield.
Cell growth was enhanced in the presence of
oleic acid at an optimal concentration of 50 PM
or in the presence of linoleic acid at an optimal
concentration of 25 PM. The growth enhance-
ment was equivalent for either fatty acid. How-
ever, concentrations of either fatty acid > 75 PM
caused growth inhibition. As a result of this ex-
periment it was decided to continue serum-free
cell growth in the presence of 25 PM oleic, linoleic
acid or an equimolar mixture of oleic and linoleic
acids.
Fig. 2. Growth of cells in serum-free media supplemented
with fatty acids. Cells (CC9ClO) were inoculated at 10’ ml-’
into 50 ml of culture media in 75 cm2 T-flasks. Daily viable
cell counts were determined from cultures supplemented with
25 PM oleic acid (W ), 25 PM linoleic acid ( A 1, 12.5 PM oleic
and 12.5 PM linoleic acid (0) or control cultures (0). Each
point represents a single independent cell count.
3.2. The effect of cell passage
Subsequent experiments were conducted in T-
flasks at an inoculum of lo5 cells per ml and cells
were sub-cultured after 4 d. Cell growth in the
presence of the unsaturated fatty acids is shown
in Fig. 2. Cell yields were higher in the presence
of fatty acids with oleic/linoleic > linoleic >
oleic > control cultures. Compared to the control
culture, the fatty acid supplemented cultures also
showed a higher growth rate, Jo (Table 1).
The relative differences in cell yield between
cultures was maintained over five passages indi-
cating a consistent enhancement of final cell den-
Table 1
Growth rate fi cd-‘) of hybridomas in media supplemented
with fatty acids (+ SEM for n = 4)
Media: control + oleic + linoleic + oleic/linoleic
0.63 f 0.03 0.71 k 0.04 0.80 k 0.07 0.85 k 0.07
+oleic acid
0 1
TimE (d)
3 4
168 M. Butler, N. Huzel/Journal of Biotechnology 39 (1995) 165-173
sity in cultures supplemented with fatty acids
(Fig. 3). Regular microscopic examination of the
cells over this growth period indicated the grad-
ual formation of cytoplasmic inclusions for cells
from fatty acid supplemented cultures. Staining
with Nile red showed that these inclusions had a
lipid content.
The antibody concentration of culture super-
natants was determined at the end of each cul-
ture period (Fig. 4). At passage 1, cells grown in
the presence of fatty acids produced a signifi-
cantly higher (58%) antibody concentration.
However, this difference was not significant after
passage 3. This was also reflected in a change in
the specific antibody productivity (qMMah) over cul-
ture passage. The qMab was not significantly dif-
ferent for the four cultures in passage 1, but by
passage 5 the productivity of cultures supple-
mented with fatty acids was significantly lower
(42%) than the control cultures (Fig. 5). Antibody
production of the controls was constant through-
out the five passages with a maximum concentra-
tion of 56.7 k 1.4 pg ml-‘.
nber
- ConfrY lZ%‘, +a,e,c CC,d _1:, +, “/,P i r<.,,i .IIl(_SI lh11>l1~~ II
Fig. 3. Cell yield in CCYClO cultures over five consecutive
passages. At each passage, cells were inoculated at 10’ ml-’
into 10 ml cultures in 75 cm* T-flasks. The final cell concen-
tration was determined after 4 d incubation. Error bars indi-
cate the SEM where n = 2 or 3.
1 3
Passage number
= C>lltrU i _,/r c “i,d , :_,, +,,nu,i c “rid , +Oi~,c,I,,,Oi~$i “-0
Fig. 4. Monoclonal antibody concentration in CC9ClO cul-
tures over five consecutive passages. At each passage, cells
were inoculated at lo5 ml-’ into 10 ml cultures in 75 cm2
T-flasks. The final antibody concentration was determined
after 4 d incubation. The error bars indicate the SEM where
n = 2 or 3. (1) Indicates determinations from single cultures.
3.3. Removal of fatty acid supplement
It was decided to investigate whether the ef-
fects of fatty acid supplementation on cell growth
and Mab productivity were reversible. Cells sub-
cultured for at least 20 passages in fatty acid
supplemented media were inoculated into control
media (without fatty acid). After two culture pas-
sages, there were still significant differences in
cell yields between cells previously grown in fatty
acid cultures and controls (Fig. 6). However, sub-
sequent passages reduced the cell yield of fatty
acid supplemented cultures, so that by passage 5
there was no significant difference in cell yields
between cultures.
The relative antibody production of cultures
was also affected by removal of the fatty acid
supplement (Fig. 7). No significant differences
were found in the final antibody concentrations
of cultures at passage 1, but by passage 2 and 3
the antibody yield from cells previously grown in
fatty acid supplemented cultures was significantly
higher (23%) compared to controls. However,
M. Butler, N. Huzel /Journal of Biotechnology 39 (1995) 165-173 169
this difference was eroded by passages 4 and 5.
These changes in final antibody concentrations
were also reflected in qMvlab values which were
significantly higher (36%) in fatty acid supple-
mented cultures at passage 3 (Fig. 8).
3.4. Utilisation rates of glucose and glutamine
Fig. 9 is representative of the pattern of glu-
cose and glutamine utilisation for a series of
cultures grown for five passages in the absence of
a fatty acid supplement or in the presence of
oleic acid, linoleic acid or a linoleic/oleic mix.
The example is of a culture supplemented with
linoleic acid at passage 5. Both glucose and glu-
tamine were almost completely utilised over the 4
d growth period. The initial rate of glutamine
utilisation (q ,,,) was calculated as 1.32 f 0.10
pmol per 10B cells per d for the cells grown in
linoleic acid supplemented media. The corre-
sponding value for the initial rate of glucose
utilisation (q,,J was 2.17 + 0.07 pmol per lo5
1
+Ol
2
Passage number
5
Fig. 5. Specific antibody productivity (qMab) in CC9ClO cul-
tures over five consecutive passages. At each passage, cells
were inoculated at 10’ ml-’ into 10 ml cultures in 75 cm2
T-flasks. The qMab was determined from the cell and anti-
body concentration changes over the 4 d of culture. The error
bars indicate the SEM where n = 2 or 3. (1) Indicates deter-
minations from single cultures.
2 4
Passage number
n
- con,ro, m +oleic acid m +ll”Olelc OCld [ +oie,c,l,n”le,c rmr
Fig. 6. The effect of fatty acid removal on final cell yields.
Cells which had been grown for at least 20 passages in fatty
acid-supplemented media were inoculated (10’ ml-‘) into
control growth media, i.e., with no fatty acid supplement. The
effect of growth in the absence of fatty acid was monitored
over five consecutive culture passages. The error bars indicate
the SEM where n = 2. (1) Indicates determinations from
single cultures.
cells per d. No significant difference was found in
the initial qgln or qaC values between the control
and fatty acid supplemented cultures. Further-
more, the qgln or qglC values were unaffected by
passage number.
3.5. The effect of initial glucose and glutamine
concentrations on cell growth
A matrix of cell cultures was established by
varying the glucose concentration (O-35 mM) in
one dimension and the glutamine concentration
(OS-24 mM) in a second dimension. The cell
inocula were taken from a single culture grown in
serum-free medium but in the absence of a fatty
acid. The experimental cultures were grown with
or without added fatty acid (Fig. 10). The relative
final cell yields were consistent under all condi-
tions with respect to fatty acid supplementation
with linoleic = linoleic/oleic > oleic =
control(+linoleic) > control. This showed that
M. Butler, N. Huzel /Journul of Biotechnology 39 (1995) 165-173
ce,,r denufi, 1111111 Passage number
m COnfrOl r2-.; tdri< I>/ Mi ;,;j tiirll k/S XII rl 1 +,-,*, ,,,IdI., /‘I
Fig. 7. The effect of fatty acid removal on monoclonal anti-
body concentrations. Cells which had been grown for at least
20 passages in fatty acid-supplemented media were inoculated
(10” mll’) into control growth media, i.e., with no fatty acid
supplement. The effect of the absence of fatty acid was
monitored over five consecutive culture passages. The error
bars shown indicate the SEM where n = 2. (1) Indicates
determinations from single cultures.
there was a significant improvement (18%) in cell
yield in cultures supplemented with linoleic acid
rather than oleic acid for cells adapted over sev-
eral passages to each respective media formula-
tion. The addition of linoleic acid to the control
cultures also improved the cell yield significantly
(60%), but only up to the level shown for cells
continuously passaged with oleic acid.
The presence of a fatty acid supplement in
cultures did not alter the concentration-depen-
dent effect of glucose or glutamine on cell growth.
The highest cell yields were found at a glucose
concentration of 17.5-25 mM and glutamine of
24 mM. Tolerance of such a high glutamine con-
centration suggests that these cells are relatively
insensitive to ammonia, a conclusion consistent
with previous experiments (not presented here).
4. Discussion
Fatty acids are often added in micromolar
concentrations to serum-free culture formulations
to maintain optimal growth of mammalian cells.
The cellular requirement for fatty acids is not
apparent in serum-supplemented cultures where
it is presumed that sufficient concentrations of
fatty acids are attached to serum proteins. In
serum-free media the fatty acids are normally
included as a complex with BSA (Ashbrook et al.,
1975).
In this report, we show that the addition of the
c&unsaturated fatty acids, oleic and linoleic, to
serum-free cultures of a murine hybridoma can
increase the growth rate and final cell yield signif-
icantly. However, the fatty acid concentration is
critical. The optimal concentrations for growth
stimulation by oleic and linoleic acids were 25
and 50 PM, respectively. At higher concentra-
tions the stimulatory effect decreased sharply and
at concentrations > 75 PM the fatty acids were
growth inhibitory.
There are extensive reports on the effects of
fatty acids on mitogen-stimulated lymphocytes
showing stimulation at low concentration (N 5
4
2 3 4
Passage number
II
= con’r”I ‘;:.‘I +o,eic “cNc i-.:2 +iilolelC “Cld 1 +o,eic,i Il”lCIC !/INS
Fig. 8. The effect of fatty acid removal on specific antibody
productivity (qMab). Cells which had been grown for at least
20 passages in fatty acid-supplemented media were inoculated
(10’ ml-‘) into control growth media, i.e., with no fatty acid
supplement, The effect of the absence of fatty acid was
monitored over five consecutive culture passages. The error
bars shown indicate the SEM where n = 2. (1) Indicates
determinations from single cultures.
M. Butler, N. Huzel /Journal of Biotechnology 39 (1995) 165-173 171
FM) (Kelly and Parker, 1979; Cuthbert and Lip-
sky, 1989) but inhibition at higher concentration
(> 30 PM; Calder et al., 1991). Similar concen-
tration effects have been reported previously for
hybridomas in serum-free medium (Kovar and
Franek, 1986), although the actual concentrations
causing stimulation and inhibition are different.
Jager et al. (1988) showed that the require-
ment of hybridomas for fatty acids is not immedi-
ately apparent following adaptation from serum-
supplemented medium. They reported that the
stimulatory effect was only shown following 10
culture passages in serum-free medium. This may
be related to a period in which an endogenous
store of fatty acid is utilised. This is consistent
with the data presented here for the CC9ClO
cells which had been grown in a serum-free, fatty
acid-free formulation for at least 10 passages
prior to the addition of a fatty acid.
At the concentrations used, linoleic acid in-
creased the growth rate of the CC9ClO cells more
than oleic acid but a synergistic effect was ob-
served with a combination of oleic and linoleic
acid providing the greatest growth stimulation
(Fig. 2). The growth stimulation of the fatty acids
was maintained over at least 20 culture passages.
When the fatty acid was removed from the cul-
ture medium, it took two culture passages before
the growth rate of the stimulated cells decreased
to the value of the control cultures (Fig. 6). Thus,
it would seem that a constant supply of fatty acid
in the medium is desirable to maintain a con-
stantly high growth rate.
The relationship between antibody production
and availability of fatty acid was more complex.
The initial addition of fatty acid caused an in-
crease in the antibody yield. However, the calcu-
lated specific antibody productivities (qMMab) were
not significantly different and the increased anti-
body yield could be explained by an increase in
cell concentration. After five passages of growth
in the presence of fatty acid, the qMulab declined
significantly compared to control cultures (Fig. 5).
This effect was reversed when the fatty acid was
removed from the medium, so that the antibody
productivity and yield increased after two pas-
sages but then declined to control levels after five
passages (Fig. 8).
The antibody productivity data is compatible
with the idea that there is an optimal intracellular
fatty acid concentration that maximises qMab. In
our experiments, the optimal intracellular fatty
acid concentration was achieved by