Increased production of human granulocyte-macrophage colony stimulating factor (hGM-CSF) by the addition of stabilizing polymer in plant suspension cultures.

Abstract

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a glycoprotein that stimulates the production of granulocytes, macrophages, and white blood cells. Secretion of human GM-CSF from cell suspension cultures of genetically modified tobacco has been facilitated using natural mammalian leader sequences. At the mid-exponential growth phase (day 4 after the initiation of cell suspension culture), GM-CSF was detected in the medium at a maximum concentration of 180 microg l(-1). However, the secreted GM-CSF was unstable in the medium, and rapidly degraded after day 5. In order to stabilize the secreted GM-CSF, three stabilizing polymers were tested, polyethylene glycol, polyvinylpyrrolidone and gelatin. Gelatin was the most effective in stabilizing the secreted GM-CSF. Following the addition of 5% (w/v) gelatin, the maximum GM-CSF concentration reached 783 microg l(-1), a 4.6-fold increase over control.

Full text

Page 1

Journal of Biotechnology 96 (2002) 205–211
Increased production of human granulocyte-macrophage
colony stimulating factor (hGM-CSF) by the addition of
stabilizing polymer in plant suspension cultures
Jae-Hwa Leea, Nan-Sun Kimb, Tae-Ho Kwonc, Yong-Suk Jangb,
Moon-Sik Yangb,*
aBasic Science Research Institute, Chonbuk National Uni?ersity, Dukjindong 664-14, Chonju, Chonbuk 561-756, South Korea
bDi?ision of Biological Sciences, Chonbuk National Uni?ersity, Dukjindong 664-14, Chonju, Chonbuk 561-756, South Korea
cInstitute for Molecular Biology and Genetics, Chonbuk National Uni?ersity, Dukjindong 664-14, Chonju,
Chonbuk 561-756, South Korea
Received 23 October 2001; received in revised form 14 January 2002; accepted 14 February 2002
Abstract
Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a glycoprotein that stimulates the production of
granulocytes, macrophages, and white blood cells. Secretion of human GM-CSF from cell suspension cultures of
genetically modified tobacco has been facilitated using natural mammalian leader sequences. At the mid-exponential
growth phase (day 4 after the initiation of cell suspension culture), GM-CSF was detected in the medium at a
maximum concentration of 180 ?g l−1. However, the secreted GM-CSF was unstable in the medium, and rapidly
degraded after day 5. In order to stabilize the secreted GM-CSF, three stabilizing polymers were tested, polyethylene
glycol, polyvinylpyrrolidone and gelatin. Gelatin was the most effective in stabilizing the secreted GM-CSF.
Following the addition of 5% (w/v) gelatin, the maximum GM-CSF concentration reached 783 ?g l−1, a 4.6-fold
increase over control. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: hGM-CSF; Tobacco; Cell suspension culture; Gelatin; Protein stabilizer
www.elsevier.com/locate/jbiotec
1. Introduction
Granulocyte-macrophage
factor (GM-CSF) is one of four specific glyco-
proteins that stimulate a population of granulo-
colony-stimulating
cyte-macrophage progenitor cells to generate
granulocytes, macrophages, and two important
types of white blood cells (Metcalf, 1985). GM-
CSF has clinical applications in the treatment of
neutropenia and aplastic anemia, and has greatly
reduced the infection risk associated with bone
marrowtransplantation
trophil count recovery (Dale et al., 1995). Murine
and human GM-CSF do not exhibit cross-species
biological activity or receptor binding, and dis-
by accelerating neu-
* Corresponding author. Tel.: +82-63-270-3339; fax: +82-
63-270-4334.
E-mailaddress: mskyang@moak.chonbuk.ac.kr
Yang).
(M.-S.
0168-1656/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S0168-1656(02)00044-5

Page 2

J.-H. Lee et al. / Journal of Biotechnology 96 (2001) 205–211 206
play only 56% amino acid sequence conservation.
Thus, they represent one of the less well-con-
served of the myeloid growth factors (Diederich et
al., 1991). GM-CSF has been expressed in various
foreign hosts, e.g. Esherichia coli (Libby et al.,
1987), yeast (Balland et al., 1998), Aspergillus
niger (Kim et al., 2000), mammalian cells (Chiou
and Wu, 1990), plant cells (James et al., 2000) and
is now produced for clinical use (Metcalf, 1991).
Although most interest during the past 15–20
years has focused on microbial and animal cell
cultures, plant cells are now considered to be
useful systems for large-scale protein production.
Plant cell cultures have several advantages com-
pared with microbial or animal cell cultures. They
have a reduced risk of mammalian viral contami-
nants, oncogenes and bacterial toxins, and they
can also produce correctly folded proteins and
assemble multimeric protein such as antibodies
(Fischer et al., 1999). As a result, reports on the
development of transgenic plants for foreign
protein expression are now accumulating (Mason
and Arntzen, 1995; Larrick et al., 1998).
Plant cell culture for foreign protein production
is not yet well-developed technologically. Foreign
proteins produced in plant cell cultures are har-
vested from plant biomass, from culture superna-
tant, or from a mixture of the two. Despite the
dilution of secreted protein to low concentrations
in a relatively large liquid volume, plant cell cul-
ture of foreign protein can be advantageous. Since
plant cell culture medium contains no added
proteins (complex medium), recovery and purifi-
cation of secreted proteins is relatively inexpen-
sive. However, accumulation of the secreted
protein depends largely on its stability in the
extracellular environment, and this is of particular
importance for proteins with complicated struc-
tures. The use of protein-stabilizing agents such as
polyvinylpyrrolidone (PVP) has been shown to
improve antibody production in suspension cul-
tures, suggesting that the medium is the site of
antibody degradation (LaCount et al., 1997;
Wongsamuth and Doran, 1997). We demon-
strated similar results in studies of bioactive
hGM-CSF production in tobacco suspension cul-
ture, where the accumulation of hGM-CSF was
drastically reduced during the exponential phase
of cell growth. In this study, we attempted to
improve the yield of hGM-CSF in tobacco sus-
pension culture by testing three different polymer
stabilizing agents.
2. Materials and methods
2.1. Plant cell line and growth medium
Tobacco (Nicotiana tabacum L. cv Havana SR)
cells were transformed with Agrobacterium tume-
faciens LBA4404 harboring the hGM-CSF gene.
The hGM-CSF cDNA was placed between a
CaMV35S promoter containing a duplicated en-
hancer region (−417 to −90), including a cDNA
sequence from the coat protein gene of tobacco
mosaic virus and a nos terminator. Following
co-cultivation, the explants were transferred to
MS medium (Murashige and Skoog, 1962) supple-
mented with 0.1 mg l−11-naphthaleneacetic acid,
1 mg l−16-benzylaminopurine, 300 mg l−1
kanamycin, and 500 mg l−1cefotaxime for selec-
tion. The explants were then transferred into fresh
medium every 2 weeks. The developed shoots
were transferred to hormone-free MS medium
containing 300 mg l−1kanamycin and 500 mg
l−1cefotaxime to induce roots. Thereafter, the
transgenic plant, O64-8, was selected based on the
highest hGM-CSF transcript level among the
transformants (Lee et al., 2001).
2.2. Flask culture
Suspension cells were obtained from leaf-
derived calli of the transgenic plant O64-8. Cells
were cultured in 50 ml (in 300 ml flask) of MS
medium containing 1 mg l−1of 2,4-dichlorophe-
noxyacetic acid, 0.05 mg l−1of kinetin, and 3%
sucrose at 25 °C in a shaking incubator with a
rotation speed of 100 rpm. The pH of the medium
was adjusted to 5.8 with 0.5 M KOH. The suspen-
sion cell culture was kept alive by transferring a
fifth of a proceeding culture into fresh medium
every 7 days.
Gelatin (type B, Sigma) and polyethylene glycol
(PEG, Mw 3500, Sigma) were added to MS
medium before autoclaving. However, stock solu-

Page 3

J.-H. Lee et al. / Journal of Biotechnology 96 (2001) 205–211 207
tion of polyvinylpyrrolidone (PVP, Mw 360000,
Sigma) was filter-sterilized using a cellulose ace-
tate membrane syringe filter described in La-
Count et al. (1997).
The cells were harvested at regular time inter-
val to analyze cell growth. Fresh cells were ob-
tained by filtering culture broth with preweighed
filter paper (whatman No 1). After measuring
fresh cell weight, the cells were dried at 60 °C
until constant weight.
2.3. Quantitati?e analysis of hGM-CSF
The cultured broth was centrifuged at 10000
rpm for 5 min, and the supernatant was col-
lected for ELISA. For the detection and quan-
tification of hGM-CSF, an ELISA kit was used
accordingto themanufacturer’s
(PharMingen, Inc., USA). Sample concentrations
were determined by comparison to a standard
curve of recombinant hGM-CSF. MS medium,
PVP, PEG, gelatin did not interfere with the
ELISA analysis. All experiments were repeated
two times.
instructions
2.4. Protease acti?ity assay
The protease activity assay was performed ac-
cording to the method of Battaglino et al.
(1991). We modified the method by using casein
hydrolysate, keeping the temperature of the hy-
drolysis at 25 °C instead of 50 °C. The modified
procedure is as following. Each 200 ?l sample
was incubated with 200 ?l 1% casein hydrolysate
(USB) in 0.2 M Tris–HCl buffer (pH 5.8) at
25 °C. At 1 h, the reaction was terminated with
600 ?l of 0.4 M trichloroacetic acid (TCA). Af-
ter incubation at 4 °C for 5 min, the precipi-
tated proteins were removed by centrifugation at
6000 rpm for 5 min, and the optical density of
the TCA-soluble fraction was measured at 280
nm. One unit (U) of protease activity was
defined as a change of one absorbency unit per
hour at 280 nm for 1 ml reaction precipitation
mixture asdescribed
protease activity was expressed as U l−1.
above. Extracellular
3. Results and discussion
3.1. Batch culture
The kinetics of transgenic cell growth and
hGM-CSF production is shown in Fig. 1. Plant
cell culture followed the standard phases of cell
growth, with a lag phase of 2 days, and an
exponential phase between days 2 and 8, followed
by a stationary phase. Cells grew well under the
experimental conditions and reached a maximum
Fig. 1. Cell growth and time course of extracellular hGM-CSF
production during batch suspension culture as detected by
ELISA. Fresh cell (?) and dry cell (?) weight during control
culture. The extracellular concentration (?) of hGM-CSF
during batch culture. Cell size index (?) was calculated by
dividing fresh cell weight by dry cell weight.

Page 4

J.-H. Lee et al. / Journal of Biotechnology 96 (2001) 205–211208
Fig. 2. Extracellular protease activity during batch suspension
culture.
3.2. Effect of stabilizing polymers on hGM-CSF
production
We examined the effect of three polymers
known to have protein stabilizing effects (Man-
ning etal., 1995;LaCount
polyethylene glycol
polyvinylpyrrolidone (PVP, Mw: 360000), and
gelatin (type B, from bovine serum albumin).
Among these, gelatin is most safe material, which
is widely used in food industry as additive. Fur-
thermore, gelatin had a significant stabilizing ef-
fect on hGM-CSF (Fig. 3), while the addition of
PVP (0.5–2 g l−1) and PEG (1–5%) did not
significantly increase the extracellular production
of hGM-CSF (data not shown). The extracellular
hGM-CSF concentration increased dramatically
with the addition of gelatin. Increasing the gelatin
concentration to 5% (w/v) resulted in a maximum
hGM-CSF concentration of 783 ?g l−1(4.6-fold
higher than that under the normal culture condi-
tion). Addition of gelatin did not affect cell
growth until the mid-exponential growth phase
(day 5). After day 5, gelatin had a negative effect
on cell growth (Fig. 4A), and at the same time
hGM-CSF production started to decrease. It ap-
peared that the culture medium environment
changed unfavorably for hGM-CSF stabilization
after the mid-exponential growth phase (day 5).
Gelatin had a significant effect on stabilizing the
etal.,1997):
3350), (PEG, Mw:
biomass concentration of 12.0 gdw l−1at day 9.
Cell size index (fresh cell weight/dry cell weight)
generally tended to increase as concentrations of
sugar were lowered. In our study, Cell size index
is proportional to water content of cells, which is
influenced mainly by osmotic pressure and cell
growth rate. As shown in Fig. 1B, cell size de-
creased during lag phase, after which cell size was
almost constant until day 4. After mid-exponen-
tial phase, cell size increased rapidly since cell
growth rate decreased (Fig. 1B). The extracellular
concentration of hGM-CSF reached 180 ?g l−1at
the mid-exponential phase (day 4), but rapidly
decreased after day 5. It is assumed that this was
mainly due to degradation of the secreted protein
by secreted protease. As shown in Fig. 2, protease
activity increased rapidly after day 5 (1.8-fold
increase) as expected. Bonner et al. (1988) also
identified protease activity in extracts from sus-
pension-cultured Nicotiana sil?estris cells. In these
cells, protease activity was highest during the
stationary and lag phases. Terashima et al. (1999)
also reported that the decrease of sucrose concen-
tration in the latter period of cell culture caused
the release of sulfhydryl protease. Thus, limiting
protease activity by altering either the medium
conditions or the physiological state of the cells
appears to be important for stabilizing secreted
hGM-CSF.
Fig. 3. Effect of various concentration of gelatin on the
extracellular hGM-CSF production during batch suspension
culture.

Page 5

J.-H. Lee et al. / Journal of Biotechnology 96 (2001) 205–211 209
Fig. 4. Effect of 2% gelatin on cell growth and intracellular
concentration of hGM-CSF during batch suspension culture.
(A) Fresh cell (?) and dry cell (?) weight during control
culture. Fresh cell (?) dry cell (?) weight during batch
culture with 2% gelatin. (B) The concentration of intracellular
hGM-CSF was calculated by measuring hGM-CSF extracted
from 1-g fresh cell without gelatin (?) and with 2% gelatin
(?).
dry cells at day 4 (Fig. 4B), which is 15% lower
than that of control. The hGM-CSF concentra-
tion then started to decrease rapidly, apparently
because the intracellular environment changed
unfavorably.
In order to confirm the positive stabilizing ef-
fect of gelatin, we examined the effect of the
timing of gelatin addition (2%, w/v) on hGM-
CSF stabilization. As shown in Fig. 4, the hGM-
CSF concentration started to increase on the day
that gelatin was added. Maximum concentrations
were similar in each case. The addition of gelatin
on days 5 and 6 could not stabilize the hGM-CSF
before the end of culture. Therefore, the best time
for gelatin addition was found to be at the begin-
ning of a culture. In other experiments, gelatin
(2%) was added twice on day 0 and 5 (data not
shown). There was no difference between a single
gelatin (2%) addition (day 0) and double addi-
tions (day 0 and 5). It appeared that gelatin could
not restabilizehGM-CSF
started (Fig. 5).
after degradation
3.3. Degradation of hGM-CSF in the presence of
a stabilizing polymer
To confirm the stabilizing effect of gelatin, we
determined the residual amount of pre-produced
hGM-CSF under various conditions in cell broth
Fig. 5. Effect of the time of gelatin addition (2%) on the
extracellular concentration of hGM-CSF during batch suspen-
sion culture.
protein, while it had a negative effect on cell
growth and physiology. As shown in Fig. 4A,
after 9 days culture, the addition of 2% gelatin
resulted in a 0.4-fold decrease in the fresh cell
weight, and a 0.7-fold decrease in dry cell weight
compared with the control culture. In contrast,
the concentration of intracellular hGM-CSF was
similar with or without the addition of gelatin.
The maximum intracellular
hGM-CSF in 2% gelatin reached 30.9 ?g g−1of
concentration of