Castilho Leda R., Moraes Angela Maria (Ed.) Animal Cell Technology: From Biopharmaceuticals to Gene Therapy

Подождите немного. Документ загружается.

directly through the mitochondrial membranes: they must ﬁrst undergo

the three enzymatic reactions of the carnitine shuttle. The ﬁrst reaction is

catalyzed by a family of isozymes (different isozymes for fatty acids

having short, intermediate, or long carbon chain) present in the outer

mitochondrial membrane, the acyl-CoA synthetases, which promote the

general reaction:

Fatty acid þ CoA þ ATP $ fatty acyl-CoA þ AMP þ PP

Thus, acyl-CoA synthetases catalyze the formation of a thioester

linkage between the fatty acid carboxyl group and the thiol group of

coenzyme A to yield fatty acyl-CoA, coupled to the cleavage of ATP to

form AMP (adenosine monophosphate) and PP (Pyrophosphate). Fatty

acyl-CoA molecules, like acetyl-CoA, are high energy compounds. Fatty

acyl-CoA esters formed at the cytosolic side of the outer mitochondrial

membrane can be transported into the mitochondria and oxidized to

produce ATP, or they can be used in the cytosol to synthesize membrane

lipids. The second reaction of fatty acyl-CoA oxidation is a transesteriﬁca-

tion catalyzed by carnitine acyltransferase I, located in the outer mem-

brane. This enzyme catalyzes the attachment of the fatty acyl-CoA to the

hydroxyl group of carnitine to form fatty acyl-carnitine. The fatty acyl-

carnitine ester then enters the matrix by facilitated diffusion through the

acyl-carnitine/carnitine transporter in the inner mitochondrial membrane.

In the third and ﬁnal step of the carnitine shuttle, the fatty acyl group is

enzymatically transferred from carnitine to intramitochondrial coenzyme

A by carnitine acyltransferase II. This isoenzyme, located on the inner face

of the inner mitochondrial membrane, regenerates fatty acyl-CoA and

releases it, along with free carnitine, into the matrix. Carnitine re-enters

the intermembrane space via the acyl-carnitine/carnitine transporter. The

carnitine-mediated entry process is the rate-limiting step for oxidation of

fatty acids in mitochondria and is a regulation point.

Mitochondrial oxidation of fatty acids takes place in three stages. In the

ﬁrst stage (-oxidation), fatty acids undergo oxidative removal of succes-

sive two-carbon units in the form of acetyl-CoA, starting from the

carboxyl end of the fatty acyl chain. The fatty acid undergoes a number of

metabolic cycles sufﬁcient for the total conversion to acetyl-CoA. In the

second stage, the acetyl groups are oxidized to CO

in the TCA cycle,

which also takes place in the mitochondrial matrix. The third stage consists

of oxidative phosphorylation, where the reduced electron carriers NADH

and FADH

donate electrons to O

with the concomitant phosphoryla-

tion of ADP to ATP. The energy released by fatty acid oxidation is thus

conserved as ATP.

Although the role of lipids in cell culture has been rather neglected, it is

recognized that lipids are important for cell proliferation in serum-free

media. In this respect the importance of serum albumin has been recog-

nized as a carrier of supplemented fatty acids and lipids. Fatty acids are

often included in serum-free media to replace the growth-promoting

properties of the lipid components of serum. However, the effect of

selective fatty acids on cell growth is variable: they can stimulate (Jager et

al., 1988; Rose and Connolly, 1990; Grammatikos et al., 1994), inhibit

92 Animal Cell Technology

(Calder et al., 1991), or have no effect on cell growth (Bailey and Dunbar,

1973; Spector et al., 1981; Cornwell and Morisaki, 1984).

The regulatory mechanism of cellular uptake of fatty acids appears to be

limited and so the composition of the intracellular lipids is likely to reﬂect

the availability of the fatty acids in the medium. This was shown for the

CC9C10 hybridoma (Butler et al., 1997) and for BHK and CHO cells

(Schmid et al., 1991). Thus, cells growing in serum-supplemented cultures

are likely to attain a fatty acid composition reﬂecting that of serum, in

which the predominant fatty acids are palmitic, stearic, oleic, and linoleic

acids at a ratio 2:1:3:1, respectively.

Over 90% of the fatty acid composition of the cells could be accounted

for by linoleic, palmitic, oleic, stearic, and arachidonic acids. Minimal

quantities of other fatty acids (C8:0, C10:0, C12:0, C14:0, C18:3) were also

determined but were less than 10 mol % in control cells and were de-

creased to less than 5 mol % after one passage of growth in either linoleic

or oleic acid (Butler et al., 1997).

Linoleic acid has been shown to enhance the proliferation of mouse

mammary epithelial cells by metabolism to arachidonic acid, which is a

precursor of prostaglandin E

(Bandyopadhyay et al., 1987). However, the

mechanism of growth promotion of the unsaturated fatty acids in culture

may be related to their importance in the synthesis of cellular membranes

(Rintoul et al., 1978; Rockwell et al., 1980), which may have a signiﬁcant

effect on membrane ﬂuidity (Calder et al., 1994).

Unsaturated fatty acids (linoleic or oleic acid) have been shown to be

essential for hybridomas in serum-free cultures, as they signiﬁcantly

enhance the cell yield and monoclonal antibody (mAb) productivity

(Butler and Huzel, 1995; Butler et al., 1997). When Butler et al. (1999)

cultured the murine hybridoma CC9C10 in spinner ﬂasks at high shear

rates to determine the effects of fatty acids, they obtained up to threefold

enhanced cell yields in 25 M linoleic acid or 50 M oleic acid compared

with fatty acid-free control cultures. The half-lives of viable cells were

2.38 and 3.63 h, respectively, for cultures containing 25 and 50 M linoleic

acid, compared with the control culture half-life of 1.97 h (Butler et al.,

1997). At a higher concentration (over 75 M), they found that cell yields

fall below the level of the control cultures. Hexanoic, lauric, margaric, and

stearic acids had no effects on cell growth over the concentration range

they tested. Arachidonic, linolenic, octanoic, and myristic acids caused a

concentration-dependent inhibition of cell growth. Palmitic and decanoic

acids enhanced cell growth marginally (9%) but signiﬁcantly at 25 M.

Linoleic acid (25 M) enhanced growth more than oleic acid (25 M), but

an equimolar mixture of oleic and linoleic acid (25 M) stimulated growth

more than either fatty acid alone (Butler et al., 1999). The most likely

mechanism for growth enhancement in hybridoma cells is that the fatty

acids are required as components of phospholipids contained in mem-

branes. The effect of growth enhancement was reversible: when cells that

had been passaged continuously in the presence of fatty acids were re-

introduced into unsupplemented medium, the growth advantage over

control cultures was lost.

The effect of linoleic or oleic acid supplementation on mAbs production

has been studied. There was a signiﬁcant initial enhancement of antibody

Cell metabolism and its control in culture 93

titers in fatty acid-supplemented cultures but this enhancement effect

gradually decreased to insigniﬁcance. Subsequent transfer of the cells to

fatty acid-free cultures resulted in a transitory state in which higher mAb

titers were re-established for cells previously grown in the presence of the

fatty acids. The observed enhancement of mAb titer was signiﬁcantly

higher in cultures supplemented with linoleic acid or linoleic/oleic acid

mix than in those supplemented with oleic acid alone (Butler et al., 1997).

Partial inhibition of glycolysis by saturated fatty acids has been ob-

served in Ehrlich ascites tumor cells. Palmitate and acetate decreased

glutamate formation from glutamine (the ﬁrst step in glutaminolysis) in

this cell line, suggesting the possible role of fatty acids as an alternative

energy source (Butler et al., 1999).

Butler et al. (1997) cultivated a murine B-lymphocyte cell line

(CC9C19) and a myeloma (SP2/0) in T-ﬂasks in a serum-free medium

supplemented with linoleic acid. They determined that, in the presence

of linoleic acid, the glutamine consumption rate was signiﬁcantly lower,

and this correlated with a signiﬁcant decrease in the speciﬁc rate of

production of ammonia. The glucose consumption rate was slightly

lower but the speciﬁc rate of production of lactate increased signiﬁ-

cantly. This can be explained by a change in the formation of glycolytic

end-products. The rate of production of alanine was lower in linoleic

acid-grown cells. Alanine is produced from pyruvate by transamination

with glutamate in response to the need for sequestration of excess

metabolic nitrogen. It is likely that the decreased glutamine utilization

in linoleic acid-grown cells would lower the level of intracellular

glutamate. This would in turn decrease the transamination reaction and

cause a greater proportion of pyruvate to be converted to lactate. These

changes were reversible on the removal of the fatty acids (Butler et al.,

1997, 1999).

There may be several mechanisms for these metabolic effects. Unsatu-

rated fatty acids have been shown to directly activate speciﬁc enzymes and

to induce DNA synthesis and cytokine release from lymphocytes (Karsten

et al., 1994). The induction of speciﬁc protein synthesis may produce the

reduction in glutamine metabolism. The increase in the robustness of the

fatty acid-grown hybridomas in agitated cultures could be explained by a

high incorporation of the available fatty acids into the cellular phospho-

lipids fraction, which is a major structural component of the outer

membrane of the cell (Butler et al., 1999).

Cell-protecting additives, such as Pluronic

F-68, are commonly in-

cluded as a media component to reduce cell damage in gas-sparged,

agitated cultures. Despite the value of Pluronic

F-68 in protecting cells,

there is a potential problem in that Pluronic

may have some cytosolic

effects: it may reduce the yield of the producer cell line in culture and

there is a possibility of complexation or co-puriﬁcation with a cell

product. The replacement of Pluronic

with fatty acids, as well as being

positive for the half-lives and production of cultured cell lines, provides an

alternative method to protect producer cells in agitated cultures (Butler

et al., 1999).

The effect of the unsaturated fatty acids on protein secretion of cultured

cells can be dissociated from growth effects. Recombinant protein produc-

94 Animal Cell Technology

tivity from BHK cells seems to be stimulated by unsaturated fatty acids

independently of cell growth (Schmid et al., 1991).

4.3 Metabolic byproducts

4.3.1 Lactate

Cells in vivo convert a proportion of the pyruvate generated by glycolysis

into lactic acid, in a reaction catalyzed by the lactate dehydrogenase

enzyme. This lactic acid is largely secreted into the blood; some passes into

the liver, where it is reoxidized to pyruvate and either further metabolized

to CO

aerobically or converted to glucose. Much lactate is metabolized

to CO

by the heart, which is perfused by blood and can continue aerobic

metabolism at times when exercising skeletal muscles secrete lactate. In the

case of mammalian cells in culture, the lactic acid is accumulated in the

medium. Lactate generation requires the use of pH control methods to

avoid the direct negative effects of medium acidiﬁcation on cell growth.

Generally, lactate concentrations below 20 mM are considered not to

show any negative effects (Table 4.1). There are several actions that can be

performed in a culture process to reduce lactate accumulation, and these

have already been mentioned.

Lao and Toth (1997) have studied the effects of lactate concentration on

the metabolism of a CHO cell line producing a recombinant glycoprotein,

cultivated in batch mode with an initial concentration of sodium lactate of

60 mM. They demonstrated a reduction of glucose and glutamine con-

sumption (decreased by 20%) and ammonia and alanine production (de-

creased by 64% and 70%, respectively), while productivity was not

affected. They showed that cultures with added lactate have no lactate

production and they hypothesized that inhibition of lactate dehydrogenase

was the cause of this decreased lactate production. Inhibition of lactate

dehydrogenase prevents the regeneration of NADH to NAD

coupled

with the pyruvate/lactate conversion, which leads to an accumulation of

NADH. This excess of NADH inhibits glycolysis in the cytosol. In-

creased efﬁciency of the malate–aspartate shuttle to transport NADH

across mitochondrial membrane would be one of the responses from the

cells to retain the glycolytic rate. The lactate inhibition of glycolysis also

leads to a lower concentration of pyruvate, which may be the cause of a

decrease in glutamine consumption. Due to the diminished energy produc-

tion from lower glycolytic and glutaminolytic rates, with the diversion of

Table 4.1 Inﬂuence of different lactate concentrations on cell growth at

constant pH (Wagner, 1997)

Lactate (mM) Growth Productivity Reference

, 20 No effect No effect Wagner et al. 1988

Miller et al. 1988

20–40 No effect Inhibition Glacken et al. 1988

40–60 Slight inhibition Inhibition Glacken et al. 1988

60 Inhibition Inhibition Glacken 1988

Cell metabolism and its control in culture 95

more energy into maintenance of the ion gradient to counter the effects of

a hypertonic environment, growth was inhibited under high concentra-

tions of lactate.

On the other hand, Cruz et al. (2000) reported that 28 mM lactate

reduced the growth of a BHK cell line by 50%, and that lactate was

consumed at concentrations above 30 mM. Increased concentrations of

lactate reduced cell growth and speciﬁc ammonia production but increased

speciﬁc glutamine and glucose consumption. The effect of lactate was at

least partially due to an increase of osmolarity. An increase in lactate from

0 to 60 mM induced a 40% reduction in speciﬁc productivity of a

recombinant fusion protein secreted by the BHK cells in both stationary

and stirred cultures.

The effects of elevated lactic acid concentration on the cell cycle kinetics

of hybridoma cell growth and antibody production in batch culture were

studied by Kromenaker and Srienc (1994). When 33 mM lactic acid was

initially present, the speciﬁc growth rate was reduced by 37% and the cell-

speciﬁc antibody production rate increased by a factor of 2.6 relative to a

control culture with no additional lactic acid.

4.3.2 Ammonia

Ammonia (NH

) and the ammonium ion (NH

) are highly toxic to

mammalian cells. In vivo, ammonium is secreted by the cells and trans-

ported to the mitochondria of hepatocytes, where it is converted into urea

via the urea cycle. Urea production occurs almost exclusively in the liver

and is the fate of most of the ammonium channeled there. The urea passes

into the bloodstream and thus to the kidneys and is excreted into the

urine. Mammalian cells in culture secrete ammonium into the culture

medium, where its concentration increases gradually because there is no

ammonium recycling pathway (Newland et al., 1990).

As previously stated, ammonium in cell culture medium is the product

of glutamine metabolism and its spontaneous decomposition at 378C.

Negrotti et al. (1989) showed that the half-life for glutamine at 378C and

pH 7.2 is only 7 days. The effects of ammonium on cell metabolism are

observed from concentrations as low as 2 mM, which is easily reached in

culture systems (Table 4.2). For example, the spontaneous decomposition

of glutamine can result in 0.1 mM ammonia per day (Butler and Spier,

1984). Anchorage-dependent cells grown on microcarriers produce be-

tween 2 and 3 mM ammonia after growth in a batch culture (Butler et al.,

Table 4.2 Effect of different medium concentrations of ammonium on

growth and productivity (Wagner, 1997)

Ammonium (mM) Growth Productivity Reference

, 2 No effect No effect Glacken 1988

2–5 Inhibition No effect Glacken et al. 1988

McQueen and Bailey

1990

Glacken 1988

5 Inhibition Inhibition Glacken 1988

96 Animal Cell Technology

1983; Butler and Spier, 1984), and hybridomas in batch suspension culture

produce between 4.5 and 5.5 mM (Reuveny et al., 1986).

Ammonium can either perturb the intracellular or intra-organelle pH

and electrochemical gradients, or directly interact with enzymes. The toxic

effects can be classiﬁed into two possibilities, as described below.

(i) Perturbation of intracellular pH and electrochemical gradients.

At the physiological pH of 7.1–7.5, which is the pH at which most

mammalian cell cultures are maintained, only about 1% of the total

concentration of ammonia/ammonium is present as ammonia (NH

the rest being ammonium (NH

). Ammonia is a small, uncharged,

lipophilic molecule, which readily diffuses across cellular mem-

branes. The small percentage of ammonia present in the extra- and

intracellular aqueous phases will diffuse across the membranes, thus

rapidly equilibrating any transmembrane gradient of ammonia. Am-

monium is produced, in the ionic form, inside the mitochondria

(Figure 4.6A) by the action of glutaminase and glutamate dehydro-

genase. The inner mitochondrial membrane is extremely imperme-

able to ions; however, ammonia readily passes from the

mitochondrial matrix into the cytoplasm. This ammonia outﬂow

from the mitochondria leads to a decrease of the pH in the matrix,

since a proton is left behind. Ammonia can diffuse out of the

cytoplasm into the environment; however, it can be transported back

as ammonium by carrier proteins. The consequence of such a cycle

is an acidiﬁcation of the cytoplasm and the mitochondrial matrix,

and an alkalinization of the environment. Ammonia/ammonium

derived from glutamine decomposition or added externally (Figure

4.6B) will transiently increase the pH of the cytoplasm due to rapid

diffusion of ammonia into the cell. This alkalinization is followed by

an acidiﬁcation due to transport of ammonium by carrier proteins.

Diffusion of ammonia into the mitochondria and other organelles

leads to an increase of the pH inside these compartments. The result

is an alkalinization of the cellular environment and of the inside of

the organelles, mitochondria included, and an acidiﬁcation of the

cytoplasm. Thus, it is very important to realize that the physiological

consequences of adding extracellular ammonium to the medium are

very different to those resulting from ammonium produced intracel-

lularly.

(ii) Ammonium and enzymatic reactions. Ammonia or ammonium can

participate in enzyme reactions and displace equilibria or interact

with regulatory sites of enzymes. The key enzyme of the glycolytic

pathway, phosphofructokinase, as well as Æ-ketoglutarate dehydro-

genase, an enzyme of the TCA cycle, have been reported to be

activated by ammonium (Uyeda and Racker, 1965; Parmeggiani et al.,

1966). Thus, elevated ammonium concentrations could lead to a high

rate of glycolysis and lactate production and a reduced TCA cycle

activity. Other enzymes affected by ammonium concentration are

those involved in the glycosylation of proteins. Gramatikos et al.

(1998) have reported that the intracellular content of UDP-N-acetyl-

hexosamines (UDP-GNAc) is substantially elevated under high am-

Cell metabolism and its control in culture 97

monium concentrations in the medium. This may lead to an increase

in the antennarity of the N-linked oligosaccharides of glycoproteins,

or a decrease in terminal sialylation (Yang and Butler, 2002) (see

Chapter 6).

NH H

⫹

NH H

⫹

Glutamine

metabolism

diffusion

CYTOPLASM

EXTRACELLULAR

SPACE

MITOCHONDRIA

diffusion

CYTOPLASM

MITOCHONDRIA

diffusion

EXTRACELLULAR

SPACE

Glutamine decomposition

Ammonium addition

active transport

⫹

NH H

⫹

NH H

⫹

active transport

NH H

⫹

NH H

⫹

Figure 4.6

Effects of ammoniu m on mammalian cell cult ures. (A) Effects of ammonium

generated by glutamine metabolism. (B) Effects of ammonium generated by

glutamine decomposition or by ammonium addition in the culture medium.

98 Animal Cell Technology

The toxic effects produced by ammonium are: (i) enzymatic inhibition

(in glycolysis, TCA cycle, glutaminolysis, and PPC); (ii) perturbation of

the transcellular ionic gradient; (iii) intracellular pH modiﬁcation; (iv)

increase of UDP-GNAc synthesis, with concomitant glycolysis modiﬁca-

tion and affecting the quality of the recombinant protein; and (v) increase

in the alanine secretion.

As shown schematically in Figure 4.7, an increase of the UDP-GNAc

concentration in the cytoplasm leads to elevated transport of UDP-GNAc

into the Golgi vesicle, by means of the sugar nucleotide transport system

I. This higher concentration of UDP-GNAc, however, inhibits the sugar

nucleotide transport system II that is responsible for CMP-NANA import

to the Golgi, and a higher CMP-NANA concentration inhibits in turn, via

feedback inhibition, the UDP-N-acetylglucosaminyl epimerase that cata-

lyzes the ﬁrst step of UDP-NANA synthesis. This block, together with

the inhibition of the transport system, reduces CMP-NANA concentra-

tion, resulting in a decrease of sialylation.

It was reported by Yang and Butler (2002) that ammonia caused an

increase in the antennarity and a decrease in the sialylation of glycans of

recombinant EPO in CHO cells. Gawlitzek et al. (1998) have studied the

ﬁnal carbohydrate structure of a human interleukin (IL)-2 mutant glyco-

protein, named IL-Mu6. Furthermore, Valley et al. (1999) demonstrated

that about 60–80% of N-acetylated sugars in N-glycan structures con-

tained

N, provided by

Cl used as a supplement in the culture

medium. This indicates that ammonium is used as a building block during

synthesis of the carbohydrate structures expressed in cultivated mamma-

lian cells. Gawlitzek et al. (2000) cultured a BHK-21 recombinant clone in

a continuously perfused double membrane bioreactor. In the presence of

15 mM exogenously added NH

Cl, a signiﬁcant and reproducible increase

in tri- and tetra-antennary oligosaccharides (45% of total) was detected in

the secretion product. They also studied the effect of ammonium on the

glycosylation pattern of the recombinant immunoadhesin tumor necrosis

factor-IgG (TNFR-IgG) produced by CHO cells. They demonstrated that

as ammonium increased from 1 to 15 mM, a concomitant decrease of up to

40% was observed in terminal galactosylation and sialylation of the

molecule. Gawlitzek et al. (1999) cultivated recombinant CHO cells

producing an immunoadhesin glycoprotein (GP1-IgG) under controlled

conditions in the presence of different ammonium N-Glycans synthesized

in the presence of

Cl revealed an N-glycan-dependent increase in

mass to charge ratio of 2.5–4.8 Da. A 60–70% of the total nitrogen

contained in the monosaccharides was

N, which suggested that

was incorporated into GlcNAc and N-acetylneuraminic acid as proposed

earlier (Ryll et al., 1994). Studies of secreted EPO from CHO recombinant

cells during cultivation in a medium supplemented with NH

Cl revealed

that the presence of ammonium caused a signiﬁcant increase in the

heterogeneity of the glycoforms, as shown by the isoelectric point, which

increased from a range of 4.06–4.67 in the control culture to 4.18–6.05,

when ammonium was added to the culture medium. A shift in molecular

weight, from 33 000–39 000 in the control culture to 27 000–37 000 in the

ammonium-supplemented medium, was further evidence for the increased

heterogeneity. Cell growth was inhibited above a culture concentration of

Cell metabolism and its control in culture 99

Glutamine NH

active transport

URIDILATE

KINASE

URIDINE

KINASE

NUCLEOSIDE

DIPHOSPHATE KINASE

Carbamoyl

phosphate

CARBAMOYL

PHOSPHATE

SYNTHETASE I

OMP

UMP

UDP

UTP

UDP-GlcNAcUDP-GalNAc

UDP- -ACETYL-D-

GLUCOSAMINE 4-EPIMERASE

Urd

nNA

NeuNAc

CMP-NANA

Glc 6-P-

Fruc 6-P-

GlcN 6-P-

GlcNAc 6-P-

GlcNAc 1-P-

MITOCHONDRIA

CYTOSOL

ACETYL GLUCOSAMINE

PHOSPHOMUTASE

GLUCOSE-6-PHOSPHATE

ACETYLASE

GLUCOSAMINE 6-

PHOSPHATE DEAMINASE

PHOSPHOFRUCTO

ISOMERASE

HEXOKINASE

UDP- -ACETYL-

GLUCOSAMINYL

EPIMERASE

TRANS-GOLGI

SI S II

NUCLEUS

PROTEIN OF INTEREST

EXTRACELLULAR

SPACE

EXTRACELLULAR SPACE (glutamine decomposition)

OR MITOCHONDRIA (glutamine metabolism)

⫹

diffusion

Glucose

⫹

Figure 4.7

Effects of increased ammonium concentrations on the glycosylation of proteins. OMP, orotidine

monophosphate; UMP, uridine monophosphate; UDP, uridine diphosphate; UTP, uridine

triphosphate; UDP-GlcNAC, UDP-N-acetylglucosamine; UDP-GalNAc, UDP-N-acetylgalactosamine;

ManNAc6P, mannose -N-acetyl-6-phosphate; NeuNAc, N-acetylsiali c acid; CMP-NANA, cytosine

monophosphate-N-acetylneuraminic acid; S I, sugar nucleoside transport system I; S II, sugar

nucleoside transport system II; Glc-6-P, glucose-6-phosphate; Fruc-6-P, fructose-6-phosphate;

GlcN-6-P, glucosamine-6-phosphate; GlcNAc-6-P , N-acetylglucosamine-6-phosphate; GlcNAc-1-P,

N-acetylglucosamine-1-phosphate. The dashed arrows indicate the inﬂuence of the increased

ammonium concentration.

100 Animal Cell Technology

5mMNH

Cl, whereas the speciﬁc production of EPO increased with the

addition of NH

Cl above this level. At 10 mM NH

Cl, the ﬁnal cell

density after 4 days in culture was signiﬁcantly lower but the ﬁnal yield of

EPO was signiﬁcantly higher. This appeared to be due to continued

protein production after cell growth had ceased. The metabolic effects of

added NH

Cl included higher speciﬁc consumption rates of glucose and

glutamine and an increased rate of production of alanine, glycine, and

glutamate (Yang and Butler, 2000).

Sun and Zhang (2001b) showed that an increase of the initial ammonia

concentration resulted in decreased cellular yields with respect to glucose,

glutamine, and other consumed amino acids. In a batch culture with an

initial ammonia concentration of 5.66 mM, the cellular yields from glucose

and glutamine reduced by 78% and 74%, respectively, compared to that

with ammonia at an initial concentration of 0.21 mM. The yields of cells

with respect to other consumed amino acids also decreased by 50–70%.

The metabolic pathways were altered in the cultures with the higher

ammonia concentrations. The glucose consumption was more prone to

form lactate. For glutamine metabolism, the reaction of glutamate to Æ-

ketoglutarate catalyzed by the glutamate dehydrogenase was inhibited by

ammonia, and that by the glutamate amino transferase was facilitated.

However, the yields of glutamate to glutamine decreased with the increase

of ammonia concentrations, showing that the reaction of glutamate to Æ-

ketoglutarate was inhibited by ammonia as a whole. Lao and Toth (1997)

reported a decrease in alanine production. While aspartate consumption

increased sixfold, glutamate changed from consumption to low-level

production and higher consumption again with added ammonium concen-

trations. Cells consumed more aspartate and glutamine at high ammonium

concentrations. There was also a decrease in TCA intermediates, for

example, malate (Lao and Toth, 1997).

Chen et al. (2005) investigated the addition of amino acids to the growth

media for CHO cell cultures as a means of mitigating the negative effects

of ammonium. Threonine, proline, and glycine additions enhanced CHO

cell growth and recombinant protein levels. Furthermore, the addition of

these amino acids positively impacted important metabolic parameters,

including glucose consumption, lactate production, glutamine utilization,

and ﬁnal ammonium levels. Additionally, threonine, proline, and glycine

increased the level of Æ-2,3-linked sialic acid, galactose--1,4-N-acetylglu-

cosamine, and Æ-2,6-linked sialic acid residues in recombinant tissue

plasminogen activator (tPA). Thus, these amino acids can be used to

mitigate some of the toxic effects of ammonium on cell growth, recombi-

nant protein productivity, and protein quality.

4.4 Factors affecting cell metabolism

There are several factors that have a large inﬂuence on metabolic path-

ways. Among the external factors, the most important are the dissolved

oxygen and carbon dioxide concentration, the pH, and the temperature of

the culture. These factors are discussed in Chapter 2, but here their effects

on metabolic pathways are described.

Cell metabolism and its control in culture 101