Hugo W.B., Russel A.D.(ed). Pharmaceutical Microbiology

Подождите немного. Документ загружается.

2.2 Expression of cloned genes

Once a gene is cloned it is necessary to convert the information contained in it into a

functional protein. There are a number of steps in gene expression: (i) transcription of

DNA into mRNA; (ii) translation of the mRNA into a protein sequence: and (iii) in

some instances, post-translational modification of the protein. In discussing these steps

in more detail, expression of a cloned insulin gene will be used as an example.

2.2.7 Transcription

Transcription of DNA into mRNA is mediated by the enzyme RNA polymerase. The

first stage is binding of the RNA polymerase to recognition sites on the DNA which are

called promoters. After binding, the RNA polymerase proceeds along the DNA molecule

until a termination signal is encountered. It follows that a gene which does not lie

between a promoter and a termination signal will not be transcribed. This would be the

case with a cloned insulin gene, since neither a cDN A gene nor an artificially synthesized

gene will carry a promoter. The solution is to clone the gene into a vector close to a

bacterial promotor. An example is shown in Fig. 24.5.

2.2.2 Translation

Translation of mRNA into protein is a complex process which involves interaction of

the messenger with ribosomes. One prerequisite for this is that the mRNA must carry a

ribosome binding site (RBS) in front of the gene to be translated. After binding, the

ribosome moves along the mRNA and initiates protein synthesis at the first AUG codon

it encounters. A synthetic insulin gene will lack an RBS and if a cDNA is used as the

starting material the RBS may be lost in the process of cloning. The solution is to

TRANSCRIPTION NO TRANSCRIPTION

Recombinant DNA technology 457

Fig. 24.5 Insertion of a cloned

insulin gene into a vector

carrying a bacterial promoter.

The arrow indicates the direction

of transcription. If we suppose

the bacterial promoter is derived

from the lactose operon then

transcription will be initiated

only in the presence of lactose.

DNA

Fig. 24.6 The use of a vector carrying a promoter and adjacent ribosome binding site (RBS) and

initiation codon to obtain synthesis of proinsulin from a synthetic gene. The arrow indicates the

direction of transcription.

utilize a vector in which the insulin gene can be inserted downstream from a promoter

and RBS (Fig. 24.6).

2.2.3 Post-translational modification

A number of proteins undergo post-translational modifications and insulin is one of

these. Proteins which are to be secreted are synthesized with an extra 15-30 amino

acids at the N-terminus. These extra amino acids are referred to as a signal sequence

and a common feature of these sequences is that they have a central core of hydrophobic

amino acids flanked by polar or hydrophilic residues. During passage through the

membrane the signal sequence is cleaved off (Fig. 24.7). If the insulin gene were cloned

by the cDNA method then the signal sequence would be present and, in Escherichia

coli at least, the insulin would be transported through the cytoplasmic membrane

{exported). Using the synthetic gene approach, a signal sequence would be present on

the protein only if the corresponding coding sequence had been incorporated at the

Human preproinsulin Human proinsulin Human insulin

Fig. 24.7 The conversion of preproinsulin to insulin by sequential removal of the signal peptide and

the C fragment.

458 Chapter 24

time of construction. Sometimes it is desirable for the bacterium to export the protein,

in which case a signal sequence is incorporated; with other proteins it may be desirable

that they are retained within the cell.

In E. coli cells, the presence of a signal sequence usually results in export of a

protein into the periplasmic space rather than into the growth medium. Unfortunately,

many recombinant proteins are rapidly and extensively degraded in the periplasmic

space because of the presence there of numerous proteases. In Gram-positive bacteria

and eukaryotic microorganisms, signal sequences direct proteins into the growth

medium. Filamentous organisms such as fungi or actinomycetes might be particularly

favourable for export because of their high surface area to volume ratio.

A small number of proteins, and again insulin is an example, are synthesized as

pro-proteins with an additional amino acid sequence which dictates the final three-

dimensional structure. In the case of proinsulin, proteolytic attack cleaves out a stretch

of 35 amino acids in the middle of the molecule to generate insulin. The peptide that is

removed is known as the C chain. The other chains, A and B, remain crosslinked and

thus locked in a stable tertiary structure by the disulphide bridges formed when the

molecule originally folded as proinsulin. Bacteria have no mechanism for specifically

cutting out the folding sequences from pro-hormones and the way of solving this problem

is described in a later section.

Another modification which can be made in vivo is glycosylation, for example that

of (3 and y interferons, although the biological role of the sugar residues is not known.

Bacteria cannot glycosylate the products of cloned mammalian genes. These non-

glycosylated proteins retain their pharmacological activity but their pharmacokinetics

and in vitro stability may be different. Yeast cells can glycosylate proteins but the

pattern of glycosylation may well be different from that seen in the normal host of the

gene. Non-glycosylated or wrongly glycosylated proteins may provoke the formation

of antibodies following administration.

2.3 Maximizing gene expression

From a commercial point of view it is desirable to maximize the yield of protein in a

fermentation. This means maximizing gene expression and important factors are:

1 the number of copies of the plasmid vector per unit cell (copy number);

2 the strength of the promoter;

3 the sequences of the RBS and flanking DNA;

4 proteolysis.

The limiting factor in expression is the initiation of protein synthesis. Increasing

the copy number of the plasmid increases the number of mRN A molecules transcribed

from the cloned gene and this results in increased protein synthesis. Similarly, the

stronger the promoter (see Fig. 24.5), the more mRNA molecules are synthesized. The

base sequence of the RBS (see Fig. 24.6) and the length and sequence of the DNA

between the RBS and the initiating AUG codon are so important that a single base

change, addition or deletion, can affect the level of translation up to 1000-fold.

Proteolysis does not affect transcription and translation but by degrading the desired

product it influences the apparent rate of gene expression. Although proteolysis can be

reduced it is difficult to eliminate it completely. One approach is to use protease-deficient

Recombinant DNA technology 459

Fig. 24.8 Release of somatostatin from a hybrid protein by cyanogen bromide cleavage.

Somatostatin can be purified free of cyanogen bromide and fragments of /J-galactosidase.

mutants and another is to protect the desired protein by fusion to an E. coli protein (see

below).

Somatostatin was the first human peptide to be synthesized in a bacterial cell. It is

only 14 amino acids long and genes for polypeptides of this size are very amenable to

direct chemical synthesis. However, small peptides are rapidly degraded in E. coli and

for this reason the synthetic gene was fused to the 5' end of the /3-galactosidase gene.

This results in the synthesis of a fusion protein which is relatively stable in E. coli.

Somatostatin does not contain any methionine residues, so the synthetic gene was

constructed in such a way that a methionine was incorporated at the junction of the

fusion peptide. By treatment with cyanogen bromide, which breaks proteins into

polypeptide fragments at methionine residues, authentic somatostatin could be recovered

(Fig. 24.8). Although in this particular instance, and in the case of insulin and fi-

endorphin, the fusion protein contained a remnant of the E. coli /3-galactosidase gene

at the N-terminus, other bacterial proteins have been used, for example tryptophan

synthetase, j3-lactamase, etc.

2.4 Choice of cloning host

A number of cloning hosts are in widespread use. Escherichia coli is still the most

popular organism for initial genetic manipulations and is used for the commercial

production of a number of therapeutic proteins. Bacillus subtilis has not lived up

to its initial promise of high-level protein secretion and interest in it is declining.

Saccharomyces cerevisiae is widely used but faces competition from recombinant animal

cells: progress with the latter has been impressive and high-level expression and secretion

systems are available. Good progress has also been made in developing cloning

systems for filamentous fungi and actinomycetes, two groups of organisms which

have long been used in the production of low molecular weight pharmaceuticals.

More recently, there has been growing interest in the development of cloning systems

460 Chapter 24

for the more unusual organisms used in the pharmaceutical industry. The advantages

and disadvantages of the main microbial cloning systems are shown in Table 24.1.

Recombinant proteins can be produced in plants and animals as well as microbes.

For example, a number of important human proteins, e.g. a,-antitrypsin, have been

produced in rats and mice and in some instances can be engineered to be secreted in the

breast milk. Clearly, small mammals are not desirable as production vehicles. However,

good expression can also be obtained with animals such as sheep and goats. Given the

history of large mammals as sources of antitoxins for human therapy they also may be

acceptable for the production of recombinant proteins. A large number of recombinant

proteins also have been produced in plants, e.g. proteins toxic to insect larvae, antibody

fragments, etc. Already some of these recombinant plants are grown commercially,

and are being consumed, so there is no reason why they cannot be sources of protein

drugs as well. In this context it is worth noting that the pharmaceutical industry is used

to manufacturing drugs from plants, as plants are the source of many of the older

medicines still in use.

3 Production of medically important

polypeptides and proteins

The overproduction of a wide variety of proteins has now been achieved in E. coli and

other cloning hosts. Many of these proteins are in clinical trials and, as indicated earlier,

over a dozen are already on the market. The current status of many of these proteins is

summarized in Table 24.2. The efficacy of many of the proteins listed remains to be

determined because until the advent of recombinant DNA technology sufficient

quantities were not available to enable clinical trials to be undertaken. It should be

noted that clinical efficacy alone is not sufficient. Market size is just as important since

it can cost up to £50 million to bring a new drug to the market place and company

shareholders expect a good return on their investment.

One of the advantages of recombinant DNA technology is that is enables analogues

of human proteins to be produced. Thus, numerous groups have produced a-a and

a-/3 hybrid interferons. Some of these hybrids have altered properties in vitro but whether

this will translate into a clinical benefit remains to be determined. In some instances

the analogues have only a single amino acid change. Thus, changing cysteine residue

17 in interferon-/3 to a serine residue yields a protein with improved half-life and in

vitro stability. Changing methionine residue 358 in a,-antitrypsin to valine yields a

more oxidation-resistant enzyme.

4 Authenticity and efficacy of drugs produced by

recombinant DNA technology

To demonstrate the safety and efficacy of any polypeptide drug, regardless of whether

it is made by recombinant DNA technology, organic synthesis or extraction from a

natural source, a number of quality criteria need to be met. Not only must the protein

be produced in accordance with good manufacturing practice but it must also meet

specification. Although the absolute specification will vary depending on the identity

of the protein, the therapeutic target and the route and period of administration, certain

Recombinant DNA technology 461

462 Chapter 24

Table 24.1 Comparison of different

Organism

Escherichia coil

Bacillus subtilis

Saccharomyces cerevisiae

Filamentous fungi

Actinomycetes

Mammalian cells

organisms as cloning hosts

Advantages

Ease of manipulation

Promoters and gene regulation

well understood

Easy to culture on large scale

Already used in manufacture of

insulin, interferon and

human somatotrophin

Many proteins naturally

exported into growth

medium

Non-pathogenic

Easy to culture

Some Bacillus enzymes

excreted at high level

(>5gM)

Widely used industrial

organism which is easy to

culture

Glycosylates proteins

Can get export into growth

medium of heterologous

proteins

High-level expression systems

developed

Heterologous proteins inside

cell do nofform inclusions

Large surface area to volume

ratio should favour protein

export

Have been used in industrial

microbiology for over 40

years

Large surface area to volume

ratio should favour protein

export

Widely used in industrial

microbiology

Good expression systems being

developed

Get export of proteins

Get desired post-translational

modifications and products

not likely to be immunogenic

to humans

Good expression systems

available

Disadvantages

Do not usually get export of

proteins into growth medium

Over-expressed foreign

proteins often form

aggregates ('inclusions') of

denatured protein

Many foreign proteins rapidly

degraded

Many post-translational

modifications do not occur

Still not much known about

gene regulation

Good, high-level expression

vectors lacking

High-level export of

heterologous proteins not

achieved

Much still to be learned about

control of gene expression

Post-translational modifications

of proteins not necessarily

the same as those in the

animal cell

Promoters/gene regulation

poorly understood but may

be similar to yeast

Good expression systems

lacking

Rheology of fermentations

important

Promoters/gene regulation still

poorly understood

Rheology of fermentations

important

Large-scale growth of animal

cells costly

Great care needed to avoid

contamination of cultures

Table 24.2 Current status of selected recombinant proteins

Protein

Human insulin

Human somatotropin

lnterferon-a

Interferon-/

Tissue plasminogen

activator

Relaxin

a-Antitrypsin

Size/structure

Two peptide chains:

A, 21 amino acids

long, and B, 30

amino acids long

191 amino acids

166 amino acids

143 amino acids,

glycosylated

530 amino acids,

glycosylated

53 amino acids;

insulin-like (two

protein chains)

394 amino acids.

glycosylated

Expression system

E. coli

Yeast

Animal cells

E. coli

Yeast

Clinical indications

Juvenile onset

diabetes

Pituitary dwarfism

Various cancers and

viral diseases

Chronic

granulomatous

disease

Acute mycocardial

infarct

Pulmonary embolism

Facilitates childbirth

Treatment of

emphysema

Comments

Approved for sale

A and B chains made separately as

fusion proteins and joined in vitro

Compared with animal insulins some

undesirable side-effects have been

noted

Approved for sale

If useful in treatment of osteoporosis

then market size will be much larger

Has additional methionine residue at

N-terminus, but technology for

removing this now available

Approved for sale

Over 80% success in treatment of hairy

cell leukemia; success with other

cancers lower and more variable

Market size may be limited

Unpleasant ('flu-like') side-effects

Approved for sale

In clinical trials for treatment of cancer

and viral diseases

Approved for sale

Animal cell culture most effective way

of producing active enzyme

Prepares endometrium for parturition

and reduces fetal distress

Pig relaxin shown to be clinically

effective

Prevents cumulative damage to lung

tissue caused by leucocyte elastase

In clinical trials

Continued on p. 464

Table 24.2 Continued

Protein

lnterleukin-2

Tumour necrosis factor

Human serum albumin

Factor VIII

Factor IX

Erythropoietin

Hepatitis B surface

antigen

Granulocyte colony

stimulating factor

Granulocyte-macrophage

colony stimulating

factor

Size/structure

133 amino acids

157 amino acids

582 amino acids; 17

disulphide bridges

2332 amino acids

415 amino acids

glycosylated;

modified residues

166 amino acids

glycosylated

Monomer has 226

amino acids

127 amino acids

Expression system

E. coli

Animal cells

E. coli

Animal cells

Yeast

Mammalian cells

Yeast

Mammalian cells

E. coli

Clinical indications

Treatment of cancer

Plasma replacement

therapy

Treatment of

haemophilia

Treatment of

Christmas disease

Treatment of

anaemia

associated with

dialysis and

AZT/AIDS

Vaccination

Adjunct to cancer

chemotherapy

Improved bone

marrow transplant

Comments

Approved for sale

Very toxic and side-effects severe

Normally obtained from plasma but

now concern over potential

contamination with AIDS virus

Normally obtained from plasma but

now concern over potential

contamination with AIDS virus

Approved for sale

Must be made in mammalian cells since

glycosylation and conversion of first

12 glutamate residues to

pyroglutamate essential for activity

Approved for sale

Without glycosylation protein is cleared

very quickly from plasma

Approved for sale

Monomer self-assembles into structure

resembling virus particles

Approved for sale

By stimulating white blood cell

formation, aids recovery

Approved for sale

HPLC, high-performance liquid chromatography; ELISA, enzyme-linked immunosorbent assay

method; RIA, radioimmunoassay method.

quality guidelines have been adopted by most countries. This core specification is shown

in Table 24.3.

Although many of the quality control tests used are designed to assess purity they

often give data which confirms the identity of the protein, e.g. chromatographic

behaviour (high-performance liquid chromatography), electrophoretic mobility and

amino acid composition. However, most of the analytical techniques in current use

give no indication of the three-dimensional structure of the protein and hence no

indication of biological activity. Thus, the absolute specific activity of the protein needs

to be determined in a biological test. Determination of the specific activity is particularly

important with proteins overproduced in E. coli, for such proteins exist in aggregates

with nucleic acid often called 'inclusions'. The protein in these aggregates has to be

extracted with denaturing agents such as urea, sodium dodecyl sulphate or guanidinium

hydrochloride and then renatured, a process akin to recreating native egg white from a

meringue.

As indicated earlier, recombinant DNA technology can be used to deliberately

produce desired analogues of natural proteins. However, undesirable analogues may

also be produced inadvertently during the production process. For example, when the

human somatotrophin gene is expressed in E. coli, the resultant protein has an additional

methionine residue at the N-terminus. Other foreign gene products may or may not

carry this additional methionine residue. Recently, methods have been developed for

enzymatically removing this N-terminal methionine and for mediating another post-

translational modification, C-terminal amidation.

Another undesirable modification is removal of some amino acids residues from

the C-terminus and/or the N-terminus by microbial exoproteases. Care needs to be

taken during the fermentation and extraction stages to minimize proteolytic damage,

and any 'nibbled' molecules should be removed during purification.

Recombinant DNA technology 465

Table 24.3 Specification of therapeutic proteins to be administered parenterally

Criterion

Greater than 95% pure

Microheterogeneity below specified level

Endotoxin below specified level

Contaminating DNA below specified

level (<10pg dose

-1

)

Toxic chemicals used in purification

below specified level

IgG below specified limit (if monoclonal

antibodies used in purification)

Absence of microorganisms

Appropriate analytical methods

Gel electrophoresis; HPLC

Polyacrylamide gel electrophoresis

C- and N-terminal analysis. Amino

acid composition

Limulus amoebocyte lysate method

(see also Chapter 18)

Hybridization

Appropriate methods

ELISA or RIA

Sterility test (see also Chapter 23)

5 Future trends with protein pharmaceuticals

Already over a dozen recombinant-derived therapeutically useful proteins are being

marketed and at least as many more are in clinical trials. So, what of the future? There

are two disadvantages with developing proteins^as therapeutic entities. First, most of

them are not active when given orally, and parenteral administration is almost de rigeur.

Some proteins can be administered in other ways, e.g. insulin can be given per rectum

but this is not a route which is favoured in many countries outside of France and Japan.

Other proteins may be active if given sublingually or if administered by aerosol. Clearly,

much work needs to be done in developing new dosage forms particularly suited to the

administration of proteins. Second, most of the proteins being marketed or currently in

clinical trials were obvious candidates for development, e.g. insulin and interferons.

Identifying the next generation of therapeutically useful proteins will be much more

difficult. There are hundreds of human proteins of which relatively little is known but

only a few are likely to be worth developing. For example, a factor which promotes

bone growth would have many clinical benefits but the candidate proteins have yet to

be identified.

5.1 Small-molecule drugs

One trend which has become obvious is that many pharmaceutical companies are turning

their attention to the application of recombinant DNA technology in the production of

small molecules. Microorganisms are widely used in the production of drugs, e.g.

antibiotics and steroid transformations. Where the rate-limiting step in production has

been identified, cloning the relevant gene could well facilitate synthesis and give

increased yields and/or decreased production times. In addition, novel metabolic

pathways can be introduced into microorganisms and this could eliminate more

conventional but complex production processes involving plants.

5.2 Anti-sense agents

Many drugs treat the symptoms of a disease rather than the cause of the disease, e.g. the

different classes of drug for the treatment of hypertension. Where the primary cause of

the disease is overexpression then anti-sense agents may be of value. These are nucleic

acids which are complementary to the 5' region of a mRNA molecule and bind to it. In

this way the translation of the mRNA into protein is reduced or eliminated and hence

the anti-sense molecule modulates expression. There are two major problems with

anti-sense nucleic acids as therapeutic entities. First, small single-stranded nucleic acids

are rapidly degraded inside cells. The solution to this problem is to use modified nucleic

acids, e.g. ones in which the phosphodiester backbone is replaced with a peptide chain

('peptide nucleic acids'). The second issue is delivering enough of the anti-sense

molecules to the target cells. This can be achieved with proper formulation, and a

number of anti-sense drugs are in clinical trials.

466 Chapter 24