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

Подождите немного. Документ загружается.

rate of polyadenylation and the nuclear export of the mRNA (Mattaj,

1990). The sequence of the introns is not conserved and the bases GU and

AG in the 59 and 39 ends, respectively, are recognized as hallmarks of

almost all introns (Petitclerc et al., 1995). Although there are genes that are

efﬁciently expressed without the presence of introns, their inclusion in

expression vectors is generally recommended. They should be placed at

the 59 end of the transcription unit to avoid aberrant splicing as a result of

inactive or cryptic splicing sites present in the gene sequence (Huang and

Gorman, 1990).

Most mRNAs from eukaryotic cells contain in their 39 end an extension

of approximately 200 adenosine molecules, known as the polyA tail.

Similar to capping and splicing, the polyadenylation of the primary

transcript is a co-transcriptional process. It occurs at the 39 end of the

RNA and consists of a site-speciﬁc cleavage followed by polymerization

(adenylation). The polyadenylation is essential for the formation, stability,

and translatability of the mRNAs. The polyadenylation signal is composed

of a highly conserved sequence (AATAAA) located about 10–30 nucleo-

tides downstream from the stop codon, and positioned immediately up-

stream of a T- or TG-rich region (T/G box; Munroe and Jacobson, 1990).

In most cases, the initial step in mRNA degradation is the shortening of

the polyA tail. This gradual aging of the mRNA takes place in the

cytoplasm and is referred to as the ‘molecular clock9 of the RNA. The

polyadenylation signals more commonly employed in expression vectors

are those derived from the bovine growth hormone, mouse -globin, SV-

40 early transcription unit, thymidine kinase (TK), herpes simplex virus

(HSV), and hepatitis B antigen (Kaufman, 1990).

The sequence signaling the culmination of the transcription reaction is

localized between the polyadenylation signal and the 39 end of the mature

RNA. Although the transcription termination region of several eukaryotic

genes has been identiﬁed – consensus sequence: ATCAAA(A/T)TAGGA

AGA – the precise site of the termination is, in most of the cases,

unknown. The termination signal is required to prevent the transcriptional

machinery extending its activity to sequences located downstream of the

gene that is being transcribed, a phenomenon known as ‘transcriptional

interference.9 It has also been proposed that the introduction of termina-

tion sequences abolishes or diminishes the transcription of the comple-

mentary DNA chain, which would give rise to the formation of antisense

RNA, with the consequent suppression of the expression of the sense

RNA (Izant and Weintraub, 1985). The expression vectors for animal cells

harbor the well-characterized transcription terminator sequences from

prokaryotes.

3.3.2 Translational control elements

In the cytoplasm the codons of mRNA are translated into amino acids via

the concerted action of the ribosomes, transfer RNA (tRNA), and a large

protein complex. In addition to the mature RNA sequence (capped and

polyadenylated) other structural elements located in the 59 and 39 untrans-

lated regions (UTRs) of the mRNA may inﬂuence its translation.

42 Animal Cell Technology

59 Untranslated region

The initiation of the translation constitutes the rate limiting step in the

synthesis of proteins. The start codon (AUG) embedded in the consensus

sequence GCC(

A/G)CCAUGG, widely known as Kozak 9s sequence,

gives the appropriate context for an optimal beginning of the translation

(Kozak, 2005). Speciﬁc purines in the Kozak9s sequence (underlined)

provide an enhancer effect on translation. It has also been noted that the

existence of additional start codons in the 59 UTR impairs severely the

translation (Grens and Schefﬂer, 1990). The formation of secondary

structures (i.e. hairpin loops) in this region may inhibit the binding and/or

the advance of the translational complex on the RNA strand due to a steric

hindrance. Regions rich in G and C are prone to form structures of this

type, which are thermodynamically very stable (Grens and Schefﬂer,

1990).

39 Untranslated region

In mammalian cells, the degradation of the mRNA is triggered after a

signiﬁcant shortening of the polyA tail and is catalyzed by 39-59 exoribo-

nuclease activity of a protein complex termed the exosome. Mechanisms

involving exonuclease activity that are independent of the polyA ‘‘aging’’

have also been described (Ina´cio and Liebhaber, 2003). In the 39 UTR of

the mRNA there are sequences able to alter the stability of the transcript.

For instance, motifs that have AU-rich elements (ARE) were found in

highly unstable mRNA. AREs were later shown to bind proteins with the

capacity to regulate either positively or negatively the decay rate of the

mRNA, that is, recruiting or blocking the exosome (Gray and Wickens,

1998).

Termination codon and codon usage

The completion of the translational process requires the binding of the

‘‘release factor’’ to the stop codon, which induces the disassociation of the

translational complex and thus, the end of polypeptide synthesis.

Although the presence of the stop codon is sufﬁcient to interrupt the

translation, exhaustive studies have demonstrated that the contiguous base

can inﬂuence signiﬁcantly the efﬁciency of this process. For example, it

has been reported that purines (A or G) occupying this position constitute

a much more effective termination signal than pyrimidines (C or U;

McCaughan et al., 1995). Two models were proposed to explain the effect

of the fourth base on the performance of the termination process: one

suggests that this base is recognized as a part of the stop codon and the

other claims that this base constitutes an independent signal for the

recruitment of the ‘‘release factor’’ (McCaughan et al., 1995).

It is worthwhile to mention that a considerable degree of preference for

certain codons has been observed for genes displaying high expression

levels, which correlated with a higher abundance of the complementary

tRNA in the cell (Fedorov et al., 2002). The optimization of codon usage

Cloning and expression of heterologous proteins in animal cells 43

can be achieved by site-directed mutagenesis to increase the expression

level of some genes (Makrides, 1999).

3.4 Systems for heterologous expression in animal cells

All expression systems consist of a vector and a host. The vector is deﬁned

as a molecule of DNA or RNA, which is genetically manipulated to carry

a molecule of foreign DNA or RNA, with the aim of producing proteins

(gene expression) or DNA (ampliﬁcation, replication) once it is introduced

into a host cell. In the nucleus of the transfected cell the expression vector

can exist: (i) as an independent replication unit, namely episome, or (ii)

integrated to the host genome through a random or non-homologous

recombination process. In order to exist as an episome the vector must

have a signal (episomal replication origin) allowing for its autonomous

multiplication, non-linked to the replication of the host genome. Episomal

vectors render high transient expression levels of recombinant protein

(discussed below). However, the cytotoxicity generated by the high

concentration of exogenous DNA, the tendency to integrate into the

cellular genome and to undergo rearrangements (mainly in long-term

cultures) have been pointed out as the major obstacles inherent to its use

(Sanders, 1990). As explained below, the clonal variation due to the

‘‘positioning effect,’’ commonly observed for integrative vectors, does not

occur with episomal vectors.

The integration of the vector into the host’s genome takes place in a

random manner and the sequences/structures from either the vector or the

site of integration can affect the expression of the target gene. The fact that

nearly 95% of the genome from animal species consists of non-coding

sequences and that the coding regions are frequently silenced (heterochro-

matin) explains the low efﬁciency in obtaining highly productive clones

from vectors that integrate into the chromosomes. Possible rearrange-

ments of the vector9s DNA giving rise to mutations, insertions or deletions

that can alter the sequence or expression of the heterologous gene, may

take place when using this type of vector.

The expression systems available for the production of recombinant

proteins in animal cells can be classiﬁed as mammalian cell/viral or plasmid

vector, or insect cell/baculovirus.

3.4.1 Viral vectors

The ﬁrst attempts to express foreign genes in mammalian cells were

carried out using viruses as vectors (Goff and Berg, 1976; Hamer et al.,

1977). According to the nature of the viral nucleic acid, these vectors are

classiﬁed as DNA or RNA viruses, or chimeras. If the genetic material of

the virus is surrounded by a protein envelope (capsid), it enters the cell

by a mechanism called infection, which generally occurs with high

efﬁciency. If the nucleic acid of the vector lacks a capsid (‘‘naked’’), its

incorporation to the cell is carried out by means of transfection techni-

ques (discussed in Section 3.7). In a permissive cell, a viral vector

develops a lytic cycle, in which the replication of the viral DNA and its

44 Animal Cell Technology

later encapsulation generates a large number of virions (progeny) that

will ﬁnally cause cell lysis and death, thus precluding the establishment

of a stable cell line. Such limitations can be overcome by using: (i)

nonpermissive cells, wherein the virus replicates as an independent entity

integrated into the host’s genome and induces only a morphological

transformation of the cells; (ii) defective viruses, whose genome has been

modiﬁed in such a way that it cannot complete the lytic cycle and

depends for its propagation on either a helper virus or a speciﬁc cell line

that provides the absent elements/functions (i.e. COS cells/SV-40 virus,

Section 3.6.1); or (iii) genetically modiﬁed permissive cells that allow the

episomal replication of the virus, increasing the productive yield while

preventing the lytic cycle, although the high concentrations of extrachro-

mosomal DNA can be cytotoxic (Sanders, 1990). The main advantage

offered by the viral systems is that tedious screening for high producing

clones is not required as is the case when using integrative plasmid

vectors. Some disadvantages displayed by viral vectors are: (i) the

physical limitations of the viral packaging restrict the size of the foreign

gene; (ii) recombinant viruses are generally defective and must be

propagated with a helper virus or a speciﬁc cell line; (iii) stable recombi-

nant cell lines cannot be established with a virus that presents a lytic

cycle; (iv) the speciﬁcity of the virus to infect certain cell types restrains

the selection of the expression host.

Vectors derived from DNA viruses

SV-40 and polyoma viruses belong to the group of the papovavirus

(Family: Polyoma virus), both presenting a small ‘‘naked’’ genome (5

kb) that is able to replicate in primate and murine cells, respectively. SV-40

is considered the pioneer vector in terms of heterologous expression in

mammalian cells (Goff and Berg, 1976; Hamer et al., 1977), and is

preferred over polyoma viruses for production of recombinant proteins

due to its higher replication rate and infection efﬁciency. The target gene

is inserted in the early or late replication regions of the viral genome,

producing a recombinant defective virus (Muller et al., 1983; Zhu et al.,

1984). The advantages of employing SV-40 as an expression vector can be

summarized in the higher expression levels and short time required to

evaluate the production of recombinant protein. The drawbacks of this

expression system are: (i) the DNA sequence to be inserted has a maxi-

mum size of 2.5 kb; (ii) the host selection is limited because the viral

infection is cell-speciﬁc; and (iii) successful recombinant expression is

variable and unpredictable (Levinson, 1990).

Adenoviruses possess a genome of about 35 kb, which allows insertion

of a larger heterologous sequence without affecting viral replication. This

expression system offers high expression levels since viral infection inhi-

bits synthesis of cellular proteins and, in the case of type-2 adenoviruses,

they allow the use of a wide range of host cells (i.e. human, primate,

rodent). Owing to the reduced number of restriction sites available in the

adenoviral genome, insertion of foreign sequences is performed by homo-

logous recombination instead of molecular cloning. In this respect, the

viral DNA is co-transfected with a plasmid containing the gene of interest

Cloning and expression of heterologous proteins in animal cells 45

ﬂanked by the regions of the viral genome where recombination will take

place. The frequency of recombination is variable and the generation of

stocks of recombinant virus requires puriﬁcation of the viral progeny.

Sequences up to 8 kb can be inserted into and expressed by adenoviruses.

Heterologous promoters (from SV-40 or cytomegalovirus) have been

successfully used to express genes with this system (Sandig et al., 1997).

The cell line HEK-293 has been generated to complement a region absent

in recombinant adenoviruses, thus obviating the need for a helper virus to

complete the viral cycle. The adenoviruses are reliable vectors for the

production of recombinant proteins. The generation of novel adenoviral

vectors is described in detail by Bourbeau et al. (2003).

The adeno-associated viruses (AAVs) belong to the Parvovirus family,

contain a small genome (5 kb), and are not pathogenic. AAVs have a

biphasic life cycle in the host cell. They can either integrate to the host

genome or persist in an episomal form, both mechanisms leading to a

latent infection in the absence of a helper virus. In the presence of a

helper virus (i.e. adenovirus or herpesvirus), AAVs undergo a productive

infection. In addition to the lack of pathogenicity, these vectors present

other advantages such as high transfection efﬁciency, reduced cytotoxi-

city at elevated multiplicity of infection, and a wide range of hosts. The

main limitation of these vectors lies in the size of the gene that can be

expressed (, 5 kb). Although the rate of integration of AAVs to the host

genome is low, undesirable chromosomal mutations can be expected

when employing the helper virus approach (Xiao, 2003). The recent

discovery of the minimal set of adenoviral genes required for efﬁcient

generation of progeny AAV particles (Matsushita et al., 1998) allows the

production of recombinant AAVs without the need of adenoviral co-

infection.

Epstein–Barr virus (EBV) is a member of the herpesvirus family that

became useful as an expression vector for mammalian cells. A peculiarity

of this virus is that it can develop either a latent infection or a lytic cycle in

the host. The ﬁrst leads to transformation/immortalization of the host cell,

and the second to cell lysis. The identiﬁcation of the elements necessary

for EBV replication extends the range of permissive cells from human B

lymphocytes to ﬁbroblastic and epithelial cells of human, primate, and

canine origin (Sugden et al., 1985; Lutfalla et al., 1989). The use of this

virus as an expression vector offers stable transfection and high levels of

gene expression, simplicity for the selection of highly producing clones,

easy recovery of the episomal DNA, and the possibility to produce

authentic human glycoproteins if human cells are used as the host

(Teshigawara and Katsura, 1992). Some disadvantages of the EBV-based

expression systems are the need for a selection marker to assure the

persistence of the DNA in an extrachromosomal form, a certain degree of

variability in the expression levels depending on the cell type, and the

probability of genomic integration with a consequent reduction in the

expression rate of the heterologous gene (Levinson, 1990).

Another member of this family is HSV (herpes simplex virus). HSV has

a genome size of 125 kb, although half of the genes are not essential for

growth, allowing large molecular inserts. HSVs infect a wide range of

hosts and show a high infectivity. The relative simplicity of preparing large

46 Animal Cell Technology

stocks of recombinant virus is another advantage presented by HSV

(Burton et al., 2003).

Papillomaviruses contain a genome of 8 kb, usually have a single host

and replicate episomally. Bovine papillomavirus (BPV) is a representative

of this family, which induces phenotypic changes in the host facilitating

the identiﬁcation and selection of transformants. Although BPV-based

vectors replicate episomally, the selection pressure achieved by the addi-

tion of selection markers (i.e. neomycin or cadmium; see Section 3.8) has

been reported to induce a stable integration of multiple copies of the viral

DNA into the host genome (Niwa et al., 1991). The susceptibility of the

virus to undergo rearrangements with the concomitant risk of altering the

sequence of interest is perhaps its main drawback (Levinson, 1990). The

optimization of the expression system based on papillomavirus has been

thwarted by the lack of progress in understanding the biology of the viral

replication and transcription.

Vaccinia virus is a representative of the Poxviruses. Members of this

family are unique among vertebrate viruses because they possess their own

transcription machinery encoded in a genome of 100–300 kb. The fact that

the host metabolism is dispensable allows vaccinia to have a high rate of

replication and protein biosynthesis, making it an excellent choice for the

expression of heterologous genes (Moss, 1996). Mammalian and avian cells

are suitable hosts for vaccinia. The viral genome can accept DNA frag-

ments up to 25 kb, which are incorporated by homologous recombina-

tion, where in general, the gene encoding for viral thymidine kinase (TK)

is replaced and selection of recombinant virus is carried out in TK-

deﬁcient cells (Mackett et al., 1982). It is worthwhile to note that the

transcription mechanism of vaccinia is unable to perform splicing; there-

fore, this vector allows only the expression of complementary DNA

(cDNA) sequences (Moss, 1996). Poxvirus-based systems are commonly

used in eukaryotic cells due to the stability of the viral genome, the wide

range of hosts, the ease of production and manipulation of the virus, and

the high expression levels (Carroll and Kovacs, 2003).

Vectors derived from RNA viruses

In retroviruses, the genetic information is encoded in the form of RNA

(5–10 kb), which is retro-transcribed to DNA in the host cell by means of

a virus-speciﬁc reversed transcriptase enzyme. Since this intermediate

DNA cannot integrate into the genome of the infected cell, retroviruses

are employed as transference and expression vectors (Holmes-Son et al.,

2001). Several cell lines that enable the production of defective recombi-

nant retroviruses have been generated to circumvent the use of helper virus

(Cone and Mulligan, 1984). The use of retrovirus as an expression vector

presents the following advantages: (i) it allows the expression of genes in

different cell types and species; (ii) stable recombinant cell lines can be

obtained since the genetic material of the virus integrates reliably and

stably in the host genome, and the viral infection does not produce cell

death; (iii) it allows increased infectivity and higher viral titers; (iv) it

allows high efﬁciency for introduction of foreign genes in animal cells

when compared with the commonly used transfection methods. Two

Cloning and expression of heterologous proteins in animal cells 47

relevant points should be considered before employing retroviral vectors:

(i) although they can express genomic sequences containing introns, the

progeny will not harbor them, only copies of the cDNA; (ii) polyadenyla-

tion signals in the target sequence must be avoided because they induce a

premature polyadenylation and the consequent absence of full mRNAs

(Shimotohno and Temin, 1981). The acceptable yields obtained with retro-

viral vectors are explained either by their capacity to integrate into

transcriptionally active loci or, once inserted, they trigger the activation of

the loci (Kaufman, 1990). However, in terms of biotechnological applica-

tions, retroviruses can hardly compete with other viral vectors and present

the following limitations: the size of the foreign DNA to be expressed

(, 6–7 kb), the moderate and variable expression level (one or few copies

of the retrovirus integrate into the chromosomes), and the incapacity to

co-express two heterologous sequences from a single retrovirus due to

epigenic effects (Emerman and Temin, 1984). Parolı´n and Palu´ (2003) have

provided an excellent review on recent progress in the design and applica-

tions of retroviral vectors.

Within the family of Togavirus, two members of the genus Alfavirus,

namely Sindbis virus and Semliki Forest virus, stand out for their use in

biotechnology. They are very simple RNA viruses with a genome encod-

ing for replication signals, structural proteins and for an RNA-dependent

polymerase. The main expression product of the infected cells is the

heterologous gene, since the viral infection down-regulates the synthesis

of host proteins (Strauss and Strauss, 1994). The generation and manipula-

tion of these vectors is simple, and well-established methods exist for the

production of recombinant proteins in different cell lines on a large scale

(Blasey et al., 1997). Lundstrom (2003) has recently outlined the applica-

tions of alfavirus as expression systems.

Among the single-strand RNA viruses, coronaviruses present the

largest genome allowing the insertion of large heterologous sequences

(27 kb). The virus replicates in the cellular cytoplasm and does not

undergo an intermediate stage as DNA. Therefore, its integration into the

host genome is an unlikely phenomenon. A description of the different

strategies available nowadays for the generation and use of coronaviruses

as expression vectors has been published by Enjuanes et al. (2003).

3.4.2 Baculoviruses

Baculoviruses encompass a family of viruses that infect different species of

arthropod insects. The name derives from the fact that the DNA genome,

with a size between 80 and 200 kb, is packed in nucleocapsids that acquire

the shape of a walking cane or baculo. A schematic representation of the

life cycle of the virus is shown in Figure 3.1. The polyhedrin gene is not

essential for virus growth in cell cultures and, if deleted, extracellular viral

particles (ECVs) will be released from the cells. This constitutes the basis

for the use of baculovirus as an expression system: the polyhedrin gene is

replaced by the sequence of interest, whose transcription will, thus, be

under the control of the potent polyhedrin promoter. The recombinant

expression unit is transferred by homologous recombination to the locus

of the polyhedrin gene of a wild-type virus.

48 Animal Cell Technology

There are three different approaches to insert genes in a baculovirus

genome: (i) by homologous recombination; (ii) by site-speciﬁc transposi-

tion to a bacmid (a modiﬁed Escherichia coli that contains a copy of the

baculovirus genome) mediated by the transposon 7 (Tn7); or (iii) by

molecular cloning (Condreay and Kost, 2003). Transcriptional promoters

that operate in insect cells are not active in mammalian cells; therefore,

baculovirus-based expression can only be achieved by replacing the poly-

hedrin promoter with CMV or SV-40 promoter sequences (pCMV,

pSV-40). These modiﬁed vectors display an extensive diversity of hosts

(hematopoietic cells excepted) but an expression level that is cell line-

dependent. Variability of expression is, at least partly, associated with

transcriptional silencing of the expression cassette as a result of its

incorporation into the nucleosomes in the genomic DNA (Condreay et

al., 1999). Besides its application for the transient expression of recombi-

nant proteins, it is also possible to obtain stable cell lines by adding to the

vector a cassette with a biochemical marker (Merrihew et al., 2001). The

baculovirus system offers: (i) versatility and diversity of hosts; (ii) absence

Fusion

Solubilization of

protein matrix

(insect gut)

Ingestion

Secondary

infection

Secondary

infection

Primary infection (insect)

Replication

Occlusion

Endocytosis

ECV

Figure 3.1

Life cycle of baculoviruses. The infection of insect cells with baculovirus produces

two types of progeny, the extracellular (ECV) and occluded (OV) viral particles,

which differ structurally and functionally. The nucleocapsid of ECVs is surrounded

byamembraneenvelopeandtheseparticlescausein vitro cell to cell infection but

are not infective to insects. In contrast, the nucleocapsid of OVs is coated by a

protein matrix, formed by polyhedrin or granulin, for the nuclear polyhedrosis or

granulosis virus, respectively. OVs are responsible for de novo infection of insects

but cannot infect c ells cultivated in vitro. The matrix protects the particles against

environmental agents (e.g. UV light, drying), but is degraded at the alkaline pH of

the insect gut. The removal of the protein coat enables the infection of the gut

cells by the virus, which migrates to the nucleus to accomplish the replication and

transcription of its genetic material. At this stage the two viral progenies can be

formed from the nucleocapsids: those accumulated in the nucleus and assembled

to the polyhedrin or granulin give origin to OVs, while those emerging from the

nucleus and cell membrane are ECVs.

Cloning and expression of heterologous proteins in animal cells 49

of cytotoxicity; (iii) transient or stable gene expression; (iv) integration as

a single copy to the genome; (v) easy manipulation and scaling-up of the

production; (vi) stability and conservation of the viral particles; (vii)

capacity to accept large inserts (40 kb); and (viii) high biosafety since

baculovirus is only able to replicate in invertebrates (King and Possee,

1992). The main limitations of this system are related to the use of insect

cells as a platform for the production of recombinant proteins. In this

respect, the expression is discontinuous (batch), because of cell lysis/death,

and the glycosylation pattern is simpler than that of mammalian cells (with

relatively unbranched sugar chains) and with a high content of mannose.

This is usually undesirable for some proteins with therapeutic application

(see Chapter 6 and Luckow, 1995).

3.4.3 Plasmid vectors

The discovery and functional characterization of several of the elements

that control the expression of genes in eukaryotic cells and viruses, in

addition to progress in the ﬁeld of genetic engineering, allowed the

generation of plasmid expression vectors. These types of vectors consist of

naked DNA with a size between 2 and 20 kb. The genetic elements they

contain dictate whether they will exist as episomes or integrated into the

host genome. The basic structure of the plasmid employed for expression

in animal cells is made up of three cassettes: (i) an eukaryotic expression

unit; (ii) a biochemical marker; and (iii) a prokaryotic replication unit (see

Figure 3.2). The ﬁrst cassette drives the expression of the exogenous gene.

The biochemical marker allows for selection of clones that have stably

integrated the plasmid DNA into their genome (see Section 3.8). In this

cassette, the expression of the selection gene is most preferably dictated by

a weak promoter to increase the chances of isolating high producer cell

lines (Wurm, 2004). All the genetic manipulations required during the

construction of the recombinant plasmids, as well as their ampliﬁcation

(replication) in sufﬁcient amounts for future transfections, are routinely

carried out in E. coli. Thus, unlike the viral vectors, plasmids also contain

sequences controlling their replication and selection in a prokaryotic host.

Some novel elements of plasmids, which have extended signiﬁcantly the

range of applications in the biotechnology area, are discussed below.

Inducible and speciﬁc promoters. The expression of a heterologous

gene from a constitutive promoter is not recommended if the recombinant

product is cytotoxic or affects cell growth. In such a case, it will be

necessary to regulate the expression levels that can be achieved by means

of an inducible promoter. An ideal inducible system should meet the

following requirements. (i) Speciﬁcity: unresponsive to endogenous acti-

vators and absence of interference with the host physiology. (ii) Efﬁciency:

null to low expression levels in the non-induced state, and quick response

to activation or induction. (iii) Dose dependency: homogenous regulation

of the expression levels. Several sequences and ligands that regulate posi-

tively or negatively the transcriptional promoters of a large number of

animal cells have been identiﬁed and incorporated into expression vectors.

These elements can respond to different external stimuli such as heat

shock, hormones, heavy metals, cytokines, or hypoxia. However, these

50 Animal Cell Technology

stimuli present pleiotropic effects in the host cells, compromising their

speciﬁcity. To overcome this, regulatory elements of evolutionary distant

species are preferred, the most common being the lactose- (Lac), tetracy-

cline- (Tet) or erythromycin- (E; isolated from E. coli), streptogramin or

pristinamycin-operon/repressor (PIP; originating from Streptomyces coeli-

color or S. pristinaespiralis). Operators are DNA sequences that, bound to

speciﬁc molecules (polypeptides, metabolites, etc.), can diminish or in-

crease the afﬁnity of the RNA polymerase for the neighboring promoter,

thus repressing or promoting the transcriptional activity of the adjacent

gene. In the case of the lactose operon, the inducer is isopropyl 1-thio -

D-galactopyranoside (IPTG), which binds to the lactose repressor protein,

alters its conformation, and results in the dissociation of the protein from

the operon, thus de-repressing the expression of the contiguous gene. One

of the disadvantages of this system is that IPTG is toxic at high concentra-

tions (50 mM), restricting its application on a large scale (Makrides, 1999).

Ward et al. (1995), using a recombinant vaccinia virus, designed a thermo-

regulable variant of the lac repressor that showed a temperature-dependent

Eukaryotic expression cassette

Eukaryotic

selection marker

Prokaryotic

cassette

polyA

SV-40 ori

Enh

PTT

Neo

Amp

ColE1

5 UTR⬘

3 UTR⬘

GCCGCC( )CCATGA/G G

MCS

ORF

PCS FUS

TAA( )A/G

Figure 3.2

Basic and optional component s of an expression vector for animal cells. The basic structure of a

plasmid vector can be divided into three cassettes: (1) a eukaryotic expression unit containing as

basic elements: promoter (p), enhancer (Enh), eukaryotic origin of replication (e.g. SV40 ori) if an

episomal replication is preferred, polyadenylation (polyA) and transcription termination signals (TT),

and a multiple cloning site (MCS), where the sequence of interest (ORF) is inserted. Other additional

elements that can be included are: SARs or LCRs (structural elements, SE), targeting signal (ts),

sequence encoding for a fusion protein or tag (FUS) and a protease cleavage site (PCS). The gene to

be expressed (ORF) should contain the Kozak sequence or translational enhancer (TE) and an

appropriate stop codon (TAA); (2) a biochemical marker (e.g. Neo – neomycin-phosphotransferase)

containing the basic elements that regulate its transcription; and (3) a prokaryotic replication unit,

composed of a replication origin (ColE1) and a selection marker (Amp, ampicillin) that allows the

genetic manipulation of the vect or in E. coli. Asterisks indicate optional elements.

Cloning and expression of heterologous proteins in animal cells 51