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

Подождите немного. Документ загружается.

A classic way to produce viral vaccines consists of cultivating cells on an

appropriate static support, infecting them with virus, collecting and

purifying the virus produced and formulating the vaccine.

Although many adherent cell lines have already been adapted to cultiva-

tion in suspension, a large number of viral infection agents do not often

allow the usage of those cells as a platform for prompt vaccinal antigen

production. For this reason, overcoming roller bottles scale limits, micro-

carrier systems have been developed as a faithful technological alternative

for simplifying the adherent cell scale-up process.

Nahapetian et al. (1986) obtained high density Vero cells cultures that

reached 3 3 10

cells/ml, when using a microcarriers system at a perfusion

rate of eight volumes of medium per day, an amount 10 times higher than

that usually obtained with repeated batch cultures. Similar results have

been described by Mendonc¸a et al. (1993, 1999, 2001) who also obtained

high density Vero cell cultures when using perfusion rates of 3 volumes of

medium per day. According to Nahapetian et al. (1986), endogenous

synthesis of cellular growth factors is probably the growth-limiting factor

of under-fed cultures, while factors related to culture conditions such as

aeration and cell metabolite production are involved in the growth limita-

tion of over-fed cultures.

Table 18.1 compares the relationship between cell culture surface area

and bioreactor volume in many different culture systems usually used with

adherent cells. For microcarriers, this coefﬁcient might reach 60 cm

/ml of

medium for culture area prepared with 10 mg of microcarriers per

milliliter. For Roux bottles, this coefﬁcient is around 3 cm

/ml. In cultures

initiated with 2 mg of microcarriers per milliliter of medium, high cell

densities of even 3 3 10

cells/ml are often reached, compared with smaller

cell densities from 2 to 3 3 10

cells/ml usually observed in Roux bottle

systems. Another great advantage of the use of microcarrier culture

systems is the possibility of preparing cell cultures with hundreds or even

thousands of liters (Montagnon et al., 1984).

According to Butler (1987), many materials have been used in the

production of microcarrier particles, and they have speciﬁc requirements

that allow appropriate cell adherence and growth:

(i) particle density should be between 1.02 and 1.05 g/cm

to facilitate

their maintenance in suspension at low agitation rates (40–150 rpm);

Table 18.1 Relationship between cell culture surface area and bioreactor

volume used with animal adherent cells

Cell culture systems Area/volume (cm

/cm

)

Roller bottle 0.2–0.7

Cell factories 1.7

Plastic bags 5.6

Fixed-bed 10–15

Hollow-ﬁber 31

Cytodex

microcarriers (10 g/l) 34

CultiSpher

G microcarriers (30 g/l) – LF 120

444 Animal Cell Technology

(ii) particles should preferably be transparent, allowing microscopic eva-

luation;

(iii) rigid materials (polystyrene and glass) are recommended because of

their low porosity;

(iv) surface charge can be positive or negative, but it should not be too

low because of the risk of difﬁculties of cell adherence, and it should

not be too high because it could inhibit cell growth; the charge should

be equally distributed throughout the surface to insure homogeneous

cell distribution.

Microcarrier particle diameter should ideally be between 100 and 400

m with a size distribution of  25 m to guarantee a homogeneous

culture. According to Butler and Spier (1984), cells tend to adhere

preferably to the smallest particles. However, Hu and Wang (1987)

demonstrated that higher microcarrier diameter tended to promote cell

growth. An increase of diameter from 185 to 265 m resulted in a longer

exponential growth phase, with a ﬁnal cell concentration four times higher

than that observed with microcarriers with smaller diameter.

Adherent cells attached to microcarriers are particularly susceptible to

damage caused by mechanical shear forces within agitated tanks. This

vulnerability to damage is usually associated with cell immobilization and

with the increase in shear force sensitivity of suspended microcarriers.

Croughan et al. (1987) demonstrated that FS-4 and Vero cells are highly

sensitive to shear forces caused by an increase in the mechanical agitation

of the cell culture. They showed a progressive reduction in cell growth

with higher agitation velocities and also cell lysis at an agitation rate over

180 rpm.

A great variety of mammal, bird, ﬁsh, amphibian, and insect cells can be

cultivated in this system. To ensure that cells from so many different

organisms and tissues can be cultivated with success in microcarriers, these

microcarriers should have a physicochemical composition and other

speciﬁc characteristics appropriate to the speciﬁc cell line. The most

commonly used microcarriers are composed of DEAE-dextran polymers,

polyacrylamide, polystyrene, gelatin, or glass. These present a great variety

of density, size, weight, or electric charge that might have a signiﬁcant

effect in cell culture (Varani et al., 1983; Reuveny et al., 1985). Some

examples are given in Figure 18.6.

Many different culture systems have demonstrated effectiveness for viral

production. As previously described, these systems are based on the

growth of cells in suspension or adherent to microcarriers, which are kept

in suspension by agitation. After achieving high density the cultured cells

can be infected by virus, allowing intracellular viral multiplication until

the viral products are ﬁnally collected and processed. After standardiza-

tion and optimization, these systems allow consistent viral particle pro-

duction, and these steps are called the synthesis or upstream phase. Figure

18.7 shows a typical cell membrane structure when rabies viral particles

are leaving the surface of an infected cell.

For viral vaccine preparation or downstream processing, the cells and

supernatant of infected cultures should go through concentration and

purifying processes. These are important steps because there could be a

Viral vaccines: concepts, principles, and bioprocesse s 445

A B

C D

E F

Figure 18.6

Scanning electron microscopy of Vero cells on microcarriers (Yokomizo et al.,

2004). (A) Preparation of a Cytopore

microcarrier without cells (original

magniﬁcation 3242). (B) Preparation of a cell-loaded Cytopore

microcarrier at

day 6 (original magniﬁcation 3330).(C)PreparationofaCultiSpher

microcarrier without cells (original magniﬁcation 3370). (D) Preparation of a cell-

loaded CultiSpher

G microcarrier at day 6 (original magniﬁcation 3295). (E)

Preparation of a Cytodex

I microcarrier without cells (original magniﬁcation

3520). (F) Preparatio n of a cell-loaded Cytodex

I microcarrier at day 6 (or iginal

magniﬁcation 3485).

446 Animal Cell Technology

potential loss of productivity. Therefore, it is not only important to fully

optimize cell growth and the viral infection phase to guarantee a high

product concentration within the synthesis phase, but also to insure that a

highly immunogenic or infective product is fully recovered in the concen-

tration and purifying process. The most relevant optimization parameters

are the multiplicity of infection (MOI) which refers to the virus amount

used in the infection, and the time of infection (TOI), which refers to the

point of infection in relation to the cell multiplication period. MOI and

TOI are parameters that can vary from one type of virus to another or

even from one cell line to another, and their impact is highly signiﬁcant

for increasing the productivity of the system.

18.4 Strategies for the production of virus-like particles

The main objective of immunization against viruses is the prevention or

modiﬁcation of the disease. However, most of the existing classical

vaccines are able to prevent the disease but are not so efﬁcient in prevent-

ing infection (Sandhu, 1994; Ellis, 1999). The development of the recombi-

nant DNA technology made possible the creation of vaccines that do not

present the typical side effects of the vaccines of attenuated or inactivated

viruses. Virus-like particles (VLPs) are one of the new vaccine strategies

arising from recombinant DNA technology.

As mentioned above, viruses are generally composed of a number of

different proteins organized in a regular three-dimensional structure

together with DNA or RNA. Viral proteins are in most cases aggregated

in icosahedric membranes on which certain proteins cooperate to form the

Figure 18.7

Transmission electron micr oscopy of Vero cell growth on a microcarrier surface,

7 days after rabies virus infection.

Viral vaccines: concepts, principles, and bioprocesse s 447

external and internal capsids. In several virus species, viral proteins are

able to form three-dimensional aggregates in the absence of nucleic acids,

thus creating VLPs. Similarly, the proteins that normally interact with

nucleic acids can also aggregate in their absence and form core-like

particles – CLPs. Therefore, in contrast to infectious viruses, VLPs consist

of viral proteins aggregated in a three-dimensional structure similar to that

of the native virus but not containing viral DNA or RNA within the

capsid (Cruz et al., 2002).

18.4.1 Advantages of VLPs

The major advantage of VLPs over individual puriﬁed antigens is the more

efﬁcient antigen presentation to the immune system, since the epitopes

resulting from the three-dimensional structure of the capsid will be avail-

able for recognition by the immune system. Also, many of the viruses for

which immunogenic VLPs were developed are difﬁcult to replicate in vitro

(e.g. HBV, parvovirus B19, Norwalk calcivirus, and human papilloma-

virus). Finally, it should be noted that VLPs are not infectious, thus

eliminating the need for inactivation and subsequent epitope modiﬁcation

due to the inactivation agent, as in the case of inactivated viruses (Cruz

et al., 2002).

18.4.2 VLP production technology

Due to the availability of the viral coding sequences, it has been possible to

develop methods for the production of the viral proteins using appropriate

expression vectors. The fact that VLPs are multimeric structures makes the

use of complex expression technologies mandatory. In addition, viruses

typically infect eukaryotic cells and thus the use of such cells is required for

correct viral protein production. Although this is not necessarily true for

the production of a single viral protein that does not have complex post-

translational modiﬁcations, such as glycosylation and phosphorylation, for

production of complete VLPs, the use of complex systems cannot be

avoided. This is due to the fact that the processing of viral proteins within

the cells and capsid assembly, require the mediation of host proteins – the

chaperones. In recent years several systems were developed with the goal of

producing viral proteins in yeast, insect, and mammalian cells. The use of

yeast was ﬁrst demonstrated in the production of the HBV surface antigen

(HBsAg) used as a vaccine against hepatitis B (Fu et al., 1996). The expres-

sion of this speciﬁc protein had been previously shown using Escherichia

coli; however, the molecules did not form VLPs and their immunogenicity

was low. The vaccine against hepatitis B is an important milestone in the use

of VLPs as a vaccination strategy, leading to replacement of the previously

existing inactivated virus vaccine.

A second milestone in the production of viral antigens was the develop-

ment of the technology associated with insect cells – the baculovirus

expression system. The Autographa californica baculovirus produces viral

particles as a part of its life cycle. These particles accumulate in the

polyhedrin protein matrix, which confers protection from inactivation due

to environmental factors. The polyhedrin is produced in large amounts

448 Animal Cell Technology

(1 mg/10

cells) under the control of a very strong promoter. The use of

this promoter to drive the expression of foreign proteins became the

obvious next step in producing recombinant baculoviruses. The cells

derived from the ovary of the caterpillar Spodoptera frugiperda (Sf-9 and

Sf-21) are the most widely used for the production of heterologous

proteins using baculoviruses (Table 18.2). Nevertheless, other insect cells

have also been used for this purpose, including Trichoplusia ni (High-

Five

) (Jiang et al., 1998; Wang et al., 2000).

18.4.3 VLP composition

The proteins to be included in VLPs should be those necessary to confer

the desired degree of immunogenicity. As a consequence, VLPs are often

composed of more than one protein. In order to use baculovirus-infected

insect cells to produce VLPs it is necessary to predeﬁne the proteins to be

included, since the expression of these proteins could determine the

stability (Hyatt et al., 1993) and location – intra- or extracellular – of the

particle (French and Roy, 1990).

The antigenicity, rather than the immunogenicity, should drive the

research and development of new VLP-based vaccines and thus antibody

reactivity tests should be performed as early in the process as possible. For

example, a porcine parvovirus vaccine is composed of a single viral protein

(VP2), which represents 95% of the native virus total protein and is able

do induce antibody production in immunized animals (Rueda et al., 1999).

In contrast, the human parvovirus B19 contains the exact same proportion

of VP2 in the native virus but VLPs made solely of VP2 are unable to

induce neutralizing antibodies (Brown et al., 1991; Tsao et al., 1996). In

this case even a VLP containing VP1 and VP2 at a ratio of 1:24,

respectively, which is very similar to that of the native virus, was not

Table 18.2 Virus-like particles (VLPs) produced in baculovirus-infected insect

cells

VLPs Cell line Reference

Human parvovirus B19 Sf-9 Bansal et al., 1993

Blue tongue virus Sf-9, Sf-21 Roy, 1990

Epizootic hemorrhagic disease Sf-9 Le Blois et al., 1991

Hepatitis B Sf-9 Lanford et al., 1989

Hepatitis C Sf-9 Baumert et al. 1998

HIV Sf-9 Cruz et al., 2002

Infectious bursal disease Sf-9, HighFive Wang et al., 2000;

Hu and Bentley, 1999

Norwalk virus Sf-9 White et al., 1997

Papillomavirus Sf-9 Kirnbauer et al., 1992

Minute virus of mice Sf-9 Hernando et al., 2000

Polyomavirus Sf-9 Montross et al., 1991

Porcine parvovirus Sf-9, Sf-21 Maranga et al., 2004;

Martinez et al., 1992

Rotavirus Sf-9, HighFive Vieira et al., 2005;

Jiang et al., 1998

SV40 Sf-9 Kosukegawa et al., 1996

Viral vaccines: concepts, principles, and bioprocesses 449

efﬁcient in inducing an immune response. The ﬁnal composition of the

antigenic VLP included VP1 at a percentage between 25 and 40%, that is,

6–10-fold higher than in the native viral particle (Tsao et al., 1996). This

clearly shows that there is still a need for a better understanding of the

immunization phenomenon.

Therefore, to ensure antigenicity, the particle composition has to be

studied and will probably lead to different baculovirus infection strategies,

using a different number of proteins per virus. For instance, the blue

tongue virus (BTV), an Orbivirus from the Reoviridae family, is a non-

enveloped virus that contains seven different structural proteins (VP1–

VP7) (Hyatt et al., 1993). The outer capsid consists of two proteins, VP2

and VP5, while the core is composed of the remaining ﬁve proteins, two

major (VP3 and VP7) and three minor (VP1, VP4, and VP6), classiﬁed

according to their abundance. The production of BTV VLPs can be

performed by co-infecting the cells with two dual-gene vectors expressing

VP3 + VP5 and VP2 + VP7 (Brown et al., 1991) or through the infection

of a multigenic vector containing all four proteins (VP2, VP3, VP5, and

VP7) (Martinez et al., 1992). The factors related to the VLP composition,

including the use of multigenic vectors, production technology, and large-

scale production VLPs, have been analyzed by Maranga et al. (2000a).

18.4.4 VLP production processes

Hepatitis B vaccine

HBV is transmitted among humans and is manifested as a chronic

infection that debilitates infected individuals and may cause severe liver

damage, primary carcinoma, and ultimately death. Although in most cases

the patients recover, there are large segments of the population that have

chronic hepatitis B (especially in African and Asian countries) and can

transmit the disease as a pandemic.

HBV is an enveloped virus whose nucleocapsid involves one single

DNA molecule. Several surface antibodies have been found against the

surface antigens (HBsAg) in the serum of HBV-infected individuals. The

HBsAg is present in the blood of infected individuals in the form of

spherical particles with a diameter of 22 nm. Since these particles are able

to induce immunization in humans, a production process based on yeast

cloned with the gene encoding the HBsAg monomer has been developed.

However, since yeasts do not secrete the VLPs, sonication is used to

extract the particles, which are subsequently puriﬁed.

The VLPs produced by yeast consist of about 100 HBsAg monomers

and confer immunization when injected into humans in spite of the

different glycosylation as compared with the native virus (Fu et al., 1996).

Nevertheless, the vaccine only confers immunity after three doses and

requires revaccination after 5 years.

Production of HIV-1 and porcine parvovirus VLPs

The production processes for HIV-1 and porcine parvovirus (PPV) VLPs

are presented in Figure 18.8. In both cases, the insect cell–baculovirus

450 Animal Cell Technology

system was used but the production strategy has some signiﬁcant differ-

ences. In the case of HIV-1 VLPs, Sf-9 cells were infected with a high

MOI (Cruz et al., 1998), while in the case of PPV VLPs, Sf-21 cells were

preferred and a low MOI was used for economic reasons, as these VLPs

are developed as a veterinary vaccine (Maranga et al., 2004). Also, at the

puriﬁcation level there are some differences, since the HIV-1 VLPs are

secreted and PPV VLPs are intracellular, thus requiring a cell disruption

step. However, it should be noted that both downstream processing

strategies are based on the relatively larger size of the VLP in comparison

with other contaminants, although chromatographic steps are used only

for HIV-1 VLPs (Cruz et al., 2002).

18.5 Development of viruses for DNA vaccines

Free DNA-mediated immunization, also referred to as genetic immuniza-

tion, consists of the administration of genetic material (DNA in the form

of a plasmid), which will lead to the in vivo expression of proteins that

induce an immune response. This vaccination method is simple and

constitutes a relevant alternative to the classical vaccination, mainly due to

two important advantages. First, the long-term expression of small

amounts of antigens is possible, avoiding the need for re-administration, as

long as no allergies, tolerance, or autoimmunity occur. Second, the

antigens are synthesized in vivo, no infection exists and therefore inad-

vertent reactions are avoided and the treatment of infected individuals

becomes possible. The ﬁrst animal studies using this type of vaccines

demonstrated their potential in the protection against the ﬂu virus. DNA

vaccines also have other advantages, namely the higher thermal, chemical,

Bioreaction

Infection (high MOI)

Cell

growth

Production

Centri-

fugation

Micro-

filtration

Ultra-

filtration

Gel exclusion

chromatography

Baculovirus

inactivation

Cell

growth

Centri-

fugation

Cell

disruption

Preci-

pitation

Ultra-

filtration

Production

(batch)

Purification

(batch) (0.45 m)µ (300 kDa)

Figure 18.8

Production and downstream processing of HIV-1(A) and PPV (B) VLPs (adapted from Cruz et al.,

2002).

Viral vaccines: concepts, principles, and bioprocesses 451

and biological stability of DNA in comparison with classical vaccines.

This lowers the requirements for storage at low temperature, thus reducing

the costs, especially for developing countries.

Nevertheless, the large-scale application in healthy individuals still

involves delicate safety issues, although clinical trials are expected for

cancer and HIV infection.

The spectrum of target diseases can be expanded through the use of a

mix of different plasmids, a strategy that is difﬁcult to implement when

using proteins due to interference. Given the ease of modiﬁcation of DNA

sequences, a single gene may include several epitopes of more than one

antigen or different pathogenic strains.

There are several methods of administration. Free DNA can be directly

injected into muscle cells (currently the most efﬁcient method), linked to

gold particles which are then bombarded onto the tissue (difﬁcult to

prepare), pulverized without injection or administered orally. In spite of

the low efﬁciency (in terms of modiﬁed cells), this new vaccination

method frequently leads to surprisingly high immune responses at both

the humoral and cellular level. The technology used in this type of

vaccination has the additional advantage of allowing research in the area of

molecular immunology, as well as in the study of the immune system and

new vaccine development due to the ease of producing plasmids encoding

different proteins. This can be performed within a few weeks as compared

with the months necessary for new viral vectors. Nevertheless, viral

vectors, especially adenoviruses, are currently being used in DNA vaccina-

tion. The ability of adenoviruses to very efﬁciently induce an immune

response has several advantages: (a) to provide high amounts of antigen to

lymphoid tissue; (b) to induce a rapid T-cell expansion and migration in

the lymphatic system; and (c) to promote and prolong T-cell response

(Yang et al., 2003). In addition, adenoviruses are non-integrating vectors

and their biology is well understood. In parallel, viral vectors based on

other virus types have been developed, for example inﬂuenza virus (Ulmer

et al., 1998). Currently, Crucell is developing two DNA vaccines against

malaria and Ebola virus based on adenovirus Ad35 and Ad11, two

serotypes that are not common, thus eliminating the problem of existing

immunity against Ad5. In this case it is possible to obtain a good immune

response and an effective protection against infection. There are several

DNA vaccines under development against a number of infectious diseases

(Table 18.3).

The application of the DNA vaccine technology is not limited to

infectious diseases and is currently being used in therapeutic vaccines

against cancer. Table 18.4 presents the therapeutic DNA vaccines under

development (Powell, 2004).

Further detail on gene therapy is provided in Chapter 21, which focuses

on the use of viruses for therapeutic applications instead of prophylaxis.

18.6 Perspectives for the evolution of viral vaccine production

Vaccination has prevented, in a safe and efﬁcient way, more diseases and

deaths due to infectious agents than any other public health policy except

452 Animal Cell Technology

for clean water supply. Among the several international programs

launched in the 20th century to eradicate diseases, the only successful one

to date was based on vaccines – the eradication of smallpox. Poliomyelitis

may be the next success within the next decade. Table 18.5 shows the

evolution of the viral vaccines available on the market from live/attenuated

vaccines to the more recent subunit and recombinant DNA vaccines.

The major problem associated with recombinant vaccines is related to

the immune response that is often only humoral (antibody production)

and not cellular (e.g. cytotoxic T-lymphocytes, CTL) (Ellis, 1999). Nor-

mally, viral vaccines are able to stimulate both types of immune response,

thus avoiding the need for revaccination (Ellis, 1996).

Another important issue concerns the number of vaccines used for

children under 2 years of age; the vaccination program includes around 15

injections causing some discomfort not only to the children but also to

their parents (Papaevangelou, 1998). It is therefore necessary to develop

combined vaccines – live, inactivated or mixed with VLPs or other

subunits – for which the allergic reactivity is minimal.

To circumvent these two problems and taking advantage of the VLP

stability, some methods have been developed to introduce multiple epi-

topes in the same VLP. One example is the use of porcine parvovirus VLP

containing epitopes to induce cellular immune response (B cells, CD4+,

and CTL). This strategy should allow the production of cheaper and more

potent vaccines that will in turn induce a more effective immune response

Table 18.3 DNA vaccines under development

Vector type Target Company Development stage

Adenovirus Ebola Crucell/NIH Preclinical

Hepatitis C Merck & Co. Preclinical

HIV Merck & Co. Phase I

Malaria Crucell/GlaxoSmithKline Preclinical

Rabies Vaxin Preclinical

West Nile virus Pﬁzer/Kimron Vet. Inst Preclinical

Plasmids Ebola NIH Vaccine Research Center Phase I

HIV Wyeth Lederle Phase I complete

HIV Wyeth Lederle Phase I

HIV Chiron Phase I

HIV Epimmune Phase I

Table 18.4 Therapeutic DNA vaccines under development

Vaccine Company Development stage

HIV-1 Corixa Phase I

Melanoma ImClone Systems Phase I

Lung cancer ImClone Systems Phase III

Prostate cancer Inovio Phase I/II

Melanoma M.D. Anderson Cancer Center Phase I/II

Melanoma Vical Phase III

Viral vaccines: concepts, principles, and bioprocesses 453