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

Подождите немного. Документ загружается.

which is ﬁtted with a tangential inlet and closed by an end plate with

an axially mounted overﬂow pipe, also called a vortex ﬁnder. The end

of the cone terminates in a circular apex opening, called an underﬂow

oriﬁce.

The liquid ﬂow path inside a conventional hydrocyclone is shown in

Figure 11.9. The culture medium is pumped through the feed pipe and

upon entry into the device, acquires a downward rotational movement

(primary vortex) trying to leave the device through the underﬂow oriﬁce.

As this opening is not large enough, only part of the liquid is able to

escape – carrying with it the larger cells, forming the concentrated stream.

The fraction of liquid that is not able to leave the hydrocyclone through

the apex opening reverses its movement, creating an upward rotational

movement (secondary vortex), which ﬁnally leaves the device through the

overﬂow pipe (diluted stream) carrying the smallest particles, including

small cells and debris.

In spite of using the same separation principle, the main difference

between a centrifuge and a hydrocyclone is that the latter does not have

moving parts. Therefore, hydrocyclones can cope well with the aseptic

operations required by the biopharmaceutical industry. Apart from that,

when separating cells, they do not require any maintenance during their

working life. This means that they can be operated during very long

periods as cell retention devices in perfusion processes.

Pinto et al. (2007) reported that a hydrocyclone specially designed for

animal cell separations (Deckwer et al., 2005) was able to produce total

separation efﬁciencies above 97% for CHO cells. The possible effects of

the device on the cells were evaluated by different techniques, including

the traditional viability measurement method based on trypan blue

exclusion. Measurements of lactate dehydrogenase activity in the culture

medium to verify if this intracellular enzyme had been released to the

extracellular medium due to cell lysis, indicated that the hydrocyclone

effect on cell viability was negligible. The concentration of apoptotic

cells monitored by ﬂuorescence microscopy showed that passing the

cells through the hydrocyclone did not induce cell death by apoptosis.

Moreover, the cell viability at the concentrated stream (97%) and at the

diluted stream (82%) indicated that the hydrocyclone could selectively

retain viable cells. This is highly desirable when conducting a perfusion

process. This hydrocyclone specially designed for animal cell separation

was also evaluated on a pilot scale by Elsayed et al. (2005). These

authors, when using the hydrocyclone coupled to a 30 L bioreactor

operating in perfusion mode with HeLa cells, found cell viabilities

always above 93%. The same hydrocyclone was also tested on an

industrial scale by the authors of this chapter. In these tests, the

bioreactor had a volume of 300 L and operated with NS0 cells produ-

cing a monoclonal antibody (mAb) (unpublished data). The results

conﬁrmed that the hydrocyclone was capable of promoting cell reten-

tion in a perfusion system, which could be operated at high cell viability

levels (Figure 11.10A), and that the separation efﬁciency for viable cells

was higher than that for dead cells (Figure 11.10B). This conﬁrmed the

capability of a hydrocyclone to selectively retain viable cells in the

bioreactor.

284 Animal Cell Technology

11.6 Filtration

Filtration is a unit operation commonly employed nowadays in biotech-

nological processes. In this unit operation, a ﬁlter medium acts as a

physical barrier to particles larger than its pores. Traditional ﬁltration

devices such as ﬁlter presses and rotary vacuum drum ﬁlters have so far

found no application for the separation of animal cells. Nevertheless,

membrane ﬁlters are commonly employed, as well as some alternative

ﬁlter designs such as spin-ﬁlters. In the next sections, the most common

types of ﬁlters used for animal cell separation will be discussed.

11.6.1 Tangential ﬂow ﬁltration with membranes

Tangential ﬁltration is distinguished from conventional ﬁltration, also

known as dead-end ﬁltration, by the fact that the main ﬂow is tangential

to the membrane surface (Figure 11.11). In dead-end ﬁltration a fast drop

in the permeate ﬂux is observed due to cake formation over the ﬁlter

(A)

(B)

0.0E 00⫹

1.0E 06⫹

2.0E 06⫹

3.0E 06⫹

4.0E 06⫹

5.0E 06⫹

6.0E 06⫹

Time (h)

Cell concentration (mL

⫺1

)

100

Cell viability (%)

X in the bioreactor

X in the perfusate

Viability

100

Time

(

)

Reduced total efficiency (%)

Viable cells

Dead cells

100

150

200

250

Figure 11.10

Results for a 300 L bioreactor operating in perfusion mode using a specially

designed hydrocyclone as a cell retention device.

Animal cell separation 285

medium, but in tangential ﬁltrations this drop in the permeate ﬂux is less

accentuated. Therefore, in systems with adequate hydrodynamic condi-

tions, tangential ﬁlters may operate continuously for relatively long

periods, under stable ﬂux conditions.

In tangential ﬁltration, membranes are used as ﬁlter media. Membranes

are deﬁned as barriers of reduced thickness, across which physical and/or

chemical gradients are established to facilitate the preferential migration of

one or more components from a given mixture, promoting their separation

(Klein, 1991). They are usually made of polymers or inorganic materials,

such as ceramic or sintered steel. In the biopharmaceutical industry,

membranes ﬁnd various applications, such as production of water for

injection (WFI), sterilization of culture media, buffer solutions and gases,

separation of cells and cell debris, and puriﬁcation and concentration of

proteins.

Microporous membranes with nominal pore sizes in the 0.1–10 m

range are usually employed in animal cell separation. Such membranes are

able to retain the cells, but unable to retain solutes of low or high molar

mass (Figure 11.12). Microﬁltration uses a pressure drop, usually in the

0.5–2 bar range, as the driving force to promote separation (Nobrega

et al., 2005).

Two main types of membrane modules are normally employed for

animal cell separation: the plate-and-frame and the hollow-ﬁber modules.

The latter type has as an important feature, a high packing density, which

results in a high permeation area for a compact module. Both module

types may be used in cell separation in batch processes, and also in

perfusion processes. The main characteristics of these two module types

are described in Table 11.1.

Tangential ﬁltration of animal cell suspensions presents several advan-

tages, such as easy scale-up, simple operation, and a permeate totally free

Permeate

(A)

Feed

Permeate

(B)

Feed

Retentate

Figure 11.11

Schematic representation of dead-end ﬁltration (A) and tangential ﬁltration (B).

∆P

Solvent

Solute with low molar mass

Solute with high molar mass

Particle

Figure 11.12

Schematic representation of a microﬁltration process. The applied pressure drop

(˜P) is usually in the range of 0.5–2 bar.

286 Animal Cell Technology

from cells, which allows, for instance, direct application of the permeate to

chromatography columns. However, it also has some serious limitations

related to concentration, polarization and membrane fouling. The ﬁrst

phenomenon is a reversible one and occurs due to the formation of a

concentration gradient of solutes on the membrane surface, at the begin-

ning of the ﬁltration process. Membrane fouling, however, is generally an

irreversible phenomenon. It occurs due to membrane modiﬁcations caused

by components present in the feed stream. These modiﬁcations arise

mainly from adsorption of some molecules onto the membrane, deposition

of suspended material onto the membrane surface and pore clogging by

suspended particles (Nobrega et al., 2005). These phenomena cause a drop

in permeate ﬂux and an increase in solute rejection with ﬁltration time

(Cheryan, 1998).

Tangential ﬁltration was developed in an attempt to reduce the drop in

permeate ﬂux, by the generation of high shear stresses on the membrane

surface, which avoids material deposition and reduces progressive mem-

brane clogging. However, when the suspension to be ﬁltered contains

animal cells, the operating ﬂow rate and, consequently, the shear stress

generated need to be low because of the fragility of the cells. Different

strategies have been proposed in the literature to reduce the progressive

drop in permeate ﬂux. These strategies include periodic back-ﬂushing, use

of a pulsating feed ﬂow, and coating the membrane with polyethylene

glycol to avoid cell and protein adhesion (Smith et al., 1991; Zhang et al.,

1993). Even so, there are few published reports of perfusion processes

employing tangential ﬁltration for long periods (. 20 days).

On the other hand, for cell separations at the end of a batch process,

maintaining a high cell viability is a less critical issue than in perfusion

processes, so that higher shear stress levels are acceptable. Moreover, the

ﬁltration periods are never longer than a few hours. Thus, in batch

processes the advantages of using tangential ﬁltration are greater than its

disadvantages.

11.6.2 Dynamic ﬁlters

The removal of particles accumulated on a ﬁlter medium is strongly

inﬂuenced by the shear rates generated on its surface (Serra et al., 1999;

Silva et al., 2000). Therefore, most membrane modules try to avoid

clogging by maximizing the shear rates (Belfort et al., 1994). However,

this is not always possible in the case of animal cells, due to their shear

Table 11.1 Comparison between hollow-ﬁber and plate-and-frame modules

Characteristics Hollow-ﬁber Plate-and-frame

Internal diameter (m) 40–500 –

External diameter (m) 80–800 –

Membrane area/module volume (m

–3

) 5000–10000 100–600

Maximum operational pressure (bar) 100 (shell), 15 (ﬁbers) 80

Adapted from Nobrega et al. (2005) and Rautenbach (1997).

Animal cell separation 287

sensitivity. Furthermore, in conventional tangential ﬂow ﬁltration, feed

ﬂow rate and shear rate levels are intrinsically coupled.

Dynamic ﬁltration modules present a relative movement between the

membrane and the module, or between the membrane and a rotor. Thus, it

is possible to adjust the shear stress independently of the feed ﬂow rate

and of the transmembrane pressure drop.

Dynamic ﬁltration modules are basically of two types: rotating disc

ﬁlter (RDF) and vortex ﬂow ﬁlter (VFF). In the latter, the ﬁltration

module has a cylindrical shape and has a rotating concentric cylindrical

mesh in its interior. The rotational movement of the internal cylinder

generates a Taylor-Couette ﬂow in the annular gap (Roth et al., 1997),

creating Taylor vortices that minimize concentration polarization and

mesh fouling. Continuous perfusion processes based on this type of ﬁlter

and operating continuously for up to 100 days have been reported

(Mercille et al., 1994).

In RDFs, a disc-shaped rotor is located inside the module. By adjusting

the rotational velocity of the rotor, it is possible to carefully control the

shear stress levels according to the sensitivity of the cells being used.

Therefore, RDFs have been successfully used to separate animal cells,

mainly in batch processes (Kempken et al., 1997; Vogel and Kroner, 1999).

Castilho and Anspach (2003) have developed an RDF (Figure 11.13)

specially designed for mammalian cell separation. These authors employed

computational ﬂuid dynamics (CFD) to optimize the ﬁlter geometry and

also to study the inﬂuence of rotor angular velocity on the shear stress

generated in the device. This ﬁlter was successfully used as a cell retention

device for perfusion cultivation of CHO cells producing a mAb. It was

also used in an integrated cell separation/product puriﬁcation process,

using afﬁnity membranes containing protein A as ligand for antibody

adsorption (Castilho et al., 2002).

11.6.3 Spin-ﬁlters

Spin-ﬁlters are cell retention devices used almost exclusively in perfusion

processes with animal cells. It is common to ﬁnd reports in the literature

referring to internal and external spin-ﬁlters. However, in this chapter the

latter have been classiﬁed as VFFs and as such were described in Section

Membranes

Feed

Filtrate

Retentate

Figure 11.13

Rotating disc ﬁlter developed by Castilho and Anpach (2003).

288 Animal Cell Technology

11.6.2. The internal spin-ﬁlter consists of a cylinder with porous walls

(normally a mesh), which rotates inside a bioreactor. It may be driven by

an independent motor or may be ﬁtted on the impeller shaft (Figure

11.14). Culture medium is continuously removed through the mesh, under

vacuum, while fresh medium is added at the same volumetric rate to the

bioreactor.

Mesh fouling is the main problem associated with the operation of a

spin-ﬁlter. As it is located inside the bioreactor vessel, it is impossible to

exchange the ﬁlter under sterile conditions once it has clogged. When this

happens, the culture must be terminated.

Apart from fouling, cell retention efﬁciency is another important factor

to be considered. An efﬁciency within a range of 63–99% is normally

achieved when working with spin-ﬁlters (Castilho and Medronho, 2002).

In reality, the mechanism of cell retention is not yet well understood. That

is why spin-ﬁlter design and optimization of operational conditions are

still performed empirically. Cells are retained inside the bioreactor as a

function of several forces. The main forces involved are the centrifugal and

lift forces, arising from spin-ﬁlter rotation, and the drag force due to

culture medium removal through the spin-ﬁlter. However, the interactions

between those forces are not well understood. Thus, there are some

reports of high retention efﬁciencies (99%) being achieved using spin-

ﬁlters with a large mesh opening (127 m) (Avgerinos et al., 1990), while

other authors found smaller efﬁciencies (75–95%) when using a mesh

opening of only 20 m (Iding et al., 2000). These discrepancies show the

necessity of a better understanding of the ﬂuid dynamics involved in cell

retention using spin-ﬁlters. In this context, CFD may be a valuable tool to

facilitate this understanding (Castilho and Medronho, 2002; Figueredo-

Cardero et al., in press).

11.7 Ultrasonic separation

Ultrasonic separators, also known as acoustic cell ﬁlters, use a plane

standing-wave to retain cells in a bioreactor. Kilburn et al. (1989) were the

Filtrate

Fresh

medium

Figure 11.14

Schematic representation of a spin-ﬁlter ﬁtted on the impeller shaft of a bioreactor.

Animal cell separation 289

ﬁrst authors to report this possibility. Since then, many reports have been

published about their use at different scales and with different cells

(Shirgaonkar et al., 2004).

The ultrasonic separation of cells occurs due to the forces that are

generated in a plane standing-wave. These forces are the result of inter-

actions between the cells and the ﬂuid, and their magnitude is a function

of density and compressibility differences between the cells and the ﬂuid.

A plane standing-wave occurs as a result of the interference between two

waves of the same wavelength and amplitude traveling in opposite direc-

tions. In such a situation, the cells are retained in pressure nodes (Figure

11.15), forming cell aggregates. The device is operated in cycles, and its

periodic shut-down allows the cell aggregates to settle back towards the

bioreactor.

This type of separator has been used basically for cell retention in

perfusion processes. It operates directly coupled to the top of a bioreactor.

To maximize retention efﬁciency and cell viability when using it, different

operational parameters should be optimized, such as cycle duration,

perfusion and recirculation ﬂow rates, acoustic force intensity, and the

separator backwash frequency (Shirgaonkar et al., 2004).

Gorenﬂo et al. (2005) employed an experimental design methodology

for optimizing the operational conditions of an acoustic ﬁlter with a

10 L d

–1

capacity. They observed a high operational stability of the device,

allowing for its use in long-term perfusion runs. Dalm et al. (2005) also

investigated the long-term stability of a 200 L d

–1

acoustic ﬁlter. The

authors veriﬁed that the separator presented a simple and stable operation

throughout 75 days of perfusion cultivation of a hybridoma.

Several different cell lines have already been investigated in perfusion

cultures using the acoustic cell ﬁlter. For instance, Shirgaonkar et al.

(2004) used NS0, HEK-293, SP2-derived hybridoma, and insect cells in

different serum-supplemented and serum-free media at different perfusion

rates and acoustic chamber volumes. According to these authors, an

adequate operation of the ﬁlter depends on optimization of operational

P piezoceramic

C glass

F polysulfone foil

AV active volume

R reflector

CA cell aggregates

CS cell suspension

H harvest (clarified fluid)

⫽

Figure 11.15

Schematic drawing of the acoustic cell ﬁlter employed by Groeschl et al. (1998).

The pressure nodes are represented by dashed lines.

290 Animal Cell Technology

conditions and strategies. For instance, when working at high cell densi-

ties, nutrient and oxygen depletion may occur during the period of time

the cells remain inside the separator. Another concern is the possible

clogging of the bottom part of the acoustic ﬁlter. Those authors developed

an efﬁcient recirculation and backwashing technique that avoids clogging.

However, these techniques increase the complexity level of the acoustic

ﬁlter operation.

References

Avgerinos GC, Drapeau D, Socolow JS, Mao J, Hsiao K, Broeze RJ (1990), Spin ﬁlter

perfusion system for high density cell culture: production of recombinant urinary

type plasminogen activator in CHO cells, Bio/Technol. 8:54–58.

Belfort G, Davis RH, Zydney AL (1994), The behavior of suspensions and macro-

molecular solutions in cross-ﬂow microﬁltration, J. Membr. Sci. 96:1–58.

Bjo

rling T, Dudel U, Fenge C (1995), Evaluation of a cell separator in large scale

perfusion culture, In: Beuvery EC, Grifﬁths JB, Zeijlemaker WP (Eds), Animal

Cell Technology: Developments Towards the 21

Century, Kluwer Academic,

Dordrecht, pp. 671–673.

Castilho LR, Anspach FB (2003), CFD-aided design of a dynamic ﬁlter for mammalian

cell separation, Biotechnol. Bioeng. 83:514–524.

Castilho LR, Medronho RA (2000), A simple procedure for design and performance

prediction of Bradley and Rietema hydrocyclones, Minerals Eng. 13:183–191.

Castilho LR, Medronho RA (2002), Cell retention devices for suspended-cell perfusion

cultures, Adv. Biochem. Eng. Biotechnol. 74:129–169.

Castilho LR, Anspach FB, Deckwer W-D (2002), An integrated process for mamma-

lian cell perfusion cultivation and product puriﬁcation using a novel dynamic

ﬁlter, Biotechnol. Progr. 18:776–781.

Centritech

(2006), Product information. Available at: http://www.pneumaticscale.

com/content/menus/psc/centritech_centrifuge.aspx (accessed March 2007).

Cheryan M (1998), Ultraﬁltration and Microﬁltration Handbook, Technomic Publish-

ing, Lancaster.

Coelho MAZ, Medronho RA (2001), A model for performance prediction of hydro-

cyclones, Chem. Eng. J. 84:7–14.

Dalm MCF, Jansen M, Keijzer, TMP, Grunsven WMJ, Oudshoorn A, Tramper J,

Martens DE (2005), Stable hybridoma cultivation in a pilot-scale acoustic perfu-

sion system: long-term process performance and effect of recirculation rate,

Biotechnol. Bioeng. 91:894–900.

Deckwer W-D, Anspach FB, Medronho RA, Luebberstedt M (2005), Method for

separating viable cells from cell suspension, Patent US 6878545 B2.

Elsayed A, Piehl G-W, Nothnagel J, Medronho RA, Deckwer WD, Wagner R (2005),

Use of hydrocyclone as an efﬁcient tool for cell retention in perfusion cultures, In:

Go` dia F, Fussenegger M (Eds), Animal Cell Technology Meets Genomics –

Proceedings of the 18

ESACT Meeting (European Society for Animal Cell

Technology), Kluwer Academic Publishers, Dordrecht, pp. 679–682.

Elsayed A, Medronho RA, Wagner R, Deckwer, W-D (2006), Use of hydrocyclones

for mammalian cell retention – I. Separation efﬁciency and cell viability, Eng. Life

Sci. 6:347–354.

Figueredo-Cardero A, Chico E, Castilho LR, Medronho RA (in press), CFD study of

the ﬂuid and particle dynamics in a spin-ﬁlter perfusion bioreactor, In: Hauser H

(Ed.), Proceedings of the 20

ESACT Meeting (European Society for Animal Cell

Technology), Springer, Dordrecht.

Gorenﬂo VM, Ritter JB, Aeschliman DS, Drouin H, Bowen BD, Piret JM (2005),

Animal cell separation 291

Characterization and optimization of acoustic ﬁlter performance by experimental

design methodology, Biotechnol. Bioeng. 90:746–753.

Gro

schl M, Burger W, Handl B (1998), Ultrasonic separation of suspended particles –

Part III: Application in biotechnology, Acustica 84:815–822.

Henzler H, Kauling J, Schmitt F, Beckers E, Boedeker B, Von Hugo H, Konstantinov

K, Naveh D (2003), Continuous high cell density fermentation unit for cultivating

suspended animal or plant cell lines uses sedimentation separator of speciﬁc

surface area, Patent WO2003020919-A2.

Iding K, Lu

tkemeyer D, Fraune E, Gerlach K, Lehmann J (2000), Inﬂuence of

alterations in culture condition and changes in perfusion parameters on the

retention performance of a 20 m spinﬁlter during a perfusion cultivation of a

recombinant CHO cell line in pilot scale, Cytotechnology 34:141–150.

Jockwer A, Medronho RA, Wagner R, Anspach FB, Deckwer WD (2001), Use of

hydrocyclones for mammalian cell retention in perfusion bioreactors, In: Olsson

EL, Chatzissavidou N, Lu

llau E (Eds), Animal Cell Technology: From Target to

Market – Proceedings of the 17

ESACT Meeting (European Society for Animal

Cell Technology), Kluwer Academic Publishing, Dordrecht, pp. 301–305.

Johnson M, Lantier S, Massie B, Lefebvre G, Kamen AA (1996), Use of the

Centritech

lab centrifuge for perfusion culture of hybridoma cells in protein-free

medium, Biotechnol. Progr. 12:855–864.

Kempken R, Preissmann A, Berthold W (1995), Assessment of a disc stack centrifuge

for use in mammalian-cell separation, Biotechnol. Bioeng. 46:132–138.

Kempken R, Rechtsteiner H, Scha

fer J, Katz U, Dick O, Weidemeier R, Sellick I

(1997), Dynamic membrane ﬁltration in mammalian cell culture harvest, In:

Carrondo MJT (Ed.), Animal Cell Technology, Kluwer, Dordrecht, pp. 379–382.

Kilburn DG, Clarke DJ, Coakley WT, Bardsley DW (1989), Enhanced sedimentation

of mammalian cells following acoustic aggregation, Biotechnol. Bioeng. 34:

559–562.

Klein E (1991), Afﬁnity Membranes – Their Chemistry and Performance in Adsorp-

tive Separation Processes, John Wiley and Sons, New York.

Luebberstedt M (2000), Separation tierischer Zellkulturen durch Einsatz von Hydro-

zyklonen, Diplomarbeit des Studienganges Biotechnologie der Technische Fach-

hochschule Berlin, Berlin.

Luebberstedt M, Medronho RA, Anspach FB, Deckwer W-D (2000), Abtrennung

tierischer Zellen mit Hydrozyklonen, Chem. Ing. Tech. 72:1089–1090.

Medronho RA (2003), Solid-liquid separation, In: Mattiasson B, Hatti-Kaul R (Eds)

Isolation and Puriﬁcation of Proteins, Marcel Dekker, New York, pp. 131–190.

Medronho RA, Schu

tze J, Deckwer W-D (2005), Numerical simulation of hydrocy-

clones for cell separation, Latin Am. Appl. Res. 35:1–8.

Mercille S, Johnson M, Lemieux R, Massie B (1994), Filtration-based perfusion of

hybridoma cultures in protein-free medium – reduction of membrane fouling by

medium supplementation with DNAse-I, Biotechnol. Bioeng. 43:833–846.

Nobrega R, Borges CP, Habert AC (2005), Processos de separac¸a

o com membranas,

In: Pessoa A Jr, Kilikian BV (Eds), Puriﬁcac¸a

o de Produtos Biotecnolo´ gicos,

Editora Manole, Barueri/SP, pp. 37–88 (in Portuguese).

Pinto RCV, Medronho RA, Castilho LR (2007), The effects of cell separation with

hydrocyclones on the viability of CHO cells, In: Smith R (Ed.), Cell Technology

for Cell Products – Proceedings of the 19

Meeting of ESACT (European Society

for Animal Cell Technology), Springer, Dordrecht, pp. 745–748.

Rautenbach R (1997), Membranverfahren – Grundlagen der Modul- und Anlagenaus-

legung, Springer Verlag, Berlin, pp. 60–77 (in German).

Roth G, Smith CE, Schoofs GM, Montgomery TJ, Ayala JL, Horwitz JI (1997), Using

an external vortex ﬂow ﬁltration device for perfusion cell culture, BioPharm.

10:30–34.

292 Animal Cell Technology

Schulz C (1996), Einsatz eines Scherﬁlters zur Separation von Tierischen Zellen aus

kontinuierlicher Kultivierung, Diplomarbeit des Studienganges Biotechnologie

der Technischen Universita

t Carolo-Wihelmina, Braunschweig (in German).

Serra CA, Wiesner MR, Laine JM (1999), Rotating membrane disk ﬁlters: design

evaluation using computational ﬂuid dynamics, Chem. Eng. J. 72:1–5.

Shirgaonkar IZ, Lanthier S, Kamen A (2004), Acoustic cell ﬁlter: a proven cell

retention technology for perfusion of animal cell cultures, Biotechnol. Adv.

22:433–444.

Silva CM, Reeve DW, Husain H, Rabie HR, Woodhouse KA (2000), Model for ﬂux

prediction in high-shear microﬁltration systems, J. Membr. Sci. 173:87–98.

Smith CG, Guillaume JM, Greenﬁeld PF, Randerson DH (1991), Experience in scale-

up of homogeneous perfusion culture for hybridomas, Bioproc. Eng. 6:213–219.

Svarovsky L (2000), Gravity clariﬁcation and thickening, In: Svarovsky L (ed), Solid-

Liquid Separation, Butterworth-Heinemann, Oxford.

Takamatsu H, Hamamoto K, Ishimaru K, Yokoyama S, Tokashiki M (1996), Large-

scale perfusion culture process for suspended mammalian cells that uses a

centrifuge with multiple settling zones, Appl. Microbiol. Biotechnol. 45:454–457.

Tebe H, Lu

tkemeyer D, Gudermann F, Heidemann R, Lehmann J (1997), Lysis-free

separation of hybridoma cells by continuous disc stack centrifugation, Cytotech-

nology 22:119–127.

Vogel JH, Kroner KH (1999), Controlled shear ﬁltration: a novel technique for animal

cell separation, Biotechnol. Bioeng. 63:663–674.

Woodside SM, Bowen BD, Piret JM (1998), Mammalian cell retention devices for

stirred perfusion systems, Cytotechnology 28:163–175.

Zhang S, Handa-Corrigan A, Spier RE (1993), A comparison of oxygenation methods

for high-density perfusion cultures of animal cells, Biotechnol. Bioeng. 41:

685–692.

Animal cell separation 293