Taxane recovery from cells of taxus in micro- and hypergravity

Acra Astronautica Vol. 42. Nos. l-8. pp. 455-463. 1998 81998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0094-5765/98 $19.00 + 0.00

Pergamon

PII:SOO94-5765(!%3)00138-6

TAXANE RECOVERY FROM CELLS OF TAXUS IN MICRO- AND HYPERGRAVITY D. J. Durzan’ F. Ventimiglia’ and L. HaveI ‘Environmental Horticulture, One Shields Ave., University of California, Davis, CA 95616-8687 [email protected] 2Botany and Plant Physiology, Mendel University of Agriculture and Forestry, 613-00, Bmo, Czech Republic. Abstract Cell suspension cultures of Tams cuspidata produce taxanes that are released from the outer surface of cells into the culture medium as Ike and bound alkaloids. Paclitaxel (TaxolW), is an anti-cancer drug in short supply. It has a taxane ring derived from baccatin III and a C- 13 phenylisoserine side-chain. This drug is produced over a wide range of gravitational forces. Monoclonal and polyclonal antibodies to paclitaxel, baccatin III, and the C- 13 phenylisoserine side chain were combined in multiple-labeling studies to localize taxanes and paclitaxel on cell surfaces or on particles released into the culture medium. Bioreactor vessel design altered the composition of taxanes recovered from cells in simulated microgravity. At ltJ2 and 2 .lW.g, taxane recovery was reduced but biomass growth and percent paclitaxel was significantly increased At 1 to 24g, growth was reduced with a significant recovery of total taxanes with low percent paclitaxel. Bound paclitaxel was also localized in endonuclease-rich fragmenting nuclei of individual apoptotic cells. A model is presented comprising TCH (touch) genes encoding enzymes that modify taxane-bearing xylan residues in cell walls, the calcium-sensing of gravitational forces by the cytoplasm, and the predisposition of nuclei to apoptosis. This integrates the adaptive physiological and biochemical responses of drug-producing genomes with gravitational forces. 0 1998 Elsevier Science Ltd. All rights reserved

Introduction

Diploid cells of Tums cuspidata synthesize free, insoluble, and covalently bound ‘natural* anti-cancer drugs (3,4). Among these, paclitaxel (Tax01~). contains a taxane ring and a phenylisoserin e C- 13 side-chain with a N-benzoyl moiety. The drug inhibits mitosis and is currently used in the treatment of ovarian and breast cancer (12). While plant cell suspension culture technology offers a potentially alternative source of paclitaxel and other valuable naturally occurring compounds in Tuxus sp ( 13), the role of gravitational forces to control drug production in this system has to our knowledge not been examined. Seedlings of T. canadensk at 1.g and in hypergravity (3.g) produced less pa&axe1 than in microgravity ( 105.g) over 30d (2). This result suggests that gravity is one of several stresses that may constrain taxane over-production. Given this notion, the hypothesis that microgravity increases both taxane and paclitaxel concentrations relative to Earth’s gravity and hypergravity was tested with cell suspensions, and with callus in hypergravity.

455

456

12th Man m Space Symposium

PACLITAXEL AND ITS FUNCTIONAL ANTI-CANCER MOIETIES ( 12)

Methods Materials: Needle callus was obtained from 3-year-old Taxus cuspidata obtained in 1995 from Zelenka Nursery (Grand Haven MI). Explants, derived needles, were surfacesterilized and cultured as a friable callus as described in (4). All cultures were grown in darkness at 20-24°C. Cells were suspended (1 g per 100 ml) in a B5 medium (5) with glucose substituted for sucrose before transfer at the same inoculation density to 50 ml high aspect rotating vessels (HARV 12.5 cm dia.) and/or to lOOmIrotating cylindrical culture vessels (RCCV 7.5 cm dia.) at 1O-‘.g(Synthecon, Houston, TX). Cells were also suspended in 1 L nippled flasks in a clinostat rotating at 1 rpm (simulated 2 x 10A.gbut with significant convectional forces). Cells on semisolid B5 medium in Petri dishes (1.g) were placed in laboratory centrifuges set at 3 and 24g. Diagnostics: Taxanes were separated by HPLC on taxi1 co1umns (Me&hem Technologies Inc., Torrance CA) (4). Taxanes and paclitaxel were localized in cells and on particles with monoclonal and rabbit polyclonal antibodies (6) to the oxetane-bearing taxane ring of paclitaxel, baccatin III, and to taxanes with a C-l 3 side chain, respectively (4). Individual apoptotic cells were distinguished morphologically and histochemically from nonapoptotic cells by a terminal deoxynucleotidyl transferase (TdT)-mediated dUTP labeling of the 3’OH ends of DNA (TUNEL reaction 7,8), which were generated by DNA nicking during cell death. Cells were also examined by laser confocal microscopy (Zeiss Invert LSM 410) in multiple-labeling studies aimed at localizing compounds having the oxetane/taxane ring and/ or the C- 13 side chain. Experimental design: All treatments (1,3,24g) were triplicated over 14 d. Analytical and immunohistochemical assays were performed in not less than triplicate. Statistical differences were supported by t-tests. Microscopic examination involved controls to ensure that localization was not due to nonspecific binding by the antibodies. All samples were corrected for background.

12thMan in SpaceSymposium

457

Results Bioreactor type for microgravity studies: An unexpectedresult was the difference in recovery of taxanes as a function of bioreactor type vir. RCCV or HARV (Figure 1). The RCCV contained a wider range of taxanes and more paclitaxel than the HARV. For all other studies, we settled on the use of RCCV bioreactors fix the study because of membrane distortion problems with the HARVs. Cell Growth and Taxaae Recovery: Cells contin& to grow even at 24g albeit slowly (Table 1). Cells also survived 24g for over 4 months without subculture (unpublished data). Exposure of cells to 1.g and to hypergravity significantly increased total taxane recovery (Table 1). This reflected differences in the ways cells were grown i.e. in a liquid suspension in simulated microgravity vs a moist semi-solid/air interface (1.g and hypergravity). Biomass was increased in microgravity (doubling rate of 7-9 days), but with decreased total taxanes.

Table 1. The recoveq of taxane from cell suspensions and needlecallusof Taxw cuspiabta exposed to gravitationa)forces after 3 weeksin darkness.

Treatment’

Air-Dry Wt 96

1

TAXANES2 mglkg DW3

% Paclitaxel

Nipple flask 1 L 2 x 104-g

7.3 f 0.3

2.44 f 0.20

21.4 f 13.2

Bioreactor RCCV 1o-2 * g

6.2 f 0.0

1.07 f 0.07

42.0 f 0.5

Semi-solid4 1 . g 3-g 24 * g

3.8 f 0.3 3.9 f 0.4 3.5 f 0.4

689.7 f 291.0 537.2 f 165.3 521.5 f 209.6

6.3 f 0.2 7.5 f 0.2 7.7 f 0.9

1. 3 wks darkness, 24OC, BS (glucose carbon source) 2. Competitive inhibition enzyme immunoasseys (mAb BAlO anti-taxane, 3C6 antipaclitaxel, 3H5 anti-baccatin III, Hawaii Biotechnology Group) 3. DW: air-dry weight, SOOC, 24 h 4. Microgravity (nipple flask, bioreactor) yields were significantly different t.= from each other and from other gravitational treatments. No differences were detected among 1, 3, and 24g.

458

12th Man in Space Symposium

T2

rulu

cam#&k nudk waur medlhd ES, KMn f24D

HAFtV

Figure 1. Bioreactor types (RCCV vs HARV, Synthecon, TX) gave different distributions of taxanes after 14 d in culture in darkness at 20-24°C. Chromatographic profiles are compared with chemical standards (bottom) to ident@ products recovered from the bioreactors.

12th Man in Space Symposium

Taraae localization: Taxanes, detected by a monoclonal antibody to the oxetane moiety and a polyclonal for the C-13 side chain were localized as ‘caps’ (c) on cell surfaces, and around spherical zones (0) of exocytosis (Fig. 2). The cap consisted mainly of taxanes reactive to both the oxetane ring (e.g. baccatin III, b green) and to the C- 13 side chain (red). Co-localization at the cap was distinguished from other sites by a yellow color. individual apoptotic cells, detected by the TUNEL reaction (Fig. 3), were found in all treatments. Apoptotic cells were dominant in browning and stressed cells. Apoptosis was characterized by TUNEL-positive and collapsing nuclei. Taxane recovery was greater from aged cells (4). Drug over-production was confirmed in a separate study with another Tuxus species (T. bmviifoliu) (unpublished, 1996). Diiuasion

Cell biomass and percent paclitaxel were enhanced in microgravity, but total taxane recovery was significantly greater at 1.g and in hypergravity. Responses extended earlier results (2) with T. curuuknsis seedlings under gravitational forces. Differences could be explained by the way callus cells were grown as suspensions in bioreacmm in contrast to the seedling body, which had more complex integrated internal correlations and &uctums. Differences in response of cells to the surrounding environm ent were also seen in chromatographic profiles of taxanes from HARV and RCCV bioreactors. The HARV, which is more or less a flat disc with a wide diameter, provided more surface for cell cluster growth than the RCCV. Under the same conditions, HARV produced very little paclitaxel. With the clinostat, the increased convection around cells and tissues reduced pa&axe1 recovery compared to the RCCV. Our study shows that 1.g and hypergravity significantly increased total taxane production primarily in stressed and apoptotic cells. The latter released membranes and particles with bound taxanes into the culture medium. Round paclitaxel was now found in subcellular compartments and in the culture medium. Given these observations, our earlier model for apoptosis (7) was modified (Figure 4). It suggested new strategies for the recovery of taxanes using gravitational forces with control of stresses leading to apoptosis. The model postulates that the adaptive response to gravity (solid box 1, Figure 4) involves touch (TM) gene expression ( 13). This comprises endoxyloglucanases and xyloglucan endotransglycosylases at sites of taxane production and/or binding that adjusted cell walls to gravitational forces. At least 8 xylose derivatives at the 7-OH group of the taxane ring of paclitaxel were already known (10). After solvent extraction of the free taxanes, xylanase treatment released some but not all of the bound taxanes from the residual cellular matter (4). Taxanes, synthesized inside stressed cells (1) reached the outer cell surface by exocytosis (fig. 2). In apoptotic cells, taxanes and particles bearing taxanes were released freely into the culture medium by diffusion in microgravity, and with the aid of convection in the clinostat. In response to gravitational stresses, a signal-transduction cascade (solid and clear boxes 2) releases calcium ions from membranes (e.g. endoplasmic reticulum) and other subcellular compartments (e.g. mitochondria, nuclei). The calcium activates proteases, endonucleases, and alters the availability of energy for divisional cycles. Golgi are activated and contribute to the exocytosis of taxanes and materials to the cell wall. Stressed cells are predisposed to apoptosis if any gravitationally induced genomic damage is not repairable, or if cells become programed for

459

460

12th Man in Space Symposium

Figure 2. Double-label studies showing the association of taxanes bearing an oxetane ring (e.g. b, baccatin III) and a C-13 side chain on the same cell at 1.g. Most paclitaxel and other taxanes were localized as caps (c) on cell surfaces by both antibodies. The taxanes appeared on the cell surface by exocytosis from circular zones (0) below which resided amyloplasts (unpublished data). Co-localization of two antibodies (e.g. c, yellow) indicated the simultaneous occurrence of oxetane-ring and C- 13 side-chain taxanes e.g. paclitaxel, baccatin III. The putative precursor for pa&axe1 viz. baccatin III (b), was localized by a green fluorescence (FITC). The C-13 moiety of paclitaxel and of other taxanes was localized by a polyclonal antibody with a red fluorescence (Cy3). The labeling pattern may also include precursors or products of taxane metabolism having the same structural moieties. Note the sloughing off of taxane-bearing particles into the culture medium (arrow). Laser confocal image. Cell diameter ca 120~.

terminal differentiation (7) (solid boxes 3 to 6). Cell suspensions fed methyl jasmonate, as a stress signal, over-produced paclitaxel(11). Our model links stress metabolism with drug production. It postulates that the stress response and unrestricted taxane binding may comprises an irreversible exit from divisional cycling leading to cell death, terminal differentiation, and necrosis (solid box 7). Regarding drug production in bioreactors, our results point to the utility of sequential process controls. First, advantage may be taken of biomass scale-up in microgravity. Second, taxanes are then over-produced in cells under stress, and without divisional cycling i.e. not growing and

12th Man in Space Symposium

Figure 3. Apoptotic nuclei in individual cells in suspension culture (2 x 10m2.g). Apoptotic nuclei with endonuclease activity were detected by the TUNEL reaction using fluorescence light microscopy. Non-apoptotic cells (C) did not react with the TUNEL assay. Apoptotic nucleus n. Nuclei were usually collapsed and/or fragmented (apoptotic bodies). Cell diameters ranged from 80 to 120 p.

dividing (solid boxes 2 to 4, Figure 4). Third, taxanes are recovered from stressed and deteriorating apoptotic cells (clear boxes 1 to 5, Figure 4). If the above were to apply to the bioproduct recovery system for the minipayload integration center (miniPIC) (9), taxane recovery, based on our model would require design modifications. Conclusioa A model is presented that identifies new steps for the design of more appropriate process controls and bioproduct recovery strategies for bioreactors suitable for microgravity studies. The use of haploid-derived Taxlls cell lines would facilitate the recovery of mutants to determine biosynthetic sequences in our model. This would identify key process-controlling steps for the design of the miniPIC in the Space Shuttle and International Space Station. Taxane recovery profiles were a function of bioreactor design. Taxanes including paclitaxel were localized to surfaces of apoptotic and to a lesser extent, nonapoptotic but stressed cells. Taxane levels were significantly increased by gravitational forces especially hypergravity. Gravitational stremes

461

12th MXI in Space Symposium

462

APDPTDTIC PATHWAY Paclitaxel

D

GRAfA$;NAL micro, 1 x g hypergravity

ryl’;;;;

mitosis/meieaisjlree nuclear cycling 1 SIGNAL TRANSDUCTION 1

poly(ADP)-rtbosyl Histone acetylation checkpoint

APDPTOSIS STARTS at exits from divisional cycles

Enxymelorming systems

8 depolymerlzation degradation

ENZYMES

DNA + DNA ladder Protein + nuclear delamination

climacteric/nonclimactericrespiration loss of osmotic integrity and organizationresistance metabolic overproductionand salvage

I

Endonucleases

g

enzyme actions

Amylates, lipates Mitochondrialeaxymes Transglutaminase, etc.

Figure 4. A model based on apoptosis in plants (7) postulating how gravitational forces affect divisional cycling in cells with the release of taxanes from growing and apoptotic cells. In this model, a key adaptive step (top left) is the expression of touch (TCH ) elicited by gravitational forces. Gene expression regulates cell wall changes through endoxyloglucanases and xyloglucan endotransglycosylases that adjust cell walls at sites for taxane production and binding as a function of cell volume changes. Gravitational stresses lead to the exocytosis of taxanes at specific sites on the plasma membrane (Fig. 2). In the gravisensing and stressed nucleus, checkpoints during divisional cycling determine whether or not genomic processes can continue and/or be repaired. If not, then cells exit divisional cycling and are predisposed to the apoptotic pathway (far right) and/or terminal differentiation. Apoptosis is evident by the start of calciumactivated lyric enzyme activity in nuclei leading to cell death (fig. 3). Cell death may also arise from the release of bound anti-mitotic taxanes that bind cytoplasmic and nuclear sites to block divisional cycles. In this case, cell death is independent of the ~53 (protein) pathway in apoptosis.

12th Man in SpaceSymposium

463

reduced biomass recovery and increased programmed cell death (apoptosis). The assembly of taxanes as caps on cell surfaces after exocytosis comprised a confluence of vesicles, membranes and particles bearing covalently bound taxanes that were released into the culture medium or retained at cell wall surfaces. As a spin-off of this work, the recovery of taxanes from other softwoods (Durzan and Ventimiglia 1996, 1997) used for pulp and paper, may if commercially feasible, add further economic value to the product stream in this industry. The result would be an improved drug supply and a rich source of other potentially useful natural products. Acknowledgements This research was supported by NASA grant NAG 9-825.

References 1. G. Appendino. The phytochemistry of the yew tree. Nat Prod Repts 12: 349-360 (1995). 2. L.J. Cseke, M. Panutich, D. Reichenbach, P. Fozo, N. S. Ghosheh and P. Kaufman, Effects of hyper-g and hypo-g stress treatments on tax01 and epicuticular wax biosynthesis in Canadian yew (Tarus can&e&r). Plant Physiol. 108s: abst. 713 p. 137 (1995). 3. D.J. Durzan and F. Ventimiglia. Taxane production in haploid-derived cell cultures. US Patent No. $5547,866 (1996). Recovery of taxanes from conifers. US Patent No. 5,670,663 (1997). 4. D.J. Durzan and F. Ventimiglia. Free taxanes and the release of bound compounds having taxane antibody reactivity by xylanase in female, haploid-derived cell suspension cultures of Tuxus brevijidiu. In Vitro Cell Biol. 30P: 219-227 (1994). 5. O.L. Gamborg, T. Murashige, T.A. Thorpe, I.K. Vasil. Plant tissue culture media. In Vitro 12: 473-478 (1976). 6. P. G. Grothaus, G. S. Bignami, S. O’Malley, K. E. Harada, J. B. Byrnes, D. F. Waller, T. J. Raybould, Taxane-specific monoclonal antibodies: Measurement of taxol, baccatin III, and ‘total taxanes’ in TUYUSbrevifXu extracts by enzyme immunoassay. J. Nat1 Prod. 58: 1003-1004 (1995). 7. L. Have1 and D.J. Durzan. Apoptosis in plants. Botanica Acta 109: 268-277 (1996). 8. L. Have1 and D.J. Durzan. Apoptosis during diploid parthenogenesis and early somatic embryogenesis of Norway spruce. Intl J. Plant Sci. 157: 8-16 (1996). 9. Gonda. Space bioreactor bioproduct recovery system. MSAD OLMSA NASA HQ and Langley Res Cntr. Advanced Technology Development ‘96. (1996). 10. D.G.I. Kingston, A.A. Molinero, J.M. Rimoldi. 1993. The taxane diterpenoids. Progress in the chemistry of organic natural products. W. Hertz, G.W. Kirby, RE. Moore, W. Steglich, C. Tamm eds. Springer Verlag NY 61: l-192 (1993). 11. J. Pezzuto. Tax01 production in plant cell culture comes of age. Nature Biotech 14: 1083 (1996). 12. M. Stiess. Overview of Taxol research: Progress on many fronts. In: Taxane anticancer agents: Basic science and current status. G. George et al. eds Am. Chem. Sot. Symp. Ser. 583: l-17 (1995). 13. W. Xu, M.M. Purugganan, D.H. Polisensky, D.M. Antosiewicz, S.C. Fry, and J. Braam. Arabidopsis TCH4, regulated by hormones and the environment, encodes a xyloglucan endotransglycosylase. The Plant Cell 7: 1555-1567 (1995).