letters to nature
`
`locus. In agreement, the wing phenotypes (thickened veins and
`wing-edge loss) of Nnd-3 mutants are completely suppressed by pk-
`sple13 (data not shown). The inhibitory effect of Sple provides a
`second mechanism for polarizing Notch signalling in R3/R4 that
`could be coordinated by Fz/Dsh (Fig. 4d).
`In conclusion, Notch signalling is activated in R3/R4 in response
`to Fz/Dsh and we propose that Fz/Dsh sets up an initial bias in
`Notch activity between R3 and R4 by promoting Dl activity and
`inhibiting Notch via Sple in a coordinated manner. Feedback in the
`Notch pathway ampli®es this bias so that even a subtle variation in
`the amount of signal received by Fz in the equatorial (pre-R3) and
`polar (pre-R4) cells in each ommatidium would generate distinct
`cell fates. This explains how a signal from the equator could be
`interpreted by the whole ®eld of ommatidia (Fig. 4e) and is likely to
`be of widespread signi®cance in the development of polarized
`structures within planar epithelia. The asymmetrical expression of
`Notch-pathway genes detected in feather primordia is consistent
`with this model24. Furthermore, this mechanism for the coordinated
`regulation of Notch signalling can also explain how neural precursors
`develop at speci®c positions within competent proneural ®elds. M
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`Methods
`Fly strains. Alleles used were: Nts1, Nnd-3, dsh1, fzR54, Dlrf , Dl6B; sple1 and
`pk-sple13. For analysis of E(spl) a combination between Df(3R)NF1P1 (removing
`NF1, E(spl)md and E(spl)mg promoter25; M.T.D.C. and S.J.B., unpublished
`data) and Df(3R)E(spl)grob32.2 (removing all E(spl) genes and NF1) was used.
`Rescue of NF1 activity25 did not modify the eye phenotypes and no defects were
`observed in NF1P2/Df(3R)NF1P1, which eliminates NF1 only25. For mis-expression
`studies we used heat-shock-inducible, intracellular Notch (hs-Nicd; ref. 15), a
`transmembrane-activated Notch derivative driven by sevenless
`enhancer
`(sevE-NDecd; ref. 17), and the Gal4/UAS-targeted expression system26. UAS lines
`were: UAS-Nicd (gift of M. Haenlin), UAS-Ndn (ECN, containing Notch extra-
`cellular and transmembrane domains16), UAS-dsh19, UAS-E(spl)md and UAS-
`E(spl)mb. These were combined with sev-Gal4 (expressed in R3, R4, R7, mystery
`and cone cells) and/or spalt-Gal4 (expressed in R3, R4 and cone cells).
`For Nts experiments, larvae were incubated at 30 8C for 6 h, returned to 25 8C for
`10 h or until eclosion. Nicd expression was induced in hs-Nicd larvae by 2 h at 37 8C.
`md0.5 transgenic lines. A 487-bp fragment from the 1.9-kb genomic HindIII
`fragment upstream of E(spl)md was ampli®ed using the primers GATCTA-
`GATGCCATCAGATGTCAGC and CTACTAGTCTTTTGGCGCACAGTCAC,
`digested with SpeI (®lled in) and XbaI, and ligated into Asp 718 (®lled in) and
`XbaI sites of HZ50PL. Transgenic lines were established by injection in cn;ry ¯ies
`using standard procedures and all lines gave identical patterns of expression.
`Immuno¯uorescence. the following antibodies were used: rabbit anti-b-
`galactosidase (Cappell), rabbit anti-Bar27, guinea-pig anti-Coracle28, rat anti-
`Elav (Developmental Studies Hybridoma Bank), rat anti-Spalt (a gift of
`R. Barrio) and mouse monoclonal antibodies against b-galactosidase (Pro-
`mega), E(spl) proteins29, Rough30 and Delta20.
`
`Received 26 October; accepted 18 December 1998.
`
`1. Tomlinson, A. Cellular interactions in the developing eye. Development 104, 183±193 (1988).
`2. Wolff, T. & Ready, D. F. in The Development of Drosophila melanogaster (eds Bate, M. & Martinez-
`Arias, A.) 1277±1325 (Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1993).
`3. Zheng, L., Zhang, J. & Carthew, R. W. frizzled regulates mirror-symmetric pattern formation in the
`Drosophila eye. Development 121, 3045±3055 (1995).
`4. Fanto, M., Mayes, C. A. & Mlodzik, M. Linking cell-fate speci®cation to planar polarity: determina-
`tion of the R3/R4 photoreceptors is a prerequisite for the interpretation of the Frizzled. Mech. Dev. 74,
`51±58 (1998).
`5. Wolff, T. & Rubin, G. M. strabismus, a novel gene that regulates tissue polarity and cell fate decisions in
`Drosophila. Development 125, 1149±1159 (1998).
`6. Theisen, H. et al. dishevelled is required during wingless signaling to establish both cell polarity and cell
`identity. Development 120, 347±360 (1994).
`7. Krasnow, R. E., Wong, L. L. & Adler, P. N. dishevelled is a component of the frizzled signaling pathway
`in Drosophila. Development 121, 4095±4102 (1995).
`8. Strutt, D. I., Weber, U. & Mlodzick, M. The role of RhoA in tissue polarity and Frizzled signalling.
`Nature 387, 292±295 (1997).
`9. Boutros, M., Paricio, N., Strutt, D. I. & Mlodzik, M. Dishevelled activates JNK and discriminates
`between JNK pathways in planar polarity and wingless signaling. Cell 94, 109±118 (1998).
`10. Wehrli, M. & Tomlinson, A. Independent regulation of anterior/posterior and equatorial/polar
`polarity in the Drosophila eye; evidence for the involvement of Wnt signaling in the equatorial/polar
`axis. Development 125, 1421±1432 (1998).
`11. de Celis, J. F. et al. Functional relationships between Notch, Su(H) and the bHLH genes of the E(spl)
`complex: the E(spl) genes mediate only a subset of Notch activities during imaginal development.
`Development 122, 2719±2728 (1996).
`
`12. Eastman, D. S. et al. Synergy between suppressor of Hairless and Notch in the regulation of Enhancer of
`split mg and md expression. Mol. Cell. Biol. 17, 5620±5628 (1997).
`13. Cagan, R. L. & Ready, D. F. Notch is required for successive cell decisions in the developing Drosophila
`retina. Genes Dev. 3, 1099±1112 (1989).
`14. Baker, N. E., Sung, Y. & Han, D. Evolution of proneural atonal expression during distinct regulatory
`phases in the developing Drosophila eye. Curr. Biol. 6, 1290±1301 (1996).
`15. Struhl, G., Fitzgerald, K. & Greenwald, I. Intrinsic activity of the Lin-12 and Notch intracellular
`domains in vivo. Cell 74, 331±345 (1993).
`16. Klein, T., Brennan, K. & Arias, A. M. An intrinsic dominant negative activity of serrate that is
`modulated during wing development in Drosophila. Dev. Biol. 189, 123±134 (1997).
`17. Fortini, M. E., Rebay, I., Caron, L. A. & Artavanis-Tsakonas, S. An activated Notch receptor blocks cell-
`fate commitment in the developing Drosophila eye. Nature 365, 555±557 (1993).
`18. Fischer-Vize, J. A., Vize, P. D. & Rubin, G. M. A unique mutation in the Enhancer of split gene complex
`affects the fates of the mystery cells in the developing Drosophila eye. Development 115, 89±101 (1992).
`19. Axelrod, J. D., Matsuno, K., Artavanis-Tasakonas, S. & Perrimon, N. Interaction between Wingless
`and Notch signaling pathways mediated by Dishevelled. Science 271, 1826±1832 (1996).
`20. Parks, A. L., Turner, F. R. & Muskavitch, M. A. T. Relationships between complex Delta expression and
`the speci®cation of retinal cell fates during Drosophila eye development. Mech. Dev. 50, 201±216
`(1995).
`21. Greenwald, I. LIN-12/Notch signaling: lessons from worms and ¯ies. Genes Dev. 12, 1751±1762
`(1998).
`22. Choi, K. W., Mozer, B. & Benzer, S. Independent determination of symmetry and polarity in the
`Drosophila eye. Proc. Natl Acad. Sci. USA 93, 5737±5741 (1996).
`23. Gubb, D. Cellular polarity, mitotic synchrony and axes of symmetry during growth. Where does the
`information come from? Int. J. Dev. Biol. 42, 369±377 (1998).
`24. Chen, C.-W. J., Jung, H.-S., Jiang, T.-X. & Chuong, C.-M. Asymmetric expression of Notch/Delta/
`Serrate is associated with the anterior-posterior axis of feather buds. Dev. Biol. 188, 181±187 (1997).
`25. The, I. et al. Rescue of a Drosophila NFl mutant phenotype by Protein Kinase A. Science 276, 791±794
`(1997).
`26. Brand, A. H. & Perrimon, N. Targeted gene expression as a means of altering cell fates and generating
`dominant phenotypes. Development 118, 401±415 (1993).
`27. Higashijima, S. et al. Dual Bar homeo box genes of Drosophila required in two photoreceptor cells, R1
`and R6, and primary pigment cells for normal eye development. Genes Dev. 6, 50±60 (1992).
`28. Fehon, R. G., Dawson, I. A. & Artavanis-Tsakonas, S. A Drosophila homolog of membrane skeleton
`protein-4.1 is associated with septate junctions and is encoded by the coracle gene. Development 120,
`545±547 (1994).
`29. Jennings, B., Preiss, A., Delidakis, C. & Bray, S. The Notch signalling pathway is required for Enhancer
`of split bHLH protein expression during neurogenesis in the Drosophila embryo. Development 120,
`3537±3548 (1994).
`30. Kimmel, B. E., Heberlein, U. & Rubin, G. M. The homeodomain protein Rough is expressed in a
`subset of cells in the developing Drosophila eye where it can specify photoreceptor subtype. Genes Dev.
`4, 712±727 (1990).
`
`Acknowledgements. We thank R. Barrio, J. de Celis, M. Dominguez, D. Gubb and M. Haenlin for
`reagents; M. Dominguez for valuable advice; the multi-imaging center for technical assistance, and
`N. Brown, K. Moses and M. Freeman for comments on the manuscript. This research was supported by a
`Project Grant from the MRC and M.T.D.C. was funded by an MRC studentship.
`
`Correspondence and requests for materials should be addressed to S.J.B. (sjb32@mole.bio.cam.ac.uk).
`
`Cyclosporine induces
`cancer progression by a
`cell-autonomous mechanism
`Minoru Hojo*², Takashi Morimoto³, Mary Maluccio*,
`Tomohiko Asano§, Kengo Morimoto², Milagros Lagman*,
`Toshikazu Shimbo² & Manikkam Suthanthiran*
`
`* Department of Transplantation Medicine and Extracorporeal Therapy, Division
`of Nephrology, and Departments of Medicine and Surgery, Weill Medical College
`of Cornell University, 525 East 68th Street, New York, New York 10021, USA
`² Department of Pediatrics, Mizonokuchi Hospital, Teikyo University School of
`Medicine, 3-8-3 Mizonokuchi, Takatsu-ku, Kawasaki 213, Japan
`³ Department of Cell Biology, New York University School of Medicine,
`550 First Avenue, New York, New York 10016, USA
`§ Department of Urology, National Defense Medical College, 3-2 Namiki,
`Tokorozawa, Saitama 359, Japan
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Malignancy is a common and dreaded complication following
`organ transplantation1±4. The high incidence of neoplasm and its
`aggressive progression, which are associated with immunosup-
`pressive therapy, are thought to be due to the resulting impair-
`ment of the organ recipient's immune-surveillance system5±9.
`Here we report a mechanism for the heightened malignancy
`that is independent of host immunity. We show that cyclosporine
`(cyclosporin A), an immunosuppressant that has had a major
`impact on improving patient outcome
`following organ
`transplantation4,5, induces phenotypic changes, including inva-
`siveness of non-transformed cells, by a cell-autonomous mechanism.
`
`530
`
`© 1999 Macmillan Magazines Ltd
`
`NATURE | VOL 397 | 11 FEBRUARY 1999 | www.nature.com
`
`NOVARTIS EXHIBIT 2070
`Breckenridge v. Novartis, IPR 2017-01592
`Page 1 of 5
`
`

`

`Our studies show that cyclosporine treatment of adenocarcinoma
`cells results in striking morphological alterations, including mem-
`brane ruf¯ing and numerous pseudopodial protrusions, increased
`cell motility, and anchorage-independent (invasive) growth. These
`changes are prevented by treatment with monoclonal antibodies
`directed at transforming growth factor-b (TGF-b). In vivo, cyclos-
`porine enhances tumour growth in immunode®cient SCID±beige
`
`mice; anti-TGF-b monoclonal antibodies but not control anti-
`bodies prevent the cyclosporine-induced increase in the number of
`metastases. Our ®ndings suggest that immunosuppressants like
`cyclosporine can promote cancer progression by a direct cellular
`effect that is independent of its effect on the host's immune cells, and
`that cyclosporine-induced TGF-b production is involved in this.
`We explored an alternative and autonomous cellular mechanism
`
`letters to nature
`
`e ~
`
`,!:
`
`0
`<ii .c
`E
`::, z
`
`1,500
`
`I
`
`:i:..
`E
`
`0)
`
`.S 1 000
`C:
`'
`
`0 e E
`
`l1l
`8
`ca. u..
`
`C:
`
`(!)
`I-
`
`Figure 1 Cyclosporine induces A-549 cells to acquire an invasive phenotype.
`Scanning electron microscopic photographs of A-549 cells grown on glass
`coverslips and incubated for 72 h with: nothing (control cells, a, b); 1 mg ml-1
`f); 1 mg ml-1
`cyclosporine (c, d); 2 ng ml-1 recombinant TGF-b1 protein (e,
`cyclosporine plus 30 mg ml-1 anti-TGF-b (g, h). Note that cyclosporine-conditioned
`cells, in a similar fashion to TGF-b1-treated cells, display membrane ruf¯ing and
`acquire exploratory pseudopodia, and that anti-TGF-b antibodies prevent these
`cyclosporine-induced morphological alterations. Scale bars, 10 mm. i, TGF-b
`concentrations (mean 6 s:d:) in supernatants obtained from untreated or cyclos-
`porine-treated A-549 cells. The cells were incubated for 72 h, in the absence or
`presence of 0.5 or 1.0 mg ml-1 cyclosporine, and a sandwich ELISA assay11 was
`used to quantify TGF-b levels.
`
`Figure 2 Cyclosporine stimulates the motility of A-549 cells. A-549 cells were
`grown for 72 h on 10-mm round polycarbonate membrane ®lters with 0.4 mm (a, b)
`or 3 mm (c, d) pore size in the absence (a, c) or presence (b, d) of 1 mg ml-1
`cyclosporine. The ®lters were removed from the culture dishes and the bottom
`surfaces were examined by scanning electron microscopy. Note that
`the
`pseudopodia of cyclosporine-treated cells grown on 0.4-mm pore size ®lters
`protrude through the pores to the bottom surface (b) whereas cyclosporine-
`treated cells grown on 3-mm pore ®lters migrate through the pores onto the
`bottom surface (d). Scale bars, 10 mm. e, Cyclosporine promotes migration of
`A-549 cells. A-549 cells were placed in the upper chamber of 12-well transwells
`(8 mm pore) at a density of 105 cells, and were incubated alone or with 0.5 or
`1.0 mg ml-1 cyclosporine or with 1.0 mg ml-1 cyclosporine plus 30 mg ml-1 anti-TGF-b
`antibodies (Ab). The cells that migrated into the lower chamber through the 8-mm
`pores of the polycarbonate membrane ®lters were counted after the cells had
`been dissociated by trypsinization. Results (mean 6 s:e:m:) are of three experi-
`ments carried out with duplicate samples.
`
`NATURE | VOL 397 | 11 FEBRUARY 1999 | www.nature.com
`
`© 1999 Macmillan Magazines Ltd
`
`531
`
`NOVARTIS EXHIBIT 2070
`Breckenridge v. Novartis, IPR 2017-01592
`Page 2 of 5
`
`

`

`letters to nature
`
`for immunosuppressant-associated tumour progression. We tested
`the hypothesis that cyclosporine, independently of any effects on the
`host immune system, would programme non-invasive cells to
`acquire an invasive phenotype. The experimental basis for our
`hypothesis was our demonstration that cyclosporine promotes the
`transcription and functional expression of the TGF-b1 gene10,11 and
`the observation by others that TGF-b can promote tumour-cell
`invasion and metastasis12±14.
`A non-transformed human pulmonary adenocarcinoma (A-549)
`cell line14 that is not invasive in vitro was used as the indicator cell to
`test the hypothesis that cyclosporine can induce an invasive phe-
`notype. A-549 cells express functional receptors for TGF-b, and
`their growth and function are regulated by TGF-b (refs 11, 14). We
`carried out these experiments ex vivo to avoid any confounding
`effects of cyclosporine-associated inhibition of in vivo immune
`surveillance mechanisms.
`Figure 1 shows the striking morphological changes observed
`following cyclosporine treatment of A-549 cells. Scanning electron
`microscopic examination revealed that untreated A-549 cells display
`a cuboidal epithelial and non-invasive phenotype (Fig. 1a, b),
`whereas cyclosporine-treated cells show phenotypic alterations
`that are characteristic of invasive cells, that is, marked membrane
`ruf¯ing and the formation of numerous pseudopodia (Fig. 1c, d).
`Additional data (Fig. 1)
`support
`the hypothesis
`that
`the
`cyclosporine-induced acquisition of an invasive phenotype is due
`
`a
`
`b
`
`• 1so~--------~
`
`CsA-treated
`
`CsA-treated
`
`0.L..J---'--~
`Anchorage-dependent
`
`Anchorage-Independent
`
`Figure 3 Anchorage-independent growth of cyclosporine-conditioned A-549
`cells. Culture medium (3 ml) containing 104 cells and 1 mg ml-1 cyclosporine (b, d)
`or no cyclosporine (a, c) was loaded onto 5 ml of a 0.3% agarose layer containing
`MEM±10% FBS in 60-mm dishes. After 96 h incubation at 37 8C, cells grown on the
`surface of the agarose layer were examined with a phase-contrast microscope (a,
`b). For scanning electron microscopic observation (c, d), the soft gels were ®xed
`and then vertical thin sections were made. These slices were processed for
`scanning electron microscopy as described in Methods. Twenty slices were
`made from each agarose plate, and representative ones are shown. Note that
`pseudopodia of cyclosporine-treated A-549 cells protrude deeply into the soft gel
`under anchorage-independent growth conditions. Scale bars, 10 mm. e, Cyclos-
`porine-associated inhibition of anchorage-dependent proliferation of A-549 cells;
`f, cyclosporine-induced stimulation of anchorage-independent growth.
`
`to TGF-b. First, cyclosporine stimulated TGF-b secretion by A-549
`cells in a concentration-dependent manner (Fig. 1i); second, anti-
`15, in contrast to
`TGF-b monoclonal antibodies (1D11.16 IgG)1
`control IgG1 monoclonal antibodies, prevented the cyclosporine-
`induced morphological alterations (Fig. 1g, h); and third, recombi-
`nant TGF-b1 induced morphological alterations in A-549 cells that
`were similar to those elicited by cyclosporine (Fig. 1e, f). Our
`®nding that cyclosporine stimulates TGF-b1 production in A-549
`cells extends earlier observations that it induces T cells10, CCL-64
`mink lung epithelial cells11 and renal cells16 to hyperexpress TGF-b1.
`The phenotypic changes elicited by cyclosporine were reversible;
`incubation of cyclosporine-treated A-549 cells in cyclosporine-free
`culture medium for 48 h resulted in the reversal of the invasive
`phenotype and a return to the original morphology (data not
`shown).
`Cells capable of locomotion and invasiveness display exploratory
`pseudopodia17,18. Because cyclosporine induced numerous, long
`pseudopodia in A-549 cells, we investigated whether such cells
`acquired motility. To explore this, A-549 cells were seeded on
`polycarbonate membrane ®lters with three pore sizes (0.4, 3 and
`8 mm) in the presence or absence of cyclosporine; and the bottom
`surfaces of the membrane ®lters were examined by scanning
`electron microscopy. Our results show that many cyclosporine-
`induced pseudopodia protrude through 0.4-mm pores onto the
`bottom surface (Fig. 2b), whereas only a few pseudopodia protrude
`in the control (Fig. 2a). When the cells were grown on 3-mm pore
`®lters, many cyclosporine-treated cells and only few untreated cells,
`migrated through the pores to the bottom surface of the membrane
`®lter (Fig. 2c, d).
`To quantify the cell motility resulting from cyclosporine treat-
`ment, we used 8-mm pore ®lters in the migration assay. We found
`that the number of A-549 cells that migrated increased in propor-
`tion to the concentration of cyclosporine used to treat the cells, and
`that the increased cell motility was suppressed by anti-TGF-b
`antibodies (Fig. 2e), but not by the control IgG1 antibodies. Thus,
`cyclosporine-induced alterations in both morphology (Fig. 1) and
`cell motility (Fig. 2) were dependent on cyclosporine-induced TGF-
`b production.
`Anchorage-independent growth in vitro is considered a correlate
`of invasive tumour growth in vivo19,20, and so we next examined
`whether cyclosporine treatment results in anchorage-independent
`growth. A-549 cells were plated on soft agarose gels and grown for
`96 h (Fig. 3, see legend for details). Phase-contrast microscopic
`examination revealed that untreated A-549 cells retained their
`spherical shape and remained suspended in the culture medium,
`whereas cyclosporine-treated cells spread and proliferated strongly
`on the soft gel (Fig. 3a, b). Because it was dif®cult to determine by
`phase-contrast microscopy whether the pseudopodia extended
`along the surface of the agarose or penetrated deeper into the
`agarose layer, we made vertical thin sections of the soft gels and
`examined them by scanning electron microscopy to obtain side
`views. This strategy revealed that many fully grown pseudopodia of
`the cyclosporine-treated cells penetrated the agarose-gel layer and
`extended vertically into the gel plate (Fig. 3d). Also,
`these
`cells appeared to be supported by the extensively invaded pseudo-
`podia, in contrast to the absence of pseudopodial extensions in the
`control A-549 cells (Fig. 3c).
`Cyclosporine's effect on A-549 cells growth was contingent
`upon whether the culture conditions were anchorage-dependent
`or -independent. It inhibited the proliferation of A-549 cells under
`anchorage-dependent conditions but
`stimulated proliferation
`under anchorage-independent conditions (Fig. 3e, f).
`We next examined whether cyclosporine induces morphological
`and functional alterations in other cell types, looking at murine
`renal cell adenocarcinoma (Renca) cells, mouse mammary gland
`epithelial (NMuMG) cells and mink lung epithelial (CCL-64) cells.
`We found that cyclosporine treatment produced phenotypic
`
`532
`
`© 1999 Macmillan Magazines Ltd
`
`NATURE | VOL 397 | 11 FEBRUARY 1999 | www.nature.com
`
`NOVARTIS EXHIBIT 2070
`Breckenridge v. Novartis, IPR 2017-01592
`Page 3 of 5
`
`

`

`letters to nature
`
`alterations in these epithelial cells as well. A representative example
`(Fig. 4a, b) shows that cyclosporine-treated Renca cells, in a similar
`fashion to cyclosporine-treated A-549 cells, display an invasive
`phenotype.
`We investigated whether cyclosporine would enhance the invasive
`and metastatic growth of tumour cells in vivo using Renca cells21,
`and two other tumour cell lines, mouse-derived Lewis lung carci-
`noma cells22 and human bladder transitional carcinoma cells23, as
`the tumour inoculum. SCID±beige mice (mice homozygous for
`both SCID and beige mutations24), which are de®cient in T cells, B
`cells and natural killer cells, were used as the host. The use of SCID±
`beige mice minimized the possibility that cyclosporine-induced
`depression of the host's immune system contributed to tumour
`progression.
`Cyclosporine increased the number of murine renal carcinoma
`metastases in SCID±beige mice (Fig. 4c, d). Data from four separate
`experiments showed that the number of renal cell cancer pulmonary
`metastases was 241 6 22 (mean 6 s:e:m:, n (cid:136) 21) in the control
`SCID±beige mice compared with 338 6 26 (n (cid:136) 18) in the cyclos-
`porine-treated mice (P (cid:136) 0:007; t-test) (Table 1). Also, the number
`of pulmonary metastases resulting from inoculation of murine
`Lewis lung carcinoma cells was 11 6 2 (n (cid:136) 9 mice) in the control
`mice compared with 28 6 4 (n (cid:136) 8 mice) with cyclosporine treat-
`ment (P (cid:136) 0:003), while the number of pulmonary metastases
`resulting from inoculation of human bladder transitional cancer
`cells was 63 6 18 (n (cid:136) 9 mice) in controls and 138 6 21 (n (cid:136) 9
`mice) in cyclosporine-treated mice (P (cid:136) 0:01) (Table 1).
`We
`investigated
`the
`effect
`of
`anti-TGF-b antibodies
`(1D11.16 IgG1)15 on the cyclosporine-induced increase in the
`metastases to determine whether in vivo tumour progression by
`cyclosporine was dependent on TGF-b1. Anti-TGF-b antibodies,
`but not control IgG1 antibodies, prevented the cyclosporine-
`induced increase in metastases. The number of pulmonary metas-
`tases was 350 6 22 (mean 6 s:e:m:, n (cid:136) 12) in the control mice,
`441 6 20 (n (cid:136) 10) in cyclosporine-treated mice, 284 6 34 (n (cid:136) 8)
`in mice treated with both cyclosporine and anti-TGF-b, and
`490 6 56 (n (cid:136) 4) in mice treated with cyclosporine and control
`IgG1 (P (cid:136) 0:0005; one-way ANOVA). The reduction in the number
`
`Table 1 Cyclosporine increases pulmonary metastases in SCID±beige mice
`
`Tumour inoculum
`
`Number of pulmonary metastases (mean 6 s:e:m:)
`
`P*
`With CsA
`Without CsA
`.............................................................................................................................................................................
`Murine Renca
`241 6 22 (n (cid:136) 21)
`338 6 26 (n (cid:136) 18)
`0.007
`.............................................................................................................................................................................
`Murine Lewis lung
`11 6 2 (n (cid:136) 9)
`28 6 4 (n (cid:136) 8)
`0.003
`carcinoma (LLC)
`.............................................................................................................................................................................
`Human bladder
`63 6 18 (n (cid:136) 9)
`138 6 21 (n (cid:136) 9)
`0.01
`cancer (T24)
`.............................................................................................................................................................................
`Tumour cells (1 3 105 or 5 3 105 in HBSS) were injected in the tail vein of SCID±beige mice.
`Cyclosporine (cyclosporin A; CsA; 20 mg per kg) was administered every other day from day
`-1 to the day of death. The mice were killed on day 19 (Renca),16 (LLC) or 23 (T24), and the
`number of metastases was counted as described30. n, Number of mice in each group; P
`value derived with t-test.
`
`of metastases found following the administration of anti-TGF-b
`antibodies to cyclosporine-treated mice was signi®cant at P , 0:01
`by ANOVA (Bonferoni P-value). In contrast, there was no signi®-
`cant difference between the number of metastases found in cyclos-
`porine-treated mice and that found in mice treated with combined
`cyclosporine and control IgG1 (P . 0:05). Our in vitro experiments
`show that the tumour cells are the sole source of TGF-b1 (Figs 1, 2).
`Many cell types, in addition to tumour cells, might contribute to
`cyclosporine-induced TGF-b1 hyperexpression in vivo.
`The malignancy-promoting effects of immunosuppressive drugs
`are thought to result from drug-induced T-lymphocyte dysfunction
`and resultant immunosuppression. On the other hand, the produc-
`tion of TGF-b by tumours represents a potential mechanism by
`which they evade the host's immune system25±28. Our demonstration
`that cyclosporine-treated, non-transformed cells acquire invasive-
`ness under in vitro conditions that allow no possible involvement of
`the host's immune system, and our in vivo data that cyclosporine
`promotes tumour growth in SCID±beige mice, suggest a cell-
`autonomous mechanism for cancer progression (Fig. 5). Speci®c
`therapeutic strategies that target pathways responsible for height-
`ened invasiveness (such as TGF-b1 regulation) are worth exploring
`and may be of value to people who are given allografts and to other
`individuals at increased risk of neoplasms.
`M
`
`Cyclosporlne
`
`_m_,Q;:..::1 ~.--1 ____ _
`
`~ . T-lymphocyte function t I
`
`+
`I Malignant transdifferentiation I
`Invasive carcinoma ! I TGF-~ H I
`~ ----1 lmmunosuppressionl
`\.------/
`ff
`
`Invasion
`Metastasis
`Recurrence
`
`Figure 5 Potential mechanisms for cyclosporine-associated tumour progression.
`In this formulation, cyclosporine induced TGF-b production by tumour cells
`promotes cell invasiveness by a cell-autonomous mechanism that is indepen-
`dent of and/or complementary to cyclosporine's immunosuppressive effect on
`the host's immune system.
`
`Figure 4 Cyclosporine induces renal cancer cells to acquire an invasive
`phenotype and promotes tumour growth in vivo. Scanning electron micrograph
`of murine renal adenocarcinoma cells incubated for 72 h in the absence (a) or
`presence of 1 mg ml-1 cyclosporine (b). Scale bars, 10 mm. Representative lungs,
`retrieved from untreated mice (c) and from cyclosporine-treated mice (d), are
`shown to illustrate the cyclosporine-associated increase in renal cell cancer
`pulmonary metastasis in SCID±beige mice.
`
`NATURE | VOL 397 | 11 FEBRUARY 1999 | www.nature.com
`
`© 1999 Macmillan Magazines Ltd
`
`533
`
`NOVARTIS EXHIBIT 2070
`Breckenridge v. Novartis, IPR 2017-01592
`Page 4 of 5
`
`

`

`letters to nature
`
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`Methods
`Cell line and culture. Human lung adenocarcinoma cells (A-549 cells; ATCC
`CCL 185; American Type Culture Collection, Rockville, MD), human bladder
`transitional carcinoma cells (ATCC HTB4, T24), mink lung epithelial cells
`(CCL-64; ATCC), mouse mammary gland epithelial cells (NMuMG; ATCC)
`and Lewis lung carcinoma cells (ATCC) were grown in minimum essential
`medium (MEM) supplemented with 10% fetal bovine serum (FBS), at 37 8C in
`a 95% air±5% CO2 atmosphere. Murine renal adenocarcinoma cells (Renca
`cells; a gift from R. H. Wiltrout, National Cancer Institute) were maintained by
`in vivo serial passages in syngeneic BALB/c mice, as described21.
`Scanning electron microscopy. Cells were seeded at a density of 105 on 12-
`mm round glass coverslips or 10-mm round polycarbonate membrane ®lters
`(0.4 or 3 mm pore size) in 12-well Transwells (Costar), and grown for 72 h in the
`presence or absence of cyclosporine. To assess the ability of TGF-b-speci®c
`antibody to inhibit cyclosporine-mediated effects, A-549 cells were incubated
`in the presence of both cyclosporine and monoclonal antibodies (Genzyme)
`that recognize TGF-b1, -b2 and -b3. Cells were ®xed with PBS, pH 7.4,
`containing 2.0% glutaraldehyde, and processed as previously described29.
`Samples were examined using a JEOL 25SIII electron microscope.
`Quanti®cation of TGF-b. TGF-b was quanti®ed using a sandwich enzyme-
`linked immunosorbent assay (ELISA) method as previously described11. In
`brief, each well of multiwell ELISA assay plates was coated with anti-TGF-b1
`antibodies (1 mg ml-1). The plates were incubated for 2 h at 37 8C after the
`addition of various amounts of TGF-b1 in PBS or conditioned medium. After
`washing with PBS containing 0.2% Tween-20 (PBST), rabbit antiserum against
`TGF-b was added to each well. The plates were incubated at 37 8C for 1 h, the
`wells were washed with PBST, and then 100 ml of goat anti-rabbit IgG±alkaline
`phosphatase conjugates was added. Absorbance at 430 nm was measured using
`an ELISA assay reader. A-549 cells were cultured in serum-free medium to
`exclude contamination of cell-free supernatants by serum-derived TGF-b.
`Cell proliferation assay. For assaying anchorage-dependent growth, A-549
`cells were grown at a density of 2 3 104 cells per well in 12-well plates in the
`presence or absence of cyclosporine. After 96 h treatment, each well received
`2 mCi of methyl-3H-thymidine, and cells were incubated for an additional 4 h.
`They were washed twice with ice-cold PBS and ®xed with methanol for 60 min.
`After washing, the ®xed cells were lysed with 0.2 M NaOH and treated with cold
`10% trichloroacetic acid (TCA) for 15±20 min on ice. The radioactivity,
`recovered as cold TCA-insoluble precipitates, was used for measuring relative
`cell proliferation by comparing the radioactivity between control and experi-
`ment. For an anchorage-independent cell growth, cells spread well on agarose
`gel were counted using a phase-contrast microscope.
`In vivo tumour growth. Murine renal cell adenocarcinoma cells (1 3 105 in
`Hank's balanced salt solution; HBSS), murine Lewis lung carcinoma cells
`(5 3 105 cells) or human bladder cancer cells (1 3 105 cells) were injected in the
`tail vein of 6-week-old male SCID±beige mice. Cyclosporine (20 mg per kg in
`0.2 ml olive oil) was administered every other day starting from day -1, to day
`19 or 23 after tumour inoculation. On day 19 or 23 after tumour inoculation,
`mice were killed and the number of pulmonary metastases was determined30
`following endotracheal insuf¯ation of lungs with 15% India ink solution and
`bleaching the collected lungs in Fekete's solution. The effect of anti-TGF-b
`antibody and the control IgG1 antibody on the cyclosporine-induced increase
`in the number of pulmonary metastases was determined by intraperitoneal
`administration of 200 mg of antibody, on a daily basis starting from day -1 to
`day 19 after tumour inoculation.
`
`Received 21 September; accepted 15 December 1998.
`
`1. Penn, I. Cancers following cyclosporine therapy. Transplantation 43, 32±35 (1987).
`2. Yokoyama, I., Carr, B., Saitsu, H., Iwatsuki, S. & Starzl, T. E. Accelerated growth rates of recurrent
`hepatocellular carcinoma after liver transplantation. Cancer 68, 2095±2100 (1991).
`3. London, N. J., Farmerry, S. M., Will, E. J., Davison, A. M. & Lodge, J. P. A. Risk of neoplasia in renal
`transp