Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 1 of 24 PageID #: 545
`Case 1:18-cv-01363-CFC Document1-15
`Filed 09/04/18
`Page 1 of 24 PagelD #: 545
`
`
`
`EXHIBIT O
`EXHIBIT O
`
`
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 2 of 24 PageID #: 546
`
`(12) U n i t ed States P a t e nt
`Andersen et al.
`
`US006610516B1
`US 6,610,516 Bl
`*Aug. 26,2003
`
`(io) Patent No.:
`(45) Date of Patent:
`
`(54) CELL CULTURE PROCESS
`
`(75)
`
`Inventors: Dana C. Andersen, Redwood City, CA
`(US); Tiffany M. Bridges, Burlingame,
`CA (US); Martin Gawlitzek, Foster
`City, CA (US); Cynthia A. Hoy,
`Hillsborough, CA (US)
`
`(73) Assignee:
`
`Genentech, Inc., South San Francisco,
`CA (US)
`
`( * ) Notice:
`
`Subject to any disclaimer, the term of this
`patent is extended or adjusted under 35
`U.S.C. 154(b) by 0 days.
`
`This patent is subject to a terminal dis(cid:173)
`claimer.
`
`(21) Appl. No.: 09/723,545
`
`(22) Filed:
`
`Nov. 27, 2000
`
`Related U.S. Application Data
`
`(62) Division of application No. 09/553,924, filed on Apr. 21,
`2000.
`(60) Provisional application No. 60/131,076, filed on Apr. 26,
`1999.
`
`(51)
`
`Int. CI.7
`
`(52) U.S. CI
`
`C12P 21/04; C12P 21/06;
`C12N 1/20; C12N 5/06; C07K 1/00
`
`435/70.1; 435/69.1; 435/252.3;
`435/358; 530/395
`
`(58) Field of Search
`
`530/395; 435/69.1,
`435/69.6, 252.3, 358, 359, 70.1
`
`(56)
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`4,560,655 A
`4,657,866 A
`4,752,603 A
`4,766,075 A
`4,767,704 A
`4,927,762 A
`5,122,469 A
`5,705,364 A
`5,721,121 A
`
`12/1985
`4/1987
`6/1988
`8/1988
`8/1988
`5/1990
`6/1992
`1/1998
`2/1998
`
`Baker
`Kumar
`Collen et al.
`Goeddel et al.
`Cleveland et al.
`Darfler
`Mather et al.
`Etcheverry et al
`Etcheverry et al
`
`OTHER PUBLICATIONS
`
`Parekh et al. Cell-Type-Specific and Site-Specific N-Gly-
`cosylation of Type I and Type II Human Tissue Plasminogen
`Activator, Biochemistry, 1989, vol. 28, No. 19, pp
`7644-7662.*
`U.S. patent application Ser. No. 09/470,849, Etcheverry et
`al, filed Jun. 6, 1995.
`U.S. patent application Ser. No. 09/705,285, Etcheverry et
`al, filed Jan. 11, 2000.
`U.S. patent application Ser. No. 09/723,625, Andersen et al.,
`Nov. 27, 2000.
`Mather, Jennie P. Mammalian Cell Culture, the Use of
`Serum-Free Hormone-Supplemented Media, Plenum Press,
`New York pps. vii-x, 1-15, 17-52, 129-150 (1984).
`
`Rudinger, J., "Characteristics of the Amino Acids as Com(cid:173)
`ponents of a Peptide Hormone Sequence" Peptide Hor(cid:173)
`mones, J.A. Parsons, Baltimore:University Park Press pps. 6
`(1976).
`Wilhelm, J., et al., "Alterations in the Domain Structure of
`Tissue-Type Plasminogen Activator Change the Nature of
`Asparagine Glycosylation" Bio/Technology 8:321-324
`(Apr. 1990).
`Castro et al., "The macroheterogeneity of recombinant
`human interferon-gamma produced by Chinese-hamster
`ovary cells is affected by the protein and lipid content of the
`culture medium" Biotechnology and Applied Biochemistry
`(abstract only) 21(1):87-100 (1995).
`Castro, P. et al., "The macroheterogeneity of recombinant
`human interferon-gamme produced by the Chinese-hamster
`ovary cells is affected by the protein and lipid content of the
`culture medium" Biotechnology and Applied Biochemistry
`21(1):87-100 (1995).
`Gebert and Gray, "Expression of FSH in CHO Cells. II.
`Stimulation of hFSH expression levels by defined medium
`supplements." Cytotechnology (abtract only) 17:13-19
`(1995).
`Gebert, CA. and Gray, P.P., "Expression of FSH in CHO
`cells. II. Stimulation of hFSH expression levels by defined
`medium supplements" Cytotechnology 17:13-19 (1995).
`Hosoi et al., "Modulation of oligosaccharide structure of a
`pro-urokinase derivative (pro-UKAGSl) by changing cul(cid:173)
`ture conditions of a lymphoblastoid cell line Namalwa
`KJM-1 adapted to serum-free medium" Cytotechnology
`19:125-135 (1996).
`Kimura and Miller, "Glycosylation of CHO-derived recom(cid:173)
`binant tPA produced under elevated pCO-c" Biotechnology
`Progress (abstract only) 13(3):311-317 (1997).
`Kimura, R. and Miller, W., "Glycosylation of CHO-derived
`recombinant tPA produced under elevated pCO-2" Biotech(cid:173)
`nology Progess 13(3):311-317 (1997).
`Lamotte et al., "Na-butyrate increases the production and
`alpha2, 6-sialylation of recombinant
`interferon-gamma
`expressed by alpha2, 6-sialyltransferase engineered CHO
`cells" Cytotechnology (abstract only) 29:55-64 (1999).
`
`(List continued on next page.)
`
`Primary Examiner—Karen Cochrane Carlson
`Assistant Examiner—Rita Mitra
`(74) Attorney, Agent, or Firm—Janet E. Hasak
`
`(57)
`
`ABSTRACT
`
`A glycoprotein is produced by a process comprising cultur-
`ing mammalian host cells expressing nucleic acid encoding
`a glycoprotein in the presence of (a) a factor that modifies
`growth state in a cell culture, (b) a divalent metal cation that
`can adopt and prefers an octahedral coordination geometry,
`and/or (c) a plasma component. In this process, the occu(cid:173)
`pancy of an N-linked glycosylation site occupied only in a
`fraction of a glycoprotein is enhanced. Such culturing is
`preferably carried out at a temperature of between about 30°
`C. and 35° C. and/or in the presence of up to about 2 mM
`of a butyrate salt and/or in the presence of a cell-cycle
`inhibitor.
`
`1 Claim, 8 Drawing Sheets
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 3 of 24 PageID #: 547
`
`US 6,610,516 Bl
`Page 2
`
`OTHER PUBLICATIONS
`
`Lamotte, D. et al., "Na-butyrate increases the production
`and alpha2, 6-sialylation of recombinant interferon-gamma
`expressed by alpha2, 6-sialyltransferase engineered CHO
`cells" Cytotechnology 29:55-64 (1999).
`Madoiwa, S. et al., "Effect of carbohydrate side chain of
`tissue-type plasminogen activator on its interaction with
`plasminogen activator inhibitor-1" Fibrinolysis & Proteoly(cid:173)
`sis 12(1): 17-22 (1998).
`Nabi and Dennis, "The extent of polylactosamine glycosy-
`lation of MDCK LAMP-2 is determined by its Golgi
`residence time" Glycobiology 8(9):947-953 (1998).
`
`West and Brownstein, "EDTA treatment alters protein gly-
`cosylation in the cellular slime mold dictyostelium-discoi-
`deum" Experimental Cell Research
`(abstract only)
`175:26-36 (1988).
`West, CM. and Brownstein, S.A., "EDTA Treatment alters
`protein glycosylation in the cellular slime mold dictyostel-
`ium-discoideum" Experimental Cell Research 175:26-36
`(1988).
`Woodfork, K. et al., "Inhibition of ATP-sensitive potassium
`channels causes reversible cell-cycle arrest of human breast
`cancer cells in tissue culture" Journal of Cellular Physiology
`162(2):163-171 (1995).
`* cited by examiner
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 4 of 24 PageID #: 548
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 1 of 8
`
`US 6,610,516 Bl
`
`TYPE I
`
`TYPE
`
`v_
`
`"V
`FIG-1
`
`J
`
`T-flask
`
`Spinner
`
`400 L
`80 L
`VESSEL SIZE
`
`2000L
`
`12000L
`
`FIG.-2
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 5 of 24 PageID #: 549
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 2 of 8
`
`US 6,610,516 Bl
`
`I
`
`l
`
`I
`
`I
`
`I
`
`I
`
`I
`
`I
`
`I
`
`l
`
`l
`
`I
`
`I
`
`o o
`
`CM
`
`o
`in
`
`o £
`-°
`111
`
`^
`I
`
`It
`
`o
`
`CO"**
`
`I
`
`I
`
`I
`
`I
`
`|
`LO
`-si"
`
`I
`
`I
`
`I
`
`I
`
`|
`o
`
`o
`LO
`
`I
`
`I
`
`I
`
`I
`
`|
`
`I
`
`I
`
`I
`
`I
`
`CO
`
`TT
`
`LO
`
`|
`o
`CO
`
`LU
`CL
`
`>•
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 6 of 24 PageID #: 550
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 3 of 8
`
`US 6,610,516 Bl
`
`LAB SCALE EXPERIMENTS
`
`MINIFERMENTOR
`EXPERIMENTS
`n
`1
`1
`1
`
`r
`
`370C
`
`330C
`T-FLASK
`
`330C
`SPINNERS
`
`FIG.-4A
`
`RUN TIME
`
`FIG.-4B
`
`0.000
`
`0.375
`BUTYRATE (mM)
`
`0.750
`
`FIG.. 5
`
`1.500
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 7 of 24 PageID #: 551
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 4 of 8
`
`US 6,610,516 Bl
`
`LU
`CL
`>
`
`LU
`CL
`>
`
`45
`
`44
`
`43
`
`42
`
`41
`
`40
`
`39
`
`38
`
`37
`
`36
`
`35
`
`370C1-
`
`37°
`C11x-
`
`37°
`C12x
`
`330C1-
`
`33°
`Cllx
`
`33°
`C12x
`
`soci(cid:173)
`
`a lo
`Cllx
`
`31°
`C12x
`
`FIGS
`
`HI
`KSffi m
`s^
`Mmi mm
`sS
`..m
`^ S?«?xi
`2sSiJ ^ ' ij
`^§
`
`0#
`$Mv
`
`I'Si 5
`
`CONTROL
`
`2x BUTYR.
`
`330C/2x
`
`330C
`FIG.. 7
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 8 of 24 PageID #: 552
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 5 of 8
`
`US 6,610,516 Bl
`
`i - ^ 80
`
`CD
`
`o
`
`RUNTIME
`
`FIG.-8
`
`t-PASITE OCCUPANCY USING CELL CYCLE INHIBITORS
`
`67%G0/G1
`
`+
`
`54%G0/G1
`
`33%G0/G1
`,....,.. T. ... -. .
`
`THYMIDINE (G2/S)
`
`i—'
`
`CONTROL CASE
`
`FIG.-9
`
`r
`QUINIDINE (GO/G1)
`
`45
`
`40 -
`
`LU
`
`35 -
`
`30
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 9 of 24 PageID #: 553
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 6 of 8
`
`US 6,610,516 Bl
`
`'"i'.i
`
`'Vii
`
`^ #1
`
`•/•TO
`ax-
`
`'ffif.'itif.'iii,
`CONTROL
`
`45
`
`44
`
`4 3-
`
`4 2-
`
`°^ 41
`
`ff 40
`>
`
`"- 3 9-
`
`3 8-
`
`37
`
`36
`35
`
`45
`
`44 -
`
`43 -
`
`S^ 42
`
`LU
`CL
`>-
`H
`2
`
`41 -
`
`40 -
`
`39
`
`38 -
`
`3 7-
`
`36 -
`
`35
`
`10 nM
`MnCI2
`
`100 nM
`MnCI2
`
`1 |LIM
`MnCI2
`
`lOH-M
`MnCI2
`
`100 LlM
`MnCI2
`
`FIG- 10A
`
`CONTROL 10nM
`MnCI2
`
`100 nM
`MnCI2
`
`1 |UM
`MnCI2
`
`FIG.-10B
`
`10jxM
`MnCI2
`
`100 uM
`MnCI2
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 10 of 24 PageID #:
`554
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 7 of 8
`
`US 6,610,516 Bl
`
`LU
`>
`2
`
`LU
`CL
`
`43
`
`42-|
`
`41
`
`40
`
`39-]
`
`38
`
`37
`
`36
`
`35
`
`34
`
`33
`
`CONTROL
`
`10 |aM Fe
`FIG.. 11
`
`50 JIM Fe
`
`100 |LlM Fe
`
`Ml
`
`CONTROL URIDINE ADENOSINE
`FIG-12
`
`-GHT
`
`GUAN/MAN
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 11 of 24 PageID #:
`555
`
`U . S. P a t e nt
`
`Aug. 26,2003
`
`Sheet 8 of 8
`
`US 6,610,516 Bl
`
`45
`
`44
`
`43
`
`LU
`Q.
`>
`
`a.
`>
`
`CONTROL
`
`1 nM
`TRIIOD.
`
`100nM
`10nM
`1nM
`100nM
`10nM
`TRIIOD. TRIIOD. THYROX. THYROX. THYROX.
`
`FIG- 13A
`
`CONTROL
`
`1 nM
`TRIIOD.
`
`100 nM
`10 nM
`1 nM
`100 nM
`10 nM
`TRIIOD. TRIIOD. THYROX. THYROX. THYROX.
`
`FIG.. 13B
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 12 of 24 PageID #:
`556
`
`US 6,610,516 Bl
`
`CELL CULTURE PROCESS
`
`to
`This is a divisional application claiming priority
`application Ser. No. 09/553,924, filed Apr. 21, 2000, which
`claims priority to U.S. Provisional Application Serial No. 5
`60/131,076, filed Apr. 26, 1999, the entire disclosure of
`which is hereby incorporated by reference.
`
`BACKGROUND OF THE INVENTION
`
`^
`
`20
`
`1. Field of the Invention
`The present invention concerns a process for the produc(cid:173)
`tion of glycoproteins in mammalian cell culture. More
`specifically, the invention provides a process for producing
`glycoproteins in mammalian cells that results in enhanced
`occupancy of an N-linked glycosylation site occupied only
`in a fraction of a glycoprotein. A process for increasing the
`fraction of Type I tissue plasminogen activator (t-PA) in a
`mammalian cell culture is specifically disclosed.
`2. Description of Related Disclosures and Technology
`Glycoproteins
`Glycoproteins, many of which have been produced by
`techniques of recombinant DNA technology, are of great
`importance as diagnostic and therapeutic agents. In a
`eukaryotic cell environment, glycosylation is attached to a 25
`secreted or membrane-spanning protein by co- and post-
`translational modification. Proteins destined for the cell
`surface are first co-translationally translocated
`into
`the
`lumen of the endoplasmic reticulum (ER) mediated by a
`signal sequence at or near the amino terminus of the nascent 30
`chain. Inside the ER, the signal sequence is usually removed
`and a high-mannose core oligosaccharide unit is attached to
`the asparagine (N) residue(s) present as part of the sequence
`Asn-X-Ser/Thr, where X is any amino acid except, perhaps,
`proline.
`The efEciency of this co-translational glycosylation step is
`dependent on the presentation of an appropriate conforma(cid:173)
`tion of the peptide chain as it enters the endoplasmic
`reticulum (Imperiali and O'Connor, Pure & Applied Chem.,
`70: 33-40 (1998)). Potential N-linked glycosylation sites 40
`may no longer be accessible after the protein has folded
`(Kornfeld & Kornfeld, Ann Rev. Biochem.
`54:631-664
`(1985)). Proteins next move from the ER to the Golgi
`apparatus where further modifications, such as sulfation and
`processing of the high-mannose oligosaccharide chain to a 45
`complex-type oligosaccharide, occur and the proteins are
`directed to their proper destinations.
`N-linked oligosaccharides can have a profound impact on
`the pharmaceutical properties of glycoprotein therapeutics
`(e.g., in vivo half-life and bioactivity). Different bioprocess 50
`parameters (e.g., bioreactor type, pH, media composition,
`and ammonia) have been shown to affect protein glycosy(cid:173)
`lation significantly. Changes
`in terminal glycosylation
`(sialylation and galactosylation) and N-glycan branching are
`the most frequently observed alterations.
`The Carbohydrate Structure of Tissue Plasminogen Activa(cid:173)
`tor
`Tissue plasminogen activator (t-PA), a glycoprotein, is a
`multidomain serine protease whose physiological role is to
`convert plasminogen to plasmin, and thus to initiate or 60
`accelerate the process of fibrinolysis. Initial clinical interest
`in t-PA was raised because of its relatively high activity in
`the presence, as compared to the absence, of fibrin. Wild-
`type t-PA is a poor enzyme in the absence of fibrin, but the
`presence of fibrin strikingly enhances its ability to activate 65
`plasminogen. Recombinant human t-PA is used therapeuti(cid:173)
`cally as a fibrinolytic agent in the treatment of acute myo(cid:173)
`
`35
`
`55
`
`cardial infarction and pulmonary embolism, both conditions
`usually resulting from an obstruction of a blood vessel by a
`fibrin-containing
`thrombus.
`In addition to its striking fibrin specificity, t-PA exhibits
`several further distinguishing characteristics:
`(a) T-PA differs from most serine proteases in that the
`single-chain form of the molecule has appreciable
`enzymatic activity. Toward some small substrates, and
`toward plasminogen in the absence of fibrin, two-chain
`t-PA has greater activity than one-chain t-PA. In the
`presence of fibrin, however, the two forms of t-PA are
`equally active (Rijken et al., /. Biol. Chem., 257:
`2920-2925 (1982); Lijnen et a l, Thromb. Haemost.,
`64: 61-68 (1990); Bennett et a l, /. Biol. Chem.,266:
`5191-5201 (1991)). Most other serine proteases exist
`as zymogens and require proteolytic cleavage to a
`two-chain form to release full enzymatic activity.
`
`(b) The action of t-PA in vivo and in vitro can be inhibited
`by a serpin, PAI-1 (Vaughan et a l . , /. Clin. Invest., 84:
`586-591 (1989); Wiman et al., /. Biol. Chem., 259:
`3644-3647 (1984)).
`(c) T-PA binds to fibrin in vitro with a Krf in the ^M range.
`(d) T-PA has a rapid in vivo clearance that is mediated by
`one or more receptors in the liver (Nilsson et al.,
`Thromb. Res., 39: 511-521 (1985); Bugelski et a l,
`Throm. Res., 53: 287-303 (1989); Morton et al.,/. Biol.
`Chem., 264: 7228-7235 (1989)).
`A substantially pure form of t-PA was first produced from
`a natural source and tested for in vivo activity by Collen et
`a l, U.S. Pat. No. 4,752,603 issued Jun. 21, 1988 (see also
`Rijken et a l, J. Biol. Chem., 256: 7035 (1981)). Pennica et
`al. (Nature, 301: 214 (1983)) determined the DNAsequence
`oft-PA and deduced the amino acid sequence from this DNA
`sequence (U.S. Pat. No. 4,766,075 issued Aug. 23, 1988).
`Human wild-type t-PA has potential N-linked glycosyla(cid:173)
`tion sites at amino acid positions 117, 184, 218, and 448.
`Recombinant human t-PA (ACTIVASE® t-PA) produced by
`expression in CHO cells was reported to contain approxi(cid:173)
`mately 7% by weight of carbohydrate, consisting of a
`high-mannose oligosaccharide at position 117, and complex
`oligosaccharides at Asn-184 and Asn-448 (Vehar et al.,
`"Characterization Studies of Human Tissue Plasminogen
`Activator produced by Recombinant DNA Technology,"
`Cold Spring Harbor Symposia on Quantitative
`Biology,
`LI:551-562 (1986)).
`Position 218 has not been found to be glycosylated in
`native t-PA or recombinant wild-type t-PA. Sites 117 and
`448 appear always to be glycosylated, while site 184 is
`thought to be glycosylated only in a fraction of the mol(cid:173)
`ecules. The t-PA molecules that are glycosylated at position
`184 are termed Type I t-PA, and the molecules that are not
`glycosylated at position 184 are termed Type II t-PA. In
`melanoma-derived t-PA, the ratio of Type I to Type II t-PA
`is about 1:1. The most comprehensive analysis of the
`carbohydrate structures of CHO cell-derived human t-PA
`was carried out by Spellman et al., /. Biol. Chem., 264:
`14100-14111 (1989), who showed that at least 17 different
`Asn-linked carbohydrate structures could be detected on the
`protein. These ranged from the high-mannose structures at
`position 117
`to di-,
`tri-, and
`tetra-antennary
`N-acetyllactosamine-type structures at positions 184 and
`448. Type I and Type II t-PAs were reported to be
`N-glycosylated in an identical way at Asn-117 and Asn-448
`positions, when isolated from the same cell line. For further
`details, see also Parekh et al., Biochemistry, 28: 7644-7662
`(1989). The specific fibrinolytic activity of Type II t-PA has
`been shown to be about 50% greater than that of Type I t-PA
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 13 of 24 PageID #:
`557
`
`US 6,610,516 Bl
`
`(Einarsson et &[.,Biochim. Biophys. Acta,830: 1-10 (1985)).
`Further, increased Type I is correlated with increased half-
`life (Cole et al., Fibrinolysis, 7: 15-22 (1993)). However,
`Type II t-PA, which lacks a portion of carbohydrate associ(cid:173)
`ated with Type I t-PA, as well as desialated t-PA, demon- 5
`strated a longer TV4 beta than standard t-PA (Beebe and
`Aronson, Thromb. Res. 51: 11-22 (1988)).
`Analysis of the sequence of t-PA has identified the mol(cid:173)
`ecule as having five domains. Each domain has been defined
`with reference
`to homologous structural or functional 10
`regions in other proteins such as trypsin, chymotrypsin,
`plasminogen, prothrombin,
`fibronectin,
`and epidermal
`growth factor (EOF). These domains have been designated,
`starting at the N-terminus of the amino acid sequence of
`t-PA, as the finger (F) domain from amino acid 1 to about 15
`amino acid 44, the growth factor (G) domain from about
`amino acid 45 to about amino acid 91 (based on homology
`with EGF), the kringle-1 (Kl) domain from about amino
`acid 92 to about amino acid 173, the kringle-2 (K2) domain
`from about amino acid 180 to about amino acid 261, and the 20
`serine protease (P) domain from about amino acid 264 to the
`carboxyl terminus at amino acid 527. These domains are
`situated essentially adjacent to each other, and are connected
`by short "linker" regions. These linker regions bring the total
`number of amino acids of the mature polypeptide to 527, 25
`although three additional residues (Gly-Ala-Arg) are occa(cid:173)
`sionally found at the amino terminus. This additional trip-
`eptide is generally thought to be the result of incomplete
`precursor processing, and it is not known to impart func(cid:173)
`tionality. Native t-PA can be cleaved between position 275 30
`and position 276 (located in the serine protease domain) to
`generate the two-chain form of the molecule.
`Each domain contributes in a different way to the overall
`biologically significant properties of the t-PA molecule.
`Domain deletion studies show that the loss of the finger, 35
`growth factor, or kringle-2 domain results in a lower-afEnity
`binding of the variant t-PA to fibrin (van Zonneveld et al.,
`Proc. Natl. Acad. Sci. USA, 83: 4 6 7 0 ^ 6 74 (1986); Ver-
`heijen et a l, EMBO J., 5: 3525-3530 (1986)); however,
`more recent results obtained with substitution mutants indi- 40
`cate that the kringle-2 domain is less involved in fibrin
`binding than earlier expected (Bennett et al., supra). The
`domain deletion studies have implicated the finger and
`growth factor domains in clearance by the liver (Collen et
`al,Blood, 71: 216-219 (1988); Kalyan et al.,/. Biol. Chem., 45
`263: 3971-3978 (1988); Fu et a l, Thromb. Res., 50: 33-41
`(1988); Refino et a l, Fibrinolysis, 2: 30 (1988); Larsen et
`a l, Blood, 73: 1842-1850 (1989); Browne et a l, J. Biol.
`Chem., 263: 1599-1602 (1988)). The kringle-2 domain is
`responsible for binding
`to lysine. The serine protease 50
`domain is responsible for the enzymatic activity of t-PA and
`contains specific regions where mutations were shown to
`affect both fibrin binding and fibrin specificity (possibly
`direct interactions with fibrin), and other regions where only
`fibrin specificity is altered (possibly indirect interactions 55
`with fibrin) (Bennett et al., supra). Studies with mutants
`resulting from site-directed alterations indicate the involve(cid:173)
`ment of the glycosylation of t-PA in clearance (Lau et al.,
`Bio/Technology
`5: 953-958 (1987); Lau et al., Bio/
`Technology, 6: 734 (1988)).
`
`60
`
`the
`An unglycosylated variant of t-PA consisting of
`kringle-2 and protease domains was described to have a
`slower plasma clearance than wild-type t-PA (Martin et al.,
`Fibrinolysis,
`4:(Suppl.3):9 (Abstract 26) (1990)). The
`effects of altering oligosaccharide structures at sites 164 and 65
`448 of
`t-PA were also examined by Howard et al.,
`Glycobiology
`1: 4 1 1 ^ 18 (1991). Hotchkiss et al. (Thromb.
`
`Haemost., 60: 255-261 (1988)) selectively removed oli(cid:173)
`gosaccharide residues from the t-PA molecule, and demon(cid:173)
`strated that the removal of these residues decreased the rate
`of clearance of t-PA. These researchers, and Lau et al.
`((1987), supra, (1988), supra) also generated the t-PA variant
`N117Q (wherein asparagine at position 117 of wild-type
`human t-PA was substituted with glutamine) to prevent
`glycosylation at position 117. This variant, similarly to that
`obtained by enzymatic removal of the high-mannose oli(cid:173)
`gosaccharide at this position, exhibited an about two-fold
`slower clearance rate than wild-type human t-PA. See also
`EP-A 238,304 published Sep. 23, 1987 and EP-A 227,462
`published Jul. 1, 1987.
`Several reports have suggested that the carbohydrate
`moieties of t-PA influence the in vitro activity of this enzyme
`(Einarsson et al., supra; Opdenakker et al., Proc. Sci. Exp.
`Biol. Med., 182: 248-257 (1986)). T-PA is endocytosed by
`mannose receptors of liver endothelial cells and by galactose
`receptors of parenchymal cells. Indeed, the in vivo clearance
`of recombinant human t-PA produced in mammalian cell
`cultures was influenced by carbohydrate structures, particu(cid:173)
`larly by the high-mannose oligosaccharides (Hotchkiss et
`al., supra). A t-PA variant (designated TNK t-PA) that has a
`glycosylation site added at amino acid position 103, the
`native glycosylation site removed at amino acid position
`117, and the sequence at amino acid positions 296-299 of
`native human t-PA replaced by AAAA, has been shown to
`have increased circulatory half-life, and markedly better
`fibrin specificity than wild-type human t-PA (Keyt et al.,
`Proc. Natl. Acad. Sci. USA, 91: 3670-3674 (1994)).
`Glycoproteins Other than Native t-PA with More than One
`Glyoform
`Cells expressing tPA-6, a molecule composed of the
`kringle-2 and serine protease domains of t-PA, process it into
`two glycoforms, a monoglycosylated form with Asn-448
`occupied, and a diglycosylated form with Asn-448 and
`A s n l 84 occupied (Berg et al., Blood, 81: 1312-1322
`(1993)).
`Plasminogen exists in two glycoforms. The more glyco(cid:173)
`sylated form, commonly referred to as "plasminogen-1,"
`"plasminogen
`I," or "Type 1 plasminogen," has a
`galactosamine-based oligosaccharide attached at amino acid
`position 345 (Thr345) and a complex glycosamine-based
`oligosaccharide at amino acid position 288 (Asn288) of a
`native human plasminogen molecule. The less glycosylated
`form, commonly referred to as "plasminogen-2," "plasmi(cid:173)
`nogen II," or "Type 2 plasminogen," has a single oligosac(cid:173)
`charide chain attached at amino acid position 345 (Thr345)
`(Hayes and Castellino,/. Biol. Chem., 254(18): 8772-8776,
`8777-8780 (1979); Lijnen et a l, Eur. J. Biochem.,
`120:
`149-154 (1981); Takada et a l, Thrombosis Research, 39:
`289-296 (1985)).
`Other glycoproteins displaying variable site occupancy
`(variations
`in N- and O-glycosylation site-occupancy)
`include granulocyte-macrophage colony-stimulating factor
`(Okamoto et al., Archives of Biochemistry and Biophysics,
`286:562-568 (1991)), interferon-gamma
`(Curling et al.,
`Biochem. J., 212: 333-337 (1990)), protein C (Miletich and
`Broze, /. Biol. Chem., 265: 11397-11404 (1990)), and
`interleukin-2. Glycosylation of gamma-interferon was stable
`throughout an optimized culture design strategy using fed-
`batch cultures, with exposure to glucose starvation possibly
`leading to a dramatic change in glycosylation efficiency (Xie
`et a l, Biotechnol. Bioeng., 56: 577-582 (1997)).
`Different factors have been discussed to be potentially
`responsible for variable site-occupancy, including availabil(cid:173)
`ity of dolichol-phosphate and nucleotide sugars (Nyberg et
`
`

`

`Case 1:18-cv-01363-CFC Document 1-15 Filed 09/04/18 Page 14 of 24 PageID #:
`558
`
`US 6,610,516 Bl
`
`15
`
`35
`
`al., Biotechnol. Bioeng., 62: 336-347 (1999)), glycosyl-
`transferase activity (Hendrickson and
`Imperiali,
`Biochemistry, 34: 9444-9450 (1995); Kaufman et al.,
`Biochemistry, 33: 9813-9819 (1994)), and variable glyco-
`sylation site accessibility due to competition with protein 5
`folding (Hoist et al., The EMBO J., 15: 3538-3546 (1996);
`Imperiali, Ace. Chem. Res., 30: 452-459 (1997); Shelikoff et
`al, Biotechnol. Bioeng., 50: 73-90 (1996)). Any of these
`factors could be influenced by cell culture conditions. T-PA
`site-occupancy usually varies within a rather narrow range
`(±5%).
`Studies of Oligosaccharyltransferase/Metal-gations in Gly-
`cosylation
`Asparagine-linked glycosylation involves the enzyme-
`catalyzed modification of an asparagine side chain in a
`tetradeca-
`nascent polypeptide with a tri-antennary
`saccharide moiety. This first committed step in the biosyn(cid:173)
`thesis of N-linked glycoproteins
`is catalyzed by
`oligosaccharyltransferase, a heteromeric membrane-
`associated enzyme complex found in the lumen of the
`endoplasmic reticulum of eukaryotic cells. See Imperiali, 20
`supra; Allen et al, J. Biol. Chem., 270: 4797-4804 (1995);
`Sharma et al, Eur. J. Biochem., 116: 101-108 (1981);
`Silberstein and Gilmore, The FASEB Journal, 10: 849-858
`(1996); Kumar el al, Biochem. Mol. Biol. Intl., 36: 817-826
`(1995); Bause et al, Biochem. J., 312: 979-985 (1995); Xu 25
`and Coward, Biochemistry, 36: 14683-14689 (1997);
`Kumar et al., Biochem. Biophys. Res. Comm., 247: 524—529
`(1998); Watt et a l, Curr. Op. Struct. Biol, 7: 652-660
`(1997).
`For optimal activity, oligosaccharyltransferase requires a 30
`small amount of manganese divalent ion, but other divalent
`metal cations with an octahedral coordination geometry will
`support transfer, although at reduced rates (Hendrickson and
`Imperiali, supra; Kaufman et al., supra; Kumar et al., Bio(cid:173)
`chem. & Mol. Biol. International, 36: 817-826 (1995)).
`The Role of Temperature in Mammalian Cell Cultures
`To simulate normal body environment, fermentor tem(cid:173)
`perature in cultivating mammalian cells is controlled almost
`exclusively at 37° C. This dogma is so widely accepted that,
`so far, little attention has been paid to varying temperature 40
`in the cell culture process. The scarce literature data suggest
`that reduced fermentor temperature results in improved cell
`viability and shear resistance, higher cell density and titer in
`batch cultures, and a reduction in glucose/lactate metabo(cid:173)
`lism (Chuppa et al., Biotechnol. Bioeng., 55: 328-338 45
`(1997)).
`Specifically, Reuveny et al., /. Immunol. Methods, 86:
`53-59 (1986) studied the effect of temperatures in the range
`of 28° C. to 37° C. on batch hybridoma cell cultures. They
`found that although at lower temperatures the cell viability 50
`was improved, this was accompanied by a decrease in
`glucose uptake and a decrease in the specific antibody
`production. Therefore, in this particular case, lower tem(cid:173)
`peratures did not enhance the overall performance of the cell
`culture process.
`Sureshkumar and Mutharasan, Biotechnol. Bioeng., 37:
`292-295 (1991) investigated the effect of the temperature
`range of 29° C. to 42° C. on the cell culture process, and
`found that maximum cell density was achieved at 33° C. In
`contrast, the glucose uptake and specific lactate production 60
`rates were dramatically lower at 33° C. than at 39° C. These
`results showed that the optimal temperatures for growth and
`productivity may considerably differ. While the viability
`increase at temperatures below 37° C. appears to be a
`general phenomenon, the effect of temperature on specific 65
`productivity has been shown to be cell-line dependent
`(Chuppa et al., supra).
`
`55
`
`Weidemann et al., Cytotechnology, 15: 111-116 (1994)
`cultivated adherent recombinant baby hamster kidney
`(BHK) cells at temperatures between 30° and 37° C. The
`low-temperature cultivation in batch and repeated batch
`mode in a two-liter bioreactor showed a lower growth rate
`and a lower glucose consumption rate (i.e., less lactate
`production). On the other hand, the maximum cell density
`and productivity were not affected by the temperature reduc(cid:173)
`tion.
`Kretzmer et al., "Cultivation Temperature-Effect on Cell
`Culture Processes and Their Optimization"
`(American
`Chemical Society Meeting, San Francisco, Calif.), abstract
`138, presented Apr. 16, 1997, disclosed the effect of culti(cid:173)
`vation
`temperature on cell culture processes and their
`optimization, but apparently no specific glycosylation analy(cid:173)
`sis.
`It has been suggested that reduced fermentor temperatures
`might have other advantages related to product quality and
`integrity, but the effect of low temperatures on product
`quality, and in particular, on protein glycosylation, has been
`scarcely studied. Chuppa et al., supra, have reported that
`fermentation temperature did not have a significant effect on
`the sialic acid content of glycoproteins. Although the total
`sugar content was somewhat lower at 37° C. than at 34° C.
`or 35.5° C, the authors viewed this difference as "not
`substantial."
`However, U.S. Pat. No. 5,705,364 described preparing
`glycoproteins by mammnalian cell culture wherein the sialic
`acid content of the glycoprotein produced was controlled
`over a broad range of values by manipulating the cell culture
`environment, including the temperature. The host cell was
`cultured in a production phase of the culture by adding an
`alkanoic acid or salt thereof to the culture at a certain
`concentration range, maintaining the osmolality of the cul(cid:173)
`ture at about 250 to about 600 mOsm, and maintaining the
`temperature of the culture between about 30 and 35° C.
`Bahr-Davidson, "Factors Affecting Glycosylation Site
`Occupancy of ASN- 184 of Tissue-Type Plasminogen Acti(cid:173)
`vator Produced in Chinese Hamster Ovary Cells," A Dis(cid:173)
`sertation submitted to the Department of Chemical Engi(cid:173)
`neering and the Committee of Graduate Studies of Stanford
`University in Partial Fulfillment of the Requirements for the
`Degree of Doctor of Philosophy, May 1995, investigated the
`effects of temperature on glycosylation site occupancy and
`reported that site occupancy was increased by exposing cells
`to 26° C. (see pages 50-51).
`Hormonal and Other Treatments to Influence Glycosylation
`The effect of various additives such as components of
`plasma to the culture media on protein production and
`glycosylation has been studied in the literature, for example,
`the effects of hormonal treatments on membrane glycosy(cid:173)
`lation in rat kidney brush broder membranes (Mittal et al.,
`Indian J. Exper. Biol, 34: 782-785 (1996)). Studies of
`Muc-1 mucin expression established the hormonal basis for
`MRNA expression (Parry et a l, /. Cell Scl, 101: 191-199
`(1992)). Thyroid hormone regulation of alpha-lactalbumin
`with differential glycosylation has been reported (Ziska et
`a l, Endocrinology,
`123: 2242-2248 (1988)). The cellular
`response to protein N-glycosylation was increased in the
`presence of thyroxine, insulin, and thrombin, and the effect
`was dose-dependent (Oliveira and Banerjee, /. Cell.
`Physiol, 144: 467-472 (1990)). Thyroxine was found to
`induce changes in the glycosylation pattern of rat alpha-
`fetoprotein (Naval et al., Int. J. Biochem., 18: 115-122
`(1986)).
`In addition to hormonal treatments, glutatb one a