200307047 玖、發明說明 【發明所屬之技術領域】 本發明係關於新穎的能夠產生蛋白質和能夠生長在無 榖氨醯胺培養基之榖氨醯胺營養缺陷的人類細胞。此外， 其係關於新穎的產生蛋白質的方法和關於在榖氨醯胺營養 缺陷之人類細胞中把榖氨醯胺合成酶（GS)當作可選擇標記 的用途。 【先前技術】 由哺乳動物細胞培養所產生的蛋白質係用來提供治療 和診斷應用的蛋白質。現今，晡乳動物細胞培養是用於人 類和動物醫藥之許多重要蛋白質的較佳來源，尤其那些相 當大的、複雜的和糖基化的（Ν· B. Finter等人，在大規模 哺乳動物細胞培養技術，1990， ed. A· S. Lubiniecki， Marcel Dekker， Inc·，紐約）0 例如，藉細胞培養產生人類蛋白質紅血球生成素（EPO) 已在WO 93/09222中描述。使用以可編碼人類EPO外生基 因轉染的人類纖維組織母細胞，人類EPO已獲得可觀的比 生成速率。在進一步產生人類EPO的方法（WO 94/12650)， 係使用以能夠活化內生編碼的EPO之DNA序列轉染的人 類纖維肉瘤細胞系HT1080。類似的方法已於WO 99/09268 中描述。永生的（Immortalised)人類細胞如Namalwa、Hela S3及HT 1080細胞係以能夠活化內生編碼的EPO之DNA 序列轉染。 200307047 描述於 WO 93/09222、WO 94/12650 及 WO 99/09268 用來產生人類EPO的細胞係都在含有榖氨醯胺的培養基中 培養。這是不利的，因爲當榖氨醯胺被培養的細胞當作能 量基質時，會產生氨這種代謝產物，其具有細胞毒性及會 抑制細胞生長。再者，因爲它在細胞的高基氏體內對pH的 影響，而會抑制蛋白質糖基化。 產生大量組織血漿素原活化劑（一種糖基化蛋白質）已在 WO 87/〇4462中描述。Gs在其中已被用來作爲放大系統， 在榖氨醯胺原質營養的（pro to trophic)中國倉鼠卵巢（CHO)細 胞用於共放大可編碼組織血漿素原活化劑（tPA)的基因，其 係藉由以可編碼GS的基因轉染細胞。然而，如同在EP-A 148 605發現的，用CHO細胞來產生糖基化的人類蛋白質 係不利的α由cH0細胞合成的蛋白質其平均碳水化合物成 分會與自然生成的糖基化人類蛋白質有所不同，這是因爲 人類細胞含有α2·3唾液酸轉移酶（sialyltransferase)及α2·6 唾液酸轉移酶酵素。CHO細胞只具有α2.3唾液酸轉移酶， 所以不能產生α2.6端唾液酸與寡糖部分（moieties)的鍵結， CHO細胞缺乏硫化（suiphation)碳水化合物結構的酵素， CHO細胞雖然有α 1 - 6巖藻糖轉移酶（fucosyltransferase)(附 著中心的巖藻糖殘基），亦缺乏al-3巖藻糖轉移酶（附著末 端的巖藻糖殘基）。人類細胞則具有兩者巖藻糖轉移酶 (Cumming D.A·，199卜 Glycobiology Vo. 1，No. 2，115 -130 ’ Jenkins Ν·及 Curling Ε·Μ·Α.，1994，enzyme and Micorbial Technology Vol. 16，354-364 ; Lee 等人，1989 200307047 ，Journal of Biological Chemistry，Vol. 264，13848-13855) 。所以，由CHO細胞合成之糖基化的蛋白質不會有希望 得到的特徵，例如如於人類細胞產生的活體內生物活性。 在WO 89/10404中，報告了製造骨髓瘤細胞例如鼠融 合瘤、鼠漿細胞瘤細胞及大鼠融合瘤細胞榖氨醯胺獨立的 方法，其藉由以GS轉化他們。其進一步證明GS可以用來 共放大可編碼免疫球蛋白分子輕及重鏈的基因及用來共放 大在骨髓瘤細胞系中可編碼纖維蛋白分解酵素的基因。然 而，齧齒類細胞系具有缺點，即以N-乙醇醯基神經氨酸殘 基的附著代替N-乙醯神經氨酸、無法進行硫化及αΐ.3半乳 糖轉移酶酵素的存在。在齧齒類細胞中合成之糖蛋白的寡 糖結構因此可以在人類產生免疫性。 【發明內容】 本發明之目的係提供一種改進的方法，其不具有用於 產生蛋白質時的上述缺點（尤其當產生糖蛋白時），及可獲得 高蛋白質效價。 利用根據申請專利範圍第1項之新穎的榖氨醯胺營養 缺陷人類細胞，及根據申請專利範圍第7項之新穎的方法 ，可以達成此目的。 根據本發明，可得到榖氨醯胺營養缺陷之人類細胞’ 其係以（第一）可編碼蛋白質之外生DNA序列或能夠改變可 編碼蛋白質之內生基因表現的外生DNA序列轉染’及進一 步以（第二）可編碼榖氨醯胺合成酶（GS)(較佳的爲哺乳動物 200307047 GS)之外生DNA序列轉染，其中這些外生的DNA序列係位 於一或一個以上DNA構築體，該轉染的細胞能夠產生該蛋 白質及能夠生長在無榖氨醯胺的培養基上。 DNA序列 ，，可編碼蛋白質的外生DNA序列，，或，，能夠改變可編碼 蛋白質之內生基因表現的外生DNA序列，，及π可編碼GS的 外生DNA序列，，通常位於DNA構築體上，例如表現載體或 感染性載體。質體可以用來作爲表現載體。作爲感染性載 體則可以使用如反轉錄病毒、泡疹病毒、腺病毒、腺病毒 伴隨病毒、流行性腮腺炎病毒及脊髓灰質炎病毒載體。較 佳的爲表現載體，尤其使用質體。 ’’可編碼蛋白質的外生DNA序列”可以包括額外序列， 例如調控序列如啓動子及/或增強子、多聚腺苷酸區域 (polyadenylation site)及剪接接頭（splice junctions)，通常用 於表現外生基因或可以包括額外一或多個分開的靶向性 (targeting)序列及視需要的可編碼可選擇的標記之DNA(如 WO 93/09222中所描述）。 ”能夠改變可編碼蛋白質之內生基因表現的外生DNA 序列’’可以包括外生的DNA序列，其不編碼蛋白質的基因 產物，但可編碼部分該基因產物（如外顯子），及可以包括額 外序列，例如調控序列及剪接接頭，通常用於表現外生 DN A序列。他們可以進一步包括靶向性序列及視需要的可 編碼可選擇的標記之DNA(如W0 93/09222中所描述）。 通常’ ”能夠改變可編碼蛋白質之內生基因表現的外生 200307047 DNA序列”係在轉染進細胞後插入細胞的染色體DNA。同 源重組或靶向性係在此用來取代或使與含有調控序列之內 生基因相關的調控區域失去作用。作爲調控序列可以使用 如啓動子及/或增強子，其造成基因表現量高於相對應之無 轉染的細胞（如WO 93/09222所描述），合適的啓動子可以 是可調控的或持續表現的啓動子。合適的啓動子可以是強 啓動子，其係取決於所使用的細胞系，如人類巨細胞病毒 主要立即早期啓重力子（major immediate early promoter， hCMV-MIE)、SV 40早期及晚期啓動子、其他腺病毒啓動 子、任何多瘤病毒或乳多泡病毒早期及晚期啓動子、干擾 素αΐ啓動子、鼠金屬硫因（metallothionein)啓動子、Rous 氏肉瘤病毒長端重複啓動子、β-珠蛋白啓動子、伴白蛋白 啓動子、卵白蛋白啓動子、鼠β-珠蛋白啓動子及人類β-珠 蛋白啓動子。 根據本發明之”可編碼GS之外生DNA序列”可以受強 啓動子及弱啓動子的控制，如果外生的DNA序列係單單需 要表現可編碼GS的基因，則使用強啓動子，當外生的 DNA序列係作爲可選擇的標記，以及若GS係用於放大作 用時，則使用弱啓動子。合適的啓動子可以是可調控的或 持續表現的啓動子。啓動子可以選擇，例如，GS係在足夠 使轉染的細胞生長的濃度中表現，但細胞培養時並不產生 高量榖氨醯胺代謝產物氨，通常不大於4 mM，較佳的不大 於2 mM，更佳的少於2 mM氨。 ”可選擇的標記’’提供了可選擇的表現型，使鑑別及分 200307047 離接受（recipient)細胞成爲可能。GS可以在本發明用來當 作可選擇的標記，以篩選出成功轉染的榖氨醯胺營養缺陷 人類細胞，其倂入和表現可編碼GS的外生DNA序列。 根據所用的細胞系，強啓動子可以是如hCMV-MIE、 SV 40早期及晚期啓動子、其他腺病毒啓動子、任何多瘤 病毒或乳多泡病毒早期及晚期啓動子、干擾素αΐ啓動子、 鼠金屬硫因（metallothionein)啓動子、Rous肉瘤病毒長端重 複啓動子、β-珠蛋白啓動子、伴白蛋白啓動子、卵白蛋白 啓動子、鼠β-珠蛋白啓動子及人類β-珠蛋白啓動子。 根據所用的細胞系，弱啓動子可以是如鼠白血病病毒 長端重複、單純疱疹病毒胸苷激酶及鼠乳房腫瘤病毒長端 重複，較佳的，可編碼GS的基因係在強啓動子控制下， 更佳的係在hCMV-MIE啓動子控制下。一項可能的具體實 例，一種來自倉鼠之可放大的（amplifiable)哺乳動物GS序 列及其在哺乳動物細胞中作爲可選擇的標記的用途已爲該 項技術所熟知，且如WO 87/04462、WO 91/06657及WO 89/01036所描述；本發明實施例使用此倉鼠GS表現單元 及個別的篩選方法如參考文獻所提及。 π可編碼蛋白質之外生DNA序列”或”能夠改變可編碼 蛋白質之內生基因表現的外生DNA序列”及”可編碼GS之 外生DNA序列”係位於一或多於一個DNA構築體上，較佳 的，這些外生的DNA序列係位於多於一個，更佳的位於兩 個DNA構築體上。若這些外生的DNA序列位於一個 DNA構築體上，他們可能功能上相組合，例如他們的表現 11 200307047 會受同一個調控序列，如啓動子及/或增強子（如wo 89/10404所描述）驅策。 人類細胞 榖氨醯胺營養缺陷人類細胞係指所有不表現GS或表現 GS很差的人類細胞，所以能夠生長在含有榖氨醯胺的培養 基，但在無榖氨醯胺培養基不能生長或生長很差。用於本 發明之榖氨醯胺營養缺陷人類細胞係必死的榖氨醯胺營養 缺陷人類細胞或永生的榖氨醯胺營養缺陷人類細胞。必死 的榖氨醯胺營養缺陷人類細胞係在培養基中具有有限壽命 的榖氨醯胺營養缺陷人類細胞，永生的（又稱永久的或建立 的）榖氨醯胺營養缺陷人類細胞係在培養基中如習於該項技 術者所熟知以適當地繼代培養和次培養，具有明顯無限壽 命之榖氨醯胺營養缺陷細胞。 必死的榖氨醯胺營養缺陷人類細胞的例子可以是人類 纖維組織母細胞及人類胎肺組織細胞，永生的榖氨醯胺營 養缺陷人類細胞的例子可以是人類纖維肉瘤細胞，如 HT1080 細胞系（如 DSMZ No· ACC-315 或 ATCC No. CCL 121)及B-淋巴母細胞人類細胞如HL60 (DSMZ No. Acc-3) 或Namalwa (DSMZ Acc-24)細胞系。用於本發明較佳的係 永生的榖氨醯胺營養缺陷人類細胞，更佳的，此永生的榖 氨醯胺營養缺陷人類細胞係用B-淋巴母細胞或纖維肉瘤細 胞，更佳的係用人類纖維肉瘤細胞，最佳的係用HT1080細 胞系（如 ATCC No. CCL 121)。 轉染 12 200307047 榖氨醯胺營養缺陷人類細胞可藉由已知的基因工程技 術以外生的DNA序列轉染。 以外生的DNA序列轉染取決於序列是否位於一或多於 一個DNA構築體上，若序列位於多於一個DNA構築體上 ，轉染可隨各序列分別發生或共轉染。當轉染隨各序列分 別發生時，序列轉染順序通常是隨意的，隨各序列分別發 生的轉染較佳先以”可編碼該蛋白質的外生DNA序列”或” 能夠改變可編碼該蛋白質之內生基因表現的外生DNA序列 ’’轉染，然後再以”可編碼GS的外生DNA序列”轉染，轉染 的榖氨醯胺營養缺陷細胞可以在各別轉染之後培養並評估 蛋白質生成。 爲了成功篩選轉染的細胞，這些係生長在無榖氨醯胺 的培養基。細胞可以直接生長在無榖氨醯胺培養基或先在 含有榖氨醯胺的培養基，然後逐步稀釋成無榖氨醯胺的培 養基，如先以榖氨醯胺濃度10 mM，然後以2 mM至0 mM 逐步稀釋。合適的篩選方法可以根據所用的細胞系來選擇 。如前述顯知，本發明能夠產生蛋白質及能夠生長在無榖 氨醯胺培養基的榖氨醯胺營養缺陷人類細胞係得自以可編 碼該蛋白質的外生DNA序列或能夠改變可編碼蛋白質之內 生基因表現的外生DNA序列及可編碼榖氨醯胺合成酶之外 生DNA序列轉染該細胞，其中這些外生的DNA序列係位 於一或多於一個DNA構築體上。 放大 可編碼蛋白質的外生DNA序列或能夠改變可編碼蛋白 13 200307047 質之內生基因表現的外生DNA序列可以在轉染後根據如 WO 94/12650所描述之基因放大已知方法來放大。可放大 的基因可編碼酵素，如DHFR(二氫葉酸還原酶）、GS、腺苷 脫氨酶、天冬醯胺合成酶、天門冬胺酸氨甲醯基轉移酶、 金屬硫因-1、鳥氨酸脫羧酶、P-糖蛋白質、核苷酸還原酶 、胸苷激酶或黃嘌呤鳥嘌呤磷酸核糖轉移酶可以用於此目 的。含有這些基因的放大複製之細胞係例如能夠在沒有酵 素代謝產物的培養基或在含有相關的選擇劑之培養基中存 活，相關的選擇劑係例如在DHFR情況下爲甲氨喋呤 (MTX)及在 GS 情況下爲 methionine sulphoximine (MSX)。 蛋白質 由本發明轉染的榖氨醯胺營養缺陷人類細胞所產生之 蛋白質係無糖基化和糖基化蛋白質，糖基化蛋白質係指具 有至少一個寡糖鏈的蛋白質。 無糖基化蛋白質的例子如無糖基化荷爾蒙，像黃體生 成素釋放激素、甲狀腺素釋放激素、胰島素、生長抑素、 催乳素、促腎上腺皮質激素、黑素細胞促黑激素、升壓素 、以及其衍生物如去氨加壓素、催產素、降鈣素、副甲狀 腺素（PTH)或其片段（如PTH (1-43))、促胃液素、胰泌素、 腸促胰酶素、激膽囊素、血管緊張素、人類胎盤促黃體激 素、人類纖毛膜促性腺激素（HCG)、caerulein及胃動素； 無糖基化的止痛劑物質如腦啡肽及其衍生物（參見US-A 4 277 394 及 EP-A 03 1567)、內啡肽、daynorphin 及 kyotorphin ;無糖基化酵素如無糖基化神經傳導物質例如蛙 14 200307047 皮素、神經緊張素、緩激肽和物質P ;神經生長因子（NGF) 家族、上皮生長因子（EGF)及纖維組織母細胞生長因子 (FGF)家族之無糖基化生長因子及無糖基化荷爾蒙及生長因 Τν?, Μΰ 子受體。 糖基化蛋白質的例子係荷爾蒙及荷爾蒙釋放因子，如 生長荷爾蒙，包括人類生長荷爾蒙、牛生長荷爾蒙、生長 荷爾蒙釋放因子、副甲狀腺素、甲狀腺刺激素、ΕΡΟ、脂 蛋白質、α -1-抗胰蛋白酶、卵泡刺激素、降鈣素、黃體生 成素、胰增血糖素、凝血因子如因子VIIIC、因子IX、組 織因子及von Willebrand氏因子、抗凝血因子例如蛋白質 C、心房利鈉因子、肺表面活性物質、血漿素原活化劑例如 尿激酶或人類尿或組織型血漿素原活化劑（t-PA)、凝血酶、 造血生長因子、腦啡肽、RANTES(調控活化自然的T-細胞 表現及分泌）、人類巨噬細胞發炎蛋白質（MIP-1-α:)、血淸 蛋白素例如人類血淸蛋白素、mulleriati氏抑制物質、鬆弛 素A鏈、鬆弛素B鏈、前鬆弛素、鼠促性腺激素相關肽、 微生物蛋白質例如/5 -lactanase、DNase、抑制素、激活素 、腎素、血管內皮生長因子（VEGF)、荷爾蒙或生長因子受 體、整合素、蛋白質A或D、類風濕性因子、神經營養因 子例如源自骨神經營養因子（BDNF)、神經營養素 (neurotrophin)-3、-4、-5 或-6 (NT-3、NT-4、NT-5 或 NT_ 6)、或神經生長因子例如NGF-β、源自血小板生長因子 (PDGF)、纖維組織母細胞生長因子例如FGF及bFGF、表 皮生長因子（EGF)、轉化生長因子（TGF)例如TGF- α及 15 200307047 TGF-Θ，包括 TGF-βΙ、TGF』2、TGF-p3、TGF-P4 或 TGF-p5、類胰島素生長因子-I及-π (IGF-I及IGF-II)、 des(l-3)-IGF-I (腦IGF-I)、類胰島素生長因子結合蛋白、 CD蛋白（一群分化蛋白質）例如CD-3、CD-4、CD-8及CD- 19、 骨誘導因子 、免疫毒素 、骨形態發生蛋白質 （BMP) 、 細 胞活素及其受體，還有嵌合（chimeric)蛋白質，包含其受體 的細胞活素，包括例如腫瘤壞死因子α及/3、他們的受體 (TNFR-1， ΕΡ 417 563 及 TNFR-2， ΕΡ 41 7 014)及其衍生 物，一種干擾素例如干擾素-α、- /5、及-7，細胞族群刺 激因子（CSFs)，如 M-CSF、GM-CSF 及 G-CSF，白介素 (ILs)如IL-1到IL-10，超氧化物歧化 ，T-細胞受體、表 面膜蛋白、降解加速因子、病毒抗原例如AIDS夾膜的一部 份、運輸蛋白、導弓丨（homing)受體、addressins、調控蛋白 質、抗體、嵌合蛋白如免疫黏連素，及任何上述所列片段 之糖基化蛋白質。較佳的，本發明產生糖基化蛋白質，更 佳的本發明產生N-糖基化蛋白質，最佳的糖基化荷爾蒙像 EPO爲N-糖基化及其生物活性係依賴其上，或尤其產生 EPO。 細胞培養 任何於該項技術已知之合適的培養程序及培養設備均 可用來生長本發明轉染的人類細胞。添加約0.1至20%(較 佳0.5至15 %)血淸之常見無榖氨醯胺基礎培養基以及無血 淸無榖氨醯胺常見基礎培養基都可作爲培養基用，此外， 也可用不含動物來源蛋白質的常見無榖氨醯胺基礎培養基 16 200307047 ，較佳的使用無血淸無榖氨醯胺常見基礎培養基。 可以用的血淸如胎牛血淸或成牛血淸，較佳係使用胎 牛血淸。常見之可用的無榖氨醯胺基礎培養基爲，例如無 榖氨醯胺Eagle氏最低基本培養基（MEM)培養基、無榖氨醯 胺Dulbecco氏修飾的Eagle氏培養基（DMEM)、無榖氨醯 胺 Iscove 氏 DMEM 培養基（N. Iscove 及 F. Melchers， Journal of Experimental Methods，1978，147，923)、無榖 氨醯胺 Ham 氏 F12 培養基（R.G. Ham，Proceedings of National Academy of Science，1965，53，288)、無榖氨醯 胺 L-15 培養基 (A. Leibovitz，American Journal of Hygiene，1963，78， 173)、無榖氨醯胺 RPMI 1640 培養 基（G.E. Morre 等人，The Journal of the American Medical Association，1967，199，519)、無榖氨醯胺專屬培養基和 其合適比例之混合物。用於高密度細胞培養之強化 (Fortification)常見細胞培養生長培養基已在該項技術所熟 知，且已描述於如GB 2251249，其也非常適合作爲本發明 之無榖氨醯胺培養基。 常見的添加物可以加到常見的無榖氨醯胺基礎培養基 。常加入的添加物包括血淸中的蛋白質以及視需要可以對 細胞生長及/或細胞生存有正面影響之進一步成分。血淸中 的蛋白質如牛血淸白蛋白（BSA)、運鐵蛋白及/或胰島素； 可以對細胞生長及/或細胞生存有正面影響之進一步成分如 大豆脂、硒及乙醇胺。取代榖氨醯胺及/或核苷的胺基酸可 以依所用的細胞系加到培養基中，胺基酸的例子如異白胺 17 200307047 酸、白胺酸、纈胺酸、離胺酸、天冬醯胺、天門冬胺酸、 榖胺酸、絲胺酸、丙胺酸。視需要，榖氨醯胺可以低濃度 加到常見的無榖氨醯胺基礎培養基中，通常少於1 mg/1， 較佳的少於〇·5 mg/1，以支持其生物合成作用（如轉胺反應) 〇 若可編碼蛋白質的外生DNA序列或能夠改變可編碼蛋 白質之內生基因表現的外生DNA序列在轉染後以可放大的 (amplifiable)基因放大，相關的選擇劑可以加到常見的無 榖氨醯胺基礎培養基中，選擇劑施用的濃度範圍依所用的 細胞系而定，通常用10 μΜ及更高的濃度。 懸浮生長於無血淸培養基之適應作用 榖氨醯胺營養缺陷人類細胞可以是固著依賴或固著獨 立’其可以作爲起始材料來獲得根據本發明之轉染的榖氨 醯胺營養缺陷人類細胞，根據本發明，該轉染的細胞能夠 產生蛋白質及能夠生長在無榖氨醯胺培養基。若使用固著 依賴人類細胞，如ΗΤ1080細胞系（ATCC No. CCL 121)， 其可以適應成爲固著獨立HT1080細胞系，能夠懸浮生長於 無血淸培養基，尙未於文獻中描述。 適應作用可以發生在以可編碼蛋白質的外生DNA序列 或能夠改變可編碼蛋白質之內生基因表現的外生DNA序列 及可編碼GS的外生DNA序列轉染之前或之後。較佳的， 細胞先以可編碼蛋白質的外生DNA序列或能夠改變可編碼 蛋白質之內生基因表現的外生DNA序列轉染，再使其適應 懸浮生長於無血淸培養基，然後進一步以可編碼GS之外生 18 200307047 DNA序列轉染。若需要，轉染的細胞可以再使其適應懸浮 生長於無血淸無榖氨醯胺的培養基。 本發明之轉染的榖氨醯胺營養缺陷人類細胞可以是固 著依賴或固著獨立，並能夠懸浮生長於無血淸無榖氨醯胺 的培養基。較佳的轉染的榖氨醯胺營養缺陷人類細胞係固 著獨立，且能夠懸浮生長在無血淸無榖氨醯胺的培養基。 適應作用使其成爲能夠懸浮生長於無血淸培養基之固 著獨立細胞，可以藉由在第一步用含有血淸的培養基適應 細胞而達成。這可以經由如胰蛋白酶處理細胞，然後接著 搖動或藉搖動釋放細胞來達成。細胞然後在第二步藉接著 減少血淸含量來適應無血淸培養基，在適應時，若使用選 擇劑，可以減少其用量，以避免抑制細胞生長。然而，細 胞也可以在第一步藉由接著減少血淸含量適應生長於無血 淸培養基，第二步適應成爲能夠懸浮生長的固著獨立細胞 ，藉由如胰蛋白酶處理細胞然後接著搖動或藉搖動釋放細 胞來達成，兩個步驟也可以同時施行。 較佳的，細胞係在第一步含血淸培養基適應成爲固著 獨立細胞用，藉由搖動釋放細胞然後第二步經由接著減少 血淸含量來適應於無血淸培養基。 如上所描述的培養基可以做爲含血淸培養基的基本。 在此所定義之選擇劑可以加到培養基中’選擇劑施用的濃 度範圍取決於所用的細胞系，通常使用ίο μΜ及更高濃度 。含有血淸的培養基通常添加約0.1到20 %，較佳的〇·2 到10 %，最佳的〇·5到5 %血淸，可以使用的血淸如上所 19 200307047 提及。接著減少血淸含量可以藉由逐步減少血淸含量，例 如從10 %到1 %到〇 %來達成。 產生蛋白質的方法 本發明轉染的榖氨醯胺營養缺陷人類細胞係用在產生 蛋白質的方法中，藉由在適合用來表現該蛋白質及取得 (recovering)該蛋白質的條件下之培養基培養該細胞，產生 的蛋白質係描述如上。如上所描述之常見的無榖氨醯胺基 礎培養基及常見的添加物可以作爲培養基，合適的培養條 件係如WO 96/39488所描述之習知的用於哺乳動物細胞體 外培養者。 蛋白質取得 蛋白質可以從細胞培養物中以習知的分離技術分離， 例如免疫親和或離子交換管柱分劃、沉澱、逆相HPLC、色 層分析、色層焦集法、SDS-PAGE、膠體過濾。習於該項技 術者應了解，適合用於有興趣的多肽之純化方法，可能需 要根據多肽在重組細胞培養中表現的特性來做修飾。 【實施方式】 實施例1 準備人類纖維肉瘤細胞系HT1080_R223 含有多人類EPO基因複製（copies)的固著依賴人類 HT1080-R223細胞系是產業上用來生產EPO的細胞，一開 始係由Transkaryotic Therapies公司所創造（劍橋，MA 02139 (US))，其係源自固著依賴人類纖維肉瘤HT 1080細 200307047 胞系。親本HT1080細胞系（ATTC No. CCL 121)具有產生 EPO的能力，其係藉由以類似DNA構築體PREP018(描述 於WO 95/3 1 560)的DNA構築體pREP022轉染，除了 DHFR基因係在相反方向以及pREP022含有比PREP018少 約600鹼基對同源序列。此細胞系接下來將以R223細胞系 提及。 實施例2 R223細胞系的適應作用以能夠懸浮生長於無血淸培養基中 藉由胰蛋白酶處理釋放在靜置瓶中以附著培養生長的 細胞，懸浮於專屬（proprietory)含有榖氛醯胺之無血淸培 養基（進一步以“HM9”提及），添加1〇 %透析的胎牛血淸 (dFBS)及500 nM MTX，然後以和震盪瓶培養。生長在六天 後開始，細胞次培養到同樣的培養基中（圖1)。一旦可信的 生長模式建立了，就減少培養基的血淸含量至1 °/◦’再一 次，可信的生長可在添加的血淸完全用盡之前被再建立。 使添加血淸和無血淸的培養物過度生長以評估產量’ 結果顯示於表1，並附有附著培養的數據。足夠的適應作 用之後，懸浮培養的比生長速率（即使沒有血淸）與添加血淸 之附著培養一樣。 在添加血淸的培養基中從附著培養到懸浮培養時的適 應作用，EPO合成比速率下降50 %，從24到12 EU/106 細胞/小時，然而，此速率在適應無血淸生長之後，大體上 會恢復至18 EU/106細胞/小時。 21 200307047 表1 在適應無血淸培養基之前或之後，R2；23細胞系在附著培養 及懸浮培養的召 長及產量。 最大細胞 數/mL χ ΙΟ'6 比生長速 率h·1 累積細胞 h/ml χ 10·6 EPO EU/ml q epo EU/106 細 胞/h 附著培養 (DMEM + 10 % dFBS 500 nM MTX) 不適用 0.0157 不適用 不適用 24 懸浮培養（HM9 + 10 % dFBS 500 nM MTX) 1.4 0.0185 334 4199 12 懸浮培養（HM9 ( 無血淸）500 nM MTX) 1.1 0.0166 210 3600 18 實施例3 HT1080 (ATCC CCL121)的適應作用以能夠懸浮生長在無血 淸培養基中 HT1080 細胞系 ATCC CCL121 是從 American Type culture Collection (Rockville， Maryland，US A)取得，一 開始細胞係以附著培養生長於含有10 %胎牛血淸（FBS)的 DMEM 中。 懸浮和無血淸的適應作用可以根據圖2的程序進行。 爲了開始懸浮培養，細胞係藉由胰蛋白酶處理從附著培養 中釋放，而釋放的細胞再懸浮於加了 2 % FBS的HM9，這 22 200307047 些接著以震盪培養，一旦細胞在加了 2 % FBS的HM9的 懸浮生長建立了，以相同的培養基稀釋培養物作子代培養 ，可信的HT1080細胞系懸浮生長在30天後建立（圖3)。 在那之後培養基血淸含量減少，最後完全不見。細胞持續 在無血淸中繼續次培養生長。 實施例4 氨對R223細胞系生長及產量的影響 當以榖氨醯胺作爲能量基質時，氨是培養的細胞產生 ® 的代謝產物，具有細胞毒性並會抑制細胞生長，除此之外 ，它會藉由對細胞高基氏體pH的影響，抑制蛋白質糖基化 。在瓶中無供應之培養的R223細胞通常產生5 mM氨，在 有養分供應之發酵槽培養則產生l〇mM。 在一開始評估氨的影響時，R223細胞係生長在震盪瓶 中，沒有加氨或加氨（2、5或10 mM)，重複培養之各氨濃 度係在不同的pH値下得到，藉改變施加氣體之C02含量。 此處之主要目標係決定氨對生長抑制的程度，及測試減低 * 的pH能不能克服此生長抑制。 氨被發現會抑制細胞生長（表2)，而減低的pH並不能 緩和氨所造成的抑制作用，雖然用來減少pH的升高之 ^02本身可能造成一些生長抑制。 培養在僅3天後結束，加長培養是無效的，因爲不揮 發性的酸代謝產物累積會造成所有培養物pH下降數個pH 單位。 23 200307047 表2 在不同的培養pH値中，氨對R223細胞生長及產量的影 響。pH係以施加 氣體之co2含量調整。 pC02 % 加入的氨 mM 起始pH 最大細胞 數/ml X ΙΟ'6 比生長速 率 h·1 EPO EU/ml Q-epo EU/106 細胞/h 5 0 7.26 0.9 0.020 1719 35 10 0 6.93 0.7 0.017 1580 40 15 0 6.75 0.4 0.012 1186 56 5 2 7.15 0.8 0.018 1796 38 10 2 6.88 0.7 0.017 1928 42 15 2 6.80 0.3 0.009 1404 78 5 5 7.18 0.6 0.018 2639 40 10 5 6.91 0.6 0.014 1663 51 15 5 6.75 0.3 0.016 1626 45 5 10 7.17 0.5 0.016 2222 43 10 10 6.90 0.6 0.014 1966 50 15 10 6.75 0.2 0.008 1891 86200307047 发明 Description of the invention [Technical field to which the invention belongs] The present invention relates to novel human cells capable of producing proteins and dystrophin of hydrazamine that can grow in sulfamamide-free medium. In addition, it relates to novel methods of protein production and to the use of amidoxamine synthetase (GS) as a selectable marker in human cells that are deficient in amidine. [Prior art] Proteins produced by mammalian cell culture are proteins used to provide therapeutic and diagnostic applications. Today, mammalian cell cultures are a better source of many important proteins used in human and animal medicine, especially those that are relatively large, complex, and glycosylated (N.B. Finter et al., In large mammals Cell culture technology, 1990, ed. A. Lubiniecki, Marcel Dekker, Inc., New York) 0 For example, the production of human protein erythropoietin (EPO) by cell culture has been described in WO 93/09222. Using human fibroblast mother cells transfected with an encoding human EPO exogenous gene, human EPO has achieved a considerable specific production rate. In a method for further producing human EPO (WO 94/12650), a human fibrosarcoma cell line HT1080 transfected with a DNA sequence capable of activating endogenously encoded EPO was used. A similar method has been described in WO 99/09268. Immortalised human cells such as Namalwa, Hela S3 and HT 1080 cell lines are transfected with DNA sequences capable of activating endogenously encoded EPO. 200307047 described in WO 93/09222, WO 94/12650, and WO 99/09268. The cell lines used to produce human EPO are all cultured in a medium containing amidine. This is unfavorable because when amoxidamine is used as an energy substrate by cultured cells, ammonia is a metabolite that is cytotoxic and inhibits cell growth. Furthermore, because of its effect on pH in the cells of the high Kellet cells, it inhibits protein glycosylation. Producing a large amount of tissue plasminogen activator, a glycosylated protein, has been described in WO 87 / 〇4462. Gs has been used as an amplification system for co-amplification of genes encoding tissue plasminogen activator (tPA) in pro to trophic Chinese hamster ovary (CHO) cells. It does this by transfecting cells with a gene that encodes GS. However, as found in EP-A 148 605, the use of CHO cells to produce glycosylated human protein lines is unfavorable. The protein synthesized by cH0 cells has an average carbohydrate composition that is somewhat different from naturally occurring glycosylated human proteins. This is different because human cells contain α2 · 3 sialyltransferase and α2 · 6 sialyltransferase enzymes. CHO cells only have α2.3 sialyltransferase, so they cannot produce the bond between α2.6 sialic acid and oligosaccharide moiety (moieties). CHO cells lack enzymes for the carbohydrate structure of suiphation. Although CHO cells have α 1-6 fucose transferase (fucosyltransferase at the center of attachment), also lacking al-3 fucose transferase (fucosyl residue at the end of attachment). Human cells have both fucose transferases (Cumming DA ·, 199 and Glycobiology Vo. 1, No. 2, 115 -130 'Jenkins Ν · and Curling EM · Α., 1994, enzyme and Micorbial Technology Vol. 16,354-364; Lee et al., 1989 200307047, Journal of Biological Chemistry, Vol. 264, 13848-13855). Therefore, glycosylated proteins synthesized by CHO cells do not have promising characteristics, such as in vivo biological activity produced by human cells. In WO 89/10404, independent methods for making myeloma cells such as mouse fused tumors, mouse plasmacytoma cells, and rat fusion tumor cells amidoxamine are reported by transforming them with GS. It further proves that GS can be used to co-amplify genes encoding light and heavy chains of immunoglobulin molecules and to co-amplify genes encoding fibrinolytic enzymes in myeloma cell lines. However, rodent cell lines have disadvantages in that the attachment of N-ethanolamylneuraminic acid residues replaces N-acetamidine neuraminic acid, the inability to undergo sulfuration, and the presence of αΐ.3 galactosyltransferase enzymes. The oligosaccharide structure of glycoproteins synthesized in rodent cells can therefore generate immunity in humans. SUMMARY OF THE INVENTION The object of the present invention is to provide an improved method which does not have the above-mentioned disadvantages when producing proteins (especially when producing glycoproteins), and can obtain high protein titers. This can be achieved by using novel amidineamine-deficient human cells according to item 1 of the scope of patent application, and novel methods according to item 7 of the scope of patent application. According to the present invention, human cells that are dystrophic of amidoxamine can be obtained 'transfected with (first) exogenous DNA sequences encoding a protein or exogenous DNA sequences capable of changing the expression of an endogenous gene encoding a protein' And further transfected with (second) an exogenous DNA sequence capable of encoding amidoxamine synthetase (GS) (preferably mammalian 200307047 GS), wherein these exogenous DNA sequences are located on one or more DNA The construct, the transfected cells are capable of producing the protein and are capable of growing on amidine-free medium. DNA sequence, an exogenous DNA sequence that encodes a protein, or, an exogenous DNA sequence that can alter the expression of an endogenous gene that encodes a protein, and an exogenous DNA sequence that can encode a GS, usually located in a DNA construct In vivo, for example, expression vectors or infectious vectors. Plastids can be used as expression vectors. As infectious vectors, vectors such as retrovirus, herpes virus, adenovirus, adenovirus-associated virus, mumps virus and polio virus can be used. A better expression vector is used, especially a plastid. "External DNA sequences that can encode proteins" may include additional sequences, such as regulatory sequences such as promoters and / or enhancers, polyadenylation sites and splice junctions, which are commonly used for performance The exogenous gene may include one or more separate targeting sequences and optionally a selectable marker-encoding DNA (as described in WO 93/09222). The exogenous DNA sequence expressed by an endogenous gene may include an exogenous DNA sequence that does not encode a gene product of a protein, but may encode a portion of the gene product (such as an exon), and may include additional sequences, such as regulatory sequences And splice adapters, which are commonly used to express exogenous DNA sequences. They may further include targeting sequences and, if desired, a DNA that encodes a selectable marker (as described in WO 93/09222). Usually "" Exogenous 200307047 DNA sequence capable of altering the expression of an endogenous gene encoding a protein "is a chromosomal DNA inserted into a cell after transfection into the cell. Homologous recombination or targeting is used here to replace or disable the regulatory regions associated with endogenous genes containing regulatory sequences. As the regulatory sequence, for example, a promoter and / or an enhancer can be used, which results in higher gene expression than corresponding non-transfected cells (as described in WO 93/09222), and a suitable promoter can be regulated or sustained Expressed promoter. A suitable promoter may be a strong promoter, which depends on the cell line used, such as the human cytomegalovirus major immediate early promoter (hCMV-MIE), SV 40 early and late promoters, Other adenovirus promoters, any polyoma virus or milk polyvesicular virus early and late promoters, interferon alphaΐ promoter, murine metallothionein promoter, Rous's sarcoma virus long-term repeat promoter, β-beads Protein promoter, albumin promoter, ovalbumin promoter, murine β-globin promoter and human β-globin promoter. According to the invention, the "encoding exogenous DNA sequence of GS" can be controlled by a strong promoter and a weak promoter. If the exogenous DNA sequence only needs to express a gene encoding GS, a strong promoter is used. Raw DNA sequences are used as selectable markers and weak promoters are used if the GS line is used for amplification. Suitable promoters may be regulatable or persistently expressed promoters. The promoter can be selected. For example, GS is expressed in a concentration sufficient to grow the transfected cells, but the cell culture does not produce a high amount of ammonium ammonia metabolite ammonia, usually not more than 4 mM, preferably not more than 2 mM, more preferably less than 2 mM ammonia. "Selectable markers" provide alternative phenotypes, making it possible to identify and separate recipient cells. 200307047 Recipient cells. GS can be used in the present invention as a selectable marker to screen out successfully transfected cells. Aminoxamine auxotrophic human cells whose invasion and expression can encode exogenous DNA sequences of GS. Depending on the cell line used, strong promoters can be, for example, hCMV-MIE, SV 40 early and late promoters, other adenoviruses Promoter, any polyoma virus or milk polyvesicular virus early and late promoter, interferon alpha 素 promoter, murine metallothionein promoter, Rous sarcoma virus long-term repeat promoter, β-globin promoter, Concomitant albumin promoter, ovalbumin promoter, murine beta-globin promoter and human beta-globin promoter. Depending on the cell line used, the weak promoter can be, for example, a murine leukemia virus long-term repeat, herpes simplex virus chest Glycoside kinase and murine breast tumor virus long terminal repeats, preferably, the gene encoding GS is under the control of a strong promoter, and more preferably under the control of the hCMV-MIE promoter. Specific examples of energy, an amplifiable mammalian GS sequence from a hamster and its use as a selectable marker in mammalian cells are well known in the art, and such as WO 87/04462, WO 91 / 06657 and WO 89/01036; the examples of the present invention use this hamster GS expression unit and individual screening methods as mentioned in the references. Π can encode protein-derived DNA sequences "or" can change the encoding protein "Exogenous DNA sequences expressed by endogenous genes" and "exogenous DNA sequences that can encode GS" are located on one or more DNA constructs. Preferably, these exogenous DNA sequences are located on more than one, more Ideally located on two DNA constructs. If these exogenous DNA sequences are located on a DNA construct, they may be functionally combined. For example, their performance 11 200307047 will be subject to the same regulatory sequences, such as promoters and / or enhancers (as described in wo 89/10404). ) Driving. Human Cells Ammonium Deficiency Nutrition Human cells refer to all human cells that do not exhibit GS or perform poorly GS, so they can grow in a medium containing amidine. However, they cannot grow or grow very poorly in amamine-free media. difference. An amlinoxamine auxotrophic human cell line used in the present invention is a dysfunctional amlinoxamine auxotrophic human cell or an immortal amlinoxamine auxotrophic human cell. Desperate amlinoxamine auxotrophic human cell line has a limited life span in the medium. Amlinoxamine auxotrophic human cells, immortal (also known as permanent or established) amlinoxamine auxotrophic human cell line in the medium As known to those skilled in the art, with appropriate subcultures and subcultures, amoxidine auxotrophic cells with significantly infinite life span. Examples of human cells that are dysfunctional amlinoxamine auxotrophs can be human fibroblasts and human fetal lung cells, and examples of human cells that are immortalized amidamine can be human fibrosarcoma cells, such as the HT1080 cell line ( Such as DSMZ No. ACC-315 or ATCC No. CCL 121) and B-lymphocyte human cells such as HL60 (DSMZ No. Acc-3) or Namalwa (DSMZ Acc-24) cell line. The preferred immortalized amidoxamine auxotrophic human cells used in the present invention, more preferably, the immortalized amoxidox auxotrophic human cell line is used for B-lymphocytes or fibrosarcoma cells, and more preferably For human fibrosarcoma cells, the best line is the HT1080 cell line (eg, ATCC No. CCL 121). Transfection 12 200307047 Amiamine-deficient human cells can be transfected with exogenous DNA sequences using known genetic engineering techniques. Transfection of exogenous DNA sequences depends on whether the sequences are on one or more DNA constructs. If the sequences are on more than one DNA construct, transfection can occur separately or co-transfect with each sequence. When transfection occurs separately with each sequence, the sequence of sequence transfection is usually random. Transfection that occurs separately with each sequence is preferably preceded by "exogenous DNA sequence that can encode the protein" or "can be changed to encode the protein The exogenous DNA sequence expressed by the endogenous gene is "transfected" and then transfected with the "exogenous DNA sequence encoding GS". The transfected amidoxamine auxotrophic cells can be cultured and transfected after each transfection. Assess protein production. In order to successfully screen transfected cells, these lines were grown in amidine-free medium. Cells can be grown directly in amidine-free medium or first in amidine-containing medium and then gradually diluted to An amlinoxamine-free medium, such as an ammoniumamine concentration of 10 mM, and then a stepwise dilution from 2 mM to 0 mM. The appropriate screening method can be selected according to the cell line used. As is apparent from the foregoing, the present invention can Production of a protein and amidoxamine auxotrophic human cell lines capable of growing in amidine-free medium obtained from exogenous DNA sequences encoding the protein or capable of altering the encoding protein The cell is transfected with an exogenous DNA sequence expressed by the endoplasmic endogenous gene and an exogenous DNA sequence encoding amidine synthase, where the exogenous DNA sequences are located on one or more DNA constructs. The exogenous DNA sequence encoding a protein or the exogenous DNA sequence capable of altering the expression of an endogenous gene encoding a protein 13 200307047 can be amplified after transfection according to a known method of gene amplification as described in WO 94/12650. May Amplified genes can encode enzymes such as DHFR (dihydrofolate reductase), GS, adenosine deaminase, asparagine synthetase, aspartate carbamate transferase, metallothionein-1, bird Alanine decarboxylase, P-glycoprotein, nucleotide reductase, thymidine kinase or xanthine guanine phosphoribosyltransferase can be used for this purpose. Cell lines containing amplified replication of these genes can, for example, be used without enzyme metabolites Survival in a medium containing a selective agent such as methotrexate (MTX) in the case of DHFR and methionine sulphoximine (MSX) in the case of GS. The protein produced by the amnoxamine auxotrophic human cells transfected with the present invention is an aglycosylated and glycosylated protein, and a glycosylated protein refers to a protein having at least one oligosaccharide chain. Examples are aglycolated hormones, like luteinizing hormone-releasing hormone, thyroxine-releasing hormone, insulin, somatostatin, prolactin, adrenocorticotropic hormone, melanocyte melanocyte stimulating hormone, vasopressin, and derivatives such as Desmopressin, oxytocin, calcitonin, parathyroid hormone (PTH) or fragments thereof (such as PTH (1-43)), gastrin, secretin, enterokinin, cholecystokinin, Angiotensin, human placental luteinizing hormone, human chorionic gonadotropin (HCG), caerulein, and motilin; aglycone-free analgesics such as enkephalin and its derivatives (see US-A 4 277 394 And EP-A 03 1567), endorphins, daynorphin and kyotorphin; aglycosylases such as aglycosylated neurotransmitting substances such as frogs 14 200307047 dermatan, neurotensin, bradykinin and substance P; nerve growth factor (NGF) Home Family, epithelial growth factor (EGF) and fibroblast growth factor (FGF) family of aglycosylated growth factors and aglycosylated hormones and growth factors Τν ?, Mΰ daughter receptors. Examples of glycosylated proteins are hormones and hormone-releasing factors, such as growth hormones, including human growth hormone, bovine growth hormone, growth hormone-releasing factor, parathyroid hormone, thyroid stimulating hormone, EP0, lipoproteins, α-1-anti-pancreas Proteases, follicle stimulating hormone, calcitonin, luteinizing hormone, glucagon, coagulation factors such as factor VIIIC, factor IX, tissue factors and von Willebrand factor, anticoagulant factors such as protein C, atrial natriuretic factor, lung Surfactants, plasminogen activators such as urokinase or human urine or tissue type plasminogen activator (t-PA), thrombin, hematopoietic growth factor, enkephalin, RANTES (regulates activated natural T-cell expression And secretion), human macrophage inflammation protein (MIP-1-α :), hemagglutinin such as human hemagglutinin, mulleriati's inhibitory substance, relaxin A chain, relaxin B chain, pre-relaxin, rat Gonadotropin-related peptides, microbial proteins such as / 5-lactanase, DNase, statin, activin, renin, vascular endothelial growth factor (VEGF) , Hormonal or growth factor receptors, integrins, proteins A or D, rheumatoid factors, neurotrophic factors such as derived from bone neurotrophic factor (BDNF), neurotrophin-3, -4, -5, or- 6 (NT-3, NT-4, NT-5, or NT_6), or nerve growth factors such as NGF-β, derived from platelet growth factor (PDGF), fibroblast growth factors such as FGF and bFGF, epidermal growth factor (EGF), transforming growth factor (TGF) such as TGF-α and 15 200307047 TGF-Θ, including TGF-β1, TGF′2, TGF-p3, TGF-P4 or TGF-p5, insulin-like growth factor-I and − π (IGF-I and IGF-II), des (l-3) -IGF-I (brain IGF-I), insulin-like growth factor binding protein, CD protein (a group of differentiated proteins) such as CD-3, CD-4 , CD-8 and CD-19, osteoinductive factor, immunotoxin, bone morphogenetic protein (BMP), cytokine and its receptor, and chimeric protein, including its receptor cytokine, Including, for example, tumor necrosis factor alpha and / 3, their receptors (TNFR-1, EP 417 563 and TNFR-2, EP 41 7 014) and their derivatives, Interferons such as interferon-α,-/ 5, and -7, cell population stimulating factors (CSFs), such as M-CSF, GM-CSF, and G-CSF, and interleukins (ILs) such as IL-1 to IL-10 , Superoxide disproportionation, T-cell receptors, surface membrane proteins, degradation acceleration factors, viral antigens such as part of the AIDS capsule, transporters, homing receptors, addressins, regulatory proteins, antibodies, Chimeric proteins such as immunoadhesin, and glycosylated proteins of any of the fragments listed above. Preferably, the present invention produces a glycosylated protein, and more preferably, the present invention produces an N-glycosylated protein. The best glycosylated hormone like EPO is N-glycosylation and its biological activity depends on it, or In particular, EPO is produced. Cell culture Any suitable culture procedure and equipment known in the art can be used to grow the transfected human cells of the present invention. Adding about 0.1 to 20% (preferably 0.5 to 15%) of common blood-free amphetamine basal medium and blood-free amphetamine-free basal medium can be used as the culture medium. In addition, it can also be used without animal sources. Common ammonium-free ammonium basal medium for proteins 16 200307047. It is preferable to use blood-free ammonium amine-based basal medium. The blood maggots that can be used are fetal bovine blood maggots or adult bovine blood maggots, and fetal bovine blood maggots are preferably used. Commonly available amphetamine-free basal medium is, for example, amphetamine-free Eagle's minimum minimal medium (MEM) medium, amphetamine-free Dulbecco's modified Eagle's medium (DMEM), and amphetamine-free Iscove's DMEM medium (N. Iscove and F. Melchers, Journal of Experimental Methods, 1978, 147, 923), Ammonium-free Ham's F12 medium (RG Ham, Proceedings of National Academy of Science, 1965, 53, 288 ), Ammonium Free Ammonium L-15 Medium (A. Leibovitz, American Journal of Hygiene, 1963, 78, 173), Ammonium Free RPMI 1640 Medium (GE Morre et al., The Journal of the American Medical Association, 1967, 199, 519), an exclusive ammonium-free medium and a mixture thereof in an appropriate ratio. Fortification of common cell culture growth media for high-density cell culture is well known in the art and has been described in, for example, GB 2251249, which is also very suitable as the amidamine-free medium of the present invention. Common additives can be added to the common amine-free ammonium basal medium. Frequently added additives include proteins in blood pupa and further ingredients that may have a positive effect on cell growth and / or cell survival as needed. Proteins in blood pupa such as bovine blood pupa albumin (BSA), transferrin, and / or insulin; further ingredients that can have a positive effect on cell growth and / or cell survival such as soybean fat, selenium, and ethanolamine. Amino acids that replace amidoxamine and / or nucleosides can be added to the culture medium depending on the cell line used. Examples of amino acids such as isoleucine 17 200307047 acid, leucine, valine, lysine, Asparagine, aspartic acid, asparagine, serine, alanine. As needed, amoxifen can be added to common amamine-free basal medium at low concentrations, usually less than 1 mg / 1, and preferably less than 0.5 mg / 1 to support its biosynthesis ( (E.g. transamination reaction) 〇 If the exogenous DNA sequence encoding the protein or the exogenous DNA sequence capable of changing the expression of the endogenous gene encoding the protein is amplified with an amplifiable gene after transfection, the relevant selection agent can be When added to the common amphetamine-free basal medium, the concentration range of the selective agent depends on the cell line used, usually 10 μM and higher. Adaptation of Suspension-Growth in Blood-Free 淸 Media 榖 Amidoxamine auxotrophic human cells can be anchor-dependent or detached-independent 'which can be used as a starting material to obtain transfected ospyromethamine-deficient human cells according to the invention According to the present invention, the transfected cells are capable of producing proteins and can be grown in amidine-free medium. If fixation-dependent human cells are used, such as the YT1080 cell line (ATCC No. CCL 121), it can be adapted to become an independent HT1080 cell line that can be grown in suspension in blood-free medium, which has not been described in the literature. The adaptation can occur before or after transfection with an exogenous DNA sequence encoding a protein or an exogenous DNA sequence capable of altering the expression of an endogenous gene encoding a protein and an exogenous DNA sequence encoding GS. Preferably, the cells are first transfected with an exogenous DNA sequence encoding a protein or an exogenous DNA sequence capable of changing the expression of an endogenous gene encoding the protein, and then adapted to suspension growth in a blood-free maggot medium, and then further encoded with GS exogenous 18 200307047 DNA sequence transfection. If desired, the transfected cells can then be adapted to suspension growth in a blood-free, ammonium-free medium. The dystrophin-deficient human cells transfected with amidoxamine according to the present invention may be fixedly dependent or fixedly independent, and capable of suspension growth in a blood-free amidoxamine-free medium. The better transfected human amalgam-deficient human cell line is stable and independent and can be grown in suspension in a blood-free amidamine-free medium. The adaptation makes it a fixed and independent cell capable of growing in suspension in a blood-free maggot medium, which can be achieved by adapting the cells with a blood maggot-containing medium in the first step. This can be achieved by, for example, treating the cells with trypsin and then shaking or releasing the cells by shaking. In the second step, the cells then adapt to the bloodless maggot medium by reducing the blood maggot content. In the adaptation, if a selection agent is used, the amount can be reduced to avoid inhibiting cell growth. However, cells can also be adapted to grow in blood-free agar medium by subsequently reducing the amount of blood maggots in the first step, and become fixed and independent cells capable of suspension growth in the second step by treating the cells with trypsin and then shaking or by shaking The cells are released to achieve this, and both steps can be performed simultaneously. Preferably, the cell line is adapted to become a stand-alone cell in the first step, and the cells are released by shaking, and then the second step is adapted to the non-blood-capped medium by reducing the blood crust content in the second step. The medium described above can be used as the basis for a blood-containing maggot medium. The selection agent defined herein can be added to the culture medium. The concentration range at which the selection agent is applied depends on the cell line used, usually ΙμΜ and higher. The medium containing blood pupa is usually supplemented with about 0.1 to 20%, preferably 0.2 to 10%, and the best 0.5 to 5% blood pudding. The blood pudding that can be used is mentioned in 19 200307047 above. Subsequent reductions in blood radon levels can be achieved by gradually reducing blood radon levels, for example from 10% to 1% to 0%. Method for producing protein The transfected amlinosamine auxotrophic human cell line is used in a method for producing a protein by culturing the cell in a culture medium under conditions suitable for expressing the protein and recovering the protein. The resulting protein line is described above. The common amidine-free ammonium base medium and common additives as described above can be used as the culture medium. Suitable culture conditions are conventionally used for mammalian cell in vitro culture as described in WO 96/39488. Protein acquisition Proteins can be separated from cell cultures using conventional separation techniques, such as immunoaffinity or ion exchange column partitioning, precipitation, reverse-phase HPLC, chromatographic analysis, chromatofocusing, SDS-PAGE, colloidal filtration . Those skilled in the art should understand that the purification method suitable for the polypeptide of interest may need to be modified based on the characteristics of the polypeptide in recombinant cell culture. [Embodiment] Example 1 Preparation of a human fibrosarcoma cell line HT1080_R223 A fixed human HT1080-R223 cell line containing multiple human EPO gene copies (copies) is an industrially used cell for the production of EPO. It was originally developed by Transkaryotic Therapies Created (Cambridge, MA 02139 (US)), its line is derived from the fixed-dependent human fibrosarcoma HT 1080 fine 200307047 cell line. The parental HT1080 cell line (ATTC No. CCL 121) has the ability to produce EPO, which was transfected with the DNA construct pREP022 similar to the DNA construct PREP018 (described in WO 95/3 1 560), except for the DHFR gene line In the opposite direction and pREP022 contains about 600 base pairs of homologous sequences less than PREP018. This cell line will be mentioned next as the R223 cell line. Example 2 Adaptation of the R223 cell line to be able to grow in suspension in a blood-free agar medium by trypsin treatment Released in a stationary bottle to attach to the cultured growth cells, suspended in a proprietary blood-free agar containing ammonium The medium (further referred to as "HM9") was supplemented with 10% dialyzed fetal bovine hemorrhage (dFBS) and 500 nM MTX, and then cultured in a shake flask. Growth began after six days, and the cells were subcultured to the same medium (Figure 1). Once a credible growth pattern is established, reduce the blood pupae content of the culture medium to 1 ° / ◦ 'again, and credible growth can be re-established before the added blood pupae are completely used up. The cultures supplemented with blood maggots and blood maggots were overgrown to evaluate the yield. The results are shown in Table 1 with the data of attached cultures. After adequate adaptation, the specific growth rate of the suspension culture (even without blood pupa) is the same as that of the attached culture with added blood pupa. In the adaptation of blood pupa to the medium, the specific rate of EPO synthesis decreased by 50% from attachment to suspension culture, from 24 to 12 EU / 106 cells / hour. However, after adapting to the growth of blood pupa, the rate was roughly the same. Will return to 18 EU / 106 cells / hour. 21 200307047 Table 1 Calls and yields of R2; 23 cell line before and after adaptation to bloodless pupa medium, in adherent and suspension cultures. Maximum number of cells / mL χ ΙΟ'6 Specific growth rate h · 1 Cumulative cells h / ml χ 10 · 6 EPO EU / ml q epo EU / 106 cells / h Attachment culture (DMEM + 10% dFBS 500 nM MTX) Not applicable 0.0157 Not applicable Not applicable 24 Suspension culture (HM9 + 10% dFBS 500 nM MTX) 1.4 0.0185 334 4199 12 Suspension culture (HM9 (no blood) 500 nM MTX) 1.1 0.0166 210 3600 18 Example 3 Adaptation of HT1080 (ATCC CCL121) HT1080 cell line ATCC CCL121 was obtained from the American Type culture Collection (Rockville, Maryland, US A). The cell line was initially grown in adherent culture in 10% fetal bovine blood pupa (FBS). In DMEM. The adaptation of suspensions and anemias can be performed according to the procedure of FIG. 2. To start suspension culture, the cell line was released from the attachment culture by trypsin treatment, and the released cells were resuspended in HM9 with 2% FBS, which was then cultured by shaking once the cells were added with 2% FBS Suspension growth of HM9 was established, and the culture was diluted with the same medium as the progeny culture. Suspicious growth of the trusted HT1080 cell line was established after 30 days (Figure 3). After that, the blood pupae content of the culture medium decreased, and finally disappeared completely. Cells continued to grow in subcultures without blood. Example 4 Effect of Ammonia on the Growth and Yield of R223 Cell Line When amoxidamine is used as the energy matrix, ammonia is a metabolite produced by cultured cells. It has cytotoxicity and inhibits cell growth. In addition, it It can inhibit protein glycosylation by affecting the high pH of the cells. R223 cells cultured without a supply in a flask usually produce 5 mM ammonia, and cultured in a fermenter with a nutrient supply produces 10 mM ammonia. In the initial evaluation of the effect of ammonia, the R223 cell line was grown in a shaker flask without ammonia or ammonia (2, 5 or 10 mM). Each ammonia concentration of repeated cultures was obtained at different pH levels, and was changed by CO2 content of the applied gas. The main goals here are to determine the degree of growth inhibition by ammonia, and to test whether a reduced pH * can overcome this growth inhibition. Ammonia was found to inhibit cell growth (Table 2), and a decrease in pH did not mitigate the inhibitory effect caused by ammonia, although ^ 02 used to reduce the increase in pH may itself cause some growth inhibition. Cultivation ends after only 3 days. Lengthening the culture is not effective because accumulation of non-volatile acid metabolites will cause the pH of all cultures to drop by several pH units. 23 200307047 Table 2 Effect of ammonia on the growth and yield of R223 cells in different culture pH 値. The pH is adjusted by the co2 content of the applied gas. pC02% added ammonia mM initial pH maximum number of cells / ml X ΙΟ'6 specific growth rate h · 1 EPO EU / ml Q-epo EU / 106 cells / h 5 0 7.26 0.9 0.020 1719 35 10 0 6.93 0.7 0.017 1580 40 15 0 6.75 0.4 0.012 1186 56 5 2 7.15 0.8 0.018 1796 38 10 2 6.88 0.7 0.017 1928 42 15 2 6.80 0.3 0.009 1404 78 5 5 7.18 0.6 0.018 2639 40 10 5 6.91 0.6 0.014 1663 51 15 5 6.75 0.3 0.016 1626 45 5 10 7.17 0.5 0.016 2222 43 10 10 6.90 0.6 0.014 1966 50 15 10 6.75 0.2 0.008 1891 86
在接下來的實驗（表3) pH係藉由調整培養基NaHC03 的含量來調整，並維持施加氣體的C02含量在無抑制作用 的濃度。測試的pH範圍從pH 7.0到7.5(要強調的是，在 瓶中培養的培養基pH不能被控制，而且會大大的下降， 即使最先兩天的培養，所指的pH値係各培養之起始pH)。 比生長速率在pH 7.5 (NaHC03於3 g/Ι)時最大，但在這 個pH下，因爲10 mM氨的存在使生長速率及最大細胞濃 度減半，EPO合成比速率減少四倍（表3)。 當 pH 7.25 (NaHC03 於 1.5 g/Ι)時，相較於 pH 7.5 的 培養，比生長速率減少，EPO合成比速率增加，然而因爲 有10 mM氨存在，生長速率較不受影響，EPO合成比速率 24 200307047 不受影響。 在最低的硏究pH，pH 7.0 (NaHC03於0.75 g/l)，比 生長速率進一步減少但仍比pH 7.5時較不受氨影響。EPO 合成比速率因爲加入的氨之存在而增加（這可能是因爲生 長速率降低）。 表3 在不同的培養pH値下，氨對R223細胞生長及產量的影響 〇 pH係藉由改變培養基中NaHC03含量來調整〇 最大細胞 比生長速 EPO Qepo 起始 數/ml 率 pH X 10·6 h·1 EU/ml EU/106個細 胞/hIn the next experiment (Table 3), the pH was adjusted by adjusting the content of NaHC03 in the culture medium, and the C02 content of the applied gas was maintained at a concentration without an inhibitory effect. The tested pH range is from pH 7.0 to 7.5 (It should be emphasized that the pH of the medium cultured in the bottle cannot be controlled and will greatly decrease, even if the first two days of culture, the pH refers to the start of each culture Starting pH). The specific growth rate was the highest at pH 7.5 (NaHC03 at 3 g / 1), but at this pH, the specific rate of EPO synthesis was reduced by four times because of the presence of 10 mM ammonia, which halved the growth rate and maximum cell concentration (Table 3) . At pH 7.25 (NaHC03 at 1.5 g / Ι), the specific growth rate is reduced and the specific rate of EPO synthesis is increased compared to the culture at pH 7.5. However, because of the presence of 10 mM ammonia, the growth rate is less affected. Rate 24 200307047 is not affected. At the lowest investigated pH, pH 7.0 (NaHC03 at 0.75 g / l), the specific growth rate was further reduced but still less affected by ammonia than at pH 7.5. The specific rate of EPO synthesis is increased by the presence of added ammonia (this may be due to a decrease in the growth rate). Table 3 Effects of ammonia on the growth and yield of R223 cells at different culture pHs. PH was adjusted by changing the NaHC03 content in the culture medium. Maximum cell specific growth rate EPO Qepo initial number / ml rate pH X 10.6 h · 1 EU / ml EU / 106 cells / h
NaHC03 3 g/l 7.49 1.3 0.030 951 16 NaHC03 3 g/l lOmM 氨 7.51 0.5 0.016 256 4NaHC03 3 g / l 7.49 1.3 0.030 951 16 NaHC03 3 g / l lOmM ammonia 7.51 0.5 0.016 256 4
NaHC03 1.5 g/l 7.26 1.1 0.027 1151 24 NaHC03 1.5 g/l 10 mM 氨 7.27 0.7 0.021 1191 26NaHC03 1.5 g / l 7.26 1.1 0.027 1151 24 NaHC03 1.5 g / l 10 mM ammonia 7.27 0.7 0.021 1191 26
NaHC03 0.75 g/l 7.02 0.9 0.020 995 26 NaHC03 0.75 g/l lOmM 氨 7.01 0.4 0.016 1009 42 25 200307047 實施例5 R223細胞系在5公升規模發酵槽培養的產量 對於R223細胞系，實施了三個5公升的發酵，控制在 介於6.95和7.15的不同pH値（參見表4)，每一個培養得 到濃縮的營養供應，其含有胺基酸和葡萄糖，設計來維持 主要的消耗養分在足夠濃度。培養物在pH 7.15及pH 7.05 生長的很好，但在pH 6.95則無法生長，可能是因爲用來 控制該pH之高濃度C02。 表4 發酵槽培養之R223細胞系的產量及代謝. 發酵條件 最大細 胞數/ml xl〇-6 倍殖時 間 小時 累積的細 胞 小時/ml X ΙΟ'6 EPO EU/ml q Ερο EU/106 細胞數 /h 最終氨 mg/1 pH 7.15 NaHC03 於 2 g/i 1.0 46 197 10795 58 163 pH 7.05 NaHC03 於 2 g/i 1.0 51 195 11414 64 157 pH 6.95 NaHC03 於 2 g/1 0.2 105 62 2995 70 未做 實施例6.以GS轉染的R223及GS轉染子在附著培養的 產量。 用來轉染的細胞係來自實施例2之R223細胞系懸浮 26 200307047 適應的無血淸儲備溶液，當這些細胞長到像大的多細胞聚 集時，再將他們以胰蛋白酶處理使其成爲實質上的單細胞 懸浮。 轉染從實施例2獲得的約1〇7單懸浮適應R223細胞系 分液（在不含鈣或鎂的磷酸緩衝鹽）係混合以20 含有GS 基因之線狀DNA(DNA序列PCMGS Bam H1描述於 Bebbington 等人，1992，Biotechnology 10，169-175)及使 用 Biorad Gene Pulser (450 伏特 ’ 250 pF)施以電穿孔。 作爲控制組者，等量分液之細胞在不加DNA下作電穿孔。NaHC03 0.75 g / l 7.02 0.9 0.020 995 26 NaHC03 0.75 g / l 10 mM ammonia 7.01 0.4 0.016 1009 42 25 200307047 Example 5 Yield of R223 cell line cultured in a 5 liter scale fermentation tank For the R223 cell line, three 5 liters were implemented Fermentation is controlled at different pH levels between 6.95 and 7.15 (see Table 4). Each culture receives a concentrated nutrient supply, which contains amino acids and glucose, and is designed to maintain the main consumption of nutrients at a sufficient concentration. The culture grew very well at pH 7.15 and pH 7.05, but could not grow at pH 6.95, probably because of the high concentration of C02 used to control this pH. Table 4 Yield and metabolism of R223 cell line cultured in a fermentation tank. Maximum number of cells in fermentation conditions / ml xl0-6 hours of doubling time accumulated hours / ml X ΙΟ'6 EPO EU / ml q Ερο EU / 106 cells / h Final ammonia mg / 1 pH 7.15 NaHC03 at 2 g / i 1.0 46 197 10795 58 163 pH 7.05 NaHC03 at 2 g / i 1.0 51 195 11414 64 157 pH 6.95 NaHC03 at 2 g / 1 0.2 105 62 2995 70 Not done Example 6. Yield of R223 and GS transfectants transfected with GS in attachment culture. The cell line used for transfection was derived from the R223 cell line suspension of Example 2 2003 200347 adapted blood-free 淸 stock solution. When these cells grew like large multicellular aggregates, they were treated with trypsin to make them essentially Single cell suspension. Transfection of approximately 107 single suspension-adapted R223 cell lines obtained from Example 2 was dispensed (in calcium or magnesium-free phosphate buffered saline) and mixed with 20 linear DNA containing the GS gene (DNA sequence PCMGS Bam H1 description Electroporation was performed in Bebbington et al., 1992, Biotechnology 10, 169-175) and using a Biorad Gene Pulser (450 volts' 250 pF). As the control group, aliquots of cells were electroporated without DNA.
細胞以HM9培養基稀釋，不加MTX，含有0或10% dFBS，及分布在96孔培養盤或25 cm2瓶，35.5至37 °C 培養。 HM9培養基一開始含有2mM榖氨醯胺，但1天後稀釋 成0.5 mM，然後十天後換成無榖氨醯胺HM9培養基，並 且在第1天或第1〇天要再加入MTX(500 nM)於培養物中。 至於在不加DNA下作電穿孔的控制組細胞培養，沒 有生長。 以DNA電穿孔的細胞，有15個GS轉染子可以在96 孔盤鑑別出來，不管起始培養基有沒有MTX，都可獲得 GS轉染子。15個GS轉染子中，有7個成功的進展到瓶培 養（表5)，剩下的8個GS轉染子呈現異常的細胞型態，或 生長的不好，遺棄之。 分離出的每一個起始的七個GS轉染子，設置一組複 製靜置瓶培養’使用10 %添加血淸之無榖氨醯胺HM9培 27 200307047 養基，每隔1到4天，犧牲培養物以計數細胞數，並以 EPO ELISA測量EPO濃度。從這些數據集成可以估計每一 個GS轉染子的EPO合成比速率，數據整理於表5及（作 爲比較）數據也包括無轉染的R223細胞系，所有GS轉染子 相較於無轉染的R223細胞系呈現提高的EPO合成比速率 ，最佳的GS轉染子，3E10，具有高於無轉染R223細胞 系五到六倍之合成速率。 表5Cells were diluted in HM9 medium without MTX, containing 0 or 10% dFBS, and distributed in 96-well culture plates or 25 cm2 flasks and cultured at 35.5 to 37 ° C. The HM9 medium initially contains 2mM amidoxamine, but it is diluted to 0.5 mM after 1 day, and then replaced with amlamine-free HM9 medium after 10 days, and MTX (500 is added on day 1 or 10). nM) in culture. As for cell culture in the control group, which was electroporated without DNA, there was no growth. For cells electroporated with DNA, 15 GS transfectants could be identified in a 96-well plate. GS transfectants could be obtained regardless of the presence or absence of MTX in the starting medium. Of the 15 GS transfectants, 7 successfully progressed to bottle culture (Table 5), and the remaining 8 GS transfectants exhibited abnormal cell types or did not grow well and were discarded. For each of the first seven GS transfectants isolated, set up a set of replicated stationary flask cultures' using 10% ammonium amidineamine HM9 supplemented with blood spores 27 200307047 nutrients, every 1 to 4 days, Cultures were sacrificed to count the number of cells, and EPO concentration was measured by EPO ELISA. From the integration of these data, the EPO synthesis specific rate of each GS transfectant can be estimated. The data are summarized in Table 5 and (for comparison) the data also includes the non-transfected R223 cell line. All GS transfectants are compared to non-transfected. The R223 cell line exhibited an increased specific rate of EPO synthesis, and the best GS transfectant, 3E10, had a synthesis rate five to six times higher than that of the untransfected R223 cell line. table 5
R223細胞系GS轉染子及無轉染的R223細胞系之EPO合 成比速率。 數據來自生長在含有10% dFBS之附著培養，在製造 轉染子的培養基中不加甲氨喋呤（Methotrexate)，但在轉染 GS轉碎^-- JU到培裳例十° 添加 MTX (500 nM)到 轉染盤的曰子 q epo (附者） EU/106細胞數/h 3B3 ^^--- 10 59 3E10 二 OT71 1 ^—---- 10 139 "Τπο ^— 10 45 HUy 10 35 8F11 -- 9T71 1 -- 1 47 or 1Z 1 79 "8G3 -- 1 52 ⑦^咅鹜中無轉染的 R223細胞系 ^——— 不適用 24Specific rate of EPO synthesis in R223 cell line GS transfectants and non-transfected R223 cell line. The data comes from an attached culture containing 10% dFBS. Methotrexate was not added to the medium used to make the transfectants, but transfected with GS. ^-JU to Pei Sang. Add MTX ( 500 nM) to the transfection tray q epo (attachment) EU / 106 cells / h 3B3 ^^ --- 10 59 3E10 two OT71 1 ^ ----- 10 139 " Τπο ^ — 10 45 HUy 10 35 8F11-9T71 1-1 47 or 1Z 1 79 " 8G3-1 52 R223 cell line without transfection in ⑦ ^ 咅 鹜
實施例7 轉染子於懸浮培養及無血淸培養基。 由實施例6所得到的七個GS轉染子，GS轉染子 28 200307047 #3E10及GS轉染子#8G3進行懸浮培養。 細胞係以附著培養生長於靜置瓶，內含無榖氨醯胺 HM9 培養基，添力口 10 % dFBS 及 500nM MTX，於 35.5 至 37 °C。細胞在五或六天後藉搖動釋放，再懸浮於同樣的培 養基中，於相同溫度以震盪瓶培養，生長在二或三天後開 始，細胞於相同培養基次培養一次，然後過度生長以評估 產量。效價及產物合成速率至少高於生長在10 % dFBS及 500 nMMTX的無轉染R223細胞系兩倍（表6)。 表6 GS轉染子3E10及8G3在震盪瓶懸浮培養的生長及產量。 無榖氨醯胺HM9培養基含有10 % dFBS及500 nM MTX。 GS轉染子 最大細胞數/ ml X 10'6 最大EPO效 價 EU/ml q_EP。(懸浮） EU/106個細 胞/h 3Ε10 0.7 11113 57 8G3 1.2 10171 33 評估在懸浮培養生長之無 轉染的R223細胞系，於 10% dFBS 及 500 nM MTX中 1.4 4199 12 GS轉染子3E10及8G3的懸浮培養，無榖氨醯胺 HM9培養基中dFBS含量係逐步減少，3E10從10 %至2 %至 1 %至 〇·2 %至 0.1 %至 0 %，8G3 從 10 %至 2 %至 1 % 29 200307047 ，並在各dFBS濃度進一步減少前，建立可信的細胞生長， 8G3沒有進一步適應於1 % dFBS。 實施例8 GS轉染子3E10在無血淸懸浮培養的產量及代謝。 實施例7中適應於無血淸生長的GS轉染子3E10係以 懸浮培養培養於無血淸無榖氨醯胺之HM9培養基中，35.5 到37 °C。對於GS轉染子3E10細胞系，培養在相同條件 於含有榖氨醯胺的HM9培養基，其具有比無轉染的R223 細胞系快2到3倍的EPO合成比速率（表7)。 表7 GS轉染子3E10細胞在無血淸懸浮培養的生長及產量。 最大細胞數/ml EPO q_EP〇 xlO·6 EU/ml EU/ 106/h 3E10 (無榖氨醯胺 HM9培養基） 1.1-1.2 9731 - 14234 40 - 67 R223 (HM9培養基加6 mM榖氨醯胺） 1.0-1.6 3075-5818 13-24 GS轉染的細胞的高EPO產量伴隨代謝氨釋放的減少， 不同於在含有6 mM榖氨醯胺HM9培養基中瓶培養典型會 產生5 mM氨的無轉染之R223細胞系，GS轉染子3E10於 無榖氨醯胺HM9培養基只產生1.8 mM氨（表8)。 30 200307047 表8 GS轉染子3E10之氨合成減少。 最大細胞數/ ml 氨 q氨 xl〇-6 mM μ莫耳/106個細胞/h R223 (含有6 mM榖氨醯 0.9 4.8 30 胺之HM9培養基） 1.2 5.0 23 1.0 5.5 27 3E10撫榖氨醯胺HM9培 養基） 0.8 1.8 11 實施例9 分析產物品質。 由GS轉染子3E10在瓶培養產生的EPO係以免疫純化 和分析其glycoform分佈，圖4顯示EPO之等電點聚焦法 (IEF)膠體分析，其係得自最高細胞濃度時及收集生長於無 榖氨醯胺HM9培養基之3E10細胞培養物，如實施例8所 描述，包括從生長在HM9培養基之無轉染R223細胞系之 可相較的樣本。從GS轉染子3E10產生的EP0(相較於無轉 染的R223細胞系）呈現更酸的異構物，分析IEF膠可以定 量異構物的分佈和計算理論上異構物的相對活性（IRA)。 31 200307047 表9 分析來自於無轉染的R223細胞系及GS轉染子3E10之EPO 異構物相對活性。 亮帶 (Band) 號碼 比活性* 更鹼 1 2 3 4 5 6 0.071 7 0.194 8 0.273 9 0.373 10 0.658 11 0.989 12 0.999 13 1.000 14 0.796 15 16 更酸 17 其他次要 亮帶 無轉染的R223 各異構 物百分 比 活性（百 分比X 比活性） 5.2 5.2 6.3 11.4 10.5 9.2 0.7 8.8 1.7 8.7 2.4 5.5 2.1 4.4 2.9 4.3 4.3 4.5 4.5 1.4 1.4 0.2 0.2 14.4 總合= 100% 總合= IRA=20. 2 GS轉染子3E10 各異構 物百分 比 活性（百 分比X 比活性 2.5 3.2 3.6 6.0 6.2 7.5 0.5 10.5 2.0 11.6 3.2 9.3 3.5 9.5 6.3 7.8 7.7 6.1 6.1 5.3 5.3 1.7 1.4 0.5 8.7 總合= 100% 總合= IRA = 36.0Example 7 The transfectants were cultured in suspension and in a blood-free medium. The seven GS transfectants obtained in Example 6, GS transfectants 28 200307047 # 3E10 and GS transfectants # 8G3 were subjected to suspension culture. The cell line was grown in a stationary bottle in an adherent culture containing ammonium-free HM9 medium, 10% dFBS and 500nM MTX, at 35.5 to 37 ° C. Cells are released by shaking after five or six days, resuspended in the same medium, cultured in shake flasks at the same temperature, growth begins after two or three days, cells are cultured once in the same medium, and then overgrown to assess yield . The titer and product synthesis rate were at least twice higher than those of non-transfected R223 cell lines grown at 10% dFBS and 500 nMMTX (Table 6). Table 6 Growth and yield of GS transfectants 3E10 and 8G3 in suspension culture in shake flasks. Ammonium-free HM9 medium contains 10% dFBS and 500 nM MTX. GS transfectants Maximum number of cells / ml X 10'6 Maximum EPO titer EU / ml q_EP. (Suspension) EU / 106 cells / h 3E10 0.7 11 113 57 8G3 1.2 10 171 33 Assess the non-transfected R223 cell line grown in suspension culture, 1.4 4199 12 GS transfectants 3E10 and 10% dFBS and 500 nM MTX In suspension culture of 8G3, the content of dFBS in the ammonium-free ammonium HM9 medium gradually decreased, with 3E10 from 10% to 2% to 1% to 0.2% to 0.1% to 0%, and 8G3 from 10% to 2% to 1 % 29 200307047, and before each dFBS concentration was further reduced, credible cell growth was established, 8G3 was not further adapted to 1% dFBS. Example 8 Yield and metabolism of GS transfectant 3E10 in suspension culture of Hemophilus. The GS transfectant 3E10 adapted to the growth of bloodless tadpoles in Example 7 was cultured in suspension culture in HM9 medium without blood tadpoles or amidamine, 35.5 to 37 ° C. For the GS transfectant 3E10 cell line, cultured under the same conditions in HM9 medium containing amidineamine, it had a specific rate of EPO synthesis that was 2 to 3 times faster than that of the non-transfected R223 cell line (Table 7). Table 7 Growth and yield of GS transfectants 3E10 cells in a bloodless maggot suspension culture. Maximum number of cells / ml EPO q_EP〇xlO · 6 EU / ml EU / 106 / h 3E10 (Amiamine-free HM9 medium) 1.1-1.2 9731-14234 40-67 R223 (HM9 medium plus 6 mM Amine-amine) 1.0-1.6 3075-5818 13-24 GS-transfected cells with high EPO production accompanied by a decrease in metabolic ammonia release, unlike bottleless cultures that typically produce 5 mM ammonia in a 6 mM ammonium ammonia-containing HM9 medium without transfection In the R223 cell line, GS transfectant 3E10 produced only 1.8 mM ammonia in amidine-free ammonium HM9 medium (Table 8). 30 200307047 Table 8 Reduced ammonia synthesis of GS transfectant 3E10. Maximum number of cells / ml ammonia q ammonia x 10-6 mM μMole / 106 cells / h R223 (HM9 medium containing 6 mM ammonia ammonia 0.9 4.8 30 amine) 1.2 5.0 23 1.0 5.5 27 3E10 HM9 medium) 0.8 1.8 11 Example 9 The product quality was analyzed. EPO lines produced from GS transfectants 3E10 in flask cultures were immunopurified and analyzed for their glycoform distribution. Figure 4 shows the isoelectric point focusing (IEF) colloid analysis of EPO, which was obtained at the highest cell concentration and collected at A 3E10 cell culture of amidine-free ammonium HM9 medium, as described in Example 8, includes comparable samples from non-transfected R223 cell lines grown in HM9 medium. EP0 produced from GS transfectant 3E10 (compared to the non-transfected R223 cell line) presents more acidic isomers. Analyzing the IEF gel can quantify the distribution of isomers and calculate the relative activity of the isomers in theory ( IRA). 31 200307047 Table 9 Analysis of the relative activity of EPO isomers from the non-transfected R223 cell line and GS transfectant 3E10. Bands are more alkaline than active * 1 2 3 4 5 6 0.071 7 0.194 8 0.273 9 0.373 10 0.658 11 0.989 12 0.999 13 1.000 14 0.796 15 16 More acidic 17 Other minor bright bands R223 without transfection Isomer percentage activity (percent X specific activity) 5.2 5.2 6.3 11.4 10.5 9.2 0.7 8.8 1.7 8.7 2.4 5.5 2.1 4.4 2.9 4.3 4.3 4.5 4.5 1.4 1.4 0.2 0.2 14.4 Total = 100% Total = IRA = 20.2 GS conversion Dye 3E10 percent isomer activity (percent X specific activity 2.5 3.2 3.6 6.0 6.2 7.5 0.5 10.5 2.0 11.6 3.2 9.3 3.5 9.5 6.3 7.8 7.7 6.1 6.1 5.3 5.3 1.7 1.4 0.5 8.7 Total = 100% Total = IRA = 36.0
*活性數據來自EP-A 0 428 267 數據指出了 GS轉染子3E10有相當大的產物品質提昇 (表9及10)，IRA幾乎兩倍高於控制組無轉染的R223細胞 系培養。 也決定了來自GS轉染子3E10之EPO的假設N-多醣電 32 200307047 荷“Z”，表10沒有決定無轉染的R223細胞系之“Z”，然而 ，GS轉染子3E10 (273於最大細胞濃度及265於收集時）之 “Z”値超過瓶培養（183-228)無轉染R223細胞系的値。“Z”係 根據 Hermentin 等人於 Glycobiology，1996，6，217-230 決 定；將個別％-唾液酸化的（sialyated)異構物之部分乘以對應 的該異構物負電荷，可得Z，視其是否缺乏唾液酸基/中性 、單涎、三-、四或五涎而定。該乘積之數學總合術語係Z 。Z數字係關於活體內給予的治療上糖蛋白質（Hermentin， ibd.)之淸除速率。 表10 來自無轉染R223細胞系及GS轉染子3E10之EP0產物品 質分析。 效價 EU / mL 異構物相對 活性 'Z' R223親本咖611^榖氨醯胺之1^9 培養基） 高峰 收集 3518 25.2 not done 6150 20.2 not done 3E10憮榖氨醯胺之HM9培養基)高峰 收集 3921 43.6 273 6862 36.0 265 實施例10 在6公升發酵槽無血淸懸浮培養之GS轉染子3E10的產量 、代謝及產物品質。 GS轉染子3E10係生長在分批培養於氣舉式（airlift)發 酵槽，培養基係無榖氨醯胺HM9，不含血淸，每一個培養 有濃縮的養分供應，其含有胺基酸及葡萄糖，設計使主要 33 200307047 消耗之養分維持在足夠濃度。結果顯示在表11，也包括無 轉染的親本R223發酵數據，用於比較。 GS轉染子呈現延長的生存能力，所以培養期間增加， 產物合成的比速率相同於無轉染的親本細胞系，但較長培 養壽命造成最大產物濃度增加，氨累積至少低於GS轉染 子四倍，GS轉染子產物品質（以IRA測量）改善。 GS轉染子3E10細胞在無血淸懸浮培養的生長及產量，產生 的EPO產物品質係以IRA定量。 發酵條件 最大 最大 收集日 氨 收集 整體 收集時 細胞數 CCH*/ mg/L 效價 q epo 的IRA /ml ml EU/IO6 X ΙΟ'6 χ ΙΟ'6 EU/ml 細胞數/h 無榖氨醯胺 1.03 306 21 41 20936 69 43.6 無榖氨醯胺 U3 360 24 20 20150 52 42.6 無榖氨醯胺 0.94 319 24 15 18680 49 48.5 以標準方法 用榖氨醯胺 之親本 R223 l.ll 216 13 165 12225 58 28 1± 表 *累積細胞時數，細胞濃度的時間積分[Renard等人，1988 ，Biotechnology Letters 10 (9) : 1-96]。 實施例11 在無血淸懸浮培養Iscove氏爲基的培養基之GS轉染子的產 量及代謝 實施例7中適應於無血淸生長的GS轉染3E10係培 34 200307047 養在震盪瓶，在無血淸無榖氨醯胺的Iscove氏培養基，親 本細胞系R223係對應生長在相等的添加榖氨醯胺之培養基 〇 所用的培養基係經Iscove氏修飾之Dulbecco氏培養基 ，含有Iscove氏的添加物（牛血淸白蛋白0.4 g/L、人類全 (holo)運鐵蛋白30 mg/L)、重組人類胰島素10 mg/L、Lutrol F68 1 g/L及乙醇胺60 kL/L。用於培養R223細胞系的培養 基含有4mM榖氨醯胺，但用於GS轉染的細胞系則不加榖 氨醯胺，改以4 mM榖氨酸鈉加4 mM天冬醯胺。 GS轉染的細胞系3E10之產物合成比速率大約50%高 於無轉染的親本系R223(表12)，轉染的細胞系3E10之氨 生成比速率低七倍（表13)。 產物係從這些個別培養純化而來，並在等電點聚焦法 (IEF)膠體上分析，此分析結果顯示於圖5,其顯示染色後 的IEF膠。IEF膠在pH 2.5到6.5變性條件下跑，並以 Coomassie Blue 染色0 圖5解說 第1、3、6及7行 空白 第2及7行 pi標記 第4行 從在含有榖氨醯胺的Iscove氏培養 基中生長的R223免疫親和純化的產 物 第5行 免疫親和純化的產物，其係從從在無 榖氨醯胺Iscoves培養基中生長的 GS轉染的R223細胞系之轉殖3E10 35 200307047 來自R223的產物至少有13條可見的亮帶，分布在整 條膠體上；來自GS轉染的細胞系3E10生成的產物，較鹼 的亮帶（膠體上方）強度比從細胞系R223產物亮帶弱，但細 胞系3E10較酸的亮帶（膠體下方）則強度增加。有至少一條 額外的酸性亮帶可在GS轉染子3E10產物中可偵測到，但 親本系R223產物偵測不到，這顯示來自GS轉染的3E10系 之產物唾液酸化（sialylation)增加。 表12 在無血淸懸浮培養於Iscove氏培養基中的GS轉染子3E10 細胞之生長及產量 最大存活細胞 數/mL EPO EU/mL EU/106細胞數 /h 生長在無榖氨醯胺 Iscove氏培養基之 3E10 0.55 1340 11.5 生長在含有榖氨醯胺 Iscove氏培養基之 R223 0.72 993 7.3 表13 在無血淸懸浮培養於Iscove氏培養基中GS轉染子3E10細 胞之氨生成減少 最大存活細胞 數/mL 氛mM q氨 η莫耳/106細胞 數/h 生長在無榖氨醯胺 Iscove氏培養基之 3E10 0.55 0.61 2.6 生長在含有榖氨醯胺 Iscove氏培養基之 R223 0.72 Γ2.39 17.4 36 200307047 【圖式簡單說明】 圖1顯示細胞系R223懸浮培養於無血淸培養基之適應 作用的細胞濃度圖，經由重複連續的次培養。 圖2顯示HT1080細胞懸浮培養於無血淸培養基之適應 作用槪要。 圖3顯示細胞系HT1080懸浮培養於無血淸培養基之適 應作用的細胞濃度圖，經由重複連續的次培養。 圖4顯示從GS轉染子3E10及從無轉染的R223細胞 系中免疫純化的EPO之IEF分析，其係生長在產業高密度 生長培養基。行2 :收集的（收集）3E10 ;行3 : 3E10峰；行 4 :無轉染的R223細胞系峰；行5 :收集的無轉染的R223 細胞系。 圖5顯示R223細胞系之GS-R223轉染子3E10及從無 轉染的R223細胞系免疫純化的EPO之IEF分析，生長在 傳統的Iscove氏培養基。行4 :自添加榖氨醯胺的Iscove 氏收集的無轉染的R223細胞系；行5 :自無榖氨醯胺 Iscove氏收集的GS-223細胞系3E10。 37* Activity data are from EP-A 0 428 267. The data indicate that the GS transfectant 3E10 has a considerable improvement in product quality (Tables 9 and 10), and the IRA is almost twice higher than that of the control group without transfection in the R223 cell line culture. The hypothesized N-polysaccharide charge of EPO from GS transfectant 3E10 32 200307047 was also determined. Table 10 does not determine the "Z" of R223 cell line without transfection. However, GS transfectant 3E10 (273 in The maximum cell concentration and 265 at the time of collection "Z" Z exceeded that of bottle culture (183-228) without transfection of the R223 cell line. "Z" is determined according to Hermentin et al. In Glycobiology, 1996, 6, 217-230; multiplying a portion of an individual% -sialyated isomer by the corresponding negative charge of the isomer to obtain Z, Depending on whether it lacks sialic acid / neutral, mono-, tri-, tetra-, or penta-sial. The mathematical term for this product is Z. The Z number refers to the rate of elimination of therapeutic glycoproteins (Hermentin, ibd.) Administered in vivo. Table 10 Qualitative analysis of EP0 products from untransfected R223 cell line and GS transfectant 3E10. Relative potency EU / mL Isomers Relative activity 'Z' R223 Parental coffee 611 ^ ammonamine 1 ^ 9 medium) Peak collection 3518 25.2 not done 6150 20.2 not done 3E10 amine ammonium HM 9 medium) peak Collection of 3921 43.6 273 6862 36.0 265 Example 10 Yield, metabolism and product quality of GS transfectants 3E10 suspension cultured in a 6-liter fermentation tank without blood mash. GS transfectants 3E10 are grown in batches and cultured in airlift fermentation tanks. The culture medium is amidine-free ammonia HM9, which does not contain blood salamander. Each culture has a concentrated nutrient supply, which contains amino acids and Glucose, designed to maintain sufficient concentrations of nutrients consumed by the main 33 200307047. The results are shown in Table 11 and also include parental R223 fermentation data without transfection for comparison. GS transfectants exhibit prolonged viability, so the culture period is increased, and the specific rate of product synthesis is the same as that of parental cell lines without transfection, but the maximum product concentration is increased due to longer culture life, and ammonia accumulation is at least lower than GS transfection The quality of the GS transfectant product (measured by IRA) was improved four-fold. The growth and yield of GS transfectants 3E10 cells in suspension cultures of A. auratus, and the quality of EPO products produced were quantified by IRA. Fermentation conditions maximum collection day ammonia collection number of cells at the time of overall collection CCH * / mg / L titer q epo IRA / ml ml EU / IO6 X ΙΟ'6 χ ΙΟ'6 EU / ml number of cells / h Amine 1.03 306 21 41 20936 69 43.6 Ammonia-free amine U3 360 24 20 20150 52 42.6 Ammonia-free amine 0.94 319 24 15 18680 49 48.5 The standard method of using amidine amine R223 l.ll 216 13 165 12225 58 28 1 ± Table * Cumulative cell hours, time integral of cell concentration [Renard et al., 1988, Biotechnology Letters 10 (9): 1-96]. Example 11 Yield and Metabolism of GS Transfectants in Iscove's Medium Based on Suspension Culture of Hematogenous Pupae in Example 7 Isamove's medium for amidoxamine, the parental cell line R223 is grown on an equivalent medium supplemented with amidoxamine. The medium used is Dulbecco's medium modified with Iscove's, containing Iscove's supplement (bovine blood)淸 Albumin 0.4 g / L, human holo transferrin 30 mg / L), recombinant human insulin 10 mg / L, Lutrol F68 1 g / L and ethanolamine 60 kL / L. The medium used to culture the R223 cell line contained 4 mM amidoxamine, but the cell line used for GS transfection did not add osmamine, and changed to 4 mM sodium tyrosine plus 4 mM asparagine. The GS transfected cell line 3E10 had a product specific rate of about 50% higher than that of the parental line R223 without transfection (Table 12), and the transfected cell line 3E10 had a seven-fold lower specific rate of ammonia production (Table 13). The products were purified from these individual cultures and analyzed on an isoelectric focusing method (IEF) colloid. The results of this analysis are shown in Figure 5, which shows the stained IEF gel. IEF gels were run under denaturing pH 2.5 to 6.5 and stained with Coomassie Blue. Figure 5 illustrates lines 1, 3, 6, and 7 blank. Lines 2 and 7 pi labeled. Line 4 from Iscove containing ammonium Immunoaffinity purified product of R223 grown in Fructus's medium Line 5 Immunoaffinity purified product, which was transfected from R223 cell line transfected with GS grown in aminamine-free Iscoves medium 3E10 35 200307047 from R223 The product has at least 13 visible bright bands distributed throughout the colloid; the product generated from the GS transfected cell line 3E10 is weaker than the alkaline bright band (above the colloid) than the bright band from the cell line R223 product. However, the cell line 3E10 was more intense than the acidic bright band (below the colloid). At least one additional acidic bright band can be detected in the GS transfectant 3E10 product, but the parental line R223 product is not detected, which shows increased sialylation of the GS transfected 3E10 line product . Table 12 Growth and yield of GS transfectants 3E10 cells cultured in suspension-free culture in Iscove's medium. Maximum number of viable cells / mL EPO EU / mL EU / 106 cells / h. 3E10 0.55 1340 11.5 Grow on R223 0.72 993 7.3 containing ampicillamine Iscove's medium Table 13 Ammonia production of GS transfectants 3E10 cells in suspension-free culture in Iscove's medium Reduces the maximum number of viable cells / mL q Ammonia mol / 106 cells / h 3E10 0.55 0.61 2.6 grown on ammonamine-free Iscove's culture medium R223 0.72 Γ2.39 17.4 36 200307047 grown on melamine-containing Iscove culture medium Figure 1 shows the cell concentration map of the adaptive effect of cell line R223 in suspension culture on blood-free maggot medium, after repeated continuous subcultures. Figure 2 shows the essentials of the adaptation of HT1080 cells in suspension culture in blood-free medium. Figure 3 shows the cell concentration map of the adaptive effect of the cell line HT1080 in suspension culture in a blood-free maggot medium, after repeated continuous subcultures. Figure 4 shows IEF analysis of immunopurified EPO from GS transfectants 3E10 and non-transfected R223 cell lines, which were grown in industrial high-density growth medium. Row 2: collected (collected) 3E10; row 3: 3E10 peak; row 4: peak of R223 cell line without transfection; row 5: collected R223 cell line without transfection. Figure 5 shows IEF analysis of GS-R223 transfectant 3E10 of R223 cell line and EPO immunopurified from non-transfected R223 cell line, grown in conventional Iscove's medium. Row 4: Non-transfected R223 cell line collected from Iscove's with amidine. Addition of GS-223 cell line 3E10 from Iscove's without amidine. 37