JP4127852B2 - Activated linker and its production and purification method - Google Patents

Description

少なくとも１つの反応性ヒドロキシル基を有する任意の化合物は本発明方法に使用できると考えられる。“反応性ヒドロキシル基”なる用語は、塩基を還元し得る水素を含む−ＯＨ基を意味しており、この−ＯＨ基は、カルボニル、ニトリルまたはスルホン基を有する試薬と反応してマイケルアクセプターを形成し得る酸素を遊離する。本発明方法に有用なヒドロキシル含有化合物としては例えば、アルコール類、ポリマー類、ヌクレオチド類、ヌクレオシド類、オリゴヌクレオチド類、ペプチド類、制御ポアガラス及びクロマトグラフ物質など、数例を挙げることができる。有用なポリマーは例えば、ポリエチレングリコール（“ＰＥＧ”）、ポリプロピレングリコール（“ＰＰＧ”）、ポリオキシエチル化グリセロール（“ＰＯＧ”）及びその他のポリオキシエチル化ポリオール類、ポリビニルアルコール（“ＰＶＡ”）及びその他のポリアルキレンオキシド類、ポリオキシエチル化ソルビトールまたはポリオキシエチル化グルコースである。
ポリマーは、少なくとも１つの活性ヒドロキシ部分を有している限り、ホモポリマー、ランダムもしくはブロックコポリマー、ターポリマーのいずれでもよく、直鎖状または分枝状、置換または未置換のいずれでもよい。重合部分はいかなる鎖長または分子量でもよいが、これらの特性値が生物学的特性に影響を与えることもある。医薬用途におけるクリアランス速度を低下させるために有用なポリマーの平均分子量は、２，０００〜３５，０００ダルトンの範囲である。特に有用なＰＥＧの分子量は３．４ｋＤａ、５ｋＤａ、１０ｋＤａ及び２０ｋＤａである。
本文中で使用された“ＰＥＧ”なる用語は、エチレングリコールのいくつかの縮合ポリマーのいずれかを意味する。ＰＥＧはまた、ポリオキシエチレン、ポリエチレンオキシド、ポリグリコール及びポリエーテルグリコールとして知られている。ＰＥＧはまた、エチレンオキシドと多くの別のモノマーとのコポリマーとして製造され得る。ＰＥＧの末端基は、ビニルスルホン部分、アルデヒド部分、メトキシ基またはその他の種類の反応性もしくは非反応性の末端基を含むように種々の方法で誘導体化され得る。医薬用またはバイオ技術用の多くの用途の場合、ビニルスルホン以外は実質的に未置換の実質的に線状の直鎖状ビニルスルホン活性化ＰＥＧが使用されるであろう。
このような化合物のヒドロキシル基は、ヒドロキシル基をマイケルアクセプターに変換し得る反応性カルボニル、ニトリルまたはスルホン基を有する試薬と反応する。反応性カルボニル、ニトリルまたはスルホン基を有する当業者に公知のいかなる試薬も本発明方法に使用できると考えられる。“反応性カルボニル、ニトリルまたはスルホン”なる用語は、２つの炭素基が結合し、約９以下のｐＨでチオール特異的結合に好適な反応性部位をカルボニル、ニトリルまたはスルホン基の第二の炭素に有しているカルボニル、ニトリルまたはスルホン基を意味している。
本発明の好ましい実施態様では、本発明のポリマー誘導体が活性スルホン部分を有している。“活性スルホン”なる用語は、２つの炭素原子が結合し、ｐＨ約以下でチオール特異的結合に好適な反応性部位をスルホン基の第二の炭素に有しているスルホン基を意味している。活性スルホンの非限定例は、ビニルスルホン及び活性化エチルスルホンである。スルホン活性化ポリマーは、約９以下のｐＨで第二の炭素のチオール特異的反応性が維持されている限り、更に別の置換基で置換されていてもよい。
このような試薬を添加すると、ヒドロキシル基が反応性マイケルアクセプターに変換される。“マイケルアクセプター”または“マイケル様アクセプター”なる用語は、互換的に使用され、マイケル付加に感受性の官能基を意味する。“マイケル付加”では、電気陰性原子を有するパイシステムに隣接の求電子性中心に対して求核性攻撃が生じる。電気陰性原子を有するパイシステムの例は、スルホキシド、スルホニル、カルボニル及び複素環芳香族である。求核剤が求電子性中心に付加される。マイケルアクセプターは式：

〔式中、Ｅは電気陰性原子を表す〕によって表される。４位に付加が生じて、式：

〔式中、Ｎ_uは４位の原子に結合された求核剤を意味する〕が形成される。マイケルアクセプター官能基の非限定例は、マレイミド及びビニルスルホンである。ダンベルを形成するために使用される活性化リンカーはマイケルアクセプターのビニルスルホン種を含んでいてもよいが、必ずしも含んでいなくてもよい。
本発明の活性化リンカーは２つ以上の反応性基を有し得る。誘導体は一官能価、二官能価または多官能価のいずれでもよい。少なくとも１つの活性スルホン部分が存在する限り、反応性基は同じでもよく（ホモ官能性）、または異なっていてもよい（ヘテロ官能性）。特に有用な２つのホモ官能性誘導体はＰＥＧ−ビス−クロロスルホン及びＰＥＧ−ビス−ビニルスルホンである。
本発明方法によって製造されたリンカーはチオール選択的結合反応に好適な安定で単離可能なリンカーである。例えば、ＰＥＧクロロエチルスルホンは約７以下のｐＨで水に安定であるが、少なくともｐＨ約９までの塩基性ｐＨ条件でチオール選択的結合に有利に使用できる。ｐＨが約９よりも高いとき、チオール選択性が低下し、スルホン部分はアミノ基に対して多少反応性になる。チオールとの反応によって形成された連鎖はまた加水分解的に安定である。
本文中で使用された“加水分解的に安定”なる用語は、ポリマーとスルホン部分との連鎖及びコンジュゲーション後のスルホン−チオール連鎖が、約１１よりも低いｐＨでは少なくとも３日間は水と反応しないことを意味する。有意な速度の加水分解が生じるとき、ポリマーと生物学的活性分子のチオールとの反応が生じる前にポリマーが失活することがあるので、加水分解的に安定であることは望ましい要件である。
活性化リンカーの普遍的な精製方法は以下の段階を含む。
（１）活性化リンカーを合成する、
（２）活性化リンカーをＨＩＣ樹脂に保持させる、次いで、
（３）種々のリンカー形態をサイズ及び／または末端基官能性に基づいて分離し得る条件、例えば逆直線状塩勾配を用いてＨＩＣ樹脂からリンカーを溶出させることによって所望のリンカー形態を単離する。
あるいは、“ワンステップ”合成手順の濃縮／透析濾過段階の直前にＨＩＣ精製段階を組込んでもよい。
本発明に使用できると考えられるＨＩＣは、メタクリル、アガロースまたはビニル主鎖を非限定例とする親水性ビニルコポリマー主鎖をもつ樹脂の表面に疎水性側基（例えば、オクチル、ブチル、フェニルまたはエーテル）が局在しているポリマー樹脂を使用する。シリカベースの主鎖も使用できる。好ましいＨＩＣカラムはメタクリル主鎖にブチルまたはフェニル配位子を有している。
上述のように、ＨＩＣ精製したスルホン活性化リンカーは、“反応性チオール部分”を有する生物学的活性分子に付着して、生物学的に活性のコンジュゲートを形成し得る。“反応性チオール部分”なる用語は、本文中に記載のような活性化リンカーと反応し得るスルフヒドリル（−ＳＨ）基を意味する。
本発明に使用できると考えられる生物学的活性分子の非限定例は、医薬、ビタミン、栄養素、核酸、アミノ酸、ポリペプチド、酵素補因子、ステロイド、糖質、ヘパリンのような有機種、金属含有物質、レセプターアゴニスト、レセプターアンタゴニスト、結合タンパク質、レセプターまたはレセプターの部分、細胞外マトリックスタンパク質、細胞表面分子、抗原、ハプテン及びキレート化剤である。
一般に、本発明に有用なＴＮＦｂｐは、ヒトＴＮＦｂｐの配列またはその先端欠失変異体（ｔｒｕｎｃａｔｅｄ ｖａｒｉａｎｔｓ）またはムテイン（ｍｕｔｅｉｎｓ）の配列を有している。３０ｋＤａのＴＮＦｂｐと命名された１つのＴＮＦｂｐは、ｐ５５ ＴＮＦレセプターの細胞外部分である。ｉｎ ｖｉｖｏではレセプターの細胞外部分が脱落し、ＴＮＦに結合する３０ｋＤａのグリコシル化タンパク質として血流中を循環する。このＴＮＦｂｐの精製及びそのアミノ酸と核酸との配列に関しては、ＥＰ３９３４３８及びＥＰ４２２３３９に記載されている。これらの特許の記載内容は参照によって本発明に含まれるものとする。これらの文献はまた、３０ｋＤａ ＴＮＦｂｐのグリコシル化形態及び脱グリコシル化形態の組換え産生を教示している。この阻害物質の脱グリコシル化形態の実際の分子量はほぼ１８ｋＤａであるが、“３０ｋＤａ ＴＮＦ阻害物質”なる用語は、グリコシル化形態と脱グリコシル化形態との双方を包含する。
４０ｋＤａ ＴＮＦｂｐと呼ばれる別のＴＮＦ阻害物質の精製及びアミノ酸と核酸配列との配列もＥＰ３９３４３８及びＥＰ４２２３３９に記載されている。この阻害物質はその天然産生形態ではｐ７５ ＴＮＦレセプターのグリコシル化細胞外部分である。脱グリコシル化形態の分子量は４０ｋＤａではないが、このＴＮＦｂｐのグリコシル化形態と脱グリコシル化との双方を“４０ｋＤａ ＴＮＦ阻害物質”と呼ぶ。これらの参考文献はまた、４０ｋＤａ ＴＮＦｂｐのグリコシル化形態と脱グリコシル化形態との組換え産生を教示している。
反応性チオール基は必要なときに生物学的活性分子に導入され得る。チオール基はまた、チオール−スルホンの連結が所望の活性を実質的に破壊しない限り、不活性分子に導入されて生物学的活性分子を形成し得る。
反応性チオール基は当業界で公知の化学的手段によって導入され得る。化学的修飾はポリペプチドまたは非ペプチド分子と共に使用でき、チオールを単独でまたは例えばシステイン残基のようなより大きい基の一部分として分子に導入する。チオールの化学的導入の一例がＪｕｅ，Ｒら，Ｂｉｏｃｈｅｍｉｓｔｒｙ，１７：５３９９−５４０６（１９７８）に記載されている。また、システインを例えばＤＴＴで化学的に還元することによってポリペプチド中に遊離システインを生じさせてもよい。
システインを含むムテインを産生するためには、必須でないアミノ酸をシステインで置換してもよくまたはポリペプチドにシステイン残基を付加してもよい。修飾前の天然型タンパク質には存在しなかった位置にアミノ酸残基を含むように修飾されたポリペプチドを“ムテイン（ｍｕｔｅｉｎ）”と命名する。非天然型システインの導入可能部位は、ポリペプチドのグリコシル化部位及びＮ末端またはＣ末端である。リシン残基はしばしば活性コンホメーションのタンパク質の表面に見出されるので、リシンからシステインへの突然変異も好適である。当業者は更に、可能な突然変異部位を選択するために、ポリペプチドの結合部位または活性部位に関する公知のいかなる情報も利用できる。
当業者はまた、システインを含むムテインを製造するために公知の組換えＤＮＡ技術を使用し得る。標準的な部位特異的突然変異誘発によって天然型ポリペプチドをコードする核酸をムテインをコードするように改変し得る。標準突然変異誘発技術の例は、Ｋｕｎｋｅｌ，Ｔ．Ａ．，Ｐｒｏｃ．Ｎａｔ．Ａｃａｄ．Ｓｃｉ．，８２：４８８−４９２（１９８５）及びＫｕｎｋｅｌ，Ｔ．Ａ．ら，Ｍｅｔｈｏｄｓ Ｅｎｚｙｍｏｌ．，１５４：６７−３８２（１９８７）に記載されている。双方の文献の記載内容は参照によって本発明に含まれるものとする。あるいは、当業界で公知の技術によってムテインをコードする核酸を化学的に合成し得る。ＤＮＡ合成機を使用することができ、これは例えばＡｐｐｌｉｅｄ Ｂｉｏｓｙｓｔｅｍｓ（Ｆｏｓｔｅｒ Ｃｉｔｙ，ＣＡ）から入手し得る。所望のムテインをコードする核酸を、動物、昆虫及び細菌系のような種々の発現系において発現させ得る。
反応性チオール基を有する生物学的活性分子とスルホン活性化ポリマーリンカーとの連鎖は共有結合であり、このようなコンジュゲートの普遍的な製造方法は以下の段階を含む。
（１）所望の生物学的活性分子を選択する、
（２）所望のスルホン活性化ポリマーリンカーを合成する、
（３）ＨＩＣクロマトグラフィーを用いてスルホン活性化ポリマーリンカーを精製する、
（４）ＨＩＣ精製したスルホン活性化ポリマーリンカーを生物学的活性分子と反応させる、
（５）当業界で公知のクロマトグラフィー技術を用いて反応生成物を単離する、次いで、
（６）形成されたコンジュゲートの生物学的活性を適切なバイオアッセイを用いて測定する。
既に製造されている３０ｋＤａ ＴＮＦｂｐのムテインとしては、ｃ１０５ ３０ｋＤａ ＴＮＦｂｐ；ｃ１ ３０ｋＤａ ＴＮＦｂｐ；ｃ１４ ３０ｋＤａ ＴＮＦｂｐ；ｃ１１１ ３０ｋＤａ ＴＮＦｂｐ；及びｃ１６１ ３０ｋＤａ ＴＮＦｂｐがある。“ｃ”はシステインを意味しており、“ｃ１０５”はシステインが１０５位に挿入されたことを意味する。これらのムテインは参照によって本発明に含まれる国際特許公開ＷＯ９２／１６２２１に記載の手順で製造され、アミノ酸番号も該特許に記載のアミノ酸配列に基づく。
ＧＢ２，２４６，５６９は、“３ドメイン”の３０ｋＤａ及び／または４０ｋＤａのＴＮＦｂｐ、即ち、第４ドメインが欠失したものを教示している。１９９５年７月９日出願の米国特許仮出願Ｎｏ．６０／０２１４４３は、“２．６ドメイン”の３０ｋＤａのＴＮＦｂｐ、即ち、第３ドメイン及び第４ドメインの一部が欠失したものを教示している。また、１９９６年１２月６日出願の米国特許仮出願Ｎｏ． は、“２．６ドメイン”の３０ｋＤａ及び／または４０ｋＤａのＴＮＦｂｐ、即ち、第３ドメイン及び第４ドメインの一部が欠失したものを教示している。
同じく既に製造されているＩＬ−１ｒａのムテインとしては、ｃ０ｓ１１６ ＩＬ−１ｒａ；ｃ８４ｓ１１６ ＩＬ−１ｒａ；ｃ８ｓ１１６ ＩＬ−１ｒａ；ｃ９ｓ１１６ ＩＬ−１ｒａ及びｃ１４１ｓ１１６ ＩＬ−１ｒａがある。ＩＬ−１ｒａムテインの製造及び精製も国際特許公開ＷＯ９２／１６２２１に開示されている。残基の番号はこの公開出願に記載された配列に基づく。“０”はアミノ酸のＮ末端に対する付加を表す。“ｃ”はシステイン、“ｓ”はセリンを表す。例えば、“ｃ０ｓ１１６”は、システインがＮ末端に付加され、セリンが１１６位に挿入されたことを意味する。天然型のＩＬ−１ｒａは６６、６９、１１６及び１２２位に遊離システイン残基を有している。
本発明の多くのコンジュゲートまたは化合物（まとめて“コンジュゲート”と呼ぶ）を含有する医薬組成物を調製し得る。これらのコンジュゲートは、本発明の医薬組成物を形成するために医薬として許容される担体中に存在し得る。本文中で使用された“医薬として許容される担体”なる用語は、有効成分または組成物の投与を受ける患者に副作用を生じない無毒性で通常は不活性の有効成分用ビヒクルを意味する。適当なビヒクルまたは担体は、標準医薬文献、例えば、“Ｒｅｍｉｎｇｔｏｎ’ｓ Ｐｈａｒｍａｃｅｕｔｉｃａｌ Ｓｃｉｅｎｃｅｓ”，１６ｔｈ Ｅｄ．，Ｍａｃｋ Ｐｕｂｌｉｓｈｉｎｎｇ Ｃｏ．，Ｅａｓｔｏｎ，ＰＡ（１９８０）に見出される。この文献の記載内容は参照によって本発明に含まれるものとする。このような担体は例えば、炭酸塩バッファ、燐酸塩バッファ、リンゲル液及び生理的食塩水のような水溶液である。更に、担体は、製剤のｐＨ、モル浸透圧濃度、粘度、清澄度、色、無菌性、安定性、溶解速度または臭味を矯正または維持するための医薬として許容される他の賦形剤を含有し得る。
医薬組成物は、例えば試薬の単なる混合のような当業界で公知の方法によって調製できる。医薬担体の選択及び組成物の適正な調製が予定の用途及び投与形態に依存することは当業者に理解されよう。
１つの実施態様によれば、担体及びコンジュゲートが生理的親和性の徐放性製剤を構成する。このような担体中の一次溶媒は水性または非水性のいずれでもよい。更に、担体は、コンジュゲートの安定性、溶解、放出または吸収の速度を矯正または維持するために生理学的に許容される他の賦形剤を更に含有し得る。このような賦形剤は、一回投与量または多数回投与量の形態の非経口投与剤形の薬剤を調製するために常用されている慣用の物質である。
医薬組成物を調製した後、溶液、懸濁液、ゲル、エマルジョン、固体、または、脱水もしくは凍結乾燥された粉末のような無菌バイアル中で医薬組成物を保存し得る。このような製剤は、そのままで使用できる形態で保存されてもよく、または、投与直前に再構成を要する形態で保存されてもよい。このような製剤の好ましい保存温度は少なくとも４℃以下、好ましくは−７０℃である。また、コンジュゲートを含有するこのような製剤は生理的ｐＨまたはこれに近いｐＨで保存及び投与されるのが好ましい。現在では、高ｐＨ（即ち８を上回る値）または低ｐＨ（即ち５未満の値）での製剤の投与は望ましくないと考えられている。
コンジュゲートを含有する製剤を全身デリバリーの目的で投与する方法には、皮下、筋肉内、静脈内、経口、鼻孔内などの経路、または膣内もしくは直腸内投与する座薬剤がある。コンジュゲートを含有する製剤を局部デリバリーの目的で投与する方法には、関節内、気管内、点滴、または、気道内吸入がある。更に、適当な製剤中のコンジュゲートの経口投与またはデバイスによって消化管の特定部分にコンジュゲートを投与することも望ましい。
経口投与のためには、コンジュゲートをカプセル封入する。固体剤形の調製に使用される慣用の医薬として許容される担体を添加するかまたは非添加でカプセル封入コンジュゲートを製剤化し得る。好ましくは、生物学的利用態が最大であり且つ浸透以前の（ｐｒｅ−ｓｙｓｔｅｍｉｃ）分解が最小となる様な胃腸管の箇所に製剤の有効部分が放出されるようにカプセルを設計する。コンジュゲートの吸収を促進する追加の賦形剤を混入してもよい。希釈剤、香味剤、低融点ワックス、植物油、潤滑剤、懸濁剤、錠剤崩壊剤及び結合剤も使用し得る。
投与形態にかかわりなく、個々の投薬用量はほぼ患者の体重に基づいて計算する。適正な投薬用量を決定する他の要因は、治療または予防すべき疾病または障害の種類、投与経路、患者の年齢、性別及び病状である。いくつかの実施態様では、投薬用量及び投与形態は、患者の血流中でコンジュゲートが予め選択された濃度範囲になるように計画される。例えば、血漿１ｍｌあたり０．０１ｎｇ未満の濃度のＴＮＦ阻害物質またはＩＬ−１阻害物質の循環を維持しても有効な組成物にはなり得ないと考えられる。逆に、１ｍｌあたり１０μｇを上回るレベルの循環を長期間維持すると望ましくない副作用を生じるであろう。上記製剤の各々を用いる治療の適正投薬用量を決定するために必要な計算をより精密に行う方法は平均的な当業者の常套的な作業であり、特に本文中に開示された投薬用量に関する情報及びアッセイを参考にして特別な実験を要せずに当業者が常套的に実施し得る作業の範囲に包含される。これらの投薬用量は、確立した投薬用量決定アッセイを適当な容量−応答データと共に使用することによって確認され得る。
本文に記載のコンジュゲート製剤が獣医薬及びヒト医薬の双方の用途に使用できること、また、“患者”なる用語を限定的に解釈してはならないことに留意されたい。獣医薬として使用する場合にも、投薬用量の範囲は上記と同様である。
本発明のコンジュゲートは、ｉｎ ｖｉｔｒｏ診断アッセイ、医薬組成物の製造、などを非限定例とする種々の用途に使用できる。ＴＮＦｂｐ、ＩＬ−１ｒａを含有するコンジュゲートは単独形態または複数のコンジュゲートを組合せた形態で、成人呼吸障害症候群、肺線維症、関節炎、乾癬、骨髄性及びその他の白血病、虚血／再潅流、喘息、悪液質／無食欲、糖尿病、敗血症ショック、炎症性腸疾患、多発性硬化症、卒中、クローン病、移植拒絶、出血性外傷のような疾病の治療に使用され得る。コンジュゲートはまた、アフィニティクロマトグラフィー（分配）のような多くのｉｎ ｖｉｔｒｏの使用目的及び当業者に周知の診断目的で使用され得る。例えば、このようなコンジュゲートを固体表面に付着させ、活性化リンカーに結合したレセプターに特異的に結合するリガンドの精製または検出に使用できる。
以下の実施例は本発明の種々の特徴をより詳細に説明する。
実施例１
この実施例は、“ワンステップ”合成手順を用いる活性化リンカー２０ｋＤａ ＰＥＧｂｖの製造を記載する。
２０ｋＤａ ＰＥＧｂｖの“ワンステップ”合成
ポリエチレングリコール（ＰＥＧ−ＤＩＯＬ^TM）は、Ｓｈｅａｒｗａｔｅｒ Ｐｏｌｙｍｅｒｓ Ｉｎｃ．（Ｂｉｒｍｉｎｇｈａｍ，ＡＬ）から購入した。ジビニルスルホン、１Ｍのカリウムｔ−ブトキシド（テトラヒドロフラン（ＴＨＦ）中）、トルエン（無水物、ｓｕｒｅ ｓｅａｌ）、ジ−ｉ−プロピルエーテル（９９＋％、２５ｐｐｍのブチル化ヒドロキシトルエン（ＢＨＴ）で阻害した）、及びテトラヒドロフラン（０．０２５％のＢＨＴを阻害物質として含有する無水物、ｓｕｒｅ ｓｅａｌ）は、Ａｌｄｒｉｃｈ Ｃｈｅｍｉｃａｌ Ｃｏｍｐａｎｙ（Ｍｉｌｗａｕｋｅｅ，Ｗｉｓｃｏｎｓｉｎ）から得られた。ジクロロメタン（Ｏｍｎｉ Ｓｏｌｖｅｎｔ）及びジエチルエーテル（ＥＭ）はＶＷＲ Ｓｃｉｅｎｔｉｆｉｃ Ｃｏｍｐａｎｙ（Ｄｅｎｖｅｒ，Ｃｏｌｏｒａｄｏ）から購入した。無水硫酸ナトリウム粉末はＪ．Ｔ．Ｂａｋｅｒ（Ｐｈｉｌｌｉｐｓｂｕｒｇ，Ｎｅｗ Ｊｅｒｓｅｙ）から購入した。
２２リットル容の４つ口（１つの口には４５／５０、３つの口には２４／４０のジョイント）の丸底フラスコに、温度計、凝縮器、ゴム隔壁、機械的変速撹拌器を取付け、水の浸漬ヒーターを入れた水浴に浸漬させた。フラスコを正のアルゴン圧力下に維持した。凝縮器を２つ口の丸底フラスコに接続し、このフラスコから真空ラインを接続した。ポリエチレングリコール（２０ｋＤａ、３００ｇ、１５ミリモル）を計量し、漏斗を介して２２リットル容のフラスコに移入し、混合物を撹拌しながら３リットルの無水トルエンをカニューレから導入した。水浴の温度を４５℃に設定し、溶液が透明になるまで（〜１時間）撹拌を継続した。真空下（〜６０ｍｍのＨｇ）に溶媒を蒸発させて乾固した。固体残渣を９リットルの無水テトラヒドロフラン（カニューレから添加）に溶解した。水浴の温度を２５℃に設定した。溶液の温度が２５℃になるまで撹拌を継続した。ジビニルスルホン（６ｍｌ、５９．９７ミリモル）をシリンジから１５分間で添加した。混合物を１０分間撹拌した（７５ｒｐｍ）。カリウムｔ−ブトキシド（ＴＨＦ中に１Ｍ、３ｍｌ、３ミリモル）を滴下漏斗から２５分間で添加した。
塩基の添加後、反応混合物を２５℃で６時間撹拌した。次に水浴を４０℃に加熱し、ＴＨＦを蒸発させた（３５０ｍｍのＨｇ）。ジクロロメタンとテトラヒドロフランとの１：１混合物１リットルを固体残渣に添加し、透明溶液が形成されるまで撹拌した。溶液を２５℃に冷却し、急冷（−２０℃）した。撹拌を継続しながらジ−ｉ−プロピルエーテル（１×５リットル）を滴下漏斗から溶液に添加した。凝縮器をフラスコからはずした。濾過用圧力漏斗を可撓針でフラスコに接続し、沈殿物を可撓針から移入して真空下の濾過用漏斗で濾過した（１リットル、７０〜１００μ）。沈殿物を２つの分量に分け、２リットルのエルレンマイヤーフラスコに移入し、１リットル（各々）のｉ−プロピルエーテルと共に撹拌（磁気撹拌器）を３時間継続した。沈殿物を濾過し（２リットル、７０〜１００μ）、８００ｍｌのジエチルエーテルで洗浄し、乾燥器に入れて真空下（５ｍｍのＨｇ）で７２時間乾燥した。測定収量は２７８．６ｇであった（ポリエチレングリコールの重量を基準として９１．６％の回収率）。
生成物をエルレンマイヤーフラスコ（２リットル）に移入し、１リットルのジクロロメタン（Ｏｍｎｉ Ｓｏｌｖｅｎｔ）を添加した。生成物が完全に溶解した後、機械的撹拌器を取付けた５リットル容のドロップボトムフラスコに溶液（シロップ形態）を移入した。エルレンマイヤーフラスコを別の５００ｍｌのジクロロメタンで洗浄し、溶液をドロップボトムフラスコに添加した。次に、溶液を５分間撹拌（６５ｒｐｍ）した。飽和塩化ナトリウム（１リットル）を添加し、５分間撹拌（７５ｒｐｍ）した。２層の分離が明瞭になるまで混合物を静置した。有機層を取出し、別の容器に保存した。このプロセスを３回繰返し、全部の有機層を一緒にした。
次に、溶液を２つの分量に分け（各３リットル）、２つのエルレンマイヤーフラスコ（４リットル）に移入し、５００ｇの無水硫酸ナトリウム粉末（Ｊ．Ｔ．Ｂａｋｅｒ，Ｐｈｉｌｌｉｐｓｂｕｒｇ，Ｎｅｗ Ｊｅｒｓｅｙ）と共に磁気撹拌器で室温で１６時間撹拌した（このプロセス中は終始フラスコをアルミホイルで完全に被覆しておく）。溶液を濾過用漏斗で濾過し（２リットル、７０〜１００μ）、水浴（４０℃）に維持しておいた２２リットル容の４つ口フラスコに濾液を移入した。真空下（２８５〜３６０ミリバール）で溶媒を蒸発させ、残渣（高度に粘性の液体）をジクロロメタンとＴＨＦとの１：１混合物（１リットル）に再溶解した。次に、前記と同様に溶液を沈殿させ、濾過し、乾燥した。測定収量は２２１．０ｇであった（ポリエチレングリコールの重量を基準として７２．６７％の回収率）。
実施例２
この実施例は、Ｂｕｔｙｌ、Ｐｈｅｎｙｌ及びＥｔｈｅｒ ＨＩＣ樹脂の２０ｋＤａ ＰＥＧｂｖに対する結合能力を試験するように設計した一連のローディング及び溶出実験を記載する。
使用アッセイ
１．ＰＥＧアッセイ
ＰＥＧ含有画分をネスラー試薬（Ｓｉｇｍａ／Ａｌｄｒｉｃｈ）によって検出した。典型的には、ガラス試験管に入れたカラム画分の０．２５ｍｌアリコートに０．０１ｍｌのネスラー試薬を添加する。ネスラー試薬を添加すると、ＰＥＧ含有画分は乳状沈殿物を形成した。
２．ＰＥＧ定量分析
ガラス試験管にネスラー試薬（１．５ｍｌ）を添加し、次いで１８．７５マイクロリットルの被験サンプルを添加した。試験管を回転させ、１０〜１５分間静置し、ネスラー試薬だけを入れたブランクに比較して６００ｎｍの混濁度を読取った（Ｂｅｃｋｍａｎ ＤＵ ７０分光光度計）。次に、１．５、１．０、０．７５、０．５０及び０．２５ｍｇ／ｍｌのＰＥＧ標準から作成した標準曲線からＰＥＧを定量した。
実験手順
５ｍｌのプレパックしたＴｏｙｏＰｅａｒｌ^▲Ｒ▼Ｂｕｔｙｌ、Ｐｈｅｎｙｌ及びＥｔｈｅｒ ６５０Ｃ ＨＩＣカラム（ＴｏｓｏＨａａｓ，Ｍｏｎｔｇｏｍｅｒｙｖｉｌｌｅ，ＰＡ）を使用した。種々の濃度のＮａＣｌまたは硫酸ナトリウム中の１０ｍｇ（ベッドボリューム１ｍｌあたり２ｍｇ）の２０ｋＤａ ＰＥＧｂｖをカラムに保持させ、重力を利用して水で溶出させた。全部の実験を室温で実施し、手動操作によって０．５〜１．０ｍｌの画分を１３×１００ｍＭのガラス試験管に集めた。ネスラーのアッセイでＰＥＧを測定した。表１は、種々のＨＩＣ樹脂に２０Ｋ ＰＥＧｂｖを完全結合（即ち、フロースルー液中に検出可能なＰＥＧｂｖが存在しない）させるために必要な塩濃度を示す。

表１に示すように、Ｂｕｔｙｌ ６５０Ｃ ＨＩＣ樹脂は、Ｐｈｅｎｙｌ ６５０Ｃ ＨＩＣ樹脂よりもＰＥＧに対する親和性がやや大きい。また、Ｅｔｈｅｒ ６５０Ｃ ＨＩＣ樹脂は２ＭのＮａＣｌまたは０．５Ｍの硫酸ナトリウムの濃度で保持されたＰＥＧに全く結合しなかった。
実施例３
この実施例では、Ｂｕｔｙｌ ＨＩＣ樹脂の、様々なＰＥＧジオールを大きさに基づき分離する能力を試験するべく計画された実験について説明する。ＰＥＧ含有画分を先に述べたＰＥＧアッセイを用いて確定し、ＰＥＧの分子量をＳＥＣ ＨＰＬＣによって確認した。
用いたアッセイ：
１．サイズ排除クロマトグラフィー（ＳＥＣ）ＨＰＬＣアッセイ
ＨＩＣ精製したＰＥＧ画分のサイズ排除クロマトグラフィーを、Ｔｏｓｏ−Ｈａａｓ Ｇ３０００_swxlカラム（Ｓｙｎｃｈｒｏｍ，Ｉｎｃ．，Ｌｉｎｄｅｎ，ＮＪ）を用いて実施した。１００〜２００μｌ試料を０．５〜１．０ｍｇ／ｍｌで注入する。カラムは室温で管理し、１０ｍＭリン酸ナトリウム緩衝液、１５０ｍＭ ＮａＣｌ、０．００４％ドデシル硫酸ナトリウム（ＳＤＳ）、ｐＨ６．５中で平衡させた。カラムからＰＥＧを、１０ｍＭリン酸ナトリウム、１５０ｍＭ ＮａＣｌ、及び０．００４％ ＳＤＳ、ｐＨ６．５を含有する緩衝液を用いて溶離した。幾つかの実験ではＲｅｆｒａｃｔｉｖｅ Ｉｎｄｅｘ検出器（ｄｅｔｅｃｔｏ）（Ｂｉｏ−Ｒａｄ，Ｈｅｒｃｕｌｅｓ，ＣＡ）を用いてＰＥＧの溶離を監視した。他の実験ではまずＰＥＧをｏ−チオ安息香酸で誘導体化し、それによって２５４ｎｍでのＵＶ吸光度により溶出液を監視できるようにした。
実験操作：
ＴｏｙｏＰｅａｒｌ^▲Ｒ▼ Ｂｕｔｙｌ ６５０Ｍ ＨＩＣカラム（１．５×１１ｃｍ）に、カラム床体積（ベッドボリュウム）１ｍｌ当たり合計で３ｍｇのＰＥＧジオールを導入（ローディング）した。カラムは４Ｍ ＮａＣｌ中で平衡させた。導入物は体積９．７ｍｌで、２．９７ｋＤａ、１１．９ｋＤａ、２０ｋＤａ及び５０ｋＤａ ＰＥＧジオールを各１．５ｍｇ／ｍｌの量で含有した（２．９７ｋＤａ及び１１．９ｋＤａ ＰＥＧジオールはＰｏｌｙｍｅｒ Ｌａｂｏｒａｔｏｒｉｅｓから購入し、２０ｋＤａ及び５０ｋＤａ ＰＥＧジオールはＳｈｅａｒｗａｔｅｒ Ｐｏｌｙｍｅｒｓから購入した）。導入後、カラムをカラム容量の２倍量の４Ｍ ＮａＣｌで洗浄し、カラム容量の１５倍量の、線形濃度勾配３→１ＭのＮａＣｌで溶離した。導入、洗浄及び溶離ステップで、流量は２．５ｍｌ／分とした。３分（７．５ｍｌ）画分を回収し、該画分をＰＥＧに関して分析した。溶離プロフィールを図１に示す。サイズ排除ＨＰＬＣクロマトグラフィーによって確認した四つのピークの分子量は、図中に示したとおりである。
上述の初期実験を完成するべく、２Ｍ ＮａＣｌフロースルー液（ｆｌｏｗ−ｔｈｒｏｕｇｈ）中に溶出したＰＥＧｂｖ及び実施例２に述べたＰｈｅｎｙｌ ６５０Ｃカラムの水溶出液中に溶出したＰＥＧｂｖをＳＥＣ ＨＰＬＣアッセイで分析した。ＳＥＣ ＨＰＬＣデータから、２Ｍ ＮａＣｌフロースルー液中の物質は２０ｋＤａ ＰＥＧｂｖに富み、一方水溶出液中の物質は主として高分子量ＰＥＧｂｖを含有することが判明した。このようなデータによって、ＨＩＣ、特にＰｈｅｎｙｌ及び／またはＢｕｔｙｌ ＨＩＣを用いてＰＥＧジオール及び２０ｋＤａ ＰＥＧｂｖを大きさに基づき分離し得ることが確認される。
実施例４
この実施例では、様々な２０ｋＤａ ＰＥＧｂｖ導入条件を試験し、かつＢｕｔｙｌまたはＰｈｅｎｙｌ ＨＩＣ樹脂からのＰＥＧｂｖの溶離に様々な線形ＮａＣｌ濃度勾配を用いた一連のＨＩＣ実験について説明する。いずれの実験も室温で、かつ通常は実施例２に述べたようにして行なった。
第１実験：
第１実験ではＴｏｙｏＰｅａｒｌ^▲Ｒ▼ Ｐｈｅｎｙｌ ６５０Ｓ ＨＩＣカラム（１．５×６．５ｃｍ）を用いた。６５０Ｓ樹脂は６５０Ｃ樹脂と、粒径が５０〜１５０μｍではなく２０〜５０μｍである点で相違する。カラムを３Ｍ ＮａＣｌ中で平衡させ、このカラムに床体積１ｍｌ当たり２ｍｇのＰＥＧｂｖを導入した。カラム容量の１０倍量の線形濃度勾配ＮａＣｌ（３→０Ｍ ＮａＣｌ）を用い、流量を１．５ｍｌ／分としてＰＥＧｂｖを溶離した。２分（３．０ｍｌ）画分を回収した。溶離プロフィールを図２に示す。その後、２０ｋＤａ ＰＥＧｂｖの収率（％）を、実施例３に述べたアッセイを用いて測定した（表２参照）。
第２実験：
第２実験ではＴｏｙｏＰｅａｒｌ^▲Ｒ▼ Ｂｕｔｙｌ ６５０Ｍ ＨＩＣカラム（１．５×１１．４ｃｍ）を用いた。６５０Ｍ樹脂は６５０Ｃ樹脂と、粒径が５０〜１５０μｍではなく４０〜９０μｍである点で相違する。カラムを２Ｍ ＮａＣｌ中で平衡させ、このカラムに床体積１ｍｌ当たり２ｍｇのＰＥＧｂｖを導入した。カラム容量の１０倍量の線形濃度勾配ＮａＣｌ（２→０Ｍ ＮａＣｌ）を用い、流量を１．５ｍｌ／分としてＰＥＧｂｖを溶離した。３分（４．５ｍｌ）画分を回収した。溶離プロフィールを図３に示す。その後、２０ｋＤａ ＰＥＧｂｖの収率（％）を、実施例３に述べたアッセイを用いて測定した（表２参照）。
第３実験：
第３実験ではＴｏｙｏＰｅａｒｌ^▲Ｒ▼ Ｂｕｔｙｌ ６５０Ｍ ＨＩＣカラム（３．２×１８．０ｃｍ）を用いた。カラムを２．５Ｍ ＮａＣｌ中で平衡させ、このカラムに床体積１ｍｌ当たり１．８ｍｇのＰＥＧｂｖを導入した。カラム容量の１０倍量の線形濃度勾配ＮａＣｌ（２→０．５Ｍ ＮａＣｌ）を用い、流量を９．６ｍｌ／分としてＰＥＧｂｖを溶離した。２．５分（２４ｍｌ）画分を回収した。その後、２０ｋＤａ ＰＥＧｂｖの収率（％）を、実施例３に述べたアッセイを用いて測定した（表２参照）。
第４実験：
第４実験ではＴｏｙｏＰｅａｒｌ^▲Ｒ▼ Ｂｕｔｙｌ ６５０Ｍ ＨＩＣカラム（１．５×１１．５ｃｍ）を用いた。カラムを５Ｍ ＮａＣｌ中で平衡させ、このカラムに床体積１ｍｌ当たり８ｍｇのＰＥＧｂｖを導入した。単位床体積当たりの導入量の増大は、大規模操作条件により近付けることを目的として行なった。導入量が増加するため、ＰＥＧｂｖが完全にカラムに結合するのにより高いＮａＣｌの濃度（２Ｍに対して５Ｍ）が必要となった。カラム容量の１０倍量の線形濃度勾配ＮａＣｌ（３．５→１Ｍ ＮａＣｌ）を用い、流量を２．０ｍｌ／分としてＰＥＧｂｖを溶離した。１．５分（３ｍｌ）画分を回収した。その後、２０ｋＤａ ＰＥＧｂｖの収率（％）を、実施例３に述べたアッセイを用いて測定した（表２参照）。
第５実験：
第５実験ではＴｏｙｏＰｅａｒｌ^▲Ｒ▼ Ｂｕｔｙｌ ６５０Ｍ ＨＩＣカラム（１．５×１１．５ｃｍ）を用いた。カラムを５Ｍ ＮａＣｌ中で平衡させ、このカラムに床体積１ｍｌ当たり８ｍｇのＰＥＧｂｖを導入した。カラム容量の１５倍量の線形濃度勾配ＮａＣｌ（２→０．５Ｍ ＮａＣｌ）を用い、流量を２．１ｍｌ／分としてＰＥＧｂｖを溶離した。画分を回収した。溶離プロフィールを図３に示す。この実験の分離度プロフィールは２ｍｇ／ｍｌ導入を用いて得られたプロフィールほど優れていなかった。しかし、この実験から得られた、測定したＳＥＣ ＨＰＬＣ純度は図４及び５に示したとおりである。図４及び５のデータは、分離度が低いにもかかわらず高分子量ＰＥＧｂｖが適宜除去されたことを証明している。
第６実験：
第６実験ではＴｏｙｏＰｅａｒｌ^▲Ｒ▼ Ｂｕｔｙｌ ６５０Ｍ ＨＩＣカラム（７．０×１４ｃｍ）を用いた。カラムを４．５Ｍ ＮａＣｌ中で平衡させ、このカラムに床体積１ｍｌ当たり８ｍｇのＰＥＧｂｖを導入した。カラム容量の１７倍量の線形濃度勾配ＮａＣｌ（３．５→１．０Ｍ ＮａＣｌ）を用い、流量を５５ｍｌ／分としてＰＥＧｂｖを溶離した。

上記データが示すように、Ｐｈｅｎｙｌ及び／またはＢｕｔｙｌ ＨＩＣ分離において線形濃度勾配ＮａＣｌを用いることは、２０ｋＤａ ＰＥＧｂｖから高分子量ＰＥＧｂｖを分離する上で有効である。更に、適当な条件下に床体積１ｍｌ当たり２ｍｇまたは８ｍｇの導入量を用いることによって約６５〜７５％の収率が得られた。最後に、上記データは、導入量を８ｍｇ／ｍｌとした実験では導入量２ｍｇ／ｍｌの実験の時ほど優れた分離度プロフィールは得られないものの、高分子量ＰＥＧｂｖを適宜除去することはできることを示している。
実施例５
この実施例では、ＴＮＦｂｐ及びＨＩＣ精製２０ｋＤａ ＰＥＧｂｖを含む生物学的に活性なコンジュゲート（ホモダンベル）の製造を説明する。上記コンジュゲートの総収率、純度及び生物学的活性を従来のＴＮＦｂｐコンジュゲートのものと比較する。
コンジュゲートの製造：
この実施例で用いたｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテインは、本明細書に参考として含まれる国際特許出願公開第９２／１６２２１号に開示されたようにして製造し得る。あるいはまた、上記ｃ１０５ムテインは次のように製造する。ｃ１０５ムテインを発現させる大腸菌細胞を遠心によって回収する。細胞スラッジを精製水の添加によって約４０湿潤重量％の固体に調節する。次に、混合物を等量の破壊（ｂｒｅａｋｉｎｇ）緩衝液（５０ｍＭトロメタミン、４ｍＭ ＥＤＴＡ、ｐＨ７．２）で更に稀釈し、それによって約２０湿潤重量％の固体を含有する懸濁液を得る。細胞スラッジを、約８，０００ｐｓｉで作動する高圧ホモジナイザーに５回通して細胞ホモジネートを製造する。ホモジナイザーに通す際、毎回事前にホモジネートを１０℃以下に冷却する。ホモジネートを遠心し、ｃ１０５を含有する固体画分を得る。固体を稀釈し、かつ再度遠心して、洗浄された封入体を得る。
次に、洗浄された封入体を５０ｍＭトリス、ｐＨ９．５中の８Ｍ尿素及び１５０ｍＭシステインの添加によって溶解させる。この混合物を再生前に室温で２時間攪拌する。前記条件下にｃ１０５ムテインは変性し、還元される。
還元された変性ｃ１０５ムテインを１．１Ｍ尿素、５０ｍＭトリスでの稀釈によって再生させ、それによって２００μｇ／ｍｌのｃ１０５ムテインと、１．５Ｍ尿素と、７．５ｍＭシステインと、５０ｍＭトリス、ｐＨ９．７とから成る最終再生溶液を得る。再生混合物を２日間６〜１０℃に維持する。再生効率を逆相ＨＰＬＣ及びカチオン交換ＨＰＬＣによって監視する。
次に、再生混合物を酢酸及びＨＣｌの添加によってｐＨ５．０とする。再生混合物を、４℃において２５ｍＭ酢酸ナトリウム、６５ｍＭ ＮａＣｌ中で予め平衡させたカチオン交換カラム（Ｓ−セファロース大型ビーズ樹脂）に導入する。導入後、カラムを上記と同じ平衡緩衝液で洗浄する。カラムを、２５ｍＭ酢酸ナトリウム、ｐＨ５中の濃度勾配６５→３５０ｍＭのＮａＣｌで溶離する。ＮａＣｌ濃度約２００ｍＭにおいてｃ１０５ムテインが溶出し、これを一つのプールに回収する。
ｃ１０５ムテインを含有する回収プールを１．５倍量の５Ｍ ＮａＣｌ、４０ｍＭリン酸ナトリウム（ｐＨ６に調節）で稀釈し、これを３Ｍ ＮａＣｌ、２０ｍＭリン酸ナトリウム、ｐＨ６中で予め平衡させた疎水的相互作用カラム（Ｔｏｙｏ Ｂｕｔｙｌ ６５０Ｍカラム）に導入する。導入終了後、カラムを平衡緩衝液で洗浄する。２０ｍＭリン酸ナトリウム、ｐＨ６中の線形塩濃度低下勾配３→１ＭのＮａＣｌをカラム容量の８倍量で用いてｃ１０５ムテインを溶離する。ｃ１０５ムテインを一つのプールに回収する。プールをｃ１０５ムテイン約３ｇ／ｌに濃縮し、その後最終的に導電率が４ｍｍｈｏ未満となるまで２０ｍＭリン酸ナトリウム、ｐＨ６に対して膜濾過する（約６倍量）。
膜濾過したプールを、２０ｍＭリン酸ナトリウム、ｐＨ６中で平衡させたＳＰ−セファロース高速カラムに導入する。導入後、カラムを追加の平衡緩衝液で洗浄し、２０ｍＭリン酸ナトリウム、５０ｍＭ ＮａＣｌ、ｐＨ６．０から２０ｍＭリン酸ナトリウム、５０ｍＭ ＮａＣｌ、ｐＨ６．５までのｐＨ／塩濃度勾配の組み合わせを有する溶離剤で溶離する。前記勾配の後半、ＮａＣｌ濃度約３５ｍＭにおいてｃ１０５ムテインが溶出する。この時点で、ｃ１０５ムテインを冷凍貯蔵し得る。
この実施例で用いた２０ｋＤａ ＰＥＧｂｖは、先の実施例１に述べた「ワンステップ」操作を用いて合成した。合成した２０ｋＤａ ＰＥＧｂｖの一部を、実施例４の第５実験に述べたようにしてＨＩＣ精製した。次に、ＨＩＣ精製しなかった２０ｋＤａ ＰＥＧｂｖとＨＩＣ精製した２０ｋＤａ ＰＥＧｂｖとの両方を用いてＰＥＧｂｖ：ＴＮＦｂｐコンジュゲートを製造した。コンジュゲートの製造は次のように行なった。ｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテインを１５℃で５〜６時間、２０ｍＭリン酸塩、ｐＨ７．６中の１．３倍モル過剰量のＤＴＴに曝露し、それによって再生過程の間にｃ１０５に結合した余分なシステインを除去した。ｐＨを６．０に調節し、試料を（４〜６ｍｇ／ｍｌ床体積の量で）Ｓ−セファロースＦａｓｔ Ｆｌｏｗカラムに導入し、カラム容量の３倍量の２０ｍＭリン酸緩衝液、ｐＨ６．０で洗浄し（それによってＤＴＴを除去し）、２０ｍＭリン酸塩、７０ｍＭ ＮａＣｌ、ｐＨ６．０で溶離した。
次に、還元ｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテイン溶液（〜４．７ｍｇ／ｍｌ）を１Ｍ ＮａＯＨでｐＨ７．４〜７．５に調節した。この還元ｃ１０５ ＴＮＦｂｐにＨＩＣ精製または非ＨＩＣ精製２０ｋＤａ ＰＥＧｂｖ（０．９ｍｇ／ｍｇ ＴＮＦｂｐ）を攪拌下に添加した。攪拌混合後、溶液を周囲温度で１２〜１８時間放置した。
コンジュゲートアッセイ：
ＰＥＧ化アッセイ
ＴＮＦｂｐの２０ｋＤａ ＰＥＧｂｖ：ＴＮＦｂｐダンベルへの変換を、ＰＯＲＯＳ^TM ＨＳ／Ｈ ４．６／５０カチオン交換ＨＰＬＣカラム（Ｐｅｒｓｅｐｔｉｖｅ Ｂｉｏｓｙｓｔｅｍｓ，Ｃａｍｂｒｉｄｇｅ，ＭＡ）を用いて定量した。５０μｌ試料をカラムに導入し、かつ２０ｍＭ酢酸ナトリウム、ｐＨ３．７５中の１０分濃度勾配０→１ＭのＮａＣｌを用いてカラムから溶離した。ｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテインモノベル（ｍｏｎｏｂｅｌｌ）、即ち一端にのみｃ１０５ ３０ｋＤａ ＴＮＦｂｐが結合したＰＥＧｂｖ、ｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテインダンベル、及び非ＰＥＧ化ｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテイン試料に関する代表的溶離プロフィールを図６に示した。ｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテインモノベルが最初に溶出し、次いでｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテインダンベル及び非ＰＥＧ化ｃ１０５ ３０ｋＤａ ＴＮＦｂｐムテインが溶出した。
結果
ＨＩＣ精製物質を用いた場合、ＴＮＦｂｐダンベルへの変化率は６６．７％であり、その際検出可能な「プレピーク」は存在しなかった（図７参照）。同一ロットから得た非ＨＩＣ精製ＰＥＧｂｖは５２．９％の変換率しかもたらさず、かつ６．７％の「プレピーク」を有した（図８参照）。際立ったこの「プレピーク」は、高分子量ＰＥＧｂｖ不純物によって発生したＴＮＦｂｐダンベルを示している。
次に、ＰＥＧ化混合物を、５０ｍＭリン酸ナトリウム、ｐＨ３．１中で平衡させた１．５×１０ｃｍ ＳＰ−セファロースＨＰカラムに導入した（４ｍｇ／ｍｌ床体積）。導入後、カラムを平衡緩衝液で洗浄し、その後５０ｍＭリン酸ナトリウム、０．２５Ｍ ＮａＣｌ、ｐＨ３．１から５０ｍＭリン酸ナトリウム、０．５Ｍ ＮａＣｌ、ｐＨ３．１までの勾配を有する溶離剤をカラム容量の２０倍量で用いてＴＮＦｂｐダンベルを溶離した。溶離プロフィールを図９及び１０に示す。各溶出物毎に幾つか異なる方法でプールを形成した。表３に、用いた様々なプール形成規準毎のＴＮＦｂｐダンベルの収率（％）及び純度（％）を示す。

表３のデータは、ＰＥＧｂｖのＨＩＣ精製が後に得られるＴＮＦｂｐダンベルの純度及び収率にとって明らかに有益であることを示している。例えば９７％の純度を達成しようとすると、非ＨＩＣ精製ＰＥＧｂｖを用いた場合はＳＰ−セファロースＨＰカラム上で３７％のＴＮＦｂｐダンベルしか回収できなかったが、ＨＩＣ精製物質を用いれば５８％もが回収でき、即ち２０ｋＤａ ＰＥＧｂｖのＨＩＣ精製によってＴＮＦｂｐダンベルの回収量を１．５〜１．６倍にも増加させることが可能となる。
実施例６
この実施例ではＨＩＣ樹脂の、ＰＥＧを末端基官能性に基づき分離する能力を評価した。
実施例４の第３実験で得られた「早期溶出」物質をＰＥＧ化アッセイにおいて試験した。「早期溶出」物質とは、２０ｋＤａ ＰＥＧｂｖを得るべくプールした画分の直前に溶出した画分のことである。この物質は、図１１に示したように、コンジュゲートの製造に用いると主にＴＮＦｂｐモノベルをもたらした。このことは、「早期溶出」物質が主としてＰＥＧ ２０ｋＤａモノビニルスルホンを含有することを示しており、即ちＢｕｔｙｌ ＨＩＣ樹脂の、２０ｋＤａ ＰＥＧｂｖを末端基官能性に基づき分離する能力を証明している。
実施例７
この実施例では、ｃ１０５ ＴＮＦｂｐムテインとＰＥＧ−ビス−ビニルスルホンとの連結の安定性を試験した。既知量のｃ１０５ ＴＮＦｂｐダンベルをＰＢＳ、ｐＨ７．４中で、３７℃で１週間インキュベートし、その間に時々ＳＤＳ ＰＡＧＥによる分析用にアリコートを取り出した。ｃ１０５ ＴＮＦｂｐダンベルの分解は実質的に観察されなかった。ｐＨ１０において３７℃で１週間インキュベートしても、５〜１０％のコンジュゲート分解が観察されただけであった。別の安定性試験では、実施例１のワンステップ法で製造した活性化ＰＥＧリンカーを、特に本明細書に参考として含まれる国際特許出願公開第９５／０７５５５号に開示されたマルチステップ法で製造した活性化ＰＥＧリンカーと比較した。
まず、ワンステップ及びマルチステップ活性化リンカーを２−メルカプト安息香酸で誘導体化し、それによってリンカーの安定性をＵＶを用いて測定した。各０．５ｍｌのリンカー溶液（２０ｍｇのリンカーを０．５ｍｌの５０ｍＭリン酸ナトリウム、ｐＨ７．５に溶解させて製造）に１ｍＭ ２−メルカプト安息香酸（オルトチオール安息香酸）を３．０ｍｌずつ添加し、渦流発生によって短時間混合した。その後、溶液を室温で１８時間以上インキュベートした。室温でのインキュベーションの時間は比較的長くすることが可能である（４８時間まで）。
次に、誘導体化リンカー溶液をＰＢＳで１：１０に稀釈し、９０℃で１時間熱処理した。熱処理したリンカー溶液を熱から遠ざけ、１００〜２００μｌの該溶液をＢＩＯ ＳＩＬ^TM ２５０−５カラム（ＢｉｏＲａｄ，Ｈｅｒｃｕｌｅｓ，ＣＡ）に注入し、流量０．５ｍｌ／分で１０ｍＭリン酸ナトリウム、１５０ｍＭ ＮａＣｌ、０．００４％ ＳＤＳ、ｐＨ６．５中へ流入させた。回収した試料を２５４ｎｍのＯ．Ｄ．においてＵＶで測定した。表４及び５にＵＶ測定結果を示す。これらの表中、生成物ＵＶピーク変化率（％）は、熱処理を施さなかった標準のＵＶ純度から９０℃熱処理試料の純度を減じた差に基づいて計算してある。純度は、平均ピーク面積を総ピーク面積で除した商として算出した。このアッセイの変動率は約２％であった。

表４及び５に示した結果の比較から明らかなように、ワンステップリンカーはマルチステップリンカーより安定である。上記結果は、ワンステップリンカーの安定性の変化が２％のアッセイ変動率の範囲内である一方で、４ステップリンカーの安定性の変化は前記アッセイ変動率を越えることも示している。
本発明についてのここまでの記述は、例解及び説明のための一例である。本発明をその思想及び範囲を逸脱することなく変更及び変形し得ることは、当業者には明らかであろう。そのような変更及び変形はいずれも以下の請求の範囲各項に包含されると解釈するものとする。 Field of Invention
The present invention relates to a method for producing an activated linker, in particular, a method for producing an activated linker having a reactive Michael acceptor, and a method for efficiently purifying the activated linker.
Background of the Invention
Polyethylene glycol (“PEG”) is a long-chain linear synthetic polymer composed of ethylene oxide units. The ethylene oxide units are variable and thus the resulting PEG compound can have a molecular weight in the range of about 200-100,000. PEG is typically colorless and odorless, water-soluble, heat-stable, does not cause hydrolysis or degradation, and is non-toxic. Therefore, extensive research has been conducted on the use of PEG for pharmaceuticals, artificial transplants, and other applications where biocompatibility is important.
Various derivatives of PEG having active moieties capable of attaching PEG to various surfaces and molecules are known. For example, PEG derivatives have been used to attach PEG to a surface to control surface wetting and charging and adhesion of other molecules to the surface. More specifically, a reagent having a reactive carbonyl group, a nitrile group or a sulfone group is reacted with PEG to convert the hydroxyl group to a reactive Michael acceptor to form an “activated linker”. To form a PEG derivative. “Activated linkers” are useful for modifying various proteins to obtain improved biologically active conjugates.
By attaching PEG to a protein ("PEG addition or PEGylation"), (1) prolonging the plasma half-life of the protein, US Pat. No. 5,349, Delgado et al. See No. 052; (2) Modification of protein biochemical and physical properties such as increased solubility, Chen et al., Biocem. Biophys. Acta,660: 293-298 (1988) and enhanced protection against proteolysis, Sada et al., J. MoI. Fermentation Bioengineering71137-139 (1991); and (3) Reduction of antigenicity, Davis et al., “Biomedical Polymers. Polymeric Materials and Pharmaceuticals for Biomedical Use 1, (Eds. Goldberg & M amapap & Nakapak & Makap. N. Y. (1980) has been proved to be obtained.
Several problems have occurred in attaching PEG to surfaces and molecules. Most of the reagents for PEG addition react with the free primary amino group of the polypeptide. Many of these free amines are ε-NH of lysine amino acid residues.₂It is a group. Since typical proteins have a large number of lysine residues, random attachment of a large number of PEG molecules, i.e. non-specific PEG addition, often occurs. Non-specific PEG addition reduces protein activity. Thus, even if PEGylated proteins are to be used for therapy, non-specific PEGylation results in a large number of species, making it difficult to produce a product that can be reproduced and characterized. Therefore, when non-specific PEG addition occurs, it is difficult to evaluate the treatment method and to determine the treatment effect and medication information. If site-selective PEG addition of such proteins can be achieved, a reproducibly modified substance can be obtained so that only the desired PEG addition is provided without reducing activity.
In particular, International Patent Publication No. WO 95/13312 describes water-soluble sulfone-activated PEGs that bind very selectively to thiol moieties rather than molecular and surface amino moieties. These PEG derivatives are stable to hydrolysis for long periods in an aqueous environment at a pH of about 11 or less, and can form chains with molecules to form hydrolytically stable conjugates as well. The linkage where the PEG and the biologically active molecule are attached includes a sulfone moiety attached to the thiol moiety and has the formula PEG-SO₂-CH₂-CH₂-S-W, where W represents a biologically active molecule and the sulfone moiety is vinyl sulfone or active ethyl sulfone.
An example of a biologically active molecule that can be modified by such a sulfone activated linker is a TNF binding protein (“TNFbp”) (also called a TNF inhibitor or TNF inhibitor). TNFbp is important as a therapeutic agent to treat TNF-mediated diseases such as rheumatoid arthritis and septic shock. TNF binding proteins are disclosed in the art. For example, EP308,378, EP393,438, EP422,339, WO90 / 13575, EP398,327, EP412,486, WO91 / 03553, EP418,014, JP127,800 / 1991, EP433,900, US Pat. 5,136,021, EP512,528, EP526,905, WO93 / 07863, EP568,928, WO93 / 21946, WO93 / 19777, EP417,563, GB2,246,569, US patent filed on July 9, 1996 Provisional application no. No. 60/021443, US Provisional Application No. filed Dec. 6, 1996. The contents of these documents are included in the present invention by reference.
WO 92/16221, which is included in the present invention by reference, has the formula R₁-X-R₂Describes the form of TNFbp represented by (referred to as "dumbbell" compound). R in the formula₁And R₂Is a biologically active TNFbp group and X is a non-peptidic polymer spacer having at least one Michael acceptor group, such as a sulfone activated PEG. One specific compound called c105 TNFbp mutein dumbbell is R₁And R₂Is a homodern bell compound in which R represents c105 30 kDa TNFbp mutein and X represents 20K polyethylene glycol bis-vinylsulfone (20 kDa PEGbv), R₁And R₂Is site-specifically bound to PEGbv at the cysteine 105 residue.
A problem often encountered by those skilled in the art when preparing mono-, di- or poly-PEG adducts is that the synthesis procedure of 20 kDa PEGbv or other activated linker has several drawbacks depending on the synthesis procedure used. is there. For example, conventional methods of activating PEG with functional groups useful as linkers attached to various molecules and surfaces are generally labor intensive and time intensive. Such methods often require a multi-step reaction procedure to produce the desired product. Accordingly, there is a need for an efficient method for producing activated linkers that can be used to form the desired conjugates.
Another disadvantage that is of particular concern is that the resulting 20 kDa PEGbv often contains high molecular weight PEG impurities, PEG monovinylsulfone species, and PEG bis-vinylsulfone species. These impurities adversely affect the overall yield of the desired compound. Accordingly, there is also a need for methods that further purify PEGbv or other activated linkers, thereby improving the purity and overall yield of the desired compound.
Many research groups have reported on the use of reverse phase high performance liquid chromatography (rpHPLC) to separate low and medium molecular weight (<5K) PEGs based on size. For example, Murphy et al. Chromat. ,211160-165 (1981); Lai et al., J. MoI. High Res. Chrom. & Chrom. Comm. ,7: 494-496 (1984); Barka & Hoffman, J. MoI. Chromat. ,389: 273-278 (1987). These techniques are relatively effective, but have the disadvantage of using expensive resins and organic solvents. Also, these methods are not manufacturing procedures but rather analytical procedures. There are no reports of successful separation of higher molecular weight PEG using HPLC. Medium and high molecular weight (1K-40K) PEGs were analyzed by size exclusion chromatography using various derivatized silica supports. Engelhardt & Mathes, J .; Chromat. ,185: 305 (1979).
Hydrophobic interaction chromatography (HIC) is well established as a purification method for the analysis and production of biomolecules. The chemical principle underlying HIC is similar to that of salt precipitation in that all treatments are based on hydrophobicity and are close to the principles associated with reverse phase liquid chromatography (RPLC). HIC adsorbs molecules to weakly hydrophobic surfaces at high salt concentrations and then elutes using a gradually decreasing salt gradient. Therefore, the feature of HIC is to combine the non-denaturing feature of salt precipitation with the accuracy of chromatography in order to obtain high activity recovery. Furthermore, since HIC can work with low binding energy, it is not necessary to use expensive organic solvents for the mobile phase.
The use of HIC resins for separating PEG, in particular for separating high molecular weight PEG, has never been reported. Since PEG has relatively the same hydrophobicity regardless of size, those skilled in the art did not expect HIC to be effective in separating PEG based on size.
Summary of the Invention
The present invention provides a novel method for the production and purification of activated linkers useful for various purposes. Briefly, the method for producing an activated linker of the present invention involves obtaining a compound having a reactive hydroxyl group and converting the hydroxyl group to a reactive Michael acceptor using tetrahydrofuran (THF) as a conversion solvent. To form an activated linker. An activated linker purification method uses HIC to separate linkers based on size and end group functionality.
More specifically, the object of the present invention is to use butyl or phenyl HIC resin to produce and purify 20 kDa polyethylene glycol bis-vinylsulfone (“20 kDa PEGbv”) from PEG bis-vinylsulfone. A method for separating high molecular weight PEG impurities and PEG mono-vinyl sulfone. Importantly, such purification methods do not require the addition of organic solvents. More importantly, purified activated linkers can be used to modify a variety of proteins, with dramatically improved purity and overall yield compared to conjugates formed by non-purified activated linkers. To provide a biologically active conjugate.
For example, compounds with at least one reactive hydroxy group can be used, such as alcohols and in particular polymers with —OH functionality. Polymers such as homopolymers, random or block copolymers or terpolymers are useful, particularly polyethylene glycol (PEG) s, polypropylene glycols, polyalkylene oxides, polyoxyethylated polyols and polyvinyl alcohol are preferred. . Among these, polyethylene glycol having a molecular weight in the range of 2 kDa to 100 kDa dalton is particularly useful, and among them, PEG having a molecular weight of 3.4 kDa, 5 kDa, 10 kDa, or 20 kDa is useful. Such PEG may have the form of PEG diol.
Non-limiting examples of useful reagents having reactive sulfone groups that can be used in the above methods are vinyl sulfone and divinyl sulfone. When used with PEG having one or two reactive hydroxy groups, the method produces PEG-vinylsulfone and PEG-bis-vinylsulfone, respectively. Such activated linkers can be used for the production of monoPEG adduct compounds and dumbbell and multimeric compounds where the linker can be attached to more than one moiety. Addition of PEG to THF ratio in the range of 1:10 to 1:50 (weight / volume) using THF as a solvent has been found to be useful in the process of the present invention. A ratio of 1:30 has been found to be particularly useful.
Biologically active molecules that could be used can be either synthetic molecules, naturally occurring molecules or modified naturally occurring molecules.
According to one feature of the invention, PEG or another activated linker having a molecular weight in the range of 2 kDa to 100 kDa is retained on and eluted from the HIC resin using conventional chromatographic methods. In a preferred embodiment, an inverse linear salt gradient is used to elute the linker from the resin. Thus, different forms of linkers are separated based on size and / or end group functionality. Preferably, the linker is a sulfone activated linker and the HIC resin consists of butyl or phenyl HIC resin. In a particularly preferred embodiment, the present invention provides Toyopearl^{▲ R ▼}A method is provided for purifying 20 kDa polyethylene glycol vinyl sulfone or bis-vinyl sulfone ("20 kDa PEGbv") using butyl-650M HIC resin.
According to yet another aspect of the present invention, an improvement over previously described compounds containing the same molecule by reacting a HIC purified non-peptidic polymer linker with a biologically active molecule. Dumbbell compounds are obtained in total purity and yield. Preferably, the HIC purified polymer linker is a sulfone activated PEG and the biologically active molecule is a TNF inhibitor, a receptor antagonist of interleukin 1 (“IL-1ra”), platelet derived growth factor (“PDGF”) ) Exon 6 peptide, interleukin 2 ("IL-2") inhibitor and receptor ("IL-2r"). In a particularly preferred embodiment, the HIC purified polymer linker is 20 kDa vinyl sulfone or PEGbv and the biologically active molecule is TNFbp.
Pharmaceutical compositions containing the compounds of the present invention are also encompassed within the scope of the present invention.
[Brief description of the drawings]
Figure 1 shows ToyoPearl^{▲ R ▼}2 shows the elution profiles of various PEG diols eluting from a butyl 650M HIC column. The PEG-containing fraction was detected with a Nessler reagent, and the molecular weight was confirmed by the procedure described in Example 3 using SEC HPLC.
Figure 2 shows ToyoPearl^{▲ R ▼}The elution profiles of 20 kDa PEGbv and high molecular weight PEGbv eluting from a phenyl 650M HIC column (2 mg loading per ml of bed volume) are shown. The PEG concentration was measured with a Nessler reagent, and the percentage of molecular weight impurities was measured using SEC HPLC according to the procedure described in Example 3.
Figure 3 shows ToyoPearl^{▲ R ▼}Shown is the elution profile of 20 kDa PEGbv (2 mg loading per ml bed volume) eluting from a butyl 650M HIC column. The PEG-containing fraction was detected with a Nessler reagent and the turbidity at 600 nm was measured.
FIG. 4 shows the elution profile of HIC unpurified 20 kDa PEGbv eluting from a SEC-HPLC column.
FIG. 5 shows the elution profile of HIC purified 20 kDa PEGbv eluting from the SEC-HPLC column. ToyoPearl for HIC purification^{▲ R ▼}A butyl 650M HIC column (8 mg loading per ml bed volume) was used.
FIG. 6 shows a typical POROS of a TNFbp PEG addition mixture (mixture using unpurified 20 kDa PEGbv).^TMThe HS cation exchange elution profile is shown. Various TNFbp peaks as shown are observed.
FIG. 7 shows POROS of PEG addition mixture of TNFbp^TMThe HS cation exchange elution profile is shown. PEG addition mixture is ToyoPearl^{▲ R ▼}Prepared as described in Example 5 using 20 kDa PEGbv purified with butyl 650M HIC.
FIG. 8 shows POROS of PEG addition mixture of TNFbp^TMThe HS cation exchange elution profile is shown. The PEG addition mixture was prepared according to the procedure described in Example 5 using HIC unpurified 20 kDa PEGbv.
FIG. 9 shows the elution profile of c105 TNFbv mutein dumbbell eluting from the SP-Sepharose HP column. A280 absorbance is plotted over time. c105 TNFbv mutein dumbbell is ToyoPearl^{▲ R ▼}Prepared using 20 kDa PEGbv purified with butyl 650M HIC.
FIG. 10 shows the elution profile of c105 TNFbv mutein dumbbell eluting from the SP-Sepharose HP column. A₂₈₀Absorbance is plotted over time. The c105 TNFbv mutein dumbbell was prepared using unpurified 20 kDa PEGbv.
FIG. 11 shows a 20 kDa PEGbv ToyoPearl^{▲ R ▼}POROS of the “early eluting” fraction obtained by butyl 650M HIC purification^TMThe HS cation exchange elution profile is shown. The PEG addition mixture was prepared according to the procedure described in Example 6.
Detailed description
Methods for the production and purification of activated polyethylene glycol and other activated linkers are discussed in more detail below and illustrated in the examples. Examples illustrate various features of the present invention, use of a “one-step” method to produce sulfone activated PEG, 20 kDa polyethylene glycol vinyl sulfone or bis-vinyl sulfone (“20 kDa EGbv”), and Figure 3 shows the results obtained by using butyl and / or phenyl HIC resin to separate high molecular weight PEG impurities. These results surprisingly show that HIC also separated PEG mono-vinylsulfone from 20 kDa PEGbv, and the resulting purified 20 kDa PEGbv was used for PEGylation of TNFbp when not using HIC unpurified 20 kDa PEGbv or multiple It has been shown to produce biologically active conjugates with dramatically improved overall yield and purity compared to conjugates prepared using the 20 kDa PEGbv synthesized by the step procedure.
In the method for producing an activated linker of the present invention, a compound having a reactive hydroxyl group is obtained, and the hydroxyl group is converted into a reactive Michael acceptor using tetrahydrofuran (THF) as a conversion solvent. Optionally, the compound having a reactive hydroxyl group may be first dissolved in toluene and then dried or evaporated before converting the hydroxyl group to a Michael acceptor. The conversion step is then completed by adding THF as a solvent to the dried compound. In the conversion step, various solvents such as dichloromethane (DCM), dimethylformamide (DMF) and DMSO can be used, but THF is preferred. Since THF is a cation chelating agent, when it is used as a solvent, chelation of the base cation is promoted during the conversion reaction to release negatively charged oxygen, which can be converted into carbonyl, nitrile or sulfone groups. It can react with a reagent containing and form a Michael acceptor. Suitable bases used in the above method are, for example, piperidine, pyridine, triethylamine, benzyltrimethylammonium hydroxide (Triton B), and potassium hydroxide. Strong bases containing cations are preferred. Examples of such strong bases are sodium or potassium tert-butoxide, sodium or potassium alkoxides, metal amides, and sodium or potassium hydrides.
Next, a reagent having a reactive carbonyl, nitrile or sulfone group is added, followed by a base. The activated linker formed by these methods is then precipitated using a suitable solvent and isolated by any means known to those skilled in the art, such as filtration, to obtain a purified activated linker. Suitable solvents are, for example, ethers, in particular diethyl ether and dipropyl ether, alcohols, hexane, xylene and the like.
The method for producing an activated linker of the present invention is referred to as a “one-step” method in contrast to known methods that require multiple steps. The one-step method can be illustrated, for example, as follows.

It is contemplated that any compound having at least one reactive hydroxyl group can be used in the method of the present invention. The term “reactive hydroxyl group” means an —OH group containing hydrogen that can reduce a base, which reacts with a reagent having a carbonyl, nitrile or sulfone group to react with a Michael acceptor. Liberates oxygen that can be formed. Examples of hydroxyl-containing compounds useful in the method of the present invention include several examples such as alcohols, polymers, nucleotides, nucleosides, oligonucleotides, peptides, controlled pore glass, and chromatographic substances. Useful polymers include, for example, polyethylene glycol (“PEG”), polypropylene glycol (“PPG”), polyoxyethylated glycerol (“POG”) and other polyoxyethylated polyols, polyvinyl alcohol (“PVA”) and Other polyalkylene oxides, polyoxyethylated sorbitol or polyoxyethylated glucose.
The polymer can be a homopolymer, random or block copolymer, terpolymer, and can be linear or branched, substituted or unsubstituted, as long as it has at least one active hydroxy moiety. The polymerisation moiety can have any chain length or molecular weight, but these property values can affect biological properties. The average molecular weight of polymers useful for reducing clearance rates in pharmaceutical applications is in the range of 2,000-35,000 daltons. Particularly useful PEG molecular weights are 3.4 kDa, 5 kDa, 10 kDa and 20 kDa.
The term “PEG” as used herein refers to any of several condensation polymers of ethylene glycol. PEG is also known as polyoxyethylene, polyethylene oxide, polyglycol and polyether glycol. PEG can also be produced as a copolymer of ethylene oxide and many other monomers. The end groups of PEG can be derivatized in various ways to include vinyl sulfone moieties, aldehyde moieties, methoxy groups, or other types of reactive or non-reactive end groups. For many applications for pharmaceutical or biotechnological use, substantially unsubstituted, substantially linear, linear vinyl sulfone activated PEGs other than vinyl sulfone will be used.
The hydroxyl group of such compounds reacts with a reagent having a reactive carbonyl, nitrile or sulfone group that can convert the hydroxyl group to a Michael acceptor. Any reagent known to those skilled in the art having a reactive carbonyl, nitrile or sulfone group could be used in the method of the present invention. The term “reactive carbonyl, nitrile or sulfone” refers to a reactive site suitable for thiol-specific binding at a pH of about 9 or less to the second carbon of the carbonyl, nitrile or sulfone group. It means carbonyl, nitrile or sulfone group possessed.
In a preferred embodiment of the invention, the polymer derivative of the invention has an active sulfone moiety. The term “active sulfone” refers to a sulfone group in which two carbon atoms are bonded and have a reactive site on the second carbon of the sulfone group that is suitable for thiol-specific binding at a pH below about . Non-limiting examples of active sulfones are vinyl sulfone and activated ethyl sulfone. The sulfone activated polymer may be further substituted with another substituent as long as the thiol-specific reactivity of the second carbon is maintained at a pH of about 9 or less.
Addition of such a reagent converts the hydroxyl group to a reactive Michael acceptor. The terms “Michael acceptor” or “Michael-like acceptor” are used interchangeably and refer to a functional group that is sensitive to Michael addition. “Michael addition” results in a nucleophilic attack on an electrophilic center adjacent to a pi-system with electronegative atoms. Examples of pi systems with electronegative atoms are sulfoxide, sulfonyl, carbonyl and heterocyclic aromatics. A nucleophile is added to the electrophilic center. Michael acceptor formula:

[Wherein E represents an electronegative atom]. Addition occurs at the 4th position, the formula:

[Where N_uMeans a nucleophile bonded to the atom at the 4 position]. Non-limiting examples of Michael acceptor functional groups are maleimide and vinyl sulfone. The activated linker used to form the dumbbell may, but need not, include the Michael acceptor vinyl sulfone species.
The activated linker of the present invention may have more than one reactive group. Derivatives may be monofunctional, bifunctional or polyfunctional. As long as at least one active sulfone moiety is present, the reactive groups may be the same (homofunctional) or different (heterofunctional). Two particularly useful homofunctional derivatives are PEG-bis-chlorosulfone and PEG-bis-vinylsulfone.
The linker produced by the method of the present invention is a stable and isolatable linker suitable for thiol selective coupling reactions. For example, PEG chloroethyl sulfone is stable to water at a pH of about 7 or less, but can be advantageously used for thiol selective coupling at basic pH conditions up to at least about pH 9. When the pH is higher than about 9, the thiol selectivity decreases and the sulfone moiety becomes somewhat reactive towards amino groups. The linkage formed by reaction with thiol is also hydrolytically stable.
As used herein, the term “hydrolytically stable” means that the linkage between the polymer and the sulfone moiety and the sulfone-thiol linkage after conjugation do not react with water at a pH below about 11 for at least 3 days. Means that. When a significant rate of hydrolysis occurs, hydrolytically stable is a desirable requirement because the polymer may be deactivated before the reaction between the polymer and the biologically active molecule thiol occurs.
A universal purification method for activated linkers includes the following steps.
(1) synthesizing an activated linker;
(2) holding the activated linker on the HIC resin, then
(3) Isolating the desired linker form by eluting the linker from the HIC resin using conditions that can separate the various linker forms based on size and / or end group functionality, eg, an inverse linear salt gradient. .
Alternatively, the HIC purification step may be incorporated immediately prior to the concentration / diafiltration step of the “one-step” synthesis procedure.
HICs that could be used in the present invention include hydrophobic side groups (eg, octyl, butyl, phenyl or ether) on the surface of a resin having a hydrophilic vinyl copolymer backbone with methacrylic, agarose or vinyl backbones as non-limiting examples. ) Is used. Silica-based backbones can also be used. Preferred HIC columns have butyl or phenyl ligands in the methacrylic backbone.
As described above, HIC purified sulfone activated linkers can be attached to biologically active molecules having “reactive thiol moieties” to form biologically active conjugates. The term “reactive thiol moiety” refers to a sulfhydryl (—SH) group that can react with an activated linker as described herein.
Non-limiting examples of biologically active molecules that may be used in the present invention include pharmaceuticals, vitamins, nutrients, nucleic acids, amino acids, polypeptides, enzyme cofactors, steroids, carbohydrates, organic species such as heparin, metal-containing Substances, receptor agonists, receptor antagonists, binding proteins, receptors or portions of receptors, extracellular matrix proteins, cell surface molecules, antigens, haptens and chelating agents.
In general, TNFbp useful in the present invention has the sequence of human TNFbp or its truncated variants or muteins. One TNFbp, named 30 kDa TNFbp, is the extracellular portion of the p55 TNF receptor. In vivo, the extracellular portion of the receptor falls off and circulates in the bloodstream as a 30 kDa glycosylated protein that binds to TNF. The purification of TNFbp and the sequences of its amino acids and nucleic acids are described in EP393438 and EP422339. The contents of these patents are hereby incorporated by reference. These documents also teach the recombinant production of a 30 kDa TNFbp glycosylated and deglycosylated form. The actual molecular weight of the deglycosylated form of this inhibitor is approximately 18 kDa, but the term “30 kDa TNF inhibitor” encompasses both glycosylated and deglycosylated forms.
The purification of another TNF inhibitor called 40 kDa TNFbp and the sequence of amino acid and nucleic acid sequences are also described in EP393438 and EP422339. This inhibitor is in its naturally produced form the glycosylated extracellular portion of the p75 TNF receptor. Although the molecular weight of the deglycosylated form is not 40 kDa, both the glycosylated and deglycosylated form of TNFbp are referred to as “40 kDa TNF inhibitors”. These references also teach recombinant production of glycosylated and deglycosylated forms of 40 kDa TNFbp.
Reactive thiol groups can be introduced into biologically active molecules when needed. Thiol groups can also be introduced into inert molecules to form biologically active molecules as long as the thiol-sulfone linkage does not substantially destroy the desired activity.
Reactive thiol groups can be introduced by chemical means known in the art. Chemical modification can be used with polypeptide or non-peptide molecules, introducing a thiol into the molecule either alone or as part of a larger group, eg, a cysteine residue. An example of chemical introduction of thiols is Jue, R et al., Biochemistry,17: 5399-5406 (1978). Free cysteines may also be generated in polypeptides by chemically reducing cysteines, for example with DTT.
To produce a mutein containing cysteine, non-essential amino acids may be substituted with cysteine or a cysteine residue may be added to the polypeptide. A polypeptide that has been modified to include an amino acid residue at a position that did not exist in the native protein prior to modification is termed “mutein”. Non-naturally occurring cysteine can be introduced at the glycosylation site of the polypeptide and at the N-terminus or C-terminus. Since lysine residues are often found on the surface of proteins in the active conformation, lysine to cysteine mutations are also suitable. One skilled in the art can further utilize any known information regarding the binding or active site of a polypeptide to select possible mutation sites.
One skilled in the art can also use known recombinant DNA techniques to produce muteins containing cysteine. Nucleic acids encoding native polypeptides can be modified to encode muteins by standard site-directed mutagenesis. Examples of standard mutagenesis techniques are described in Kunkel, T .; A. , Proc. Nat. Acad. Sci. ,82: 488-492 (1985) and Kunkel, T .; A. Et al., Methods Enzymol. ,15467-382 (1987). The contents of both documents are included in the present invention by reference. Alternatively, nucleic acids encoding muteins can be chemically synthesized by techniques known in the art. A DNA synthesizer can be used and can be obtained, for example, from Applied Biosystems (Foster City, CA). Nucleic acids encoding the desired muteins can be expressed in various expression systems such as animal, insect and bacterial systems.
The linkage between the biologically active molecule having a reactive thiol group and the sulfone activated polymer linker is a covalent bond, and a universal method for producing such conjugates includes the following steps.
(1) selecting a desired biologically active molecule;
(2) synthesize the desired sulfone activated polymer linker;
(3) purify the sulfone activated polymer linker using HIC chromatography;
(4) reacting the HIC purified sulfone activated polymer linker with a biologically active molecule;
(5) isolating the reaction product using chromatographic techniques known in the art, then
(6) The biological activity of the formed conjugate is measured using an appropriate bioassay.
Already produced 30 kDa TNFbp muteins include c105 30 kDa TNFbp; c1 30 kDa TNFbp; c14 30 kDa TNFbp; c111 30 kDa TNFbp; and c161 30 kDa TNFbp. “C” means cysteine, and “c105” means that cysteine was inserted at position 105. These muteins are produced by the procedure described in International Patent Publication WO 92/16221, which is included in the present invention by reference, and the amino acid number is also based on the amino acid sequence described in the patent.
GB 2,246,569 teaches a “3-domain” 30 kDa and / or 40 kDa TNFbp, ie a deletion of the fourth domain. U.S. Provisional Application No. 1 filed July 9, 1995. 60/021443 teaches a “2.6 domain” 30 kDa TNFbp, ie, a portion of the third and fourth domains deleted. In addition, US Patent Provisional Application No. Teaches a “2.6 domain” of 30 kDa and / or 40 kDa TNFbp, ie, a portion of the third and fourth domains deleted.
The IL-1ra muteins that have already been produced include c0s116 IL-1ra; c84s116 IL-1ra; c8s116 IL-1ra; c9s116 IL-1ra and c141s116 IL-1ra. The production and purification of IL-1ra muteins is also disclosed in International Patent Publication No. WO 92/16221. Residue numbers are based on the sequences described in this published application. “0” represents addition to the N-terminus of an amino acid. “C” represents cysteine and “s” represents serine. For example, “c0s116” means that cysteine was added to the N-terminus and serine was inserted at position 116. Natural IL-1ra has free cysteine residues at positions 66, 69, 116 and 122.
Pharmaceutical compositions containing many conjugates or compounds of the present invention (collectively referred to as “conjugates”) may be prepared. These conjugates can be present in a pharmaceutically acceptable carrier to form a pharmaceutical composition of the invention. The term “pharmaceutically acceptable carrier” as used herein refers to a non-toxic, usually inert, active ingredient vehicle that does not cause side effects in patients receiving the active ingredient or composition. Suitable vehicles or carriers are described in standard pharmaceutical literature, for example, “Remington's Pharmaceutical Sciences”, 16th Ed. , Mack Publishing Co. , Easton, PA (1980). The contents of this document are included in the present invention by reference. Such carriers are, for example, aqueous solutions such as carbonate buffer, phosphate buffer, Ringer's solution and physiological saline. In addition, the carrier contains other pharmaceutically acceptable excipients to correct or maintain the pH, osmolarity, viscosity, clarity, color, sterility, stability, dissolution rate or odor of the formulation. May be contained.
The pharmaceutical composition can be prepared by methods known in the art, such as simple mixing of reagents. Those skilled in the art will appreciate that the choice of pharmaceutical carrier and the proper preparation of the composition will depend on the intended use and mode of administration.
According to one embodiment, the carrier and conjugate constitute a physiologically compatible sustained release formulation. The primary solvent in such a carrier may be either aqueous or non-aqueous. In addition, the carrier may further contain other physiologically acceptable excipients to correct or maintain the stability, dissolution, release or absorption rate of the conjugate. Such excipients are conventional substances commonly used to prepare pharmaceuticals in parenteral dosage forms in single or multiple dose forms.
After preparing the pharmaceutical composition, the pharmaceutical composition can be stored in sterile vials such as solutions, suspensions, gels, emulsions, solids, or dehydrated or lyophilized powders. Such formulations may be stored in a form that can be used as is, or may be stored in a form that requires reconstitution immediately prior to administration. The preferred storage temperature of such formulations is at least 4 ° C. or less, preferably −70 ° C. Also, such formulations containing the conjugate are preferably stored and administered at or near physiological pH. Currently, administration of formulations at high pH (ie, values above 8) or low pH (ie, values below 5) is considered undesirable.
Methods for administering a formulation containing the conjugate for systemic delivery include subcutaneous, intramuscular, intravenous, oral, intranasal, or other suppositories for intravaginal or rectal administration. Methods for administering a formulation containing the conjugate for the purpose of local delivery include intra-articular, intratracheal, infusion, or intra-airway inhalation. Furthermore, it may also be desirable to administer the conjugate to a specific portion of the gastrointestinal tract by oral administration or device of the conjugate in a suitable formulation.
For oral administration, the conjugate is encapsulated. Encapsulated conjugates can be formulated with or without the addition of conventional pharmaceutically acceptable carriers used in the preparation of solid dosage forms. Preferably, the capsule is designed so that the active portion of the formulation is released at the location of the gastrointestinal tract where bioavailability is greatest and pre-system degradation is minimized. Additional excipients that facilitate the absorption of the conjugate may be incorporated. Diluents, flavorings, low melting point waxes, vegetable oils, lubricants, suspending agents, tablet disintegrating agents, and binders may also be used.
Regardless of the dosage form, individual dosages are calculated based approximately on the patient's weight. Other factors that determine the proper dosage are the type of disease or disorder to be treated or prevented, the route of administration, the age, sex and condition of the patient. In some embodiments, dosages and dosage forms are planned such that the conjugate is in a preselected concentration range in the patient's bloodstream. For example, it is considered that an effective composition cannot be obtained by maintaining circulation of a TNF inhibitor or IL-1 inhibitor at a concentration of less than 0.01 ng per ml of plasma. Conversely, maintaining a level of circulation above 10 μg per ml for an extended period of time will cause undesirable side effects. The method of making the calculations necessary to determine the appropriate dosage for treatment with each of the above formulations is a routine task of the average person skilled in the art, especially information regarding dosages disclosed herein. And a range of operations that can be routinely performed by those skilled in the art without special experimentation with reference to the assay. These dosages can be ascertained by using established dosage determination assays with appropriate volume-response data.
It should be noted that the conjugate formulations described herein can be used for both veterinary and human pharmaceutical applications, and that the term “patient” should not be construed restrictively. When used as a veterinary medicine, the dosage range is the same as above.
The conjugates of the invention can be used in a variety of applications, including non-limiting examples such as in vitro diagnostic assays, production of pharmaceutical compositions, and the like. Conjugates containing TNFbp, IL-1ra, alone or in combination with multiple conjugates, adult respiratory disorder syndrome, pulmonary fibrosis, arthritis, psoriasis, myeloid and other leukemias, ischemia / reperfusion, It can be used to treat diseases such as asthma, cachexia / anorexia, diabetes, septic shock, inflammatory bowel disease, multiple sclerosis, stroke, Crohn's disease, transplant rejection, hemorrhagic trauma. Conjugates can also be used for many in vitro uses such as affinity chromatography (partition) and diagnostic purposes well known to those skilled in the art. For example, such conjugates can be attached to a solid surface and used to purify or detect a ligand that specifically binds to a receptor bound to an activated linker.
The following examples illustrate various features of the present invention in more detail.
Example 1
This example describes the preparation of an activated linker 20 kDa PEGbv using a “one-step” synthesis procedure.
“One-step” synthesis of 20 kDa PEGbv
Polyethylene glycol (PEG-DIOL^TM) Shearwater Polymers Inc. (Birmingham, AL). Divinyl sulfone, 1 M potassium t-butoxide (in tetrahydrofuran (THF)), toluene (anhydrous, seal), di-i-propyl ether (99 +%, inhibited with 25 ppm butylated hydroxytoluene (BHT)), Tetrahydrofuran (anhydrate containing 0.025% BHT as an inhibitor, true seal) was obtained from Aldrich Chemical Company (Milwaukee, Wisconsin). Dichloromethane (Omni Solvent) and diethyl ether (EM) were purchased from VWR Scientific Company (Denver, Colorado). Anhydrous sodium sulfate powder is described in J. Org. T.A. Purchased from Baker (Phillipsburg, New Jersey).
A thermometer, a condenser, a rubber septum, and a mechanical variable speed stirrer are attached to a round bottom flask with a capacity of 4 liters (45/50 for one mouth and 24/40 joint for 3 necks). It was immersed in a water bath containing a water immersion heater. The flask was maintained under positive argon pressure. The condenser was connected to a two-necked round bottom flask and a vacuum line was connected from this flask. Polyethylene glycol (20 kDa, 300 g, 15 mmol) was weighed and transferred via a funnel into a 22 liter flask and 3 liters of anhydrous toluene was introduced through the cannula while stirring the mixture. The water bath temperature was set at 45 ° C. and stirring was continued until the solution became clear (˜1 hour). The solvent was evaporated to dryness under vacuum (˜60 mm Hg). The solid residue was dissolved in 9 liters of anhydrous tetrahydrofuran (added via cannula). The water bath temperature was set at 25 ° C. Stirring was continued until the temperature of the solution reached 25 ° C. Divinyl sulfone (6 ml, 59.97 mmol) was added via syringe over 15 minutes. The mixture was stirred for 10 minutes (75 rpm). Potassium t-butoxide (1M in THF, 3 ml, 3 mmol) was added via an addition funnel over 25 minutes.
After the addition of the base, the reaction mixture was stirred at 25 ° C. for 6 hours. The water bath was then heated to 40 ° C. to evaporate the THF (350 mm Hg). 1 liter of a 1: 1 mixture of dichloromethane and tetrahydrofuran was added to the solid residue and stirred until a clear solution was formed. The solution was cooled to 25 ° C. and quenched (−20 ° C.). Di-i-propyl ether (1 x 5 liters) was added to the solution from the addition funnel with continued stirring. The condenser was removed from the flask. A filtration pressure funnel was connected to the flask with a flexible needle and the precipitate was transferred from the flexible needle and filtered through a filtration funnel under vacuum (1 liter, 70-100 μ). The precipitate was divided into two portions and transferred to a 2 liter Erlenmeyer flask and stirring (magnetic stirrer) with 1 liter (each) of i-propyl ether was continued for 3 hours. The precipitate was filtered (2 liters, 70-100μ), washed with 800 ml diethyl ether, placed in a dryer and dried under vacuum (5 mm Hg) for 72 hours. The measured yield was 278.6 g (91.6% recovery based on the weight of polyethylene glycol).
The product was transferred to an Erlenmeyer flask (2 liters) and 1 liter of dichloromethane (Omni Solvent) was added. After the product was completely dissolved, the solution (syrup form) was transferred to a 5 liter drop bottom flask equipped with a mechanical stirrer. The Erlenmeyer flask was washed with another 500 ml of dichloromethane and the solution was added to the drop bottom flask. The solution was then stirred (65 rpm) for 5 minutes. Saturated sodium chloride (1 liter) was added and stirred for 5 minutes (75 rpm). The mixture was allowed to stand until the two layers were clearly separated. The organic layer was removed and stored in a separate container. This process was repeated three times and all organic layers were combined.
The solution is then divided into two aliquots (3 liters each), transferred to two Erlenmeyer flasks (4 liters), and magnetically combined with 500 g of anhydrous sodium sulfate powder (JT Baker, Phillipsburg, New Jersey). Stir at room temperature for 16 hours with a stirrer (the flask is completely covered with aluminum foil throughout the process). The solution was filtered through a filter funnel (2 liters, 70-100 μ) and the filtrate was transferred to a 22 liter four-necked flask maintained in a water bath (40 ° C.). The solvent was evaporated under vacuum (285-360 mbar) and the residue (highly viscous liquid) was redissolved in a 1: 1 mixture (1 liter) of dichloromethane and THF. The solution was then precipitated as before, filtered and dried. The measured yield was 221.0 g (72.67% recovery based on the weight of polyethylene glycol).
Example 2
This example describes a series of loading and elution experiments designed to test the binding ability of Butyl, Phenyl and Ether HIC resins to 20 kDa PEGbv.
Assay used
1. PEG assay
PEG-containing fractions were detected by Nessler's reagent (Sigma / Aldrich). Typically, 0.01 ml of Nessler's reagent is added to a 0.25 ml aliquot of the column fraction placed in a glass test tube. Upon addition of the Nessler reagent, the PEG containing fraction formed a milky precipitate.
2. PEG quantitative analysis
Nessler's reagent (1.5 ml) was added to a glass test tube, followed by 18.75 microliters of the test sample. The tube was rotated and allowed to stand for 10-15 minutes, and the turbidity at 600 nm was read (Beckman DU 70 spectrophotometer) compared to a blank containing only the Nessler reagent. PEG was then quantified from standard curves generated from 1.5, 1.0, 0.75, 0.50 and 0.25 mg / ml PEG standards.
Experimental procedure
5ml pre-packed ToyoPearl^{▲ R ▼}Butyl, Phenyl and Ether 650C HIC columns (TosoHaas, Montgomeryville, PA) were used. 10 mg (2 mg / ml bed volume) of 20 kDa PEGbv in various concentrations of NaCl or sodium sulfate was retained on the column and eluted with water using gravity. All experiments were performed at room temperature and 0.5-1.0 ml fractions were collected manually into 13 × 100 mM glass test tubes. PEG was measured by Nessler's assay. Table 1 shows the salt concentration required to fully bind 20K PEGbv to the various HIC resins (ie, there is no detectable PEGbv in the flow-through).

As shown in Table 1, Butyl 650C HIC resin has a slightly higher affinity for PEG than Phenyl 650C HIC resin. Also, Ether 650C HIC resin did not bind to PEG retained at a concentration of 2M NaCl or 0.5M sodium sulfate.
Example 3
This example describes an experiment designed to test the ability of Butyl HIC resin to separate various PEG diols based on size. PEG-containing fractions were determined using the PEG assay described above and the molecular weight of PEG was confirmed by SEC HPLC.
Assay used:
1. Size exclusion chromatography (SEC) HPLC assay
Size exclusion chromatography of the HIC purified PEG fraction was performed using Toso-Haas G3000._swxlThis was performed using a column (Synchrom, Inc., Linden, NJ). 100-200 μl samples are injected at 0.5-1.0 mg / ml. The column was maintained at room temperature and equilibrated in 10 mM sodium phosphate buffer, 150 mM NaCl, 0.004% sodium dodecyl sulfate (SDS), pH 6.5. PEG was eluted from the column using a buffer containing 10 mM sodium phosphate, 150 mM NaCl, and 0.004% SDS, pH 6.5. In some experiments, elution of PEG was monitored using a Refractive Index detector (detecto) (Bio-Rad, Hercules, CA). In other experiments, PEG was first derivatized with o-thiobenzoic acid so that the eluate could be monitored by UV absorbance at 254 nm.
Experimental operation:
ToyoPearl^{▲ R ▼}  A Butyl 650M HIC column (1.5 × 11 cm) was loaded with a total of 3 mg PEG diol per ml column bed volume (bed volume). The column was equilibrated in 4M NaCl. The introduction was 9.7 ml in volume and contained 2.97 kDa, 11.9 kDa, 20 kDa and 50 kDa PEG diols in amounts of 1.5 mg / ml each (2.97 kDa and 11.9 kDa PEG diols were purchased from Polymer Laboratories. , 20 kDa and 50 kDa PEG diols were purchased from Shearwater Polymers). After introduction, the column was washed with 2 column volumes of 4M NaCl and eluted with 15 column volumes of linear gradient 3 → 1M NaCl. The flow rate was 2.5 ml / min during the introduction, washing and elution steps. Three minute (7.5 ml) fractions were collected and analyzed for PEG. The elution profile is shown in FIG. The molecular weights of the four peaks confirmed by size exclusion HPLC chromatography are as shown in the figure.
To complete the initial experiments described above, PEGbv eluted in 2M NaCl flow-through and PEGbv eluted in the water eluate of the Phenyl 650C column described in Example 2 were analyzed by SEC HPLC assay. . From the SEC HPLC data, it was found that the material in the 2M NaCl flow-through was rich in 20 kDa PEGbv, while the material in the water eluate contained mainly high molecular weight PEGbv. Such data confirms that PEG diol and 20 kDa PEGbv can be separated based on size using HIC, especially Phenyl and / or Butyl HIC.
Example 4
This example tests various 20 kDa PEGbv introduction conditions and describes a series of HIC experiments using various linear NaCl concentration gradients for elution of PEGbv from Butyl or Phenyl HIC resins. All experiments were performed at room temperature and usually as described in Example 2.
First experiment:
In the first experiment, ToyoPearl^{▲ R ▼}  A Phenyl 650S HIC column (1.5 × 6.5 cm) was used. The 650S resin differs from the 650C resin in that the particle size is 20-50 μm instead of 50-150 μm. The column was equilibrated in 3M NaCl and 2 mg PEGbv per ml bed volume was introduced into the column. PEGbv was eluted using a linear concentration gradient NaCl (3 → 0 M NaCl) 10 times the column volume at a flow rate of 1.5 ml / min. A 2 minute (3.0 ml) fraction was collected. The elution profile is shown in FIG. The yield (%) of 20 kDa PEGbv was then measured using the assay described in Example 3 (see Table 2).
Second experiment:
In the second experiment, ToyoPearl^{▲ R ▼}  A Butyl 650M HIC column (1.5 × 11.4 cm) was used. The 650M resin differs from the 650C resin in that the particle size is 40-90 μm instead of 50-150 μm. The column was equilibrated in 2M NaCl and 2 mg PEGbv per ml bed volume was introduced into the column. PEGbv was eluted using a linear concentration gradient NaCl (2 → 0 M NaCl) 10 times the column volume at a flow rate of 1.5 ml / min. A 3 minute (4.5 ml) fraction was collected. The elution profile is shown in FIG. The yield (%) of 20 kDa PEGbv was then measured using the assay described in Example 3 (see Table 2).
Third experiment:
In the third experiment, ToyoPearl^{▲ R ▼}  A Butyl 650M HIC column (3.2 × 18.0 cm) was used. The column was equilibrated in 2.5 M NaCl and 1.8 mg PEGbv per ml bed volume was introduced into the column. PEGbv was eluted using a linear concentration gradient NaCl (2 → 0.5 M NaCl) 10 times the column volume at a flow rate of 9.6 ml / min. A 2.5 minute (24 ml) fraction was collected. The yield (%) of 20 kDa PEGbv was then measured using the assay described in Example 3 (see Table 2).
Fourth experiment:
In the fourth experiment, ToyoPearl^{▲ R ▼}  A Butyl 650M HIC column (1.5 × 11.5 cm) was used. The column was equilibrated in 5M NaCl and 8 mg PEGbv per ml bed volume was introduced into the column. The amount of introduction per unit bed volume was increased for the purpose of getting closer to large-scale operating conditions. Due to the increased amount introduced, higher NaCl concentrations (5M vs 2M) were required for PEGbv to fully bind to the column. PEGbv was eluted using a linear concentration gradient NaCl (3.5 → 1 M NaCl) 10 times the column volume at a flow rate of 2.0 ml / min. A 1.5 minute (3 ml) fraction was collected. The yield (%) of 20 kDa PEGbv was then measured using the assay described in Example 3 (see Table 2).
Experiment 5:
In the fifth experiment, ToyoPearl^{▲ R ▼}  A Butyl 650M HIC column (1.5 × 11.5 cm) was used. The column was equilibrated in 5M NaCl and 8 mg PEGbv per ml bed volume was introduced into the column. PEGbv was eluted using a linear gradient NaCl (2 → 0.5 M NaCl) 15 times the column volume and a flow rate of 2.1 ml / min. Fractions were collected. The elution profile is shown in FIG. The resolution profile of this experiment was not as good as that obtained with the 2 mg / ml introduction. However, the measured SEC HPLC purity obtained from this experiment is as shown in FIGS. The data in FIGS. 4 and 5 demonstrate that high molecular weight PEGbv was removed as appropriate despite the low resolution.
Experiment 6:
In the sixth experiment, ToyoPearl^{▲ R ▼}  A Butyl 650M HIC column (7.0 × 14 cm) was used. The column was equilibrated in 4.5M NaCl and 8 mg PEGbv per ml bed volume was introduced into the column. PEGbv was eluted using a linear concentration gradient NaCl (3.5 → 1.0 M NaCl) 17 times the column volume at a flow rate of 55 ml / min.

As the above data shows, the use of linear concentration gradient NaCl in Phenyl and / or Butyl HIC separation is effective in separating high molecular weight PEGbv from 20 kDa PEGbv. In addition, yields of about 65-75% were obtained by using 2 mg or 8 mg loading per ml of bed volume under appropriate conditions. Finally, the above data show that the experiment with the introduced amount of 8 mg / ml does not give a better resolution profile than the experiment with the introduced amount of 2 mg / ml, but the high molecular weight PEGbv can be removed as appropriate. ing.
Example 5
This example describes the preparation of a biologically active conjugate (Homone Bell) containing TNFbp and HIC purified 20 kDa PEGbv. The total yield, purity and biological activity of the conjugate is compared to that of a conventional TNFbp conjugate.
Conjugate production:
The c105 30 kDa TNFbp mutein used in this example can be prepared as disclosed in WO 92/16221, incorporated herein by reference. Alternatively, the c105 mutein is produced as follows. E. coli cells that express the c105 mutein are recovered by centrifugation. The cell sludge is adjusted to about 40% wet weight solids by the addition of purified water. The mixture is then further diluted with an equal volume of breaking buffer (50 mM tromethamine, 4 mM EDTA, pH 7.2), thereby obtaining a suspension containing about 20 wet weight percent solids. The cell sludge is passed 5 times through a high pressure homogenizer operating at about 8,000 psi to produce a cell homogenate. When passing through a homogenizer, the homogenate is cooled to 10 ° C. or lower in advance every time. The homogenate is centrifuged to obtain a solid fraction containing c105. The solid is diluted and centrifuged again to obtain washed inclusion bodies.
The washed inclusion bodies are then lysed by the addition of 8M urea in 50 mM Tris, pH 9.5 and 150 mM cysteine. The mixture is stirred for 2 hours at room temperature before regeneration. Under the above conditions, c105 mutein is denatured and reduced.
Reduced denatured c105 mutein is regenerated by dilution with 1.1 M urea, 50 mM Tris, whereby 200 μg / ml c105 mutein, 1.5 M urea, 7.5 mM cysteine, 50 mM Tris, pH 9.7 A final regeneration solution consisting of is obtained. The regeneration mixture is maintained at 6-10 ° C. for 2 days. Regeneration efficiency is monitored by reverse phase HPLC and cation exchange HPLC.
The regeneration mixture is then brought to pH 5.0 by addition of acetic acid and HCl. The regeneration mixture is introduced onto a cation exchange column (S-Sepharose large bead resin) pre-equilibrated in 25 mM sodium acetate, 65 mM NaCl at 4 ° C. After introduction, the column is washed with the same equilibration buffer as above. The column is eluted with a gradient of 65 → 350 mM NaCl in 25 mM sodium acetate, pH 5. The c105 mutein elutes at a NaCl concentration of about 200 mM and is collected in one pool.
The recovery pool containing the c105 mutein was diluted with 1.5 volumes of 5M NaCl, 40 mM sodium phosphate (adjusted to pH 6), which was pre-equilibrated in 3M NaCl, 20 mM sodium phosphate, pH 6 Introduce into working column (Toyo Butyl 650M column). After the introduction is complete, the column is washed with equilibration buffer. The c105 mutein is eluted using a linear salt concentration gradient 3 → 1M NaCl in 20 mM sodium phosphate, pH 6 at 8 column volumes. Collect c105 muteins in one pool. The pool is concentrated to about 3 g / l of c105 mutein and then membrane filtered (approximately 6 volumes) against 20 mM sodium phosphate, pH 6, until the final conductivity is less than 4 mmho.
The membrane filtered pool is introduced onto a SP-Sepharose high speed column equilibrated in 20 mM sodium phosphate, pH 6. After introduction, the column is washed with additional equilibration buffer and has an eluent with a pH / salt gradient combination from 20 mM sodium phosphate, 50 mM NaCl, pH 6.0 to 20 mM sodium phosphate, 50 mM NaCl, pH 6.5. Elute with. In the latter half of the gradient, c105 mutein elutes at a NaCl concentration of about 35 mM. At this point, the c105 mutein can be stored frozen.
The 20 kDa PEGbv used in this example was synthesized using the “one-step” operation described in Example 1 above. A portion of the synthesized 20 kDa PEGbv was HIC purified as described in the fifth experiment of Example 4. Next, a PEGbv: TNFbp conjugate was prepared using both 20 kDa PEGbv that was not HIC purified and 20 kDa PEGbv that was HIC purified. The conjugate was produced as follows. c105 30 kDa TNFbp mutein was exposed to a 1.3-fold molar excess of DTT in 20 mM phosphate, pH 7.6 at 15 ° C. for 5-6 hours, thereby extra cysteine bound to c105 during the regeneration process Was removed. The pH was adjusted to 6.0 and the sample was introduced (in an amount of 4-6 mg / ml bed volume) onto an S-Sepharose Fast Flow column and 3 column volumes of 20 mM phosphate buffer, pH 6.0. Wash (thus removing DTT) and elute with 20 mM phosphate, 70 mM NaCl, pH 6.0.
Next, reduced c105 30 kDa TNFbp mutein solution (˜4.7 mg / ml) was adjusted to pH 7.4-7.5 with 1M NaOH. To this reduced c105 TNFbp, HIC purified or non-HIC purified 20 kDa PEGbv (0.9 mg / mg TNFbp) was added under stirring. After stirring and mixing, the solution was left at ambient temperature for 12-18 hours.
Conjugate assay:
PEGylation assay
Conversion of TNFbp to 20 kDa PEGbv: TNFbp dumbbell, POROS^TM  Quantification was performed using an HS / H 4.6 / 50 cation exchange HPLC column (Perseptive Biosystems, Cambridge, Mass.). A 50 μl sample was introduced onto the column and eluted from the column using a 10 minute gradient 0 → 1 M NaCl in 20 mM sodium acetate, pH 3.75. Representative elution profiles for c105 30 kDa TNFbp mutein monobell, ie, PEGbv, c105 30 kDa TNFbp mutein dumbbell, c105 30 kDa TNFbp mutein dumbbell attached to one end only, and a representative elution profile for unPEGylated c105 30 kDa TNFbp mutein sample. The c105 30 kDa TNFbp mutein monobell eluted first, followed by the c105 30 kDa TNFbp mutein dumbbell and the non-PEGylated c105 30 kDa TNFbp mutein.
result
When the HIC purified material was used, the conversion rate to the TNFbp dumbbell was 66.7%, and there was no detectable “pre-peak” (see FIG. 7). Non-HIC purified PEGbv from the same lot yielded only 52.9% conversion and had a “pre-peak” of 6.7% (see FIG. 8). This prominent “pre-peak” indicates a TNFbp dumbbell generated by high molecular weight PEGbv impurities.
The PEGylated mixture was then loaded onto a 1.5 × 10 cm SP-Sepharose HP column equilibrated in 50 mM sodium phosphate, pH 3.1 (4 mg / ml bed volume). After introduction, the column is washed with equilibration buffer, after which eluent with a gradient from 50 mM sodium phosphate, 0.25 M NaCl, pH 3.1 to 50 mM sodium phosphate, 0.5 M NaCl, pH 3.1 is added to the column volume. Was used to elute the TNFbp dumbbell. The elution profile is shown in FIGS. Pools were formed in several different ways for each eluate. Table 3 shows the yield (%) and purity (%) of TNFbp dumbbells for the various pooling criteria used.

The data in Table 3 shows that HIC purification of PEGbv is clearly beneficial for the purity and yield of the later obtained TNFbp dumbbell. For example, when trying to achieve a purity of 97%, only 37% TNFbp dumbbells could be recovered on the SP-Sepharose HP column when using non-HIC purified PEGbv, but as much as 58% when using HIC purified material In other words, the recovery amount of TNFbp dumbbell can be increased by 1.5 to 1.6 times by HIC purification of 20 kDa PEGbv.
Example 6
In this example, the ability of the HIC resin to separate PEG based on end group functionality was evaluated.
The “early eluting” material obtained in the third experiment of Example 4 was tested in a PEGylation assay. “Early eluting” material is the fraction that eluted immediately before the fraction pooled to obtain 20 kDa PEGbv. This material resulted primarily in TNFbp monobell when used in the manufacture of conjugates, as shown in FIG. This indicates that the “early eluting” material contains primarily PEG 20 kDa monovinyl sulfone, ie, the ability of Butyl HIC resin to separate 20 kDa PEGbv based on end group functionality.
Example 7
In this example, the stability of the linkage between c105 TNFbp mutein and PEG-bis-vinylsulfone was tested. A known amount of c105 TNFbp dumbbell was incubated for 1 week at 37 ° C. in PBS, pH 7.4, during which time aliquots were removed for analysis by SDS PAGE. Substantially no degradation of the c105 TNFbp dumbbell was observed. Only 5-10% conjugate degradation was observed after 1 week incubation at 37 ° C. at pH 10. In another stability test, an activated PEG linker prepared by the one-step method of Example 1 was prepared by the multi-step method disclosed in WO 95/07555, specifically incorporated herein by reference. Compared to the activated PEG linker.
First, one-step and multi-step activated linkers were derivatized with 2-mercaptobenzoic acid, whereby the linker stability was measured using UV. 3.0 ml of 1 mM 2-mercaptobenzoic acid (orthothiolbenzoic acid) was added to each 0.5 ml of linker solution (produced by dissolving 20 mg of linker in 0.5 ml of 50 mM sodium phosphate, pH 7.5). The mixture was mixed for a short time by vortex generation. The solution was then incubated at room temperature for over 18 hours. The incubation time at room temperature can be relatively long (up to 48 hours).
The derivatized linker solution was then diluted 1:10 with PBS and heat treated at 90 ° C. for 1 hour. Keep the heat-treated linker solution away from heat and add 100-200 μl of the solution to BIO SIL^TM  A 250-5 column (BioRad, Hercules, CA) was injected and flowed into 10 mM sodium phosphate, 150 mM NaCl, 0.004% SDS, pH 6.5 at a flow rate of 0.5 ml / min. The collected sample was subjected to 254 nm O.D. D. At UV. Tables 4 and 5 show the UV measurement results. In these tables, the product UV peak change rate (%) is calculated based on the difference between the standard UV purity without heat treatment and the purity of the 90 ° C. heat treated sample. Purity was calculated as the quotient obtained by dividing the average peak area by the total peak area. The variability of this assay was about 2%.

As is apparent from a comparison of the results shown in Tables 4 and 5, the one-step linker is more stable than the multi-step linker. The results also show that the change in stability of the one-step linker is within the range of 2% assay variability, while the change in stability of the 4-step linker exceeds the assay variability.
The foregoing description of the invention is an example for purposes of illustration and description. It will be apparent to those skilled in the art that the present invention can be changed and modified without departing from the spirit and scope thereof. All such changes and modifications are to be construed as being encompassed by the following claims.

Claims (6)

A method for purifying polyethylene glycol (PEG) diol or PEGbv , comprising:
(A) applying the PEG to a hydrophobic interaction chromatography (HIC) column, wherein the hydrophobic side group selected from the group consisting of octyl, butyl, phenyl and ether is hydrophilic. Using a polymer resin located on the surface of the resin having a vinyl copolymer backbone;
(B) The method comprising the step of selectively isolating a given form of PEG using reverse linear salt gradient elution conditions. The method of claim 1, wherein the HIC comprises a butyl or phenyl ligand on the methacrylic acid backbone. The method of claim 1, wherein the polyethylene glycol has a molecular weight of 5 kDa to 100 kDa. The method of claim 1, wherein the polyethylene glycol has at least one active sulfone moiety. The method of claim 4 , wherein the active sulfone moiety is vinyl sulfone. The method of claim 4 , wherein the polyethylene glycol is polyethylene glycol-bis-vinylsulfone.

Images