穿越時空，與五位科學家共話合成生物學

前言：雖然題目中是穿越，但是我們穿越的時間並不是很長，我們將回到2014年，也就是3年前。3年似乎並不是很長的時間，但是對於合成生物學這門新興的學科來說卻是一段發展的黃金時期。在這幾年裡，合成生物學快速發展。2014年CRISPR-Cas9技術似乎剛剛開始在全球受到關注，而如今這項技術幾乎已經成為一項每個分子實驗室必備的技術，同時新型的基因編輯技術也在一個接一個的被開發出來。

在這幾年裡，合成生物學的設計能力不斷增加，不僅僅是在複雜的基因電路方面，在基因組層面，我們也有了長足的進步和鼓舞人心的進展，如Craig Venter的Synthia 3.0，酵母基因組計劃Sc2.0，人類基因組合成計劃——HGP-Wirte；合成生物學的應用也在不斷的增加，基因編輯，天然產物，癌症療法，器官移植......

但是把一篇舊聞翻出來有什麼用呢？首先這篇文章中的訪談人物是在合成生物學領域做出過突出貢獻的頂尖科學家。同時我認為即使過去了3年，在這篇報道中5位科學家所提到的合成生物學所面臨的機遇與挑戰仍然對我們有指導意義。這也算是品味經典吧。我希望通過這篇文章可以讓大家了解到這個領域頂尖的科學家的正在思考什麼，也希望讓各位能夠進一步了解合成生物學。

文末有5位科學家的簡要介紹。

同時真心感覺我翻譯的水平不是很高，各位如果覺得中文翻譯蹩腳，可以直接閱讀後面的英文。

歡迎大家轉發，與我一起討論，並提出批評意見~

----------------------------------------------------------------------------------------------

儘管還在襁褓中成長，合成生物學正在為生物醫療、生物技術和基礎分子研究持續注入源源不斷的動力。從細胞生物學家到工程師，這個綜合性的研究團體正在探索著合成生物學的巨大潛力，同時他們也在應對著合成生物學所帶來的種種挑戰。2014年，Nature出版社對合成生物學領域5位頂尖科學家進行了一次訪談，在這次訪談中，五位科學家就合成生物學的未來，合成生物學在基礎和應用科學的主要成就，以及合成生物學帶來的生物倫理問題分享了他們的觀點和見解。

問：現在越來越多的的出版物和研究機構正在關注合成生物學。從最初關注合成基因電路的設計開始，該領域是如何進一步發展的？今天我們所談的合成生物學與其他的學科有什麼關係呢？比如系統生物學和數學建模。

Ron Weiss. 簡單合成基因電路設計如雙穩態開關、震蕩器和人造細胞通訊系統在細菌中的成功代表著合成生物學的正式創立。通過使用非轉錄調控邏輯包括microRNAs，蛋白質磷酸化和DNA編輯技術，合成生物學的關注點已經從小型轉錄調控網路轉向了更加複雜的多細胞系統諸如酵母、哺乳動物細胞和植物。詳細、精確的刻畫與可預測的元件設計已經成為構建具有複雜多信號輸入邏輯功能、複合多態基因電路的關鍵。這使得細胞感測器（Sensor）和執行器（Actuator）的能力得以提高，多樣性也有所增加。

我認為，合成生物學和系統生物學是一種互補的關係：前者基於正向工程學，而後者基於反向工程學。系統生物學對自然過程的研究成果可以提高合成生物學對生物系統的設計能力，同時合成生物學創建的小型人工網路有助於檢驗系統生物學對自然系統提出的假設。計算機建模工具則成為人工調控網路設計中的必要條件，同時它也在多種生物學領域發揮著重要的作用。

最後，合成生物學研究人員對生物系統複雜性的認識也在逐漸提高。生物本身的複雜性讓我們認識到我們需要跨學科交叉性研究、新的基因電路設計原理以及程序設計範式來克服合成生物學所遇到的問題和瓶頸，例如代謝負荷（metabolic load），干擾（crosstalk），資源分配（resource sharing）和基因表達雜訊（gene expression noise）。

George M. Church. 在我的觀點中，合成生物學從來沒有完全專註在遺傳電路設計上，而是專註在讓生物快速成熟為工程學科的領域，包括計算機輔助設計，安全系統構建，整合模型，基因組編輯和生物進化。合成生物學不完全像高度模塊化（或「開關式」）的電氣工程和計算機科學，其更像是土木工程和機械工程。

Michael B. Elowitz. 通常來說，合成生物學將生物學從現有物種和細胞系統的空間擴展到更大的非自然但可行的物種和系統空間。雖然一開始我們通過基因電路來實現一些最簡單的動態行為，但是目前合成生物學研究方法已廣泛的應用到從代謝到多細胞發育等各種各樣的生物領域中。合成生物學使得我們能夠清楚的知曉何種類型的基因電路設計能夠實現何種的細胞行為。通過在活細胞中構建和測試這些基因電路，我們也逐漸明白了不同設計之間的優劣。

這樣一來，除了傳統的反向工程學的手段，我們還可以利用正向工程學的方法來實現「Building to Understand It」。除了許多重要的直接應用之外，我認為正是對於生物系統思考和研究方式的轉變，大大的激發了更多人的想像力，進而促進了合成生物學領域的快速發展。

處於合成生物學核心的基礎性問題很大程度上與系統生物學是相互重疊的，因為兩個領域都在致力於理解基因電路的設計原理 。而且我預計這些領域將會相互交叉影響，並越來越難以在未來分別開來。

Christina D. Smolke. 合成生物學和系統生物學都與其發展而來的領域在方法上都發生了根本性轉變。相比於傳統的基因工程研究，合成生物學在設計、構造和刻畫生物系統方面更加強調工程學原理和方法學。而相對於在傳統生物研究中的還原論方法，系統生物學則是在研究整體系統組成的方面發生了根本性轉變。計算機建模在兩個領域都是重要的工具，但是在兩個領域的應用有所不同。在系統生物學中，計算模型用於預測系統的行為，而在合成生物學中，建模則用於指導人工系統的設計。

合成生物學已經從最初的相當狹窄的關注點中大大擴展開來。研究人員更加充分地認識到了天然生物系統中機制的多樣性，同時也在充分利用著天然系統的多樣性。例如，絕大多數早期的工作集中在基於轉錄因子的調控網路設計上，以展現人工系統的動態行為（像震蕩子和雙穩態開關），然而現在的設計一般都會包含其它層次的調控設計，包括基於RNA的調控子，翻譯後調控修飾和分子支架等等。

另一個例子，突變和進化之前被廣泛認為是系統性能的一個阻礙，需要將其影響降到最低。然而目前最新的研究正在探索生物系統這個獨特的性質，利用生物的進化和適應性質進行設計。

Christopher A. Voigt. 這個領域早期的一些目標都圍繞在創造可執行一系列程序化任務的細胞展開。例如，Adam Arkin設想了一種可以在人體內識別微環境並執行治療功能的工程菌。想要做到這一點，需要控制相關的基因在特定時間特定環境狀態下開啟，這需要合成生物學的調控方式，也就是說，通過基因電路來調控。這也一直是傳統基因工程領域難以解決的問題之一。

合成生物學是一個工程學科，在這個領域研究人員渴望去創造一些自然界中不存在的東西。而系統生物學是一個基礎學科，它的目標是更好的理解自然的生物學。這種關係類似於生物工程和生物學，或化學工程和化學。學科之間的「語言」可能是類似的，但是彼此的動機卻是不同的。例如，合成生物學和系統生物學都對模塊化非常感興趣。對於自然的細胞，基因和調控網路可能是模塊化的，也有可能不是模塊化的，這取決於你如何定性這個問題。相比之下，工程師則更加熱衷於創造具有更高模塊化程度的合成基因電路。

數學模型是一種工具，它能夠進行定量的預測或者對數據進行詮釋。數學建模對兩個領域都非常有用，就像其他的工具如測定蛋白水平的質譜或者將DNA導入的細胞的轉化方法一樣在很多領域都有應用。

問：在基礎研究中目前合成生物學的主要成就有哪些？這個領域正在面臨這哪些挑戰？

R.W. 在過去的十年里，合成生物已經幫助生物科學轉變為一門真正的工程學科。顯著的成就包括創造了可組合的基因元件庫，具有革命性進展的基因合成技術和更快更高效的模塊化DNA組裝方法。通過將快速發展的生物技術和計算機建模結合起來，合成生物學家現在可以以一種可預測的方式（至少在一定程度上可以這麼講）自上而下對生物系統進行設計和工程化改造。

然而，合成生物學真正缺少的，就我的實驗室目前正在關注的哺乳動物合成生物學來說，是現實世界中的實際應用。我們必須做出努力將合成生物學研究從「玩弄玩具」轉變成讓整個社會能夠從中受益的實際應用，比如癌症治療，病毒疫苗和改造的生物組織。值得注意的是，現在主要的挑戰是如何將合成的基因電路安全的遞送到哺乳動物體內。我認為基於RNA的遞送方法的發展將會改變遊戲規則：它將會克服政府監管的限制，同時滿足傳統基因療法所要求的安全保障。

G.M.C. 除了一些應用，合成生物學的成果包括指數程度發展的基因組測序技術，晶元合成DNA文庫技術，基因組的多重自動化基因組工程技術（Multiplex automated genome engineering, MAGE）（主要在大腸桿菌中）和幾乎能在任何基因組中進行編輯的CRISPR-Cas9技術。

隨著基因組「讀」與「寫」技術的不斷進步，新的挑戰在於新功能的系統設計和篩選，以及高效的複製和生產能力。此外的挑戰還包括如何預測，模擬和改進高度多樣化或個性化的技術及探索其對複雜的生理和生態系統的影響。

M.B.E. 我們在這個領域已經走過很長的一段路了：我們已經構建好幾代版本的震蕩子和基因開關。它們具有多樣的細胞成分和調節機制，並且能夠和內源的基因電路相互作用。

複雜的代謝通路被改造用來生產有用的產品，信號通路以可預測的方式被重新設計改變其動態行為。然而，由於技術的挑戰和我們對於生物基因電路設計理解的不足，合成生物學仍然處於極其原始的狀態。

在技術方面，合成基因電路並將它們導入到細胞內的過程仍然十分緩慢並且問題頗多，尤其是在哺乳細胞中，然而，一些新的技術，例如基於CRISPR的基因編輯系統，是極其鼓舞人心的。

在基礎研究方面，我們對基因電路如何在細胞和組織中有效的發揮作用所知甚少，我們仍然需要從自然界了解更多。具體而言，最大的挑戰之一就是將合成生物學在單個微生物中作用的基因電路轉移到多細胞系統中。例如，能夠實現細胞自我模式化的基因電路。如果能夠成功的話，我們就可能從一個全新的角度理解多細胞生物的發育過程，這可以為組織改造和再生提供有用的信息。

C.D.S. 合成生物學對於基礎研究重要的影響之一是其驅動了多種方法的進步，以支持大規模遺傳編碼程序和基因組規模的工程化改造，包括DNA合成。其他的進展包括闡明了支持理性設計和精確基因調控與酶促活動的框架，同時我們還在推動定量標準化和數據報告標準化方面做出了初期的努力。

這個領域一個很大的挑戰是開發高通量、無破壞性的量化技術，以便能夠對無法通過熒光蛋白信號測定的生命活動進行定量。雖然這個領域正在轉向多樣性的設計思路，但是至今在開發刻畫多樣性基因電路的技術方面我們仍然投入不足。正因如此，我們無法從當前的設計方法中有所收穫，這限制了我們向更加「更智能設計」方法的轉變（與「更多設計」的方法相比）。

C.A.V. 合成生物學領域近期多項進展革命性的改變了工程化改造細胞的方式。DNA合成和組裝技術使得研究人員可以對上百萬鹼基的任意一個鹼基進行操控。我2003年做博士後時還沒有這個能力。但我們利用這樣的能力來做什麼也是一個不小的挑戰。同時我們在基因組工程化改造和DNA轉化進入細胞等技術方面也有了很大的進步，包括George Church開發的MAGE技術，Craig Venter構建的人造基因組，基於CRISPR方法的基因組編輯技術等。

設計方法也有了進一步的發展。許多實驗室一直在重新思考遺傳系統的設計，以便能夠建立關聯更多元件、複雜度更高的系統。許多實驗室，包括美國勞倫斯伯克利國家實驗室（LBNL）的BIOFAB，已經構建了大量刻畫好的基因元件，並且這些元件可以整合進這些複雜的設計之中。

基因電路的設計能力也得到了提高。我們已經構建了轉錄因子的大型文庫，並且將他們轉換為了邏輯門基因電路。同時我們也更加理解了如何讓他們彼此絕緣從而他們能夠更好的組成更大更複雜的基因電路。CRISPR干擾技術（CRISPRi）為我們提供了關閉基因表達的新思路，同時其系統本身的正交性使得更多的調控元件可以應用到同一個細胞中。

就實際應用而言，市場上很多產品都是由基因改造的細胞生產的。來自美國Biodesic公司的Rob Carlson估測這個市場每年會有3500億美元的增長，大約佔美國經濟的2%。合成生物學無論是對現在還是下一代產品的貢獻是非常大的。大公司中的一些值得注意的例子包括美國Dow AgroSciences的殺蟲劑（例如spinosyn），美國的Amyris和法國的Sanofi的抗瘧青蒿素。 小公司也在開發產品，例如，美國的Genomatica公司生產丁二醇（BDO）和美國的Refactored Materials生產重組蜘蛛絲。其實有很多例子，我在這裡只是強調了一些。

在應用層次一個主要的挑戰是利用細胞創造更加複雜的功能材料。細胞是天然的原子結構，現在很多正在使用的材料均源自生物本身。上面提到的例子相對來說都是簡單的化學物質，天然產物或者單一的蛋白質。如果想要獲得更加複雜的產品則需要合成更加複雜的基因電路，同時需要對很多，甚至成百上千個的基因進行操控。

問：合成生物學的一些應用引領了臨床醫療、藥物與生物燃油相關的生物技術的發展。合成生物學方法的優勢有哪些？你們是如何看待這些應用的未來的？

R.W. 顯然合成生物學已經革命性的改變了代謝工程領域，這使得生產藥物或者小分子成為可能。儘管在這個領域近期有了很大的進展，但是，我之前也說過，其實至今只有很少的應用最終實現，特別是對於哺乳動物合成生物學領域。

然而，這其中的原因可能在於這個領域如此之新。我認為哺乳動物合成生物學未來將會在兩個領域有著廣闊的應用前景。

第一個領域是利用眾多近期開發的基因組工程工具創造下一代編程化的哺乳動物細胞系，以產生更加量身定製的先進生物製劑。哺乳動物合成生物學界和生物醫藥工業的緊密合作會讓這個領域極大的受益。

第二個領域是創建能夠匹配生物系統複雜度的基因電路療法。系統生物學正在幫助我們揭秘生物網路之間的調控方式和相互作用關係，從而破譯生物體的複雜性。不過這也教導我們：像癌症、代謝疾病、免疫疾病、神經疾病或者心理紊亂這些生物疾病本身是相關深奧和複雜的。我認為複雜的疾病最好是利用複雜的合成基因電路調節身體來達到最好的治療。隨著研究步伐的加速，哺乳動物合成生物學很快就會有標誌性的應用從而贏得大眾的支持和信任。

G.M.C. 具體的應用實例包括Amyris和Sanofi對抗瘧葯青蒿素的高效生物生產、美國LS9和Joule Unlimited對石油的「綠色化學」替代方案。未來，我們將看到越來越多的合成生物學用於醫療：從隨機轉基因插入轉向精準基因組編輯; 從抗生素地毯轟炸到腸道微生物療法; 生物納米材料能夠實現原子級別的精確度，比當前電子電路處理信息的能力高出一百萬倍以上。

M.B.E. 所有這些領域的工作是極其令人興奮的。儘管我們已經成功利用工程化改造的微生物來生產重要的產品，但是這些平台都是在嚴格控制的實驗室條件下設計的。一個關鍵的挑戰是弄清楚如何使工程化改造細胞能夠在複雜的自然環境中發揮作用，包括人類的身體內。我們需要掌握設計多基因基因電路的能力，將它們精確的整合進細胞內部，並且能夠定量的控制它們的行為。目前，進展過程仍然乏味緩慢，並且這極大地限制了我們可探索的想法和設計方案。

話雖如此，我認為很多問題其實都是可以解決的。想像一下，當研究人員可以常規的讀取和控制細胞中多層相互作用的基因和蛋白質系統時將會發生什麼——這將顯著的改變基因工程領域的現狀，將常規基因工程的單位從基因改變為基因電路！

C.D.S. 合成生物學正在為臨床療法和生物技術帶來新的工具和方法。合成生物學不僅在推進生物合成過程的複雜度，同時也在提高通過生物過程生產的分子的多樣性。利用生物自身強大的生產能力在未來將會是一個巨大的機遇，這將為我們帶來更加複雜的生物產品。

我們還可以通過以一個動態的、局域性的方式控制患者治療過程中藥物的輸送和劑量來開發更加安全和高效的臨床療法。目前的努力集中在基於細胞的平台包括免疫療法和益生菌。隨著技術的不斷進步，合成生物學在臨床上的長期應用還將延伸到組織工程和再生醫學領域。

C.A.V. 在短期內，合成生物學開發的工具可以直接被用來開發天然產物，包括生物醫藥，人類增強藥物，工業化學產品，能源和農業產品（例如，殺蟲劑和農藥）。

利用自然多樣性的潛力是巨大的。自然的多樣性存在於大量的菌種和DNA序列資料庫中。我們現在可以很容易獲得生物的DNA信息。生物信息學可以預測能夠用來生產高附加值化學物質的基因簇。DNA合成技術使得序列信息能夠轉換為物理DNA，然而，在生物中這些合成的DNA往往容易丟失或者不能夠被操控或者基因本身在生物體內就是被沉默掉的。克服這些困難需要合成生物學開發相應的調控工具。 我們正處於挖掘資料庫以開發有價值產品這種淘金熱的早期階段。

同時對酶的深度探索（資料庫基因的高通量搜尋、列印和篩選）也是很重要的，這使得可生產的化學物質呈現多樣性。這將帶領我們實現這樣的夢想：酶催化就像化學合成一樣靈活多變，可以催化合成任意化學結構。同時細胞療法將會擁有巨大的優勢：其不僅具有感應和處理信號的能力並且能夠在目標靶點合成多種化學和蛋白效應因子。

問：技術的進步使得「書寫」整個基因組成為可能。然而，創造人造生命形式的可能性已經引起公眾廣泛的擔憂。合成生物學領域面臨著什麼樣的挑戰，哪些是必須解決的倫理和監管問題？

R.W. 創造一個新的生命形式是很有趣的——並且我認為這最終是有可能的。然而，大多數合成生物學項目都是在細菌，酵母或哺乳動物細胞添加新的功能，因為這對生物技術工業有很大用處。

我們試圖通過正向基因工程來提高標準化的，刻畫優良的細胞系的多樣性和應用價值。在哺乳動物合成生物學中，生物治療方案將需要在任何監管批准之前保證安全性，同時病毒載體能夠有效的將DNA基因電路遞送到目標靶點。我們已經看到一個有前景的結果：RNA基因電路可以作為一個安全的替代方案。此外，相比於用於藥物開發和診斷的動物模型，可編程的「organs-on-a-chip」可以提供一個更接近人類實際情況的替代方案。

像其他負責任的科學領域一樣，合成生物學有一種根深蒂固的文化：我們一直在預測、避免和降低技術帶來的風險。

G.M.C. 合成生物學現在面臨的問題可以追溯到DNA重組技術的開始階段，那時候Paul Berg和Rudy Jaenisch在1973年將SV40腫瘤病毒插入到了細菌和小鼠中。幾十年來，分子生物學家在複雜的生物系統中用基礎的技術研究單個基因。現在合成生物學則將系統設計和工程安全性引入生物學中。

我們可以最終構建基因組重編程的生物，這些生物無法與自然的基因組相互交流。另一方面，用來控制害蟲、疾病或者入侵生物的基因也無法與天然的基因組產生關係。

M.B.E. 我們對於合成生物學技術將會帶來什麼和不會帶來什麼了解的還有遠遠不夠。我尤其擔心公眾對於合成生物學的誤導，同時這種誤導可能會隨著炒作而逐漸增加。斯洛恩基金會和美國國家科學院等已經作出了巨大的努力來改善和大眾之間的信息鴻溝。我們科學家必須一起行動來「Keep it real」，以現實為基礎，將關注點放在最關鍵的風險上，讓更大的團體參與進來，利用好我們現今擁有的每一項資源為合成生物學的發展做出努力。

C.D.S. 通過DNA合成技術創造遺傳系統的進展已經遠遠超過了設計這些系統的進展。儘管我們能夠構建合成的基因組，但是我們還遠遠沒有達到設計新的基因組水平基因程序的能力，更不用提合成全新的生命。

然而，在道德和監管層面對工程化改造的生物的擔心仍然是需要的。道德和管理層面的擔心取決於合成生物學潛在的應用。例如，如果生物被用在實驗外的環境中，那麼我們就應該考慮其對於環境和生態系統的潛在影響。

C.A.V. 這個領域需要用不同的方法來進行風險分類以及產品管理。例如，目前FDA和USDA的管理結構並不利於有效評估高度工程化的生物對人體或者相關醫療領域的影響。

這個領域正在採取積極的方法來解決這些問題。這方面的討論和研究一直很積極。

（結束）

訪談中5位科學家的簡要介紹：

George M. Church (G.M.C.)

• 哈佛大學遺傳學教授, 美國國立衛生研究院（NIH）基因科學卓越中心負責人，美國國家科學院和美國國家工程院院士，http://PersonalGenomes.org網站(世界上唯一的開放式人類基因組和特徵資料庫)的負責人，共發表論文300餘篇，專利60餘項，出版過圖書一本：《Regenesis》。他的博士學位（1984年授予）所使用的方法推動了1994年的第一個微生物基因組測序的完成。他在「下一代」測序，寡核苷酸合成和細胞工程等方面的工作促進了醫療診斷，生物醫療和合成化學品的發展，同時他也在推動人類隱私，生物安全等相關政策中做出貢獻。

Michael B. Elowitz (M.B.E.)

• 美國帕薩迪納加利福尼亞理工學院（Caltech）霍華德·休斯醫學研究所和生物學，生物工程和應用物理學教授的研究員。他創建了一個經典的振蕩基因電路Repressilator，並展示了如何在單個細胞中分析基因表達「噪音」及其功能。他的實驗室利用合成生物學，定量延時影像和數學建模來理解細胞和組織中基因電路的設計原理，研究範圍包括從細菌到哺乳動物細胞等多種系統。他是麥克阿瑟獎及人類前沿科學計劃（HFSP）中曾根獎的獲得者。

Christina D. Smolke (C.D.S.)

• 美國加州斯坦福大學生物工程系副教授，教育副主席和William M. Keck基金會學者。她的研究開發了基礎工具，推動了我們工程化改造和設計生物的能力的進步。例如，她的團隊領導開發了一種新型的生物輸入/輸出（I / O）設備，從而從根本上改變了我們與生物進行交互和編程的方式。 她的團隊使用這些工具來推動細胞療法，天然產物生物合成和藥物發現等不同領域的變革。她是十多項專利的發明人，她的研究計劃已獲得多項殊榮，其中包括美國國立衛生研究院（NIH）主任的先鋒獎，世界技術網路（WTN）生物技術獎和TR35獎

Christopher A. Voigt (C.A.V)

• 自2010年起一直擔任美國馬薩諸塞州劍橋市麻省理工學院（MIT）生物工程系的教授，在那裡他是合成生物學中心的共同負責人並成立了MIT-Broad Foundry。他擔任美國化學學會期刊ACS Synthetic Biology的主編。 在麻省理工之前，他在美國舊金山加利福尼亞大學藥物化學系任教。

Ron Weiss (R.W.)

• 美國麻省理工學院（MIT）生物工程系和電氣工程與計算機科學系的教授。他創立並擔任麻省理工學院合成生物學中心主任。 他於2001年獲得麻省理工學院博士學位，並於2001年至2009年在美國新澤西州普林斯頓大學任教。他的研究幫助開拓了合成生物學領域，通過計算設計，構建並實驗測試了能夠進行數字邏輯，模擬控制和細胞間通信等幾個基礎合成基因電路。最近，Weiss實驗室還專註於哺乳動物合成生物學，重點是治療應用領域，包括組織工程，糖尿病和癌症治療。

英文原文：

Q1: An increasing number of publications and institutions are dedicated to synthetic biology. From the initial focus on the design of synthetic genetic circuits, how has the field expanded and evolved? And how does synthetic biology today relate to other disciplines such as systems biology and mathematical modelling?

Ron Weiss. Synthetic biology was seeded when bacterial cells were programmed with basic circuits — a toggle switch, an oscillator and cell–cell communication. The focus has evolved from small transcriptional regulatory networks into complex multicellular systems that are embedded in a variety of organisms such as yeast, mammalian cells and plants, using non-transcriptional logic, including microRNAs, protein phosphorylation and DNA editing. Detailed and precise characterization and predictable part composition have become essential for the efficient creation of sophisticated multi-input logic functions, composite states and analogue circuits. This has led to varied and improved interfaces for cellular sensors and actuators along with advancements such as subcellular compartmentalization of logic operations.

Synthetic biology complements systems biology: the former is based on forward engineering and the latter on reverse engineering. For instance, insight gained from systems biology investigations of natural processes leads to improved designs of synthetic systems, and the creation of small artificial networks helps to analyse hypotheses on the function of natural ones. Computational modelling tools have become essential for the design of artificial networks and are also finding application in other areas of biology that require advanced observation and correlation. Finally, synthetic biology researchers are developing an ever-growing appreciation for biological complexity, which requires interdisciplinary research, new circuit design principles and programming paradigms to overcome barriers such as metabolic load, crosstalk, resource sharing and gene expression noise (and sometimes actually utilize these barriers to create more robust systems).

George M. Church. In my view, synthetic biology was never focused on 『genetic circuits』, but rather on biology rapidly maturing as an engineering discipline, including computer-aided-design (CAD), safety systems, integrating models, genome editing and accelerated evolution. Synthetic biology is less like highly modular (or 『switch-like』) electrical engineering and computer science and more like civil and mechanical engineering in its use of optimization of modelling of whole system-level stresses and traffic flow.

Michael B. Elowitz. At the most general level, synthetic biology expands the subject matter of biology from the (already enormous) space of existing species and cellular systems that have evolved to the even larger space of non-natural, but feasible, species and systems.

Although we started with circuits to carry out the simplest kinds of dynamic behaviours, synthetic approaches can be applied broadly to all types of bio-logical functions from metabolism to multi-cellular development.

Synthetic biology allows us to figure out what types of genetic circuit designs are capable of implementing different cellular behaviours, and what trade-offs exist between different designs, by building and testing these circuits in living cells.

We can thus use forward design instead of (or rather in addition to) more traditional reverse engineering approaches, effectively 『building to understand』. Beyond the many important immediate applications, I think it is this transformation in the way of thinking about, and working with, biological systems that stimulates the imagination of so many people and explains the rapid growth of the field.

The fundamental questions that are at the heart of synthetic biology overlap considerably with systems biology, as both fields seek to understand principles of genetic circuit design, and I expect that these fields will cross-stimulate each other and become increasingly difficult to disentangle in the future.

Christina D. Smolke. Both synthetic biology and systems biology represent fundamental shifts in approaches from the fields they grew out of. Synthetic biology emphasizes engineering principles and methodology in designing, constructing and characterizing biological systems from traditional genetic engineering research; systems biology represents a shift in studying integrated components from the more traditional reductionist approach taken in biological research.

Computational modelling is an important tool in both fields but used to achieve different objectives. In systems biology, computational models are used to make predictions about the behaviour of a system, whereas modelling is used to direct design in synthetic biology.

Synthetic biology has expanded and evolved substantially from its initial rather narrow focus to appreciate and use more fully the diversity of mechanisms found in natural biological systems.

For example, early work focused largely on transcription factor-based regulatory networks designed to exhibit dynamic behaviour (such as oscillator and 『toggle switch』 behaviours), whereas designs now routinely incorporate other levels of regulation, including RNA-based regulators, post-translational modifications and molecular scaffolds.

As another example, mutation and evolution were widely viewed as an obstacle to system performance and something to be minimized, whereas newer approaches are beginning to exploit this unique aspect of biological systems and to design for evolution and adaptation.

Christopher A. Voigt. The early ambitions in the field were around the creation of cells that could go through a series of programmed tasks; for example, Adam Arkin envisioned an engineered bacterium that could move through the human body and could identify microenvironments, and carry out therapeutic functions.

Doing this requires control over when and under what conditions genes are turned on, which in turn requires synthetic regulation — in other words, circuitry. This was, and remains, one of the most difficult aspects of genetic engineering.

Synthetic biology is an engineering discipline — there is a desire to build things that do not yet exist. Systems biology is a basic science, where the goal is to better understand natural biology.

The relationship is similar to biological engineering and biology, or chemical engineering and chemistry. There can be similar language, but the motivation is different. For example, both synthetic and systems biology are interested in modularity.

For natural cells, genetics and regulatory networks may or may not be modular, perhaps depending on how you frame the question. By contrast, engineers can continue to strive to create synthetic genetics and circuits that are increasingly modular.

Mathematical modelling is a tool that enables quantitative predictions or the understanding of data. It is applicable to both areas, as are other tools such as mass spectroscopy to measure protein levels or transfection methods to move DNA into cells.

Q2: What have been the main achievements of synthetic biology so far in basic research, and what are the challenges to be met?

R.W. Over the past decade or so, synthetic biology has helped to transform the bio-logical sciences into a true engineering dis-cipline. Notable achievements include the creation of a registry of composable parts, revolutionary advances in gene synthesis technologies and faster and more efficient modular DNA assembly methods.

By integrating these rapidly developing tech-nologies with computational modelling approaches, a synthetic biologist can now engineer biological systems top-down in a (somewhat) predictable manner.

What is critically lacking, however, with respect to mammalian synthetic biology (an area that my laboratory is currently most focused on), are real-world applications.

We must now make an effort to transition from developing 『toy』 applications to creating actual applications for cancer therapies, vaccination and engineered tissues that society can directly benefit from.

Arguably, the main challenge here is how to safely deliver synthetic circuits into mammalian organisms. I predict that the development of RNA-based delivery methods could be a game changer in overcoming the regulatory hurdles and required safety guarantees associated with traditional gene therapy.

G.M.C. In addition to some applications (discussed below), the achievements include new tools enabling rapid exponential improvements in genome sequencing, harvesting DNA from chips for libraries, multiplex-automated genome engineering (MAGE) of natural and artificial chromosomes (mainly in Escherichia coli) and Cas9–CRISPR technology for genome editing of nearly any genome.

As reading and writing genomes progresses, the new challenges lie in system design and selection for new functions, and efficient replication and production. Additional challenges arise from how to anticipate, simulate and improve highly diverse or personalized technologies and their impact on complex physiological and ecological systems (discussed below)

M.B.E. We have come a long way as a field: we have built several generations of oscillators and genetic switches that work with diverse cellular components and regulatory mechanisms, and which interact with endogenous gene circuits.

Complex metabolic pathways have been engineered to produce useful products, and signalling pathways have been rewired to alter their dynamic behaviours in predictable ways. However, synthetic biology remains extremely primitive owing both to technical challenges and, even more, to fundamental inadequacies in our understanding of biological circuit design.

On the technical side, synthesizing genetic circuits and transferring them into cells remains far too slow and idiosyncratic, especially in animal cells. How-ever, several new methods, such as those based on the CRISPR system, are extremely encouraging.

On the fundamental side, we still have little understanding of how circuit designs can function effectively in cells and tissues and much to learn from natural examples. In particular, one of the greatest challenges is to move synthetic biology from circuits operating in individual microorganisms to circuits that function in a truly multi-cellular fashion, for example, circuits sufficient to implement self-patterning of cells. If successful, we may be able to understand multicellular development from a totally new point of view that could inform tissue engineering and regeneration.

C.D.S. One of the important effects of synthetic biology on basic research is that it has driven the advancement of a variety of methods to support the construction of large-scale genetic programmes and genome engineering, including DNA synthesis.

Other advances include elucidating frameworks that support the rational design and precise control over activities, such as gene regulatory and enzyme activities, as well as early steps in pushing standards in metrology and data reporting.

One big challenge in the field remains to develop measurement technologies that enable the high-throughput, non-invasive quantification of activities that are not encoded in fluorescent reporter proteins.

As the field has shifted towards favoring design approaches that generate diversity within the genetic programmes, there has not been a corresponding emphasis on the development of technologies that enable the characterization of the resulting diversity of genetic designs.

As such, the shift towards a 『design smarter』 approach (versus a 『design more』 approach) is limited by our inability to learn from current design approaches.

C.A.V. Many recent advances that have emerged from synthetic biology are revolutionizing the ways that we engineer cells. DNA synthesis and assembly methodologies make it routine to build constructs in which the designer has full operational control over every base pair for megabases of DNA.

This is not a capability that I had as a postdoctoral researcher in 2003 and it remains a challenge to know what to do with such a capability. There have also been many advances in genome engineering and the transfer of DNA into cells, including MAGE developed by G.M.C., the synthetic genome construction by the John Craig Venter Institute (JCVI), USA, and CRISPR-based methods to introduce directed changes into genomes.

Advances have also been made in design methods to build up to these capacities. A number of laboratories have been rethinking the process of the design of genetic systems to make it possible to build more sophisticated systems involving the connection of many parts.

Many laboratories, including BIOFAB at the Lawrence Berkeley National Laboratory (LBNL), USA, have built a large number of well-characterized genetic parts that can be incorporated into these designs.

The design of genetic circuits has also improved. We have built large libraries of transcription factors, have converted them into gates, and better understand how to insulate them so that they can be compose d into larger circuits. CRISPR interference (CRISPRi) offers a new way to turn off genes, and the intrinsic orthogonality of the system may allow many more synthetic regulators to be used in one cell.

In terms of applications, many products on the market are produced by genetically modified cells. Rob Carlson of Biodesic, USA, has estimated that it adds up to US$350 billion per year, or roughly 2% of the US economy.

The contribution of synthetic biology to the current and next generation of products is large. Some notable examples produced by large companies include insecticides (for example, spinosyn) by Dow AgroSciences, USA, and the anti-malarial artemisinin by Amyris, USA, and Sanofi, France. Small companies are developing products; for example, Genomatica, USA, produces butanediol (BDO) and Refactored Materials, USA, produces recombinant spider silk. There are many examples, and I am only highlighting a few.

One of the challenges in applications is harnessing cells to build complex functional materials. Cells are natural atomic architects and we already exploit this, as many in-use materials are from biology. The examples above are all relatively simple chemicals, natural products or individual proteins. Obtaining more complex products will require synthetic gene circuits and the ability to control many, possibly hundred s, of genes simultaneously.

Q3: Some applications of synthetic biology have steered development of new clinical therapies and of biotechnology for the production of drugs and fuels. What advantages have such approaches brought, and what do you foresee for the future of such applications?

R.W. Synthetic biology has certainly revolutionized the field of metabolic engineering, enabling the production of drugs and other small biomolecules inspired by natural products.

With respect to mammalian synthetic biology in particular, as I mentioned above, despite the great recent progress in the field, very few applications have actually come to fruition so far.

However, this is probably in large part owing to the field being so new. I see largely two areas to which mammalian synthetic biology will contribute applications in the future.

The first is the use of numerous recently developed genome engineering tools to create next-generation programmed mammalian cell lines for the production of more tailored advanced biologics. Progress in this area may benefit from a closer connection between the mammalian synthetic biology community and the biopharmaceutical industry.

The second is the creation of synthetic gene circuit therapies that match the complexity of biological systems. Systems biology is helping us to decipher the complexity of living organisms by unravelling regulatory networks and crosstalk between these networks.

It has also taught us that biological diseases themselves, such as cancer, or metabolic, immunological, neurological or psychiatric disorders, are equally sophisticated and complex.

I believe that such complex diseases will be best treated by modulating our bodies with correspondingly sophisticated synthetic gene circuit encoded in therapeutic agents. With its incredible pace of research, mammalian synthetic biology will soon be able to find flagship applications that will win over the support and trust of the general public.

G.M.C. Examples include the efficient bio-production of the anti-malarial drug artemisinin by Amyris and Sanofi and the 『green-chemistry』 replacement of petrochemicals by LS9, USA, and Joule Unlimited, USA.

In the future, we will see increased use of synthetic biology for therapeutics: moving from random transgenic insertions to precise Cas9-based genome editing; from antibiotic carpet-bombing to skin and gut microbial therapies; from blunt brain ablations and drugs to optogenetics and other bionano neurotherapies. Bio-nanomaterials are atomically precise and more than a million times more energy efficient than current electronic circuits for handling information.

M.B.E. The work in all of these areas is extremely exciting. Although we have been successful in engineering microorganisms to produce important products, these platforms are designed to work under well-controlled laboratory conditions.

One key challenge is figuring out how to engineer cells that can act as programmable devices and function safely in complex natural environments, including the human body.

Recent cell-based therapy results suggest the potential power of such approaches. However, we need the ability to routinely synthesize multi-gene circuits, integrate them precisely in cells and control their behaviour quantitatively.

Right now, this design cycle remains tediously slow, and this dramatically limits the diversity of designs and ideas that we can explore. Having said that, I think that a lot of these problems are solvable. Just imagine what will happen when researchers can routinely control and read out systems of multiple interacting genes and proteins in living cells and organisms — effectively changing the unit of routine genetic engineering from the gene to the gene circuit!

C.D.S. Synthetic biology is bringing new tools and approaches to applications in clinical therapies and biotechnology. For the production of drugs and fuels, synthetic biology is both advancing the complexity of biosynthetic processes that can be engineered and increasing the diversity of molecules that can be manufactured through biological processes.

Harnessing the full manufacturing capacity of biology is a big future opportunity, providing us access to more complex scaffolds and materials.

Synthetic biology tools also bring the potential to develop safer and more effective clinical therapies, by enabling control over the delivery and dosing of therapeutic activities in a dynamic temporal and spatial manner within a patient.

Current efforts have focused on cell-based platforms such as immunotherapies and probiotics. As the field progresses to tackle spatiotemporal programming and pattern formation, long-term applications of synthetic biology in the clinic should extend to tissue engineering and regenerative medicine.

C.A.V. In the short term, the tools of synthetic biology can be directly applied to exploit natural products, including pharmaceuticals, human performance enhancers, industrial chemicals, energetic materials and agricultural products (for example, insecticides and pesticides).

The potential of accessing the natural diversity that is already present in large strain banks and the sequence databases is extraordinary. The DNA sequence information is available, and bioinformatics can predict gene clusters that encode the machinery for producing high value chemicals.

DNA synthesis enables the conversion of the sequence information back to physical DNA. However, the host organism is often long lost, cannot be manipulated or the genes are 『silent』.Overcoming this requires synthetic regulation and tools from synthetic biology. We are at the early phase of a gold rush of mining the databases and strain banks for valuable products. There will also be deep enzyme mining (that is, the high-throughput identification, printing and screening of genes from databases) to diversify the chemical structures.

This may lead to 『the dream』 of having enzymes that can be as flexible as synthetic chemistry in building arbitrary chemical structures. The use of cells as therapeutics offers the advantage of being able to harness the sensing and signal processing capabilities of the cell, as well as the capability to synthesize multiple chemical and protein effectors at a site of interest.

Q4: Technical advances have made it possible to write entire synthetic genomes. However, the possibility of creating synthetic forms of life has raised public concerns about the potential development of infectious agents or effects on biodiversity. What challenges lie ahead for this field, and what are the key ethical and regulatory issues that must be addressed?

R.W. It is interesting to consider entirely new forms of life — and these may eventually be possible. However, the majority of synthetic biology projects are focused on improving existing strains by adding new functionality to bacterial, yeast or mammalian cells that are useful to the biotechnology industry.

We seek to increase the diversity and application of standardized, well-characterized cell lines through forward genetic engineering. In mammalian synthetic biology, therapies and treatments will require viruses that are safe but effective at targeted delivery of DNA circuits before any regulatory approval.

We have seen promising results with delivery of RNA synthetic circuits as a safer alternative. In addition, programmable 『organs-on-a-chip』 may provide a more humane alternative to animal models for drug development and diagnosis. Like other responsible scientific fields of endeavour, synthetic biology has an ingrained culture of seeking to reduce risks with existing technological solutions, while seeking to predict and avoid problems with proposed new solutions.

G.M.C. Examples of synthetic genomes include interleukin-4 (IL-4)-ectromelia-pox virus, human endogenous retrovirus (HERV) and resurrected influenza. The issues that are raised now for synthetic biology date back to the dawn of recombinant DNA when Paul Berg and Rudy Jaenisch inserted SV40 tumour virus into bacteria and mice in 1973.

For decades, molecular biologists focused on single genes with only rudimentary skills in complex biological systems. Synthetic biology is now positioned to embrace systems design and safety engineering with proactive, community-based, scenario planning for ecosystems and physiological nuances (US Environ-mental Protection Agency (EPA) and US Food and Drug Administration (FDA)).

We can finally construct genomically recoded organisms (GROs) that cannot exchange functions with natural genomes, and at the other end of the spectrum, genes that intentionally spread to control parasites, disease vectors and invasive species (for example, malaria, mosquitoes, kudzu and carp).

M.B.E. There is a lot we do not yet know about what will or will not become possible with the technologies of synthetic biology. I worry in particular about the potential for misunderstanding, sometimes increased by a misleading hype. Great efforts have been made to improve the information gap, for example by the Sloan Foundation, USA, and the US National Academy of Sciences.

Ultimately, we scientists must collectively 『keep it real,』 focus on the most critical risks seriously and thoughtfully in a way that is grounded in reality, engage with the larger community and make the most of the unique opportunity we now have for fundamental discovery and understanding.

C.D.S. Advances in the fabrication of genetic systems (through DNA synthesis technologies) have far outpaced advances in the design of these systems.

While we can build synthetic genomes, we are nowhere near being able to design novel genome-scale genetic programmes and thus completely synthetic life forms.

However, ethical and regulatory concerns still apply to engineered organisms that incorporate some synthetic genetic information in addition to a native genome. The ethical and regulatory concerns depend on the potential uses and applications. For example, if the organism is to be used outside of a contained environment (such as an enclosed bioreactor), then potential effects on the environment and existing ecosystems should be addressed, as well as related issues such as evolution, adaptation, containment and removal. Part of the challenge for the field is being able to design in accordance with quantitative design, performance specifications and tolerance.

C.A.V. The new applications promised by the field require different approaches for the categorization of risk and how the products are regulated. For example, the current regulatory structures of the FDA and US Department of Agriculture (USDA) are not conducive for evaluating the impact of highly engineered organisms as therapeutics into the human body or in a field, respectively. The field has taken a proactive approach to addressing these issues and it is a very active area of discussion and research.

----------------------------------------------------------------------------------------------

Glossary:

CRISPR(Clustered regularly interspaced short palindromic repeats).

An adaptive immune system that is found in bacteria and archaea, which is based on an RNA-guided nuclease (Cas9). Components of the CRISPR system are being repurposed to provide powerful, flexible and precise genome engineering, and regulatory systems across diverse species.

Genetic switches

Natural or synthetic systems for regulating gene expression in response to one or more external or internal signals. The output of genetic switches is often a complex logical function of input signals that in many cases can provide a persistent response to transient inputs or

other capabilities.

Genomically recoded organisms (GROs)

Changing every instance in a genome of one or more of the 64 codons in the genetic code for higher safety and productivity.

Multiplex-automated genome engineering

(MAGE). Efficient genome editing that is capable of making dozens of changes per genome and billions of genomes by inserting short (90 bases long) single-stranded DNA into the cellular replication fork with one or more DNA changes.

Optogenetics

A technique to control and perturb cellular behaviour using light and genetically encoded light-sensitive proteins. It has been extensively used to precisely control neuronal activity spatially and temporally through light.

Oscillator

Produces oscillations that underlie diverse biological behaviours from neurobiology to multicellular development. Synthetic biology has shown that remarkably simple circuit designs can produce clock-like oscillations of protein levels in individual living cells.

----------------------------------------------------------------------------------------------