Ep199: Martin Burke on Making Small Molecule Medicines for the AI Era

所以我觉得，正是从那时起，我开始转变方向，心想：好吧。

So I think that actually started my transition to say, okay.

我拿到了医学博士学位。

I I finished the MD.

我想完成这个目标，但同时也在想，能不能设计出一些分子，去解决那些还没人碰过的难题？

I wanted to do that, but could I could I think about how to make molecules that could impact in places where they hadn't had a chance to do it yet?

正是这个问题，把我引向了这个方向。

And that was the problem that kinda drew me into that direction.

对。

Mhmm.

没错。

Mhmm.

那会儿你是不是正在哈佛跟斯图·施赖伯一起工作？

And was this around the time you were working with Stu Schreiber there at Harvard?

是的。

Yeah.

Stu Schreiber 真的是太棒了。

So Stu Schreiber was amazing.

他还没正式把我招进化学系，就让我先开始做实验了，这让我特别感激。

He let me get started even before I was, like, officially joining the chemistry department, which was tremendously appreciated.

我那时候下午都泡在他实验室，早上还得跑去轮转的实验室。

So I was spending, like, afternoons in his lab and then running over and doing my rotations.

早上跑一趟轮转，下午又冲去他那儿。

You know, know, in the morning, running over to his lab in the afternoon.

那段时间真是两边同时干，让我能提前开始做论文的工作。

So it's it's an amazing time where I was kinda doing both at the same time, which gave me a chance to get a head start on my thesis work.

对，没错。

And, and yeah.

那时候我还在上一门课。

So, and at the same time, I was also taking a class.

Andy Myers 在教高级有机合成。

Andy Myers was teaching, advanced organic synthesis.

那时候我还在上医学院的课，同时就把安迪的课也给上了。

So while I was finishing up my med school classes, I was also able to take Andy's class kind of in parallel.

我永远记得有一天上课，安迪在黑板上画了两性霉素，那是一种特别神奇的天然产物。

And actually, I'll never forget one day in class, Andy drew up on the board, amphotericin, which is this amazing natural product.

他当时在讲凯西·尼科劳对它的全合成。

And actually, he was talking about Casey Nicholau's total synthesis of it.

不过他顺口提了一句，两性霉素能在酵母和人体细胞里形成离子通道。

But he'd mentioned in passing that amphotericin forms ion channels in yeast and human cells.

我一听就竖起耳朵了——因为我刚接触过一个囊性纤维化的小姑娘，一直在想，怎么才能设计一个小分子，像离子通道一样替代她缺失的CFTR蛋白。

And I remember my ears had perked up because I kinda just had this experience with this young woman with cystic fibrosis and was trying to dream about how would you make a small molecule that would act like an ion channel to replace her missing CFTR.

听到安迪提到两性霉素的这个特性，我简直兴奋坏了。

And when he mentioned that about amphotericin, I was so excited.

后来这直接成了我第一个学术提案：想办法把两性霉素改造成治疗囊性纤维化的分子假体。

It actually became my first academic proposal, was to figure out to turn amphotericin into a molecular prosthetic for CF.

所以，那时候我两边都在忙，后来做完轮转，就全身心投入了，跟着斯图特做我的博士研究。

So, yeah, I was I was kinda doing both at the same time and then eventually finished the rotations and went full in to do my PhD with Stuart.

居然真有分子能形成离子通道。

That that there was a molecule that could make ion channels.

而囊性纤维化的问题，说白了就是这个。

And in CF, you know, that's the the problem.

细胞膜上的物质根本没法正常转运。

The the molecules are not transported across the cell membrane the right way.

要是能造出个更管用的通道，就能直击问题根源。

If you could make a more functional channel, you could address the the underlying problem.

没错。

That's right.

这正是我们的想法。

That was the idea.

后来我们特别兴奋，我在想怎么改造两性霉素，让它能实现你刚才说的这种理解和优化。

And and then we got very excited about I was thinking about how to make amphotericin in a way that would enable that kind of understanding and then optimization that you just kind of articulated.

这又引出了一个想法：干脆把它做得模块化，甚至反复用同一个反应，把各个小块像搭积木一样拼起来。

That, in turn, led to this idea of trying to make it very modular and perhaps even use the same reaction over and over again to to stitch the blocks together.

我觉得，这其实就是我们后来发展出如今这套模块化化学方法的关键起点。

And I think, you know, that was the kinda key takeoff point for us, which would ultimately became the block chemistry that we're working on today.

所以，说到底，这一切都源于我在研究生阶段的经历，正是那些经历让我对这个领域产生了兴趣。

So, yeah, it kinda all comes back to those experiences in graduate school, which really got me interested in these spaces.

是的。

Yeah.

这正是我们所说的化学生物学。

This this is a description of what we call chemical biology.

对吧？

Right?

就是这两个主要科学分支的交汇之处。

How these two major branches of science intersect.

说真的，这一点特别关键。

Well, and that's the thing.

而且说实话，当时能跟斯图尔特一起工作，真是难得的机会，那会儿他正在开创整个领域。

And look, I mean, I was it was an incredible chance to work with Stuart at a time where he was trailblazing that whole space.

我从他那儿学到了太多东西了。

And and actually, I learned so much from him.

真是太了不起了。

It's incredible.

他跟我说过几句话，我一辈子都忘不了。

A couple of things that he said to me that I'll never forget.

他说，只要不违背物理定律，那只是时间问题。

One was, if it doesn't break the laws of physics, it's just a matter of time.

作为化学家，那时候能亲手在实验室里合成分子，我记得那种感觉——这简直把合成的威力放大到了极致。

And as a chemist, with the capacity to actually go in the lab and make molecules at that time with your hands, I remember feeling like, I mean, that really amplified this kind of kind of power factor associated with synthesis.

他还说过另一句特别有意思的话：要创造方法，把整个世界变成你的实验室。

You know, the other one he said, which is also, I think, super interesting was that, you know, kinda create methods to help make the world into your research lab.

这话也让我觉得特别有力量。

That was also felt very powerful.

别只想着自己实验室里那几个学生埋头解决问题，而要把自己当成全球科学共同体的一员，让每个人都能参与进来，共同解决更广泛的问题——我觉得这才是真正厉害的地方。

So rather than a scientist just thinking about, with your small cluster of students in your own lab space trying to solve problems, really trying to think of yourself as a global scientific community and making it possible for everyone to kind of participate in problem solving more broadly, I think that was really powerful.

那时候我还突然经历了一次‘四分之一人生危机’。

And then I think the other thing that happened to me at that point is that I kinda had, like, my quarter life crisis.

我记得当时我们正在学解剖和病理，早上还要做尸体解剖和尸检。

So it's just like, I remember, we were doing anatomy and and pathology and, you know, dissecting a cadaver and then seeing, autopsies in the morning.

就在那一刻，我突然意识到：就算一切顺利，我也就剩下大概六十年了，顿时感到一种强烈的紧迫感，必须做点什么。

And I think I had this moment of realization like, okay, best case scenario, I got like sixty years left, and I just felt this intense sense of urgency to do something.

所有这些想法在我脑子里慢慢汇合，于是我下定决心：我要当一名化学家，但我会珍惜在医学院学到的一切，尤其是那种理解医学难题的视角。

And I think all that kind of coming together in my own mind, I made the decision, you know, I'm gonna be a chemist and treasure all the experience about medical school and I think access to understanding problems that might be really interesting to solve.

那些经历我都带着，但从那以后，我就全身心投入，开始申请化学领域的教职。

All of that I took with me, but at that point, I went in full force and decided to jump and apply for academic positions for chemistry.

明白了。

Okay.

所以你是想自己研发新药，而不是一个个给病人开药，对吧？

So you're gonna make new medicines rather than administer those to patients one on one.

这真是个重大的决定。

That's the big decision there.

那就是我的选择。

That was the choice.

对。

Yeah.

而且

And

所以你最后去了伊利诺伊大学。

so you end up going to University of Illinois.

你怎么会去那儿的？

How did you end up going there?

很棒。

Great.

对。

Yeah.

在美国，当然，你要经历申请教职的整个流程。

So in The US, of course, you go through this process where you apply for academic positions.

你要提交申请材料，比如研究计划、个人背景之类的。

You send in your proposals and your, you know, your background, etcetera.

我很幸运，当时有几个选择。

And, I was I was fortunate to have some options.

可当我一到伊利诺伊，天哪，真是吓到了。

And and but when I came to Illinois, oh my goodness.

这地方太棒了，资源简直没得说。

Like, this place is phenomenal, and the resources here are just incredible.

学生也都特别出色。

The students are incredible.

跟同事们聊一聊，你会发现伊利诺伊真有一些特别的地方。

And just talking with colleagues, I mean, there's some some things about Illinois that make it, I think, very special.

对我来说，最重要的一点是，我清楚地知道，只要我来了，限制我的只会是我自己。

The one for me that that was the most important is it was clear that if I came here, I was only gonna be limited by myself.

那时候的我，特别渴望这种感觉——知道自己的上限由自己说了算。

And at that point in my life, I just craved that so much to know that I didn't have a ceiling.

我就试着看看自己到底能走多远，结果真的一点都没让我失望。

And I tried to find out what I was capable of, and and I'm very happy that proved to be true.

这就是我为什么还在这儿，这段经历简直太棒了。

I mean, that's why I'm still here, and it's been an amazing run.

这儿搞科研真的特别棒。

It's a it's a really great place to do science.

是的。

Yeah.

对。

Yeah.

资源丰富，学生和教授都来自世界各地。

And all kinds of resources, great students and faculty from all over the world.

这其实是美国生物医学领域一个不大为人知的秘密。

It's it's one of the great secrets of the American biomedical enterprise.

我觉得我们拥有的这些优势，根本不是什么新鲜事。

I I think that we have it it's not just news flash.

不只是哈佛、斯坦福或者伯克利。

It's not just Harvard or Stanford or Berkeley.

外面的世界大着呢。

There's there's a wider world out there.

太棒了。

Awesome.

对。

Yeah.

不是的。

No.

我的意思是，你想想看。

I mean, think that look.

在美国，我们真的很幸运。

In The US, we're really lucky.

对吧？

Right?

我的意思是，咱们国家的学术研究体系真的特别独特。

I mean, our kind of academic research infrastructure is so special.

它简直太能赋能人了。

It's so incredibly enabling.

作为国家，我觉得我们特别珍惜这一点。

And as a country, I think we just treasure this.

对吧？

Right?

你想想，这种公共投入带来的影响有多大——不仅建起了公立研究型大学，还带动了整个生态系统，包括私立研究型大学。从人类历史的角度看，咱们国家能拥有这一切，简直太了不起了。

And if you look at all of the impact that that kind of public investment in creating both, you know, obviously, public research universities and then contributing to the entire ecosystem, which includes the private research universities, I mean, across the course of human history, it's a phenomenal thing that we have as a country.

我们真该珍惜这份财富，继续培育它，让它成为我们持续发展的强大基石。

We wanna treasure that and make sure we continue to foster that as a great strength, that we can continue to build on.

所以你后来去了厄巴纳-香槟，一切从零开始。

So you moved off to Urbana Champaign and you had a clean sheet of paper.

你有支持，也能自己定研究方向。

You had support and, and you could write your own research agenda.

那你当时决定要研究什么？是什么让你特别兴奋？

So what did you decide you wanted to work on, and and what really excited you then?

是的。

Yeah.

这就是美国助理教授最棒的地方。

And that is that's the amazing thing at the assistant professor in The US.

这可不是给别人当助手的意思。

It doesn't mean you're an assistant to anybody else.

完全是靠自己闯，能成能败。

It's like you sink or swim.

对吧？

Right?

这船是你自己的，舵你来掌。

It's it's your ship, and you get to drive it.

我觉得这种处境既让人害怕，又特别鼓舞人心，刚起步的时候，我两种感觉都有。

And I think that's actually both frightening and incredibly inspiring situation to be in, and I felt both at the same time when we got started.

所以我们彻底投入了这个分子假体的想法。

And so we actually went all in on this molecular prosthetics idea.

我们当时想搞清楚：小分子能不能像蛋白质一样发挥作用？如果能，我们能不能利用这个发现，开发出一类全新的药物，来替代体内缺失的蛋白质，比如囊性纤维化那种病？

And so the the question we wanted to answer is could we understand the capacity for small molecules to perform protein like function, and then could we leverage that understanding to transform them into hopeful a new class of medicines that would kind of replace missing proteins, kinda like cystic fibrosis?

要启动这个发现引擎，最关键的是合成环节。

And the key thing in order to be able to even start that discovery engine was the synthesis part.

那时候，我觉得合成这事儿是次要的。

And and at the time, that was in my mind, that was kind of secondary.

最核心的还是分子假体这个方向，但我们知道，必须用一种能快速理解、优化这些分子的方式来合成它们。

The the number one thing was the molecular prosthetics thing, but we knew we'd have to make these molecules in a way that was very enabling to understand them and to optimize them.

所以我们没打算走那种冗长、线性的合成路线，而是想搞出一种模块化的方法，方便快速修改、深入理解、不断优化。

So rather than try to think about a a very long, kind of linear synthesis of a molecule, we wanted to try to figure out to make them in a very modular way so that we could modify them quickly, understand them, and then optimize them.

我们一开始选的分子是两性霉素。

And we actually started with that molecule amphotericin.

我们没想着从头设计一个能形成通道的分子，而是想：嘿，大自然已经给了我们十亿年的领先优势。

And so rather than try to invent a channel forming molecule, we were like, okay, nature gave us like a billion year head start.

我的意思是，这个分子是在巨大的选择压力下进化出来的，它能自动在酵母和人类细胞里组装成离子通道，还能传导阴离子——这正是我们替换CFTR蛋白所需要的。

I mean, this is a molecule evolved over the course of tremendous selective pressure, and it already autonomously self assembles to ion channels in both yeast and human cells and conducts anions, which is exactly what we need in terms of replacing CFTR.

顺便说一句，它也能传导阳离子，所以我们觉得这或许是个可以改进的机会。

It also conducts cations, by the way, so we saw maybe an opportunity to to improve that.

所以我们一开始重点放在合成上，琢磨怎么用模块化化学的方法做出两性霉素的衍生物。

And, so we focused initially on the synthesis and how could we make, you know, kind of amphotericin derivatives through modular chemistry.

然后我们就特别兴奋了。

And then we got very excited.

也许我们可以用同一个反应反复操作来实现。

Maybe we could do it where you use the same reaction over and over again.

这感觉太酷了，也太有力量了。

That felt very cool and powerful.

长话短说，我们渐渐意识到，这种模块化的方法甚至可能实现自动化。

And then long story short, we started to realize that kind of block approach could perhaps even be automated.

于是我们对这门化学本身变得无比兴奋。

And then we got very, very excited about the chemistry itself.

所以我们当时是两边同时推进的。

So we kinda pursued the two in parallel.

明白了。

Okay.

与此同时，整个生物技术行业正在经历一场革命，大家都在搞重组蛋白、酶替代疗法、单克隆抗体，做各种各样的蛋白质替代治疗，就跟你说的差不多。

Now, at the same time while you're doing this, the whole biotech industry revolution was underway and people were making, you know, recombinant proteins, enzyme replacement therapies, monoclonal antibodies to do all kinds of things, replacing proteins, kinda like what you say.

但你从一个小分子的角度来看，觉得其实每种疗法都有自己的优势。

But you're looking at this from a small molecule perspective and saying, actually, you know, each of those modalities has has its advantages.

小分子呢，你可以合成出来，还能口服。

But small molecules, you know, they you can make them, you can take them orally.

而且相对便宜，大规模生产也更容易。

They are are relatively cheaper and easier to manufacture at scale.

还能全球分发。

You can distribute them around the world.

小分子作为药物的基石，已经有几十年的历史了，这方面的经验太丰富了。

There's a whole lot of, you know, history, decades of of of small molecules being kind of the backbone of of medicine.

所以你是想，用传统的化学方法，实现那些生物技术能达到的目标？

And so you're thinking, we achieve those kind of like biotech like objectives or or but with the conventional chemistry underneath it?

是的。

Yeah.

没错。

Exactly.

再想想我上医学院的时候，有位老太太，我那时候负责帮她照顾病情。

So I think again, back to my med school experience, I'll never forget there was a grandmother I was trying to tell help take care of.

她刚做完肾移植。

She had had a kidney transplant.

作为医学生，每天早上你得挨个给病人发药。

And as the medical student, your job is to kinda go around and give out their pills in the morning.

我记得当时递给她一粒1毫克的雷帕霉素片，再配上一杯水。

So I remember handing her this little one milligram tablet of rapamycin and then her little cup of water.

她一口吞下去，喝口水，就开始跟我聊她的孙子孙女。

And, you know, she just pops in her mouth, drinks the water, and then starts telling me about her grandkids.

我心里直嘀咕：天哪。

And I'm thinking to myself, holy mackerel.

因为我太熟悉这个分子了，你知道的，斯图尔特早就搞清楚了它的作用机制。

Because I knew that molecule really well because, you know, Steward had figured out how it works.

这结构复杂得简直难以想象。

And and that is one of the most complex structures you can imagine.

可它居然能口服吸收，穿过肠道，精准找到她的T细胞，轻轻踩下刹车，阻止她的免疫系统攻击体内那颗来自他人、却已存活的肾脏。

And here it is orally absorbed, going in, getting through her gut, going and finding her T cells, and gently tapping the brakes and and blocking her immune system from rejecting this other person's kidney that's living inside her.

你想想，小分子的力量真是惊人。

And you just kinda think about, like, small molecules are just phenomenally powerful things.

一个人病了，你给他一颗化学药丸，他就能慢慢好起来。

It's like you have a person who's sick, you give them a chemical and then watch them get better.

想想看，人体多复杂啊，尤其是得了病的人，这种药居然真能起效，太不可思议了。

Like, it is an amazing thing that they work given how complex a person is, especially a person who's got some type of disease.

我那时候就一直在想，这类化学物质，真是了不起。

And I just I do I remember thinking about, like, this class of chemical matter is amazing.

你想想进化的过程，大自然选择这条路肯定有它的道理。

And, if you think about the course of evolution, there's a reason why nature went there.

要是光靠DNA、RNA和蛋白质就能解决所有问题，大自然早就不折腾了。

Like, if nature could have solved all of her problems with DNA, RNA, and protein, she would have stopped.

对吧？

Right?

但正因某种特殊的功能需求，自然选择推动它发展出了次级代谢。

But then selective pressure for this very special functional capacity pushed nature to develop secondary metabolism.

而正是在这个过程中，大自然造出了这些信息量极大的小分子天然产物。

And that's where we get these amazing small molecule natural products that are so information dense.

它们能轻松穿过细胞膜。

They can slip through membranes.

还不容易被免疫系统发现。

They don't get recognized by the immune system.

而且特别稳定。

They're stable.

而且还能大规模生产。

They're scalable.

所有这些优点。

All of those things.

所以尽管生物科技发展得这么快，我们最好的药物还是小分子，而且大多数都源自天然产物。

And that's why still, despite all the advances in biotech, most of our best medicines are small molecules, and most of them are inspired by natural products.

我觉得这一类化学物质真的非常特别。

So I think there's something extremely special about that class of chemical matter.

我们当然也想过：能不能把这类化学物质变成一种给生物体‘加功能’的工具，而不仅仅是用来‘关掉’它？

And, absolutely, we asked, could we turn that class of chemical matter into a way to add function as opposed to just try to turn it off?

现在正值财报季，投资者们被一家又一家公司的财务报告淹没了。

It's earnings season, which means investors are deluged with financial reports from one company after another.

要消化这么海量的信息，还得抽空琢磨它们意味着什么，实在太难了。

It's hard to absorb the tidal wave of information and find time to think about what it means.

为了帮大家更好地应对财报季，Alfasense最近办了一场线上研讨会，邀请了两位投资人和一位前诺华竞争情报总监。

To help people better prepare for earnings season, Alfasense recently hosted a webinar with a pair of investors and a former Novartis competitive intelligence director.

AlphaSense的AI平台整合了海量的监管文件、分析师报告、专家访谈记录等数据。

AlphaSense's AI platform draws on a vast pool of regulatory filings, analyst reports, expert interview transcripts, and more.

想获取《财报准备与分析的AI指南》？请访问temperamentreport.com的节目摘要页面，点击链接注册。

Register for the AI Playbook for Earnings Prep and Analysis by following the link on the show summary on temperamentreport.com.

TR的订阅用户还可以直接通过今天节目邮件提醒里的链接进入这场网络研讨会。

TR subscribers can also go directly to the webinar from a link in the email alert about today's show.

生物分析的结果不该取决于你碰上CRO的哪位科学家。

And Bioanalysis should not rely on which scientists at the CRO you get.

这是实验检测，不是咨询服务。

These are assays, not consulting.

不管谁来做，结果都该一模一样。

Your results should be the same, regardless of who does the work.

我们确实喜欢跟CRO合作，但前提是得碰上靠谱的项目负责人。

We love working with our CRO when we get the right PI.

这根本不是优点，而是警告信号。

Is a red flag, not a compliment.

如果你的生物分析结果取决于哪个首席研究员恰好负责你的项目，那这根本不是和CRO的合作。

And if the quality of your bioanalysis depends on which principal investigator happens to be assigned to your study, you don't have a CRO partnership.

这简直就是买彩票。

You have a lottery ticket.

DASH从设计之初就围绕自动化和标准化流程，就是为了消除这种不确定性。

DASH was built around automation and standardized workflows specifically to eliminate that variability.

无论是ELISA、MSD、LCMS还是PCR，每个项目都能享受到同样的严谨标准和快速交付——以天计，不是以周计。

Every study gets the same rigor and the same rapid turnaround days not weeks across ELISA MSD, LCMS, and PCR.

符合GLP规范，价格透明，结果有保障。

GLP compliant, transparent pricing, guaranteed outcomes.

你根本不用操心谁在做你的项目。

You shouldn't have to ask who's running your study.

你只需要拿到高质量的数据就行。

You should just get great data.

访问 www.dash.bio 了解更多。

Visit www.dash..bio.

我们还是回头说说阿洛特拉辛B这个例子吧，你之前也看到了，它其实是一种抗真菌药。

Let's come back to the afloteracin b example because you saw that it had all well, I mean, it was an antifungal.

我觉得这主要是它的主要用途。

That was, I think, its primary use.

能杀灭真菌，治疗真菌感染。

Killed fungus, fungal infections.

但它副作用挺严重的。

But it had serious side effects.

我记得会损伤肾脏，还有别的问题。

I think damage to the kidneys and some other things too.

所以你是从一开始就想着怎么改造它，去掉这些缺点，还是后来才想到的？

So you thinking about how to engineer some of those properties out of it early on, or or did that come later?

说得对。

Great.

没错。

Yeah.

其实我们一开始是从分子假体这个方向入手的。

So we started actually with the molecular prosthetics thing.

我们想把两性霉素改造成CFTR的替代品。

We wanted to turn amphotericin into a replacement for CFTR.

一旦我们搞清楚了怎么用模块化的方法合成它，能开始调整结构、理解它的机制，最终就取得了一个关键突破：发现了你刚才提到的、它那臭名昭著的毒性的真正机制。

Once we had figured out how to make it in a modular fashion and could start modifying its structure and to try to understand it, we ultimately were able to make an important breakthrough, which was the discovery that the mechanism by which it causes its toxicity that you just mentioned and its notorious toxicity.

我的意思是，临床上大家都管它叫‘Amphoterrible’，原因你也都说了。

I mean, people call it amphoterrible on the clinic for all the reasons you pointed out.

它对肾脏的毒性太强了。

It's extremely toxic to the kidney.

大家都以为——其实有五千多篇论文都这么说——毒性是那个离子通道造成的。

Everybody thought, and actually, there's like 5,000 people saying this, that the toxicity was caused by the channel.

如果真是这样，那想把它改造成帮囊性纤维化患者康复的药，简直难如登天。

And if that were true, that's a very tough starting point for trying to turn it into something that helps you get better if you have cystic fibrosis.

但我们直觉觉得，这事儿没那么简单。

We had a hunch that that wasn't true.

我当时为了准备学术职位申请，读了那五千篇论文，发现文献里有些地方对不上号。

And I had read those 5,000 papers as I was trying to get ready for academic position proposals, and there were hints in the literature that didn't fit.

而且大家都知道这些分子会结合固醇，还有一种叫纳他霉素的短链分子，它不形成通道，却照样能杀死真菌细胞。

And it's also known that those molecules bind sterols, and there was a shorter one called natamycin that doesn't form channels but still kills fungal cells.

所以我们当时琢磨的假说是：毒性机制可能根本不是靠通道。

And so the hypothesis that we were chasing was maybe the mechanism of toxicity is not primarily the channels.

而是靠结合固醇。

It's the sterile binding.

于是我们用模块化合成法，把两性霉素的尾巴上去掉了一个原子——那个羟基，以前大家都觉得它对形成通道至关重要。

So then we used our block chemistry to make a derivative of amphotericin where we we removed a single atom from the tail, and that atom, the hydroxyl group, was thought to be critical for forming channels.

我们做出这个衍生物后发现，它确实不形成通道，但依然能杀死真菌细胞。

So once we made this derivative, we showed that molecule does not form channels, but it still kills fungal cells.

那一刻，我们真的迎来了一个重大突破。

And so we we then had a really exciting breakthrough.

我们终于明白，以前的机制理解全错了。

We knew the mechanism was wrong.

所有教科书都搞错了。

All the textbooks were wrong.

两性霉素确实能杀死细胞，但不是靠形成通道。

Amphotericin does kill cells, but not because of the channel.

而是靠和固醇结合。

It's because of the sterile binding.

最让人兴奋的是，这意味着两件事现在成为可能。

And the really exciting thing is that meant two things were now possible.

第一，我们可以把通道和杀伤作用分开了。

Number one, we could separate the channel from the killing.

明白了。

Okay.

所以我们决定放手一搏。

So we said, let's go for it.

咱们试试为囊性纤维化开发一种分子替代疗法吧。

Let's try to develop molecular prosthetic for CF.

但我们当时也有了一个关键的洞见。

But we also had an insight.

我们知道它为什么有毒，而且对这个问题特别感兴趣。

We knew now why it was toxic, and we felt drawn to that problem.

我们能不能想办法理性地消除它对人类细胞的毒性，却保留它对真菌细胞的杀伤力？

Could we figure out how now to rationally tune out its toxicity for human cells but keep its toxicity for fungal cells?

我们能不能开发出一种更安全的抗真菌药，不伤肾脏？

Could we develop a better antifungal that spares the kidney?

于是我们分两条路并行推进，我很兴奋地告诉你，这两条路现在都已进入临床，有望成为全新的疗法。

So we kind of then bifurcate along those two tracks, and I think I'm excited to tell you both of those now made it to the clinic as a potential new therapy.

这就是我们开启这段探索的起点。

So that's where we ended up starting that adventure.

明白了。

Okay.

你描述的这个问题很有意思，背后牵涉到生物学和医学上的重大挑战。

So you're describing, you know, an interesting problem set here that you wanna solve with biological and medical implications.

但正如你所说，你们确实得开发一种新方法来研究这个问题，你们管这叫‘阻断化学’。

But as you said, you did need to develop a new method to, like, begin to interrogate this, and this is what you call the block chemistry.

你能简单讲讲这到底是什么吗？关键突破在哪里？

Can you talk a little just a little bit about what what is that, and what was the the key innovation there?

当然可以。

Sure.

合成有机化学这门领域已经有大约两百年历史了。

So the field of synthetic organic chemistry is about 200 years old.

具体怎么算起点，看法不太一样。

It depends on where you define the the starting point.

要是以尿素的合成为起点，那我们差不多就快到两百年了。

But if you use the synthesis of urea as the starting point, we are almost there.

我们正站在这个门槛上。

We're right on the cusp.

要是以科里合成乙酸算起——那是人类第一次人工形成碳碳键，那就要从1845年算起了。

If you use, Colby Sync synthesis of acetic acid, which is the first carbon carbon bond made by a human, then it's, okay, 1845.

所以我们现在离那个目标也就几十年吧。

So we're, you know, a couple decades away.

但不管怎么说，这门学科已经有两百年历史了。

But either way, we're about 200 years old as a field.

说真的，放在人类历史长河里，这简直就跟一眨眼似的，所以它还特别年轻。

By the way, that's a blink of an eye, of course, in the course of human history, so it's a very young field.

但在整个这两百年里，合成有机化学一直是个高度手工艺化的过程。

But through the course of that entire period, okay, of about two centuries, synthetic organic chemistry has been an highly artisanal process.

全靠那些受过严格训练的专家，几乎都成了我们文化的一部分。

It's highly trained specialists, and it's almost part of our culture.

我觉得伍德沃德对这一切影响特别大。

I think Woodward, inspired a lot of this.

我们总是想方设法让每个分子都做得不一样，当成一种自我表达的艺术。

You know, we go out of our way to make every molecule as differently as possible as, like, an artistic expression of ourselves.

现在可选的反应多得像菜单一样，上千种反应，每种都有成千上万种条件组合，还有上百万种起始原料。

And the menu now includes, like it's like a thousand different reactions on the menu, each one run under thousands of possible different conditions and, like, a million different starting points.

明白吧？

Okay?

所以这种手工艺式的方法确实带来了巨大变革。

So that artisanal process has been transformative.

它对科学和社会的影响难以置信，但有个大问题。

The impact on science and society has been unbelievable, but and it's a really big but.

如果你开始想象自动化，或者现在人工智能已经介入其中，这种做法就太受限了。

It's a extremely limiting approach if you start dreaming about things like automation or now in that AI is on the scene interfacing with artificial intelligence.

最受限的一点是，只有经过严格训练的专家才能做。

And perhaps the most limiting thing is that only highly trained specialists can do it.

对吧？

K?

与其他领域不同，那些领域早就让他们的发现机制大众化了，因此受益于大量外行涌入带来的创意和创新——每次发生这种事都会这样，但合成有机化学从来没经历过这个。

Unlike other fields who've managed to democratize their discovery engine and thereby benefit from this rush of outsider effect into their space and all the interesting creativity innovation that comes with that every time it ever happens, Synthetic organic chemistry didn't have that.

当我们开始研究两性霉素衍生物的时候，我们就特别兴奋，因为感觉这可能是一种通用的方法，能用来合成大量其他物质。

And we started to get very excited when we were making those amphotericin derivatives that this felt like a potential general way to make a lot of other stuff.

所以我们对这个想法产生了浓厚兴趣。

And so we got very interested in the idea.

我们能不能只用一个反应，反复循环，用那些预先设计好所有官能团氧化态和立体化学的模块来合成小分子？

Could we make small molecules using just one reaction over and over and over again from blocks that are prefabricated with everything dialed in upfront, all the functional group oxidation, stereochemistry?

这个过程对机器人来说非常友好。

And that process looked very friendly to robots.

于是我们就动手干了。

So we went for it.

我们试着拼了个机器人出来看看行不行，结果还真成功了。

We tried to decide to kinda put a robot together and see if we could, and we were successful in doing that.

那时候我们特别激动，心想：天啊。

And and that was for us a very exciting time to say, wow.

也许现在我们可以利用这个发现。

Maybe we could leverage this now.

我们不再想着去造能做化学的机器，而是意识到：我们找到了一种机器能做的化学。

Rather than try to build machines that can do chemistry, we realized we had discovered chemistry that machines can do.

这真是一个特别激动人心的起点。

And that was a very exciting takeoff point.

这可是很久以前的事了，虽然还没到二十年，但也算挺久的了。

And this is all the way back, I mean, not quite twenty years ago, but a long time ago.

是的。

Yeah.

我们2015年发了一篇论文，证明了自动化是可行的。

We post that paper in in 2015 showing you could automate.

我们第一次展示迭代式碳-碳键形成是在2007年。

The first time we showed the iterative carbon carbon bond formation was 2007.

对。

Mhmm.

没错。

Yeah.

所以你们为了打造这个模块化系统、这些预制分子，差不多折腾了二十年，那这样能省下多少时间呢？

So you were working on this for close to twenty years to create this modular system, these prefabricated molecules that you could how much time could you save doing this?

这种事，随便找个研究生，短时间就能学会吗？

And and, like, is this the sort of thing that, like, you could train any graduate student to do in a short amount of time?

已经不像以前那样靠手工操作了？

It was no longer so artisanal?

太棒了。

Awesome.

对。

Yeah.

不只是研究生，谁都行。

Not just a grad student.

谁都能教会做。

You can teach anyone to do it.

这才是我们最兴奋的地方。

So that's what we're really excited about.

我直接跳过细节，给你讲个故事。

I'll fast forward and tell you the story.

有个特别厉害的学生，叫赛夫·威廉姆斯，他还在八年级的时候就加入了分子里子制造实验室，现在他已经上高中了。

This amazing, student, Saif Williams, who joined the molecule maker lab institute, when he was an eighth grader, and now he's in high school.

他开发了一个类似《我的世界》的分子构建游戏，教印度浦那的朋友们怎么玩这个。

He created a Minecraft for molecules and just taught friends in Pune, India, how to do this.

更厉害的是，他们真的从浦那把分子结构打印出来，发到了8000英里外的美国厄巴纳。

And, actually, they just printed molecules from Pune to Urbana, USA, 8,000 miles away.

这八个来自浦那的初中生，正在我们的分子实验室里打印分子。

These are eight middle school students from Pune printing molecules at our molecular lab.

所以我们觉得，化学这门学科，完全有可能变成每个人都能参与的领域。

So we think we have a chance to reimagine the field of chemistry as anyone's space.

所以你是想把它平民化。

So you're gonna democratize it.

但那时间成本呢？

But now what about the time parameter?

你刚才说的那个流程，传统做法要花多久？比如说，你是个制药公司，想针对某个分子靶点合成小分子化合物。

Because that process you described, the tradition of it, how long would it like, let's say you're a pharmaceutical company and you wanna make, you know, small molecules against a certain molecular target.

我们得花上几周甚至几个月，才能搞出这么一大套备选分子库。

We're talking like weeks or months to to create this, you know, this bounteous library of of options.

对吧？

Right?

没错。

Yep.

其实大家找新药的方法五花八门，对吧？

So there's lots of different ways people go about that process, right, of how to how to discover a new medicine.

但不管用啥法子，合成这一步总是卡脖子的环节。

But no matter how you do it, the synthesis part ends up being a major bottleneck.

明白吗？

Okay?

比如说，你第一步想先造一大波分子库，然后挨个筛一遍。

So let's say, yeah, let's say your first goal is to maybe make a very large library and screen it.

对。

Okay.

这个过程可能会花上很长时间。

Then that that process can take a long time.

但我觉得更关键、也更有趣的是迭代环节。

The the really important part, I think, though, that's even more interesting is it's the iteration.

当然，你不可能靠一次筛选就直接得到药。

So, of course, the you don't get a drug from a screen.

对吧？

Right?

你只是得到一个起点。

You get, a starting point.

接下来就得经过一个至关重要的过程：不断迭代和优化。

So then you gotta go through this really important process of iterating and optimizing.

每次这么做，都得去合成一些新的衍生物。

And every time you do that, you have to go about making, you know, some new derivatives.

这不就是先导化合物和先导优化嘛。

And if that's lead and then lead optimization.

没错。

Exactly.

对极了。

Exactly.

而且

And

没人关心你合成了上亿个化合物，关键在于那几个有效的。

that's So it doesn't nobody cares if you've made a billion compounds.

他们只想要那一个或两个真正起效的。

They want the one or two that work.

没错。

Exactly.

关键是要找到一种真正对人体有效、能改善健康、还不伤身体，同时具备其他所有理想特性的东西。

And and it's all about trying to find something that could actually work in a person and make them better and not hurt them and, have all the other properties.

这本质上是个复杂的多目标优化问题。

I mean, it's a complex multi objective optimization.

对吧？

Right?

所以通常来说，最好的情况下，这个周期要三到六个月，化学家会说：‘咱们有这几个新想法了。’

And so typically, the cycle time for that, okay, in a best case scenario, is like three to six months where the chemist say, oh, we've got these ideas.

这些是我们觉得可能更好的接下来十种化合物。

These are the next 10 things that we think could be better.

对吧？

Right?

然后你就得开始动手合成这些化合物。

And then so then you go through the process of trying to make them.

合成这部分耗时很长，之后还得测试。

The synthesis part takes a long time, then you gotta test them.

所以整个流程才这么慢。

And then that's why the whole process takes so long.

对吧？

Right?

如果你的周期要三到六个月，那难怪要花好几年、几千万甚至上百亿美元才能发现新药。

If your if your cycle time is three to six months, I mean, it takes that's why it takes years and many, many millions, sometimes billions of dollars to discover them.

所以我觉得关键不在于你多快能合成一个分子。

So I think the key, it's not necessarily about how fast you can make one molecule.

而在于你多快能不停地迭代。

It's how fast can you turn the crank.

而且这还没算上人工智能呢。

And and that's not even yet thinking about AI.

现在真正有意思的地方来了。

So now where things I think get super interesting.

说实话，我们早就预见到自动化要来了。

And I'll be honest, like, we saw the automation thing coming.

我们早就开始幻想，让机器来干化学这活儿。

Like, we we started dreaming about chemistry that machines can do.

说实话。

I'll be honest.

我们真没料到AI会这么快到来。

We didn't see the AI thing coming.

我们算是撞上好运了。

We got lucky.

说真的，运气正好赶上了——最近大家才真正意识到AI有多厉害，尤其是聊天机器人和语言模型这些玩意儿。

And and the stars just kind of aligned for us because, of course, recently, we've all learned how amazing AI can be and chatty bitty and discovering language models, etcetera.

所以我们后来明白，我们的模块化化学不仅适合机器人，也特别适合AI，因为你是在用模块化的方式合成分子，同时生成大量高质量的模块化数据，AI能像用语言一样用这些数据。

So what we realized then was that our block chemistry is not only friendly to robots, it's very friendly to AI because you're making the molecules in a modular fashion and generating high quality modular data that AI can use, kinda like language.

这就跟语言里的单词一样。

It's like the words in a language.

所以我们的这些模块，现在就是AI的‘词元’，完全可以拿它们来训练语言模型。

So our blocks now equal tokens, and you can build language models this way.

一旦你有了——

So once You have

给AI提供高质量的数据才行。

to give the AI good material good data to work with.

现在这个情况就是

And in this case,

最基础的原材料本身就是那些小分子物质，而这些物质是有潜力做成药物的。

it is the starting material here are the small molecules themselves that that could be drugs.

没错。

Correct.

那用这套自动化流程之后，整个研发周期能压缩到多短呢？

And how how quickly can you, shrink down these cycle times now with this more automated process?

说得对。

Exactly.

所以说，我能走到今天真的特别幸运。

So, yeah, this is where I've been super fortunate.

我有好几次机会都能和行业内的伙伴合作，还创办了自己的公司。

Multiple times, I have a chance to team up with industry and spin out companies.

我最新创办的这家公司叫Excelsior Sciences，我现在提起都特别激动。

This is where, the latest company I just started, which I'm so excited about called Excelsior Sciences.

现在这项技术已经实现工业化，整个周期缩短到仅仅一两周。

They've now industrialized this, and it's gotten to the point where the entire cycle now can just be a week or two.

AI 可以自己决定该合成哪些分子。

So, you know, the AI can decide what molecules to make.

机器人现在能自动合成这些分子了。

Robots can now make the molecules.

合成完后自动测试，数据再反馈给 AI 系统。

They can then be tested automatically, and the data goes back to the AI agent.

然后 AI 就会判断下一个该合成什么，好在下一轮学到最多东西。

And then, you know, the AI agent decides what to make next to learn as much as possible in the next cycle.

所以，人类历史上第一次——我知道这话听起来挺夸张，但我觉得真没夸张。

So you can now, for the first time in human history I know that sounds like a dramatic statement, but I think it's true.

对于碳碳键为基础的化学物质，我们现在能通过主动学习构建基础模型了。

With carbon carbon bond based chemical matter, you can now build foundation models through active learning.

这可是头一回，我觉得这正是我们社会现在拥有的、特别了不起的能力，我们正迫不及待想好好利用它。

And that's a first, and I think that's a very exciting superpower we now have as a society, and we're really looking forward to trying to leverage that.

你楼下走廊就有个动物实验室，养着小鼠和大鼠，能马上验证你的假设，当场就知道对不对？

And you have a wet lab right down the hall with mice and rats where you can test these hypotheses and get real world data, like, right away on whether you're right or not?

没错。

Yes.

‘治愈’大楼的四楼，就是吉姆·弗林为了在纽约推动生物科技而建的。

So there's the 4th Floor of The Cure, which is the building that Jim Flynn, built to build out biotechs in New York City.

迈克·福利是Excelsior Sciences的CEO，整个楼层的设计都是他一手操办的。

Mike Foley is the CEO of Excelsior Sciences, and he has designed this entire floor that he created.

左边是合成分子的机器人，右边是做测试的机器人，全都安排得明明白白。

And you've got the robots to make the molecules on the left, and you've got all the robots to do the testing on the right.

整个流程就在一个房间里跑起来，AI在云端驱动，亲眼看着真叫人震撼。

And in one room, the whole cycle runs with the AI in the cloud, and and it's an incredible thing to witness.

我想再多问问你和产业界的合作，当初在伊利诺伊大学，这事儿是怎么开始的？

Well, I wanna ask you a little bit more about your relationships with industry, but how how has that gone at or how did it get started, I guess, at University of Illinois?

他们那时候好说话吗？

Were were they friendly?

当时的管理人员或者其他人，支持你们和企业合作、把这些想法落地吗？

Were were were the administrators or people open to you working with companies to apply these ideas?

是的。

Yeah.

再说说我的运气吧，我确实挺幸运的。

So to add to my list of things, I've been very fortunate.

伊利诺伊大学在这方面特别有远见。

Illinois is very forward looking about this type of thing.

而且说到底，我们是一所赠地大学。

And in fact, we are a land grant institution.

我们的使命就是为社会做点真正重要的事。

So our kind of mission is do something important for society.

所以我们特别推崇基础研究。

And so we love basic science.

我的意思是，我们能钻得特别深，直抵地球核心。

I mean, we dig in as deep, journey to the center of the earth.

我的意思是，这固然是基础，但没必要非此即彼。

I mean, this is, of course, the foundation, but it doesn't have to be one or the other.

有时候你深入探索，挖到地球中心，反而发现了一些能真正改善社会的东西。

And sometimes you journey to the center of the earth and you find something that has a chance to help, you know, make society better.

所以在这里，把研究成果转化成实际应用，用来扩大影响、服务社会，是被大力鼓励的。

And so being able to kind of spin stuff out as a vehicle for scaling that and trying to make an impact on society, it's celebrated here.

这真的给了我极大的支持。

And that's just been tremendously enabling.

我知道很多其他地方的朋友，他们可没这种感觉。

I know a lot of my friends in other places, they don't feel that.

我觉得我得把这个故事讲出来：作为科学家，在实验室做基础研究，如果你真有热情，想把成果产业化，这同样值得骄傲。

And I think it's something that I wanna make sure to tell the story because I'll say, like, as a scientist, doing basic science in the lab and knowing that if you get excited and try to spin something out, that can also be celebrated.

关键是得营造这样的环境，让这一切成为可能。

You know, it's a really important thing to set up, you know, ecosystems where that's true.

我在伊利诺伊，一路走来，一直都拥有这样的支持，真的特别幸运。

And I've been very fortunate in Illinois to have that every you know, all along the way.

听到这个我真高兴，因为我知道你的很多工作都得到了国立卫生研究院的支持。

Well, I'm really glad to hear it because I'm sure a lot of your work is supported by the National Institutes of Health.

说真的，谁先开的这个玩笑我也不知道，但人家叫‘国立卫生研究院’，可不是什么‘国立《自然》论文研究所’之类的。

And, you know, I don't know who cracked the joke first, but it is the National Institutes of Health, not, you know, the National Institute of Published Papers in Nature or something like that.

当然，做这些事对推动领域内的知识进步很重要。

I mean, it's important for you to do those things to advance the knowledge in the field.

但是但是……

But but

是的。

Yeah.

我想再补充一点你刚才说的。

And I would add one more thing to what you're saying.

我经常在想一件事。

I mean, something I always think a lot about.

对吧？

Right?

我的意思是，这些钱可是纳税人的血汗钱。

I mean, these are taxpayer dollars.

对，就是这么回事。

I mean, they're, you know Yeah.

咱们国家有三亿多人口，人人都在交税。

Three hundred three hundred million people in our country that are always paying taxes.

作为科学家，我们真的很幸运。

We're very fortunate as scientists.

我们有机会做这些事。

We get a chance.

对吧？

Right?

能掌控这么大一笔钱，简直不可思议，而这些钱全都是纳税人辛辛苦苦挣来的。

I mean, it's an crazy amount of money to be able to, like, have control over it, and all of which came from taxpayers working hard and contributing.

所以，我确实这么认为。

So, actually, I do.

我有一种强烈的责任感，我想我们很多人都有。

I feel this tremendous sense, and I think many of us do.

这种责任感就是要把这些资金转化为对社会的实际贡献。

Tremendous sense of responsibility to turn that back into societal impact.

当然，能有一所大学支持并推崇这种理念，真是巨大的助力，我们非常感激。

And, of course, have a university that celebrates that has been a huge asset and very much appreciated.

那Revolution Medicines是你第一个参与的公司吗？

Now was Revolution Medicines the first company that you got engaged with?

是的。

Yes.

明白了。

Okay.

那你给大家讲讲，这个公司是怎么回事？

So tell folks the the story, how that one came to be.

太棒了。

Awesome.

是的。

Yeah.

所以，我第一次尝试搞个校内衍生公司，其实就从校园里开始的。

So my first adventure in trying to do something spin out actually started right here on campus.

我们这儿有个小孵化器叫Enterprise Works，当时对那种阻断化学特别感兴趣，觉得它潜力很大。

We have a little incubator space called Enterprise Works, and we were excited about the kind of block chemistry and what it might be able to do.

于是我们就创办了一家叫Midasyn的公司，M-I-D-A-S-Y-N，想显得有点创意——用我们发现的一种配体，叫mida硼酸酶，来实现对硼酸的开关控制。

And so, we started a little company called Midasyn, m I d a s y n, trying to be clever, about synthesis enabled by mida boronase, which is the kind of ligand we discovered to turn to be able to turn a boronic acid on and off.

当时我们就在校园里，带着两个员工启动了这家公司。

So we'd actually started that here on campus with two two employees.

我一个学生直接跳进来，成了第一个员工，后来我又从哈佛招了个研究生。

One of my students jumped and kind of was employee number one, and then I was able to hire a grad student from Harvard.

这两个人就在校园的孵化器里埋头干。

So there was the two of them working at the incubator space here on campus.

我们想推点东西出来，攒点势头，但那时候真就是小打小闹。

And, you know, we were trying to get some stuff going and and build some momentum, but it was it was small potatoes.

我记得当初一起搞这个创业的那个人，我们俩直接刷了三万五千美金的信用卡。

I remember the other person who helped me start this, we both put 35 k on our credit card.

这就是我们启动Mitterstadtas的方式。

That was how we started Mitterstadtas.

不过我也特别幸运，因为在读研的时候认识了一堆干大事的人。

And and but then I was also very fortunate because I had, you know, been in grad school with a lot of people who did amazing things.

很多人后来都跑去做了风投。

Many of them went on to do venture capital.

所以这里我又想说，真是睁大眼睛看清楚了。

And so this is where, again, I think it's eyes wide open.

我真的很幸运，能有这些机会。

I've I've been very, very lucky to have these opportunities.

其中一个是迈克·福利，我刚又跟他重新联系上了。

One of them, was Mike Foley, who I just looped back to.

没想到我们现在正一起合作搞Excelsior。

Turns out we're teaming up now on Excelsior.

但迈克主动引荐我认识了一些做风险投资的人。

But Mike went ahead and introduced me to some people who had gone on and done venture venture capital.

其中一位是亚历克西斯·博尔西，当时他在第三岩资本工作。

And, one of them was Alexis Borsey, who was at the time at Third Rock Ventures.

迈克做了介绍后，我就跟亚历克西斯讲了我的故事，他一听就特别兴奋。

And, Mike made the introduction, and I started telling Alexis the story, and it was so exciting.

我的天，他立马就上头了。

I mean, he jumped all over it.

他还邀请我去波士顿，给整个第三岩团队做路演，那真是一次难得的经历。

He invited me to come to Boston and pitch it to the whole Third Rock team, which was an amazing experience.

他们二话不说就决定投，我真是运气太好了。

And, I was so fortunate they jumped.

当时马克·戈德史密斯也在第三岩，后来他干脆跳槽出来，亲自当CEO，创办了革命医药公司。

Mark Goldsmith was also at Third Rock at the time, and then he chose to to jump ship and actually be the CEO and and launch Revolution Medicines.

他原本在帕洛阿尔托，就把整个项目搬到了加州，这整个故事简直太传奇了。

He's from Palo Alto, so he moved the whole thing out to California, and it's just been an unbelievable story.

我从他和他团队身上学到了太多，他们确实做了很多了不起的事。

I've learned so much from him and the team there, and they've obviously done some really incredible things.

这些成就很大程度上要归功于马克·戈德史密斯，是他让这一切成为现实。

And a lot of that credit goes to Mark Goldsmith who ended up making all that happen.

所以这家公司是在Third Rock的支持下起步的。

Well and so the company got started, with Third Rock support.

我记得当时写过，阿莫特ericin B 是你们当时在谈的、革命医药平台的核心应用。

And the Amphotericin b, as I remember writing about it at the time, that was the center project you were talk Revolution Medicines was talking about at that time as a key application of the block chemistry platform.

不过Third Rock更看重的是整个平台，以及它在其他领域可能带来的突破。

But, of course, Third Rock was interested in the platform and what it might be able to do for lots of other applications too.

后来他们把另一家Third Rock旗下的公司Warp Drive Bio合并了，那家公司原本在做RAS抑制剂，现在革命医药就在做RAS抗癌靶点——这可是癌症领域最顶尖的难题之一。

Turns out that they ended up merging another one of the third rock companies, Warp Drive Bio, which was working on RAS inhibitors, and now Revolution Medicines is working on RAS for cancer, one of the really big holy grail targets.

这段历史挺长的，你不用全讲出来。

Now there's a long history there, you don't need to go into all of it.

但说到底，这事儿跟你个人关系没那么大，也不是你主导的项目。

But, like, that's not your project or your thing necessarily.

但作为化学家，看到业界发生这样的事，你是什么感受？

But, like, as a chemist, when you look at things like that happening in in the field, what what's your reaction?

是啊。

Yeah.

亲眼看着这一切发生，真是令人惊叹。

It was it was a phenomenal thing to watch.

对吧？

Right?

业界在优先级上的投入程度，简直不可思议。

So the intensity for prioritization in industry is is incredible.

对吧？

Right?

这真的很重要。

And it's important.

对吧？

Right?

毕竟，归根结底，要想在这么大的规模上做点事，真正产生影响，你就得愿意把优先级放在第一位。

Because, obviously, at end of the day, to do something on that scale and really try to make an impact, you have to be, you know, willing to be super priority driven.

RevMed 最后能走到这一步，是因为他们手上有好几个方向可以选。

And, you know, RevMed ended up in a really great spot because they had multiple options of what to pursue.

明白吧？

Okay?

看到 KRAS 项目这么快就起来了，我真的特别激动，这可能会给那些由 RAS 基因突变引发的癌症患者带来巨大希望。

And, of course, it's just been phenomenal to see the KRAS program take off, and I'm just so excited about all of the potential impact that might have on people suffering from cancers caused by kind of Raspians.

不过话说回来，作为创始人，你刚才说得特别对，这真的会触动很多深层的情绪。

That said, as a founder, and you make a really important point, you actually bring up a lot of visceral emotions.

那时候特别难熬，因为 RevMed 最后决定把抗真菌项目放一放。

It was a very tough time because RevMed decided to deprioritize the antifungal program, Okay?

真的很难。

And it was hard.

那段时间真的特别煎熬。

It was a very tough time.

我记得有一天下午，我冲着马克在电话里大声争辩，想让他明白，抗真菌项目还是值得继续做的。

I remember one afternoon, I'll say I was speaking very, very loudly on the phone to Mark, trying to, you know, make my case that it was still worth working on the antifungal.

但那场仗我输了。

And I lost that that battle.

值得称赞的是，他做出了一个极其明智的决定，把所有资源都集中到了当时最该投入的地方。

Much to his credit, he made an incredibly smart decision to throw every weight into where it needed to be at the time.

于是，RevMed就彻底砍掉了抗真菌项目，那真的很难受。

And so then the the the antifungal program got killed at RevMed, and that was tough.

真的特别难熬。

That was very tough.

后来我们把项目又带回了大学。

And we brought it back to the university.

他们把知识产权还给了我们。

They gave the IP back.

但我心里很清楚，这项目还有希望。

And I believed in my heart of hearts, there was still potential there.

所以我特别幸运，能把这个项目带回实验室，继续往下做。

And so I was very fortunate to then bring it back into the lab, and we kept working on it.

后来我们终于取得了突破，找到了一条新路，研发出了一些远超以往的全新化合物。

And then we were ultimately able to make a breakthrough and find a path forward that led us to some really exciting newer compounds that outperformed anything we'd made before.

接着我又很幸运地搭上了Deerfield Management，一起创办了一家新公司。

And then I was very fortunate again to now team up with Deerfield Management and spin out a new company.

这家公司叫Funga Therapeutics，现在改名叫Elyon了。

This was Funga Therapeutics, which now has become Elyon.

我特别兴奋地告诉你，他们现在那个新一代化合物已经进入二期临床了。

And I'm very excited to tell you they've now got that next generation compound in phase two.

对我来说，这曾经是我一次重大的失败，但它教会了我坚持的力量——只要你心里觉得还有希望，就别轻易放弃。

So for me, that was one of my big failures that actually, you know, taught me the power of persistence and and how to continue to follow something if you feel like if you feel in your heart of hearts, there's still something there.

毕竟这是《长期主义》播客，花了二十年才走到今天，这也不算意外。

Well, this is the long run podcast, and so it's no big surprise that it's taken twenty years to get to where you think you need to be.

那你们这次的新化合物，最大的突破点到底在哪？

What are the the big advantages that you've been able to, crack open here with, the newest the newest compound?

是的。

Yep.

第一代化合物在消除肾毒性方面还算不错，但还不够理想。

So the first series was pretty good at eliminating the kidney toxicity, but not good enough.

简而言之，那段时间我们一直在钻研作用机制。

And long story short, in the inner in a time period, we kept working on the mechanism.

最终我们搞清楚了：efatercin主要是通过在细胞膜表面自组装成一种海绵状结构，这种海绵会比细胞自身合成速度更快地把胆固醇抽走。

And what we ultimately were able to understand is that the way efatercin kills cells primarily is through self assembly into a sponge on the surface of the membrane, and that sponge rapidly extracts sterols from cells faster than the cells can make more.

这其实是个特别有意思的基础科学故事——它甚至根本不是三维的问题。

So it's a very interesting story of fundamental science where it's it's actually not even a three-dimensional solution.

这是个四维的问题。

It's a four dimensional solution.

关键就在于提取的速度。

It's all about the rate of extraction.

一旦我们真正理解了这一点，又很幸运地和全球顶尖的固态核磁共振专家查德·林斯特拉合作，拿到了这种‘海绵’的高分辨率结构。

And once we fundamentally understood that, we were also very fortunate to team up with Chad Rinstra, who's a world leading solacean NMR spectroscopist, and we got a high resolution structure of this sponge.

基于这些发现，我们设计了一种衍生物，完全去除了它结合胆固醇的能力——也就是我们人体里的固醇——但保留了它结合并提取真菌固醇麦角固醇的能力。

With all of that, we could then design a derivative where we knocked out completely its ability to bind cholesterol, which is, of course, the sterile in you and me, but yet keep its ability to bind and extract ergosterol, which is the fungal sterile.

这一系列化合物为如今的AM2-19或EL2-19奠定了基础。

That series of compounds then set the stage for what now has become AM two dash nineteen or EL two nineteen.

这个化合物现在已经进入临床阶段了。

That's the compound that's now in the clinic.

你觉得这对抗真菌治疗会带来什么改变？

And what do you think this is gonna do for antifungal therapy?

说实话，药物研发太难了，我们始终怀着敬畏之心，知道前面的路还很长。

Well, I think if you're drug discovery is really hard, and it's with a lot of humility and understanding that we still have a long road ahead.

二期临床可不像我大学时以为的那样，是货架上随手就能拿的普通药片。

So phase two is not the Glendale wafer on the shelf as I remember from undergrad.

但至少，我觉得我们已经真正入局了。

But beyond I feel like we're on our we're we're in the game.

你懂的吧？

You know?

我们正在路上。

Like, we're on the journey.

从我们看到的所有数据来看，真的有很多让人充满希望的理由。

And and from all the data we can see, there's a lot of reason to be hopeful.

所以我愿意用‘谦逊而充满希望’来形容现在的状态。

And so kind of humbly hopeful is how I'll describe it.

而且我们全身心投入，就是要搞清楚：怎么才能把这事真正做成。

And and fully committed to figure out is, you know, how can we try to get something across the finish line.

你一直在这个平台技术上深耕，我能听出来你借鉴了不少其他技术。

Now, you've continued to work on this platform technology, and I can hear you borrowing from some of the other technologies.

你手头已经有了很多能实现原子级分辨率的工具。

You got a lot of these tools for atomic level resolution of targets.

就像你有——

Like, you you've got a

你已经掌握了所有必要的工具和方法，去攻克那些最难的问题。

lot of the the the tools and the methods you need to really go after hard problems.

我知道其中一个，咱们就花几分钟聊聊吧，在西雅图那边，有一家叫Cajal Therapeutics的公司正在研发它。

I know one of them, and I I don't we'll just talk about this for a couple minutes, but one in my area in Seattle is being developed by a company called Cajal Therapeutics.

这是一种铁螯合剂。

This is an iron mobilizer.

你能说说吗，咱们在怎么让铁在体内更有效运输这方面，到底学到了些什么？

Can you talk a little bit about, like, what you've been able to learn about the how how we transport iron more effectively in the body?

太棒了。

Awesome.

对。

Yeah.

所以呢，我特别幸运，我实验室里有一群特别出色的研究生。

So this is where, again, I was super fortunate also to have some amazing students in my lab.

我们当时在琢磨分子假体这个想法——能不能用两性霉素里的一个小分子通道，来替代CFTR蛋白？

And so we were working on the idea of molecular prosthetics and could we replace CFTR with a small molecule channel in amphotericin?

我有个特别出色的研究生。

And I had this phenomenal student.