​Making Drugs Out of Dirt Is Really Hard

If you're looking for new medicines, a good place to search is in the dirt. Many of the molecules that have made human life simpler—like those used i​n antibiotics—have been found by simply analyzing the ground, where microorganisms work relentlessly to synthesize precious, possibly curative stuff.

The good news is that there are very likely still some useful products buried in the world's dirt. The less good news is that coming across them is very hard.

A team of biologists of Rockefeller University in New York recently launched the website Drug​s from Dirt. The site is a clarion call for people around the world to grab a shovel and send samples of soil to the folks at the university, who will scour them for interesting elements. The same team also published a pa​per this month in which they underline how recent analyses "suggest the existence of an enormous untapped reservoir of natural-product-encoding biosynthetic gene clusters in the environment."

I got in touch with Sea​n Brady, one of the authors of the paper and head of the university's Laboratory of Genetically Encoded Small Molecules, to ask him what the deal was with dirt and drugs.

MOTHERBOARD: How come you're looking into dirt to find drugs?
​Sean Brady: Well, the majority of clinically used substances, like antibiotics, anti-cancer agents, immunosuppressants… I mean, if you go down the list, the majority of them actually come from nature, and the most productive source for them, traditionally, is bacter​ia. And the most productive way of finding bacteria is looking into dirt. Historically, that's been the primary way we used to get drugs.

What we have learned over the past couple of decades is that we probably missed most of what bacteria can actually do for us in producing these [medicinal] molecules, because we can't culture most of them, and they actually don't like to produce molecules in the lab. So we've been working on ways to just directly go into dirt and access the genes that make those molecules, so that we're circumventing those limitations.

How does the whole process work out?
​So, essentially, we take a spoon and we dig out some dirt and then we keep that in the presence of a detergent that lyses [i.e. disintegrates] the bacteria that are in there. Then we freeze the DNA and we begin to sequence that DNA using some specific strategies like polymerase chain r​eaction. It usually takes a couple of hours to work up the sample, and about a day to sequence it. What we want to do is find genes that indicate they might be making either molecules that look like things we know about—like antibiotics, so we can make new versions of those—or genes that don't look like anything we've seen before and might be new molecules we still don't know about, and which need further research.

Have you found any good drugs so far?
​Although the project is called "Drugs from Dirt," we're not actually aiming at getting to the drug-production stage. It's more an attempt to sequence genes that you can find in dirt, so we can get some vision of what's out there. But then individual countries should control their own biodiversity.

We have a purely scientific interest in what's out there. We want to see how things differ around the world: Are the chemistry and the genes you find in Texan dirt the same as those you see in Africa? We don't think so, but we would like to see it in more detail. We want to get enough information that other people in those countries can go look. And then we could also collaborate with them.

So you're trying to create a sort of international database of the molecules you find in dirt.
​A database of genes, actually. Genes that we know can make useful molecules. It's just a starting point of a process that could eventually lead to getting drugs. It's also about inspiring people to realize there's so much stuff under your feet, which we can begin to see if we sequence it.

Let's get into the details of the sample-gathering. You're asking people around the world to send you samples of dirt. How does it work? Do they just send you a box of dirt?
​In some cases yes, but it depends. It's pretty easy to send dirt samples around the United States; there are not many restrictions. So one of our main focuses is getting a nice distribution of samples from the US. But ultimately, we'd love to have samples from all over the world. And there are many more restrictions to that: There are import restrictions, there are legal restrictions to taking biodiversity out to other countries. They are issues we have to deal with, but we hope that people can get around them. We also provide them with the legal assistance necessary to import dirt samples to the United States.

How many samples have you got so far?
​For our latest paper we have used 185 samples, a small fraction of which­­—I'd say a quarter—came from citizen scientists' efforts. Now, with the website, we've launched the first large-scale attempt to get samples from around the world. We'll see how it goes. It's a big experiment in citizen science, which I think is a great model to use for large-scale, global efforts that want to look at diversity.

What's the hardest part of what you are doing?
​As for the "making drugs from dirt" aspect, the hardest thing is turning a gene into a molecule. You've got to have a kind of bacteria to express a gene, to make a protein, and then to make an antibiotic from that. You are cloning DNA out of dirt and asking the bacteria to do a whole lot with those foreign genes, and it doesn't always work out. As for the sequencing aspect, the technical procedure is quite simple, and the hard part is getting a lot of samples from different places.

That's why it's important to get the public involved to send samples. We can't travel around the world to get samples, it would be too cumbersome and too expensive. But everyone lives everywhere now. We are especially interested in samples from strange places where the bacteria and the genes might be different from the ones we've sequenced before.

What's next for this research?
​At some point we'll need to do a more detailed map of global biodiversity. We have already created a map out of the 185 samples we've received. But it's very sparse: Less than 200 samples around the world don't really cover anything. We want a detailed map of the kinds of genes you can find around the world. And then we want to figure out where are the best places to get good molecules, and start to get molecules from there, and make drug discoveries out of it.