Identifying Drug Targets—And Then the Drugs

To make new drugs that treat a particular illness, scientists first must understand what’s going wrong in the body.

At Yale School of Medicine (YSM), researchers are uncovering the fundamental mechanisms underlying how the body works, how diseases arise, and how drugs interact with the body. They are using these insights to create new drugs for conditions ranging from kidney diseases to neurodevelopmental disorders.

“It’s important for us to understand how the world works, including living systems,” says Carl-Mikael Suomivuori, PhD, assistant professor in the Department of Pharmacology at YSM. “Understanding these systems enables us down the line to make better therapies.”

Suomivuori specializes in physics-based methods that predict how biomolecules move. By understanding biological processes at the atomic level, he hopes to create safer and more effective therapeutics. One area of interest in his laboratory is “binding kinetics,” in which his team adjusts how slowly or quickly drugs bind to receptors in our body.

“Rates of binding and unbinding are actually very important for determining what effects occur in a biological system,” Suomivuori says.

Suomivuori’s team simulates how molecules bind to G-protein coupled receptors (GPCRs), which are the largest class of proteins in the body and are found on the surface of cells. They play a role in a wide range of bodily functions and are a major target of interest for drug developers. As much as a third of all drugs work by acting on these receptors, including medications for allergies, treatments for blood pressure, and opioids. Suomivuori hopes his work will lead to the identification of further therapeutics that target GPCRs.

He is also interested in a class of transporter molecules that enables crosstalk between neurons. One neuron communicates to another by releasing chemical messengers called neurotransmitters through a small gap between the cells. In order to travel across this gap, or synapse, the neurotransmitters must first be contained in small organelles called synaptic vesicles. Suomivuori’s laboratory is focused on the transporters that package the neurotransmitters.

“These transporters play a huge role in regulating our brain chemistry, but they’re poorly understood,” Suomivuori says.

His laboratory is conducting simulations to better understand how these transporters work. Vesicular monoamine transporter 2 (VMAT2), for example, is responsible for packaging neurotransmitters like serotonin and dopamine, and VMAT2 inhibitors can help treat involuntary movement disorders such as tardive dyskinesia.

“It’s an important drug target, but people don’t know how it works,” Suomivuori says. “We’re trying to figure out the mechanisms of transport and how to design new drugs that modulate VMAT2 in desired ways.”

The emergence of artificial intelligence (AI) could dramatically speed up the drug discovery process and far surpass what can be achieved with experimental methods alone. Assaf Alon, PhD, assistant professor of pharmacology, explores the use of an AI-powered algorithm, known as AlphaFold, for predicting the three-dimensional structures of membrane proteins.

“If you get a good picture of the structure, you can infer a lot of information about function,” he says.

Newer generations of AlphaFold not only help researchers study protein structure, but also predict how various molecules bind with proteins. Alon has been collaborating with scientists at The Rockefeller University who have modeled the interactions of billions of compounds with specific protein receptors. Comparisons of Alphafold’s predictions with the outcomes of traditional experimental methods have shown that AI is effective in identifying promising drug candidates.

Alon and his collaborators are honing the algorithm to further improve its reliability. “Being able to do almost all of your research from your laptop instead of going into the lab saves a lot of time and a lot of money,” Alon says. “The goal is to make the drug discovery process better and more efficient.”

For instance, Alon’s team is especially interested in a family of proteins known as EXPERA, which has been implicated in a range of diseases. TM6SF2 is a member of this family that helps regulate the secretion of very-low-density lipoproteins (VLDLs), a type of “bad” cholesterol, from the liver. Certain mutations in this protein are associated with a greater risk of liver disease.

To better understand how to target TM6SF2 with potential therapeutics, Alon’s team is studying its structure and using artificial intelligence to identify compounds that bind with the protein. “We were able to find better and better drugs that bind to this specific protein,” Alon says.

Now, they are studying whether the binding of these compounds with TM6SF2 has any biological impacts, such as changes in VLDL secretion.

Identifying key targets has led to potential new therapies for diseases. Several promising therapies for autosomal dominant polycystic kidney disease (ADPKD), for example, have arisen from collaborations between researchers in molecular physiology and nephrology at YSM.

“Twenty-five years ago, scientists had little molecular understanding of what was going on in ADPKD,” says Stefan Somlo, MD, C.N.H. Long Professor of Medicine (Nephrology). “Now, we are using basic molecular genetic discovery to actually develop real treatments.”

Each of our kidneys has more than 1 million nephrons—the tubules that filter blood and remove waste products. ADPKD causes some of these nephrons to turn into cysts. Over time, these cysts expand and displace the surrounding healthy tissue, causing significant health consequences. While the average kidney is about the size of a fist, a kidney affected by ADPKD can grow as large as a football.

“These cysts only form on a small subset of cells, but they become so large that the normal parts of the kidney get scarred and compressed,” says Lloyd Cantley, MD, C.N.H. Long Professor of Medicine (Nephrology). “We really need ways to target that cyst growth and slow it down so that the remaining healthy cells can do their jobs.”

The vast majority of cases of ADPKD (80%) are driven by mutations in the gene PKD1, which encodes the protein polycystin-1. YSM researchers have been investigating the function of the protein and what goes wrong when it’s absent. They discovered that polycystin-1 regulates cellular metabolism by undergoing a cleavage that releases a small piece of the protein, which then travels to mitochondria.

“Cellular metabolism shows major dysregulation in polycystic kidney disease and may be a therapeutic target,” says Michael Caplan, MD, PhD, chair and C.N.H. Long Professor of Cellular and Molecular Physiology.

A team led by Caplan and Somlo then created mouse models of ADPKD by turning off PKD1. When they delivered this small piece of polycystin-1 into the animals, they found that its expression helped reduce the progression of the disease. They are now working with a biotech company to explore the use of gene therapy targeting the polycystin-1 piece.

Diseased cells with ADPKD contain high levels of a signaling molecule that drives cell proliferation called cyclic adenosine monophosphate (cAMP). Additionally, a signaling pathway that also regulates cell proliferation called mechanistic target of rapamycin (mTOR) is always turned on in ADPKD. “We don’t know why these things happen, but we know that if we could stop them, maybe we could control the growth of cysts,” says Caplan.

Caplan’s and Somlo’s teams also identified an enzyme in cells called adenosine monophosphate-activated protein kinase (AMPK), which acts as an energy sensor. When energy levels in the cell fall, the enzyme turns off cellular pathways using up energy, including those involved in growth and proliferation. YSM researchers wondered if activating AMPK could help turn off the pathways driving ADPKD like cAMP and mTOR.

Metformin, a commonly used drug for type 2 diabetes, activates AMPK. Treating animal models of ADPKD with metformin slowed cystic growth, the researchers discovered. And the findings led to a large clinical trial, conducted by researchers in Australia, testing the ability of metformin to treat ADPKD in humans. The Phase III trial is still ongoing.

Through this kind of work, YSM has established itself as a major center of ADPKD research. Researchers are exploring many other targets for therapies. Somlo and Cantley, for example, are interested in the role of cilia, the small, hair-like sensory organelles that help cells sense their surroundings and play a role in signaling. Cantley is also exploring the mechanisms underlying the dysfunctional enlargement of tubules that leads to cysts. “There are parallel but distinct pathways that can be targeted in different ways therapeutically,” says Cantley.

Today, patients with ADPKD have limited treatment options. There is one medication available, tolvaptan, but its benefit is modest and it only works in a subset of patients. The researchers are optimistic that their work will lead to better options for patients. “There really is an opportunity here to find treatments or a cure,” Somlo says.

For decades, Anthony Koleske, PhD, Ensign Professor of Molecular Biophysics and Biochemistry, has been working to understand the mechanisms underlying neuronal development and signaling. He is particularly interested in a protein named triple functional domain (TRIO). Mutations that alter the function of this protein are associated with neurodevelopmental disorders such as autism spectrum disorder and schizophrenia.

Koleske’s research on the biochemical and cellular functions of this protein has revealed potential avenues for new treatments.

“The idea is to use a drug-like small molecule to repair the signaling defects resulting from mutations in TRIO,” he says.

There are two main types of mutations of TRIO associated with developmental delays and intellectual disability: Gain-of-function mutations hyperactivate certain parts of TRIO and loss-of-function mutations hinder its activity. Gain-of-function mutations are associated with macrocephaly, or a large head circumference, while loss-of-function mutations are associated with microcephaly, or small head circumference.

Koleske’s lab has generated mice bearing these gain- and loss-of-function alleles and showed that they replicate these changes in head size and also result in deficits in behaviors known to be impacted in humans with TRIO-related disorders. They’ve also found that these mutations are linked to deficits in synaptic transmission, the process through which neurons communicate.

To identify potential therapeutics, Koleske teamed up with the Yale Center for Molecular Discovery (YCMD), a Yale core facility that helps screen for potential drugs. In partnership with Yulia Surovtseva, PhD, director of YCMD, Koleske’s team has identified a series of promising compounds that inhibit the protein’s activity and could help treat dysfunction caused by gain-of-function mutations.

“If we can improve the potency a bit and get them into the brain, they could potentially be therapeutics for treating individuals with the hyperactive form of mutations,” Koleske says.

The researchers are optimistic that they will one day develop a therapeutic that restores synaptic activity.

“This is where having spent many decades trying to understand mechanisms of controlled neuronal development and signaling pays off,” Koleske says. “This is a clear example of where we could repair mutations with small molecule drugs. We could make a real difference.”