Ever wondered how medicines work? Here, I am not trying to quiz your pharmacology background but rather have some fun triggering your first-principle thinking. What is the common mechanism of action that applies to all drugs, new and old, and is that an absolute prerequisite for any drug to work? The very simple answer is that drugs must bind to a target to work. The target here is a biological molecule associated with the disease via enzymatic, structural, or genetic activity. So, now we have established that any drug must bind to a relevant biological target, usually on or inside our cells. Building on this first principle, how could better drugs be made?

The limitations of conventional drug discovery

Identifying molecules that optimally bind the disease targets is the obvious answer that jumps to mind, and this is what has been keeping thousands of industry and research labs busy for over a century. The conventional drug discovery process hones in on disease targets identified from basic research and starts by screening libraries of tens to hundreds of thousands of molecules to identify the candidates that best bind the target. Once the best binding candidates are identified, testing in tubes, cells, and animals is initiated to establish potential efficacy and marginal safety. Good-performing candidates transition into clinical trials to ensure efficacy and safety in patients; these have a dismal success rate of 7.9% despite the huge investment and fine-tuning in all the preclinical stages.

Going back to our first principle of drug-target binding or interaction as the foundation of drug efficacy, we can easily deduce the reason for the high failure rate of the drug discovery process. The entire pipeline focuses on optimizing binding to the target while largely neglecting the specificity of binding. For drugs to work properly, it is equally essential that they bind only the intended target with minimal "crosstalk" to other targets. Most conventional small-molecule drugs lack specificity, which causes side effects. They often get processed by the body, "metabolized," to make other molecules that might have other unintended effects and/or high toxicity.

How does nature manage it perfectly?

How does biology manage to avoid unintended interactions, though? How do enzymes, transcription regulators, structural proteins, genomic enhancers, RNAs, and the wide tapestry of the molecules of life elegantly find their interaction partners in the complex soup of molecules within cells on the cell surface? The size of life molecules is part of the secret here. Most biological molecules are large and complex, and this complexity allows for higher statistical variety, which, in turn, allows for coevolving interaction partners with maximum specificity. Meanwhile, small molecules, due to their lower size and complexity, are more likely to fit multiple targets.

Paradigm shifts, modern, and future approaches in drug discovery

Despite this limitation, brilliant small molecules have been identified that make for reliable drugs, especially when the affinity to the desired target is much higher than that of the unintended ones, making safer and more tolerable drugs. Emerging approaches in small-molecule drug discovery are set to flip the inefficient drug discovery paradigm by starting with clinically relevant drugs and working backward to generate leads with increased safety and efficacy. Technological developments allow for screening molecular libraries, predicting molecular docking, and target binding. Specialized generative AI platforms like Chemistry42 can even come up with entirely new small-molecule drug leads with optimal target binding. Similarly, generative AI potential, coupled with the emerging power of predicting protein folding, extends to the use of AI platforms in designing entirely new proteins. These unlimited possibilities synergize and extend the potential of complex biological therapies, including gene and cell therapies, which will be the focus of the remainder of this article.

Guided by first principles and inspired by nature

Let us now summarize the first principles we have discussed so far. To function, drugs need to bind to their targets. Statistically, in biological molecules, increased molecular size and complexity increase the potential set of molecular interacting partners and reduce the potential of unspecific interactions. Side and reverse interactions of small molecules are mainly due to unwanted interactions enabled by their small size and low complexity. It logically follows that increasing the molecular size and complexity should give rise to better medicinal molecules, a concept well known to medicinal chemists who work on improving drugs by derivation, addition, or alteration of chemical groups in the molecular structure.

Furthermore, nature has given rise to complex medicinal molecules, whose complex structure enables them to target very complex biological targets. Rapamycin is a remarkable example as a macromolecule in its ability to specifically target mTOR complex 1, one of the most sophisticated regulatory protein complexes in eukaryotic cells, which I will be addressing in detail in a future article.

Molecular biology heralds the dawn of biological medicines

Aiming to achieve optimal target matching, the next generation of drugs is taking inspiration from nature. The elucidation of the DNA → RNA → Protein central model of molecular biology, along with the basic technologies of PCR and molecular cloning, has allowed the introduction of recombinant proteins as essential and life-saving medications. Being easy to produce in bacteria, recombinant human insulin has changed the lives of millions of diabetics. Meanwhile, more complex proteins, including antibodies, are produced in mammalian cells or organisms.

The example of insulin and other hormonal drugs administered into the circulation represents the simplest example of biological medicines since these work by reaching the bloodstream. Nonetheless, the required tight regulation of insulin blood levels to mimic healthy physiology is still an open challenge, with cutting-edge cell therapies offering a potential solution.

Returning to the initial premise of drugs working by binding to their biological targets, what could be done in diseases where the entire target is missing or faulty? This is the basic premise of genetic and hereditary diseases, such as hemophilia, sickle cell disease, and more devastating early-onset diseases such as muscular dystrophies and hereditary blindness. A major distinction can be drawn here between diseases like hemophilia, where replacing the missing clotting factors VIII or IX in the blood is enough, akin to the case with insulin, and debilitating diseases that see affected tissues or organs lose function due to the missing or faulty genes. For these diseases, the only conceivable solution is to introduce the missing gene or interfere with its expression to eliminate the faulty part, the mutation.

Either case requires the introduction of genetic material specifically into the affected tissues and organs while ensuring minimal interference with non-target tissues and cell types. As we all learned in school, viruses are the most efficient hackers of cells, with remarkable abilities to introduce their genetic material into cells and hijack their DNA replication and protein synthesis machinery to reproduce and spread.

Gene therapy, hacking our own cells

Ironically, we humans have learned how to harness the remarkable cell entry and expression abilities of viruses to our advantage. Viral vectors, as they came to be called, are "hollowed-out" viruses whose original genomes, including the parts essential for replication, have been replaced with the missing genetic elements needed to treat human diseases. This is possible since viral coats (known as capsids) can package any genetic material of the right size, provided it is labeled with the same "flanks" as the viral genomes. The superstars of viral vectors are a group of very small viruses called Adeno-Associated Virus (AAV). These are non-pathogenic viruses that were accidentally discovered in the 1960s and have since been engineered in many ways to improve their tissue and cell targeting capacity and accuracy.

By now, there are already five approved AAV-based gene therapies treating hereditary disorders ranging from hemophilia and hereditary retinal disorders to debilitating muscular and spinal muscular conditions. These vectors continue to offer hope for patients with monogenic disorders that can be pinpointed to a single missing or malfunctioning gene.

In contrast to the stealthy and low-profile small AAVs, larger and more virulent viruses, such as Adeno and Herpes simplex, can be "vectorized" for local use against some cancers or topical genetic diseases. Hence, these viruses can deliver larger genetic cargos more efficiently without having to worry about systemic toxicity due to topical or localized application.

DNA vs. RNA gene therapies, it is all about durability

Going back to our first principles, gene therapy is essentially the ideal answer for diseases where the target itself is missing or faulty. Instead of providing a ligand for an existing target, gene therapies depend on delivering the genetic instructions to cells to make the correct target from scratch or to correct the expression of a faulty gene to express the normal target.

What viral gene therapies also have in common is the delivery of genetic instructions in the form of ultra-stable DNA molecules that have been shown to persist for over a decade in target cells. This has the advantage of a lasting effect, which is desired in diseases arising from a complete lack of the target gene. However, long-term persistence of DNA turns into a disadvantage in other cases, in which a transient effect is desired due to toxicity, dose dependence, or other long-term safety concerns.

Therefore, the field of RNA biology holds massive potential for introducing gene-based therapies for non-genetic diseases where transience and reversibility of expression are desired. This is because RNA has a much shorter life in the cell as it gets degraded. Furthermore, the durability of the delivered messenger RNA can be modified by manipulating the sequence of the non-expressing molecular "tail" of the RNA.

The coincidental rock star of gene therapies, RNA, goes global

The discovery of RNA nanoparticles that are preferably taken up by specialized immune cells called dendritic cells that prime the adaptive immune system against new targets has inspired the possibility of RNA vaccinations. Though this started and remains a promising avenue in treating and vaccinating against cancers based on their aberrant changes that can be recognized and targeted by the immune system, the largest experiment up to date has been the mRNA vaccines against the COVID-19 viruses, with over 12 billion COVID vaccine doses dispensed worldwide, most of which are RNA vaccines. Thus, it constitutes the largest-ever global proof of concept for the safety and efficacy of a gene-based therapy, despite the imperfections that are yet to be overcome and improved upon. The pandemic emergency has allowed for an unprecedented acceleration of messenger RNA delivery, and this boost is already reflecting positively on the entire fields of gene and cell therapy as their promise seems closer in reach than ever.

From genes to cell therapies, why repair what you can make?

As we established, gene therapies deliver a correct target or "repair" a faulty one in cells, assuming that the target cells exist and are waiting for the delivery of the correct gene to function. This leaves open the cases in which the entire cell type or tissue is destroyed, faulty beyond repair, or simply missing altogether. Examples of these cases are many acquired conditions such as type 1 diabetes, degenerative diseases including neurodegeneration, and cancers. The cancer example needs further clarification, since most would think that, if anything, cancer is characterized by too many cells and not missing ones. However, from the perspective of the immune system, we could define cancer as the lack of adaptive immune cells that can eliminate the tumour. This is because in a healthy individual, cancer cells constantly form and are eliminated by the immune system before they progress to forming a tumour.

Indeed, cancer turns out to be the first and most promising target for engineered cell therapies. These therapies make use of T-cells with synthetic receptors called chimeric antigen receptors, which have the ability to bind abnormal targets unique to cancerous cells. This allows CAR-T cells to specifically eliminate cancerous cells without wreaking havoc on healthy ones.

The ability to program cells to respond to physiological and/or synthetic cues, such as hormones, nutrients, specific chemicals, light, or even music, gives rise to theoretically limitless potential for cell therapies. They will soon enable the restoration of tissue function with full control over the time, location, and magnitude of the desired therapeutic response. Though "functional" cell therapies are still in their infancy, the recent FDA approval of the first cell therapy for type 1 diabetes and the promising ongoing clinical trials of stem-cell-based therapies for Parkinson's offer much promise for unlocking the vast potential of cells as living medicine. As the line between disease and age-related decline blurs, it is easy to imagine a future in which cells are engineered to replace declining tissues and reverse aging altogether. This is another exciting topic that I will be addressing in a future article.

The century of biology

From fire to atoms, the story of human ingenuity has always revolved around first understanding a phenomenon, interacting with and manipulating it, and eventually becoming able to reverse engineer and build it from scratch. When it comes to biology, our mastery is coming of age as we transition from tweaking and manipulating biological systems using low-specificity approaches to a full understanding of diseases on a molecular and genetic level and the development of a comprehensive toolbox to engineer perfect solutions for medical and countless other applications.

As Craig Venter and Daniel Cohen put it, "If the 20th century was the century of physics, the 21st century will be the century of biology."


1 FDA Drug Discovery and Development.
2 In-Silico Medicine Chemistry42.
3 Intermolecular Interaction an Overview.
4 Li Y, Zhao D. Basics of Molecular Biology, Molecular Imaging, July 2013, 19:541–601, doi: 10.1007/978-3-642-34303-2_16, PMCID: PMC7122053.
5 Approved Cellular and Gene Therapy Products.
6 Samulski, R. Jude, and Nicholas Muzyczka, "AAV-mediated gene therapy for research and therapeutic purposes, " Annual review of virology 1, 2014, 427-451.
7 Barbier, A.J., Jiang, A.Y., Zhang, P. et al, The clinical progress of mRNA vaccines and immunotherapies, Nat Biotechnol 40, 840–854, 2022.
8 Sterner, R.C., Sterner, R.M. CAR-T cell therapy: current limitations and potential strategies, Blood Cancer J. 11, 69, 2021.
9 Bloomberg Covid Vaccine Tracker.
10 The Century of Biology.