The idea of treating the human body with light is nothing new. Herodotus of Halicarnassus asserted that sunlight is essential to our health, and he was already attempting to restore health through the use of “heliotherapy” in ancient Greece. Today, light is recognized as essential for the production of vitamin D (which helps preventosteoporosis) and melanin (a protective skin pigment), and it is still used to treat certain conditions, such as some forms of neonatal jaundice.
But if we know that light can interact with elements in the body, can we use it to influence the medications we take? Certain molecules can, in fact, change their conformation (their three-dimensional shape) in the presence of light. In cases where only one of these forms exhibits biological activity, it is therefore possible to activate or deactivate the drug’s effect on demand through light irradiation. Furthermore, certain types of light (red and infrared, in particular) can penetrate the skin and tissues, allowing for highly precise targeting of the affected organ, while remaining minimally invasive and easy to administer.
This idea is of considerable interest: it would make it possible—to cite just two examples of global health issues—to prevent the emergence of bacterial resistance caused by the uncontrolled use of antibiotics, or to reduce the side effects associated with the toxicity of chemotherapy.
As is often the case at the interface between chemistry and biology, it is based on the observation of a natural phenomenon: the mechanism that enables animal vision, in which a molecule called retinal plays a role (a derivative of vitamin A, which the body produces from β-carotene—hence the idea, which is partially true, that eating carrots is good for your eyesight). In the dark, retinal is bound to a protein called opsin in a bent conformation. But when the eyes are exposed to light, it suddenly adopts a linear conformation, causing it to detach from opsin. This phenomenon triggers a series of reactions and ultimately leads to the brain receiving an electrical signal, which it interprets as a visual image.
Retinal is therefore what is known as a “photoswitch”—a light-sensitive molecule capable of changing its shape in response to light: depending on the light, it may or may not interact with a protein.
What if this concept could be extended to certain drugs, which, depending on their form, might or might not exert their effect? Indeed, after administration, a conventional drug spreads throughout the body and, while it certainly acts on its target, it often affects other parts of the body as well, leading to side effects. In the case of a “photoswitch” drug, it is administered in an inactive form, and only when exposed to light does it change form and exert its effect. Its impact is therefore limited to the irradiated area for a specific period of time, and the rest of the body and the environment can more easily be spared. This is referred to as photopharmacology, as opposed to classical pharmacology.
This field is still in its early stages, and “photoswitches” have only just reached the first phase 1 clinical trials, such as the compound “KIO-301” from Kiora Pharmaceuticals, which was developed to treat retinitis pigmentosa, a serious genetic disorder that causes progressive vision loss.
But for the past decade or so, several research teams have been working on developing applications for this concept, primarily by introducing an azobenzene moiety into existing bioactive molecules. Azobenzene is one of a class of chemical structures capable of changing shape when exposed to light; it is notably used in the manufacture of azo dyes and pigments traditionally used in dyeing and painting to produce color ranges from yellow to red.
The chemical modification of antibiotics such as ciprofloxacin to make them light-sensitive is a prime example described by the team led by Nobel Prize in Chemistry laureate Ben L. Feringa. While inactive or only slightly active against bacteria in their original form, these molecules acquire antibiotic properties when exposed to UV light. Thus, the drug’s action can be controlled in both time and space, and the short half-life of the irradiated form (a few hours, after which the molecule spontaneously reverts to its original form) prevents any residual antibiotic activity from persisting in the environment.
Similarly, modifying the structure of methotrexate—a drug widely used in cancer treatment but one that causes severe side effects (depression, liver cirrhosis, pneumonia, etc.)—led to the discovery of phototrexate, its “photoswitch” version.
While the original form of phototrexate has no effect, exposure to UV light causes it to adopt another form, whose toxicity is similar to that of methotrexate. In vivo experiments conducted by the teams led by Soler and Gorostiza on zebrafish embryos treated with phototrexate for five days showed that while embryos kept in the dark had a low mortality rate after five days, other individuals irradiated with UV light twice a day experienced an eightfold increase in mortality, indicating that the drug was activated by light (light alone had no significant effect over five days). Molecules such as phototrexate could make it possible to deliver a cytotoxic effect by specifically irradiating a tumor, without affecting other parts of the body, which would receive only the inactive form of the drug.
Other fields would benefit from the ability to precisely control a drug’s effect. In the case of Alzheimer’s disease, for example, levels of acetylcholine—a neurotransmitter involved in memory—are reduced, which triggers memory problems. One solution is to inhibit the action of cholinesterases, the enzymes that break down acetylcholine (this is the mechanism of action of donepezil, for example, one of the few commercially available treatments for the disease), but precise control of their activity is difficult to achieve with conventional therapies. Recently, a team succeeded in developing a “photoswitch”-type molecule capable of inhibiting the action of butyrylcholinesterase when exposed to UV light. This molecule appears to be ten times more active in its irradiated form than in its original form in tests on the isolated enzyme, but it produces an “all-or-nothing” effect in mouse models exhibiting the characteristics of Alzheimer’s disease. Indeed, while injection of the non-irradiated product showed no noticeable improvement in memory in the affected mice, treatment with the irradiated form resulted in a complete return to normal behavior, opening up promising avenues for more finely tuning the chemical balance in the brains of patients with the disease.
These various studies have validated decisive proof-of-concept results regarding the efficacy of photopharmacological treatments in vitro and in vivo. Nevertheless, translating these results into clinical use presents a significant challenge. Indeed, the systematic use of the azobenzene moiety in current “photoswitches” is accompanied by certain drawbacks that could be deal-breakers in a therapeutic context, such as the use of UV light, which does not penetrate the body well and can be harmful.
Together with other teams of chemists, we are trying to develop alternative structures to overcome these challenges, such as hemiindigoids or the phenylazothiazoles, which can be activated and deactivated by visible light. Ideally, the target molecules must meet three criteria that are as crucial as they are difficult to achieve. They must be able to change shape efficiently in the water and/or in a biological environment; this change in form must occur upon exposure to visible light at a wavelength of as far as possible on the red side of the rainbow, in order to penetrate deep into the body and reach, for example, internal organs; and the irradiated form must result in a a sharp increase in biological activity compared to the initial form. This delicate balance between the various properties desired in an optimal photopharmacological drug is one of the main challenges in the field and requires true molecular craftsmanship.
