Published on 15 July 2019
Des virus tueurs de cancers

Armed with oncolytic viruses, researchers of Université Paris-Saclay are attacking tumors. The vesicular stomatitis virus and adenoviruses figure among the subjects of their investigations.

Using viruses to destroy tumors is a promising therapeutic approach developed by researchers to supplement conventional cancer therapies. Virotherapy in cancer treatment capitalizes on the fact that certain viruses are able to specifically infect tumor cells and induce their cell death by means of direct lysis and/or the response of the immune system, without affecting healthy cells. This preferential replication, called “oncotropism”, is naturally present in viruses or acquired after genetic modification.

“As a general rule, a virus does not infect all types of cells, because its spectrum of hosts is relatively limited. In order to penetrate a cell, specific molecules present on its surface must recognize a specific receptor on the cell surface” explains Yves Gaudin, a researcher at the Institute of Integrative Cell Biology (I2BC – CEA/CNRS/Université Paris-Sud). “Often prone to deregulation of their cellular mechanisms, tumor cells express a number of unusual membrane proteins, some of which, if we knew how to target them, could play the role of viral receptor.”

The VSV, a multi-talented oncolytic virus

An excellent laboratory model, the vesicular stomatitis virus (VSV) is a naturally oncolytic virus that is the subject of many studies and widely used in gene therapy. Benign for humans, this “enveloped” virus infects many species of mammal and certain insects. “The problem is that this virus lacks targeting specificity and rapidly causes the death of the infected cell,” notes Yves Gaudin. The glycoprotein G present in its envelope ensures recognition of the targeted cell receptors, then the release – thanks to the fusion of the viral envelope and the cell membrane – of the viral genome inside the cell, both of which are stages required for infection to continue. These receptors belong to the same family as those of low density lipoproteins (LDL), responsible for carrying lipids in the blood and present at the surface of many cells. All feature a structural cysteine-rich repeat motif encaging a calcium ion.

Thanks to a collaboration with researchers at Synchrotron SOLEIL, France’s national synchrotron facility, Yves Gaudin and Aurélie Albertini recently identified, using X-ray crystallography, the precise points at which the VSV glycoprotein anchors to the LDL receptors as well as the regions of glycoprotein G involved. Located in two cysteine-rich domains of the receptor, these binding sites involve two amino acids in the glycoprotein G. “Their mutation leads to a total failure to recognize the LDL receptor. The virus cannot bind to the cell and loses its infectious capability. On the other hand, the membrane fusion activity, essential for cargo release, is maintained,” remarks Yves Gaudin.

Being able to decouple receptor recognition and viral fusion opens up numerous perspectives. “What we’re seeing is the construction of new glycoproteins that no longer recognize their natural receptor, but target other receptors at the surface of cells and conserve their membrane fusion properties,” points out Yves Gaudin, who is testing the association of glycoproteins with nanobodies, i.e. small single-chain antibodies derived from llamas. “The idea is to put these antibodies into our viral glycoprotein to confer a specificity that it did not originally have and direct the virus to new targets.”

Genetic manipulations and the oncotropism of adenoviruses

At the Vectorology and Anticancer Therapies Laboratory (CNRS/Université Paris-Sud) at Gustave Roussy, Karim Benihoud and his team are working for the most part on another type of virus, adenovirus serotype 5, and, more recently, on that of serotype 3. These replicating viruses lyse cells at the end of their cycle. His focus is on controlling their toxicity and tropism when these viruses are administered by a systemic route. “Normally, an adenovirus preferentially targets liver cells. Once injected into the bloodstream, it becomes covered with coagulation factors that function like adaptors between itself and the receptor on the surface of the hepatocytes. By genetically modifying the capsid proteins, it is possible to eliminate the pathway by which the virus naturally enters the cell or, on the contrary, create a new one after adding a ligand for addressing purposes.”

As adenoviruses are not naturally oncolytic, researchers have two strategies that they can use to render these viruses tumor-selective. “The first way is to put the entire viral genome under the control of a tumor-specific promoter, such as that of PSA (prostate-specific antigen), CEA (carcinoembryonic antigen) or telomerase, known to function in tumoral cells,” explains Karim Benihoud. “The other involves mutating certain regions of the viral genome so that, once the virus has entered the cell, it only replicates in cycling cells, i.e. tumoral cells for the most part.” In the laboratory, he works with adenoviruses with mutated E1A proteins, responsible for initiating the viral cycle. In normal cells, the mutated protein E1AΔ24 is not capable of trapping the Rb protein, a tumor suppressor, or of removing cell cycle inhibition. On the contrary, the adenovirus will be able to replicate its genome in the tumoral cells, often mutated for Rb, and the new virions produced will lead to the lysis of the cells.

The other benefit of oncolytic viruses resides in the activation of the antitumor immune response. The tumor environment, generally immunosuppresive, changes in contact with them. The injection of a virus into a tumor creates a pro-inflammatory context: infection of the cells induces the production of cytokines (IL-1, IL-6, interferons and TNF-α) that will activate the immune system. “Once adenoviruses enter tumor cells, they begin to replicate and kill cells, releasing tumor antigens that are processed and presented by antigen presenting cells.”

Ready to wage war on tumors from every angle, researchers are investigating the use of oncolytic viruses in association with chemotherapies, with encouraging results. “As far as colon carcinomas are concerned, we have shown the synergistic effect of combining oncolytic adenovirus and valproic acid in a murine model,” said Karim Benihoud.



∙ Jovan Nikolic et al., Structural basis for the recognition of LDL-receptor family members by VSV glycoprotein. Nature Communications. 2018 ; 9 : 1029

∙ Bressy C. et al., Combined therapy of colon carcinomas with an oncolytic adenovirus and valproic acid. Oncotarget, 2017, 8 (57) : 97344 – 97360.


Portrait : Karim Benihoud

“The genome of the adenovirus can be easily manipulated using the tools of molecular biology and the vectors derived from it can be produced in large quantity.ˮ

Karim BenihoudKarim Benihoud is a professor of cell signaling and immunology at Université Paris-Sud and heads the anti-tumor viral vectorology group of the Vectorology and Anticancer Therapies Laboratory (CNRS/Université Paris-Sud) at the Gustave Roussy cancer campus. His work focuses on vectors and oncolytic viruses derived from human adenoviruses, seeking to control their tropism and develop new vaccination strategies. He has applied for, and been granted, several patents.


By Véronique Meder.

The original version of this article was published in L'Edition #10.