Lean, mean machines

A virus can’t be said to be truly alive until it gets inside a cell.

By itself, a virus is less a life-form than a lean, mean machine. That’s because viruses travel lightly. All of them are obligate parasites, each typically consisting of not much more than a protein-shell-encased strand or two of genetic material, enough to serve as instructions for several or (with a few notable upside exceptions) several dozen bespoke proteins — all of which it will force the cell it invades to make for it.

The virus’s mission is to latch onto the machinery of the infected cell to gain admittance, get its string of genes copied and its proteins reproduced in bulk and then assembled into completed copies of itself, and watch its ungrateful progeny bust out of the cell that built them, the better to infect other cells and, with luck, other people.

The virus accomplishes this by hijacking the infected cell’s normally workaday enzymes — proteins that, like carpenters or electricians, perform skilled labor within and at the surfaces of cells — and diverting their skills to its own purposes.

The standard pharmaceutical strategy is to disable the virus by gumming up one of its mission-critical proteins. A handful of these drugs, called direct-acting antivirals, have proven successful — sometimes spectacularly so: Millions of people with hepatitis C and HIV/AIDS, for example, owe their lives to one or another (or a combination) of them.

But the microscale marauders have ways of wriggling out of that hit-’em-where-they-live trap. Their superpower is their ability to evolve at rates orders of magnitude faster than we do. They can generate many thousands of copies of themselves in each cell they successfully infect.

That, plus a slightly sloppy replication process, enable a viral lineage to, in a relatively short time, acquire mutations in one or more of its genes. A single random mutation can change a viral protein’s shape or chemical properties just enough to evade the grip of a drug that worked great until it didn’t: The targeted viral protein no longer gets glommed onto, and the virus is back in business.

The time it takes to push such a drug through approval may be long enough for it to be useless against the bug it was intended to squash.

Another problem: Most approved antiviral drugs target a protein, or a patch of it, that may be unique to a single viral species. Other species, and even other strains — not to mention other viral families — might lack that protein entirely or carry versions of it that the narrowly focused pharmaceutical fails to foil.

Not much of a profit incentive in that equation. Short shelf lives and narrow ranges of efficacy make drug development prohibitively costly.

A virus’s Achilles’ heel

But there’s another way to attack viruses — fight them on your home turf. Their utter dependence on the internal machinery of the cells they invade is their Achilles’ heel. That leaves them especially vulnerable to an alternative therapeutic game plan being explored in several Stanford Medicine labs including Einav’s, which she established in 2011.

What if, Einav asks, instead of trying to disable the virus directly, we try a more roundabout strategy? Why not temporarily dial down the efficiency or abundance of one of our own enzymes whose activity the virus needs for some aspect of its own “enter, make copies, exit, rinse, repeat” agenda?

Temporary disabling of one of our enzymes that we can live without for a week or two but that the virus can’t live without at all could deliver a knockout punch. (Einav calls this route the “host-targeted antiviral” approach.)

The Darwinian doctrine of survival of the fittest dictates that a virus will be rewarded with increased survival of its offspring for evolving a mutant one of whose own proteins is consequently less susceptible to a drug targeting that protein. But Darwin’s doctrine doesn’t do a darn thing to encourage any of our proteins to evolve in a way that helps a pathogen infect us. Evolution doesn’t work that way. Our proteins evolve (granted, at a relatively glacial pace) to help us, not them.

Bonus points: Lots of otherwise strikingly different viruses may depend on the same cellular enzyme for their successful proliferation, opening the door to broad-spectrum antiviral strategies.

Bunching up several uses for the same drug cuts down the time and cost of early-stage drug development, diminishes the odds of a safety belly flop in more advanced stages of development, and contributes to readiness for future outbreaks of newly emerging pathogens. Such a drug could be administered before a viral threat has been accurately diagnosed (with favorable implications for frontline health care providers and military personnel).

Objection: Won’t a compound that messes up one of your own proteins be toxic? Presumably that protein is there for a reason.

Well, true … but guess what? Virtually every drug we use to treat every disease we get that’s not infectious (all drugs, that is, except antimicrobials such as antibiotics and current antivirals) works by blocking, however briefly, the activity of one of our own proteins. Why should it be any different for infections?

Sure, there can be side effects. (Many direct-action antivirals are toxic, Einav pointed out.) The solution to that problem? Find the right dose. At least initially, Einav said, it might also be smart to go after acute infections requiring only several days of administration rather than chronic ones.

Recalcitrance to the development of viral escape mutants, and the possible shot at targeting multiple viral types, offer the prospect of bigger, longer-lasting markets — and therefore a potentially lower price — for the drug.

An added bonus: “There are a lot of approved drugs that target our own proteins,” Einav said. “We may be able to repurpose some of those existing drugs as antivirals.”

In a 2023 review article in The Journal of Clinical Investigation, Einav and colleagues noted that metformin — a cheap, long-approved diabetes drug with an excellent safety profile — has shown signs of potent antiviral activity against several viruses. Ditto for statins — taken by many millions to tweak their cholesterol profiles and reduce cardiovascular risk — in combination with angiotensin II blockers (ARBs), safe, popular blood-pressure-lowering drugs.

A new drug possibility

In 2025, Einav and colleagues including postdoctoral researchers Marwah Karim, PhD, and Manjari Mishra, PhD, published a study in Nature Communications showing that a small-molecule compound, named RMC-113, suppressed the replication of viruses spanning four viral families, without inducing any undue toxic effects in infected cells.

One was SARS-CoV-2, the cause of COVID-19 and a member of the coronavirus family from which a big chunk of common colds spring. Others were dengue virus, a flavivirus responsible for 14.4 million cases in 2024, mostly in the Western Hemisphere (although it’s endemic in more than 100 countries); the brain-inflaming Venezuelan equine encephalitis virus, an alphavirus; and Ebola and Marburg viruses, two lethal hemorrhagic-fever-inducing filoviruses that are largely confined to sub-Saharan Africa at the moment.

RMC-113 simultaneously blocks activities of a pair of enzymes SARS-CoV-2 needs, the study showed. One of these enzymes is required for SARS-CoV-2 to get inside a cell, while the other enzyme is required not only for the virus’s entry but also its replication, assembly and exit. Inhibition of this enzyme, the study discovered, suppresses viral replication via induction of a cellular degradation mechanism that SARS-CoV-2 suppresses. With both enzymes' actions blocked, the virus is a goner.

Presumably, RMC-113’s effectiveness against the other viruses whose replication it prevents is due to the same one-two punch.

Because RMC-113 is a “dual inhibitor,” even a hypothetical viral mutant that could get around the loss of one enslaved enzyme’s availability would be hard put to simultaneously develop a mutation compensating for a second on-the-blink enzyme required for carrying out the virus’s program. Recall, the virus has absolutely no control over our enzymes’ evolution. Developing resistance to such a compound is extremely unlikely, Einav said.

RMC-113 itself isn’t metabolically stable enough to use as a drug but could be tweaked by medicinal chemists to be sturdier.

Similar drugs might even work against bacteria that, rather than just swim around in our bloodstream or in wounds, infect and hang out inside cells. (Examples are mycobacterium, the cause of tuberculosis; chlamydia, the notorious sexually transmitted pathogen; and salmonella, the noted food-poisoning agent.)

Einav is walking down a promising yet not very beaten path in drug discovery.

“We’re seeing a growing interest from pharmaceutical companies and biotechs,” she said. We’re in touch with several companies that seem very motivated to advance our host-targeted approach.

“It takes time to achieve a change in mindset, to make people view viral infections as any other human disease and, hence, see the utility of treating them by targeting our own cellular functions,” Einav said. “Yet the real world keeps showing us the need for out-of-the-box perspectives.”