Skip to main content


Novobiocin Returns? But Not as an Antibiotic.

Here’s some interesting work from Dana-Farber (earlier BioRxiv version here) that may turn into a rare example of repurposing an old drug, if it works out. The team was studying an enzyme called DNA polymerase theta (also known by its corresponding gene name, POLQ), which is one of the lesser-known members of that functional family. Some of these seem to be involved in DNA replication under high-stress conditions, and polymerase theta seems to be required when there is a deficiency in homologous recombination (HR) mechanisms in a cell (involved in the repair of double-strand breaks). Loss of both of those pathways leads to “synthetic lethality”, that is, a lethal effect from a combination of gene or protein knockdowns when neither of the individual targets are lethal by themselves.

There are tumor types that have that HR deficiency already operating, so there’s been a lot of interest in completing the synthetic lethality combo for those. One big push has been into PARP (poly ADP-ribose polymerase) inhibitors, several of which have reached the clinic and the market, but (as with most any oncology mechanism), PARP inhibitor resistance also develops. So you’d want another way to attack this mechanism, and polymerase theta has been the subject of a lot of interest.

In this case, the team found by screening that a longtime antibiotic molecule, novobiocin, turns out to be a good polymerase theta inhibitor. It’s long been known as a DNA gyrase inhibitor in bacteria, although not a clinically useful one, where it works by hitting an ATP-binding site. I spent some time messing around with it and some derivatives in my own antibiotic days. More recently, it’s been shown to hit other such sites, such as on human HSP90. And that’s what seems to be going on here: it seems to be working deep into an ATP pocket on the polymerase. There’s an impressive amount of cellular and in vivo validation in the paper, showing that novobiocin does indeed have significant effects on HR-deficient tumor lines (but not those that still have that pathway intact), that it potentiates the effects of PARP inhibitors in such tumors, and that indeed it can overcome PARP resistance once that has set in, at least through several of its known mechanisms. This may or may not bring novobiocin itself back as a drug – although it’s definitely something to look into – but it really shores up the whole idea of polymerase-theta inhibitors as drug candidates in general.

Novobiocin itself was withdrawn from the US market back in the 1980s, due to an unfavorable efficacy/safety profile. But that was in the context of other effective antibiotics – oncology is another matter. You would imagine that you’d want to treat patients with HR-deficient tumors with a dual regimen of PARP and polymerase-theta inhibitors, in order to give the cells fewer ways to escape. This paper demonstrates that the two pathways do seem to be independent, making it an obvious cocktail combination, and polymerase theta expression (the authors show) may well be a prospective marker for selecting the patients you’d want to treat.

Interestingly, novobiocin was actually the subject of an oncology trial in the early 1990s, since it had been shown to potentiate the effects of DNA-damaging agents in animal studies. Some of the patients in this Phase I showed responses, although the patients were in no way selected for HR-deficient tumor types, so a trial in those patients would seem to be a straightforward thing to mount, and I hope we see one underway soon.

There are other polymerase-theta inhibitors being investigated – here’s a paper that just came out with an allosteric inhibitor – but I don’t see how any of those will reach the clinic more quickly than novobiocin. And there’s a lot of clinical work to be done. Working out the combinations with various PARP inhibitors (with and without other DNA-targeting therapies) and figuring out patient selection will all need work, and novobiocin might be the tool to do it.

17 comments on “Novobiocin Returns? But Not as an Antibiotic.”

  1. Orange says:

    How would it be commercialized? In combination with a patented drug to increase revenue of that or are there other, less obvious ways to expect a return of investment?

    1. David Kochman says:

      Method of use patent

    2. David says:

      Method of use patent

      1. David says:

        It’s called a ‘Second use patent’, in cases when the compound already had a first use. In some cases, new combinations of known compounds can also be patented.

  2. anon the II says:

    Novobiocin is a fancy coumarin. My feeling is that these suckers tend to stick to lots of stuff so I would suggests that you tread carefully in interpreting any results you get.

    For me, I’d walk away.

    1. Grant Seiler says:

      Tell that to warfarin.

  3. Sjur Holm Hartvedt says:

    Is Novobiocin currently in production anywhere? Is it complicated to start producing again? A combo of Olaparib and NVB for those having biallelic Brac2 in breast, ovarian, prostate and pancreatic CA could prove effective, I understand POL Theta inhibits NHEJ ?

    1. Derek Lowe says:

      It’s commercially available, but I don’t know if there’s anything that’s approved for human dosing. I would think not. . .

      1. Sjur Holm Hartvedt says:

        Thank you!
        I also understand that this could be a nice combo for those who can utilize luPSMA in PCA (Vision trial), NVB radiosenzitise the cells by inhibiting MMEJ, for those who are biallelic BRAC2 and block homologuous recombination – it’s perfect.

  4. metaphysician says:

    Silly question, but does anybody ever revisit old, rejected, or discontinued antibiotics for reconsideration against bacterial resistance? An insufficient safety/efficacy profile in 1980, might not look so terrible in 2020 as a last line antibiotic against multiresistant diseases.

    1. WetDogq says:

      Chloramphenicol is an example of this. Has this nasty side effect of aplastic anemia which is rare but has a high mortality rate. But it’s very effective and cheap and there are few resistant variants known.

      In the US, use is typically restricted to Infectious Disease specialists who can best determine if its benefits outweigh the costs.

      And there have been proposals to stop production of some older antibiotics to let them ‘cool off’ in terms of resistance but this approach has been criticized as unworkable for various reasons.

      1. Jim Van Zandt says:

        Chloramphenicol was mentioned in the science fiction series “1632” as something with risks, but easy to synthesize and much better than nothing.

    2. Adam Shapiro says:

      That is what happened with colistin. It was rarely used because of safety issues, until multidrug-resistant bacterial infections brought it back into frequent use.

  5. Michael M says:


    What portion of novobiocin were you making derivatives of? I assume you were making GyrB inhibitors?

  6. gippgig says:

    I wonder how effective coumermycin would be against POLQ.

  7. Tore says:

    Relatedly, DNA polymerase theta was reported to work as a reverse transcriptase in a recent paper (10.1126/sciadv.abf1771).

  8. Lizzy says:

    Parp inhibitors have revolutionised treatment of ovarian cancer over the last 6 years. It’s been truly amazing. What’s been happening more recently are attempts to integrate Parp inhibitors with chemotheraputic agents to achieve synergy. There are many ways this can be done, with platinum based drugs, with alkylating agents, with immunotherapies, and with anti-angiogenesis agents. Overall, most of these combinations work, the problem has been unacceptable toxicities.
    This is a recent excellent review article on the subject: “Choosing wisely: Selecting PARP inhibitor combinations to promote anti-tumor immune responses beyond BRCA mutations” Jennifer Taylor Veneris, Gynecologic Oncology 2020.
    She talks about “Inducing BRCAness” (making a tumor change from HR proficient to deficient) as well as “Increasing stress response” (includes heat shock proteins) as 2 of 4 approaches.

    Sorry, don’t know how to make this article a simple click. But it is free on Sciencedirect.

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.