Low-Dose Naltrexone restored TRPM3 ion channel function in Natural Killer cells from long COVID patients, 2025, Martini et al

Low-Dose Naltrexone restored TRPM3 ion channel function in Natural Killer cells from long COVID patients

Etianne Martini Sasso, Natalie Eaton-Fitch, Peter Kenneth Smith, Muraki Katsuhiko, Sonya Marshall-Gradisnik

[Line breaks]

Abstract
Long COVID is a multisystemic condition that includes neurocognitive, immunological, gastrointestinal and cardiovascular manifestations, independent of the severity or duration of SARS-CoV-2 acute infection.

Dysfunctional Transient Receptor Potential Melastatin 3 (TRPM3) ion channels are associated with long COVID pathophysiology due to a reduced calcium (Ca 2+ ) influx, negatively impacting cellular processes in diverse systems.

Accumulating evidence suggests the potential therapeutic benefits of low-dose naltrexone (LDN) for people suffering from long COVID. Our study aimed to investigate treatment efficacy with LDN in restoring the TRPM3 ion channel function in natural killer (NK) cells from long COVID patients.

NK cells were isolated from nine people with long COVID, nine healthy controls and nine people with long COVID taking LDN (3 -4.5 mg/day). Electrophysiological experiments were used to assess TRPM3 ion channel functions modulated with pregnenolone sulfate and ononetin.

The findings from this current research are the first to demonstrate that long COVID patients treated with LDN have restored TRPM3 ion channel function and also validate previous findings of TRPM3 ion channel dysfunction in NK cells from people with long COVID not on treatment.

There was no significant difference in TRPM3 currents between long COVID patients taking LDN and HC in either PregS-induced current amplitude (p>0.9999) or resistance to ononetin (p>0.9999).

Overall, our findings support LDN as a potentially beneficial treatment for long COVID patients by restoring TRPM3 ion channel function and reestablishing adequate Ca 2+ influx for homeostatic cellular processes to occur.

Link (Frontiers in Molecular Biosciences) [Provisionally accepted, currently only abstract]

 

Has this been researched for ME patients with low NK cells? And if yes, was it helpful?

 

This has come from this group's research on ME/CFS, e.g., https://pmc.ncbi.nlm.nih.gov/articles/PMC8313851/

 

First time I have seen a p>0.9999.
And it is the one piece of data we do not need a p value for.
We want p values for differences involving the untreated groups.

 

Do researchers doing this kind of research i.e. with Low Dose Naltrexone, who get some beneficial results, ever try higher doses so that they are just researching Naltrexone? Or some dose in between the two? I'm just curious. It seems an obvious next step to try. Just because it has been used for those with addictions doesn't mean it can't be useful for other things.

Naltrexone is sold in 50mg tablets. Low Dose Naltrexone doesn't appear in the British National Formulary but I've read that the standard dose is between 0.5mg - 4.5 mg.

Perhaps doses between the 4.5mg and the 50mg should be tested.

 

Thread here:
https://www.s4me.info/threads/poten...atin-3-ion-channels-2021-cabanas-et-al.21440/

 

You can get 3mg tablets in Norway.

 

These seem like very high p values to be getting from 18 people.

I looked at the paper @InitialConditions posted and they did something similar there, also with 9 plus 9 participants, and they got p=.9020.

So they're comparing 49+56 observations since there are many from each person, but I don't see anything in the methods that says they dealt with non-independent observations.

Edit: Maybe that means they just averaged the observations from each participant first? In any case, the other paper doesn't have a "before" LDN difference either, so as Jonathan said, this doesn't really mean much.

 

Higher doses have some nasty side effects, which would be problematic both in terms of wanting to avoid harming trial participants and also possibly creating confusion in the data.

 

It is a problem I have never come across before. Google says:
In statistics, a p-value represents the probability of obtaining results as extreme as, or more extreme than, the observed results, assuming the null hypothesis is true.

A value of 0.9999 indicates that the two results from two populations are so extraordinarily similar that there is a 0.9999 chance of them being more different even assuming there is no real difference between populations. It is extraordinarily unlikely that the results should be so similar just by chance.

That seems a bit worrying. If you sampled a billion subjects in each population maybe you should expect something like this? But not 18.

 

Could it be an issue with their measuring, that there is some kind of upper limit that all healthy people are above?

 

Given the dosages used in the UK I was wondering about testing with doses of 5mg, 8mg, 10 mg, 15mg, etc might be worth doing. The huge leap in dose from the smallest doses straight up to 50mg means that benefits might be missed at the lower doses.

 

I rarely see ME/CFS or LC patients mention doses higher than 5mg in various SoMe groups. Many are at doses well below 3mg and spend months or years titrating up from e.g. starting doses between 0.1mg - 0.75mg.

 

Those are doses used wordwide, not just in the UK. 50mg of naltrexone has a completely different action to LDN. Normal doses of naltrexone does not produce endogenous opioids or lower inflammation.
The original low doses ranges are from researcher Ian Zagon -

https://www.ldnscience.org/resources/interviews/interview-ian-zagon

How did your research with Naltrexone proceed?

In 1980, we began an experiment in which we injected three different dosages of Naltrexone (0.1, 1, and 10 mg/kg) on a daily basis into mice that had been inoculated with cancer cells. We had no funding at this stage, so this was all done on our own time and at our own expense. We expected that the mice receiving the higher dose of Naltrexone would develop fewer tumors than those on the lower doses. After all, Naltrexone is an opioid antagonist that is far more potent and long-lasting than naloxone - and we already had great success with naloxone as an anticancer agent. However, much to our surprise, we discovered that the animals getting the 10 mg/kg each day were coming down with tumors far earlier than the controls. Yet there were no tumors in the group of mice receiving 0.1 mg/kg! One month after tumor injection all of the mice in the 10 mg/kg group and control group had tumors, but none of the mice in the 0.1 mg/kg group. By two and a half months, still only one-third of the mice in the 0.1 mg/kg group had a tumor. Mice receiving 10 mg/kg survived for a significantly shorter time than control animals. Among mice in the 0.1 mg/kg group that developed a tumor, these animals had a survival time of almost 50% longer than controls.

How did you begin to resolve these confusing observations?


I decided to investigate how long the Naltrexone acted on the opioid receptors. To do this, we use a small machine (like a hot plate) that measures how long it takes for an animal to detect heat. Exogenous opioids are known to suppress pain. The idea is to inject an exogenous opioid (morphine) which diminishes pain in the animals. If the morphine works, there is a delay before the animal experiences pain (and in some cases it does not feel pain at all). If an opioid antagonist like naloxone or Naltrexone is injected, then the animal feels pain because the morphine is displaced from the opioid receptors in the animal and becomes ineffective. Of course, we do not let the animals stay on the hotplate so they burn themselves, and we stop at 45 seconds (well before tissue damage occurs). We found out that at a dosage of 0.1 mg/kg, animals felt pain for the first 4-6 hours; in other words Naltrexone blocked the receptors for this period of time. After this time, we found that animals began to feel no pain when the morphine was injected. This means that after 4-6 hours, the Naltrexone was removed (metabolized), allowing the morphine once again to bind to the receptors. When we injected 10 mg/kg, the animals felt pain during the entire 24-hour duration of the experiment. This meant that at a dose of 10 mg/kg, Naltrexone was permanently bound to the receptors ('continuous opioid receptor blockade'), preventing morphine from binding at any stage. The 1 mg/kg had pain for around 12 hours and then no pain for the next 12 hours.

I know some of this is a little confusing, so let me just recapitulate. Animals with the highest dosage of Naltrexone (10 mg/kg) had increased tumor formation (tumorigenesis) and the drug lasted in its ability to block the opioid receptors for the entire day (continuous/full opioid receptor blockade). The animals with the lowest dosage (0.1 mg/kg) had decreased - or no - tumors and tumor growth, and the drug only lasted 4-6 hr/day (we called this intermittent opioid receptor blockade). The animals that received 1 mg/kg Naltrexone had cancer characteristics that resembled controls, and they had drug lasting around 12 hours per day.

How do you manage to explain these counter-intuitive findings?

Well, I was extremely perplexed! I gave real thought to these experiments, then one morning, shortly after the final data came in, it all dawned on me. We had observed that if Naltrexone blocked the opioid receptors for the entire 24 hours, endogenous opioids (those formed in the body, the 'endorphins') were blocked from the receptors and cancers were grew rapidly. If the blockade was for only 4-6 hours, the cancers either did not appear or were markedly delayed. This indicated to me that 1) endogenous opioids were really regulating the growth of these cancers and were antitumor agents, 2) endogenous opioids are tonically active - they work all the time and if you block them from receptors all the time then cancer growth is less controlled, 3) opioid antagonists can be used to modulate this process, 4) if you block opioid receptors part of the day, you get an exaggerated action of the opioids after the blockade, and 5) the intermediate dosage of drug produced animals that had tumor growth acceleration for 12 hours (when blockade was in effect) and tumor growth inhibition for the next 12 hours (when blockade was no longer in place) - giving you the net effect of tumor growth that looked like normal.

We also knew from previous scientific literature on the action of Naltrexone that opioid receptor blockade by such an opioid antagonist causes cells to have a compensatory increase in endogenous opioids and opioid receptors because they sense they are deprived of these elements. After the opioid antagonist (Naltrexone) is no longer present (having been metabolized by the liver), there is a period of super-sensitivity of the elevated levels of receptors to the increased levels of the endogenous opioids and one sees a more pronounced functional effect. If, the endogenous opioids were really inhibitory to growth (increase in cell number), then more opioids interacting with more opioid receptors would lead to a greater decrease in cell proliferation - hence, tumor cells would be depressed in replication. So, in the 4-6 hr period each day, we elicited what we call an upregulation in endogenous opioids and opioid receptors, and cell division was increased. But in the remaining 18-20 hr when the Naltrexone was no longer present, cell proliferation was greatly diminished by the high levels of opioids interfacing with the increased number of receptors. And, why did tumor growth in the 1 mg/kg Naltrexone group resemble control levels. Because for 12 hr each day there was an acceleration (Naltrexone blocked the "good" opioids from interacting with receptors and cell number increased), and for the next 12 hours when the Naltrexone was no longer present the increased opioids reacted with the increased opioid receptors and cell proliferation was reduced. Overall, therefore, this equaled out to a "normal" looking tumor growth.

How did you interpret these observations?

The interpretation of these experiments was that the duration of opioid receptor blockade by the opioid antagonist determined tumorigenic response, rather than drug dosage per se. To prove this point, we designed an experiment in which a dosage of drug (0.4 mg/kg) was given once a day to mice receiving transplanted neuroblastoma. This produced a reaction that we had seen earlier with 0.1 mg/kg - animals either not displaying tumors, or cancers that grew slowly. We then divided this dosage of 0.4 mg/kg into four individual dosages of 0.1 mg/kg (remember, this dosage proved very potent in arresting tumor expression in our first experiments). But, we now gave the 0.1 mg/kg not once a day, but every 6 hours; in essence, the animals were receiving 0.1 mg/kg Naltrexone for the entire 24-hour period. If it was drug dosage that was key, then we would predict that four times 0.1 mg/kg should yield the same results as 0.4 mg/kg given once daily. However, if it was duration rather than dosage that was important, then the 4 times/day of 0.1 mg/kg Naltrexone would accelerate tumorigenesis (you would be blocking endogenous opioids from opioid receptors continuously each day). The results showed that 4 times daily of 0.1 mg/kg accelerated tumor growth in the same way as a dosage of 10 mg/kg! This provided an important principle of opioid antagonist action how long the blockade of opioids from opioid receptors lasted was a determinant of the outcome.

Yet another test that distinguished an opioid-receptor mediated action from a non-opioid receptor effect of opioid antagonists on proliferating cells was to examine the stereospecificity of these compounds. A fast chemistry lesson...Nature has given our universe biological substances that exist as two mirror image forms of one another. Thus, 19 of the 20 naturally-occurring amino acids that make up proteins are "chiral", meaning that each can be arranged in two orientations around a carbon atom. The result is a mixture of "mirror-image compounds" called L- and D-amino acids. A great mystery of life has been the almost exclusive finding of the L-forms in proteins, the building blocks of life. Now, it turns out that the L-form of opioid antagonists like naloxone is three to four orders of magnitude more active than the D-form in its ability to antagonize the physiological actions of opioids. Therefore, opioid interactions at the receptor level are stereospecific, with isomeric forms showing markedly different affinities for opioid receptors. So, we designed an experiment in which tumor inoculated animals were given daily doses of either the L-form (-) or D-form (+) of naloxone. Remember, our very first studies with neuroblastoma taught us that naloxone was an antitumor agent. We found that tumorigenic events (the incidence of tumors, time to tumor appearance, and survival time) for mice receiving the L-form of naloxone indicated that this form acted like our earlier anticancer effects. However, animals given the D-form on a daily basis had characteristics of tumorigenesis comparable to control subjects that received injections of the vehicle without the drug. We went ahead and did the same experiments investigating the development of body and organ growth in rats using stereoisomers, and once again discovered that the L-form but not the D-form was the active ingredient. All of this pointed to an opioid receptor effect with respect to cancer or animal growth rather than a non-opioid receptor action.

Once this epiphany came to pass, we immediately asked if such a system only applied to cancer or to other phenomena as well, such as development. So we did similar experiments in rats. We found that baby rats receiving Naltrexone grow far more rapidly than those getting sterile saline if the dosage of Naltrexone blocks the opioid receptors for the entire day. Those developing animals who received a low dose of Naltrexone grew more slowly than controls. We repeated all the experiments described earlier (with the hotplate, for example), and found the same results we had observed with mice and cancer.

Likewise, the experiments that divided one dosage into four separate dosages given each day revealed again that duration of opioid receptor blockade was a major factor in determining growth effects. In 1981-1982 when we filed for patents on the use of opioid antagonists such as Naltrexone (and naloxone) as growth regulators (e.g., cancer, development) I had to provide the patent office with the dosages and regimens of Naltrexone and naloxone that humans could take. Using a weight basis of rats and mice, and knowing the pharmacokinetics/metabolism of Naltrexone in humans, we proposed that 3-10 mg/day of Naltrexone would yield a 'Low Dose Naltrexone' effect (growth retardation). A dosage of 30-50 mg/day or higher would block opioid receptors all day and enhance growth - this we knew from previous reports.

 

I have a bit of a trouble interpreting this value.
With a frequentist approach, the p-value measures the probability of obtaining the observed results assuming that the null hypothesis is true.

So if the null hypothesis is "there is no difference between LDN LC and HC", the probability of obtaining the observed results is >99.99%.
However, in that sense the statement "It is extraordinarily unlikely that the results should be so similar just by chance." doesn't make sense as well?

From my view as a particle physicist, the question arrises why one would actually test such a hypothesis.

 

The abstract says this:

Could that mean that the null hypothesis was that there is a difference? @Kronos

And welcome to the forum!

 

Thank you!

Well in that case, the p-value reported should be 1-p.
Can't really say more based on the abstract.

Maybe - hate to say it - this ties into the "casual" usage of statistics which is common in medical papers.
Arguably the general topic is very complicated if one digs deeper and wants to do it "right".
Though at the same time it doesn't take years of study to realize that one p<0.05 is to be expected if one runs 20 tests of hypotheses (example).

 

Here's how I understand it. Assuming the null hypothesis is true, you're equally likely to get any p value between 0 and 1.

https://davidlindelof.com/how-are-p-values-distributed-under-the-null/

It doesn't actually matter what the sample size is, p values are always uniformly distributed under the null hypothesis. So there's a 1 in 10,000 chance of getting a p value of 0.9999 or higher if there's no real difference between the groups. Such a high p value isn't an indicator that the two groups are similar. [Edit: On second thought, scratch that, it does mean they are very, very similar, but it's unlikely this is due to chance.] It's an indicator that the means of the two groups are extraordinarily close considering the high variance in the groups, and such a situtation should just happen due to chance 1 in 10,000 times. That or there could have been an error in the analysis like comparing one group to itself by accident.

 

Could it be an issue of faulty measurement devices? Maybe they all capped out in some way?

 

I'm not totally sure, but I think the only way that could influence higher p values is if the distribution of the data returned by the faulty device violated the assumptions of the statistical test they were using, for example a distribution where all the high values were squished into what would have been the middle of a bell curve being used in a t test that expects a normal distribution.

I tried simulating it by creating random normal distributions of 9 samples (group sizes in the study) with a mean of 0, but where any values greater than 0 were replaced with 0:


They all look very left skewed so might not be appropriate to use a t-test and might influence the p values. I ran t-tests with pairs of these random distributions 10,000 times. The p values still look fairly uniform where it would be very unlikely to get a value over .9999. If anything, there's a slight bias towards low values:


Edit: Maybe it's possible there's some other test assumption violation that does heavily bias p values towards 1.0, I'm not sure. But they would be expected to make sure the assumptions weren't violated before running a statistical test, even if the equipment was faulty.