The buspirone challenge test clearly distinguishes ME/CFS patients from healthy controls: why is it not being developed and deployed?

[hotblack]

This is all really interesting. Lots of great ideas and thanks for digging into the papers @forestglip

I haven’t gone into depth and need to spend more time reading but agree it’s definitely something that needs following up on. Some thoughts/questions while reading, some of which it looks like others are already asking and digging around in.

How does this fit into the eMSN and wider genetic (DecodeME etc) findings?

What other tests or studies can help isolate what’s happening and where? (Narrowing down to 5HT or D2 receptors for instance or particular cells or networks)

The stuff around TIDA and the tuberoinfundibular pathway is intruiging!

I’m getting the impression this would indicate less of a problem with a specific neurotransmitter level or a hypersensitivity of a particular receptor and more a network problem or instability?

Could this help explain why we’ve had difficulties with other theories or trials? Could other factors which affect this network further confuse things? If people in a study are male or female, at different points in life or hormonal cycles, different levels of sleep amount or quality, different other triggers like sexual activity or stress or physical activity etc.. there’s so many potentially confounding factors…

I think I flagged this before but it seems potentially significant that the prolactin receptor is a cytokine receptor and shares and uses familiar immune pathways like JAK-STAT (okay I know lots of things do, but the potential for some crossover seems worth exploring)?

 

[ScoutB]

How does this fit into the eMSN and wider genetic (DecodeME etc) findings?

I'm sure I'm not alone in having pondered this, but don't have a great answer yet. The TIDA neurons controlling prolactin release (via dopamine) live in the hypothalamus, which the GTEx analysis from DecodeME said may be relevant. (Tho I think we're taking the algorithms that turn the GWAS results into brain areas with a small grain of salt atm since more brain studies may be needed).

The hypothalamus seems potentially interesting in general. The CRH autopsy study is looking at a different part of it. Certain places in the hypothalamus lack a standard blood brain barrier (because they want to sample immune signals and such in the blood), maybe making those areas more vulnerable to insult than other parts of the brain. I've been collecting info but probably should wait until I'm less foggy to try and write it up..

What other tests or studies can help isolate what’s happening and where? (Narrowing down to 5HT or D2 receptors for instance or particular cells or networks)

forestglip suggested using different medications (beyond buspirone) with more receptor selectivity and that seems like a good idea to me. Also more PET scans probing dopamine. There are all these recent studies using optogenetics to figure out exactly which brain circuits in mice are involved in which experiences and responses. So maybe if we read some of those that touched on this area of the brain we might get some more ideas of what could be going wrong specifically/what to test.

I’m getting the impression this would indicate less of a problem with a specific neurotransmitter level or a hypersensitivity of a particular receptor and more a network problem or instability?

Yea, that sort of thing would be my guess. Mainly I am trying to keep my mind open to how complicated the underlying system is.

 

[forestglip]

How does this fit into the eMSN and wider genetic (DecodeME etc) findings?

Medium spiny neurons have dopamine receptors, and prolactin release is also controlled by dopamine receptors, so if the MSN idea turns out to be right, that might be a connection.

 

[forestglip]

I haven't read this in detail, but I thought it was worth linking on the thread. This is Abdel Bakheit's thesis that includes the same study as the 1992 postviral fatigue syndrome prolactin challenge paper.

Hypothalamic Dysfunction and Neutrotransmitter Abnormalities in the Postviral Fatigue Syndrome (1992)

The final paper only included tables, but the thesis shows the prolactin response to buspirone as plots. Females and males, respectively:

View attachment 33356 View attachment 33357

The thesis also includes details about another study about vasopressin in postviral fatigue syndrome, which was published as: Abnormal arginine-vasopressin secretion and water metabolism in patients with postviral fatigue syndrome 1993, Bakheit, Behan et al

I made a separate thread for the Bakheit 1992 thesis: Hypothalamic Dysfunction and Neurotransmitter Abnormalities in the Postviral Fatigue Syndrome, 1992, Bakheit

I'll copy over the individual level prolactin response plot I made, since I think it's intriguing how clear the difference is between groups in this study:

And also interesting how the difference is not nearly as clear in the 1996 Majeed thesis:

Neuroendocrine Alterations in Chronic Fatigue Syndrome (Majeed, 1996, Thesis)

There was a significant difference between groups for prolactin response. Here is delta prolactin, which is the difference between maximum prolactin and baseline prolactin for each individual:

Edit: I think the difference is not as clear in Majeed partly due to analyzing males and females together, since even in healthy people, females have a larger response than males.

 

[hotblack]

I'm sure I'm not alone in having pondered this, but don't have a great answer yet. The TIDA neurons controlling prolactin release (via dopamine) live in the hypothalamus, which the GTEx analysis from DecodeME said may be relevant. (Tho I think we're taking the algorithms that turn the GWAS results into brain areas with a small grain of salt atm since more brain studies may be needed).

Yea, that sort of thing would be my guess. Mainly I am trying to keep my mind open to how complicated the underlying system is.

Medium spiny neurons have dopamine receptors, and prolactin release is also controlled by dopamine receptors, so if the MSN idea turns out to be right, that might be a connection.

Great, we seem to be thinking along similar lines.

My current chain of thought… (speculation and analogy warning!)

So rather than a smoothed electrical system we have a noisy and spiky one. Not good on the national grid or in our brain. This is only in a specific location, in specific cells, but a key one and has spillover into other areas.

I was thinking of the analogy the national grid, or any electrical system really. In a healthy system there are dampeners (capacitors and management circuits) to absorb spikes in demand or supply and keeping voltage output smooth.

If we think of eMSNs in the striatum and these TIDA neurons as playing a similar role, smoothing signals, and they are malfunctioning for some reason…. then the neural circuits managing this interface become noisy and we get spiky responses.

When there is high demand (physical or mental exertion, immune or hormonal changes, certain emotions, or challenges like the one in this paper) the response is erratic and spills over in to the surrounding networks triggering downstream autonomic/endocrine problems.

It sort of fits into some other ideas but presents them a bit differently and may account for some of the problems there’s been with consistency of results too. Given all the variables and that this isn’t a straight hypersensitivity or deficit.

 

[hotblack]

I can also see how his may be a bit close to some of the dysregulation of the hpa-axis stuff we hear from some. The irony may be that they could have been close but spent decades not looking for biological evidence or real treatments. They instead decided people could think their way out of it or that neuroplasticity was the answer when it looks like there may be a biological problem and the sort of ability to adapt they rely on may even be impaired at a cellular level.

Going back to the power grid analogy it’s like thinking that when there are power supply problems with the grid, to get it working again we should tell people across the country how to manage their usage and when they turn their kettles. That we draw up big grid management plans on individual usage and somehow manage our way out of it. Rather than just isolate the substation playing up and send someone to fix a few knackered capacitors.

 

[hotblack]

Slightly self indulgent tangent but it was digging around and reading these papers (my last post probably sums things up) that previously got me interested in some of these networks and how they’re simulated and could go awry. May be of interest to people.

And I wonder if as well as pharmacological tests to isolate responses and genetic/co-expression tests to see what cells may be implicated, there is a possibility for someone to do some similar neurological network simulation to see effects of the changes we think may be happening?

 

[SugarSquared]

Slightly self indulgent tangent but it was digging around and reading these papers (my last post probably sums things up) that previously got me interested in some of these networks and how they’re simulated and could go awry. May be of interest to people.

And I wonder if as well as pharmacological tests to isolate responses and genetic/co-expression tests to see what cells may be implicated, there is a possibility for someone to do some similar neurological network simulation to see effects of the changes we think may be happening?

Would you be interested in joining the Google doc group chat ans becoming an editor on there? You could add on to the mechanisms section and future study brainstorming. There’s no pressure to contribute more than you are able to. Just wanted to extend the invitation in case you were interested.

 

[hotblack]

Would you be interested in joining the Google doc group chat ans becoming an editor on there? You could add on to the mechanisms section and future study brainstorming. There’s no pressure to contribute more than you are able to. Just wanted to extend the invitation in case you were interested.

Thanks @SugarSquared I hadn’t read that bit of the thread yet but collating info and ideas sounds like a really good idea. I’m not up to adding more editing to my plate atm but please feel free to take any of my ramblings which are useful.

 

[SugarSquared]

Thanks @SugarSquared I hadn’t read that bit of the thread yet but collating info and ideas sounds like a really good idea. I’m not up to adding more editing to my plate atm but please feel free to take any of my ramblings which are useful.

Sounds good! You’re always welcome to join if you have more time later.

 

[forestglip]

I don't know what the content of this letter is, but this is a comment on Yatham et al. 1995:

Re: Endocrine responses to fenfluramine challenge in chronic fatigue syndrome (Cleare; O'Keane, 1996)

I found where to read this letter to the editor. It starts at the end of page 129 here: https://journals.sagepub.com/doi/epdf/10.1177/070674379604100215

This is a response to Yatham et al's 1995 study of 11 ME/CFS patients (8 female, 3 male) and 11 age and sex matched controls, where administration of dl-fenfluramine (60 mg) did not lead to significantly different prolactin or cortisol responses between groups.

Cleare and O'Keane argue that since nearly all of the patients in the study had current or past psychiatric diagnoses, this could mask a difference, as studies have found the opposite of ME/CFS findings in depression: blunted prolactin response. They also say that the dl- formulation of fenfluramine is less specific for serotonin release than d-fenfluramine, and additionally affects dopamine and norepinephrine pathways, which makes it less of a pure test of serotonin function. Their own study of d-fenfluramine found a significant difference.

Yatham et al. respond to say that only two of eleven patients in the study had active depression, and their prolactin response was similar to the rest of the patients. They also say that dl-fenfluramine does not appear to affect dopamine at the dosage used in the study.

 

[hotblack]

Thinking about some of the analysis, normal range and intraday variance in prolactin (and maybe other hormone) levels due to sleep/wake, stress, other hormonal events seems like it could be a big confounding factor. I wonder if looking at individual rate of change and comparing those would give a different or clearer picture? I haven’t looked through all the studies but did any do this, or were they all comparing absolute change? Would it be worth taking this approach in future studies? Can we do so retrospectively on any? I guess we’d need studies with multiple measurements to get s full picture of the profile of change in levels.

 

[bobbler]

Thinking about some of the analysis, normal range and intraday variance in prolactin (and maybe other hormone) levels due to sleep/wake, stress, other hormonal events seems like it could be a big confounding factor. I wonder if looking at individual rate of change and comparing those would give a different or clearer picture? I haven’t looked through all the studies but did any do this, or were they all comparing absolute change? Would it be worth taking this approach in future studies? Can we do so retrospectively on any? I guess we’d need studies with multiple measurements to get s full picture of the profile of change in levels.

This is the question …I also stumbled on a little while back (then beat myself up - about it as maybe people think it’s just me ‘making things more complicated’ etc)

I’ve felt really poorly over heat wave onwards so getting brain to heady heights of thinking about this stuff even for moments at the detail and communicate it succinctly level has been hard tho. But the question still lingers in the simmer it together/ chew on it picture-wise in my brain type level whilst I’m stuck not doing much.

I think with the medication stuff it’s interesting thinking it as theoretically a x+y=10 but 2x+z=50 type thing with the different doses (seems to matter for buspirone as to what receptors it affects and what it’s used for) and comparator drugs in same me/cfs cohort (which gets complicated when it’s old studies with old academics who had certain pet. Hypotheses potentially as to why they were doing these back then). And the comparing the same to those with different illnesses.

But now I’ve realised prolactin spike is related to a lot of things that are so fundamental ‘symptoms patterns of me/cfs’ (and symptom feels wrong to explain them because it’s the getting the body doing the wrong response after eg an exhausting day by then not sleeping despite being exhausted type thing) it both gets really interesting but a little bit more complicated to control for those effects (abd the question is do we need to or do we need to just note them or study them etc - abd that’s the bit I’m struggling to get my brain round the size of that conundrum on)

That so many things cause spike too makes measuring things like basal or ‘level it’s at for today’ (if it can be risen for longer due to PEM causing lack of sleep for x days) interesting in an individual.

There is a post I put on @hutans thread on cortisol where lightman talked about new measuring more constantly etc for that that’s springing to mind - thinking eg of a study over a week like an insulin monitor (to see if it spikes all over the place in comparison to eg controls throughout a week due to those other things like body temp or sleep or exertion making the concept of ‘basal’ different to what you might get as a baseline on a given day from a me/cfs cohort).

 

[forestglip]

I wonder if looking at individual rate of change and comparing those would give a different or clearer picture? I haven’t looked through all the studies but did any do this, or were they all comparing absolute change?

Do you mean like looking at percentage increase in prolactin compared to baseline? The Bakheit 1992 and Richardson 1995 studies look like they used that measure. (So if someone had a baseline of 100 and increased to 200, that'd give the same response value as someone with a baseline of 300 that increased to 600.)

 

[hotblack]

Do you mean like looking at percentage increase in prolactin compared to baseline? The Bakheit 1992 and Richardson 1995 studies look like they used that measure. (So if someone had a baseline of 100 and increased to 200, that'd give the same response value as someone with a baseline of 300 that increased to 600.)

Yes exactly. Thanks for the info!

 

[forestglip]

Decreased tryptophan availability but normal post-synaptic 5-HT2c receptor sensitivity in chronic fatigue syndrome (Vassallo et al. 2001)

Just one note about this study of ME/CFS that did not find a signifincantly increased prolactin response to the serotonin agonist meta-Chlorophenylpiperazine (mCPP) is that the dose they used (0.25 mg/kg) might have been too small.

The discussion of a paper about prolactin response in migraine (Cassidy et al. 2003) points out that one migraine study used 0.25 mg/kg and did not see a difference in prolactin response, while another study used 0.5 mg/kg and did see a difference:

Gordon et al. (25) found no difference in hormonal (cortisol and prolactin) responses to m-chlorophenylpiperazine (mCPP- 0.25 mg/kg) a 5-HT agonist drug, in eight subjects with migraine (with and without aura) and 10 normal controls. [...] The dose of mCPP may have been inappropri-
ately low as suggested by the authors [...]

Leone et al. (26) overcame some of these methodological problems. They found an increased maximal serum prolactin response to a higher dose of mCPP (0.5 mg/kg) in a group of 12 subjects with migraine without aura compared to 14 healthy controls.

Spoiler: References

25 Gordon ML, Lipton RB, Brown SL, Nakraseine C, Russell M, Pollack SZ et al. Headache and cortisol responses to m- chlorophenylpiperazine are highly correlated. Cephalalgia 1993; 13:400–5. DOI

26 Leone M, Attanasio A, Croci D, Ferraris A, D’Amico D, Grazzi L et al. 5-HT1A receptor hypersensitivity in migraine is suggested by the m-chlorophenylpiperazine test. Neuroreport 1998; 9:2605–8 Article

 

[forestglip]

I also found another interesting paper that used the larger 0.5 mg/kg dose of mCPP, where prolactin response was higher in males with gambling addiction than in healthy males:

Serotonin Dysfunction in Pathological Gamblers: Increased Prolactin Response to Oral m-CPP Versus Placebo (Pallanti et al. 2014)

While there was no significant difference at baseline between the groups, pathological gamblers' m-CPP prolactin response was significantly increased (14.77+4.56) relative to controls (10.56±4.01) at 180 minutes (F=9.14, df=1.50, P<.01) and at 210 minutes (15.14±5.15 vs 10.25±4.87) (F=12.26, df=1.50, P<.01) post-administration (Table 3).

The bottom two lines are placebo. The top two lines show the prolactin levels after the drug in those with gambling addition and healthy controls.

Plasma m-CPP levels did not differ between groups:

Although mean peak plasma m-CPP levels did not significantly differ between pathological gamblers (30.5±16.7) and controls (28.54±15.7) (F=2.03, df=1.5'\, P=0.87), great variability was observed in the scores from each group.

There was a correlation between severity of gambling addiction and prolactin response:

Current clinical severity, as assessed by the Y-BOCS-PG total score in the pathological gambling group, correlated positively with m-CPP peak change prolactin levels (r=.481, P<.02),

The gambling group, but not healthy controls, reported a "high" after m-CPP administration:

To evaluate changes in subjective conditions, the Affect Analogue Scale (AAS) was administered. [...] The AAS-High subscale was the primary behavioral measure for this study

Compared with controls, PG patients showed a mean peak "high" response greater after m-CPP (F=11.89, cff=1.50, P<.05), not significant after placebo (F=0.34, d/^1.50, NS).

They suggest future studies should measure beyond 3 hours:

A longer (>3 hours) test time should be utilized in future studies to determine if prolactin response further increases, rendering even more robust findings.

---------

mCPP sounds like an interesting drug. From Wikipedia:

Despite its advertisement as a recreational substance, mCPP is actually generally considered to be an unpleasant experience and is not desired by drug users. It lacks any reinforcing effects, but has "psychostimulant, anxiety-provoking, and hallucinogenic effects." It is also known to produce dysphoric, depressive, and anxiogenic effects in rodents and humans, and can induce panic attacks in individuals susceptible to them. It also worsens obsessive–compulsive symptoms in people with the disorder.

 

[forestglip]

Basically every study I've read about altered prolactin response to a serotonergic drug concludes that the findings suggest altered serotonin function, but they don't bring up dopamine as potentially involved as another step in the pathway to prolactin response. For example, from the above gambling study:

These results provide additional evidence for 5-HT disturbance in pathological gamblers

I'd like to know if I'm missing something, because they all seem to treat it as if it's obvious. I don't know if they know something I don't or if they just see that a "serotonin drug" increases prolactin too much, and thus think that the serotonin system must be impaired.

Here is a paper suggesting that serotonin does not act on the pituitary gland directly, and instead probably acts through the hypothalamus to affect prolactin:

The Interaction of the Serotonergic and Dopaminergic Systems on Prolactin Secretion in the Rat: The Mechanism of Action of the “Specific” Serotonin Receptor Antagonist, Methysergide (Lamberts et al. 1978)

In this study, the mechanisms governing the secretion of PRL by the serotonergic system and the inhibitory effects of methysergide were analyzed. The injection of 5-hydroxytryptophan (5-HTP) into rats greatly increased the serum PRL concentration. In vitro, the pituitary gland from these rats synthesized and secreted amounts of PRL similar to those produced by nontreated rats. [...] Neither 5-HTP nor serotonin had a direct in vitro effect on PRL secretion. It is suggested that serotonin precursors exert their stimulatory effect on PRL secretion via the hypothalamus and not at the pituitary gland.

The mechanisms through which serotonin and methysergide act to modify PRL secretion are very complex and may act by influencing the dopaminergic system.

So just from what little I've read about the mechanism, it seems possible that mCPP and other serotonergic agents lead to an abnormal prolactin response like this:

mCPP activates serotonin receptors in the hypothalamus. It is a normal amount of activation in both groups. This leads to the hypothalamus decreasing its output of dopamine. Again, a normal amount of decreased dopamine. But then the same change in dopamine between the groups leads to different prolactin responses because of alterations at the level of the pituitary dopamine receptors or beyond.​

​

 

[bobbler]

I also found another interesting paper that used the larger 0.5 mg/kg dose of mCPP, where prolactin response was higher in males with gambling addiction than in healthy males:

Serotonin Dysfunction in Pathological Gamblers: Increased Prolactin Response to Oral m-CPP Versus Placebo (Pallanti et al. 2014)

View attachment 33537
The bottom two lines are placebo. The top two lines show the prolactin levels after the drug in those with gambling addition and healthy controls.

Plasma m-CPP levels did not differ between groups:


There was a correlation between severity of gambling addiction and prolactin response:


The gambling group, but not healthy controls, reported a "high" after m-CPP administration:


View attachment 33538

They suggest future studies should measure beyond 3 hours:


---------

mCPP sounds like an interesting drug. From Wikipedia:

When its gamblers one thinks dopamine so are they sure m-cpp isn’t actually ‘dirty’ in some way that means it’s triggering those receptors and not serotonin?

 

[forestglip]

When its gamblers one thinks dopamine so are they sure m-cpp isn’t actually ‘dirty’ in some way that means it’s triggering those receptors and not serotonin?

Yes, the authors mention dopamine's involvement in addiction too:

gambling behavior can be induced in individuals with Parkinson's disease during dopamine agonist treatment,13 consistent with the involvement of dopamine in the mechanisms of addiction.

I'm not sure if mCPP might also affect dopamine receptors. But my point in the last post is that even if all it affects is serotonin receptors, dopamine could still be involved as part of the pathway between serotonin receptors and prolactin release.

I'm having trouble thinking of a good analogy, but maybe something like: Imagine a study of allergic people and healthy people. They ask both groups to inhale pollen dust through their nose (nose as an analogy for serotonin receptors). Only the allergic group shows increased symptoms. The authors conclude that people with allergies have overly sensitive noses, as opposed to anything downstream relating to the immune system (analogy for dopamine receptors).