PAQ-ing more punch for Parkinson’s


In the 1990, scientists identified some fruits that they suspected could give people Parkinson’s. 

These fruit are bad, they reported.

More recently, researchers have identified chemicals in that exist in those same fruits that could potential be used to treat Parkinson’s. 

These fruit are good, they announce.

In today’s post, we will explain why you should avoid eating certain members of the Annonaceae plant family and we will also look at the stream of research those plants have given rise to which could provide a novel therapy for Parkinson’s.


Guadeloupe. Source: Bluefoottravel

In the late 1990s, researchers noticed something really odd in the French West Indies.

It had a very strange distribution of Parkinsonisms.

What are Parkinsonisms?

‘Parkinsonisms’ refer to a group of neurological conditions that cause movement features similar to those observed in Parkinson’s disease, such as tremors, slow movement and stiffness. The name ‘Parkinsonisms’ is often used as an umbrella term that covers Parkinson’s disease and all of the other ‘Parkinsonisms’.

Parkinsonisms are generally divided into three groups:

  1. Classical idiopathic Parkinson’s disease (the spontaneous form of the condition)
  2. Atypical Parkinson’s (such as multiple system atrophy (MSA) and Progressive supranuclear palsy (PSP))
  3. Secondary Parkinson’s (which can be brought on by mini strokes (aka Vascular Parkinson’s), drugs, head trauma, etc)

Source: Parkinsonspt

Some forms of Parkinsonisms that at associated with genetic risk factors, such as juvenile onset Parkinson’s, are considered atypical. But as our understanding of the genetics risk factors increases, we may find that an increasing number of idiopathic Parkinson’s cases have an underlying genetic component (especially where there is a long family history of the condition) which could alter the structure of our list of Parkinsonisms.

So what was happening in the French West Indies?

Between August of 1996 and August of 1998, a total of 87 people with features of Parkinson’s disease were referred to the Neurology Department at the University Hospital in Pointe-à-Pitre, Guadeloupe.

Of the 87 individuals referred, a total of 22 were diagnosed with classical idiopathic Parkinson’s, while 31 had Progressive supranuclear palsy (PSP), 30 had atypical parkinsonisms, and the remaining four had atypical parkinsonism associated with motor neuron disease (basically the doctors had a hard time diagnosing them)

Ok, but what is strange about that?

The curious feature of this distribution is that less than 20% of those cases had idiopathic Parkinson’s disease.


Well, compare that with the typical distribution (around the rest of the world) of 70% of cases having idiopathic Parkinson’s, and less than 30% having atypical parkinsonisms.

Naturally the doctors in Pointe-à-Pitre were scratching their heads. Why did they have such high rates of atypical parkinsonisms and PSP?

What did they think was causing this?

Well, the researchers of the Caribbean Parkinsonism Study Group made an analysis of the affected individuals and they identified something in the diet of the affected individuals: they consumed a lot of a particular fruit.

They published their observations in this report:


Title: Possible relation of atypical parkinsonism in the French West Indies with consumption of tropical plants: a case-control study. Caribbean Parkinsonism Study Group.
Authors: Caparros-Lefebvre D, Elbaz A.
Journal: Lancet. 1999 Jul 24;354(9175):281-6.
PMID: 10440304

The researchers found that of the 31 individuals diagnosed with PSP in their cohort (or affected population), 29 of them reported regular consumption of Soursop fruit (also known as corossol), and 26 of them drank tea infusions made from the leaves of plant on a frequent basis.

Soursop fruit. Source: Wikipedia

Coincidence? Maybe.

But the investigators also found that all of the atypical Parkinsonisms cases regularly consumed Soursop fruit, and 24 of them (out of the 30) frequently drank tea made from the fruit.

In fact, both of the PSP and atypical Parkinsonisms groups consumed significantly more of the Soursop fruit than the individuals diagnosed with Parkinson’s and a large group of healthy control subjects. These observations led the researchers to conclude that the Parkinsonisms they were observing might be related to a toxin in the plant.

The plant, Annona muricata (Spanish name: guanábana), is a species of the Annonaceae family. They are native to the tropical regions of the Americas.

The investigators decided to test the idea that Annonaceae plants may have neurotoxins, and in 2002 they published their research results:

Title: Toxicity of Annonaceae for dopaminergic neurons: potential role in atypical parkinsonism in Guadeloupe.
Authors: Lannuzel A, Michel PP, Caparros-Lefebvre D, Abaul J, Hocquemiller R, Ruberg M.
Journal: Mov Disord. 2002 Jan;17(1):84-90.
PMID: 11835443

When the investigators exposed dopamine neurons being grown in cell culture to the total extract of Annona muricata rootbark, they observed 50% loss of dopamine neurons after 24 hours (at 18 micrograms/ml). Dopamine neurons being one of the populations of cells in the brain that are badly affected by Parkinson’s. The nuclei of the dying neurons exposed to the extract exhibited DNA condensation and fragmentation, which suggested that the cells were dying via programmed cells death (or apoptosis). This indicated to the researchers that there was a toxin in the extract of the Annona muricata plant.

One year later, the same researchers identified the chef toxic compound within the Annonaceae plants: Annonacin (Source)

Annonacin. Source: Wikipedia

It is a member of the class of compounds known as acetogenins. Annonacin acts on the mitochondria (the power stations of the cell), functioning as an inhibitor of normal function, thus impairing energy metabolism. Investigators have also established how much Annonacin each Soursop fruit contains:

Title: Quantification of acetogenins in Annona muricata linked to atypical parkinsonism in guadeloupe.
Authors: Champy P, Melot A, Guérineau Eng V, Gleye C, Fall D, Höglinger GU, Ruberg M, Lannuzel A, Laprévote O, Laurens A, Hocquemiller R.
Journal: Mov Disord. 2005 Dec;20(12):1629-33.
PMID: 16078200

As I mentioned above, the ‘median lethal dose’ of annonacin – that is, the dose that kills 50% of the dopamine neurons in a petri dish – is 18 micrograms/ml.

And I don’t expect that statement to mean anything to any non-scientist types like my mum, so let me explain it to you like this: when compared to other known neurotoxins, annonacin is 100-times more toxic than 1-methyl-4-phenylpyridinium (or MPP, which is a commonly used toxin for modelling Parkinson’s in the lab).

And the average Soursop fruit contains 15 mg of annonacin.

That’s rather a lot!

Time to remove Soursop fruit from your diet.

Ok, so Annonaceae fruits are bad – message received loud and clear. No longer one of my five-per-day. What next?

So here is the twist in this story.

You see, Annonaceae are not all bad.

In fact some parts of Annonaceae plants are good. REALLY good. So good in fact that they have been the focus of a lot of research for Parkinson’s disease.

Que? I’m confused. What do you mean?

This is Prof Bruno Figadère:


Bruno Figadère. Source: NLSpharma

He is a director and head of chemical research at the National Center for Scientific Research in Paris.

For a long time his lab has been working on chemicals that were originally derived from Annonaceae. Those chemicals are called tryptaminic alkaloids.

And they are particularly interesting.


Well aside from other reasons, they share several structural similarities with melatonin.

Remind me again, what is melatonin?

Source: Talkaboutsleep

Melatonin (or N-acetyl-5-methoxy tryptamine) is a natural neurotransmitter-like compound produced primarily by the pineal gland. This chemical is involved with many aspects of normal biological function, particularly in circadian rhythms and sleep disorders (Click here to read more about this).

The location of the Pineal gland. Source: Anatomypal

But melatonin is also recognised as a neurotransmitter that possesses neuroprotective properties for dopamine neurons (Click here and here to read more about this), and it was this feature of tryptaminic alkaloids that Prof Figadère and his colleagues were interested in.

They asked the question: If tryptaminic alkaloids share many structural similarities to melatonin, could they also exhibit neuroprotective properties?

And this lead to a series of experiments which resulted in this publication in 2010:


Title: Tryptamine-derived alkaloids from Annonaceae exerting neurotrophin-like properties on primary dopaminergic neurons.
Authors: Schmidt F, Le Douaron G, Champy P, Amar M, Séon-Méniel B, Raisman-Vozari R, Figadère B.
Journal: Bioorg Med Chem. 2010 Jul 15;18(14):5103-13.
PMID: 20579892

Prof Figadère and his colleagues started this endeavour by extracting many different forms of natural tryptaminic alkaloid compounds from Annonaceae plants, and investigated their neurotrophic properties. These ‘structure associated relationship’ (SAR) investigations resulted in 18 compounds being tested on dopamine neurons grown in culture. These experiments identified a 6 compounds that demonstrated the ability to help the dopamine cells to survive. The researchers further investigated for antioxidant activity of those 6 compounds, before then evaluating their ability to cross the blood-brain-barrier (a protective membrane that surrounds the brain, blocking the entrance of many compounds). This identified one compound (called 4C) which exhibited interesting properties.


Source: Sciencedirect

The brain penetrance of this compound was not as robust as hoped so the researchers turned their attention to some compounds that were structural similar compounds to tryptaminic alkaloids which are called Quinoxalines.

They conducted a similar series of analysis on these compounds, which resulted in this research report:


Title: Neuroprotective effects of a brain permeant 6-aminoquinoxaline derivative in cell culture conditions that model the loss of dopaminergic neurons in Parkinson disease.
Authors: Le Douaron G, Schmidt F, Amar M, Kadar H, Debortoli L, Latini A, Séon-Méniel B, Ferrié L, Michel PP, Touboul D, Brunelle A, Raisman-Vozari R, Figadère B.
Journal: Eur J Med Chem. 2015 Jan 7;89:467-79.
PMID: 25462259

In this study the researchers screened 11 molecules and they found two very active molecules, one of which they called MPAQ (2- methyl-3-phenyl-6-aminoquinoxaline, compound 3c). It provided substantial protection for dopamine neurons grown in cell culture, operating indirectly through a repressive effect on the helper cells in the brain (called astrocytes – these cells help nurture neurons and keep them alive). 

The neuroprotective effect of MPAQ on dopamine neurons was found to be independent of the neuroprotective chemical GDNF (or glial cell-derived neurotrophic factor – Click here to read a previous post about this). In fact, when the researchers deprived mature dopamine neurons of GDNF, they found that MPAQ could efficiently rescue the cells from dying. MPAQ was also found to be as effective as iron chelators deferoxamine and apotransferrin, demonstrating significant antioxidant properties (Click here to read a previous post about iron chelators and antioxidants in Parkinson’s)

As interesting as MPAQ was, however, Prof Figadere and his colleagues thought that they could improve on the compound even further. So they set up another set of ‘structure associated relationship’ (SAR) investigations to explore MPAQ-like compounds. And that research was published in mid-2016:


Title: New 6-Aminoquinoxaline Derivatives with Neuroprotective Effect on Dopaminergic Neurons in Cellular and Animal Parkinson Disease Models.
Authors: Le Douaron G, Ferrié L, Sepulveda-Diaz JE, Amar M, Harfouche A, Séon-Méniel B, Raisman-Vozari R, Michel PP, Figadère B.
Journal: J Med Chem. 2016 Jul 14;59(13):6169-86.
PMID: 27341519

To improve the neuroprotective activity of compound MPAQ, the researchers designed and synthesised similarly structured molecules (37 all together) and carried out their screening procedure on dopamine neurons grown in cell culture.


Source: Acs

That screen identified compound 4c (or what they decided to call PAQ). PAQ was closely related structurally to MPAQ, but it had no effect on the astrocyte cells, rather it operated directly on the dopamine neurons themselves. Its effect was partly due to the activation of calcium release channels. This mechanism of action has been previously described for a compound called paraxanthine, which is a structurally similar molecule to caffeine (Click here and here to read more about this).

PAQ also exerted substantial neuroprotection to dopamine neurons that were deprived of supportive factors (like GDNF), and demonstrated significant antioxidative stress features. Of particular interest, these neuroprotective properties were also observed in a mouse model of Parkinson’s (using the neurotoxin MPTP). These results suggested that the investigators were gradually narrowing in on a compound that could be considered for clinical evaluation for Parkinson’s.

The concentrations at which PAQ was being protective, however, were relatively high. And so, once again Prof Figadere and his colleagues conducted another set of ‘structure associated relationship’ (SAR) investigations to find PAQ-like compounds.

And earlier this year they announced to the world that they had found one:


Title: Identification of a Novel 1,4,8-Triazaphenanthrene Derivative as a Neuroprotectant for DopamineNeurons Vulnerable in Parkinson’s Disease.
Authors: Le Douaron G, Ferrié L, Sepulveda-Diaz JE, Séon-Méniel B, Raisman-Vozari R, Michel PP, Figadère B.
Journal: ACS Chem Neurosci. 2017 Feb 15.
PMID: 28140556

The researchers again took their compound of interest (PAQ), synthesised similarly structured molecules (19 in total), and tested them on their ability to protect dopamine neurons in cell culture. This identified a compound that they called PPQ. As the image below illustrates, PPQ was able to provide a more powerful neuroprotective effect on cells in culture than PAQ at much lower doses. And this was also evident in a mouse model of Parkinson’s.


Source: Acs

While the researchers conclude by saying that “Additional preclinical studies will be needed, however, to confirm that PPQ is of potential interest for PD treatment“, they are obviously very interested in its potential as a patent has been filed for it (Click here to read that patent). I will be keeping an eye out for the follow up pre-clinical research on PPQ.

So what does it all mean?

In this post we have explored how a fruit that could potential cause Parkinson’s-like conditions, can also provide the means by which we can hopefully prevent them. While it is unlikely that all cases of Parkinsonisms are caused by neurotoxins absorbed from eating fruit, it is a wondrous thing that researchers can derive chemicals from the seemingly toxic fruit and – via a repeated process of analysing similarly structured compounds – derive molecules that can exert powerful neuroprotective effects.

I have a great deal of admiration for the efforts of researchers like Prof Figadere and his colleagues. Their ‘try and try again’ approach has gradually identified and improved upon molecules that demonstrated very potent neuroprotective properties. Have they found a drug in PPQ that can be taken to the clinic? Only time will tell. Hopefully in the new year there will be follow up research telling us whether they are close. Yet more to look forward to in 2018.

One last word: 200!

The banner for today’s post was sourced from Mymonline

8 thoughts on “PAQ-ing more punch for Parkinson’s

  1. Any additional info on Asimina annonacin ?
    I just planted 95 Sunflower / Susquehanna pawpaw seeds for trials in Hawaii.
    Apparently annonacin levels vary in pawpaw cultivars & efforts are underway for lower & higher annonacin types.

    That neuroprotective bit blew my mind !


    1. Hi Eli,
      Thanks for your comment. Unfortunately I don’t have much more information than what I have written here. What kind of trials are you conducting?
      Kind regards,


  2. Simon,

    What an incredible post! I’m a scientist and health-conscious fruit collector who owns several annona fruit trees, and this topic just popped up into my radar. I loved your chronological investigation of this subject, and how it was extremely easy to follow :).

    Something seems extremely important is that the acetogenins research continued on after the 2005 paper that you found! Did you see ‘Identification of annonaceious acetogenins in the ripe fruit of the north american pawpaw (asmina triloba)’ by Pomper et. al in 2009? They look at the ‘bioactivity’ of the toxins extracted from soursop fruit, cherimoya fruit, several pawpaw fruits, peach fruit, and banana fruit. They measure bioactivity by looking at the LC50 of the extracted material to brine shrimp (not as compelling as neurons, but certainly easier to work with!). They found that the cherimoya was almost as harmless as the peach and banana. Good news for fruit collectors!

    Their results were a bit erratic throughout their paper, but the ‘NC-1’ varietal of pawpaw came up as ~100x more toxic than soursop. A homeowner binge-eating fruit from their pawpaw tree could easily end up eating ~3lbs of pawpaw in a day (or 3~5 pawpaw fruits). Assuming a typical soursop weighs 6lbs, that’s roughly like eating 50 soursop fruits in a single day! A single tree could yield 30~80 pounds of fruit, so that could amass to eating the equivalent of 500~1300 soursops per tree. That’s a dose higher than eating a full 6lbs of soursop every day *and* drinking the soursop leaf tea every day for a year, but administered all in 1~2 months! If that’s the case, fruit fanatics and pawpaw lovers in the american midwest (seriously, there are entire annual festivals around this fruit) should be suffering just as much or more than the people of french west indies due to the Indiana banana!

    One of the concluding points of that paper was that there have not yet been any links between consumption of pawpaw and the incidence rates of Parkinsons vs. Parkinsonisms. They point to another paper in the concluding remarks, and suggest that people who seem to be affected by this toxin are likely genetically predisposed. This is perhaps backed up during a presentation by the same authors with an acknowledgement of Caparros-Lefebvre et al, who estimated in 2005 that there are approximately 640 cases of parkinsons and parkinsonisms combined in Guadeloupe out of a population of 420,000, which is an incidence rate that is identical to the united states and europe. That paper amazingly also found that younger folks who had symptoms managed to reverse symptoms when they stopped annona products. It seems that this toxin doesn’t cause parkinsons, but it seems like there is clear evidence that it could make existing parkinsons worse.

    It is extremely concerning that Champy found in 2004 that injecting rats for 1 month with equivalent scaled amounts of the toxin injested from soursop induced neurodegeneration of basal ganglia and mesencephalon. However, this does not investigate bioavailability upon digestion; 1 month is really dramatic, and I would imagine that people would have quite a bad reaction if it hit them that hard every month.

    All that said, Annonas are my favorite fruits! I totally admit my bias that I’m swinging for a solution that lets me eat them, rather than a balanced viewpoint. PLEASE comment on whether I’m being too optimistic about Annonas being OK!


    1. Thanks for the very insightful comment to a very insightful blog post. I would love to hear if you have any more updated thinking on the matter, ditto if Simon does as well. I quite like eating soursop (and pawpaw) and its heartbreaking to have to stop!


  3. Great article. I’m curious about how long Annonacin stays in the brain, or is it cumulative? Yes, heartbreaking for those who love pawpaws; however, it would be difficult for most to engage in regular consumption of the fruit since it’s seasonal. Now I’m glad I wasn’t able to attend the PawPaw festival this year!


  4. Not A Valid Model Of Parkinson’s

    I checked the most recent of papers cited:
    Identification of a Novel 1,4,8-Triazaphenanthrene Derivative…
    abstract located at

    Under the materials and methods section of the fulltext:
    “male mice… were divided into four groups (n = 12/group) for testing the effects of compound 5m in vivo. … Two groups received MPTP together with 5M hydrochloride … THE ADMINISTRATION STARTING 3 DAYS BEFORE INITIATION OF MPTP TREATMENT AND CONTINUING UNTIL SACRIFICE TIME [emphasis added]. ”

    In the case of actual Parkinson’s, diagnosis generally occurs years after commencement of the illness. Whatever initiated the disease process is usually long gone. Studies such as this that use pretreatment to demonstrate protection against the toxicant are irrelevant to the reality of actual Parkinson’s. What is relevant are substances that demonstrate ability to alleviate the Parkinson’s disease process AFTER administration of the toxicant. Such studies do exist but are rare in comparison to studies such as this. Studies utilizing pretreatment are useless and should be ignored.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

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