Autoantibodies in the Postural Tachycardia Syndrome (P1.272)

[Gijs]

Abstract

OBJECTIVE: To ascertain the frequency of clinically-validated autoantibodies in sera of patients with Postural Tachycardia Syndrome (POTS), by systematically assessing organ-specific autoantibody profiles. BACKGROUND: We have occasionally noted IgG neuronal autoantibodies specific for ganglionic acetylcholine receptors, voltage-gated potassium channels (Kv1-type), calcium channels, and peripherin filament in clinically evaluating patients with POTS. It has recently been suggested that autoantibodies targeting vascular and cardiac adrenergic receptors are a major cause of POTS.

To continue reading Methods and Results, go here.
http://www.neurology.org/content/84/14_Supplement/P1.272.short?rss=1

 

[Hip]

comprehensive autoantibody testing:
neural cation channels,
thyroglobulin,
thyroperoxidase and GAD65 (radio-immunoprecipitation assays),
neural-specific plasma membrane,
cytoplasmic and nuclear antibodies (standardized tissue-based indirect immunofluorescence assay on mouse cerebellum, midbrain, gut, and kidney)
muscle sarcomeric antibodies (ELISA)

From the above list of autoantibodies they tested for in this study, I am not sure if it included tests for adrenergic receptor autoantibodies, which this previous 2014 study found in POTS patients.

 

[Jonathan Edwards]

It is so sad to see badly written abstracts like this. This is the text of a poster presentation at a meeting - almost certainly by some very junior trainee wanting to show their willingness to do some research. It is crucially dependent on proper controls. There are no controls in the methods but hey presto they appear in the results. Twenty years ago an abstract like this would be rejected by the society because it is just not well enough organised. It is a bit like a thank you letter to aunty Mary that says half way through 'Oh and I hope you are well, and uncle Jack' (a properly composed thank you letter would always have that at the beginning).

The problem I see here is that they give a figure of 4% for autoantibodies in controls. But when I did a quick poll on PR of thyroid antibodies and checked the standard control rate it was 20-30%. The long and short is that this study failed to find any interesting antibodies in POTS. They threw in the thyroid ones in the hope of getting a poster accepted, which would mean they would get travel expenses for the conference. Everyone has to start somewhere but everyone else does not need to take too much notice!! And as Hip points out they did not actually look for the new antibodies that have been reported.

 

[Gijs]

@Hip, no this is another study. The study you refer too is much more interesting. Maybe professor Edwards can look into this.

J Am Heart Assoc. 2014 Feb 26;3(1):e000755. doi: 10.1161/JAHA.113.000755.
Autoimmune basis for postural tachycardia syndrome.
Li H1, Yu X, Liles C, Khan M, Vanderlinde-Wood M, Galloway A, Zillner C, Benbrook A, Reim S, Collier D, Hill MA, Raj SR, Okamoto LE, Cunningham MW, Aston CE, Kem DC.
Author information

• 1Endocrinology and Harold Hamm Diabetes Center, University of Oklahoma Health Sciences Center & Veterans Affairs Medical Center, Oklahoma City, OK.

Abstract
BACKGROUND:
Patients with postural tachycardia syndrome (POTS) have exaggerated orthostatic tachycardia often following a viral illness, suggesting autoimmunity may play a pathophysiological role in POTS. We tested the hypothesis that they harbor functional autoantibodies to adrenergic receptors (AR).
METHODS AND RESULTS:
Fourteen POTS patients (7 each from 2 institutions) and 10 healthy subjects were examined for α1AR autoantibody-mediated contractility using a perfused rat cremaster arteriole assay. A receptor-transfected cell-based assay was used to detect the presence of β1AR and β2AR autoantibodies. Data were normalized and expressed as a percentage of baseline. The sera of all 14 POTS patients demonstrated significant arteriolar contractile activity (69±3% compared to 91±1% of baseline for healthy controls, P<0.001) when coexisting β2AR dilative activity was blocked; and this was suppressed by α1AR blockade with prazosin. POTS sera acted as a partial α1AR antagonist significantly shifting phenylephrine contractility curves to the right. All POTS sera increased β1AR activation (130±3% of baseline, P<0.01) and a subset had increased β2AR activity versus healthy subjects. POTS sera shifted isoproterenol cAMP response curves to the left, consistent with enhanced β1AR and β2AR agonist activity. Autoantibody-positive POTS sera demonstrated specific binding to β1AR, β2AR, and α1AR in transfected cells.
CONCLUSIONS:
POTS patients have elevated α1AR autoantibodies exerting a partial peripheral antagonist effect resulting in a compensatory sympathoneural activation of α1AR for vasoconstriction and concurrent βAR-mediated tachycardia. Coexisting β1AR and β2AR agonistic autoantibodies facilitate this tachycardia. These findings may explain the increased standing plasma norepinephrine and excessive tachycardia observed in many POTS patients.
KEYWORDS:
adrenergic receptor; autoantibody; autonomic function; postural tachycardia syndrome

 

[Jonathan Edwards]

@Hip, no this is another study. The study you refer too is much more interesting. Maybe professor Edwards can look into this.

J Am Heart Assoc. 2014 Feb 26;3(1):e000755. doi: 10.1161/JAHA.113.000755.
Autoimmune basis for postural tachycardia syndrome.
Li H1, Yu X, Liles C, Khan M, Vanderlinde-Wood M, Galloway A, Zillner C, Benbrook A, Reim S, Collier D, Hill MA, Raj SR, Okamoto LE, Cunningham MW, Aston CE, Kem DC.

I think we discussed this one before and it did seem very interesting. To be fair to the new abstract at least they are trying to do something sensible. I just think it needs to be done with more attention to detail. I would like to believe that thyroid antibodies are higher in ME, as a broad signal of B cell disturbance, but it needs careful definition of disease cohort and controls I think.

 

[halcyon]

It has recently been suggested that autoantibodies targeting vascular and cardiac adrenergic receptors are a major cause of POTS

So they even acknowledged the 2014 paper and then they don't bother trying to look for the same antibodies? What a waste of time.

The finding of β1AR and β2AR agonist autoantibodies is kind of interesting to me. Apparently a side effect of beta agonist drugs is hypoglycemia due to insulin release caused by beta-2 receptor stimulation. I couldn't find anything official but anecdotally it sounds like some POTS only patients do suffer from hypoglycemia. Obviously those with ME do as well, as noted by Dr. Ramsay.

 

[Gijs]

The study of 2014 will be replicated in a large cohort soon. I wonder if Propranolol will be helpfull.

 

[Hip]

I wonder if Propranolol will be helpfull.

Propranolol is a standard treatment for POTS. Other POTS treatments are listed here: POTS - What Helps

 

[Hip]

@Jonathan Edwards
As a slight aside: has any autoimmune research been conducted on the idea that in chronic intracellular infections (and ME/CFS is frequently linked to these), when the major histocompatibility complex (MHC) is presenting peptide fragments of the pathogens inside the cell to the immune system, could it be at this time that some errors are made, such that the MHC starts presenting fragments of normal cellular components to the immune system as well, so that autoimmune reactions are triggered?

I know very little about the operations of the MHC, but this thought occurred to me. Obviously in normal viral infections, where the virus only enters the cell briefly for reproduction purposes before destroying the cell through lysis and breaking out, there may be little time for these autoimmune errors to occur. But in intracellular infections that continue for years (such coxsackievirus B infections or Chlamydia pneumoniae infections), there may be greater opportunity for the MHC to make these errors, and start displaying peptide fragments of normal cellular components.

Also, I believe the general idea in these intracellular infections is that the pathogen starts to take some control over the cell, so this would mean that cellular function would not operating as normal anyway, thus perhaps creating greater propensity for MHC errors. Coxsackievirus B3 has been shown to actually down-regulate surface expression of MHC I, using various ruses, so this is an example of how intracellular infections take some control over the cell.

And for example, if we had a chronic intracellular infection of neurons or glial cells, which I presume will be rich with the proteins and peptides from neurotransmitter receptors such as the adrenergic receptor, you might perhaps expect autoimmune reactions to these receptors to be the most likely ones to occur, if the MHC is malfunctioning.

 

[Jonathan Edwards]

@Jonathan Edwards
As a slight aside: has any autoimmune research been conducted on the idea that in chronic intracellular infections (and ME/CFS is frequently linked to these), when the major histocompatibility complex (MHC) is presenting peptide fragments of the pathogens inside the cell to the immune system, could it be at this time that some errors are made, such that the MHC starts presenting fragments of normal cellular components to the immune system as well, so that autoimmune reactions are triggered?

The MHC presents normal components all the time anyway - and the immune system is designed to ignore it. We have no evidence that it ever makes mistakes at this level. The mistakes in the common forms of autoimmunity are B cell mistakes, not T cell mistakes. The errors are not at the level of MHC.

Coxsackievirus B3 has been shown to actually down-regulate surface expression of MHC I, using various ruses, so this is an example of how intracellular infections take some control over the cell.

Lots of viruses trigger downregulation of MHCI but this is generally thought to be a good thing because if a cell does not express MHCI it is recognised as 'not proper self' by NK cells and killed.

And for example, if we had a chronic intracellular infection of neurons or glial cells, which I presume will be rich with the proteins and peptides from neurotransmitter receptors such as the adrenergic receptor, you might perhaps expect autoimmune reactions to these receptors to be the most likely ones to occur, if the MHC is malfunctioning.

This is the old molecular mimicry theory which from my point of view should be pensioned off. It has been around for 50 years and nobody ever found much evidence to support it (and people have found lots of evidence that does not).

 

[halcyon]

And for example, if we had a chronic intracellular infection of neurons

My understanding is the majority of neurons don't actually display MHC molecules on their surface which gives them some of their immune privilege.

 

[Hip]

The MHC presents normal components all the time anyway - and the immune system is designed to ignore it.

Right. I did not know that.

Lots of viruses trigger downregulation of MHCI but this is generally thought to be a good thing because if a cell does not express MHCI it is recognised as 'not proper self' by NK cells and killed.

That very interesting. So I might guess that from the virus's perspective, in order to maximize its survival chances, it might want to down-regulate MHC I just enough to usefully reduce CD8 T cell surveillance, but not enough so as to rouse NK cell responses, which would kill the cell.

I guess if anyone were designing an antiviral to target chronic coxsackievirus B infections, an antiviral strategy might be either try to reduce production the viral proteins that down-regulate MHC 1, with the idea that then CD8 T cells will then be allowed to do their antiviral job; or alternatively, actually further promote production of these viral proteins that down-regulate MHC 1, in which case the NK response will finish off the virally infected cell.

This is the old molecular mimicry theory which from my point of view should be pensioned off. It has been around for 50 years and nobody ever found much evidence to support it (and people have found lots of evidence that does not).

If the molecular mimicry theory does not look viable, do you have some suggestions, even just ball park ones, as to how type II autoimmunity to self might arise (I understand that autoimmunity to cellular surface components is classed as type II autoimmunity, and sometimes referred to as type V autoimmunity when cellular surface receptors are involved)?

The adaptive immune system must have had some type of initiating process where it learn to make antibodies to these receptors.

 

[Hip]

My understanding is the majority of neurons don't actually display MHC molecules on their surface which gives them some of their immune privilege.

That's useful to know.

I think when it comes to coxsackievirus B, the nervous system cells this virus infects are astrocytes, and also neural progenitor cells (brain stem cells).

 

[Jonathan Edwards]

That very interesting. So I might guess that from the virus's perspective, in order to maximize its survival chances, it might want to down-regulate MHC I just enough to usefully reduce CD8 T cell surveillance, but not enough so as to rouse NK cell responses, which would kill the cell.

Maybe, but I am not sure we have any reason to think these are dissociable.

If the molecular mimicry theory does not look viable, do you have some suggestions, even just ball park ones, as to how type II autoimmunity to self might arise (I understand that autoimmunity to cellular surface components is classed as type II autoimmunity, and sometimes referred to as type V autoimmunity when cellular surface receptors are involved)?

The adaptive immune system must have had some type of initiating process where it learn to make antibodies to these receptors.

Yes, I have very specific suggestions, proposed in a review article in Immunology 1999, authored by myself, G Cambridge and V Abrahams. It is probably open access now.

Firstly, type II and type V are types of hypersensitivity, not autoimmunity. They were invented by Robin Coombs, who I go to know quit well in his retirement. This was before people talked of 'autoimmunity' and Robin, very reasonably refused to use that word, since he said it was silly to say you were 'immune' to yourself. He talked of autosensitivity. But hypersensitivity types are types of inflammatory effector mechanism that tend to target foreign rather than self antigens. Type I is allergy to pollen etc and involves IgE antibody bound to receptors on mast cells that are activated when foreign allergen is present. Type II is antibody attacking cell surface, as in mismatched blood transfusion reaction (self versus foreign) or rhesus disease in babies (maternal versus self) or autoimmune haemolytic anaemia (self antibody versus sefl antigen - true autoimmunity). Type III turns out to be two quite different mechanisms involving antibody binding free antigen to form complexes. In one type these deposit in blood vessels (as in lupus or other nephritis) and in the other they activate macrophages (as in RA). Type IV is the only T cell one. Type V was never in the original classification and is really only autoimmune hyperthyroidism.

All the antibody mediated types that can target self (II,IIIa, IIIb, V) are due to autoantibodies and differ only in the nature of the self antigen - whether cell surface or soluble. The explanation for production of autoantibodies that we derived in 1996 is in the Immunology paper. Basically, for about 50 common self proteins there is a mechaism whereby an autoantibody -producing B cell clone can subvert what should be a negative feedback loop to a positive loop because the binding of antibody to antigen creates an aberrant immunomodulatory signal. The easiest way to understand this is that most of the antigens involved are themselves immune signals (IgG Fc, CIq, TRIM21, peanut agglutinin receptor, AChR, thyroxine receptors, DNA, etc) and when they are bound to antibody their signalling function gets misdirected.

The key feature of this theory is that you need no trigger for autoimmunity to set up other than the chance generation of a B cell that just happens to be able to create an abnormal positive loop. This explains why autoimmunity appears almost entirely randomly in susceptible individuals, since B cells generate antibodies randomly.

Molecular mimicry was never any good as a theory because there was never an explanation for why it should happen when it does. The theory says that the immune system never makes antibodies to self, but sometimes it does. That is a feeble theory. If every time we saw a foreign protein that looked like a self protein we made autoantibodies we would all have autoimmunity in infancy. There are millions of foreign proteins in the environment that closely resemble self proteins in their surface epitopes. Moreover, autoimmune disease are the same world over, despite microbial populations being very different - etc. Molecular mimicry also fails to take into account the fact that B cells and T cells recognise completely different forms of antigen - and there is no reason for both to mimic so the theory does not even work that way. In the theory we have proposed the T cells involved always recognise something quite different from the B cells, in fact a different protein - so there is no T cell reactivity to self.

 

[Snow Leopard]

The aforementioned paper:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2326840/

 

[Hip]

Basically, for about 50 common self proteins there is a mechaism whereby an autoantibody -producing B cell clone can subvert what should be a negative feedback loop to a positive loop because the binding of antibody to antigen creates an aberrant immunomodulatory signal. The easiest way to understand this is that most of the antigens involved are themselves immune signals (IgG Fc, CIq, TRIM21, peanut agglutinin receptor, AChR, thyroxine receptors, DNA, etc) and when they are bound to antibody their signalling function gets misdirected.

The key feature of this theory is that you need no trigger for autoimmunity to set up other than the chance generation of a B cell that just happens to be able to create an abnormal positive loop. This explains why autoimmunity appears almost entirely randomly in susceptible individuals, since B cells generate antibodies randomly.

It seems like a very intriguing idea and theory (although of course I don't really have the knowledge to understand it to any depth).

Can I ask a few question to see if I have understood the general concept:

By this theory, would autoimmune reactions only occur with self antigens that contain one of these 50-odd common self proteins? So for example, if we take the case of autoantibodies to adrenergic receptors, according to this theory, does that imply that the adrenergic receptors must contain one of those 50 proteins?

And likewise, does the theory imply any self antigen which does not contain one of these 50 self proteins cannot become a target for an autoimmune reaction?



In this video tutorial of how B cells respond to antigens, the video narrator says (at timecode 7:40) that:

During their development they [the B cells] weed out all the combinations that would bind to things that are essentially you [self]

Is this where the theory of autoimmunity yourself, Cambridge and Abrahams are proposing says that it can all go horribly wrong, and lead to an autoimmune state?

Is this where the theory says that instead of weeding out B cells that make antibodies to self, this weeding out mechanism can sometimes go wrong in the case of these specific 50 self proteins, due to the negative feedback loop that would normally do the wedding out being inadvertently turned into positive feedback loop, and thus disastrously promoting the production of B cells that make antibodies to self?



Is this unfortunate promotion of B cells that make antibodies to self seen as a one-off random accident, which once it occurs, creates a long lasting pool of such B cells by polyclonal expansion, and these B cells thereafter cause autoimmunity by their constant production of antibodies to self? In other words, after this unfortunate one-off random accident occurs, is it only the pool of B cells that causes and maintains the autoimmune state?

Or is the idea that under certain (perhaps unknown) conditions which predispose the body to a higher risk of developing autoimmunity, these random accidents of promotion of B cells making antibodies to self happen more frequently, and it is a continual stream of these accidents which underpins and maintains the autoimmune state?



And regarding conditions that increase the risk of triggering autoimmunity: are there any factors the theory might predict would increase the likelihood of the normal negative feedback loop being inadvertently turned into a positive loop, for these 50 self proteins?

I understand that there is a significant increase in autoimmune diseases worldwide, suggesting some environmental factors are playing a role in the triggering process. So possibly these factors are directly affecting the inadvertent conversion of a negative feedback loop into the positive one that the theory says triggers autoimmunity?

 

[Jonathan Edwards]

It seems like a very intriguing idea and theory (although of course I don't really have the knowledge to understand it to any depth).

Can I ask a few question to see if I have understood the general concept:

By this theory, would autoimmune reactions only occur with self antigens that contain one of these 50-odd common self proteins? So for example, if we take the case of autoantibodies to adrenergic receptors, according to this theory, does that imply that the adrenergic receptors must contain one of those 50 proteins?

And likewise, does the theory imply any self antigen which does not contain one of these 50 self proteins cannot become a target for an autoimmune reaction?



In this video tutorial of how B cells respond to antigens, the video narrator says (at timecode 7:40) that:


Is this where the theory of autoimmunity yourself, Cambridge and Abrahams are proposing says that it can all go horribly wrong, and lead to an autoimmune state?

Is this where the theory says that instead of weeding out B cells that make antibodies to self, this weeding out mechanism can sometimes go wrong in the case of these specific 50 self proteins, due to the negative feedback loop that would normally do the wedding out being inadvertently turned into positive feedback loop, and thus disastrously promoting the production of B cells that make antibodies to self?



Is this unfortunate promotion of B cells that make antibodies to self seen as a one-off random accident, which once it occurs, creates a long lasting pool of such B cells by polyclonal expansion, and these B cells thereafter cause autoimmunity by their constant production of antibodies to self? In other words, after this unfortunate one-off random accident occurs, is it only the pool of B cells that causes and maintains the autoimmune state?

Or is the idea that under certain (perhaps unknown) conditions which predispose the body to a higher risk of developing autoimmunity, these random accidents of promotion of B cells making antibodies to self happen more frequently, and it is a continual stream of these accidents which underpins and maintains the autoimmune state?



And regarding conditions that increase the risk of triggering autoimmunity: are there any factors the theory might predict would increase the likelihood of the normal negative feedback loop being inadvertently turned into a positive loop, for these 50 self proteins?

I understand that there is a significant increase in autoimmune diseases worldwide, suggesting some environmental factors are playing a role in the triggering process. So possibly these factors are directly affecting the inadvertent conversion of a negative feedback loop into the positive one that the theory says triggers autoimmunity?

I have had too good a Spanish evening meal to reply tonight but your questions indicate that you have locked on to the main idea better than most immunologists. The answers get complicated but you are on track. I do not personally think we have evidence for autoimmunity becoming more common. I have not seen any significant data on that. Our main model plays down environment, although we know that smoking was an important factor in RA as a threshold lowering agent. But another key feature of the theory is that every disease works a little bit differently, because the way the vicious cycle gets going is different in each case. That allows for a variant on the theme in which environmental agents like viruses could step in and dance in a new style.

The point about only the proteins that can trip an immune signal themselves being involved is certainly central to the theory. And that is a very interesting test of the theory in the context of adrenergic receptors. So far we know that B cells have adrenergic receptors and cortisol receptors so there would be a way to make it fit, but to feel confident that we could explain this I would like to see a more detailed story of the sort we have for RA and lupus. It needs a bit of thought.

 

[Jonathan Edwards]

As an addendum to the above: there is of course no hard and fast rule about just 50 proteins being potential autoantigens, just as there is no hard and fast rule about whether any given protein can 'modulate the immune response'. The most plausible approach I think is to look at it statistically. Some proteins are so obviously major immune signals, like IgG Fc and C1q, that they are obvious candidates as autoantigens. Others like hormone receptors, that are present on lymphoid cells as well as other cells would be less obvious candidates so might be expected to turn up as autoantigens less often. Still others might have such a tenuous link to the immune system that they only become autoantigens very rarely (1 person in a million or ten million). The situation is, however, complicated by the fact that almost certainly evolution will have built in protection against the most obvious mistakes (like IgG Fc), so relative likelihood for each protein may be skewed. What I have always thought interesting is just how often thyroid related proteins become autoantigens. There is a lot more to understand. What does seem true is that there are large numbers of common proteins like albumin and transferrin and even immune signallers like MHC that have never been found to act as autoantigens - which they should if a simple molecular mimicry theory was true maybe.

 

[Snow Leopard]

In this video tutorial of how B cells respond to antigens, the video narrator says (at timecode 7:40) that:
During their development they [the B cells] weed out all the combinations that would bind to things that are essentially you [self]

This isn't completely true.

From one of my previous posts:

"B-cell selection and the development of autoantibodies"
http://arthritis-research.com/content/14/S4/S1

It was suggested that the bone-marrow selection processes are not as specific as say, for T-Cells in the thymus.

The following rates of polyreactivity (auto-reactive to self antigens, though not necessarily with high affinity) have been shown (see above study for references)

75% very immature CD20+CD27- CD38hi CD24hi CD10+ b-cells are polyreactive

25 to 45% of immature blood B cells CD20+ CD10+ CD21lo IgMhi CD27-

After transition to mature B-Cells: CD20+ CD10- CD21-hi IgM-hi CD27- 10 to 20%

It was therefore suggested that the two key checkpoints are in the Bone Marrow and there is an additional checkpoint in periphery as B cells transition from immature to mature.

So there certainly is the possibility of self-reactive B-cells, but this on its own does not lead to autoimmune processes, unless those autoreactive B-cells are activated by a T-cell and continue to receive further survival signals.

The next major check point is where the naive B-cell is activated by an antigen (binding to the B cell receptor, which is then internalised and fragments re-presented on the surface, bound to MHC class II receptors. It is then possible for the B-cell to be activated, and proliferate (though not yet as antibody secreting plasma cells).

http://www.ncbi.nlm.nih.gov/books/NBK27142/

Compared to B-Cells, T-Cells are far more extensively screened against self-reactivity and thus if the B-cell was to present peptides from self-proteins, then nothing would happen.

But what would happen if the antigen was a combination of a self protein/molecule (which the BCR bound to), and a non-self protein. What would happen if the B-Cell presented peptides from this non-self protein to a T-cell and was subsequently activated?

This alone does not necessarily lead to autoimmune disease either, as the B-Cell clones still need to receive further survival signals.

That is why there needs to be some sort of feedback loop that allows these B-cell clones to continue to receive survival signals (and ultimately end up as plasma cells). Otherwise the process would effectively end in a short period of time and no long term disease would result.

Those 50 proteins that @Jonathan Edwards mentions are just examples of where this can go wrong, and in principle there may be other additional possibilities that we do not yet know of.

Environmental factors can certainly be an issue (eg infections!), but in terms of understanding the resultant autoimmune disease, these environmental factors might not be specific enough to be important.

If I've got any of this wrong then certain people can/will correct me.