The Microbial Metagenome

First some background.  I was first directed to the Marshall protocol by a reader who wondered about the information the found on the web.  So I went to the web and looked at the available information, much as any patient would, and discussed what I found there.

I have subsequently been lead to believe that none of the information on the website can be considered up to date or accurate.  As as result of,  I have told that my post is chockablock with errors, although, outside of writing doxycycline where I should have put minocycline, I am left in the dark as to exactly what my errors are.  I am told that it is my responsibility to locate the errors in the last post, yet I can find none when compared to the website.

However, to remedy the deficiency of having reviewed inaccurate and out of date material,  I have been sent 6 articles that I am informed represent the state of the art in understanding the science behind the Marshall protocol.  Ah, the peer reviewed medical literature.  An opportunity to carefully read and critique new  ideas.  It is one of the reasons people publish: to see if their ideas can withstand the scrutiny of others.

Several of these papers concern Vitamin D,  the Vitamin D receptor, and olmesartan which I will review, perhaps, another time.  I don’t find them a compelling read, but it not an area about which I have more than a standard medical knowledge. The other papers concern the role of infection in autoimmune diseases, which I will discuss here.  It is easier as an infectious disease doctor  to read this literature as I am, as least as far as the American Board on Internal Medicine is concerned, a specialist in the field.  Alternatively, I am a closed minded tool of the medical industrial complex who only seeks to push his own twisted, narrow agenda at the expense of suffering patients (1).  We can’t all be perfect.

The main article, which was published in July of this year, is “Autoimmune disease in the era of the metagenome (2)” will be discussed along with several (2, 3) references that are, as of the writing of this post, not yet published. For the clarity, footnotes in ‘()’ are mine, footnotes in ‘[]” are from the referenced material. I added any bold face type.

The article starts with an interesting sentence.

“According to the NIH, a mere 10% of the cells that comprise the organism known as Homo sapiens are human cells. The remaining 90% are bacterial in origin. Thus, Homo sapiens is best described as a superorganism in which a large number of different organisms coexist as one (1).”

That is certainly a curious way to think about the microbial ecosystem we carry around with us.  It is true that there are 10 to 100 times more bacteria on us than there are cells that make us up (for the sake of clarity, bacteria on the skin and in the GI tract I will refer to as on us even though from one perspective the bacteria in our gastrointestinal tract are in us. Prepositions can be imprecise.   But the GI tract is contiguous with the environment, so I will use the word on.  To be in us, the bacteria will have had to breech the cells  that line us and be in the blood, the cytoplasm, the nucleus, the interstitial spaces, etc.).

We are covered in bacteria and carry a complex ecosystem that we are just beginning to understand and of which we can grow or, by molecular techniques, identify a minority.  However those cells are not us.  While we do have some interaction with these bacteria, they are not the cells that make us.  It would be the same as saying that the cells of the fleas on a dog are part of what constitutes a dog or the cells of a monkey in a tree are part of what constitutes a tree.  There is self, and there is not-self, and our bacteria are not-self. Our genome is what we pass on to our children with reproduction.  So the opening sentence does not inspire confidence in the authors understanding of what constitutes being human.  From my understanding of human-microbial interaction the opening sentence is wrong.

“Medicine is now comfortable with the bacterial populations that exist in the gut and areas of the body in contact with the external environment, such as the mouth, ears, nose and skin. Yet, components of the human microbiota also likely persist in many other body tissues, including those which become inflamed in autoimmune disease [5] (1).”

Reference 5 is a video of a lecture of one of the authors which is a keynote address summarizing the ideas found in the articles discussed here. You evaluate a review by the quality of the arguments and the quality of the referenced material.  It has more credence if  the information is based on first class, reproduced research in high impact journals. Linking to your own lectures?  Well,  no so much.   I should put up YouTube video of me reading this blog and link to it as proof of the assertions in this entry as I suppose an on line video constitutes now constitutes valid, scholarly reference material. See. It’s on the interwebs, so it must be true.

What they appears to be saying is the complex microbiology found on us is also to be be found in us, in our cells and tissues,  and are involved in the pathogenesis of autoimmune diseases. We shall see.

They continue with

“Such bacteria can persist inside the very cells of the immune system that are supposed to kill them [6], or in biofilm communities in which they are protected from the immune response by a self-created polymeric matrix [7].”

The reference in regards to persisting bacteria is  scanning electron microscopy (EM) of white cells in three patients with eye disease and the alleged pathogens were never shown as living organisms nor have the findings of the paper ever been reproduced or verified by others.   Dr. Witosko, the author of reference 6,  published a number of studies in the 1980’s where he found what he considered to be mycoplasms in several diseases, the proof of an infection being based on EM and response to antibiotics.  Unfortunately since that time, his findings have not been confirmed by other researchers.  Searching Pubmed for mycoplasms and uveitis, sarcoid, Crohn’s and JRA yield next to nothing.  If the primary references to support your hypothesis, at least the ones that are not internet videos, are one off’s, never reproduced or verified, then the foundation upon which your are building your argument may be less than solid.

Bacteria can persist for short times in phagocytic cells during acute infections.  However, extrapolating from this paper to  bacteria living for years in phagocytes for the lifetime of the patient with an autoimmune disease  is a leap across the Grand Canyon of medical hypothesis.  Biofilms are important in persisting infections associated with prosthetic material and osteomyelitis but the role of biofilms in other diseases not associated with inert material is uncertain.

So they are building an argument for disease pathogenesis, based in large part on examples and parallel, for infections due to colonies of as of yet undiscovered organisms that evade the immune system and live in biofilms.

Because some organisms cannot be isolated, ours exist and cannot be isolated.
Because some organisms can evade the immune system, ours evade the immune system.
Because some organisms live in biofilms, ours live in biofilms.

The next paragraph concerns the easy exchange of bacterial DNA and how calling an organism E. coli may by old fashioned.  I am sympathetic  to this view of bacteria, because at one level, given their ability to exchange and utilize DNA, bacteria are a metagenome and our classification of bacteria can be imprecise.  I tend to think of bacterial identification as probability functions rather than an absolute characteristic.  They are fuzzy, and not in a good way.

“Due to their small size, hundreds, or even thousands, of bacterial cells can fit inside a human cell [6]. The combined genetic contributions of these microbes inevitably provide myriad gene products not encoded by our own relatively small genomes. This means that the human genome is only one of the many genomes that affect Homo sapiens function. In reality, the organism we call Homo sapiens is controlled by a metagenome, a tremendous number of different genomes working in parallel. “

Sorry, this seems ludicrous to me.  Sure.  We have covered with millions of bacterial genes and in some places there is an ongoing interaction between the bacteria and us.  But to suggest that these genes in bacteria are part of our genes and controlling  us is an unwarranted extrapolation and, in my understanding, more science fiction than science fact. And just because thousands of bacteria could fit into a cell it doesn’t mean they do.  One would think, given the last 50 years of light and other forms of microscopy and the ongoing search for infectious etiologies of most autoimmune diseases (a search that keeps coming up negative) that someone, some time, would have noticed huge numbers of bacteria stuffed into cells.  And that such findings would be easily reproduced by others.  Curiously, that hasn’t happened.  So if the bacteria don’t seem to fit, we must acquit…. sorry for that one.

Then another odd example:

“Bacterial gene products can be very similar to our own. For example, the metabolism of glucose-6-phosphate by both the human body and E.coli is nearly identical, so that remarkably similar metabolites are produced by both species [11]. With this in mind, the interaction between an E.coli genome and the human genome, as they exchange nutrition and toxins, increases the complexity of transcription and translation for both species.”

Thanks to evolution, we do share many common metabolic pathways with many bacteria; it is why some antibiotics will affect human cells and some anticancer drugs will affect bacteria.  However, the role of  the interaction of bacterial metabolic products, out side of locally,  is currently speculative in disease.  They then mention that in a test tube system that probiotic bacteria grown in skim milk make enzymes that inhibit angiotensin 1 converting enzyme.  How this is relevant to human systems requires  another huge leap.  It is always a question whether test tube examples have clinical relevance.  In this case, unlikely, as we are not skim milk and the bacterial strains in probiotics are usually different than the strains found in us. Possible, but I would not generate a treatment plan based on it.

The authors then give examples of how two genes are altered by bacteria in the test tube: one up regulated, one down regulated. No argument here.  When infected the body responds in a variety of ways.  Genes get up regulated and down regulated.  It is always a question whether these test tube phenomena are of clinical relevance or if they are lost in  the complex human metabolic system. Often  these effects are the equivalent of tossing a rock in the Oregon ocean and to increase the size of the waves in Hawaii.   The effects are too small to be noticed, butterflies and hurricanes not withstanding.

Then the move on to capnine:

“Created by the gliding bacteria that are present in biofilm, the sulfonolipid Capnine provides a specific example of how a bacterial metabolite could manipulate human gene expression in order to dramatically alter the progression of autoimmune and other chronic diseases.Capnine has the capacity to disrupt transcription by the VDR, one of the body’s most prolific nuclear receptors [15].”

Note could, not does. And the proof?  The reference is from a paper that is from computer modeling of the interaction of capnine and the VDR. Again with the computer modeling. No real data to show an interaction of this molecule and the VDR.  No data to show that this molecule has ever been found outside of biofilm where it could have pharmacologic effects.  Going from computer modeling to disease pathogenesis and treatment is another huge stretch.  As a clinician I am not impressed.  Computer models are interesting but have to be compared to that pesky reality before they can be taken seriously.  It is a shame that one cannot jump straight from the computer model to real life, as I would love a computer model that would makes me a Powerball winner.

“Thus, if Capnine was deregulating the VDR, it would greatly hamper the innate immune response. VDR dysfunction would cause the active vitamin D metabolite 1,25-D to rise to excessively high  levels where it could inhibit expression by the bulk of the body’s other nuclear receptors -including alpha thyroid, the glucocorticoid receptor, and the androgen receptor [20]. This would result in hormonal imbalances and also interfere with expression of the dozens of other AmPs expressed by these receptors. In vivo, the microbiota appears to gradually shut down the innate immune response over a person’s lifetime, resulting in the increased accumulation of chronic bacteria and other pathogens [21]. “

If wishes were horses, Beggars would ride: If turnips were watches, I would wear one by my side. If capnine was pathogenic, Bacteria would more than glide.

All an unproven speculation presented as fact, all following from could and if.  There is no in vitro or in vivo data to suggest that capnine is involved in anything but the production of biofilms.  Reference 21 leads to a speculative review that will be discussed later,  not original research.  But none of this is based on original research in real biologic systems.  The shutting down of the immune system by bacteria,  while stated as fact, is yet more unproven speculation.

“Eventually, genes from the accumulating microbial metagenome may determine a clinical disease symptomatology such as an autoimmune diagnosis, or simply drive the inflammation associated with the aches and pains of aging [5]. This accumulation is an extremely logical evolutionary survival mechanism. Components of the human microbiome have evolved to dysregulate the VDR receptor that would otherwise activate a potent immune response against its presence.”

No.  Most bacteria in our microbial metagenome are simply not pathogenic and lack the capacity to invade and cause disease. Bacteria need to have the capacity to cause disease, and most do not. They live outside of us, on us, and in symbiosis with us. Even if you wipe out most of the immune system with chemotherapy or AIDS and patients become ill with their own bacteria invading, the number of pathogens in these patients is limited relative to the number of bacteria  they have.  Of our thousand bacterial species, I only have to worry about a couple dozen.  Of the millions of microbial species in the world, less than a thousand have the ability to infect humans.   Most bacteria simply can’t cause disease.  They do not need to have evolved to dysregulate the Vitamin D receptor to survive.  Unless, of course, they are forming large, unseen, intracellular colonies, which as best as I can tell, they do not.

“As Royston Goodacre comments in Journal of Nutrition, we are born with a genome that, aside from the genes of species that can survive in the womb and endometrium [22], is largely human. But we inevitably die with a genome that is at least 90% bacteria.”

Nonsense.  The genome that is on us is not the genome that is in us. Our genome, what makes us human, is what we pass on with reproduction and those genes are most certainly not bacterial in origin.  I suppose under this paradigm,  that if you were to drown in a sewage pit, then you would die with an even greater genome.

They move on to discuss antibody production as a result of infection with the metagenome.

“Yet now that Homo sapiens is understood to be the product of multiple genomes, it is equally possible that the antibodies observed in autoimmune disease result from alteration of human genes and gene products by the bacterial metagenome.”

If we take one unsupported hypothesis as true then anther unsupported hypothesis is equally true. We are not the product of multiple genomes, that is hyperbole.  But if you repeat something frequently enough,  maybe it will be considered true despite a lack of evidence.  Works for politicians.   Are they saying that the metagenome alters our genes to cause disease?  That the bacteria lead to genetic mutations?  They are not talking natural selection here.

They are saying that the DNA is altered by the bacteria. News to me, and, if true, leads to a new understanding of disease pathophysiology.   It is only “equally possible,”  although I would classify that speculation as extremely unlikely.

Next is a discussion of how the bacterial flora varies in time and in location and as a result the products of that bacterial flora will change with time.  No argument here, the ongoing ebb and flow of out bacterial flora is just beginning to be elucidated.  How this alters disease manifestation is hinted at in the text, but I infer that different microbiota in people will lead to different autoimmune diseases.  Which would be good excuse for the not finding a consistent pathogen causing  a given disease in each patient. Each patient has a different microbial metagenome causing their sarcoid or RA or whatever disease is present.   I admit to being wedded to the idea that one bug causes one disease.  For over a century that concept,  at the core of Koch’s postulates (5), have helped to discover the cause of everything from AIDS  to Zygomycetes.  If inference is true, it may make the pathophysiology espoused in this paper unable to be tested or falsified.  It would also make the it impossible to isolate the infecting organisms and test it to see what is the optimal antibiotic to be used to kill it.  A clinician could never say that the patient DIDN’T have a disease, which is often important information to give a patient.

Then they move on to where and how the microbiota cause disease:

“Persistent bacteria including Francisella tularensis [25], Mycobacterium tuberculosis [26], Rickettsia massiliae [27], Brucella spp. [28], Listeria monocytogenes [29], Salmonella typhimurium [30] and others, use a variety of mechanisms to evade the immune response and survive inside macrophages and other phagocytic cells.”

It is the rare pathogen that survives, usually for a relatively short time, in the cytoplasm of cells.  Eventually it either kills the host in relatively short order or is cured. Chronic infection of cells is not the norm of these organisms, although extracellular persistent infections can occur.   As will be discussed later, it is postulated that cell wall deficient organisms persist in the cytoplasm to cause autoimmune disease.  What is curious about the organisms above is that is the mechanisms by which these organisms evade the immune system are often part of, well, er, the cell wall.  Often what allows bacteria to evade the immune system and survive intracellularly is due to constituents on the cell wall and  I would suppose that cell wall deficient organisms would lack that mechanism.

“Furthermore, various species of bacteria have been detected inside the cells of patients with juvenile rheumatoid arthritis [31],sarcoidosis [6], and other inflammatory diseases [32].”

The search for a reproducible infection in these and other autoimmune diseases continues and I will not be surprised if one day  one or more microbes are found at the center of a sarcoid granuloma. Or in a JRA joint.  But the consistent, reproducible demonstration of an infectious agent in any of these autoimmune diseases is still pending.   When they discover an infectious cause of some of these autoimmune diseases, it will probably be one lone organism, not a colony.  If it takes a village of bacteria to cause a metropolis of diseases, that would be a novel pathophysiology.

“This suggests that disease-causing microbiota largely persist in the cytoplasm of nucleated cells, where it has access to both the DNA transcription and protein translation machinery of Homo sapiens. “

Really? It does? If so, where are they?  Colonies of bacteria should be seen by light or electron microscopy and yet for 100 years these bacteria have evaded detection. Romulan cloaking technology is probably beyond their capacity. It might be suggested, but it has never been noticed.  There are the occasional reports of quasi bacterial products found in the cells on EM, never reproduced, never confirmed by alternative techniques.  Just where are they in all the specimens in all the pathology labs all over the world?  It should be easy enough to find the microbiota, should they not?

“Microorganisms are also capable of integrating their DNA with our own [36]. This results in alteration of the human DNA by the microbiota over time, potentially leading to genetic mutations associated with autoimmune diagnoses. Genetic haplotypes observed in autoimmune disease frequently have very low statistical significance, as would be expected based on knowledge that the metabolome varies from population to population and individual to individual.  In addition, host DNA repair mechanisms are susceptible to modification by the products of the metabolome. In fact, bacteria may hijack DNA repair mechanisms to generate genetic diversity without losing genomic stability [37, 38].”

The microorganisms referred to is a herpes virus, and there are a few viruses, herpes and retrovirus, that can insert its DNA into ours.  The paper refers to bacteria as the constituents of the microbiota and then they switch to a virus and do not bother to mention it.  That is disingenuous at best and misleading to anyone who did not bother to go to the footnotes.  This kind of scholarship is not reassuring to the reader.   However, higher organisms (bacteria, mycoplasma etc) do not put their DNA into ours.  There is zero data to suggest that the accumulation of bacterial DNA in humans accounts for autoimmune disease.  This is stated as if it were fact, when it is closer to science fiction. Reference 37 is in an insect and reference 39 is computer modeling.  Quite a leap from a bug and a PC to the cause of human disease.

So what do we have?

Humans have in them colonies of unseen and undiscovered bacteria that persist undetected in the cytoplasm and alter our genome in ways as of yet proven to suppresses the immune system by way of uncharacterized and unproven metabolites interacting with  the Vitamin D receptor to cause a wide variety of autoimmune diseases ranging from sarcoid to diabetes. Hey, I’m convinced. Nothing like lots of speculation and extrapolation from primary sources that are poor in quality and never reproduced or validated by alternative lines of data, and are unsupported by primary in vivo or in vitro studies.

Maybe the reader can see why I am less than enamored of the validity of the pathophysiology noted above. I read dozens of papers a month to keep up in my field. This one is not going to rank high of the compelling review list.

So how does that translate into treating infections?  For that we must turn to “Reversing Bacteria-Induced Vitamin D Receptor Dysfunction  is Key to Autoimmune Disease. (2)”, which is in press. Sorry.  I have a copy and you do not.

Here is the basic hypothesis from the introduction:

“An intraphagocytic bacterial microbiota is proposed to be the primary cause of VDR dysfunction.  Because the VDR is key to the innate immune response, its dysfunction would lead to chronic infections with a wide range of pathogens, leading to inflammation and frequent elevation in autoimmune disease markers.”

The first part of the article is a review of the same issue discussed above.  Intracytoplamsic colonization of colonies of bacteria make products that interfere with the VDR.  No new data, no primary research, many of the same references.   I am again going to focus on the infectious disease aspect of the paper, rather than the vitamin D.

“The use of pulsed, subinhibitory, bacteriostatic antibiotics is an important part of this approach. Research indicates that most standard antibiotic protocols used so far to treat chronic inflammatory diseases are ineffective,  and so new ways of using antibiotics are called for to deal with the relative treatment resistance of the bacteria involved.”

I read this and my jaw dropped.  I promised myself not to be snotty when I evaluated  these papers, but such statements really do deserve ridicule.   Maybe they are ineffective because the disease do not have infectious etiology? Maybe? The new ways of giving antibiotics? Dose them so that they cannot be effective.

In the practice of infectious diseases when patients do not respond to therapy, there can be a multitude of reasons for the failure.  One is that the patient does not have an infection.  Another is that the organism is resistant to the antibiotics being used.  In the context of the e pathophysiology of the paper, what they say is that since standard antibiotics at standard doses do not work, they will give the same antibiotic, but dosed intermittently and in small amounts, or in such a way that it should have no effect on the infecting organism.

Your infection is not getting better, so we will continue the antibiotic that is not working, but we will give you less of it.  That would only make sense to a homeopath but if I were to treat infections this way we would have no shortage of cases for morgue conference. However, novel infections get novel therapies, right?

That’s assuming that bacteria are involved, which as we have noted, while interesting, is far from proved. Antibiotics are presumed not to work against the bacteria in autoimmune disease as the bacteria are cell wall deficient and the antibiotics are dosed too high.
The first part I understand.  If there are indeed L forms persisting in the cytoplasm, which have never been demonstrated definitively in autoimmune diseases, then cell wall active agents will not work. Which is why the tetracycline class (minocycline) and macrolides are used against these bugs.

There is a long history in infectious disease of the concept of the minimal inhibition concentration.  It takes many more than one antibiotic molecule to inhibit an organism.  How many depends on the antibiotic and the organism, but is takes hundreds of antibiotic molecules to have an effect on a bacterium.  If you fall below the MIC, the organism still grows, although perhaps more slowly.  Using subtherapeutic concentrations of antibiotics are the quick and easy way to develop resistance to the antibiotics.  What doesn’t kill it makes it stronger.  Currently for most infections you give maximal dose for as short a period of time to maximize killing and minimize the development of resistance.  The opposite of what is suggested in this paper. Of course I am biased by 60 years of the successful treatment of infectious diseases and am stuck in my ways.

“The survival of “persister” cells when constant dosages are used means that pulsed antibiotics are likely to be more effective.[18]  The ability of bacteriostatic antibiotics to be effective at low dosage levels has been documented.[37]”

To date, there is not an infection treated with pulsed antibiotics.  Intermittent antibiotics helps breed resistance. Reference 37 does not concern the ability of low doses of antibiotics to be effective against bacteria, it is an article about the effect on antibiotics on the phagocytosis of bacteria, the ability of cells to ingest bacteria.  In the context of the paper, this appears to be a misrepresentation of primary sources.  Since the alleged bacteria in autoimmune disease are already intracellular, one wonders about the relevance of this reference as it applies to the use of antibiotics as proposed in this paper.

“The existence of communities of multiple bacterial species, like those occurring in biofilms, means that combinations of antibiotics are likely to be necessary to fully target all species. Thus, there is increasing support for the use of pulsed, low dosages of combinations o bacteriostatic antibiotics.”

Where are these biofilms?  In the cells? Extracellular? Never quite certain where the biofilm is in the context of autoimmune diseases.  Where are the multiple bacterial species?  One does not necessarily need multiple antibiotics just because there are multiple species of bacteria.  Often the judicious and rational use of antibiotics allows one to kill multiple organisms with one antibiotic.  Rather than delaying the development of resistance in bacteria (bacteria, not mycobacteria, so do not use Tb as an example of why I am wrong), multiply antibiotics may intensify the evolutionary pressure on bacteria to become resistant.   There is no support outside the extrapolations of this paper for the use of pulsed, low dose combination antibiotics in any disease that I treat (and cure) for a living.

“The waxing and waning of symptoms that typically follows each of the pulsed antibiotic doses supports the model’s contention that bacteria are involved in autoimmune disease. The fact that these immunopathological responses decline over time, as the patient improves during treatment, provides evidence of disease reversal.”

This is the most risible statement in the paper. An equally unjustified conclusion from a similar paper:

“Perez showed that 81% of a group of 54 patients representing 20 different autoimmune diagnoses reported continual improvement after treatment durations of 18-53 months with aVDR agonist and antibiotics – further pointing to bacteria as a causative agent in autoimmune disease.[21] (4)”

Here is the reasoning: we dose antibiotics in a manner that should have no effect on bacteria (pulsed and subinhbitory).
We therefore conclude that because the ineffectively dosed antibiotics are having an effect, the diseases have an infectious etiology.
That’s a clinical paradox that would shut down an M-5 multitronic unit.

I emphasize with the residents that unless you have firm evidence of infection (culture, serology, PCR etc), response to therapy is not a good way to prove an infectious disease etiology.  If they were to make such a  suggestion, I would pat them on the head, smile patronizingly and point out that there are four solutions to an alleged antibiotic response:

The patient improved because they had in infection that responded to antibiotics.

The patient improved because they had a self limited infection and the antibiotics did nothing.  Treating a viral pneumonia with ampicillin as an example.

The patient improved because they had a non infectious disease and the antibiotics did nothing.  They received antibiotics for their stasis changes in their legs misdiagnosed as cellulitis but responded to elevation.

The patient improved because they had a non infectious disease that responded to the antibiotics.

It is the last that makes the treatment success problematic to prove.  I used to think that the immunomodulatory effects of antibiotics and other agents were too minor to be of clinical importance.  Now I am not so sure.  The association of NSAID use and the development of group A streptococcus necrotizing fascitis and the possible decreased mortality with infections in patients who are on statins are non-antibiotic examples of possibly clinically relevant effects of minor immunomodulation.

Both the tetracycline and the macrolide classes of antibiotics are immunomodulatory, and several recent studies have demonstrated decreased mortality for infections resistant to macrolides that are treated with adjunctive macrolides.  An animal model suggests that the decrease in mortality from macolides is due to a decrease in cytokines.  The minor immunomodulatory effects may, much to my surprise, be clinically relevant.

So assuming that tetracycline and macrolide antibiotics are indeed effective in decreasing the symptoms of autoimmune diseases, in the absence of a well defined infection, you cannot know for sure of the result is from the antimicrobial drug treating an infection or the result of an immunomodulatory drug altering immune function.  I bet, since they are dosed at levels to low to be effective against bacteria, that it is the latter that will be the mechanism of action, if, indeed, they are demonstrated in clinical trials to be effective.

“Diseases responding to treatment include systemic lupus erythematosus, rheumatoid arthritis, scleroderma, Sjogren’s syndrome, autoimmune thyroid disease, psoriasis, ankylosing spondylitis, Reiter’s syndrome, type I and II diabetes mellitus and uveitis8 and sarcoidosis. [3] Fibromyalgia, though not generally established as an autoimmune disease, has also responded favorably to this approach.”

If there is a recurrent theme in this blog, it is the difficulty of assigning causality to therapeutic interventions, especially when the underlying treatment has no or unproven biologic plausibility. The acupuncturists say their interventions are effective and that chi exists because of anecdotes of efficacy and poorly controlled studies.  Yet when rigorous trials are done, acupuncture is no better than twirling toothpicks on the skin.  The plural of anecdote is anecdotes, not data.  With diseases that can naturally wax and wane there is always a question of true causality with anecdotal information.  Until you do a placebo controlled trial, anecdotal evidence may be interesting but is not definitive.

So perhaps  you can see why, as an infectious disease doctor, I am less than convinced about the infectious diseases component of the pathogenesis of autoimmune diseases as presented in these papers.

No reproducible, credible evidence that colonies of organisms and their biofilms exist inside of human cells.
No evidence outside of in silico (I love that term.  It means computer modeling, but suggests parallel validity with in vivo and in vitro) data to suggest that bacteria make molecules to alter the VDR to suppress immune function.
 No controlled trials to suggest that patients are having a response to the antibiotics due its antimicrobial, rather than immunomodulatory, effects. 
Antibiotics are used in a manner that has a long history of being ineffective and aids in the development of resistance.

Lots of extrapolation, hypothesis and anecdote. To me it is all,  as Shakespeare might say, all sound and fury, signifying nothing.

All aspects of the pathophysiology are amenable to testing, so I could easily be proven wrong.  I can easily  imagine a  clinical trial that would prove or disprove the therapies independent of the theory behind it. It could be that they have the right therapeutic idea for all the wrong reasons. It wouldn’t be the first time in medicine. Take a well defined population of sarcoid or another illness.  4 treatment groups:  2 placebos, one placebo plus antibiotics, one placebo plus olmesartan and both antibiotics and olmesartan. Everyone blinded.   It also makes me wonder: what research would the proponents consider valid in falsifying the pathyphysiology and treatment in the Marshall protocol?

Much is made by the fact that these articles are peer reviewed.  After reading the papers closely, what peers and what they were reviewing, I am uncertain. Perhaps spelling and punctuation? They didn’t pay a whole lot of critical attention to the arguments or the references.   Certainly, peer reviewed does not mean true; much of the published literature in medicine is subsequently shown to be incorrect.  It is good to have speculative ideas published, it drives innovation.  To use speculative  ideas as the basis of a complex treatment protocol is not, to mind, the safe and prudent practice of medicine.  I have in my mind a factoid, and I am not sure where I read it,  it so it may be in error, that if you have no data to support a therapeutic intervention, the odds are you will do more harm than good.  I heard that, I think, as a resident almost 25 years ago and could be misremembering, but I am sure the commentators will set me straight.

“It should be noted that the immunopathological reactions resulting from this approach are sometimes strong enough to be life-threatening in patients with more advanced disease.[3]”

It is good to note that you may be killing people with an intervention.  The ethics of almost killing people with unproven treatments is problematic.  The ethical guidelines for doing research are found in the Declaration of Helsinki, I will quote from the Wikpedia summary, which I sure will lead to derision.

“Research should be based on a thorough knowledge of the scientific background (Article 11), a careful assessment of risks and benefits (Articles 16, 17), have a reasonable likelihood of benefit to the population studied (Article 19) and be conducted by suitably trained investigators (Article 15) using approved protocols, subject to independent ethical review and oversight by a properly convened committee (Article 13). The protocol should address the ethical issues and indicate that it is in compliance with the Declaration (Article 14). Studies should be discontinued if the available information indicates that the original considerations are no longer satisfied (Article 17). Information regarding the study should be publicly available (Article 16). Ethical publications extend to publication of the results and consideration of any potential conflict of interest (Article 27). Experimental investigations should always be compared against the best methods, but under certain circumstances a placebo or no treatment group may be utilized (Article 29). The interests of the subject after the study is completed should be part of the overall ethical assessment, including assuring their access to the best proven care (Article 30). Wherever possible unproven methods should be tested in the context of research where there is reasonable belief of possible benefit (Article 32).”

I am not so certain that these (nonbinding) guidelines are being adhered to.  But if you are almost killing people, it really behooves the proponents to subject the therapies to careful clinical trials to make sure the potential for death is worth the benefit.

Which brings us to China, an internationally recognized center of human rights and ethics.  A press release has noted that the Marshall protocol is to be used in the biggest hospital in China. Bigger, of course, is always synonymous with quality.  What is being done in China is not clear:  a clinic to make money and turn out more anecdotes? Like the fake stem cell clinics? Or is a protocol finally to be put to the test with a placebo control blinded trial in a well defined patient population.  If the former, well, I can go to Mexico for a cheaper alternative.  If the latter, I look forward to eating my words when a well done clinical trial demonstrates the efficacy of the Marshall protocol.



(1) For the  record, I have not talked to a drug rep not accepted anything, not even pizza at a conference, for over 25 years.  I am, however, the Head of the Vitamin D subcommittee of the trilateral commission.
(2) Autoimmune disease in the era of the metagenome
Amy D. Proal, Paul J. Albert, Trevor G. Marshall
Autoimmunity Reviews.  Jul;8(8):677-81. Epub 2009 Feb 13.
(3) Reversing Bacteria-Induced Vitamin D Receptor Dysfunction  is Key to Autoimmune Disease
Joyce C. Waterhouse,  Thomas H. Perez,  Paul J. Albert, 
in press.
(4) Dysregulation of the Vitamin D Nuclear Receptor may contribute to the higher prevalence of some autoimmune diseases in women
Amy D. Proal, Paul J. Albert, Trevor G. Marshall, 
in press.
(5) From Wikipedia
Koch’s postulates are:
1.    The microorganism must be found in abundance in all organisms suffering from the disease, but should not be found in healthy animals.
2.    The microorganism must be isolated from a diseased organism and grown in pure culture.
3.    The cultured microorganism should cause disease when introduced into a healthy organism.
4.    The microorganism must be reisolated from the inoculated, diseased experimental host and identified as being identical to the original specific causative agent.
There are ethical and practical limitations to these postulates, but they serve as a reasonable starting point for determining causality.

Posted in: Basic Science, Medical Ethics, Science and Medicine

Leave a Comment (30) ↓