Why Moss Nutrition is no Longer Selling L-Tryptophan Supplements (at least for now) – Part I

Some of you may remember or may have heard about a very high profile event involving L-tryptophan supplements that occurred in 1989.  Very basically, what is widely believed to be a contaminant that was found in certain batches of L-tryptophan supplements produced by a manufacturing firm in Japan lead to a constellation of signs and symptoms in consumers of this batch of L-tryptophan supplements collectively called eosinophilia-myalgia syndrome (EMS).  Within a fairly short time after the connection was made between those suffering from EMS and the contaminated L-tryptophan supplements, the FDA issued a directive that L-tryptophan could no longer be sold as a dietary supplement in the US.  This absolute ban on L-tryptophan supplements continued until about five years ago, when, because of a very compelling body of evidence that uncontaminated L-tryptophan supplements are very safe for human consumption as a dietary supplement, the FDA reversed its directive from 1989.  As L-tryptophan started to reappear in the supplemental marketplace, requests from all of you were of sufficient numbers that we began to stock the supplement again.  However, about two months ago a somewhat strange and interesting set of circumstances occurred that led us to discontinue sales of L-tryptophan supplements even though there has been no change in the FDA policy of allowing their sales as a dietary supplement.

Of course, as you might expect, what you have just read is the “six o’clock news sound byte” version of the story.  In reality the “story behind the story” is so much more interesting, strange, somewhat complicated, and very different from what you may have heard, not only recently, but during the last 25 years since the L-tryptophan connection with EMS first surfaced. Therefore, I would like to tell that story now. Where should I begin?  I feel the best place to begin would be about two months ago, or May 21, 2013, to be precise.


On that day I received a letter from the law firm Hyman, Phelps & McNamara located in Washington, DC.  The law firm was writing to me on behalf of the International Council on Amino Acid Sciences (ICAAS), which is an association of organizations that have a vested interest in maintaining the safety and quality of amino acids used as dietary supplements.  In this letter it was pointed out that Uber Rest™, the L-tryptophan product we were selling, was falsely labeled as U.S. Pharmacopoeia (USP), which violated several state and federal laws.  Because of this, the ICAAS was requesting that we cease mentioning on our website or any other marketing materials that UberRest™ is USP grade and cease all sales of falsely labeled product.

The letter went on to describe an elaborate body of laboratory tests that were performed by the ICAAS on the product on more than one occasion that demonstrated levels of impurities that are well above the limits set by the USP.  Even more disturbing was a statement in the letter pointing out that the levels of total impurities seemed to correlate with levels of L-tryptophan contaminants linked with the EMS outbreak in 1989 – l’Ethylidenebis-(L-tryptophan) (EBT) and 3-(Phenhylamino)alanine (PAA).

The letter concluded with several paragraphs describing the potential legal consequences if we did not comply with the ICAAS requests.

Hopefully, you know me well enough by now to guess that legal threats were not necessary to lead me to remove a product that appeared to contain unacceptable levels of contaminants from our inventory.  Within hours of receiving this letter, all mention of the mislabeled product was removed from our website and all other marketing venues and arrangements were made to send our remaining inventory back to the manufacturer.

As you might expect, after returning this batch of contaminated product, I had every intention of finding a source of L-tryptophan supplements that were in compliance with USP guidelines.  However, before I did so, I decided to reread the letter from Hyman, Phelps & McNamara.  Upon reading the letter again, I noticed a reference to a paper published in 2011 in a reputable medical journal suggesting that UberRest™ had been linked to a case of EMS in the Chicago area well after 1989.

Reading this paper, “Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan” by Allen et al (1) disturbed me for two reasons.  The first, most obvious reason was that I was never informed by the manufacturers of UberRest™ about the existence of this paper.  However, the second reason was a bit more subtle.  It had to do with some uncertainty as to whether the patient suffering from EMS described in this paper consumed capsules from a batch of UberRest™ that had actually been contaminated in the manner described in the letter I had received.  Because of this uncertainty, I began to wonder if it was possible, under very specific circumstances, for uncontaminated L-tryptophan to cause EMS.  As I hope to demonstrate, my literature search to address this possibility, even though it was far from conclusive, raised enough doubt to lead me to, in the interests of erring on the side of caution, a decision, until more definitive research is obtained, to refrain from supplying supplemental L-tryptophan in any form, no matter how pure.

In part II of this series you will read my review, not only of the Allen et al (1) paper on the EMS case involving UberRest™, but the papers suggesting an EMS link with uncontaminated L-tryptophan supplements.  However, to truly understand and appreciate the nature of this controversy, I feel it is important to go back to 1989 and relive what actually happened, using literature that is more detailed and more historically precise than current sound bytes.  For, all too often, these sound bytes employ recollections or secondhand retellings that lack the full sense of gravity of a truly monumental and frightening series of events that occurred over 20 years ago.


When the relationship between EMS and L-tryptophan supplements first became known in 1989, there was an almost immediate flood of media reports that seemed to fit into two major groupings.  The first group came from opponents of nutritional supplements who seemed to take never ending joy in sensationalizing an already tragic situation, creating both an overblown sense of fear among the general populace and an undeserved portrayal of everyone in the supplement industry as greedy, uncaring, and careless charlatans.  The second group came from those connected with the supplement industry who, because of what appeared to be an excessive desire to preserve image rather than protect the public and find the truth, seemed to be doing everything in their power to minimize and “whitewash” the whole debacle.  Of course, with this conflicting “war of words” going at full power, I’m sure I was just one of many who felt one major emotion after reading many of these reports: confusion.

Therefore, I was overjoyed to find, upon receiving the 1992 Annual Review of Nutrition, that one of the reviews was entitled “The eosinophila-myalgia syndrome and tryptophan.”  In this review by Belongia et al (2), a very detailed, dispassionate and objective description was presented that not only chronicled all the events in the order they occurred but described with great detail the chemistry and physiology of why a certain batch of tryptophan supplements created so much suffering for an unfortunate group of people.

The paper begins by discussing some observations made in the summer and fall of 1989 of a new malady that seemed to be afflicting many Americans:

“During the summer and fall of 1989, an epidemic of a new, multisystem illness occurred in the United States.  The disease was characterized by severe muscle pain and profound eosinophilia.  It was initially recognized in October when physicians in New Mexico treated three women with similar clinical findings.  They suspected an association with tryptophan consumption after observing that all three had consumed the food supplement prior to the onset of illness.  This finding was publicized by the local news media, and additional cases were reported by other New Mexico physicians.  Shortly thereafter, cases were also recognized in other regions of North America and Europe.  The major clinical features formed the basis for the name of the new disease: eosinophilia-myalgia syndrome, or EMS.”

After further investigation a case definition was created:

“The case definition was developed based on review of the clinical findings of the initial cases.  It included (a)eosinophil count of greater than 1,000/mm3, (b) generalized debilitating myalgia, and (c) no evidence of infection or neoplasm that would explain the clinical findings.”

Based on the above, the FDA took action:

“On November 11, 1989, the US Food and Drug Administration issued a warning that advised consumers to discontinue use of tryptophan food supplements.  The agency subsequently requested a nationwide recall of all over-the-counter food supplements that contained at least 100 milligrams of tryptophan in a daily dose.”

Was this action by the FDA effective?  The authors state:

“Although over 1500 EMS cases were identified, the epidemic was essentially halted by the removal of tryptophan from the consumer market.”

Next, the authors discuss why so many people were ingesting tryptophan supplements in the first place:

“Medical research of the 1970s and early 1980s suggested that tryptophan might be useful for treatment of depression.  Since then a number of investigations have examined its efficacy for a variety of other conditions, including insomnia, chronic pain, schizophrenia, premenstrual syndrome, affective disorders, and behavioral disorders.  The emphasis on treatment of psychiatric and behavioral disorders stemmed in part from the observation that brain serotonin content could be altered by changes in plasma tryptophan levels.”

Based on this research, the popular media in the 1980s carried information on the use of tryptophan supplements therapeutically:

“During the 1980s, reports in the popular press encouraged consumers to use tryptophan for therapeutic purposes.  The product was widely available without a prescription, and it was promoted as an over-the-counter remedy for a variety of reasons.  A 1990 survey of tryptophan used in the Minneapolis-St. Paul area found that 4% of households had at least one person who had used tryptophan between 1980 and 1989.  The prevalence of use increased markedly between 1985 and 1989 and was highest in women.”

Biochemistry and metabolism of tryptophan

To truly understand the basis for my suggestion that uncontaminated tryptophan might contribute to the development of EMS, it is important to understand how tryptophan is metabolized.  While many papers, which I will be discussing, have been written on this subject, I would like to begin my review of this key issue by presenting the overview contained in the Belongia et al (2) paper.  Because of the complexity of the tryptophan metabolic pathways, you will find, in addition to the text that follows, a diagrammatic overview from the Belongia et al (2) paper on the last page of this newsletter.

To truly understand tryptophan metabolism and why it could lead to adverse health sequelae even in an uncontaminated state, it is important to first appreciate that tryptophan is metabolized via two major pathways: (1) The pathway that ends with serotonin and melatonin, and (2) The kynurenine pathway that ends with niacin.  While many have portrayed the serotonin pathway as the “good” pathway because it leads to the production of serotonin and melatonin and the kynurenine pathway as the “bad” pathway because it can lead to excessive production of potentially neurotoxic metabolites, both are, in fact, essential to optimum health and, in reality, problems occur only when there is an imbalance in these pathways, very often induced by chronic inflammation.  Interestingly, even in an ideal health situation, the vast majority of tryptophan is metabolized down the kynurenine pathway.  Much more will be presented on the relationship between chronic inflammation and the creation of imbalances between the serotonin and kynurenine metabolic pathways later.

“Tryptophan is an essential amino acid.  It is catabolized in mammals along two main pathways, resulting in the formation of kynurenine and serotonin.  Most ingested tryptophan is degraded via the kynurenine pathway and provides precursors for the biosynthesis of niacin (nicotinic acid) and nicotinamide adenine dinucleotide.  In this pathway, tryptophan is first oxidized to N-formylkynurenine by tryptophan 2,3-dioxygenase (TDO) or indolamine 2,3-dioxygenase (IDO).  This first enzymatic step is the rate-limiting step in the degradation of tryptophan.  TDO (also called tryptophan pyrrolase) is localized in the liver while IDO is distributed throughout various tissues.  Induction of IDO or TDO increases tryptophan catabolism and the formation of kynurenine and its metabolites.”

Before continuing with the text of the description of tryptophan metabolism by Belongia et al (2), I would like to request that you pay particular attention to the next quote.  For, it is the basis of not only the suggestion that EMS might be related to uncontaminated tryptophan but of literatally hundreds of papers addressing metabolic/lifestyle causes of chronic illnesses ranging from depression to fibromyalgia to chronic fatigue syndrome and so many more:

“TDO activity is partially regulated by the hypothalamic-pituitary-adrenal axis.  The enzyme is induced by glucocorticoids and adrenocorticotrophic hormones; it is down-regulated by growth hormone.  TDO activity is also increased by tryptophan loading.  IDO activity is induced by gamma interferon (IFN-γ), and administration of this cytokine leads to increased levels of tryptophan metabolites in vitro and in vivo.  Administration of interleukin 2 also induced IDO.  Unlike TDO, IDO is not induced by either glucocorticoids or tryptophan loading.”

As I hope you can see from the above quote, tryptophan metabolism can be shifted away from the serotonin pathway towards the kynurenine pathway via two factors that are very common in our chronically ill patients; stress-induced increases in cortisol production and chronic inflammation.   Why is this a problem?  If you glance at the tryptophan metabolic pathway chart that follows the text, you will see that one of the metabolites from the kynurenine side of tryptophan metabolism is quinolinic acid, a neurologic irritant that, as I suggested above, may be a major contributing factor to increasingly prevalent chronic ailments such as depression, fibromyalgia, and chronic fatigue.

The text from the Belongia et al (2) paper continues with a brief description of kynurenine metabolism:

“Kynurenine is catabolized through several routes.  The major pathway involves hydroxylation to 3-hydroxykynurenine, followed by degradation to 3-hydroxyanthranilic acid.  Tranaminases convert small portions of kynurenine and 3-hydroxykynurenine to kynurenic acid and xanthurenic acid, respectively.”

Next the authors briefly discuss the serotonin side of tryptophan metabolism:

“A small portion of ingested tryptophan is converted to serotonin, a neurotransmitter.  This metabolic pathway is found primarily in the central nervous system.  Serotonin is degraded by monoamine oxidase and is excreted as 5-hydroxyindoleacetic acid.”

Those of you who perform organic acids testing on your patients may recognize 5-hydroxyindoleacetic acid.  Abbreviated as 5-HIAA, it is used as a measurement on organic acids testing to ascertain serotonin levels.

In concluding their description of tryptophan metabolism, Belongia et al (2) describe a tryptophan metabolite that some you may remember as an in-office urine test to evaluate digestive capacity, indican:

“A small proportion of ingested tryptophan is also metabolized by bacteria in the large intestine to indole, skatole, and other indole derivatives.  Indole is converted to indican, which is excreted in the urine along with other indole compounds such as tryptoamine, indole pyruvic acid, and indole acetic acid.”

More data on EMS prevalence

Next, Belongia et al (2) discuss EMS prevalence subsequent to the FDA action in 1989:

“By mid-1990, a total of 1531 EMS cases had been reported to the Centers of Disease control, including 27 deaths.  Eighty-four percent of patients were female, 97% were non-Hispanic white, and 86% were over 34 years old (median age, 49 years).  Surveillance data demonstrated a dramatic increase in the incidence of EMS during the summer and fall of 1989.

The prevalence of EMS was higher in the western United States than in other parts of the country, possibly because of a higher rate of tryptophan consumption in those states.  High prevalence rates were also found in states that carried out investigations of EMS, including Minnesota, South Carolina, New Mexico, and Oregon.  The high prevalence in these states may be partly attributed to more active surveillance and case identification.

In concluding this section on EMS prevalence, the authors present compelling evidence that the numbers above only represent a fraction of the actual numbers of EMS cases:

“The true prevalence of EMS is underestimated by surveillance reports.  Persons with mild disease were excluded by the surveillance case definition even if the clinical diagnosis was consistent with EMS.  In addition, surveillance data were compiled from reports submitted by physicians, and it likely that a number of cases were diagnosed but not reported to state or federal health agencies.”

Why did tryptophan supplements cause EMS?

Next, Belongia et al (2) discuss theories as to why tryptophan supplements caused EMS.  As I suggested in the beginning of this monograph, there were two primary theories:

“Two hypotheses were initially advanced to explain the association.  According to one hypothesis, tryptophan itself triggered EMS in susceptible individuals, possibly owing to abnormalities of tryptophan metabolism.  According to the other hypothesis, EMS was triggered by a contaminant that was present in some lots of manufactured tryptophan.  The latter hypothesis was consistent with the sudden appearance of the outbreak after tryptophan had been marketed for several years with no apparent effects.”

Ultimately, it was concluded that contamination was the most likely culprit:

“Epidemiologic investigations subsequently demonstrated that EMS was not triggered by tryptophan per se, but rather by exposure to a contaminant in tryptophan manufactured by one company.”

This was established via research on asymptomatic tryptophan users (controls) and EMS-afflicted tryptophan users (cases).  The findings from this research were as follows:

“Analysis of the tryptophan source for case patients and controls demonstrated a strong association between EMS and consumption of tryptophan manufactured by Showa Denko, K.K. (Tokyo, Japan).  Twenty-nine (97%) of 30 case-patients consumed tryptophan (during the month before onset) that was manufactured by this company, compared to 21 (60%) of 35 in the combined control groups…The tryptophan consumed by the 29-case patients was manufactured by Showa Denko between October 1988 and June 1989.”

The authors continue with a discussion of the manufacturing methods used by Showa Denko to manufacture tryptophan.  These methods involved fermentation of a strain of the bacteria Bacillus amyloliquefaciens.  In late 1988, Showa Denko modified the manufacturing process:

“In December 1988, the company introduced a new strain of B. amyloliquefaciens (strain V) that had been genetically modified to increase the synthesis of intermediates in the tryptophan synthetic pathway.  In 1989 the company also processed some fermentation batches with a reduced amount of powdered activated carbon (10 kg) in one of the purification steps.”

Nevertheless, the company insisted that these changes did not create contamination concerns:

“According to the company, these changes did not significantly alter the purity of the tryptophan produced, which was maintained at 99.6% or greater.”

Unfortunately, analysis of the batches linked with EMS demonstrated some disturbing findings:

“Chemical analyses of bulk tryptophan lots provided additional support for the epidemiologic findings.  High performance liquid chromatography demonstrated a unique pattern, or ‘fingerprint,’ for tryptophan manufactured by different companies.”

What was unusual about the fingerprint from Showa Denko tryptophan?  The authors note:

“The chromatogram for Showa Denko tryptophan was distinctive and included 5 ‘signature’ peaks that were present in all tryptophan manufactured by this company.  Comparison of individual peaks in case and control lots demonstrated one peak (‘peak E’) that was significantly associated with case lots.  This peak was present in 9 (75%) of 12 case lots and 3 (27%) of control lots.  The presence of peak E was also associated with the manufacturing changes described earlier.  The chemical structure of peak E was subsequently determined to be 1,1′-ethylidenebis[tryptophan] (EBT).  The chemical is hydrolyzed under acidic conditions, and its biologic activity is uncertain.  However, preliminary results from animal studies suggest that EBT may cause abnormalities of the fascia and microvasculature.”

Two other contaminants were found but there was no suggestion that they were involved in creating EMS.

Risk factors for EMS

Since there were many individuals that consumed the contaminated tryptophan batches without developing EMS symptoms, Belongia et al (2) next explored risk factors that may have acted in cooperation with EBT to create EMS.  First, was dosage an issue?  The authors point out:

“In Minnesota, EMS patients consumed a median of 40.5 grams of tryptophan per month, compared to 6.0 grams for random controls and 15.0 grams for self-referred controls.  In Oregon, EMS patients consumed an average of 1.3 grams of tryptophan per day, while the average in the two control groups was less than half that amount.  In the South Carolina cohort, the risk of EMS was 3.5 times higher for persons taking more than 4 grams of implicated tryptophan per day than for those using 0.5 to 1.5 grams daily.  This demonstrates a dose-response relationship between the amount of implicated tryptophan ingested and the risk of EMS.”

Age was also found to be a risk factor:

“Age was also found to be a risk factor for EMS in two investigations.  In the Minnesota study, the median age of case patients (45 years) was significantly older than the median age of randomly selected tryptophan users.”

Other risk factors were considered and found to not be relevant:

“A variety of factors were examined in both the Minnesota and South Carolina studies, including preexisting illnesses, asthma, smoking, alcohol consumption, and use of specific food supplements or prescription medications (e.g. nonsteroidal anti-inflammatory drugs, tricyclic antidepressants, benzodiazepines, pyridoxine).  None of these factors were significant when EMS patients were compared with controls who consumed tryptophan manufactured by Showa Denko.”

Clinical features and pathophysiology

In the next section of their paper, Belongia et al (2) go into great detail on the specifics of the clinical presentation of EMS and the underlying pathophysiology.  While a complete review of this section is beyond the scope of what I wanted to address, I do want to highlight two key paragraphs.  The first highlights the most common symptoms:

“In addition to myalgia, the most commonly reported early symptoms included arthralgias, weakness or fatigue, dyspnea or cough, rash, headache, peripheral edema, fever, and paresthesia.”

The second paragraph highlights the similarities between EMS and a clinical entity called eosinophilic fasciitis:

“The clinical and histopathologic findings of EMS overlap those of esoinophilic fasciitis.  The latter is a scleroderma-like syndrome characterized by tender swelling and induration of the subcutaneous tissue, primarily in the arms and legs.  Eosinophilia and hypergammaglobulinemia are characteristic findings, along with inflammatory infiltrates in the fascia and dermis.  Eosinophilic fasciitis is distinguished from systemic sclerosis (scleroderma) by the relative absence of visceral involvement, digital ulcerations, and Raynaud’s phenomenon, although the two diseases share many common features and may be variants of the same pathologic process.”

Could EMS and eosinophilic fasciitis be the same illnesses from a histopathologic standpoint?  The authors comment:

“…the histopathologic features of EMS and eosinophilic fasciitis are identical and…some cases of eosinophilic fasciitis were caused by tryptophan consumption.”

Why am I emphasizing the similarities between EMS and eosinophilic fasciitis?  The answer to this question will become more apparent when I review more current research that addresses my concern that uncontaminated tryptophan may, under certain circumstances, contribute to signs and symptoms consistent with EMS.

Interestingly, Belongia et al (2) did comment on the idea that disturbances on the kynurenine side of tryptophan metabolism caused by inflammation may have contributed to EMS.  As you will see, they discount this possibility:

“Abnormalities of tryptophan metabolism have been reported in patients with EMS, the toxic oil syndrome, and other scleroderma-like conditions, leading to speculation that one or more metabolites may play a role in the pathogenesis of these diseases.  In patients with EMS, both kynurenine and quinolinic acid levels are elevated compared to those of controls, and quinolinic acid is a potential neurotoxin.  However, the same abnormalities of tryptophan metabolism are found in unrelated conditions that involve chronic immune system activation, including HIV infection.  These abnormalities appear to be mediated by IFN-γ, which induces IDO, the major rate-limiting enzyme in the kynurenine pathway.  Administration of IFN-γ to cancer patients causes increased serum levels of kynurenine and quinolinic acid and decreased levels of tryptophan.  However, it does not produce scleroderma-like changes or eosinophilia.  Overall, studies of tryptophan metabolism suggest that the observed abnormalities are secondary to immune system activation; there is no evidence that they contribute to the specific pathologic changes seen in EMS.”

However, despite this very compelling rationale, research performed within the last 10-15 years, that I will be presenting in the next installment of this series, suggests that the biochemical picture may not be as crystal clear as suggested above.  In fact, as you will see, several researchers now appear to be of the opinion that supplemental tryptophan, independent of any contaminants, could contribute to the creation of EMS.

Some ideas on specific biochemical mechanisms as to why tryptophan supplements caused EMS

The last section of the Belongia et al (2) paper addresses specific biochemical mechanisms underlying the tryptophan-EMS connection.  After discussing the underlying inflammatory findings seen with EMS, the authors, as you might expect, state that the inflammatory picture described is consistent only with EBT contamination:

“This framework is consistent with epidemiologic evidence for a dose-response relationship between tryptophan consumption (a surrogate for exposure to the etiologic agent) and risk of EMS.  It is also consistent with recent reports that EBT may stimulate IL-5 production in a dose-dependent manner and that IL-5 is elevated in the serum of some patients with EMS.  In this model, the degree of eosinophil activation would be correlated with the amount of etiologic agent consumed.  Clinical signs and symptoms of EMS would presumably develop in persons who were exposed above a critical threshold level; the latter could vary depending on individual susceptibility.  However, it is unclear why the inflammatory process can become self-sustaining in individuals after exposure to the etiologic agent is stopped (i.e. tryptophan use discontinued.”

The authors continue this discussion on mechanism with an intriguing hypothesis on EBT metabolism in relation to EMS:

“Another general hypothesis involves incorporation of the etiologic agent into metabolic or biosynthetic pathways that utilize tryptophan.  EBT and tryptophan have obvious structural similarities.  If EBT is the etiologic agent, it might function as a tryptophan analogue with adverse immunologic effects.  For example, if EBT is recognized by the transfer RNA that is specific for tryptophan, it might be incorporated into a nascent protein molecule, stimulating an autoimmune response.  Alternatively, EBT might be incorporated into either the serotonin or kynurenine pathway of tryptophan degradation, leading to production of one or more toxic metabolites.”

Unfortunately, this hypothesis does not account for all findings in EMS:

“However, the hypothesis that EBT acts as a tryptophan analogue does not explain the eosinophilia or the predilection for involvement of the fascia.”


Belongia et al (2) seem to make it completely clear that a contaminant, EBT, found in a certain batch of supplemental tryptophan produced by Showa Denko is completely and totally responsible for the EMS debacle of 1989.  Or do they?

In part II of this series I will review papers on the subject published after the Belongia et al (2) paper was released in 1992 that question whether the issue is really as black and white as suggested by Belongia et al (2).

Moss Nutrition Report #252 – 08/01/2013 – PDF Version


  1. Allen JA et al. Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan. Arthritis & Rheumatism. 2011;63(11):3633-9.
  2. Belongia EA et al. The eosinophilia-myalgia syndrome and tryptophan. Ann Rev Nutr. 1992;12:235-56.