Treatment Study of Patients with Chronic Fatigue Syndrome and Fibromyalgia, based on the Glutathione Depletion–Methylation Cycle Block Hypothesis

by Neil Nathan, MD, and Richard A. Van Konynenburg, PhD

Introduction

At the 2007 conference of the International Association for Chronic Fatigue Syndrome, one of us (RVK) proposed a hypothesis for the pathogenesis and pathophysiology of CFS, called the glutathione depletion–methylation cycle block hypothesis^].¹ This hypothesis was based on research that had been done by James et al. in autism, and a recognition of similarities between the biochemical abnormalities in autism and those in CFS.² Deth et al. have published a somewhat similar hypothesis for autism.³

Shortly after the conference, with the help of a CFS patient (name withheld to protect privacy), RVK extracted part of the comprehensive treatment program developed by Amy Yasko, PhD, ND, for autism and adult neurological diseases, and suggested the use of the resulting seven supplements for treating CFS.⁴ Initial experience with this treatment led to a further reduction to five supplements.

After obtaining some positive clinical experience with the five-supplement protocol in his private practice, one of us (NN) proposed a treatment study to evaluate it in a more controlled manner. Funding for lab testing was obtained from a private source.

This article is an abridged version of a poster paper that was presented at the 2009 conference of the IACFS/ME in Reno, Nevada. The full paper can be found on the internet^].⁵ Longer-term results from this study are discussed in chapter 14 of a recently published book by one of us (NN)^].⁶

There were two overall objectives for the study:

1. 1. to test the glutathione depletion–methylation cycle block hypothesis for CFS;

2. to assess the potential efficacy of a nonpharmacologic treatment for chronic fatigue syndrome (CFS) that is based on this hypothesis and is designed to support the methylation cycle.

The Glutathione Depletion

–

Methylation Cycle Block (GD-MCB) Hypothesis for the Pathogenesis and Pathophysiology of CFS

• An individual inherits a genetic predisposition (polymorphisms in several of certain genes) toward developing CFS. (This genetic factor has not yet been fully elucidated, and is apparently more important for the sporadic cases than for the cluster cases of CFS.)

• The person experiences some combination of a variety of possible stressors (physical, chemical, biological, and/or psychological/emotional) that place demands on glutathione, the combination being different for each case.

• Glutathione levels drop, producing oxidative stress, interfering with the intracellular metabolism of cobalamin (vitamin B12) and allowing toxins to accumulate

• A functional deficiency of vitamin B12 is produced, resulting in lowering of methylcobalamin and adenosylcobalamin.

• Lack of sufficient methylcobalamin inhibits the activity of methionine synthase, placing a partial block in the methylation and folate cycles.

• Methylfolate drains from the cells into the blood via the “methyl trap” mechanism.

• Sulfur metabolites drain excessively through the transsulfuration pathway and are excreted.

• An interaction (vicious circle) is established between the partial block in the methylation cycle and glutathione depletion, and the disorder therefore becomes chronic.

• A wide range of symptoms results from these chronic abnormalities in the basic biochemistry of the cells.

• The dysfunction of the detoxication system and the immune system that results from this combination allows toxins and infections to accumulate over time, which increasingly produce effects of their own.

• Treatment should be directed primarily at increasing the activity of methionine synthase. The resulting normalization of the methylation cycle, the folate metabolism and glutathione levels will restore function to the immune system and the detoxication system as well as to a wide range of other parts of the overall biochemistry.

• It can be expected that die-off of pathogens and mobilization of stored toxins will initially produce some exacerbation of symptoms, but improvements will be experienced as the body burdens of toxins and active infections are decreased.

Characteristics of the Study

This was an open-label clinical study in a single private practice in Springfield, Missouri. Patients signed forms for informed consent.

The treatment period lasted 6 months (Note, however, that after the 6-month study period, individualized treatments were added to the basic protocol for an additional 3 months. Outcomes after this final period were not analyzed as part of the study, because treatments were not uniform during the final period.)

Objective outcome measures were obtained by lab testing (see below). The patients also performed self-rating of symptoms.

There were no restrictions on medications and additional supplements, except that they and their dosages were not to be changed without informing one of us (NN).

Subjects

Number: Thirty patients who presented sequentially to the practice of one of us (NN) were started on the treatment. One dropped out at three months for a reason unrelated to response to the treatment. The remaining 29 patients were treated for the six-month uniform-protocol period of this study. Of these 29 patients, eight did not meet the selection criteria (see below). Therefore, 21 patients were included in the statistical analysis of the results for the six-month period.

Selection criteria: Those selected to be included in the statistical analysis of results were required to satisfy the Fukuda et al. 1994 case definition for CFS and were also required to have experienced postexertional fatigue and malaise.⁷ They were not required to meet the American College of Rheumatology 1990 criteria for fibromyalgia, but 18 of the 21 patients who were selected for the statistical analysis also met these criteria.⁸

The subjects were all female and Caucasian. Their ages ranged from 33 to 84, with a mean of 52 years. They had been ill for times ranging from 1 to more than 20 years.

Additional diagnoses:

Migraine headaches – 15 patients

Irritable bowel syndrome – 13

Chronic sinus infections – 11

Endometriosis – 6

Restless leg syndrome – 5

Mononucleosis (Epstein-Barr virus) – 5

Mold exposure and/or toxicity – 5

Multiple chemical sensitivity – 4

Lyme disease (previously treated) – 2

Interstitial cystitis – 1

Mycoplasma infections – 1

Chronic vulvitis – 0

Histories of previous treatment: These patients had exhibited partial response to treatment ranging from one to twelve years in duration with a protocol that included evaluation and treatment of adrenal, thyroid and sex hormones; food allergies; intestinal dysbiosis; heavy metal toxicity; infections (EBV, Lyme disease, mycoplasma); mold exposure; magnesium deficiency; and other nutritional imbalances.

Supplement Protocol

1. Activated B12 Guard (hydroxocobalamin)⁹: 1 sublingual lozenge (2000 micrograms) daily

2. FolaPro (5-methyl tetrahydrofolate)¹⁰: ¼ tablet (200mcg) daily

3. Intrinsi B12/folate¹¹: ¼ tablet daily – combination of folic acid, 5-methyl tetrahydrofolate, and folinic acid (200 mcg); cyanocobalamin (125 mcg); calcium (22.5 mg); phosphorus (17.25 mg); and intrinsic factor (5 mg)¹²

4. General Vitamin Neurological Health Formula (a multivitamin, multimineral supplement including antioxidants, trimethylglycine, nucleotides, supplements to support the sulfur metabolism, a high ratio of magnesium to calcium, and no iron or copper)¹³: starting with ¼ tablet and increasing the dosage as tolerated, to 2 tablets daily

5. Phosphatidyl Serine Complex (phospholipids and fatty acids)¹⁴: 1 soft gel capsule daily

Methylation Pathways Panel

This panel was run initially and at 3 and 6 months. It was the main objective diagnostic used in this study. It evaluates the status of glutathione, the methylation cycle and the folate metabolism, and was used to determine the effects of the treatment on these aspects of the metabolism. This panel includes the following¹⁵:

Glutathione (plasma)

Oxidized glutathione (plasma)

Adenosine (plasma)

S-adenosylmethionine (RBC)

S-adenosylhomocysteine (RBC)

5-methyl tetrahydrofolate (plasma)

10-formyl tetrahydrofolate (plasma)

5-formyl tetrahydrofolate (plasma)

Tetrahydrofolate (plasma)

Folic acid (plasma)

Folinic acid (whole blood)

Folic acid (RBC)

It is important to note that the sample collection kit for this panel incorporates enzyme inhibitors to prevent the reduced form of glutathione from oxidizing during shipping and storage of samples. This is necessary because glutathione readily oxidizes after removal from the body.

Enumeration of Symptoms and Self-Rating of Outcome Measures

The patients were asked to mark their symptoms on a checklist (initially and at 6 months) that included 38 symptoms. The patients were also asked to rate five outcome measures initially and at 3 and 6 months on visual analog scales ranging from 1 to 10. These measures consisted of energy, sleep, mental clarity, freedom from pain, and overall feeling of well-being. In addition, at 3 and 6 months they were asked to estimate their percentage of improvement.

Results

Various patients reported some early exacerbation of symptoms, which in most cases was followed by a greater improvement in symptoms. Three of the patients found it necessary to decrease their dosage frequency to every second or third day for several days, until they could tolerate the full daily dosage schedule.

Sixteen of 30 patients (53%) reported an initial worsening of symptoms, beginning in most of these cases within 3 or 4 days, but in some cases beginning at up to 2 weeks. Most of the symptoms were mild, and none of the patients discontinued usage of the supplements during the first 3 months. The most common side effects were gastrointestinal (pain, cramps, constipation, or diarrhea), reported by 6 out of 30 patients or 20%; increase in pain, reported by 4 out of 30 or 13%; and increase in fatigue, reported by 3 out of 30 or 10%. Other symptoms, reported by one patient each, were a decrease in appetite, poor sleep, weak legs, flu-like symptoms, and an increase in anxiety and depression.

For those who experienced improvement, the time to self-reported improvement on the protocol was an average of 5.6 weeks, with a range from immediate improvement (which was rare) to as long as 8 weeks before improvement was experienced.

Results of the methylation pathways panel are shown in Table 1. The numbers shown are mean values for 21 patients, with the standard deviations in parentheses (except that the values for one patient were omitted from the red blood cell folate distributions, because of a clearly spurious value). The laboratory reference values shown in this table were determined by Vitamin Diagnostics Inc. (now called Health Diagnostics and Research Institute). They are each based on measurements in at least 120 volunteer male and female medical students between ages 20 and 40, nonsmoking and with no known chronic diseases.¹⁶ They are shown as mean values with standard deviations, except that the lab reference range is given for 5-methyl-tetrahydrofolate. This folate vitamer’s reference range is quite broad, presumably because of variations between individuals in the prominence of the so-called methyl trap phenomenon, in which lowered activity of methionine synthase causes 5-methyl-tetrahydrofolate to leak out of cells into the blood plasma.

The reference values for the ratios GSH/GSSG and SAM/SAH were calculated from the mean laboratory reference values of the parameters, but standard deviations for these ratios were not known. The p-values were determined using the one-tailed Student’s t distribution. Because the distributions of values for 5-methyl-tetrahydrofolate and tetrahydrofolate were highly skewed, these values were transformed to their reciprocals before the statistical analysis was done on them, in order to obtain distributions more nearly normal. Note that folinic acid and 5-formyl-tetrahydrofolate are the same chemical species, measured both in plasma and in whole blood in this panel.

As can be seen from the table, the initial mean values of glutathione (GSH), oxidized glutathione (GSSG), S-adenosylmethionine (SAM), and four of the folate vitamers were significantly different from their laboratory reference values. In particular, GSH was significantly depleted, as was SAM.

Table 1: Methylation Pathways Panel Results

Numbers shown in parentheses are standard deviations.

* p<0.0005 with respect to the reference value.

The following symbols indicate the p values of measured parameters with respect to their values at time 0: ** p<0.0005; *** p<0.005; **** p<0.02;

# p<0.04.

It can also be seen that significant increases in the levels of GSH, SAM and 10-formyl-tetrahydrofolate were observed after 3 months of treatment. After 6 months of treatment, significant increases were observed in GSH, adenosine, SAM, 5-methyl-tetrahydrofolate, 10-formyl-tetrahydrofolate, tetrahydrofolate, plasma folic acid and red blood cell folinic acid. Adenosine and plasma folic acid started below, and rose to values above, their laboratory reference values. The ratios GSH/GSSG and SAM/SAH were initially below their laboratory reference values. Although the former rose and the latter decreased during the treatment, these changes were not statistically significant.

Glutathione is plotted in Figure 1, S-adenosylmethionine in Figure 2, and tetrahydrofolate in Figure 3. Tetrahydrofolate is plotted because of its role as the hub of the folate metabolism.

The coenzyme forms of folate are derived from it. Its level can therefore be used as a gauge of the degree of normalcy of the folate metabolism. Because its distribution is highly skewed, as noted earlier, the median values rather than the mean values are shown. After 6 months of treatment, though there was a large spread in the values, the mean value of tetrahydrofolate had risen significantly.

As these figures show, glutathione, S-adenosylmethionine, and tetrahydrofolate all monotonically approached their laboratory reference values during the 6 months of treatment, but it appears that this period of treatment was not long enough for their mean values to reach the mean laboratory reference values. (As mentioned earlier, in fact treatment was continued for an additional 3 months, but since individualized treatments were added during this period, the results at 9 months were not included in the statistical analysis. It is interesting to note, however, that after 9 months of treatment, the mean glutathione and SAM levels had reached and exceeded their laboratory reference values.)

In response to the question asking for a subjective estimate of percentage of improvement after 6 months of treatment, 15 out of 21 (71%) reported improvement, 5 reported no improvement (24%), and 1 did not respond. Of those who reported a percentage of improvement, the percentages ranged from 5 to 98%, with a mean value of 47.5% and a standard deviation of 25%.

Time on treatment

Figure 1: Plasma reduced glutathione in 21 CFS patients

initially and after 3 and 6 months of treatment,

compared with the laboratory reference range

Time on treatment

Figure 2

:

Plasma S-adenosylmethionine in 21 CFS patients

initially and after 3 and 6 months of treatment,

compared with the laboratory reference range

Figure 3: Plasma tetrahydrofolate in 21 CFS patients

initially and after 3 and 6 months of treatment,

compared with the laboratory reference range

Results of statistical analysis of the enumeration of symptoms and the self-rating of outcome measures are shown in Table 2 as mean values and standard deviations, and the results of enumeration of symptoms are plotted in Figure 4.

Table 2: Enumeration of Symptoms

and Self-Rating of Outcome Measures

The following symbols indicate the p values of measured parameters with respect to their values at time 0: ** p<0.0005; *** p<0.005; **** p<0.02;

# p<0.025; ## p<0.04

Figure 4: Number of symptoms reported by 21 patients before and after 6 months of treatment

In response to the symptoms checklist, 20 out of 21 patients (95%) reported a decrease in their number of symptoms at 6 months, compared with their number at the start of treatment, and 1 reported an increase from 19 to 20 symptoms.

In the self-rating of outcome measures at 6 months, 16 of 21 (76%) reported improvement in energy, 16 of 21 (76%) reported improvement in sleep, 15 of 21 (71%) reported improvement in mental clarity, 15 of 21 (71%) reported greater freedom from pain, and 14 of 21 (67%) reported improvement in their overall feeling of well-being.

As can be seen, the mean number of symptoms reported by the patients dropped by nearly half after 6 months of treatment, and the mean ratings of all five of the symptomatic outcome measures were significantly improved at both 3 months and 6 months of treatment. The rating of overall feeling of wellbeing decreased between 3 and 6 months, but this decrease was not statistically significant.

Discussion

The GD-MCB hypothesis predicts that CFS patients are depleted in glutathione and have a partial block in their linked methylation cycle and folate metabolism. It furthermore predicts that the abnormalities in glutathione, methylation, and the folate metabolism are linked together, and that the key to correcting these abnormalities is to stimulate the activity of the enzyme methionine synthase.

The reduced glutathione level in the blood plasma was found to be significantly depleted in the patients before the treatment was begun, which is consistent with the hypothesis. It is important to consider how this result compares with past reports of measurements of glutathione in CFS patients. Some background information may be helpful in making this comparison:

Glutathione is present in all cells of the body, and in the blood plasma, the bile and the epithelial lining fluid of the lungs, but is compartmentalized, having different concentrations in different locations in the body. Of most interest in testing the GD-MCB hypothesis would be the levels of reduced glutathione inside the cells that are most closely associated with the symptoms of CFS, particularly those of the skeletal muscles, the heart muscle, the brain and nervous system, the immune system, and glands that secrete hormones found to be deficient in CFS.

However, though muscle biopsies can be performed, the most practical medium for clinical testing is blood. Within blood, the concentration of glutathione is about three orders of magnitude higher in the red blood cells than in the blood plasma, and it is also much easier to evaluate the glutathione level in whole blood or in red blood cells than in plasma. As a result, this is what has most commonly been done. Because of the great abundance of red blood cells in whole blood and the much higher concentration of glutathione in red blood cells than in plasma, the whole blood glutathione level is dominated by the red blood cell glutathione, so that a whole blood measurement of glutathione is essentially equivalent to a measurement of red blood cell glutathione.

In evaluating the relevance of measurements of glutathione in red blood cells or whole blood to CFS, it is necessary to consider to what degree these measurements reflect the levels of glutathione inside the cells that are actually of interest. Red blood cells normally have excess capacity for producing glutathione, and they are net exporters of it, as are the cells of the liver.¹⁷ Red blood cells have systemic roles to play in controlling oxidative stress and in conjugating toxins in addition to controlling their own redox status, and this probably accounts for their excess capacity.

This is in contrast to the situation of the cells of interest, which are net users of glutathione. These cells utilize reduced glutathione from the blood plasma, or cysteine or cystine that originated in glutathione, and they export oxidized glutathione when the amounts are in excess of their abilities to recycle it.¹⁸ As a result, the levels of reduced and oxidized glutathione in the blood plasma can be expected to more closely reflect the glutathione status in the cells of interest than will the levels in the red blood cells. The overall system of producing and distributing glutathione in the body should be kept in mind when comparing the present glutathione measurements in blood plasma to literature reports of past glutathione measurements in red blood cells or whole blood in CFS patients.

It is our view that the present measurements of reduced glutathione in the blood plasma are more indicative of the reduced glutathione levels in the cells of interest in CFS than are the measurements reported in the literature, for these reasons.

In other results in the present study, S-adenosylmethionine was found to be significantly low initially, suggesting dysfunction in the methylation cycle, and there were also significant initial abnormalities in several of the folate vitamers, indicating problems in the folate metabolism as well. These results are also consistent with the GD-MCB hypothesis.

It can be seen from the results of the present study that treatment directed specifically at increasing the activity of methionine synthase brought significant improvement in S-adenosylmethionine and several folate vitamers. This is evidence that a partial block of this enzyme was responsible for the observed methylation cycle and folate cycle dysfunctions in CFS, as proposed in the GD-MCB hypothesis.

It is particularly noteworthy that treatment directed at assisting the methylation cycle produced a significant increase in the level of glutathione, even though glutathione was not supplemented directly. This is evidence for a vicious-circle type of interaction between these two phenomena, as predicted by the GD-MCB hypothesis, and as consistent with what was found earlier by James et al. in autism^].²

As noted in the results above, the ratios GSH/GSSG and SAM/SAH were below normal initially and did not change significantly during the treatment period. A low value of GSH/GSSG indicates a state of oxidative stress, and a low value of SAM/SAH indicates lower than normal capacity for performing methylation reactions.

The fact that improvements in several of the measured parameters were still occurring after 6 months of treatment and that the GSH/GSSG and SAM/SAH ratios continued to be low at that time suggests that a longer duration of treatment would produce additional benefit (and in fact, this was observed in the subsequent 3-month treatment period that included additional individualized treatments beyond the protocol used in this study).

The initial exacerbation of symptoms that was experienced by over half of the patients may have resulted from mobilization of stored toxins, and well as toxins produced in the die-off of pathogens. Since the methylation cycle operation, the folate cycle operation, and the glutathione status were all improved, it can be expected that both the detoxication system and the immune system moved toward more normal operation, because their operation depends to a large degree on these parts of the overall biochemistry. During their extended Illnesses, it can be expected that toxins and pathogens had accumulated in the bodies of the patients, and that there would therefore have been considerable mobilization of toxins into the blood stream when these systems began functioning more normally. Because the rates of excretion into the stools, urine and perspiration are limited, the resulting elevation in toxin levels in the blood can be expected to impact tissue cells, leading to a variety of detoxication-related symptoms.

Though there was (mostly mild) initial exacerbation of symptoms in over half the patients, the symptomatic improvement in at least two thirds of the patients was the dominant effect of the treatment. There was a significant decrease (by nearly half) in the average number of symptoms after 6 months of treatment, and significant improvements in all five of the self-rated symptomatic outcome measures. Four of the self-rated measures showed monotonic improvement with treatment time. The rating of overall feeling of wellbeing at 6 months was lower than the value to which it had risen at 3 months, but this decrease was not statistically significant.

The fact that treatment of this type produced improvement in the whole range of symptoms experienced in CFS is evidence that the partial block at methionine synthase is fundamental to the pathophysiology of CFS, and this is consistent with the central feature of the GD-MCB hypothesis.

Conclusions

The results of this study are consistent with the predictions of the glutathione depletion–methylation cycle block hypothesis for the pathogenesis of chronic fatigue syndrome. This hypothesis appears to be a good candidate for more detailed testing.

A treatment based on this hypothesis and directed at supporting the methylation cycle was found to produce significant improvements in the levels of glutathione, metabolites in the methylation cycle and folate vitamers in this group of CFS patients. There were also a significant decrease in the number of symptoms and significant self-rated improvements in energy, sleep, mental clarity, freedom from pain, and overall feeling of wellbeing. Treatment to support the methylation cycle in CFS is promising, and should receive more controlled study.

Acknowledgments

This work was sponsored by an anonymous donor as part of the effort of the Ratna Ling Working Group. The authors express their appreciation for this sponsorship. We also wish to acknowledge the help of Drs. Tapan Audhya, Amy Yasko, Jacob Teitelbaum, and Ritchie Shoemaker in planning this study. Management was provided by Kevin Joyce, and nursing support was provided by Neva Dix. We are very grateful to them as well. Finally, we extend special thanks to the patients who were willing to participate in this study.

Disclaimers

The authors have no financial interest in the tests or supplements discussed in this paper. Although the treatment described in this paper consists only of food supplements, it has not yet been widely tested in chronic fatigue syndrome, and it must be entered upon only under the supervision of a licensed physician, because adverse effects are possible in patients whose general health has become quite debilitated, or those who have certain respiratory, cardiac, endocrine, or autoimmune conditions.

Notes

1. Van Konynenburg RA. Glutathione depletion–methylation cycle block, a hypothesis for the pathogenesis of chronic fatigue syndrome. Poster paper presented at: 8th Intl. Conf. on CFS, Fibromyalgia and Other Related Illnesses, IACFS; Fort Lauderdale, Florida; January 10–14, 2007.

2. James SJ, Cutler P, Melnyk S, et al. Metabolic biomarkers of increased oxidative stress and impaired methylation capacity in children with autism. Clin Nutr. 2004;80:1611–1617.

3. Deth R, Muratore C, Benzecry J, Power-Charnitsky V-A, Waly M. How environmental and genetic factors combine to cause autism: A redox/methylation hypothesis. NeuroToxicology. 2008;29:190–201.

4. Yasko A. The Puzzle of Autism: Putting it All Together. 4th ed. Bethel, ME: Neurological Research Institute; 2006.

5. Nathan N, Van Konynenburg RA. Treatment study of methylation cycle support in patients with chronic fatigue syndrome and fibromyalgia. Poster paper presented at: 9th Intl. IACFS/ME Conference; Reno, NV; March 12–15, 2009. Available at

http://aboutmecfs.org.violet.arvixe.com/Trt/TrtMethylStudy09.pdf or http://www.mecfs-vic.org.au/sites/www.mecfs-vic.org.au/files/Article-2009VanKonynenburg-TrtMethylStudy.pdf.

6. Nathan N. On Hope and Healing: For Those Who Have Fallen Through the Medical Cracks. Little Rock, AR: Et Alia Press; 2010.

7. Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff A, and the International Chronic Fatigue Syndrome Study Group. The chronic fatigue syndrome: a comprehensive approach to its definition and study. Ann Intern Med. 1994;121:953–959.

8. Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 criteria for the classification of fibromyalgia. Arthritis Rheum. 1990;33:160–172.

9. Activated B12 Guard is a registered trademark of Perque LLC.

10. FolaPro is a registered trademark of Metagenics Inc.

11. Intrinsi B12/Folate is a registered trademark of Metagenics Inc.

12. The composition of the supplement called “Intrinsic B12/folate” was changed by Metagenics after the study was completed. It no longer contains folinic acid. Actifolate (a registered trademark of Metagenics Inc.) was substituted for it in subsequent use of the protocol. The most recent revision of the protocol substitutes Hydroxo B12 Megadrops (Holistic Health Consultants LLC.) for Activated B12 Guard, Methyl Mate B drops (Holistic Health Consultants LLC) for FolaPro, and folinic acid for Actifolate. Lecithin was made an option in place of phosphatidyl serine complex for those with low cortisol. These changes were made to lower the amount of folic acid (which is difficult for some to utilize), to make the protocol more convenient, and to avoid further lowering of cortisol, which is already low in many CFS patients. (See Bailey SW, Ayling JE. The extremely slow and variable activity of dihydrofolate reductase in human liver and its implications for high folic acid intake. Proc Natl Acad Sci USA. 2009;106[36]:15424–15429.)

13. General Vitamin Neurological Health Formula is formulated and supplied by Holistic Health Consultants LLC.

14. Phosphatidyl Serine Complex is a product of Vitamin Discount Center.

15. Available from Health Diagnostics and Research Institute, New Jersey, USA; phone 732-721-1234; and European Laboratory of Nutrients, Netherlands.

16. Tapan Audhya, PhD, research director, Health Diagnostics and Research Institute. Personal communication; 2008.

17. Giustarini D, Milzani A, Dalle-Donne I, Rossi R. Red blood cells as a physiological source of glutathione for extracellular fluids. Blood Cells Mol Dis. 2008;40(2):174–179.

18. Lomaestro BM, Malone M. Glutathione in health and disease: pharmacotherapeutic issues. Ann Pharmacother. 1995;29:1263–1273.