- Email: [email protected]
Enzyme-based therapy for autism spectrum disorders – Is it worth another look?
Medical Hypotheses (2002) 58(5), 422±428 & 2002 Published by Elsevier Science Ltd. 1
doi: 10.1054/mehy.2001.1513, available online at http://www.idealibrary.com on
Enzyme-based therapy for autism spectrum disorders ± Is it worth another look? Mark A. Brudnak,7 Bernard Rimland,1 Roy E. Kerry,2 Margaret Dailey,2 Robert Taylor,3 Bruce Stayton,4 Frank Waickman,5 Michael Waickman,5 Jon Pangborn,6 Ilene Buchholz6 1
Autism Research Institute, San Diego CA; 217 6th Avenue Greenville PA 16125; 3Triangle ENT Services, Durham, NC 27704; 2116 South Sterling Ave Independence, MO 64052; 5544 White Pond Dr., Ste B. Akron, OH 44320; 6Bionostics Inc., St Charles, IL 60175 and 7 Makwood Inc., Grafton, WI, USA 4
Summary Autism is a developmental disease usually manifesting within the first three years of life. To date, no causative agent has been found. Similarly, treatment options have been limited. Of the treatment options available, a number of them have been nutritionally based in an attempt to address one or more of the theories regarding the etiology of the disease. An example would be enzyme therapy for the digestion of purported offending neuroactive peptides collectively known as exorphins. This paper discusses the exorphin theory of autism and subsequent treatment with dietary enzyme therapy. Novel data are presented in support of the theory that enzymes play a critical role in autism. Forty-six patients between the ages of 5 and 31 were selected for inclusion in the study based on a diagnosis placing them in the category of the autism spectrum disorders (ASD). The diets were supplemented with a novel dietary enzyme formulation, ENZYMAID, for a period of 12 weeks. Progress was tracked according to the Symptom Outcome Survey (SOS) (1) form method of symptom charting and presented in a table for further analysis. The novel enzyme formula, ENZYMAID, beneficially and safely affected all 13 of the parameters measured. Improvements ranged from 50±90%, depending on the parameter measured. Enzyme therapy to treat ASD may indeed a viable option in treatment protocols. These results indicate that further controlled studies are warranted. & 2002 Published by Elsevier Science Ltd.
INTRODUCTION It has been estimated that 5 : 10,000 to 1 : 80 (clusters) suffer from autism with an initial manifestation of symptoms by age three (2). In reality, the incidence of autism may be several fold higher than previously thought. While
Received 18 July 2001 Accepted 29 October 2001 Correspondence to: Mark A. Brudnak, Ph.D., N.D., Vice President Technology ± MAK Wood, Inc., 1235 Dakota Dr., Units E-F. Grafton, WI 53024-9429. E-mail: [email protected]
422
the exact cause of autism remains elusive, considerable advances have been made in recent years. These advances come from a study of the geographic localization, biological, and psychological aspects of the disease. From these studies, several theories have emerged. One such theory is that autistics suffer from several maladies with the accumulated symptoms being categorized into the autism spectrum disorders, herein simply referred to as autism or ASD. Typically, the autistic's digestive tract is thought to function sub-optimally. Two important pioneers of this work, Reichelt and Shattock, observed a significant correlation between the symptoms of autism and an impaired ability to adequately digest
Enzyme-based therapy for autism spectrum disorders
peptides/proteins from dairy (casein) and wheat (gluten). During digestion, pre-opioid-type compounds in the diet, typically from casein and gluten, are thought to be activated due to an incomplete breakdown of proteins (3,4). These exorphins (i.e., casomorphins and gluteomorphins or gliadorphin) are then easily transferred across the lumen of the gut into the circulation where they exert opioid-type action on the brain. The transfer of peptides across the lumen of the gut is thought to occur at high levels due to the `leaky' nature found to be associated with the autistic's gastrointestinal tract. According to the exorphin theory of autism, the attenuated level of DPPIV could manifest as autistic symptoms (5). Enzyme therapy has recently been employed to try to remove exorphins from the system. This has been based on supplementation with large amounts of proteases from the different categories of proteolytic enzymes and these have included acid or carboxyl peptidases, serine, cystein and zinc proteases. Additionally, peptidases with both endo- and exo-peptidase activity have been favored. One approach has been to use a large amount of peptidase to specifically digest the exorphins. While sound in theory, it has only met with limited success. The data below address the question of whether or not a more proactive enzyme formulation can have a positive effect on various parameters in ASD. While previous attempts to address a nutritional-based therapy have tried to completely eliminate all contributions of exorphins from the diet via exclusion of the causative agents, it has been determined that it is not entirely practical or 100% guaranteed (6). Those that are not removed may go on to exert a neurotoxic effect. Previous attempts to deal with this situation via enzyme therapy have been to digest the peptides after they are made, while they are in the small intestines, via a strong peptidase enzyme. In the present study, the enzyme formula was designed with a very high acid-stable protease to thoroughly digest the proteins to the fullest extent possible in the shortest period of time. The idea being that if the exorphins are going to form anyway from dietary constituents, driving the reaction forward with large amounts of acid-stable enzyme in the stomach allows more time for the body's endogenous enzymes, that have evolved to specifically digest exorphins, to work. Continuing with that rationale, the following enzymes were selected for inclusion in ENZYMAID. The enzyme AFP was used as a first line of attack in the breakdown of proteins into small pieces. This enzyme functions in the stomach where it hydrolyzes large proteins into smaller peptides. The presence of a large amount of this enzyme is crucial to forcing the reaction to happen faster. The net result is that any exorphins that are going to form, will do so in the stomach, before they reach a point of absorption. & 2002 Published by Elsevier Science Ltd.
423
To assist with this digestion, large amounts of bromelain were also used. Bromelain was chosen over papain because research has suggested that bromelain has a wider specification and function than papain. Also, it is common knowledge that papain is naturally high in sulfites and the researchers attempted to limit any potential negative effects. Bromelain has been demonstrated to be an effective anti-inflammatory, which has also been implicated as important in austism (6). The CASE-GLUTENASE is a peptidase concentrate providing the next to last assistance with exorphin digestion. This enzyme has both high exo-peptidase as well as endo-peptidase activity and contains known DPPIV activity (Table 2). This provides an overall digestion profile suitable to any extraneous peptide development. Phytase was added for its ability to digest phytic acid, which is present in plants such as corn, rice, wheat, soybean, and other beans, etc. Phytic acid can negatively affect absorption of minerals such as zinc, calcium, magnesium, copper, manganese, and iron. Phytase supplementation results in greater bioavailability of these important minerals. It has been suggested that autistics are deficient in zinc, so the extension was to try to increase the bioavailability of dietary zinc. The only non-enzyme constituent of the blend is galactose. Galactose was included to function as a genomeceutical. Genomeceuticals are naturally occurring compounds, which are able to beneficially affect gene expression. The inclusion of galactose in the formulation is based on three different works. The first was the report by Smith et al. (7) that galactose can increase the expression of dipeptidyl peptidase-IV (DPPIV or CD26) in murine enterocytes. The second is based on the finding that adding glucosamine, could increase the level of transcription and translation of important genes (8). Both transcription and translation are important for a nutrient to be able to affect the levels of a protein made from the DNA. Wang's work demonstrates clearly, elegantly, and definitively, that the concept of genomeceuticals is sound. By adding glucosamine to the diet, not only did they get more RNA, but they got more protein as well. The corollary is that by adding galactose, more DPPIV can be made. This is the third work and is based on Brudnak's theory of genomeceuticals suggesting that because the addition of glucosamine to a diet can increase the expression of leptin (fat hormone) in the body, the addition of galactose may have a similar effect on DPPIV (9). Genomeceuticals describe nutrients which can affect the structure of a gene, how well the gene products (protein and sometimes RNA) work, and/or how much of it is made. That is to say, genomeceuticals can not just replace substances, which may be missing (e.g., an enzyme diminished by mutation), but actually alter the expression Medical Hypotheses (2002) 58(5), 422±428
424 Brudnak et al.
and functionality of gene products in, and resulting from, genomic multi-level nutrient-sensing pathways. The last category is where the inclusion of the galactose comes into play. In this case, the genomeceutical would be galactose. Galactose appears to be able to increase the expression of the Dipeptidyl peptise IV gene. This means adding galactose can increase the amount of DPPIV that is present. If in autistics, the situation is not that there is absolutely zero DPPIV made, due to a mutated gene or regulated element, but rather that the gene has been silenced or attenuated (down-regulated), then the addition of galactose has the potential to reverse or circumvent that. The caveat being that the gene needs to at least be functional and this is discussed below. Indeed, DPPIV has been suggested to be down-regulated in autistics and is currently being used as a diagnostic marker for the disease. However, that is not the only place DPPIV, and DPPIV-like enzymes, are expressed and may prove to be secondary to other sources. For instance, it is well-known that more than one of the probiotic species inhabiting the human GI tract expresses an analog capable of digesting exorphins via DPPIV analogs. It is equally well-known that there are more probiotics in the GI tract than there are cells in the body. Probiotics are a major contributor to the metabolic activities of the human host, including digestion of potentially offensive substances. Galactose is a known prebiotic (substance capable of selectively stimulating the growth of probiotics verses pathogenic organisms) (10,11). As such, the addition of galactose to the formula was a natural selection. This paper addresses the theory that foods are not just nutrients but bioactive substances. The data below are presented to support that hypotheses by determining the effectiveness of the rationale above as gauged by 13 markers assessed in 22 patients. The results of the application of ENZYMAID to autistics are presented below. Statistical analyses suggest that the enzyme formulation is both sound and effective for the treatment of autism spectrum disorders. Additionally, data are presented on the DPPIV enzyme analog levels of various endogenous enteric bacteria to support Brudnak's theory that probiotic bacteria can assist in alleviation of austistic parameters.
Study protocol and human subjects The study protocol was approved by a Medical Review Board. The legal guardians of all participants signed appropriate `informed consent' forms. The study was carried out with the highest standards of ethics. Initially, 46 patients between the ages of two and 21 were chosen for the study (12). Of those that started the study, 22 remained for the entire 12-week period. For data analysis, only those subjects who completed the entire course of therapy were included. All raw data are kept on file for public inspection at the Autism Research Institute (ARI), San Diego, CA. Enzymes All enzymes were manufactured by MAK Wood, Inc. (Grafton, WI). They were manufactured specially for Kirkman Laboratories (Lake Oswego, OR) to be free of milk, casein, wheat, gluten, gliadin, corn, soy, egg, yeast, sugar, starch, MSG, sterates, palmitates, artifical sweeteners, colors or flavors, preservatives, salicylates and other common allergens. Kirkman Laboratories kindly supplied all the enzymes. The following enzymes and activities were included: CASO-GLUTENASE 10,000 AU; Bromelain 230 BTU; Acid Fast Protease 100 SAPU; Lactase 330 LacU; Phytase 125 U; Galactose (as Genomeceutical) 100 mg. Dosing The enzymes were to be taken at the beginning of each meal (or early on into the meal). The capsules were swallowed whole or pulled apart and the contents mixed with the first few bits of food or in beverages. One-half to one capsule with each meal was suggested with a gradual increase. Assay of enzyme effect Parents, guardians, teachers, or therapists were asked to complete an SOS form every two weeks. They evaluated the 13 different parameters of function and behavior in the child. DPPIV assay
MATERIALS AND METHODS All reagents were processed and approved by the MAKTechTM process of material qualification which includes DNA fingerprint analysis, strain vertification, optimal growth conditions, etc. for bacterial cultures, endotoxin assay, sample purity, heavy metal analysis, and microbial plate counting. Whenever possible, all reagents are supported by published literature. Medical Hypotheses (2002) 58(5), 422±428
A modification of Tachi et al. (1992, Phytochemistry, 3: 3707±3709). Principle: Gly-Pro p-nitroanilide is hydrolyzed not Gly-Pro and p-nitroaniline by X-Prolyl dipeptidyl-aminopeptidase. The p-nitroaniline liberated from the substrate is measured at 405 nM. Unit Definition: One unit of enzyme activity is defined as the enzyme quanitity that liberates 1 mmole of p-nitroaniline from the substrate per minute. Reagents: Substrate solution: 0.45 nM Gly-Pro & 2002 Published by Elsevier Science Ltd.
Enzyme-based therapy for autism spectrum disorders
p-nitroanilide (in the form gly-Pro-p nitroanilide p-toluenesulfonate slat Sigma Chemical #G-2901) in 50 mM potassium phosphate buffer (pH 7.5). Standards solution: 1.448 mM p-nitroanaline solution in ddH2O. Stop reagent: 0.1 N hydrochloric acid. Standard preparation and assay: Pipet 1-, 2-, 3-, and 4-ml portions of p-nitroanilide solution into a series of 50 ml volumetric flasks, dilute each to volume with 50 mM phosphate buffer and mix. These dilutions contain 0.029, 0.058, 0.87, and 0.116 mmol/ml, respectively. Determine the absorbance of each dilution at 405 nM. Plot absorbance against mmol/ml of p-nitroanlinine. A straight line through the origin must be obtained. Divide mmol/ml by the absorbance of each dilution to obtain the factor (F) value (mmol/ml). Average the four values. Approximate value for F 0.10 mmol/ml. Sample preparation and assay: Dilute sample to a concentration of 0.005±0.001 units/ml with water. Add 0.5 ml of substrate solution to a test tube and pre-incubate for 5 minutes at 30 C. Add 0.1 ml of sample solution to the tube and incubate with agitation for 30 min. Sample blank to contain 0.1 ml of distilled water in place of sample solution. After 30 minutes incubation, add 0.5 ml of 0.1 N HCl (stop reagent) to tube. Determine absorbance at 405 nM (sample absorbance A1; sample blank absorbance A2). Calculation:
425
1.1 Sample solution volume (ml); 30 Incubation time (min); C the sample solution concentration (g/ml). Statistical analysis A Student's t-test for two groups was used for the statistical analysis. A Java Implementation by Bryan Lewis, Kent State University, 1997 (13). The significance is shown as the `P value' for the two groups. The two groups were considered to be Before and After treatment with enzymes. The analysis was run in two ways. The first is the column designated `Best Case' in Table 1. This column represents a comparison of before and after treatment assuming that all patients scored an initial zero for the before treatment score. The second is shown in the column designated `Worst Case' in Table 1. This column represents a comparison of before and after treatment assuming that all patients scored an initial marking equal to the 1±2 week period score. This was done to elevate the baseline. RESULTS Of the 46 patients that initially started the treatment, 17 dropped out during the first few weeks of the 12-week trial. Those that remained for the entire 12 weeks were included in the data analysis. For each of the 13 parameters, patients were assayed by the following 0±4 scale: 0 none; 1 possible; 2 moderate; 3 significant; 4 great. The results in Table 1 list the percentage of patients that scored between a moderate and significant improvement by a guardian.
X-prolyl dipeptidyl-aminopeptidase mg/g A1 ÿ A2 F 1:1 0:1 30 C A1 Absorbance of enzyme sample; A2 Absorbance of reagent blank; F p-nitroaniline factor (mmol/ml);
Table 1 Percentage of patients showing improvement Start
Eye contact Socialization Attention Mood Hyperactivity Anxiety/Compulsion Stimming Comprehension Speech Sound sensitivity Digestion Sleep Perseveration
0 0 0 0 0 0 0 0 0 0 0 0 0
Weeks %
P value
W1±2
W3±4
W5±6
W7±8
W9±10
W11±12
Best case
Worst case
37 42 40 36 31 20 27 40 27 17 35 23 33
47 67 54 52 31 41 38 45 41 17 50 36 38
43 71 63 57 50 46 27 58 53 18 56 43 44
56 76 59 60 75 47 38 55 47 42 50 50 50
67 81 73 56 75 41 31 50 38 25 56 64 39
67 90 68 59 80 60 50 63 44 50 50 57 53
8.08Eÿ04 7.66Eÿ04 6.05Eÿ04 5.03Eÿ04 0.002315 0.001284 8.75Eÿ04 4.92Eÿ04 6.77Eÿ04 0.004429 4.64Eÿ04 0.001454 5.38Eÿ04
0.01361 0.00348 0.00451 0.00237 0.01879 0.00397 0.03593 0.00947 0.0047 0.05748 0.00269 0.00671 0.01246
An observer scored 22 patients for the 13 parameters listed. The column `Start' is the total initial score as a percentage for improvement for each parameter. These were not actually scored and are assumed to be zero for each. Scores were taken every two weeks, totaled, and indicated as a percentage of the total. `Best case' P value represents the signi®cance based on a Student's t-test using the initial `Start' of zero. The `Worst case' P value represents the signi®cance based on a Student's t-test using the initial Week 1±2 reading as the baseline. The two groups analyzed for both Best-case and Worst-case were the initial Start vs. W 1±2, 3±4, 5±6, 7±8, 9±10, 11±12 percentages. P values < 0.05 are considered signi®cant.
& 2002 Published by Elsevier Science Ltd.
Medical Hypotheses (2002) 58(5), 422±428
426 Brudnak et al.
Patients who stopped the study in the first few weeks, reported a number of reasons including: need to start other therapies; family illness; personal issues; changed physicians; did not tolerate the taste of the supplement and/or no change in symptoms. Additionally, there were six patients who had `adverse responses' including hyperactivity/increased aggression, increased stimming (one), diarrhea/loose stools (two), provocation of red ears and cheeks (one), increased hunger (one), and stopped eating (one). Approximately 40% (data not shown) of the participants were following a gluten-free and casein-free diet but all were instructed not to change their diet during the treatment. An overwhelmingly positive trend is seen for each of the parameters. The two greatest improvements were seen in socialization and hyperactivity with 90% and 80% improvements, respectively. The lowest improvements were for stimming, speech, and sound sensitivity, each scoring 50%. Statistical analysis (Student's t-test) using zero for the base-line in the `Best case' column indicated that all scores were significant. The P values ranged from 0.004429, for sound sensitivity, to 8.75 10ÿ04 for stimming. Because the zero point was an assumed number, a further round of analysis was pursued. Here, as indicated in column `Worst case' the baseline was chosen to be the same as that of the initial 1±2 week rating. All parameters except for one, sound sensitivity, were again shown to be statistically significant. Due to the addition of galactose in EnzymAidTM and its known prebiotic effect on lactic acid bacteria, an analysis was done on several products that are commonly used to support GI function. Each product ranged from a single lactobacillus strain (Acidophilus PlusTM) to a mixed culture of several organisms (L. acidophilus, L. rhamnosus, Bifidobacterium BB12, S. thermophilus, L. bulgaricus). Table 2 presents the results for each. As can be seen, there was detectable DPPIV activity in all the products tested. Additionally, a newly developed (Via MAK TechTM) DPPIV Table 2 DPPIV enzyme activity of various gastrointestinal bacteria Enzyme or strain TM
MAK Tech Rhamnosus MAK TechTM DPPIV Bifido-ComplexTM Acidophilus PlusTM Pro-Bio GoldTM Multi-Flora SpectrumTM Pro-Culture GoldTM Pro-Bio Balance Flora maintenance
DPPIV level (mg/g) 1.07 4.28 0.60 0.60 3.55 3.04 1.99 0.77 0.39
Several common probiotics were assayed for DPPIV activity according to the modi®ed procedure outlined in the Results. The range of activity was from 0.39 to 4.28 mg/g for the bacteria and enzymes and was 0 mg/g for the negative (buffer) control (data not shown). Medical Hypotheses (2002) 58(5), 422±428
enzyme (MAK Tech DPPIV) was tested and showed the highest level of activity. Further experiments are planned to test this product singly in a similar treatment protocol for ASD. DISCUSSION There were a number of special conditions experienced with the study, which may have contributed to the dropout rate. Most of those that dropped out did so early on in the study. The majority of those that dropped out had reasons other than negative effects of the study on the patient. Also, the families were instructed not to start any new therapies while in the study. This proved to be an unexpected problem as several families apparently had a strong desire to try other approaches. This was an interesting development and will need to be considered when designing follow-up studies. Despite the dropout rate, it is believed that the ENZYMAID formulation was generally well-tolerated as only a few respondents reported adverse effects which may be directly attributed to the enzyme formulation. While other studies using enzymes have been undertaken, a Medline search revealed that none have been reported in the literature at this time. While the present study is similar in structure, the novelty of the enzyme formulation is pronounced. In addition to adding several new enzymes, this is the first reporting of the therapeutic potential of genomeceuticals. The functioning of galactose is believed to be at two levels. The first is as a genomeceutical. Here, it is believed to be increasing the gut expression of the DPPIV gene. This increased expression allows for a greater level of DPPIV enzyme in the enterocytes, promoting the more thorough breakdown of any exorphins produced by the proteases. For a review of this, the reader is referred elsewhere (9). The second and equally interesting possibility is that galactose is serving as a fuel source (prebiotic) of the beneficial microflora (i.e., probiotics) in the gut. This is important because the probiotic organisms themselves contain enzymes capable of breaking down such exorphins. Varmanen et al. (2000, pp. 146±154) recently showed that probiotic organisms, currently utilized as health supplements, contain analogs of the Dipeptidyl peptidase IV enzyme (e.g., PepX14) which is known to be able to digest exorphins. With over 1011 microorganisms in the gut, their contribution of enzymatic activity can far exceed that of the enterocytes. It is well documented that galactose is a prebiotic (i.e., stimulates growth of probiotics) and can increase the number of probiotics in the gut (15±18). The importance of the gut-associated microflora can not be overstated in discussions of enzyme therapy theory. & 2002 Published by Elsevier Science Ltd.
Enzyme-based therapy for autism spectrum disorders
Increasingly, with the use of mono- or bi-strain products, there are reports of a reduction of clinical effects seen over time. For instance, in the case of autism, a child that pre-treatment had either diarrhea, or some other gastrointestinal distress, improves dramatically when first supplemented with those products. The child starts producing well-formed stools on a regular basis. However, in some cases, the child can undergo a reversion to the previous distressed state. Two approaches seem to have found a resolution for this consternating phenomenon. First, is the `pulse' method. Here, when the clinical effect begins to change, the physician can cease the probiotic supplementation for a period of time, followed by a repeat application. The body appears to be `resetting' itself after cessation of the therapy. During this period, changes are occurring in the gastrointestinal tract that may be making the local environment more hospitable to the probiotics. Additionally, there are certainly changes in the immune functioning of the gut. It is well established that the intestines are a major source of immune competence. It is equally well established that there are genetic elements that respond to heavy metals, and in particular to mercury. What may be happening is that upon initial exposure to the probiotics, the gut-associated immune system is mobilized and may be attacking the probiotics. After withdrawal of the probiotics, the immune reaction tapers off until any subsequent exposure. This is not a bad thing and can be used efficaciously by pulsing the probiotics. Also, a variation on that theme can be employed. That is to say, instead of withdrawing all probiotics for any period of time, a substitution of another product, which contains different organisms would seem the logical next step. For instance, if a product that just L. rhamnosus is being used and the above condition presents, a product of just Bifidobacterium could be employed. If a single strain of Bifidobacterium is used, and again the same situation occurs, then a third product with several strains could be administered. It most certainly is going to have to go on a case-by-case basis, according to the physician's observations. In the foreseeable future, there will be products designed specifically to address the issue in this way. This `probiotic rotation' will be applicable in a variety of situations. For these reasons, it is expected that the inclusion of a probiotic regimen or pulsing or rotating strains may be of great benefit. It is theorized here that what is happening is both the level of DPPIV in the gut is increased and the level of DPPIV-type activity, contributed by a large increase in the bacterial flora, is also increased. The net result is that due to the rapid formation of any exorphins, from the high level of acid-stable protease, the levels of absorbed exorphins dropped below a threshold required for & 2002 Published by Elsevier Science Ltd.
427
manifestation of the parameters measured. To test this hypothesis, further studies are planned to assay urinary polypeptide levels on retained samples both before and after treatment (11). Also, future studies will look at blood peptide levels both before and after treatment. Additionally, further studies are planned to determine the genomeceutical contribution to the formula. Again, increasing levels of galactose will be assessed for against all the parameters measured in the present study, along with blood levels of exorphins and stool levels of various probiotic species. The only known contraindication, for the use of galactose, is in a fairly rare metabolic disorder, galactosemia. However, the amount of galactose in the formula, 100 mg, is far less than that present upon full digestion of a glass of milk. A typical glass of milk contains around 12 grams of lactose, which, when broken down, would contribute 6 grams of galactose to the diet. That is approximately 60 times the amount in a capsule of ENZYMAID. Data has been presented merely to support the hypothesis that enzymes can have a beneficial therapeutic effect on ASD. Additionally, the theory of genomeceuticals and their potential beneficial effects as a part of a treatment protocol for ASD was supported. Finally, novel data on the DPPIV level of various probiotics were presented in support of their use in an enzyme therapy-based model for the treatment of ASD. As an uncontrolled pilot clinical study the present work has tremendous value for the support of these hypotheses. Not only did the overwhelming majority of parameters measured show a significant benefit of the enzyme blend, but there were also very few associated negative reactions. Also, this is the first time a genomeceutical approach has been therapeutically applied, not to just autism, but to any disease state. While the etiology of autism and the ASDs still remains elusive, clearly the present study advances the knowledge base for efficacious treatment protocols. Several theories have been addressed in this work. First, the idea that foods, such as enzymes and monosaccharides, are really bioactive substances and not just nutrients has been addressed and supported by their measurable effects. Further the theories of genomeceuticals, exorphin theory of the etiology of autism also appear to be substantially supported. Further the theories that probiotic organisms can significantly contribute to not just the overall metabolic activities of an organism but also assist in the alleviation of disease (e.g., autistism) symptoms via their contribution of enzymatic activities has also been explored. Taken as a whole, the entire area of genomeceuticals and their effects on ASD appears to deserve a closer look and may indeed prove to eventually be the breakthrough those in the area have been searching for. Medical Hypotheses (2002) 58(5), 422±428
428 Brudnak et al.
ACKNOWLEDGEMENTS Special thanks go to Ronda Dukes for data collection and tabulation. Vector Laboratories for technical assistance. The Autism Research Institute for meaningful discussions and data retention.
REFERENCES 1. Dr. Bernard Rimland. Personal Communication. Autism Research Institute. San Diego, CA. 2. Turner M., Barnby G., Bailey A. Genetic clues to the biological basis of autism. Mol Med Today 2000 Jun; 6(6): 238±244. 3. Haileselassie S. S., Lee B. H., Gibbs B. F. Purification and identification of potentially bioactive peptides from enzymemodified cheese. J Dairy Sci 1999; 82: 1612±1617. 4. Reichelt K. L., Hole K., Hamberger A., Saelid G., Edminson P. D., Braestrup C. B., Lingjaerde O., Ledaal P., Orbeck H. Biologically active peptide-containing fractions in schizophrenia and childhood autism. Adv Biochem Psychopharmacol 1981; 28: 627±643. 5. Drs. Alan Friedman and Jon Pangborn. Personal Communication. 6. Brudnak M. A. Application of genomeceuticals to the molecular and immunological aspects of autism. Medical Hypotheses. In Press. 7. Smith M. W., James P. S., Peacock M. A. Galactose effects on enterocyte differentiation in the mouse jejunum. Biochim Biophys Acta 1991 Jul 10; 1093(2±3): 1446.
Medical Hypotheses (2002) 58(5), 422±428
8. Wang J., Liu R., Hawkins M., Barzilai N., Rossetti L. A nutrient-sensing pathway regulates leptin gene expression in muscle and fat. Nature 1998; 393: 684±688. 9. Brudnak M. A. Genomic multi-level nutrient-sensing pathways. Medical Hypotheses 2001 Feb; 56(2): 194±199. 10. Langhendries J. P., Detry J., Van Hees J., Lamboray J. M., Darimont J., Mozin M. J., Secretin M. C., Senterre J. Effect of a fermented infant formula containing viable bifidobacteria on the fecal flora composition and pH of healthy full-term infants. J Pediatr Gastroenterol Nutr 1995 Aug; 21(2): 177±181. 11. Mozzi F., Rollan G., de Giori G. S., Font de Valdez G. Effect of galactose and glucose on the exopolysaccharide production and the activities of biosynthetic enzymes in Lactobacillus casei CRL 87. J Appl Microbiol 2001 Jul; 91(1): 160±167. 12. Kirkman Laboratories. Personal communications. Wilsonville, OR. 13. http://nimitz.mcs.kent.edu/blewis/stat/tTest.html 14. Varmanen P., Savijoki K., Avall S., Palva A., Tynkkynen S. 54X-prolyl dipeptidyl aminopeptidase gene (pepX) is part of the glnRA operon in Lactobacillus rhamnosus. J Bacteriol 2000 Jan; 182(1): 146±154. 15. Hove H., Norgaard H., Mortensen P. B. Lactic acid bacteria and the human gastrointestinal tract. Eur J Clin Nutr 1999 May; 53(5): 339±350. Review. 16. Hove H., Nordgaard-Andersen I., Mortensen P. B. Effect of lactic acid bacteria on the intestinal production of lactate and short-chain fatty acids, and the absorption of lactose. Am J Clin Nutr 1994 Jan; 59(1): 74±79. 17. Mortensen P. B., Hove H., Clausen M. R., Holtug K. Fermentation to short-chain fatty acids and lactate in human faecal batch cultures. Intra- and inter-individual variations versus variations caused by changes in fermented saccharides. Scand J Gastroenterol 1991 Dec; 26(12): 1285±1294.
& 2002 Published by Elsevier Science Ltd.
doi: 10.1054/mehy.2001.1513, available online at http://www.idealibrary.com on
Enzyme-based therapy for autism spectrum disorders ± Is it worth another look? Mark A. Brudnak,7 Bernard Rimland,1 Roy E. Kerry,2 Margaret Dailey,2 Robert Taylor,3 Bruce Stayton,4 Frank Waickman,5 Michael Waickman,5 Jon Pangborn,6 Ilene Buchholz6 1
Autism Research Institute, San Diego CA; 217 6th Avenue Greenville PA 16125; 3Triangle ENT Services, Durham, NC 27704; 2116 South Sterling Ave Independence, MO 64052; 5544 White Pond Dr., Ste B. Akron, OH 44320; 6Bionostics Inc., St Charles, IL 60175 and 7 Makwood Inc., Grafton, WI, USA 4
Summary Autism is a developmental disease usually manifesting within the first three years of life. To date, no causative agent has been found. Similarly, treatment options have been limited. Of the treatment options available, a number of them have been nutritionally based in an attempt to address one or more of the theories regarding the etiology of the disease. An example would be enzyme therapy for the digestion of purported offending neuroactive peptides collectively known as exorphins. This paper discusses the exorphin theory of autism and subsequent treatment with dietary enzyme therapy. Novel data are presented in support of the theory that enzymes play a critical role in autism. Forty-six patients between the ages of 5 and 31 were selected for inclusion in the study based on a diagnosis placing them in the category of the autism spectrum disorders (ASD). The diets were supplemented with a novel dietary enzyme formulation, ENZYMAID, for a period of 12 weeks. Progress was tracked according to the Symptom Outcome Survey (SOS) (1) form method of symptom charting and presented in a table for further analysis. The novel enzyme formula, ENZYMAID, beneficially and safely affected all 13 of the parameters measured. Improvements ranged from 50±90%, depending on the parameter measured. Enzyme therapy to treat ASD may indeed a viable option in treatment protocols. These results indicate that further controlled studies are warranted. & 2002 Published by Elsevier Science Ltd.
INTRODUCTION It has been estimated that 5 : 10,000 to 1 : 80 (clusters) suffer from autism with an initial manifestation of symptoms by age three (2). In reality, the incidence of autism may be several fold higher than previously thought. While
Received 18 July 2001 Accepted 29 October 2001 Correspondence to: Mark A. Brudnak, Ph.D., N.D., Vice President Technology ± MAK Wood, Inc., 1235 Dakota Dr., Units E-F. Grafton, WI 53024-9429. E-mail: [email protected]
422
the exact cause of autism remains elusive, considerable advances have been made in recent years. These advances come from a study of the geographic localization, biological, and psychological aspects of the disease. From these studies, several theories have emerged. One such theory is that autistics suffer from several maladies with the accumulated symptoms being categorized into the autism spectrum disorders, herein simply referred to as autism or ASD. Typically, the autistic's digestive tract is thought to function sub-optimally. Two important pioneers of this work, Reichelt and Shattock, observed a significant correlation between the symptoms of autism and an impaired ability to adequately digest
Enzyme-based therapy for autism spectrum disorders
peptides/proteins from dairy (casein) and wheat (gluten). During digestion, pre-opioid-type compounds in the diet, typically from casein and gluten, are thought to be activated due to an incomplete breakdown of proteins (3,4). These exorphins (i.e., casomorphins and gluteomorphins or gliadorphin) are then easily transferred across the lumen of the gut into the circulation where they exert opioid-type action on the brain. The transfer of peptides across the lumen of the gut is thought to occur at high levels due to the `leaky' nature found to be associated with the autistic's gastrointestinal tract. According to the exorphin theory of autism, the attenuated level of DPPIV could manifest as autistic symptoms (5). Enzyme therapy has recently been employed to try to remove exorphins from the system. This has been based on supplementation with large amounts of proteases from the different categories of proteolytic enzymes and these have included acid or carboxyl peptidases, serine, cystein and zinc proteases. Additionally, peptidases with both endo- and exo-peptidase activity have been favored. One approach has been to use a large amount of peptidase to specifically digest the exorphins. While sound in theory, it has only met with limited success. The data below address the question of whether or not a more proactive enzyme formulation can have a positive effect on various parameters in ASD. While previous attempts to address a nutritional-based therapy have tried to completely eliminate all contributions of exorphins from the diet via exclusion of the causative agents, it has been determined that it is not entirely practical or 100% guaranteed (6). Those that are not removed may go on to exert a neurotoxic effect. Previous attempts to deal with this situation via enzyme therapy have been to digest the peptides after they are made, while they are in the small intestines, via a strong peptidase enzyme. In the present study, the enzyme formula was designed with a very high acid-stable protease to thoroughly digest the proteins to the fullest extent possible in the shortest period of time. The idea being that if the exorphins are going to form anyway from dietary constituents, driving the reaction forward with large amounts of acid-stable enzyme in the stomach allows more time for the body's endogenous enzymes, that have evolved to specifically digest exorphins, to work. Continuing with that rationale, the following enzymes were selected for inclusion in ENZYMAID. The enzyme AFP was used as a first line of attack in the breakdown of proteins into small pieces. This enzyme functions in the stomach where it hydrolyzes large proteins into smaller peptides. The presence of a large amount of this enzyme is crucial to forcing the reaction to happen faster. The net result is that any exorphins that are going to form, will do so in the stomach, before they reach a point of absorption. & 2002 Published by Elsevier Science Ltd.
423
To assist with this digestion, large amounts of bromelain were also used. Bromelain was chosen over papain because research has suggested that bromelain has a wider specification and function than papain. Also, it is common knowledge that papain is naturally high in sulfites and the researchers attempted to limit any potential negative effects. Bromelain has been demonstrated to be an effective anti-inflammatory, which has also been implicated as important in austism (6). The CASE-GLUTENASE is a peptidase concentrate providing the next to last assistance with exorphin digestion. This enzyme has both high exo-peptidase as well as endo-peptidase activity and contains known DPPIV activity (Table 2). This provides an overall digestion profile suitable to any extraneous peptide development. Phytase was added for its ability to digest phytic acid, which is present in plants such as corn, rice, wheat, soybean, and other beans, etc. Phytic acid can negatively affect absorption of minerals such as zinc, calcium, magnesium, copper, manganese, and iron. Phytase supplementation results in greater bioavailability of these important minerals. It has been suggested that autistics are deficient in zinc, so the extension was to try to increase the bioavailability of dietary zinc. The only non-enzyme constituent of the blend is galactose. Galactose was included to function as a genomeceutical. Genomeceuticals are naturally occurring compounds, which are able to beneficially affect gene expression. The inclusion of galactose in the formulation is based on three different works. The first was the report by Smith et al. (7) that galactose can increase the expression of dipeptidyl peptidase-IV (DPPIV or CD26) in murine enterocytes. The second is based on the finding that adding glucosamine, could increase the level of transcription and translation of important genes (8). Both transcription and translation are important for a nutrient to be able to affect the levels of a protein made from the DNA. Wang's work demonstrates clearly, elegantly, and definitively, that the concept of genomeceuticals is sound. By adding glucosamine to the diet, not only did they get more RNA, but they got more protein as well. The corollary is that by adding galactose, more DPPIV can be made. This is the third work and is based on Brudnak's theory of genomeceuticals suggesting that because the addition of glucosamine to a diet can increase the expression of leptin (fat hormone) in the body, the addition of galactose may have a similar effect on DPPIV (9). Genomeceuticals describe nutrients which can affect the structure of a gene, how well the gene products (protein and sometimes RNA) work, and/or how much of it is made. That is to say, genomeceuticals can not just replace substances, which may be missing (e.g., an enzyme diminished by mutation), but actually alter the expression Medical Hypotheses (2002) 58(5), 422±428
424 Brudnak et al.
and functionality of gene products in, and resulting from, genomic multi-level nutrient-sensing pathways. The last category is where the inclusion of the galactose comes into play. In this case, the genomeceutical would be galactose. Galactose appears to be able to increase the expression of the Dipeptidyl peptise IV gene. This means adding galactose can increase the amount of DPPIV that is present. If in autistics, the situation is not that there is absolutely zero DPPIV made, due to a mutated gene or regulated element, but rather that the gene has been silenced or attenuated (down-regulated), then the addition of galactose has the potential to reverse or circumvent that. The caveat being that the gene needs to at least be functional and this is discussed below. Indeed, DPPIV has been suggested to be down-regulated in autistics and is currently being used as a diagnostic marker for the disease. However, that is not the only place DPPIV, and DPPIV-like enzymes, are expressed and may prove to be secondary to other sources. For instance, it is well-known that more than one of the probiotic species inhabiting the human GI tract expresses an analog capable of digesting exorphins via DPPIV analogs. It is equally well-known that there are more probiotics in the GI tract than there are cells in the body. Probiotics are a major contributor to the metabolic activities of the human host, including digestion of potentially offensive substances. Galactose is a known prebiotic (substance capable of selectively stimulating the growth of probiotics verses pathogenic organisms) (10,11). As such, the addition of galactose to the formula was a natural selection. This paper addresses the theory that foods are not just nutrients but bioactive substances. The data below are presented to support that hypotheses by determining the effectiveness of the rationale above as gauged by 13 markers assessed in 22 patients. The results of the application of ENZYMAID to autistics are presented below. Statistical analyses suggest that the enzyme formulation is both sound and effective for the treatment of autism spectrum disorders. Additionally, data are presented on the DPPIV enzyme analog levels of various endogenous enteric bacteria to support Brudnak's theory that probiotic bacteria can assist in alleviation of austistic parameters.
Study protocol and human subjects The study protocol was approved by a Medical Review Board. The legal guardians of all participants signed appropriate `informed consent' forms. The study was carried out with the highest standards of ethics. Initially, 46 patients between the ages of two and 21 were chosen for the study (12). Of those that started the study, 22 remained for the entire 12-week period. For data analysis, only those subjects who completed the entire course of therapy were included. All raw data are kept on file for public inspection at the Autism Research Institute (ARI), San Diego, CA. Enzymes All enzymes were manufactured by MAK Wood, Inc. (Grafton, WI). They were manufactured specially for Kirkman Laboratories (Lake Oswego, OR) to be free of milk, casein, wheat, gluten, gliadin, corn, soy, egg, yeast, sugar, starch, MSG, sterates, palmitates, artifical sweeteners, colors or flavors, preservatives, salicylates and other common allergens. Kirkman Laboratories kindly supplied all the enzymes. The following enzymes and activities were included: CASO-GLUTENASE 10,000 AU; Bromelain 230 BTU; Acid Fast Protease 100 SAPU; Lactase 330 LacU; Phytase 125 U; Galactose (as Genomeceutical) 100 mg. Dosing The enzymes were to be taken at the beginning of each meal (or early on into the meal). The capsules were swallowed whole or pulled apart and the contents mixed with the first few bits of food or in beverages. One-half to one capsule with each meal was suggested with a gradual increase. Assay of enzyme effect Parents, guardians, teachers, or therapists were asked to complete an SOS form every two weeks. They evaluated the 13 different parameters of function and behavior in the child. DPPIV assay
MATERIALS AND METHODS All reagents were processed and approved by the MAKTechTM process of material qualification which includes DNA fingerprint analysis, strain vertification, optimal growth conditions, etc. for bacterial cultures, endotoxin assay, sample purity, heavy metal analysis, and microbial plate counting. Whenever possible, all reagents are supported by published literature. Medical Hypotheses (2002) 58(5), 422±428
A modification of Tachi et al. (1992, Phytochemistry, 3: 3707±3709). Principle: Gly-Pro p-nitroanilide is hydrolyzed not Gly-Pro and p-nitroaniline by X-Prolyl dipeptidyl-aminopeptidase. The p-nitroaniline liberated from the substrate is measured at 405 nM. Unit Definition: One unit of enzyme activity is defined as the enzyme quanitity that liberates 1 mmole of p-nitroaniline from the substrate per minute. Reagents: Substrate solution: 0.45 nM Gly-Pro & 2002 Published by Elsevier Science Ltd.
Enzyme-based therapy for autism spectrum disorders
p-nitroanilide (in the form gly-Pro-p nitroanilide p-toluenesulfonate slat Sigma Chemical #G-2901) in 50 mM potassium phosphate buffer (pH 7.5). Standards solution: 1.448 mM p-nitroanaline solution in ddH2O. Stop reagent: 0.1 N hydrochloric acid. Standard preparation and assay: Pipet 1-, 2-, 3-, and 4-ml portions of p-nitroanilide solution into a series of 50 ml volumetric flasks, dilute each to volume with 50 mM phosphate buffer and mix. These dilutions contain 0.029, 0.058, 0.87, and 0.116 mmol/ml, respectively. Determine the absorbance of each dilution at 405 nM. Plot absorbance against mmol/ml of p-nitroanlinine. A straight line through the origin must be obtained. Divide mmol/ml by the absorbance of each dilution to obtain the factor (F) value (mmol/ml). Average the four values. Approximate value for F 0.10 mmol/ml. Sample preparation and assay: Dilute sample to a concentration of 0.005±0.001 units/ml with water. Add 0.5 ml of substrate solution to a test tube and pre-incubate for 5 minutes at 30 C. Add 0.1 ml of sample solution to the tube and incubate with agitation for 30 min. Sample blank to contain 0.1 ml of distilled water in place of sample solution. After 30 minutes incubation, add 0.5 ml of 0.1 N HCl (stop reagent) to tube. Determine absorbance at 405 nM (sample absorbance A1; sample blank absorbance A2). Calculation:
425
1.1 Sample solution volume (ml); 30 Incubation time (min); C the sample solution concentration (g/ml). Statistical analysis A Student's t-test for two groups was used for the statistical analysis. A Java Implementation by Bryan Lewis, Kent State University, 1997 (13). The significance is shown as the `P value' for the two groups. The two groups were considered to be Before and After treatment with enzymes. The analysis was run in two ways. The first is the column designated `Best Case' in Table 1. This column represents a comparison of before and after treatment assuming that all patients scored an initial zero for the before treatment score. The second is shown in the column designated `Worst Case' in Table 1. This column represents a comparison of before and after treatment assuming that all patients scored an initial marking equal to the 1±2 week period score. This was done to elevate the baseline. RESULTS Of the 46 patients that initially started the treatment, 17 dropped out during the first few weeks of the 12-week trial. Those that remained for the entire 12 weeks were included in the data analysis. For each of the 13 parameters, patients were assayed by the following 0±4 scale: 0 none; 1 possible; 2 moderate; 3 significant; 4 great. The results in Table 1 list the percentage of patients that scored between a moderate and significant improvement by a guardian.
X-prolyl dipeptidyl-aminopeptidase mg/g A1 ÿ A2 F 1:1 0:1 30 C A1 Absorbance of enzyme sample; A2 Absorbance of reagent blank; F p-nitroaniline factor (mmol/ml);
Table 1 Percentage of patients showing improvement Start
Eye contact Socialization Attention Mood Hyperactivity Anxiety/Compulsion Stimming Comprehension Speech Sound sensitivity Digestion Sleep Perseveration
0 0 0 0 0 0 0 0 0 0 0 0 0
Weeks %
P value
W1±2
W3±4
W5±6
W7±8
W9±10
W11±12
Best case
Worst case
37 42 40 36 31 20 27 40 27 17 35 23 33
47 67 54 52 31 41 38 45 41 17 50 36 38
43 71 63 57 50 46 27 58 53 18 56 43 44
56 76 59 60 75 47 38 55 47 42 50 50 50
67 81 73 56 75 41 31 50 38 25 56 64 39
67 90 68 59 80 60 50 63 44 50 50 57 53
8.08Eÿ04 7.66Eÿ04 6.05Eÿ04 5.03Eÿ04 0.002315 0.001284 8.75Eÿ04 4.92Eÿ04 6.77Eÿ04 0.004429 4.64Eÿ04 0.001454 5.38Eÿ04
0.01361 0.00348 0.00451 0.00237 0.01879 0.00397 0.03593 0.00947 0.0047 0.05748 0.00269 0.00671 0.01246
An observer scored 22 patients for the 13 parameters listed. The column `Start' is the total initial score as a percentage for improvement for each parameter. These were not actually scored and are assumed to be zero for each. Scores were taken every two weeks, totaled, and indicated as a percentage of the total. `Best case' P value represents the signi®cance based on a Student's t-test using the initial `Start' of zero. The `Worst case' P value represents the signi®cance based on a Student's t-test using the initial Week 1±2 reading as the baseline. The two groups analyzed for both Best-case and Worst-case were the initial Start vs. W 1±2, 3±4, 5±6, 7±8, 9±10, 11±12 percentages. P values < 0.05 are considered signi®cant.
& 2002 Published by Elsevier Science Ltd.
Medical Hypotheses (2002) 58(5), 422±428
426 Brudnak et al.
Patients who stopped the study in the first few weeks, reported a number of reasons including: need to start other therapies; family illness; personal issues; changed physicians; did not tolerate the taste of the supplement and/or no change in symptoms. Additionally, there were six patients who had `adverse responses' including hyperactivity/increased aggression, increased stimming (one), diarrhea/loose stools (two), provocation of red ears and cheeks (one), increased hunger (one), and stopped eating (one). Approximately 40% (data not shown) of the participants were following a gluten-free and casein-free diet but all were instructed not to change their diet during the treatment. An overwhelmingly positive trend is seen for each of the parameters. The two greatest improvements were seen in socialization and hyperactivity with 90% and 80% improvements, respectively. The lowest improvements were for stimming, speech, and sound sensitivity, each scoring 50%. Statistical analysis (Student's t-test) using zero for the base-line in the `Best case' column indicated that all scores were significant. The P values ranged from 0.004429, for sound sensitivity, to 8.75 10ÿ04 for stimming. Because the zero point was an assumed number, a further round of analysis was pursued. Here, as indicated in column `Worst case' the baseline was chosen to be the same as that of the initial 1±2 week rating. All parameters except for one, sound sensitivity, were again shown to be statistically significant. Due to the addition of galactose in EnzymAidTM and its known prebiotic effect on lactic acid bacteria, an analysis was done on several products that are commonly used to support GI function. Each product ranged from a single lactobacillus strain (Acidophilus PlusTM) to a mixed culture of several organisms (L. acidophilus, L. rhamnosus, Bifidobacterium BB12, S. thermophilus, L. bulgaricus). Table 2 presents the results for each. As can be seen, there was detectable DPPIV activity in all the products tested. Additionally, a newly developed (Via MAK TechTM) DPPIV Table 2 DPPIV enzyme activity of various gastrointestinal bacteria Enzyme or strain TM
MAK Tech Rhamnosus MAK TechTM DPPIV Bifido-ComplexTM Acidophilus PlusTM Pro-Bio GoldTM Multi-Flora SpectrumTM Pro-Culture GoldTM Pro-Bio Balance Flora maintenance
DPPIV level (mg/g) 1.07 4.28 0.60 0.60 3.55 3.04 1.99 0.77 0.39
Several common probiotics were assayed for DPPIV activity according to the modi®ed procedure outlined in the Results. The range of activity was from 0.39 to 4.28 mg/g for the bacteria and enzymes and was 0 mg/g for the negative (buffer) control (data not shown). Medical Hypotheses (2002) 58(5), 422±428
enzyme (MAK Tech DPPIV) was tested and showed the highest level of activity. Further experiments are planned to test this product singly in a similar treatment protocol for ASD. DISCUSSION There were a number of special conditions experienced with the study, which may have contributed to the dropout rate. Most of those that dropped out did so early on in the study. The majority of those that dropped out had reasons other than negative effects of the study on the patient. Also, the families were instructed not to start any new therapies while in the study. This proved to be an unexpected problem as several families apparently had a strong desire to try other approaches. This was an interesting development and will need to be considered when designing follow-up studies. Despite the dropout rate, it is believed that the ENZYMAID formulation was generally well-tolerated as only a few respondents reported adverse effects which may be directly attributed to the enzyme formulation. While other studies using enzymes have been undertaken, a Medline search revealed that none have been reported in the literature at this time. While the present study is similar in structure, the novelty of the enzyme formulation is pronounced. In addition to adding several new enzymes, this is the first reporting of the therapeutic potential of genomeceuticals. The functioning of galactose is believed to be at two levels. The first is as a genomeceutical. Here, it is believed to be increasing the gut expression of the DPPIV gene. This increased expression allows for a greater level of DPPIV enzyme in the enterocytes, promoting the more thorough breakdown of any exorphins produced by the proteases. For a review of this, the reader is referred elsewhere (9). The second and equally interesting possibility is that galactose is serving as a fuel source (prebiotic) of the beneficial microflora (i.e., probiotics) in the gut. This is important because the probiotic organisms themselves contain enzymes capable of breaking down such exorphins. Varmanen et al. (2000, pp. 146±154) recently showed that probiotic organisms, currently utilized as health supplements, contain analogs of the Dipeptidyl peptidase IV enzyme (e.g., PepX14) which is known to be able to digest exorphins. With over 1011 microorganisms in the gut, their contribution of enzymatic activity can far exceed that of the enterocytes. It is well documented that galactose is a prebiotic (i.e., stimulates growth of probiotics) and can increase the number of probiotics in the gut (15±18). The importance of the gut-associated microflora can not be overstated in discussions of enzyme therapy theory. & 2002 Published by Elsevier Science Ltd.
Enzyme-based therapy for autism spectrum disorders
Increasingly, with the use of mono- or bi-strain products, there are reports of a reduction of clinical effects seen over time. For instance, in the case of autism, a child that pre-treatment had either diarrhea, or some other gastrointestinal distress, improves dramatically when first supplemented with those products. The child starts producing well-formed stools on a regular basis. However, in some cases, the child can undergo a reversion to the previous distressed state. Two approaches seem to have found a resolution for this consternating phenomenon. First, is the `pulse' method. Here, when the clinical effect begins to change, the physician can cease the probiotic supplementation for a period of time, followed by a repeat application. The body appears to be `resetting' itself after cessation of the therapy. During this period, changes are occurring in the gastrointestinal tract that may be making the local environment more hospitable to the probiotics. Additionally, there are certainly changes in the immune functioning of the gut. It is well established that the intestines are a major source of immune competence. It is equally well established that there are genetic elements that respond to heavy metals, and in particular to mercury. What may be happening is that upon initial exposure to the probiotics, the gut-associated immune system is mobilized and may be attacking the probiotics. After withdrawal of the probiotics, the immune reaction tapers off until any subsequent exposure. This is not a bad thing and can be used efficaciously by pulsing the probiotics. Also, a variation on that theme can be employed. That is to say, instead of withdrawing all probiotics for any period of time, a substitution of another product, which contains different organisms would seem the logical next step. For instance, if a product that just L. rhamnosus is being used and the above condition presents, a product of just Bifidobacterium could be employed. If a single strain of Bifidobacterium is used, and again the same situation occurs, then a third product with several strains could be administered. It most certainly is going to have to go on a case-by-case basis, according to the physician's observations. In the foreseeable future, there will be products designed specifically to address the issue in this way. This `probiotic rotation' will be applicable in a variety of situations. For these reasons, it is expected that the inclusion of a probiotic regimen or pulsing or rotating strains may be of great benefit. It is theorized here that what is happening is both the level of DPPIV in the gut is increased and the level of DPPIV-type activity, contributed by a large increase in the bacterial flora, is also increased. The net result is that due to the rapid formation of any exorphins, from the high level of acid-stable protease, the levels of absorbed exorphins dropped below a threshold required for & 2002 Published by Elsevier Science Ltd.
427
manifestation of the parameters measured. To test this hypothesis, further studies are planned to assay urinary polypeptide levels on retained samples both before and after treatment (11). Also, future studies will look at blood peptide levels both before and after treatment. Additionally, further studies are planned to determine the genomeceutical contribution to the formula. Again, increasing levels of galactose will be assessed for against all the parameters measured in the present study, along with blood levels of exorphins and stool levels of various probiotic species. The only known contraindication, for the use of galactose, is in a fairly rare metabolic disorder, galactosemia. However, the amount of galactose in the formula, 100 mg, is far less than that present upon full digestion of a glass of milk. A typical glass of milk contains around 12 grams of lactose, which, when broken down, would contribute 6 grams of galactose to the diet. That is approximately 60 times the amount in a capsule of ENZYMAID. Data has been presented merely to support the hypothesis that enzymes can have a beneficial therapeutic effect on ASD. Additionally, the theory of genomeceuticals and their potential beneficial effects as a part of a treatment protocol for ASD was supported. Finally, novel data on the DPPIV level of various probiotics were presented in support of their use in an enzyme therapy-based model for the treatment of ASD. As an uncontrolled pilot clinical study the present work has tremendous value for the support of these hypotheses. Not only did the overwhelming majority of parameters measured show a significant benefit of the enzyme blend, but there were also very few associated negative reactions. Also, this is the first time a genomeceutical approach has been therapeutically applied, not to just autism, but to any disease state. While the etiology of autism and the ASDs still remains elusive, clearly the present study advances the knowledge base for efficacious treatment protocols. Several theories have been addressed in this work. First, the idea that foods, such as enzymes and monosaccharides, are really bioactive substances and not just nutrients has been addressed and supported by their measurable effects. Further the theories of genomeceuticals, exorphin theory of the etiology of autism also appear to be substantially supported. Further the theories that probiotic organisms can significantly contribute to not just the overall metabolic activities of an organism but also assist in the alleviation of disease (e.g., autistism) symptoms via their contribution of enzymatic activities has also been explored. Taken as a whole, the entire area of genomeceuticals and their effects on ASD appears to deserve a closer look and may indeed prove to eventually be the breakthrough those in the area have been searching for. Medical Hypotheses (2002) 58(5), 422±428
428 Brudnak et al.
ACKNOWLEDGEMENTS Special thanks go to Ronda Dukes for data collection and tabulation. Vector Laboratories for technical assistance. The Autism Research Institute for meaningful discussions and data retention.
REFERENCES 1. Dr. Bernard Rimland. Personal Communication. Autism Research Institute. San Diego, CA. 2. Turner M., Barnby G., Bailey A. Genetic clues to the biological basis of autism. Mol Med Today 2000 Jun; 6(6): 238±244. 3. Haileselassie S. S., Lee B. H., Gibbs B. F. Purification and identification of potentially bioactive peptides from enzymemodified cheese. J Dairy Sci 1999; 82: 1612±1617. 4. Reichelt K. L., Hole K., Hamberger A., Saelid G., Edminson P. D., Braestrup C. B., Lingjaerde O., Ledaal P., Orbeck H. Biologically active peptide-containing fractions in schizophrenia and childhood autism. Adv Biochem Psychopharmacol 1981; 28: 627±643. 5. Drs. Alan Friedman and Jon Pangborn. Personal Communication. 6. Brudnak M. A. Application of genomeceuticals to the molecular and immunological aspects of autism. Medical Hypotheses. In Press. 7. Smith M. W., James P. S., Peacock M. A. Galactose effects on enterocyte differentiation in the mouse jejunum. Biochim Biophys Acta 1991 Jul 10; 1093(2±3): 1446.
Medical Hypotheses (2002) 58(5), 422±428
8. Wang J., Liu R., Hawkins M., Barzilai N., Rossetti L. A nutrient-sensing pathway regulates leptin gene expression in muscle and fat. Nature 1998; 393: 684±688. 9. Brudnak M. A. Genomic multi-level nutrient-sensing pathways. Medical Hypotheses 2001 Feb; 56(2): 194±199. 10. Langhendries J. P., Detry J., Van Hees J., Lamboray J. M., Darimont J., Mozin M. J., Secretin M. C., Senterre J. Effect of a fermented infant formula containing viable bifidobacteria on the fecal flora composition and pH of healthy full-term infants. J Pediatr Gastroenterol Nutr 1995 Aug; 21(2): 177±181. 11. Mozzi F., Rollan G., de Giori G. S., Font de Valdez G. Effect of galactose and glucose on the exopolysaccharide production and the activities of biosynthetic enzymes in Lactobacillus casei CRL 87. J Appl Microbiol 2001 Jul; 91(1): 160±167. 12. Kirkman Laboratories. Personal communications. Wilsonville, OR. 13. http://nimitz.mcs.kent.edu/blewis/stat/tTest.html 14. Varmanen P., Savijoki K., Avall S., Palva A., Tynkkynen S. 54X-prolyl dipeptidyl aminopeptidase gene (pepX) is part of the glnRA operon in Lactobacillus rhamnosus. J Bacteriol 2000 Jan; 182(1): 146±154. 15. Hove H., Norgaard H., Mortensen P. B. Lactic acid bacteria and the human gastrointestinal tract. Eur J Clin Nutr 1999 May; 53(5): 339±350. Review. 16. Hove H., Nordgaard-Andersen I., Mortensen P. B. Effect of lactic acid bacteria on the intestinal production of lactate and short-chain fatty acids, and the absorption of lactose. Am J Clin Nutr 1994 Jan; 59(1): 74±79. 17. Mortensen P. B., Hove H., Clausen M. R., Holtug K. Fermentation to short-chain fatty acids and lactate in human faecal batch cultures. Intra- and inter-individual variations versus variations caused by changes in fermented saccharides. Scand J Gastroenterol 1991 Dec; 26(12): 1285±1294.
& 2002 Published by Elsevier Science Ltd.