clinical study effectiveness of methylcobalamin and folinic ...effectiveness of methylcobalamin and...

Hindawi Publishing CorporationAutism Research and TreatmentVolume 2013 Article ID 609705 9 pageshttpdxdoiorg1011552013609705

Clinical StudyEffectiveness of Methylcobalamin and Folinic Acid Treatment onAdaptive Behavior in Children with Autistic Disorder Is Relatedto Glutathione Redox Status

Richard E Frye1 Stepan Melnyk1 George Fuchs1 Tyra Reid1

Stefanie Jernigan1 Oleksandra Pavliv1 Amanda Hubanks1 David W Gaylor2

Laura Walters1 and S Jill James1

1 Department of Pediatrics Arkansas Childrenrsquos Hospital Research Institute University of Arkansas for Medical SciencesLittle Rock AR 72202 USA

2Department of Biostatistics Arkansas Childrenrsquos Hospital Research Institute University of Arkansas for Medical SciencesLittle Rock AR 72202 USA

Correspondence should be addressed to Richard E Frye refryeuamsedu

Received 15 July 2013 Accepted 4 September 2013

Academic Editor Klaus-Peter Ossenkopp

Copyright copy 2013 Richard E Frye et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Treatments targeting metabolic abnormalities in children with autism are limited Previously we reported that a nutritionaltreatment significantly improved glutathione metabolism in children with autistic disorder In this study we evaluated changes inadaptive behaviors in this cohort and determined whether such changes are related to changes in glutathione metabolism Thirty-seven children diagnosed with autistic disorder and abnormal glutathione and methylation metabolism were treated with twiceweekly 75 120583gKgmethylcobalamin and twice daily 400120583g folinic acid for 3months in an open-label fashionTheVineland AdaptiveBehavior Scale (VABS) and glutathione redox metabolites were measured at baseline and at the end of the treatment period Overthe treatment period all VABS subscales significantly improved with an average effect size of 059 and an average improvement inskills of 77 months A greater improvement in glutathione redox status was associated with a greater improvement in expressivecommunication personal and domestic daily living skills and interpersonal play-leisure and coping social skills Age gender andhistory of regression did not influence treatment responseThe significant behavioral improvements observed and the relationshipbetween these improvements to glutathione redox status suggest that nutritional interventions targeting redox metabolism maybenefit some children with autism

1 Introduction

Autism is a neurodevelopmental disorder characterized bysignificant impairment in reciprocal social interaction andcommunication as well as restricted interests and repetitivebehaviors An estimated 1 of 88 individuals in the UnitedStates is affected with an autism spectrum disorder (ASD)[1] Although several genetic syndromes are associated withASD all of these genetic syndromes together only accountfor a minority of ASD cases [2] Other areas of novel researchhave concentrated on systemic physiological abnormalitiessuch as mitochondrial dysfunction [3 4] oxidative stress [5ndash7] and inflammationimmune dysregulation [8ndash10] These

novel areas of research have substantially grown over the lastdecade [11] These emerging areas of research have provideda new understanding of the diverse mechanisms involved inASD and have promoted the idea that the autism spectrum iscomposed of several subgroups or endophenotypes [12]

Several lines of evidence support the notion of anASD endophenotype with abnormal redox and methylationmetabolism In two case-control studies we reported thatredox and methylation metabolism in children diagnosedwith autism was abnormal compared to that of unaffectedcontrol children [6 13] Briefly the mean ratio of plasmaS-adenosyl methionine (SAM) to S-adenosyl homocysteine(SAH) a reflection of cellular methylation capacity was

2 Autism Research and Treatment

significantly reduced Also the mean concentration of freereduced glutathione (GSH) the major intracellular antiox-idant and mechanism for detoxification was also signifi-cantly decreased and the oxidized disulfide form of glu-tathione (GSSG) was significantly increased This imbalancein reduced as compared to oxidized glutathione resulted ina 2-fold reduction in the GSHGSSG redox ratio an indexknown as the glutathione redox status Several precursorsfor methylation and glutathione synthesis were also lowerin the autistic children suggesting an insufficiency in theproduction of methylation and glutathione metabolites atleast in a subgroup of ASD children In addition we havealso demonstrated that GSH is significantly reduced in thecytosol and isolated mitochondria from lymphoblastoid celllines and fresh peripheral blood mononuclear cells derivedfrom children with autistic disorder (AD) [14 15]

The glutathione provides the essential intracellular reduc-ing environment required for normal immune functiondetoxification capacity redox-sensitive enzyme activity andmembrane redox signaling [16ndash20] Oxidative stress occurswhen antioxidant defense mechanisms fail to counterbal-ance reactive oxygen species generated from endogenousoxidative metabolism or from prooxidant environmentalexposuresThe notion that abnormal glutathionemetabolismis associated with oxidative cellular damage is consistent withstudies from our laboratory which demonstrate an increasein oxidative damage to protein and DNA in peripheralblood mononuclear cells and postmortem brain derivedfrom ASD individuals [15 21] Several recent reviews andresearch studies also lend support to the hypothesis that redoximbalance and oxidative stress may be a contributing factorto autism pathophysiology [6 22]

These findings have particular clinical relevance sinceabnormalities in redox metabolism are potentially amenableto treatment Glutathione is a tripeptide of cysteine glycineand glutamate that is synthesized de novo in all cellsGlutathione synthesis is tightly connected to methylationmetabolism where it derives its cysteine precursor As bothmethylation and glutathionemetabolism have been shown tobe abnormal in children with ASD and due to the fact thatabnormal methylation metabolism will result in a depletionof the precursors needed for glutathione production wepreviously conducted an open-label intervention trial wheretargeted metabolic cofactors for methylation and glutathionemetabolismwere provided Specifically in our previous studywe determine if glutathione redox status could be improvedwith a three-month intervention of subcutaneously injectedmethylcobalamin and oral folinic acid in children with ADand metabolic evidence of abnormal methylation and glu-tathione metabolism [23] Although we measured adaptivebehavior using the Vineland Adaptive Behavior Scale (VABS)during the trial this behavioral data has not been previouslyreportedWe now report the change in communication dailyliving and social skills domains of the VABS associated withthe three-month intervention as well as investigate whetherimprovement in glutathione redox status was related to anychange in VABS behavior scores

Screened and met inclusion criteria

N = 65

Entered clinical trialN = 48

Normal metabolic profile N = 17

Lost to follow-up N = 4

Started treatmentN = 44

Discontinued treatmentUncomfortable with injections N = 2

Hyperactivity and reduced sleep N = 2Completed treatment

N = 40

Did not complete second behavioral evaluation N = 3

Behavioral data for analysis N = 37

Figure 1 Flowchart of patients who met inclusion criteria for thestudy

2 Methods and Materials

21 Participants Forty-eight children with ADwere enrolledin an open-label trial to test whether a three-month treatmentof methylcobalamin (methyl B12) and folinic acid wouldimprove glutathione synthesis as well as adaptive behaviorInclusion criteria were a diagnosis of AD as defined by theDiagnostic and Statistical Manual of Mental Disorders FourthEdition (DSM-IV) criteria (2990) and a Childhood AutismRating Scales (CARS) score greater than 30 Exclusion cri-teria included Aspergerrsquos disorder pervasive developmentaldisorder-not otherwise specified (PDD-NOS) genetic disor-ders epilepsy severe gastrointestinal symptoms recent infec-tions andor current use of high-dose vitamin or mineralsupplements (ie above the recommend daily allowance)Metabolic and behavioral assessments were conducted atbaseline and at the end of the 3-month intervention A nursecontacted each family every two weeks to ensure complianceand monitor adverse effects

Figure 1 provides a flowchart of the patients who met cri-teria for possible inclusion in the study In order to determineif abnormal redox metabolism could be improved with theintervention we selected children with abnormal glutathioneredox status and methylation capacity for entry into thestudy Normal glutathione redox status and methylationcapacity were defined by measuring these same metabolitesin typically developing control children Sixty-five autisticchildrenmet the inclusion criteria and were initially screenedfor metabolic evidence of abnormal methylation capac-ity (SAMSAH lt 30) and glutathione redox metabolism(GSHGSSG lt 60) Of these 48 (75) children were foundto have both abnormal methylation capacity and glutathioneredox metabolism thus meeting metabolic qualifications for

Autism Research and Treatment 3

inclusion The remaining 17 (25) were excluded becausetheir baseline metabolic profile was within normal rangeFour children dropped out during the study and another fourchildren were lost to follow-up Two children dropped outof the study because the parents were uncomfortable givingthe methylcobalamin injections and 2 children dropped outbecause of hyperactivity and reduced sleep For the four thatwere lost to follow-up every attempt was made to contactthe family but they either did not response to such requestsor moved out of the area and could not return for follow-up Of the 40 remaining children who completed the studythree of the parents did not complete the second behavioralevaluation leaving 37 children who completed both theintervention and the behavioral evaluation Characteristics ofthe children that completed the study are given in Table 1Adverse effects were minimal with occasional reports ofhyperactivity as summarized in our previous study [23] andbelow All parents signed informed consent approved by theInstitutional Review Board at the University of Arkansas forMedical Sciences

22 Intervention Methylcobalamin was provided as a sterileinjectable liquid from Hopewell Compounding Pharmacy(Hopewell NJ) Tuberculin syringes fitted with a 14 inch 31gauge needle were prefilled with 75 120583gKg methylcobalaminand prepared individually based on each childrsquos weight Ademonstration and instructions for the sterile subcutaneousinjection of methylcobalamin in the fatty tissue of the but-tocks were given to all parents Parents were instructed to givemethylcobalamin every third day for 3 months Folinic acid(400120583g) obtained from Custom Compounding (Little RockAR) was administered orally twice each day as a powdermixed with a convenient food

The subcutaneous injectable route of administration ofmethylcobalaminwas selected based onpreliminary observa-tions that it resulted in improvement in speech and cognition[13] and the potential for support of methionine synthaseactivity under conditions of oxidative stress by substitutingfor the oxidized inactive form of coenzyme B12 [Cob (II)]Folinic acid (5-formyltetrahydrofolate) was selected in lightof the fact that it is absorbed as the reduced metaboliteis rapidly polyglutamated and is more readily available forfolate-dependent reactions than the synthetic vitamin formof folic acid In our previous study we demonstrated that thistreatment significantly improved GSH and GSSG concentra-tions and glutathione redox status (ie the GSHGSSG ratio)[23]

23 Glutathione Measurement Fasting blood samples werecollected into EDTA-Vacutainer tubes and immediatelychilled on ice before centrifuging at 4000timesg for 10 minutesat 4∘C To prevent metabolite interconversion the ice-coldsamples were centrifuged within 15 minutes of the bloodcollection and the plasma stored at minus80∘C until high-pressureliquid chromatography quantification within 2 weeks afterreceipt Details of the methodology for high-pressure liquidchromatographywith electrochemical detection andmetabo-lite quantitation have been previously described [24 25]

Table 1 Characteristics of the 37 participants with autistic disorderwho completed the trial

Gender ( male) 81Age (mean plusmn standard deviation) 51 plusmn 14 yearsRegression ( regression) 35Childhood Autism Rating Scale score(mean plusmn standard deviation) 392 plusmn 78

24 Behavioral Evaluation The VABS (1st Edition) is avalidated parent interview that provides a numerical scorefor adaptive functioning in the domains of communicationsocialization and daily living skills which has been shown tohave good reliability and validity [26] Nine subscales repre-sented these three domains specifically receptive expressiveand writing communication personal domestic and com-munity daily living skills and interpersonal relations play-leisure and coping skills Raw scores were used in the statis-tical analysis while age equivalent scores are used to illustratethemagnitude of the changes in behavioral development overthe treatment period and domain scaled scores are providedas a comparison of the participant sample to the generalpopulation Higher raw age equivalent and scaled scoresrepresented better performance on thesemeasuresTheVABSwas administered by a trained nurse at the baseline visit andagain at the end of the treatment period

25 Statistics Mixed-effects regression models were con-ducted via SAS version 93 (Cary NC) ldquoglmmixrdquo procedure[27] Raw scores were used in the statistical analysis to repre-sent the change in behavior over the intervention period Asimilar approach was used in our recent study on the effectof sapropterin treatment in children with ASD [28] Rawscores were used due to the fact that transformation of rawscores into scaled scores is not statistically valid to followchanges in adaptive behavior especially when examining themoderating effect of a biomarker on the change in behaviorand especially when there is no reference control group Forexample many children with ASD have slow or stagnantdevelopment so their development may not change overa typical three-month interval With an intervention theirdevelopment may approach the rate of a typically developingchild However a child whose development changes overthe three-month period at a rate equal to that of a typicallydeveloping child would have a stable unchanging scaledscore Thus in such a case the scaled score would appearto reflect no developmental gains over the treatment periodeven though the child was making developmental gainsSince the null hypothesis is that there is no change overthe treatment period an unchanging scaled score wouldnot be statistically significant even though it would indicatedevelopmental progress

The first statistical model included a linear effect of timeon raw scores A random intercept was used to account foreach individualrsquos symptom level Two-tailed alpha of 005was used in all analyses In these initial models we testedwhether age gender andor a history of regression influenced


Table 2 Age equivalent scores from the Vineland Adaptive Behavior Scales at baseline before and after 3-month intervention withmethylcobalamin and folinic acid The change in age equivalent scores with 95 confidence interval (CI) is also givenThe overall average ofall subscales is also given in the last row of the table

Vineland subscale Baseline age equivalent (months)(mean plusmn SE)

Postintervention age equivalent(months) (mean plusmn SE)

Change in age equivalent(months) (mean 95 CI)

Receptive language 231 plusmn 18 314 plusmn 34 83 (29 137)Expressive language 206 plusmn 19 275 plusmn 29 60 (33 94)Written language 405 plusmn 38 467 plusmn 40 62 (34 90)Personal skills 305 plusmn 23 405 plusmn 38 100 (38 162)Domestic skills 303 plusmn 41 393 plusmn 59 90 (minus14 194)Community skills 329 plusmn 29 361 plusmn 38 20 (minus30 69)Interpersonal skills 187 plusmn 27 241 plusmn 39 54 (00 109)Playleisure skills 220 plusmn 45 340 plusmn 41 120 (41 196)Coping skills 258 plusmn 25 343 plusmn 40 115 (49 180)Overall skills 266 plusmn 23 343 plusmn 36 77 (34 120)

the change in the outcome measure To illustrate the relativestrength of the effect we provide the Cohenrsquos 119889 effect sizefor each significant change in outcome variables Next wedetermined whether baseline andor change in glutathioneredox status (ie the free reduced-to-oxidized glutathioneratio) moderated changes in the outcome variables The glu-tathione redox status was investigated because it significantlychanged with treatment in our previous study [23] We didnot investigate the moderating effect of other metabolitesas multiple statistical tests would have increased our type Ierror rate We conducted two sets of analyses one examinedthe moderating effect of the overall glutathione redox statusand another that examined the moderating effect of thechange in glutathione redox status over the treatment periodThe former analysis was conducted to determine whetherglutathione redox status in general was related to VABSscores while the latter analysis was conducted to determine ifthe change in glutathione status was related to the change inVABS scores over the treatment period Age gender andorregression factors were included in the analyses examiningthe moderating effect of glutathione if they were significantin the initial model

3 Results

31 Behavior Change over the Intervention Period The VABSwas used to examine the impact of the intervention on mea-sures of core behaviors associated with autism The changesin VABS subscales with treatment were not influenced bygender age or history of regression and neither gender norhistory of regression was related to overall VABS subscalescores Age was significantly related to overall raw scores forall subscales except for interpersonal and coping skills

Significant increases were found in all VABS subscalescores including receptive (119865(1 36) = 1190 119875 = 0001 119889 =059) expressive (119865(1 36) = 3280 119875 lt 00001 119889 = 097)and written (119865(1 36) = 1107 119875 lt 0005 119889 = 056) com-munication skills personal (119865(1 36) = 1469 119875 lt 00005119889 = 065) domestic (119865(1 36) = 485 119875 lt 005 119889 = 037) and

10203040506070

Actu

al ag

e

Rece

ptiv

e lan

guag

e

Expr

essiv

e lan

guag

e

Writ

ten

lang

uage

Pers

onal

skill

s

Dom

estic

skill

s

Com

mun

ity sk

ills

Inte

rper

sona

l ski

lls

Play

-leisu

re sk

ills

Cop

ing

skill

s

Ove

rall

skill

s

Age

(mon

ths)

Figure 2 Histogram of actual age and age equivalents for theVineland Adaptive Behavior Scale subscales Error bars depictstandard error

community (119865(1 36) = 951 119875 lt 0005 119889 = 052) daily livingskills and interpersonal (119865(1 36) = 645 119875 lt 005 119889 = 043)play-leisure (119865(1 36) = 1236 119875 = 0001 119889 = 059) andcoping (119865(1 36) = 1568 119875 = 00005 119889 = 066) social skills

To illustrate the magnitude of the change in develop-ment with the intervention Table 2 and Figure 2 present theaverage age equivalent for all subscales before and after theintervention Table 2 also demonstrates the change in ageequivalent with the intervention including confidence inter-vals As evident from this table many subscales particularlythose with larger effect sizes demonstrate large gains withthe three-month intervention This is particularly true forthe communication domain where skills improved between60 and 83 months on average and in the social skillsdomain where skills improved between 54 and 120 monthson average over a three-month period

Baseline pretreatment scores of the individual domainsand overall composite of the VABS (mean (SD) commu-nication 654 (130) daily living skills 664 (134) socialskills 673 (93) adaptive behavioral composite 661 (93))were found to be within the range of previously publishedVABS scores for children with autistic disorder [15 29]Following intervention all individual domains and the overall


minus4minus20

0

4

8

12

0 20 40 60Change in expressive communication

raw score

Chan

ge in

free

GSH

GSS

G

(a)

minus4minus20

0

4

8

12

0 20 40 60Change in personal daily living skills

raw score

(b)

minus4

0

4

8

12

minus20 0 20 40 60Change in domestic daily living skills

raw score

(c)

0

4

8

12

0 20 40 60Change in interpersonal social skills

minus4minus20

raw score

Chan

ge in

free

GSH

GSS

G

(d)

0

4

8

12

0 20 40 60minus4

minus20Change in play-leisure social skills

raw score

(e)

0

4

8

12

0 20 40 60Change in coping social skills

minus4minus20

raw score

(f)

Figure 3 Significant relationships between the change in the glutathione redox status (reduced-to-oxidized glutathione ratio) and changein subscales of the Vineland Adaptive Behavior Scale (VABS) subscales An improvement in glutathione redox status was associated withimprovement in (a) expressive communication (119865(1 33) = 966 119875 lt 001) (b) personal daily living skills (119865(1 34) = 1284 119875 = 0001) (c)domestic daily living skills (119865(1 34) = 469 119875 lt 005) (d) interpersonal social skills (119865(1 34) = 1047 119875 lt 0005) (e) play-leisure socialskills (119865(1 34) = 816 119875 lt 001) and (f) coping social skills (119865(134) = 609 119875 lt 005) The moderating effect of glutathione redox status onthe VABS subscale is provided in the graph as a representation of the relationship between the variables Since the linear models examiningthe moderating effect of glutathione redox status on the VABS subscale takes into account age simple correlation coefficients would not beaccurate for inclusion in the graphs

composite of the VABS markedly increased (Mean (SD)communication 721 (157) daily living skills 760 (180) socialskills 756 (165) adaptive behavioral composite 738 (174))

32 Relation between Behavior and Glutathione Redox StatusThe overall glutathione redox status was not related toVABS subscales indicating that overall development did notappear to be related to overall glutathione redox status Inorder to determine whether changes in VABS subscales wererelated to changes in glutathione redox status we examinedwhether the glutathione redox status moderated the changein the VABS subscales We found that glutathione redoxstatus moderated expressive communication (119865(1 33) =966 119875 lt 001) personal (119865(1 34) = 1284 119875 = 0001)and domestic (119865(1 34) = 469 119875 lt 005) daily living skillsand interpersonal (119865(1 34) = 1047 119875 lt 0005) play-leisure(119865(1 34) = 816 119875 lt 001) and coping (119865(1 34) = 609119875 lt 005) social skills such that a greater increase in the glu-tathione redox status (ie greater improvement in glu-tathione) was associatedwith a greater improvement inVABSsubscale scores (see Figure 3)

33 Adverse Effects The adverse effect rates are outlined inTable 3 Out of the forty-four children that started the studyand were not lost to follow-up 72 reported no adverseeffects of the treatment For the four children that droppedout of the study two children (5) dropped out becausethe parents were uncomfortable giving the methylcobalamin

Table 3 Adverse effects reported of intervention with methylcobal-amin and folinic acid All 44 childrenwho entered the study butwerenot lost to follow-up were included

Adverse effect (119873)Hyperactivity 14 (6)Reduced sleep 7 (3)Discomfort with injections 5 (2)Insomnia 2 (1)Impulsivity 2 (1)Irritability 2 (1)

injections and 2 children (5) dropped out because ofhyperactivity and reduced sleep Of the 4 families whoreported hyperactivity but did not drop out of the study thehyperactivity resolved with decreasing the folinic acid dose to400 120583g per day

4 Discussion

This intervention trial was undertaken to determine whethera nutritional intervention consisting of treatment withmetabolic precursors for methionine and glutathione syn-thesis would improve autism symptoms in children withAD Children meeting the inclusion criteria were initiallyscreened for metabolic evidence of abnormal methylationcapacity and glutathione redox status as an indication of


potential benefit from the interventionOur goalwas to deter-mine whether nutritional support with precursors for theabnormalmetabolic pathways could result in improvement inASD associated behavior as measured by the VABS Overallsignificant improvement was noted on all subscales of theVABS during the three-month intervention In additionimprovement on several VABS subscales including thoserelated to communication and social skills was related tothe improvement in glutathione redox status Overall thisstudy provides preliminary evidence that targeted nutritionalsupport in children with AD who have metabolic abnor-malities in glutathione and methylation pathways may beassociatedwith improvements in certain behaviors associatedwith ASD at least in a subset of AD children In addition alow rate of adverse effects with methylcobalamin and folinicacid treatment indicates that these supplements are safe andwarrants further study as an intervention

41 Behavioral Improvements with Intervention A simple andsafe nutritional intervention of methylcobalamin and folinicacid resulted in significant increases in VABS scores for alldomains including daily living social and communicationskills with an average effect size of 059 which is in themedium-to-large range In fact even the smallest effect size(domestic daily living skills 119889 = 037) was in the mediumrange and the largest effect size (expressive language skills119889 = 097) was in the very large range When consideringchanges in age equivalent skills over the treatment period itwas found that skills improved an average of 77 months overthe three-month treatment period with the lower end of therange being an average gain of 20 months for communitydaily living skills to the higher end of the range being anaverage gain of 120 months improvement for play-leisureskillsHowever whenwe examine the overall change in scaledscores across the three-month treatment period we find thaton average the scaled score for the individual domains andthe adaptive behavioral composite at the end of the treatmentwere still well below the average for typically developingchildren In fact these scaled scores were on average at thevery lower end of normal range Thus although significantlyimproved the behavior scores after treatment remained onaverage at the very low end of normal and for many childrenbelow the normal range This suggests that further therapymay be required to promote continued improvement throughcontinued nutritional treatment with metabolic precursorsover a prolonged time period the addition of intensivebehavioral and educational therapy or both

42The Relationship betweenMetabolic and Behavioral Chan-ges Change in glutathione redox status with treatment butnot baseline glutathione redox status prior to treatmentwas found to be related to improvements in several VABSsubscales including all of the subscales in the social skillsdomain two of three subscales in the daily living skillsdomain and one of the subscales in the communicationdomain Interestingly many of the VABS subscales that wererelated to glutathione redox status were also subscales thatdemonstrated large effect sizes and larger age equivalent

gains (eg personal daily living skills and play-leisure andcoping social skills) This suggests that the marked changesin behaviors observed in these particular subscales werenot simply due to a placebo-effect and may indeed berelated to improvements in glutathione redox status overthe three-month treatment period although we will haveto await double-blind placebo-controlled studies to validatethese findings The lack of a relationship between overallglutathione redox status and VABS scores most likely reflectsthe preselection of childrenwith low glutathione redox statusthus decreasing the overall range in glutathione redox status

In our previous study we demonstrated that this open-label intervention significantly improved metabolites of glu-tathionemetabolism including total and free GSH andGSSGconcentrations and the glutathione redox status (ie theGSHGSSG ratio) [23] However despite this improvementtotal and free GSH concentrations and the glutathione redoxstatus remained significantly below those of age-matchedcontrol children Additionally methionine SAM and SAHconcentrations did not significantly changewith the interven-tion despite the fact that both methylcobalamin and folinicacid provide methyl groups for the methionine cycle Thefact that the treatment improved but did not normalizemethionine SAMand glutathione concentrationsmay reflectongoing metabolic compensation for incompletely resolvedoxidative stress Thus continued prooxidant conditions maypromote glutathione synthesis as the metabolic priority atthe expense of methionine transmethylation Treatment withmethylcobalamin and folinic acid appears to have rescuedglutathione synthesis in this cohort of children albeit incom-pletely at the expense of transmethylationmetabolismGiventhe fact that some behavioral improvements were related toimprovements in glutathione status it is possible that longer-term treatment or treatments that include additional precur-sors or methylcobalamin and folinic acid at higher doses mayresult in additional behavioral improvements For exampleone recent open-label study demonstrated that high-dosefolinic acid demonstrated potential efficacy in children withASD [30] and another double-blind placebo controlled studydemonstrated that N-acetylcysteine a glutathione precur-sor resulted in decreased irritability and improved socialfunction [31] However neither of these aforementionedstudies measured redox metabolism in order to determinewhether the observed behavioral effects were related tochanges in redox metabolism Further studies will be neededto determine the optimal combination of precursors for thetreatment of redox abnormalities in children with ASD

43 The Use of Novel Interventions in Autism Spectrum Disor-der Our findings are consistent with smaller [31 32] andlarger [33] randomized control trials and large case series[30] demonstrating that nutritional interventions for childrenwith ASD that target oxidative stress are associated withimprovements in core ASD symptoms [30 31] sleep and gas-trointestinal symptoms [32] and hyperactivity tantrumingand parental impression of general functioning [33] Besidesthe methylcobalamin and folinic acid treatment used in thisstudy [23] the beneficial effect of nutritional treatments on


redox metabolism in children with ASD has been docu-mented in a randomized control trial [33] and a cross-sectional study [34] Such studies suggest that glutathionemetabolism can be improved by vitamin and mineral [33]and antioxidants coenzyme Q10 and B vitamins supplemen-tation [34] and sapropterin treatment [28] Thus althoughthere is evidence for nutritional treatments improving redoxmetabolism and ASD symptoms and behaviors in childrenwith ASD there is no specific established treatment forcorrecting redox abnormalities in children with ASD andthe evidence for efficacy of nutritional treatments whichaddress oxidative stress on core and associatedASDbehaviorsis preliminary Clearly more studies are needed to clarifythe potential beneficial effect of nutritional treatments forchildren with ASD

Thenutritional treatment used in this study is consideredby some as a complementary and alternative medicine(CAM) treatment [35] However CAM treatments refer toa much wider variety of treatments including nonnutri-tional treatments such as acupuncture and homeopathyThe prevalence of CAM in children diagnosed with a ASDin western countries is estimated to be between 32 and87 [35] which is believed to be much higher than CAMtherapies used in typically developing children [36] Childrenwith ASD with a greater number of medical and behavioralproblems receive more CAM treatments than those withfewer medical and behavioral problems [37] Parents believethat most CAM-associated treatments are either helpful orwithout effect but not harmful [35]Themain reasons parentscite for choosing to use CAM for ASD are concerns aboutthe safety and side effects of available medications for ASDand a need to be involved in decisions involving care oftheir child [38] Because the majority of CAM therapiesare based on anecdotal evidence there is a clear need forclinical trials to evaluate the efficacy of these treatments[35] In addition it should be noted that for ASD of alltreatments considered CAM nutritional treatments have themost evidence for effectiveness and efficacy Thus it maynot be proper to consider nutritional treatments which havebeen used in mainstream medicine for over a century asequivalent to CAM treatments which have undergone muchfewer investigations and have less support for their use

44 Limitations The reported improvement in behaviorscores by parent report should be interpreted with cautionin an open-label trial because of expectation bias [38] Inaddition without a control group it is difficult to know ifand towhat extent developmentwould have changedwithouttreatment As this is a group with significant delays it isunlikely that they would have demonstrated typical gainsover the three-month period but it is possible they mayhave shown some developmental progress However it isreassuring that a significant number of improvements weremoderated by improvements in glutathione metabolismsuggesting that changes in metabolism were indeed relatedto improvements in behavior

This study prescreened and selected a subgroup of chil-dren with ASD that demonstrated abnormal glutathione

redox status and methylation capacity as a means to increasesensitivity to treatment The broad heterogeneity of clinicaland behavioral symptoms in autistic children predicts thatno single treatment will benefit every autistic child Thusthe definition and characterization of subgroups of childrenwho respond positively or negatively to intervention will benecessary to more sensitively identify those children who aremost likely to benefit from a given treatment or medication

45 Conclusions This study demonstrates that a three-month treatment with methylcobalamin and folinic acidis associated with significant improvements in behavioralsymptoms associated with ASD in a group of children withAD and metabolic markers of abnormal redox and methyla-tion metabolism Our previous study demonstrated that thistreatment also improves metabolic markers of glutathionemetabolism in these same children This provides conver-gence of independent measures demonstrating the benefi-cial effect of this simple and safe nutritional interventionClearly this intervention requires further study in a double-blind placebo-controlled trial to eliminate the potential biasassociated with an open-label study

Conflict of Interests

None of the authors declare a conflict of interests

Authorsrsquo Contribution

Richard E Frye was responsible for data analysis interpre-tation and writing of the paper S Jill James was the PIfor this project and was responsible for the conduct of thestudy interpretation of the data and the writing of the paperStepanMelnyk is the laboratory director and provided HPLCexpertise formetabolite analysis George Fuchs andTyra Reidprovided clinical advice assisted with the interpretation ofdata and provided critical review of the paper OleksandraPavliv provided technical assistance for the HPLC analysisAmandaHubanks was the study nurse who recruited patientsand administered the Vineland behavior testing David WGaylor provided statistical support for data analysis

Acknowledgments

The authors gratefully acknowledge the support of theArkansas families affected by autism whose participationmade this study possible They also thank the pediatriciansand nurses at the University of Arkansas for Medical Sci-ences Dennis Developmental Center for patient referralsand Hopewell Compounding Pharmacy (Hopewell NJ) fortheir generous donation of methylcobalamin for the studyThis research was supported in part with funding from theNational Institute of Child Health and Development (RO1HD051873) to S Jill James and by grants from the Univer-sity of Arkansas for Medical Sciences Childrenrsquos UniversityMedical Group and the Arkansas Biosciences Institute (S JillJames Richard E Frye)


References

[1] C Rice ldquoAutism Developmental Disabilities Monitoring Net-work Surveillance Year Principal I Centers for Disease CPrevention prevalence of autism spectrum disordersmdashautismand developmental disabilities monitoring network UnitedStates 2006rdquo Morbidity and Mortality Weekly Report vol 58no 10 pp 1ndash20 2009

[2] G B Schaefer and N J Mendelsohn ldquoClinical genetics evalua-tion in identifying the etiology of autism spectrum disordersrdquoGenetics in Medicine vol 10 no 4 pp 301ndash305 2008

[3] R E Frye and D A Rossignol ldquoMitochondrial dysfunction canconnect the diverse medical symptoms associated with autismspectrum disordersrdquo Pediatric Research vol 69 no 5 pp 41Rndash47R 2011

[4] D A Rossignol and R E Frye ldquoMitochondrial dysfunctionin autism spectrum disorders a systematic review and meta-analysisrdquoMolecular Psychiatry vol 17 no 3 pp 290ndash314 2012

[5] S J James S Melnyk S Jernigan A Hubanks S Rose andD W Gaylor ldquoAbnormal transmethylationtranssulfurationmetabolism and DNA hypomethylation among parents ofchildren with autismrdquo Journal of Autism and DevelopmentalDisorders vol 38 no 10 pp 1966ndash1975 2008

[6] S J James S Melnyk S Jernigan et al ldquoMetabolic endopheno-type and related genotypes are associated with oxidative stressin children with autismrdquo American Journal of Medical GeneticsPart B vol 141 no 8 pp 947ndash956 2006

[7] S Melnyk G J Fuchs E Schulz et al ldquoMetabolic imbalanceassociated with methylation dysregulation and oxidative dam-age in childrenwith autismrdquo Journal of AutismandDevelopmen-tal Disorders vol 42 no 3 pp 367ndash377 2012

[8] P Ashwood B A Corbett A Kantor H Schulman J van deWater and D G Amaral ldquoIn search of cellular immunopheno-types in the blood of children with autismrdquo PLoS ONE vol 6no 5 Article ID e19299 2011

[9] P Ashwood P Krakowiak I Hertz-Picciotto R Hansen IN Pessah and J Van de Water ldquoAssociations of impairedbehaviors with elevated plasma chemokines in autism spectrumdisordersrdquo Journal of Neuroimmunology vol 232 no 1-2 pp196ndash199 2011

[10] P Ashwood P Krakowiak I Hertz-Picciotto R Hansen I NPessah and J Van deWater ldquoAlteredT cell responses in childrenwith autismrdquo Brain Behavior and Immunity vol 25 no 5 pp840ndash849 2011

[11] D A Rossignol and R E Frye ldquoA review of research trendsin physiological abnormalities in autism spectrum disordersimmune dysregulation inflammation oxidative stress mito-chondrial dysfunction and environmental toxicant exposuresrdquoMolecular Psychiatry vol 17 no 4 pp 389ndash401 2012

[12] E Viding and S-J Blakemore ldquoEndophenotype approachto developmental psychopathology implications for autismresearchrdquo Behavior Genetics vol 37 no 1 pp 51ndash60 2007

[13] S J James P Cutler S Melnyk et al ldquoMetabolic biomarkers ofincreased oxidative stress and impaired methylation capacity inchildren with autismrdquo American Journal of Clinical Nutritionvol 80 no 6 pp 1611ndash1617 2004

[14] S J James S Rose SMelnyk et al ldquoCellular andmitochondrialglutathione redox imbalance in lymphoblastoid cells derivedfrom children with autismrdquo FASEB Journal vol 23 no 8 pp2374ndash2383 2009

[15] S Rose S Melnyk T A Trusty et al ldquoIntracellular andextracellular redox status and free radical generation in primary

immune cells from children with autismrdquo Autism Research andTreatment vol 2012 Article ID 986519 10 pages 2012

[16] W Droge and R Breitkreutz ldquoGlutathione and immune func-tionrdquo Proceedings of the Nutrition Society vol 59 no 4 pp 595ndash600 2000

[17] A Pastore G Federici E Bertini and F Piemonte ldquoAnalysisof glutathione implication in redox and detoxificationrdquo ClinicaChimica Acta vol 333 no 1-2 pp 19ndash39 2003

[18] S Biswas A S Chida and I Rahman ldquoRedox modificationsof protein-thiols emerging roles in cell signalingrdquo BiochemicalPharmacology vol 71 no 5 pp 551ndash564 2006

[19] F Q Schafer and G R Buettner ldquoRedox environment of thecell as viewed through the redox state of the glutathione disul-fideglutathione couplerdquo Free Radical Biology andMedicine vol30 no 11 pp 1191ndash1212 2001

[20] M Reid and F Jahoor ldquoGlutathione in diseaserdquoCurrent Opinionin Clinical Nutrition and Metabolic Care vol 4 no 1 pp 65ndash712001

[21] S Rose S Melnyk O Pavliv et al ldquoEvidence of oxidativedamage and inflammation associated with low glutathioneredox status in the autism brainrdquo Translational Psychiatry vol2 article e134 2012

[22] A Chauhan and V Chauhan ldquoOxidative stress in autismrdquoPathophysiology vol 13 no 3 pp 171ndash181 2006

[23] S J James S Melnyk G Fuchs et al ldquoEfficacy of methylcobal-amin and folinic acid treatment on glutathione redox status inchildren with autismrdquo American Journal of Clinical Nutritionvol 89 no 1 pp 425ndash430 2009

[24] S Melnyk M Pogribna I Pogribny R J Hine and S J JamesldquoA new HPLC method for the simultaneous determination ofoxidized and reduced plasma aminothiols using coulometricelectrochemical detectionrdquo Journal of Nutritional Biochemistryvol 10 no 8 pp 490ndash497 1999

[25] S Melnyk M Pogribna I P Pogribny P Yi and SJ James ldquoMeasurement of plasma and intracellular S-adenosylmethionine and S-adenosylhomocysteine utilizingcoulometric electrochemical detection alterations with plasmahomocysteine and pyridoxal 51015840-phosphate concentrationsrdquoClinical Chemistry vol 46 no 2 pp 265ndash272 2000

[26] S Sparrow D Cicchetti and D Balla Vineland AdaptiveBehavior Scales P Assessment Minneapolis Minn USA 2ndedition 2005

[27] N M Laird and J H Ware ldquoRandom-effects models forlongitudinal datardquo Biometrics vol 38 no 4 pp 963ndash974 1982

[28] R E Frye R DeLatorre H B Taylor et al ldquoMetaboliceffects of sapropterin treatment in autism spectrum disorder apreliminary studyrdquo Translational Psychiatry vol 3 article e2372013

[29] S M Melnyk K F More and E F Miles ldquoIdiopathic radiationrecall dermatitis developing nine months after cessation ofCisplatin therapy in treatment of squamous cell carcinoma ofthe tonsilrdquo Case Reports in Oncological Medicine vol 2012Article ID 271801 3 pages 2012

[30] R E Frye J M Sequeira E V Quadros S J James and DA Rossignol ldquoCerebral folate receptor autoantibodies in autismspectrum disorderrdquoMolecular Psychiatry vol 18 no 3 pp 369ndash381 2013

[31] A Y Hardan L K Fung R A Libove et al ldquoA randomizedcontrolled pilot trial of oral N-acetylcysteine in children withautismrdquo Biological Psychiatry vol 71 no 11 pp 956ndash961 2012


[32] J B Adams and C Holloway ldquoPilot study of a moderate dosemultivitaminmineral supplement for children with autisticspectrum disorderrdquo Journal of Alternative and ComplementaryMedicine vol 10 no 6 pp 1033ndash1039 2004

[33] J B Adams T Audhya S McDonough-Means et al ldquoNutri-tional and metabolic status of children with autism vs neu-rotypical children and the association with autism severityrdquoNutrition and Metabolism vol 8 article 34 2011

[34] R E Frye R DeLatorre and H B Taylor ldquoRedox metabolismabnormalities in autistic children associated with mitochon-drial diseaserdquo Translational Psychiatry 2013

[35] N Lofthouse RHendren EHurt I E Arnold andE Butter ldquoAreview of complementary and alternative treatments for autismspectrum disordersrdquo Autism Research and Treatment 2012

[36] D A Rossignol ldquoNovel and emerging treatments for autismspectrum disorders a systematic reviewrdquo Annals of ClinicalPsychiatry vol 21 no 4 pp 213ndash236 2009

[37] J M Perrin D L Coury S L Hyman L Cole A M Reynoldsand T Clemons ldquoComplementary and alternative medicine usein a large pediatric autism samplerdquo Pediatrics vol 130 pp S77ndashS82 2012

[38] S L Hyman and S E Levy ldquoIntroduction novel therapiesin developmental disabilitiesmdashhope reason and evidencerdquoMental Retardation and Developmental Disabilities ResearchReviews vol 11 no 2 pp 107ndash109 2005

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


significantly reduced Also the mean concentration of freereduced glutathione (GSH) the major intracellular antiox-idant and mechanism for detoxification was also signifi-cantly decreased and the oxidized disulfide form of glu-tathione (GSSG) was significantly increased This imbalancein reduced as compared to oxidized glutathione resulted ina 2-fold reduction in the GSHGSSG redox ratio an indexknown as the glutathione redox status Several precursorsfor methylation and glutathione synthesis were also lowerin the autistic children suggesting an insufficiency in theproduction of methylation and glutathione metabolites atleast in a subgroup of ASD children In addition we havealso demonstrated that GSH is significantly reduced in thecytosol and isolated mitochondria from lymphoblastoid celllines and fresh peripheral blood mononuclear cells derivedfrom children with autistic disorder (AD) [14 15]

The glutathione provides the essential intracellular reduc-ing environment required for normal immune functiondetoxification capacity redox-sensitive enzyme activity andmembrane redox signaling [16ndash20] Oxidative stress occurswhen antioxidant defense mechanisms fail to counterbal-ance reactive oxygen species generated from endogenousoxidative metabolism or from prooxidant environmentalexposuresThe notion that abnormal glutathionemetabolismis associated with oxidative cellular damage is consistent withstudies from our laboratory which demonstrate an increasein oxidative damage to protein and DNA in peripheralblood mononuclear cells and postmortem brain derivedfrom ASD individuals [15 21] Several recent reviews andresearch studies also lend support to the hypothesis that redoximbalance and oxidative stress may be a contributing factorto autism pathophysiology [6 22]

These findings have particular clinical relevance sinceabnormalities in redox metabolism are potentially amenableto treatment Glutathione is a tripeptide of cysteine glycineand glutamate that is synthesized de novo in all cellsGlutathione synthesis is tightly connected to methylationmetabolism where it derives its cysteine precursor As bothmethylation and glutathionemetabolism have been shown tobe abnormal in children with ASD and due to the fact thatabnormal methylation metabolism will result in a depletionof the precursors needed for glutathione production wepreviously conducted an open-label intervention trial wheretargeted metabolic cofactors for methylation and glutathionemetabolismwere provided Specifically in our previous studywe determine if glutathione redox status could be improvedwith a three-month intervention of subcutaneously injectedmethylcobalamin and oral folinic acid in children with ADand metabolic evidence of abnormal methylation and glu-tathione metabolism [23] Although we measured adaptivebehavior using the Vineland Adaptive Behavior Scale (VABS)during the trial this behavioral data has not been previouslyreportedWe now report the change in communication dailyliving and social skills domains of the VABS associated withthe three-month intervention as well as investigate whetherimprovement in glutathione redox status was related to anychange in VABS behavior scores

Screened and met inclusion criteria

N = 65

Entered clinical trialN = 48

Normal metabolic profile N = 17

Lost to follow-up N = 4

Started treatmentN = 44

Discontinued treatmentUncomfortable with injections N = 2

Hyperactivity and reduced sleep N = 2Completed treatment

N = 40

Did not complete second behavioral evaluation N = 3

Behavioral data for analysis N = 37

Figure 1 Flowchart of patients who met inclusion criteria for thestudy

2 Methods and Materials

21 Participants Forty-eight children with ADwere enrolledin an open-label trial to test whether a three-month treatmentof methylcobalamin (methyl B12) and folinic acid wouldimprove glutathione synthesis as well as adaptive behaviorInclusion criteria were a diagnosis of AD as defined by theDiagnostic and Statistical Manual of Mental Disorders FourthEdition (DSM-IV) criteria (2990) and a Childhood AutismRating Scales (CARS) score greater than 30 Exclusion cri-teria included Aspergerrsquos disorder pervasive developmentaldisorder-not otherwise specified (PDD-NOS) genetic disor-ders epilepsy severe gastrointestinal symptoms recent infec-tions andor current use of high-dose vitamin or mineralsupplements (ie above the recommend daily allowance)Metabolic and behavioral assessments were conducted atbaseline and at the end of the 3-month intervention A nursecontacted each family every two weeks to ensure complianceand monitor adverse effects

Figure 1 provides a flowchart of the patients who met cri-teria for possible inclusion in the study In order to determineif abnormal redox metabolism could be improved with theintervention we selected children with abnormal glutathioneredox status and methylation capacity for entry into thestudy Normal glutathione redox status and methylationcapacity were defined by measuring these same metabolitesin typically developing control children Sixty-five autisticchildrenmet the inclusion criteria and were initially screenedfor metabolic evidence of abnormal methylation capac-ity (SAMSAH lt 30) and glutathione redox metabolism(GSHGSSG lt 60) Of these 48 (75) children were foundto have both abnormal methylation capacity and glutathioneredox metabolism thus meeting metabolic qualifications for


















3 Results



10203040506070

Actu

al ag

e

Rece

ptiv

e lan

guag

e

Expr

essiv

e lan

guag

e

Writ

ten

lang

uage

Pers

onal

skill

s

Dom

estic

skill

s

Com

mun

ity sk

ills

Inte

rper

sona

l ski

lls

Play

-leisu

re sk

ills

Cop

ing

skill

s

Ove

rall

skill

s

Age

(mon

ths)






minus4minus20

0

4

8

12


raw score

Chan

ge in

free

GSH

GSS

G

(a)

minus4minus20

0

4

8

12


raw score

(b)

minus4

0

4

8

12


raw score

(c)

0

4

8

12


minus4minus20

raw score

Chan

ge in

free

GSH

GSS

G

(d)

0

4

8

12

0 20 40 60minus4


raw score

(e)

0

4

8

12


minus4minus20

raw score

(f)








4 Discussion




















Acknowledgments



References














































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

































3 Results



10203040506070

Actu

al ag

e

Rece

ptiv

e lan

guag

e

Expr

essiv

e lan

guag

e

Writ

ten

lang

uage

Pers

onal

skill

s

Dom

estic

skill

s

Com

mun

ity sk

ills

Inte

rper

sona

l ski

lls

Play

-leisu

re sk

ills

Cop

ing

skill

s

Ove

rall

skill

s

Age

(mon

ths)






minus4minus20

0

4

8

12


raw score

Chan

ge in

free

GSH

GSS

G

(a)

minus4minus20

0

4

8

12


raw score

(b)

minus4

0

4

8

12


raw score

(c)

0

4

8

12


minus4minus20

raw score

Chan

ge in

free

GSH

GSS

G

(d)

0

4

8

12

0 20 40 60minus4


raw score

(e)

0

4

8

12


minus4minus20

raw score

(f)








4 Discussion




















Acknowledgments



References














































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















minus4minus20

0

4

8

12


raw score

Chan

ge in

free

GSH

GSS

G

(a)

minus4minus20

0

4

8

12


raw score

(b)

minus4

0

4

8

12


raw score

(c)

0

4

8

12


minus4minus20

raw score

Chan

ge in

free

GSH

GSS

G

(d)

0

4

8

12

0 20 40 60minus4


raw score

(e)

0

4

8

12


minus4minus20

raw score

(f)








4 Discussion




















Acknowledgments



References














































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of


































Acknowledgments



References














































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















References














































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





























of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















clinical study effectiveness of methylcobalamin and folinic ...effectiveness of methylcobalamin and...

Documents