THANK YOU FOR YOUR ORDERThank you for your recent purchase. If you need further assistance, please contact Customer Service at customerservice@infotrieve.com. Please include the Request ID, so we can better assist you.

Delivered by

CUSTOMER INFORMATION ORDER INFORMATION

DOCUMENT INFORMATION

Request ID:

Ordered For:

Ordered For Email:

Ordered:

Ordered For:

Company:

Address:

Country:

Phone:

Fax:

Email:

Publication:

Vol(Iss) Pg:

Title:

Genre:

Urgency:

Copies:

Type:

Usage:

Date

Std. Num.:

Publisher:

This is not an invoice.

Deliver Via:

Delivery Address: Tracking Info.:

Author(s):

Client ID:

(800) 422-4633 Toll Free (203) 423-2175 Direct

Email: customerservice@infotrieve.com

Total Fee:

Copyright Terms:

You have not secured permission through Copyright Clearance Center, Inc. for any other purpose but may have other rights pursuant to other arrangements you may have with the copyright owner or an authorized licensing body. To the extent that a publisher or other appropriate rights-holder has placed additional terms and conditions on your use of this document, such terms and conditions are specified herein under “Copyright Terms”. If you need to secure additional permission with respect to this content, please purchase the appropriate permission via RightFind.

The contents of the attached document are copyrighted works. You have secured permission to use this document for the following purpose:

Molecular Genetics and Metabolism 122 (2017) 51–59

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

j ourna l homepage: www.e lsev ie r .com/ locate /ymgme

Anaplerotic therapy in propionic acidemia

Nicola Longo a,b,⁎, Leisa B. Price a, Eduard Gappmaier c, Nancy L. Cantor d, Sharon L. Ernst a,Carrie Bailey a, Marzia Pasquali b

a Division of Medical Genetics, Department of Pediatrics, University of Utah, Salt Lake City, UT, USAb Department of Pathology, University of Utah, ARUP Laboratories, Salt Lake City, UT, USAc Department of Physical Therapy, University of Utah, Salt Lake City, UT, USAd Primary Children's Hospital, Salt Lake City, UT, USA

⁎ Corresponding author at: Division of Medical GeneUniversity of Utah, 295 Chipeta Way, Salt Lake City, UT 84

E-mail address: Nicola.Longo@hsc.utah.edu (N. Longo)

http://dx.doi.org/10.1016/j.ymgme.2017.07.0031096-7192/© 2017 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 6 July 2017Accepted 7 July 2017Available online 12 July 2017

Background: Propionic acidemia is a raremetabolic disorder caused by a deficiency of propionyl- CoA carboxylase,the enzyme converting propionyl-CoA to methylmalonyl-CoA that subsequently enters the citric acid cycle assuccinyl-CoA. Patients with propionic acidemia cannotmetabolize propionic acid, which combineswith oxaloac-etate to form methylcitric acid. This, with the defective supply of succinyl-CoA, may lead to a deficiency in citricacid cycle intermediates.Purpose: The objective of this studywas to determinewhether supplementswith glutamine (400mg/kg per day),citrate (7.5mEq/kg per day), or ornithineα-ketoglutarate (400mg/kg per day) (anaplerotic agents that could fillup the citric acid cycle) would affect plasma levels of glutamine and ammonia, the urinary excretion of Krebscycle intermediates, and the clinical outcome in 3 patients with propionic acidemia.Methods: Each supplementwas administered daily for fourweekswith a twoweekwashout period between sup-plements. The supplement that produced themost favorable changes was supplemented for 30weeks followingthe initial study period and then for a 2 year extension.Results: The urinary excretion of the Krebs cycle intermediates, α-ketoglutarate, succinate, and fumarate in-creased significantly compared to baseline during citrate supplementation, but not with the other two supple-ments. For this reason, citrate supplements were continued in the second part of the study. The urinaryexcretion of methylcitric acid and 3-hydroxypropionic acid did not change with any intervention. No significantchanges in ammonia or glutamine levels were observed with any supplement. However, supplementation withany anaplerotic agents normalized the physiological buffering of ammonia by glutamate, with plasma glutamateand alanine levels significantly increasing, rather than decreasing with increasing ammonia levels. No significantside effects were observed with any therapy and safety labs (blood counts, chemistry and thyroid profile)remainedunchanged.Motor and cognitive developmentwas severely delayedbefore the trial and didnot changesignificantly with therapy. Hospitalizations per year did not change during the trial period, but decreased signif-icantly (p b 0.05) in the 2 years following the study (when citrate was continued) compared to the 2 years beforeand during the study.Conclusions: These results indicate that citrate entered the Krebs cycle providing successful anaplerotic therapyby increasing levels of the downstream intermediates of the Krebs cycle: α-ketoglutarate, succinate and fuma-rate. Citrate supplements were safe and might have contributed to reduce hospitalizations in patients withpropionic acidemia.

© 2017 Elsevier Inc. All rights reserved.

Keywords:Propionic acidemiaOrganic acidemiaOutcomeSodium citrateAnaplerosisClinical trial

1. Introduction

Propionic acidemia is an autosomal recessive disorder caused by de-ficiency of propionyl CoA carboxylase (EC 6.4.1.3), the enzyme that con-verts propionyl CoA to methymalonyl CoA with the help of the cofactor

tics, Department of Pediatrics,108, United States..

biotin [1]. This conversion, which occurs in mitochondria, is part of thepathway for degradation of the amino acids isoleucine, methionine,threonine, and valine, odd chain fatty acids, and cholesterol [1].Propionic acid also originates from the catabolism of the nucleotidesthymine and uracil and from bacterial production of propionate frompyruvate in the gut [1]. Propionyl CoA is eventually converted into suc-cinyl CoA and enters the citric acid (Krebs) cycle for energy production.Propionyl CoA carboxylase is composed of two distinct subunits: α andβ, either of which can be defective in propionic acidemia [1]. As a result

52 N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

of defective propionyl CoA carboxylase, propionyl CoA accumulates andcombines with oxaloacetate, another intermediate of the citric cycle, toform methylcitric acid, the diagnostic metabolite of propionic acidemia(Fig. 1). Most cases of propionic acidemia present.

with lethargy progressing to coma from 16 h to weeks after birth,depending on the severity of the enzyme impairment caused by the ge-netic lesion [1]. Patients can have severe hyperammonemia associatedor not with metabolic acidosis [1,2]. Even when patients are rescuedfrom the hyperammonemic coma, the prognosis is poor since patientscan develop life-threatening complications such as pancreatitis or car-diomyopathy [2,3]. These complications, whose mechanism is un-known, cause severe morbidity and mortality even in optimallytreated patients limiting the benefits of early diagnosis by newbornscreening programs [2,3].

The mechanism at the basis of these long-term complications is un-known. It might be related to toxic effects of metabolites accumulatingas a result of themetabolic block (including chronic hyperammonemia)and/or to decreased energy production (by inhibition of the Krebscycle). In propionic acidemia, glutamine levels are loweven inwell-con-trolled patients and decrease (rather than increase) withhyperammonemia [4]. We have proposed that this glutamine paradox,also seen by other groups [5–7], could be due to a dysfunctional Krebscycle, with deficiency of α-ketoglutarate [4]. In propionic acidemia,glutamate dehydrogenase, normally a cataplerotic enzyme favoringthe exit of α-ketoglutarate as glutamate [4], works in reversegenerating α-ketoglutarate from glutamate [4]. Low levels ofglutamate favor the release of ammonia from glutamine to generate

Fig. 1. Anaplerotic therapy in propionic acidemia. The supplements used in this study (ornithinacidmetabolism and the tricarboxylic acid cycle. Propionyl CoA is normally carboxylated by proCoA racemase and mutase, this produces succinyl CoA that can enter the Krebs cycle and cocondenses with oxaloacetate to produce methylcitric acid. The decrease in oxaloacetate and suThis can be repleted to the expense of glutamine and glutamate. Ornithine can be converted t5-carboxylate dehydrogenase. GDH: glutamate dehydrogenase; MAAT: mitochondrial asparglutaminase.

glutamate, explaining the association between high ammonia andlow glutamine levels (Fig. 1).

Propionic acid is important in the anaplerosis (filling-up) of theKrebs cycle as indicated by the clinical improvement of patients withfatty acid oxidation disorders and pyruvate carboxylase deficiencytreated with heptanoin, an odd-chain fatty acid that is metabolized topropionic acid and converted to succinyl-CoA [4,8–10]. Since such amechanism is so effective in replenishing the Krebs cycle, its completeabsence in patients with propionic acidemia, coupled with the seques-tration of oxaloacetate by propionyl-CoA to form methylcitrate, shouldresult in a severe deficiency of all intermediates of the citric acid cycle.In muscle, α-ketoglutarate is the intermediate with the lowest concen-tration (0.05 mmol/kg) after oxaloacetate (0.012 mmol/kg, [11,12]). α-Ketoglutarate concentration further declines with exercise [11] and isregenerated from glutamine and glutamate [11,12]. If glutamine/gluta-mate andα-ketoglutarate are too low, such as in propionic acidemia [4],the process might become ineffective in generating energy, possiblycontributing to hypotonia and progressive organ failure.

Supplements can replenish substrates to the Krebs cycle. In patientswith argininosuccinic aciduria, citrate, the intermediate with thehighest concentration in the Krebs cycle (0.362 mmol/kg in muscle,[11,13]), can generate cytoplasmic aspartate to increase conjugationwith citrulline and the urinary excretion of the water-solubleargininosuccinic acid [14,15]. This action does not require entry of cit-rate into mitochondria and this therapy is not routinely used given itsmodest efficacy and possible side effects (metabolic alkalosis, [14]). Cit-rate in combination with aspartate stabilized the metabolic control of

e alpha-ketogluratate, glutamine and citrate) are shown underlined in respect to propionicpionyl CoA carboxylase to become D-methylmalonyl CoA.With the action ofmethymalonylntribute to energy metabolism. In propionic acidemia, propionyl CoA accumulates andccinyl CoA can impair the Krebs cycle, reducing the concentration of alpha-ketoglutarate.o glutamate by the action of two enzymes, ornithine amino transferase and Δ1pyrroline-tate amino transferase; OAT: ornithine amino transferase; PDG: phosphate-dependent

53N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

one patient with pyruvate carboxylase deficiency [16] and increasedplasma glutamine levels in another patient [9]. Both patients had severeneurological compromise and citrate effects on the overall outcomewere difficult to assess given the concomitant use of other medica-tions/supplements [9,16].

Other natural supplements include glutamate, glutamine, and α-ketoglutarate. Glutamate is mostly (N90%) metabolized by the splanch-nic bed, while N50% of oral glutamine can reach the circulation and in-crease glutamate production [17]. Glutamine has been usedextensively in the treatment of very low birth weight infants in whomit is extremely safe (up to 0.6 g/kg/day) and prevents protein catabolism[18]. Glutamine (2 g/m2 BID) decreases narcotic requirements and thenumber of days of intravenous hyperalimentation in cancer patients re-ceiving chemotherapy [19]. α-Ketoglutarate is available as a salt or incombination with the positively charged amino acids arginine andornithine. The salt form and arginine α-ketoglutarate are widely avail-able as supplements, but only limited clinical trials have been reported[20]. Ornithine α-ketoglutarate has been used since the early 1960sand is more effective than α-ketoglutarate or ornithine alone inincreasing glutamine levels in healthy subjects [20]. This is because or-nithine can be converted into glutamate through ornithine amino trans-ferase and Δ1pyrroline-5-carboxylate dehydrogenase (Fig. 1) [20,21].Ornithine α-ketoglutarate improves the nutritional status in patientswith burns and in children receiving total parental nutrition [22,23].The combination of ornithine and α-ketoglutarate can spareendogenous glutamine (increasing its plasma levels), feed directly intothe Krebs cycle, and participate in the synthesis of urea cycle aminoacids [21]. Since in propionic acidemia there is a direct correlation be-tween plasma levels of glutamine/glutamate and levels of the ureacycle amino acids ornithine and arginine [4], the simultaneousadministration of ornithine and α-ketoglutarate could have asynergistic effect. Both glutamine and ornithine α-ketoglutarate areextremely well tolerated in humans at doses up to 0.6 g/kg/day andhave not caused any known adverse reaction. They are freely availableas nutritional supplements to the general public and are marketedmostly as daily supplements for body builders or for general health.The interaction of these supplements (citrate, glutamine and alpha-ketoglutarate) with the Krebs cycle and the metabolism of propionicacid is summarized in Fig. 1.

Here we evaluate the effect of anaplerotic supplements (citrate, glu-tamine, or ornithine α-ketoglutarate) on biochemical parameters andoutcome of patients with propionic acidemia.

2. Patients and methods

The study was approved by the University of Utah InstitutionalReview Board and parents signed an informed consent prior to anystudy procedure. The trial is registered at ClinicalTrials.gov underidentifier NCT00645879. Three patients with propionic acidemiafollowed at the University of Utah since birth were enrolled in thisstudy. Their ages at time of enrollment ranged from 5.9 to 11 yearsof age. Their diagnosis was confirmed by enzyme assay in fibroblastsor white blood cells showing absent activity (0% of controls) ofpropionyl-CoA carboxylase and normal activities of pyruvatecarboxylase and methylcrotonyl–CoA carboxylase. They allpresented within 1 month of age with acute hyperammonemia andrequired intubation and admission to the intensive care unit. Theirdevelopmental quotient/IQ ranged from moderately to severelydelayed with the gross motor area being the most affected. Eachpatient in the 2 years prior to the trial required 2–6 hospitaladmissions per year to control metabolic decompensation or inter-current illnesses.

Prior to beginning and at the conclusion of the study, each patientreceived a cognitive evaluation using the Stanford-Binet IntelligenceScales (SB5), Fifth Edition and a motor evaluation. The motor activity

assessment included hand-held myometry studies, time-functiontests, and an accelerometry evaluation.

The study occurred in two parts. In the first part, individualsupplements (ornithine α-ketoglutarate, glutamine, and citrate)were evaluated for their ability to raise plasma glutamine levels,or to cause other biochemical changes as compared to baseline(before the study and during wash-out periods). In the secondpart, the diet of the patients was supplemented with the supple-ment judged to produce the most favorable changes (citrate) for30 weeks. Doses of supplements (ornithine α-ketoglutarate,glutamine, and disodium citrate) were based on patients' weightsat weeks 0, 6, 12, and 18 and were not changed during thecorresponding treatment period (4 weeks or 30 weeks). Ornithineα-ketoglutarate 400 mg/kg (2 mmol/kg ornithine, 1 mmol/kg α-ketoglutarate) per day was started at week 0 and continued untilweek 4. Week 4–6 was a washout period with no supplements for2 weeks. Glutamine 400 mg/kg (2.74 mmol/kg) per day was startedat week 6 and continued throughweek 10. A secondwashout periodoccurred from week 10–12 with no supplements for 2 weeks. Atweek 12, disodium citrate (7.5 mEq/kg, 2.5 mmol/kg) was startedand continued through week 16. The final washout period occurredfrom week 14–16. At week 18, citrate was started at the samedosing level and continued through week 48 when measurementswere repeated, the study was officially concluded, and patientswere offered to continue citrate supplements on a clinical basis.All patients continued citrate supplements for N2 years followingthe conclusion of the study.

Study visits occurred each week starting at baseline andcontinued until week 16, with the exception of two week washoutperiods. During the final treatment period, week 18–48, study visitsoccurred every 6 weeks. Each study visit included a physical examalong with anthropometric and vital sign measurements. Dietrecords were collected and analyzed during the final visit of eachtrial period (weeks 4, 10, 16, 48). Concomitant medications andtherapies were collected and recorded.

After the initial evaluation, parents were given the nutritional sup-plement to mix in the child's metabolic formula containing all aminoacids except propiogenic amino acids (isoleucine, methionine, threo-nine and valine). Patients were also receiving a measured amount ofnatural proteins from regular food. All patients had gastrostomy tubesand received part or all their daily nutrition through bolus or continuousfeeds. Following the initial study period, lab results were drawn period-ically at clinic visits 2–4 h after feeding. There were no changes in thediet (percent of calories from fats, carbohydrates, natural protein andspecial formulas) during the study period except for adjustments forthe changing body weight.

Study labs included plasma amino acids, acylcarnitine profile, am-monia, lactic acid, urine acylglycines, and urine organic acids (as firstvoidmorning urine). Biochemical testswere performed at ARUP labora-tories of the University of Utah. Plasma amino acids were analyzed byion-exchange chromatography with post-column ninhydrin detection(Biochrom 30 amino acid analyzer) [24]. Urinary organic acids were an-alyzed by gas chromatography–mass spectrometry (HP/Agilent) [25].Plasma acylcarnitine levels and profile were determined by tandemmass spectrometry (Waters Quattro Premiere) using stable isotopestandards for quantitation. Analytes not routinely reported(methylcitrate, propionylglycine in urine) were consistently quantifiedin each sample.

Plasma amino acids were compared before and after treatmentwithanaplerotic agents. Baseline study labs (week 0, 6, 12, and 18) were av-eraged and compared to those obtained during treatment with orni-thine α-ketoglutarate (n = 4), glutamine (n = 4) and citrate (n = 4).Data were compared using analysis of variance with p b 0.05 as a statis-tical cutoff. Regression analysiswas used to correlate different biochem-ical parameters. Calculationswere performed usingMicrosoft Excel andSigmaPlot software, version 13.0.

Fig. 2. Effect of anaplerotic therapy on (A) plasma levels of glutamic acid (Glu), Glutamine(Gln) their sum (Glu + Gln and ammonia and (B) selected urinary organic acids inpatients with propionic acidemia. Each value is the average ± SD of at least 12observations. *p b 0.05 versus baseline.

54 N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

3. Results

3.1. Patients and safety

Table 1 summarizes the clinical and anthropometric data of patientswith propionic acidemia. Three patients participated in this study (ageat the beginning of the study ranged from 5.9 to 11 years). Anthropo-metric measures (weight, height, head circumference, and BMI) didnot change significantly from the beginning to the end of the study.BMI improved in all 3 patients during the study,with percentiles that in-creased in two patients who were underweight at the beginning of thestudy (from the 22nd to the 50th and from the 3rd to the 46th percen-tile, respectively) and decreased in a third patient who was overweight(from the 95th to the 91st percentile), though these improvementswere not statistically significant.

Safety laboratory studies (hemoglobin, white blood cell counts,platelets, transaminases, alkaline phosphatase, total proteins, albumin,free T4, TSH, amylase, lipase) also did not change during treatmentwith ornithine α-ketoglutarate, glutamine or citrate. Serum lactatelevels also did not change significantly during the study, with very fewvalues above the normal range. As noted in Fig. 2A, ammonia levelsdid not change significantly during the trial as did the urinary excretionof lactate and pyruvate (Fig. 2B). Overall, routine lab values were notsignificantly affected by any of the treatments indicating overall safetyof the supplements.

3.2. Amino acids and organic acids

In the first part of the study, patients with propionic acidemia re-ceived three different supplements (ornithine α-ketoglutarate (OKG),glutamine (Gln) and citrate) for 4 weeks with blood and urine testingevery week. Values obtained in the presence of supplementswere com-pared to baseline (before the initiation of the study and during wash-out periods). None of the treatments significantly increased [glutamine]or [glutamine] + [glutamate] concentration (Fig. 2A). Ammonia levelswere mildly elevated in all patients (mean value 56.6 ± 21.2 μmol/L,normal for age 21–50 μmol/L) and none of the supplements was effec-tive in normalizing them (Fig. 2A). The administration of OKG and glu-tamine provided extra amino groups, yet there was no increase inammonia levels (Fig. 2A) indicating that the extra nitrogen generatedfrom them was normally disposed.

Since there was no change in glutamine or ammonia levels, welooked at the urinary excretion of Krebs cycle intermediates (urine or-ganic acids) for evidence that the supplements entered mitochondria.The urinary excretionof ketoglutarate did not increase significantly dur-ing treatment with ornithine α-ketoglutarate or glutamine, but did in-crease significantly during citrate therapy (p b 0.01) (Fig. 2B). The

Table 1Characteristics of patients with propionic acidemia. Measurements taken at baseline and at the

Patient 1 2

Gender F F

Race Hispanic White

Age (yr) Before 5.9 11After 7.3 12.2

Weight, kg (centile) Before 15.6 (2) 45.2 (80)After 19.7 (9) 48.9 (74)

Height, cm (centile) Before 104.5 (2) 136.5 (14)After 112.5 (1) 143.3 (10)

Head circumference, cm (centile) Before 47.2 (b3) 51 (10)After 48 (b3) 51.5 (10)

Body mass index, kg/m2 (centile) Before 14.3 (22) 24.3 (95)After 15.6 (50) 23.8 (91)

Mutation Not done PCCBc.429 + 3_c.1218_12

urinary levels of fumarate and succinate also increased significantlyduring citrate treatment compared to baseline (p b 0.001), but notwith the two other supplements. Levels of lactate, pyruvate, and citratealso increased during citrate treatment, though this increase was notstatistically significant compared to baseline.

Some of the supplements might have affected the accumulation oftoxic metabolites in propionic acidemia. The urinary excretion ofmethylcitrate, the characteristicmetabolite of propionic acidemia, 3-hy-droxy propionic acid and propionylglycine remained elevated and didnot change significantly with any supplement (Fig. 3A, B, C). In addition,the plasma concentration of propionyl (C3−) carnitine remained ex-tremely elevated (normal 0–0.83 μmol/L) and was not modified byany of the therapies (Fig. 3D).

In summary, citrate, but not glutamine nor ornithine α-ketoglutarate significantly increased the urinary excretion ofketoglutarate, succinate and fumarate, all metabolites produced aftercitrate entry into mitochondria and its modification by enzymes of the

conclusion of the study period. Results reported as means ± SD.

3 Average

M

White

6.8 7.9 ± 2.77.8 9.1 ± 2.720.2 (18) 27 ± 15.9 (33 ± 41)27.2 (67) 31.9 ± 15.2 (50 ± 36)122 (57) 121 ± 23 (24 ± 29)132 (80) 128 ± 22 (30 ± 43)49.5 (3) 49.2 ± 1.950 (3) 49.8 ± 1.813.6 (3) 17.4 ± 6 (40 ± 48.6)15.6 (46) 18.3 ± 4.7 (62 ± 25)

+ 6delAAGT31delins12 (p.G407Rfs*14)

PCCAc.1284 + 1G N Ac.2119 − 1G N C

Fig. 3. Effect of anaplerotic therapy on the urinary excretion of methylcitrate (A), 3-hydroxy-propionate (B), propionylglycine (C), and ketone bodies (sum of 30hydroxybutyric acid andacetoacetate) (D) in patients with propionic acidemia. Each value is the average± SD of at least 12 observations. There were no statistically significant difference between any treatmentand baseline.

55N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

Krebs cycle. At the same time, there was no statistically significant in-crease in the excretion of lactate and pyruvate, which accumulatewhen the Krebs cycle is dysfunctional nor a change in the classic metab-olites that accumulate in propionic acidemia. Thus, citrate was supple-mented in the second part of the study for 30 weeks.

Fig. 4. Correlation of ammonia levels with selected plasma amino acids in patients withpropionic acidemia. Correlation of ammonia levels with select plasma amino acids inpatients with propionic acidemia during the clinical trial. Data were analyzed by linearregression with results indicated in the figure. The shaded area represents the normalrange of variation.

To evaluate changes in plasma amino acids, the results of both partsof the studywere grouped together and are summarized in Tables 3 and4 and Fig. 4. Table 3 shows the average values of the 22 amino acids col-lected at baseline and during each treatment period. The majority ofamino acids remained unchanged throughout the study. As expected,

Fig. 5. Correlation of ammonia levels with selected plasma amino acids in patients withpropionic acidemia. Data were analyzed by linear regression with results indicated inthe figure. The shaded area represents the normal range of variation. Note that severallysine levels were above the normal range.

Table 3Plasma amino acids (μmol/L) in patients with propionic acidemia. Values are averages± SD of at least 12 observations. *p b 0.05 (or better) versus baseline. Values outside thenormal range are indicated in bold.

Amino acid Normal Baseline Citrate Glutamine OKG

Alanine 240–600 349 ± 85 331 ± 91 328 ± 79 362 ± 65Arginine 40–160 42.8 ± 11.7 36.5 ± 10.2 45.3 ± 21.4 41.4 ± 10.9Asparagine 15–40 31.2 ± 13.1 27.6 ± 9.9 26 ± 9.5 29.2 ± 5.0Aspartate 0–20 11.1 ± 4.6 9.9 ± 2.8 9.7 ± 2.5 11.7 ± 2.8Citrulline 10–60 19.6 ± 4.2 17.5 ± 4.9 21.4 ± 7.9 21.9 ± 4.8Cysteine 7–70 34 ± 7.5 29.7 ± 6.9 33.5 ± 7.1 32.1 ± 7Glutamine 410–700 365 ± 55 375 ± 49 367 ± 55 391 ± 74Glutamate 10–120 51.2 ± 14.2 50.7 ± 15.6 49.7 ± 12.6 52 ± 7.3Glycine 140–490 1268 ± 177 1173 ± 244 1249 ± 229 1376 ± 158Histidine 50–130 76.9 ± 20.3 67.7 ± 22.1 60.8 ± 20.2 71.9 ± 20.4Isoleucine 30–130 20.7 ± 10.6 20.6 ± 9.5 20.7 ± 11 23.3 ± 9.5Leucine 60–230 69.5 ± 35.7 60.8 ± 21.7* 64.3 ± 22.6 50.1 ± 15.8Lysine 80–250 227 ± 76 226 ± 68 244 ± 103 224 ± 82

Table 2Safety Laboratory studies in patients with propionic acidemia. Values are averages ± SD of at least 12 observations. OKG: ornithine α-ketoglutarate.

Labs Normal Baseline Citrate Glutamine OKG

Hemoglobin (g/dL) 11–13.3 13.1 ± 2.1 12.8 ± 1.4 12.6 ± 1.1 12.1 ± 1.0White blood cells (k/uL) 4.5–10.5 5.3 ± 1.5 5.2 ± 1.8 6.0 ± 2.5 5.4 ± 1.0Platelets (k/uL) 204–405 234 ± 70 233 ± 70 273 ± 90 302 ± 94AST (U/L) 20–60 60.3 ± 24.4 59.9 ± 22.4 47.9 ± 11.4 44.9 ± 6.7ALT (U/L) 5–45 45.8 ± 29.6 54.2 ± 32.4 33.5 ± 15.7 35.4 ± 11.5Alk. phos (U/L) 145–320 189 ± 57 207 ± 44 191 ± 53 175 ± 51Total protein (g/dL) 5.9–7.0 6.7 ± 0.7 6.6 ± 0.5 6.9 ± 0.6 6.7 ± 0.7Albumin (g/dL) 3.1–4.2 3.9 ± 0.4 3.8 ± 0.2 3.9 ± 0.3 4.0 ± 0.3

56 N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

ornithine levels increased significantly (p b 0.001 compared to baseline)during ornithine α-ketoglutarate supplementation. This indicated thatthe chemical was successfully absorbed in the gut. However,ketoglutarate excretion in urine did not increase during the ornithineα-ketoglutarate treatment period (Fig. 2B) indicating that α-ketoglutarate was either metabolized prior to entry into mitochondriaor did not enter the Krebs cycle in sufficient amounts to increase its uri-nary excretion or that of downstreammetabolites.

Some amino acids (leucine, phenylalanine and threonine) decreasedsignificantly compared to baseline (Table 3) during supplementationwith ornithineα-ketoglutarate (p b 0.01, p = 0.014, and p= 0.037, re-spectively) and leucine decreased significantly (p=0.02) during citratesupplementation. These changes were relatively small and probablyreflected mild changes in protein restriction during the trial. As expect-ed, glycine levels were significantly above the normal range and did notchange significantly with any treatment. Similarly, levels of isoleucine,methionine and valine were below normal reflecting their restrictionin medical foods and did not change with any intervention.

In propionic acidemia, glutamine levels are low even in well-con-trolled patients and decrease (rather than increase) withhyperammonemia [4]. This might be due to low levels of α-ketoglutarate favoring release of nitrogen from glutamine and gluta-mate to generate theKrebs cycle intermediate [4]. Since all supplementsused in the study could have potentially increasedα -ketoglutarate pro-duction and reversed this relationship, we explored whether the nega-tive correlation between ammonia and glutamine/glutamate andalanine were still observed during the study. As shown in Fig. 4 andTable 4, glutamine levels did not decrease with increasing ammonialevels (panel B) and levels of glutamate (A) and alanine (C) significantlyincreased (even though the correlation was minimal), rather than de-creased with increasing ammonia levels. This suggests that these thera-pies might have re-established, at least in part, ketoglutarate levels andthe physiological mechanism of ammonia buffering by the centrilobularhepatocytes.

In propionic acidemia, levels of ammonia positively correlated withan increase in the amino acids leucine, tyrosine and phenylalanine [4].During this clinical trial we observed the same trend (Table 4, Fig. 5C,E, F). Since these amino acids are essential and provided both by naturalproteins and the special formula that patients with propionic acidemiaconsume, these results suggest that ammonia levels correlated to over-all protein intake. In addition to these large neutral amino acids, argi-nine, citrulline, and lysine increased with ammonia levels (Fig. 5A, B,D), with lysine rising in many occasions above the normal range as am-monia increased (Fig. 5D).

There was no significant correlation between ammonia levels andthe remaining amino acids (Table 4).

Methionine 17–53 12.6 ± 3 13.5 ± 3.6 13.9 ± 3.9 12.7 ± 2.9Ornithine 20–135 40.7 ± 16.9 32.6 ± 16 37.3 ± 12.7 94.3 ± 39.2*Phenylalanine 30–80 42.6 ± 9.5 41 ± 7 39.4 ± 7.9 34.3 ± 5.2*Proline 110–500 252 ± 92 308 ± 79 264 ± 64 284 ± 70Serine 60–200 142 ± 22 132 ± 29 132 ± 33 147 ± 7Taurine 25–80 67.8 ± 15.6 60.8 ± 8.8 61.2 ± 7.1 66.5 ± 10.2Threonine 60–220 65.3 ± 15.2 74 ± 26 55.4 ± 4.2 54.1 ± 8.7*Tyrosine 30–120 79.8 ± 20.8 67.4 ± 20 80.1 ± 21 66.1 ± 17.6Valine 140–350 60.7 ± 25.6 67.8 ± 28.2 64 ± 26.6 55.5 ± 4.8

3.3. Cognitive and motor assessment

The cognitive assessment (Table 5) indicated that all patients had se-verely compromised intellectual abilities before the study. There wereno changes in nonverbal IQ, verbal IQ, or full scale IQ from the beginningto the end of the study in patients 1 or 2. Patient 3 was too severely

affected and unable to complete the evaluation at the beginning andend of the study.

Table 6 shows the results from the motor assessment of patients 1and 2. Patient 3 was unable to perform the tasks of the assessment.The hand-held myometry composite score increased for both patient 1and 2 during the course of the study. Results of the time function testswere variable between the patients. Time required to walk/run 10 mand to climb 4 stairs improved for patient 1, but the time to rise from su-pine to standworsened. Patient 2was unable to complete all testing be-cause she did not have her walker during the final day of testing. Thetime to climb four stairs worsened for patient 2 during the study.Accelerometry studies showed that both patient 1 and 2 became less ac-tive from the beginning to the end of the study with decreases in aver-age steps in 7 days, averageminutes of activity in 7 days, and increase inaverage % of inactivity in 7 days.

3.4. Hospitalizations

To evaluate the effect of treatment on overall health, we counted thenumber of days that our patients spent in the hospital before, duringand after the clinical trial while they were continuing on the supple-ments (Table 7). Hospitalization days were normalized per year to ac-count for the shorter time span during the study compared to the2 years before and after the study. The cause of hospitalizations was re-lated mostly to respiratory problems, pancreatitis, or viral infections. Inno case were adverse events deemed related to treatment, but rather tothe underlyingmedical condition or intercurrent illnesses. Patientswithpropionic acidemia did not have any change in the number of days inthe hospital during the study. However, the average number of days/

Table 4Correlation of ammonia levels with plasma amino acids in patients with propionicacidemia during the clinical trial. Statistical significance was calculated using analysis ofvariance. R2 and p are indicated. *p b 0.05 or better (indicated in bold).

Amino acid R2 p

Alanine 0.07* 0.041Arginine 0.29* b0.001Asparagine 0.01 0.380Citrulline 0.13* 0.006Cysteine 0.01 0.388Glutamine 0.02 0.282Glutamine + glutamate 0.05 0.112Glutamine + glutamate + alanine 0.09* 0.026Glutamate 0.12* 0.008Glycine 0.06 0.074Histidine 0.02 0.240Isoleucine 0.01 0.369Leucine 0.11* 0.012Lysine 0.52* b0.001Methionine 0.02 0.278Ornithine 0.05 0.108Phenylalanine 0.10* 0.014Proline 0.006 0.582Serine 0.02 0.246Threonine 0.04 0.148Tyrosine 0.13* 0.007Valine 0.005 0.608

57N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

year that patients were hospitalized significantly decreased during the2 years after the study, while the patients continued to supplementwith open-label citrate, as compared to the 2 years prior to the study(p = 0.04) or during the study (p = 0.03). This suggests that citratetherapy might be clinically effective, but might require prolonged timebefore generating measurable changes. The number of ER visits notleading to hospital admissions during and after the study was alsolower than the number of ER visits per year during the 2 years prior tothe study, though the changes were not statistically significant.

4. Discussion

The objective of this study was to determine whether citrate, gluta-mine, or ornithine α-ketoglutarate were safe and effective in raisingplasma glutamine and reducing plasma ammonia levels compared tobaseline in patients with propionic acidemia. All supplements weresafe, with no adverse events attributed to them nor changes in safetylaboratory values (Table 2). Growth remained appropriate during thestudy. BMI (kg/m2) increased in patients 1 and 3, who started with aBMI lower than average, and decreased in patient 2, who started withan above average BMI (Table 1). Overall, there were no indicationsthat any of the supplements were detrimental to the patients' health.Our results in this small patient population are slightly different fromthose reported in patients with the related condition methylmalonicacidemia in which medical foods have been associated with defectivegrowth [26].

In the first part of the study, none of the supplements significantlyraised plasma glutamine or [glutamine + glutamate] concentration(Fig. 2A). Ammonia levels also did not change significantly during ther-apy with any of the 3 supplements and remained slightly elevated. De-spite lack of significant changes in glutamine levels, urine organic acids

Table 5Cognitive assessment using the Stanford Binet fifth edition in patients with propionic acidemia

Nonverbal IQ Verbal IQ

Pt. Before (percentile) After (percentile) Before (percentil

1 42 (b0.1) 42 (b0.1) 43 (b0.1)2 42 (b0.1) 42 (b0.1) 43 (b0.1)3 Child not sufficiently cooperativeAverage 42 (b0.1) 42 (b0.1) 43 (b0.1)

indicated that treatment with citrate, but not the other supplements,significantly increased the excretion of ketoglutarate, succinate and fu-marate (Krebs cycle intermediates, Fig. 2B), indicating that citrate en-tered mitochondria and was converted to subsequent intermediates inthe Krebs cycle possibly providing successful anaplerotic therapy. Forthis reason, treatment with citrate was continued over time(30 weeks) to determine its effects on clinical outcomes and biochemi-cal parameters.

Even with a longer period of time on citrate, therapy failed to signif-icantly increase glutamate or glutamine levels (Table 3). There were nochanges in the urinary excretion of methylcitric acid, 3-OH-propionicacid, or propionylglycine, metabolites that accumulate in propionicacidemia and remained markedly elevated (Fig.3). Similarly, plasmalevels of propionylcarnitine remained markedly increased, indicatingthat none of the therapies reduced the production of toxic metabolites(Fig. 3). The only changes that were observed in the plasma aminoacids were an increase in ornithine in patients receiving ornithine α-ketoglutarate and a reduction in some essential amino acids, possiblyreflecting more strict dietary protein restriction (Table 3). With respectto plasma ammonia, the inverse relationship between plasma ammoniaand glutamine levels [4]was lost during the trial (Fig. 5), suggesting thatmore substrate (ketoglutarate) may have been available to hepatocytesto conjugate ammonia escaping theurea cycle in periportal hepatocytes.At the same time, there was a striking direct correlation between am-monia and lysine levels (R2= 0.52, p b 0.0001) (Fig. 5). Lysine levels in-crease in many forms of hyperammonemia and in urea cycle defects.Since lysine is an essential amino acid, its increased levels likely reflectdecreased degradation. There are two pathways devoted to lysinebreakdown: the saccharopine and the pipecolic acid routes that ulti-mately converge at the level of al α-aminoadipic semialdehyde [27].In the saccharopine pathway, that is believed to be the main catabolicroute in extracerebral tissue [28], L-lysine is converted to saccharopineby α-aminoadipic semialdehyde synthase by condensation with α-ketoglutaric acid [27]. In the pipecolic acid pathway, transaminationalso requires an acceptor for ammonia that, in most cases, is α-ketoglutarate. In urea cycle defects, ammonia combines with α-ketoglutarate to generate glutamate and then glutamine, leading to α-ketoglutarate depletion and a secondary increase in lysine levels [29].In propionic acidemia, it would be the primary depletion of α-ketoglutarate that results in a stable increase in lysine levels [30],which further increase when residual α-ketoglutarate is conjugatedwith ammonia to generate glutamic acid (Fig. 5). If this is the mecha-nism, normalization of lysine levels would become a further indicatorof successful therapy in patients with propionic acidemia. In addition,mild restriction of dietary lysine could spare ketoglutarate foranaplerotic functions.

Most patients with propionic acidemia have developmental delays.We tested the effects of supplements on cognitive and motor delays.Our patients were so severely affected that not all testing could be com-pleted and the small sample size precluded any statistical analysis(Tables 5 and 6).

While there was no effect on cognitive or motor development, theoverall health of patients with propionic acidemia might have benefit-ted from citrate supplementation. In the 2 years prior to the study, pa-tients were hospitalized 15.3 days per year on average (Table 7).During the study, the number remained at 15.8 days per year. In the

.

Full scale IQ

e) After (percentile) Before (percentile) After (percentile)

43 (b0.1) 40 (b0.1) 40 (b0.1)43 (b0.1) 40 (b0.1) 40 (b0.1)

43 (b0.1) 40 (b0.1) 40 (b0.1)

Table 6Evaluation ofmuscle activity in patientswith propionic acidemia. Results reported at base-line and at the conclusion of the study. *Patient 3was unable to complete the evaluation ofmuscle activity. **Patient 2 did not bring herwalker on thefinal day of testing andwas un-able to perform these tasks.

Pt 1 2

Hand-heldmyometry

Composite score (lbs) Before 10.85 12.6After 14.7 12.85

Time function test 10 m walk/run (sec) Before 7.1 14.6After 5 NA**

Time to rise from supine to stand(sec)

Before 3.3 NA**After 4.2 NA**

Time to climb 4 stairs (sec) Before 4.3 12.9After 2.7 14.7

Accelerometry tests Average steps in 7 days Before 7679 2192After 5599 1891

Average minutes of activity in 7days

Before 496 247After 485 227

Average % of inactivity in 7 days Before 65.5 82.8After 66.3 84.21

58 N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

2 years post-study, all three patients remained on open label citrate sup-plements and the average number of days hospitalized per year de-creased significantly to 1.8 (p = 0.04 as compared to baseline). Thesignificant reduction in the number of hospitalization days during thistime suggests that citrate supplements might improve the generalhealth of patients with propionic acidemia, but that the effects of sup-plementsmight require time to become evident. The decreased numberof hospitalizations could bedue to factors other than therapy. In general,patients with propionic acidemia become more clinically stable as theyget older. However, the patient who improved the most (patient 2) inthis case was the oldest of the group and was already 11 years oldwhen the study started. All patients improved regardless of age. Patientsand their families also had extended contacts with the healthcare teamconsisting of physicians and dietitians, during the study. Improved edu-cation about the disease and consistent monitoring could have helpedthemanagement of propionic acidemia in the long term. Our clinical im-pression was that patients seemed to have shorter hospital stays andimproved more rapidly while on citrate supplements.

There are several limitations in our study. Our study involved onlythree patients with propionic acidemia, a rare metabolic disorder, af-fecting approximately 1:240,000 in the USA [31]. The results of thispilot study results should be extended to a larger population of patientswith a multicenter trial enabling enrollment of an adequate number ofpatients. Another factor that may have affected the results of the studymay be the dose of the supplement since a single dose was used. On amolar basis, OKG provided 2 mmol/kg ornithine and 1 mmol/kg α-ketoglutarate, glutaminewas supplemented at 2.7mmol/kg, and citratewas supplemented at 2.5 mmol/kg. Therefore, considering that orni-thine can be converted into ketoglutarate (Fig. 1), the dosage of the sup-plements was almost equivalent. However, these dosing levels may not

Table 7Hospitalizations in patients with propionic acidemia before, during and after anaplerotictherapy. Average number of days hospitalized and ER visits in patients with propionicacidemia two years prior to study, during study, and two years following study. The num-ber of days spent in the hospital was normalized per year. *p b 0.05 versus prior to studyusing analysis of variance.

Two years prior tostudy

During study Two years followingstudy

Pt Dayshospitalized

ERvisits

Dayshospitalized

ERvisits

Dayshospitalized

ERvisits

1 19 3.5 10 2.1 4.5 0.52 25 0.5 32 0 0 0.53 2 0 5.5 0 1 0Average± SD

15.3 ± 11.9 1.3 ±1.9

15.8 ± 14.2 0.7 ±1.2

1.8 ± 2.4* 0.3 ±0.3

Total 46 4 47.5 2.1 5.5 1

have been sufficient to increase glutamine levels. Specifically, the doseof citrate might have increased intermediates of the citric acid cycle,butmight have not been sufficiently high to increase glutamate and glu-tamine levels. In a fish model of propionic acidemia (the medaka fish,Oryzias latipes), supplements of sodium citrate, ornithine α-ketoglutarate, and glutamine resulted in significant improvements ofboth survival and locomotor activity [32]. For this reason, higher dosesof citrate and possibly other supplements should be tested in future pa-tients with dose-escalation.

5. Conclusion

Propionic acidemia is a rare metabolic disorder caused by deficiencyof propionyl-CoA carboxylase. Patients with propionic acidemia gener-ally have a poor prognosis due to health complications including pan-creatitis or cardiomyopathy. Deficient breakdown of propionyl-CoAmay lead to deficient levels of Krebs cycle intermediates and decreasedglutamine levels. Current treatment in patients with propionic acidemiaincludes a low protein diet with an artificial formula that does not con-tain the amino acids threonine, valine, leucine, andmethionine. Patientsreceive carnitine supplements to replace propionylcarnitine that is lostin urine. Adding an anaplerotic supplement to their treatment regimenwould be a simple, effective way to improve the prognosis.

By supplementing citrate in patients with propionic acidemia, wehave shown their ability to increase Krebs cycle intermediates throughanaplerosis and (for all supplements) to normalize the relationship be-tween ammonia and glutamate levels. This treatment, whose effectsseem incremental to those of existing treatments, appears to be safeand effective in the long term in reducing hospitalizations in patientswith propionic acidemia. Further testing of citrate therapy at a higherdose on a larger population might prove useful not only in treating pa-tients with propionic acidemia, but also patients with other organicacidemias in which there is depletion of Krebs cycle intermediates.

Conflict of interest notification

The authors have no conflicts of interest.

Grant support

National Institutes of Health grant 1 R21 DK 077415. Clinical trialNCT00645879.

Acknowledgements

This work was supported in part by NIH grant 1 R21 DK 077415.The authors thank all the patients and their families for their partic-

ipation in this complex protocol.

References

[1] N. Carrillo-Carrasco, C. Venditti, Propionic Acidemia, in: R.A. Pagon, M.P. Adam, H.H.Ardinger, S.E. Wallace, A. Amemiya, L.J.H. Bean, T.D. Bird, C.T. Fong, H.C. Mefford,R.J.H. Smith, K. Stephens (Eds.),GeneReviews(R), Seattle (WA), 1993.

[2] F. Deodato, S. Boenzi, F.M. Santorelli, C. Dionisi-Vici, Methylmalonic and propionicaciduria, Am. J. Med. Genet. C: Semin. Med. Genet. 142C (2006) 104–112.

[3] F. D., C. Dionisi-Vici, W. Röschinger, W.J. Rhead, B. Wilcken, “Classical” organicacidurias, propionic-methylmalonic- and Isovaleric-aciduria, long-term outcomeand effects of expanded newborn screening using tandemmass spectrometry, J. In-herit. Metab. Dis. 29 (2006) 383–389.

[4] H.R. Filipowicz, S.L. Ernst, C.L. Ashurst, M. Pasquali, N. Longo, Metabolic changes as-sociated with hyperammonemia in patients with propionic acidemia, Mol. Genet.Metab. 88 (2006) 123–130.

[5] L. Ierardi-Curto, P. Kaplan, S. Saitta, A. Mazur, G.T. Berry, The glutamine paradox in aneonate with propionic acidaemia and severe hyperammonaemia, J. Inherit. Metab.Dis. 23 (2000) 85–86.

[6] Z.N. Al-Hassnan, S.A. Boyadjiev, V. Praphanphoj, A. Hamosh, N.E. Braverman, G.H.Thomas, M.T. Geraghty, The relationship of plasma glutamine to ammonium andof glycine to acid-base balance in propionic acidaemia, J. Inherit. Metab. Dis. 26(2003) 89–91.

59N. Longo et al. / Molecular Genetics and Metabolism 122 (2017) 51–59

[7] M. Tuchman, M. Yudkoff, Blood levels of ammonia and nitrogen scavenging aminoacids in patients with inherited hyperammonemia, Mol. Genet. Metab. 66 (1999)10–15.

[8] C.R. Roe, L. Sweetman, D.S. Roe, F. David, H. Brunengraber, Treatment of cardiomy-opathy and rhabdomyolysis in long-chain fat oxidation disorders using ananaplerotic odd-chain triglyceride, J. Clin. Invest. 110 (2002) 259–269.

[9] F. Mochel, P. DeLonlay, G. Touati, H. Brunengraber, R.P. Kinman, D. Rabier, C.R. Roe,J.M. Saudubray, Pyruvate carboxylase deficiency: clinical and biochemical responseto anaplerotic diet therapy, Mol. Genet. Metab. 84 (2005) 305–312.

[10] J. Vockley, B. Burton, G.T. Berry, N. Longo, J. Phillips, A. Sanchez-Valle, P. Tanpaiboon,S. Grunewald, E. Murphy, R. Humphrey, J. Mayhew, A. Bowden, L. Zhang, J. Cataldo,D.L. Marsden, E. Kakkis, UX007 for the treatment of long chain-fatty acid oxidationdisorders: safety and efficacy in children and adults following 24weeks of treat-ment, Mol. Genet. Metab. 120 (2017) 370–377.

[11] O.E. Owen, S.C. Kalhan, R.W. Hanson, The key role of anaplerosis and cataplerosis forcitric acid cycle function, J. Biol. Chem. 277 (2002) 30409–30412.

[12] M.J. Gibala, D.A. MacLean, T.E. Graham, B. Saltin, Anaplerotic processes in humanskeletal muscle during brief dynamic exercise, J. Physiol. 502 (Pt. 3) (1997)703–713.

[13] H. Brunengraber, C.R. Roe, Anaplerotic molecules: Current and future, J. Inherit.Metab. Dis. 29 (2006) 327–331.

[14] C. Renner, A.C. Sewell, K. Bervoets, H. Forster, H. Bohles, Sodium citrate supplemen-tation in inborn argininosuccinate lyase deficiency: a study in a 5-year-old patientunder total parenteral nutrition, Eur. J. Pediatr. 154 (1995) 909–914.

[15] A.K. Iafolla, D.S. Gale, C.R. Roe, Citrate therapy in argininosuccinate lyase deficiency,J. Pediatr. 117 (1990) 102–105.

[16] A. Ahmad, S.G. Kahler, P.S. Kishnani, M. Artigas-Lopez, A.S. Pappu, R. Steiner, D.S.Millington, J.L. Van Hove, Treatment of pyruvate carboxylase deficiency with highdoses of citrate and aspartate, Am. J. Med. Genet. 87 (1999) 331–338.

[17] D.E. Matthews, M.A. Marano, R.G. Campbell, Splanchnic bed utilization of glutamineand glutamic acid in humans, Am. J. Phys. 264 (1993) E848–E854.

[18] S.C. Kalhan, P.S. Parimi, L.L. Gruca, R.W. Hanson, Glutamine supplement with paren-teral nutrition decreases whole body proteolysis in low birth weight infants, J.Pediatr. 146 (2005) 642–647.

[19] V.M. Aquino, A.R. Harvey, J.H. Garvin, K.T. Godder, M.L. Nieder, R.H. Adams, G.B.Jackson, E.S. Sandler, A double-blind randomized placebo-controlled study of oralglutamine in the prevention of mucositis in children undergoing hematopoieticstem cell transplantation: a pediatric blood and marrow transplant consortiumstudy, Bone Marrow Transplant. 36 (2005) 611–616.

[20] L. Cynober, C. Coudray-Lucas, J.P. de Bandt, J. Guechot, C. Aussel, M. Salvucci, J.Giboudeau, Action of ornithine alpha-ketoglutarate, ornithine hydrochloride, and

calcium alpha-ketoglutarate on plasma amino acid and hormonal patterns inhealthy subjects, J. Am. Coll. Nutr. 9 (1990) 2–12.

[21] L.A. Cynober, The use of alpha-ketoglutarate salts in clinical nutrition and metaboliccare, Curr. Opin. Clin. Nutr. Metab. Care 2 (1999) 33–37.

[22] A.A. Moukarzel, O. Goulet, J.S. Salas, C. Marti-Henneberg, A.L. Buchman, L. Cynober,R. Rappaport, C. Ricour, Growth retardation in children receiving long-term totalparenteral nutrition: effects of ornithine alpha-ketoglutarate, Am. J. Clin. Nutr. 60(1994) 408–413.

[23] C. Coudray-Lucas, H. Le Bever, L. Cynober, J.P. De Bandt, H. Carsin, Ornithine alpha-ketoglutarate improves wound healing in severe burn patients: a prospective ran-domized double-blind trial versus isonitrogenous controls, Crit. Care Med. 28(2000) 1772–1776.

[24] S.E. Moller, Quantification of physiological amino acids by gradient ion-exchangehigh-performance liquid chromatography, J. Chromatogr. 613 (1993) 223–230.

[25] K. Tanaka, A. West-Dull, D.G. Hine, T.B. Lynn, T. Lowe, Gas-chromatographic methodof analysis for urinary organic acids. II. Description of the procedure, and its applica-tion to diagnosis of patients with organic acidurias, Clin. Chem. 26 (1980)1847–1853.

[26] I. Manoli, J.G. Myles, J.L. Sloan, N. Carrillo-Carrasco, E. Morava, K.A. Strauss, H.Morton, C.P. Venditti, A critical reappraisal of dietary practices in methylmalonicacidemia raises concerns about the safety of medical foods. Part 2: cobalamin C de-ficiency, Genet. Med. (2015).

[27] S.M. Houten, H. Te Brinke, S. Denis, J.P. Ruiter, A.C. Knegt, J.B. de Klerk, P.Augoustides-Savvopoulou, J. Haberle, M.R. Baumgartner, T. Coskun, J. Zschocke,J.O. Sass, B.T. Poll-The, R.J. Wanders, M. Duran, Genetic basis of hyperlysinemia,Orphanet J. Rare Dis. 8 (2013) 57.

[28] A. Hallen, J.F. Jamie, A.J. Cooper, Lysine metabolism in mammalian brain: an updateon the importance of recent discoveries, Amino Acids 45 (2013) 1249–1272.

[29] P. Kamoun, V. Richard, D. Rabier, J.M. Saudubray, Plasma lysine concentration andavailability of 2-ketoglutarate in liver mitochondria, J. Inherit. Metab. Dis. 25(2002) 1–6.

[30] S. Scholl-Burgi, J.O. Sass, J. Zschocke, D. Karall, Amino acid metabolism in patientswith propionic acidaemia, J. Inherit. Metab. Dis. 35 (2012) 65–70.

[31] B.L. Therrell Jr., M.A. Lloyd-Puryear, K.M. Camp, M.Y. Mann, Inborn errors of metab-olism identified via newborn screening: ten-year incidence data and costs of nutri-tional interventions for research agenda planning, Mol. Genet. Metab. 113 (2014)14–26.

[32] V. Ginocchio, E. DeFelice, F.T. Barrows, R.M. Sepe, P. Sordino, F.G. Salierno, I. Conte, N.Brunetti-Pierri, A fish model for propionic acidemia: increased survival and im-provement of neurological phenotype by anaplerotic diet, J. Inherit. Metab. Dis. 39(Suppl. 1) (2016) 0–022.