NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Adam MP, Ardinger HH, Pagon RA, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2018.

Cover of GeneReviews®

GeneReviews® [Internet].

Show details

Glycine Encephalopathy

Synonyms: NKH, Nonketotic Hyperglycinemia

, MD, PhD, , MS, MBe, and , MD, PhD.

Author Information

Initial Posting: ; Last Update: July 11, 2013.


Clinical characteristics.

Glycine encephalopathy, also known as nonketotic hyperglycinemia (NKH), is an inborn error of glycine metabolism defined by deficient activity of the glycine cleavage enzyme and, as a consequence, accumulation of large quantities of glycine in all body tissues including the brain. The majority of glycine encephalopathy presents in the neonatal period (85% as the neonatal severe form and 15% as the neonatal attenuated form). Of those presenting in infancy, 50% have the infantile attenuated form and 50% have the infantile severe form. Overall, 20% of all children presenting as either neonates or infants have a less severe outcome, defined as developmental quotient greater than 20. A minority of patients have mild or atypical forms of glycine encephalopathy. The neonatal form manifests in the first hours to days of life with progressive lethargy, hypotonia, and myoclonic jerks leading to apnea and often death. Surviving infants have profound intellectual disability and intractable seizures. The infantile form is characterized by hypotonia, developmental delay, and seizures. The atypical forms range from milder disease, with onset from late infancy to adulthood, to rapidly progressing and severe disease with late onset.


Glycine encephalopathy is suspected in individuals with elevated glycine concentration in blood and CSF. An increase in CSF glycine concentration together with an increased CSF-to-plasma glycine ratio suggests the diagnosis. Enzymatic confirmation of the diagnosis relies on measurement of glycine cleavage system (GCS) enzyme activity in liver obtained by open biopsy or autopsy. The majority of affected individuals have no detectable enzyme activity. The three genes in which biallelic pathogenic variants are known to cause glycine encephalopathy are: GLDC (encoding the P-protein component of the GCS complex and accounting for 70%-75% of disease), AMT (encoding the T-protein component of the GCS complex and accounting for ~20% of disease), and GCSH (encoding the H-protein component of the GCS complex and accounting for <1% of disease). About 5% of individuals with enzyme-proven glycine encephalopathy do not have a pathogenic variant in any of these three genes and have a variant form of glycine encephalopathy.


Treatment of manifestations: No effective treatment exists for severe glycine encephalopathy; preliminary evidence suggests that children with pathogenic variants associated with residual GCS enzyme activity treated aggressively in the first two years of life with sodium benzoate to reduce plasma concentration of glycine and N-methyl D-aspartate (NMDA) receptor site antagonists have improved outcome compared to late-treated or untreated controls. Other: antiepileptic drugs and/or ketogenic diet for seizure control; gastrostomy tube for feeding problems; therapy for gastroesophageal reflux; physical therapy.

Prevention of secondary complications: For those on sodium benzoate, monitor plasma carnitine concentration and supplement with carnitine as needed.

Surveillance: Neurologic assessments in the first year to identify spasticity in severely affected infants and chorea in more mildly affected infants; developmental assessment routinely in infancy/childhood; monitoring for scoliosis, swallowing dysfunction, and spasticity in severely affected individuals; monitoring those on benzoate treatment with serial plasma glycine concentrations.

Agents/circumstances to avoid: Valproate, as it increases blood and CSF glycine concentrations and may increase seizure frequency and even result in coma in attenuated glycine encephalopathy.

Evaluation of relatives at risk: Testing of at-risk sibs to promote early diagnosis and treatment.

Genetic counseling.

Glycine encephalopathy is inherited in an autosomal recessive manner. At conception, each sib of an affected individual has a 25% chance of being affected, a 50% chance of being an asymptomatic carrier, and a 25% chance of being unaffected and not a carrier. De novo pathogenic variants occur in approximately 1% of individuals with glycine encephalopathy; thus, carrier status in parents should be confirmed by molecular genetic testing rather than be assumed or inferred. Most individuals with glycine encephalopathy do not reproduce. Carrier testing for at-risk relatives and prenatal testing for pregnancies at increased risk are possible if the pathogenic variants in the family are known.


Neonates or infants presenting with hypotonia, unexplained coma, and seizures and children with seizures, hypotonia, and developmental delay should be suspected of having glycine encephalopathy (nonketotic hyperglycinemia, NKH).

If glycine encephalopathy is clinically suspected, the following diagnostic testing algorithm is recommended:

  • First-tier testing should consist of biochemical screening including measurement of glycine levels in both plasma and CSF. Urine organic acid analysis should also be performed to exclude ketotic hyperglycinemia. Elevated glycine levels in plasma and CSF constitute a high likelihood of glycine encephalopathy and confirmatory tests should be pursued (see details).
  • Second-tier testing consists of molecular genetic testing of the constituent genes GLDC, AMT, and GCSH (see details).
  • Third-tier testing in patients with an unresolved suspicion of glycine encephalopathy consists of measurement of the GCS enzyme activity following a liver biopsy (see details).

First Tier Details

Quantitative amino acid analysis. Measurement of glycine concentration in CSF and plasma samples obtained simultaneously is advised to establish the diagnosis of glycine encephalopathy. An isolated elevation of CSF glycine and resulting abnormal CSF-to-plasma glycine ratio can suggest the diagnosis of glycine encephalopathy (see Table 1).

Note: (1) In CSF the serine concentration can be low, and the threonine concentration should not be elevated. (2) To establish the diagnosis, the CSF must be completely free of contamination by blood or serum (which is not visible to the eye), as evidenced by a normal RBC (red blood cell) count and protein concentration. (3) The presence of blood or elevated protein in the CSF invalidates the results.

Table 1.

CSF and Plasma Glycine Concentration (µmol/L) in Glycine Encephalopathy

Glycine Encephalopathy PhenotypeNormal Control
Neonatal FormAtypical Form
CSF glycine concentration>80 µmol/L>30 µmol/L<20 µmol/L 1
Plasma glycine concentrationVaries 2Varies 2125-450 1, 3
CSF/plasma glycine ratio 4>0.080.04-0.2<0.02

Normal values vary with age. Both CSF and plasma glycine concentrations are higher in the neonatal period and decrease rapidly in the first months of life (e.g., at age >1 year, normal values for CSF glycine concentration are <12 µmol/L and for plasma glycine concentration are <350 µmol/L).


Can be normal [Steiner et al 1996]


Samples must be obtained simultaneously.

Note: Current expanded newborn screening (NBS) using amino acid analysis by tandem mass spectrometry detects a fraction of the severely affected neonates, but has a high false positive rate and, thus, is not an effective screening method at this time [Tan et al 2007]. No newborn screening centers offer newborn screening based on glycine levels.

Urine organic acid analysis. Urine organic acid profile obtained simultaneously with the plasma and CSF glycine concentration is expected to be normal. Small elevations of multiple acylglycine esters can occasionally be noticed.

Second Tier Details

Molecular genetic testing. GLDC, AMT, and GCSH are the three genes in which biallelic pathogenic variants are known to cause glycine encephalopathy.

Currently the most common testing strategy is to perform concurrent testing of all three genes. In the absence of pathogenic variants identified through traditional sequencing, targeted deletion analysis for intragenic deletions within these genes must be performed, particularly given the high frequency of exon deletions in GLDC.

Note: (1) When two apparent disease-causing alleles have been identified in a proband, parental samples should be obtained in a timely manner to ensure proper identification of both alleles. Patients who appeared to have two GLDC pathogenic variants or two AMT pathogenic variants have turned out to have both pathogenic variants on the same parental allele with an unrecognized deletion on the second allele [Author, personal observation], a situation which can cause serious problems during prenatal testing unless recognized ahead of time. (2) For patients with a variant of unknown significance, enzyme testing is the recommended next step. (3) Up to 5% of persons with deficient glycine cleavage enzyme activity do not have a pathogenic variant in any of the three genes known to cause glycine encephalopathy. These individuals are candidates for pathogenic variants in genes encoding proteins involved in the addition of the cofactors lipoate and pyridoxal-P (see Differential Diagnosis) and in the transport of glycine into the astrocyte (GLYT1) where the enzyme is located.

Table 2.

Molecular Genetic Testing Used in Glycine Encephalopathy

Gene 1Proportion of Glycine Encephalopathy Attributed to Pathogenic Variants in This GeneTest MethodVariants Detected 2
GLDC70%-75% 3Sequence analysis 4Sequence variants
Deletion/duplication analysis 5Exon and whole-gene deletions 6, 7
AMT20%Sequence analysis 4Sequence variants
Deletion/duplication analysis 5Exon and whole-gene deletions 8
GCSH<1% 9Sequence analysis 4Sequence variants
Deletion/duplication analysis 5Exon and whole-gene deletions 8

See Molecular Genetics for information on allelic variants.


Sequence analysis detects variants that are benign, likely benign, of uncertain significance, likely pathogenic, or pathogenic. Pathogenic variants may include small intragenic deletions/insertions and missense, nonsense, and splice site variants; typically, exon or whole-gene deletions/duplications are not detected. For issues to consider in interpretation of sequence analysis results, click here.


Testing that identifies deletions/duplications not readily detectable by sequence analysis of the coding and flanking intronic regions of genomic DNA; a variety of methods including quantitative PCR, multiplex ligation-dependent probe amplification (MLPA), or targeted chromosomal microarray analysis (gene/segment-specific) may be used. A full chromosomal microarray analysis that detects deletions/duplications across the genome may also rarely include this gene/segment, but typically is not sufficient to detect the majority of deletions/duplications associated with glycine encephalopathy.


Approximately 20% of GLDC mutated alleles are (multi)exon deletions or duplications [Kanno et al 2007]. These occur on various haplotypes and in different ethnic groups. They appear to arise as a result of non-homologous allelic recombination of Alu repeats.


Sensitivity is increasing as new test methods (e.g., targeted chromosomal microarray analysis) to detect deletions are developed.


Very rarely, deletions or duplications involving AMT have been identified; no deletions or duplications in GCSH have been identified as causative of glycine encephalopathy. (Note: By definition, deletion/duplication analysis identifies rearrangements that are not identifiable by sequence analysis of genomic DNA.)


One individual was identified with deficient H-protein enzyme activity in 1981; however, no pathogenic variant was identified in the comprehensive analysis of pathogenic variants by Kure et al [2006a].

Third Tier Details

Enzyme testing: glycine cleavage system (GCS) enzyme activity. The diagnosis of glycine encephalopathy may be confirmed by measurement of GCS enzyme activity in liver. In its major degradative pathway, glycine is metabolized by the GCS (see Figure 1).

Figure 1. . Metabolism of glycine by glycine cleavage enzyme (GCS).

Figure 1.

Metabolism of glycine by glycine cleavage enzyme (GCS). Glycine enters the four-protein enzyme complex at the lower left, where it is decarboxylated by the P-protein (also known as glycine decarboxylase). A defect of the P-, H-, or T-proteins of this (more...)

Reliable enzymatic confirmation of the diagnosis of glycine encephalopathy requires 80 mg of liver to measure the activity of the glycine cleavage enzyme GCS. Liver is usually obtained by surgical endoscopy as a wedge biopsy or as soon as possible at autopsy. The enzyme is labile and rapid processing and deep freezing are essential for proper enzyme assay.

  • The vast majority of affected individuals have no detectable enzyme activity.
  • Patients with a T-protein defect tend to have activity up to 25% of normal values. Conversely, 50% of individuals with residual GCS enzyme activity in liver have AMT pathogenic variants [Toone et al 2003].
  • Patients with a P-protein defect do not tend to have residual activity [Toone et al 2000] except for mildly affected patients. The highest residual activity of an affected individual tested in one series is 0.4 units (normal 2.1-11.9) [Steiner et al 1996].

Note: Although enzymatic confirmation of glycine encephalopathy using Epstein-Barr virus cultured lymphoblasts from peripheral blood samples was reported in six individuals with P-protein defects, others have obtained overlapping GCS enzyme activity in both controls and individuals with glycine encephalopathy, making this method unreliable [Applegarth et al 2000b].

Other Testing (not included in the Algorithm)

Identification of specific protein deficiency: glycine exchange reaction. The glycine exchange reaction measures the combined activity of P- and H-proteins without the need for T-protein activity (see Figure 1) and thus can be used to distinguish between those with a P-protein defect (~70% of affected individuals) and those with a T-protein defect (≤20% of affected individuals) [Toone et al 2000]. Thus, this test can help determine the order in which to test specific genes if that approach is used. H-protein deficiency is rare, having been documented in only one individual in whom H-protein was found to be devoid of lipoic acid [Hiraga et al 1981].

Note: (1) 200 mg of liver, which can be obtained by surgical endoscopic wedge biopsy, is required to perform the glycine exchange assay and the glycine cleavage enzyme assay. (2) The order in which testing is used in establishing the diagnosis of a symptomatic individual greatly differs according to clinical circumstances. For example, when the patient is deceased and a rapid postmortem liver biopsy is available, the first test is GCS enzyme assay to confirm the diagnosis followed by glycine exchange assay to guide molecular genetic testing.

13C-glycine breath test. The 13C-glycine breath test shows promise in rapid confirmation of the diagnosis of glycine encephalopathy [Kure et al 2006b]. No data on the differential diagnostic value currently exist.

Magnetic resonance spectroscopy (MRS). MRS has been used for non-invasive measurement of brain glycine levels in patients with glycine encephalopathy [Heindel et al 1993, Gabis et al 2001]. The use of MRS has been suggested to aid in the diagnosis of patients with glycine encephalopathy. No data on the differential diagnostic value of MRS currently exists.

Clinical Characteristics

Clinical Description

Glycine encephalopathy, also known as nonketotic hyperglycinemia (NKH), is an inborn error of glycine metabolism defined by deficiency of the glycine cleavage enzyme and, as a consequence, accumulation of large quantities of glycine in all body tissues including the brain.

Children with the classic form of the disease can present in the neonatal period or in infancy. They can progress to develop a severe form of the disease (developmental quotient [DQ] <20) or a milder attenuated form of the disease (DQ >20). Classification of the typical disease is thus based on both age at initial presentation and ultimate clinical outcome and includes the following categories: neonatal severe, neonatal attenuated, infantile attenuated, and infantile severe [Hennermann 2006, Hennermann et al 2012]. Very mildly affected persons are seen, usually with later presentation. Some (rare) atypical forms exist.

The majority of individuals with glycine encephalopathy present as neonates. Of those presenting as neonates, 85% have the neonatal severe form and 15% have the neonatal attenuated form. Of those presenting in infancy, 50% have the infantile attenuated form and 50% have the infantile severe form. Overall, 20% of all infants presenting in either the neonatal or the infantile period have a mild outcome [Hoover-Fong et al 2004, Hennermann et al 2012].

Neonatal (classic) presentation. The neonatal form of glycine encephalopathy manifests in the first hours to days of life with progressive lethargy, hypotonia, and myoclonic jerks leading to apnea and often death if not supported by intubation and ventilation. The vast majority of infants regain spontaneous respiration within the first three weeks of life. Many, but not all, show some spontaneous improvement in alertness in the first month of life. They develop further intellectual disability and seizures (see Severe outcome and Attenuated outcome). The initial EEG usually shows a burst-suppression pattern that evolves into hypsarrhythmia and/or multifocal spikes by about age three months.

Infantile presentation. Infants do not have lethargy and coma in the neonatal period, but often have a history of hypotonia from early on. They present with developmental delay and infantile-onset seizures that can be mild or increasingly difficult to treat.

Severe outcome. Affected individuals do not make developmental progress or at most learn to sit with limited interaction with their environment (DQ <20). They develop increasingly difficult-to-treat seizures over the first year of life, usually requiring multiple anticonvulsants. They develop progressive and early spasticity with positive Babinski sign before age six months. They have a tendency to develop scoliosis, and often have swallowing dysfunction requiring tube feeding. Occasionally patients are found to have club feet [Hennermann et al 2012].

Attenuated outcome. Affected individuals make developmental progress to a variable degree (DQ 20-65). They can learn to walk, have sign language, and interact with caregivers and attend special education classes. In a recent retrospective study of 65 individuals with NKH, up to 20% of surviving children were able to walk and say or sign words [Hoover-Fong et al 2004]. They have little spasticity. They develop a seizure disorder, which is often relatively easy to treat with either benzoate or dextromethorphan alone or with the addition of a single anticonvulsant [Van Hove et al 2005] (see Management). Individuals tend to be hyperactive [Wiltshire et al 2000, Hennermann 2006]. They usually have choreic movements, which is a good prognostic sign [Hall & Ringel 2004, Hennermann 2006, Hennermann et al 2012].

Individuals can have intermittent episodes of severe lethargy; some of these cases have been reported in the past as a mild episodic form. One adult with the mild episodic form experienced acute decompensation while on valproate (which is contraindicated in this disorder because it raises serum glycine concentration) [Hall & Ringel 2004] (see Management).

Mild outcome. Rare individuals with a very mild form of glycine encephalopathy have been described. They often present after age one year, have an IQ of >60, and sometimes can attend normal class in school. They have attention deficit and hyperactivity disorder (ADHD), typically have no (or rare) seizures, and can have episodes of severe lethargy with infections [Brunel-Guitton et al 2011].

The best outcome ever reported is normal intelligence, observed in individuals with a genotype associated with residual enzyme activity (p.Ala802Val) who received early and aggressive treatment in the first two years of life [Korman et al 2004].

The recognition of the episodes of lethargy led to the description of the mild episodic form, reported in four children with mild intellectual disability and episodes of chorea, agitated delirium, and vertical gaze palsy associated with febrile illness [Steiner et al 1996]. One individual had confirmed GCS deficiency [Steiner et al 1996].

Note: The clinical spectrum of the condition is continuous and occasionally patients have an outcome intermediate between severe and attenuated.

Other findings

  • A number of patients have had delayed gastric emptying and poor gastrointestinal motility, leading to very severe problems in at least one.
  • Several patients have reported dysuria with difficulty emptying the bladder. It is unclear if this is a side effect of dextromethorphan or a manifestation of the disorder [Van Hove, unpublished observations].
  • MRI may be normal. In almost all neonatal severe cases, diffusion-weighted MR images show high-signal lesions in symmetric weight matter consistent with vacuolating myelin. The lesions are present in myelin that is already myelinated at birth, such as the posterior limb of the internal capsule and long tracts in the brain stem [Khong et al 2003, Sener 2003, Seo et al 2003].
  • In infants with neonatal presentation and severe outcome, brain malformations may be observed, the most common of which is thinning or agenesis of the corpus callosum. Delayed myelination and atrophy are later findings. Less common findings include retrocerebellar cyst with subsequent development of hydrocephalus [Van Hove et al 2000], gyral simplification [Van Hove, unpublished observations], and rarely, gyral malformations.
  • MRS can show an elevated peak consistent with glycine on long echo time, usually present in those with a severe outcome.

Atypical forms range from milder disease, with onset from late infancy to adulthood, to rapidly progressing and severe disease with late onset.

  • The late-onset form has been described in seven individuals: four had spastic paraparesis and optic atrophy without seizures or cognitive impairment, two had mild intellectual disability with choreoathetosis, and one had severe intellectual disability and rare seizures.
  • Progressive neurologic deterioration with pulmonary hypertension. Two unrelated children who developed pulmonary hypertension with progressive neuronal deterioration and vacuolating leukoencephalopathy died before age 18 months. Both had normal EEGs and no seizures. GCS enzyme activity in frozen liver was abnormal, but no pathogenic variants in GLDC, AMT, or GCSH were identified [Riudor et al 2001, del Toro et al 2006]. Pathogenic variants in NFU1 were later identified in these patients [Navarro-Sastre et al 2011].

Intrafamilial variability. Discordance between affected children in the same family was reported for age at presentation, the clinical severity of disease, and/or CSF glycine concentration [Applegarth et al 2000b]; however, more recently a retrospective study showed a consistent phenotype within seven families with two or more affected children [Hoover-Fong et al 2004]. The familial concordance for outcome has been observed in several additional families [Wiltshire et al 2000; Van Hove, Scharer, Hennermann, unpublished observations].

Genotype-Phenotype Correlations

Preliminary data suggest that the degree to which the pathogenic variants impair enzyme activity predicts the outcome of the disease, but not for the age at presentation. Pathogenic variants associated with residual enzyme activity appear to be associated with attenuated or mild outcome, and two pathogenic variants with no residual enzyme activity appear to be associated with severe outcome [Author, personal observation]. However, the usually private nature of many pathogenic variants and the early demise without evaluation of ultimate neurodevelopmental outcome makes genotype-phenotype correlations difficult.


The birth incidence of glycine encephalopathy is 1:55,000 newborns in Finland (1:12,000 in an area of Northern Finland) and 1:63,000 in British Columbia, Canada [Applegarth et al 2000a].

Because it is often the finding of an increased plasma glycine concentration that triggers the request for measurement of CSF glycine concentration, glycine encephalopathy may be underdiagnosed.

Several different pathogenic variants in GLDC and AMT account for the high incidence of glycine encephalopathy in several small consanguineous Arab villages in Israel [Boneh et al 2005, Korman et al 2005]. A recurrent variant is present in New Zealand Maori [Van Hove et al 2007, Wilson et al 2007] and in the southern part of the Netherlands [Van Hove et al 2007].

The calculated carrier frequency is approximately 1:125 in the British Columbia, Canada population (predominantly a population of northern European origin at the time of data collection for disease incidence).

Differential Diagnosis

Transient glycine encephalopathy was the term used initially to describe the findings in seven neonates [Boneh et al 1996] who presented with seizures and/or a burst suppression pattern on EEG. All had the biochemical features of glycine encephalopathy, which then resolved. Five developed normally; two had neurologic impairment. GCS enzyme activity was not measured in these children.

Aliefendioğlu et al [2003] reported two additional unrelated children who were born to consanguineous parents; one had a positive family history. At follow-up plasma glycine concentrations were normal; CSF glycine concentrations were not determined in either child. In the child with the positive family history, psychomotor development was severely retarded at age eight months; the other child died of pneumonia at age three months.

Three infants with transient glycine encephalopathy and heterozygosity for either GLDC or GCSH pathogenic variants were reported by Kure et al [2002]. One was subsequently shown to have persistent glycine encephalopathy with severe seizures and intellectual disability [Hamosh & Van Hove 2003].

Currently the diagnosis of transient glycine encephalopathy requires that subsequent CSF and plasma glycine concentrations be normal when the affected individual is not on medication that may lower glycine levels, whereas the diagnosis of glycine encephalopathy requires identification of biallelic pathogenic variants in GLDC, AMT, or GCSH or deficient GCS enzyme activity. The diagnosis of transient glycine encephalopathy remains a rare and controversial diagnosis.

In one patient with transient glycine encephalopathy, the GCS enzyme activity was normal [Lang et al 2008] and in another patient no pathogenic variants were identified in AMT, GLDC, or GCSH. A large review study identified elevated glycine levels consistent with transient glycine encephalopathy in a series of patients with varying diagnoses such as hypoxic-ischemic injury [Aburahma et al 2011]. These data indicate that transient glycine encephalopathy is a biochemical phenocopy of glycine encephalopathy that does not constitute glycine encephalopathy.

Hyperglycinemia. The differential diagnosis for hyperglycinemia includes ketotic hyperglycinemia and valproate treatment, both of which cause a secondary decrease in liver GCS enzyme activity and can mimic the laboratory findings of glycine encephalopathy.

Ketotic hyperglycinemia can be seen in propionic acidemia, methylmalonic acidemia, isovaleric acidemia, and ß-ketothiolase deficiency.

A single patient with glyceric aciduria was reported with elevated glycine levels and deficient GCS enzyme activity. This finding was not present in any subsequent patients with glyceric aciduria, and may be a coincidental finding.

All of the foregoing can be distinguished from glycine encephalopathy by determination of urine organic acids by gas chromatography/mass spectrometry. The diagnosis of glycine encephalopathy cannot be established by amino acid analysis of plasma or CSF alone in an individual receiving valproate, which can reversibly increase the CSF glycine concentration to more than 60 µmol/L in persons who do not have glycine encephalopathy [Jaeken & Van Hove, unpublished observations].

Several neonates have been shown on newborn screening to have persistent very high concentrations of glycine in serum (>1000 µmol/L). The infants have been asymptomatic, although long-term follow up is not yet available. CSF glycine concentration measured in one child was normal. The cause for this asymptomatic presentation is currently unknown.

Hyperglycinuria. Hyperglycinuria can be seen in type I or type II hyperprolinemia, familial iminoglycinuria, and benign hyperglycinuria, which is a common transient finding caused by immaturity of renal glycine reabsorption. Patients who are homozygous for a null allele in the high affinity transporter for proline, hydroxyproline, and glycine in the proximal renal tubule SLC36A2 have iminoglycinuria, whereas heterozygotes (carriers) can have isolated glycinuria. Persons with a pathogenic variant in SLC36A2 and a polymorphism in SLC6A18 may also have glycinuria [Bröer et al 2008]. These conditions are generally asymptomatic and the previously reported association with symptoms likely reflects ascertainment bias [Coşkun et al 1993, Bröer et al 2008].

PNPO deficiency. Persons with pathogenic variants in PNPO, which encodes pyridox(am)ine 5’-phosphate oxidase, have deficient glycine cleavage enzyme activity resulting from insufficient co-factor pyridoxal-5-phosphate for the P-protein. They present in the neonatal period with severe seizures, burst suppression pattern, and coma; apnea may be present, providing a phenotypic overlap. Patients respond clinically to treatment with pyridoxal-5’-phosphate (but not to pyridoxine). Biochemically they can have elevated glycine in both serum and CSF (usually more in serum than CSF, but this phenotype is rapidly expanding), elevated threonine concentration in plasma and CSF, and sometimes decreased monoamine metabolites [Clayton et al 2003, Mills et al 2005, Hoffmann et al 2007]. The concentration of pyridoxal-5’-phosphate in CSF is very low.

Neonatal seizures. The differential diagnosis for neonatal seizures is wide and involves several genetic and metabolic conditions [Van Hove & Lohr 2011]. It includes:

  • Peroxisome biogenesis disorders, Zellweger syndrome spectrum (including the phenotypes of neonatal adrenoleukodystrophy and infantile Refsum syndrome), all of which cause an increase in plasma concentration of very long chain fatty acids;
  • Sulfite oxidase deficiency and molybdenum cofactor deficiency, which are suggested by very low concentration of cysteine on quantitative plasma amino acid determination and low concentration of serum uric acid as well as elevated urine sulfites;
  • Pyridoxine-dependent seizures and pyridoxal-P-responsive encephalopathy, both of which may present with a neonatal epileptic encephalopathy and sometimes with elevated glycine levels;
  • Phosphoglycerate dehydrogenase deficiency, a disorder of serine metabolism.

Several genetic conditions, such as pathogenic variants in CDKL5, represent a frequent cause of neonatal seizures. Because of the numerous causes of neonatal seizures, multigene panels are often used in their diagnostic evaluation.

Unrecognized perinatal hypoxic-ischemic injury can present with progressive neonatal coma and seizures. The breakdown of the blood-brain barrier can result in seepage of serum into the CSF with elevation of CSF glycine concentration and an elevated CSF: plasma glycine ratio, but also with elevated CSF protein concentration.


Evaluations Following Initial Diagnosis

To establish the extent of disease and needs in an individual diagnosed with glycine encephalopathy, the following evaluations are recommended if they have not already been completed:

  • MRI of the brain in neonates. The presence of brain malformations is a prognostic indicator of severe outcome. Abnormalities in the posterior fossa should raise suspicion for the development of hydrocephalus. In a single case study, Mourmans et al [2006] showed that serial follow up of the diffusion-weighted images/apparent diffusion coefficient (ADC) and the diffusion tensor images can provide evidence for the progressive loss of axonal structures seen in children with severe outcome.
  • EEG. The presence of a burst suppression pattern, rather than multifocal seizures, is a poor prognostic indicator.
  • Developmental assessment throughout the first years of life
  • Neurologic assessment in the first year to identify early development of spasticity in severely affected patients and of chorea in more mildly affected patients
  • Other. Consultation with a clinical geneticist and/or genetic counselor

Treatment of Manifestations

No effective treatment for severe glycine encephalopathy exists. Circumstantial evidence is however accumulating that in patients with pathogenic variants associated with residual enzyme activity (although not in patients with two severe pathogenic variants [Korman et al 2006]) early and aggressive treatment (benzoate and NMDA receptor blockade) results in improved neurodevelopmental outcome [Korman et al 2004; Flusser et al 2005; Van Hove et al, unpublished].

Current treatment of glycine encephalopathy consists of reduction of plasma concentration of glycine through treatment with sodium benzoate and blocking of glycinergic receptors, most commonly at the N-methyl D-aspartate (NMDA) receptor site.

Sodium benzoate. Oral administration of sodium benzoate at doses of 250-750 mg/kg/day can reduce the plasma glycine concentration into the normal range (see Table 1); however, this treatment does not normalize CSF glycine concentration. Lowering the plasma glycine concentration into the low normal range (between 120 and 300 µmol/L) is often required in order to observe beneficial effects. Benzoate is also a useful anticonvulsant agent in this disorder; it increases alertness. In more mildly affected individuals, it may also improve behavior.

In patients with the attenuated or mild phenotype, benzoate eliminates the intermittent episodes of severe lethargy often seen in this group. However, in patients with the severe phenotype, even high doses of benzoate administered early in the disease course do not affect the natural progression toward severe intellectual disability/seizure disorder.

The dose of sodium benzoate required varies and thus must be tailored to the individual patient. The dose should be gradually increased (usually by 50 mg/kg/day) until the plasma glycine concentration is within treatment range. The higher dose of this range (500-750 mg/kg/day) is frequently associated with gastritis, which may require oral administration of antacids, H2 antagonists, or proton pump inhibitors.

  • Patients with a mild form of the disease require a lower dose (200-450 mg/kg/day).
  • Patients with a severe form of the disease require a higher dose (550-750 mg/kg/day) [Van Hove et al 2005].

Dosing of benzoate in excess of the individual requirement is dangerous: benzoate toxicity has high morbidity and mortality [Van Hove et al 2005]. Hypocalcemia and low plasma glycine concentration (<150 µmol/L) are frequent early signs of benzoate overdose. Measurement of plasma benzoate concentration can be helpful in evaluating potential toxicity.

NMDA receptor site antagonists. Antagonists at the NMDA receptor site include dextromethorphan, ketamine, and felbamate. Use of each has resulted in improved seizure control.

In contrast, antagonism of a presumably overstimulated N-methyl-D-aspartate (NMDA) receptor channel complex with use of dextromethorphan, ketamine, or felbamate has been of limited benefit to the ultimate neurodevelopmental outcome of severely affected children; this may be different in mildly affected children.

Dextromethorphan doses commonly range from five to 15 mg/kg/day, but individual variability is substantial. Blood concentration can be monitored; the therapeutic level is not defined, but should be greater than 0 and lower than 100 nmol/L. Overdose of dextromethorphan causes increased sleepiness and while awake, more movement.

Note: Cimetidine slows the metabolism of dextromethorphan and should not be used in dextromethorphan slow metabolizers (a separate pharmacogenetic phenotype) as it may cause toxicity.

Seizure control is important for symptomatic benefit.

Control of severe seizure disorders such as burst suppression pattern or hypsarrhythmia tends to result in improved developmental progress.

Children with myoclonic seizures, such as newborns and infants, may benefit from benzodiazepines.

Standard antiepileptic drugs (AEDs) such as phenobarbital or phenytoin have limited efficacy for control of the seizures of glycine encephalopathy by themselves in neonates with this condition. However, because the nature of the epilepsy changes in late infancy, phenobarbital is often useful in seizures in older affected children.

Various antiepileptic drugs have been used with variable success. Felbamate has been successfully used in some children with difficult-to-treat seizures. This treatment must be closely monitored for signs of liver or hematopoietic toxicity.

Ketogenic diet has been used in some patients with variable success for the treatment of seizures. Ketogenic diet always lowers the amount of glycine substantially and the dose of benzoate should be reduced accordingly to avoid benzoate toxicity [Cusmai et al 2012].

For some older patients with severe glycine encephalopathy and difficult-to-control seizures, a vagal nerve stimulator has been used with varying (sometimes very high) levels of success [Tsao 2010].

Other. Gastrostomy tube placement should be considered early in the management of patients with swallowing dysfunction associated with severe disease.

Gastroesophageal reflux is common, and a Nissen procedure can be very helpful in reducing the risk of aspiration pneumonia.

Most affected individuals need physical therapy.

Scoliosis is managed with standard techniques.

Note: Although strychnine improves tone and respiration, its use has been abandoned because of serious side effects that result from its long-term use.

Prevention of Primary Manifestations

There is no way to prevent primary manifestations of this disorder. Note: The effect of therapy in mildly affected children with pathogenic variants that result in residual enzyme activity is still under investigation.

Prevention of Secondary Complications

Patients on sodium benzoate should have plasma carnitine concentration monitored. Those with free carnitine below the lower limit of normal should receive supplementation to maintain normal plasma concentrations.


Developmental assessment should be performed throughout the first years of life.

Neurologic assessments in the first year can identify early development of spasticity in severely affected patients and early development of chorea in more mildly affected patients.

Severely affected individuals should be monitored for scoliosis.

Agents/Circumstances to Avoid

Valproate is contraindicated in glycine encephalopathy as an antiepileptic drug (AED). It raises blood and CSF glycine concentrations and may increase seizure frequency. It has resulted in severe lethargy, coma, severe seizures, and chorea particularly in mildly affected patients [Hall & Ringel 2004; Author, personal observation].

Evaluation of Relatives at Risk

Biochemical genetic testing to promote early diagnosis and treatment of at-risk newborn sibs is indicated. If the pathogenic variants in the family are known, such testing can be followed with molecular genetic testing.

See Genetic Counseling for issues related to testing of at-risk relatives for genetic counseling purposes.

Pregnancy Management

Pregnancy has been reported in one woman with mild glycine encephalopathy. No obvious teratogenic effect was observed [Ellaway et al 2001].

There is currently no effective treatment for pregnancies in which the fetus is affected with glycine encephalopathy.

Therapies Under Investigation

Search in the US and in Europe for access to information on clinical studies for a wide range of diseases and conditions. Note: There may not be clinical trials for this disorder.

Genetic Counseling

Genetic counseling is the process of providing individuals and families with information on the nature, inheritance, and implications of genetic disorders to help them make informed medical and personal decisions. The following section deals with genetic risk assessment and the use of family history and genetic testing to clarify genetic status for family members. This section is not meant to address all personal, cultural, or ethical issues that individuals may face or to substitute for consultation with a genetics professional. —ED.

Mode of Inheritance

Glycine encephalopathy is inherited in an autosomal recessive manner.

Risk to Family Members

Parents of a proband

  • The parents of an affected individual are obligate heterozygotes and therefore carry one mutated allele.
  • Heterozygotes are asymptomatic.
  • De novo pathogenic variants occur in approximately 1% of individuals with glycine encephalopathy; thus, carrier status in parents should be confirmed by molecular genetic testing rather than be assumed or inferred.
  • Molecular genetic testing of parental samples should be performed to confirm that both pathogenic variants have been properly identified in the proband. Affected individuals who appeared to have two GLDC variants or two AMT variants have turned out to have both variants on the same parental allele with an unrecognized deletion on the second allele [Author, personal observation], a situation which can result in interpretation errors of carrier and prenatal testing.

Sibs of a proband

  • When each parent is heterozygous for one pathogenic variant, each sib of an affected individual has at conception a 25% chance of being affected, a 50% chance of being an asymptomatic carrier, and a 25% chance of being unaffected and not a carrier.
  • Heterozygotes (carriers) are asymptomatic.

Offspring of a proband

  • Most individuals with glycine encephalopathy do not reproduce. However, some individuals with milder forms may be fertile (see Pregnancy Management).
  • The offspring of affected individuals are obligate heterozygotes (carriers) for a pathogenic variant.

Other family members. Each sib of the proband's parents is at a 50% risk of being a carrier.

Carrier (Heterozygote) Detection

Carrier testing using molecular genetic techniques is possible once the pathogenic variants have been identified in the family.

Carrier testing using biochemical methodologies is not reliable. However, Finnish obligate heterozygotes for glycine encephalopathy were reported to have slightly higher plasma glycine concentrations than controls. Others observed somewhat increased plasma glycine concentrations in at-risk sibs monitored soon after birth.

  • The initial report of measurement of GCS enzyme activity in lymphoblasts included ten obligate heterozygotes for P-protein deficiency. A subsequent report showed overlap of GCS enzyme activity in lymphoblasts in normal controls and carriers [Kure et al 1999], making this assay unreliable.
  • No attempts have been made to measure GCS enzyme activity in liver biopsies from obligate heterozygotes.

Related Genetic Counseling Issues

See Management, Evaluation of Relatives at Risk for information on evaluating at-risk relatives for the purpose of early diagnosis and treatment.

Family planning

  • The optimal time for determination of genetic risk, clarification of carrier status, and discussion of the availability of prenatal testing is before pregnancy.
  • It is appropriate to offer genetic counseling (including discussion of potential risks to offspring and reproductive options) to young adults who are affected, are carriers, or are at risk of being carriers.

Prenatal Testing and Preimplantation Genetic Diagnosis

Prenatal diagnosis for at-risk pregnancies is possible by either molecular genetic testing or GCS enzyme assay.

Molecular genetic testing. Once the pathogenic variants have been identified in an affected family member, prenatal diagnosis for a pregnancy at increased risk and preimplantation genetic diagnosis (PGD) are possible. PGD has been performed successfully in several families with glycine encephalopathy.

Note: If the gene in which pathogenic variants are causative has been properly assigned but only a single pathogenic variant has been identified and if material from the proband and both parents is available, intragenic SNPs can often be used to provide intragenic linkage analysis.

Biochemical testing. Earlier prenatal testing using measurement of amniotic fluid glycine concentration and the glycine/serine ratio was unreliable because normal and affected values overlapped.

Enzymatic testing. If the gene in which pathogenic variants are causative has not been assigned, intragenic SNPs are not informative, or parental samples are not available, prenatal diagnosis is possible by assay of GCS enzyme activity in uncultured CVS material.

Note: Prenatal diagnosis by GCS enzyme assay has at least a 1% false negative rate [Applegarth et al 2000b]. The false negative cases appear particularly to involve T-protein deficiency or mildly affected cases, in which residual GCS enzyme activity is present and overlap has been observed between affected individuals and carriers in the range of enzyme activity.


GeneReviews staff has selected the following disease-specific and/or umbrella support organizations and/or registries for the benefit of individuals with this disorder and their families. GeneReviews is not responsible for the information provided by other organizations. For information on selection criteria, click here.

  • National Library of Medicine Genetics Home Reference
  • NKH International Family Network
    2236 Birchbark Trail
    Clearwater FL 33763
    Phone: 727-799-4977
    Fax: 727-441-4942

Molecular Genetics

Information in the Molecular Genetics and OMIM tables may differ from that elsewhere in the GeneReview: tables may contain more recent information. —ED.

Table A.

Glycine Encephalopathy: Genes and Databases

Data are compiled from the following standard references: gene from HGNC; chromosome locus from OMIM; protein from UniProt. For a description of databases (Locus Specific, HGMD, ClinVar) to which links are provided, click here.

Table B.

OMIM Entries for Glycine Encephalopathy (View All in OMIM)



Gene structure. GLDC comprises 25 exons and intron/exon boundaries [Takayanagi et al 2000]. Its molecular analysis has been hampered by the presence of a processed full-length pseudogene with 97.5% homology to the true gene, differing in single-nucleotide variants along its length [Takayanagi et al 2000]. Primers based on intronic sequence are required to avoid amplification of the pseudogene. For a detailed summary of gene and protein information, see Table A, Gene.

Pathogenic variants (see Table 3)

  • The p.Ser564Ile and p.Gly762Arg variants have been identified in 70% of alleles and 8% of alleles, respectively, in individuals from a particular area of Finland [Kure et al 1999].
  • The p.Arg515Ser variant was identified in nine unrelated individuals and in 5% of glycine encephalopathy alleles in a series of 50 non-Finnish individuals [Toone et al 2001b]. It tends to be present in higher frequency in the British Isles.
  • Approximately 20% of GLDC mutated alleles are (multi)exon deletions or duplications [Kanno et al 2007]. These occur on various haplotypes and in different ethnic groups. They appear to arise as a result of non-homologous allelic recombination of Alu repeats.
  • Many other pathogenic variants have been described in single individuals [Kure et al 1999, Takayanagi et al 2000, Applegarth & Toone 2001, Toone et al 2002, Kure et al 2006a].

Table 3.

Selected GLDC Pathogenic Variants

DNA Nucleotide ChangePredicted Protein ChangeReference Sequences

Note on variant classification: Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.

Note on nomenclature: GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen​ See Quick Reference for an explanation of nomenclature.

Normal gene product. The normal gene product is glycine decarboxylase or P-protein, a pyridoxal-5-phosphate containing homodimer of approximately 200 kd. It completes the first of four steps in glycine degradation (see Figure 1), by P protein-catalyzed decarboxylation of glycine with CO2 as product.

Abnormal gene product. The occurrence of nonsense variants and deletions strongly suggests that the disorder results from lack of sufficient functional glycine decarboxylase.


Gene structure. AMT is small, comprising only nine exons; the coding region can be sequenced in six PCR fragments. Genomic sequence for T-protein is found under Genbank accession number D14681-86. For a detailed summary of gene and protein information, see Table A, Gene.

Pathogenic variants. Three recurrent pathogenic variants have been found (see Table 4):

Table 4.

Selected AMT Pathogenic Variants

DNA Nucleotide Change
(Alias 1)
Predicted Protein ChangeReference Sequences

Note on variant classification: Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.

Note on nomenclature: GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen​ See Quick Reference for an explanation of nomenclature.


Variant designation that does not conform to current naming conventions

Normal gene product. The normal gene product is a 45-kd monomeric aminomethyltransferase or T-protein, which requires tetrahydrofolate (THF) as cofactor. It transfers the methyl group from glycine to tetrahydrofolate and is the second step of the GCS.

Abnormal gene product. Many pathogenic variants cause loss of gene product/function. With some pathogenic missense variants, the protein is stable but measurable residual enzyme activity is decreased.


Gene structure. GCSH comprises five exons and spans 13.5 kb. The lipoic acid-binding site is in exon 4. Three highly homologous pseudogenes have been found [Kure et al 2001]. For a detailed summary of gene and protein information, see Table A, Gene.

Pathogenic variants. Hiraga et al [1981] showed an H-protein that was devoid of lipoic acid. No GCSH pathogenic variants have been identified in persons with glycine encephalopathy.

Normal gene product. Glycine cleavage system H protein, a 14-kd lipoamide-containing non-enzyme protein, interacts as a co-substrate with all three enzyme proteins by providing the central arm for substrate binding on which the glycine cleavage cycle depends.

Abnormal gene product. Unknown at present.


Literature Cited

  • Aburahma S, Khassawneh M, Griebel M, Sharp G, Gibson J. Pitfalls in measuring cerebrospinal fluid glycine levels in infants with encephalopathy. J Child Neurol. 2011;26:703–6. [PubMed: 21335543]
  • Aliefendioğlu D, Tana Aslan A, Coşkun T, Dursun A, Cakmak FN, Kesimer M. Transient nonketotic hyperglycinemia: two case reports and literature review. Pediatr Neurol. 2003;28:151–5. [PubMed: 12699870]
  • Applegarth DA, Edelstein AD, Wong LT, Morrison BJ. Observed range of assay values for plasma and cerebrospinal fluid amino acid levels in infants and children aged 3 months to 10 years. Clin Biochem. 1979;12:173–8. [PubMed: 519849]
  • Applegarth DA, Toone JR. Nonketotic hyperglycinemia (glycine encephalopathy): laboratory diagnosis. Mol Genet Metab. 2001;74:139–46. [PubMed: 11592811]
  • Applegarth DA, Toone JR, Lowry RB. Incidence of inborn errors of metabolism in British Columbia 1969-1996. Pediatrics. 2000a;105:e10. [PubMed: 10617747]
  • Applegarth DA, Toone JR, Rolland MO, Black SH, Yim DK, Bemis G. Non-concordance of CVS and liver glycine cleavage enzyme in three families with non-ketotic hyperglycinaemia (NKH) leading to false negative prenatal diagnoses. Prenat Diagn. 2000b;20:367–70. [PubMed: 10820402]
  • Boneh A, Degani Y, Harari M. Prognostic clues and outcome of early treatment of nonketotic hyperglycinemia. Pediatr Neurol. 1996;15:137–41. [PubMed: 8888048]
  • Boneh A, Korman SH, Sato K, Kanno J, Matsubara Y, Lerer I, Ben-Neriah Z, Kure S. A single nucleotide substitution that abolishes the initiator methionine codon of the GLDC gene is prevalent among patients with glycine encephalopathy in Jerusalem. J Hum Genet. 2005;50:230–4. [PubMed: 15864413]
  • Bröer S, Balley C G, Kowalczuk S, et al. Iminoglycinuria and hyperglycinuria are discrete human phenotypes resulting from complex mutations in proline and glycine transporters. J Clin Invest. 2008;118:3881–92. [PMC free article: PMC2579706] [PubMed: 19033659]
  • Brunel-Guitton C, Casey B, Coulter-Mackie M, Vallance H, Hewes D, Stockler-Ipsiroglu S, Mercimek-Mahmutoglu S. Late-onset nonketotic hyperglycinemia caused by a novel homozygous missense mutation in the GLDC gene. Mol Genet Metab. 2011;103:193–6. [PubMed: 21411353]
  • Clayton PT, Surtees RAH, DeVile C, Hyland K, Heales SJR. Neonatal epileptic encephalopathy. Lancet. 2003;361:1614. [PubMed: 12747882]
  • Coşkun T, Ozalp I, Tokatly A. Iminoglycinuria: a benign type of inherited aminoaciduria. Turk J Pediatr. 1993;35:121–5. [PubMed: 7504361]
  • Cusmai R, Martinelli D, Moavero R, Dionisi Vici C, Vigevano F, Castana C, Elia M, Bernabei S, Bevivino E. Ketogenic diet in early myoclonic encephalopathy due to non ketotic hyperglycinemia. Eur J Paediatr Neurol. 2012;16:509–13. [PubMed: 22261077]
  • del Toro M, Arranz JA, Macaya A, Riudor E, Raspall M, Moreno A, Vazquez E, Ortega A, Matsubara Y, Kure S, Roig M. Progressive vacuolating glycine leukoencephalopathy with pulmonary hypertension. Ann Neurol. 2006;60:148–52. [PubMed: 16802295]
  • Ellaway CJ, Mundy H, Lee PJ. Successful pregnancy outcome in atypical hyperglycinaemia. J Inherit Metab Dis. 2001;24:599–600. [PubMed: 11757588]
  • Flusser H, Korman SH, Sato K, Matsubara Y, Galil A, Kure S. Mild glycine encephalopathy (NKH) in a large kindred due to a silent exonic GLDC splice mutation. Neurology. 2005;64:1426–30. [PubMed: 15851735]
  • Gabis L, Parton P, Roche P, Lenn N, Tudorica A, Huang W. In vivo 1H magnetic resonance spectroscopic measurement of brain glycine levels in nonketotic hyperglycinemia. J Neuroimaging. 2001;11:209–11. [PubMed: 11296595]
  • Hall DA, Ringel SP. Adult nonketotic hyperglycinemia (NKH) crisis presenting as severe chorea and encephalopathy. Mov Disord. 2004;19:485–6. [PubMed: 15077252]
  • Hamosh A, Van Hove J. Concerns regarding transience and heterozygosity in neonatal hyperglycinemia. Ann Neurol. 2003;53:685. [PubMed: 12731008]
  • Heindel W, Kugel H, Roth B. Noninvasive detection of increased glycine content by proton MR spectroscopy in the brains of two infants with nonketotic hyperglycinemia. AJNR Am J Neuroradiol. 1993;14:629–35. [PubMed: 8517351]
  • Hennermann JB. Clinical variability in glycine encephalopathy. Fut Neurol. 2006;1:621–30.
  • Hennermann JB, Berger JM, Grieben U, Scharer G, Van Hove JL. Predicition of long-term outcome in glycine encephalopathy: a clinical survey. J Inherit Metab Dis. 2012;35:253–61. [PubMed: 22002442]
  • Hiraga K, Kochi H, Hayasaka K, Kikuchi G, Nyhan WL. Defective glycine cleavage system in nonketotic hyperglycinemia. Occurrence of a less active glycine decarboxylase and an abnormal aminomethyl carrier protein. J Clin Invest. 1981;68:525–34. [PMC free article: PMC370827] [PubMed: 6790577]
  • Hoffmann G F, Schmitt B, Windfuhr M, Wagner N, Strehl H, Bagci S, Franz AR, Mills PB, Clayton PT, Baumgartner MR, Steinmann B, Bast T, Wolf NI, Zschocke J. Pyridoxal 5’-phosphate may be curative in early-onset epileptic encephalopathy. J Inherit Metab Dis. 2007;30:96–9. [PubMed: 17216302]
  • Hoover-Fong JE, Shah S, Van Hove JL, Applegarth D, Toone J, Hamosh A. Natural history of nonketotic hyperglycinemia in 65 patients. Neurology. 2004;63:1847–53. [PubMed: 15557500]
  • Jaeken J, de Koning T, van Hove J. Disorders of GABA, glycine, serine and proline. In: Blau N, Duran M, Blaskovics ME, Gibson KM, eds. Physician’s Guide to the Laboratory Diagnosis of Metabolic Diseases. 2 ed. Berlin: Springer; 2002:123-40.
  • Kanno J, Hutchin T, Kamada F, Narisawa A, Aoki Y, Matsubara Y, Kure S. Genomic deletion within GLDC is a major cause of non-ketotic hyperglycinaemia. J Med Genet. 2007;44:e69. [PMC free article: PMC2598024] [PubMed: 17361008]
  • Khong PL, Lam BC, Chung BH, Wong KY, Ooi GC. Diffusion-weighted MR imaging in neonatal nonketotic hyperglycinemia. AJNR Am J Neuroradiol. 2003;24:1181–3. [PubMed: 12812951]
  • Korman SH, Boneh A, Ichinohe A, Kojima K, Sato K, Ergaz Z, Gomori JM, Gutman A, Kure S. Persistent NKH with transient or absent symptoms and a homozygous GLDC mutation. Ann Neurol. 2004;56:139–43. [PubMed: 15236413]
  • Korman SH, Waterham HR, Gutman A, Jakobs C, Wanders RJ. Novel metabolic and molecular findings in hepatic carnitine palmitoyltransferase I deficiency. Mol Genet Metab. 2005;86:337–43. [PubMed: 16146704]
  • Korman SH, Wexler ID, Gutman A, Rolland MO, Kanno J, Kure S. Treatment from birth of nonketotic hyperglycinemia due to a novel GLDC mutation. Ann Neurol. 2006;59:411–5. [PubMed: 16404748]
  • Kure S, Kato K, Dinopoulos A, Gail C, DeGrauw TJ, Christodoulou J, Bzduch V, Kalmanchey R, Fekete G, Trojovsky A, Plecko B, Breningstall G, Tohyama J, Aoki Y, Matsubara Y. Comprehensive mutation analysis of GLDC, AMT, and GCSH in nonketotic hyperglycinemia. Hum Mutat. 2006a;27:343–52. [PubMed: 16450403]
  • Kure S, Kojima K, Ichinohe A, Maeda T, Kalmanchey R, Fekete G, Berg SZ, Filiano J, Aoki Y, Suzuki Y, Izumi T, Matsubara Y. Heterozygous GLDC and GCSH gene mutations in transient neonatal hyperglycinemia. Ann Neurol. 2002;52:643–6. [PubMed: 12402263]
  • Kure S, Kojima K, Kudo T, Kanno K, Aoki Y, Suzuki Y, Shinka T, Sakata Y, Narisawa K, Matsubara Y. Chromosomal localization, structure, single-nucleotide polymorphisms, and expression of the human H-protein gene of the glycine cleavage system (GCSH), a candidate gene for nonketotic hyperglycinemia. J Hum Genet. 2001;46:378–84. [PubMed: 11450847]
  • Kure S, Korman SH, Kanno J, Narisawa A, Kubota M, Takayanagi T, Takayanagi M, Saito T, Matsui A, Kamada F, Aoki Y, Ohura T, Matsubara Y. Rapid diagnosis of glycine encephalopathy by 13C-glycine breath test. Ann Neurol. 2006b;59:862–7. [PubMed: 16634033]
  • Kure S, Mandel H, Rolland MO, Sakata Y, Shinka T, Drugan A, Boneh A, Tada K, Matsubara Y, Narisawa K. A missense mutation (His42Arg) in the T-protein gene from a large Israeli-Arab kindred with nonketotic hyperglycinemia. Hum Genet. 1998a;102:430–4. [PubMed: 9600239]
  • Kure S, Shinka T, Sakata Y, Osamu N, Takayanagi M, Tada K, Matsubara Y, Narisawa K. A one-base deletion (183delC) and a missense mutation (D276H) in the T- protein gene from a Japanese family with nonketotic hyperglycinemia. J Hum Genet. 1998b;43:135–7. [PubMed: 9621520]
  • Kure S, Takayanagi M, Kurihara Y, Leisti J, Zalai D, Chuck G, Tada K, Matsubara Y, Narisawa K. Nonketotic hyperglycinemia: mutation spectra of the GLDC and AMT genes in Finnish and non-Finnish populations. Am J Hum Genet. 1999;65:A2406.
  • Lang TF, Parr JR, Matthews EE, Gray RG, Bonham JR, Kay JD. Practical pitfalls in the diagnosis of transient non-ketotic hyperglycinemia. Dev Med Child Neurol. 2008;50:157–9. [PubMed: 18201306]
  • Mills PB, Surtees RA, Champion MP, Beesley CE, Dalton N, Scambler PJ, Heales SJ, Briddon A, Scheimberg I, Hoffmann GF, Zschocke J, Clayton PT. Neonatal epileptic encephalopathy caused by mutations in the PNPO gene encoding pyridox(am)ine 5'-phosphate oxidase. Hum Mol Genet. 2005;14:1077–86. [PubMed: 15772097]
  • Mourmans J, Majoie CB, Barth PG, Duran M, Akkerman EM, Poll-The BT. Sequential MR imaging changes in nonketotic hyperglycinemia. AJNR Am J Neuroradiol. 2006;27:208–11. [PubMed: 16418385]
  • Navarro-Sastre A, Tort F, Stehling O, Uzarska MA, Arranz JA, Del Toro M, Labayru MT, Landa J, Font A, Garcia-Villoria J, Merinero B, Ugarte M, Gutierrez-Solana LG, Campistol J, Garcia-Cazoral A, Vaquerizo J, Riudor E, Briones P, Elpeleg O, Ribes A, Lill R. A fatal mitochondrial disease is associated with defective NFU1 function in the maturation of a subset of mitochondrial Fe-S proteins. Am J Hum Genet. 2011;89:656–67. [PMC free article: PMC3213398] [PubMed: 22077971]
  • Riudor E, Arranz JA, del Toro M, Castello F, Briones P, Ribes A, Roig M. A new presentation of nonketotic hyperglycinemia with primary pulmonary hypertension and branched chain acylglycines with fatal outcome in three families. J Inherit Metab Dis. 2001;24 Suppl 1:35.
  • Sener RN. Nonketotic hyperglycinemia: diffusion magnetic resonance imaging findings. J Comput Assist Tomogr. 2003;27:538–40. [PubMed: 12886139]
  • Seo DW, Na DG, Na DL, Moon SY, Hong SB. Subcortical hypointensity in partial status epilepticus associated with nonketotic hyperglycemia. J Neuroimaging. 2003;13:259–63. [PubMed: 12889174]
  • Steiner RD, Sweetser DA, Rohrbaugh JR, Dowton SB, Toone JR, Applegarth DA. Nonketotic hyperglycinemia: atypical clinical and biochemical manifestations. J Pediatr. 1996;128:243–6. [PubMed: 8636821]
  • Takayanagi M, Kure S, Sakata Y, Kurihara Y, Ohya Y, Kajita M, Tada K, Matsubara Y, Narisawa K. Human glycine decarboxylase gene (GLDC) and its highly conserved processed pseudogene (psiGLDC): their structure and expression, and the identification of a large deletion in a family with nonketotic hyperglycinemia. Hum Genet. 2000;106:298–305. [PubMed: 10798358]
  • Tan ES, Wiley V, Carpenter K, Wilcken B. Non-ketotic hyperglycinemia is usually not detectable by tandem mass spectrometry newborn screening. Mol Genet Metab. 2007;90:446–8. [PubMed: 17207649]
  • Toone JR, Applegarth DA, Coulter-Mackie MB, James ER. Biochemical and molecular investigations of patients with nonketotic hyperglycinemia. Mol Genet Metab. 2000;70:116–21. [PubMed: 10873393]
  • Toone JR, Applegarth DA, Coulter-Mackie MB, James ER. Identification of the first reported splice site mutation (IVS7-1G>A) in the aminomethyltransferase (T-protein) gene of the glycine cleavage complex in 3 unrelated families with nonketotic hyperglycinemia. Hum Mutat. 2001a;17:76. [PubMed: 11139253]
  • Toone JR, Applegarth DA, Coulter-Mackie MB, James ER. Recurrent mutations in P- and T-proteins of the glycine cleavage complex and a novel T-protein mutation (N145I): a strategy for the molecular investigation of patients with nonketotic hyperglycinemia (NKH). Mol Genet Metab. 2001b;72:322–5. [PubMed: 11286506]
  • Toone JR, Applegarth DA, Kure S, Coulter-Mackie MB, Sazegar P, Kojima K, Ichinohe A. Novel mutations in the P-protein (glycine decarboxylase) gene in patients with glycine encephalopathy (non-ketotic hyperglycinemia). Mol Genet Metab. 2002;76:243–9. [PubMed: 12126939]
  • Toone JR, Applegarth DA, Levy HL, Coulter-Mackie MB, Lee G. Molecular genetic and potential biochemical characteristics of patients with T-protein deficiency as a cause of glycine encephalopathy (NKH). Mol Genet Metab. 2003;79:272–80. [PubMed: 12948742]
  • Tsao CY. The efficacy of vagus nerve stimulation in intractable epilepsy associated with nonketotic hyperglycinemia in two children. J Child Neurol. 2010;25:375–8. [PubMed: 19841478]
  • Van Hove JLK, Lohr N. Metabolic and monogenic causes of seizures in neonates and young infants. Mol Genet Metab. 2011;104:214–30. [PubMed: 21839663]
  • Van Hove JL, Kishnani PS, Demaerel P, Kahler SG, Miller C, Jaeken J, Rutledge SL. Acute hydrocephalus in nonketotic hyperglycinemia. Neurology. 2000;54:754–6. [PubMed: 10680820]
  • Van Hove J, Scharer G, Spector E, Mahieu V, Schollen E, Matthijs G, Freehauf C, Wilson C. Mutation spectrum in non-ketotic hyperglycinemia. Hamburg, Germany: Society for the Study of Inborn Errors of Metabolism Annual Symposium; 2007.
  • Van Hove JL, Vande Kerckhove K, Hennermann JB, Mahieu V, Declercq P, Mertens S, De Becker M, Kishnani PS, Jaeken J. Benzoate treatment and the glycine index in nonketotic hyperglycinaemia. J Inherit Metab Dis. 2005;28:651–63. [PubMed: 16151895]
  • Wilson C, Prosser D, Wong M, Percival T, Heron P, Scharer G, Van Hove J. The high incidence of non-ketotic hyperglycinemia in Maori neonates. Poster session. Auckland, NZ: Australasian Society for Inborn Errors of Metabolism Annual Meeting; 2007.
  • Wiltshire EJ, Poplawski NK, Harrison JR, Fletcher JM. Treatment of late-onset nonketotic hyperglycinaemia: effectiveness of imipramine and benzoate. J Inherit Metab Dis. 2000;23:15–21. [PubMed: 10682304]

Suggested Reading

  • Hamosh A, Johnston MV. Nonketotic hyperglycinemia. In: Scriver CR, Beaudet AL, Sly WS, Valle D, Vogelstein B, eds. The Online Metabolic and Molecular Bases of Inherited Disease (OMMBID). New York: McGraw-Hill. Chap 90.

Chapter Notes

Author History

Derek A Applegarth, PhD, FCCMG; University of British Columbia (2002-2005)
Curtis Coughlin II, MS, MBe (2013-present)
Ada Hamosh, MD, MPH; Johns Hopkins University School of Medicine (2002-2013)
Gunter Scharer, MD, PhD (2009-present)
Jennifer Toone, BSc, RT; Children's and Women's Health Centre of British Columbia (2002-2005)
Johan Van Hove, MD, PhD (2009-present)


The authors thank Julia Hennermann, MD, Berlin, for continuing collaboration on the clinical phenotype and Drs. Rolland, Hutchin, and Kure for collaboration on the molecular basis.

Revision History

  • 11 July 2013 (me) Comprehensive update posted live
  • 24 November 2009 (me) Comprehensive update posted live
  • 26 July 2005 (ah) Revision: molecular genetic testing clinically available
  • 14 December 2004 (me) Comprehensive update posted to live Web site
  • 16 May 2003 (cd) Revision: enzymatic prenatal testing no longer available
  • 14 November 2002 (me) Review posted to live Web site
  • 7 March 2002 (da) Original submission
Copyright © 1993-2018, University of Washington, Seattle. GeneReviews is a registered trademark of the University of Washington, Seattle. All rights reserved.

GeneReviews® chapters are owned by the University of Washington. Permission is hereby granted to reproduce, distribute, and translate copies of content materials for noncommercial research purposes only, provided that (i) credit for source ( and copyright (© 1993-2018 University of Washington) are included with each copy; (ii) a link to the original material is provided whenever the material is published elsewhere on the Web; and (iii) reproducers, distributors, and/or translators comply with the GeneReviews® Copyright Notice and Usage Disclaimer. No further modifications are allowed. For clarity, excerpts of GeneReviews chapters for use in lab reports and clinic notes are a permitted use.

For more information, see the GeneReviews® Copyright Notice and Usage Disclaimer.

For questions regarding permissions or whether a specified use is allowed, contact: ude.wu@tssamda.

Bookshelf ID: NBK1357PMID: 20301531


  • PubReader
  • Print View
  • Cite this Page
  • Disable Glossary Links

Tests in GTR by Gene

Related information

  • MedGen
    Related information in MedGen
  • OMIM
    Related OMIM records
  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed
  • Gene
    Locus Links

Similar articles in PubMed

See reviews...See all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...