| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Diabetes Forschungsinstitut and
2
Kinderklinik, Heinrich-Heine-Universität, D-40225 Düsseldorf, Germany.
a Address correspondence to this author at: Diabetes-Forschungsinstitut, Klinische Biochemie, Auf'm Hennekamp 65, D-40225 Düsseldorf, Germany. Fax 49-211-3382-603; e-mail schadewa{at}uni-duesseldorf.de
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: Branched-chain L-amino acids were measured by automatic amino acid analysis.
Results: Alloisoleucine reference values in plasma were established in healthy adults [1.9 ± 0.6 µmol/L (mean ± SD); n = 35], children 3–11 years (1.6 ± 0.4 µmol/L; n = 17), and infants <3 years (1.3 ± 0.5 µmol/L; n = 37). The effect of dietary isoleucine was assessed in oral loading tests. In controls receiving 38 µmol (n = 6; low dose) and 1527 µmol (n = 3; high dose) of L-isoleucine per kilogram of body weight, peak increases of plasma isoleucine were 78 ± 24 and 1763 ± 133 µmol/L, respectively; the peak increase of alloisoleucine, however, was negligible for low-dose (<0.3 µmol/L) and minor for high-dose (5.5 ± 2.1 µmol/L) load. In patients with diabetes mellitus, ketotic hypoglycemia, phenylketonuria, and obligate heterozygous parents of MSUD patients, alloisoleucine was not significantly different from healthy subjects. Therefore, a plasma concentration of 5 µmol/L was used as a cutoff value. In patients with classical MSUD (n = 7), alloisoleucine was beyond the cutoff value in 2451 of 2453 unselected samples. In patients with variant MSUD (n = 9), alloisoleucine was >5 µmol/L in all samples taken for establishment of diagnosis and in 94% of the samples taken for treatment control (n = 624). With the other branched-chain amino acids, the frequency of diagnostically significant increases was <45%.
Conclusions: The present findings indicate that plasma L-alloisoleucine above the cutoff value of 5 µmol/L is the most specific and most sensitive diagnostic marker for all forms of MSUD.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the classical form, severe neurological symptoms and a maple syrup-like odor appear during the first week of life. Demonstration of grossly increased concentrations of branched-chain amino acids, especially leucine, in plasma firmly establishes the diagnosis.
In patients with the variant form of MSUD, the onset of metabolic derangements associated with ketoacidosis and cerebral symptoms is generally delayed. In these patients, overt clinical symptoms may be absent for months, years, or even decades. Some patients are admitted for medical examination because of psychomotor retardation and are diagnosed incidentally without having a history of ketoacidotic episodes. In other patients, intermittent episodes may arise in infancy and childhood during catabolic states, which are often triggered, in an apparently unpredictable manner, by intercurrent illnesses (1)(2)(3)(4). We, for example, recently experienced diagnosis of variant MSUD in two 4- and 5-year-old German patients who were experiencing severe metabolic crises. One of these patients remained undetected although traceably subjected to neonatal screening.
Detection of MSUD variants can be difficult. In the absence of evident clinical symptoms, the patients often exhibit near normal or moderately increased plasma concentrations of leucine, valine, and isoleucine, which are similar to the concentrations observed in secondary amino acid disturbances such as ketotic hypoglycemia, diabetes mellitus, starvation, and other catabolic states (5)(6)(7)(8)(9)(10). Early diagnosis of MSUD is essential, however, to maintain patients under metabolic control during intercurrent episodes to prevent permanent brain damage. Thus, a suitable indicator specific for MSUD is needed that permits early identification of variant MSUD.
We therefore examined the (patho)physiological significance of alloisoleucine plasma concentrations for the differential diagnosis of MSUD. This nonprotein amino acid is formed from isoleucine in vivo. It is consistently present in human plasma and can be reliably determined along with the other branched-chain amino acids (11). In the present study, we established alloisoleucine reference ranges and investigated the effect of dietary isoleucine on plasma L-alloisoleucine. Based on these results, a cutoff value was defined. This value was then used to estimate the sensitivity and specificity of increased alloisoleucine plasma concentrations for the diagnosis of MSUD.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. In all the
patients, blood was collected on occasion of routine clinical
examination or for metabolic monitoring.
|
analytical procedures
Branched-chain amino acid concentrations in plasma were measured
on an automatic amino acid analyzer (LC 5000, LC 6000;
Biotronik), using ninhydrin detection and a short program as
detailed previously (11). The limit of detection for the
branched-chain amino acids was <0.05 nmol (equivalent to plasma
concentrations of <0.2 µmol/L). The range for reliable
quantification of alloisoleucine was 0.1–10 nmol (equivalent to
0.5–50 µmol/L in plasma). Over this range, the molar response (area
per nmol) varied <5%. Typically, the CV was well below 10%, e.g.,
for alloisoleucine in plasma at 1 (250) µmol/L, the CV within
(n = 10) and between runs (n = 9) was 7 (2) % and 8 (2) %,
respectively. When the concentration exceeded 50 µmol/L in plasma,
the sample was diluted appropriately before analysis. Thus, in
specimens from non-MSUD subjects, concentrations of alloisoleucine and
the other branched-chain amino acids were generally measured in
separate analytical runs. Analytical data of external MSUD
patients were provided by the respective attending metabolic
centers, which applied an equivalent methodology for the
determination of plasma amino acids. Residual activity of
branched-chain 2-oxoacid dehydrogenase complex in our laboratory was
assessed in cultured fibroblasts using
L-[1-14C]leucine as
described previously (12).
loading tests
After an overnight fast, six healthy subjects (five males, one
female; 31 ± 5 years) received 5 mg of L-isoleucine
per kilogram of body weight (dissolved in 50 mL of diluted citric acid
solution; low-dose loading) orally. Three volunteers (two males, one
female; 29 ± 9 years) underwent a high-dose loading and ingested
200 mg of L-isoleucine per kilogram of body weight
thoroughly mixed with 150 mL of yogurt. The L-isoleucine
(from Bachem) was essentially free from L-alloisoleucine
(<0.03%). Venous blood was collected into EDTA tubes just before
(basal values) and after ingestion of the loading dose according to the
time schedules depicted in Fig. 1
. Plasma was obtained by centrifugation and analyzed for
branched-chain amino acids as described above. Written informed consent
was obtained from the participants. The experimental protocol had been
approved by the Ethikkommission of the Heinrich-Heine-Universität
Düsseldorf.
|
calculations
Unless otherwise noted, the results are presented as means ±
SEM with the number of determinations in parentheses. Correlations were
checked by simple linear regression analysis (least-squares method).
For examination of differences, the Mann–Whitney U-test
(two-tailed) was applied. Sensitivity estimates were based on the
relationship between the number of plasma samples with alloisoleucine
concentrations beyond the cutoff value and the total number of plasma
analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
loading tests
To assess the effect of dietary isoleucine on plasma
alloisoleucine concentrations, loading tests were performed. When
healthy volunteers ingested 38 µmol of L-isoleucine per
kilogram of body weight (low-dose loading), the mean peak increase of
plasma isoleucine over basal was 78 ± 10 µmol/L (121% ± 10%;
n = 6). However, the increase of alloisoleucine was scarcely
measurable (<0.3 µmol/L; Fig. 1
). High-dose L-isoleucine
loading (1527 µmol/kg of body weight) induced considerable peak
increases in plasma isoleucine (1763 ± 133 µmol/L; n = 3)
but only minor peak increases in plasma alloisoleucine (5.5 ±
1.2 µmol/L; Fig. 1
).
non-msud patients
Alloisoleucine plasma concentrations were also evaluated in
several metabolic defects. In patients with diabetes mellitus
exhibiting significantly increased mean leucine (53%), valine (36%),
and isoleucine (35%) plasma concentrations (P <<0.001 vs
adult controls), alloisoleucine concentrations were comparable to
control concentrations (see Table 1
). In patients with phenylketonuria
and ketotic hypoglycemia, the plasma concentrations of all
branched-chain amino acids, including alloisoleucine, were comparable
to control concentrations. As checked by regression analysis,
the plasma concentrations of alloisoleucine and its metabolic
precursor, isoleucine, were statistically not correlated in all
non-MSUD study groups, including adults, children, and infants [e.g.,
linear regression for controls: y = 0.002 (±
0.002)x + 1.39 (± 0.20); coefficient of determination =
0.008; n = 102].
msud patients
In the non-MSUD study groups, plasma alloisoleucine was always
below 5 µmol/L. The loading experiments also suggest that in controls
and non-MSUD patients receiving branched-chain amino acids in typical
amounts in their diet, plasma alloisoleucine concentrations should not
exceed that value. Therefore, this value was taken as a reasonable
cutoff value for the discrimination of MSUD and non-MSUD subjects.
Of note is that alloisoleucine concentrations were below the cutoff in
obligate heterozygous parents of patients with the classical form of
MSUD (Table 1
). In classical MSUD patients, alloisoleucine
concentrations >5 µmol/L were found in 99.9% of a representative
number of unselected plasma samples (n = 2453 from seven patients;
Table 1
). There was a statistically highly significant linear
relationship between isoleucine (x) and alloisoleucine
(y) concentrations [y = 0.40 (±
0.01)x + 54.6 (± 2.0); P <<0.001] with but a
somewhat low coefficient of determination (coefficient of determination
= 0.419).
When plasma branched-chain amino acids were within ranges indicating good to moderate metabolic control in the patients (leucine, valine, and isoleucine concentrations within 50–500, 100–700, and 30–300 µmol/L, respectively), alloisoleucine was consistently >7 µmol/L (mean, 120 ± 2 µmol/L; range, 7–443 µmol/L; n = 1085). In 208 samples, normal branched-chain amino acid concentrations were found (i.e., leucine, valine, and isoleucine concentrations within 50–250, 100–400, 30–150 µmol/L), but the alloisoleucine concentration was 90 ± 3 µmol/L (range, 7–244 µmol/L).
Altogether, there were just two exceptions in classical MSUD with alloisoleucine concentrations <5 µmol/L. In these samples (from one patient), isoleucine was virtually absent from plasma (<1 µmol/L). We noticed six additional plasma samples in which isoleucine had incidentally dropped to such low concentrations. In the latter cases, however, the alloisoleucine concentrations were 69 ± 9 µmol/L (range, 38–92 µmol/L), which was in good agreement with the values predicted by the results of the correlation analysis given above.
Appropriate observations in variant MSUD are far less abundant. Data
procured for nine patients with variant MSUD of different severity are
compiled in Table 2
. In all patients, alloisoleucine was >5 µmol/L in the blood
samples taken for the establishment of diagnosis by quantitative amino
acid analysis before the start of dietary treatment. Under good to
moderate metabolic control (see above), five patients with more severe
variants generally exhibited clearly increased alloisoleucine
concentrations (patients D.G., Y.M, S.C., H.H., and T.R.). In the mild
variants, the plasma concentrations were somewhat lower and the
individual frequency of increased alloisoleucine was more variable. The
frequency, however, was never below 78% (patient L.F; Table 2
).
Of note is that, in variants, alloisoleucine was increased in 153
(86%) of a total of 179 plasma specimens that exhibited otherwise
normal branched-chain amino acid concentrations (for ranges, see
above). Altogether, the incidence of alloisoleucine above the cutoff
value in variant MSUD was 94% (n = 624; Table 2
).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The results in diabetic patients show that moderately increased branched-chain amino acid concentrations do not in themselves lead to increased alloisoleucine concentrations. Likewise, transient branched-chain aminoacidemia that occurs, for example, in ketotic hypoglycemia and starvation appears not to be associated with increased alloisoleucine concentrations (5)(6)(7)(8)(9)(10). Similarly, the transient approximately twofold increase of plasma isoleucine in the present low-dose isoleucine loading studies exerted no significant effect on plasma alloisoleucine. In the high-dose loads, the isoleucine equivalents administered corresponded to ~4 g of protein/kg of body weight. This amount grossly exceeded the uptake in a typical meal. The effect on increases in alloisoleucine, however, was only slight despite the ~30-fold peak increase in plasma isoleucine. Taken together, these findings indicate that increased plasma concentrations and the routine dietary supply of isoleucine have negligible influence on the alloisoleucine plasma concentrations in non-MSUD subjects. With exception of the high-dose isoleucine loads, we never observed plasma alloisoleucine concentrations >5 µmol/L in the non-MSUD study groups. Therefore, this concentration was taken as a tentative cutoff value for a retrospective analysis in MSUD patients.
The general knowledge that the presence of alloisoleucine is characteristic of MSUD (1)(16)(17) has apparently never been substantiated in quantitative terms. According to the present data, alloisoleucine concentrations below the cutoff value are extremely rare in classical MSUD and occur only when the patients are on a too strict dietary regimen. Generally, increased alloisoleucine persisted even when the isoleucine concentrations were extremely low (<1 µmol/L). The latter was not unexpected in MSUD because it has been shown that plasma alloisoleucine increases within hours after an isoleucine challenge but decreases with sluggish plasma kinetics within days or even weeks (18)(19)(20). In the 2 samples (of 2453) in which plasma alloisoleucine was <5 µmol/L, the patient was on a stringently restricted diet, and isoleucine was practically absent from the plasma. Most likely, there was a somewhat prolonged isoleucine deficiency in this patient, which finally led to the disappearance of plasma alloisoleucine.
In our patients showing a rather representative spectrum of variant MSUD, alloisoleucine was below the cutoff value in several plasma samples although isoleucine was generally beyond 30 µmol/L. The overall incidence of increased alloisoleucine decreased with the severity of the disease, in agreement with previous findings showing a graded enhancement of plasma alloisoleucine clearance in MSUD variants (20). In the majority of samples, however, alloisoleucine was >5 µmol/L. In the absence of grossly increased branched-chain amino acid concentrations and any clinical symptoms, increased alloisoleucine was found in at least 78% of the samples taken from an individual patient. Even when the branched-chain amino acid concentrations were normal, the incidence of alloisoleucine beyond the cutoff value was never below 70% (patient L.F.; data not shown).
Regarding the significance of increased plasma branched-chain amino acid concentrations for the diagnosis of MSUD, leucine, valine, and isoleucine should exceed ~400, 600, and 250 µmol/L, respectively, to allow reliable differentiation of MSUD-induced increases from secondary disturbances of branched-chain amino acid metabolism that occur, e.g., in ketotic hypoglycemia (5)(6). In our classical MSUD patients, the overall percentages of samples exhibiting plasma concentrations above these threshold values were 43% with leucine, 2% with valine, and 20% with isoleucine compared with >99% of the samples showing increased alloisoleucine. In variants, the overall percentages of increased plasma concentrations were 28% with leucine, 3% with valine, and 14% with isoleucine, compared with 94% with alloisoleucine. Simultaneous increases of leucine, valine, and isoleucine over these threshold values were found in only 2% of the specimens from classical MSUD patients and in 3% of the samples obtained from the variants.
Taken together, the present findings indicate that plasma alloisoleucine is the most sensitive and most specific general diagnostic marker for classical as well as variant forms of MSUD. Alloisoleucine analysis in plasma should allow a differential diagnosis of MSUD even in episodes of mild clinical symptoms and before the development of severe metabolic crises. Urinary analysis cannot be recommended for this purpose. Because of the generally low and rather variable fractional renal clearance of branched-chain compounds, analysis in urine is far less sensitive than in plasma specimens (21).
When based on an alloisoleucine cutoff value of 5 µmol/L, the sensitivity estimates for detection of variant and classical MSUD in the absence of clinical symptoms were >90% and >99%, respectively. The sensitivity in the presence of symptoms and the overall specificity can be expected to be almost 100%.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  D. Oglesbee, K. A. Sanders, J. M. Lacey, M. J. Magera, B. Casetta, K. A. Strauss, S. Tortorelli, P. Rinaldo, and D. Matern Second-Tier Test for Quantification of Alloisoleucine and Branched-Chain Amino Acids in Dried Blood Spots to Improve Newborn Screening for Maple Syrup Urine Disease (MSUD) Clin. Chem., March 1, 2008; 54(3): 542 - 549. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
