Developmental Brain Research 109 Ž1998.

179–186

Research report

Reversal of abnormalities of myelination by thyroxine therapy in

congenital hypothyroidism: localized in vivo proton magnetic resonance
spectroscopy ž MRS/ study 1
N.R. Jagannathan a , N. Tandon b, P. Raghunathan a , N. Kochupillai b,)

a
Department of NMR, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India
b
Department of Endocrinology and Metabolism, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India
Accepted 12 May 1998

Abstract

Deficiency of thyroid hormone during central nervous system ontogeny results in a variety of clinical, anatomical and biochemical
defects. Delay in thyroxine therapy in newborns with congenital hypothyroidism leads to irreversible brain damage. We have used
localized in vivo proton magnetic resonance spectroscopy ŽMRS. to assess biochemical changes in different regions of brain in three
patients with congenital hypothyroidism before and after thyroxine therapy. An abnormal lipid peak which disappeared with thyroxine
therapy was observed in cerebellum and frontal lobe in one patient. Statistically significant reduction of NAArŽCr q PCr. w P - 0.009x
and elevation of ChorŽCr q PCr. w P - 0.008x ratios in comparison to controls were documented in all three patients which tended to
normalise with thyroxine therapy. A variety of biochemical abnormalities relatable to myelin maturation were documented and these were
found to be reversible on thyroxine therapy. Reversibility was documented even though thyroxine therapy was initiated at ages beyond
which abnormalities in myelinogenesis are considered irreversible. Also, proton MRS revealed biochemical heterogeneity between
patients with congenital hypothyroidism. q 1998 Elsevier Science B.V. All rights reserved.

Keywords: Congenital hypothyroidism; Proton magnetic resonance spectroscopy; Central nervous system ontogeny; Thyroxine; Myelination

1. Introduction myelination. Post-natally thyroid hormones promote myeli-

The role of the thyroid hormones in prenatal and imme-
diate post-natal brain has been documented w41x. Follow-up
data of neonatal thyroid screening programmes support
this view w41x. Brain development in humans continues
until 2 years of post-natal life. Untreated congenital hypo-
thyroidism causes irreversible brain damage with mental
retardation and psychomotor disabilities w26x.
In the ontogeny of central nervous systems ŽCNS.,
thyroxine plays different roles at different stages w41x. In
the first trimester when most of cerebral and brain stem
neurogenesis occurs, the role of thyroid hormones has not
been established. During the second and third trimesters
thyroid hormones influence cerebellar neuronal prolifera-
tion, migration and differentiation and, to some extent,
) biochemistry w38,53x. The metabolites detected in human
Corresponding author. Fax: q91-11-686-2663
1
A preliminary report was presented at the 5th scientific meeting of
the International Society for Magnetic Resonance in Medicine, Vancou-
ver, Canada, April 1997.
nation and gliogenesis.
Deficiency of thyroid hormone causes biochemical ab-
normalities, resulting in neuronal and glial dysfunctions. In
the neonatal hypothyroid rat, the biosynthesis of cerebro-
side, sulfatide and sphingomyelin, and components of
myelin are transiently suppressed w41x. Enzymes associated
with myelin formation are also regulated by thyroid hor-
mones in the neonate w1,5,16x. In addition, a decrease in
the rate of brain protein and RNA synthesis has been
reported in thyroidectomised neonatal rats w41x. Evidence
also exists for retarded development of neuronal processes
w10x, reduced nerve terminal development w41x and retarded
development of cerebral oxidative enzymes in the hypothy-
roid brain w41x.
In-vivo proton magnetic resonance spectroscopy ŽMRS.
is used to obtain information on brain physiology and
brain include N-acetylaspartate ŽNAA., creatine phospho-
creatine ŽCr q PCr., choline containing compounds ŽCho.
and inositol ŽIno.. In general, relative concentration of

0165-3806r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.

PII S 0 1 6 5 - 3 8 0 6 Ž 9 8 . 0 0 0 8 1 - 9
180 N.R. Jagannathan et al.r DeÕelopmental Brain Research 109 (1998) 179–186
these metabolites are normally expressed by the peak areas
of the corresponding resonances. Determination of the
concentration and relative levels of proton-containing
metabolites help in assessing normal and abnormal tissues
and their response to therapy. Several articles have ap-
peared in the recent past on the use of proton MRS in
different brain disorders w7,22,25,32,33,35,54x. Recently, a
few studies have also appeared on the phosphorus ŽP-31.
MRS w6,19,24,36,49x and magnetic resonance imaging in
patients with hypothyroidism w4,28x.
In this paper, we report the results on the area specific
proton MR studies carried out before and after thyroxine
therapy in three patients with congenital hypothyroidism.
Patients with congenital hypothyroidism are usually identi-
fied early and treatment recommended soon after birth. It
may be emphasized here that since it is rare to find such
patients left untreated, our study is restricted to only three
patients. To the best of our knowledge, this is the first
report of MR spectroscopic observation indicating
myelinogenesis in the human brain beyond 5 years of age
in response to thyroxine therapy.
2. Materials and methods
2.1. Patients
Three patients with congenital hypothyroidism were
studied after obtaining informed consent. The clinical eval-
uation and investigative work up of the three patients
studied is provided in Table 1. Two patients were residents
of known iodine deficient areas. Post-treatment follow-up
and MRS studies for these patients were done at varying
intervals: before therapy, after 6 weeks ŽPost-therapy I.
and 6 months ŽPost-therapy II. of initiation of thyroxine
therapy. In addition, patient a1 was reassessed after 1 year
of treatment. Even at 1 year of treatment, the IQ score
remained unchanged Žscore: 35–40.. Thyroid function tests
revealed that serum T3 and T4 levels were normalised at 6
weeks for all three patients. The same treatment schedule
was then continued.
Table 1
Assessment of patients with congenital hypothyroidism
Patient a1
Chronological age in year 14
Height in cm Žpercentile. 106 Ž- 5th.
US:LS ratio 59:47
Bone age in years 6.1
IQ 35–40
Thyroid scan Sublingual thyroid
Basal thyroid function
T3 Žnmolrl. 0.22
T4 Žnmolrl. UD
TSH ŽmIurl. ) 100
Normal values: T3 Ž1.23–2.76 nmolrl.; T4 Ž51.4–154.3 nmolrl; TSH Ž0.5–4.5 mIurl..
2.2. MRI r MRS technique
MRIrMRS investigation were carried out at 1.5 T
using a Siemens 63-SP4000 whole body scanner. MR
images as well as localized proton MR spectra were
acquired using a CP head coil. Prior to proton MRS study,
multislice T1-weighted wrepetition time ŽTR. s 550 ms,
echo time ŽTE. s 15 msx images in the three orthogonal
planes and multislice T2-weighted ŽTR s 2500 ms; TE s 90
ms. axial images of the brain were obtained. These images
were subsequently used to guide the MRS study. Magnetic
field shimming at both the global and voxel levels was
performed prior to MRS, and the line-width after voxel
shimming was - 5 Hz. A STEAM pulse sequence w17x
was used to acquire the proton MR spectra with water
suppression. Three 908 slice-selective r.f. sync pulses with
a duration of 2.56 ms were applied using a slice-selection
gradient strength of 2 mTrm. Water suppression was
achieved by a single Gaussian shaped CHESS pulse of
bandwidth" 130 Hz for a duration of 25.6 ms followed by
spoiler gradients for efficient suppression of the water.
Spectra were acquired at an echo time ŽTE. of 135 ms, and
in some cases at 270 ms with a repetition time ŽTR. of 3 s.
To get a reasonably good spectrum, 256 averages were
accumulated. The signal ŽFID. was sampled by 2048 com-
plex data points with a sample frequency of 1000 Hz.
Spectral processing involved zero filling of the complex
time-domain data ŽFID. to 4096 data points and multiplica-
tion by a Gaussian function corresponding to a line broad-
ening of 3 Hz. Elimination of residual eddy current effects,
if any, were achieved by correcting pertinent phase distor-
tions of the time-domain signal with reference to the phase
evolution of the non-suppressed water proton signal from
the same voxel. The frequency spectra were obtained by
Fourier transformation of the corrected raw data ŽFID.
followed by zero and first order phase correction to yield
pure absorption mode spectra. Neither spectral baseline
correction nor resolution enhancement was applied. The
area under various metabolite peaks were measured by line
integration which was eventually used for further calcula-
tions. The total time for imaging and spectroscopy was
Patient a2 Patient a3
8 31
89 Ž- 5th. 114 Ž- 5th.
52:37 62:52
2 13
Severe retardation Severe retardation
Lingual thyroid No thyroid tissue localized
0.99 0.3
7.68 8.0
70.0 ) 100
 N.R. Jagannathan et al.r DeÕelopmental Brain Research 109 (1998) 179–186
Fig. 1. Representative in-vivo proton MR spectra of a volunteer from Ža. cerebellum, Žb. frontal lobe, and Žc. thalamus region of the brain.
between 45 and 60 min. Chemical shifts were measured
relative to tetramethylsilane ŽTMS; 0 ppm. with the NAA
peak fixed at 2.01 ppm. MR spectra from the cerebellum
were obtained for all the three patients, while for two
patients Ža1 and a3. frontal lobe and thalamus regions of
the brain were also investigated.
Seven healthy control subjects were recruited for the
study. To assess the reproducibility of MR ŽFig. 1. data,
controls were studied on two or more time points at least 3
months apart.
The data obtained from three patients was analysed and
compared with the control subjects for determining the
statistical significance Ž P - 0.05. between different groups
using the unpaired t-test.
3. Results
Brain MRI scans appeared normal except for mild
cerebellar atrophy. The in-vivo proton MR spectrum of
patient a1 acquired prior to therapy from the cerebellum
ŽFig. 2a. consisted of four major resonances from a broad
peak in the 0.8 to 1.5 ppm region assigned to lipids, NAA
181
Ž2.01 ppm., wCr q PCrx Ž3.01 ppm. and Cho Ž3.20 ppm..
The experiment was repeated with a reduced voxel size of
4.9 ml ŽFig. 2a. to rule out contamination from extraneous
lipids from the skull vault. However, lipid resonance per-
sisted even with the reduced voxel size which disappeared
at longer echo times ŽTE s 270 ms. due to the short T2
relaxation of lipids. In contrast, the spectrum of brain from
normal volunteers show no lipid peak w38x. In addition,
minor resonances due to myoinositol ŽmI; 3.54 ppm. and
methylene protons of Cr Ž3.9 ppm. were also observed.
Analysis of proton metabolite ratios acquired from the
cerebellum of all three patients showed that NAArŽCr q
PCr. and NAArCho ratios were lower compared to con-
trols Ž P - 0.01—see Table 2. while the ChorŽCr q PCr.
ratio was significantly elevated.
The proton MR spectrum from a 4.9 ml voxel from the
frontal lobe of patient a1 shown in Fig. 2b was markedly
different from that obtained from the cerebellum. Qualita-
tively, a large lipid peak Žwhich was reproducible when the
experiment was performed on two subsequent occasions.
dominated the spectrum with little contribution from other
metabolites. Therefore, prior to therapy, quantitative as-
182 N.R. Jagannathan et al.r DeÕelopmental Brain Research 109 (1998) 179–186

Fig. 2. In-vivo localized proton MR spectra obtained at an echo time ŽTE. of 135 ms from a 4.9 ml voxel prior to initiation of thyroxine therapy Ža, b, c.
and after 6 months of therapy Žd, e, f. from cerebellum, frontal lobe and thalamus region of the brain of patient a1.

sessment of these metabolites was not possible Žfor the
spectra recorded at TE s 135 ms. in the frontal lobe.
However, expectedly, the ratios calculated for the spec-
trum acquired at TE s 270 ms were also much lower
compared to controls Ždata not shown.. In contrast, the
spectrum from the thalamus ŽFig. 2c. did not show any
lipid peak. However, the metabolite ratios calculated from
the thalamus NAArŽCr q PCr. and NAArCho were also
lower compared to normals.
Interestingly, no lipid peak was demonstrable in any of
the brain regions of the other two patients prior to com-
mencement of therapy. However, similar to patient a1, the
metabolite NAArŽCr q PCr. and NAArCho ratios were
significantly less compared to controls while the ChorŽCr
q PCr. ratio was significantly elevated.
Spectra obtained from patient a1, 1 year after treat-
ment, are shown in Fig. 2d–f. The lipid peak observed in
frontal lobe and cerebellum completely disappeared. In-

Table 2
Statistical analysis of the metabolite ratios in different regions of the brain in patients with congenital hypothyroidism before and after thyroxine therapy
Cerebellum Frontal lobe Thalamus
NAA r (Cr q PCr)
Control 1.79 " 0.25 Ž n s 7. 2.00 " 0.26 Ž n s 7. 1.93 " 0.19 Ž n s 7.
Pre-therapy 1.33 " 0.12 Ž n s 3; P - 0.009. 1.55 Ž n s 1. 1.60 " 0.06 Ž n s 2; P - 0.03.
Post-therapy I 1.74 " 0.21 Ž n s 3. 1.73 " 0.41 Ž n s 2. 1.94 Ž n s 1.
Post-therapy II 1.77 " 0.05 Ž n s 2. 2.03 " 0.02 Ž n s 2. 1.88 " 0.14 Ž n s 2.

NAA r Cho
Control 2.05 " 0.28 Ž n s 7. 2.32 " 0.36 Ž n s 7. 2.13 " 0.34 Ž n s 7.
Pre-therapy 1.42 " 0.30 Ž n s 3; P - 0.007. 1.02 Ž n s 1. 1.33 " 0.01 Ž n s 2; P - 0.008.
Post-therapy I 2.03 " 0.26 Ž n s 3. 1.79 " 0.15 Ž n s 2; P - 0.04. 2.53 Ž n s 1.
Post-therapy II 2.06 " 0.20 Ž n s 2. 2.14 " 0.01 Ž n s 2. 1.88 " 0.14 Ž n s 2.

Cho r (Cr q PCr)

Control 0.84 " 0.10 Ž n s 7. 0.83 " 0.10 Ž n s 7. 0.87 " 0.10 Ž n s 7.
Pre-therapy 1.02 " 0.06 Ž n s 3; P - 0.008. 1.51 Ž n s 1. 0.82 " 0.04 Ž n s 2; P - 0.3.
Post-therapy I 0.81 " 0.09 Ž n s 3. 0.99 " 0.31 Ž n s 2. 0.72 Ž n s 1.
Post-therapy II 0.86 " 0.06 Ž n s 2. 0.98 " 0.02 Ž n s 2; P - 0.04. 0.89 " 0.02 Ž n s 2.

Ži. Except where indicated, P-values were non-significant. Žii. Post-therapy I corresponds to 6 weeks after initiation of therapy. Žiii. Post-therapy II
corresponds to 6 months after initiation of therapy.
 N.R. Jagannathan et al.r DeÕelopmental Brain Research 109 (1998) 179–186
deed, the disappearance of these peaks was evident even at
6 weeks after therapy by when serum T3 and T4 values
had normalised. Metabolite ratios from patient a1 for the
three regions showed near normal values after 1 year of
treatment. The trend of improvement was visible by 6
weeks in all these patients Žsee Table 2..
4. Discussion
A variety of biochemical abnormalities were docu-
mented from different brain regions of patients with con-
genital hypothyroidism. The most striking among these
was the presence of a large lipid peak in cerebellum and
frontal lobe of patient a1 which disappeared on treatment.
Since adequate care was taken to prevent interference by
factors, such as patient movement and displacement of
voxel position, which are known to degrade spectral qual-
ity w37x, it is unlikely that the broad resonances could be
consequent to such technical inadequacies. Also, spectra
associated with patient movement are hard to phase, but
such difficulty was not encountered in this patient. Proteins
and other macromolecules are also known to resonate in
the same chemical shift region as lipid but their resonances
are known to be of low intensity and observed at short
echo times ŽTE s 20 or 30 ms. when performed with J
editing w9x or metabolite nulling method w23x. Since our
experiments were carried out at an echo time of 135 ms,
the contribution from non-lipid macromolecules reso-
nances are not expected to be present due to then short
T2-relaxation times. Normally, the highly organised lipids
of myelin are invisible on MR spectroscopy w38x. How-
ever, in states characterised by breakdown of myelin and
other membrane structures mobile lipids are released which
are MR visible w54x. The pre-treatment lipid peak in patient
a1 may thus indicate presence of disorganised and imma-
ture myelin in cerebellum and frontal lobe. However, the
lipid peak was not observed in the thalamus probably
because it predominantly comprises of grey matter. Since
thyroxine is known to regulate enzymes involved in
myelinogenesis w1,5,16,41x post-treatment disappearance of
lipid peak suggests thyroxine induced myelin maturation.
Experimental hypothyroidism has been documented to
cause a variety of abnormalities in myelinogenesis
w20,29,31,40,52x. Cerebral hypomyelination in mice with
congenital hypothyroidism is reversed only when thyrox-
ine treatment is started before 20 days of life w39x, which is
equivalent to 1 year of post-natal life in man w41x. On the
basis of these current perceptions, abnormalities of myelin
function in humans with congenital hypothyroidism would
be reversible on T4 treatment only till the age of 1 year.
Disappearance of the lipid peak and normalisation of
metabolite ratios even when thyroxine therapy was started
at late ages Žbetween 7 and 31 years. in the present group
of patients suggests that disorganised myelination can be
reversed beyond this post-natal window period.
183
Recently, Bangur and Katyare w8x have adduced evi-
dence to support the concept of regional heterogeneity in
thyroxine dependent lipid anabolic processes relevant to
myelin formation in the developing rat brain. Our data in
humans from three discrete brain areas wfrontal lobe Žfore-
brain., thalamus Žforebrain. and cerebellum Žhindbrain.x
show different MRS patterns in untreated hypothyroid
state an observation analogous to the above animal experi-
ment.
The other important finding is the low level of NAA as
reflected by the reduced NAArŽCr q PCr. ratio, a finding
also documented by a recent proton MRS study on the
perchloric acid extract of neonatal hypothyroid rat brain
w47x. Since Cr is relatively constant throughout the normal
brain tissue and in different pathological conditions w15,43x,
reduction of this ratio can reliably be interpreted to be
equivalent to a decline in the levels of NAA. The levels of
NAA improved with thyroxine therapy.
In the rat brain, the concentration of NAA rises rapidly
to adult levels between days 10 and 20 of life w48x. This
fact has also been demonstrated by in-vivo 1 H MRS in
Ref. w21x. This is also the period of myelination in this
animal. Since myelination is a period of rapid lipid synthe-
sis, the simultaneous increase in NAA concentration sug-
gests that NAA might function as an acetyl donor to CoA
or to the acyl carrier protein involved in lipid synthesis.
D’Adamo et al. w12,13x have verified this hypothesis. Their
work has clearly demonstrated that NAA provides acetyl
groups for lipid synthesis in myelinating rat brain in-vivo.
In man, evidence for the role of NAA in myelin synthe-
sis is available from the study of Canavan’s disease, an
autosomal recessive leukodystrophy w42x. This disease is
characterized by spongy degeneration of white matter
largely because of extensive vacuolation and demyelina-
tion. There is no abnormality of neurons in this disorder.
This disease results from a deficiency of the enzyme
N-acetyl aspartoacylase which cleaves NAA into acetate
and aspartate w45x. Children with this disease have an
increase in NAArŽCr q PCr.and NAArCho ratios consis-
tent with the enzyme deficiency w14x. CT and MRI show
only diffuse white matter degeneration especially in the
cerebral hemispheres w30x. The brain has a spongy appear-
ance with defective myelin or lack of myelin. Brian biopsy
shows spongy degeneration of myelin fibres, astrocyte
swelling and enlarged mitochondria w11x. There is no men-
tion in literature of neuronal involvement. Biochemical
analysis reveals a loss of myelin lipids and considerable
reduction of phospholipids, cholesterol, cerebrosides and
sulfatides w11x. Further, in a recent study w50x, it was
suggested that deficient NAA catabolism affects myelin
metabolism. All these data suggest that NAA has a role in
myelin maturation.
As per prevailing concepts, NAA is recognised to be a
specific marker for neurons w54x or oligodendrocyte precur-
sors w51x. Therefore, the reduction in NAA may be at-
tributed to neuronal loss w7,22,25,54x. However, the post-
184 N.R. Jagannathan et al.r DeÕelopmental Brain Research 109 (1998) 179–186
treatment increase in NAA is unlikely to be due to increase
in neuronal numbers because neogenesis of neurons has
not been documented in the adult brain w44x. There are
reports which suggest that neurogenesis occurs in the
hippocampus, olfactory bulb and subventricular zone of
the adult mouse, though no such observation have been
recorded in humans or higher primates w2,3,27x. However,
recent studies of patients with multiple sclerosis reveal that
NAA concentration in viable neurons might change in a
reversible fashion w7,46x. A more acceptable explanation
would be that in the untreated state, there is cerebral
hypometabolism leading to a diminished synthesis of NAA
by viable neurons. With adequate thyroxine therapy, a
eumetabolic state is achieved, permitting an increase in the
synthesis of NAA by the same neurons. High levels of
NAA have also been reported in oligodendrocyte type-2A
Ž02A. progenitor cells w51x. Cells of this lineage are in-
volved in myelinogenesis. The initial lipid peak and low
NAA followed by a post-treatment increase in NAA and
disappearance of the lipid peak are features compatible
with onset of myelin synthesis and maturation suggesting
that thyroxine mediated activation of 02A progenitors may
also be occurring.
The other important metabolite ratio analysed was
ChorŽCr q PCr.. The pre-treatment ratio was higher in the
cerebellum than that observed in normals. After therapy,
there was a sharp decline in the ChorŽCr q PCr. ratio, the
value reaching close to that of controls. In proton MRS,
choline is a composite peak which has contributions from
free choline, phosphocholine and glycerophosphocholine.
The major part of choline is present as a polar head group
of the lipids in membranes and myelin w34x. Due to its low
mobility, only part of the choline is MR visible. Mobility
of the polar choline head group in the membranes is
affected by hydration and the presence of ions. Hence, the
decrease in the MRS choline peak may be consequent to
changes in dynamic behaviour of the membrane lipids or a
decrease in choline concentration. Myelin in the neonatal
brain is rich in choline containing compounds. However,
with maturation, the concentration of choline declines and
myelin is largely composed of galactolipid and sphingo-
myelin. Therefore, the post-treatment decline in the choline
peak can be interpreted to reflect thyroxine induced myelin
maturation. In a recent report on patients with congenital
hypothyroidism, a prominent choline resonance was ob-
served in the left parietal lobe which normalised with
thyroxine therapy w18x.
The NAArCho ratio was also calculated and found to
be low in all three regions of the brain in the untreated
state. Once again, with treatment, the ratio showed a
tendency to revert to normal levels. As discussed earlier,
the change may be secondary to the increased NAA peak
and reduced Cho resonance. The presence of lipid peak in
only one out of three patients indicates biochemical hetero-
geneity between patients of congenital hypothyroidism on
MRS. The presence of lipid peak may well be indicative of
severe degrees of myelination abnormalities. Even this was
corrected with thyroxine treatment. However, the function-
ing status of the brain appears unaltered as evidenced by
the unchanged IQ score of patient a1, 1 year after therapy.
These observations provide fresh insight into the relation-
ship between thyroid hormones and myelination in human
brain.
Acknowledgements
The authors wish to thank Prof. P.N. Tandon for many
useful discussion and Prof. P.A. Narayana, Department of
Radiology, The University of Texas Health Science Center
at Houston, U.S.A. for critically evaluating the manuscript.
Ms. P. Ghosh is thanked for her typographical assistance.
References
w1x G. Almazan, P. Honegger, J. Matthieu, Triiodothyronine stimulation
of oligodendroglial differentiation and myelination, Dev. Neurosci. 7
Ž1985. 45–54.
w2x J. Altman, G.D. Das, Autoradiographic evidence of post-natal hip-
pocampal neurogenesis in rats, J. Comp. Neurol. 124 Ž1965. 319–
336.
w3x J. Altman, Autoradiographic and histological studies of post-natal
neurogenesis: IV. Cell proliferation and migration in the anterior
forebrain with special reference to persistory neurogenesis in the
olfactory bulb, J. Comp. Neurol. 137 Ž1969. 433–458.
w4x C. Alves, M. Eidson, H. Engle, J. Sheldon, W.W. Cleveland,
Changes in brain maturation detected by magnetic resonance imag-
ing in congenital hypothyroidism, J. Pediatr. 115 Ž1989. 600–603.
w5x S.G. Amur, G. Shanker, R.A. Pieringer, Regulation of myelin basic
protein Žarginine. methyltransferase by thyroid hormone in myelino-
genic cultures of cells dissociated from embryonic mouse brain, J.
Neurochem. 43 Ž1994. 494–498.
w6x Z. Argov, P.F. Renshaw, B. Boden, B. Winokur, W.J. Bank, Effects
of thyroid hormones on skeletal muscle bioenergetics. In vivo
phosphorus-31 magnetic resonance spectroscopy study of humans
and rats, J. Clin. Invest. 81 Ž1988. 1695–1701.
w7x D.L. Arnold, P.M. Mathews, G. Francis, J. Antel, Proton magnetic
resonance spectroscopy of human brain in vivo in the evaluation of
multiple sclerosis: assessment of the load of disease, Magn. Reson.
Med. 14 Ž1990. 154–159.
w8x C.S. Bangur, S.S. Katyare, Effect of thyroidectomy and subsequent
replacement with L-thyroxine on lipid content and phospholipid
composition of discrete brain regions in the rat, Ann. Neurosci. 5
Ž1995. 33–39.
w9x K.L. Behar, D.L. Rothman, D.D. Spencer, O.A.C. Petroff, Analysis
of macromolecule resonances in 1 H NMR spectra of human brain,
Magn. Reson. Med. 32 Ž1994. 294–302.
w10x J.A. Cocks, R. Balazs, A.L. Johnson, J.T. Eayrs, Effect of thyroid
hormone on the biochemical maturation of rat brain: conversion of
glucosecarbon into amino acids, J. Neurochem. 17 Ž1970. 1275–
1285.
w11x J.N. Cumings, The lipidoses, in: P.J. Vinken, G.W. Bruyn ŽEds..,
Handbook of Clinical Neurology, American Elsevier Publishing,
NY, 1970, pp. 325–361.
w12x A.F. D’Adamo, F.M. Yatsu, Acetate metabolism in the nervous
system. N-acetyl-L-aspartic acid and biosynthesis of brain lipids, J.
Neurochem. 13 Ž1966. 961–965.
w13x A.F. D’Adamo, L.I. Gidez, F.M. Yatsu, Acetyl transport mecha-
 N.R. Jagannathan et al.r DeÕelopmental Brain Research 109 (1998) 179–186
nisms. Involvement of N-acetyl aspartic acid in de novo fatty acid
biosynthesis in developing rat brain, Exp. Brain Res. 5 Ž1968.
267–273.
w14x V. Englebrecht, M. Rassek, J. Gartner, J. Kahn, U. Modder, Mag-
netic resonance tomography and localized proton spectroscopy in 2
siblings with Canavan’s disease, Rofo Fortschr Geb Ront. Neven
Bildgeb verfahr 163 Ž1995. 238–244.
w15x G. Fein, D.J. Meyerholf, V. Disclafani, F. Ezekiel, N. Poole, S.
Mackay, W.P. Dillon, J.M. Constans, M.W. Weiner, 1 H magnetic
resonance spectroscopic imaging separates neuronal from glial
changes in alcohol-related brain atrophy, Alcohol and Glial Cells,
Research Monograph-27, National Institutes of Health, Bethesda,
MD, 1994, pp. 227–241.
w16x V. Ferret-Sena, A. Sena, F. Besnard, C. Fressinand, G. Rebel, L.L.
Sarlieve, Comparison of the mechanisms of action of insulin and
triiodothyronine on the synthesis of cerebroside sulfotransferase in
cultures of cells dissociated from brains of embryonic mice, Dev.
Neurosci. 12 Ž1990. 89–105.
w17x J. Frahm, K.D. Merboldt, W. Hanicke, Localized proton spec-
troscopy using stimulated echoes, J. Magn. Reson. 72 Ž1987. 502–
508.
w18x R.K. Gupta, V. Bhatia, H. Poptani, R.B. Gujral, Brain metabolite
changes on in vivo proton magnetic resonance spectroscopy in
children with congenital hypothyroidism, J. Pediatr. 126 Ž1995.
389–392.
w19x K.D. Hagspiel, C. von-Weymarn, G. McKinnon, R. Haldemann, B.
Marincek, G.K. Von Schulthess, Effect of hypothyroidism on phos-
phorus metabolism in muscle and liver: in vivo P-31 MR spec-
troscopy study, J. Magn. Reson. Imaging 2 Ž1992. 527–532.
w20x R.A. Harris, H.H. Loh, Brain sulfatide and non-lipid sulfate
metabolism in hypothyroid rats, Res. Commun. Chem. Pathol. Phar-
macol. 24 Ž1979. 169–179.
w21x K. Hida, I.L. Kwee, T. Nakata, 1 H and 31 P NMR spectroscopy of
the developing rat brain, Magn. Reson. Med. 23 Ž1992. 31–36.
w22x Hiehle Jr., R.E. Lenkinski, R.I. Grossman, V. Dousset, K.N. Ramer,
M.D. Schnall, J.A. Cohen, S. Genzalez, Correlation of spectroscopy
and magnetization transfer imaging in the evaluation of demyelinat-
ing lesions and normal appearing white matter in multiple sclerosis,
Magn. Reson. Med. 32 Ž1994. 285–293.
w23x J.H. Hwang, G.D. Graham, K.L. Behar, J.R. Alger, J.W. Prichard,
D.C. Rothman, Short-echo time proton magnetic resonance imaging
of macromolecule and metabolite signal intensities in the human
brain, Magn. Reson. Med. 35 Ž1996. 633–639.
w24x P. Kaminsky, B. Robin-Lherbier, F. Brunotte, J.M. Escanye, M.
Klein, J. Robert, M. Duc, Energetic metabolism in hypothyroid
skeletal muscle as studied by phosphorus magnetic resonance spec-
troscopy, J. Clin. Endocrinol. Metab. 74 Ž1992. 124–129.
w25x H.B.W. Larson, P. Christiansen, M. Jensen, J. Fredeviksen, A.
Heltberg, J. Olesen, O. Henriksen, Localized in vivo proton spec-
troscopy in the brain of patients with multiple sclerosis, Magn.
Reson. Med. 22 Ž1991. 23–31.
w26x P.R. Larsen, S.H. Ingbar, The thyroid gland, in: J.D. Wilson, D.W.
Foster ŽEds.., Williams Textbook of Endocrinology, W.B. Saunders,
New York, 1992, pp. 357–487.
w27x C. Lois, A. Alvarez-Buylla, Proliferating subventricular zone cells in
the adult mammalian forebrain can differentiate into neurons and
glia, Proc. Natl. Acad. Sci. USA 90 Ž1993. 2074–2077.
w28x T. Ma, Z.C. Lian, S.P. Qi, E.R. Heinz, G.R. DeLong, Magnetic
resonance imaging of brain and the neuromotor disorder in endemic
cretinism, Ann. Neurol. 34 Ž1993. 91–94.
w29x M.J. Malone, N.P. Rosman, M. Szoke, D. Davis, Myelination of
brain in experimental hypothyroidism. An electron microscopic and
biochemical study of purified myelin isolates, J. Neurol. Sci. 26
Ž1975. 1–11.
w30x R.K. Matalon, Defects in metabolism of amino acids: aspartic acid,
in: R.E. Behrman, R.M. Kliegman, A.M. Arvon ŽEds.., Nelson
Textbook of Pediatrics, W.B. Saunders, 1996, pp. 358–360.
185
w31x J.M. Matthieu, P.J. Reier, J.A. Sawchak, Proteins of rat brain myelin
in neonatal hypothyroidism, Brain Res. 84 Ž1975. 443–451.
w32x D.J. Meyerhoff, S. Mackay, N. Grossman, C. Van Dyle, N. Dillon,
G. Fein, M.W. Weiner, Effects of normal aging and Alzheimer’s
disease on cerebral 1 H metabolites, 10th Soc. Magn. Reson. Med.,
1992, p. 752.
w33x B.L. Miller, R.A. Moats, T. Shonk, T. Ernst, S. Wolley, B.D. Ross,
Alzheimer’s disease: depiction of increased cerebral myoinositol
with proton MR spectroscopy, Radiology 187 Ž1993. 433–437.
w34x B.L. Miller, A review of chemical issues in 1 H NMR spectroscopy:
N-acetyl-L-aspartate creatine and choline, NMR Biomed. 4 Ž1991.
47–50.
w35x R.A. Moats, T. Ernst, T.K. Shonk, B.D. Ross, Abnormal cerebral
metabolite concentration in patients with probable Alzheimer’s dis-
ease, Magn. Reson. Med. 32 Ž1994. 110–115.
w36x D. Moka, P. Theissen, A. Linden, W. Waters, H. Schicha, The effect
of hyper and hypothyroidism on the energy metabolism of skeletal
muscle—a study using P-31 magnetic resonance spectroscopy, Nuk-
learmedizin 30 Ž1991. 77–83.
w37x P.A. Narayana, J.J. Doyle, D. Lai, J.S. Wolinsky, Serial proton
magnetic resonance spectroscopic imaging contrast-enhanced mag-
netic resonance imaging and quantitative lesion volumetry in multi-
ple sclerosis, Ann. Neurol. 43 Ž1998. 56–71.
w38x P.A. Narayana, E.F. Jackson, Image-guided in vivo proton magnetic
resonance spectroscopy in human brain, Curr. Sci. 61 Ž1991. 340–
351.
w39x T. Noguchi, T. Sugisaki, I. Satoh, M. Kudo, Partial restoration of
cerebral myelination of the congenitally hypothyroid mouse by
parenteral or breast milk administration of thyroxine, J. Neurochem.
45 Ž1985. 1419–1426.
w40x J.M. Pasquini, I.A. Faryna de Raveglia, N. Capitman, E.F. Soto,
Neonatal hypothyroidism and early undernutrition in the rat: defec-
tive maturation of structural membrane components in the central
nervous system, Neurochem. Res. 6 Ž1981. 979–991.
w41x S.P. Porterfield, C.E. Hendrich, The role of thyroid hormones in
prenatal and neonatal neurological development—current perspec-
tives, Endo. Rev. 14 Ž1993. 94–106.
w42x J.M. Powers, A. Rubio, Selected leukodystrophics, Semin. Pediatr.
Neurol. 2 Ž1995. 200–210.
w43x M.C. Preul, D.L. Collius, D.L. Arnold, Differential diagnosis of
human intracranial tumours in vivo using 1 H-MR spectroscopic
imaging and feature space for spectral pattern recognition, in: S.D.
Saorvon, D.R. Fish, F. Andermann, G.M. Bydder, H. Stefan ŽEds..,
Magnetic Resonance Scanning and Epilepsy, Plenum, New York,
1993, pp. 299–313.
w44x P. Rakre, Limits of neurogenesis in primates, Science 227 Ž1985.
1054–1055.
w45x A. Shaag, Y. Anikster, E. Christensen, J.Z. Glustein, A. Fois, H.
Michelakakis, F. Nigro, E. Pronicka, M.T. Zabot et al., The molecu-
lar basis of Canavan Žasparto acylase deficiency. disease in Euro-
pean non-Jewish patient, Am. J. Hum. Genet. 57 Ž1995. 572–580.
w46x N.D. Stefano, P.M. Matthews, D.L. Arnold, Reversible decreases in
N-acetylaspartate after acute brain injury, Magn. Reson. Med. 34
Ž1995. 721–727.
w47x H. Sugie, S. Tsurui, A. Ishikawa, F. Matsuda, Y. Sugie, Y. Igarashi,
Y. Fujise, Effects of neonatal hypothyroidism on brain development:
analysis of brain metabolites using high resolution phosphorus and
proton magnetic resonance ŽNMR. spectroscopy, No. To. Hattatsu
ŽJapanese. 22 Ž1990. 166–172.
w48x H.H. Tallan, Studies on the distribution of N-acetyl-L-aspartic acid
in brain, J. Biol. Chem. 224 Ž1957. 41–45.
w49x D.J. Taylor, B. Rajagopalan, G.K. Radda, Cellular energetics in
hypothyroid muscle, Eur. J. Clin. Invest. 22 Ž1992. 358–365.
w50x P.B. Toft, R. Geiss-Holtorff, M.O. Rolland, O. Pryds, W. Muller-
Forell, E. Chritensen, W. Lehnert, H.C. Lou, D. Ott, J. Henning,
Magnetic resonance imaging in juvenile Canavan disease, Eur. J.
Pediatr. 152 Ž1993. 750–753.
186 N.R. Jagannathan et al.r DeÕelopmental Brain Research 109 (1998) 179–186

w51x J. Urenjak, S.R. Willams, D.G. Gadian, M. Noble, Specific expres-
sion of N-acetylaspartate in neurons oligodendrocyte type-2 astro-
cyte progenitors and immature oligodendrocytes in vitro, J. Neu-
rochem. 59 Ž1992. 55–61.
w52x T. Valcana, E.R. Einstein, J. Csejtey, K.B. Dalal, P.S. Timiras,
Influence of thyroid hormones on myelin proteins in the developing
rat brain, J. Neurol. Sci. 25 Ž1975. 19–27.
w53x M.S. Vanderknaap, B. Ross, J. Valk, Uses of MR in inborn errors of
metabolism, in: J. Kucharczyk, M. Moseley, A.J. Barkovich ŽEds..,
Magnetic Resonance Neuroimaging, CRC Press, Boca Raton, FL,
1994, pp. 245–318.
w54x J.S. Wolinsky, P.A. Narayana, M.J. Fenstermacher, Proton magnetic
resonance spectroscopy in multiple sclerosis, Neurology 40 Ž1990.
1764–1769.