[ Pobierz całość w formacie PDF ]
10
Spinal Dysraphism:
The Search For Magic
Timothy M. George and David Corey Adamson
CONTENTS
10.1 INTRODUCTION: WHY MAGIC?
Although pediatric neurosurgery is relatively young as a formal subspecialty of
general neurosurgery (the first meeting of the Section of Pediatric Neurological
Surgery was held in 1972 and the American Society of Pediatric Neurosurgery first
met in 1978), it has been practiced for millennia. Trephined pediatric skulls were
excavated in Peru and at other ancient sites.
1
The father of neurosurgery, Sir Victor
Horsley, performed his first epilepsy surgery on a child in 1886.
2
Harvey Cushing
wrote extensively about the unique disorders of childhood.
3,4
Many other notables
followed these icons, ensuring the momentum for further progress and refinement
in the surgical care of children with disorders of the nervous system.
5
Congenital spinal nervous system abnormalities continue to be the mainstays
and also the pitfalls of pediatric neurosurgery. Paralysis, incontinence, obesity, endo-
crinopathy, hydrocephalus, short stature, social stigmata, and shortened lifespans are
still the norms for children with open neural tube defects (NTDs).
6,7
The number of
children born with myelomeningocele has decreased over the past several decades
and 3 of every 10,000 children born in the U.S. are handicapped by open spinal
NTDs.
8
Additionally, improved imaging techniques have diagnosed even more chil-
dren who suffer from spinal cord dysfunctions secondary to closed dysraphisms.
9
The future treatment objectives are clear: congenital spinal defects must be prevented
or their neurological sequelae must be cured. Imagine that a pill or procedure could
© 2005 by CRC Press LLC
prevent or cure neurological deficits. Attaining such a goal often seems impossible.
It would seem to require magic — but what if magic did exist?
10.2 WHY MAGIC IS NEEDED: CAUSATIVE FACTORS
OF NTDS
Current hypotheses suggest that NTDs are caused by complex interactions between
extrinsic (drugs, environmental toxins, temperature, etc.) and intrinsic (genetic,
metabolic, etc.) variables. Clinical and epidemiological studies in humans have
implicated maternal illnesses, medications, environmental toxins, and dietary factors
such as folic acid that play causative or at least contributing roles in NTD develop-
ment.
10
Evidence that mutant genes cause NTDs has been supported by epidemio-
logical studies revealing an increased incidence of NTDs in certain families. In
familial cases, the trait for a NTD is semi-dominant, with apparent maternal inher-
itance.
8,11
Thus, NTDs represent examples of complex genetic disorders in which
genes and the environment interact through an unknown relationship.
Extrinsic factors causally related to NTDs have been studied extensively. Vita-
mins in general and folates in particular have been shown to significantly reduce
children’s risks of NTD,
12
particularly when siblings have NTDs.
13
The protective
mechanism of folate is unknown. Mothers have not been shown to be folate-deficient
or have defective intestinal uptakes of folate.
11
Studies of mutant enzymes in the
folate metabolic pathway, particularly, methylenetetrahydrofolate reductase
(MTHFR), suggest a possible association with NTDs.
14–16
However, Speer et al.
could not demonstrate MTHFR as a major risk factor.
17
Other cellular interactions,
such as cellular transport mechanisms, are currently under investigation.
18
Folate
can act as a methyl donor, permanently altering gene function via an epigenetic
mechanism
19
or interfering with a metabolic pathway such as homocysteine conver-
sion to methionine.
20–25
Although folic acid is irrelevant to the predominant basic
mechanism of action of folate, supplementation with folic acid has reduced the risk
of NTDs worldwide.
26–32
Other teratogenic agents and maternal diseases have been identified as causal
factors for NTD development.
10
Maternal diabetics have greater risks of having
children with diabetic embryopathies consisting of NTD and other organ system
anomalies.
33–35
Epileptic mothers using valproic acid as an anticonvulsant have a 1
to 2% risk.
36,37
Those exposed to carbamazepine face approximately 0.5% risk of
having children with NTDs.
38
Obesity has been associated with increased risk of
having a child born with a NTD.
39–41
A twofold increase in NTD incidence was also
found in obese versus non-obese mothers, regardless of use of vitamin, folate, and
other nutritional supplements. Febrile illnesses and hyperthermia produced by the
use of a sauna or hot tub early in pregnancy have been also suggested as causes of
NTDs.
10
The exact risk of occurrence due to maternal hyperthermia is not known.
Although strongly implicated, the specific genetic factors that cause NTD are
not known. It is proposed that many different genes are involved in neural tube
development. Some genes may confer strong genetic components and others may
© 2005 by CRC Press LLC
only exert minimal direct effects or require interaction with other genes. Environ-
mental factors may act as triggers to genetic susceptibility.
Several lines of evidence point to a genetic component. Empiric studies have
shown that the recurrence risk for NTD is greatest among first-degree relatives of
an affected patient and decreases for more distant relatives. The recurrence risk for
siblings of an affected patient is 2 to 5%, representing a 25- to 50-fold increase in
recurrence risk compared to the general population.
27
Techniques for identification
of specific genes are based on identifying populations at high risk, such as twins,
investigating the recurrence risks of NTD, and identifying mutated genes.
Mouse mutants have provided many of the genes investigated as candidates for
human NTDs. More than 40 mouse species have been described,
42
and the specific
gene identified in only 6 species. The six well-known mutations are splotch (Sp),
43–45
extra toes (Xt),
46
short tails (T),
47
patch (Ph),
48
and targeted mutations in apolipo-
protein B (ApoB)
49
and Hox-a1.
50
These mouse mutants provided clues to the
embryopathies of NTDs and identified potential candidate genes for human inves-
tigation. For example, the Pax-3 mutation in splotch mice mirrored the mutated Pax-
3 human homologue in Waardenburg’s syndrome.
51
Furthermore, several Waarden-
burg’s patients have been reported to have spina bifida.
23
Brachyury, when mutated,
is responsible for short-tailed mice, and has been shown to have an association with
human spina bifida.
52
Greig’s cephalopolysyndactyly corresponds to mouse Xt, with
patients revealing mutations in the Gli3 gene.
53
Pax-3, brachyury, and Gli3 have not
been shown to be major candidates for human NTDs.
54
10.3 WHERE MAGIC HAPPENS: DEVELOPMENT OF
THE EMBRYO
Normal nervous system development of an embryo requires proper formation of
embryonic axes. Determination of dorsoventral (DV) and anteroposterior (AP)
domains during gastrulation appears critical for normal neural development. Axis
patterning is reliant upon positional signals that provide DV and AP specifications.
55
Furthermore, positional signals appear essential to neural tube induction and pat-
terning.
56–64
Early embryonic axis determination is dependent on specification of anterior
axial mesoderm followed by posterior axial mesodermal induction.
55
A specified
group of cells (organizers) are known to function to organize the AP domains of the
embryonic axis.
65–68
Several genes (brachyury, goosecoid, noggin, XLIM-1, Not-1)
and diffusible morphogens (retinoids, activins, fibroblast growth factors) appear to
be important in the regulation of organizer activity, specifically in posterior devel-
opment of the axis.
56,58–60,62,69,70
The anterior axial mesoderm (chordamesoderm)
induces competent ectoderm to form archencephalic structures (telencephalon, dien-
cephalon, optic rudiment). The posterior mesoderm (notochord) induces competent
ectoderm to form the deuterencephalon (metencephalon, myelencephalon, cerebel-
lum) and spinal cord. Induced ectoderm forms the brain, hindbrain, and spinal cord
by the process of neurulation.
© 2005 by CRC Press LLC
Neurulation follows two stages: primary and secondary. Primary neurulation
71
begins after gastrulation when the primitive ectoderm is induced by the axial meso-
derm to form a neural plate. The neural plate undergoes further elevation, folding,
and fusion to form the neural tube. Neural crest cells migrate from the dorsal aspect
of the neural tube. Primary neurulation forms all functional levels of the brain and
spinal cord to the second sacral level in humans.
The caudal elements of the spinal cord, conus medullaris and filum terminale,
are formed by secondary neurulation,
72–78
which begins at a transitional zone where
the dorsally located primary neural tube overlaps the more ventral mesenchymal cells
of the tail bud in the future lumbosacral area. In this overlap zone, randomly arranged
mesenchymal cells condense to form the medullary cord. Radially oriented peripheral
cells surround a cellular central core in the medullary cord. Cavitation occurs centrally,
forming multiple lumina that coalesce to form a secondary neural tube.
The source of secondary neural tube cells is under scrutiny. Recent evidence in
chick embryos suggests that cells may migrate from more rostral neural plates to
attain their proper positions in the secondary neural tubes.
79,80
Normal caudal spinal
cord patterning in humans has been described
81
and abnormal patterning has been
demonstrated in dysraphic states.
82,83
Aberrant positional identity of caudal spinal
cord cells may be a consequence of disrupted positional signals, faulty differentia-
tion, or improper migration. Governing factors in the caudal neural tube pattern such
as the brachyury and Pax-3 patterning genes have not been identified as major factors
in spinal dysraphism.
54
10.4 MAGIC PILLS
Exciting and provocative evidence demonstrates that some manifestations of NTDs
are preventable or reversible at any one of numerous steps along the pathway from
preconception to childhood, and possibly even into adulthood (
Several
different therapeutic interventions (or “magic pills”) may be developed to treat the
remaining types of NTDs. These pills may target genetic loci, proteins, or any of
several metabolites involved in NTD development.
We now understand a great deal about the development of the neural tube, and
are quickly approaching a more complete genetic characterization of the process.
Ideally, NTDs could be detected early enough in development to target the defects
before any permanent manifestations occurred. The epidemiological studies
described definitively implicate maternal risk factors as well as inheritable and/or
acquired genetic influences that may be targeted. The combination of genetic, epi-
genetic, and environmental factors offers numerous targets for interventions.
Preconception would be the optimal time for prevention. Mothers with modifi-
able risk factors should be identified and counseled. Perhaps one of the most remark-
able advances in NTD treatment has been the introduction of periconceptional folic
acid supplementation for the prevention of myelodysplasias. Whether taken in pill
form or supplemented in dietary flour, this simple and inexpensive measure has cut
the incidence and devastating sequelae of myelomeningocele by more than half.
Despite this extraordinary achievement, it is still a challenge to prevent this unfor-
tunate disorder of aberrant neural tube closure.
© 2005 by CRC Press LLC
Pre-Conception
Magic Pill
Intra-Uterine Early
Intra-Uterine Late
Magic Repair
Postnatal
Lifetime
FIGURE 10.1
The magic phases of spinal dysraphism.
Other maternal risk factors that may prove important include good control of
diabetes, reduction of obesity and infections, vitamin supplementation (folate, inos-
itol, and vitamin B
12
), and avoidance of over-heated environments like saunas.
Additionally, mothers taking valproate and carbamazapine antiepileptic medications
should discontinue use or take other medications if possible to eliminate the
increased risk.
It may be possible in some cases to identify mothers with inheritable genetic
predispositions and counsel them during the preconception period in preparation for
possible treatment during pregnancy. Several possible medications could be devel-
oped to provide genetic targeting during early fetal development. Tools for targeting
candidate genes at the DNA, RNA, or protein level are all plausible possibilities.
These tools could target defects in genes involved in proper neural tube patterning,
folate-dependent and -independent mechanisms, or healing mechanisms. The next
decade certainly will see attempts at
in vitro
correction of genetic defects during the
blastocyst stage or manipulation of these genes
in utero
via delivery systems like
viral vectors.
Several studies with animal models have elucidated some of the genes involved
in the induction of proper neural tube development, for example, Wnt-1, Gnot1 (a
notochord family homeobox gene), HOX-1, and activin.
50,56,60
Activin and retinoic
acid regulate Gnot1 expression prior to gastrulation. The neural tube-inducing prop-
erties of sonic and bone morphogenic protein genes are also under intense investi-
gation. The Sp mouse model has defects in neural tube closure due to mutations in
the Pax-3 paired box gene.
44,45
When genes are deleted or mutated, the fetal cells
© 2005 by CRC Press LLC
[ Pobierz całość w formacie PDF ]