|
 When
important advances in the neurosciences relating to childhood disorders
occur, the scheduled series of columns will be interrupted. Recently,
the gene that causes Rett syndrome was identified. The significance and
relevance of this finding to our work as clinicians will be discussed
in this column.
Rett
syndrome is a neurodevelopmental disorder within the autism spectrum.
The prevalence is estimated at 1 in 10,000. Because it is a sporadic illness,
initially it was difficult to establish its genetic basis. Females are
almost exclusively affected. These children develop normally for the first
year or so of life. Milestones are reached on time, and no abnormalities
are apparent. Then a rapid regression occurs, including loss of purposeful
hand movements. Other clinical findings include growth retardation, loss
of speech, microcephaly, ataxia, and often a severe interruption of normal
cognitive progression. Affected children then enter a period of relative
stability that lasts for decades after the initial regression.
The
simplest explanation for the genetic findings is that the responsible
gene lies on the X chromosome. If the mutated gene is present in males,
they die as fetuses, as no additional X chromosome is available to produce
functional protein. This explains the absence of the disorder among males.
In females, the second X chromosome carries a complementary copy of the
gene that provides some degree of protection. However, an insufficient
amount of the protein is available and symptoms develop after the first
few months of life.
The first task was to determine the chromosomal location for the responsible
gene. Sufficient numbers of affected sisters had been characterized to
allow investigators to focus their attention on an area of the X chromosome
(Xq28). The search then began to identify the genes present in the region.
A number of candidate genes were carefully examined for mutations, and
these genes were systematically excluded. A recently isolated gene was
found to be mutated in several individuals with the disorder. The gene
encodes the methyl CpG-binding protein 2 (MeCP2). Mutations in 2 critical
domains of the protein were discovered to be present in several affected
probands.
What is the normal function of this protein? In higher organisms, such
as vertebrates, a large number of genes are expressed in a tissue-specific
manner. In fact, more than one third of genes are expressed only within
the brain. Many of these genes are needed during critical periods of CNS
development, and their expression must then be turned off. Other genes
are required only after birth, and turning these genes on at the appropriate
time is critical to normal and proper development. In addition, some genes
are expressed only if they lie on either the maternal or paternal chromosome,
a phenomenon called genomic imprinting (see recent columns). Disruption
of this normal genetic mechanism is responsible for several human disorders,
including Prader-Willi and Angelman syndromes.
A
mechanism has evolved to regulate which genes are expressed and which
must be repressed in both genomic imprinting and normal development. The
stable silencing of a large fraction of genes allows cells to transcribe
only those genes that are essential for a particular cell type. One of
the first steps in this process involves the modification of chromosomal
DNA by the addition of methyl groups to the cytosine nucleotide. Furthermore,
the methylation that occurs is particularly enriched within stretches
of DNA containing numerous cytosine and guanine pairs, so-called CpG islands.
Such CpG islands are present throughout the genome, but they are most
numerous within the promoters of genes, which are the regions of a gene
known to regulate its expression pattern.
Initially,
it was speculated that DNA methylation alone was sufficient to prevent
the initiation of transcription. This turns out not to be the case. Several
additional modifications are required to interfere with the binding of
the transcriptional machinery to the promoter region of genes. Specific
methyl-binding proteins must first recognize the methylated regions. The
chromatin structure is then modified so that the promoter is no longer
accessible to transcription factors or other enzymes required to initiate
transcription. This regulatory mechanism allows the amounts of specific
proteins to be finely tuned over the life of the organism.
The
relationship between DNA methylation, chromatin structure, and gene expression
has been recognized for some time. The protein that was recently implicated
in Rett syndrome, MeCP2, links 2 steps in gene regulation: DNA methylation
and histone deacetylation. MeCP2 has at least 2 functional domains that
carry out these dual functions. At one end, a domain of the protein recognizes
methylated CpG islands and allows the protein to bind tightly to these
sequences. The second functional domain is then activated. This region
is responsible for the recruitment of another group of proteins, termed
the histone deacetylase complex, to the segment of DNA that must be repressed.
The protein complex chemically modifies the structure of certain histones
that are themselves highly enriched within the nuclei of all cells. Histones
have long been known to associate with DNA and are required for compaction
of DNA into packaged chromatin. Once histones are modified by deacetylation,
they compact the DNA even further, making the underlying genes less accessible
to transcriptional activators.
The
discovery of mutations in a key player in this process suggests that patients
with Rett syndrome suffer from inappropriate transcriptional activity.
This hypothesis, while intriguing, requires more experimental findings.
A number of laboratories are actively working to test it. However, it
is reasonable to conclude that mutations in MeCP2 lead either directly
or indirectly to inappropriate amounts of transcription of downstream
genes that MeCP2 would normally silence.
The
recent findings raise a number of questions. Why arenšt symptoms noted
before 1 year of age? A similar delay occurs in a number of neurodegenerative
disorders, including Huntingtonšs chorea. One possible explanation for
the time lags has been that mutations of a number of different genes results
in the production of toxic compounds, such as free radicals. Mutations
in proteins needed to inactivate these compounds result in an inability
to detoxify them, and they accumulate over time, eventually damaging neuronal
structure and function.(Fig.
1).
It
is possible that a similar mechanism is occurring in the brains of Rett
syndrome patients, and neuroanatomical findings might help to clarify
this issue. Unfortunately, only a single report exists in the literature
of the neuropathological findings in Rett syndrome. Abnormalities were
found in the architecture of cortical pyramidal neurons in layer II and
III, which had a smaller number of dendrites than is typically seen. Moreover,
the normally occurring arborization of these neurons was absent, with
a corresponding decrease in the number of dendritic spines.
Another
question that remains unanswered relates to predominance of neurological
symptoms. MeCP2 is expressed in a number of tissues in the body as well
as in the brain, and yet disruptions of these other organ systems are
not a significant part of the clinical findings. It is possible that the
brain is particularly sensitive to disruption in the activity of MeCP2,
although this will need to be determined experimentally. Finally, if mutations
in MeCP2 lead to inappropriate expression of downstream genes, what are
these target genes? It is possible that mutations will be found in these
downstream genes that might result in the appearance of other developmental
disorders.
Several
additional points should be made. MeCP2 is one member of a larger family
of proteins. Two other members of this family have been shown to bind
to DNA and prevent transcription of downstream genes. These family members
as well as their target genes are now additional candidate genes that,
when mutated, may cause similar disruptions to normal brain development
and functioning. Moreover, as was discussed above, several enzymatic steps
are required before genes can be repressed. Proteins that participate
in this process include those present in the histone deacetylase complex
as well as MeCP2 and probably a number of other yet-to-be-identified proteins.
It is likely that mutations in some of these genes will have an effect
on gene expression through inappropriate regulation of transcription.
Six
different types of mutations were found in the 7 children who had mutations
present in their MeCP2 gene. Two of these children were half-sisters
and carried the same mutation. Several mutations are missense mutations
that substitute one amino acid for another. These missense mutations are
found in the methyl-binding domain and are thought to disrupt its ability
to recognize and interact with methyl groups. The remaining mutations
are found within the second functional region, the domain that recruits
the deacetylation complex. Mutations in this domain also result in disruption
of transcriptional repression. One of these mutations is caused by the
insertion of a single nucleotide that leads to a shift in the downstream
nucleotide sequence. This results in a completely different amino acid
sequence in the portion of the protein beyond the mutation. The final
mutation that was found results in a novel stop codon in the DNA sequence
and causes the production of a shortened or truncated protein. Both of
these mutations lead to severely impaired or nonfunctional protein being
synthesized.
Separate
mutations in a single gene are termed allelic heterogeneity. In the majority
of disease-causing mutations studied so far, when different regions of
a single gene are mutated, they lead to the same clinical phenotype among
affected individuals. Occasionally, however, different mutations within
the same gene will produce different clinical presentations. For example,
mutations to the receptor for an important family of growth factors, fibroblast
growth factors, lead to several clinical syndromes depending on where
the mutation occurs.
It is important to note that mutations within the MeCP2 gene were
found in only 7 of the 21 Rett syndrome patients studied. Several possibilities
might explain these findings. Only the exonic DNA sequences that encode
the MeCP2 protein were sequenced. Neither the intervening DNA segments,
called introns, nor the promoter or additional regulatory regions were
sequenced. It is likely that mutations in these regulatory areas will
be found in some of the remaining patients with Rett syndrome in whom
the coding sequence of the MeCP2 gene appears normal. If these
additional mutations are found, they would be further examples of allelic
heterogeneity in which mutations within a single gene disrupt its ability
to produce sufficient amounts of functional protein.
It
is also possible that mutations will be found in genes other than MeCP2,
and yet the clinical symptoms of Rett syndrome will again be seen. Mutations
that are present in separate genes within a single enzymatic pathway often
lead to very similar clinical symptoms among affected individuals. These
types of mutations are examples of genetic heterogeneity. In the case
of Rett syndrome, this might be caused by mutations present in some of
the other proteins required for the proper methylation of DNA sequences,
or the deacetylation of histones. It will be interesting to determine
whether disruptions in the functional activity of these proteins are responsible
for some of the other cases of Rett syndrome not caused by mutations in
the MeCP2 gene. A number of laboratories are actively working on
this question.
top of page
|
