Research
Interests
With the completion of the human genome project, it has become clear that the sheer number of genes cannot account for the complexity of the human proteome. Among several proposed mechanisms, alternative pre-mRNA splicing is considered to be one of the most efficient and widespread avenues to generate multiple protein isoforms from individual genes. Current estimates indicate that over 60% of all human genes undergo alternative splicing, thus greatly increasing the coding potential of our genome. In addition, defects in splicing have been linked with a variety of human genetic diseases. Understanding the basic mechanisms of pre-mRNA splicing and splice site recognition is therefore fundamental to understanding the expression of genes and human diseases.
In higher eukaryotes, alternative splicing has been associated with essentially all facets of biology. Research in the Hertel Laboratory focuses on understanding the mechanisms that allow for the generation of alternative splicing patterns. Specifically, we are interested in gaining insights into the most critical step of generating mRNA diversity; the processes of splice-site selection and pairing. We are taking a number of quantitative and computational approaches to
(1) investigate how splicing regulatory elements and the exon/intron architecture influence exon inclusion,
(2) determine when and how the spliceosome commits
to splice site pairing,
(3) determine how the coupling of Pol II transcription and pre-mRNA splicing influences splice site choice, and
(4) develop molecular strategies to modulate
alternative splicing in vivo.
The unifying theme in these lines of investigation is
the study of biochemical events that control alternative splicing
patterns. Our long-term goals are
to relate these basic mechanisms of splice-site recognition to biological
processes and to identify strategies to manipulate the expression of splicing
isoforms in disease genes.
The
role of splicing regulatory elements and the exon/intron architecture in
influencing exon inclusion. Our overall goal is to develop a quantitative
framework to predict exon inclusion from sequence analysis. In the past years we have investigated
the mechanisms of exonic splicing enhancer (ESE)-dependent alternative 5’
splice site choice by using an in vitro system. Our analysis
showed that ESEs function similarly in activating regulated 5' and 3' splice
sites, suggesting that exonic splicing enhancers recruit multiple spliceosomal
components required for the initial recognition of 5' and 3' splice sites. Indeed, we demonstrated that an
individual enhancer complex is sufficient to activate both weak splice sites
(see paper). Thus, ESEs recruit a complex that minimally contains factors
necessary for both 3' and 5' splice site recognition. However, it has become apparent that the regulation of
alternative splicing is much more complex than ESE-dependent splice-site
activation. Intronic enhancers
(ISE), exonic silencers (ESS), and intronic silencers (ISS) appear to be as
abundant as ESEs. In addition, our
laboratory has demonstrated that the exon/intron architecture significantly
influences the nature and likelihood of alternative splicing within the human
and Drosophila genomes (see paper).
Thus,
exon inclusion depends on the combinatorial activities of multiple cis-acting
elements. It is not known to what
extent the downstream or upstream splices site, the exon/intron architecture,
or splicing regulators influence the overall probability of exon
recognition. Using our strength in
quantitative analysis, we are carrying out a systematic investigation to
determine how the relative contributions of splice-site strength, exon length,
and splicing regulatory elements alter the probability of exon inclusion.
Determine when
and how the spliceosome commits to splice site pairing. Recent
work from our laboratory answered a critical question that has remained elusive
for almost 20 years of studying pre-mRNA splicing – when does splice site
pairing occur? After definition of
the exons, the spliceosome is activated by a series of sequential structural
rearrangements. Formation of the
first ATP-independent spliceosomal complex commits the pre-mRNA to the general
splicing pathway. However, the
time at which a commitment to a specific splice site choice and pairing is made
has been unknown. We were able to
demonstrate that alternative splicing patterns are irreversibly chosen at a
kinetic step different from the ATP-independent commitment to splicing. Splice sites become committed at the
first ATP-dependent spliceosomal complex when rearrangements lock U2 snRNP onto
the pre-mRNA. Thus, commitment to
the splicing pathway and commitment to splice site pairing are separate steps
during spliceosomal assembly and ATP hydrolysis drives the irreversible
juxtaposition of exons within the spliceosome (see paper).
The results set the stage for further investigations geared towards
understanding the molecular and physical events that lead to irreversible
pairing. Our current studies
address the biochemical nature of splice-site pairing to evaluate the generality
of our findings, to determine the role of ATP hydrolysis during splice-site
pairing, and to determine which components of the splicing machinery are the
executioners of splice-site pairing.
The
coupling of Pol II transcription and pre-mRNA splicing. The
in vitro analysis of splicing is
normally carried out with pre-synthesized RNA transcripts. However, RNA processing in vivo is carried out in close proximity to the site of
transcription, allowing for co-transcriptional regulation of alternative pre-mRNA
splicing. We developed an in
vitro assay for transcription and
splicing to assay the kinetics of pre-mRNA and mRNA formation. Our analysis demonstrates that an
association of RNA Pol II transcription and pre-mRNA splicing is required for
efficient gene expression. Newly
synthesized RNAs containing functional splice sites are protected from nuclear
degradation, presumably because the local concentration of the splicing
machinery is sufficiently high to ensure its association over interactions with
nucleases. Furthermore, the
process of transcription influences alternative splicing of newly synthesized
pre-mRNAs. Because other RNA
polymerases do not provide similar protection from nucleases and their RNA
products display altered splicing patterns, the link between transcription and
RNA processing is RNA Pol II specific (see paper). We are currently investigating how the
connection between transcription by Pol II and pre-mRNA splicing guarantees an
extended half-life and proper processing of nascent pre-mRNAs.
Regulation of
pre-mRNA splicing and human genetic diseases. Proximal spinal muscular
atrophy (SMA) is a common human genetic disease that is the leading cause of
hereditary infant mortality. It is
characterized by the progressive degeneration of the anterior horn stem cells
of the spinal cord with consequent paralysis of the trunk and limbs. Three clinical groups of the disease (I
- III) have been described based on the decreasing severity of the symptoms. SMA has been linked to deletions or
mutations of the Survival of Motor Neuron (SMN) gene that has been mapped as an
inverted repeat to chromosome 5 at 5q13.
Homozygous absence of the telomeric copy (SMN1) correlates with
development of SMA. By contrast,
alterations within the centromeric SMN gene (SMN2) do not produce any known
phenotype. A genomic sequence
comparison of the two genes revealed that SMN1 and SMN2 encode for the
identical protein. However, three
alternatively spliced transcripts generated with different efficiencies have
been described for each locus.
SMN1 primarily produces the full-length form of SMN, whereas
differential splicing of the SMN2 pre-mRNA predominantly produces an isoform lacking
exon 7 (SMN∆7). Comparison
of SMN transcripts revealed a direct relationship between SMA and exon 7
skipping. The only critical
nucleotide change between SMN1 and SMN2 affecting the inclusion of exon 7 has
been pinpointed to a single C to T base difference located six nucleotides
inside exon 7. This transition
disrupts a putative ESE and fortuitously creates a splicing silencer
element. As a consequence, exon 7
is skipped in SMN2 and SMN∆7 is the product of an alternative processing
event. Because all individuals
with SMA have retained their SMN2 allele, therapy directed towards increasing
SMN2 exon 7 inclusion could provide a promising tool to lower the clinical
severity of SMA. Our research has
focused on testing molecular strategies that redirect the SMN2 splicing pattern
to produce full length, viable SMN protein. We have developed an antisense oligonucleotide-based
strategy that alters the splicing pattern of the SMN2 transcript. We demonstrated that the application of
antisense oligonucleotides targeting the 3' splice site of SMN2 exon 8
efficiently tilted the balance of splice site competition in favor of exon 7
inclusion (see paper). Thus, we have a technology at hand that
allows SMN2 to be turned into a true backup gene. As a natural extension of these promising results, we are
now focusing on testing the antisense strategy developed in SMA animal models.