Introduction
The disease epilepsy has been known for ages. Around 1% of the world's population
is estimated to have epilepsy. The term idiopathic epilepsy is no longer used. Antiseizure
medications (ASMs) are the mainstay in the treatment of epilepsy. Most patients with
epilepsy respond well to the existing ASMs and remain seizure free. Many patients
also attain complete remission and remain seizure free for the rest of their lifetimes.
Despite all this, some patients respond poorly to the existing treatment modalities,
and treating them becomes challenging. The reason for variations in response among
patients with epilepsy is still unknown. So, this remains an area of research to date.
Numerous agents were repurposed and tried with mixed results. And still, conventional
ASMs remain the mainstay of treatment. One among the alternative agents that were
newly tried is the antisense molecules. Targeting intracellular targets that are usually
not druggable is possible with these agents. This review focuses on the antisense
molecules targeted for epilepsy and their progress in epilepsy therapeutics.
What Are Antisense Molecules
The term antisense molecules comprise several classes of oligonucleotide molecules.
This contains a sequence complementary to target RNA (tRNA) molecules, such as messenger
RNA (mRNA), viral RNA, or other RNA ([Fig. 1]) species that inhibit the function of their tRNA after sequence-specific binding.
In the meantime, at least four classes of antisense molecules have been described:
antisense oligodeoxyribonucleotide, that is, single-stranded DNA molecules, small
interfering RNA (siRNA) molecules, ribozymes, and DNA zymes ([Table 1]).
Fig. 1 Depiction of the mechanism by which the antisense molecules act. (Adapted from Robinson
R. RNAi Therapeutics: how likely, how soon? PLoS Biol 2004;2(1): e28).
Table 1
The challenges faced and the effects of various antisense molecules
|
Antisense DNA (ODN)
|
siRNA
|
DNA zyme
|
RNA zyme
|
|
Oligo type and structure
|
Linear ssDNA
|
Linear dsRNA
|
ssDNA with two binding domains surrounding a central catalytic domain
|
Complex RNA with single and double-strand section
|
|
Size
|
12–25 bases
|
21–25 base pairs
|
30–35 bases
|
>30 bases (hammerhead RNA zymes)
>50 bases (hairpin RNA zymes)
|
|
Inherent enzymatic activity
|
No
|
No
|
Yes
|
Yes
|
|
Recruited enzymatic activity
|
RNAses H, L, or P
|
RISC (RNA-induced silencing complex)
|
No
|
No
|
|
Modifications
|
Multiple backbone modifications were established, such as PT, 2-O-methyl, 2-O-methoxyethyl,
locked nucleic acids
|
Several changes, preferentially of terminal nucleotide overhangs found, such as 39-inverted
thymidine
|
39-inverted thymidine to enhance stability modifications in binding arms possible
|
Often unmodified (mainly when expressed in vivo), several modifications are possible
|
|
Advantages
|
• Easy to produce
• Easy to modify
• Good cell penetration, especially for modified short antisense DNAs
|
• Can be generated intracellularly from more significant precursors of the whole mRNA
|
• Easy to produce
• Multiple potential cleavage sites in target RNAs
• Good in vivo cell penetration
• No significant off-target effects
|
• Coding DNA sequences can express multiple potential cleavage sites in target RNAs
in vivo
|
|
Disadvantages
|
• Off-target effects
• Potential protein-binding (aptamer) activities
• Activity dependent on intracellular enzymes
|
• Off-target effects
• More challenging to produce limited cell penetration
• Activity dependent on intracellular enzymes
|
• Empiric selection process
• Activity depends on intracellular Mg21 levels
|
• Large size and complex structure
• Empiric selection process
• Challenging due to limited in vivo stability
|
The mechanism by which the antisense molecules act, as explained above, is depicted
below
Why Are They Important?
Epilepsy is a common disorder affecting many people worldwide.[1] Currently, more than 20 small-molecule drugs are used for epilepsy. The clinical
outcomes are good for two-third of the patients, but the rest show a poor response.[2] In many patients with epilepsy, the response to ASMs reduces with time. Patients
who respond poorly to the existing therapies need multiple ASMs to attain seizure
control. One classic example of such a scenario is temporal lobe epilepsy (TLE)[3]
[4]—many patients with TLE progress to a stage where they require epilepsy surgery.
Most ASMs today are designed to target ion channels. They mostly rely on modulating
the conduction in neurons. Long-term use of these agents can develop resistance and
poor response in those patients.[5] Novel approaches that can modify multiple targets up to the genetic level may be
necessary to combat this problem. ASO's strategy in treating epilepsy can help us
overcome our current challenges in treating drug-resistant epilepsy.[6]
Epilepsy and Antisense Oligonucleotides
Developmental Epileptic Encephalopathies
Developmental epileptic encephalopathies (DEEs) comprise a group of disorders characterized
by epileptic seizures, which are difficult to treat with current-day ASMs.[7] The gene implicated in these conditions is SCN2A (sodium channel 2A) that encodes
the αsubunit of the voltage-gated sodium channel (Nav1.2). Nav1.2 is involved in the
initiation and conduction of action potentials. In DEE patients, Nav1.2 is the predominant
isoform that leads to excessive excitation compared to people with other isoforms
(Nav1.1, 1.3, and 1.6).[8]
[9] DEE is associated with de novo mutations, which constantly increase as the disease
progresses.[10] Functional analysis has shown a gain of function mutation in the SCN2A gene that
encodes for Nav1.2 associated with this disease, so gene therapy targeted to reduce
SC2NA overexpression will be promising.[11]
[12] This strategy was successful in preclinical studies done in a mouse model by Li
et al. Li et al administered an antisense oligonucleotide (ASO) targeting mouse Scn2a
(Scn2a ASO) to a mouse model of DEE. The mice models of DEE used were engineered with
human equivalent SCN2A p.R1882Q mutation. Intracerebroventricular Scn2a ASO administration
into the mutated mice between postnatal days 1 to 2 significantly extended lifespan
and reduced Scn2a mRNA levels by 50%.[13]
Dravet Syndrome
Dravet syndrome (DS) is another disorder that belongs to the group of diseases called
epileptic encephalopathies. DS is caused by de novo mutations SCN1A gene. The mutations result in the haploinsufficiency of the voltage-gated sodium
channel α subunit NaV1.1, which leads to abnormal channel function that manifests
as epileptic seizures.[14] Multiple seizure types characterize DS. Patients with DS are resistant to usual
ASMs. The disease is characterized by seizures, cognitive deficits, ataxia, and increased
mortality.[15]
[16] There is a risk of sudden unexpected death in epilepsy (SUDEP) in all people with
epilepsy. In DS, SUDEP is increased by 20% compared to other people with epilepsy.[17] Recently developed and repurposed small molecules show only a partial improvement
in the seizure in patients with DS. This includes stiripentol, Epidiolex, and fenfluramine.[18]
[19]
[20] Again, these drugs do not target the direct pathophysiology of DS. The haploinsufficiency
of the SCN1A is essential in understanding the underlying pathology. ASOs employed
in DS targets the exon sequence. Such binding helps remove the nonfunctional exon
sequences interrupting with the normal exon, thereby helping produce more abundant
normal functioning SCN1A mRNA. ASO approach was found to be successful in various
preclinical studies. In the F1:129S-Scn1a ± × C57BL/6J mouse model of DS, a single
intracerebroventricular dose of ASO at postnatal day 2 or 14 reduced the incidence
of seizures. SUDEP was also reduced in the ASO group. Increased expression of the
normal SCN1A transcript and NaV1.1 protein was seen in the brains of mice treated
with ASO.[21] There was a reduction in SUDEP by 97% in mice treated with specific ASOs. Anti-SCN1A
antagonists are antisense molecules against noncoding RNA, which are tried in DS.[22] Another mutation detected by Lenk et al is the SCN8A mutation.[23] The SCN8A gene encodes for Nav1.6. Mutations in SCN8A lead to encephalopathy and
DS. In the experiment by Lenk et al ASOs targeted against the SCN8A exon sequence
were used in mouse models of DS. The mean survival increased from 3 weeks to more
than 5 months in DS.[24] Targeted augmentation of nuclear gene output (TANGO) introduced by Stoke therapeutics
STK-001 has shown promising results in mouse models of DS. The first human trial in
18 children with DS, called the MONARCH trial, was conducted to assess the safety
of STK-001 in the case of DS. Results of this phase I trial have shown an excellent
tolerability profile up to doses of 30 mg in children. TANGO uses an ASO that works
by binding to splice pre-mRNA to promote a process called nonsense-mediated mRNA decay
exon exclusion. TANGO ASOs reduce the DNA that encodes for nonproductive proteins,
restoring the protein output to near-normal or sometimes normal levels.[25]
Another selective gene therapy by encoded therapeutics, ETX101, aims to increase SCN1A
gene expression, which has shown promising preclinical results in mouse models with
DS.[26] CRISPR-associated protein 9 in a deactivated form increases the gene expression
of SCN1A in preclinical cell line-based studies and mouse models.[27] Zogenix is another tRNA-based therapeutic to increase SCN1A expression. The first
therapy to target tRNA, overcome nonsense mutations, and ensure the production of
functional Nav1.1 is Zogenix. This tRNA-targeting molecule is currently under preclinical
evaluation.[28]
Channelopathies and Epilepsies
Our conduction system in the brain functions via various channels. Sodium, potassium,
and calcium channels (CACN) are the most important and are integral to normal neurotransmission.
When any disease affects these channels, it is referred to as channelopathy. These
channelopathies can manifest as epileptic seizures and are often difficult to treat.
Potassium Channels in Epilepsy
Recent research has suggested that potassium channels (KCNA) play a significant role
in epilepsy. The study done in preclinical animal models indicated that the KCNA mutations
associated with KCNA1 are involved in SUDEP in the case of epilepsy. Kv1.1 deficiency
leads to impaired neural control and cardiac rhythmicity. The pathology is assumed
to be due to aberrant parasympathetic neurotransmission. This abnormality suggests
KCNA1 could be a strong candidate gene for human SUDEP.[29] Another gene, LGI1, is found to be responsible for the proper functioning of KCNA
in neurons. LGI1 is associated with autosomal dominant lateral temporal lobe epilepsy
and autosomal dominant partial epilepsy with auditory features.[30] Patients affected by these disorders carry LGI1 mutations.[31]
[32] Since these genes were identified, antisense molecules can be developed and tried
in these conditions.
Calcium Channels in Epilepsy
Another vital channel involved in epilepsy is the brain's CACN, which is implicated
in the pathogenesis of epilepsy. The CACN called CaV2.1 medium is a P/Q channel involved
in epilepsy. The others include CaV3.1 and CaV3.2b and are found to be strongly associated
with absence seizures in preclinical studies.[33] In the Genetic Absence Epilepsy Rats of Strasbourg model, the rats did not manifest
absence seizure episodes with 7–9-Hz spike-wave discharges. An R1584P missense mutation
in CACNA1H, which encodes CaV3.2, was identified in the rats. CACNA1H, if expressed above a certain threshold,
shows a gain of T-type channel function with the appropriate splice variant (+exon
25). This expression increases with age.[34]
[35] Again, this could be a potential target for future ASOs development.
Lafora Disease and Antisense Molecules
Lafora disease is a severe progressive form of myoclonic epilepsy. Lafora disease
manifests itself in the adolescent age group. The disease is characterized by the
formation of polyglucosan inclusions called Lafora bodies. The mutations, namely EPM2A
(laforin) and EPM2B (Malin) genes, are associated with this disease.[36] Laforin is an essential dual specificity phosphatase crucial for neurons' survival.
Mutations in the EPM2A gene that encodes for laforin affect the exon sequence. Due
to this, dysfunctional laforin proteins are produced that get deposited in the form
of Lafora bodies.
Additionally, the EMP2B gene encodes for an E3 ubiquitin ligase involved in the polyubiquitylation
of laforin and its degradation. Dysfunctional malin fails to clear laforin and leads
to the accumulation of Lafora bodies within neurons. Conventional ASMs are the mainstay
for treating Lafora disease. Valproate mainly controls myoclonus in Lafora condition.[37] The new emerging treatment for Lafora disease still in the preclinical stage is
Asos. The one ASO tried is Gys1- ASO, which targets the mRNA in the brain and causes
a significant reduction in Lafora body production in mouse models. Intracerebroventricular
injection of Gys1-ASO prevented Lafora body formation in young mice that had not yet
formed them. In older mice that already exhibited Lafora bodies, Intracerebroventricular
injection of Gys1- ASO inhibited further accumulation, markedly preventing large Lafora
bodies characteristic of advanced disease.[38] So, hopefully, this might be promising in the case of Lafora disease.
Temporal lobe Epilepsy and Antisense Molecules
TLE is the most common type of acquired epilepsy in adulthood.[39] According to seizure symptomatology, this is divided into mesial TLE and a rare
variant called lateral TLE. ASMs are the mainstay in TLE, but the primary concern
is their long-term use and resistance.[39] Our understanding of TLE is mainly based on imaging, clinical, and electrophysiological
data. The most common finding in the case of TLE is hippocampal sclerosis.[40] TLE can occur primarily in some instances or secondary to head trauma or prolonged
febrile seizures. It is suggested that hyperexcitability leads to seizure attacks
secondary to gliosis.[41] There is a need for new drugs and treatment strategies for TLE. Understanding molecular
mechanisms and the genetic basis of the disease is essential to develop ASOs for TLE.
miRNAs in TLE
It is estimated that around 60% of human proteins can be directly regulated by microRNA
(miRNAs) by binding to complementary sites on mRNAs and decreasing mRNA stability
and translation.[42] In TLE, there is altered gene expression in the hippocampal region for which the
miRNAs are partly responsible, followed by short ncRNA. Recent research has subjected
this to various changes, and now miRNAs are believed to be the major players in the
case of TLE.[43] Twenty miRNAs were altered in TLE, of which 19 were upregulated, and one was downregulated.
From this information, it is undoubtful that miRNAs play an essential role in the
pathogenesis of TLE. But the multipathway regulation and interconnections make it
a challenging target for drug development ([Fig. 2]). With the knowledge from preclinical studies, it has been found that among the
miRNAs, miR-134 is involved.
Fig. 2 Schematic representation of ncRNA mechanisms of action. ncRNAs regulate gene expression
both at transcriptional and post-transcriptional levels. miRNA, lncRNA, and circRNA
act in various ways, promoting or inhibiting the expression of specific targets. Mechanisms
of action of ncRNAs are summarized within each box. (Adapted from Fernandes JCR, Acuña
SM, Aoki JI, Floeter-Winter LM, Muxel SM. Long non-coding RNAs in the regulation of
gene expression: physiology and disease. Non-Coding RNA 2019;5(1):17).
Studies have shown elevated levels of miRNA-134 in patients with TLE. So miR-134 antagomirs/ASOs
might help in treating TLE.[44] This effect was successfully demonstrated in preclinical studies. Pretreatment of
mice with miR-134 antagomirs reduced the proportion of animals that developed status
epilepticus (SE). miR-134 antagomir was assessed in the pilocarpine-induced seizure
model, which showed increased survival. In antagomir-treated mice that did develop
SE, seizure onset was delayed, and total seizure power was also reduced.[45]
Another target of interest is miR-132, associated with hyperexcitability in the hippocampus.
Silencing this via specific ASO targets has successfully reduced hippocampal damage
in postseizure animal models. The damage to the hippocampus was evaluated with the
help of a biopsy.[46] Another molecule, miR-46a, serves as an ASO target for TLE and SE. In this study,
the C57BL/6 TLE mouse model, with the help of lithium-pilocarpine protocol, was used
to study the role of miR-46a. The Racine scale was used to evaluate the seizure severity
after intranasal delivery of miR-146a. There was an increase in latency and reduced
seizure severity. miR-146a is also found to modulate inflammatory cytokines supporting
its anti-inflammatory role via the toll-like receptor pathway in the brain.[47]
lncRNAs in TLE
As their name suggests, they have a nucleotide length of around 200. There is excellent
preclinical evidence regarding the role of long non-coding RNAs (lncRNAs) in TLE.
They are known to influence the development of seizures by different mechanisms.[48] In preclinical studies, 497 lncRNAs were expressed in hippocampal sclerosis, 294
lncRNAs were upregulated, and 203 were downregulated. Furthermore, 399 differentially
expressed mRNAs were identified. Among them, 236 were upregulated, and 163 were downregulated.
In recent years, lncRNAs also regulate proinflammatory cytokines in the hippocampus,
which is elevated in preclinical models of TLE.[49] The role of lncRNAs in synaptic plasticity is well known. lncRNAs were found to
have a significant relationship with brain-derived neurotrophic factor (BDNF).[50] Preclinical studies show that BDNF levels are high in the cerebral cortex of TLE
mouse models, which can be cleared by lncRNAs called BDNF-antisense RNA.[51] So, targeting the latter and developing compounds analogous to it can help treat
patients with TLE ([Fig. 2]).
lncRNA is ubiquitously expressed in the brain and regulates several genes involved
in dendritic and synaptic development.[50] A study by Wan et al highlighted that lncRNA nuclear enriched abundant transcript-1
(NEAT1) is known to play a significant role in the inflammatory response in TLE models.
Preclinical studies have also demonstrated markedly increased levels of NEAT1 in hippocampus
sclerosis, which suggests this could be a potential target for ASOs.[52] Another lncRNA of interest is lncMEG3 that is highly expressed in TLE. MEG3 plays
a significant role in regulating the inflammatory cytokines in our brain, and enhancers
of MEG3 could help halt the progression of TLE.[53] In the studies by Han et al, H19 plays a significant role in casp3 expression and
apoptosis. Increased lncH19 suggests that H19 plays a substantial role in inducing
microglial activation through JAK/STAT pathway.[54] NEAT2 depletion by anti-NEAT2 siRNA resulted in the loss of hippocampal neurons.
LY294002 (2-4-morphonilyl-8-phenlchromone) is an inhibitor of the PI3K/Akt signaling
pathway, which could reverse such neuronal loss and is found to be successful in rat
models with epilepsy.[55] The advent of ASOs has made lncRNAs druggable targets. In the future, this approach
could hopefully become a therapeutic approach for TLE.
circRNAs in TLE
Circular RNAs (circRNAs) are the newcomers in this domain and are known to regulate
various aspects of gene expression and post-translational aspects. CircRNAs contain
miRNA binding sites called miRNA response elements (MREs). This enables circRNAs to
sequestrate the target miRNA, a process known as the “miRNA sponge effect.” They are
called miRNA sponges, which can bind to them and inhibit their effects that play a
vital role in pharmacoresistant TLE.[56]
[57] Even though the understanding regarding circRNAs is limited, studies show the regulatory
part of circRNAs in the brain. They play vital roles in transmission events, synaptic
plasticity, apoptosis, and other aspects of neuronal activity.[58] circRNA-0067835, associated with miRNA for regulation, was decreased in preclinical
models of TLE. These levels correlated with increased seizure frequency, reinforcing
the sponge effect.[59]
In the hippocampus, the overall expression profile of circRNAs showed 43 circRNAs.
Among them, 26 were upregulated and 17 were downregulated. Various studies have found
an inverse association between MREs and target miRNA expression.[60] circEFCAB2 with miR-485-5p and circ-DROSHA with miR-1252-5p were strongly associated
with the expression of epilepsy-associated genes CLCN6 and ATP1A2, respectively.[61] Circulating circ-DROSHA might be a promising biomarker for the clinical diagnosis
of TLE.[62] In a study by Chen et al, circ-0003170 was upregulated in animal models of TLE.
Circ-0003170 was found to play a crucial role in cell viability, oxidative stress,
and apoptosis. Knockout models of circ-0003170 ameliorated oxidative stress damage
induced by circ-0003170, which indicates a potential ASO target.[63] A recent interesting experiment by Zheng et al found that increased circ-DROSHA
weakened the neural injury of the TLE cell model. Circ-DROSHA could bind to miR-106b-5p
to mediate the expression of myocyte-specific enhancement factor 2C (MEF2C); circ-DROSHA
regulated MEF2C expression via sponging miR-106b-5p. Circ-DROSHA alleviated cytotoxicity
in the TLE cell model by enhancing cell proliferation and repressing cell apoptosis.[64] All these can be exploited as potential ASO targets in the future. No drug targets
have been developed targeting these circRNAs, which might be feasible once the knowledge
regarding this strengthens.
Status Epilepticus and Antisense Molecules
SE is a medical emergency characterized by prolonged seizures. This condition results
from the failure of the mechanisms responsible for seizure termination. SE could also
be due to excessive initiation complexes in the epileptic foci leading to abnormally
prolonged seizures. According to International League Against Epilepsy, SE can be
fatal in 10 to 20% of cases who experience this condition despite treatment.[65] So, a targeted prophylactic measure is necessary for this population. Antagomir
miRNA-134 successfully prevents the emergence of seizures in preclinical studies.
Pretreatment of mice with miR-134 antagomirs reduced the proportion of animals that
developed SE. miR-134 antagomir was assessed in the pilocarpine-induced seizure model,
which showed increased survival. In antagomir-treated mice that did develop SE, seizure
onset was delayed, and total seizure power was also reduced.[45]
[46] ASOs could help develop suitable preventive measures for seizure control in patients
predisposed to developing SE.
Current Status of ASOs in Epilepsies
These promising molecules could change the fate of patients with epilepsy, especially
those with resistant epilepsy. Many rare epilepsy syndromes are resistant to treatment
by conventional ASMs, and ASOs are being tried to combat this approach. Some other
conditions ASOs are tested for include encephalopathy. Since our review is confined
to epilepsy, we have just listed them.
Ataluren the ASO in a Clinical Trial
Ataluren, previously known as PTC124, is a bioactive molecule that modulates the translation
machinery.[66] The compound allows for the readthrough of PTCs during mRNA translation, facilitating
the production of full-length functional proteins.[67]
A phase 3 study (NCT02758626) on 16 patients in a double-blind crossover manner failed
to reduce the seizure frequency[68] effectively. The results of Ataluren seem disappointing after successful preclinical
studies, but this is unavoidable in the case of drug discovery and therapeutics. Further
studies and future perspectives might rectify this and aid in developing successful
ASOs. So, they could be practically used in the treatment of epilepsy.
Conclusion
ASOs have proven themselves to be potential agents for the treatment of epilepsy.
The various ASOs discussed above have certain limitations and challenges. Further
research is required to unleash their application in therapeutics. Epilepsy therapeutics
have different therapeutic options but still pose a challenge in many cases. ASOs
can be a better replacement for such patients in the future. Despite various pharmacological
and pharmaceutical challenges, we believe ASOs will find a place in future epilepsy
therapeutics. However, time and determined research will have a significant role in
bringing them to reality.
