Exon skipping therapy for dystrophic epidermolysis bullosa
Bremer, Jeroen
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Bremer, J. (2018). Exon skipping therapy for dystrophic epidermolysis bullosa. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
10
Discussion
Short recap of the aims and conclusions of this thesis
With this thesis, I made an effort to encapsulate the feasibility and impact of antisense oli-gonucleotide-mediated exon skipping as a potential therapeutic approach for dystrophic epidermolysis bullosa (DEB). The thesis is divided into two parts; in the first part,
chap-ters 2-5, we looked into the feasibility of exon skipping as therapy. In the second part of
the thesis, chapters 6-9, we investigated the potential impact of exon skipping as thera-py. In the first part, we showed pre-clinical proof of concept of exon skipping for DEB, by systemic delivery of antisense oligonucleotides (AONs) that induced exon skipping and restoration of protein synthesis in recessive DEB (RDEB) patient skin grafts (Chapter 2). We showed that the essential functional characteristics of the resulting type VII collagen protein are not impaired by specific exon deletion (Chapter 3). The importance of spe-cies-specific type VII collagen analysis in two widely used human skin graft mouse mod-els was emphasized in Chapter 4. Further, we reviewed the literature to provide insight into RNA-based therapeutic strategies for genodermatoses (Chapter 5). The second part of this thesis anticipated on the therapeutic effect of exon skipping for DEB, by examin-ing the genotype-phenotype correlation of naturally occurrexamin-ing exon skippexamin-ing in COL7A1; we concluded that the most clinical benefit is expected for patients suffering from RDEB caused by null mutations (Chapter 6). Further, a case of junctional epidermolysis bullo-sa, caused by a null-variant in COL17A1, exemplified the potential of exon skipping. The phenotype of the patient was ameliorated due to natural exon skipping, revealing the po-tential of exon skipping as therapy for other EB genes as well (Chapter 7). In contrast, by studying a case of epidermolysis bullosa simplex caused by natural exon skipping in the
KRT5 gene, we showed that for some EB genes exon skipping is most likely not a suitable
approach (Chapter 8). Finally, we showed that physiological alternative splicing can alter prognosis, by presenting a novel physiological splice isoform of PLEC (Chapter 9). Chapter 9 underlines the need to fully understand both constitutive and alternative splicing of genes to adequately anticipate on-target and off-target effects of AONs.
10
What does the future of exon skipping for EB look like?
This thesis provides evidence in favour of exon skipping for EB by providing the in
vit-ro and in vivo pvit-roof-of-concept for exon skipping for DEB. The future of exon skipping
for EB is, of course, difficult to predict. There are several important questions that need to be addressed before the results of this thesis can be translated into the clinic in the future. The most important topics are discussed in this chapter in order from “bench to bedside”. The questions are diverse, and vary from questions related to the optimisation of the exon skipping strategy itself to questions related to the clinical setting, such as which patients are expected to benefit the most from this therapy. By addressing these topics, I will sketch the near future of research into exon skipping therapy for EB.
1. Which cells, keratinocytes and/or fibroblasts, are reached with systemically ad-ministered AONs?
In Chapter 2, we subcutaneously injected AONs in athymic nude mice bearing patient skin grafts. We observed re-expression of type VII collagen in these skin grafts, however, be-cause both patient fibroblasts and keratinocytes were used to generate the skin grafts, the source of this newly expressed type VII collagen, i.e. patient fibroblasts or keratinocytes, is unknown. The question as to which cells are reached is important from a future clinical perspective, as there is a difference in expression level of type VII collagen between fibro-blasts and keratinocytes. Although fibrofibro-blasts are believed to be an easier target than keratinocytes for gene therapy of DEB,1 keratinocytes express type VII collagen in higher
quantities in vitro.2 Therefore, being able to reach keratinocytes compared to only
reach-ing fibroblasts might have a positive impact on the therapeutic effect of exon skippreach-ing for DEB, with respect to the expected quantities of re-expression of type VII collagen. Moreo-ver, reaching basal keratinocytes, might make exon skipping an option for other EB types caused by mutations in proteins expressed solely by keratinocytes, like type XVII collagen.
In order to partly address this question, we investigated the mdx-mouse-model for Duchenne muscular dystrophy. Skin biopsies (whole skin, epidermis, dermis) were tak-en from mdx-mice treated with an AON against the dystrophin gtak-ene (Apptak-endix 2). Exon skipping for the Dmd gene was subsequently observed in samples of the whole skin, as well as the dermis only, of these treated mdx-mice. Based on these data, we concluded that the AONs are at least capable of inducing exon skipping in the dermis. No conclusions could, however, be made regarding exon skipping in the epidermis, as the Dmd gene is only expressed in the dermis, and not in the epidermis. In the skin graft experiments de-scribed in Chapter 2, it is therefore likely that at least the dermal fibroblasts were targeted by the COL7A1 AONs. However, no conclusion can yet be drawn as to whether the AONs can also penetrate the basement membrane zone and target epidermal basal keratino-cytes.
two different RDEB patients carrying mutations in different COL7A1 exons can be envi-sioned. In this experiment, fibroblasts from patient 1 carrying mutation X and keratino-cytes from patient 2 carrying mutation Y would be used to generate skin grafts and vice versa. This way, two different kinds of skin grafts can be generated. Subsequently, the skin grafts are treated using an AON that skips the exon harbouring mutation Y, but not mutation X. This way, type VII collagen re-expression in the grafts can be only attributed to the cell layer generated with cells of patient 2, carrying mutation Y. The same experiment could be repeated with AONs targeting mutation X, but not Y. These experiments would provide insights into whether the basal keratinocytes contribute to the re-expression of type VII collagen seen in our skin grafts. Obviously, it would be essential that the skin cells of both RDEB patients do not produce any level type VII collagen protein, as the readout would primarily be type VII collagen expression. RNA analysis is less favourable in such a study, as the AONs will induce skipping of the target exon regardless of the presence of the mutation and would therefore give inconclusive results, as only skipping of the exon harbouring the null-mutation will result in protein re-expression.
2. Would targeting of fibroblasts only, have a therapeutic benefit?
In Chapter 4, we showed that endogenous murine fibroblasts growing in human skin grafts can produce quantities of type VII collagen that are sufficient to form detectable anchoring fibrils. Further, pre-clinical fibroblast cell therapy studies in mice showed that intradermal or intravenous injection of type VII collagen expressing fibroblasts, led to homing of the fibroblasts to wounds, type VII collagen expression, and improved skin integrity.3-5 The effect of fibroblast therapy was observed in three different mouse
mod-els: a wild type wound healing mouse model,3 a type VII collagen hypomorphic mouse
model,4 and a type VII collagen knockout mouse model,5 In contrast to these pre-clinical
data, in a clinical study performed by Wong and colleagues in which patients were treated with fibroblast injections,6 homing of the fibroblasts to wounds was not observed. Two
weeks post-injection, injected fibroblasts were no longer detectable in the patient. Nev-ertheless, at the two-week measurement an increase in type VII collagen and improved wound healing were observed. This effect was predominantly observed in patients that already showed residual expression of type VII collagen at baseline and was suggested by additional in vivo studies to be an effect secondary to the fibroblast injections.7 In
an-other randomized vehicle-controlled phase-II clinical trial, fibroblast injections also led to improved wound healing compared to the vehicle control,8 exposing that fibroblasts do
impact wound healing and type VII collagen expression in skin. Ideally, both keratinocytes and fibroblasts are reached by therapeutic strategies to obtain the highest clinical benefit. However, pre-clinical and clinical data suggest that reaching fibroblasts alone might suf-fice to obtain a clinically relevant benefit.
10
3. How can delivery of AONs to specific cells or tissues be optimised?
In Chapter 2, we showed expression of type VII collagen in RDEB-gen sev skin grafts after AON treatment. Compared to healthy control skin, the expression was much lower, de-spite the 8 weeks 5 times per week AON injections at a high dose. Even though a small increase in type VII collagen production could be clinically relevant, the expression of type VII collagen in the skin grafts needs further optimization. One way to increase the expres-sion of type VII collagen and expected treatment effect may be to enhance the delivery of AONs to fibroblasts and/or keratinocytes specifically. Increased uptake of AONs by specific target cells is currently a main topic of oligotherapeutics research,9, 10 and continues to
be the prominent theme at the annual oligotherapeutics society meetings (http://www. cvent.com/events/13th-annual-meeting-of-the-oligonucleotide-therapeutics-society/ agenda-1ffd7395a30e4bfc99060bede83ea639.aspx).
One of the most promising methods of enhancing selective uptake of AONs is conjugation of the AON to a molecule that binds receptors on the target cell, so-called AON-conjugates. For example, conjugation with N-acetyl galactosamine (GalNAc) has been clearly demonstrated to improve uptake in liver, as GalNAc binds to the asialogly-coprotein receptor highly expressed specifically by hepatocytes. After binding to the receptor, the GalNAc oligo conjugate is internalized much more efficiently via the en-docytosis pathway compared to the naked (unconjugated) AON.11 Another successful
example of an AON-conjugation was recently presented at the 13th Annual Meeting of
the Oligonucleotide Therapeutics Society. (Andersson et al. Advances in targeted deliv-ery of nucleotides beyond the liver, http://www.oligotherapeutics.org/wp-content/up-loads/2017/09/2017Speakers_Session_SEPT17_2017.pdf; p.7) Andersson and colleagues used peptide-AON conjugates to specifically target pancreatic beta-cells, which they were not able to reach with the naked AON. The AONs were conjugated to a peptide that binds to the GLP1-receptor, which is highly expressed in pancreatic beta-cells. After a single sub-cutaneous administration of the AON-peptide conjugate, a dose-dependent effect of the AON conjugate was observed, while the naked AON was not active.
Because conjugation to small peptides and specific receptor ligands seems suc-cessful for several cell and tissue types, such an approach might also work for skin basal keratinocytes. To specifically target the basal keratinocytes, a peptide or molecule would be needed that binds to a receptor or membrane protein that is uniquely or highly ex-pressed by basal keratinocytes. Numerous options as target are theoretically available among which the keratinocyte growth factor receptor.12
Since an infinite amount of possibilities of peptide-oligo modifications is con-ceivable, thorough in silico and in vitro screening of peptides and oligos is essential. An example of such in vitro screening method involves phage-display in combination with high throughput sequencing in which a rigorous in vitro selection can be made.13 In short,
mixture of 109 different M13 bacteriophages. The binding peptides of these
bacteriophag-es are modified and linked to random sequencbacteriophag-es of 7 amino acid long peptidbacteriophag-es. Repeated selection for binding to the target cells of the phage display library followed by phage-DNA isolation and sequencing, yields the phage-DNA sequences of 7 amino acid long peptides that are able to bind to the target cells. Subsequently, a selection of peptides is produced and fluorescently labelled for downstream in vitro binding assays. Obviously, after in vitro selection of peptides that bind to the target cell or protein, the peptides should be thor-oughly analysed in vivo, as increased nonspecific uptake in other cells than the target cells could be induced.
To limit nonspecific cellular uptake, antibody-AON conjugation might be consid-ered with antibodies that are directed against membrane proteins specifically expressed by basal keratinocytes. Antibody-drug conjugates for specific delivery are used in several therapy approaches, like breast cancer, where monoclonal antibodies are used to deliver cytotoxic loads specifically to cancer cells.14 As this approach would increase
target-spe-cific delivery, it might be an interesting approach for targeting the skin as well. Impor-tant for cellular uptake via membrane proteins is that the targeted membrane protein, like most receptors, has a short turnover life span, as this has been shown to enable ef-ficient internalisation of the antibody drug conjugate.15 To target the epidermis with a
keratinocyte-specific antibody-AON conjugate, one could think of desmosomal proteins that are highly expressed by keratinocytes.16 An important matter to take into account
when developing such AON-conjugates is the anticipated increase in production cost of AON-conjugates versus the increased efficacy of the AON-conjugates in comparison to naked AONs.
4. Which skin areas can we reach with AONs, and can we reach mucosae?
Systemically administered AONs end up, and exert their exon skipping effect, in the skin. A drawback of the study in Chapter 2 was that the grafting model could not be analysed systemically, as the AON could only exert its effect in the skin graft. To take the next step towards the clinic, two important questions should be answered: 1) Which skin areas can we reach with systemically administered AONs, and how efficiently? And 2), do these AONs reach and induce exon skipping in mucosal tissue? These questions are obviously important, as the disease systemically affects the patients.17, 18 To answer these questions
in a pre-clinical setting, an AON targeting the murine Col7a1 gene could be used to induce exon skipping in Col7a1 wild-type mice. After treatment, each tissue of interest could be analysed separately for exon skipping. Subsequently, an overview could be made in which skin areas and tissues the AONs induce exon skipping.
Over the last decades, several chemical modifications have been developed that have a great impact on the pharmacokinetics and pharmacodynamics of AONs.19 The
10
potential to bind to serum proteins, and their binding affinity to the complementary tar-get.19 It could very well be that the 2’-O-Methyl phosphorothioate (2OMePS) nucleic acid
modification that we used in our study is not the most efficient chemistry to target skin. Therefore, the two most promising alternative chemistries used in splice modulation, i.e. the phosphorodiamidate (PMO) and 2’-O-Methoxyethoxy (MOE) chemistries, should be investigated for their efficacy to target skin, as these chemistries are widely tested and successfully used in clinical settings.20, 21 See also Figure 6 in the Introduction (Chapter 1)
for the structural differences of these AON chemistries. The biggest difference in pharma-cokinetics is expected between the PMO and the other two (2OMePS and MOE). The most influencing factor is the charge of the molecules, i.e. the 2OMePS and MOE are negative-ly charged, whereas the PMO is neutralnegative-ly charged. Negativenegative-ly charged backbones have shown to have positive impact on the ability to bind to serum proteins, and thereby a longer serum half-life and a better bioavailability of the AON.19, 22-24
5. How is the COL7A1 gene physiologically spliced and how is this influenced by AONs?
Another key question is how the COL7A1 gene is physiologically spliced. For exon skip-ping, especially the order in which introns are spliced out is important, because it could affect the skippability of exons. In order to understand this, another study performed in the DMD field uncovered a few intriguing splicing principles.25 Gazolli and colleagues,25
studied the order in which the DMD gene introns were spliced out from the pre-mRNA. Interestingly, they found that the introns were not sequentially spliced, sequential mean-ing in order from the first to the last transcribed intron. In fact, 40% of the introns were not sequentially spliced. It was hypothesized that intron length correlated to the splicing speed, however, a rather unexpected finding was that the order of splicing of the DMD introns was not dependent on intron length. Some introns spliced faster than others, but not in a random pattern. The fast and slowly spliced introns could be arranged in blocks of one, two, or more than two fast spliced introns followed by a slowly spliced intron. This revealed a multi-step splicing process of the DMD introns and exon blocks of two, three or more exons, which led to the fundamental understanding as to why some exons could not be skipped alone. For example, AONs targeting exon 8 of the DMD gene always led to the skipping of exons 8 and 9 together,26 and indeed, exon 8 and 9 turned out to form
such an exon block. The reasons why introns can be spliced out slowly or fast are currently unknown, however it might be affected by the rate in which RNA polymerase II transcribes the gene, in combination with the abundance of splicing factors, as this has been shown to cause, so-called, co-transcriptional splicing effects.27
Because mutations are scattered throughout the COL7A1 gene, AONs need to be developed against multiple exons in order to treat multiple patients. It might be, as well as for the DMD gene, that due to this principle of non-sequential splicing, some exons of
the COL7A1 gene cannot be skipped alone, which might hinder the development of AONs for those exons. Therefore, it is important to investigate the order of splicing for the
CO-L7A1 gene. To investigate these physiological splicing processes, whole transcriptome
se-quencing (RNA-seq) might be a viable method. As described by Gazzoli and colleagues,25
the number of RNA-seq reads (normalized for intron length) mapped to an intron corre-lates to the relative speed of splicing. Highly simplified: the lower the coverage of intron x, the faster intron x is presumably spliced and vice versa.
In addition to the order of splicing, the RNA-seq method should be used in the early pre-clinical design stages of AONs to look at their effect on splicing in general. Alter-native splicing of the COL7A1 transcript is important to consider when designing antisense strategies. For example, cryptic splice sites might become active, or more active, when adjacent splice signals are blocked by AONs. In pre-clinical settings, alternative splicing could be tested by performing RT-PCR surrounding the exon that is targeted by the AON. However, the tertiary structure of RNA could have an influence on the effects of the AON. This is caused by the basic principle of steric hindrance of AONs, for example, AON x tar-geting exon 4 could actually in the tertiary structure be in close proximity of exon 12, and thus AON x could have peripheral effects on exon 12 splicing whilst being designed for exon 4.28 This will obviously be missed by classical RT-PCR surrounding the targeted exon
alone and, therefore, employing a high throughput RNA screening method, like RNA-seq, is crucial in preclinical studies to identify splice variation throughout the gene. Moreover, the data generated by RNA-seq could be used to screen for other off-target effects of the AONs on the entire transcriptome,29 as in silico analysis of potential off-target binding of
the AON may not be reflecting its in vivo effects.
6. How does the high variant density of the COL7A1 gene affect splicing?
In general, in the COL7A1 gene, multiple splice signals are spread throughout the exons and introns. Some exons and introns comprise more splicing signals, whereas others con-tain only a few splicing signals, like exon 87 (Chapter 6). Reduced splice site recognition caused by a mutation, without the presence of other splice signals has been shown to result more frequently in complete skipping of the exon.30 The splice sites in the COL7A1
gene, in general, seem weakly defined when predicted by splice site prediction analysis.31
The Exome Aggregation Consortium (ExAC, http://exac.broadinstitute.org) data-base contains DNA sequence data from 60,706 humans, as per August 2017.32 A search for
“COL7A1” in this database, yielded 3857 unique variants recorded for the COL7A1 gene of which 2053 were located in the coding region of COL7A1. On average, the human exome contains one variant every eight bases.32 In contrast, the COL7A1 exons contain a higher
average variant density of one variant every four bases. In addition, more than 700 unique DEB causing mutations are described,33, 34, which corresponds roughly to 2´10-2 disease
den-10
sity in the COL7A1 gene. An explanation for this observation could be that the replication of the gene is more difficult and prone to error than average, which might be caused by the repetitive nature and high GC-content of the gene, due to the glycine repeat. The high number of variations in the COL7A1 gene, in combination with weakly defined intron-ex-on borders, suggest that it is almost certain that more of these variants will affect splicing than we observed in Chapter 6.
I personally believe that there are many more variants that lead to alternative splicing but fully functional type VII collagen. This expectation is based on the high num-ber and density of variants found in the COL7A1 gene and the fact that there are so many variants found that lead to alternative splicing and a recessive phenotype (Chapter 6). Therefore, the probability that there are more undiscovered variants that lead to alterna-tive splicing but no pathogenesis or recessive phenotype, is indisputable.
Identification of variants that lead to alternative splicing is not only essential for the development of exon skipping therapy, but also for diagnostic purposes. For example, the effects of deep intronic variants remain difficult to predict. Effects of the variants in close vicinity of splice sites could be analysed by RT-PCR. However, today, high through-put genome wide sequencing is standard in genome diagnostics,35, 36 and adding RNA-seq
to routine diagnostics should be carefully considered for the near future.
7. Can pre-clinical data obtained in mice be translated into the clinic?
Mouse models remain an important tool in drug discovery and validation. Therefore, a mouse model in which the murine Col7a1 gene is replaced by the human COL7A1 gene would be a perfect starting point to create a model for DEB. Chapter 3 and other stud-ies in which the function of recombinant human type VII collagen is analysed in type VII collagen deficient mice,37, 38 suggest that human type VII collagen is functional in murine
skin. Thereby, from a functional point of view, replacing the murine by the human gene would provide a useful animal DEB model. However, there may be differences in splicing in general between human and mouse, and differences in splicing between the COL7A1 and Col7a1 genes in particular. More insight into such splicing differences between the
COL7A1 and the Col7a1 genes is crucial for assessing the translatability of pre-clinical
data obtained in mice. Therefore, the splicing process in such a humanized mouse model should be fully characterized by, for example, RNA-seq analysis.
We identified one example of differences in splicing between the human COL7A1 gene and the murine Col7a1 gene. We designed an AON that is complementary to both the human and murine exon 73. We observed efficient exon 73 skipping in human kerat-inocytes but not in murine cells (Appendix 2). Interestingly, in murine cells, exon 73 was not skipped alone but always together with exons 74 and 75, indicating that the sequence of splicing of exons 73, 74, and 75 likely differs between mice and men. Most likely, intron 72 is a more slowly spliced intron in mice than in human, resulting in introns 73 and 74
be-ing spliced out before intron 72, leadbe-ing to exons 74 and 75 skippbe-ing together with exon 73. However, we do not have experimental proof yet that confirms this hypothesis. Wheth-er this diffWheth-erence in splicing is sequence based, or due to diffWheth-erences in splicing between human and mouse in general, is unknown and should be thoroughly investigated for the proposed generation of a humanized COL7A1 mouse model. It is known that alternative splicing is not a highly conserved mechanism between human and mouse. Three inde-pendent studies found that alternative splicing of highly conserved exons was species specific and not conserved between human and mouse.39-41 Therefore, it is conceivable
that also for the COL7A1 and Col7a1 genes the observed difference in splicing of exons 73-75 is based on the general differences in splicing between human and mouse. However, it might also be partly based on differences in the sequence of exons-introns 73 to exon 75. The coding sequence of exon 73 to exon 75 shows 85% homology between human and mouse (Clustal Omega global alignment42). Intron 73 and intron 74 show only 67% and
53% homology, respectively, and, as expected, the introns are thus less conserved than the exons. Although the total coding sequence of the COL7A1 gene is conserved, intron length varies between mouse and human. The question whether the difference in splicing observed is species or sequence based, is essential to answer prior to further preclinical development of exon skipping therapy for DEB.
8. Which patients are expected to benefit the most?
In Chapter 6, we showed that RDEB patients carrying bi-allelic null mutations that do pro-duce low amounts of type VII collagen as a result of natural exon skipping of an exon har-bouring (one of) the null-mutations have milder than expected phenotypes. In contrast, dominant DEB (DDEB) phenotypes caused by natural exon skipping, are indistinguishable from other dominant DEB cases caused by glycine-substitutions. This predicts that exon skipping as therapy might not be a suitable approach for DDEB, not even if it would be possible to selectively skip the mutant allele alone. Additionally, because an additional al-lele is introduced by exon skipping therapy, i.e. the skipped transcript is added to the mix-ture of wild-type and mutant transcripts, it might even deteriorate the phenotype. In the case of exon skipping for EB, the patient group that has the highest chance of benefiting the most from exon skipping therapy thus seem to be carriers of bi-allelic null mutations in the COL7A1 gene. In my opinion, the RDEB patients carrying bi-allelic null mutations that express type VII collagen as result of natural exon skipping, described in Chapter 6, are a great example of what exon skipping aims to do. In Chapter 6, we see that lifelong expression of low amounts of exon skipped type VII collagen can result in dramatically less severe phenotypes.43-45
Similar to natural exon skipping, induced exon skipping is likely to lead to accu-mulation of type VII collagen to a certain steady-state level over time and lead to clinical beneficial amounts of type VII collagen. Accumulation of type VII collagen to a
steady-10
state level is foreseen due to two mechanisms: first, AON treatment has been shown to result in an increasing tissue concentration of AONs to a certain maximum saturation lev-el, over time.19 Therefore, the effect of the AONs in the skin is expected to increase until
that maximum tissue concentration is achieved. And secondly, the long half-life of type VII collagen in skin of about a month,46 will help build up type VII collagen levels in skin to a
steady-state, over time. At what point this steady-state quantity of type VII collagen in skin is achieved, and at what treatment regimen the AON concentration in skin tissue plateaus, should be the first questions to be answered in the clinical phase of exon skipping trials.
9. Can exon skipping be used for other EB subtypes?
In Chapter 7 and Chapter 8, we learned that natural exon skipping in the COL17A1 gene could lead to amelioration of the phenotype, whereas exon skipping in the KRT14 gene seems to have rather detrimental effects on protein functionality and, consequently, the phenotype. The COL17A1 gene structure is highly similar to the COL7A1 gene, especially in respect to their collagenous regions. Therefore, it is not surprising that exon skipping in the COL17A1 gene also has positive effects on disease outcome, as described in Chapter 7 and by Pasmooij and colleagues.47 Feasibility of AON-mediated exon skipping for other
EB genes is difficult to predict. However, from the literature we know that for the LAMB3 gene natural exon skipping can lead to milder than expected phenotypes.44 In these cases
of junctional EB caused by null mutations in exon 30 of the COL17A1 gene and exon 17 of the LAMB3 gene, respectively, exon skipping of these exons resulted in re-expression of functional type XVII collagen or laminin-332. Together with the data presented in Chap-ter 7, where natural in-frame skipping of COL17A1 exon 49 led to improvement of the phenotype, these cases demonstrate that at least exons 30 and 49 of COL17A1 and exon 17 of LAMB3 are potential exon skipping candidates. However, all EB genes, except for
FLG2, EXPH5, and CD151, carry several exons that are theoretically skippable (Table 1), and
therefore I predict that the exon skipping approach will be much wider applicable than
COL7A1 and DEB alone.
Theoretically, an exon is skippable if the removal of the exon leads to an in-frame transcript, however thorough investigation on the functional consequences should be performed, like we did for the COL7A1 gene and DEB in Chapter 3. Importantly, for all oth-er types of EB, delivoth-ery to koth-eratinocytes in the basal and suprabasal layoth-er of the epidoth-ermis is essential for exon skipping to be considered, as genes causing other types of EB are expressed by keratinocytes only. Therefore, the question whether or not keratinocytes can be targeted with AONs (point 2) is additionally crucial to solve the question as to whether exon skipping can be considered for other EB types.
Table 1. Number of in-frame exons of all 21 EB genes.
Gene No. of coding exons Skippable exons* Percentage
TGM5 13 5 38 % DSP 24 12 50 % JUP 13 3 23 % PKP1 14 4 29 % KLHL24 6 3 25 % FLG2 2 0 0 % KRT5 9 4 44 % KRT14 8 3 48 % EXPH5 6 0 0 % PLEC 39 17 44 % CD151 7 0 0 % DST 24 12 50 % LAMA3 76 27 36 % LAMB3 22 5 23 % LAMC2 23 9 39 % COL17A1 55 53 96 % ITGB4 40 21 53 % ITGA6 25 11 44 % ITGA3 26 12 46 % COL7A1 118 107 91 % FERMT1 14 6 43 %
*Skippable means that the number of nucleotides comprised by the exon is dividable by three, i.e. the reading frame is left intact if the exon is skipped. It does not reflect experimental evidence that it is feasible to skip the exon with AONs, nor that the resulting protein will be functional. The latter should be experimentally analysed for each exon individually. The first and last exons of genes are never considered candidates, even if they are dividable by three.
10. Which lessons can we learn from other splice modulation drugs?
Over the recent years, major steps have been taken in the field of antisense oligonucle-otide-mediated splice modulation.48 The largest milestones for antisense
oligonucle-otide-mediated exon skipping and other splice switching methods were the American Food and Drug Administration (FDA) approval of Exondys 51 (active substance eteplirsen) for Duchenne muscular dystrophy (DMD) in 2016, and both FDA and European Medicines Agency (EMA) approval of Spinraza (nusinersen) for spinal muscular atrophy (SMA) in 2016 and 2017. See also Table 4 in Chapter 1 for the complete overview of approved splice modulating antisense drugs, to date.
Two medicinal products were developed at the same time for the skipping of exon 51 of the DMD gene by two different companies. Exondys 51 (eteplirsen), devel-oped by Biomarin Pharmaceuticals, is a 30mer phosporodiamidate AON (PMO).49 After
its FDA approval, it is currently under Committee for Medicinal Products for Human use (CHMP) evaluation in Europe
(http://www.ema.europa.eu/docs/en_GB/document_li-10
brary/Minutes/2017/06/WC500229137.pdf). A decision is expected by the end of 2017. Kyndrisa (drisapersen), on the other hand, is a 20mer 2’-O-methyl phosphorothioate AON (2OMePS).23 Kyndrisa was developed by Biomarin International Limited, which applied for
FDA marketing authorization in the United Stated in late 2015. Despite promising data in preclinical and early clinical studies, the company had to withdraw its application in a later stage in the spring of 2016, as Biomarin International Limited was not able to ad-dress the CHMP concerns regarding the results of the clinical studies within the expected timeframe. The main concerns of the CHMP (http://www.ema.europa.eu/docs/en_GB/ document_library/Medicine_QA/2016/06/WC500209339.pdf), were that treatment with Kyndrisa did not show any significant differences in clinical effect between placebo and Kyndrisa in the main study. Additionally, the safety profile of the drug was not consid-ered satisfactory, especially due to persisting reactions at the injection site and the risk of thrombocytopenia. Therefore, at the time of withdrawal, the CMPH was of the opinion that the benefit of Kyndrisa did not outweigh the risks of treatment.
The clinical studies with Spinraza (nusinersen) reveal the great potential of AON-mediated splice modulation. Spinraza targets intronic silence sequences (ISS) of in-tron 7 of the SMN2 gene, and thereby induces the inclusion of exon 7 of the SMN2 gene to mimic the SMN1 gene. In essence, exon inclusion could be seen as the opposite of exon skipping, however, it makes use of the same basic principle of steric hindrance by AONs of RNA binding proteins that are involved in splicing. The 2MOE AONs are administered via intrathecal injections show long lasting effects, as the AONs have an intrathecal half-life of 4-6.50-54 From the clinical studies with Spinraza, we can learn that if the affected tissue, or
affected cells can be targeted with AONs, we can expect high efficiency and great impact on clinical outcome.
From the DMD clinical studies, important lessons have been learned about poten-tial adverse effects and route of administration. In general, treatment was well tolerated, although, transient thrombocytopenia was observed sporadically and unpredictably.55-58
Crucially though, injection site reactions upon subcutaneous injections of AONs were very commonly observed in all clinical trials and in a dose dependent manner. In high dose treatment groups, injection site reactions were observed in 100% of the patients.59
Injection site reactions comprised symptoms as erythema, induration, itching, discomfort and pain, but occasionally even more severe adverse events like ulceration or necrosis were observed.59 These injection site reactions are most likely caused by recognition of
nucleic acids by Toll-Like receptors, mainly TLR-7 and TLR-9 that are predominantly ex-pressed in endosomes.59, 60 Obviously, the last thing one would want is to elicit such
reac-tions by administration of AONs either by subcutaneous injection or via a spray or cream, in severely affected EB skin. Therefore, it is recommended to investigate the intravenous route of administration for exon skipping in EB, as intravenously injected AONs have so far not been reported to elicit such responses.60
Important lessons can be learned from the phase II clinical trial with eteplirsen and the phase III clinical trial with drisapersen.61-63 The six minutes walking test, which
was the primary clinical measurement for disease severity in the clinical trials for DMD, was not suitable for the more severely affected patients at the start of the study, as in advanced disease state patients always require the use of a wheelchair. Therefore, it is essential to have a well described natural history of disease, in order to be able to design the proper clinical measurements. The main lessons were; (1) validate a primary endpoint measurement based on natural history of the disease, (2) validate a biomarker for the dis-ease progression or severity, and (3) subcutaneously injected AONs provoke injection site reactions, whereas intravenously injected AONs do not. These lessons drawn from these clinical trials have provided invaluable insight for future development of AON-mediated therapies for DMD and other diseases, like DEB.64
With regard to (lesson 1) the natural history of disease, it can be stated that, al-though the clinical course of DEB is well known, the natural history of the disease has not yet been systematically documented as such.17, 18, 65 Regarding (lesson 2) a disease-specific
biomarker, there is a strong correlation between the amount of type VII collagen and the severity of the disease.66 Therefore, in clinical trials that are focussed on re-expression of
type VII collagen, type VII collagen could be an important biomarker. The level of type VII collagen in skin that can prevent skin blistering is estimated to be around 30 % in mice,4
however, lower levels of type VII collagen might help reduce the risk of developing ag-gressive squamous cell carcinomas or aid wound healing, as type VII collagen is known to be an important factor in both.3, 18, 67, 68 A small increase in the level of type VII collagen
by exon skipping could therefore already be clinically relevant, not only because of the strong correlation between expression of the structural protein and disease severity, but also because of the complex role of type VII collagen in immunoregulatory pathways and wound healing.69, 70 .
10
Concluding remarks
In this thesis, first pre-clinical proof of concept has been shown for exon skipping as sys-temic therapy for DEB. These positive results warrant further research into the develop-ment of exon skipping, especially into the optimization of exon skipping for DEB. Impor-tant questions have been raised and should be answered before the first patient can be treated. Questions like the fibroblast/keratinocyte conundrum, or the effect of chemical modifications of the AON to its pharmacokinetics in skin, are universal for antisense strat-egies for skin targets, and are therefore of high value for the entire research field. In addi-tion, the detection of exon skipping on the RNA and protein level, should be extensively validated to provide solid read-outs for future research and diagnostics.
References
1. M Goto, D Sawamura, K Ito, M Abe, W Nishie, K Sakai, A Shibaki, M Akiyama, H Shimizu. Fibroblasts show more potential as target cells than keratinocytes in COL7A1 gene therapy of dystrophic epidermolysis bullosa. J Invest Dermatol 2006, 126, 766.
2. YQ Chen, A Mauviel, J Ryynanen, S Sollberg, J Uitto. Type VII collagen gene expression by human skin fibro-blasts and keratinocytes in culture: influence of donor age and cytokine responses J Invest Dermatol 1994, 102, 205.
3. DT Woodley, J Remington, Y Huang, Y Hou, W Li, DR Keene, M Chen. Intravenously injected human fibroblasts home to skin wounds, deliver type VII collagen, and promote wound healing. Mol Ther 2007, 15, 628.
4. A Fritsch, S Loeckermann, JS Kern, A Braun, MR Bosl, TA Bley, H Schumann, D von Elverfeldt, D Paul, M Erlacher, D Berens von Rautenfeld, I Hausser, R Fassler, L Bruckner-Tuderman. A hypomorphic mouse model of dystrophic epidermolysis bullosa reveals mechanisms of disease and response to fibroblast therapy J Clin Invest 2008, 118, 1669.
5. J Remington, X Wang, Y Hou, H Zhou, J Burnett, T Muirhead, J Uitto, DR Keene, DT Woodley, M Chen. Injection of recombinant human type VII collagen corrects the disease phenotype in a murine model of dystrophic epi-dermolysis bullosa. Mol Ther 2009, 17, 26.
6. T Wong, L Gammon, L Liu, JE Mellerio, PJ Dopping-Hepenstal, J Pacy, G Elia, R Jeffery, IM Leigh, H Navsaria, JA McGrath. Potential of fibroblast cell therapy for recessive dystrophic epidermolysis bullosa J Invest Dermatol 2008, 128, 2179.
7. JS Kern, S Loeckermann, A Fritsch, I Hausser, W Roth, TM Magin, C Mack, ML Muller, O Paul, P Ruther, L Bruck-ner-Tuderman. Mechanisms of fibroblast cell therapy for dystrophic epidermolysis bullosa: high stability of col-lagen VII favors long-term skin integrity Mol Ther 2009, 17, 1605.
8. SS Venugopal, W Yan, JW Frew, HI Cohn, LM Rhodes, K Tran, W Melbourne, JA Nelson, M Sturm, J Fogarty, MP Marinkovich, S Igawa, A Ishida-Yamamoto, DF Murrell. A phase II randomized vehicle-controlled trial of intra-dermal allogeneic fibroblasts for recessive dystrophic epidermolysis bullosa. J Am Acad Dermatol 2013, 69, 898. 9. C Godfrey, LR Desviat, B Smedsrod, F Pietri-Rouxel, MA Denti, P Disterer, S Lorain, G Nogales-Gadea, V Sardone, R Anwar, S El Andaloussi, T Lehto, B Khoo, C Brolin, WM van Roon-Mom, A Goyenvalle, A Aartsma-Rus, V Arecha-vala-Gomeza. Delivery is key: lessons learnt from developing splice-switching antisense therapies. EMBO Mol Med 2017, 9, 545.
10. JC Kaczmarek, PS Kowalski, DG Anderson. Advances in the delivery of RNA therapeutics: from concept to clinical reality. Genome Med 2017, 9, 60.
11. Y Huang. Preclinical and Clinical Advances of GalNAc-Decorated Nucleic Acid Therapeutics. Mol Ther Nucleic Acids 2017, 6, 116.
12. V de Giorgi, S Sestini, D Massi, I Ghersetich, T Lotti. Keratinocyte growth factor receptors. Dermatol Clin 2007, 25, 477.
13. PA 't Hoen, SM Jirka, BR Ten Broeke, EA Schultes, B Aguilera, KH Pang, H Heemskerk, A Aartsma-Rus, GJ van Ommen, JT den Dunnen. Phage display screening without repetitious selection rounds. Anal Biochem 2012, 421, 622.
14. PA Trail, GM Dubowchik, TB Lowinger. Antibody drug conjugates for treatment of breast cancer: Novel targets and diverse approaches in ADC design. Pharmacol Ther 2017, .
15. BE de Goeij, D Satijn, CM Freitag, R Wubbolts, WK Bleeker, A Khasanov, T Zhu, G Chen, D Miao, PH van Berkel, PW Parren. High turnover of tissue factor enables efficient intracellular delivery of antibody-drug conjugates. Mol Cancer Ther 2015, 14, 1130.
16. A Ishida-Yamamoto, M Kishibe, M Honma. Desmosomes and corneodesmosomes and their relevance to ge-netic skin diseases. G Ital Dermatol Venereol 2017, 152, 148.
10
17. JD Fine, JE Mellerio. Extracutaneous manifestations and complications of inherited epidermolysis bullosa:part I. Epithelial associated tissues J Am Acad Dermatol 2009, 61, 367.
18. JD Fine, JE Mellerio. Extracutaneous manifestations and complications of inherited epidermolysis bullosa: part II. Other organs J Am Acad Dermatol 2009, 61, 387.
19. RS Geary, D Norris, R Yu, CF Bennett. Pharmacokinetics, biodistribution and cell uptake of antisense oligonu-cleotides. Adv Drug Deliv Rev 2015, 87, 46.
20. HA Heemskerk, CL de Winter, SJ de Kimpe, P van Kuik-Romeijn, N Heuvelmans, GJ Platenburg, GJ van Ommen, JC van Deutekom, A Aartsma-Rus. In vivo comparison of 2'-O-methyl phosphorothioate and morpholino anti-sense oligonucleotides for Duchenne muscular dystrophy exon skipping J Gene Med 2009, 11, 257.
21. F Rigo, SJ Chun, DA Norris, G Hung, S Lee, J Matson, RA Fey, H Gaus, Y Hua, JS Grundy, AR Krainer, SP Henry, CF Bennett. Pharmacology of a central nervous system delivered 2'-O-methoxyethyl-modified survival of motor neuron splicing oligonucleotide in mice and nonhuman primates. J Pharmacol Exp Ther 2014, 350, 46.
22. V Arora, GR Devi, PL Iversen. Neutrally charged phosphorodiamidate morpholino antisense oligomers: up-take, efficacy and pharmacokinetics. Curr Pharm Biotechnol 2004, 5, 431.
23. NM Goemans, M Tulinius, M van den Hauwe, AK Kroksmark, G Buyse, RJ Wilson, JC van Deutekom, SJ de Kimpe, A Lourbakos, G Campion. Long-Term Efficacy, Safety, and Pharmacokinetics of Drisapersen in Duchenne Muscular Dystrophy: Results from an Open-Label Extension Study. PLoS One 2016, 11, e0161955.
24. AE Felber, N Bayo-Puxan, GF Deleavey, B Castagner, MJ Damha, JC Leroux. The interactions of amphiphilic antisense oligonucleotides with serum proteins and their effects on in vitro silencing activity. Biomaterials 2012, 33, 5955.
25. I Gazzoli, I Pulyakhina, NE Verwey, Y Ariyurek, JF Laros, PA 't Hoen, A Aartsma-Rus. Non-sequential and mul-ti-step splicing of the dystrophin transcript. RNA Biol 2016, 13, 290.
26. A Aartsma-Rus, CL De Winter, AA Janson, WE Kaman, GJ Van Ommen, JT Den Dunnen, JC Van Deutekom. Functional analysis of 114 exon-internal AONs for targeted DMD exon skipping: indication for steric hindrance of SR protein binding sites. Oligonucleotides 2005, 15, 284.
27. NJ Proudfoot. Dawdling polymerases allow introns time to splice. Nat Struct Biol 2003, 10, 876.
28. TA Vickers, JR Wyatt, SM Freier. Effects of RNA secondary structure on cellular antisense activity. Nucleic Acids Res 2000, 28, 1340.
29. HK Yalamanchili, YW Wan, Z Liu. Data Analysis Pipeline for RNA-seq Experiments: From Differential Expression to Cryptic Splicing. Curr Protoc Bioinformatics 2017, 59, 11.15.1.
30. M Krawczak, NS Thomas, B Hundrieser, M Mort, M Wittig, J Hampe, DN Cooper. Single base-pair substitutions in exon-intron junctions of human genes: nature, distribution, and consequences for mRNA splicing. Hum Mutat 2007, 28, 150.
31. FO Desmet, D Hamroun, M Lalande, G Collod-Beroud, M Claustres, C Beroud. Human Splicing Finder: an on-line bioinformatics tool to predict splicing signals Nucleic Acids Res 2009, 37, e67.
32. M Lek, KJ Karczewski, EV Minikel, KE Samocha, E Banks, T Fennell, AH O'Donnell-Luria, JS Ware, AJ Hill, BB Cummings, T Tukiainen, DP Birnbaum, JA Kosmicki, LE Duncan, K Estrada, F Zhao, J Zou, E Pierce-Hoffman, J Berghout, DN Cooper, N Deflaux, M DePristo, R Do, J Flannick, M Fromer, L Gauthier, J Goldstein, N Gupta, D How-rigan, A Kiezun, MI Kurki, AL Moonshine, P Natarajan, L Orozco, GM Peloso, R Poplin, MA Rivas, V Ruano-Rubio, SA Rose, DM Ruderfer, K Shakir, PD Stenson, C Stevens, BP Thomas, G Tiao, MT Tusie-Luna, B Weisburd, HH Won, D Yu, DM Altshuler, D Ardissino, M Boehnke, J Danesh, S Donnelly, R Elosua, JC Florez, SB Gabriel, G Getz, SJ Glatt, CM Hultman, S Kathiresan, M Laakso, S McCarroll, MI McCarthy, D McGovern, R McPherson, BM Neale, A Palotie, SM Purcell, D Saleheen, JM Scharf, P Sklar, PF Sullivan, J Tuomilehto, MT Tsuang, HC Watkins, JG Wilson, MJ Daly, DG MacArthur, Exome Aggregation Consortium. Analysis of protein-coding genetic variation in 60,706 humans. Nature 2016, 536, 285.
Zam-bruno, D Castiglia, JE Mellerio, JA McGrath, AJ van Essen, RM Hofstra, MA Swertz. The international dystrophic epidermolysis bullosa patient registry: an online database of dystrophic epidermolysis bullosa patients and their COL7A1 mutations Hum Mutat 2011, 32, 1100.
34. K Wertheim-Tysarowska, A Sobczynska-Tomaszewska, C Kowalewski, M Skronski, G Swieckowski, A Kutkows-ka-Kazmierczak, K Wozniak, J Bal. The COL7A1 mutation database Hum Mutat 2012, 33, 327.
35. B Sikkema-Raddatz, LF Johansson, EN de Boer, R Almomani, LG Boven, MP van den Berg, KY van Spaen-donck-Zwarts, JP van Tintelen, RH Sijmons, JD Jongbloed, RJ Sinke. Targeted next-generation sequencing can replace Sanger sequencing in clinical diagnostics. Hum Mutat 2013, 34, 1035.
36. T Vrijenhoek, K Kraaijeveld, M Elferink, J de Ligt, E Kranendonk, G Santen, IJ Nijman, D Butler, G Claes, A Costessi, W Dorlijn, W van Eyndhoven, DJ Halley, MC van den Hout, S van Hove, LF Johansson, JD Jongbloed, R Kamps, CE Kockx, B de Koning, M Kriek, R Lekanne Dit Deprez, H Lunstroo, M Mannens, OR Mook, M Nelen, C Ploem, M Rijnen, JJ Saris, R Sinke, E Sistermans, M van Slegtenhorst, F Sleutels, N van der Stoep, M van Tienho-ven, M Vermaat, M Vogel, Q Waisfisz, J Marjan Weiss, A van den Wijngaard, W van Workum, H Ijntema, B van der Zwaag, WF van IJcken, J den Dunnen, JA Veltman, R Hennekam, E Cuppen. Next-generation sequencing-based genome diagnostics across clinical genetics centers: implementation choices and their effects. Eur J Hum Genet 2015, 23, 1142.
37. J Remington, X Wang, Y Hou, H Zhou, J Burnett, T Muirhead, J Uitto, DR Keene, DT Woodley, M Chen. Injection of recombinant human type VII collagen corrects the disease phenotype in a murine model of dystrophic epi-dermolysis bullosa. Mol Ther 2009, 17, 26.
38. Y Hou, LT Guey, T Wu, R Gao, J Cogan, X Wang, E Hong, W Vivian Ning, D Keene, N Liu, Y Huang, C Kaftan, B Tangarone, I Quinones-Garcia, J Uitto, OL Francone, DT Woodley, M Chen. Intravenously Administered Recom-binant Human Type VII Collagen Derived from Chinese Hamster Ovary Cells Reverses the Disease Phenotype in Recessive Dystrophic Epidermolysis Bullosa Mice. J Invest Dermatol 2015, 135, 3060.
39. Q Pan, MA Bakowski, Q Morris, W Zhang, BJ Frey, TR Hughes, BJ Blencowe. Alternative splicing of conserved exons is frequently species-specific in human and mouse. Trends Genet 2005, 21, 73.
40. GW Yeo, E Van Nostrand, D Holste, T Poggio, CB Burge. Identification and analysis of alternative splicing events conserved in human and mouse. Proc Natl Acad Sci U S A 2005, 102, 2850.
41. B Modrek, CJ Lee. Alternative splicing in the human, mouse and rat genomes is associated with an increased frequency of exon creation and/or loss. Nat Genet 2003, 34, 177.
42. F Sievers, A Wilm, D Dineen, TJ Gibson, K Karplus, W Li, R Lopez, H McWilliam, M Remmert, J Soding, JD Thompson, DG Higgins. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 2011, 7, 539.
43. A Schwieger-Briel, L Weibel, N Chmel, J Leppert, K Kernland-Lang, G Gruninger, C Has. A COL7A1 variant lead-ing to in-frame skipplead-ing of exon 15 attenuates disease severity in recessive dystrophic epidermolysis bullosa. Br J Dermatol 2015, 173, 1308.
44. JA McGrath, GH Ashton, JE Mellerio, JC Salas-Alanis, O Swensson, JR McMillan, RA Eady. Moderation of phe-notypic severity in dystrophic and junctional forms of epidermolysis bullosa through in-frame skipping of exons containing non-sense or frameshift mutations. J Invest Dermatol 1999, 113, 314.
45. R Gardella, D Castiglia, P Posteraro, S Bernardini, N Zoppi, M Paradisi, G Tadini, S Barlati, JA McGrath, G Zam-bruno, M Colombi. Genotype-phenotype correlation in italian patients with dystrophic epidermolysis bullosa. J Invest Dermatol 2002, 119, 1456.
46. T Kuhl, M Mezger, I Hausser, LT Guey, R Handgretinger, L Bruckner-Tuderman, A Nystrom. Collagen VII Half-Life at the Dermal-Epidermal Junction Zone: Implications for Mechanisms and Therapy of Genodermatoses J Invest Dermatol 2016, 136, 1116.
47. AM Pasmooij, S van Zalen, AM Nijenhuis, AJ Kloosterhuis, J Zuiderveen, MF Jonkman, HH Pas. A very mild form of non-Herlitz junctional epidermolysis bullosa: BP180 rescue by outsplicing of mutated exon 30 coding for the COL15 domain. Exp Dermatol 2004, 13, 125.
10
48. CA Stein, D Castanotto. FDA-Approved Oligonucleotide Therapies in 2017. Mol Ther 2017, 25, 1069.49. JR Mendell, LR Rodino-Klapac, Z Sahenk, K Roush, L Bird, LP Lowes, L Alfano, AM Gomez, S Lewis, J Kota, V Malik, K Shontz, CM Walker, KM Flanigan, M Corridore, JR Kean, HD Allen, C Shilling, KR Melia, P Sazani, JB Saoud, EM Kaye, Eteplirsen Study Group. Eteplirsen for the treatment of Duchenne muscular dystrophy. Ann Neurol 2013, 74, 637.
50. MJA Wood, K Talbot, M Bowerman. Spinal muscular atrophy: antisense oligonucleotide therapy opens the door to an integrated therapeutic landscape. Hum Mol Genet 2017, 26, R151.
51. BS Wilfond, C Morales, HA Taylor. Expanded Access for Nusinersen in Patients With Spinal Muscular Atropy: Negotiating Limited Data, Limited Alternative Treatments, and Limited Hospital Resources. Am J Bioeth 2017, 17, 66.
52. JJ Weaver, N Natarajan, DWW Shaw, SD Apkon, KSH Koo, GM Shivaram, EJ Monroe. Transforaminal intrathecal delivery of nusinersen using cone-beam computed tomography for children with spinal muscular atrophy and extensive surgical instrumentation: early results of technical success and safety. Pediatr Radiol 2017, .
53. RS Finkel, E Mercuri, BT Darras, AM Connolly, NL Kuntz, J Kirschner, CA Chiriboga, K Saito, L Servais, E Tizzano, H Topaloglu, M Tulinius, J Montes, AM Glanzman, K Bishop, ZJ Zhong, S Gheuens, CF Bennett, E Schneider, W Farwell, DC De Vivo, ENDEAR Study Group. Nusinersen versus Sham Control in Infantile-Onset Spinal Muscular Atrophy. N Engl J Med 2017, 377, 1723.
54. CA Chiriboga. Nusinersen for the treatment of spinal muscular atrophy. Expert Rev Neurother 2017, 17, 955. 55. U Flierl, TL Nero, B Lim, JF Arthur, Y Yao, SM Jung, E Gitz, AY Pollitt, MT Zaldivia, M Jandrot-Perrus, A Schafer, B Nieswandt, RK Andrews, MW Parker, EE Gardiner, K Peter. Phosphorothioate backbone modifications of nucleo-tide-based drugs are potent platelet activators. J Exp Med 2015, 212, 129.
56. T Voit, H Topaloglu, V Straub, F Muntoni, N Deconinck, G Campion, SJ De Kimpe, M Eagle, M Guglieri, S Hood, L Liefaard, A Lourbakos, A Morgan, J Nakielny, N Quarcoo, V Ricotti, K Rolfe, L Servais, C Wardell, R Wilson, P Wright, JE Kraus. Safety and efficacy of drisapersen for the treatment of Duchenne muscular dystrophy (DEMAND II): an exploratory, randomised, placebo-controlled phase 2 study. The Lancet Neurology 2014, 13, 987.
57. CL Tanganyika-de Winter, H Heemskerk, TG Karnaoukh, M van Putten, SJ de Kimpe, J van Deutekom, A Aarts-ma-Rus. Long-term Exon Skipping Studies With 2'-O-Methyl Phosphorothioate Antisense Oligonucleotides in Dystrophic Mouse Models Mol Ther Nucleic Acids 2012, 1, e44.
58. X Chi, P Gatti, T Papoian. Safety of antisense oligonucleotide and siRNA-based therapeutics. Drug Discov Today 2017, 22, 823.
59. L van Meer, M Moerland, J Gallagher, MB van Doorn, EP Prens, AF Cohen, R Rissmann, J Burggraaf. Injection site reactions after subcutaneous oligonucleotide therapy. Br J Clin Pharmacol 2016, 82, 340.
60. AM Krieg, SM Efler, M Wittpoth, MJ Al Adhami, HL Davis. Induction of systemic TH1-like innate immunity in normal volunteers following subcutaneous but not intravenous administration of CPG 7909, a synthetic B-class CpG oligodeoxynucleotide TLR9 agonist. J Immunother 2004, 27, 460.
61. Q Lu, S Cirak, T Partridge. What Can We Learn From Clinical Trials of Exon Skipping for DMD? Mol Ther Nucleic Acids 2014, 3, e152.
62. JR Mendell, LR Rodino-Klapac, Z Sahenk, K Roush, L Bird, LP Lowes, L Alfano, AM Gomez, S Lewis, J Kota, V Malik, K Shontz, CM Walker, KM Flanigan, M Corridore, JR Kean, HD Allen, C Shilling, KR Melia, P Sazani, JB Saoud, EM Kaye, Eteplirsen Study Group. Eteplirsen for the treatment of Duchenne muscular dystrophy. Ann Neurol 2013, 74, 637.
63. S Cirak, V Arechavala-Gomeza, M Guglieri, L Feng, S Torelli, K Anthony, S Abbs, ME Garralda, J Bourke, DJ Wells, G Dickson, MJ Wood, SD Wilton, V Straub, R Kole, SB Shrewsbury, C Sewry, JE Morgan, K Bushby, F Muntoni. Exon skipping and dystrophin restoration in patients with Duchenne muscular dystrophy after systemic phosphoro-diamidate morpholino oligomer treatment: an open-label, phase 2, dose-escalation study. Lancet 2011, 378, 595. 64. M Veltrop, A Aartsma-Rus. Antisense-mediated exon skipping: taking advantage of a trick from Mother
Na-ture to treat rare genetic diseases. Exp Cell Res 2014, 325, 50.
65. JD Fine, L Bruckner-Tuderman, RA Eady, EA Bauer, JW Bauer, C Has, A Heagerty, H Hintner, A Hovnanian, MF Jonkman, I Leigh, MP Marinkovich, AE Martinez, JA McGrath, JE Mellerio, C Moss, DF Murrell, H Shimizu, J Uitto, D Woodley, G Zambruno. Inherited epidermolysis bullosa: updated recommendations on diagnosis and classifi-cation J Am Acad Dermatol 2014, 70, 1103.
66. PC van den Akker, AJ van Essen, MM Kraak, R Meijer, M Nijenhuis, G Meijer, RM Hofstra, HH Pas, H Scheffer, MF Jonkman. Long-term follow-up of patients with recessive dystrophic epidermolysis bullosa in the Netherlands: expansion of the mutation database and unusual phenotype-genotype correlations. J Dermatol Sci 2009, 56, 9. 67. JD Fine, LB Johnson, M Weiner, KP Li, C Suchindran. Epidermolysis bullosa and the risk of life-threatening cancers: the National EB Registry experience, 1986-2006 J Am Acad Dermatol 2009, 60, 203.
68. Martins, Vyas, Chen, Purdie, Mein, South, Storey, McGrath, O'Toole. Increased invasive behaviour in cutaneous squamous cell carcinoma with loss of basement-membrane type VII collagen. J Cell Sci 2009, 122, 1788. 69. F Cianfarani, G Zambruno, D Castiglia, T Odorisio. Pathomechanisms of Altered Wound Healing in Recessive Dystrophic Epidermolysis Bullosa. Am J Pathol 2017, 187, 1445.
70. A Nystrom, D Velati, VR Mittapalli, A Fritsch, JS Kern, L Bruckner-Tuderman. Collagen VII plays a dual role in wound healing. J Clin Invest 2013, 123, 3498.
