Gene editing in dermatology: Harnessing CRISPR for the treatment of cutaneous disease

Genodermatoses

Most genodermatoses are monogenic in nature and therefore serve as an attractive disease model for gene therapy¹⁴. Because there are no widely available effective treatments for these disorders, current therapies are focused primarily on symptom management. Early success in the use of gene therapy for the treatment of the monogenic inherited epidermolysis bullosa (EB) disorders provided particular promise for the development of curative therapies for genodermatoses. In 2006, a patient suffering from nonlethal junctional EB (JEB) underwent successful long-term skin transplantation with epidermal sheets made from gene-corrected autologous keratinocytes^15,16. These keratinocytes were corrected using a retroviral vector encoding the beta 3 chain of laminin-332, compensating for the mutated version of the gene in this patient. Such successes laid the groundwork for further research in gene replacement and gene editing strategies for cutaneous disorders. Unlike classically defined gene therapy, which involves the random insertion of one or more exogenous genes into cells to replace the function of a missing or mutated gene, gene editing involves the direct, site-specific manipulation of the genome using targeted nucleases such as ZFNs, TALENs, and CRISPR-Cas enzymes (Figure 1)¹⁷. The therapeutic potential of direct genome manipulation is immense. Researchers have leveraged CRISPR-Cas constructs to develop diverse treatment strategies for genodermatoses including the targeted addition of genes to specific genomic sites, the correction of disease-causing point mutations, and the removal of disease-causing genes or genomic sequences. Such strategies allow for corrective gene editing and the targeted ‘knockout’ of mutant alleles in dominant negative disorders. It is also hoped that targeted gene editing strategies will reduce the risks associated with random insertion of exogenous transgenes.

EB. Of the genodermatoses, the inherited EB disorders have been the most extensively studied as potential candidates for gene editing therapy³. The EB disorders are a diverse group of inherited blistering diseases that affect the skin and, in some subtypes, mucous membranes and other organs¹⁸. They are caused by mutations in over 20 different genes that code for different proteins expressed at the cutaneous basement membrane zone¹⁹.

Two EB subtypes—EB Simplex (EBS) and dominant dystrophic EB (DDEB)—are caused by dominant negative mutations that cannot be corrected with traditional additive gene therapies. Thus, these disorders are particularly well-suited for treatment with CRISPR-Cas genome editing, which allows for the direct modification of the dominant, disease-causing allele. EBS is caused by dominant negative missense mutations in either the keratin 14 (KRT14) or keratin 5 (KRT5) genes that code for intermediate filaments (IFs) expressed in the basal layer of the epidermis^20,21. Kocher et al.²² used CRISPR-Cas9-induced HDR to correct the disease-causing mutated KRT14 allele in EBS patient keratinocytes in vitro. Gene-corrected clones showed normal phenotype without characteristic mutant cytoplasmic aggregates in cell culture. DDEB is caused by dominant negative mutations in COL7A1, which codes for collagen 7 (C7)^23,24. Shinkuma et al.²⁵ successfully used CRISPR-Cas9 to induce site-directed mutagenesis of the mutated COL7A1 allele in iPSCs derived from DDEB patient keratinocytes. Edited cells expressed a truncated version of C7 that was incapable of forming deleterious trimers with WT C7 and would hypothetically allow for normal anchoring fibril formation at the DEJ²⁵.

Unlike EBS and DDEB, JEB and RDEB are inherited in an autosomal recessive manner. Therefore, CRISPR-Cas therapeutics for these disorders must achieve gene correction to allow for production of new, functional proteins. This can be accomplished through precise correction of the disruptive mutation (i.e., via CRISPR-Cas9 induced HDR with a gene-corrected donor template) or through methods that produce a functional protein without fully correcting the disease-causing mutation. For example, in specific cases, targeted deletions can be used to either remove a premature stop codon or causative mutation directly or to disrupt splicing signals, allowing for therapeutic exon skipping. JEB is caused by autosomal recessive mutations in genes encoding subunits of the heterotrimeric laminin-5 (laminin-332) protein (e.g., LAMA3, LAMB3, and LAMC2)^26,27, and RDEB is caused by autosomal recessive mutations in COL7A1^28,29. For both of these conditions, CRISPR-Cas9-mediated HDR has been used to successfully correct disease-causing mutations ex vivo in patient-derived primary keratinocytes^30–33 and iPSCs³⁴. When grafted onto immunodeficient mice, gene-corrected keratinocytes displayed restored WT functionality and adhesion at the DEJ^31–33. However, the efficiency of HDR during ex vivo gene modification remains low, particularly without antibiotic section protocols or other methods of enriching for modified cells. Exon skipping approaches have been explored for the treatment of RDEB. Exon 80 contains a common disease-causing point mutation in the COL7A1 gene³⁵. By inducing targeted Cas9-mediated DSBs on either side of the mutation-bearing exon 80 in COL7A1, it is possible to mediate complete excision of the disease-causing mutation. Bonafant et al. delivered paired Cas9/sgRNA RNPs to patient keratinocytes ex vivo through electroporation. They were able to achieve efficient rates of deletion of exon 80, generating cells with restored C7 expression that, when grafted to immunodeficient mice, showed long-term dermal-epidermal adhesion³⁶.

Notably, Cas9-mediated therapies for RDEB have recently expanded to include in vivo approaches, with Wu and colleagues³⁷ restoring C7 function in an RDEB mouse model using the exon skipping approach. Wu et al. designed two sgRNAs targeting the 5’ and 3’ side of exon 80 and delivered them as an sgRNA/Cas9 ribonucleoprotein complex (RNP) via intradermal injection into mouse tail skin. They then directly electroporated the mouse tail to facilitate penetration of RNPs into epidermal stem cells. By causing DSBs on either side of the mutation-bearing exon 80 in COL7A1, they were able to achieve complete excision of the disease-causing mutation in epidermal stem cells. Electroporated mice displayed restored C7 function, and their dermal-epidermal adhesion area improved from 30% to 60% after one treatment³⁷. The success of this approach demonstrated the potential for CRISPR-Cas9-induced gene correction of epidermal stem cells in vivo without the cost and technical challenge of ex vivo cell modification. Still, however, several limitations of this method are apparent. Only 2% of epidermal cells were capable of being targeted with this novel in vivo delivery method, and no long-term follow-up to assess sustainability of the treatment could be performed. Moreover, potential off-target effects of the Cas9/sgRNA RNPs at sites other than exon 80 were not analyzed. Whether transdermal delivery of Cas9/sgRNA RNPs via electroporation would be safe and effective in humans has yet to be explored. However, it is likely that considerable pain and collateral damage may be associated with transdermal electroporation techniques³⁸, particularly for EB patients. Electroporation protocols for EB patients would require high voltages to penetrate the nuclei of epidermal stem cells and would need to be administered over large surface areas of skin.

Epidermolytic palmoplantar keratoderma (EPPK). EPPK is an autosomal dominant keratin disease of the hereditary palmoplantar keratoderma (PPK) group characterized by an abnormal thickening of skin on the palms and soles³⁹. The disease is caused by dominant-negative missense mutations in the KRT9 gene, leading to the formation of a mutant keratin 9 (K9) protein that interferes with the function of the wild-type K9⁴⁰. EPPK has relatively localized disease manifestations that are well suited for targeted delivery of CRISPR therapeutics.

Luan and colleagues showed that through local delivery of a lentiviral (LV) vector carrying an sgRNA and Cas9 directed to the mutant KRT9 allele, they could disrupt the formation of the mutant K9 protein in vivo in an EPPK mouse model and induce phenotypic correction of the disease⁴⁰. Mice who received nine subcutaneous injections of the Cas9/sgRNA-containing LV particles into their right forepaw over the course of 24 days displayed restored epidermal proliferation of the forepaw and a reduction in disease-associated K9 expression. Off-target effects of the genome editing system were reportedly minimal, but analyses were limited to 10 predicted off-target sites in the mouse genome⁴⁰. In addition, long-term effects of the treatment on the mice were not recorded, and there was no analysis of potential immune response to the LV Cas9 vector—an occurrence which has been reported for other LV vector systems⁴¹. Still, this study demonstrates the potential for robust in vivo effects of CRISPR-Cas9-mediated gene editing, particularly for diseases that have localized manifestations.

Cutaneous viruses

CRISPR-Cas systems, which evolved in bacteria to fight invading bacteriophages, have also been repurposed to serve a similar function in virally infected human cells. In human cells, CRISPR-Cas enzymes have the potential to target latent viruses that are able to escape eradication by immune surveillance and standard antiviral therapies. As such, there has been extensive research into the ability of CRISPR-Cas systems to target specific viral genomic sequences, enabling targeted disruption and even complete excision of components of the viral genome⁴². In addition, researchers have leveraged the high sensitivity of certain Cas enzymes for the purposes of viral pathogen detection in human tissue samples⁴³. Cas12 and Cas13—two Cas enzymes that demonstrate indiscriminate trans-cleavage of ssDNA when activated by their guide-complementary target nucleotide sequence—enable ultrasensitive nucleic acid detection in viral biosensing systems such as DETECTR⁴⁴, SHERLOCK/SHERLOCKv2⁴⁵ and HOLMES/HOLMESv2^46,47

Though most CRISPR-mediated antiviral research to date has focused on systemic viruses that lack primary cutaneous involvement, the unique accessibility of the skin suggests that CRISPR therapeutics and diagnostics might be more effectively applied to cutaneous viruses. Current research suggests CRISPR-Cas technology may be effective in detecting and modifying human papillomavirus (HPV), herpes simplex virus (HSV), and Kaposi sarcoma-associated herpesvirus (KSHV).

HPV. HPV is a dsDNA virus that infects the basal cells of stratified epithelium. Here, the virus has the potential to integrate its viral DNA into the host genome. The HPV E6 and E7 proteins of certain high-risk strains, including 16, 18, 31, 33, cause malignant transformation of epithelial cells by inactivating p53 and Rb, respectively^48,49, and can lead to anogenital squamous cancers and other neoplasms. E7 expression from lower risk strains of HPV, particularly strains 6 and 11, is associated with the uncontrolled proliferation of epithelial cells that cause genital warts^50,51.

Several researchers have effectively used CRISPR-Cas9 to disrupt of E6 and E7 genes in cervical cancer cells, both in vitro and in vivo in animal models^{47–49,52–54}. One of the first antiviral CRISPR-Cas9 clinical trial in humans will involve in vivo targeting of E6/E7 in HPV-infected neoplastic cervical cells⁵⁵.

Though fewer studies have been performed with the aim of using CRISPR-Cas to cure the dermatological manifestations of HPV, researchers have begun to develop CRISPR-Cas constructs targeting the virus in HPV-associated anal cancer and genital warts. Hsu et al.⁵⁶ were able to successfully decrease tumor burden in a mouse model of HPV-16-associated anal cancer. Using an adeno-associated virus (AAV) vector, they delivered Cas9 nuclease in conjunction with two gRNAs: one gRNA specific for HPV-16 E6 and the other specific for HPV-16 E7. When delivered via three intratumoral injections over one week to mice with patient-derived xenografts of HPV-16 anal cancer cells, the CRISPR-Cas9 and dual-gRNA caused a twofold decrease in tumor volume, providing proof of concept for a novel in vivo gene editing strategy for HPV-16-associated anal cancer. For HPV-associated genital warts, CRISPR-Cas9 has been used to successfully target the E7 gene in vitro, promoting apoptosis of HPV-6 and -11-infected keratinocyte cell lines⁵⁷. Notably, however, the Cas9 and gRNAs were transfected transiently into the keratinocyte cell line and achieved only incomplete E7 deactivation. In addition, efficacy of the genome editing construct was not confirmed in vivo.

CRISPR-Cas systems hold potential for not only HPV therapeutics, but also HPV diagnostics. DETECTR is a nucleic acid biosensing system that uses the enzyme Cas12a and a ssDNA fluorescent reporting scheme to identify specific sequences in amplified dsDNA from human samples⁴⁴. Chen et al.⁴⁴ showed that by using their DETECTR platform, they could detect human papillomavirus (HPV) DNA in patient anal swab samples with attomolar sensitivity, even reliably distinguishing between different genotypes of the virus. Moreover, their assay was performed in just one hour and involved only isothermal amplification of DNA, suggesting that DETECTR could serve as a rapid, low-cost, point-of-care detection assay for HPV with similar sensitivity and specificity to conventional diagnostic PCR.

Herpesviruses. The herpesviruses are large dsDNA viruses that establish lifelong infection in human hosts^58,59. HSV-1 and HSV-2 classically infect the oral and genital mucosal epithelium, respectively, leading to the local production of ulcers. After undergoing partial clearance by the host immune system, HSV establishes latent infection in the form of episomal DNA in sensory ganglia. KSHV, or human herpesvirus 8, infects human endothelial cells and is the causative agent of Kaposi’s sarcoma⁶⁰. Like HSV and other herpesviruses, KSHV remains latent in most infected cells. Latent herpes infections avoid immunological surveillance by limiting viral gene transcription⁶¹ and are extremely difficult to treat.

As conventional therapies targeting viral replication are ineffective against latent virus, CRISPR-mediated targeting of viral DNA has emerged as an alternative approach for HSV, KSHV, and other herpesviruses. Roehm and colleagues⁶² were able to completely abrogate HSV-1 replication in human fibroblast cells in vitro by disrupting two essential viral genes using CRISPR-Cas9. Other researchers have achieved similar success against HSV-1, inhibiting viral replication in vitro in oligodendroglioma cells and epithelial cells using Cas9/gRNA editing complexes without evident off-target effects^62,63. For KSHV, researchers have demonstrated an ability to reduce the burden of KSHV in latently infected epithelial and endothelial cell lines by AAV-CRISPR-Cas9-mediated disruption of the KSHV latency-associated nuclear antigen (LANA)⁶⁴.

Still, challenges remain in the use of CRISPR-Cas for herpesviruses. First, despite demonstrated ability to abrogate active viral replication of HSV-1, it remains to be seen whether CRISPR-Cas can be effectively used to eradicate latent HSV-1 in neurons⁶⁵. Oh et al. were able to disrupt quiescent HSV-1 genomes, but at a much lower rate compared to HSV-1 virions in the lytic cycle⁶⁶. Because other targeted nucleases, such as MNs, have been shown to successfully target latent HSV-1 infection⁶⁷, it is suspected that current failure to eliminate latent HSV-1 using CRISPR-Cas9 may be due to epigenetic modifications of the latent HSV-1 genome that block Cas9 activity^65,68. Moreover, further demonstration of the effectiveness of anti-HSV and anti-KSHV CRISPR-Cas systems in vivo will also be important. For HSV-1 and -2, in vivo delivery may be simpler than for other systems because latent HSV-1 and 2 have very specific tropism for the trigeminal and sacral ganglia, respectively, and possible delivery strategies could be designed to specifically home to these ganglia.

Cutaneous bacterial infections

In addition to viral infections, CRISPR-Cas technology also shows promise against resistant bacterial infections^69,70. Bacterial infections that are resistant to common antibiotics represent a growing public health concern^71,72. Yet, despite this risk, antibiotics continue to be overprescribed⁷³ while the development of novel antimicrobials for new pathogens lags behind⁷⁴. Recently, CRISPR-Cas antimicrobials have emerged as a novel treatment strategy against bacterial infections^69,70. Specifically, CRISPR-Cas9 has been leveraged to selectively remove antimicrobial resistance genes from populations of bacteria, re-sensitizing populations of bacteria to common antimicrobials^75,76. Though systemic delivery of CRISPR-based antimicrobials remains a challenge, the accessibility of the skin enables the delivery of CRISPR constructs via convenient topical formulations⁶⁹ and places cutaneous bacterial infections at the forefront of CRISPR antimicrobial research.

Staphylococcus aureus. Staphylococcus aureus, a common cutaneous bacterial pathogen known for its antimicrobial resistance, is responsible for 76% of all skin and soft tissue infections⁷⁷ and is associated with high morbidity and mortality⁷⁸. Antimicrobial resistance to S. aureus continues to emerge as the pathogen gains plasmids and other mobile genetic elements that confer antibiotic resistance and virulence genes⁷⁹. Moreover, outbreaks remain common as the pathogen maintains a high prevalence in the population, asymptomatically colonizing the nostrils of 20–30% of healthy adults⁸⁰.

Bikard and colleagues⁷⁵ developed a novel approach to target virulent strains of S. Aureus using CRISPR-Cas9. They developed gRNAs targeted to specific S. Aureus antimicrobial resistance genes, including the methicillin resistance gene, mecA. When delivered with Cas9 via a phage capsid to mixed populations of bacteria in vitro, these gRNA/Cas9 constructs were able to eradicate resistant S. Aureus strains and completely remove specific plasmids carrying antimicrobial resistance genes. Moreover, when delivered topically to a mouse model of S. Aureus skin colonization in vivo, these gRNA/Cas9 constructs were able to significantly decrease colonization by resistant bacteria. These results demonstrated promise for the topical application of CRISPR antimicrobials in vivo and laid the foundation for future multiplexed CRISPR antimicrobials designed to simultaneously target either several bacterial species or multiple gene sequences within the same bacterium. These results also demonstrated the potential for CRISPR antimicrobials to modulate the cutaneous microbiome. The role of the cutaneous microbiome in dermatologic disease continues to unfold as researchers analyze metagenomic sequencing data from varied skin samples⁸¹. For example, reduced microbiome diversity and increased S. aureus skin colonization has been implicated in the pathogenesis of atopic dermatitis⁸². The potential role for CRISPR antimicrobials in atopic dermatitis patients has yet to be explored, but antimicrobials targeting pathogenic S. aureus strains may complement strategies designed to augment commensal bacterial populations in the cutaneous microbiome⁸³.

Melanoma

Some of the first CRISPR-Cas clinical trials in humans have involved the use of CRISPR-Cas technology in immunotherapy for cancers, including melanoma and non-small-cell lung cancer (NSCLC)⁴. Much of this research has centered on the use of gene editing to inactivate key immune checkpoint inhibitors such as programmed cell death protein-1 (PD-1) and cytotoxic T-lymphocyte-associated protein-4 (CTLA-4)—two proteins that normally inhibit the anti-tumor cytotoxic effect of endogenous and exogenous T cells⁸⁴.

Melanoma is well-known to have exceptional immunogenic potential, owing largely to the high mutational burden that drives the formation of immune-stimulating neoantigens⁸⁵. Thus, under optimal conditions, melanoma cells are particularly susceptible to destruction by the human immune system⁸⁶. Yet, clinically, the tumor microenvironment in melanoma is highly immunosuppressive⁸⁷, and advanced disease has exceptionally poor treatment response⁸⁸. Consequently, melanoma serves as an ideal target for immunotherapies that are designed to relieve tumor immunosuppression.

The first human trial designed to test the use of CRISPR-Cas for melanoma builds on the demonstrated success of previous immunotherapies, including PD-1 inhibitors⁸⁹ and T cells transduced with the NY-ESO-1 T-cell receptor (TCR)⁹⁰. Specifically, researchers aim to amplify the therapeutic effects of these existing approaches by using CRISPR-Cas9 to knock out PD-1 gene loci in autologous NY-ESO-1 TCR-transduced T cells. Autologous T cells are first taken from a patient and transduced with a LV vector that expresses the NY-ESO-1 TCR, priming them to recognize a highly immunogenic NY-ESO-1 antigen expressed on melanoma cells⁹¹. They are then electroporated with RNA-guided CRISPR-Cas9 nucleases designed to disrupt the expression of both PD-1 and the endogenous TCR subunits, TCRα and TCRβ⁴. Disrupting PD-1 prevents immune suppressive signaling, and blocking the endogenous TCR subunits inhibits aberrant immune responses that may result from TCR-mediated targeting of unknown antigens. With re-introduction of these melanoma-targeted, immune-avid T cells, researchers hope to obtain a more robust tumor-specific immune response—a response that has been difficult to achieve with previous T-cell therapies for solid tumors⁹². Moreover, beyond having an enhanced anti-tumor effect, such an approach is also expected to minimize unwanted treatment side effects. Using this treatment model, PD-1 blockade will be limited to the specific TCR-transduced T-cells that are designed to home to and attack NY-ESO-1-expressing melanoma cells. This stands in contrast to traditional anti-PD1 receptor antibodies which, when administered systemically, have the potential to affect T-cells more broadly and cause widespread immunogenic side effects⁹³.