Original Article | Published:

siRNA silencing of keratinocyte-specific GFP expression in a transgenic mouse skin model

Gene Therapy volume 16, pages 963972 (2009) | Download Citation

Subjects

Abstract

Small interfering RNAs (siRNAs) can be designed to specifically and potently target and silence a mutant allele, with little or no effect on the corresponding wild-type allele expression, presenting an opportunity for therapeutic intervention. Although several siRNAs have entered clinical trials, the development of siRNA therapeutics as a new drug class will require the development of improved delivery technologies. In this study, a reporter mouse model (transgenic click beetle luciferase/humanized monster green fluorescent protein) was developed to enable the study of siRNA delivery to skin; in this transgenic mouse, green fluorescent protein reporter gene expression is confined to the epidermis. Intradermal injection of siRNAs targeting the reporter gene resulted in marked reduction of green fluorescent protein expression in the localized treatment areas as measured by histology, real-time quantitative polymerase chain reaction and intravital imaging using a dual-axes confocal fluorescence microscope. These results indicate that this transgenic mouse skin model, coupled with in vivo imaging, will be useful for development of efficient and ‘patient-friendly’ siRNA delivery techniques and should facilitate the translation of siRNA-based therapeutics to the clinic for treatment of skin disorders.

Introduction

The discovery of RNA interference (RNAi) and the observation that short interfering RNAs (siRNAs) largely evade the immune response have opened up new therapeutic opportunities.1, 2, 3 The potency (IC50 in the picomolar range) and selectivity (single-nucleotide (nt) discrimination4, 5) of siRNAs make these inhibitors attractive drug candidates. Clinical trials of siRNAs are currently underway targeting the liver, kidney, lung, eye and skin.6, 7 Skin represents an ideal target for siRNA therapies due to its accessibility and the large number of monogenic dominant negative skin disorders.8, 9, 10 It is estimated that in the general population as many as one person in 3000 suffers from diseases resulting from pathogenic keratin mutations.11

The first siRNA clinical trial in skin (for treatment of the rare skin disorder pachyonychia congenita) has recently been completed, with encouraging results (unpublished data and Leachman et al.10). Pachyonychia congenita is an autosomal dominant disorder caused by mutations (often single-nt changes) in the KRT6, KRT16 and KRT17 genes, resulting in thickened dystrophic nails, leukokeratosis and hyperkeratosis with painful blistering.12, 13 The outcome of the clinical trial included improvement at the site of siRNA treatment, but not the paired control injection site on the opposite foot receiving vehicle alone. Importantly, no adverse events were observed. In the clinical trial, siRNA was delivered to a callused and painful region on the foot sole through intradermal injection. The intense pain associated with this delivery method necessitated regional nerve blocks and oral pain medication to make treatment bearable, underscoring the need for development of ‘patient-friendly’ delivery technologies (that is, efficient siRNA delivery with little or no pain).

Despite the accessibility of the skin, the stratum corneum represents a significant barrier, and the ability to study delivery of functionally active siRNAs is currently hampered by the lack of good animal models. To address this deficiency, we developed a transgenic reporter mouse (transgenic click beetle luciferase/humanized monster green fluorescent protein (Tg CBL/hMGFP)), in which hMGFP from Montastrea cavernosa and CBL expression is confined to the epidermis. Keratinocytes expressing hMGFP are readily visualized in vivo by a recently developed dual-axes confocal (DAC) fluorescence microscope14 or alternatively by fluorescence microscopy of biopsied skin sections. The siRNA treatment of the Tg CBL/hMGFP mice through intradermal injection resulted in localized decreased reporter expression in the epidermis. This mouse model should prove useful for testing of a variety of siRNA skin delivery technologies and facilitate development of patient-friendly methodologies that can be rapidly translated to the clinic.

Results

Generation and characterization of a transgenic reporter mouse skin model

A skin-specific dual-reporter Tg CBL/hMGFP mouse was generated by breeding a silenced dual-reporter mouse (CBL and hMGFP flanked by loxP sites) with a mouse expressing Cre recombinase driven by the keratinocyte-specific keratin 14 (K14) promoter (Figure 1a). The resultant reporter mouse expressed CBL and hMGFP from a synthetic promoter (chick β-actin (CAG))15 comprised of the β-actin promoter and the human cytomegalovirus (CMV) immediate early enhancer elements. Unexpectedly, under the control of this hybrid β-actin promoter, hMGFP expression localized predominantly as aggregates in the granular layer of the epidermis as well as uniformly throughout the stratum corneum (Figure 1b). No hMGFP expression was detected in non-bioluminescent (no CBL) mice (used as a negative control) from the same litter (Figure 1c). Standard histological examination of transgenic mouse paw skin expressing hMGFP revealed no gross difference when compared with negative control paw skin (Supplementary Figure S1), suggesting that hMGFP transgene expression at these levels is not affecting skin morphology.

Figure 1
Figure 1

Transgenic mouse model in which hMGFP reporter is expressed in the epidermis. (a) Schematic of hMGFP reporter transgene. A transgenic mouse expressing Cre recombinase driven by the keratinocyte-specific K14 promoter was crossed with a silenced reporter mouse, as described in Materials and methods, resulting in a reporter mouse in which hMGFP expression is driven by a modified CAG promoter. (b and c) The hMGFP expression in mouse footpad skin sections is confined to the epidermis. Skin vertical cross sections (10 μm) were obtained from positive (b) and negative (c) littermates and hMGFP signal detected by fluorescent microscopy. The majority of hMGFP expression is localized in the granular layer and stratum corneum. The scale bar is 20 μm. Nuclei are visualized by DAPI stain (blue). Left panels show bright field overlay.

To confirm that the expression pattern of hMGFP protein paralleled reporter mRNA expression, in situ hybridization analysis was performed. Figure 2 shows that the mRNA encoding hMGFP is also confined to the upper layers of the epidermis, coincident with the localization of the protein, whereas K14 mRNA (control) is restricted to the basal layer.

Figure 2
Figure 2

In situ hybridization of hMGFP/CBL mRNA in mouse epidermis. Tg CBL/hMGFP mouse footpad skin was fixed in paraformaldehyde and frozen in OCT, as described in Materials and methods. Sections (10 μm) were prepared and fixed again before hybridization with hMGFP and K14 Dig-labeled riboprobes. Washed sections were incubated with anti-Dig antibodies and subsequently developed with NBT and BCIP. Stratum corneum (SC), granular layer (G) and basal layer (B). The scale bar is 50 μm.

Identification of a potent siRNA that inhibits reporter gene expression

The Tg CBL/hMGFP mouse expresses a CBL/hMGFP bicistronic mRNA. Therefore, an siRNA that targets CBL will result in degradation of the entire mRNA, including the hMGFP coding region. The siRNAs targeting the CBL coding region (CBL1-10) were designed and tested by co-transfection with a plasmid expressing the reporter target (see Materials and methods) in 293T cells (Figure 3a). The most potent siRNAs identified in this screen (CBL1, 3, 5 and 10) were further analyzed in a dose-response study (Figure 3b, data not shown for CBL1, 5 and 10). Figure 3b shows the potency of CBL3 siRNA in the tissue culture assay (IC50 value was 40 pM) and was selected for use in the subsequent in vivo experiments.

Figure 3
Figure 3

Identification of potent and selective siRNAs targeting reporter expression. (a) siRNA screening. Candidate siRNAs (CBL1-10) were identified using the Dharmacon siRNA prediction algorithm and evaluated in human 293T cells by co-transfection with a phMGFP/CBL plasmid that expresses hMGFP and CBL from a bicistronic mRNA (CMV promoter). The effectiveness of each analyzed siRNA (1 nM) was evaluated 48 h after co-transfection by quantitating bioluminsescence after luciferin substrate addition using the IVIS 50 Imaging system, as described in Materials and methods. An irrelevant siRNA (targets unrelated EGFP with no sequence similarity to hMGFP) was used as a negative control. (b) Dose response of CBL3 siRNA. CBL3 siRNA, at the indicated concentrations, was cotransfected with phMGFP/CBL plasmid in 293T cells and its effectiveness determined, as described above. Red depicts highest luciferase expression levels, blue the lowest.

RNA silencing in footpad epidermis of the Tg CBL/hMGFP mouse

Mouse footpad skin was chosen for siRNA treatment due to its thickness, similarity in layer composition to human skin and relevance to the monogenic skin disorder pachyonychia congenita.10 The left footpads of six Tg CBL/hMGFP mice were treated daily with intradermally injected CBL3 siRNA for 14 days, whereas an irrelevant siRNA (K6a_513a.12, which targets a single-nt mutation in human keratin 6a (K6a)4) was injected into the right footpads (control). After the last siRNA treatment, the footpad skin was removed, sectioned and analyzed by fluorescence microscopy. Footpad skin treated with CBL3 siRNA (with or without carrier plasmid DNA) exhibited areas (corresponding to the injection site) of dramatically reduced hMGFP expression. These areas were defined as regions where at least 80% of the cells exhibited little or no hMGFP fluorescence and included reduced hMGFP aggregates in the upper layers of the epidermis as well as reduction of hMGFP signal in squames of the stratum corneum (Figures 4b and d; Supplementary Figure S2B, D, F, H). No significant reduction in hMGFP signal was observed in the footpad skin of mice treated with irrelevant siRNA (Figures 4a and c; Supplementary Figure S2A, C, E, G) or untreated controls (see Figure 1b and data not shown).

Figure 4
Figure 4

CBL3 siRNA potently and specifically inhibits pre-existing hMGFP expression in mouse epidermis. Transgenic hMGFP mouse footpads were treated daily by intradermal injection of 60 μg of CBL3 or non-specific control K6a_513a.12 siRNAs for 14 days. The mice were killed and frozen skin sections (10 μm) prepared. The hMGFP expression (or lack thereof) was visualized by fluorescence microscopy. (a and b) Cross sections from mouse footpad intradermally injected with the non-specific siRNA (K6a_513a.12) (a) or CBL3 (b) siRNA. (c and d) Same conditions as in panels a and b, but in combination with 5 μg of pUC19. Scale bar is 20 μm. Nuclei are visualized by DAPI stain (blue).

To show that siRNA-mediated reduction of hMGFP expression occurs at the mRNA level and to quantitatively measure the extent of the silencing effect, total RNA was extracted from the skin surrounding the injection sites of similarly treated mice (80 μg siRNA injected daily for 12 days), and CBL/hMGFP mRNA levels were analyzed by real-time quantitative polymerase chain reaction (RT-qPCR) using K14 mRNA as the endogenous control (Figure 5). Treatment of footpad skin with CBL3 siRNA resulted in a 33% decrease of reporter expression (mean of three mice) when compared with irrelevant siRNA treatment of the opposing footpad.

Figure 5
Figure 5

RT-qPCR analysis of Tg CBL/hMGFP mice treated with CBL3 siRNA. Total RNA, isolated from footpad skin of three mice (M1, M2 and M3) treated with CBL3 siRNA (right footpad) or an irrelevant siRNA (K6a_513a.12, left footpad), was reverse transcribed and hMGFP levels were quantified by qPCR. The hMGFP levels were normalized to K14 levels (endogenous control). Each bar corresponds to the mean of three experiments (three replicates per experiment). Bars indicate standard error.

Recently, siRNA-mediated immunostimulation has been reported and brought into question some results ascribed earlier to RNAi.16, 17, 18 To investigate the ability of CBL3 siRNA to stimulate immune cells, freshly prepared human peripheral blood mononuclear cells (PBMCs) were treated with the siRNAs used in this study, either in the presence or absence of the transfection reagent DOTAP, and induction of interleukin 6 (IL-6) or tumor necrosis factor (TNF) alpha measured.19 As expected, no immunostimulation was observed in the absence of DOTAP (data not shown), consistent with an uptake requirement of the siRNAs (for example, the main PBMC siRNA toll-like receptors appear to be toll-like receptor 7/8, which are present on the endosomes20, 21, 22). In the presence of DOTAP, the siRNAs used in this study showed peak induction of IL-6 and TNF mRNA levels at the 8- and 16-h time points (Supplementary Figures S3 and S4), with CBL3 siRNA treatment resulting in the least amount of activation. These results suggest that CBL3 siRNA has minimal potential to stimulate mRNA expression for IL-6 and TNF in human PBMC under conditions in which the poly I:C positive controls show dramatic induction. Coupled with the observation that no differences were observed between untreated and control siRNA mice, these findings suggest that the CBL3 siRNA-mediated inhibition results are not the result of non-specific immunostimulation.

In vivo three-dimensional imaging of hMGFP silencing

Intravital imaging is an emerging set of tools that enable pathologic conditions to be monitored over time with cellular resolution within living tissue. Such measures can provide information that is not accessible by other means and enable new perspectives into biological mechanisms. For these studies, a white Tg CBL/hMGFP mouse was treated for 18 days (80 μg siRNA intradermally injected daily), and footpad skin of the live mouse was imaged over time using DAC fluorescence imaging.14 (A white mouse was used to avoid the attenuation of optical signals due to absorption of light by melanin pigment.) After treatment with CBL3 siRNA, hMGFP signals in both the stratum corneum and granular layer were markedly decreased, whereas the irrelevant siRNA control showed no changes (Figure 6; Supplementary material Video Files 1 (irrelevant control) and 2 (CBL3 siRNA)). Localized areas (10) lacking hMGFP were detected in the CBL3 siRNA-treated footpad (Figure 4) as well as areas of strong hMGFP signal (data not shown), consistent with the results obtained in histology sections.

Figure 6
Figure 6

Intravital imaging of siRNA-mediated hMGFP inhibition in mouse footpad skin using DAC fluorescence microscopy. Mouse footpads were treated daily for 18 days with an irrelevant siRNA (K6a_513a.12) (left) or CBL3 siRNA (right) as described in Materials and methods. The images show a side view of a 3D volume rendering of 80 × 250 μm en face images through a depth of 100 μm. Images were taken at day 19 using the same gain setting for both footpads. The hMGFP signal strongly accumulates in the stratum corneum (SC) of the control footpad, and cytoplasmic granules can be detected in the granular layer (G). Strong inhibition of hMGFP signal is observed in the CBL3 siRNA-treated footpad for both SC and G. The 2D image stack videos of these 3D images are provided as Supplementary material.

Discussion

RNAi has generated keen interest as a technology that potentially can be exploited for treatment of genetic diseases, if delivery obstacles can be overcome. Short interfering RNAs have been designed to silence defective genes, including mutant genes that differ by only a single nt from wild type.4, 5 Furthermore, allele-specific discrimination based on single-nt polymorphisms has been achieved.5, 23, 24 The RNAi-based therapeutics are currently in clinical trials for a number of non-skin diseases including age-related macular degeneration as well as respiratory syncytial, hepatitis B and human immunodeficiency viruses.3, 6, 7 A wide spectrum of heritable skin diseases have been found, with more than 80 distinct genes involved in more than 100 disease phenotypes,25 a high proportion of which is keratin diseases without effective treatment options. Many keratin mutations act in a dominant fashion, in which the mutant allele encodes an aberrant protein causing the disease phenotype.26 Silencing the mutant protein through RNAi is one of the most promising therapeutic options for these dominant monogenic skin disorders.

Recently, a Phase 1b clinical trial using an siRNA (TD101) that targets a mutant version of K6a (N171K) was completed for the rare dominant genodermatosis pachyonychia congenita (unpublished results10). This siRNA was shown earlier to robustly and specifically inhibit mutant gene expression (with single-nt discrimination between the wild type and mutant mRNAs) both in tissue culture cells and in mice (data not shown and Leachman et al.4, 10). In the clinical trial, TD101 siRNA was administered twice weekly by intradermal injection6, 10 and resulted in resolution (‘healing’) of pachyonychia congenita symptoms at the injection treatment site (unpublished data). The intense injection site pain (for both siRNA and vehicle control), even after administration of oral pain medication and regional nerve blocks, precludes intradermal injection as a patient- and physician-acceptable delivery option. Therefore, for therapeutic skin siRNAs to become a viable option in a clinical setting, less painful delivery methodologies must be considered and/or developed.

In the present work, the Tg CBL/hMGFP mouse model, which expresses a bioluminescent/fluorescent reporter gene confined to the epidermis, was prepared as a model to efficiently study alterations in pre-existing gene expression (CBL and/or hMGFP) after siRNA treatment. The ability to monitor gene expression in real time using non-invasive intravital imaging methods in the same mouse reduces the mouse-to-mouse variability commonly observed in these types of experiments and greatly reduces the number of animals needed. Furthermore, once treatment is terminated, the time needed for recovery of gene expression can be observed. Using intravital imaging, we observed modest inhibition of hMGFP signal after 5 days of treatment increasing to significant inhibition at day 19 (1 day after the completion of the entire treatment course, Figure 6). Similar reduction in hMGFP expression was observed at day 21, but there appeared to be an increase in signal at day 23, and restoration to near baseline levels of reporter signal at day 40 (22 days after the last injection) (data not shown). The delay in the detection of an inhibitory response (5 days) may be a function of the time required for siRNA uptake and processing, the half life of the CBL/hMGFP protein (estimated to be 1–2 days) and/or the time (or injections) required for accumulation of a sufficient number of inhibited keratinocytes to allow detection by the DAC intravital imaging system. Importantly, the demonstration that intradermally injected CBL3 siRNA can reproducibly inhibit hMGFP expression by multiple independent assays including intravital imaging, RT-qPCR and ex vivo microscopy lays the groundwork for future experiments in which other delivery methodologies can be evaluated and improved. Two such patient-friendly methodologies, hollow dissolvable micro-needle arrays, which can be loaded with siRNA cargo and cream formulations that enable siRNA penetration through the stratum corneum, are under development.

In recent years, immunostimulatory activities of siRNAs, particularly unmodified siRNAs, have been reported and caution advised in interpreting siRNA results.16, 17, 18 Immunostimulation resulting from siRNA treatment continues to be an active area of investigation, with some published animal reports suggesting no immunostimulatory activity following treatment with unmodified siRNAs,27, 28 whereas other studies clearly warrant caution.16, 17, 29 In our hands, minimal immune stimulation was observed as assayed by TNF and IL-6 levels in CBL3 siRNA-treated PBMC cultures, suggesting that the diminished reporter gene activity (GFP) was not solely attributable to an immunostimulatory effect. Furthermore, no differences were observed in untreated vs control (K6a_513a.12) siRNA-treated footpad skin.

The Tg CBL/hMGFP mouse model design (see Figure 1a) predicted that transgene expression should occur throughout the epidermis as expression is driven by the strong, ubiquitous and constitutively expressed modified β-actin (CAG) promoter.30 To create this mouse, the silenced reporter transgene was activated through breeding with a transgenic mouse expressing Cre recombinase under the control of the K14 promoter (K14 is normally expressed in the basal layer of the epidermis31, 32, 33). Unexpectedly, the resulting hMGFP transgene expression (both protein and mRNA, see Figures 1b and 2) was mainly observed in the granular layer and stratum corneum. Our results contrast with those obtained by Sawicki et al. (1998), which showed that reporter gene expression (β-galactosidase) was expressed throughout mouse epidermis when driven by a similar CAG promoter. Other researchers have also noted that reporter gene expression is mainly observed in the upper layers of the epidermis with other promoters including CMV.34

The hMGFP aggregate formation (Figures 1b and 4) in the granular layer does not appear to be related to the electron dense keratohyalin granules (for which the stratum granulosum is named35), as immunohistochemical analysis performed with anti-filaggrin antibodies does not show co-localization (data not shown). The hMGFP aggregates may, however, result from misfolding of the fusion protein as similar aggregates have been observed in non-skin tissues of an independent Tg CBL/hMGFP mouse (silenced reporter gene mouse crossed with a transgenic mouse expressing Cre under the control of the ubiquitin promoter) expressing an identical reporter transgene (data not shown). Similarly, tissue culture cells transiently expressing the CBL/hMGFP construct also show aggregate formation (data not shown). The reason for aggregate dispersion, as cells further differentiate from the granular layer to the stratum corneum, is also unknown, but may be due to proteolytic cleavage by enzymes present and responsible for cleavage of pro-proteins (for example, pro-filaggrin granules are dephosphorylated, which allows their proteolysis into filaggrin molecules in the squames) during this differentiation process.36 The presence of hMGFP aggregates does not appear to adversely affect the mice and no differences in skin structure are detectable between positive and negative littermates using standard hematoxylin and eosin staining of skin sections (see Supplementary Figure S1)

Several groups have reported earlier that ‘naked’ plasmid DNA can be taken up by mouse or human keratinocytes and that intradermally injected reporter plasmid is expressed by these cells.34, 37, 38 In this study, we present data demonstrating that phosphate-buffered saline (PBS)-formulated unmodified siRNAs are capable of silencing pre-existing reporter gene expression in the Tg CBL/hMGFP mouse model. In the earlier work, we showed knockdown of reporter gene expression by coinjection of reporter plasmids and siRNAs into mouse skin;4, 34, 39, 40 therefore, the ability of plasmid DNA (as a ‘carrier’) to facilitate keratinocyte uptake of siRNA was examined. A similar number of areas (and percentage of cells per area) with strongly siRNA-mediated hMGFP silencing were observed when siRNA was injected with or without plasmid DNA (compare Figures 4d with b), suggesting that siRNA can be taken up by keratinocytes without the need of a carrier.

The combination of the Tg CBL/hMGFP mouse model with DAC fluorescence intravital imaging opens up new opportunities to monitor the effectiveness of siRNAs over time and at cellular resolution in living tissues of mice and human beings. Importantly, using both these emerging technologies together presents a powerful approach to evaluate the effectiveness of siRNA delivery, a major hurdle that must be overcome for efficient development of these agents as therapeutics. If delivery issues can be resolved, siRNA technology is poised to become an effective approach for treatment of skin disorders resulting from dominant mutations or overexpression of disease-causing genes.

Materials and methods

Animals

All animal work was carried out according to the guidelines for Animal Care of both National Institutes of Health and Stanford University.

Plasmid constructs

The phMGFP/CBL plasmid was generated by PCR amplification of the CBL coding sequence from the plasmid pCBR-Control Vector (Promega, Madison, WI, USA) using the primers ‘G4S-CBL+’ (capital letters correspond to the coding regions) 5′-gctctagaGGCGGTGGTGGATCCGGTGGCGGTGGATCAGGTGGAGGTGGATCCggaATGGTAAAGCGTGAGAAAAATGTC and ‘CBL- XbaI-’ 5′-cgtctagattaCTAACCGCCGGCCTTCACCAACAATTGTTTCAACAGC introducing upstream and downstream Xba I sites (shown in bold). The PCR product was ligated into the pCR4 Blunt TOPO plasmid using the Zero Blunt PCR Cloning kit (Invitrogen, Carlsbad, CA, USA). The inserts were cut out of this vector and inserted into the unique Xba I site downstream of the hMGFP sequence in the Monster GFP (phMGFP) plasmid (Promega). The hMGFP stop codon, the upstream cysteine residue and the sequence upstream of the G4S linker (CTAATAGTTCTAGAG) were removed and replaced with ‘TACC’ using QuikChange site-directed mutagenesis (Stratagene, La Jolla, CA, USA) using the following primers: ‘GFP-CBL+’ 5′-GCACGCCGAAGCCCACAGCGGACTACCCCGCCAGGCCGGtaccGGCGGTGGTGGATCCGGAGGCGGTGGATCAGGTGG and ‘GFP-luc-’ 5′-CCACCTGATCCACCGCCTCCGGATCCACCACCGCCggtaCCGGCCTGGCGGGGTAGTCCGCTGTGGGCTTCGGCGTGC.

The pCBL/hMGFP plasmid was created using a similar strategy. The CBL coding region was amplified through PCR using the following primers: ‘CBLRed-5’ 5′-gcgctagcccgggctcgagatctgcgatctaagtaagc and ‘CBLRed-linker’ 5′-cgctagcGGATCCACCTCCACCTGATCCACCGCCACCGGATCCACCACCGCCGGTACCGCCGGCCTTCACCAACAATTG to remove the stop codon and to introduce upstream and downstream Nhe I sites (shown in bold). This product was cut and ligated into the Nhe I site that exists upstream of the hMGFP coding region in the phMGFP plasmid. The undesired nts left between the coding regions were removed with the QuikChange kit using the primers ‘Luc-GFP fusion-’ 5′-GGATCCGGTGGCGGTGGATCAGGTGGAGGTGGATCCggaATGGGCGTGATCAAGCCCGACATGAAGATCAAGCTGCGG and ‘Luc-GFP fusion+’ 5′-CCGCAGCTTGATCTTCATGTCGGGCTTGATCACGCCCATtccGGATCCACCTCCACCTGATCCACCGCCACCGGATCC. The cloning strategy introduced a 15-amino-acid linker (G4S)3 between the luciferase and fluorescent protein in both plasmids (phMGFP/CBL and pCBL/hMGFP). All plasmids were confirmed by dideoxy sequencing before use.

PhK1441 plasmid (K14 coding region cloned into the pCRII-TOPO dual-promoter plasmid (Invitrogen)) was a gift from Dr Mohammed Ikram (Institute for Cell and Molecular Science, Barts and the London School of Medicine and Dentistry). A DNA construct for hMGFP detection was generated by removing the K14 coding region from phK14 by digestion with Hind III/Acc 65I. The hMGFP coding region was generated from pCBL/hMGFP by digestion with Nsi I. After digestion, the products were separated by agarose gel electrophoresis and the resultant fragments were purified from gel plugs using the Zymoclean Gel DNA Recovery kit (Zymo Research, Orange, CA, USA). The overhanging ends were filled in to make blunt ends using DNA polymerase I Klenow (New England Biolabs, Beverly, MA, USA) and joined with T4 ligase (New England Biolabs) generating pTD137 (hMGFP coding region in the pCRII-TOPO dual-promoter vector). The sequence and orientation of the hMGFP insert were verified before use.

Generation of the reporter mouse skin model

A silenced dual-reporter transgenic mouse was created to enable in vivo bioluminescence imaging and fluorescence imaging after breeding with Cre-expressing transgenic mice. The CBL/hMGFP cassette from the pCBL/hMGFP vector was cloned downstream of the Renilla luciferase (rLuc) separated by a strong translation stop sequence in the following order: loxP-rLuc-stop-loxP-CBL-hMGFP. This construct was used to generate the transgenic mouse using standard methods of pronuclear injection. The transgene contains the CAG (CMV enhancer-CAG) promoter,30 the coding sequence of rLuc (flanked by loxP sites) and the coding sequence of CBL fused to hMGFP. The silenced reporter transgenic mouse was crossed with a Tg(KRT14-cre)1Amc transgenic mouse (The Jackson Laboratory, Bar Harbor, ME, USA) expressing Cre recombinase driven by the keratinocyte-specific K14 promoter. Mice resulting from this cross were tested for the expression of CBL by bioluminescence imaging in an IVIS 200 Imaging System (Xenogen product of Caliper Life Sciences, Alameda, CA, USA) after intraperitoneal injection of luciferin (30 mg kg–1 body weight, data not shown). A bioluminescent mouse (considered as positive) and a non-bioluminescent (negative) mouse from the same litter were killed and the footpad skin was removed, embedded in O.C.T. compound (Tissue-Tek, Torrance, CA, USA) and frozen on dry ice. Vertical cross sections (10 μm) were imaged with the GFP filter set (470 nm excitation, 525 nm emission) in an Axio Observer Inverted Fluorescence Microscope (Zeiss, Thornwood, NY, USA) equipped with an AxioCam MRm camera using AxioVs40 V4.6.3.0 software to visualize transgene fluorescence.

In situ hybridization of reporter mRNA in transgenic mouse skin

Antisense digoxigenin (Dig)-labeled riboprobes were prepared from pTD137 (hMGFP coding region cloned into pCRII-TOPO dual-promoter plasmid) and phK14 (K14 coding region cloned into pCRII-TOPO dual-promoter plasmid) linearized with Bsg I and Hind III, respectively. Linearized plasmids were in vitro transcribed with T7 polymerase (Promega) and Dig RNA-labeling mix (Roche Applied Science, Indianapolis, IN, USA) resulting in 693 and 1200 nt riboprobes for hMGFP and K14, respectively. Reactions were treated with DNase I (Invitrogen) to remove template DNA (according to manufacturer's instructions) and NTPs were removed by gel filtration (G-50 column; GE Healthcare, Piscataway, NJ, USA). Riboprobes were stored in RNase-free water at −80 °C.

For sectioning and in situ hybridization, skin from the Tg CBL/hMGFP mouse footpad was excised and fixed overnight at 4 °C in ice-cold 4% PFA in PBS. After fixation, tissues were placed into OCT compound and frozen at −20 °C. Skin vertical cross sections (10 μm) were obtained in a cryotome (Jung Frigocut 2800E, Leica, Nussloch, Germany). Sections were fixed for 10 min in ice-cold 4% PFA in PBS, digested with proteinase K (1 μg ml–1 in PBS) for 12.5 min at room temperature and again fixed for 10 min in ice-cold 4% PFA in PBS. Sections were dehydrated by incubation in increasing methanol solutions in PBS (25, 50, 75 and 100% twice) for 2 min each. Sections were pre-hybridized in hybridization buffer (50% formamide (Sigma, St Louis, MO, USA), 5 × SSC, 5 mM ethylene diamine tetraacetic acid, 0.1% CHAPS (Sigma), 50 μg ml–1 heparin (Sigma), 1 mg ml–1 yeast tRNA (Sigma), 0.1% Triton-X 100, 2% Blocking Reagent (Roche Applied Science)) for 1 h at the hybridization temperature (52 °C for hMGFP and 62 °C for K14). Riboprobes were heat denatured by addition of an equal volume of formamide and incubated at 65 °C for 5 min, followed by immediate cooling on ice. Sections were hybridized with heat-denatured hMGFP or K14 Dig-labeled riboprobes (3 μg ml–1 final concentration in hybridization buffer) overnight at the respective hybridization temperatures. Sections were washed in 50% formamide with 2 × SSC followed by washing in 2 × SSC at the hybridization temperatures for 30 min each followed by washing twice for 30 min each with 0.5 × SSC or 0.2 × SSC at the hybridization temperatures, respectively. Sections were rinsed in PBS with 0.1% Triton-X 100 (PBT) and blocked in the same solution with 5% heat-inactivated sheep serum (Sigma). After incubation with 1:2000 anti-Dig antibody (sheep, conjugated to alkaline phosphatase (Roche Applied Science)) in PBT overnight at 4 °C, sections were rinsed in PBT, followed by incubation in 1 mM levamisol (Sigma) in PBT to inhibit endogenous alkaline phosphatase activity that may contribute to background. Sections were then rinsed in NTMT buffer (100 mM NaCl, 50 mM MgCl2, 0.1% Triton-X 100, 100 mM Tris–HCl, pH 9.5) and developed in NTMT buffer containing 175 μg ml–1 BCIP (Promega) and 350 μg ml–1 NBT (Promega) for 8 h (hMGFP) and 2 h (K14), respectively.

In vitro screening of siRNAs targeting CBL

Using the Dharmacon siDesign Center ‘http://www.dharmacon.com/DesignCenter/DesignCenterPage.aspx’, 10 siRNAs were designed and synthesized (Thermo Fisher Scientific, Dharmacon Products, Thermo Fisher Scientific, Dharmacon Products Lafayette, CO, USA) that target the CBL coding region. Guide-strand sequences: CBL1, 5′-UUGGCGGUAAUUUCUGUACuu; CBL2, 5′-AUACCGAUAUACCAUGCGGuu; CBL3, 5′-UAACGAUCCACGACGUAAAuu; CBL4, 5′-UCUUCGUCGUAAUAUCCAAuu; CBL5, 5′-AAUCACGCGGAGACCGACCuu; CBL6, 5′-UUCGUCGGGAAUGUAGCUCuu; CBL7, 5′-UUCGCAACCGUGAAUAUUCuu; CBL8, 5′-UUCAUCGCCGACCACAUCGuu; CBL9, 5′-AGUUCGUCGGGAAUGUAGCuu; CBL10, 5′-AUACGCUUAAUAAAGUUGGuu. The functional activity of each of these siRNAs was tested by co-transfection with the reporter plasmid phMGFP/CBL in 293T cells (in triplicate) using lipofectamine 2000 (Invitrogen) according to manufacturer's instructions in 48-well plates (80% confluent at time of transfection) with 1 nM of each siRNA and 100 ng of phMGFP/CBL, supplemented with pUC19 to give a final nucleic acid concentration of 400 ng per transfection. The most potent siRNAs were selected for a dose-response study using the following concentrations: 0.015, 0.06, 0.25, 1 and 4 nM siRNA. Silencing efficiency was evaluated by analyzing the bioluminescent signal 5 min after addition of 50 μl of luciferin (3 mg ml–1) per well in the IVIS 50 Imaging System (Caliper). The collected data were processed using the Living Image 2.50.1 software.

siRNA immunostimulation assay in PBMCs

Human ‘buffy coats’ (6 ml) were obtained from the Stanford Blood Bank and transferred to 50 ml conical tubes. Serum-free RPMI-1640 media (29 ml) was added to each tube. The diluted samples were overlayed on 15 ml of Ficoll-Plaque Plus (GE Healthcare). The samples were centrifuged at 2000 r.p.m. for 20 min and the ‘buffy coat’ PBMC layer was removed and transferred to a clean 50 ml conical tube. The cells were diluted three-fold with serum-free RPMI-1640 media and centrifuged at 1500 r.p.m. for 10 min. The cells were washed three times by resuspension in serum-free RPMI-1640 and centrifugation at 1500 r.p.m. for 10 min. The cells were ultimately resuspended in RPMI-1640 with 10% fetal bovine serum and penicillin (100 U ml–1)/streptomycin (100 μg ml–1). PBMCs (1 × 106 cells ml–1 per well) were seeded in a 24-well plate. The siRNAs (50 pmol), with and without DOTAP (12 pmol siRNA per μg DOTAP, Roche Applied Science), were added to each well. The synthetic single-strand RNA poly(cytidylic-inosinic) acid (poly I:C-ss, Sigma, St Louis, MO, USA) and double-stand RNA polyriboinosinic:polyribocytidylic acid (poly I:C-ds, Sigma) were used as positive controls (1.5 μg poly I:C per μg DOTAP). At the indicated time points, total RNA was isolated using the RNeasy kit from Qiagen (Qiagen, Valencia, CA, USA), according to manufacture's instructions on the on-column DNase digestion. Total RNA was reverse transcribed using the Superscript III First Strand Synthesis system (Invitrogen) using 1–2 μg of total RNA and random hexamer primers. The RT enzyme was heat denatured at 85 °C for 5 min and qPCR reactions were prepared as follows: 1 μl 20 × primer/probe (TNF, Hs99999043_m1, and IL-6, Hs99999032_m1, Applied Biosystems) and 10 μl 2 × master mix were combined in a 96-well plate. A measure of 9 μl cDNA (0.5 ng μl–1) was then added and analyzed using the ABI standard 7500 procedure. The data were analyzed with the Applied Biosystems Sequence Detection software (version 1.4) and reported as the relative quantitation using GAPDH (Hs99999905_m1, Applied Biosystems) as the endogenous control. All data points reported are the mean of three replicate assays and error is reported as the standard error.

Intradermal injection of siRNAs into mouse footpads and analysis of skin sections

CBL3 siRNA (60 μg; 4.5 nmol) per footpad in 50 μl PBS alone or in combination with pUC19 (5 μg) as a potential ‘carrier’ was intradermally injected daily into the footpad, as described earlier,34 for 14 days. As a control, an equivalent quantity of irrelevant siRNA (K6a_513a.124) was intradermally injected into the counterpart footpad. Three treated mice were killed, skin tissues removed from the footpad, embedded in O.C.T. compound and frozen directly in dry ice. Vertical cross sections (10 μm) were taken and mounted with Hydromount (National Diagnostic, Highland Park, NJ, USA) containing DAPI for nuclear staining. Slides were examined using an Axio Observer Inverted Fluorescence Microscope as described above. The percentage of silenced cells was estimated in overlapping images by counting nuclei (blue stained with DAPI) with or without green fluorescence signal in the surrounding cytoplasm in 17 areas of 3 different treated mice, in which reduction of hMGFP was detected.

Analysis of hMGFP mRNA levels by RT-qPCR

Three mice were treated with CBL3 siRNA daily (80 μg (6.0 nmol) per footpad in 50 μl PBS) by intradermal injection into the footpad for 12 days. As a negative control, an irrelevant siRNA (K6a_513a.12) was intradermally injected into the counterpart footpad at the same timepoints and concentrations. The mice were killed, skin tissues were removed from the footpad and frozen directly in dry ice. Tissue was homogenized in a ‘bead beater’ instrument (FastPrep-24, FP24, from MP Biomedicals, Solon, OH, USA) using D matrix to mechanically lyse the cells for 40 s at speed setting 6 (m s–1) and total RNA was isolated using the Fibrous RNeasy RNA isolation kit (Qiagen) following the manufacturer's instructions. RNA was reverse transcribed using the Superscript III First Strand Synthesis system (Invitrogen) as described above. RT-qPCR was run in the ABI 7500 Fast Sequence Detection system (Applied Biosystems, Foster City, CA, USA) using standard procedures. A Taqman Gene Expression Assay was specifically designed for hMGFP (hMGFP-F: 5′-CCCCAAGGACATCCCTGACT; hMGFP-R: TGCTTCGCTCCCACGAGTA and probe 6FAM-TCAAGCAGACCTTCCCCGA-MGBNFQ; Applied Biosystem). The pre-designed 20 × Taqman Gene Expression Assay for mouse K14 (Applied Biosystems Catalog #Mm00516876) was used as an endogenous control. The hMGFP primers and probe were prepared such that the final primer and probe concentrations were 200 and 120 nM, respectively. The data were analyzed as described above and reported as the relative quantitation. All data points reported are the mean of three replicate assays in three independent experiments (nine data points per sample in total) and error is reported as the standard error.

Intravital DAC fluorescence imaging of Tg CBL/hMGFP mouse skin

Tg CBL/hMGFP mice were bred with Balb/c mice (a white strain) to obtain a white background to facilitate in vivo detection of hMGFP signal using a DAC microscope. CBL3 siRNA (80 μg) was delivered through 18 daily intradermal injections into the left footpad of white mice that had earlier shown luciferase expression in the skin (data not shown). As a control, an equal amount of irrelevant siRNA (K6a_513a.12) was intradermally injected into the contralateral footpad. The footpads were analyzed by intravital imaging at the indicating time points during and after treatment using the DAC microscope equipped with a fiber-coupled 488 nm wavelength laser.14 A microelectromechanical systems scanner42 within the DAC microscope performs two-dimensional (2D) en face imaging at four frames per second. Three-dimensional (3D) images were acquired by translating the microelectromechanical systems scanner in the depth direction with a piezoelectric actuator to sequentially save image stacks. The footpads were scanned at a single depth of 20 μm to localize both those regions expressing hMGFP and those in which the signals were absent. At these sites, image stacks were collected, processed and reconstructed into 3D volumes and video files (as Supplementary material) using Amira software (Visage Imaging, Carlsbad, CA, USA). Intravital imaging was performed with the mouse under isoflurane anesthesia according to institutional guidelines. Optical gel (NyoGel OC-431A-LVP, Nye Lubricants Inc., Fairhaven, MA, USA, index of refraction=1.46) was used as a coupling agent between the footpad skin and the microscope.

References

  1. 1.

    , . Strategies for silencing human disease using RNA interference. Nat Rev Genet 2007; 8: 173–184.

  2. 2.

    , , , . Interfering with disease: a progress report on siRNA-based therapeutics. Nat Rev 2007; 6: 443–453.

  3. 3.

    , . Knocking down disease with siRNAs. Cell 2006; 126: 231–235.

  4. 4.

    , , , , , et al. Single-nucleotide-specific siRNA targeting in a dominant-negative skin model. J Invest Dermatol 2008; 128: 594–605.

  5. 5.

    , , , , , et al. Designing siRNA that distinguish between genes that differ by a single nucleotide. PLoS Genet 2006; 2: e140.

  6. 6.

    , , , . RNAi therapeutics: an update on delivery. Curr Opin Mol Ther 2008; 10: 158–167.

  7. 7.

    , , , . Delivering silence: advancements in developing siRNA therapeutics. Curr Opin Drug Discov Devel 2008; 11: 217–224.

  8. 8.

    . Gene therapy progress and prospects: the skin—easily accessible, but still far away. Gene Therapy 2006; 13: 1555–1563.

  9. 9.

    , , . Cutaneous gene transfer for skin and systemic diseases. J Intern Med 2002; 252: 1–10.

  10. 10.

    , , , , , et al. Therapeutic siRNAs for dominant genetic skin disorders including pachyonychia congenita. J Dermatol Sci 2008; 51: 151–157.

  11. 11.

    , . Keratins and skin disorders. J Pathol 2004; 204: 355–366.

  12. 12.

    , , , , , et al. Clinical and pathological features of pachyonychia congenita. J Investig Dermatol Symp Proc 2005; 10: 3–17.

  13. 13.

    , , , , . Pachyonychia congenita. Gene Rev 2006, .

  14. 14.

    , , , , , et al. Three-dimensional in vivo imaging by a handheld dual-axes confocal microscope. Opt Express 2008; 16: 7224–7232.

  15. 15.

    , , . Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 1991; 108: 193–199.

  16. 16.

    , , , , , et al. Misinterpreting the therapeutic effects of siRNA caused by immune stimulation. Hum Gene Ther 2008; 19: 991–999.

  17. 17.

    , , , , , et al. Sequence- and target-independent angiogenesis suppression by siRNA via TLR3. Nature 2008; 452: 591–597.

  18. 18.

    , . Activation of the mammalian immune system by siRNAs. Nat Biotechnol 2005; 23: 1399–1405.

  19. 19.

    , , , , , et al. Determinants of cytokine induction by small interfering RNA in human peripheral blood mononuclear cells. J Interferon Cytokine Res 2008; 28: 221–233.

  20. 20.

    , . Overcoming the innate immune response to small interfering RNA. Hum Gene Ther 2008; 19: 111–124.

  21. 21.

    . RNA interference and innate immunity. Adv Drug Deliv Rev 2007; 59: 153–163.

  22. 22.

    . Does the understanding of immune activation by RNA predict the design of safe siRNAs? Front Biosci 2008; 13: 4379–4392.

  23. 23.

    , , , , , et al. Allele-specific silencing of dominant disease genes. Proc Natl Acad Sci USA 2003; 100: 7195–7200.

  24. 24.

    , , , , , . Identification and allele-specific silencing of the mutant huntingtin allele in Huntington's disease patient-derived fibroblasts. Hum Gene Ther 2008; 19: 710–719.

  25. 25.

    , . The genodermatoses: candidate diseases for gene therapy. Hum Gene Ther 2000; 11: 2267–2275.

  26. 26.

    , . Human keratin diseases: the increasing spectrum of disease and subtlety of the phenotype-genotype correlation. Br J Dermatol 1999; 140: 815–828.

  27. 27.

    , , , . Inhibition of respiratory viruses by nasally administered siRNA. Nat Med 2005; 11: 50–55.

  28. 28.

    , , , , . Lack of interferon response in animals to naked siRNAs. Nat Biotechnol 2004; 22: 1579–1582.

  29. 29.

    , , , , , . Astrocytes are very sensitive to develop innate immune responses to lipid-carried short interfering RNA. Glia 2009; 57: 93–107.

  30. 30.

    , , , , . A composite CMV-IE enhancer/beta-actin promoter is ubiquitously expressed in mouse cutaneous epithelium. Exp Cell Res 1998; 244: 367–369.

  31. 31.

    , . Evidence for posttranscriptional regulation of the keratins expressed during hyperproliferation and malignant transformation in human epidermis. J Cell Biol 1986; 103: 1945–1955.

  32. 32.

    , , , , . The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 1982; 31: 11–24.

  33. 33.

    , , , , , et al. Monoclonal antibody studies of mammalian epithelial keratins: a review. Ann NY Acad Sci 1985; 455: 307–329.

  34. 34.

    , , , , , et al. Delivery and inhibition of reporter genes by small interfering RNAs in a mouse skin model. J Invest Dermatol 2007; 127: 2577–2584.

  35. 35.

    , . The chemical nature of keratohyalin granules of the epidermis. J Cell Biol 1970; 47: 593–603.

  36. 36.

    , , , . Filaggrin in the frontline: role in skin barrier function and disease. J Cell Sci 2009; 122 (Pt 9): 1285–1294.

  37. 37.

    , , , , . Efficient expression of naked plasmid DNA in mucosal epithelium: prospective for the treatment of skin lesions. J Invest Dermatol 1998; 111: 605–608.

  38. 38.

    , , . Expression of naked DNA in human, pig, and mouse skin. J Clin Invest 1996; 97: 2911–2916.

  39. 39.

    , , , , , et al. Stability study of unmodified siRNA and relevance to clinical use. Oligonucleotides 2008; 18: 345–354.

  40. 40.

    , , , , , et al. Development of therapeutic siRNAs for pachyonychia congenita. J Invest Dermatol 2008; 128: 50–58.

  41. 41.

    , , , , , et al. GLI2 is expressed in normal human epidermis and BCC and induces GLI1 expression by binding to its promoter. J Invest Dermatol 2004; 122: 1503–1509.

  42. 42.

    , , , , , . Two-dimensional MEMS scanner for dual-axes confocal microscopy. J Microelectromech Syst 2007; 16: 969–976.

  43. 43.

    , , , , , . Sequence-dependent stimulation of the mammalian innate immune response by synthetic siRNA. Nat Biotechnol 2005; 23: 457–462.

  44. 44.

    . Induction of inflammatory cytokines and interferon responses by double-stranded and single-stranded siRNAs is sequence-dependent and requires endosomal localization. J Mol Biol 2005; 348: 1079–1090.

Download references

Acknowledgements

This study was funded in part by grants from the National Institutes of Health (R43AR055881, RLK and U54 CA105296-01, CHC). We thank Irwin McLean and Birgitte Lane for critical review of the manuscript, Heini Ilves, Manuel Flores and Maria Fernanda Lara for technical support, and Mohammed Ikram and Mike Philpott for the kind gift of phK14 plasmid. EGG is the recipient of a PC Project fellowship.

Author information

Affiliations

  1. Molecular Imaging Program at Stanford (MIPS), Stanford University School of Medicine, Stanford, CA, USA

    • E Gonzalez-Gonzalez
    • , Q Wang
    • , W Piyawattanametha
    • , T C Doyle
    •  & C H Contag
  2. Department of Pediatrics, Stanford University School of Medicine, Stanford, CA, USA

    • E Gonzalez-Gonzalez
    • , Q Wang
    • , W Piyawattanametha
    • , T C Doyle
    •  & C H Contag
  3. Department of Radiology, Stanford University School of Medicine, Stanford, CA, USA

    • C H Contag
  4. Department of Microbiology and Immunology, Stanford University School of Medicine, Stanford, CA, USA

    • C H Contag
  5. Department of Electrical Engineering, Stanford University, Stanford, CA, USA

    • H Ra
    • , W Piyawattanametha
    • , M J Mandella
    • , G S Kino
    •  & O Solgaard
  6. TransDerm Inc., Santa Cruz, CA, USA

    • R P Hickerson
    •  & R L Kaspar
  7. NECTEC, Pathumthani, Thailand

    • W Piyawattanametha
  8. Thermo Fisher Scientific, Dharmacon Products, Lafayette, CO, USA

    • D Leake
  9. Centre for Cutaneous Research, Cancer Research UK, Institute of Cell and Molecular Science, Barts and the London School of Medicine and Dentistry, Whitechapel, London, UK

    • A A Avilion

Authors

  1. Search for E Gonzalez-Gonzalez in:

  2. Search for H Ra in:

  3. Search for R P Hickerson in:

  4. Search for Q Wang in:

  5. Search for W Piyawattanametha in:

  6. Search for M J Mandella in:

  7. Search for G S Kino in:

  8. Search for D Leake in:

  9. Search for A A Avilion in:

  10. Search for O Solgaard in:

  11. Search for T C Doyle in:

  12. Search for C H Contag in:

  13. Search for R L Kaspar in:

Corresponding author

Correspondence to R L Kaspar.

Supplementary information

About this article

Publication history

Received

Revised

Accepted

Published

DOI

https://doi.org/10.1038/gt.2009.62

Supplementary Information accompanies the paper on Gene Therapy website (http://www.nature.com/gt)

Further reading