**3.1 Origin of flaky tail mice**

6 Atopic Dermatitis – Disease Etiology and Clinical Management

pyrrolidone-5-carboxylic acid (PCA). The concentrations of UCA and PCA in SC in the carriers of *FLG* mutations were significantly lower than those in healthy donors (Kezic, et al., 2009).

An *in vitro* experiment using filaggrin knocked down human organotypic skin cultures showed enhanced penetration of hydrophilic dye Lucifer yellow, smaller lamellar bodies, and deficiency of their typical lamellae without altered lipid composition (Mildner, et al., 2010). In addition, UCA, one of the filaggrin-derived free amino acids and as an important UV absorbent within SC, was decreased following filaggrin knocked down, leading to increased

The SC serves as a biosensor of the external environment and a link between innate and adaptive immune systems (Vroling, et al., 2008). The critical association between the abnormal barrier in AD and Th2 polarization may in part be explained by the production of the cytokine, thymic stromal lymphopoietin (TSLP) (Ebner, et al., 2007). TSLP is expressed by epithelial cells, with the highest levels seen in lung-derived and skin-derived epithelial cells (Soumelis, et al., 2002, Ziegler, 2010), and is highly detected in the lesional skin of AD (Soumelis, et al., 2002) Inducible TSLP transgene specifically in the skin leads to the development of a spontaneous Th2-type skin inflammatory disease with the hallmark

TSLP has been shown to activate dendritic cells to drive Th2 polarization, through upregulation of the co-stimulatory molecules CD40, CD80, and OX40L, triggering the differentiation of allergen-specific naïve CD4+ T cells to Th2 cells that produce IL-4, IL-5,

Patients with Netherton syndrome (NS), a severe ichthyosis in which affected individuals experience a significant predisposition for AD, have elevated levels of TSLP in their skin. Upregulated kallikrein (KLK) 5 in the skin of NS patients directly activates proteinaseactivated receptor 2 (PAR-2) and induces nuclear factor kappaB-mediated overexpression of TSLP, intercellular adhesion molecule 1, TNF-α, and IL-8. This phenomenon occurs independently of the environment, adaptive immune system and underlying epithelial barrier defect (Briot, et al., 2009, Briot, et al., 2010). *In vitro* study using human keratinocyte cell line HaCaT cells and reconstituted human epidermal layers transfected with filaggrin siRNA showed increased production of TSLP via toll-like receptor (TLR) 3 stimulation (Lee, et al., 2011). These findings suggest that reduced filaggrin levels may influence innate

immune response via TLR stimuli and elevate TSLP, leading to AD-like skin lesions.

AD is one of the emerging diseases in which epidermal dysfunction increases allergen and microbial penetration in the skin, with the consequent development of adaptive Th2 immune responses (Kondo, et al., 1998) within regional lymphoid tissue. The resultant Th2 cells may then home back to the skin or lungs, where they recognize allergen in the skin (McPherson, et al., 2010), which leads to local Th2 inflammation, reduced antimicrobial peptide expression (Nomura, et al., 2003), and filaggrin downregulation (Howell, et al., 2007). Indeed, the induction of circulating allergen-specific CD4+ T cells may be an important prerequisite underlying the pathogenesis of the atopic march (O'Regan, et al., 2009). Among moderate-to-severe AD patients, the *FLG* mutation carriers showed a greater number of house dust mite Der p1-specific IL-4 producing CD4+ T cells, suggesting that filaggrin mutations predispose to the development of allergen-specific CD4+ Th2 cells. The

Therefore, filaggrin deficiency is sufficient to impair epidermal barrier formation.

sensitivity to UVB-induced keratinocyte (KC) damage (Mildner, et al., 2010).

**2.3 Filaggrin and altered immunobiology** 

and IL-13 (Ebner, et al., 2007, Soumelis, et al., 2002).

features of AD (Yoo, et al., 2005).

The above findings indicate the involvement of filaggrin in the development of AD. Therefore, the impact of filaggrin deficiency on cutaneous biological functions *in vivo* should be analyzed in detail. To address this issue, animal models are of great value.

Flaky tail mice (*Flgft*), first introduced in 1958, are spontaneously mutated mice with smaller ears, tail constrictions, and a flaking tail skin appearance (Lane, 1972). *Flgft* mice were outcrossed onto B6 mice at Jackson Laboratory (Bar Harbor, ME, USA) (Lane, 1972, Presland, et al., 2000) (Note: Although this strain was crossed with B6, it is not a B6 congenic strain but rather a hybrid stock that is probably semi-inbred). Homozygous *Flgft* mice have dry, flaky skin which expresses reduced amounts of profilaggrin mRNA and abnormal profilaggrin protein that is not processed to filaggrin monomers (Fallon, et al., 2009, Presland, et al., 2000).

Recently, it has been revealed that the gene responsible for the characteristic phenotype of *Flgft* mice is a single nucleotide deletion at position 5303 in exon 3 (5303delA) of the profilaggrin gene, resulting in a frameshift mutation and premature truncation of the predicted protein product. The copy number of the filaggrin repeat contained within this gene varies depending on the background strain. This mutant occurs in an allele with 16 copies of the filaggrin repeat (Fallon, et al., 2009).

*Flgft* mouse carries double gene mutation, *Flg* and matted (*ma*) in which the locations of the mutated genes are within close linkage to one another (Lane, 1972). The *ma* gene characteristic reported by Searle & Spearman (1957) causes the body-hair of affected mice to be brittle and inflexible, which results in longitudinal splitting and breaking due to friction against the cage and other objects. This mutation is a fully penetrant recessive house-mouse mutant which belongs to the "naked" category (i.e., a house-mouse with baldness resulting from the breaking of hairs or from hereditary hairlessness). This mutation can be identified morphologically by (1) erection of hairs, (2) matting of hair in clumps, (3) a tendency towards baldness, (4) a change from black- to brown-colored melanin in old hairs. The age at which this mutant is first identified based on external appearance varies from between two to four weeks (Jarret A, 1957, Searle A.G., 1957).

Recognition of the features of this mouse is more evident between 5 and 14 days of age when constricted, flaking tail skin and thickened short pinna of the ears are observed. In addition, *Flgft* mice are often smaller than their normal siblings at this age. Routine histological sections stained with hematoxylin and eosin showed that the stratum granulosum in *Flgft* mice at 1, 2, 4, and 8 days of age does not contain as many granular layers as that of non-*Flgft* mice (Lane, 1972). Mice of the *Flgft* genotype express an abnormal profilaggrin polypeptide that does not form normal keratohyalin F-granules and is not proteolytically processed to filaggrin. Therefore, filaggrin is absent from the cornified layers in the epidermis of the *Flgft* mouse (Fallon, et al., 2009, Presland, et al., 2000, Scharschmidt, et al., 2009). Consistently, we and others have described that *Flgft* mice express a truncated and smaller profilaggrin protein that is not processed to filaggrin (Fallon, et al., 2009, Moniaga, et al., 2010, Presland, et al., 2000) (Fig.1).

Flaky Tail Mouse as a Novel Animal Model of Atopic Dermatitis:

the gender of *Flgft* mice throughout the period (Moniaga, et al., 2010).

*Flgft* mice (Scharschmidt, et al., 2009).

scores (right panel)

Fig. 3. Characteristics of the clinical skin lesions.

Possible Roles of Filaggrin in the Development of Atopic Dermatitis 9

age (Oyoshi, et al., 2009), and the third contained no notice of any spontaneous dermatitis in

The fourth paper by Moniaga et al. have demonstrated that *Flgft* mice showed spontaneous dermatitis with skin lesions mimicking human AD as early as 5 weeks of age with mild erythema and fine scales and the cutaneous manifestations advanced with age in a steady state under SPF conditions (Moniaga, et al., 2010) (Fig. 2). The first manifestations to appear when mice were young were erythema and fine scaling; pruritic activity, erosion, and edema followed later (Fig. 3). In contrast, no cutaneous manifestation was observed in either C57BL/6 mice, studied as a control, or heterozygous mice intercrossed with *Flgft* and B6 mice kept under SPF conditions. There was no apparent difference in terms of clinical manifestations based on

Fig. 2. Clinical photographs of 20-week-old *Flgft* mice (left panel) and total clinical severity

Histological examination of the skin of *Flgft* mice stained with H&E revealed epidermal acanthosis, increased lymphocyte and mast cell infiltration and dense fibrous bundles in the dermis, in both younger (8-week-old) and older (18-week-old) *Flgft* mice; none of these conditions were observed in B6 mice (Fig. 4) (Moniaga, et al., 2010). These features were also reported in other studies (Fallon, et al., 2009, Oyoshi, et al., 2009) with more total cells,

Fig. 1. *Flgft* mouse has a truncated and smaller profilaggin and a lack of filaggrin protein.

#### **3.2 Flaky tail mouse and ichtyosis vulgaris**

Ichthyosis vulgaris (IV) is a heterogeneous autosomal skin disease characterized by dry and scaly skin, mild hyperkeratosis, and a decreased or absent granular layer that either lacks, or contains morphologically abnormal, keratohyalin granules (Manabe, et al., 1991). Several lines of evidences point to a genetic defect in a gene encoding *FLG* in IV. Immunoblotting studies showed that filaggrin protein was absent or markedly reduced in the epidermis of individuals with IV (Fleckman, et al., 1987, Sybert, et al., 1985). In line with this, it was proposed that *Flgft* mice could provide insight into the molecular basis of the filaggrindeficient human skin disorder IV. The epithelia of *Flgft* mice showed defects in tissue organization especially in the tail, an attenuated granular layer, reduced profilaggrin and a lacked of filaggrin granules in SC. In addition, keratinocytes culture from *Flgft* mice synthesized reduced amounts of profilaggrin mRNA and protein (Presland, et al., 2000).

#### **3.3 Flaky tail mouse in a steady state**

An early report demonstrated that *Flgft* mice without the *ma* mutation showed flaky skin as early as postnatal day 2, but became normal in appearance by 3 to 4 weeks of age without spontaneous dermatitis except for their slightly smaller ears (Lane, 1972). Later, the lack of filaggrin in the epidermis was proposed in the commercially available strain of *Flgft* mice, which has both *Flg* and *ma* mutations, as a model of IV, and therefore there was no discussion about the cutaneous inflammatory conditions from the perspective of AD (Presland, et al., 2000).

There have been four recent papers of *Flgft* mice as a model of filaggrin deficiency: the first paper used *Flgft* mice from which the *ma* mutation had been eliminated with four additional backcrosses to B6 mice (Fallon, et al., 2009), and the others used the commercially available *Flgft* mice (Moniaga, et al., 2010, Oyoshi, et al., 2009, Scharschmidt, et al., 2009). The first report showed only histological abnormality without clinical manifestation (Fallon, et al., 2009), and the second demonstrated spontaneous eczematous skin lesions after 28 weeks of

Fig. 1. *Flgft* mouse has a truncated and smaller profilaggin and a lack of filaggrin protein.

Ichthyosis vulgaris (IV) is a heterogeneous autosomal skin disease characterized by dry and scaly skin, mild hyperkeratosis, and a decreased or absent granular layer that either lacks, or contains morphologically abnormal, keratohyalin granules (Manabe, et al., 1991). Several lines of evidences point to a genetic defect in a gene encoding *FLG* in IV. Immunoblotting studies showed that filaggrin protein was absent or markedly reduced in the epidermis of individuals with IV (Fleckman, et al., 1987, Sybert, et al., 1985). In line with this, it was proposed that *Flgft* mice could provide insight into the molecular basis of the filaggrindeficient human skin disorder IV. The epithelia of *Flgft* mice showed defects in tissue organization especially in the tail, an attenuated granular layer, reduced profilaggrin and a lacked of filaggrin granules in SC. In addition, keratinocytes culture from *Flgft* mice synthesized reduced amounts of profilaggrin mRNA and protein (Presland, et al., 2000).

An early report demonstrated that *Flgft* mice without the *ma* mutation showed flaky skin as early as postnatal day 2, but became normal in appearance by 3 to 4 weeks of age without spontaneous dermatitis except for their slightly smaller ears (Lane, 1972). Later, the lack of filaggrin in the epidermis was proposed in the commercially available strain of *Flgft* mice, which has both *Flg* and *ma* mutations, as a model of IV, and therefore there was no discussion about the cutaneous inflammatory conditions from the perspective of AD

There have been four recent papers of *Flgft* mice as a model of filaggrin deficiency: the first paper used *Flgft* mice from which the *ma* mutation had been eliminated with four additional backcrosses to B6 mice (Fallon, et al., 2009), and the others used the commercially available *Flgft* mice (Moniaga, et al., 2010, Oyoshi, et al., 2009, Scharschmidt, et al., 2009). The first report showed only histological abnormality without clinical manifestation (Fallon, et al., 2009), and the second demonstrated spontaneous eczematous skin lesions after 28 weeks of

**3.2 Flaky tail mouse and ichtyosis vulgaris** 

**3.3 Flaky tail mouse in a steady state** 

(Presland, et al., 2000).

age (Oyoshi, et al., 2009), and the third contained no notice of any spontaneous dermatitis in *Flgft* mice (Scharschmidt, et al., 2009).

The fourth paper by Moniaga et al. have demonstrated that *Flgft* mice showed spontaneous dermatitis with skin lesions mimicking human AD as early as 5 weeks of age with mild erythema and fine scales and the cutaneous manifestations advanced with age in a steady state under SPF conditions (Moniaga, et al., 2010) (Fig. 2). The first manifestations to appear when mice were young were erythema and fine scaling; pruritic activity, erosion, and edema followed later (Fig. 3). In contrast, no cutaneous manifestation was observed in either C57BL/6 mice, studied as a control, or heterozygous mice intercrossed with *Flgft* and B6 mice kept under SPF conditions. There was no apparent difference in terms of clinical manifestations based on the gender of *Flgft* mice throughout the period (Moniaga, et al., 2010).

Fig. 2. Clinical photographs of 20-week-old *Flgft* mice (left panel) and total clinical severity scores (right panel)

Fig. 3. Characteristics of the clinical skin lesions.

Histological examination of the skin of *Flgft* mice stained with H&E revealed epidermal acanthosis, increased lymphocyte and mast cell infiltration and dense fibrous bundles in the dermis, in both younger (8-week-old) and older (18-week-old) *Flgft* mice; none of these conditions were observed in B6 mice (Fig. 4) (Moniaga, et al., 2010). These features were also reported in other studies (Fallon, et al., 2009, Oyoshi, et al., 2009) with more total cells,

Flaky Tail Mouse as a Novel Animal Model of Atopic Dermatitis:

and/or matted deficiency.

Oyoshi, et al., 2009).

al., 2009).

barrier disruption.

solution compared to its control littermate.

Possible Roles of Filaggrin in the Development of Atopic Dermatitis 11

The extent of severity of AD is known to be correlated with elevated serum IgE levels (Novak, 2009). Serum IgE and IgG1 levels in *Flgft* mice were significantly higher than those in control mice in the steady state under SPF conditions (Moniaga, et al., 2010, Oyoshi, et al., 2009). In addition, the numbers of CD4+ and CD8+ cells in the skin draining LNs in *Flgft* mice were significantly higher than those in control mice, but those of the spleen were similar for both groups. Thus, an enhanced cutaneous immune reaction seems to be induced in *Flgft* mice due to the condition of their skin induced by filaggrin

AD is thought to be mediated by helper T cell subsets, such as Th1, Th2, and Th17 (Bieber, 2008, Hattori, et al., 2010, Koga, et al., 2008). In the steady state, the skin of *Flgft* mice showed no difference of Th1 cytokine IFN-γ and Th2 cytokines IL-4 and IL-13 compared to the control. In contrast, there is a significant increase in mRNA expression of the Th17 cytokine IL-17, IL-17 promoting cytokines IL-6 and IL-23 (p19), and IL-17 inducible neutrophil attractant chemokine CXCL2 in *Flgft* mice (Moniaga, et al., 2010,

Since the barrier dysfunction is a key element in the establishment of AD, it is necessary to evaluate outside-to-inside barrier function from the perspective of invasion of external stimuli. Scharschmidt et al. reported increased bidirectional paracellular permeability of water-soluble xenobiotes by ultrastructural visualization in *Flgft* mice suggesting a defect in the outside-to-inside barrier. The ultrastructural visualization of tracer perfusion was analyzed by water-soluble, low molecular weight, electron-dense tracer lanthanum nitrate or fluorophore calcium green with enhanced penetration in *Flgft* mice. The data demonstrated that filaggrin deficiency leads to alterations in basal barrier function through a defect in the SC extracellular matrix and greater permeability through the same paracellular pathway that is used by water itself when exiting the skin (Scharschmidt, et

A new method for evaluating outside-to-inside barrier function quantitatively by measuring the penetrance of fluorescein isothiocyanate isomer 1 (FITC) through the skin has been developed (Moniaga, et al., 2010). The epidermis of *Flgft* mice contained a higher amount of FITC than that of B6 mice did (Fig.6 left panel). Consistently, fluorescence intensities observation in the epidermis of both mice showed stronger fluorescence in *Flgft* mice (Fig.6 right panel). In addition, the *Flgft* embryo was entirely dye permeable to toluidine blue

Another AD-like dermatitis model to test allergen priming of the skin in these mice was performed by application of ovalbumin (OVA) (Oyoshi, et al., 2009). Non tape-stripped skin of *Flgft* mice exposed to OVA exhibited significantly increased epidermal thickening, hyperkeratosis, spongiosis, acanthosis, and cellular infiltrates, as well as TEWL compared to control mice. mRNA levels for IL-17, IL-6, IL-23, IL-4, IFN-γ and CXCL2 but not IL-5 and IL-13 in the skin of *Flgft* mice after OVA exposure were significantly higher than those of control mice. The systemic immune response following cutaneous exposure revealed increased specific IgG and IgE to OVA, and splenocytes proliferated and produced OVAspecific Th1, Th2, Th17 and regulatory T cell cytokines (Fallon, et al., 2009, Oyoshi, et al., 2009). These findings demonstrate that *Flgft* mice tend to generate allergen-specific IgE and cytokine following cutaneous allergen challenge to the skin even without additional

**3.4 Flaky tail mouse showed enhanced percutaneous allergen priming** 

lymphocytes, eosinophils, and mononuclear cells in *Flgft* mice compared to control mice. These data support the diagnosis of AD-like dermatitis in *Flgft* mice in the steady state under SPF conditions.

Fig. 4. Hematoxyllin and eosin (H&E)-stained sections in 8- and 18-week old mice. Scale bar, 100µm

Therefore, there exist discrepancies among the results of four recent papers on the cutaneous manifestation in the steady states. It seems to be related to the presence or absence of the *ma* mutation and/or variation in the genetic backgrounds of the different strains used, and to environmental factor. It has been reported that Japan carries a higher morbidity of AD than other countries (1998, Williams, et al., 1999), possibly due to environmental factors such as pollen. Because barrier dysfunction is a common characteristic of AD (Elias, et al., 2008, Nomura, et al., 2007, Palmer, et al., 2006), TEWL is commonly measured as an indicator of barrier function (Gupta, et al., 2008). TEWL was significantly higher in *Flgft* mice than in B6 mice from an early age (4 weeks) to an older age (16 weeks) (Fig. 5) (Moniaga, et al., 2010).

Fig. 5. TEWL through dorsal skin of 5-, 8-, and 16-week-old B6 and *Flgft* mice.

Flowcytometry analysis of cells isolated from ear skin confirmed that *Flgft* skin contained significantly increased percentages of CD4+ T cells and Gr-1+ neutrophils, but not CD11c+ dendritic cells, compared with ear skin from controls (Moniaga, et al., 2010, Oyoshi, et al., 2009).

lymphocytes, eosinophils, and mononuclear cells in *Flgft* mice compared to control mice. These data support the diagnosis of AD-like dermatitis in *Flgft* mice in the steady state under

Fig. 4. Hematoxyllin and eosin (H&E)-stained sections in 8- and 18-week old mice. Scale

Therefore, there exist discrepancies among the results of four recent papers on the cutaneous manifestation in the steady states. It seems to be related to the presence or absence of the *ma* mutation and/or variation in the genetic backgrounds of the different strains used, and to environmental factor. It has been reported that Japan carries a higher morbidity of AD than other countries (1998, Williams, et al., 1999), possibly due to environmental factors such as pollen. Because barrier dysfunction is a common characteristic of AD (Elias, et al., 2008, Nomura, et al., 2007, Palmer, et al., 2006), TEWL is commonly measured as an indicator of barrier function (Gupta, et al., 2008). TEWL was significantly higher in *Flgft* mice than in B6 mice from an early age (4 weeks) to an older

Fig. 5. TEWL through dorsal skin of 5-, 8-, and 16-week-old B6 and *Flgft* mice.

Flowcytometry analysis of cells isolated from ear skin confirmed that *Flgft* skin contained significantly increased percentages of CD4+ T cells and Gr-1+ neutrophils, but not CD11c+ dendritic cells, compared with ear skin from controls (Moniaga, et al., 2010, Oyoshi, et al., 2009).

SPF conditions.

bar, 100µm

age (16 weeks) (Fig. 5) (Moniaga, et al., 2010).

The extent of severity of AD is known to be correlated with elevated serum IgE levels (Novak, 2009). Serum IgE and IgG1 levels in *Flgft* mice were significantly higher than those in control mice in the steady state under SPF conditions (Moniaga, et al., 2010, Oyoshi, et al., 2009). In addition, the numbers of CD4+ and CD8+ cells in the skin draining LNs in *Flgft* mice were significantly higher than those in control mice, but those of the spleen were similar for both groups. Thus, an enhanced cutaneous immune reaction seems to be induced in *Flgft* mice due to the condition of their skin induced by filaggrin and/or matted deficiency.

AD is thought to be mediated by helper T cell subsets, such as Th1, Th2, and Th17 (Bieber, 2008, Hattori, et al., 2010, Koga, et al., 2008). In the steady state, the skin of *Flgft* mice showed no difference of Th1 cytokine IFN-γ and Th2 cytokines IL-4 and IL-13 compared to the control. In contrast, there is a significant increase in mRNA expression of the Th17 cytokine IL-17, IL-17 promoting cytokines IL-6 and IL-23 (p19), and IL-17 inducible neutrophil attractant chemokine CXCL2 in *Flgft* mice (Moniaga, et al., 2010, Oyoshi, et al., 2009).
