Vol.47, No.11 November 2004

CONTENTS

Skin Diseases

Development of Skin Measurement Instruments

Hachiro TAGAMI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495

Animal Models of Atopic Dermatitis

Hitoshi MIZUTANI et al. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501

The Molecular Basis of Keratinizing Disorders

Motomu MANABE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508

Updates on Autoimmune Skin Bullous Diseases

Masayuki AMAGAI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514

Latest Information on Alopecia

Satoshi ITAMI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520

Recent Progress in Diagnosis and Treatment of Melanoma

Toshiro KAGESHITA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 524

Postherpetic Neuralgia

Treatment of Postherpetic Neuralgia

Akira OZAWA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529

JMAJ, November 2004—Vol. 47, No. 11 495

This article is a revised English version of a paper originally published inthe Journal of the Japan Medical Association (Vol. 129, No. 9, 2003, pages 1405–1408).

Introduction

Gross observation has been and is an essen-tial part of the examination of skin in ourroutine practice. Descriptive dermatology hasa long history, and experienced dermatologistsare usually reputed for their capability in iden-

Development ofSkin Measurement InstrumentsJMAJ 47(11): 495–500, 2004

Hachiro TAGAMI

Professor Emeritus, Department of Dermatology,Tohoku University School of Medicine

Abstract: The observation of skin changes has long been conducted with anemphasis on the visually identifiable and palpable lesions. However, because ofthe limitation of sensory evaluation, efforts have been made during the last30 years toward the introduction of various types of instrumental measurement.Skin conditions that have conventionally been categorized as normal can now beclassified into various types on the basis of numerical values. Achievements fromsuch studies are widely utilized not only for the treatment of diseases, but also inthe field of cosmetics. Of particular importance is the measurement of the barrierfunction of the stratum corneum in the outermost layer of the skin, as well as itswater content. Such measurement enables us to perform numerical assessment ofskin irritation and abnormal cornification. It also facilitates the quantitative evalua-tion of the action of topical drugs in softening and smoothing the skin surface. Skincolor, surface topography, and stiffness can also be evaluated numerically. Themagnified observation of the skin assists the differential diagnosis of malignanttumors. Recent developments are enabling us to perform in-situ non-invasiveobservation of the internal structures of the skin, eliminating the need for invasivebiopsy and histopathological studies to some extent.

Key words: Barrier function; Biophysical measurement; Hydration state;Instrumental measurement; Skin; Stratum corneum

tifying subtle changes in the patient’s skin.Typically, the presence of a skin rash is obvi-ous. A dermatologist makes a diagnosis byobserving the characteristic morphology of theskin lesion, applies appropriate topical drugs,and confirms the disappearance of macro-scopic lesions. The whole process of treatment

� Skin Diseases

496 JMAJ, November 2004—Vol. 47, No. 11

stratum corneum all over the human bodysurface is the basic need to retain water in thebody living in a dry environment and wardingoff the invasion of microbes and other harmfulsubstances.

As compared with the healthy skin, diseasedskin looks whitish and is rough and dry tothe touch. The stratum corneum of the skinshowing such pathological appearance lackssufficient barrier functions. Because it loseswater to dry air and becomes hard and brittle,cracks develop on the skin surface as a resultof body movement. This leads to the formationof hard, thick lumps of horny tissues calledscales. From the standpoint of maintaining abeautiful, healthy skin, the moisture contentof the stratum corneum is one of the majorfactors affecting the properties of both healthyand diseased skin, and this tissue is consideredto be one of the most important targets ofinstrumentation in dermatology. In addition,this measurement is important to evaluate theeffectiveness of topical drugs and skin carecosmetics.

1. Measurement of the barrier function ofstratum corneum

Recently, it has become possible to examinein situ the barrier function of the stratumcorneum using a non-invasive method calledtransepidermal water loss (TEWL) measure-ment. This method uses an electric hygrometerto measure the amount of water being lostslowly from the body through the stratumcorneum.

Because the measurement is affected byperspiration, it is performed at an ambienttemperature of 22°C or less and under condi-tions that cause no sweating. The skin withinflammatory conditions forming scales consist-ently shows high values of TEWL reflectingthe loss of barrier function. A study using thismeasurement has revealed a loss of barrierfunction in the skin damaged by detergents, aswell as in atopic xerosis (areas of skin in atopicdermatitis patients that are simply dry but have

H. TAGAMI

is performed with the naked eye. The use ofinstruments has traditionally been limited tothe use of a magnifier by presbyopic physiciansand the use of a microscope for mycologicalexamination.

However, the field of dermatology is notlimited to overt pathologic changes, but alsoshares a number of elements with cosmeticfield. Even the skin of healthy individualsshows subtle variations in tone and the degreeof softness, which are very difficult to describeverbally. Needless to say, the characteristics ofdiseased skin are much more diverse.

When we measure various characteristics ofthe skin using appropriate instruments, we findthat the normal skin shows wide variationsthat are difficult to discern with the naked eye.It is now clear that there are even “invisibledermatoses,” which are conditions detected onlyby the use of instruments.1) What we see withthe naked eye is not everything.

Reflecting this situation, there has been amovement toward the introduction of mea-surement instruments in dermatology and thedevelopment of skin biometry in the past quar-ter of a century. Specialized academic societieshave been established and dedicated scientificjournals have come into publication. Nowadays,more and more instruments are being used inour daily practice.

Measurement ofSkin Surface Properties

The outermost layer of the skin consistsof the stratum corneum, which functions as abarrier preventing the free passage of sub-stances and maintains the suppleness of theskin. The stratum corneum is composed ofcorneocytes, which are dead epidermal keratino-cytes commonly called scurf. The spaces amongthese cells are filled with intercellular lipidsarranged in a special lamellar structure. Thestratum corneum as a whole is a biologicallyproduced membranous structure less than 20�mthick. The reason for the presence of the

JMAJ, November 2004—Vol. 47, No. 11 497

DEVELOPMENT OF SKIN MEASUREMENT

no visible lesions).2) The situation differs in thexerosis of elderly people. While inflammatoryskin lesions involve the formation of patholog-ical stratum corneum and appearance of scalesas a result of enhanced epidermal metabolism,the xerosis of elderly people is characterizedby the presence of intact barrier function, andthe dryness results from the lack of moisturesupply from inside after moisture is lost to thedry air of winter. Despite the similar drynessin clinical appearance, the latter is character-ized by the thick accumulation of the stratumcorneum and reduced TEWL.3)

On the other hand, fresh scars, hypertrophicscars, and keloids with inflammation and fibroustissue proliferation in deep layers of the dermisshow a reduction of barrier function, even ifno scales are observed. This is also the casewith the topical applications of retinoic acidfor photo-aging in the exposed skin areas ofelderly patients and the oral use of retinoidsfor dyskeratotic dermatoses.4) Naturally, normalskin in different parts of the body shows dif-ferent values of TEWL. The skin of the faceand the external genitalia present high valuesof TEWL approximating those in skin lesionsin other locations.5)

2. Measurement of hydration state ofstratum corneum

Normal stratum corneum binds water andperforms the important hydration (moistureretention) function for maintaining the softnessand smoothness of the skin and allowing theunrestricted movement of the skin. A loss ofthis function may result in dryness, rough skin,and even the scales and cracks that frequentlyaccompany skin diseases. Thus, the hydrationstate of the stratum corneum is a very impor-tant measure both for cosmetic scientists, whoattempt to design fundamental cosmetics thatwould enhance the water content of a normalstratum corneum, and for dermatologists, whotreat skin diseases by the use of topical drugsincreasing the water content of an abnormalstratum corneum.

In 1980, the authors6) reported that we couldperform in-vivo measurement of the hydrationstate of the stratum corneum surface instantlyand non-invasively by the measurement of thehigh-frequency conductance or electric capaci-tance. Since then, many types of instrumentsbased on this principle have been developedand used widely for diagnosis and the develop-ment of moisturizing agents and other topicaldrugs.

Along with the measurement of the barrierfunction of the stratum corneum, that of thehydration state of the stratum corneum providesthe basis for the functional analysis of thestratum corneum. This measurement also shouldbe performed at an ambient temperature of22°C or less and a relative humidity of about50% to avoid the influence of perspiration.

Among the parts of the body, the hydrationvalue tends to be high in the face and the upperparts of the body, where sebum secretion is high,and low in the lower limbs.5) Elderly peoplewho show a loss of hydration function of thestratum corneum often suffer from dryness inthe lower limbs in winter, and this may lead tosenile xerosis characterized by itchiness due tosurface cracking.3) Normal individuals withoutskin rashes also show lower barrier function andhigher TEWL of the stratum corneum in winterthan in summer, when the same individuals areexamined under the same conditions.7) Thismeans that the skin in winter is more prone todermatitis and aggravation of conditions suchas atopic dermatitis.

The water content of the stratum corneumdecreases when there are scales or crusts andincreases promptly after the application ofmoisturizing agents. The oral use of retinoidfor abnormal cornification and the topical useof retinoic acid for acne and photo-agingincrease the softness of the skin within 1 or 2weeks.4) The afore-mentioned keloids and freshscars also show similar changes associated withthe increase in the water content of the stratumcorneum.

498 JMAJ, November 2004—Vol. 47, No. 11

3. Measurement of skin surface topographyThe most widely used method for detailed

observation of skin surface topography seemsto be the replica (microrelief) method. Thesurface topography of the skin is transferred toa soft nitrocellulose or silicone rubber material,transferred further from it to a resin or otherappropriate material, and then observed underthe light or electron microscope. Attemptshave been made to visualize the surface topog-raphy in 2-dimensional representation and toachieve 3-dimensional reconstruction.

At present, specimens are usually observedunder oblique light and the resulting shadowsare read into an image analyzer connected to acomputer, which calculates the area of detectedshadows. For example, in Paget’s disease andmycosis fungoides, a type of T lymphoma, wecan accurately define the borderline betweenthe margin and normal tissues based on thedifference in the form of minute wrinkles.8)

Measurement of Skin Color

The measurement of the redness, whiteness,and pigmentation of the skin has been showingextremely remarkable progress. One of themost notable developments is Minolta ChromaMeter®, which has the capability to analyzeskin tone based on hue, lightness, and chroma.With the convenience for routine use, it is fairlyclose to the ideal of instruments of this type.The facial skin of Japanese people shows adecrease in lightness (L*) and an increase inyellowness (b*) with the advancement of age.

Magnified Observation of Skin Surface

The observation of the skin surface undera magnification of about 20� using a portableoptical device called a dermatoscope facilitatesthe evaluation of superficial tumors, pigmentlesions, and blood vessel network. It improvedthe accuracy of clinical diagnosis of seborrheickeratosis, solar lentigo, pigmented mole, basal

cell carcinoma, and malignant melanoma,which often present difficulties in differentialdiagnosis.9)

Non-Invasive In-Vivo Observation ofMicroscopic Skin Structures

While histopathological examination of theskin can be performed by simple procedures,it has a problem of scars remaining on the skin.The use of instruments enables us to conductin-situ observation of microscopic changes inskin tissues. The construction of the skin can beexamined by the use of high-frequency ultra-sound. The numerical data of the hypertrophyand atrophy of the skin and the tumor locationmay be useful in assessing the extent of lesionsor the effectiveness of treatment.

With respect to non-invasive observation ofsmaller structures, confocal laser microscopesare useful to examine the shallower layers ofthe skin, i.e., the epidermis and the papillarylayer of the dermis. This method can provideinformation on the nature and thickness ofcorneocytes, nature of epidermal cells, andhemodynamics in capillary vessels.10) However,in view of the fact that we have been accus-tomed to the observation of artifacts in HEstained histopathological specimens, the valueof this method in the diagnosis of diseases hasyet to be established.

Identification of Substances inSkin Tissues

Microdialysis is a method in which physio-logical saline is injected via a thick injectionneedle inserted intradermally or subcutaneouslywhile the tissue fluid is recovered via anotherneedle to analyze and identify substancesoccurring in it. This method enables us toanalyze various substances in the tissue fluidsampled in real time and to study the roles andmetabolism of these substances in normal andpathological skin.11)

H. TAGAMI

JMAJ, November 2004—Vol. 47, No. 11 499

Measurement of Viscoelasticity andStiffness of Skin

Cutometers are commercially available devicesthat comprehensively measure the mechanicaland physical properties of the skin, such asextensibility and viscoelasticity. These instru-ments apply suction to the skin and measureits displacement and time to recovery. Becausechanges in the skin detected by this methodreflect not only the hydration state of thestratum corneum, but also the thickness ofepidermis, the extent of fibrous changes, thestate of the dermal matrix, etc., it is necessaryto understand the characteristics of each deviceand select appropriate models to suit the pur-pose of the study.

We recently developed a method of measur-ing skin stiffness with robotic tactile sensorsand analyze it into superficial stiffness due todrying of the stratum corneum and deep stiff-ness reflecting fibrotic changes in the dermis.12)

Conclusion

Subjective observation that can be recognizedonly by the observer himself lacks reproducibilityand reliability. The instrumental measurementof the living skin has been developed based onthe idea of analytical science to understandskin conditions through analyzing instrumentalmeasurement data. The application of biom-etry technology to the skin has enabled us tograsp numerically the properties of the skinthat cannot be evaluated by the human senses.This approach has become essential to theunderstanding of not only the changes causedby skin diseases, but also the characteristics ofnormal skin reflecting age, sex, and anatomicallocation. With respect to normal individuals,we need to pay particular attention to the skinof elderly people, who represent an increasinglylarge part of our society. It is hoped that thisapproach will facilitate the measurement ofthe age-related changes in the skin and thescientific evaluation of the efficacy of treatment

DEVELOPMENT OF SKIN MEASUREMENT

for aged skin.

REFERENCES

1) Kligmian, A.M.: The invisible dermatoses.Arch Dermatol 1991; 127: 1375–1382.

2) Watanabe, M., Tagami, H., Horii, I. et al.:Functional analyses of the superficial stratumcorneum in atopic xerosis. Arch Dermatol1991; 137: 1689–1692.

3) Hara, M., Kikuchi, K., Watanabe, M. et al.:Senile xerosis: functional, morphological, andbiochemical studies. J Geriatr Dermatol 1993;1: 111–120.

4) Suetake, T., Sasai, S., Zhen, Y.X. et al.: Func-tional analyses of the stratum corneum inscars. Sequential studies after injury and com-parison among keloids, hypertrophic scars,and atrophic scars. Arch Dermatol 1996; 132:1453–1458.

5) Ya-Xian, Z., Suetake, T. and Tagami, H.:Number of cell layers of the stratum corneumin normal skin-relationship to the anatomicallocation on the body, age, sex and physicalparameters. Arch Dermatol Res 1999; 291:555–559.

6) Tagami, H., Ohi, M., Iwatsuki, K. et al.: Evalu-ation of the skin surface hydration in vivoby electrical measurement. J Invest Dermatol1980; 75: 500–507.

7) Kikuchi, K., Kobayashi, H., Le Fur, I. et al.:The winter season affects more severely thefacial skin than the forearm skin: Comparativebiophysical studies conducted in the sameJapanese females in late summer and winter.Exog Dermatol 2002; 1: 32–38.

8) Kikuchi, K., Aiba, S., O’Goshi, K. et al.:Demonstration of characteristic skin surfacecontours of extramammary Paget’s diseasearid parapsoriasis en plaque by image analysisof negative impression replicas. J Dermatol Sci2002; 30: 20–28.

9) Soyer, H.P., Smolle, J., Leitinger, G. et al.:Diagnostic reliability of dermoscopic criteriafor detecting malignant melanoma. Derma-tology 1995; 190: 25–30.

10) Huzaira, M., Rius, F., Rajadhyaksha, M. et al.:Topographic variations in normal skin, asviewed by in vivo reflectance confocal micro-scopy. J Invest Dermatol 2001; 116: 846–852.

11) Rukwied, R., Zeck, S., Schmelz, M. and

500 JMAJ, November 2004—Vol. 47, No. 11

McGlone, F.: Sensitivity of human scalp skinto pruritic stimuli investigated by intradermalmicrodialysis in vivo. J Am Acad Dermatol2002; 47: 245–250.

12) Sasai, S., Ya-Xian, Z., Suetake, T. et al.:

H. TAGAMI

Palpation of the skin with a robot finger: anattempt to measure skin stiffness with a probeloaded with a newly developed tactile vibra-tion sensor and displacement sensor. Skin ResTechnol 1999; 5: 237–246.

JMAJ, November 2004—Vol. 47, No. 11 501

Animal Models of Atopic DermatitisJMAJ 47(11): 501–507, 2004

Hitoshi MIZUTANI*, Takeshi NISHIGUCHI**, and Takaaki MURAKAMI***

***Professor, Department of Dermatology, Mie University, Faculty of Medicine***Assistant Professor, Department of Dermatology, Mie University, Faculty of Medicine***Assistant Professor, Department of Dermatology, Mie University Hospital

Abstract: Atopic dermatitis (AD) is a chronic, recurrent eczematous skin diseasewith itching. Many AD patients have a family history of asthma, allergic rhinitis, anda tendency to IgE overproduction. AD is recognized as a form of persistent antigen-specific dermatitis with a Th2 cytokine profile. Various antigens induce AD, butuniformity of their clinical manifestations suggest the involvement of a common non-antigen-specific pathway. We have generated a keratinocyte-specific caspase-1transgenic mouse (KCASP1Tg) using the keratin 14 promoter. KCASP1Tg over-secreted IL-1alpha/beta and IL-18 from the skin and developed persistent itchingdermatitis with marked serum IgE and histamine elevation at age 8 weeks inspecific pathogen-free conditions. We also produced a keratin 14-driven matureIL-18 transgenic mouse (KIL-18Tg). KIL-18Tg developed similar AD-like skinchanges with IgE elevation at 24 weeks. KCASP1Tg crossed with stat6-/- devel-oped skin manifestations at age 8 weeks without IgE elevation; however,KCASP1Tg crossed with IL-18-/- did not develop dermatitis. KCASP1Tg crossedwith IL-1alpha/beta-/- developed dermatitis at age 24 weeks. Thus, epidermalIL-18 secretion induces AD-like dermatitis, and epidermal IL-1s enhances AD-likechanges. Both KCASP1Tg and KIL-18Tg showed Th2-type cytokine profiles andsatisfied the human AD criteria. These mice models indicated the involvement ofan antigen-independent innate-type pathway in addition to the antigen-specificacquired-type pathway in AD. These models responded to various therapeuticagents for AD, and are ideal, potent tools for the development of gene therapy andother new therapies for AD.

Key words: Atopic dermatitis; Interleukin 18 (IL-18); Mast cell; Mouse;Caspase

Introduction

According to the definition by the Japanese

This article is a revised English version of a paper originally published inthe Journal of the Japan Medical Association (Vol. 129, No. 9, 2003, pages 1409–1413).

Dermatological Association (JDA), atopicdermatitis (AD) is “a chronic and recurrenteczematous skin disease with itching; many AD

� Skin Diseases

502 JMAJ, November 2004—Vol. 47, No. 11

mouse has been found to show an elevatedplatelet factor 4 and eotaxin, an eosinophilchemotactic factor, in the skin.3) However, thegenetic background of this mouse has not beenestablished. While backcrossing with Th2-dominant BALB/c mice results in the develop-ment of symptoms, backcrossing with Th1-dominant mice such as C57BL/6 results in lowoccurrence of dermatitis. Because of theseproblems, it has been necessary to develop newgenetically modified AD mouse models.

Clinically, AD has been regarded as a per-sistant antigen-specific eczematous reaction toenvironmental antigens such as food antigensand mites. The pathogenesis of AD is believedto involve Langerhans cells and activated Tcells. The overproduction of IgE antibodies inAD has led researchers to pay attention to theroles of B cells, mast cells, and basophils, andAD has been considered a Th2-type diseaseinvolving cytokines such as IL-4, IL-10, IL-13,and IL-5. Although various antigens induceAD, the uniformity of their clinical manifesta-tions suggests the presence of a non-antigenspecific pathway in the pathogenesis of AD.

The authors produced a mouse model thatshowed continuous secretion of proinflamma-tory cytokines in the skin,5) which developedAD-like dermatitis under SPF conditions.6) Weidentified IL-18 as a cytokine involved in ADpathogenesis,7) and proposed a new concept ofAD.8)

Skin-Specific Caspase 1 TransgenicMouse (KCASP1Tg)

Inflammation is believed to involve pro-inflammatory cytokines including IL-1�, IL-1�,tumor necrosis factor (TNF) �, and IL-18.These substances are produced in the cell asinactive precursors and released from the cellafter enzymatic activation. The IL-1�-convertingenzyme that activates IL-1� was later found tobe coded by a member of a gene family relatedto apoptosis. This enzyme, named caspase, is acysteine protease that mediates proteolysis at

H. MIZUTANI et al.

patients have a family history of atopic pre-disposition (bronchial asthma, AD or allergicrhinitis) and a tendency to IgE overproduc-tion.”1) Understanding AD pathology is verydifficult because it involves immediate typeallergic reactions despite its basically eczema-tous reaction.

The clinical efficacy of topical steroids hasbeen remarkable, but the uncritical use of cor-ticosteroids has resulted in frequent adverseeffects. A flood of incorrect information thatensued created mistrust of corticosteroids ther-apy, which continues to hinder the propertreatment of AD. In this situation, animalmodels of AD are essential for answering ques-tions about AD and developing new therapies.The mouse is the most suitable species for thispurpose since it is a mammal, the reproductioncycle is short, the immunological backgroundof this species has been established, and diag-nostic reagents are available.

Several AD mouse models have been usedfor years. Mutant mouse models include theNc/Nga mouse,2) which is constantly infestedby mites on the skin, and the NOA (NarutoResearch Institute Otsuka Atrichia) mouse.3)

Other models have been produced by repeatedsensitization with antigens.4) The production ofa sensitization model takes a long time, and canbe achieved only in a few strains of mice such asBALB/c. These mice have a different geneticbackground from commonly used geneticallymanipulated mice based on C57BL/6 mice,and this causes difficulty in cross breedinganalysis. The genetic background of Nc/Ngamice has not been well characterized, and thesemice do not develop dermatitis under specificpathogen-free (SPF) conditions because derma-titis in this model depends on the presence ofmite infestation.2) Thus, it is difficult to usethese mice to develop therapeutic drugs.

The NOA mouse, a hairless strain created bymutation, is considered an AD model becausethe mouse develops ulcerating skin lesionsmainly on the trunk, shows scratching behavior,and presents elevated serum IgE levels. This

JMAJ, November 2004—Vol. 47, No. 11 503

The frequency of scratching behavior in-creased remarkably with the onset of the skinlesions. In week 10 and after, the frequency ofscratching per 10 minutes was 10 times as highas that in normal mice (Fig. 2b).

Histopathological Changes inModel Mice

The skin before the onset of skin symptomsshowed no histopathological changes. Afteronset, it presented remarkable epidermotropiccell infiltration, papillomatous proliferation andthickening of the epidermis, partial epidermaldefects, crusts, and parakeratosis. These changesresembled those observed in the acute stage ofhuman atopic dermatitis (Fig. 1). The dermisshowed an increase in the number of infiltrat-ing CD4� T cells and a remarkable increase intoluidine blue-positive mast cells, similar to thefindings in AD6) (Fig. 2c).

ANIMAL MODELS OF ATOPIC DERMATITIS

an aspartate site.Monocytes produce caspase 1, which pro-

motes the activation and secretion of IL-1� andIL-18. While epidermal keratinocytes producepro-IL-1� and IL-18, they do not secrete thesecytokines because they normally lack caspase 1activity.9) We attached a keratin14 promoter tohuman caspase 1 DNA segments and intro-duced the DNA into fertilized eggs so thatcaspase 1 would be expressed only in the epi-dermal basal cells.5)

Clinical Symptoms in Model Mice

About 8 weeks after birth, the mice started toshow dermatitis around the eyes and neck. Thelesions extended to the ears and legs and finallybecame generalized. The dermatitis started aserosive erythematous patches and moist crustedareas. After repeated re-epithelization andinflammation, the lesions progressed to chroniclichenified plaques (Fig. 1). Almost all the micedeveloped cataract, as is the case with AD.

KCASP1Tg KIL-18Tg WT

Fig. 1 Clinical and histopathological findings in KCASP1Tg and KIL-18Tg

504 JMAJ, November 2004—Vol. 47, No. 11

IL-5, which are not detected in normal mice,and IFN-� was not detected. The increase inIL-4 production was confirmed with an enzyme-linked immunosorbent assay (ELISA) for theculture condition medium of the lesions.

In addition, we measured the cytokine pro-duction in splenic cells after stimulation withanti-CD3 antibodies. We detected elevation ofIL-3, IL-4, and IL-5, as well as the suppressionof IFN-�, indicating a systemic Th2 shift ofcytokines. These mice showed an elevation ofCD40L expression, implying the promotion ofIgE production from B cells.

Appropriateness as a Model ofAtopic Dermatitis

Because the definition of atopic dermatitis isbased on clinical findings in humans, it is some-what controversial whether the same criteriacan be applied to mice. The definition bythe Japanese Dermatological Association, i.e.,recurrence, itching eczema, family history, andelevated serum IgE levels, applies to our mousemodel.1)

Serological Changes in Model Mice

Serum histamine levels correlate with theseverity of AD in humans. The serum histaminelevels in the model mice started to increasebefore the onset of symptoms, and it increasedaccording to the extension of the skin lesions(Fig. 2a). Serum histamine levels correspondedwith an increase in mast cells in the skin. SerumIgE levels were as low as that of wild-type mice.It increased by the time of symptom onset,and showed remarkable elevation to 50�g/mlor more, correlating with the severity of theskin rash (Fig. 2a). IgG1 increased remarkably,however, increases in IgG2 and IgM werelimited, suggesting the allergic nature of theobserved changes.

Cytokine Production andImmunological Profile of Model Mice

The sera of the model mice were found tocontain high levels of active IL-1� and IL-18.The expression of cytokine mRNAs in theaffected skin showed an elevation of IL-4 and

H. MIZUTANI et al.

Fig. 2 Immunological profiles of KCASP1Tg

0

25

50

75

100

0

500

1,000

1,500

2,000

2,500

100 2 4 6 8 1612 14Age: weeks-old

IgE (�g/ml)Histamine (nM)

a: Clinical course and serum IgE and histamine levels in KCASP1Tg

c: Number of mast cells in affected dermis in KCASP1Tg (PCS/40F)

b: Scratching times per 10 min in KCASP1Tg

Dermatitis

40

30

20

10

0

0

60

40

20

WT KCASP1Tg

WT KCASP1Tg

JMAJ, November 2004—Vol. 47, No. 11 505

Our model satisfies all the essential elementsof the internationally accepted diagnostic cri-teria by Hanifin-Rajka,10) i.e., (1) itching, (2)typical distribution of skin rash, (3) chronicrecurrent dermatitis, and (4) family history.At least three of these major criteria arerequired. Our model satisfies 11 of 22 minorelements, including xeroderma, elevated serumIgE, early onset, dermatitis of the hands andfeet, cheilitis, conjunctivitis, cataract, facialerythema, lesions around the eyes, and aggra-vation due to environmental changes. We couldnot detect environmental antigen specific IgEin their sera.

As discussed above, these mice are qualifiedas an animal model of AD.

Onset Mechanism of AD Studied inModel Mice

There has been a long debate as to whetherthe onset of AD involves type I or type IVallergy. However, it has been accepted that thedistinction between type I and type IV does nothave much significance in this context. The roleof IgE and the mechanism for IgE elevation arenevertheless important issues.

KCASP1Tg mice produce IgE withoutantigen-specific reactions, and the skin of thesemice secretes IL-1 and IL-18. IL-18 was origi-nally discovered as a factor inducing IFN-� inhepatitis, and thus had been considered a Th1cytokine. This assumption contradicts the symp-toms seen in these mice.5) To answer this ques-tion, we injected recombinant IL-18 into wild-type mice, and observed the elevation of IgE.The results indicated that IL-18 is a Th2 cyto-kine and suggest the possibility that Th2 induc-tion in these mice may be mediated by IL-18.7)

On the other hand, the class switching andproduction of IgE essentially requires signaltransduction via the stat6 system locating thedownstream of IL-4. We therefore crossedKCASP1Tg with stat6 knockout mice. Becausethis resulted in the complete suppression ofIgE production, we concluded that the IgE

induction by IL-18 is mediated by stat6. On theother hand, we crossed our mice with IL-18knockout mice to see whether IgE productionin KCASP1Tg depends on IL-18. This experi-ment showed that the IgE level in KCASP1TgIL-18-/- mice is as low as one-tenth of thereference IgE level. These mice possessed IL-4,suggesting that there are minor IgE inductionpathways other than those involving IL-18.

While the stat6-dependence of IgE produc-tion was confirmed, this study produced thesurprising results that the stat6-/-KCASP1Tgmice lacking IgE developed similar dermatitiswith the same timing as the KCASP1Tg mice.Furthermore, the IL-18-/-KCASP1Tg micedid not develop dermatitis.

The above findings indicate that IgE is notessential to the development of dermatitis, andIL-18 plays a key role.

IL-18 Transgenic Mice

Next, it was necessary to examine the in vivofunction of epidermal IL-18. We producedepidermis-specific mature IL-18Tg (KIL-18Tg)for this purpose. While KCASP1Tg developedsymptoms within 8 weeks, KIL-18Tg took 6months before the onset of symptoms.8) Thesemice developed lichenification of the skin fromthe early stage of the skin rash (Fig. 1).

Because KCASP1Tg mice also secrete IL-1�, we crossed the model mice with IL-1�/�knockout mice to examine the effect of IL-1�.While the IL-1s-/-KCASP1Tg mice presenteda similar phenotype to the KCASP1Tg mice,they resembled KIL-18Tg mice in that 6 monthswas required before the onset of symptoms.These observations suggest that IL-18 is essen-tial for the onset of skin rash and that IL-1s actsas an onset booster.

Suggestions from AD Model Studiesand Future Prospects

The model mice described above introduceda new concept of AD. High IgE levels and

ANIMAL MODELS OF ATOPIC DERMATITIS

506 JMAJ, November 2004—Vol. 47, No. 11

RAST (radioallergosorbent test) scores havebeen emphasized in the clinical management ofAD. These indices sometimes cause excessivedietary restriction and limitation on beddingand clothing, and these may be exploited as anopportunity for home reform businesses. Inaddition to the conventional antigen-specificmechanism, it is possible that epidermal dam-age due to scratching promotes the release ofepidermal proinflammatory cytokines (in par-ticular, IL-18), that activates mast cells, andresults in aggravation of AD.

We proposed a new concept that the part ofAD caused by environmental antigen-specificeczema reaction as “acquired-type AD” andthe part consisting of non-antigen specificdermatitis seen in these mice as “innate-typeAD.”8) In Europe, the term “extrinsic AD” wasproposed to describe AD showing antigen-specific IgE antibodies and the term “intrinsicAD” for AD lacking these antibodies, and thelatter has been reported to be increasingrecently.11) In our daily practice, about 10 to20% of AD cases do not have high serum IgElevels. It is likely that these cases are affectedby epidermal injury due to scratching causedby psychological factors or ichthyosis.

Therapeutically, KCASP1Tg mice respondedwell to topical steroids and steroid injections.These mice showed disappearance of erosionsand remarkable improvement in dermatitis.The effectiveness of the topical use of anti-allergic drugs and immunosuppressive agentswas also confirmed. Because KCASP1Tg miceare tested in SPF conditions, they are expectedto be ideal tools for the development of newdrugs and the evaluation of existing drugs.

The past pococurante use of corticosteroidsand the flood of incorrect information aboutadverse effects once caused much confusionabout the treatment of AD. Owing to theefforts of the JDA to promote correct infor-mation, we have overcome most of the mis-conceptions. The development of gene therapyand other new future therapies will requireanimal models. These two types of model

H. MIZUTANI et al.

mouse discussed here are expected to opennew possibilities in the development of newAD therapies.Acknowledgement:

This study was conducted in cooperationwith Professor Kenji Nakanishi of Hyogo Col-lege of Medicine.

This work was supported in part by a Grant-for-Aid Core Research for Evolutional Scienceand Technology and Grant-in-Aid for ScientificResearch by JSPS.

REFERENCES

1) Kawashima, M., Takikawa, M., Nakagawa, H.et al.: “Guidelines for Treatment of AtopicDermatitis” ed. by Japanese DermatologicalAssociation. Journal of Japanese Dermato-logical Association 2000; 110: 1099–1104. (inJapanese)

2) Vestergaard, C., Yoneyama, H., Murai, M. etal.: Overproduction of Th2-specific chemo-kines in NC/Nga mice exhibiting atopicdermatitis-like lesions. J Clin Invest 1999; 104:1097–1105.

3) Watanabe, O., Tamari, M., Nakamura, Y. andShiomoto, Y.: A search for causative genes ofatopic dermatitis. Journal of Japanese Derma-tological Association 1999; 109: 1743–1747. (inJapanese)

4) Kitagaki, H., Ono, N., Hayakawa, K. et al.:Repeated elicitation of contact hypersensi-tivity induces a shift in cutaneous cytokinemilieu from a T helper cell type 1 to a T helpercell type 2 profile. J Immunol 1997; 159: 2484–2491.

5) Yamanaka, K., Tanaka, M., Tsutsui, H. et al.:Skin-specific caspase-1-transgenic mice showcutaneous apoptosis and pre-endotoxin shockcondition with a high serum level of IL-18.J Immunol 2000; 165: 997–1003.

6) Yamanaka, K., Konishi, H., Murakami, T. etal.: IL-18 and mouse models of atopic derma-titis. Allergology (Kagaku-hyoronsha) 2002;13: 500–504. (in Japanese)

7) Yoshimoto, T., Mizutani, H., Tsutsui, H. et al.:IL-18 induction of IgE: dependence on CD4�T cells, IL-4 and STAT6. Nat Immunol 2000; 1:132–137.

JMAJ, November 2004—Vol. 47, No. 11 507

8) Konishi, H., Tsutsui, H., Murakami, T. et al.:IL-18 contributes to the spontaneous develop-ment of atopic dermatitis-like inflammatoryskin lesion independently of lgE/stat6 underspecific pathogen-free conditions. Proc NatlAcad Sci USA 2002; 99: 11340–11345.

9) Mizutani, H., Black, R. and Kupper, T.S.:Human keratinocytes produce but do not pro-cess pro-interleukin-1 (IL-1) beta. Differentstrategies of IL-1 production and processing

in monocytes and keratinocytes. J Clin Invest1991; 87: 1066–1071.

10) Hanifin, J.M. and Rajka, G.: Diagnostic fea-tures of atopic dermatitis. Acta Dermatovener1980; S92: 44–47.

11) Schafer, T., Kramer, U., Vieluf, D. et al.: Theexcess of atopic eczema in East Germany isrelated to the intrinsic type. Br J Dermatol2000; 143: 992–998.

ANIMAL MODELS OF ATOPIC DERMATITIS

508 JMAJ, November 2004—Vol. 47, No. 11

The Molecular Basis ofKeratinizing DisordersJMAJ 47(11): 508–513, 2004

Motomu MANABE

Professor, Department of Sensory Medicine, Division of Dermatology andPlastic Surgery, Akita University School of Medicine

Abstract: Technological advances in molecular biology have brought about anumber of major scientific achievements in the field of normal and abnormalepidermal differentiation. This article reviews the molecular defects that causemany types of keratinizing disorders, and highlights the roles of affected genesencoding a variety of epidermal proteins.

Key words: Keratinizing disorders; Pathogenic gene; Mutation

Introduction

Keratinizing disorders are conditions inwhich qualitative or quantitative abnormalitybased on a certain genetic background affectsthe pathway of epidermal differentiation fromthe birth of basal cells in the epidermis to theirexfoliation as cornified cells at the skin sur-face. This paper reviews biological aspects ofpathogenic genes encoding cytoskeletal pro-teins (keratins), desmosomal proteins (desmo-plakin, desmoglein, plakoglobin, and plako-philin), gap junction proteins (connexins),cornified envelope protein (loricrin), and cer-tain enzymes/enzyme inhibitors

Keratinizing Disorders and PathogenicGenes (Table 1)

1. Enzymes and their inhibitors(1) Lipoxygenase

Nonbullous ichthyosiform erythrodermais caused by the mutation of the lipoxy-genase-3 (ALOXE3) or the 12R-lipoxygenase(ALOX12B) gene.1) ALOXE3 and ALOX12Bare both enzymes involved in the metabolismof arachidonic acid. The loss of activity of theseenzymes results in a shortage of essential fattyacids and compromises the hydration ability ofthe horny layer.(2) Transglutaminase 1 (TGase 1)

Lamellar ichthyosis, a condition resemblingnonbullous ichthyosiform erythroderma, iscaused by the mutation of the TGase 1 gene.2)

TGase 1 catalyzes the bridging of proteins atthe �-carboxyl groups in glutamine residuesand at the lysine residues. Through thesereactions, TGase 1 bridges the components ofthe cornified cell envelope, which is a structurethat lines the cell membrane, and plays animportant role in maintaining the constructionof cornified cells. In this disease, the activity of

This article is a revised English version of a paper originally published inthe Journal of the Japan Medical Association (Vol. 129, No. 9, 2003, pages 1414–1418).

� Skin Diseases

JMAJ, November 2004—Vol. 47, No. 11 509

TGase 1 is reduced, and the ability to bridgethe cornified cell envelope precursor proteinsis insufficient. As a result, the ceramide thatnormally binds to the cornified cell envelopecomponents is accumulated in intercellular

spaces, and the decomposition of desmosomesby protease is inhibited.(3) Sarco/endoplasmic reticulum Ca2� ATPase

(SERCA)Darrier’s disease is caused by the mutation

Table 1 Molecular Basis of Keratinizing Disorders

Pathogenic genes Keratinizing disorders

KeratinsK1 palmoplantar keratoderma Unna-Thost

ichthyoisis hystrix Curth-MacklinK1/K10 bullous congenital ichthyosiform erythroderma

annular epidermolytic ichthyosisK2e bullous ichthyosis of SiemensK16 focal palmoplantar keratodermaK6a/K16 pachyonychia congenital, type 1K6b/K17 pachyonychia congenital, type 2K9 palmoplantar keratoderma VornerhHa1, hHb1, hHb6 monilethrix

Desmosomal proteinsdesmoglein 1 striate palmoplantar keratodermadesmoplakin striate palmoplantar keratodermaplakoglobin Naxos diseaseplakophilin skin fragility syndrome

Gap junction proteinsConnexin 31 Erythrokeratoderma variabilisConnexin 30 Clouston syndromeConnexin 26 Vohwinkel’s syndrome with deafness

keratitis-ichthyosis-deafness syndrome

Cornified envelope proteinloricrin Vohwinkel’s syndrome with ichthyosis

progressive symmetric erythrokeratoderma

Enzymes and their inhibitorslipoxygenase non-bullous congenital ichthyosiform erythrodermatransglutaminase 1 lamellar ichthyosisSERCA 2 Darier disease3beta-hydroxysteroid dehydrogenase CHILD syndromecathepsin C Papillon-Lefevre syndromesteroid sulfatase X-linked ichthyosisfatty aldehyde dehydrogenase Sjögren-Larsson syndromephytanoyl-CoA hydroxylase Refsum diseaseSPINK5 Netherton syndrome

OthersSLURP-1 palmoplantar keratoderma Mal de Meledadyskerin X-linked recessive dyskeratosis congenitatelomerase RNA autosomal dominant dyskeratosis congenital

Unknown18q21.3 harlequin fetus17q25 palmoplantar keratoderma Howel-Evans

THE MOLECULAR BASIS OF KERATINIZING DISORDERS

510 JMAJ, November 2004—Vol. 47, No. 11

of the sarco/endoplasmic reticulum Ca2�-ATPase type 2 isoform (SERCA2) gene.3)

SERCA2 belongs to the large family of P-typecation pumps that couple ATP hydrolysis withcation transport across membranes. SERCApumps specifically maintain low cytosolic Ca2�

concentrations by actively transporting Ca2�

from the cytosol into the sarco/endoplasmicreticulum lumen. In this disease, the level ofSERCA2b that has a normal pump function isreduced by half, and this results in the impair-ment of Ca2�-mediated signal transduction andcell adhesion.(4) 3beta-hydroxysteroid dehydrogenase

CHILD (congenital hemidysplasia with ich-thyosiform erythroderma and limb defects)syndrome is caused by the mutation of the3beta-hydroxysteroid dehydrogenase gene.4) Thisenzyme functions in the pathway for choles-terol biosynthesis. Since cholesterol is impor-tant in embryonic development controlled bythe hedgehog signaling pathway, this led to thenotion that a cholesterol precursor without thepotential to form esters might impair signalingthrough hedgehog proteins and thus explainthe dysplasias observed in CHILD syndrome.(5) Cathepsin C

Papillon-Lefèvre syndrome is caused by themutation of the cathepsin C gene.5) Cathepsin Cis a cysteine protease of the papain family andis a dipeptidyl aminopeptidase capable ofremoving dipeptides from amino terminus ofproteins. The Cathepsin C knockout mice showinsufficiency in the cytotoxic activity of CTLcells and NK cells through activation impair-ment of serine proteases called granzymes Aand B. In this disease, the immune response toinfections becomes abnormal via the samemechanism seen in the knockout mice andcharacteristically induces severe periodontitis.(6) Steroid sulfatase (STS)

X-linked ichthyosis vulgaris is caused by themutation of the STS gene.6) STS is an enzymehydrolyzing the sulfate group at the sterol ring3� position in steroid hormones and choles-terols. In this disease, the degradation of cho-

lesterol sulfate, the substrate for this enzyme,to cholesterol, does not take place, and choles-terol sulfate accumulates in the spaces betweencornified cells. As a result, cholesterol sulfateinhibits the activity of chymotrypsin-type serineproteases and hinders the decomposition ofdesmosome components.(7) Fatty aldehyde dehydrogenase (FALDH)

Sjögren-Larsson syndrome is caused by themutation of the FALDH gene.7) FALDH is acomponent of the fatty alcohol: NAD oxido-reductase enzyme complex that catalyzes thesequential oxidation of fatty alcohol to alde-hyde and fatty acid. In this disease, FALDHdeficiency causes the accumulation of fattyalcohol and wax esters in the intercellularmembrane lamella, which may disrupt theepidermal water barrier.(8) Phytanoyl-CoA hydroxylase (PAHX)

Refsum syndrome is caused by the mutationof the PAHX gene.8) PAHX is involved in the�-oxidation of phytanic acid. In this disease,PAHX deficiency leads to the accumulationof phytanic acid followed by replacement ofessential fatty acid with phytanic acid in thelipid moieties of various tissues.(9) LEKTI (lymphoepithelial Kazal-type-

related inhibitor)Netherton syndrome is caused by the muta-

tion of the SPINK5 (serine protease inhibitor,Kazal-type 5) gene, which codes LEKTI, aserine protease inhibitor.9) LEKTI is involvedin T-cell differentiation in thymic epitheliumas well as anti-inflammatory and bactericidalactivities in mucosal epithelium. DefectiveLEKTI expression may cause unbalanced Th2immune response with markedly elevated IgElevels, increased susceptibility to infection andimpaired desquamation of cornified cells.

2. Adhesion Molecules(1) Desmosome components

Striate palmoplantar keratoderma, Naxosdisease, and skin fragility syndrome are causedby mutations of the desmoplakin gene or thedesmoglein 1 gene, the plakoglobin gene, and

M. MANABE

JMAJ, November 2004—Vol. 47, No. 11 511

the plakophilin gene, respectively.10–13) Desmo-somes are intracellular adhesion apparatusesoccurring in epithelial cells and myocardialcells. The protein component of a desmosomeis composed of a group of transmembrane glyco-proteins and a group of attachment plaqueproteins binding to them on the cytoplasm side.Desmoglein 1 belongs to the former group, anddesmoplakin, plakoglobin, and plakophylinbelong to the latter. Mutations in these desmo-some components are considered to causeabnormalities in cell adhesion, but little isknown about their relation to palmoplantarkeratosis, the common clinical manifestation ofthe above-mentioned diseases.(2) Connexins

Erythrokeratoderma variabilis, Clouston syn-drome, Vohwinkel’s syndrome with deafnessand KID (keratitis-ichthyosis-deafness) syn-drome are caused by mutations of the connexin31 gene, the connexin 30 gene and the connexin26 gene, respectively.14–17)

Connexins are the main protein componentsof the gap junction. Six connexin moleculesassembled in a ring form a doughnut-shapedstructure. The tunnel at the center of this struc-ture opens and closes to adjust the transporta-tion of small molecules between adjacent cells,which play an important role in intercellularinteraction. Mutant connexins are likely to beincapable not only of forming the normal gapjunction but also of playing a crucial role inepithelial homeostasis and differentiation.

3. Cornified Cell Envelope Proteins(1) Loricrin

Both Vohwinkel syndrome with ichthyosisand progressive erythrokeratodermia arecaused by the mutation of the loricrin gene.18, 19)

Loricrin deposits on the 15-nm-thick layer ofthe cytoplasmic surface of the cell periphery(cornified cell envelope). Frameshift mutationsin the loricrin gene produce amino acidsequences that contain motifs resemblingnuclear localization signals. As a result, themutant loricrin moves into the nucleus being

associated with normal loricrin, and the forma-tion of the cornified cell envelope is impaired.

4. Cytoskeletal Proteins (Table 1)(1) Keratins

Various diseases collectively called keratindiseases are caused by the mutation of variouskeratin genes.20) Keratins are an extremelydiversified group of proteins that play animportant role in cell shape maintenance. Inkeratin diseases, the ability of keratins to main-tain cell shape is weakened, and cells are easilydestroyed by even slight external force.

5. Others(1) Dyskerin and telomerase RNA

X-linked recessive dyskeratosis congenitaand autosomal dominant dyskeratosis con-genita are caused by mutations of the dyskeringene and telomerase RNA gene, respectively.21,22)

These components are suspected to play a rolein the activity adjustment of telomerase, whichgoverns the life span of cells. However, theirfunctional significance in these diseases isunknown.(2) SLURP-1 (secreted Ly6/uPAR-related

protein 1)Mal de Meleda type of palmoplantar

keratosis is caused by the mutation of theSLURP-1 gene.23) SLURP-1 belongs to theleukocyte antigen-6 (Ly6)/urokinase-type plas-minogen activator (uPAR) protein family, andthe structure of SLURP-1 suggests its relation-ship with snake and frog toxins. SLURP-1receptors are assumed to play an importantrole in signal transduction, cell growth, celladhesion, etc., but their functional significancein this disease is unknown.

6. Diseases Being Studied for Identificationof Pathogenic Genes

The genes for harlequin fetus and pal-moplantar keratoderma Howel-Evans mapto chromosomes 18q21.3 and 17q25, re-spectively.24,25)

THE MOLECULAR BASIS OF KERATINIZING DISORDERS

512 JMAJ, November 2004—Vol. 47, No. 11

Conclusion

These findings of genes responsible forkeratinizing disorders are expected to providevaluable information in the elucidation ofpathogenic mechanisms and the search for newtreatment modalities.

REFERENCES

1) Jobard, F., Lefevre, C., Karaduman, A. et al.:Lipoxygenase-3 (ALOXE3) and 12 (R)-lipoxygenase (ALOX12B) are mutated innon-bullous congenital ichthyosiform eryth-roderma (NCIE) linked to chromosome17p13.1. Hum Mol Genet 2002; 11: 107–113.

2) Huber, M., Rettler, I., Bernasconi, K. et al.:Mutations of keratinocyte transglutaminase inlamellar ichthyosis. Science 1995; 267: 525–528.

3) Sakuntabhai, A., Ruiz-Perez, V., Carter, S.et al.: Mutations in ATP2A2, encoding a Ca(2�) pump, cause Darier disease. Nat Genet1999; 21: 271–277.

4) Konig, A., Happle, R., Bornholdt, D. et al.:Mutations in the NSDHL gene, encoding a3-beta-hydroxysteroid dehydrogenase, causeCHILD syndrome. Am J Med Genet 2000; 90:339–346.

5) Toomes, C., James, J., Wood, A.J. et al.: Loss-of-function mutations in the cathepsin C generesult in periodontal disease and palmo-plantar keratosis. Nat Genet 1999; 23: 421–424.

6) Bonifas, J.M., Morley, B.J., Oakey, R.E. et al.:Cloning of a cDNA for steroid sulfatase:frequent occurrence of gene deletions inpatients with recessive X chromosome-linkedichthyosis. Proc Natl Acad Sci USA 1987; 84:9248–9251.

7) De Laurenzi, V., Rogers, G.R., Hamrock, D.J.et al.: Sjögren-Larsson syndrome is caused bymutations in the fatty aldehyde dehydro-genase gene. Nat Genet 1996; 12: 52–57.

8) Jansen, G.A., Ofman, R., Ferdinandusse, S.et al.: Refsum disease is caused by mutationsin the phytanoyl-CoA hydroxylase gene. NatGenet 1997; 17: 190–193.

9) Chavanas, S., Bodemer, C., Rochat, A. et al.:Mutations in SPINK 5, encoding a serine pro-tease inhibitor, cause Netherton syndrome.

Nat Genet 2000; 25: 141–142.10) Rickman, L., Simrak, D., Stevens, H.P. et al.:

N-terminal deletion in a desmosomal cadherincauses the autosomal dominant skin diseasestriate palmoplantar keratoderma. Hum MolGenet 1999; 8: 971–976.

11) Armstrong, D.K., McKenna, K.E., Purkis, P.E.et al.: Haploinsufficiency of desmoplakincauses a striate subtype of palmoplantarkeratoderma. Hum Mol Genet 1999; 8: 143–148.

12) McKoy, G., Protonotarios, N., Crosby, A. et al.:Identification of a deletion in plakoglobin inarrhythmogenic right ventricular cardiomyo-pathy with palmoplantar keratoderma andwoolly hair (Naxos disease). Lancet 2000; 355:2119–2124.

13) McGrath, J.A., McMillan, J.R., Shemanko,C.S. et al.: Mutations in the plakophilin 1 generesult in ectodermal dysplasia/skin fragilitysyndrome. Nat Genet 1997; 17: 240–244.

14) Richard, G., Smith, L.E., Bailey, R.A. et al.:Mutations in the human connexin gene GJB3cause erythrokeratodermia variabilis. NatGenet 1998; 20: 366–369.

15) Maestrini, E., Korge, B.P., Ocana-Sierra, I.et al.: A missense mutation in connexin26,D66H, causes mutilating kerato-derma withsensorineural deafness (Vohwinkel’s syn-drome) in three unrelated families. Hum MolGenet 1999; 8: 1237–1243.

16) Richard, G., Rouan, F., Willoughby, C.E. et al.:Missense mutations in GJB2 encodingconnexin-26 cause the ectodermal dysplasiakeratitis-ichthyosis-deafness syndrome. Am JHum Genet 2002; 70: 1341–1348.

17) Lamartine, J., Munhoz Essenfelder, G., Kibar,Z. et al.: Mutations in GJB6 cause hidroticectodermal dysplasia. Nat Genet 2000; 26:142–144.

18) Maestrini, E., Monaco, A.P., McGrath, J.A. etal.: A molecular defect in loricrin, the majorcomponent of the cornified cell envelope,underlies Vohwinkel’s syndrome. Nat Genet1996; 13: 70–77.

19) Ishida-Yamamoto, A., McGrath, J.A., Lam, H.et al.: The molecular pathology of progressivesymmetric erythrokeratoderma: a frameshiftmutation in the loricrin gene and perturba-tions in the cornified cell envelope. Am J HumGenet 1997; 61: 581–589.

M. MANABE

JMAJ, November 2004—Vol. 47, No. 11 513

20) Yoneda, K., Takahira, Y., Idemitsu, T. et al.:Keratin diseases. Pharma Medica 1999; 17: 83–87.

21) Heiss, N.S., Knight, S.W., Vulliamy, T.J. et al.:X-linked dyskeratosis congenita is caused bymutations in a highly conserved gene withputative nucleolar functions. Nat Genet 1998;19: 32–38.

22) Vulliamy, T., Marrone, A., Goldman, F. et al.:The RNA component of telomerase ismutated in autosomal dominant dyskeratosiscongenita. Nature 2001; 413: 432–435.

23) Fischer, J., Bouadjar, B., Heilig, R. et al.:

Mutations in the gene encoding SLURP-1 inMal de Meleda. Hum Mol Genet 2001; 10:875–880.

24) Stewart, H., Smith, P.T., Gaunt, L. et al.: Denovo deletion of chromosome 18q in a babywith harlequin ichthyosis. Am J Med Genet2001; 102: 342–345.

25) Risk, J.M., Evans, K.E., Jones, J. et al.: Charac-terization of a 500kb region on 17q25 and theexclusion of candidate genes as the familialTylosis Oesophageal Cancer (TOC) locus.Oncogene 2002; 21: 6395–6402.

THE MOLECULAR BASIS OF KERATINIZING DISORDERS

514 JMAJ, November 2004—Vol. 47, No. 11

Updates on Autoimmune SkinBullous DiseasesJMAJ 47(11): 514–519, 2004

Masayuki AMAGAI

Assistant Professor, Department of Dermatology, Keio University School of Medicine

Abstract: Autoimmune skin bullous diseases are a group of diseases induced byautoantibodies against adhesion molecules in the skin. The understanding ofthese diseases has progressed greatly through studies combining cell and molecu-lar biological investigation of cell adhesion. A technique for producing recombinantpemphigus antigen proteins became available, and the development of the ELISAmethod facilitated serodiagnosis and monitoring of disease activity. Furthermore,a mouse model of pemphigus was developed through a new methodology usingautoantigen knockout mice produced by gene manipulation. The pemphigus modelmice produce the IgG antibodies against desmoglein3 (Dsg3) for over 6 months.This model is useful for the study of the mechanisms for antibody production andimmune tolerance to peripheral antigens, as well as the evaluation of variousimmunosuppression therapies. In addition, monoclonal antibodies against Dsg3were isolated from the pemphigus model mouse, and the studies using theseantibodies expanded our knowledge of the molecular basis for blister formation. Itis hoped that the study on autoimmune diseases of the skin will lead to elucidationof the pathophysiological mechanisms of autoimmune diseases and the develop-ment of treatment with minimal side effects.

Key words: Pemphigus; Bullous pemphigoid; ELISA; Model mouse

Introduction

Autoimmune skin bullous diseases are agroup of diseases induced by autoantibodiesagainst adhesion molecules in the skin. Typicalexamples of these diseases are pemphigus,which impairs adhesion between epidermalcells, and bullous pemphigoid, which impairs

adhesion in the basement membrane area. Theunderstanding of autoimmune bullous derma-toses has progressed greatly through studiescombining cell and molecular biological inves-tigation of cell adhesion. Based on the findingsfrom these studies, an ELISA (enzyme-linkedimmunosorbent assay) method using recom-binant antigen proteins was developed as a

This article is a revised English version of a paper originally published inthe Journal of the Japan Medical Association (Vol. 129, No. 9, 2003, pages 1425–1429).

� Skin Diseases

JMAJ, November 2004—Vol. 47, No. 11 515

diagnostic tool for clinical application. Further-more, a mouse model of pemphigus was devel-oped through a new methodology using auto-antigen knockout mice produced by genemanipulation.

Focusing on these topics, this article reviewsthe updates on autoimmune skin bullous dis-eases.

Pemphigus and Bullous Pemphigoid

Pemphigus is divided into two major forms,pemphigus vulgaris (PV) and pemphigus folia-ceus (PF). PV is further classified into mucosaldominant type mainly affecting the oral mucosaand mucocutaneous type forming blisters anderosion not only in the oral mucosa but also inthe skin. PF does not affect mucous mem-branes, but forms erosion and scaly erythemaonly in the skin.

Patients with pemphigus produce IgGautoantibodies against desmoglein (Dsg), acadherin-type cell-cell adhesion molecule thatplays important roles in the adhesion betweenepidermal cells. Dsg1 and Dsg3 are mainlyexpressed in stratified squamous epitheliumincluding the skin and mucous membranes.Patients with mucosal dominant type of PVshow anti-Dsg3 antibodies alone, while thosewith mucocutaneous type of PV show bothanti-Dsg1 and anti-Dsg3 IgG autoantibodies.Patients with PF produce anti-Dsg1 antibodiesalone. An indirect immunofluorescence (IF)assay using normal human skin as the substratehas demonstrated that the IgG autoantibodiesin pemphigus react only with the cell surfacesof keratinocytes.

Bullous pemphigoid (BP) is an autoimmunedisease in which the target antigens are the230 kD BP antigen (BP230, BPAG1) and180 kD BP antigen (BP180, BPAG2) found inhemidesmosomes, which are adhesion struc-tures connecting the epidermal basal cells andthe basement membrane. BP is prone todevelop in aged persons. Typical clinical symp-toms of BP include tense blisters on itchy

urticarial erythema. Histopathologically, BPshows subepidermal blisters.

In addition to typical BP that produces skinrashes all over the body, there are variousclinical subtypes such as localized BP (the typeshowing skin rashes in limited anatomical areassuch as anterior tibial area, head, face, orneck); vesicular BP (the type showing onlysmall vesicles); nodular BP (the type showingskin rashes resembling nodular purigo); andvegetative BP (the type showing papillomatousgrowth and rising of the intertriginous areas).In any type, an IF assay using healthy humanskin as the substrate demonstrates that the IgGautoantibodies in pemphigoid react with thebasement membrane zone.

Serodiagnosis with ELISA UsingRecombinant Antigen Proteins

ELISA can be applied to any antigen, pro-vided that highly purified antigen has beenobtained. A major advantage of this assay is theability to measure antibody titers, because theassay quantifies the reactivity of autoanti-bodies by means of a color reaction, and theresult is read on a spectrophotometer. Inter-assay and inter-facility comparisons of assayresults can be achieved by the use of theELISA score (index value) calculated from thecomparison with the reactivity of a standardspecimen.

As a tool for serodiagnosis of pemphigus, anELISA method using recombinant Dsg1 andDsg3 as coated antigens has been developedand put into clinical use.1–4) The recombinantproteins used in this method are produced inbaculovirus expression system using culturedinsect cells, and have been confirmed to haveproper 3-dimensional structures. Serodiagnosisis made as described below, based on thecombination of ELISA result using Dsg1 andDsg3 (Fig. 1).8)

If the tested serum is positive for Dsg3 andnegative for Dsg1, the diagnosis is mucosaldominant type of PV. If the serum is positive

AUTOIMMUNE SKIN DISEASES

516 JMAJ, November 2004—Vol. 47, No. 11

converted to the true antigen titer.8)

The Dsg1 and Dsg3 ELISA has been cov-ered by the health insurance system and theresults of basic studies have been widely uti-lized in daily clinical practice.

A Novel Mouse Model for Pemphigus

Why are autoantibodies attacking the selfcomponents formed in autoimmune diseases?We have not clearly answered this question asyet even after the turn of the century. The treat-ment for autoimmune diseases is mainly basedon steroids, immunosuppressants, and plasma-pheresis. These therapies suppress the immunesystem in general, and thus have the problemof causing severe side effects. The study ofthe mechanisms for autoantibody productionand the development of new antigen-specificimmunosuppressive therapies will require ani-mal models simulating diseases in humans.

Conventionally, mouse models of auto-immune diseases have been produced byforced immune reactions using repeated immu-nization of various wild-type mice with antigenproteins in various immune adjuvants. Whileseveral models have been produced by thismethod, the success of model preparationstrongly depends on empirical factors, because

M. AMAGAI

for both Dsg3 and Dsg1, the diagnosis is muco-cutaneous type of PV. If the serum is negativefor Dsg3 and positive for Dsg1, the diagnosis isPF. It should be noted that serodiagnosis is notthe final diagnosis. Final diagnosis should bemade based on the clinical as well as histo-pathological findings.

For serodiagnosis of bullous pemphigoid, anELISA method using recombinant protein ofNC16a domain, where the main epitope ofBP180 is located, was developed.6,7)

Because the ELISA method quantitativelymeasures antibody titers, it is useful for moni-toring disease progression.1,2,8) ELISA scorescan be used as a guide in the determination ofa steroid tapering schedule. In plasmapheresis,clearance rate can be calculated from theELISA score of the serum before treatmentand that of the waste fluid, and this provides anindex for objective evaluation of the antibodyremoval rate. In patients showing repeatedremissions and aggravations, an increase in theELISA score sometimes precedes an aggrava-tion and aids early treatment.

The measurement in ELISA is based on anenzyme reaction. When the patient has a veryhigh antibody titer, this means that the serumdilution factor must be increased more thanusual 100 fold and the result of ELISA must be

Fig. 1 Serodiagnosis of pemphigusThe serodiagnosis of pemphigus should preferably be conducted by a combination of indirect immunofluorescence(IIF) assay and ELISA. A serodiagnosis of pemphigus is made, if the reactivity of IgG to epidermal cell surfacesby IIF and the presence of IgG antibodies against Dsg1 or Dsg3 are confirmed by ELISA. Differentiation ofmucosal dominant type of PV, mucocutaneous type of PV, and PF is made by the combination of the presence orabsence of anti-Dsg1 and anti-Dsg3 antibodies.(Modified from Amagai, M.: Newest Dermatology Series Vol.6, Nakayama Shoten, 2002; pp.21–26.)

Indirect Immunofluorescence(IIF) Assay ELISA Serodiagnosis

Anti-Dsg1 IgG Ab Anti-Dsg3 IgG Ab

�

�

�

�

�

�

�

�

Pemphigus vulgaris (mucosal dominant)

Pemphigus vulgaris (mucocutaneous)

Pemphigus foliaceus

Normal or diseases other than pemphigus

Positive onepidermal cell surface

Negative onepidermal cell surface

JMAJ, November 2004—Vol. 47, No. 11 517

immune responses vary depending on mousestrains and the type of adjuvants. The auto-immune response of these model mice is oftentransient, and few achieve a persistent auto-immune state similar to actual diseases.

In addition, organ-specific autoimmune dis-eases in humans usually involve antigen-specific disruption of autoimmune tolerance,and other immune functions remain in normalconditions. However, the conventional methodusing forced immunity induces the activation ofthe immune system in general, and oftenresults in great differences from actual dis-eases, such as the manifestation of systemicinflammatory reactions.

A new method for producing mouse modelsof organ-specific autoimmune diseases hasbeen invented recently.9) In autoantigen knock-out mice, the immune system in the process ofdevelopment does not encounter an auto-antigen that is missing in such mice. Therefore,the lymphocytes that are reactive to thisautoantigen are not removed or inactivated inthese mice, and the immune tolerance to theknocked out gene product is not established.

Utilizing this fact, the splenocytes (T and Bcells) of immunized autoantigen knockout miceare transferred into mice that express theautoantigen. To prevent the rejection of trans-planted splenic cells, this procedure is con-

ducted in a Rag2 knockout mouse, which isan immunodeficient mouse lacking mature Tand B cells. The transferred T and B cellsderived from the autoantigen knockout mouseencounter the autoantigen in the recipientmouse, and a persistent autoimmune responseis expected to take place.

Based on this principle, a mouse model ofpemphigus was produced using Dsg3 knockoutmice9) (Fig. 2). When Dsg3 knockout micewithout immune tolerance to Dsg3 were immu-nized with rDsg3, this procedure easily inducedantibodies that could bind to Dsg3 in vivo.Next, the splenocytes from the immunizedDsg3 knockout mice were adoptively trans-ferred into Rag2 knockout mice expressingDsg3. The production of the IgG antibodiesagainst Dsg3 was detected in the blood of therecipient mice 4 to 7 days after transfer. Theantibody titer reached a peak after 21 days, andpersistent antibody production was observedfor over 6 months. The antibodies did not showreactivity with Dsg1, and were specific to Dsg3.

Deposition of mouse IgG was observed inthe cell membranes of Dsg3-expressing strati-fied squamous epithelium in the skin, oralmucosa, and esophagus of the recipient mice.In addition, impairment of cell adhesion wasseen in the epidermis and mucous epithelium,as well as blister formation just above the basal

A B

Fig. 2 Pemphigus model mousePemphigus model mice present feeding impairment due to extensive erosion in theoral cavity and body weight loss (A: the pemphigus model mouse above is smallerthan the normal control mouse below). Some pemphigus model mice develop crustederosions in areas constantly receiving external force, such as the soles (B).

AUTOIMMUNE SKIN DISEASES

518 JMAJ, November 2004—Vol. 47, No. 11

cell layer, which is a histopathological changecharacteristic of PV. Extensive erosions in theoral cavity caused inhibition of food intake,and body weight loss was observed from 7 daysafter the adoptive transfer, when antibodyproduction became evident. Some recipientmice showed crusted erosions in areas usuallyscratched by the mice, such as the area aroundthe nose.

Based on these findings, the mice producedhere were considered a model showing charac-teristic clinical, histopathological, and immu-nological features of pemphigus. Among thetypes of pemphigus, these mice were con-sidered a model of mucosal dominant typeof PV, because they produced only anti-Dsg3IgG antibodies and the main symptoms wereobserved in mucous membranes.

Isolation of Monoclonal AntibodiesInducing Pemphigus

Do all IgG autoantibodies that can bind toDsg3 in vivo equally induce blisters, or dodifferent antibodies have different pathogenicactivity to induce blisters? The answer to thisquestion is important for understanding themechanism of blister formation in pemphigusand the reason why severity varies amongcases. Taking advantage of the fact that thepemphigus model mice produced antibodieswith pathogenic activity, we isolated severaltypes of anti-Dsg3 IgG monoclonal antibodiesand analyzed the relation between their patho-genic activity and the conformational epitopesrecognized by the antibodies.10) We obtained 9monoclonal antibodies (AK mAb) that werereactive with Dsg3.

When the hybridoma cells of each AK mAbwere inoculated into the intraperitoneal cavityof mice, only the hybridoma of AK23 mAbinduced the phenotype of PV in mice. Next, weanalyzed the conformational epitopes of thesemAbs. Dsg1/Dsg3 chimeric molecules andpoint mutated molecules were used for thisanalysis. AK23 mAb was found to occur in V3,

K7, P8, and D59 considered to comprise theadhesive interface formed between Dsg3molecules. The other mAb types that did notinduce blister formation were found to recog-nize functionally unimportant parts that arenot directly related to the interaction betweenDsg3 molecules.

These results indicated that not all auto-antibodies recognizing Dsg3 have equal patho-genic activity, and autoantibodies recognizingdifferent sites on the Dsg3 molecule havedifferent degrees of pathogenicity. A possibilitywas suggested that the difference in severitymight be explained by the different epitopes ofautoantibodies in different cases.

In the future, the use of pemphigus modelmice is expected to facilitate the elucidation ofautoantibody production mechanisms and thedevelopment of new disease-specific therapies.In addition, the method using autoantigenknockout mice demonstrated in this study canbe applied widely to other autoimmune diseases.

Conclusion

The pathogenetic mechanism of auto-immune diseases has not been clarified as yet.The dermatoses discussed in this paper areexamples of those few diseases that can beexplained in terms of molecular biology. It ishoped that the findings in dermatoses will opena path for true understanding of the patho-physiological mechanisms of autoimmunediseases.

REFERENCES

1) Ishii, K., Amagai, M., Hall, R.P. et al.: Charac-terization of autoantibodies in pemphigususing antigen-specific ELISAs with baculo-virus expressed recombinant desmogleins. JImmunol 1997; 159: 2010–2017.

2) Amagai, M., Komai, A., Hashimoto, T. et al.:Usefulness of enzyme-linked immunosorbentassay (ELISA) using recombinant desmo-gleins 1 and 3 for serodiagnosis of pemphigus.Br J Dermatol 1999; 140: 351–357.

M. AMAGAI

JMAJ, November 2004—Vol. 47, No. 11 519

3) Nishikawa, T.: Desmoglein ELISAs: a noveldiagnostic tool for pemphigus. Arch Dermatol1999; 135: 195–196.

4) Harman, K.E., Gratian, M.J., Seed, P.T. et al.:Diagnosis of pemphigus by ELISA: a criticalevaluation of two ELISAs for the detection ofantibodies to the major pemphigus antigens,desmoglein 1 and 3. Clin Exp Dermatol 2000;25: 236–240.

5) Amagai, M.: Diagnosis of pemphigus. InTamaki, K. et al. (general editor): NewestDermatology Series Vol. 6, Bullous Dermatosisand Pustulosis, Nakayama Shoten, Tokyo,2002; pp.21–26. (in Japanese)

6) Zillikens, D., Mascaro, J.M., Rose, P.A. et al.:A highly sensitive enzyme-linked immuno-sorbent assay for the detection of circulatinganti-BP 180 autoantibodies in patients withbullous pemphigoid. J Invest Dermatol 1997;109: 679–683.

7) Kobayashi, M., Amagai, M., Kuroda- Kinoshita,K. et al.: BP 180 ELISA using bacterial recom-binant NC 16a protein as a diagnostic andmonitoring tool for bullous pemphigoid. JDermatol Sci 2002; 30: 224–232.

8) Cheng, S.W., Kobayashi, M., Tanikawa, A. etal.: Monitoring disease activity in pemphiguswith enzyme-linked immunosorbent assayusing recombinant desmogleins 1 and 3. Br JDermatol 2002; 147: 261–265.

9) Amagai, M., Tsunoda, K., Suzuki, H. et al.:Use of autoantigen-knockout mice in devel-oping an active autoimmune disease model forpemphigus. J Clin Invest 2000; 105: 625–631.

10) Tsunoda, K., Ota, T., Aoki, M. et al.: Inductionof pemphigus phenotype by a mouse mono-clonal antibody against the amino-terminaladhesive interface of desmoglein 3. J Immunol2003; 170: 2170–2178.

AUTOIMMUNE SKIN DISEASES

520 JMAJ, November 2004—Vol. 47, No. 11

Latest Information on AlopeciaJMAJ 47(11): 520–523, 2004

Satoshi ITAMI

Associate Professor, Department of Dermatology, Graduate School of Medicine,Osaka University

Abstract: In most of mammals, hair plays a very important role and protectsthem against physical forces, defends them against parasites and UV rays,provides heat conservation, and acts as a sensory organ. During the process ofevolution, human hair lost these essential life-supporting functions. Human hair isnow viewed as an object of aesthetic interest. However, androgenetic alopecia(male pattered hair loss) has been a source of distress for many men since ancienttimes. Hair loss due to the side effect of anticancer agents and extensive alopeciaareata often drastically impair the QOL of affected persons. Recent rapid progressin molecular biology has clarified the molecular mechanisms for hair generationand regeneration. The pathogenetic mechanism of androgenetic alopecia, reportedto affect about one-third of all men in their 40’s, and the nature of alopecia areata(as an autoimmune condition), have recently been clarified. New therapies basedon these findings are expected to emerge.

Key words: Hair cycle; Male pattern alopecia; Alopecia areata

Introduction

The hair follicle is one of the smallest organsin the body. In many mammals, hair plays avery important role and protects the animalagainst physical forces, defends it against para-sites and UV rays, provides heat conservation,and acts as a sensory organ. For human beings,hair has lost these essential life-supportingfunctions, and it is now viewed as an object ofaesthetic interest that only serves to enhancephysical attractiveness and to provide camou-flage. However, unexpected hair loss causes

This article is a revised English version of a paper originally published inthe Journal of the Japan Medical Association (Vol. 129, No. 9, 2003, pages 1431–1434).

serious psychological trauma in the affectedpersons, and it is sometimes described as a life-altering disease.

This paper reviews new topics on andro-genetic alopecia and alopecia areata, which areclassic types of hair loss.

Hair Cycle

Biologically, the hair follicle is a unique organin the body where the process of tissue regen-eration and involution is repeated throughoutthe lifespan. This process is called the hair cycle

� Skin Diseases

JMAJ, November 2004—Vol. 47, No. 11 521

in vitro co-culture system using epithelial cellsand dermal papilla cells from hair follicles. Inthe growth of sexual hair during puberty, wefound that insulin-like growth factor I (IGF-I)is the hair growth factor released from bearddermal papilla cells (paracrine growth factor)in an androgen-dependent manner. In andro-genetic alopecia in humans, we found that der-mal papilla cells release transforming growthfactor (TGF) �1 in an androgen-dependentmanner. TGF-�1 suppressed the growth ofkeratinocytes in hair follicles and shortened theduration of anagen in male pattern alopecia(Fig. 2).3,4)

For the treatment of this common diseaseaffecting 30% of Japanese males in their 40’s, a1% minoxidil solution was marketed in Japanseveral years ago, which generated great inter-est. Minoxidil was originally developed as a Kchannel opener to treat hypertension. Becausethis agent was found to cause the adverse effectof hypertrichosis, it is used as a topical medica-tion for alopecia. Although the mode of actionhas not been clarified, minoxidil is suspectedto promote hair growth because it activatesprostaglandin endoperoxide synthase-1 in hairpapilla cells and enhances PGE2 synthesis.5)

In Western countries, a type-II 5� reductaseinhibitor called finasteride is currently pre-scribed.6) A clinical trial of this agent is ongoingin Japan, it will probably be marketed in Japanbefore next year. While this agent causes

(from anagen to catagen to telogen) (Fig. 1).Because the regulation of the hair cycle involvesmany factors, various pathological conditionsresult in alopecia.1)

Effects of Androgen on Hair Cyclesin Humans

Interestingly, most of the steroid receptorfamily affect hair cycles in humans, as exem-plified by hypertrichosis due to glucocorticoids,alopecia due to thyroid hormone, retinoid, andvitamin D abnormalities, and hirsutism andmale pattern alopecia due to androgen. Whilemany of these abnormalities are caused by theaction to epithelial cells of the hair follicle, thedifference in androgen sensitivity is defined bydermal papilla cells, which are mesenchymalcells of hair follicles.

The dermal papilla is the only part of a hairfollicle that has androgen receptors and type-II 5�-reductases (known isozymes of 5�-reductase), which are needed for the action ofandrogen in male sexual organs.2) Strong malephenotype expressions, such as beards andmale pattern alopecia, require both androgenreceptors and type-II 5� reductase. Becauseepithelial cells of hair follicle lack androgenreceptors, hair growth seems to be controlledby the androgen-dependent release of signalsfrom dermal papilla cells.

To find out the signals, we established an

Fig. 1 Hair cycle

Anagen Catagen Telogen Anagen

Signals forcatagen initiation

Signals foranagen initiation

Shaft formingsignals

Fig. 2 Action mechanism of androgen in hair follicle

Apoptosis ofmatrix

Androgeneticalopecia:TGF-�1

Activation ofmatrix cells

Hair follicle

Matrix cells

Dermal papilla cellsAndrogen

Growth signalSuppressive signal

Beards: IGF-I

Prolongation ofanagen

Transition totelogen

RECENT ADVANCE IN HAIR BIOLOGY AND DISEASE

522 JMAJ, November 2004—Vol. 47, No. 11

DQB1*0301 (DQ7) has been shown to causeextensive development and prolongation ofthe disease.8)

In the affected area of the skin, CD4lymphocyte infiltration is observed around theanagen hair roots, CD8 lymphocytes invadeinto hair roots, and matrix cells undergoapoptosis. However, no lymphocyte infiltrationis seen in the vicinity of the hair bulge, wherehair follicle stem cells are believed to belocated. For this reason, permanent hair lossdoes not occur in alopecia areata. When analopecic scalp skin from a patient is trans-planted to a SCID mouse (an immunodeficientmouse), hair will regenerate. When T lympho-cytes isolated from the affected skin are stimu-lated by a homogenate of hair follicles andintroduced locally into the transplanted site,the disease is reproduced.9) These results indi-cate that alopecia areata is an autoimmunedisease in which T lymphocytes recognize andattack autoantigens in hair follicles.

Local immunotherapy is the most effectivetreatment for intractable multiple alopeciaareata at present (Fig. 3). Though details of themode of action have not been clarified, thistherapy uses local application of hapten toinduce allergic contact dermatitis aimed atstimulating hair generation. When the expres-sion of cytokine mRNAs in locally infiltratinglymphocytes was compared before and after

S. ITAMI

hypogonadism among a few percent of themales, oral administration of finasteride iscontraindicated in females, because it causesfeminization of male fetuses. We expect thedevelopment of new therapeutic agents tar-geted at TGF-� in the future.

Alopecia Areata

Alopecia areata is a common disease repre-senting 2% to 5% of all new patients visitingdermatology departments. According to statis-tics in the U.S., the prevalence is about 2,000per 100,000 persons, and the lifetime risk isreported to be 1.7%.7) This disease is character-ized by clearly defined patches of hair losschiefly on the head. Alopecia areata is classifiedinto 4 types: single, multiple, totalis, and uni-versalis (i.e., total body hair loss). Usually, thedisease appears as a single or a few coin-shapedpatches of hair loss, but a total loss of hair onthe head and body occurs in 7% of all cases. Nodifferences between the sexes are observed.While this disorder develops in a wide range ofages, about 25% of all cases develop before theage of 15. About 20% of the patients have afamily history of this disease. As noted above,alopecia areata is often regarded as a trivialproblem, but it may cause serious psycho-logical trauma in the affected persons. In thecase of minors, it can be a cause of schooltruancy, and the patient may become the targetof bullying by classmates. In adults, this diseasecan result in limited employment choices.

The pathogenesis of this disorder wasexplained by various theories in the past,including infection, nutritional disturbance,neurological (stress), and endocrine theories.Nowadays, it is considered an organ-specificautoimmune disease with a genetic back-ground, because about 20% of patients arepositive for antinuclear antibodies, thyro-globulin, and microsome antibodies. Humanleukocyte antigen (HLA) DQB1*03 (DQ3)is commonly detected in this disease, andthe presence of DQB1*1104 (DR11) and

Fig. 3 Local immunotherapy for alopecia areata(universal)

A: Before treatment B: After treatment

JMAJ, November 2004—Vol. 47, No. 11 523

dermal papilla cells from beard and occipitalscalp hair. J Invest Dermatol 1991; 96: 57–60.

3) Itami, S., Kurata, S. and Takayasu, S.: Andro-gen induction of follicular epithelial cellgrowth is mediated via insulin-like growthfactor-I from dermal papilla cells. BiochemBiophys Res Commun 1995; 212: 988–994.

4) Inui, S., Fukuzato, Y., Nakajima, T. et al.:Androgen-inducible TGF-�1 from baldingdermal papilla cells inhibits epithelial cellgrowth: a clue to understand paradoxicaleffects of androgen on human hair growth.FASEB J 2002; 10: 1096/fj. 02–0043fje.

5) Michelet, J.F., Commo, S., Billoni, N. et al.:Activation of cytoprotective prostaglandinsynthase-1 by minoxidil as a possible explana-tion for its hair growth-stimulating effect. JInvest Dermatol 1997; 108: 205–209.

6) Price, V.H., Menefee, E., Sanchez, M. et al.:Changes in hair weight and hair count in menwith androgenetic alopecia after treatmentwith finasteride, 1mg, daily. J Am AcadDermatol 2002; 46: 517–523.

7) Safavi, K.H., Muller, S.A., Suman, V.J. et al.:Incidence of alopecia areata in OlmstedCounty, Minnesota, 1975 through 1989. MayoClin Proc 1995; 70: 628–633.

8) Colombe, B.W., Price, V.H., Khoury, E.L. etal.: HLA class II antigen associations help todefine two types of alopecia areata. J Am AcadDermatol 1995; 33: 757–764.

9) Gilhar, A., Ullmann, Y., Berkutzki, T. et al.:Autoimmune hair loss (alopecia areata) trans-ferred by T lymphocytes to human scalpexplants on SCID mice. J Clin Invest 1998;101: 62–67.

10) Hoffmann, R., Wenzel, E., Huth, A. et al.:Cytokine mRNA levels in Alopecia areatabefore and after treatment with the contactallergen diphenylcyclopropenone. J InvestDermatol 1994; 103: 530–533.

11) Friedli, A., Labarthe, M.P., Engelhardt, E. etal.: Pulse methylprednisolone therapy forsevere alopecia areata: an open prospectivestudy of 45 patients. J Am Acad Dermatol1998; 39: 597–602.

local immunotherapy, the TH1 type was seenbefore treatment, and an increase in IL-10and a decrease in IL-1� were observed aftertreatment.10)

Systemic administration of steroids has longbeen criticized because the life prognosis ofthis disease does not warrant such treatment.However, this therapy is attracting new atten-tion as understanding and recognition of thisdisease as an autoimmune condition and as alife-altering disease have grown. The efficacyof steroid pulse therapy has been reported inpatients with multiple alopecia areata withinone year of its onset.11)

Conclusion

Alopecia is often disregarded as a minordisease, but the distress of affected persons isserious. This article focuses on the mode ofonset of male pattern alopecia and alopeciaareata, which have been elucidated recently.The mechanism of the hair cycle is being clari-fied rapidly, since molecular biologists, in addi-tion to dermatologists, have become interestedin hair as an ideal tissue regeneration model.

Based on such findings, promising resultshave been reported from animal experimentsto prevent chemotherapy induced alopecia. Inaddition, the study of the hair follicle, thesmallest organ in the body, is expected to pro-vide new information that may contribute tothe progress of regenerative medicine of otherorgans.

REFERENCES

1) Cotsarelis, G. and Millar, S.E.: Towards amolecular understanding of hair loss and itstreatment. Trends Mol Med 2001; 7: 293–301.

2) Itami, S., Kurata, S., Sonoda, T. et al.: Charac-terization of 5�-reductase in cultured human

RECENT ADVANCE IN HAIR BIOLOGY AND DISEASE

524 JMAJ, November 2004—Vol. 47, No. 11

Introduction

Melanoma is the malignancy of melanocytes(pigment cells) that produce the melaninpigment. Since the 1970s, the incidence ofmelanoma has doubled in Europe and the U.S.,and this increase has become a serious socialconcern. Changes in lifestyle and the increasein UV radiation due to ozone depletion havebeen identified as causes that contribute to

this increase.On the other hand, the incidence of mela-

noma in Japan has been 1/10 to 1/20 of that inEurope and the U.S. While about 40% of mela-noma cases in Japan used to have lesions in theplantar area, which is not associated with UVradiation, recent cases show an increase inlesions on the trunk and lower limbs that areexposed to sunlight. This trend is alarming.

Melanoma tends to undergo lymph node

Recent Progress in Diagnosis andTreatment of MelanomaJMAJ 47(11): 524–528, 2004

Toshiro KAGESHITA

Associate Professor, Department of Dermatology,Graduate School of Medical Sciences, Kumamoto University

Abstract: New methods for the diagnosis and treatment of melanoma areoutlined. Dermoscopy provides a simple and effective means of diagnosing mela-noma, and it is expected to be used widely in primary care settings. The CGHmethod that detects abnormalities in the number of DNA copies in the genome isvery useful where differential diagnosis between melanoma and benign pigmentedlesions is difficult in histopathological study. The effectiveness of sentinel lymphnode biopsy has been confirmed with respect to surgery. This method eliminatesthe need for lymph node dissection when the sentinel lymph node is negative formetastasis, and can improve the patient’s QOL. The clinical use of leading-edgetherapies for advanced cases, such as peptide