Letters to the Editor / Journal of Dermatological Science 61 (2011) 60–81 69

miR-29a at the SSD stage, but maintained by other factors in SScpatients.

Table 1a shows the association of serum miR-29a levels withthe clinical features in SSc patients. Considering that collagenexpression in SSc patients is up-regulated as described above, weregarded reduction of miR-29a levels as the meaningful change.We found that SSc patients with reduced miR-29a levels hadsignificantly higher right ventricular systolic pressure by Dopplerechocardiography than those with normal levels. Though the causeof pulmonary hypertension in SSc is still uncertain, our resultssuggested that miR-29a also takes part in the pathogenesis ofpulmonary hypertension. On the other hand, as shown in Table 1b,clinical features of each SSD patient were not correlated with thelevels of miR-29a. Taken together, serum miR-29a levels is notlikely to be a specific marker for clinical manifestations of SSD, butthe miRNA may play an important role in the pathogenesis of thisdisease.

The concept of SSD has been proposed by Maricq et al.originally to unify typical SSc, early forms of SSc and closelyrelated disorders including mixed connective tissue disease(MCTD) [8,9]. Thereafter, Ihn et al. established a new diagnosticmethod using a points system to distinguish patients with SSDfrom those with early SSc [10]. Although the point system has notbeen accepted in the world wide basis, because progressivefibrosis of SSc is often irreversible, at least clinically, there is anurgent need to develop new strategies to diagnose patients asearly as possible and follow-up carefully. For that purpose, theconcept of SSD should be further understood and characterized.Our study is the first to examine serum miRNA levels using serafrom SSD as well as SSc, and is shedding new light on the definitionof SSD. To note, it may be difficult to distinguish early stage SScfrom SSD, because skin sclerosis is sometimes not apparent inearly SSc, especially in lcSSc. Serum levels of miR-29a levels maybe useful for the differentiation of SSc from SSD.

As the limitation of this study, we could not collect largenumber of SSD patients because of the rarity of this condition.However, our approach may be effective to clarify the property ofSSD. Larger studies are needed in the future.

Acknowledgements

This study was supported in part by a grant for scientificresearch from the Japanese Ministry of Education, Science, Sportsand Culture, by project research on intractable diseases from theJapanese Ministry of Health, Labour and Welfare.

References

[1] Korn JH. Immunologic aspects of scleroderma. Curr Opin Rheumatol1989;1(4):479–84.

[2] Mauch C, Krieg T. Fibroblast-matrix interactions and their role in the patho-genesis of fibrosis. Rheum Dis Clin North Am 1990;16(1):93–107.

[3] Jelaska A, Arakawa M, Broketa G, Korn JH. Heterogeneity of collagen synthesisin normal and systemic sclerosis skin fibroblasts: increased proportion of highcollagen-producing cells in systemic sclerosis fibroblasts. Arthritis Rheum1996;39(8):1338–46.

[4] LeRoy EC. Increased collagen synthesis by scleroderma skin fibroblasts invitro: a possible defect in the regulation or activation of the sclerodermafibroblast. J Clin Invest 1974;54(4):880–9.

[5] Kroh EM, Parkin RK, Mitchell PS, Tewari M. Analysis of circulating microRNAbiomarkers in plasma and serum using quantitative reverse transcription-PCR(qRT-PCR). Methods 2010;50(4):298–301.

[6] Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked byadenosines, indicates that thousands of human genes are microRNA targets.Cell 2005;120(1):15–20.

[7] Ihn H, Sato S, Fujimoto M, Kikuchi K, Igarashi A, Soma Y, et al. Measurement ofanticardiolipin antibodies by ELISA using b2-glycoprotein I (b 2-GPI) insystemic sclerosis. Clin Exp Immunol 1996;105(3):475–9.

[8] Maricq HR, McGregor AR, Diat F, Smith EA, Maxwell DB, LeRoy EC, et al. Majorclinical diagnoses found among patients with Raynaud phenomenon from thegeneral population. J Rheumatol 1990;17(9):1171–6.

[9] Maricq HR, Weinrich MC, Keil JE, Smith EA, Harper FE, Nussbaum AI, et al.Prevalence of scleroderma spectrum disorders in the general population ofSouth Carolina. Arthritis Rheum 1989;32(8):998–1006.

[10] Ihn H, Sato S, Tamaki T, Soma Y, Tsuchida T, Ishibashi Y, et al. Clinicalevaluation of scleroderma spectrum disorders using a points system. ArchDermatol Res 1992;284(7):391–5.

Yoshio KawashitaMasatoshi Jinnin*

Takamitsu MakinoIkko Kajihara

Katunari MakinoNoritoshi Honda

Shinich MasuguchiSatoshi Fukushima

Yuji InoueHironobu Ihn

Department of Dermatology and Plastic Surgery, Faculty of Life

Sciences, Kumamoto University, 1-1-1 Honjo, Kumamoto, Japan

*Corresponding author. Tel.: +81 96 373 5233;fax: +81 96 373 5235

E-mail address: jinjin1011@hotmail.commjin@kumamoto-u.ac.jp

14 October 2010

doi:10.1016/j.jdermsci.2010.11.007

Letter to the Editor

Differential hyaluronan homeostasis and expression ofproteoglycans in juvenile and adult human skin

Cutaneous ageing is a complex biological process affecting allskin components. It consists of two independent, clinically andbiologically, distinct processes. The first is intrinsic or innateageing, which affects the skin in the same pattern as it affects allinternal organs. The second is extrinsic ageing, which is the resultof exposure to external factors, mainly ultraviolet (UV) irradia-tion [1].

Extracellular matrix molecules are highly implicated in theageing process and exhibit specific alterations in extrinsic andintrinsic skin ageing [2]. Among them, hyaluronic acid (HA) is ofhigh importance since it has the unique capacity of binding water,

thus providing viscosity and hydration to the dermis. In photo-ageing process, HA homeostasis shows specific alterations since inphoto-aged skin HA of reduced size is elevated and exhibitsabnormal deposition [3].

In this study, we have tried to elucidate alterations in HAhomeostasis and proteoglycan expression associated with intrinsicskin ageing. We employed juvenile skin tissue specimens (n = 10,mean age 5 years, 4 mm punch biopsies) collected from foreskin ofchildren undergoing surgery for phimosis. Adult photo-protectedskin tissue specimens (n = 16, mean age 72 years, 4 mm punchbiopsies) were collected from the area behind the ear lobe. Totalglycosaminoglycans were isolated and purified from skin tissuespecimens, as previously described [3]. Aliquots of total glycosa-minoglycans were assayed for HA content by ELISA. Gene

Letters to the Editor / Journal of Dermatological Science 61 (2011) 60–8170

expression of HYAL, HAS, CD44, RHAMM and proteoglycans wasassessed by RT-PCR [3]. Statistical analyses were performed usingSPSS 12.0 for Windows.

We report that the relative amount of HA in adult skin tissuespecimens was significantly less (p-value: 0.0001) compared withjuvenile skin (160 � 19 versus 310 � 27 ng/mg of uronic acids).Furthermore, gene expression of HAS-1, HAS-2, HYAL-2 and HYAL-3was significantly down-regulated in adult photo-protected skin(Fig. 1), indicating lower HA metabolic rate in intrinsic skin ageing,resulting in decreased content of HA in the adult skin. Thisobservation is in agreement with previous studies, employing tissuestaining, which demonstrated that there is a progressive reduction inthe number of electron-dense granules of HA and their filamentswith age [4]. The reduced HA content could account for some of themost striking alterations of the aged skin, including decreasedturgidity, less support for micro vessels, wrinkling and alteredelasticity.

The biological effects of HA are mediated followingbinding of HA to its receptors CD44 and RHAMM. In thepresent study, we showed that gene expression of bothreceptors is decreased following intrinsic ageing (Fig. 1). Thismay reflect to reduced or abnormal function of HA in the agedskin. Furthermore, it has been shown that binding of HA toCD44 is a prerequisite step for HA degradation. Thus, the

ig. 1. Gene expression of HAS, HYAL, CD44 and RHAMM. (A) Representative RT-PCR analysis of HAS, HYAL, CD44 and RHAMM. Densitometric ratios of HAS-1/b-actin (B),

AS-2/b-actin (C), HAS-3/b-actin (D), HYAL-2/b-actin (E), HYAL-3/b-actin (F), CD44/b-actin (G), and RHAMM/b-actin (H). Data are presented as boxplot analysis of RT-PCR

sults obtained from juvenile (n = 10) and from adult (n = 16) photo-protected skin tissue specimens. The length of the box contains 50% of cases. The line across the inside of

e box represents the median value. Lines protruding from the box go out to the variable’s smallest and largest values. *p-Value between 0.05 and 0.01, **p-value between

.01 and 0.005, ***p-value < 0.005.

[()TD$FIG]

FH

re

th

0

reduced expression of HAS, HYAL and CD44 that we report here,results to reduced turnover of HA and to extracellulardeposition of lower amounts of HA which may be regardedas ‘‘aged’’ and probably less functional.

We have previously shown that extrinsic skin ageing isassociated with alterations in the expression of HA and itsmetabolizing enzymes [3]. However, HA homeostasis in extrinsicskin ageing exhibits a different profile than in intrinsic skin ageingreported in this study. Extrinsic ageing is characterized byincreased amounts of HA of lower molecular mass, reduced HASand increased HYAL expression, whereas HA receptors are down-regulated [3]. These findings suggest that intrinsic and extrinsicskin ageing are two distinct biological processes, both associatedwith abnormal HA turnover.

In the present study, we further investigated the expression ofvarious proteoglycans in intrinsic skin ageing, since these moleculeshave been shown to regulate the mechanichal properties of the skin,tissue hydration and resiliency. We found that with the exception ofaggrecan which was significantly up-regulated in the aged skin, allother proteoglycans were significantly down-regulated (Fig. 2).

It has been shown that the mRNA level of aggrecan isup-regulated in skin fibroblasts of Hutchinson–Gilford ProgeriaSyndrome, a disorder of accelerated ageing which exhibits similarcharacteristics to intrinsic skin ageing [5]. The increased expres-

[()TD$FIG]

Fig. 2. Gene expression of proteoglycans. (A) Representative RT-PCR analysis of proteoglycans in 2 different juvenile and 2 different adult skin tissue specimens. (B)–(H)

Densitometric ratios of proteoclycans/b-actin. Data are presented as boxplot analysis of RT-PCR results obtained from juvenile (n = 10) and from adult (n = 16) photo-

protected skin tissue specimens. The length of the box contains 50% of cases. The line across the inside of the box represents the median value. Lines protruding from the box

go out to the variable’s smallest and largest values. *p-Value between 0.05 and 0.01, ***p-value < 0.005.

Letters to the Editor / Journal of Dermatological Science 61 (2011) 60–81 71

sion of aggrecan in the aged skin that we report here, indicates thatthis proteoglycan is related with skin ageing and its role in theageing process should be further investigated.

Versican has been identified in the dermis in association withelastic fibers [6]. The decreased expression of versican that wereport here is in agreement with previous studies showing thatversican decreases with age and this may reflect to reduced elastinlevels in papillary dermis found in the aged skin [7].

Decorin is the major proteoglycan in the skin, which plays acrucial role in collagen fiber formation and remodelling [8].Therefore, the decreased expression of decorin in aged skin that wereport here, may be associated with decreased collagen synthesisrelated with skin ageing.

Biglycan is present in low amounts in normal human skin.However, biglycan is highly expressed in models of tissue repair andis required for endothelial cell migration [9]. The decreasedexpression of biglycan in aged skin that we report in the presentstudy, indicates impaired tissue repair associated with intrinsic skinageing.

Perlecan, is essential for the formation of the epidermis. It iswell known that in intrinsic skin ageing, epidermis is thinner andthe basement membrane zone is flattened with reduced function.Therefore, the low expression of perlecan in aged skin reported in

this study may account for the reduced function of the epidermisassociated with intrinsic skin ageing.

Syndecan-3, mediates the adhesion of schwann cells to aheparan sulfate binding site on type V collagen [10]. The decreasedexpression of syndecan-3 in aged skin that we report here,indicates an abnormal schwann cell function in intrinsic skinageing.

Overall, the lower metabolic rate of HA and the reducedexpression of proteoglycans that we report in the present study,reflects the general deterioration of skin function in the aged skin.These findings provide a better understanding of intrinsic skinageing and could contribute to the refinement of current drugs andthe development of novel pharmacological opportunities toconfront skin ageing.

References

[1] Uitto J, Bernstein EF. Molecular mechanisms of cutaneous aging: connectivetissue alterations in the dermis. J Invest Dermatol Symp Proc 1998;3:41–4.

[2] Rabe JH, Mamelak AJ, McElgunn PJ, Morison WL, Sauder DN. Photoaging:mechanisms and repair. J Am Acad Dermatol 2006;55:1–19.

[3] Tzellos TG, Klagas I, Vahtsevanos K, Triaridis S, Printza A, Kyrgidis A, et al.Extrinsic ageing in the human skin is associated with alterations in theexpression of hyaluronic acid and its metabolizing enzymes. Exp Dermatol2009;18(12):1028–35.

Letters to the Editor / Journal of Dermatological Science 61 (2011) 60–8172

[4] Ghersetich I, Lotti T, Campanile G, Grappone C, Dini G. Hyaluronic acid incutaneous intrinsic aging. Int J Dermatol 1994;33:119–22.

[5] Lemire JM, Patis C, Gordon LB, Sandy JD, Toole BP, Weiss AS. Aggrecanexpression is substantially and abnormally upregulated in Hutchinson–Gilford Progeria Syndrome dermal fibroblasts. Mech Ageing Dev 2006;127:660–9.

[6] Zimmerman DR, Dours-Zimmermann MT, Schubert M, Bruckner-Tuderman L.Versican is expressed in the proliferating zone in the epidermis and in associa-tion with the elastic network of the dermis. J Cell Biol 1994;124:817–25.

[7] Francis C, Robert L. Elastin and elastic fibres in normal and pathologic skin. Int JDermatol 1984;23:166–79.

[8] Danielson KG, Barribault H, Gomes DF, Graham H, Kadler KE, Iozzo RV.Targeted disruption of decorin leads to abnormal collagen fibril morphologyand skin fragility. J Cell Biol 1997;136:729–53.

[9] Kinsella MG, Tsoi CK, Jarvelainen HT, Wight TN. Selective expression andprocessing of biglycan during migration of bovine aortic endothelial cells. Therole of endogenous basic fibroblasts growth factor. J Biol Chem 1997;272:318–25.

[10] Erdman R, Stahl RC, Rothblum K, Chernousov MA, Carey DJ. Schwann celladhesion to a novel heparan sulfate binding site in the N-terminal domain ofalpha 4 type V collagen is mediated by syndecan-3. J Biol Chem 2002;277:7619–25.

Thrasivoulos G. Tzellos2nd Department of Pharmacology, School of Medicine,

Aristotle University of Thessaloniki, Greece

Xenophon Sinopidis1st Department of Pediatric Surgery, G. Gennimatas Hospital,

Aristotle University of Thessaloniki, Greece

Athanassios KyrgidisKonstantinos Vahtsevanos

Department of Oral and Maxillofacial Surgery,

Theageneion Anticancer Hospital, Thessaloniki, Greece

Stefanos TriaridisAthanasia Printza

1st Academic Otorhinolaryngology Department, School of Medicine,

Aristotle University of Thessaloniki, Greece

Ioannis KlagasGeorge Karakiulakis

Eleni Papakonstantinou*2nd Department of Pharmacology, School of Medicine,

Aristotle University of Thessaloniki, Greece

*Corresponding author. Tel.: +30 2310 999367;fax: +30 2310 999367

E-mail address: epap@med.auth.gr (E. Papakonstantinou)

10 June 2010

doi:10.1016/j.jdermsci.2010.10.010

Letter to the Editor

The expression of b-defensin-2, 3 and LL-37 induced by Candida

albicans phospholipomannan in human keratinocytes

Antimicrobial peptides (AMPs) like b-defensin-2, 3 and LL-37are expressed in skin exposed to a variety of stimuli. AMPs play animportant role not only in host defense against the invadingpathogens, but also in wound healing, release of chemokines andpro-inflammatory cytokines [1,2]. Lopez-Garcıa et al. have shownthat the expression of LL-37 was upregulated in the skin lesion ofCandida albicans infection [3]. The production of b-defensin-2 invaginal epithelial cells could be triggered by C. albicans [4].However, the role of the interaction between human keratinocytesand C. albicans components in the production of AMPs remainsunclear. Here, we investigated whether the exposure of humankeratinocytes to native phospholipomannan form the cell wall of C.

albicans (PLM) could lead to the release of b-defensin-2, 3 and LL-37 and to determine the underlying mechanisms.

PLM was prepared from C. albicans (ATCC32354, serotype A,Rockville, MD, USA) according to the method previously described[5]. Real-time PCR was used to determine the time course of mRNAexpression of b-defensin-2, 3 and LL-37 in human keratinocyteschallenged with PLM (50 mg/ml), PGN (10 mg/ml, positive control,Sigma, St. Louis, MO), PLM hydrolyzed with b-D-mannosidemannohydrolase (Hydrolyzed PLM, Sigma, St. Louis, MO) anduntreated cells as the control for 1 hr, 4 hr and 6 hr. The mRNAexpression of b-defensin-2, 3 and LL-37 in keratinocytes wassignificantly up-regulated by PLM and PGN, starting at 1 hr afterstimulation and maintained the high level at 6 hr post stimulation(Fig. 1a).

The concentration of b-defensin-2, 3 (Phoenix Pharmaceuticals,Burlingame, CA) and LL-37 (Hycult Biotechnology, Uden, The

Netherlands) in the cell free culture media of keratinocytesexposed to three different concentrations (10, 50, and 100 mg/ml)of PLM, 10 mg/ml PGN and Hydrolyzed PLM for 8 hr was measuredby ELISA. The secretion of b-defensin-2, 3 and LL-37 inkeratinocytes was upregulated by PLM in a dose-dependentmanner (Fig. 1b).

The results showed that PLM hydrolyzed with b-D-mannosidemannohydrolase failed to upregulate the mRNA expression andsecretion of b-defensin-2, 3 and LL-37 in keratinocytes. Thesefindings suggested that gene expression and secretion of b-defensin-2, 3 and LL-37 in keratinocytes was dependent on thestructural integrity of the b-1, 2-linked mannose polysaccharidemoiety of PLM.

To clarify whether TLR2, TLR4 and related signal transductionpathways (NF-kB, p38MAPK) were involved in the secretion of b-defensin-2, 3 and LL-37 in keratinocytes challenged with PLM, wedetected the effect of TLR2 and TLR4 neutralizing antibodies andinhibitors of NF-kB, p38. The secretion of b-defensin-2, 3 and LL-37induced by PLM was significantly inhibited by neutralizing anti-human TLR2 mAb (TL2.1, 10 mg/ml, eBioscience San Diego, CA)whereas anti-human TLR4 (HTA125, 10 mg/ml, eBioscience SanDiego, CA) did not show any similar inhibitory effect and wasblocked by NF-kB inhibitor (isohelenin, 25 mM, Calbiochem, LaJolla, CA) and p38 inhibitor (SB239063, 5 mM, Sigma, St. Louis, MO)(Fig. 2a). We also showed that this response of keratinocytesinduced with PLM was not modified by LPS neutralizer (PolymyxinB, 10 mg/ml, Sigma, St. Louis, MO), thus excluding LPS contamina-tion. Since these three AMPs have fungicidal activities against C.

albicans [4,6], we further investigate whether PLM-TLR2-NF-kBand p38MAPK pathway contributed to the killing of C. albicans bykeratinocytes. C. albicans killing assay by keratinocytes was