View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector COMMENTARY

A variety of important scientific Oncogene 22:8590–607 of TRAIL receptors 1 and 2. J Invest Dermatol issues arise from the data of Farley and Farley SM, Dotson AD, Purdy DE, Sundholm AJ, 121:149–55 colleagues (2006). Is CD95L an ini- Schneider P, Magun BE et al. (2006) Fas ligand Locksley RM, Killeen N, Lenardo MJ (2001) tial activator, or rather an amplifier of elicits a caspase-independent proinflammatory The TNF and TNF receptor superfamilies: response in human keratinocytes: implications integrating mammalian biology. Cell 104:487– inflammation? If caspase activity is not for dermatitis. J Invest Dermatol 126:2438– 501 required for CD95-mediated nonapop- 2451 Thome M, Tschopp J (2001) Regulation of totic signals, are there conditions that Goebeler M, Trautmann A, Voss A, Brocker EV, lymphocyte proliferation and death by FLIP. allow for or block caspase activation in Toksoy A, Gillitzer R (2001) Differential and Nat Rev Immunol 1:50–8 the absence of apoptosis? This regulat- sequential expression of multiple chemokines Trautmann A, Akdis M, Kleemann D, Altznauer during elicitation of allergic contact F, Simon HU, Graeve T et al. (2000) T cell- ed caspase activity may potentially be hypersensitivity. Am J Pathol 158:431–40 mediated Fas-induced keratinocyte apoptosis crucial for the proinflammatory signal Kreuz S, Siegmund D, Rumpf JJ, Samel D, plays a key pathogenetic role in eczematous by CD95. In contrast to the findings by Leverkus M, Janssen O et al. (2004) NFkappaB dermatitis. J Clin Invest 106:25–35 Farley et al. (2006) with pharmacologi- activation by Fas is mediated through FADD, Vaux DL, Silke J (2005) IAPs, RINGs and cal peptide inhibitors of caspases, the caspase-8, and RIP and is inhibited by FLIP. J ubiquitylation. Nat Rev Mol Cell Biol 6:287– Cell Biol 166:369–80 97 physiological inhibitor cFLIP appears to κ Krueger A, Baumann S, Krammer PH, Kirchhoff S Viard-Leveugle I, Bullani RR, Meda P, Micheau O, inhibit death receptor-mediated NF- B (2001) FLICE-inhibitory : regulators of Limat A, Saurat JH et al. (2003) Intracellular activation and gene induction (Kreuz death receptor-mediated apoptosis. Mol Cell localization of keratinocyte Fas ligand explains et al., 2004; Wachter et al., 2004). These Biol 21:8247–54 lack of cytolytic activity under physiological variations hint at substantially different Lavrik IN, Golks A, Krammer PH (2005) Caspases: conditions. J Biol Chem 278:16183–8 outcomes of CD95 receptor trigger- pharmacological manipulation of cell death. J Wachter T, Sprick M, Hausmann D, Kerstan Clin Invest 115:2665–72 A, McPherson K, Stassi G et al. (2004) ing depending on the type of inhibitor Leverkus M, Sprick MR, Wachter T, Mengling cFLIPL inhibits tumor necrosis factor-related used. Moreover, it could be envisioned T, Baumann B, Serfling E et al. (2003a) apoptosis-inducing ligand-mediated NF- that physiological regulation of cFLIP Proteasome inhibition results in TRAIL kappaB activation at the death-inducing might be highly relevant to intracellu- sensitization of primary keratinocytes by signaling complex in human keratinocytes. J Biol Chem 279:52824–34 lar signal deviation upon CD95 trigger- removing the resistance-mediating block of effector caspase maturation. Mol Cell Biol Wajant H, Pfizenmaier K, Scheurich P (2003) ing. Likewise, intracellular inhibition 23:777–90 Non-apoptotic Fas signaling. Cytokine Growth of effector caspases or mitochondrial Leverkus M, Sprick MR, Wachter T, Denk A, Factor Rev 14:53–66 signaling pathways of apoptosis down- Brocker EB, Walczak H et al. (2003b) TRAIL- Wehrli P, Viard I, Bullani R, Tschopp J, French LE stream of caspase-8 activation by Bcl-2 induced apoptosis and gene induction in (2000) Death receptors in cutaneous biology family members or inhibitor-of-apop- HaCaT keratinocytes: differential contribution and disease. J Invest Dermatol 115:141–8 tosis proteins (IAPs) (Cory et al., 2003; Vaux and Silke, 2005) might inter- fere with apoptotic, but not nonapop- See related article on pg 2377 totic, signals by death receptors. This scenario may ultimately result in an The New Nomenclature uncontrolled activation of CD95-medi- ated inflammation in the skin, but this Rebecca M. Porter1 hypothesis awaits further experimental studies. Taken together, it appears that When the first nomenclature of the keratin family was published over CD95 is not only needed for a silent 20 years ago, only 19 were thought to exist. Sequencing of the human end of the life of keratinocytes during genome has now revealed that there are 54 keratin genes. As a consequence, eczema but may rather contribute to the nomenclature needed revision to apply a logical numbering system that a “going out with an (inflammatory) includes the more recently identified keratins of the hair follicle. bang.” It will be interesting to deter- Journal of Investigative Dermatology (2006) 126, 2366–2368. doi:10.1038/sj.jid.5700532 mine under which conditions apoptosis may predominate over CD95-mediated inflammation in keratinocytes. This dif- ference — subtle at first glance — of a The intermediate filaments of the cyto- lial cell types of the epidermis and hair death receptor-mediated signal might skeleton in epithelial cells are made up follicle as well as to determine their dif- prove to be highly relevant to the quan- of a wide variety of subtly different kera- ferentiation state. titative response in the skin as the target tin proteins. Antibodies to keratins have The first nomenclature of keratins was organ of eczematous inflammation. been extensively used in dermatology published 24 years ago by Moll et al. research. They are useful tools to defini- (1982). At this time the concept of there CONFLICT OF INTEREST The authors state no conflict of interest. tively identify the many different epithe- being two types of keratin proteins, type I

1 REFERENCES Department of Dermatology, Wales College of Medicine, Cardiff University, United Kingdom Cory S, Huang DC, Adams JM (2003) The Bcl-2 Correspondence: Dr. Rebecca M. Porter, Department of Dermatology, Wales College of Medicine, family: roles in cell survival and oncogenesis. Cardiff University, Heath Park, Cardiff CF14 4XN, Wales, United Kingdom. E-mail: [email protected]

2366 Journal of Investigative Dermatology (2006), Volume 126 COMMENTARY

The keratin genes are clustered at two loci in the genome, type I on chro- mosome 17q21.2 and type II on chro- mosome 12q13, with the exception of the K18, which resides in the type II locus. In an attempt to iden- tify all of the human keratins, both loci have now been completely sequenced (Hesse et al., 2004; Rogers et al., 2005). Interestingly, this revealed only four keratins that had not been previously published. One of these was a keratin found only in the eccrine sweat gland, suggesting that perhaps their identity was not revealed before because of limited expression of these keratins (Langbein et al., 2005). The expression of the other three remains to be examined in detail. As it is unlikely that there are other keratins outside the two loci, it is now timely to revise the nomenclature. At the Gordon Conference on Intermediate Filaments in 2004, in Oxford, a nomen- Figure 1. Changes to the nomenclature of the major epidermal keratins. The only change to the clature committee was convened. The epidermal keratins is from K2e to K2. No changes in nomenclature apply to the outer root sheath or full details of the new nomenclature are sebaceous gland. *The keratins of the sebaceous gland are variable in the acini and in different regions published in the Journal of Cell Biology of the duct, and not all keratins have been indicated in the diagram. The majority of the changes apply to (Schweizer et al., 2006). Langbein et al. the keratins of the hair shaft and inner root sheath, which have all been renamed. (2006) use this new nomenclature for the first time, and it is proposed that the new nomenclature should be adopted for all keratins, with an acidic charge, and type molecular biology techniques that finally future keratin publications. II keratins, with a neutral-basic charge, revealed the extent of the diversity of ker- The new keratin nomenclature can be was only just beginning to emerge. The atins required to make a hair shaft. Jürgen viewed as an amendment of the one cre- requirement for equimolar amounts of Schweizer’s group played a large part in ated by Moll et al. (1982), as it tries to both types to be present to form an inter- identifying the eleven type I and six type keep, as far as possible, identical names mediate filament was not yet confirmed. II human genes (Langbein et for the keratins K1–K19, with the excep- The 18 keratins identified were thought al., 1999; Langbein et al., 2001). to be close to the full number present. Finally, inner root sheath keratins It is proposed that The numbering of the keratins was based were the last major group to be discov- this new keratin on the location of each protein after two- ered. The first of these was in mice (Sato nomenclature should dimensional polyacrylamide gel electro- et al., 1999), but it was not immunolocal- phoresis according to molecular weight ized until later (Steinert et al., 2003). In be adopted for all future and isoelectric point. The neutral-basic the meantime, inner root sheath keratins | publications. keratins were numbered from largest to were discovered in sheep and humans smallest, K1–K8, and the acidic keratins (Bawden et al., 2001). There are now tion of K2e, which is now K2. The main were numbered similarly, K9–K19. K11 four type I and four type II keratins identi- changes are to the newer keratins of turned out to be a polymorphic variant fied in the human inner root sheath. The the hair shaft and inner root sheath (the of K10 that migrated faster because of a localization of all the human type II inner changes are summarized in Figure 1). different number of glycine loops in the root sheath keratins has been previously The type I inner root sheath keratins have tail domain (Korge et al., 1992). reported by Langbein and colleagues undergone several name changes by dif- Although keratins of the hair shaft and (Langbein et al., 2003), and, as they ferent groups but have now been des- inner root sheath had been investigated report in this issue (Langbein et al., 2006), ignated K25–K28. There is then a short before 1982, it was not clear that they they have now completed the expression gap before the hair keratins start at K31– were proteins. pattern of the human type I inner root K40. This makes the conversion of hair Hard keratins of the hair shaft are more sheath keratins using both in situ hybrid- keratins previously called Ha1–Ha8 to insoluble than the soft epithelial kera- ization and immunofluorescence micros- K31–K38 easier to remember. Similarly, tins, which makes them more difficult to copy with antibodies raised to peptide the type II hair keratins previously called study (Lynch et al., 1986). It was therefore sequences unique to the four keratins. Hb1–Hb6 are now K81–K86. The type

www.jidonline.org 2367 COMMENTARY

II inner root sheath keratins, previously 83:19–26 Sun T-T (1986) Acidic and basic hair/nail K6irs1–K6irs4, are now named K71–K74 Korge BP, Gan SQ, McBride OW, Mischke (“hard”) keratins: their colocalisation in upper cortical and cuticle cells of the human and are followed by K6hf, the type II D, Steinert PM (1992) Extensive size polymorphism of the human chain hair follicle and their relationship to “soft” keratin located in the companion layer. resides in the C-terminal V2 subdomain due to keratins. J Cell Biol 103:2593–606 K77–K80 are the type II keratins, discov- variable numbers and sizes of glycine loops. Moll R, Franke WW, Schiller DL, Geiger B, Krepler ered by sequencing of the type II kera- Proc Natl Acad Sci USA 89:910–14 R (1982) The catalog of human : tin locus. The gene designation for each Langbein L, Rogers MA, Praetzel S, Cribier patterns of expression in normal epithelial tumors and cultured cells. Cell 31:11–24 keratin carries the same number as the B, Peltre B, Gassler N et al. (2005) Characterization of a novel human type II protein with the prefix KRT. Rogers MA, Edler L, Winter H, Langbein epithelial keratin K1b, specifically expressed L, Beckmann I, Schweizer J (2005) Although we can be almost 100% in eccrine sweat glands. J Invest Dermatol Characterization of new members of the confident that all the human keratins 125:428–44 human type II keratin gene family and a have been identified, enough flexibility Langbein L, Rogers MA, Praetzel S, Winter H, general evaluation of the keratin gene domain in the nomenclature has been provided Schweizer J (2003) K6irs1, 2, 3, and 4 represent on chromosome 12q13.13. J Invest Dermatol the inner root sheath (IRS)-specific type II 124:536–44 to allow for the discovery of new kera- epithelial keratins of the human hair follicle. Sato H, Koide T, Sagai T, Ishiguro S-I, Tamai tins in other mammals. Because keratins J Invest Dermatol 120:512–22 M, Saitou N et al. (1999) The genomic are cell type and differentiation specific, Langbein L, Rogers MA, Praetzel-Wunder S, organisation of type I keratin genes in mice. it is not hard to imagine that the vast dif- Helmke B, Schirmacher P, Schweizer J (2006) Genomics 56:303–9 ferences in morphology of the skin that K25 (K25irs1), K26 (K25irs2), K27 (K25irs3), Schweizer J, Bowden PE, Coulombe PA, Langbein and K28 (K25irs4) represent the type I inner have evolved between mammals (for L, Lane EB, Magin TM et al. (2006) New root sheath keratins of the human hair follicle. consensus nomenclature for mammalian example, elephant, dolphin, armadillo, J Invest Dermatol 126:2377–2386 keratins. J Cell Biol 174:169–74 duck-billed platypus, deer, porcupine) Langbein L, Rogers MA, Winter H, Praetzel S, Steinert PM, Parry DA, Marekov LN (2003) might coincide with a requirement for Beckhaus U, Rackwitz H-R et al. (1999) The Trichohyalin mechanically strengthens the keratins with different structural prop- catalog of human hair keratins. I. Expression hair follicle: multiple cross-bridging roles in of the nine type I members in the hair follicle. erties. Certainly several type II keratins the inner root sheath. J Biol Chem 278:41409– J Biol Chem 274:19874–84 19 have been identified that do not exist in Langbein L, Rogers MA, Winter H, Praetzel S, Winter H, Langbein L, Krawczak M, Cooper humans, including a keratin in the goril- Schweizer J (2001) The catalog of human DN, Jave-Suarez LF, Rogers MA et al. (2001) la and chimpanzee that is redundant in hair keratins. II. Expression of the six type II Human type I hair keratin pseudogene ϕhHaA humans (Winter et al., 2001). members in the hair follicle and the combined has functional orthologs in the chimpanzee catalog of human type I and type II keratins. In the last 24 years, Moll et al. (1982) and gorilla: evidence for recent inactivation J Biol Chem 276:35123–32 of the human gene after the Pan-Homo has been cited more than 4,000 times. A Lynch MH, O’Guin WM, Hardy C, Mak L, divergence. Hum Genet 108:37–42 lot has happened in that time, including many expression studies, the identifica- tion of almost 20 genetically inherited diseases caused by keratin mutations, See related article on pg 2525 the generation of more than 30 trans- genic mice (either keratin knockouts or A New Nail in the CTCL Coffin mice carrying altered keratin genes), and numerous in vitro structural studies and Stuart R. Lessin1 cellular functional studies. For Schweizer et al. (2006) to be cited as often in the The impact of immunotherapy on the natural progression of cutaneous next 24 years will depend on the future T-cell lymphoma (CTCL), particularly the mycosis fungoides and Sézary syn- of research into keratins. Certainly the drome variants, has been based on our evolving understanding of the dis- legacy of the last few years is the ques- ease’s immunobiology. tion: why are there are so many keratins? Journal of Investigative Dermatology (2006) 126, 2368–2369. doi:10.1038/sj.jid.5700558 This question alone should keep scien- tists approaching keratin biology from all angles busy for some time to come. The characterization of the T-helper 2 The Th2 phenotype is associated with (Th2) cytokine phenotype (IL-4, IL-5, IFN signal transduction pathway defects CONFLICT OF INTEREST The author states no conflict of interest. and IL-10 predominance) of the malig- (Sun et al., 1998), thus rendering the nant CD4+ T lymphocytes of cutaneous CTCL cell devoid of endogenous IFN T-cell lymphoma (CTCL) has provided immunoregulation. These observa- REFERENCES Bawden CS, McLaughlan C, Nesci A, Rogers G an immunologic basis for the immune tions provided the rationale leading to (2001) A unique type I keratin intermediate dysfunction correlated with advancing the therapeutic use of IFNs (INF-α and filament gene family is abundantly expressed stages of CTCL (Vowels et al., 1994). INF-γ) in CTCL (Olsen, 2003). Further in the inner root sheaths of sheep and human hair follicles. J Invest Dermatol 116:157–66 1 Hesse M, Zimek A, Weber K, Magin TM (2004) Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania, USA Comprehensive analysis of keratin gene Correspondence: Dr. Stuart R. Lessin, Fox Chase Cancer Center, 333 Cottman Avenue, Philadelphia, clusters in humans and rodents. Eur J Cell Biol Pennsylvania 19111, USA. E-mail: [email protected]

2368 Journal of Investigative Dermatology (2006), Volume 126