See related article on page 715

A Hard Core Look at Rod Packing in the Skin

David S. RubensteinDepartment of Dermatology, University of North Carolina-Chapel Hill, Chapel Hill, North Carolina, USA

Experimental approaches to determine the position andmomentum of an electron by necessity require applyingenergy to the system. In 1927, the physicist Werner Hei-senberg noted that it was not possible to simultaneouslyknow both an electron’s position and momentum becausethe experiment itself alters either the electron’s position ormomentum. In quantum mechanics, the inability to com-pletely define the system due to observation-inducedperturbations of the system under study subsequently be-came known as Heisenberg’s Uncertainty Principle.

Keratinocytes undergo marked structural transitions asthey migrate from the basal layer up through the epidermisand terminally differentiate to form the stratum corneum.To the extent that experimental approaches can often per-turb the system under study, Heisenberg’s uncertainty prin-ciple could be equally well applied to structural studies ofthe stratum corneum cornified envelope. Typical approach-es to study this relatively insoluble structure include use ofdenaturants and subsequent fractionation or imaging tech-niques such as electron microscopy that employ dehydrat-ing protocols during sample fixation. The denaturing and/ordehydrating effects of these preparative approaches maygive rise to artifacts in structure that deviate from the nativesituation in vivo. Despite these limitations, much has beenlearned about the formation of the cornified envelope.

In a calcium-dependent differentiation process, assem-bly of the cornified envelope (for a review, see Kalinin et al,2002) is thought to be initiated by desmosomal nucleationwith subsequent transglutaminase 1-mediated covalent in-corporation into the maturing structure of envoplakin, pe-riplakin, and involucrin. Ceramides, along with fatty acidsand cholesterol, are integral components of the cornifiedenvelope. Fusion of lamellar body granules with the cellmembrane results in increased incorporation of o-(OH)-ceramides in the membrane with subsequent transglutami-nase 1-mediated formation of covalent ester linkages be-tween o-(OH)-ceramides and glutamate residues of thecornified envelope proteins. Cystatin a, elafin, loricrin, andsmall proline-rich proteins act to reinforce the structure.Transglutaminase 3 is thought to catalyze the covalentcross-linking of loricrin to small proline-rich proteins.

In nucleated keratinocytes, keratin intermediate filamentslink desmosomes and hemidesmosomes at the cell surfaceto interior structural elements as the keratin filaments form athree dimensional lattice that traverses the cell interior.Functions of the intermediate filament lattice are structuraland regulatory. Keratin defects have been linked to geneticdiseases of mechanical skin fragility including epidermolysisbullosa simplex (Coulombe et al, 1991) and epidermolytic

hyperkeratosis (Cheng et al, 1992) as well as disordersof differentiation such as steatocystoma multiplex (Smithet al, 1997).

Keratins are also incorporated into the terminally differ-entiated cornified envelope and similarly undergo markedstructural/conformational changes in which they contributeto this structure that functions as a protective barrier tomechanical and environmental perturbations. The maturecornified envelope maintains a remarkable degree of struc-tural flexibility that permits transient deformations of thestructure without mechanical failure and barrier compro-mise. The structure of keratin in the stratum corneum hasbeen somewhat elusive. Several possible orientations forkeratin filament organization within the stratum corneumhave been proposed including (i) random organization, (ii)‘‘sandwich-like’’ organization of planes of filaments witheach plane at defined angles to the other (e.g., plywood),and (iii) three fold axis symmetry that results in isotropicdistribution of mechanical loads.

As the cornified envelope forms during keratinocyte ter-minal differentiation, keratin intermediate filaments undergoa structural transition resulting in the keratin intermediatefilaments transitioning from a three-dimensional cage-likenetwork to a compact lattice like structure. How this struc-tural transition occurs is the subject of a report in thismonth’s issue of the Journal. On p 715, Norlen and Al-Amoudi (2004) present a provocative model in whichphospholipid bilayer membranes within the cell provide atemplate for keratin filament association and compactionduring formation of the terminally differentiated stratumcorneum. The dehydration that accompanies conventionalelectron microscopic fixation protocols results in loss ofmembrane-limited organelles and subcellular structure,leading to potential experiment-induced artifacts that donot reflect the true in vivo structure. To address this prob-lem, Norlen and Al-Amoudi utilized cryotransmission elec-tron microscopy that enabled them to examine fullyhydrated skin sections. Using this approach, they wereable to visualize subcellular organelle and membrane-lim-ited structures not visualized by traditional electron micro-scopic fixation techniques.

In cryopreserved samples, the authors observed hexa-gonally arranged groupings of keratin filaments in patternsthat suggested body-centered cubic rod packing. Suchpacking would allow for uniform distribution of mechanicalstress. The observation that the intermediate filament cyto-skeleton is insoluble in aqueous buffers and the close phys-ical association of keratin intermediate filaments withmembrane lipids suggested to the authors a mechanism

Copyright r 2004 by The Society for Investigative Dermatology, Inc.

ix

for organization of stratum corneum keratin in which keratinintermediate filaments are organized by assembly onto cu-bically symmetric endoplasmic reticulum-like membranestructures. Subsequent removal of the lipid component al-lows for compaction of the keratin filaments such that theymove into close proximity to one another in three-dimen-sional space. This model explains many of the physiologictransitions that occur during keratinocyte terminal differen-tiation and formation of the stratum corneum and addi-tionally makes predictions that can be approachedexperimentally. For example, in vitro reconstitution of keratinintermediate filament cornified envelope-like structurescould be attempted using lipid vesicles as templating sur-faces.

The model suggested by Norlen and Al-Amoudi has po-tential applications beyond understanding the biology of theskin. Their theory of membrane templating and keratin rodcompaction suggests a manufacturing process for the de-sign of intermediate filament-based materials. Membranetemplating may provide an energy-efficient mechanism forkeratin/intermediate filament-based materials with biomed-ical and/or commercial applications. Two- or three-dimen-

sional membrane surfaces could be designed to allow forassembly of intermediate filaments; subsequent controlledremoval of the membrane bilayer would result in definedcompaction of the intermediate filaments to generate amaterial with specific tensile and elastic properties.

DOI: 10.1111/j.0022-202X.2004.23242.x

References

Cheng J, Syder AJ, Yu QC, Letai A, Paller AS, Fuchs E: The genetic basis of

epidermolytic hyperkeratosis: A disorder of differentiation-specific epi-

dermal keratin genes. Cell 70:811–819, 1992

Coulombe PA, Hutton ME, Letai A, Hebert A, Paller AS, Fuchs E: Point mutations

in human keratin 14 genes of epidermolysis bullosa simplex patients:

Genetic and functional analyses. Cell 66:1301–1311, 1991

Kalinin AE, Kajava AV, Steinert PM: Epithelial barrier function: Assembly and

structural features of the cornified envelope. BioEssays 24:789–800, 2002

Norlen L, Al-Amoudi A: Stratum corneum keratin structure, function and forma-

tion-the cubic rod-packing and membrane templating model. J Invest

Dermatol 123:715–732, 2004

Smith FJ, Corden LD, Rugg EL, et al: Missense mutations in keratin 17 cause

either pachyonychia congenita type 2 or a phenotype resembling

steatocystoma multiplex. J Invest Dermatol 108:220–223, 1997

x RUBENSTEIN THE JOURNAL OF INVESTIGATIVE DERMATOLOGY