744 Review Article DOI: 10.1111/j.1610-0387.2009.07033.

Mosaic manifestations of monogenic skin diseases

Peter Itin1, Bettina Burger2
(1) Department of Dermatology, University of Basel, Switzerland
(2) Research Laboratory Dermatology, Department of Biomedicine, Basel, Switzerland

JDDG; 2009 7:744748 Submitted: 8.12.2008 | Accepted: 29.12.2008

Keywords Summary
genetic mosaicism A genetic mosaic is defined as an organism which is composed of genetically
epigenetics different cell lines which originate from a homogeneous zygote. Etiologically,
X-inactivation cutaneous mosaics can be divided into two large categories, epigenetic mosai-
postzygotic mutation cism and genomic mosaicism. Genomic mosaics which have two or more gene-
linear dermatoses tically different cell populations are not inherited with the exception of parado-
minant inheritance pattern. Epigenetic mosaics have a structurally
homogeneous cell population but there are functional differences induced by
modifying factors in the form of gene-steering retroviral elements that can be
inherited. We distinguish five different manifestation patterns of mosaicism, in-
cluding the Blaschko lines pattern, patchy pattern without midline separation,
checkerboard pattern, phylloid pattern and lateralization pattern. All forms of
epigenetic mosaicism, including the various patterns of X-inactivation, appear
to be caused by the action of retrotransposons. A new concept is functional au-
tosomal mosaicism transmittable through the action of retrotransposons.

Introduction
Every disease is a reflection of the inter-
action between gene activities and envi-
ronmental effects. In monogenic disor-
ders the role of gene mutations is more
obvious and predictable in comparison
to polygenic diseases or paradominant
inheritance patterns. The skin and its ap-
pendages is the organ that is most often
involved in monogenic disorders. More
than 1/3 of all monogenic diseases affect
the skin [1]. For this reason, the skin is
predestined to study mosaic states.
Mosaic patterns of the skin
Mosaics on the skin are relatively easy to
recognize due to characteristic patterns
of involvement. Cutaneous mosaics oc-
cur in five different forms of distribu-
tion: the checkerboard patterns as in e. g.
melanosis neviformis Becker or nevus
spilus, the phylloid type in mosaic tri-
somy 13, the patch pattern without mid-
line demarcation as seen in giant congen-
ital melanocytic nevi, the lateralization
pattern in CHILD syndrome and the
Blaschko line pattern in incontinentia
pigmenti or the linear forms of autoso-
mal dominantly inherited mosaics [2].
Rarely a combination of individual patterns
is present in a single person (Figure 1).
Significance and development of
mosaics
In order to understand why every human
is a unique individual, it is important to
realize that the genome is a composite of
innumerable gene segments which unite
to a mosaic-like puzzle [3]. In medical
terminology a genetic mosaic denotes an
organism composed of genetically differ-
ent cell lines but which developed out of
a genetically homogeneous zygote. Ge-
netic alterations (mutations) can occur
in the germline with the first manifesta-
tion of this mutation becoming apparent
in the offspring. Mutations occurring in
up to the 8/16 cell stage can result in
germline as well as somatic mutations.
Mutations outside of the germline as so-
matic mutations are seen more fre-
quently and can lead to circumscribed
malformations, nevi or tumors. Muta-
tions during embryogenesis but also later

JDDG | 9 2009 (Band 7) The Authors Journal compilation Blackwell Verlag GmbH, Berlin JDDG 1610-0379/2009/0709
Mosaics of genodermatoses
Figure 1: Phacomatosis pigmentokeratotica.
Combination of Blaschko-linear pattern with
checkerboard pattern.
in life result in mosaic states. In tumors
the secondary mutations occur during
the course of life. Genetic mosaics are
usually not inherited with the exception
of paradominant inheritance (e. g. Klip-
pel-Trenaunay syndrome). In epigenetic
mosaics all cells possess the structurally
identical genome, but due to the effects
of steering genes functionally diverse
clones develop. These functional mosaics
are heritable. Postzygotic mutations may
occur in nuclear or mitochondrial DNA
or epigenetic factors can manifest in the
form of a functional mosaic, e. g. in the
form of X-inactivation. Retrotrans-
posons can influence the activity of
neighboring genes via demethylation or
methylation either in terms of activation
or deactivation. A chimera is an organ-
ism containing different cell clones that
derive from two or more zygotes. This
does not correspond to the proper defi-
nition of mosaic.
Functional mosaicism due to random
X-inactivation
Mosaics caused by X-inactivation are
also termed functional forms of mo-
saicism. X-inactivation in the female or-
ganism is regulated by the gene XIST, a
functional RNA that does not code for a
protein [4]. X-inactivation occurs during
the blastocyst stage and random distribu-
tion of both differing cell clones results
in the great phenotypic variability of X-
linked diseases in females.
The Authors Journal compilation Blackwell Verlag GmbH, Berlin JDDG 1610-0379/2009/0709
Epigenetics
Decoding individual variability is a
burning issue in current molecular ge-
netics [5]. The human genome project,
which had the aim of decoding the hu-
man genome, has shown that the entire
genome contains innumerable segmental
duplications that have been transferred
from one chromosome to another just
like a backup system. This explains why
similar disorders can be located on di-
verse gene loci in different families [6].
In recent years the study of epigenetics
has moved to the epicenter of modern
medicine. Here the individual genetic
background, environmental factors, the
aging process and disease pathogenesis
interact [7]. It is noteworthy that cellular
alterations can be remembered by the
mentioned processes and influence the
next cell generation in the sense of repro-
gramming gene activities. Knowledge of
these epigenetic processes is important
for understanding disease-modifying fac-
tors and on the other hand for rational
prevention and therapy. In a study by
Fraga et al. [8] it was shown that in 3-
year-old identical twins the gene activity
pattern was practically identical. At the
age of 50 years, identical twins demon-
strated distinct differences in gene activ-
ity, which can be attributed to epigenetic
factors.
Epidermal stem cells and their
significance in mosaicism
Dermatologists are visually oriented
people and their attention is often
drawn by special distribution and man-
ifestation patterns. The cutaneous pat-
tern can indicate special genetic inheri-
tance forms, for example, monogenic
mosaicism. We refer to a clonal ectoder-
mal unit that is supplied by an ectoder-
mal stem cell and follows a distribution
pattern along the Blaschko lines.
Within this clonal ectodermal unit epi-
dermal stem cell units are found that
supply a smaller area within which the
actual epidermal proliferation unit is lo-
cated. This consists of a stem cell that
supplies about 10 basal cells of the epi-
dermis and controls their subsequent
differentiation. Depending on the point
of time that localized postzygotic muta-
tion occurs, either the clonal ectoder-
mal unit or the epidermal stem cell unit
or even only the epidermal proliferation
unit is affected resulting in different
clinical features [9].
Review Article 745
Monogenic cutaneous mosaics
Cutaneous mosaic manifestations in
monogenic skin diseases can result from
postzygotic mutations of an autosomal
dominant gene that is either lethal and
can only survive as a mosaic (McCune-
Albright syndrome, Proteus syndrome,
epidermal nevus syndrome etc.) or is
known as a non-lethal form (neurofibro-
matosis, tuberous sclerosis etc.) [10].
Further, X-linked dominant mutations
can occur as functional mosaics and re-
sult in the phenotypes of focal dermal
hypoplasia, incontinentia pigmenti,
Conradi-Hnermann-Happle syndrome,
CHILD syndrome etc. Other X-linked
recessive skin diseases can cause skin ab-
normalities in the form of a functional
mosaic in heterozygous female carriers. A
form of genetic mosaicism known only
for a short time is reversion mutation of
genetic diseases. This is so to say a natural
form of gene therapy as has been docu-
mented repeatedly for epidermolysis bul-
losa, for example, but also for other dis-
orders with cutaneous involvement [11].
Autosomal dominant disorders
Autosomal dominant disorders can also
form mosaics and display segmental
manifestations. We differentiate a type 1
segmental manifestation in diseases in-
herited in an autosomal dominant fash-
ion, with this manifestation type being
characterized by a circumscribed linear
expression of a phenotype inherited in
an autosomal fashion. The remaining
skin is normal and molecular genetic
methods detect no mutation. The muta-
tion is only found in the altered segmen-
tal zone. Recently Happle described a
second form of segmental manifestation
in disorders inherited in an autosomal
dominant fashion and termed this as
type 2 form [12]. In this situation a
germline mutation for an autosomal
dominant disorder exists and in addition
a second postzygotic mutation occurs, so
that there is a loss of heterozygosity (Fig-
ure 2 4). Clinically, there is a more pro-
nounced manifestation in the affected
zone and molecular genetic examination
reveals a loss of heterozygosity in this
segmental area. This concept has, in the
meantime, been confirmed clinically as
well as with molecular genetic methods
in several diseases. A mechanism for the
loss of heterozygosity appears to be cross-
ing-over which corresponds to the con-
cept of twin spotting. We observed a
JDDG | 9 2009 (Band 7)
746 Review Article Mosaics of genodermatoses

patient with Darier disease with type 2

segmental manifestation where on the
back segments with excessive expression
alternated with hardly involved skin
areas documenting a crossing-over
mechanism as suggested in the patho-
genesis of twin spotting [13] (Figure 3).
In 2004 it was possible for the first time
to prove type 2 segmental manifestation
in autosomal dominant disorders in the
sense of the described concept by mole-
cular genetic methods [14].
It is assumed today that giant caf-au-lait
macules in neurofibromatosis 1 also do-
cument a type 2 segmental inheritance
form [15] (Figure 4). The same holds
true for plexiform neurofibromas where
the loss of heterozygosity has already
been documented. In 2007 it was shown
Figure 2: (a, b) Segmental Recklinghausen disease with type 2 manifestation (numerous neurofibromas that a loss of heterozygosity appears in
exclusively in a segmental distribution in a patient with several caf-au-lait spots also contralaterally. neurofibromas of patients with Reck-
linghausen disease and that a second
hit within the framework of a postzy-
gotic mutation can be found in the
melanocytes of the caf-au-lait macules
[16]. Segmental type 2 manifestations of
a PTEN mutation are found in Cowden
syndrome [17]. In the past this entity
was termed Proteus syndrome with
epidermal nevus and later diagnosed as
SOLAMEN syndrome standing for seg-
mental overgrowth, lipomatosis, arteri-
ovenous malformation and epidermal
nevus [18]. Happle and coworkers are
convinced that Proteus syndrome can
never correspond to a germline muta-
tion, but must always represent a
postzytgotic mutation. The observation
of identical twins where one develops
Proteus syndrome, but not the other,
supports this hypothesis [19]. Mosaic
Figure 3: Darier disease with type 2 manifestation. states of keratin 1 gene mutation lead to
epidermolytic hyperkeratotic nevi along
the Blaschko lines. The germline muta-
tion of keratin 1 results in ichthyosis
with palmoplantar keratoderma. Arin
and Roop showed that chimeric mice
with a keratin 10 mutation displayed
persistence of hyperkeratosis, while
chimeric mice with a keratin 14 or 5 mu-
tation with resulting blistering healed
spontaneously. This fact leads to the con-
clusion that a keratin 10 mutation is not
a negative selection factor in the mosaic
state, while this is so in a keratin 5 or 14
mutation [20].

Linear genodermatoses in females

X-linked dominant disorders lead to lin-
Figure 4: Type 2 distribution of Recklinghausen disease in a patient with giant caf-au-lait spot. ear cutaneous manifestations that follow

JDDG | 9 2009 (Band 7) The Authors Journal compilation Blackwell Verlag GmbH, Berlin JDDG 1610-0379/2009/0709
Mosaics of genodermatoses
the skin development lines described by
Blaschko. These include incontinentia
pigmenti, focal dermal hypoplasia, X-
linked dominant chondrodysplasia
punctata, MIDAS syndrome, CHILD
syndrome and oral-facial-digital syn-
drome. The responsible gene has been
described for all of these entities. Steijlen
and coworkers [21] have shown that the
variability of the phenotype in Conradi-
Hnermann-Happle syndrome is due to
genetic mosaicism. Shirahama et al. [22]
reported that the random and at times
extreme lyonization is responsible for the
intrafamilial variability of the phenotype
in Conradi-Hnermann-Happle syn-
drome. Recently the gene responsible for
focal dermal hypoplasia was identified as
a mutation in PORCN. This gene is an
important regulator of the Wnt cascade
which plays a central role in embryology
[23]. The group from Marburg has doc-
umented a family with CHILD syn-
drome over three generations with the
phenotype being highly variable within
this family. The disease is caused by a
mutation in the NSDHL gene that is of
significance in cholesterol metabolism
[24]. In incontinentia pigmenti Bloch-
Sulzberger the mutated NEMO gene
causes a disturbance of the NF-kappa B
cascade with consecutive effects on em-
bryogenesis and cellular apoptosis.
Linear manifestations are also found
among X-linked recessive dermatoses.
Especially the female carriers have seg-
mental manifestations of the disease,
such as in IFAP syndrome (ichthyosis
follicularis with atrichia and photopho-
bia) [25]. In hypohidrotic ectodermal
dysplasia also female carriers display hy-
pohidrotic zones following the Blaschko
lines that can be visualized by the starch
iodide test.
In the past years it was shown that epi-
dermal nevi are the result of a postzy-
gotic mutation in the fibroblast growth
factor receptor 3. It is remarkable that
the same phenomenon can also occur in
adulthood and lead to seborrheic warts
[26]. Recently it was shown for the first
time that an unusual epidermal nevus
syndrome with cerebral involvement can
also be caused by a mutation in the fi-
broblast growth factor receptor 3 [27].
The fact is interesting that somatic mo-
saicism also can lead to full-blown fea-
tures of autosomal dominant disorders,
where molecular genetic analyses no
longer possess 100% probability of suc-
The Authors Journal compilation Blackwell Verlag GmbH, Berlin JDDG 1610-0379/2009/0709
cess. For example in a patient with
Costello syndrome the mutation was
only detected in the buccal mucosa but
not in the blood [28]. This is of signifi-
cance for future molecular genetic dia-
gnostics as negative mutation analyses of
the blood cannot exclude a somatic muta-
tion. Patients also exist that can develop
several independent second-site mutations
during embryogenesis as has been docu-
mented in Wiskott-Aldrich syndrome [29].
Reversion of mutations was described as
a form of natural gene therapy by
Jonkman and has been confirmed again
and again in the meantime [11]. This
phenomenon is well-studied in Wiskott-
Aldrich syndrome and has been con-
firmed in Fanconi anemia.
Mosaicism in monogenic skin disorders
reflect how the skin can react to gene-
modifying factors during the postzygotic
and postnatal periods. Understanding of
the discussed facts is not only helpful for
the interpretation of clinical phenotypes,
but also can be considered in the goals of
gene therapy.
Conflicts of interest
None.
Correspondence to
Prof. Dr. Peter Itin
Chairman Dermatology
University Hospital
Petersgraben 4
CH-4031 Basel, Switzterland
Tel.: + 41-61-265-4084
Fax: +41-61-265-4885
E-mail: pitin@uhbs.ch
References
1 Scriver CR, Beaudet AL, Valle D, Sly
WS. The metabolic basis of inherited
disease. 8th ed. New York: McGraw-
Hill, 2001.
2 Happle R. Dohi Memorial Lecture.
New aspects of cutaneous mosaicism. J
Dermatol 2002; 29: 681692.
Review Article 747
3 Pbo S. The mosaic that is our ge-
nome. Nature 2003; 421: 409412.
4 Sun BK, Tsao H. X-chromosome inac-
tivation and skin disease. J Invest Der-
matol 2008; 128: 27532759.
5 Pennisi E. Breakthrough of the year.
Human genetic variation. Science
2007; 318: 18421843.
6 Subramanian G, Adams MD, Venter
JC, Broder S. Implications of the hu-
man genome for understanding human
biology and medicine. JAMA 2001;
287: 22962307.
7 Feinberg AP. Epigenetics at the epicen-
ter of modern medicine. JAMA 2008;
299: 13451350.
8 Fraga MF, Ballestar E, Paz MF, Ropero
S, Setien F, Ballestar ML, Heine-Suer
D, Cigudosa JC, Urioste M, Benitez J,
Boix-Chornet M, Sanchez-Aguilera A,
Ling C, Carlsson E, Poulsen P, Vaag A,
Stephan Z, Spector TD, Wu YZ, Plass
C, Esteller M. Epigenetic differences
arise during the lifetime of monozygo-
tic twins. Proc Natl Acad Sci USA
2005; 102: 1060410609.
9 Strachan LR, Ghadially R. Tiers of clo-
nal organization in the epidermis: the
epidermal proliferation unit revisited.
Stem Cell Rev 2008; 4: 149157.
10 Itin PH, Buechner SA. Segmental
forms of autosomal dominant skin dis-
orders: the puzzle of mosaicism. Am J
Med Genet 1999; 85: 351354.
11 Pasmooij AM, Pas HH, Bolling MC,
Jonkman MF. Revertant mosaicism in
junctional epidermolysis bullosa due to
multiple correcting second-site mutati-
ons in LAMB3. J Clin Invest 2007;
117: 12401248.
12 Happle R. A rule concerning the seg-
mental manifestation of autosomal do-
minant skin disorders. Review of clini-
cal examples providing evidence for
dichotomous types of severity. Arch
Dermatol 1997; 133: 15051509.
13 Itin PH, Happle R. Darier disease with
paired segmental manifestation of eit-
her excessive or absent involvement: a
further step in the concept of twin spot-
ting. Dermatology 2002; 205: 344
347.
14 Poblete-Gutirrez P, Wiederholt T, K-
nig A, Jugert FK, Marquardt Y, Rbben
A, Merk HF, Happle R, Frank J. Allelic
loss underlies type 2 segmental Hailey-
Hailey disease, providing molecular
confirmation of a novel genetic con-
cept. J Clin Invest 2004; 114: 1467
1474.
JDDG | 9 2009 (Band 7)
748 Review Article
15 Yang CC, Happle R, Chao SC, Yu-Yu
Lee J, Chen W. Giant caf-au-lait ma-
cule in neurofibromatosis 1: a type 2
segmental manifestation of neurofibro-
matosis 1? J Am Acad Dermatol 2008;
58: 493497.
16 Maertens O, De Schepper S, Van-
desompele J, Brems H, Heyns I, Jans-
sens S, Speleman F, Legius E, Messiaen
L. Molecular dissection of isolated di-
sease features in mosaic neurofibroma-
tosis type 1. Am J Hum Genet 2007;
81: 243251.
17 Happle R. Linear Cowden nevus: a new
distinct epidermal nevus. Eur J Derma-
tol 2008; 17: 133136.
18 Caux F, Plauchu H, Chibon F, Faivre L,
Fain O, Vabres P, Bonnet F, Selma ZB,
Laroche L, Grard M, Longy M. Seg-
mental overgrowth, lipomatosis, arte-
riovenous malformation and epidermal
nevus (SOLAMEN) syndrome is rela-
ted to mosaic PTEN nullizygosity. Eur
J Hum Genet 2007; 15: 767773.
19 Brockmann K, Happle R, Oeffner R,
Knig A. Monozygotic twins discor-
dant for Proteus syndrome. Am J Med
Genet 2008; 146A: 21222125.
20 Arin MJ, Roop DR. Inducible mouse
models for inherited skin diseases: im-
plications for skin gene therapy. Cells
Tissues Organs 2004; 177: 160168.
JDDG | 9 2009 (Band 7)
21 Steijlen PM, van Geel M, Vreeburg M,
Marcus-Soekarman D, Spaapen LJ, Ca-
stelijns FC, Willemsen M, van Steensel
MA. Novel EBP gene mutations in
Conradi-Hnermann-Happle syn-
drome. Br J Dermatol 2007; 157:
12251229.
22 Shirahama S, Miyahara A, Kitoh H,
Honda A, Kawase A, Yamada K, Mabu-
chi A, Kura H, Yokoyama Y, Tsutsumi
M, Ikeda T, Tanaka N, Nishimura G,
Ohashi H, Ikegawa S. Skewed X-chro-
mosome inactivation causes intra-fami-
lial phenotypic variation of an EBP
mutation in a family with X-linked do-
minant chondrodysplasia punctata.
Hum Genet 2003; 112: 7883.
23 Grzeschik KH, Bornholdt D, Oeffner
F, Knig A, del Carmen Boente M, En-
ders H, Fritz B, Hertl M, Grasshoff U,
Hfling K, Oji V, Paradisi M, Schuch-
ardt C, Szalai Z, Tadini G, Traupe H,
Happle R. Deficiency of PORCN, a re-
gulator of Wnt signaling, is associated
with focal dermal hypoplasia. Nature
Genet 2007; 39: 833835.
24 Bittar M, Happle R, Grzeschik KH,
Leveleki L, Hertl M, Bornholdt D, K-
nig A. CHILD syndrome in 3 generati-
ons: the importance of mild or minimal
skin lesions. Arch Dermatol 2006; 142:
348351.
The Authors Journal compilation Blackwell Verlag GmbH, Berlin JDDG 1610-0379/2009/0709
Mosaics of genodermatoses
25 Knig A, Happle R. Linear lesions re-
flecting Lyonization in women heterozy-
gous for IFAP syndrome (ichthyosis fol-
licularis with atrichia and photophobia).
Am J Med Genet 1999; 85: 365368.
26 Hafner C, van Oers JM, Vogt T, Landt-
haler M, Stoehr R, Blaszyk H, Hofstaed-
ter F, Zwarthoff EC, Hartmann A. Mo-
saicism of activating FGFR3 mutations
in human skin causes epidermal nevi. J
Clin Invest 2006; 116: 22012207.
27 Garca-Vargas A, Hafner C, Prez-Ro-
drguez AG, Rodrguez-Rojas LX,
Gonzlez-Esqueda P, Stoehr R,
Hernndez-Torres M, Happle R. An
epidermal nevus syndrome with cere-
bral involvement caused by a mosaic
FGFR3 mutation. Am J Med Genet
2008; 146A: 22752279.
28 Gripp KW, Stabley DL, Nicholson L,
Hoffman JD, Sol-Church K. Somatic
mosaicism for an HRAS mutation cau-
ses Costello syndrome. Am J Med Ge-
net 2006; 140: 21632169.
29 Boztug K, Germeshausen M, Avedillo
Dez I, Gulacsy V, Diestelhorst J, Ball-
maier M, Welte K, Mardi L, Chernys-
hova L, Klein C. Multiple independent
second-site mutations in two siblings
with somatic mosaicism for Wiskott-
Aldrich syndrome. Clin Genet 2008;
74: 6874.