pluslighttherapy
Enzyme
to repairDNAdamage
in
ultravioIet-B-irradiated humanskin
Helger Stege*, Len Rozat, Arie A. Vinkt, Markus Grewe*, Thomas Ruzicka*, SusanneGrether-Beck*,
and JeanKrutmann**
*Cfinicaland ExperimentalPhotodermatology,Departmentof Dermatology,HeinrichHeine University,D-4O225
Düsseldorf,Germany;tskin EffectsTesting
and Photobiology.Netherlandscentral Organizationfor Applied ScientificResearchNutrition and Food ResearchInstitute,3700 AJ Zeist,The Netherlands
Communicatedby RichardB. Setlow, BrookhavenNational Laboratory Upton, NY, December6, 1999(receivedfor review September10, 1999)
Ultraviolet-B (UVB) (290-320 nm) radiation-induced cyclobutane pyrimidine dimerswithin the DNA of epidermalcellsare detrimentalto
human heahh by causing mutations and immunosuppressiveeffects
that presumably contribute to photocarcinogenesis.Conventional
photoprotection by sunscreensis exclusively prophylactic in nature
and of no value once DNA damage has occurred.In this paper, we
havethere{ore assessedwhether it is possibleto repair UVBradiationinduced DNA damagethrough topical applicationof the DNA-repair
enzyme photof yase,derived f rom Anacystisnidulans,that specifically
converts cyclobutane dimers into their original DNA structure after
exposure to photoreactivating light. When a dose of UVB radiation
sufficientto induce eryrthemawas administered tothe skin of healthy
subjects,significant numbers of dimers were formed within epidermal cells. Topical application of photolyase-containing liposomesto
UV&irradiated skin and subsequent exposure to photoreactivating
light decreased the number of UVB radiation-induced dimers by
4o-45%. No reduction was observed if the liposomeswere not filled
with photolyase or if photoreactivating exposure preceded the application of filled liposomes.The UVB dose administered resulted in
suppressionof intercellular adhesion molecule-1(ICAM-1),a molecule
requiredfor immunity and inflammatory events in the epidermis.In
addition, in subjectshypersensitiveto nickel sulfate, elicitation of the
hypersensitivity reaction in irradiated skin areas was prevented.
Photolyase-induced dimer repair completely prevented these UVB
radiation-induced immunosuppressive effects as well as erythema
and sunburn-cell formation. These studies demonstrate that topical
application of photolyase is effective in dimer reversal and thereby
leads to immunoprotection.
fl ecause of increased leisure time, the growing popularity of
IJslayint outdoors, and of holidays in the sun, it has become
more and more important to study the molecular and cellular
effects that ultraviolet radiation exposure to the sun exerts on
human skin (1). In this context, a critical observation has been
the appreciation that the induction of DNA damage by ultraviolet-B (UVB) radiation (290-320 nm) is detrimental to human
health. Among the DNA lesions induced by UVB irradiation,
cyclobutane pyrimidine dimers predominate. Dimer formation is
thought to be of crucial importance for the initiation of skin
cancer, because it was found to be closely linked to the generation of mutations in tumor-suppressor genes expressed in
UV-induced skin cancer (2). From recent animal work, it has
been concluded that dimers also contribute to photocarcinogenesis through suppression of the skin's immune system, allowing
transformed cells to grow unimpeded (3, 4). Strategiesdirected
at the prevention of detrimental effects resulting from dimers in
human skin are, thus, of paramount concern for human health.
Sunscreens provide protection against erythema of human
skin through absorption or reflection of UV radiation (5). This
widely used photoprotective approach, however, is not effective
once damage to skin cells has been generated after sun exposure.
In healthy human skin, cyclobutane pyrimidine dimers are
repaired through nucleotide excision repair. In addition to
nucleotide excision repair, cyclobutane pyrimidine dimers in
some biological systems may be effectively removed from cel1790-1795| PNASI february15,2000| vol.97 | no 4
lular DNA by the DNA-repair enzymephotolyase(6, 7). Photolyase recognizesand specificallybinds to cyclobutane pyrimidine dimers, and exposureof the photolyase-dimer complex
to photoreactivatinglight (300-500 nm) convertsthe dimerized
pyrimidinesto their monomericform. Dimer-specificphotolyase
is presentin an activeform in numerousprokaryotesand certain
eukaryotes,including fish and marsupials.
The existenceof photoreactivationin humansis controversial.
Although recentreportshavepresentedevidenceofphotoreactivating activity in human white blood cells (8) and fibroblasts
(9), many negativereports havequestionedthe validity of these
findings(ref. 10,seereviewin ref. 1L).The recentdiscoveryof
a human homologue of a Drosophila photolyase gene with
unknown function (I2, L3) has again raised the possibilityof
photoreactivationin human skin.
In this paper, we have assessedwhether dimer repair in
UVB-irradiated human skin can be enhancedthrough topical
applicationof a cyclobutanepyrimidine dimer-specificphotolyase that has been prepared from Anacystis nidulans and
in
encapsulatedinto liposomes.Becauseimmunosuppression,
contrastto mutagenesis,
canbe detectedin UVB-irradiatedskin
within hours after UVB radiationexposure,we alsowantedto
determine whether this approachwas capable of providing
immunoprotection.
Materials and Methods
The local ethical committee (Ethikkommission
HumanSubjects,
der Medizinischen Einrichtungen der Heinrich-HeineUniversität,Düsseldori Germany)approvedthe protocol,advertisements,and consentdocumentfor the study. Nineteen
healthy,adult humanvolunteerswere studied.Individualswere
of skin type II or III (14),without a historyof chronic disease,
and not currently on medication.
UVB Radiation.For administration of biologically equivalent
doses,the individual minimal erythemadose (MED) was determined for each subject 24 h after irradiation by exposing
previouslyunirradiatedbuttock skin to increasingdosesof UVB
radiation from a UVB phototherapydevicecontainingsix FS-40
bulbs (WestinghouseElectric, Pittsburgh),which are known to
emit primarily in the UVB range (280120 nm). The UVB
output wasmonitoredby meansof an IL1700researchradiometer and SEE240 UVB photodetector (International Light,
Newburyport, MA) and was approximately0.24 mWf cm2at a
Abbreviations:ICAM-1,intercellularadhesion molecule-l; i.c, intracutaneously;MED,
minimal erythemadose;rh, recombinanthuman; UVB,ultraviolet-8.
To whom reprint requests should be addressedat: Clinical and Experimental Photodermatology, Department of Dermatology, Heinrich Heine University, Moorenstrasse 5,
0-40225 Düseldorf, Germany. E-mail:[email protected].
The publication costs of this article were defrayed in part by page charge payment. This
article must therefore be hereby marked "adverfisement" in accordancewith 18 U.s.C.
5 | 734 solely to indicate this fact.
Articfe pubfished online before print Proc- Natl. Acad. ki usA, 10.1073/pnas.030528697.
pnas.org/cai/doi/10.1073/pnas.030528897
Article and publiGtion date are at w
source-to-skindistanceof 22 cm.For all individualsincludedin
the study,the MED was in the normal range (30-50 mJfcmz
UVB), and in particular,no abnormalreactionto UVB radiation
was observed.To assessthe effect of UVB radiation on intercellular adhesionmolecule-1(ICAM-1) expressionin human
keratinocytes,previouslyunirradiated buttock skin (16 cmz)
subsequentlywas exposed to 1 MED of UVB radiation or
sham-irradiated.Except for the I6-cmzexposureareas,all areas
of the body were draped.
lmmunostaining
of Dimersin HumanSkinSections.
Biopsieswere
taken as indicated,either immediatelyafter photoreactivationor
24 h after UVB irradiation,and skin sampleswere embeddedin
paraffin.Sections(5 g.m)were deparaffinizedby subsequent
2-min
incubationsin rylene (two times),l00Vo ethanol(two times),96%
ethanol,70Voethanol,and PBS.The slideswerethen boiledfor L0
min in 10mM citratebuffer (pH 6.0),rinsedwith PBS(two times),
and usedfor immunostaining.The skin sectionswere stainedwith
monoclonal antibodies against cyclobutane thymine dimers (hybridoma clone II3) and goat anti-mouseIgG fluorescein-labeled
secondaryantibodies.The H3 mouse monoclonalantibody has
beendevelopedagainsta TT-containingoligonucleotide(15) and
hashigh affinity for 5'-T-containingdimers,e.g.,TT and TC (L.R.
and A.A.V., unpublisheddata). Thesephotoproductsare representativefor the cyclobutanepyrimidine dimers inducedby UV
radiation.Nuclei of skin cellswere counterstainedwith propidium
iodide. Nucleargreenfluorescencein ttre epidermalcellsproporwith a scanning
tional to the level of thymine dimers was assessed
laser microscope(Zeiss I,SM-41) by usingimage processingand
image analysis.The procedurefor the immunostainingand measurementof the fluorescencehasbeen described(15, 16). Photographsweretakenon 35-mm400ASA KodakElite chromefilm for
color slides.
For invivo photoreactivation,liposomes
ln VivoPhotoreactivation.
containing functionally active DNA-photolyasewere employed
(Photosomes,a gift from Applied Genetics, Freeport, NY).
Briefly, DNA-photolyase was prepared from Escherichiacoli
expressingthe photolyasefrom Anacystisnidulans and subsequentlyincorporatedinto liposomes.The liposomecomponent
eggphoswas formed from the lipids egg phosphatidylcholine,
phatidylethanolamine,
and oleic acid, and from the membrane
stabilizercholesterolhemisuccinate.
Photolyase-containingliposomes were applied immediately
after UVB irradiation as a lotion [PhotosomeDaytime Formula
without sunscreen(1% DNA-photolyase-containingliposomes
in a 1Vo hydrogel)] onto UVB-irradiated human skin (0.5
mllcmz) for 60 min to achieve sufficient penetration of the
DNA-repair enzyme into human skin. The treated skin area
subsequentlywas covered with aluminum foil to prevent photoreactivationat this time point. After t h, the aluminum foil was
radiation
removedand the skinwasexposedto photoreactivating
from a UVASUN 3000 irradiation device (340-450 nm, maximum at 365 nm; Muzhas,Munich, Germany)for 30 min at 9.6
mW/cm2. As controls, selectedUVB-irradiated skin areasfirst
were exposedto photoreactivatingradiation and subsequently
were treatedwith Photosomelotion as described.To limit the
degreeof spontaneousphotoreactivationto a minimum, Photosome-treatedhuman skin was either kept in the dark or, for
obtainingbiopsies,exposedto yellow light only.
on the surfaceof most keratinocytes24 h after application(17).
At the indicatedtime points, 6-mm punch biopsieswere obtained,snapfrozen immediately,and storedin liquid nitrogen.
For immunohistochemicaldetectionof keratinocyteICAM-I
wereembeddedin Optimum
surfaceexpression,
frozenspecimens
Cutting Medium (OCT, Mile$, and 5-pcmserial sectionswere
preparedusinga Cryocut 2000(Reichert& Jung,Nußbach,Germany).For immunohistochemistry,
the labeledstreptavidin-biotin
methodwasused,employinga commerciallyavailablekt (Dako)
and a L:400dilution of monoclonalanti-ICAM-l antibody84H10
(mIgG1;Immunotech,Marseille,France)or an equivalentconcentration of a mlgGl isotypecontrol antibody(Immunotech).After
staining, the slides were mounted with aqueousmounting gel
(Permafluor,Immunotech).Positivestainingwasidentifiedby light
microscopyas brown reactionproducts.
PatchTesting.Three male subjects(mean age, 32+12 years)
hypersensitiveto nickel sulfate (NiSO+), as proven by earlier
patch testing,were enrolled. Irradiated (2 MED) and noniradiatedskin siteswere alwaystestedin parallel.For the evaluation
reactions,nickel sulfatewasapplied
of delayedhypersensitivity
immediately after UVB irradiation to the skin of the back by
usingFinn chambers(18).The testchamberswereremoved24 h
after application,and the test reactionwas evaluated48 h later
and scoredas positiveif there were erythema,infiltration,and
papules.From each test site, 4-mm punch biopsieswere obtained, fixed in formalin, embeddedin paraffin, stainedwith
hematoxylin-eosin,and subsequentlyhistologically evaluated.
assessment
of lymphocytesin the epiderFor semiquantitative
mis and dermis,three serial sectionsper specimenwere analyzed, and the number of lymphocytesin three high-power
(x200) view fieldswas counted(19).
Results
PyrimidineDimersfrom UVB-lrradiated
Removalof Cyclobutane
Clclobutanepyrimidinedimer
HumanSkinby Photoreactivation.
formation wasstudiedin previouslysunprotectedhumanbuttock
skin 24 h after exposureto 1 MED of-UVB radiation(Fig. 1).
Significantcyclobutanepyrimidine dimer formation was observedin biopsies,which were obtainedfrom UVB-irradiated
skin areas(Fig. 1b), but could not be detectedin unirradiated
skin areas(Fig. Ia). The administrationof photolyaseencapsuIated in liposomesto UVB-irradiated human skin and the
subsequentexposureto photoreactivatingradiation reducedthe
number of UVB radiation-inducedcyclobutanepyrimidine
dimers within the epidermis (Figs. lc and 28.).In contrast, no
significant decreasein the number of epidermal cyclobutane
pyrimidinedimerswas observedin UVB-irradiatedskin areas,
which had first been exposedto photoreactivatingradiation and
subsequentlytreated with photolyase(Figs. Ld and2g).Thus,we
did not detectendogenousphotoreactivationunder conditionsin
which exogenousphotoreactivatingenzymewas active in irradiatedskin.When the time of exposureto photoreactivatinglight
increasein the
wasincreasedfrom 0 to 30 min, a time-dependent
removal of cyclobutanepyrimidine dimers was observed (Fig.
1e).This effect was maximal after a 30-min exposureto photoreactivatinglight; by this time 40-45Vo of cyclobutanepyrimidine dimers were removed from UVB-irradiated human skin.
Therefore, a 30-min photoreactivationperiod was used for all
further experiments.The efficacy of photolyase-induceddimer
removal was essentiallyidentical immediately after (Fig. le,
*42Vo) or 225 h after (Fig. 29, :44%) photoreactivation,
indicating that the repair effect was alreadymaximal at the end
of the photoreactivatingperiod.
To induce the exAssessment
of KeratinocyteICAM-IExpression.
pressionin keratinocytesof ICAM-I, sham-irradiated
or UVBirradiated areas of buttock skin were intracutaneously (i.c.)
injectedwith 38,000units of recombinanthuman (rh) IFN-7
of
Suppression
PreventsUVBRadiation-lnduced
(a gift from Dr. Rentschler, Laupenheim, Germany; specific Photoreactivation
We next asked
KeratinocyteICAM-IExpression.
IFN-y-Mediated
activity,I.2 x t07 units/mg). This dosewas appliedbecauseit
previouslyhad beenshownto induceintenseICAM-I expression whether the induction of cyclobutanepyrimidine dimers after
Stege et a/.
P N A S I F e b r u a r1
y5,2000 | vol.97 | no.4 | 1791
T
=
ct
o
I
c
o
(J
o
at
o
l!-
100
90
80
70
60
50
40
30
20
10
0
Fig. 1. Removalof pyrimidinedimersfrom UvB-irradiatedhumanskinthrough
photoreactivation.(a-d) Human buttock skin was left untreated (a), UVBirradiated( 1 M ED)(b),UVB-irradiated
andsubsequentlytreated
with photolyasecontainingliposomesfollowed by a 30-minexposureto photoreactivatinglight
(c), or UVB-|rradiatedand subsequentlyexposedto photoreactivatinglight
for 30 min and then treated topicallywith photolyase-containing
liposomes(@.
The presenceof cyclobutanepyrimidinedimerswas assessed
22.5h after photoreactivationby immunofluorescence
microscopywith mAb H3 as describedin
Mateilals and Mefhods. Data representone of sevenessentiallyidentical experiments. (e) Dose responseof photoreactivation-induced
dimer removal from
UVB-irradiatedhuman skin.To assess
the dose responsebetween photoreactivation and dimer removal,humanbuttockskinwaseither left unirradiated(A)or
exposedto 2 MED of UVB radiationand subsequentlyleft untreated(B),UVBirradiatedand treated with photolyase-containing
liposomesplus photoreactivating light for 1 min (C), 5 min (D), or 30 min (E).The presenceof dimersin
1792 | m,pnas
org
exposureto 1 MED of UVB radiation was capableof exerting
immunosuppressive
effectsin human skin. For this purpose,we
assessed
the capacityof UVB radiationto suppresslFN-7-induced
keratinocyteICAM-I expressionin vivo in human skin. Buttock
skin was exposedto 1 MED of UVB radiation,subsequently
i.c.
injectedwith rh IFN-7, and,24h later,IFN-y-inducedkeratinocyte
ICAM-I protein expressionwas assessed
by immunohistochemistry. In controlbiopsyspecimens
obtainedfrom vehicle-injected
skin
or UVB-irradiatedskin, positivestainingfor ICAM-I waslimited
to dermal endothelialcells(Fig. 2 a andc) aspreviouslydescribed
(17).Injectionof rh IFN-7 inducedintenseICAM-I expression
on
the surfaceof almostall keratinocytes,and stainingwas maximal
along the basal cell layer (Fig. 2ö). Injection of the identical
concentration of rh IFN-7 into a buttock skin area of the same
individual,whichprior to cytokinestimulationhad beenexposedto
UVB radiation,did not induceor onlyweaklyinducedkeratinocyte
ICAM-1 surfaceexpression(Fig. U). Thesedata indicatethat rn
vrvo exposureof human skin to 1 MED of UVB radiation is
effectivein suppressing
IFN-yinduced keratinocyteICAM-I surface expression.
This observationallowedus to askwhetherremovalof cyclobutanepyrimidinedimersfrom UVB-irradiatedhumanskin through
photoreactivation
wascapableof restoringIFN-7-inducedkeratinocyteICAM-I expression.
Administrationof rh IFN-7 into UVBirradiatedskin,whichhadbeentreatedwith photolyase-containing
liposomesfollowed by exposureto photoreactivatingradiation,
induced keratinocyteICAM-1 expression(Fig. k). In contrast,
IFN-yinduced keratinocyteICAM-1 expressionwas inhibited in
UVB-irradiatedskin areasthat hadbeenleft untreated(Fig.2ä) or
that had beenexposedfirst to photoreactivating
radiationandthen
treatedwith photolyase(Fig. 2fl.
To further determinethe specificityof the photolyasetreatment, the numberof dimerspresentin the epidermalcompartment of thesesameskin areaswas simultaneouslymeasured.
Intracutaneous
injectionof rh IFN-7 did not affectthe formation
of dimers in UVB-irradiated skin (pig. 2d. Restorationof
IFN-y-mediated keratinocyteICAM-I expressionin UVBirradiated skin areasthat had been treated with photolyasecontainingliposomesfollowedby exposureto photoreactivating
Iight was associated
with a reductionof epidermalcyclobutane
pyrimidinedimer expression
by aboutaa% Gig.29).In contrast,
no significantdimer reductionwasobservedin UVB-irradiated
skin areasthat had beenexposedfirst to photoreactivatinglight,
followedby topical treatmentwith photolyase-containing
liposomesand subsequenti.c. injectionof rh IFN-7 Fig 2d.
Photoreactivation
PreventsUVBRadiation{nduced
Suppression
of
ElicitationPhaseof ContactHypersensitivityResponseto Nickel
Sulfatein HumanSkin.In subjectshypersensitive
to nickelsulfate,
positivepatch test reactionswere observedif patch testingwas
performedin unirradiatedskin areas(data not shown).Histological examinationof the sameskin area revealedspongiosis
and the presenceof an intraepidermaland intradermalinflammatory infiltrateconsistentwith a delayed-typehypersensitivity
reaction (Fig. 3a). Elicitation of this hypersensitivityreaction
was completelypreventedin the sameindividualsif patch tests
wereappliedto a skin areaimmediatelyafter exposureto 2 MED
of UVB radiation.Histologicalexaminationdid not indicatethe
presence of a significant inflammatory infiltrate in UVBirradiated, erythematousskin, but revealedthe presenceof
sunburncells (Fig. 3b). Topical treatmentof UVB-irradiated
immediatelyafter photoreactivationin triplicates
theseskin areaswas assessed
and quantified by immunfluorescencemicroscopywith mAb H3 as describedin
Materials and Methods. Data are given as histogramsof specificf luorescencein
arbitrary units (a. u.) as described(16) versusskin area and representmean
valuesa SDof three experiments.
Stege et a/.
f
60
-
f
:50
840
(,
12
30
v,
o
o20
=
LL
10
f
!
zg
L
!
c
!
2
Discussion
In this study,it hasbeen demonstratedthat-topicaltreatment of
shri wittr liposomescontainingbiologicallyactivephoh;;.
Stege et a/
P N A S I F e b r u a r y 1 5 , 2 0 0 0I v o l 9 7 | n o ' 4 | 1 7 9 3
F i g .3 . R e s t o r a t i o n
o f t h e e l i c i t a t i o np h a s eo f t h e c o n t a c t - h y p e r s e n s i t i vriet ys p o n s teo n i c k e sl u l f a t ei n U V B - i r r a d i a t ehdu m a ns k i nt h r o u g h p h o t o r e a c t i v a t i o n .
H i s t o l o g i c apl i c t u r e s( x ' l 0 0 )o f p a t c ht e s t r e a c t i o ntso n i c k e sl ul f a t e i n h u m a ns k i n .S k i na r e a sh a d b e e ni e f t u n i r r a d i a t e d( a ) ;U V B - I r r a d i a t e(d2 M E D )b e f o r ep a t c h
ini ng Ii posomes,fol lowed by exposureto photoreactivating Iig ht before patchtesting
testing (b);or UVB-irradiated,the n topicaIly treated with photolyase-conta
( c ) .B i o p s i e w
s e r e t a k e n 4 8 h a f t e r a p p l i c a t i o no f t h e p a t c ht e s t .
tolyase and subsequent exposure to photoreactivating radiation
is effective in partially removing UVB radiation-induced cyclobutane pyrimidine dimers from the epidermis of treated skin
areas, thereby diminishing erythema, sunburn-cell formation,
and suppression of ICAM-1, a molecule required for immunity
and inflammatory events in the epidermis. Photolyase is a
DNA-repair enzyme that is functionally defined by its capacity
to monomerize pyrimidine dimers in cellular DNA through a
repair mechanism that requires exposure of the enzyme-DNA
complex to radiation at 300-500 nm and has therefore been
termed photoreactivation (6, 7). The recent cioning of a human
photolyase gene homologue and reports of photoreactivation in
human white blood cells and fibrobiasts have suggested that
endogenous photoreactivation may offer a natural means of
photoprotection in human skin (6, 8, 9, 13). However, we found
no evidence of endogenousphotoreactivation, becauseexposure
of UVB-irradiated human skin to photoreactivating radiation
before topical treatment with photolyase-containing liposomes
(Fig. 1d) or treatment of human skin with empty liposomes
followed by exposure to photoreactivating radiation (data not
shown) did not decreasethe number of cyclobutane pyrimidine
dimers present in UVB-irradiated human skin. This discrepancy
might be because of the possibility that wavelengths and doses
used to effect photoreactivation by Anacystis nidulans might
differ from those used by human skin. In addition, using
different techniques to measure dimers might give different
results. In the present study, the only condition under which
photoreactivation was observed was when exogenousphotolyase
in liposomes (Photosomes)was applied before photoreactivating
light. These studies demonstrate that UVB radiation-induced
DNA damage can be efficiently repaired in normal human skin,
after it has been exposed to UV radiation, through the topical
administration of photolyase, a DNA repair enzyme that, under
Table 1. Semiquantitativeassessmentof lymphorytic infiltration
i n F i g .3
Cellcountaftertreatment
Localization
N|SO4
Ep i d e r m i s
U p p e rd e r m i s
48+7
tz=z
Niso4
UVB
)=
I
24+4
Ni5o4,uvB,
photolyase-conta
i ning
UVA
liposomes,
9+2
35+5
T h e n u m b e r o f e p i d e r m a l a n d d e r m a l l y m p h o c y t e sw a s q u a n t i f i e d a s
d e s c r i b e di n M a t e r i a l sa n d M e t h o d s .D a t ar e p r e s e n t t h em e a nc e l lc o u n t( + 5 9 1
f r o m t h r e e v i e w f r e l d si n t h r e e s e r i a ls e c t i o n s .
1794 | M.pnasorg
our experimental conditions, cannot be detected in human skin.
Similarly, topical application of the DNA-repaft enzyme T4
endonucleaseV was shown previously to effect dimer repair in
skin of patients suffering from the DNA-repair deficiency syndrome xeroderma pigmentosum (20). In these previous studies,
the average dimer removal enhanced by T4 endonuclease V
treatment was approximately 20Voin 6 h, whereas in the present
study, photolyase treatment resulted in removal of 40-457a of
cyclobutane pyrimidine dimers present in UVB-irradiated normal human skin immediately after photoreactivation, indicating
that the latter approach has a greater efficiency. Also, instant
reversal of cyclobutane pyrimidine dimers by photolyase treatment might prevent saturation of the natural repair systemsand
thereby facilitate or enhance nucleotide excision repair. At the
dose used in our experiment, however, this was not the case,and
the dimer-repair rates in UVB-exposed skin either treated with
photolyase or untreated are very similar. Consequently, the
relative reductions in DNA-damage content immediately after
photolyase treatment and after 24 h were similar.
Removal of cyclobutane pyrimidine dimers from UVBirradiated human skin through photoreactivation restored the
capacity of keratinocytes to up-regulate expression of the adhesion molecule ICAM-I on stimulation with rh IFN-y. ICAM-I
expression on human skin cells is thought to be of critical
imoortance for the elicitation of cell-mediated immune respons"s in human skin (21). In previous studies,the induction of
ICAM-1 expressionin human cells, including keratinocytes, has
been shown to be suppressedupon exposure to sublethal doses
of UVB radiation (22). This immunosuppressiveeffect of UVB
irradiation was found to be dose-dependent, and this assay
system has therefore been used as a photoimmunological endpoint to assessthe susceptibiiity of human skin cells toward UVB
radiation-induced immunosuppression (23). The present study
demonstrated that, similar to the in vilro situation, UVB radiation is capable of suppressinglFN-7-induced ICAM-1 expression in vivo as well. Moreover, restoration of IFN-7-induced
keratinocyte ICAM-1 expressionin UVB-irradiated human skin
through photoreactivation strongly suggeststhat this immunosuppressiveeffect is causally linked to the formation of dimers
within UVB-irradiated skin areas.
The U\{B doses required to achieve suppression of IFN-yinduced keratinocyte ICAM-I up-regulation in human skin were
essentiallyidentical to those that previously were found to suppress
T cell-mediated immune responses against contact allergens in
human skin Q\. In those experiments, UVB radiation-induced
reduction of immunization rates in human skin was found to be
MED-dependent, and significant immunomodulatory effects could
be observed at levels of in vivo UVB exposure that were equivalent
Stege et a/
to thoseusedin the presentstudyfor inhibition of IFN-y-induced
keratinocyteICAM-1 expression.Cyclobutanepyrimidine dimer
formation in humankeratinocytesin UVB-inadiated humanskin
not only maybe relevantfor UVB radiation-induced
suppression
of
keratinocyte ICAM-I expression,but also may be of general
importancefor the capacityof UVB radiationto affectthe immune
function of keratinocytesand thereby the generationof cellmediatedimmune resDonses
in human skin.
One consequence
of thisgeneralimmunosuppressive
abilityof
UVB radiationis that exposureof humanskinto 2 MED of UVB
radiation effectively inhibited the elicitation phase of the delayed-typehypersensitivityreactionto nickel sulfatein sensitized
individuals (25-27). This immunosuppressive
effect could be
overcomethroughthe topical applicationof photolyaseand the
subsequentexposureof treatedskin areasto photoreactivating
radiation.Thesedata indicate that DNA damageis linked to
UVB radiation-inducedsuppression
of the elicitationphaseof
allergiccontactdermatitis.Photolyasetreatmentalsoinhibited
the formation of UVB-induced erythemaaswell as of sunburn
cells.Theseresultsclearly link DNA damagewith the clinical
symptomsof immunosuppression
and erythema.They extend
observationsmade in the opposumthat photoreactivationof
pyrimidine dimers reduced UV radiation-inducederythema
(28). These data explain why the action spectrum for human
erythemaso closelyparallelsthe actionspectrumfor the induction of cyclobutanedimers in DNA (29).
In aggregate,the presentstudy indicatesthat topical application of photolyase encapsulatedinto liposomes to UVBirradiatednormalhumanskin followedby exposureof theseskin
areas to photoreactivatingradiation was efficient in providing
this approach
DNA repair and immunoprotection.
Interestingly,
providedcompleterestorationof keratinocyteICAM-L expression,whereasthe numberof epidermalthyminedimerswithin
the same skin areaswas reducedby only 40-45%. A similar
discrepancybetweencomplete immunoprotectionand incomplete dimer removalhad beennoted previouslyafter treatment
of mouseskinwith T4 endonuclease
V-containingliposomes(4).
mightbe that not all
A possibleexplanationfor this discrepancy
dimersare equal,but that repair of a fraction of dimersmight
1 Coldiron, B. M. (7992)J. Am. Acad Dematol. 27, 653-662.
2 Elmeß, C A. & Mukhtar, H. (1996)Prog.Dematol. 30, 7-L6
3 Applegate,L. A., Ley, R. A., AJcalay,J. & Kripke, M. L. (1989)J. Exp.Med
170, rrr7-t131.
4 Vink, A. A, Strickland,F M., Bucana,C, Cox,P A., Roza,L, Yarosh,D. B.
& Kripke, M. L. (1996)l. Exp. Med. r83, 1491-1500.
5. Gasparro,F. P., Mitchnick, M. & Nash,J. F. (L998)PhotochemPhotobiol.68,
243-256
J. C.
6. Sutherland,BM,Hacham, H, Gange,R.W., Maytum,D & Sutherland,
(1990) in DNA Damageand Repairin Human Tissues,
eds Sutherland,B. M.
& Woodhead,A D. (Plenum,New York), pp. 149-159.
7. Sancar,G. B. (1990)Mutat. Res.236, L47-760.
8. Sutherland,B. M. & Bennett, P. V. (1995) Proc. Natl. Acad. Sci. USA 92,
9732-9736.
9. Zölzer, F, Belisheva,N. Ii, Irvin, V. L. & Samoilova,L A. (1993) J
Photochem.Photobiol.B lE, 87-89
10. de Gruijl, F R. & Roza, L. (1993) J. Photochem. Photobiol. B l0'
501-5tz
11. Iry, R. D. (1993)Proc.NatI.Acad. Sci.USA90,4331.
L2. Kanai, S., Kikuno, R., Toh, H , Ryo, H. & Todo, T (1997)I. Mol. Evol. 45,
535-548
13. Todo, T., Tsuji, H., Otoshi,E., Hitomi, IC, Kim, S.-T.& Ikenaga,M (1997)
Mutat. Res.384, 195-204.
14. Pathak, M. A., Nghiem, P & Fitzpatrick, T. B. (1999)in Fitzpatrick'sDermatology in GeneralMedicine, eds. Freedberg,I M., Eisen, A. 2., Wolff, K,
Austen, IC F, Goldsmith,L. A, Katz, S. I. & Fitzpatrick,T. B. (McGraw-Hill,
New York, NY), 1598-1607.
15- Roza,L, Van der Wulp, I( J. M., Macfarlane,S J , Lohman,P H. M. & Bam,
R. A. (1988)PhotochemPhotobiol. 48,627-634
16. Roza, L., de Gruijl, F. R., Bergen Henegouwen,J. B A, Guikers, K, Van
Weelden,H., Van Der Schans,G. P. & Baan,R. A. (I99I') J. Invest.Dennatol.
96, 903-907.
Steoeet a/.
havedisproportionateeffectson biologicalresponses.
This hypothesisis supportedby the recentobservationthat only photoproductsin activelytranscribedDNA, which arepreferentially
repaired (30) and constituteonly a small fraction of the total
DNA, are importantin signaltransduction(31). PerhapsDNA
regionsinvolvedin the transcriptionalregulationof the human
ICAM-I gene may be repairedmore rapidly than othersmay.
Alternatively, UVB radiation effects may require a critical
number of cyclobutanepyrimidine dimers within epidermal
keratinocytes,
and if the numberof dimersis reducedbelowthis
threshold,completeprotectionagainstUVB radiation-induced
immunomodulationand apoptosismay be achieved.This would
alsoimply that thereis no linearcorrelationbetweenthe number
of dimers and the developmentof immunosuppression.
In
support of this view, UVB radiation-inducedsuppressionof
immune responsesin human skin was shown to be a function of
biologicallyrather than physicallyequalizedUVB radiation
doses(24). Moreover,recentevidencedemonstrates
that keratinocyteapoptosisinducedby UV (sunburncells)is mediatedby
DNA damageand can be preventedeven if not all cyclobutane
pyrimidinedimersin DNA are repaired(32).
In conclusion,the presentstudy indicatesthat topical application of exogenousphotolyaseto human skin is an approach
that is highly efficient in protecting human skin from the
deleteriouseffectsthat result from the presenceof UVB radiation-inducedpyrimidinedimers.Exogenousapplicationof photolyasediffers from conventionalphotoprotectionthrough its
ability to removedamagethat hasalreadyoccurred.This enzyme
therapyapproachcould thus be ideally combinedasan after-sun
to provide photoprotecstrategywith conventionalsunscreens
tion and repair at the sametime.
We thank Paulavan den Berg for her excellenttechnicalassistanceand
Dr. MossadMegahedof Düsseldorf,Germany, and Dr. Daniel Yarosh
This work wassupportedby
of Applied Geneticsfor helpful discussions.
grants trom the European Commission(ENV4-CT95-0174a);the
Bundesministeriumfur Bildung, Wissenschaft,Forschung und TechForsnologie (BMBF 07UV8570); and the Biologisch-Medizinische
chungszentrumof the Heinrich He ine University, Düsseldorf,Germany.
1.7.Barker,J. N. W. N, Allen, M. H. & MacDonald,D M. (1989)J. Invest.
Dematol 93,439-442.
18. Rietschel,R. L. & Fowler, J. F., eds. (1995) in Fßher's ContactDematitis
(Williams & Wilkins, Baltimore), pp. ll-32.
19 Morita, A., Werfel, T., Stege,H., Ahrens, C, Karmann, K., Grewe, M.,
Grether-Beck,S , Ruzicka,T , K^pp, A., KJotz,L O.,et al. (199Tf. Exp Med.
186,7763-1768.
20. Yarosh,D., Klein, J., Kibitel,J , Alas,L., O'Connor,A, Cummings,B , Grob,
D., Gerstein, D, Gilchrest, B. A., Ichihashi,1., et al. (L996) Photodematol.
Photoimmunol. Photomed. 12, 722-130.
21 Nickoloff, B. J. (1989)in TheSkin Immune System,ed. Bos,J. D. (CRC, Boca
Raton,FL), pp.49-74
22 Krutmann, J., Köck, A, Schauer,E , Parlow,F., Kapp,4., Förster,E., Schöpf,
E. & Luger, T. A (1990)./. Invest.DermatoL 95, tn-f3L
23. Ahrens,C., Grewe,M, Berneburg,M., Grether-Beck,S.,Quilliet, X.,Mezzina,
M., Sarasin,A.,I-ehmam, A. R., Arlett, C. F. & Krutmarn,J. (1997)Proc.Natl.
Acad. Sci. USA 94,6837-6841.
24 Cooper,IC D., Oberhelman,L., Hamilton, T A, Baadsgaard,O , Terhune,M.,
Levee, G., Ardenon, T. & Koren, H. (1992)Proc. Natl. Acad. Sci. USAE9,
8497-8s01
25 Kafimo, I(, Koulu, I( & Jansen,C T (1983)ArclL Dematol Res.275, 374-378.
26. Sjovall,P. & Christensen,
O. B. (1986),actaDem. Venereol.66,290-294.
27. Damia, D L, Halliday, G. M. & Barnetson,R. S. (1997)I Invest.Dematol.
109.146-151.
28 Ley, R. (1985)Proc. Natl. Aacd. Sci. USA 82,2409-24L7.
29 Hacham,H., Freeman,S E. & Gange,R.W. (1991)Photochem.Photobiol.53,
559-565
30 Hanawalt,P. C. (1994)Science266,1957-1.958.
31 Blattner, C., Bender,K.. Herrlich, P. & Rahmsdorf,H. J (1998)Oncogene16,
2827-2834.
32 Kulms,D., Pöppelmam,8., Yarosh.D-, Luger, T. A., Krutmam, J. & Schwarz,
T (1999)Proc.Natl.Acad Sci.USA 96,7974-79i9.
PNA5 | February15,2000 I vol 97 I ^o.4 | 1795