III
I
Photoprotection by melanin black and Caucasian skin
I
a comparison of
Kays H. Kaidbey, M.D., Patficia Poh Agin, M.S., Robert M. Sayre, Ph.D., and Albert M. Kligman, M.D., Ph.D.
Philadelphia, PA, and Memphis, TN The photoprotective role of melanin was evaluated by comparing the transmission of ultraviolet (UV) radiation through skin samples of blacks and Caucasians, using both biologic and spectroscopic techniques. UVA transmission was measured using fluoranthene, which causes a phototoxic response to UVA wavelengths. UVB was measured by monitoring erythema produced by either a 150-watt xenon arc or FS-20 sunlamps. It was found that on the average, five times as much ultraviolet light (UVB and UVA) reaches the upper dermis of Caucasians as reaches that of blacks. Differences in transmission between the stratum corneum of blacks and of Caucasians were far less striking. The main site of UV filtration in Caucasians is the stratum corneum, whereas in blacks it is the malpighian layers. Melanin acts as a neutral density filter, reducing all wavelengths of light equally. The superior photoprotection of black epidermis is due not only to increased melanin content but also to other factors related to packaging and distribution of melanosomes. Not only are these data consistent with epidemiologic evidence, but they also may indicate why blacks are less disposed to phototoxic drug responses as well as less susceptible to acute and chronic actinic damage. (J AM ACAD DERMATOL1:249-260, 1979.)
The evidence for the photoprotective role of melanin is compelling. ~-a Both epidemiologic and experimental studies have already shown that blacks and darkly pigmented Caucasians are far less susceptible to acute and chronic actinic damage than are fair-skinned individuals. There is a clear inverse correlation between the degree of histologic solar elastosis and skin complexion at every decade of life. a In blacks, the hyperplasia of elastic fibers is greatly delayed and of lesser magReprint requests to: Dr. Kays H. Kaidbey, Duhring Laboratories, Department of Dermatology, University of Pennsylvania, 3500 Market St., Suite 203, Philadelphia, PA 19104.
0190-9622/79/030249+ 12501.20/0 9 1979 Am Acad Dermatol
nitude. Squamous and basal cell cancer, believed to be at least partially due to chronic exposure to actinic radiation, is very rare in blacks living in the sunbelt areas. Black albinos, on the other hand, are very susceptible to actinic damage and readily develop precancerous and cancerous lesions. McFadden 1 and KeeleP detected keratoses, atrophy, and telangiectasias in the albino Cuna Indians as early as the age of 2 years. By the age of 7, the incidence of keratoses approached 100% and many had developed squamous cell carcinomas and actinic cheilitis. Normally pigmented Indians did not develop these changes. Oettle 5 also found a high incidence of precancerous keratoses 249
250
Journal of the American Academy of Dermatology
Kaidbey et al
Abbreviations used E
whole epidermis (including stratum corneum) EPI epidermis MAL malpighian layer MED minimal erythema dose MPD minimum phototoxic dose PF protection factor STC stratum corneum UV ultraviolet UV spectrum in vitro forward scattering spectrum using integrating sphere UVA-SS solar simulator UVA transmission in human volunteers UVB-FS FS sunlamp UVB transmission on human volunteers UVB-SS solar simulator UVB transmission in human volunteers
among hypopigmented individuals living in subtropical belts. Another example is vitiliginous skin, which is unusually susceptible to sunburn. The natural photoprotective role o f melanin is beyond doubt. More direct evidence of a quantitative nature has come from in vitro spectroscopic studies on excised skin specimens. Pathak and Fitzpatrick 2 compared the transmission characteristics of both stratum corneum and whole epidermis from blacks and fair-skinned Caucasians. Caucasoid epidermis was more transparent to ultraviolet and visible wavelengths. At 300 nm, there was 10% to 15% transmission through Caucasian epidermis, as compared to only 2% to 5% through black epidermis. Everett et al, 6 in a comprehensive investigation of transmission spectroscopy by human skin specimens, studied one sample of black epidermis. By using integrating spheres along with recording spectrophotometry, they were able to measure forward scattering and hence total transmission more accurately than had been accomplished before. They also found considerably greater transmission of light over the entire 300 to 600 nm wavelength range than had been reported previously. This study also showed that values for a skin specimen from an American Indian were
intermediate to those of Caucasoid and black epidermis. Although the protective function of melanin seems abundantly clear, quantitative data on the comparative contribution of the horny layer and epidermis in blacks and whites are few. In the past, some investigators concluded that in Caucasians, thickening of the stratum corneum was more important than pigmentation in photoprotective adaptation. 7-9 This question can be resolved in part by accurate measurements of transmission through the various skin layers. Although spectroscopic studies of excised skin are quite informative, there has been some concern about the relevance of in vitro measurements to actual transmission in vivo. it is difficult to measure all of the components of a light beam after it has traversed an optically heterogenous medium such as skin: reflection, refraction, absorption, backward and forward scattering all have to be measured and accounted for before meaningful conclusions can be drawn. The object of the present investigation was to assess quantitatively the natural photoprotective role of melanin. This was done by comparing transmission of ultraviolet (U V) radiation through excised skin specimens of blacks and whites. Biologic measurements were obtained by using skin sheets as filters and the values correlated with those obtained by spectroscopic techniques. MATERIALS AND METHODS Chemicals
Trypsin (type II, hog pancreas) was purchased from Sigma Chemical Co., St. Louis, MO. Fluoranthene was obtained from Eastman Chemicals, Rochester, NY. Skin specimens
Specimens of black and Caucasian skin were obtained from the abdominal region of cadavers and quick-frozen with liquid nitrogen until use. After thawing at room temperature, the specimens were placed in a water bath at 60 C for 2 minutes, and the epidermis was peeled off the dermis. To obtain stratum corneum, the epidermal sheets were floated in 0. 01% trypsin solution (buffered at pH 8) overnight. Epidermal cells were then wiped away with cotton-tip applicators. The corneum sheets were rinsed in water, allowed to air-dry
Volume 1 Number 3 September, 1979
Photoprotection by melanin
251
T a b l e I. Skin s p e c i m e n identification Tests performed Specimen
Race
Age
Sex
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA BB CC DD EE FF GG HH II JJ
B B W B W W B W W W W W B B B B B B B B B B W W W W W W W W B B W W W W
47 50 57 50 57 61 30 67 78 72 60 51 69 50 24 29 41 46 60 60 61 75 22 40 53 54 60 62 72 74 48 71 70 58 69 48
F F M M F M M F F F F F M F F M F F F F M M F F F F M M M F F F F F M F
UVB-SS Biol.
UVB-FS Biol.
STC & E STC & E STC & E
UVA-SS Biol.
E E E
STC & E
STC & E
STC STC STC STC STC STC STC STC STC STC STC STC STC STC STC STC
& & & & & & & & & & & & & & & &
E E E E E E E E E E E E E E E E
E
STC & E E E E E E E
STC STC STC STC STC STC
& & & & & &
UV spectrum Forward scattering
STC & STC & STC & STC & STC & STC STC STC STC E E E E E
E E E E E
E E E E E E E E E E E E E E E E
E E E E E E
B: Black skin, W: Caucasian skin. at room temperature, and then refrigerated until use. Each skin sample (epidermis and corneum) was used as a filter for transmission measurements in only one experiment to avoid possible U V-induced alteration of the tissue, The possibility of this had previously been suggested and is known to be common with gelatin protein filters.10,11 Table I identifies each skin sample and the tests performed on skin specimens from that individual. Not all samples used in the in vivo studies were suitable
for spectroscopic forward scattering examination, The specimens were examined histologically after formalin fixation. Biologic determination of UVB transmission
Transmission of UVB radiation (290-320 nm) was measured by comparing the doses required to produce minimal erythema dose (MED) in the untanned skin of fair-skinned Caucasian subjects (types I. II, and IlI)
252
Journal of the American Academy of Dermatology
Kaidbey et al
T a b l e I I . UVB p r o t e c t i o n - - s o l a r simulator Black skin filters
White skin filters
Specimen
EPI PF
STC PF
Specimen
EPI PF
O P Q R A B S T U V Mean • SD
12 18 18 16 10 10 13 10 I5 12 13.4 __- 3.16
3 4 4 3 3 2 4 3 4 3 3.3 • 0.67
W X Y Z C AA BB J CC DD
4 3 5 3 3 4
2 2 2 2 2 3
3
2
3 3 3
2 2 2 2.1 +-_ 0.31
[ I
STC PF
3.4 •
T a b l e I I I . UVB protection using FS-20 sunlamps Black skin filters
White skin filters
Specimen
EPI PF
STC PF
Specimen
EPI PF
P G EE FF Mean • SD
29 30 32 30 30.25 ___ 1.26
4 5 5 5 4.75 --+ 0.5
GG HH II JJ
5 5 4 4 4.5 ___ 0.58
with and without filtering by the test skin specimens. Informed consent was obtained prior to testing. On unprotected skin, the MED was first determined by administering a series of geometrically increasing exposures to the lower back from a 150-watt xenon solar simulator,* so that each dose was 25% larger than the previous one. The MED was the smallest dose required to produce visible erythema 24 hours later. Then 1.52. 0-cm squares of stratum corneum or epidermis were taped to a thin plastic frame with I x 1-cm windows. The frame was then securely positioned over the lower back. One-tenth milliliter of physiologic saline was pipetted under each square to wet the specimen and assure intimate contact with the skin. Increasing multiples of the control MED energy were then given through the specimens until erythema appeared. The ratio of the MED through the test specimen to the MED of unprotected normal skin represents a protection factor (PF). Transmission of the biologically effective radiation is the reciprocal of the PF. *Solar Light Co., Philadelphia, PA; filtered with 2.0 mm Sehott WG320 filter.
[
STC PF
3 2 2 3 2.5 - 0.58
To evaluate the potential effects produced by a different UVB light distribution, four specimens from each group were subjected to the same procedure just outlined, except that the U VB source was a bank of five closely set fluorescent sunlamps (FS-20, Westinghouse) mounted in a reflector housing. The lamp-toskin distance was 12 cm. Biologic d e t e r m i n a t i o n o f U V A t r a n s m i s s i o n
The transmission of U V A radiation (320-400 nm) can be measured by comparing the minimum phototoxic exposure of skin pretreated with fluoranthene, a hydrocarbon photosensitizer activated by U VA wavelengths. Fluoranthene in 95% ethanol (5 t z l / c m 2 of 0.5%) was delivered to 1-cm squares of skin over the midback of white volunteers. The sites were allowed to air-dry; then they were covered by square patches of nonwoven cotton cloth (Webfil; Curity) and fastened to the skin with occlusive tape (Blenderm). Two hours later the patches were removed and the sites exposed to increasing doses of U VA from the xenon source using a 2-mm WG345 filter to eliminate the UVB wavelengths.
Volume 1 Number 3 September, 1979
The minimum phototoxic dose (MPD), the smallest dose of UVA necessary to produce minimal visible erythema, was determined 24 hours later. The MPD was determined again in an identical fashion on the opposite side of the back through the test skin specimens as already described. A U VA protection factor (PF) in this instance is the ratio of the MPD through the test specimen to the MPD of normal unprotected skin. Again, the reciprocal of the PF represents the transmission of UVA-effective radiation through the test skin samples. Energy output measurements were made using a calibrated thermopile (Eppley Laboratories Calibration No. 11609) attached to a Keithly millimicrovoltmeter. UVB intensity from the xenon solar simulator was 19.70 mw/cm2; UVA irradiance was 15.82 mw/cmL The U VB irradiance at the skin surface from the bank of fluorescent FS-20 sunlamp tubes (lamp-to-skin distance, 12 cm) was 0.6 m w / c m 2.
Spectroscopic determination of UV transmission: In vitro forward scattering Dried samples of human epidermis or stratum corneum measuring at least 4 cm 2 were rehydrated by immersion in distilled water at room temperature for approximately one minute. The hydrated sample was floated onto a quartz carrier plate and inserted into the forward scattering compartment of a Beckman Acta MVI spectrophotometer fitted with a diffuse reflectance sphere. Each sample was scanned from 250 to 400 nm. In both the sample and the reference beams, 2-ram UG-5 filters were used to remove possible visible fluorescence. Baseline absorbance was subtracted from each sample scan, and protection factors were calculated as previously described. TM RESULTS
Biologic measurements
Photoprotection by melanin 253
Table IV. Protection factors of w h o l e epidermis against U V A (solar simulator)
Black epidermalfilter Specimen [ PF
1 Whit..eepidermalfilt.er .... Specimen
O 7.0 P 10.6 Q 4.0 R 3.5 A 3.5 B 3.3 S 4.3 T 6.0 U 6.7 V 8.0 Mean +__SD 5.69 • 2.41
W X Y Z C AA BB J CC DD
PF
2.0 1.5 2.3 1.3 2.0 2.4 1.3 1.6 2.0 1.6 1.8 _-_ 0.39
Table V . Comparison o f in vitro and in vivo protection
Specimen A B C A D C J
I I
Race Sampletype B B W B B W W
STC STC STC EPI EPI EPI EPI
vtro (predicted) PF
In vivo PF
2.3 2.3 2,2 13.6 12.4 4.8 3.6
3 2 2 10 10 3 3
B: Black skin. W: Caucasian skin.
stratum corneum was 3. 3 and for Caucasians, 2. 1. Hence, the mean transmission by corneum for blacks was 30.3% (range, 25% to 5 0 % ) and 47.6% (range, 33.3% to 50%) for Caucasians.
Transmission of UVB with solar simulator.
UVB transmission with FS-20 sunlamps.
The individual and m e a n PF values of epidermal and stratum c o r n e u m specimens from ten black and ten Caucasian subjects are shown in Table II. The mean PF for black epidermis was 13.4, as compared to 3.4 for white epidermis. The mean transmission by black epidermis was therefore 7.4% (range, 5.5% to 10%) compared to 29.4% (range, 20% to 33.3%) for Caucasian epidermis. The difference was less p r o n o u n c e d with stratum corneum specimens. The mean PF of black
Higher PF values were obtained with this light source (Table III). T h e mean P F for black epidermis was 30.25 (compared to 13.4 with the solar simulator) and 4.5 for white epidermis (compared to 3.4). T h e mean transmission was therefore 3.3% for black epidermis compared to 22.2% for Caucasian epidermis. T h e mean PF for the black stratum c o r n e u m was also greater with this source (4.75 c o m p a r e d to 3. 3 with the solar simulator).
254
Journal of the American Academy of Dermatology
Kaidbey et al
8TRATUlit CORNEUM ..,....,o'~
DI~K
.,, .,.' ~".. ",., 1.1
//
\
G.""
j. C ,8
R* A" E |
"',. '.,, .,~
..
"'4.% "",.
"%
.3
WAVELENGTH
Fig. 1. Absorption spectra of the STC samples studied by forward scattering spectroscopy. Between 290 and 320 nm, the two groups are almost indistinguishable, except for sample G. Sample identification is taken from Table I.
Transmission of U V A . Table I V shows that the PF of black epidermis is about 5.7, significantly higher than that of white epidermis at 1.8 (p < 0.005, Student's t test). T h e mean U V A transmission, therefore, by black epidermis was 17.5% compared to 55.5% for Caucasians. In vitro spectroscopic measurements Figs. 1 and 2 show the absorption spectra of all of the stratum corneum and epidermis samples studied. Samples from the same individual are identified by identical labeling in both figures. Several samples do not have corresponding matches in both figures. In Fig. 1, the stratum corneum samples do not segregate into distinguishable groups by melanin content at wavelengths shorter than 320 nm. Fig. 2, however, clearly demonstrates that there are great differences in the optical qualities of black and white epidermis. It is quite striking that the two groups are nonoverlapping at all wavelengths, from 250
to 400 nm. Fig. 3 shows the average absorption spectrum for each type of skin sample examined: black epidermis, black stratum corneum, white epidermis, and white stratum corneum. White epidermis and black stratum corneum are indistinguishable in the sunburn (UVB) range 290320 nm. White stratum corneum is only slightly less protective than black stratum corneum. Black epidermis, however, is very photoprotective at all wavelengths. Table V compares the protection values of skin samples determined by both forward scattering and in vivo studies. The predicted protection factors obtained from spectroscopic measurements were calculated from the model described by Sayre et al TM using the skin specimen absorbance in place of a sunscreen. The stratum corneum samples, black and white, are indistinguishable by this method and give a protection factor averaging about 2. 2. The epidermal samples, however, are quite different. The black
Volume 1 Number 3 September, 1979
Photoprotection by melanin 255
3.f
|PlDEflldl8
~ .......
\
29
2.l
-
.........~.,,~~
_
willie
.................
black
,g ,~
Pt B ~
IA
t
i
,q
',.
t!
'%
,
~
iiiiiiiiiii,iiiiiiiiiiiiiiiiiii ii WAVELENGTH
Fig, 2, Absorption spectra of the epidermal samples studied by forward scattering spectroscopy. Note that the two groups are distinct at all wavelengths. Sample identification is taken from Table I. epidermal samples are three to four times as protective as the white epidermal samples. The predicted in vitro protection correlates well with the in vivo values. Table VI summarizes the results of light transmission values through the stratum corneum and whole epidermis, and gives a predicted value for malpighian layers. Transmission of light through the entire epidermis is equivalent to the transmission of light through the stratum corneum multiplied by the transmission of light by the malpighian layers. High, low, and average values are given for each set of samples examined. Again, comparison of the values obtained by human testing to
those obtained by in vitro spectroscopic methods shows good correlation using both UVA and UVB. For Caucasian skin, between 30% and 50% of incident UVB radiation is transmitted by the stratum corneum. Yet, predictions from each method regarding the transmission by the malpighian layers indicate tbat 60% to 70% of the available light is transmitted, that is, more UVB radiation is removed by the Caucasian stratum corneum than is removed by the underlying epidermal cells. For U VA, approximately the same amount of light is transmitted by both major skin layers. For black skin, Table VI illustrates that
256
Journal of the American Academy of Dermatology
Kaidbey et al
Maak epMeNNs
................ ~ strotwn~fneum ........ w~Ite e p I d e ~ l s ..... white ~4ratum cwmeum
/\'
~2s
;
9" ,75 i
I
\ -\
I
~ : I
.50
~n'~'n"%" .25
9
.~~ ..r.:.... ...... -~.::::..:.: .............
,,|
250
27S
300
325 350 WAVELENGTH
311'5
400 NM
Fig. 3. Average absorption spectrum for each type of skin specimen examined. Black stratum corneum and white stratum corneum are quite similar, whereas the epidermal samples are very different. However, from 300 to 400 nm, black stratum corneum and white epidermis are almost identical. there is wide variability between individual skin samples, depending upon tile test method used by a factor of two to four. Yet the average transmission of U V B is 20% to 30% regardless of method. In all cases, less light must be transmitted through the malpighian layers than through the stratum corneum if the 3% to 7% penetrating the entire epidermis is to be realized. This means the mal-
pighian layers of black skin remove twice as much U VB radiation as their stratum corneum. For U VA, possibly even greater removal of radiation occurs in black malpighian layers. DISCUSSION Other factors besides melanin contribute to natural photoprotection. These include absorption
Volume 1 Number 3 September, 1979
Table
Photoprotection by melanin
257
VI. Summary of transmission of light through skin layers* White skin (predicted)
Wave band
Range
Black skin (predicted)
STC
MAL
EPI
STC
0.618
0.333 0.294 0.200
0.500 0.303 0.250
0.250 0.222 0.200 0.500 0.286 0.164
0.250 0.210 0.200 0.526 0.278 0.135
0.539 0.322 0.166
0.537 0.325 0.134
UVB (solar simulator --human)
High AV Low
0.500 0.476 0.333
UVB (FS-20 sunlamp - - human) UVB (predicted) in vitro)
High AV Low High AV Low
0.500 0.400 0.333 0.556 0.435 0.370
UVB 305 nm (in vitro)
High AV Low
0.575 0.461 0.380
UVA (fluoranthene phototoxicity --human)
High AV Low
UVA 360 nm (in vitro)
High AV Low
0.555
0.657
0.698
0.248
0. I56
0.147
0.157
0.769 0.556 0.417 0.957 0.824 0.707
0.769
0.782 0.637 0.213
0.100 0.075 0.056 0.034 0.033 0.031 0.088 0,041 0.019 0.089 0.051 0.031 0.303 0.176 0.094
0.926 0.617 0.467
0.198
0.199 0.122 0.039
*Transmission values shown in the table are the reciprocal of the PFs.
and scattering by keratin, collagen, and other macromolecules, absorption by aromatic amino acid residues in proteins, and probably absorption due to the presence of some carotenoid pigments. The present work, however, is concerned primarily with the role of melanin. Since there is no evidence suggesting major differences in the thickness or composition of the epidermal layers in the two racial groups, TM 14 we assume that the observed differences in transmission are mainly due to this pigment. By using skin specimens as filters, and living skin as the detector, photoprotection was measured in much the same way as is used to assess the efficacy of a sunscreening agent. The protection by black epidermis against sunburn was impressive indeed. A PF of 13.4 is matched only by a few of the most effective commercially available sunscreens. The average spectrum shown in Fig. 3 demonstrates why this is so. The total transmission of UVB was less than one fourth of that by Caucasian epidermis. Wide variations were seen, however, in the individual values, which almost
certainly reflect differences in the intensity of pigmentation among blacks. Blacks are no more a homogeneous group than whites. It is interesting to note that the difference in U VB transmission by black and white stratum corneum was far less striking. Caucasian specimens on the average transmitted only about 1.5 times more U V B than did blacks. It is clear, therefore, that in blacks, filtration of sunburning radiation occurs primarily within the living epidermal layers, and not in the stratum corneum. This is also evident from the transmission data outlined in Table VI. In contrast, the primary site of filtration in Caucasians is the stratum corneum, where over 50% of the incident radiation is absorbed. Clearly, then, melanin appears to exhibit much greater photoprotection in the black malpighian layers than in the stratum corneum. This is true for both the U VA and U VB wavebands. When melanin in black skin reaches the stratum corneum, it appears to lose much of its protective capabilities and does not absorb and scatter radiation as effectively as in the malpighian layers. Whether this loss of protective ability is
258
Kaidbey et al
due to a biochemical or physical process is not known. Thomson '5 used a photographic technique and radiation from a mercury vapor lamp to compare transmission by stratum corneum of blacks mad Caucasians. The stratum corneum specimens were obtained from cantharidin blister roofs. Although he found no difference in the thickness of the stratum corneum between the two groups, the total transmission over the 300 to 400 nm range was about 3.5 times greater in Caucasians and was less marked, about 2.5 times, in the erythemal range (290 to 320 nm). He attributed the lessened transmission to the presence of melanin. We found an even smaller difference between the two groups with solar-simulating radiation. These results indicate that in blacks as well as in whites the keratinocytes of the upper epidermis are exposed to substantial radiation levels. Seemingly low transmission values were obtained in tests with the FS-20 sunlamp tubes. This is due to the presence in this source of lower UV wavelengths extending into the U VC range. These shorter wavelengths are more effectively absorbed and scattered by the upper epidermal layers, even though they are more erythemogenic. Since this occurs more efficiently in blacks than in Caucasian skin, U VC-rich sources tend to yield elevated PF values. Similarly, differences in the threshold dosages for U V erythema between the two groups will thereby be accentuated. It is imperative, therefore, to carefully interpret such results depending upon the radiation source chosen when transmission is measured by this technique. There was good agreement between the estimated protection values obtained by in vitro spectroscopic measurements and those obtained biologically (Table V), providing further evidence for the equivalence of both in vitro and in vivo measurements in predicting transmission and protection. The collection and measurement of the forward scattered component, in addition to directly transmitted light, is essential for accurate estimation of total transmission, as pointed out by Everett et al. GTheir data indicate that transmission between 290 and 320 nm through Caucasian stratum corneum is about 30% to 60% and through
Journal of the American Academy of Dermatology
whole epidermis, about 10% to 35%. Our values of 47.6% and 29.4%, respectively, for solarsimulated radiation fail within these ranges. They obtained a value of 5% or less for one specimen of black epidermis, compared to our average value of 7.4%. Determination of the MED in deeply pigmented blacks is highly uncertain owing to the difficulty of visualizing redness. Often, the only clue that sunburn had developed comes days later when desquamation becomes evident. Our measurements indicate that in blacks the MED is about thirteen times the average MED of fair-skinned Caucasians. This value is, of course, highly variable and determined to a large extent by the intensity of pigmentation present. Blacks display a wide range of complexion, and, as shown by Olson et al, '6 the MED is directly proportional to the color intensity. Hausser and Vahle 17 found that the MED in one black subject was ten times that of Caucasians. Olson et al, 'G on the other hand, reported that the average MED of dark-complexioned blacks was about thirty-three times that of Caucasians. We now believe this to be an excessi.vely high figure, due to their use of an unfiltered highpressure mercury lamp rich in U VC wavelengths, which would greatly accentuate the difference between the two groups as shown with the FS-20 sunlamp tests in this study. The high protection against sunburning radiation by the epidermis of blacks deserves further comment. Photoprotection by melanin is accomplished through two mechanisms: absorption and scattering. Both processes will be influenced by the density and dis~ibution of melanosomes within keratinocytes. The larger, singly dispersed and heavily melanized melanosomes of blacks TM can absorb considerably more energy than the smaller, less dense, lightly melanized melanosomes of Caucasians. Furthermore, melanosomes in blacks are more numerous in the basal and parabasal layers than in the stratum corneum and are not degraded by lysosomal enzymes as in whites, ~9 Absorption by melanin is nonspecific and extends through the UV into the visible ranges, but it is most pronounced toward the
Volume 1 Number 3 September, 1979
shorter end of the U V spectrum. 2~ Specific absorption bands are present only in the infrared region. For the same reasons, attenuation of the incident radiation by scattering would also occur more efficiently in blacks. Hence, both the amount of melanin and the way it is distributed and packaged account for the excellent photoprotection by black epidermis. There is other evidence to support the importance of racial distribution of melanosomes. Maximal stimulation of melanin synthesis in Caucasians by repeated exposure to U VA provides only a modest increase in sunburn protection. 2~ The MED increases by a factor of 2 to 3. This procedure does not significantly increase the thickness of the stratum corneum, nor does it alter the natural distribution pattern of melanosomes. 22 Increased concentration of melanin in Caucasians, therefore, does not begin to approach the photoprotection conferred by natural melanin in blacks. On the other hand, stimulation of melanin by psoralen increases the MED by a factor of six. 23 This treatment is known to alter the distribution of melanosomes in whites to the racial pattern found in blacks. 24 We used a polycyclic hydrocarbon photosensitizer (fluoranthene) to estimate the transmission of U VA. The action spectrum lies within the U VA range (320-400 nm), and, therefore, our transmission values relate mainly to these wavelengths. U V A transmission was significantly higher than U V B in both racial groups. Our values of 17.5% and 55.5% through black and Caucasian epidermis, respectively, correspond well with spectroscopic measurements. Everett et al's ~ data indicate U VA transmission of about 5% to 15% through black epidermis and 35% to 50% through Caucasian specimens. Hence, about three or four times more U VA reaches the upper dermis of Caucasians than that of blacks. The appreciable degree of transmission through black epidermis can account for the not-infrequent occurrence of UVAmediated allergic photocontact dermatitis in this race. Finally, these results support not only the theoretical formalism but also the epidemiologic data cited by Beadle 2"~ in a study of vitamin D
Photoprotection by melanin
259
formation. If vitamin D is indeed formed in the lower epidermal layers, then for a corresponding exposure to U V light, only one-quarter as much could be formed in black skin as could be formed by Caucasians. If vitamin D supplementation in foods is not as effective as is natural vitamin D or is lacking entirely, then blacks living in Northern latitudes primarily indoors could be at risk for developing vitamin D deficiency.
REFERENCES
1. McFadden AW: Skin diseases in the Cuna Indians. Arch Dermatol 84: 1013- 1023, 1961. 2. Pathak MA, Fitzpatrick TB: The role of natural photoprotective agents in human skin, in Fitzpatrick TB, Pathak MA, Harber LC, Seiji M, Kukita A, editors: Sunlight and man. Tokyo, 1974, University of Tokyo Press, pp. 725-750. 3. KligmanAM: Solar elastosis in relation to pigmentation, in Fitzpatrick TB, et al, editors: Sunlight and man. Tokyo, 1974, University of Tokyo Press, pp. 157-163. 4. Keeler CE: Albinism, xeroderma pigmentosum and skin cancer, in Urbach F, editor: National Cancer Institute Monograph No. 10, 1963, pp. 349-359. 5. Oettle AG: Skin cancer in Africa, in Urbach F, editor: National Cancer Institute Monograph No. I0, 1963, pp. 197-214. 6. Everett MA, Yeargers E, Sayre RM, Olson RL: Penetration of epidermis by ultraviolet rays. Photochem Photobiol 5:533-542, 1966. 7. Guillaume AC: Le pigment epidermique la penetration des rayons u.v. mechanisme de l'organisme visa vis de ces radiation. Bull Soc Med Hop Paris 50:1133-1135, 1926. 8. Blum HF: Does the melanin pigment of human skin have adaptive value? Rev Biol 36:50-63, 1961. 9. Miescher G: Das problem des Lichtshutzer and der lichtwohnung. Strahlentherapie 35:403-443, 1930. I0. Anglin JH Jr, Raz AI: Diffuse spectral reflectance modification of skin by U V radiation. Photochem Photobiol 10:273-281, 1969. 11. Anglin JH Jr, Sayre RM, Batten WH: Change in total diffuse spectral reflectance of amino acids and proteins after ultraviolet irradiation. Photochem Photobiol 15: 534-537, 1972. 12. Sayre RM, Agin PP, LeVee GJ, Marlowe E: A comparison of in vivo and in vitro testing of sunscreening formulas. Photochem Photobiol 29:559-566, 1979. 13. Kligman AM: Biology of the stratum corneum, in Montana W, Lobitz WC, editors: The epidermis. New York, 1964, Academic Press, Inc. 14. Wiegand DA, Haygood C, Gaylor JR: Cell layers and density of Negro and Caucasian stratum corneum. J Invest Dermatol 62:563-568, 1974. 15. Thomson ML: Relative efficiency of pigment and horny layer thickness in protecting the skin of Europeans and
260
16. 17. 18. 19. 20.
K a i d b e y et al
Africans against solar ultraviolet radiation. J Physiol 127:236-246, 1955. Olson RL, Gaylor J, Everett MA: Skin color, melanin and erythema. Arch Dermatol 108:541-544, 1973. Hausser KW, Vahle W: Sunburn and suntanning, in Urbach F, editor: The biologic effect of ultraviolet radiation. London, 1969, Pergamon Press, pp. 3-21. Szabo G, Gertad AB, Pathak MA: Racial difference in the fate of melanosomes in human epidermis, Nature 22:1081- 1082, 1969. Olson RL, Nordquist J, Everett MA: The role of lysosomes in melanin physiology. Br J Dermatol 83: 189199, 1970. Edwards EA, Finklestein NA, Duntley SQ: Spectrophotometry of living human skin in the ultraviolet range. J Invest Dermatol 16:311-321,1951.
Journal of the American Academy of Dermatology
21. Kaidbey KH, Kligman AM: Sunburn protection by longwave ultraviolet radiation-induced pigmentation. Arch Dermatol 114:46-48, 1978. 22. Pathak MA, Jimbow K, Parrish JA, Kaidbey KH, Kligman AM, Fitzpatrick TB: Effect of UV-A, UV-B and psoralen on in vivo haman melanin pigmentation. Pigment Cell 3:291-298, 1976. 23. Gschnait F, Brenner W, Wolff K: Photoprotective effect of a psoralen-UVA induced tan. Arch Dermatol Res 263: 181-188, 1978. 24. Toda K, Pathak MA, Parrish JA, Fitzpatrick TB: Alteration of racial differences in melanosome distribution in human epidermis after exposure to ultraviolet light. Nature [New Biol] 236:143-145, 1972. 25. Beadle PC: The epidermal biosynthesis of cholecalciferol (vitamin D~). Photochem Photohiol 25:519-527, 1977.
I
Photoprotection by melanin black and Caucasian skin
I
a comparison of
Kays H. Kaidbey, M.D., Patficia Poh Agin, M.S., Robert M. Sayre, Ph.D., and Albert M. Kligman, M.D., Ph.D.
Philadelphia, PA, and Memphis, TN The photoprotective role of melanin was evaluated by comparing the transmission of ultraviolet (UV) radiation through skin samples of blacks and Caucasians, using both biologic and spectroscopic techniques. UVA transmission was measured using fluoranthene, which causes a phototoxic response to UVA wavelengths. UVB was measured by monitoring erythema produced by either a 150-watt xenon arc or FS-20 sunlamps. It was found that on the average, five times as much ultraviolet light (UVB and UVA) reaches the upper dermis of Caucasians as reaches that of blacks. Differences in transmission between the stratum corneum of blacks and of Caucasians were far less striking. The main site of UV filtration in Caucasians is the stratum corneum, whereas in blacks it is the malpighian layers. Melanin acts as a neutral density filter, reducing all wavelengths of light equally. The superior photoprotection of black epidermis is due not only to increased melanin content but also to other factors related to packaging and distribution of melanosomes. Not only are these data consistent with epidemiologic evidence, but they also may indicate why blacks are less disposed to phototoxic drug responses as well as less susceptible to acute and chronic actinic damage. (J AM ACAD DERMATOL1:249-260, 1979.)
The evidence for the photoprotective role of melanin is compelling. ~-a Both epidemiologic and experimental studies have already shown that blacks and darkly pigmented Caucasians are far less susceptible to acute and chronic actinic damage than are fair-skinned individuals. There is a clear inverse correlation between the degree of histologic solar elastosis and skin complexion at every decade of life. a In blacks, the hyperplasia of elastic fibers is greatly delayed and of lesser magReprint requests to: Dr. Kays H. Kaidbey, Duhring Laboratories, Department of Dermatology, University of Pennsylvania, 3500 Market St., Suite 203, Philadelphia, PA 19104.
0190-9622/79/030249+ 12501.20/0 9 1979 Am Acad Dermatol
nitude. Squamous and basal cell cancer, believed to be at least partially due to chronic exposure to actinic radiation, is very rare in blacks living in the sunbelt areas. Black albinos, on the other hand, are very susceptible to actinic damage and readily develop precancerous and cancerous lesions. McFadden 1 and KeeleP detected keratoses, atrophy, and telangiectasias in the albino Cuna Indians as early as the age of 2 years. By the age of 7, the incidence of keratoses approached 100% and many had developed squamous cell carcinomas and actinic cheilitis. Normally pigmented Indians did not develop these changes. Oettle 5 also found a high incidence of precancerous keratoses 249
250
Journal of the American Academy of Dermatology
Kaidbey et al
Abbreviations used E
whole epidermis (including stratum corneum) EPI epidermis MAL malpighian layer MED minimal erythema dose MPD minimum phototoxic dose PF protection factor STC stratum corneum UV ultraviolet UV spectrum in vitro forward scattering spectrum using integrating sphere UVA-SS solar simulator UVA transmission in human volunteers UVB-FS FS sunlamp UVB transmission on human volunteers UVB-SS solar simulator UVB transmission in human volunteers
among hypopigmented individuals living in subtropical belts. Another example is vitiliginous skin, which is unusually susceptible to sunburn. The natural photoprotective role o f melanin is beyond doubt. More direct evidence of a quantitative nature has come from in vitro spectroscopic studies on excised skin specimens. Pathak and Fitzpatrick 2 compared the transmission characteristics of both stratum corneum and whole epidermis from blacks and fair-skinned Caucasians. Caucasoid epidermis was more transparent to ultraviolet and visible wavelengths. At 300 nm, there was 10% to 15% transmission through Caucasian epidermis, as compared to only 2% to 5% through black epidermis. Everett et al, 6 in a comprehensive investigation of transmission spectroscopy by human skin specimens, studied one sample of black epidermis. By using integrating spheres along with recording spectrophotometry, they were able to measure forward scattering and hence total transmission more accurately than had been accomplished before. They also found considerably greater transmission of light over the entire 300 to 600 nm wavelength range than had been reported previously. This study also showed that values for a skin specimen from an American Indian were
intermediate to those of Caucasoid and black epidermis. Although the protective function of melanin seems abundantly clear, quantitative data on the comparative contribution of the horny layer and epidermis in blacks and whites are few. In the past, some investigators concluded that in Caucasians, thickening of the stratum corneum was more important than pigmentation in photoprotective adaptation. 7-9 This question can be resolved in part by accurate measurements of transmission through the various skin layers. Although spectroscopic studies of excised skin are quite informative, there has been some concern about the relevance of in vitro measurements to actual transmission in vivo. it is difficult to measure all of the components of a light beam after it has traversed an optically heterogenous medium such as skin: reflection, refraction, absorption, backward and forward scattering all have to be measured and accounted for before meaningful conclusions can be drawn. The object of the present investigation was to assess quantitatively the natural photoprotective role of melanin. This was done by comparing transmission of ultraviolet (U V) radiation through excised skin specimens of blacks and whites. Biologic measurements were obtained by using skin sheets as filters and the values correlated with those obtained by spectroscopic techniques. MATERIALS AND METHODS Chemicals
Trypsin (type II, hog pancreas) was purchased from Sigma Chemical Co., St. Louis, MO. Fluoranthene was obtained from Eastman Chemicals, Rochester, NY. Skin specimens
Specimens of black and Caucasian skin were obtained from the abdominal region of cadavers and quick-frozen with liquid nitrogen until use. After thawing at room temperature, the specimens were placed in a water bath at 60 C for 2 minutes, and the epidermis was peeled off the dermis. To obtain stratum corneum, the epidermal sheets were floated in 0. 01% trypsin solution (buffered at pH 8) overnight. Epidermal cells were then wiped away with cotton-tip applicators. The corneum sheets were rinsed in water, allowed to air-dry
Volume 1 Number 3 September, 1979
Photoprotection by melanin
251
T a b l e I. Skin s p e c i m e n identification Tests performed Specimen
Race
Age
Sex
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA BB CC DD EE FF GG HH II JJ
B B W B W W B W W W W W B B B B B B B B B B W W W W W W W W B B W W W W
47 50 57 50 57 61 30 67 78 72 60 51 69 50 24 29 41 46 60 60 61 75 22 40 53 54 60 62 72 74 48 71 70 58 69 48
F F M M F M M F F F F F M F F M F F F F M M F F F F M M M F F F F F M F
UVB-SS Biol.
UVB-FS Biol.
STC & E STC & E STC & E
UVA-SS Biol.
E E E
STC & E
STC & E
STC STC STC STC STC STC STC STC STC STC STC STC STC STC STC STC
& & & & & & & & & & & & & & & &
E E E E E E E E E E E E E E E E
E
STC & E E E E E E E
STC STC STC STC STC STC
& & & & & &
UV spectrum Forward scattering
STC & STC & STC & STC & STC & STC STC STC STC E E E E E
E E E E E
E E E E E E E E E E E E E E E E
E E E E E E
B: Black skin, W: Caucasian skin. at room temperature, and then refrigerated until use. Each skin sample (epidermis and corneum) was used as a filter for transmission measurements in only one experiment to avoid possible U V-induced alteration of the tissue, The possibility of this had previously been suggested and is known to be common with gelatin protein filters.10,11 Table I identifies each skin sample and the tests performed on skin specimens from that individual. Not all samples used in the in vivo studies were suitable
for spectroscopic forward scattering examination, The specimens were examined histologically after formalin fixation. Biologic determination of UVB transmission
Transmission of UVB radiation (290-320 nm) was measured by comparing the doses required to produce minimal erythema dose (MED) in the untanned skin of fair-skinned Caucasian subjects (types I. II, and IlI)
252
Journal of the American Academy of Dermatology
Kaidbey et al
T a b l e I I . UVB p r o t e c t i o n - - s o l a r simulator Black skin filters
White skin filters
Specimen
EPI PF
STC PF
Specimen
EPI PF
O P Q R A B S T U V Mean • SD
12 18 18 16 10 10 13 10 I5 12 13.4 __- 3.16
3 4 4 3 3 2 4 3 4 3 3.3 • 0.67
W X Y Z C AA BB J CC DD
4 3 5 3 3 4
2 2 2 2 2 3
3
2
3 3 3
2 2 2 2.1 +-_ 0.31
[ I
STC PF
3.4 •
T a b l e I I I . UVB protection using FS-20 sunlamps Black skin filters
White skin filters
Specimen
EPI PF
STC PF
Specimen
EPI PF
P G EE FF Mean • SD
29 30 32 30 30.25 ___ 1.26
4 5 5 5 4.75 --+ 0.5
GG HH II JJ
5 5 4 4 4.5 ___ 0.58
with and without filtering by the test skin specimens. Informed consent was obtained prior to testing. On unprotected skin, the MED was first determined by administering a series of geometrically increasing exposures to the lower back from a 150-watt xenon solar simulator,* so that each dose was 25% larger than the previous one. The MED was the smallest dose required to produce visible erythema 24 hours later. Then 1.52. 0-cm squares of stratum corneum or epidermis were taped to a thin plastic frame with I x 1-cm windows. The frame was then securely positioned over the lower back. One-tenth milliliter of physiologic saline was pipetted under each square to wet the specimen and assure intimate contact with the skin. Increasing multiples of the control MED energy were then given through the specimens until erythema appeared. The ratio of the MED through the test specimen to the MED of unprotected normal skin represents a protection factor (PF). Transmission of the biologically effective radiation is the reciprocal of the PF. *Solar Light Co., Philadelphia, PA; filtered with 2.0 mm Sehott WG320 filter.
[
STC PF
3 2 2 3 2.5 - 0.58
To evaluate the potential effects produced by a different UVB light distribution, four specimens from each group were subjected to the same procedure just outlined, except that the U VB source was a bank of five closely set fluorescent sunlamps (FS-20, Westinghouse) mounted in a reflector housing. The lamp-toskin distance was 12 cm. Biologic d e t e r m i n a t i o n o f U V A t r a n s m i s s i o n
The transmission of U V A radiation (320-400 nm) can be measured by comparing the minimum phototoxic exposure of skin pretreated with fluoranthene, a hydrocarbon photosensitizer activated by U VA wavelengths. Fluoranthene in 95% ethanol (5 t z l / c m 2 of 0.5%) was delivered to 1-cm squares of skin over the midback of white volunteers. The sites were allowed to air-dry; then they were covered by square patches of nonwoven cotton cloth (Webfil; Curity) and fastened to the skin with occlusive tape (Blenderm). Two hours later the patches were removed and the sites exposed to increasing doses of U VA from the xenon source using a 2-mm WG345 filter to eliminate the UVB wavelengths.
Volume 1 Number 3 September, 1979
The minimum phototoxic dose (MPD), the smallest dose of UVA necessary to produce minimal visible erythema, was determined 24 hours later. The MPD was determined again in an identical fashion on the opposite side of the back through the test skin specimens as already described. A U VA protection factor (PF) in this instance is the ratio of the MPD through the test specimen to the MPD of normal unprotected skin. Again, the reciprocal of the PF represents the transmission of UVA-effective radiation through the test skin samples. Energy output measurements were made using a calibrated thermopile (Eppley Laboratories Calibration No. 11609) attached to a Keithly millimicrovoltmeter. UVB intensity from the xenon solar simulator was 19.70 mw/cm2; UVA irradiance was 15.82 mw/cmL The U VB irradiance at the skin surface from the bank of fluorescent FS-20 sunlamp tubes (lamp-to-skin distance, 12 cm) was 0.6 m w / c m 2.
Spectroscopic determination of UV transmission: In vitro forward scattering Dried samples of human epidermis or stratum corneum measuring at least 4 cm 2 were rehydrated by immersion in distilled water at room temperature for approximately one minute. The hydrated sample was floated onto a quartz carrier plate and inserted into the forward scattering compartment of a Beckman Acta MVI spectrophotometer fitted with a diffuse reflectance sphere. Each sample was scanned from 250 to 400 nm. In both the sample and the reference beams, 2-ram UG-5 filters were used to remove possible visible fluorescence. Baseline absorbance was subtracted from each sample scan, and protection factors were calculated as previously described. TM RESULTS
Biologic measurements
Photoprotection by melanin 253
Table IV. Protection factors of w h o l e epidermis against U V A (solar simulator)
Black epidermalfilter Specimen [ PF
1 Whit..eepidermalfilt.er .... Specimen
O 7.0 P 10.6 Q 4.0 R 3.5 A 3.5 B 3.3 S 4.3 T 6.0 U 6.7 V 8.0 Mean +__SD 5.69 • 2.41
W X Y Z C AA BB J CC DD
PF
2.0 1.5 2.3 1.3 2.0 2.4 1.3 1.6 2.0 1.6 1.8 _-_ 0.39
Table V . Comparison o f in vitro and in vivo protection
Specimen A B C A D C J
I I
Race Sampletype B B W B B W W
STC STC STC EPI EPI EPI EPI
vtro (predicted) PF
In vivo PF
2.3 2.3 2,2 13.6 12.4 4.8 3.6
3 2 2 10 10 3 3
B: Black skin. W: Caucasian skin.
stratum corneum was 3. 3 and for Caucasians, 2. 1. Hence, the mean transmission by corneum for blacks was 30.3% (range, 25% to 5 0 % ) and 47.6% (range, 33.3% to 50%) for Caucasians.
Transmission of UVB with solar simulator.
UVB transmission with FS-20 sunlamps.
The individual and m e a n PF values of epidermal and stratum c o r n e u m specimens from ten black and ten Caucasian subjects are shown in Table II. The mean PF for black epidermis was 13.4, as compared to 3.4 for white epidermis. The mean transmission by black epidermis was therefore 7.4% (range, 5.5% to 10%) compared to 29.4% (range, 20% to 33.3%) for Caucasian epidermis. The difference was less p r o n o u n c e d with stratum corneum specimens. The mean PF of black
Higher PF values were obtained with this light source (Table III). T h e mean P F for black epidermis was 30.25 (compared to 13.4 with the solar simulator) and 4.5 for white epidermis (compared to 3.4). T h e mean transmission was therefore 3.3% for black epidermis compared to 22.2% for Caucasian epidermis. T h e mean PF for the black stratum c o r n e u m was also greater with this source (4.75 c o m p a r e d to 3. 3 with the solar simulator).
254
Journal of the American Academy of Dermatology
Kaidbey et al
8TRATUlit CORNEUM ..,....,o'~
DI~K
.,, .,.' ~".. ",., 1.1
//
\
G.""
j. C ,8
R* A" E |
"',. '.,, .,~
..
"'4.% "",.
"%
.3
WAVELENGTH
Fig. 1. Absorption spectra of the STC samples studied by forward scattering spectroscopy. Between 290 and 320 nm, the two groups are almost indistinguishable, except for sample G. Sample identification is taken from Table I.
Transmission of U V A . Table I V shows that the PF of black epidermis is about 5.7, significantly higher than that of white epidermis at 1.8 (p < 0.005, Student's t test). T h e mean U V A transmission, therefore, by black epidermis was 17.5% compared to 55.5% for Caucasians. In vitro spectroscopic measurements Figs. 1 and 2 show the absorption spectra of all of the stratum corneum and epidermis samples studied. Samples from the same individual are identified by identical labeling in both figures. Several samples do not have corresponding matches in both figures. In Fig. 1, the stratum corneum samples do not segregate into distinguishable groups by melanin content at wavelengths shorter than 320 nm. Fig. 2, however, clearly demonstrates that there are great differences in the optical qualities of black and white epidermis. It is quite striking that the two groups are nonoverlapping at all wavelengths, from 250
to 400 nm. Fig. 3 shows the average absorption spectrum for each type of skin sample examined: black epidermis, black stratum corneum, white epidermis, and white stratum corneum. White epidermis and black stratum corneum are indistinguishable in the sunburn (UVB) range 290320 nm. White stratum corneum is only slightly less protective than black stratum corneum. Black epidermis, however, is very photoprotective at all wavelengths. Table V compares the protection values of skin samples determined by both forward scattering and in vivo studies. The predicted protection factors obtained from spectroscopic measurements were calculated from the model described by Sayre et al TM using the skin specimen absorbance in place of a sunscreen. The stratum corneum samples, black and white, are indistinguishable by this method and give a protection factor averaging about 2. 2. The epidermal samples, however, are quite different. The black
Volume 1 Number 3 September, 1979
Photoprotection by melanin 255
3.f
|PlDEflldl8
~ .......
\
29
2.l
-
.........~.,,~~
_
willie
.................
black
,g ,~
Pt B ~
IA
t
i
,q
',.
t!
'%
,
~
iiiiiiiiiii,iiiiiiiiiiiiiiiiiii ii WAVELENGTH
Fig, 2, Absorption spectra of the epidermal samples studied by forward scattering spectroscopy. Note that the two groups are distinct at all wavelengths. Sample identification is taken from Table I. epidermal samples are three to four times as protective as the white epidermal samples. The predicted in vitro protection correlates well with the in vivo values. Table VI summarizes the results of light transmission values through the stratum corneum and whole epidermis, and gives a predicted value for malpighian layers. Transmission of light through the entire epidermis is equivalent to the transmission of light through the stratum corneum multiplied by the transmission of light by the malpighian layers. High, low, and average values are given for each set of samples examined. Again, comparison of the values obtained by human testing to
those obtained by in vitro spectroscopic methods shows good correlation using both UVA and UVB. For Caucasian skin, between 30% and 50% of incident UVB radiation is transmitted by the stratum corneum. Yet, predictions from each method regarding the transmission by the malpighian layers indicate tbat 60% to 70% of the available light is transmitted, that is, more UVB radiation is removed by the Caucasian stratum corneum than is removed by the underlying epidermal cells. For U VA, approximately the same amount of light is transmitted by both major skin layers. For black skin, Table VI illustrates that
256
Journal of the American Academy of Dermatology
Kaidbey et al
Maak epMeNNs
................ ~ strotwn~fneum ........ w~Ite e p I d e ~ l s ..... white ~4ratum cwmeum
/\'
~2s
;
9" ,75 i
I
\ -\
I
~ : I
.50
~n'~'n"%" .25
9
.~~ ..r.:.... ...... -~.::::..:.: .............
,,|
250
27S
300
325 350 WAVELENGTH
311'5
400 NM
Fig. 3. Average absorption spectrum for each type of skin specimen examined. Black stratum corneum and white stratum corneum are quite similar, whereas the epidermal samples are very different. However, from 300 to 400 nm, black stratum corneum and white epidermis are almost identical. there is wide variability between individual skin samples, depending upon tile test method used by a factor of two to four. Yet the average transmission of U V B is 20% to 30% regardless of method. In all cases, less light must be transmitted through the malpighian layers than through the stratum corneum if the 3% to 7% penetrating the entire epidermis is to be realized. This means the mal-
pighian layers of black skin remove twice as much U VB radiation as their stratum corneum. For U VA, possibly even greater removal of radiation occurs in black malpighian layers. DISCUSSION Other factors besides melanin contribute to natural photoprotection. These include absorption
Volume 1 Number 3 September, 1979
Table
Photoprotection by melanin
257
VI. Summary of transmission of light through skin layers* White skin (predicted)
Wave band
Range
Black skin (predicted)
STC
MAL
EPI
STC
0.618
0.333 0.294 0.200
0.500 0.303 0.250
0.250 0.222 0.200 0.500 0.286 0.164
0.250 0.210 0.200 0.526 0.278 0.135
0.539 0.322 0.166
0.537 0.325 0.134
UVB (solar simulator --human)
High AV Low
0.500 0.476 0.333
UVB (FS-20 sunlamp - - human) UVB (predicted) in vitro)
High AV Low High AV Low
0.500 0.400 0.333 0.556 0.435 0.370
UVB 305 nm (in vitro)
High AV Low
0.575 0.461 0.380
UVA (fluoranthene phototoxicity --human)
High AV Low
UVA 360 nm (in vitro)
High AV Low
0.555
0.657
0.698
0.248
0. I56
0.147
0.157
0.769 0.556 0.417 0.957 0.824 0.707
0.769
0.782 0.637 0.213
0.100 0.075 0.056 0.034 0.033 0.031 0.088 0,041 0.019 0.089 0.051 0.031 0.303 0.176 0.094
0.926 0.617 0.467
0.198
0.199 0.122 0.039
*Transmission values shown in the table are the reciprocal of the PFs.
and scattering by keratin, collagen, and other macromolecules, absorption by aromatic amino acid residues in proteins, and probably absorption due to the presence of some carotenoid pigments. The present work, however, is concerned primarily with the role of melanin. Since there is no evidence suggesting major differences in the thickness or composition of the epidermal layers in the two racial groups, TM 14 we assume that the observed differences in transmission are mainly due to this pigment. By using skin specimens as filters, and living skin as the detector, photoprotection was measured in much the same way as is used to assess the efficacy of a sunscreening agent. The protection by black epidermis against sunburn was impressive indeed. A PF of 13.4 is matched only by a few of the most effective commercially available sunscreens. The average spectrum shown in Fig. 3 demonstrates why this is so. The total transmission of UVB was less than one fourth of that by Caucasian epidermis. Wide variations were seen, however, in the individual values, which almost
certainly reflect differences in the intensity of pigmentation among blacks. Blacks are no more a homogeneous group than whites. It is interesting to note that the difference in U VB transmission by black and white stratum corneum was far less striking. Caucasian specimens on the average transmitted only about 1.5 times more U V B than did blacks. It is clear, therefore, that in blacks, filtration of sunburning radiation occurs primarily within the living epidermal layers, and not in the stratum corneum. This is also evident from the transmission data outlined in Table VI. In contrast, the primary site of filtration in Caucasians is the stratum corneum, where over 50% of the incident radiation is absorbed. Clearly, then, melanin appears to exhibit much greater photoprotection in the black malpighian layers than in the stratum corneum. This is true for both the U VA and U VB wavebands. When melanin in black skin reaches the stratum corneum, it appears to lose much of its protective capabilities and does not absorb and scatter radiation as effectively as in the malpighian layers. Whether this loss of protective ability is
258
Kaidbey et al
due to a biochemical or physical process is not known. Thomson '5 used a photographic technique and radiation from a mercury vapor lamp to compare transmission by stratum corneum of blacks mad Caucasians. The stratum corneum specimens were obtained from cantharidin blister roofs. Although he found no difference in the thickness of the stratum corneum between the two groups, the total transmission over the 300 to 400 nm range was about 3.5 times greater in Caucasians and was less marked, about 2.5 times, in the erythemal range (290 to 320 nm). He attributed the lessened transmission to the presence of melanin. We found an even smaller difference between the two groups with solar-simulating radiation. These results indicate that in blacks as well as in whites the keratinocytes of the upper epidermis are exposed to substantial radiation levels. Seemingly low transmission values were obtained in tests with the FS-20 sunlamp tubes. This is due to the presence in this source of lower UV wavelengths extending into the U VC range. These shorter wavelengths are more effectively absorbed and scattered by the upper epidermal layers, even though they are more erythemogenic. Since this occurs more efficiently in blacks than in Caucasian skin, U VC-rich sources tend to yield elevated PF values. Similarly, differences in the threshold dosages for U V erythema between the two groups will thereby be accentuated. It is imperative, therefore, to carefully interpret such results depending upon the radiation source chosen when transmission is measured by this technique. There was good agreement between the estimated protection values obtained by in vitro spectroscopic measurements and those obtained biologically (Table V), providing further evidence for the equivalence of both in vitro and in vivo measurements in predicting transmission and protection. The collection and measurement of the forward scattered component, in addition to directly transmitted light, is essential for accurate estimation of total transmission, as pointed out by Everett et al. GTheir data indicate that transmission between 290 and 320 nm through Caucasian stratum corneum is about 30% to 60% and through
Journal of the American Academy of Dermatology
whole epidermis, about 10% to 35%. Our values of 47.6% and 29.4%, respectively, for solarsimulated radiation fail within these ranges. They obtained a value of 5% or less for one specimen of black epidermis, compared to our average value of 7.4%. Determination of the MED in deeply pigmented blacks is highly uncertain owing to the difficulty of visualizing redness. Often, the only clue that sunburn had developed comes days later when desquamation becomes evident. Our measurements indicate that in blacks the MED is about thirteen times the average MED of fair-skinned Caucasians. This value is, of course, highly variable and determined to a large extent by the intensity of pigmentation present. Blacks display a wide range of complexion, and, as shown by Olson et al, '6 the MED is directly proportional to the color intensity. Hausser and Vahle 17 found that the MED in one black subject was ten times that of Caucasians. Olson et al, 'G on the other hand, reported that the average MED of dark-complexioned blacks was about thirty-three times that of Caucasians. We now believe this to be an excessi.vely high figure, due to their use of an unfiltered highpressure mercury lamp rich in U VC wavelengths, which would greatly accentuate the difference between the two groups as shown with the FS-20 sunlamp tests in this study. The high protection against sunburning radiation by the epidermis of blacks deserves further comment. Photoprotection by melanin is accomplished through two mechanisms: absorption and scattering. Both processes will be influenced by the density and dis~ibution of melanosomes within keratinocytes. The larger, singly dispersed and heavily melanized melanosomes of blacks TM can absorb considerably more energy than the smaller, less dense, lightly melanized melanosomes of Caucasians. Furthermore, melanosomes in blacks are more numerous in the basal and parabasal layers than in the stratum corneum and are not degraded by lysosomal enzymes as in whites, ~9 Absorption by melanin is nonspecific and extends through the UV into the visible ranges, but it is most pronounced toward the
Volume 1 Number 3 September, 1979
shorter end of the U V spectrum. 2~ Specific absorption bands are present only in the infrared region. For the same reasons, attenuation of the incident radiation by scattering would also occur more efficiently in blacks. Hence, both the amount of melanin and the way it is distributed and packaged account for the excellent photoprotection by black epidermis. There is other evidence to support the importance of racial distribution of melanosomes. Maximal stimulation of melanin synthesis in Caucasians by repeated exposure to U VA provides only a modest increase in sunburn protection. 2~ The MED increases by a factor of 2 to 3. This procedure does not significantly increase the thickness of the stratum corneum, nor does it alter the natural distribution pattern of melanosomes. 22 Increased concentration of melanin in Caucasians, therefore, does not begin to approach the photoprotection conferred by natural melanin in blacks. On the other hand, stimulation of melanin by psoralen increases the MED by a factor of six. 23 This treatment is known to alter the distribution of melanosomes in whites to the racial pattern found in blacks. 24 We used a polycyclic hydrocarbon photosensitizer (fluoranthene) to estimate the transmission of U VA. The action spectrum lies within the U VA range (320-400 nm), and, therefore, our transmission values relate mainly to these wavelengths. U V A transmission was significantly higher than U V B in both racial groups. Our values of 17.5% and 55.5% through black and Caucasian epidermis, respectively, correspond well with spectroscopic measurements. Everett et al's ~ data indicate U VA transmission of about 5% to 15% through black epidermis and 35% to 50% through Caucasian specimens. Hence, about three or four times more U VA reaches the upper dermis of Caucasians than that of blacks. The appreciable degree of transmission through black epidermis can account for the not-infrequent occurrence of UVAmediated allergic photocontact dermatitis in this race. Finally, these results support not only the theoretical formalism but also the epidemiologic data cited by Beadle 2"~ in a study of vitamin D
Photoprotection by melanin
259
formation. If vitamin D is indeed formed in the lower epidermal layers, then for a corresponding exposure to U V light, only one-quarter as much could be formed in black skin as could be formed by Caucasians. If vitamin D supplementation in foods is not as effective as is natural vitamin D or is lacking entirely, then blacks living in Northern latitudes primarily indoors could be at risk for developing vitamin D deficiency.
REFERENCES
1. McFadden AW: Skin diseases in the Cuna Indians. Arch Dermatol 84: 1013- 1023, 1961. 2. Pathak MA, Fitzpatrick TB: The role of natural photoprotective agents in human skin, in Fitzpatrick TB, Pathak MA, Harber LC, Seiji M, Kukita A, editors: Sunlight and man. Tokyo, 1974, University of Tokyo Press, pp. 725-750. 3. KligmanAM: Solar elastosis in relation to pigmentation, in Fitzpatrick TB, et al, editors: Sunlight and man. Tokyo, 1974, University of Tokyo Press, pp. 157-163. 4. Keeler CE: Albinism, xeroderma pigmentosum and skin cancer, in Urbach F, editor: National Cancer Institute Monograph No. 10, 1963, pp. 349-359. 5. Oettle AG: Skin cancer in Africa, in Urbach F, editor: National Cancer Institute Monograph No. I0, 1963, pp. 197-214. 6. Everett MA, Yeargers E, Sayre RM, Olson RL: Penetration of epidermis by ultraviolet rays. Photochem Photobiol 5:533-542, 1966. 7. Guillaume AC: Le pigment epidermique la penetration des rayons u.v. mechanisme de l'organisme visa vis de ces radiation. Bull Soc Med Hop Paris 50:1133-1135, 1926. 8. Blum HF: Does the melanin pigment of human skin have adaptive value? Rev Biol 36:50-63, 1961. 9. Miescher G: Das problem des Lichtshutzer and der lichtwohnung. Strahlentherapie 35:403-443, 1930. I0. Anglin JH Jr, Raz AI: Diffuse spectral reflectance modification of skin by U V radiation. Photochem Photobiol 10:273-281, 1969. 11. Anglin JH Jr, Sayre RM, Batten WH: Change in total diffuse spectral reflectance of amino acids and proteins after ultraviolet irradiation. Photochem Photobiol 15: 534-537, 1972. 12. Sayre RM, Agin PP, LeVee GJ, Marlowe E: A comparison of in vivo and in vitro testing of sunscreening formulas. Photochem Photobiol 29:559-566, 1979. 13. Kligman AM: Biology of the stratum corneum, in Montana W, Lobitz WC, editors: The epidermis. New York, 1964, Academic Press, Inc. 14. Wiegand DA, Haygood C, Gaylor JR: Cell layers and density of Negro and Caucasian stratum corneum. J Invest Dermatol 62:563-568, 1974. 15. Thomson ML: Relative efficiency of pigment and horny layer thickness in protecting the skin of Europeans and
260
16. 17. 18. 19. 20.
K a i d b e y et al
Africans against solar ultraviolet radiation. J Physiol 127:236-246, 1955. Olson RL, Gaylor J, Everett MA: Skin color, melanin and erythema. Arch Dermatol 108:541-544, 1973. Hausser KW, Vahle W: Sunburn and suntanning, in Urbach F, editor: The biologic effect of ultraviolet radiation. London, 1969, Pergamon Press, pp. 3-21. Szabo G, Gertad AB, Pathak MA: Racial difference in the fate of melanosomes in human epidermis, Nature 22:1081- 1082, 1969. Olson RL, Nordquist J, Everett MA: The role of lysosomes in melanin physiology. Br J Dermatol 83: 189199, 1970. Edwards EA, Finklestein NA, Duntley SQ: Spectrophotometry of living human skin in the ultraviolet range. J Invest Dermatol 16:311-321,1951.
Journal of the American Academy of Dermatology
21. Kaidbey KH, Kligman AM: Sunburn protection by longwave ultraviolet radiation-induced pigmentation. Arch Dermatol 114:46-48, 1978. 22. Pathak MA, Jimbow K, Parrish JA, Kaidbey KH, Kligman AM, Fitzpatrick TB: Effect of UV-A, UV-B and psoralen on in vivo haman melanin pigmentation. Pigment Cell 3:291-298, 1976. 23. Gschnait F, Brenner W, Wolff K: Photoprotective effect of a psoralen-UVA induced tan. Arch Dermatol Res 263: 181-188, 1978. 24. Toda K, Pathak MA, Parrish JA, Fitzpatrick TB: Alteration of racial differences in melanosome distribution in human epidermis after exposure to ultraviolet light. Nature [New Biol] 236:143-145, 1972. 25. Beadle PC: The epidermal biosynthesis of cholecalciferol (vitamin D~). Photochem Photohiol 25:519-527, 1977.
