Effect of topical indomethacin on UVR-induced redness and prostaglandin E levels in sunburned guinea pig skin

PROSTAGLANDINS

EFFECT

OF TOPICAL

INDOMETHACIN

ON UVR-INDUCED

REDNESS

AND PROSTAGLANDIN

E LEVELS

IN SUNBURNED

GUINEA

PIG SKIN

Diane Sekura Snyder Dermatology Department School of Medicine

University of Miami Miami, Fla. 33136

ABSTRACT Prostaglandin E (PGE) levels in the skin have been shown to be elevated during the 24 hr. period following exposure of guinea pig skin to ultraviolet radiation from 280-320 nm in the so-called UVB or “sunburn spectrum”. The development of increased PGE levels paralleled the development of the delayed phase of erythema. When applied immediately after UVB exposure, one topical application of 2.5% Indomethatin (IM) simultaneously decreased the redness and PGE levels in the The IM effect persisskin to near normal within one hr. of treatment. ted for approximately 24 hrs. By 48 hrs. the PGE level in the sunburned skin was near normal whether or not the skin had been treated The redness which was evident at 48 hrs. did not repond to with IM. This suggests that a substance(s) other additional topical IM treatment. than PGE may be involved in this latter portion of UVB-induced erythema. A major role for PGE as an early mediator or potentiator of UVBinduced erythema has been deomonstrated. Any involvement of PGE with UVB-induced cell death or alterations in DNA synthesis seems unlikely.

ACKNOWLEDGEMENTS The author wishes toacknowledge Ms. Maria Green for her excellent technical assistance and Dr. Vincent A. Ziboh for his critical review of the manuscript. This work was supported maceuticals.

APRIL

in part by a grant from

1976 VOL. 11 NO. 4

Westwood

Phar-

631

PROSTAGLANDINS

INTRODUCTION Prostaglandins of the E series (PGE) have been implicated as mediators of the delayed phase of erythema in skin exposed to sunlight or ultraviolet radiation (UVR). PG or PG-like compounds have been isolated from human dermal perfusates 16 to 48 hrs. following UVR exposure (1) and from rat skin one hr. after UVR exposure (2). In both humans and guinea pigs, intradermal injections of nonsteroidal anti-inflammatory agents, known inhibitors of PC synthetase, will decrease and delay UVR-induced redness in skin exposed to the “sunburn spectrum” of 280 nm to 320 nm known as UVB (3). Topically applied indomethacin (IM) has been demonstrated to alleviate the dermal symptoms of sunburn (4). Direct evidence that IM diminishes redness by decreasing PG levels in the skin has been lacking. It is the purpose of this study to determine-the levels of PGE in sunburned skin during the 48 hrs. following UVB exposure and to correlate these levels with the appearance of UVB-induced erythema. The relationship between PGE levels and UVB-induced cell death and altered DNA synthesis was also explored. The ability of topically applied IM to influence PGE levels in the skin and to influence UVBinduced redness and epidermal damage has been evaluated.

MATERIALS

AND METHODS

MATERIALS PGE2 (kindly supplied by Dr. John Pike or purchased from Analabs) was stored in absolute ethanol at a concentration of 10 mg/ml at -2O’C. Tritiated PGE2 (New England Purity was checked by TLC each month, specific activity 117-210 C/mMole) was also stored in absolute Nuclear, ethanol at -2O’C. Tritiated PGE2 was used only if the radiochemical purity exceeded 90 70 as determined by TLC. Antiserum to PGE was prepared, supplied by Dr. Marc Goldyne (Mayo

characterized Clinic).

All other chemicals were of reagent grade. reagent grade and freshly distilled before use. PRODUCTION

AND

EVALUATION

All

(5) and generously

solvents

were

of

OF ERYTHEMA

Fourteen English short hair female guinea pigs weighing 300 to 600 They received 5 times g were prepared as previously reported (3). their minimal erythema dose of UVB from a Westinghouse FS 20 sunImmediately following lamp to discrete areas on their dorsal surfaces.

632

APRIL

1976

VOL. 11 NO. 4

PROSTAGLANDINS

irradiation

2. 5 70 (w/v)

IM solution

in ethanol:propylene

glycol:dimethyl-

acetamide

(19:19:2) was applied to a portion of the UVB exposed skin Another portion was treated with vehicle only, or in some (5 I.tl/cm2). cases received no treatment.

At t = 0, 0.25, 1, 4, 24 and 48 hrs. following UVB exposure, erythema was evaluated using the following subjective scale: 0 = no erythema; 0.5 = bearly detectable erythema; 1 = definite but faint erythema; 2 = moderate erythema and 3 = brisk, bright erythema. IN VIVO TRITIATED

THYMIDINE

INCORPORATION

At 1, 4 and 48 hrs. following UVB exposure, tritiated thymidine incorporation was determined autoradiographically (4) in normal skin, sunburned skin treated with IM or vehicle and sunburned skin which received no treatment. The labeling index (L. I. ) is defined as the number of heavily labeled basal cells divided by the total number of basal cells. A minimum of 1000 basal cells was counted for each histologic specimen. EXTRACTION

OF PG’S

FROM

SKIN

At t = 0, ‘0.25, 1, 4, 24 and 48 hrs. following UVB exposure a 4disposable skin biopsy punch was used to obtain specimens in duplicate of unirradiated, normal skin, skin exposed to UVB and skin exposed to UVB then immediately treated with IM or vehicle. The skin was kept cold until weighed (within 10 min. of excision) then 1-2 ml of CHC13: methanol (2:l) were added to stop enzymic release of precursor fatty acid and subsequent in vitro biosynthesis of PGE by the skin. Immediately before homogenization m 6000 dpm of tritiated PGE2 were added to the tissue in order to calculate the 70 recovery. Skin was homogenized in a minimal volume of CHCl3:methanol (2:l) at 4OC with a glass, hand homogenizer. Tissue debris was removed by filtering through a scintered glass filter. The clear filtrate was evaporated to dryness under nitrogen to yield a residue containing total lipids. SEPARATION PHY

OF PG’s

BY SILICIC

ACID

COLUMN

CHROMATOGRA-

Silicic acid column chromatography was similar to methods previously described (6). The PGE containing fraction eluted from the column in 12 ml of benzene:ethyl acetate:methanol (60:40:2) and was evaporated to dryness with nitrogen. The PGE containing residue was dissolved in 0.15 M NaC1, 0.01 M Tris HCl buffer, pH 7.4 containing 0.1% gelatin. An aliquot was counted in a Packard TriCarb Scintillation counter at ~35 % efficiency to determine the percent recovery for each sample. The remaining solution was analyzed for PGE by a

APRIL

1976

VOL. 11 NO. 4

633

PROSTAGLANDINS

radioimmunoassay

(RIA) procedure.

RADIOIMMUNOASSAY The RIA was performed in 12 x 75 mm disposable glass tubes and is similar to other previously reported procedures which utilize a dextran coated charcoal separation step (6). The final antiserum dilution needed to bind 50 “/o of the tritiated PGE2 in the absence of unlabeled PGE2 was 1:1500. All samples were run in duplicate in at least 2 concentrations. Serial dilutions of the samples yielded curves parallel to the standard PGE curve. Each RIA consisted of a standard curve ranging from 70 to 600 pg PGE2, tissue samples and one or more internal PGE2 standards (a known amount of PGE2 spiked with tracer, chromatographed and otherwise treated identically to the tissue samples). This internal standard provided a check on the overall methodology. Periodically, column blanks tained by running the appropriate and assaying the fractions which tion. These blanks detected any terial picked up from the column

These blanks were obwere assayed. solvents through silicic acid columns correspond to the PGE containing porimpurities in the solvents and any mawhich would interfere with the RIA.

A quench correction curve was generated using the automatic external standard and all data were corrected to disintegrations per miThe RIA standard curve is expressed as the amount of added nute. PGE2 vs. B/B, x 100, where B is the amount of tritium bound in the presence of unlabeled PGE and B. is the amount of tritium bound in the Since this antiserum shows substantial absence of unlabeled PGE. cross reactivity with PGEl (5), all results are expressed as PGE. The overall accuracy of the methods used to measure PGE levels is reflected by the assay values of 97.8 f 5.9 % (Mean 3~SEM, n = 12) for the internal PGE2 standards which were processed identically to the tissue samples through silicic acid chromatography and RIA. The amount of radioactivity which eluted from the silicic acid column in the PGE fraction ranged from 65 to 90 ‘% of the tritiated PGE2 tracer Less than initially added to the tissue sample or internal standard. 1.5 70 of 3H PGA2 and less than 4.6 ‘% of 3H PGFZa added to selected This separation of PG’s into samples was eluted in the PGE fraction. classes by column chromatography minimized the problem of crossreactivity between the antiserum and PC’s other than PGE. Protein

634

was determined

by the method

of Lowry

APRIL

et al.,

1976

(7).

VOL. 11 NO. 4

PROSTAGLANDINS

RESULTS ERY THEMA

DEVELOPMENT

The kinetics of UVB-induced erythema development and the effects of a single topical IM treatment on that redness are seen in Figure 1. Immediately following UVB exposure, there was a transient redness which faded quickly to a barely perceptable level before the delayed phase of erythema began. By one hour following irradiation and treatment, the ability of topical IM to decrease UVB-induced redness was apparent and by 4 hrs., IM reduced the erythema to near unirradiated, normal skin levels. Twenty-four hrs. after UVB exposure, the effect of IM on redness remained evident, while at 48 hrs. the entire irradiated skin was red and there was no significant difference in the extent of erythema, whether or not the skin was treated with IM. Treatment of sunburned skin with the vehicle alone had no effect on the redness. When 1 pg of PGE8 was injected into normal skin and nonerythematous, IM-treated, sunburned skin at 4 hrs. after UVB, marked erythema which extended beyond the injection sites became evident within minutes and persisted approximately 1 hr. in each injected area. Injected saline produced only slight, transient redness which was limited to the site.

3.0

0

4

I

0.25 TIME

@0JRS

AFTER

UVB

24

48

EXPOSURE)

Figure 1. The effect of topical IMqq_UVB-induced erythema in the Skin exposed to UVB = u . Skin treated with IM immeguinea pig. diately after exposure to UVB =s.

APRIL

1976

VOL. 11 NO. 4

635

PROSTAGLANDINS

MEASUREMENT

OF PGE

The PGE level in normal guinea pig skin was determined to be 106.8 f 15.0 rig/g tissue or 690 f 69 rig/g protein (Mean f SEM, n = 14). This value probably reflects some -in vitro synthesis of PGE since lower values were found in skin which had been frozen immediately after excision. The trauma of excision also contributed to increased PGE levels. Normal skin was pre-treated with IM 4 hrs. prior to biopsy in order to inhibit trauma-induced PGE synthesis. The PGE level in this tissue was 20% lower than the value determined for corresponding skin untreated with IM. PGE levels determined in skin exposed to UVB 4 hrs. earlier are listed in Table I. The absolute values at the top of Table I show that UVB exposure caused marked elevation of PGE levels in the guinea pig skin. A single, topical treatment of the sunburned skin with 2.5% IM reduced PGE levels to that of the normal, unirradiated skin. Sunburned skin treated with the vehicle alone showed no decrease in PGE levels. Using a matched comparison test, the differences between UVB and UVB To minimize the + IM or control are significant at the p = 0.03 level. effect of animal variation and -in vitro PGE synthesis, the results from the data obtained at four hrs. after UVB exposure were calculated on a relative basis for each of the five animals. Expression of the data in ratios in the lower portion of Table I yields the same conclusions as the absolute values, but with much more acceptable standard deviations. All further data are reported as ratios to emphasize that the differences between the experimental and control tissues are more important that the absolute values.

636

APRIL

1976 VOL. 11 NO. 4

PROSTAGLANDINS

TABLE PGE LEVELS

I

IN SKIN FOUR HRS.

AFTER

UVB IRRADIATION

PGE Skin Content UVB

UVB + IM

Control

320.0 1055.6 360.2 180.0 1137.0

12 3.0 635.7 88.8 50.9

110.0 568.6 86.8 66.9 470.0

-GP 1 2 3 4 5 Mean * SD

610.6

224.4

* 449.4 PGE

-GP

UVB/ Control

1 2 3 4 5

(rig/g tissue)

SD

2.81 f .85

APRIL

1976 VOL. 11 NO. 4

260.5

f 239.3

Ratio

JUVB + IM)/Control

(UVB + IM)/UVB

1.11 1.11 1.02 0.76

2.91 1.86 4.15 2. 69 2. 42

Mean&

f 275.6

1.00

* .17

0. 38 0.60 0.25 0. 28

0.38

zt

. 16

637

PROSTAGLANDINS

PGE levels present in the skin during the 48 hr. period after UVB exposure are summarized in Figure 2. In the absence of topical medication, within minutes after UVB exposure, the PGE levels (striped bars) were elevated and continued to increase until they reached apBy 24 hrs. proximately three times the normal control value by 4 hrs. the PGE levels were declining and at 48 hrs., the PGE levels were essentially identical in sunburned and normal skin. When a single topical application of IM was applied immediately after UVB exposure, the PGE levels (clear bars) were significantly reduced by one hr. The PGE level in the skin remained near the control value (ratio = 1) for the entire 24 hr. period. The PGE levels appeared to parallel the erythema To (which is a reflection of the increased amount of blood in the skin). determine whether the PGE level was simply reflecting the amount of blood present in the tissue, blood was added to normal skin and the combination was assayed. No elevation in PGE level was obtained.

0

--c---L---l-

0

0.25 TIME

I

4

48

24

0

(HOURS AFTER UVB EXPOSURE)

Figure 2. The effects of a single topical application of IM on PGE levels induced in skin by UVB irradiation. The data are expressed as the ratio of PGE levels in control, unirradiated skin (C), UVB irradiated skin treated with vehicle (UVB) and UVB irradiated skin treated with IM (UVB + IM) The ratios are UVB/C =a; (UVB + IM)/C =m; (UVB + IM)/UVB = :$!:>sz. The bars show the SEM. The numbers in parenthesis indicate the number of animals used for each determination.

.

638

APRIL

1976

VOL. 11 NO. 4

PROSTAGLANDINS

A direct comparison between PGE ratios and UVB-induced erythema is seen in Figure 3. It is apparent that both redness and PGE levels increased in a parallel manner from 0 to 4 hrs. after irradiation. After that time, the redness (clear bars) remained high while the PGE ratio (shaded bars) began to decline. At 48 hrs. definite redness was still very evident, while the PGE ratio was near normal, approaching a value of 1.

4.0

30 0 5 L w :: 20

: 0

T,ME

4

I

0.25

(HOURS

AFTER

UVB

24

1.0

48

EXPOSURE)

Figure 3. A comparison of the development of erythema and relative PGE levels in guinea pig skin exposed to UVB. The clear bars indicate the average erythema grade in UVB irradiated skin using the scale on the left. Shaded bars indicate the ratio of PGE (rig/g protein) in UVB exposed skin/PGE (rig/g protein) in control, unirradiated skin as shown on the scale on the scale on the right.

APRIL

1976

VOL. 11 NO. 4

639

PROSTAGLANDINS

Table II compares the labeling index (L. I. ) for sunburned guinea pig skin treated with vehicle or IM to the PGE ratios determined under these same conditions. In the absence of topical medication, 4 hrs. after UVB, the L. I. in sunburned skin was depressed below the normal value of 0.05. By 48 hrs. after UVB, the L. I. had increased to approximately 6 times the normal value. These typical UVB-induced changes in L. I. were not influenced by application of vehicle to the sunburned skin. From Table II there appears to be an inverse relationship between the PGE ratio and the L. I. in sunburned skin. This would suggest that fluctuations in PGE level might affect DNA synthesis as measured by the incorporation of tritiated thymidine into basal cells. This possibility was not confirmed by the results obtained from sunburned skin treated with IM. In this case by 4 hrs. after UVB, the PGE level was essentially normal and yet the L. I. was depressed. In spite of the maintenance of normal PGE levels throughout the first 24 hr. period in sunburned skin treated with IM, the L. I. in that skin injust as it did in the untreated suncreased dramatically at 48 hrs., burned skin. TABLE

II

A COMPARISON OF PGE RATIOS AND LABELING INDICES IN SUNBURNED GUINEA PIG SKIN TREATED WITH INDOMETHACIN OR VEHICLE

UVB + Vehicle

UVB + IM

PGE Ratio Hrs. After UVB

“L.

640

L. I. +

UVB + Vehicle Control >

PGE Ratio L. I. t

UVB + IM Control

1

0.047

2.44

0.022

1. 25

4

0.015

2.99

0.013

1.10

48

0.334

1.17

0. 332

1.19

I. in normal

>

skin = 0.05.

APRIL

1976

VOL. 11 NO. 4

PROSTAGLANDINS

DISCUSSION A single topical application of IM applied immediately after UVB exposure was demonstrated to simultaneously decrease the PGE content and the UVB-induced redness of sunburned skin to near normal, unirradiated skin levels. The rapidity of the effect implies quick, percutaneous absorption of physiologically active amounts of IM. The decrease in redness caused by IM treatment does not appear to be the result of a decreased skin response to PGE since sunburned skin treated with IM is still capable of responding to injected PGE2 with persistent The decrease in PGE leerythema in a manner similar to normal skin. vel is most probably due to the ability of IM to inhibit the PC synthetase system rather than to some ability to accelerate the already rapid metabolic inactivation of PGE (8). The immediate, transient erythema which fades within 15 min. after cessation of UVB has been reported to be mediated by histamine (9). The second or delayed phase of erythema begins about one hr. after UVB exAnalysis of the lag time between UV irradiation and the appearposure. ance of erythema allowed Van der Leun (10) to show that the development of the delayed erythema is consistent with the formation of one or more mediators in the epidermis in response to UV exposure and then subsequent diffusion of the mediator(s) into the dermis to cause vasodilation. The kinetics of the delayed erythema appearance were not consistent with a direct effect of UV on the vasculature. This present study has shown that exposure of guinea pig skin to UVB causes elevation of PGE levels in the skin in a manner which parallels the development of the delayed phase of erythema and suggests that PGE may be the diffusible compound responsible for mediating or potentiating the first 24 hrs. of the delayed phase of UVB-induced erythema. Evidence already exists that UV irradiation can labilize epidermal lysosomal membranes (11) and that the enzymes capable of synthesizing PC’s are associated with the microsoma1 fraction of the epidermis (12). Beyond the first 24 hrs., PGE’s role in erythema mediation becomes questionable, since redness remains Earevident while the PGE content in the skin has returned to normal. lier work has shown that repeated applications of IM throughout the first 48 hrs. after irradiation were no more effective in reducing redness than was a single treatment immediately following UVB exposure (4). These additional applications should provide sufficient PC synthetase inhibitor to the skin to rule out the possibility that the redness apparent at 48 hrs. was due to new biosynthesis of PGE following the depletion of IM. Other explanations for the 48 hr. redness are currently under investigation. PGE does not appear to play a significant role in non-vascular UVBinduced skin responses. “Sunburn cells” present in UVB irradiated skin at 24 hrs. after exposure are considered to be degenerating, dying

APRIL

1976 VOL. 11 NO. 4

641

PROSTAGLANDINS

A previous study showed that IM treated sunburned keratinocytes (9). skin had the same number of “sunburn cells” as the untreated irradiated Since IM depressed PGE levels to near normal during those skin (4). 24 hrs., this suggests that the keratinocyte death occurs independently of PGE levels and is probably the result of damage due to absorption of UVB energy. A direct role for PGE in UVB-induced alterations in DNA synthesis is not obvious. Although an inverse relationship between DNA synthesis and PGE levels in sunburned skin exists (Table II), it was possible to reduce the PGE content to normal without any corresponding change in In untreated sunburned skin the PGE level was elethe labeling index. vated before the labeling index was decreased. This could suggest that synthesis of PGE resulted from UVB-induced disruption of the cell membrane and then the PGE was available to diffuse to the cell nucleus to influence subsequent DNA synthesis. In the presence of topical medication, an amount of PGE sufficient to trigger alterations in DNA synthesis may have been produced by UVB irradiation before effective levels of IM were absorbed. This could account for the observed lack of effect of IM on UVB-induced alterations in DNA synthesis as measured by -in vivo incorporation of tritiated thymidine. It is apparent that more work is required before the role of PGE as modulator of DNA synthesis is understood. REFERENCES 1.

2. 3. 4.

5.

6.

7.

8.

642

Sondergaard, J. and Greaves, M. W. Pharmacological Studies in Inflammation Due to Exposure to Ultraviolet Radiation. J. Pathol. 101:93-97, 1970. Mathur, G. P. and Gandhi, V. M. Prostaglandin in Human and Albino Rat Skin. J. Invest. Dermatol. 58:291-295, 1972. Snyder, D. S. and Eaglstein, W. H. Intradermal Anti-Prostaglandin Agents and Sunburn. J. Invest. Dermatol. 62:47-50, 1974. Snyder, D. S. Cutaneous Effects of Topical Indomethacin, an Inhibitor of Prostaglandin Synthesis, on UV Damaged Skin. J. Invest. Dermatol. 64:322-326, 1975. Goldyne, M. E., Winkelmann, R. K. and Ryan, R. J. Prostaglandin Activity in Human Cutaneous Inflammation: Detection by Radioimmunoassay. Prostaglandins 4:737-749, 1973. Jaffe, B. M., Behrman, H. R. and Parker, C. W. Radioimmunoassay Measurement of Prostaglandins E, A and F in Human Plasma. J. Clin. Invest. 52:398-405, 1973. Lowry, O.H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem. 193:265-275, 1951. znggard, E. andSamuelsson, B. Metabolism of Prostaglandin El in Guinea Pig Lung: The Structure of Two Metabolites. J. Biol. Chem. 239:4097-4102, 1964.

APRIL

1976 VOL. 11 NO. 4

PROSTAGLANDINS

9. Johnson,

Daniels, F. Jr. and Magnus, B. E., logy IV (A. C. Giese, Editor) Academic Press,

10. 11. 12.

I.A. in PhotophysioNew York, 1968,

pp. 139-202. Theory of Ultraviolet Erythema. Photochem. Van der Leun, J. C. Photobiol. 4:453-458, 1965. Ultraviolet Radiation and Lysosomes in Skin. Johnson, B.E. Nature 219:1258-1259. 1968. Ziboh, V. A. Biosynthesis of Prostaglandin E2 in Human Skin: Subcellular Localization and Inhibition by Unsaturated Fatty Acids and Anti-Inflammatory Drugs. J. Lipid Res. 14:377-384, 1973.

APRIL

1976 VOL. 11 NO. 4

643