Ultrastructural alteration of tape‐stripped normal human skin after photodynamic therapy

ARTICLE

Auteur(s) : Jacek BARTOSIK*, Ida-Marie STENDER, Takasi KOBAYASI, Magnus S. GREN¹

Department of Dermatology D92, University of Copenhagen, Bispebjerg Hospital, Copenhagen, Denmark 
* Department of Dermatology and Venereology, University of Lund, Lund, Sweden. 
¹ Department of Surgical Gastroenterology K, Bispebjerg Hospital, University of Copenhagen, Bispebjerg Bakke 23, DK-2400 Copenhagen NV, Denmark

Article accepted on 06/01/2004

Abbreviations: PDT, photodynamic therapy; KC, keratinocyte(s); LC, Langerhans cell(s); MC, melanocyte(s). 

Photodynamic therapy (PDT) is based on uptake of photosensitizer and subsequent light activation [1]. In this process highly reactive oxygen intermediates are generated [2, 3] and oxidize tissue and cell components leading to cell damage [4]. PDT offers an alternative regimen in the management of certain malignant and benign skin diseases [5, 6]. Preferential accumulation of photosensitizers in rapidly growing cells provides the mechanism for the selectivity in damaging skin tumours. The extent of cell injury depends on the type of PDT and on the type of the target cell [7]. It is believed that cell and mitochondrial membranes are the principal targets for PDT photo-oxidative processes in cells [8], but the mechanisms responsible for their destruction are still poorly understood. Little is also known about the extent of the injury brought about by PDT on various normal skin cells, when this therapy is applied i.e. for tumours. 
To learn more about this issue and thus to be able to optimize PDT in future, we investigated the reactivity of epidermal and dermal cells to standard PDT with 5-aminolevulinic acid (ALA) as photosensitizer in tape-stripped normal human skin using transmission electron microscopy.

Material and methods

Volunteers and skin tape stripping

Three healthy Caucasian volunteers (2 men and 1 woman), aged 40-45 years (median age 42 years), participated in the study. Two volunteers had skin type II and one skin type III according to Fitzpatrick.

On each volunteer, one skin area on the back, measuring 5 cm × 5 cm, was tape-stripped 10 times consecutively with an adhesive dressing (Tegaderm^®, 3M Health Care, St. Paul, MN, USA).

PDT treatment, biopsy sampling and control biopsies

Twenty-four hours after skin tape stripping, 20% ALA (5-aminolevulinic acid hydrochloride; Sigma-Aldrich, St. Louis, MO, USA), incorporated in an oil-in-water-based cream without parabenes, was applied topically under occlusion (Tegaderm^®) for 4 hours. Thereafter, the skin was irradiated for 30 minutes with a Waldman PDT 1200 lamp (570-650 nm) with a total dose of 70 J/cm².

Four-mm full-thickness punch biopsies were obtained under local anaesthetic (1% lidocaine containing epinephrine) immediately after irradiation (0 hour), and at 3 and 24 hours of the ALA-treated area. In addition, four control 4-mm biopsies were obtained immediately after the irradiation in the three volunteers from adjacent untreated skin (control I), tape-stripped skin (control II), tape-stripped skin treated with ALA under Tegaderm^® but non-irradiated (control III), and from tape-stripped, vehicle-treated under Tegaderm^® and irradiated (control IV).

Preparation and examination of biopsies by electron microscopy

The skin specimens were fixed in 6% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) at 4 C overnight, osmicated, dehydrated and then embedded in epoxy resin. Two to 3 series of 20 consecutive ultra-thin sections were cut from each specimen and were contrasted by uranyl acetate and lead citrate. The specimens were scrutinized in a Jeol 100 CX transmission electron microscope independently by two investigators who had no prior knowledge of group affiliation.

Results

Effect of tape stripping, ALA treatment and irradiation on skin morphology

The tape-stripped skin was essentially free from corneocytes (control II). As opposed to normal non-treated and non-irradiated skin (control I), in the tape-stripped skin specimens (control II) keratinocytes (KC) extending from basal to the spinous layers exhibited slightly rounded and swollen nuclei with little perinuclear condensation of chromatin (Fig. 1). ALA treatment alone (control III) did not alter any ultrastructural features of tape-stripped skin (pictures not shown). Tape-stripped, occluded and irradiated skin but not treated with ALA (control IV) was found to be similar with controls II and III, but some KC revealed enlarged intracytoplasmic vacuoles and electronlucent cytoplasm in specimens obtained immediately after irradiation (pictures not shown).

Effect of PDT on skin morphology

Keratinocytes (KC)

There was a wide reaction spectrum to PDT and the degree of the cell injury varied throughout the epidermis. Areas with only mild morphological alterations could be seen close to regions where the KC displayed severe cell damage.
Immediately after the light exposure (0 hour) the intracellular space enlarged in the lower epidermis. The KC nuclei became swollen and the perinuclear space strongly dilated. Bundles of tonofilaments were thin and dispersed, but the desmosomes were unchanged. Melanin granules were preserved. We observed some KC containing vacuoles of varying sizes, swollen and enlarged mitochondria, and extremely dilated cisternae of the endoplasmatic reticulum. Many of the KC also displayed a very electron-lucent cytoplasm (Fig. 2). An interesting observation was the relative normal appearance of basal KC containing large amounts of melanin (Fig. 2). Cell debris was often seen in the adjacent, intercellular space.
Three hours after the light exposure, there were fewer KC with extremely electronlucent cytoplasm. The intracytoplasmic vacuoles were seen less prominent and less frequent than immediately after light exposure (0 hour) (Fig. 3).
All these alterations disappeared after 24 hours in two of the individuals and were much less pronounced in the third individual. Swollen nuclei, as seen after tape stripping, remained in many KC in all the biopsies after 24 hours.

Langerhans cells (LC)

The majority of LC preserved normal cell structure and position in the epidermis. There were, however, a few LC that had prominent intracellular vacuoles (Fig. 4a). Also, some LC were located in the high level of the spinous cell layers. In 2 biopsies taken 3 hours after the irradiation, a few LC in the high level of the spinous cell layers displayed morphological alterations in the nucleus. The chromatin was condensed and the nucleus partitioned, while the cell organelles were still unaltered and the cell membrane preserved (Fig. 4b).

Melanocytes (MC)

Almost all the MC preserved their normal cell structure even in the most severely damaged epidermal areas. A few MC appeared to possess large intracellular vacuoles but at higher magnification, it could be clearly demonstrated that these structures were in fact extracellular (Fig. 5). A few MC showed slightly swollen mitochondria.
The ultrastructural changes observed in epidermal cells are summarised in table I.

Table I. Summary of ultrastructural changes of epidermal cells as a result of PDT

Dermis

No changes were observed at 0 hour. At 3 and 24 hours after the irradiation, perivascular inflammatory cell infiltrates and dilated lymph vessels were observed. Otherwise, there were no apparent ultrastructural alterations in the dermal cells.

Discussion

The knowledge about tumour response to PDT has been steadily growing during recent years [7], but few investigators have paid attention to the effects of PDT on normal skin. Normal epidermis and upper dermis with their well-established and firm morphology may provide a model for the study of action of PDT on skin cells and can offer a model for comparing various PDT techniques.
The biological effects of PDT vary widely depending on the types of the target cells, the photosensitizer and the light source [7]. In different tumour cells the initial sites of photodynamic response have been linked to various cell structures, such as cytoplasmic membrane systems, cell membrane, lysosomes, mitochondria and nuclei. Active oxygen, which is produced during PDT, can mediate oxidation of lipids and/or cross-link protein components of cell/organelle membranes. This probably leads to alterations in membrane permeability and to inactivation of membrane-associated enzymes. These alterations may result in cell lysis [9]. PDT triggers different morphological alterations, such as swelling and vacuolization in tumour and endothelial cells, damage to endoplasmatic reticulum, damage to the nuclear envelope and single-stranded DNA, as well as disruptions to the basal lamina of vascular endothelium [9, 10]. Apoptosis has also been reported after PDT in cultured human epidermal keratinocytes and in certain tumour cell lines [11-13].
In this study we applied standard PDT procedure with the commonly used photosensitizer, 5-aminolevulinic acid (ALA). Because the hydrophilic ALA penetrates poorly through intact epidermis [7], we used tape stripping to enhance penetration into the skin. Tape stripping of the normal skin appeared not to be completely harmless for the epidermal cells since it caused morphological changes in KC nuclei. Enlarged basal KC nuclei in response to tape stripping of normal human skin were also reported by Pinkus [14]. ALA is a precursor of porphyrins, mainly protoporphyrin IX in mitochondria [15] and maximal levels of protoporphyrin IX are reached 3-5 hours after a single topical ALA application in PDT [10]. To assure optimal effect of PDT in this study, the healthy individuals were treated with ALA for 4 hours before irradiation.
The present study demonstrated that the epidermal cells react differently to PDT. Varying severities of the ultrastructural alterations from area to area in the epidermis may explain this phenomenon. Such uneven distribution of reaction pattern throughout the epidermis has previously been described for chemical stimuli and has been referred to as “microfocussed reaction principle” [16]. Tape stripping possibly elicited mitosis to a varying degree that also contributed to the heterogeneous response to PDT in epidermis [14].
Despite signs of significant cellular damage, such as electronlucent cytoplasm, enlargement of perinuclear space, and cell debris in the intercellular space, the KC appeared to recover within less than 24 hours after PDT. This fact suggests that the type and doses of PDT used in the present study did not provoke any widespread irreversible damage to the KC.
The two dendritic cell types Langerhans cells (LC) and melanocytes (MC) appeared to react differently to PDT. We found nuclear morphological changes in some LC 3 hours after illumination of tape-stripped, ALA-treated skin. These changes resemble those of apoptotic LC after ultraviolet radiation described by Hollis et al. [17]. Because there were no morphological alterations in LC in irradiated, tape-stripped but non-ALA-treated skin, it seems unlikely that light exposure alone was sufficient to trigger apoptosis in the present study. The epidermal MC appeared to be unperturbed by PDT in this study. Thus, it can be speculated that PDT can be pro-apoptotic for certain epidermal cells. Our findings are supported by those of Monfrecola et al. [18] who found an opposite response of LC and MC to PDT with 20% ALA. The percentage of LC decreased from 6.7% to 3.8% while the percentage of MC increased from 21.3% to 42.8% in non-tape-stripped human skin 7 days after a single ALA-PDT treatment session [18].
Pigmented malignant melanomas have previously been found unresponsive to PDT [19]. However, by using other photosensitizers and other light sources, the progressive growth of some pigmented tumours can be reduced, though the tumour cannot be totally eradicated [20]. This is in contrast to non-pigmented tumours, which can be totally eradicated after PDT [10].
It has been shown that altered membrane potentials in a cell may change its sensitivity to PDT [21]. The lack of response of MC in our study may depend on either modified membrane potentials, altered metabolism of ALA or other, yet unknown, cell qualities in MC protecting them from injuries during PDT. Another possibility is that melanin granules were responsible for the nuclear protection as less nuclear changes were observed in epidermal keratinocytes rich in melanin. More in-depth knowledge on the protective mechanism(s) of melanocytes may assist us in the future to develop new PDT regimens for treatment of melanomas. n

References

1. Moan J, Berg K. Photochemotherapy of cancer: experimental research. Photochem Photobiol 1992; 55: 931-48. Review.

2. Kimel S, Tromberg BJ, Roberts WG, Berns MW. Singlet oxygen generation of porphyrins, chlorins, and phthalocyanines. Photochem Photobiol 1989; 50: 175-83.

3. Athar M, Mukhtar H, Bickers DR. Differential role of reactive oxygen intermediates in photofrin-I- and photofrin-II-mediated photoenhancement of lipid peroxidation in epidermal microsomal membranes. J Invest Dermatol 1988; 90: 652-7.

4. Kreimer-Birnbaum M. Modified porphyrins, chlorins, phthalocyanines, and purpurins: second-generation photosensitizers for photodynamic therapy. Semin Hematol 1989; 26: 157-73.

5. Itoh Y, Ninomiya Y, Tajima S, Ishibashi A. Photodynamic therapy of acne vulgaris with topical δ-aminolaevulinic acid and incoherent light in Japanese patients. Br J Dermatol 2001; 144: 575-9.

6. Stender IM, Na R, Fogh H, Gluud C, Wulf HC. Photodynamic therapy with 5-aminolaevulinic acid or placebo for recalcitrant foot and hand warts: randomised double-blind trial. Lancet 2000; 355: 963-6.

7. Peng Q, Warloe T, Berg K, Moan J, Kongshaug M, Giercksky KE, Nesland JM. 5-Aminolevulinic acid-based photodynamic therapy. Clinical research and future challenges. Cancer 1997; 79: 2282-308. Review.

8. Henderson BW, Dougherty TJ. How does photodynamic therapy work? Photochem Photobiol 1992; 55: 145-57. Review.

9. Peng Q, Moan J, Nesland JM. Correlation of subcellular and intratumoral photosensitizer localization with ultrastructural features after photodynamic therapy. Ultrastruct Pathol 1996; 20: 109-29.

10. Peng Q, Moan J, Warloe T, Nesland JM, Rimington C. Distribution and photosensitizing efficiency of porphyrins induced by application of exogenous 5-aminolevulinic acid in mice bearing mammary carcinoma. Int J Cancer 1992; 52: 433-43.

11. Zaidi SI, Oleinick NL, Zaim MT, Mukhtar H. Apoptosis during photodynamic therapy-induced ablation of RIF-1 tumors in C3H mice: electron microscopic, histopathologic and biochemical evidence. Photochem Photobiol 1993; 58: 771-6.

12. He XY, Sikes RA, Thomsen S, Chung LW, Jacques SL. Photodynamic therapy with photofrin II induces programmed cell death in carcinoma cell lines. Photochem Photobiol 1994; 59: 468-73.

13. Ketabchi A, MacRobert A, Speight PM, Bennett JH. Induction of apoptotic cell death by photodynamic therapy in human keratinocytes. Arch Oral Biol 1998; 43: 143-9.

14. Pinkus H. Examination of the epidermis by the strip method. II. Biometric data of regeneration of the human epidermis. J Invest Dermatol 1952; 19: 431-47.

15. Rimington C. Haem biosynthesis and porphyrias: 50 years in retrospect. J Clin Chem Clin Biochem 1989; 27: 473-86. Review.

16. Mikulowska A, Bartosik J, Andersson A, Björnberg A, Falck B. Microfocussed distribution of reactive events in the epidermal Langerhans cell system. In: Thivolet J, Schmitt D, eds. The Langerhans Cell. Colloque INSERM/John Libbey Eurotext Ltd., 1988; 172: 285-92.

17. Hollis DE, Scheibner A. Ultrastructural changes in epidermal Langerhans cells and melanocytes in response to ultraviolet irradiation, in Australians of Aboriginal and Celtic descent. Br J Dermatol 1988; 119: 21-31.

18. Monfrecola G, Procaccini EM, D’Onofrio D, Roberti G, Liuzzi R, Staibano S, Manco A, De Rosa G, Santoianni P. Hyperpigmentation induced by topical 5-aminolaevulinic acid plus visible light. J Photochem Photobiol B 2002; 68: 147-55.

19. Pass HI. Photodynamic therapy in oncology: mechanisms and clinical use. J Natl Cancer Inst 1993; 85: 443-56. Review.

20. Woodburn KW, Fan Q, Kessel D, Luo Y, Young SW. Photodynamic therapy of B16F10 murine melanoma with lutetium texaphyrin. J Invest Dermatol 1998; 110: 746-51.

21. Moorehead RA, Armstrong SG, Wilson BC, Singh G. Cross-resistance to cisplatin in cells resistant to photofrin-mediated photodynamic therapy. Cancer Res 1994; 54: 2556-9.