MEDICAL UV EXPOSURES AND HIV ACTIVATION

This paper presents the first attempt to evaluate the potential of cli nical UV exposures to induce the human immunodeficiency (HIV) promoter and, thus, to upregulate HIV growth in those skin cells that are dire ctly affected by the exposure. Using the data for HIV promoter activat ion in vitro, we computed UVB and psoralen plus UVA (PUVA) doses that produce 50% of the maximal promoter activation (AD(50)). Then, using ( a) literature data for UV transmittance in the human skin, (b) a compo site action spectrum for HIV promoter and pyrimidine dimer induction b y UVB and (c) an action spectrum for DNA synthesis inhibition by PUVA, we estimated the distribution of medical UVB and PUVA doses in the sk in. This allowed us to estimate how deep into the skin the HIV-activat ing doses might penetrate in an initial and an advanced stage of UVB o r PUVA therapy. Such analysis was done for normal type II skin and for single exposures. The results allow us to predict where in the skin t he HIV promoter may be induced by selected small and large therapeutic UVB or PUVA doses. To accommodate changes in skin topography due to d isease and UV therapy, our considerations would require further refine ments. For UVB we found that, when the incident dose on the surface of the skin is 500 J/m(2) (290-320 nm) (initial stage of the therapy), t he dose producing 50% of the maximal HIV promoter activation (AD(50)(U VB)) is limited to the stratum corneum. However, with an incident dose of 5000 J/m(2) (an advanced stage of the therapy), AD(50)(UVB) may be delivered as far as the living cells of the epidermis and even to som e parts of the upper dermis. For PUVA we found that, when the incident UVA doses are 25 or 100 kJ/m(2) (320-400 nm) (an initial and an advan ced stage of therapy, respectively), and the 8-methoxypsoralen concent ration in the blood is 0.1 mu g/mL (the desired level), the combined d oses to the mid epidermis (and some areas of the upper dermis) are wel l below the 50% HIV promoter-activating PUVA dose (AD(50)(PUVA)). Only under the worst scenario conditions, i.e. an exceptionally high drug concentration in the patient's tissues and localization of HIV in the nearest proximity to the skin surface, would the combined PUVA dose ex pected during photochemotherapy exceed AD(50)(PUVA). These results sug gest that the probability of HIV activation in the epidermis by direct mechanisms is higher for UVB than for PUVA treatment. However, comple xities of the UV-inducible HIV activation and immunomodulatory phenome na are such that our results by themselves should not be taken as an i ndication that UVB therapy carries a higher risk than PUVA therapy whe n administered to HIV-infected patients.