SEDAC Socioeconomic Data and Applications Center
Environmental Effects of Ozone Depletion 1998 Assessment


Hazards for Humans

Humans have three major organ systems whose cells and tissues are commonly exposed to sunlight: the eye, the immune system and the skin, and it is in these three systems that the effects of sunlight on health have been documented. The cells/tissues exposed in the eye are principally those associated with the cornea; the iris and the lens, those of the skin include the outermost layer of the skin, the stratum corneum, and the epidermis; and those of the immune system are the Langerhans (or antigen-presenting) cells that reside in, or migrate through the epidermis.

    Each of the different types of UV-exposed tissues contains a collection of chemical substances whose light-absorbing properties can contribute to the process shown in Fig. 2.1. Furthermore, the organ systems are structured such that some tissues/cells will absorb part of the UV energy before it reaches others. Thus the spectrum of light which first hits the surface of an organ such as the skin, is not the same as that reaching tissues/cells located deeper, e.g., in the basal layer of the epidermis. As a result, the wavelength dependence, or action spectrum, of a particular end-point of concern rarely looks exactly like the absorption spectrum of a particular chromophore. Figure 2.3 shows action spectra for several of the more important effects, sunburn, DNA damage (dimers) and carcinogenesis, that will be discussed in detail below. Note again that only absorbance above 290 is relevant to environmental exposures.

Fig. 2.3. Action spectra of key UV-associated health effects.

Effects on the Eye

As indicated above, the eyes are a principal route of exposure to UVR. As illustrated in Fig. 2.4, when sunlight (and the UVR it contains) impinges on the normal eye, the cornea is encountered first, then the lens, the vitreous humor, and the retina. Studies indicate that due to its absorption by various molecules in the cornea and the lens, most UVR never reaches the retina in the normal adult eye. In the case of ambient UVR (i.e., UV-B and UV-A), the shorter wavelengths are absorbed preferentially, with the cornea absorbing most of the radiation below 300 nm, and the lens absorbing almost all of the rest of the UVR below about 370 nm (Merriam, 1996). Lens removal (as for the treatment of cataract) does place the retina at risk for UV damage; it is for this reason that many artificial replacement lenses are made with UV-absorbing materials.

Fig. 2.4 Absorption of UVR by the eye

Effects on the Cornea/Conjunctiva

The ocular effect most directly attributable to environmental exposures to UVR is photokeratitis. The ocular equivalent of sunburn, this effect occurs after an acute, i.e., short term, exposure and is characterized by reddening (inflammation) of the eyeball, gritty feeling of severe pain, tearing, photophobia (avoidance of light) and blepharospasm (twitching). Frequently diagnosed in skiers as "snowblindness", photokeratitis is also seen in beach-goers and others involved in outdoor recreation. (McCarty and Taylor, 1996). Mechanistic studies have revealed that human corneal stromal cells when exposed to low doses (10-100 mJ/cm2) either in vitro or in situ show significant increases in the production of a number of biologically-active chemicals, i.e., cytokines. The cytokines detected, interleukin (IL)-1, IL-6, IL-8 and tumor necrosis factor alpha (TNFa ), are proinflammatory and may be responsible for the inflammation which accompanies photokeratitis (Kennedy et al., 1997).

    Additional ocular effects on the cornea/conjunctiva attributed to solar exposure are climatic droplet keratopathy (CDK), pinguecula, pterygium, and squamous cell carcinoma (SCC) of the cornea and conjunctiva. CDK is a degeneration of the fibrous layer of the cornea with the accumulation of droplet shaped deposits. Pterygium results from an outgrowth of the conjunctiva (outermost mucous layer) over the cornea, which results in the loss of transparency, and pinguecula is a raised opaque mass just adjacent to the cornea (Hollows, 1989) and SCC is a malignant neoplasm similar to those found on sun-exposed skin. Data supporting the relationship between solar exposure and disease are strongest with CDK and pterygium; epidemiologic studies indicate that chronic exposure to the sun, and, most probably UV-B, is an important factor in development of these diseases. Both are associated with outdoor living or working in environments with high surface reflectance, e.g., water, sand, concrete (Hollows, 1989; Taylor et al., 1989; Mackenzie et al., 1992). Interestingly, in a recent study of pterygium, a much-increased risk (36-fold) was found in people with early, intense exposures (residing at 30 degrees south latitude or less for the first 5 years of life). This was independent of the almost 40-fold increase in risk associated with a work environment below 30 degrees south between the ages of 20-29 (Mackenzie et al., 1992). Data linking pinguecula to solar exposure are largely anecdotal or based on case reports (Bergmanson and Sheldon, 1997), although Taylor and his colleagues (1989) found a weak association between pinguecula and UVR exposure in the Maryland Watermen study.

    A recent study of SCC of the eye (Newton et al., 1996) examined incidence data from across the world in order to assess whether solar UVR is a risk factor for this disease. The study focused on conjunctival and corneal lesions, excluding those on the eyelid, and developed estimated daily UV-B exposures weighted using the erythemal action spectrum. Because HIV infection increases the risk of conjunctival SCC, and two African centers have seen substantial increases in this tumor in the past 5-10 years, the analysis looked at two data sets, one including data from Africa and one excluding these data. With the African data, the study found an increase in incidence of SCC of the eye with UV-B exposure that is equivalent to almost a 50% increase in incidence for every 10° decrease in latitude. Without the African data, an equivalent 40% decline was found for each 10° increase in latitude.

Effects on the Uveal Tract

The uveal tract consists of the iris, ciliary body and choroid. Malignant melanoma of the uveal tract is the most commonly occurring primary ocular malignancy. Rare in Blacks, in white patients it most commonly occurs in the choroid (Berkow, 1992), but also occurs in the iris. Several epidemiologic studies (Holly et al., 1996), found an increased risk of intraocular melanomas associated with sensitivity to UV. In these studies, sensitivity to sunburn was associated with almost a two-fold increase in risk of intense UV exposures, e.g., prior history of a welding burn or snow blindness was associated with more than a seven-fold increase in risk (6et al., 1990). Others have found the evidence less than compelling (Dolin and Johnson, 1994; Schwartz et al., 1997). The latter analysis, however, examined only the hypothesis of a direct effect of UV on the affected cells (i.e., a DNA-damaging effect) and did not evaluate the role of indirect effects that might contribute to carcinogenesis such as immunosuppression associated with increased production of cytokines (Kennedy et al., 1997). That such processes can occur following UV exposure of the eyes is suggested by the recent finding that in mice, high ocular doses of UV-B can produce systemic immunosuppression equivalent to that obtained by skin irradiation. Furthermore, severing the optic nerve prevented this immunosuppression (Hiramoto et al., 1997).

Effects on the Lens

Of all of the ocular diseases associated with solar exposure, that which affects the lens, cataract, is by far the most important from a public health perspective. Characterized by a gradual loss in the transparency of the lens (due to the accumulation of oxidized lens proteins) (McCarty and Taylor, 1996), the end-result is frequently blindness, unless the affected lens is surgically removed.

    Several different kinds of cataract are distinguished based on their location in the lens. Cortical cataracts develop in the outer layers of lens protein, commonly called the cortex of the lens. Nuclear cataracts occur in the inner layers of lens protein, i.e., the nucleus of the lens. And, posterior subcapsular cataracts (PSC) occur at the back (posterior) interface of the lens and its epithelial capsule. A fourth form of cataract is mixed, i.e., combining elements of two or more of the aforementioned forms. In a recent Italian case-control study of individuals aged 45-79 years old, pure cortical cataract accounted for slightly less than 50% of cases, pure nuclear cataract accounted for about 10%, pure PSC for less than 3%, and mixed for about 40% with the majority of the mixed having a cortical component (Italian American Cataract Study Group, 1991; 1994).

The epidemiologic evidence identifying exposure to UV-B as a risk factor for cataract suggests that the risk may be limited to pure or mixed cortical cataracts and PSC with the exception of the nuclear/cortical combination (Italian American Cataract Study Group, 1994). In the case of cortical cataract, a number of studies have indicated that the relative risk associated with increased sun exposure is between about 1- and 3fold (Italian American Cataract Study Group, 1991; Klein et al., 1995; West and Valmadrid, 1995, West et al. 1998). In the Beaver Dam Eye Study, this places heavy sun exposure about on a par with diabetes or heavy drinking as a risk factor (Klein et al., 1995).

The economic and social importance of cataract is enormous. It is be leading cause of blindness in the world(West and Valmadrid, 1995), with public health care costs for cataract surgery in the U.S. exceeding $3 billion in 1992. With the prevalence of cataract after age 30 approximately doubling each decade, anything that accelerates onset by 10 years (e.g., the increase in UV achieved in moving from the northernmost to the southernmost regions of the US) would double the number of operations (Javitt and Taylor, 1995).

Effects on the Immune System

In humans, the skin is the principal barrier to the outside world, and thus the first line of defense against foreign agents that may threaten health. In order to fulfill this role, the skin hosts a number of cells from the immune system that can mount or modify immune responses against such 'foreign invaders' or against skin cells that have become 'strange', e.g., by virus infection or transformation into a cancer cell. However, to function optimally, the immune system needs to be able to discriminate between 'self' and ‘strange’ or 'non-self', and eliminate only the latter, especially if it is (potentially) harmful.


As mentioned above, the skin contains a wide range of molecules, including both proteins and DNA, which undergo photochemical reactions upon absorbing UVR. It is quite likely that a great many of the cell-surface proteins which are used to determine 'self' are modified in such photochemical reactions so that at certain UV doses, the skin becomes swamped with ‘non-self’ cells. Were the immune system to react to all of these cells, the resulting inflammatory response might compromise other important skin functions. For this reason, it is believed that the decreased immune responses observed after UV irradiation serve to prevent excessive inflammation and damage to the skin that has been exposed to the sun. The drawback of this postulated beneficial physiological response is that it may be detrimental when it coincides with the entry of an infectious agent, or the development of a cancer cell, against which a forceful immune reaction needs to be mounted. These immunosuppressive effects of UV exposure can thus result in adverse circumstances: i.e., implants of UV-induced tumors between genetically identical mice, are rejected in a naive, unexposed host, but they fail to be rejected in a UV exposed host (Fisher and Kripke, 1977). Such a UV-induced immune suppression has also been found for contact hypersensitivity (CH) reactions (a type of immune reaction seen following skin contact with certain reactive chemicals, e.g., poison ivy) (Kripke, 1984; Yoshikawa, et al., 1990; Cooper et al., 1992). It also is seen in delayed type hypersensitivity (DTH) reactions (the kind of immune response made against virus-infected cells and certain microorganisms) (Howie et al., 1986). Figure 2.5 shows the two phases of the contact allergy response. UV irradiation can decrease both the induction of new responses through immunization and the elicitation of established immunity. Furthermore, immune suppression can occur locally, within UV-irradiated skin, or systemically at distant sites, depending on the dose of UV and the type of immune response.

Fig. 2.5 Multiple phases of contact hypersensitivity responses.

Clearly, the switch from an immune reaction to UV-induced suppression of it needs to be well tuned; at one extreme, too much suppression could render an individual susceptible to infections, whereas at the other extreme, too little could result in skin-damaging inflammatory reactions upon UV exposure. The latter would resemble what has commonly been referred as an allergic reaction to sunlight (a 'sun allergy'), that a physician would diagnose as a photodermatosis, e.g. UV-B-induced polymorphic light eruption, PLE. This disease can often be treated successfully by subjecting the skin to a series of gradually increasing UV-B irradiations, which are thought to permit the skin to adapt slowly to the effects of UVB (van der Leun and van Weelden, 1986). Thus, PLE patients appear to suffer from a compromised adaptation response; their immune response can adapt to small changes in UV-B, but is overwhelmed by large ones. This could explain why PLE is more common toward the poles (van der Leun and De Gruijl, 1993), where the seasonal UV modulation is greatest. If, as projected, large decreases in ozone occur during the wintertime at higher latitudes, one would expect to see a decrease in the seasonal UV modulation at these locations and a lowered incidence of PLE.


Immunological reactions tend to be rather complex because they involve multiple simultaneous processes that can act in concert or in opposition to one another. The impact of UV irradiation appears principally to be on cellular immune responses which are mediated through direct cell contact, and usually do not affect humoral immunity that is mediated through blood-borne proteins, e.g., so-called 'antibodies'. However, within the cellular immune response, there are multiple reactive sub-pathways that are affected differently by UV radiation. Current research on UV immunosuppression has had to recognize and account for these differences in order to understand the many, some of them seemingly contradictory, results. It is beyond the scope of this chapter to deal with this matter in any great detail. However, some of this information, in particular, the relevant action spectra, the role of antigen presenting cells and the genetic factors that can modify these responses, are necessary to the development of risk management strategies and so is briefly summarized here.

    A critical first step in understanding UV-induced immunosuppression is knowledge of the important chromophore(s). Unfortunately, recent research has increased rather than decreased the list of possibilities. Initially, there appeared to be two major chromophores of interest: urocanic acid (UCA; DeFabo and Noonan 1983), and DNA (Kripke et al., 1992), each playing a distinct role in the immune response. Finding from a number of different groups now contribute to the conclusion that both UCA and DNA are important to UV-induced systemic immunosuppression, and that under ambient exposures involving both UV-A and UV-B, a number of interacting events probably contribute to the final outcome. (For recent summaries see Streilein 1996 and Strickland and Kripke 1997). In addition, interest is now focused on chromophores that (perhaps through oxidation) alter cell membrane components (affecting internal cell signal transduction pathways (Devary et al., 1993), as well as on provitamin D3 that, through its active metabolite [1,25-dihydroxivitamin D3 (Müller and Bendtzen, 1996)] may become immunosuppressive.

    Much of the interest in the roles of DNA damage and UCA isomerization in immune suppression relates to the fact that these events affect the production or expression of a variety of biologically active chemicals that can modify immune reactions. Some are released into circulation, i.e., cytokines, while others are displayed on the cell surface, e.g., cell-surface receptors; however, they are key candidates for the development of interventions.

    The most prominent cellular target involved in the immunosuppressive action of UVR appears to be the Langerhans or antigen-presenting cell (APC). Large numbers of APCs reside in the epidermis and act as the skin’s security force, catching and processing foreign intruders, e.g., antigens, microorganisms, then migrating to the draining lymph nodes to activate the T lymphocytes that will mount the final immune response. Like cis UCA, UV irradiation diminishes the number of Langerhans in the epidermis, and disturbs the proper priming of T cells, often leading to the generation of suppressor T cells that can specifically block the development of an effective immune response against the invading agent. These suppressive cells induce a lasting tolerance toward the invading agent (i.e. the immune system is rendered 'blind' for this specific agent). Recent research suggests that for these aspects of UV-induced immunosuppression: two cytokines, TNF-a and IL10 are responsible for the induction of the transient and persistent tolerances, respectively (Niizeki and Streilein, 1997)

    Finally there is good evidence that susceptibility to UV-induced immunosuppression is under some degree of genetic control. Animal experiments initially demonstrated certain mouse strains to be resistant to UV-induced immunosuppression(Streilein and Bergstresser, 1988; Yoshikawa and Streilein, 1990); subsequent work indicated that this distinction was based on relatively high doses of antigen, with lower doses all animals became susceptible (Yamawaki et al., 1997). Humans also show differences in sensitivity to local suppression of the CH response (Cooper et al., 1992) which are largely independent of skin pigmentation. In fact, pigmentation provides surprisingly little protection against UV-induced immune suppression (Screibner et al., 1987) These findings suggest that vaccination programs carried out under conditions of high UV may want to evaluate their dosage regimens carefully in order to avoid providing doses that induce tolerance instead of immunity.

Infectious Diseases

Because of the experimental evidence that UV affected cellular immunity, concern arose about the implications of UV-induced immunosuppression for infectious diseases. Cellular immune responses are of paramount importance in the defense against a wide variety of infections. The 1903 Nobel laureate Niels Finsen first reported that UV radiation could heal skin tuberculosis (Lupus vulgaris) but adversely affected smallpox (Finsen, 1901), and others later found that lung tuberculosis was also adversely affected. Much more recent work using animal models of human diseases has confirmed that UV-B can affect different infectious diseases (and even differing manifestations of the same diseases) differently, as well as indicated that certain diseases, e.g., schistosomiasis (Noonan and Lewis, 1995), are unaffected.

    Human infectious diseases that in animal models have shown an effect of UV-B include herpes, tuberculosis, leprosy, trichinella, candidiasis, leishmaniasis, listeriosis, and Lyme disease. Reported effects have included suppression of immune responses to the organisms or their antigens, reactivation of latent infections, increased body burdens of organisms, decreased resistance to re-infection and reduced survival (Howie et al., 1986; Giannini and DeFabo, 1987; Denkins et al., 1989; Jeevan and Kripke, 1993; Goettsch et al., 1994).

    UV-B has been shown to activate viruses such as herpes, HIV and HPV (Spruance, 1985; Perna et al,. 1987; Zmudzka and Beer, 1990) as well as affect the immune response to herpes (Norval and el-Ghorr, 1996). In animal models of human tuberculosis, leprosy, listeriosis, trichinosis and Lyme disease, UV-B treatments suppressed DTH responses to the organisms, depressed clearance of the organisms, and in certain instances, increased mortality (Jeevan and Kripke, 1993, Goettsch et al., 1994; Goettsch et al., 1996).

    The information summarized above has raised concerns that UV-induced immune suppression could adversely affect the course of some infectious diseases in human populations. However, with the exception of HIV infection, there appear to be no recent published studies that have explored this issue through epidemiologic analysis. In the case of HIV infection, a recent report from the Multicenter AIDS Cohort Study (Saah et al. 1997) found no evidence that solar UV exposure exacerbated HIV infections in white homosexual males. Indeed, it was found that in men who were HIV positive at baseline, those who purposely sought sun exposure were less likely to have progressed to AIDS. As these authors noted, however, these findings need to be confirmed in a large, prospective study of HIV-infected individuals before any conclusions are made as to the beneficial effects of solar exposure on AIDS progression (Saah et al. 1997).

Lupus and other Autoimmune Diseases

The impact of UV-B exposures on APCs, and on the production and release of cytokines, opens up the possibility that increases in UV-B could either exacerbate or ameliorate autoimmune diseases. Data exist in support of both possibilities and seem to suggest that diseases likely to be exacerbated will be those involving aberrant humoral immunity, e.g., systemic lupus erythematosus (SLE), whereas those likely to be ameliorated will be those involving aberrant cellular immunity, e.g., multiple sclerosis (MS) and psoriasis. In the case of SLE it has long been known that UV-B exposures can exacerbate certain symptoms, probably in part due to the production of TNFa (McGrath et al., 1994).

    More recent studies of SLE, however, also suggest that exposures to UV-B could not only exacerbate SLE, but might even serve as initiating events. First, studies of SLE have demonstrated a spectrum of APC and T-cell defects consistent with loss of the cytokines that help regulate cell mediated responses (Lucey et al., 1996) — exactly what is observed following UV exposures (Ullrich, 1995; Boonstra and Savelkoul, 1997). The loss of cytokines also removes a regulatory constraint on the cells important to humoral immunity, thereby leading to an increase in the production of antibodies of the type implicated in the pathogenesis of SLE (Ullrich, 1995 ). Second, although the SLE antigens are mostly derived from internal components of the cell (e.g., the nucleus and the cytoplasm), the autoimmune response that is made to them is of the type normally reserved for components found outside the cell or on the cell membrane. This paradoxical finding may be explained by a recent observation that several of the nuclear and cytoplasmic autoantigens important to SLE pathogenesis are found on the surfaces of dying (apoptotic) skin cells, but not normal cells. If this means that these antigens are now presented to APC as if they were membrane components (Koh and Levine, 1997), this could explain the anomalous immune response seen in SLE. Since induction of apoptosis in keratinocytes is a consequence of UV exposure (Aragane et al.,1998), increases in UV could provide greater opportunities for encouraging an SLE type immune response.

    In the case of MS, a degenerative disease of the nervous system with autoimmune characteristics, it has long been recognized that there is a latitude gradient for the disease, with incidence increasing with increasing distance from the equator (exactly the opposite of what is found for skin cancer.) More than a decade ago, it was proposed that this could be explained as a result of the immunosuppressive effect of sunlight, and cross-reactivity between melanocyte antigens and those of the nervous system important to MS development (Sharpe, 1986). Subsequent work has strengthened the hypothesis that a sunlight-mediated immunosuppression may explain the latitude gradient, although it is thought now that the important antigen(s) are viral proteins that are cross-reactive with myelin basic protein (McMichael and Hall, 1997).


CIESIN Copyright © 1997 For more information about CIESIN and our activities contact CIESIN User Services. E-mail: Tel.: (914) 365-8988