Cario-Andre等人（2004）报道了皮损周围角质形成细胞色素沉积，表明了色素沉积通常出现在初期阶段。EM研究显示，与雀斑不同的是，在雀斑样痣和黑素细胞的相邻部位中黑素小体的大小是正常的。另外，与相邻部位的黑素细胞相比，雀斑样痣中的黑素细胞具有更多的线粒体和更发达的内质网（Nakagawa等人，1984；Noblesse等人，2006）。雀斑样痣的基底角质形成细胞含黑素小体复合物（聚黑素小体），在细胞核形成大量的色素递质（Cario- Andre等人，2004；Noblesse等人，2006），这表明能将黑色素小体从黑素细胞高效产生并转运到相邻的角质形成细胞。EM研究还表明了，与正常皮肤相比，雀斑样痣的基底膜存在差异。雀斑样痣中，真皮-表皮连接处结构紊乱、分裂，并且致密板与正常相比更薄（Noblesse等人，2006）。此外，还观察到基底膜轻微内陷到角质形成细胞，并且与正常相比，黑色素细胞陷入真皮更深，形成‘黑素细胞沉积床’（Cario- Andre 等人，2004；Nakagawa等人，1984；Noblesse等人， 2006）。Andersen等人（1997）对51例面部雀斑样痣病例进行了研究，并根据形态学将其分为三类。第一类表现为扁平表皮，第二类表现为增生性，少数第三类表现出以上两种临床特征。取自手臂的雀斑样痣并没有观察到这种临床特征（Hodgson，1963；Rhodes等人，1983），这表明其具有部位特异性差异。Cario-Andre等人（200
Table 2. Genes involved in freckling
雀斑与日光暴晒的关系同样复杂，其皮损通常到2-3岁才出现。雀斑对阳光有反应，通过紫外线照射非曝光部位皮肤，可诱发雀斑或使隐藏的雀斑显现（Wilson 和Kligman，1982）（图 1A, C）。 Ezzedine的研究显示，面部雀斑与频繁晒伤有关。有趣的是，Garbe等人（1994）的研究则表明，日光性黑子与青春期雀斑发展趋势有关，认为雀斑患者在晚年时期更易患SL。然而，此结果既不与Bastiaens（1999）和Monestier等人（2006）的研究一致，也不符合雀斑更常见于皮肤白皙的个体的观察结果；相反，SL更常见于深色皮肤。在这些研究中，雀斑与SL的定义可能并不总是明确。
雀斑大多数与浅色皮肤、金发/红发相关，常出现于儿童早期，表明青少年雀斑的形成主要由遗传决定。Bataille等人（2000）认为附加的遗传效应可以解释91%雀斑的不同类型，与环境因素无关。与此一致的是，已证实许多基因对雀斑的形成很重要，包括MC1R、 IRF4、ASIP、 TYR和BNC2（表 2）。大多数红发个体是纯合子或黑皮质素1受体（MC1R）基因突变的复合杂合子（Box 等，1997；Flanagan等，2000；Smith等，1998； Valverde等，1995）。MC1R变异也可导致皮肤苍白。更重要的是，已证实MC1R基因是雀斑形成的主要因素（Bastiaens等，2001； Flanagan等，2000；Rees, 2004）。总的来说，这被称为RHC（红头发，白皙的皮肤，缺乏晒黑能力和雀斑倾向）表型，MC1R基因的高渗透突变称为R-等位基因的和低渗透突变则称为r-等位基因（Beaumont等，2011）。MC1R是7个跨膜域的G蛋白偶联受体，位于黑素细胞的细胞膜（Chhajlani和 Wikberg，1992；Mountjoy等，1992）。它将不同配体与相反特性结合。配体a-黑素细胞促进激素（AMSH）和促肾上腺皮质激素（ACTH）通过受体激活信号，导致环状AMP（cAMP）增加，继而导致cAMP反应元件结合蛋白（CREB）的磷酸化，从而激活MITF，黑色素细胞主要调控元件。多个基因的MITF激活表达需要产生黑色真黑素，包括TYR，催化合成黑色素的酶（Busca和Ballotti，2000）。刺鼠信号蛋白（ASIP）是MC1R的拮抗剂，防止AMSH结合，从而产生真黑素。进而产生红色或黄色褐黑素，调节黑素细胞所产生的真黑素和褐黑素的比率（Cone等，1996；Rees，2003）。在欧洲血统的人群中，已报道超过80个的MC1R突变编码（Gerstenblith等，2007；Makova 和Norton, 2005）。这些突变造成色素性特征，如人类和动物的白皙皮肤，红色和金色头发，对紫外线的敏感度和雀斑（Kijas等，1998；Robbins等，1993；Rouzaud等，2006；Vage等，1999）。MC1R突变影响受体功能（Beaumont等，2007， 2009；Garcia-Borron等，2005）或启动子区域（Motokawa等，2008）；在没有MC1R功能的个体中发现红发和雀斑（Beaumont等，2008）。激活MC1R突变可以使蛋白质和深色素永久激活，至今仅有动物的研究报道（例如，小鼠，猪和羊），人类尚未涉及。
Cario-Andre et al. (2004) reported pigment accumulation in keratinocytes in perilesional skin suggesting that pigment accumulation happens early in the process. EM studies showed that, unlike in ephelides, melanosomes are of normal size in the lentigines as well as in melanocytes of adjacent regions. On the other hand, melanocytes in lentigines have more mitochondria and a better-developed endoplasmic reticulum than melanocytes from adjacent regions (Nakagawa et al., 1984; Noblesse et al., 2006). Basal keratinocytes of lentigines contain melanosome complexes (polymelanosomes) that form massive pigmented caps over the nuclei (Cario- Andre et al., 2004; Noblesse et al., 2006), suggesting efficient production and transport of melanosomes from melanocytes to neighboring keratinocytes. The EM studies also showed differences in the basement membrane of lentigines as compared to normal skin. In the lentigines, the dermal-epidermal junction was disorganized and disrupted, and the lamina densa was thinner than normal (Noblesse et al., 2006). Furthermore, micro-invaginations into the keratinocytes were observed, and the melanocytes seemed to drop further into the dermis than normally, forming ‘pendulum melanocytes’ (Cario- Andre et al., 2004; Nakagawa et al., 1984; Noblesse et al., 2006). Andersen et al. (1997) studied 51 facial lentigines and showed that they form three classes based on morphology. The first class showed a flattened epidermis, the second exhibited hyperplasia and the third and smallest group showed characteristics of both. This variation was not observed in lentigines taken from the arm (Hodgson, 1963; Rhodes et al., 1983), suggesting that there are region-specific differences. Cario-Andre et al. (2004) have suggested, based on association with age of their subjects, that the three classes represent lesional progression.
Unfortunately, careful side-by-side comparisons of the histopathology of ephelides and lentigines, including molecular markers, are not available. It is clear that ephelides and lentigines share important characteristics, including increased melanin production and extended epidermal ridges (Table 2 and Figure 2). Differences include the increased size of the melanosomes in ephelides and the effects observed on mitochondria, endoplasmic reticulum and basement membrane in lentigines, which have not been reported in ephelides. This suggests that these pigmentation traits have different origins.
To characterize the epidemiology of ephelides and lentigines, Bastiaens et al. (2004) studied 962 individuals from the Leiden Skin Cancer Study. They showed that SLs on the face were associated with increased age, whereas ephelides were inversely associated with age. Pigment lesions such as skin cancer and nevi usually develop more easily in light-skinned people. Ephelides are also associated with skin type I or II and blond/red hair color (Bastiaens et al., 2004; Ezzedine et al., 2013). Lentigines, however, seem to be more associated with darker skin types. Studying 118 cases with SL on the face, and equal number of controls, all between 60 and 80 yr of age, Monestier et al. (2006) showed that having multiple senile lentigos on the face (what they called lentigo aging pattern) was associated with skin types III and IV. They speculated that this might be due to a more active melanocyte system in those skin types. A similar association with darker skin was observed by Ezzedine et al. (2013). Multiple studies suggest that the formation of SLs is linked to sun exposure in one form or another and photodamage to the skin. In the Monestier study, SLs were associated with frequent sunburns and with recreational sun exposure, but not with occupational or lifetime sun exposure. In the study of Bastiaens et al. (2004), however, SLs on the face were associated with cumulative lifetime sun exposure and not with history of sunburns, whereas SLs on the back were associated with cumulative sun exposure and with history and number of sunburns before the age of 20. Ezzedine et al. (2013) studied 523 French middle-aged women and found that SLs were associated with lifetime sun exposure but not with sunburn during either childhood or adulthood. Derancourt et al. (2007) studied 145 cases and controls and found that SLs were associated with sunburn during adolescence and this was dose-dependent. These studies also suggest that there might be region-specific differences in the etiology of SL formation. No relationship was found between the formation of SLs and ephelides, suggesting that these pigment spots are etiologically unrelated (Bastiaens et al., 1999; Monestier et al., 2006).
The relationship of ephelides to sun exposure is also complicated. Freckles usually do not appear until the age of 2–3 yr of age. Ephelides respond to sunlight and can be induced, or more likely made visible in those that already carry them, by UV-irradiating non-exposed skin (Wilson and Kligman, 1982) (Figure 1A, C). The Ezzedine study showed that facial ephelides were associated with frequent sunburns. Interestingly, Garbe et al. (1994) showed that actinic lentigines were related to a tendency to develop ephelides in adolescence, suggesting that frecklers are more prone to having SLs later in life. This is not, however, consistent with the studies of Bastiaens et al. (1999) and Monestier et al. (2006), nor with the observation that ephelides are more frequently observed in lightskinned individuals, whereas SLs are more observed in darker skin types. In these studies, the definition of ephelides versus SLs may not always be clear.
An increased frequency and area of pigmented birthmarks in freckled individuals have been reported in a study of Australian schoolchildren aged 7–17 yr (Nicholls, 1968). Furthermore, freckling is clearly associated with an increased risk of melanoma (Bliss et al., 1995; Dubin et al., 1990; Mackie, 1998; Titus-Ernstoff et al., 2005). Bliss et al. performed a meta-analysis of 10 case–control studies and found that all seven studies that analyzed freckles reported an association with the formation of melanoma (Bliss et al., 1995). The risk was higher the more dense the freckles were. They also found association of blond/red hair, blue eyes and light skin with melanoma. This has been confirmed in numerous studies, including a study of 423 primary melanomas and 678 controls in the US where blond/red hair, blue eyes and the presence of freckles before the age of 15 were associated with melanoma (Titus-Ernstoff et al., 2005). Clearly, light pigmentation and freckles are a risk factor for melanoma.
Pigmented spots can also be induced by drug treatment, as has been described for the so-called PUVA lentigines, which arise upon treatment of patients with psoriasis with psoralens and ultraviolet A light (PUVA). This leads to the appearance of lentigines in otherwise sun-protected skin (Abel et al., 1985; Rhodes et al., 1983). Abel et al. found atypical melanocytes containing large hyperchromatic nuclei in 57% of the PUVA lentigines and in 70% of the non-lesional PUVA-exposed skin. A comparison with 24 samples from lentigo simplex or SL patients showed only two cases of atypic melanocytic nuclei. They further discovered binucleated melanocytes and the presence of giant melanosomes in the PUVA treated patients, which can cause further complications as melanocytic dysplasia or malignancy (Abel et al., 1985). When Nakagawa et al. (1984) compared PUVA lentigines from light-protected regions (the patients had not been treated with UV radiation) to solar lentigines, they found that the melanocytes from the PUVA patients had longer and more numerous dendrites and showed more active melanogenesis. The basal keratinocytes of the PUVA lentigines showed a significantly higher frequency of large, single melanosomes.
Ephelides are mostly associated with light skin and blond/-red hair color and appear during early childhood, suggesting that the formation of freckles in juveniles is largely genetically determined. Bataille et al. (2000) showed that additive genetic effects explained 91% of the variance in freckle counts, with no measurable environmental effects. Consistent with this, a number of genes have been shown to be important for the formation of freckles, including MC1R, IRF4, ASIP, TYR and BNC2 (Table 2). The majority of red-haired individuals are homozygous or compound heterozygous for variants in the melanocortin-1-receptor (MC1R) gene (Box et al., 1997; Flanagan et al., 2000; Smith et al., 1998; Valverde et al., 1995). MC1R variants also lead to pale skin. More importantly, the MC1R gene has been shown to be a major contributor to the formation of freckles (Bastiaens et al., 2001; Flanagan et al., 2000; Rees, 2004). Overall, this is referred to as the RHC (red hair, fair skin, lack of tanning ability and propensity to freckle) phenotype, with highly penetrant variants of the MC1R gene designated as R-alleles and lower penetrant alleles as r-alleles (Beaumont et al., 2011). MC1R is a seven pass transmembrane G protein-coupled receptor, which is located in the cell membrane of melanocytes (Chhajlani and Wikberg, 1992; Mountjoy et al., 1992). It binds different ligands with opposite characteristics. The ligands a-melanocyte stimulating hormone (aMSH) and the adrenocorticotropic hormone (ACTH) activate signaling through the receptor, leading to increased production of cyclic AMP (cAMP), which then leads to phosphorylation of the cAMP responsive element-binding protein (CREB), resulting in activation of expression of MITF, the melanocyte master regulator. MITF activates expression of several genes required for production of black eumelanin, including TYR, the enzyme that catalyzes synthesis of eumelanin (Busca and Ballotti, 2000). The agouti signaling protein (ASIP) is an antagonist of MC1R and prevents binding of aMSH and thus production of eumelanin. This leads to production of red or yellow pheomelanin, thus modulating the ratio of eumelanin and pheomelanin produced by melanocytes (Cone et al., 1996; Rees, 2003). More than 80 coding variants in MC1R have been described in populations of European ancestry (Gerstenblith et al., 2007; Makova and Norton, 2005). These variations lead to pigmentation traits such as fair skin, red and blond hair, UV sensitivity and freckles in both humans and animals (Kijas et al., 1998; Robbins et al., 1993; Rouzaud et al., 2006; Vage et al., 1999). The variants in MC1R affect receptor function (Beaumont et al., 2007, 2009; Garcia-Borron et al., 2005) or the promoter region (Motokawa et al., 2008); red hair and ephelides are found in the absence of MC1R function (Beaumont et al., 2008). Activating mutations in MC1R can lead to permanent activation of the protein and dark pigmentation, which has so far only been described for animals (e.g. mice, pigs and sheep) and not for humans.