Clinical dermatology • Original article CED

Clinical and Experimental Dermatology

Characteristics of keratinocytes in facial solar lentigo with flattened

rete ridges: comparison with melasma
J. Shin, J.-Y. Park, S. J. Kim and H. Y. Kang
Department of Dermatology, Ajou University School of Medicine, Suwon, Korea

doi:10.1111/ced.12621

Summary Background. Flattened rete ridges occurring on the face are not uncommon in
solar lentigo (SL).
Aim. To investigate the morphological changes of keratinocytes in facial SL with
flattened rete ridges and melasma.
Methods. In total, 25 patients with facial SL showing flattened rete ridges and 20
patients with melasma, in which rete ridge flattening is also a common feature, were
included in the study. Skin biopsies were performed on lesional skin and perilesional
normal skin.
Results. The histological findings showed that the epidermis was significantly
thicker in SL skin with flattened rete ridges compared with perilesional normal skin.
Comparative quantitative analysis of epidermal morphology revealed that the indi-
vidual keratinocytes were larger in size in lesional skin, without a significant change
in the number of cells. Anti-p16 antibody staining was more intense in lesional epi-
dermis, suggesting senescence of keratinocytes in the thickened epidermis. By con-
trast, melasma samples showed no significant difference in epidermal thickness or
keratinocyte morphology in terms of number or size compared with normal skin.
Conclusions. These findings suggest that senescent changes in keratinocytes are
important in the development of SL, even in the absence of rete ridge elongation,
and the removal of keratinocytes harbouring melanin could be a possible strategy
for SL treatment.

Introduction more often on the face compared with other anatomi-

Solar lentigo (SL) is a common acquired hyperpigmen-
tary disorder, frequently occurring on the face. The
most histologically distinguishable features that differ-
entiate SL from other pigmentary disorders are the
increase in the number of melanocytes and in content
of melanin in the lower epidermis, and the downward
elongation of rete ridges.1,2 Nonetheless, flattened rete
ridges are a frequent finding in SL, and tend to occur
Correspondence: Dr Hee Young Kang, Department of Dermatology, Ajou
University School of Medicine, 206, Worldcup-ro, Yeongtong-gu, Suwon,
Gyeonggi-do, 443–749, Korea
E-mail: hykang@ajou.ac.kr
Conflict of interest: the authors declare that they have no conflicts of
interest.
Accepted for publication 16 September 2014
cal sites. In an earlier report,2 more than half of cases
of facial SL showed flattening rather than elongation of
rete ridges. In a more recent study carried out in Japa-
nese patients, more severe solar elastosis and fewer
Langerhans cells were seen in the epidermis in the 20
cases (out of 40) that showed flattened rete ridges,
compared with the budding (elongated) group.2 The
authors suggested that more severely sun-damaged
solar lentigines might show the changes observed in
the flattened rete ridge group. These studies support
that elongation of the rete ridges is not an essential
histological finding in SL.
In cases of facial SL showing flattened rete ridges,
the histology is comparable to that of melasma, in
which rete ridge flattening with solar elastosis is a
common feature.3–6 Therefore, facial SL and melasma

ª 2015 British Association of Dermatologists Clinical and Experimental Dermatology (2015) 40, pp489–494 489
Keratinocytes in solar lentigo  J. Shin et al.
share the histological features of conspicuous pigmen-
tation within the epidermis due to higher expression of
melanogenesis-associated proteins, as well as flattened
rete ridges with solar elastosis.7,8 However, despite the
general histological resemblance, there is a consensus
that SL responds relatively well to conventional laser
or light treatment, whereas melasma does not. This
suggests that there may be subtle biological differences
between the two conditions, despite their apparently
similar histology.
In this study, we focused on the morphological and
immunohistochemical changes of keratinocytes in
facial SL, especially in patients with flattened rete
ridges, which were compared with perilesional normal
skin and also with melasma. This study may also pro-
vide a new way of distinguishing flattened SL from
melasma.
Methods
This study was approved by the institutional review
board of Ajou University Hospital (AJIRB-GEN-GEN-
10–319), and informed consent was obtained from all
participants for collection of biopsy specimens.
Biopsy collection
A retrospective study of 25 facial SL samples showing
flattened rete ridges in haematoxylin and eosin (H&E)-
stained sections and of 20 melasma samples were per-
formed. The flattened rete ridges were defined by absence
of a single elongation of rete ridge under 9 200 magnifi-
cation by two independent dermatologists.
All subjects were Korean women with Fitzpatrick skin
types III or IV. The average ages of SL and melasma
groups were 46.5 and 40 years, respectively. Diagnosis
of each disease was based on physical examination and
confirmed by histopathological findings. Skin biopsy
samples (2 mm) were obtained from lesional and adja-
cent perilesional normal skin (usually within 10 mm
from the lesion margin) from all patients.
Stains
Tissue sections were embedded in paraffin wax, and
cut into sections 3 lm thick, then processed for light
microscopy examination. H&E stain was used to study
the general histopathological changes in the lesional
skin of SL and melasma, compared with the adjacent
normal facial skin.
To allow counting of the number of melanocytes in
SL sections, monoclonal antibodies to microphthalmia-
490 Clinical and Experimental Dermatology (2015) 40, pp489–494
associated transcription factor (MITF) (Leica Biosys-
tems, Newcastle, UK) were used for immunohisto-
chemical staining. Mouse monoclonal anti-p16
antibody (1 : 100; Santa Cruz Biotechnology, Dallas,
TX, USA) was used for immunohistochemistry of SL
sections. Cellular enlargement, or hypertrophy, is a
well-known feature of ageing cells, and p16 expression
can be used as a marker of senescence.9
Analysis and image analysis
A digital camera (SPOT Flex; SPOT Imaging Solutions,
Sterling Heights, MI, USA) mounted on a microscope
(BX50; Olympus Optical, Tokyo, Japan) was used to
capture images of the sections. The images were eval-
uated using Image Pro Plus (v4.5; Media Cybernetics
Co., Silver Spring, MD, USA), on a representative area
of each specimen.
The number of melanocytes was measured as the
number of MITF+ melanocytes per 1 mm length of
rete ridge. Epidermal area was measured by manually
tracing the borders of the epidermis and rete ridges,
but the stratum corneum and the epidermal area con-
taining hair follicles were excluded from this tracing.
The number of keratinocytes in the cross-section was
counted by detecting the nuclei. Using the data
obtained from the image analysis, relative epidermal
thickness was estimated by dividing the area by the epi-
dermal length in the trace. To determine whether alter-
ation of epidermal area was caused by hypertrophy
(cellular enlargement) or hyperplasia (numerical
increase) of keratinocytes, the two-dimensional area
per keratinocyte was calculated by dividing the epider-
mal area by the number of keratinocytes in the traced
area. Consequently, the number of keratinocytes was
adjusted to per 100 lm of epidermal length to compen-
sate for the irregularity of traced areas.
For p16 staining, positivity was determined semi-
quantitatively through visual grading by two indepen-
dent observers, using a four-point scale (0 = negative;
1 = weak; 2 = moderate; 3 = strongly positive).
Statistical analysis
Data are expressed as mean values with their stan-
dard deviation. Data had a Gaussian distribution,
thus paired t-test was performed to compare the
morphological features of normal and lesional skin
of each subjects quantitatively. P < 0.05 was consid-
ered statistically significant. GraphPad Prism (v5;
GraphPad, La Jolla, CA, USA) was used for statistical
analysis.
ª 2015 British Association of Dermatologists
 Keratinocytes in solar lentigo  J. Shin et al.

Results Comparative quantitative analysis

Lesional epidermis of SL was significantly thicker than

Histological staining
H&E staining demonstrated flattened rete ridges with
basal hyperpigmentation in all SL and melasma speci-
mens (Fig. 1a). The number of melanocytes was
increased in lesional SL skin, but not in lesional mel-
asma skin. These findings were confirmed by immuno-
histochemical staining using monoclonal antibodies to
MITF (Fig. 1b). There were no significant differences
in the level of lesional epidermal pigmentation
between SL and melasma (data not shown). Variable
degrees of solar elastosis were evident in the facial
specimens, without significant inflammatory infiltra-
tion.
that of perilesional normal skin (66.64  3.87 vs.
51.22  2.99 lm, P < 0.001). The area of individual
keratinocytes was also significantly larger in the le-
sional skin (181.13  9.72 vs. 128.38  5.79 lm2,
P < 0.001). Despite the differences in epidermal thick-
ness and area occupied by each keratinocyte, the
number of keratinocytes per 100 lm of epidermal
length remained fairly constant between SL and nor-
mal skin (38.27  2.27 vs. 40.27  1.72, P = 0.41).
By contrast, there were no significant differences in
epidermal thickness, epidermal area per keratinocyte
or number of keratinocytes between lesional and nor-
mal melasma skin (Table 1).

(a)

(b)

Figure 1 (a) Thickened epidermis without elongation of rete ridges in facial solar lentigo, whereas in melasma, the rete ridges are flat-
tened without epidermal thickening (haematoxylin and eosin, original magnification 9 200). (b) Immunohistochemical staining using
monoclonal antibodies to microphthalmia-associated transcription factor; the number of melanocytes was increased in the lesional skin
of solar lentigo, but not in the lesional skin of melasma (original magnification 9 200).

ª 2015 British Association of Dermatologists Clinical and Experimental Dermatology (2015) 40, pp489–494 491
Keratinocytes in solar lentigo  J. Shin et al.

Immunohistochemistry
p16 staining was positive in 10 of 25 SL samples.
Semi-quantitative visual analysis demonstrated signifi-
cantly increased staining intensity in the lesional epi-
dermis (1.78 vs. 0.90, P < 0.001) (Fig. 2). p16
staining was negative in the epidermis of all melasma
samples (Fig. 2b).
Discussion
The present study showed that although flattening of
rete ridges is a feature found in facial SL, the epidermis
in SL was still significantly thickened compared with
perilesional normal skin. The thickened epidermis in
SL might be due to enlargement of individual
keratinocytes, not to a change in their numbers. By

Table 1 Epidermal thickness and mor-

Epidermal Epidermal area per Keratinocytes per phological features of keratinocytes in
Skin sample thickness, lm keratinocyte, lm2 100 lm epidermis, n facial solar lentigo and melasma.
Solar lentigo
Normal 51.22  2.99 128.38  5.79 40.27  1.72
Lesional 66.64  3.87* 181.13  9.72* 38.27  2.27
Melasma
Normal 48.24  1.86 167.36  6.17 29.13  1.22
Lesional 51.33  2.43 168.80  4.08 30.40  1.16

*P < 0.05.

(a)

(b)

Figure 2 Anti-p16 antibody staining of normal and lesional skin from (a) solar lentigo and (b) melasma (original magnification 9 200).

492 Clinical and Experimental Dermatology (2015) 40, pp489–494 ª 2015 British Association of Dermatologists
 Keratinocytes in solar lentigo  J. Shin et al.

contrast, melasma samples in our study showed no Conclusion

significant difference in epidermal thickness or
Our study suggests that changes in keratinocytes are
keratinocyte morphology in terms of number or size
important in the development of SL, even in the
compared with normal skin.
absence of the rete ridge elongation. Only a few stud-
These findings suggest that certain alterations in
ies of SL have been carried out on facial skin,
keratinocytes are implicated in SL and that enlarge-
although the face is the most common site of occur-
ment of keratinocytes might have a major role in the
rence and is the cosmetically most problematic area.
development of SL. In contrast to SL, the emerging
Further investigations focusing on the exact patho-
key pathophysiological finding in melasma is the inter-
mechanism underlying development SL are warranted,
action between melanocytes and dermal components,
which may also enable more effective therapies and
such as vascular structure, fibroblasts and extracellu-
even prevention of SL.
lar matrix.10–12 Despite the similarity in histological
features between SL with flattened rete ridges and mel-
asma, there are major difference in their underlying Acknowledgement
mechanisms.
This work was supported by a grant (NRF-
In the present study, we found that p16 staining
2013R1A2A2A01013421) of National Research
was positive in the 10 of 25 SL samples and seemed
Foundation to Hee Young Kang.
to increase in line with the epidermal thickness (data
not shown). p16 protein was found to be focally
expressed in a few epidermal cells, which is consis-
tent with previous reports.13,14 p16 expression in the What’s already known about this topic?
cell occurs heterogeneously, and therefore, p16 levels
• Flattened rete ridges are a common finding in
might be dependent on the different stages of senes-
SL occurring on the face.
cence in individual keratinocytes. It has been sug-
• There is a consensus that SL responds relatively
gested that keratinocyte expression of p16 increases
well to conventional laser or light treatment,
steadily with each passage.14 In our samples, we
whereas the result is disappointing in melasma.
found that p16 expression was almost doubled in le-
• This suggests that there may be subtle biologi-
sional SL skin. These findings suggest that senescence
cal differences between SL and melasma, despite
of keratinocytes might explain the hypertrophy of
the apparently similar histology.
keratinocytes in the thickened epidermis. Indeed, it
has been suggested that SL has a shared genetic
basis with seborrhoeic keratosis, in which the
enhanced expression of p16 indicates that keratino-
cytes are in a senescent condition.15,16 What does this study add?
Recently, attempts to establish the evolution of SL • Contrary to melasma, SL is probably initiated
have been made.17–19 It has been suggested that by keratinocyte proliferation followed by quies-
keratinocyte proliferation is responsible for the initiation cence of enlarged senescent keratinocytes with
of SL, and that during the disease evolution, melanin higher melanin burden, which could explain the
accumulation correlates with reduced keratinocyte persistence of SL.
proliferation.16 Melanin-loaded keratinocytes in SL • Removal of keratinocytes harbouring melanin
show lower proliferative capacity and altered differenti- could be a possible strategy for SL treatment.
ation, and are resistant to cell apoptosis.19
Taken together, these results indicate that SL may
be initiated by keratinocyte proliferation, followed by
quiescence of enlarged senescent keratinocytes with
higher melanin burden, which could explain the
References
persistence of SL. Therefore, removal of keratinocytes 1 Andersen WK, Labadie RR, Bhawan J. Histopathology of
harbouring melanin could be a possible strategy in SL solar lentigines of the face: a quantitative study. J Am
treatment. However, we cannot conclude that the Acad Dermatol 1997; 36: 444–7.
positive p16 staining seen in this study is a proof of 2 Yonei N, Kaminaka C, Kimura A et al. Two patterns
senescence of keratinocytes in the SL. of solar lentigines: a histopathological analysis of

ª 2015 British Association of Dermatologists Clinical and Experimental Dermatology (2015) 40, pp489–494 493
Keratinocytes in solar lentigo  J. Shin et al.
40 Japanese women. J Dermatol 2012; 39:
829–32.
3 Hernandez-Barrera R, Torres-Alvarez B, Castanedo-
Cazares JP et al. Solar elastosis and presence of mast
cells as key features in the pathogenesis of melasma.
Clin Exp Dermatol 2008;
33: 305–8.
4 Kang WH, Yoon KH, Lee ES et al. Melasma:
histopathological characteristics in 56 Korean patients.
Br J Dermatol 2002; 146: 228–37.
5 Sarvjot V, Sharma S, Mishra S et al. Melasma: a
clinicopathological study of 43 cases. Indian J Pathol
Microbiol 2009; 52: 357–9.
6 Torres-Alvarez B, Mesa-Garza IG, Castanedo-Cazares JP
et al. Histochemical and immunohistochemical study in
melasma: evidence of damage in the basal membrane.
Am J Dermatopathol 2011; 33: 291–5.
7 Aoki H, Moro O, Tagami H et al. Gene expression
profiling analysis of solar lentigo in relation to
immunohistochemical characteristics. Br J Dermatol
2007; 156: 1214–23.
8 Kang HY, Suzuki I, Lee DJ et al. Transcriptional profiling
shows altered expression of wnt pathway- and lipid
metabolism-related genes as well as melanogenesis-
related genes in melasma. J Invest Dermatol 2011; 131:
1692–700.
9 Soroka Y, Ma’or Z, Leshem Y et al. Aged keratinocyte
phenotyping: morphology, biochemical markers and
effects of Dead Sea minerals. Exp Gerontol 2008; 43:
947–57.
10 Kang HY, Hwang JS, Lee JY et al. The dermal stem cell
factor and c-kit are overexpressed in melasma. Br J
Dermatol 2006; 154: 1094–9.
494 Clinical and Experimental Dermatology (2015) 40, pp489–494
11 Kim EH, Kim YC, Lee ES et al. The vascular
characteristics of melasma. J Dermatol Sci 2007; 46:
111–16.
12 Lee DJ, Park KC, Ortonne JP et al. Pendulous
melanocytes: a characteristic feature of melasma and
how it may occur. Br J Dermatol 2012; 166: 684–6.
13 Hodges A, Smoller BR. Immunohistochemical
comparison of p16 expression in actinic keratoses and
squamous cell carcinomas of the skin. Mod Pathol 2002;
15: 1121–5.
14 Natarajan E, Omobono JD 2nd, Jones JC et al. Co-expression
of p16INK4A and laminin 5 by keratinocytes: a wound-
healing response coupling hypermotility with growth arrest
that goes awry during epithelial neoplastic progression.
J Investig Dermatol Symp Proc 2005; 10: 72–85.
15 Hafner C, Stoehr R, van Oers JM et al. FGFR3 and PIK3CA
mutations are involved in the molecular pathogenesis of
solar lentigo. Br J Dermatol 2009; 160: 546–51.
16 Nakamura S, Nishioka K. Enhanced expression of p16 in
seborrhoeic keratosis; a lesion of accumulated senescent
epidermal cells in G1 arrest. Br J Dermatol 2003; 149:
560–5.
17 Cario-Andre M, Lepreux S, Pain C et al. Perilesional vs.
lesional skin changes in senile lentigo. J Cutan Pathol
2004; 31: 441–7.
18 Lin CB, Hu Y, Rossetti D et al. Immuno-histochemical
evaluation of solar lentigines: the association of KGF/
KGFR and other factors with lesion development. J
Dermatol Sci 2010; 59: 91–7.
19 Pollefliet C, Corstjens H, Gonzalez S et al. Morphological
characterization of solar lentigines by in vivo reflectance
confocal microscopy: a longitudinal approach. Int J
Cosmet Sci 2013; 35: 149–55.
ª 2015 British Association of Dermatologists