Control of melanocyte
differentiation by a MITF–
PDE4D3 homeostatic circuit
Mehdi Khaled, Carmit Levy, and David E. Fisher1
Department of Dermatology, Cutaneous Biology Research
Center, Massachusetts General Hospital, Harvard Medical
School, Massachusetts 02114, USA
Cyclic AMP (cAMP) is a ubiquitous second messenger
that regulates a variety of biological processes. The mag-
nitude and duration of cAMP expression are regulated by
both production and hydrolysis. Melanocyte-stimulating
hormone (MSH) plays a crucial role in pigment cell differ-
entiation via cAMP-regulated expression of the master
transcription factor MITF. We report the identification
of phosphodiesterase 4D3 as a direct target of the MSH/
cAMP/MITF pathway. This creates a negative feedback
loop that induces refractoriness to chronic stimulation
of the cAMP pathway in melanocytes. This homeostatic
pathway highlights a potent mechanism controlling me-
lanocyte differentiation that may be amenable to phar-
macologic manipulation for skin cancer prevention.
Supplemental material is available at http://www.genesdev.org.
Received April 15, 2010; revised version accepted August 30,
Cyclic AMP (cAMP) is a second messenger that regulates
key processes (such as cell growth, differentiation, and
movement) and specialized actions unique to specific cell
lineages (Houslay et al. 2007). cAMP is typically metab-
olized by phosphodiesterases (PDEs), which control the
magnitude, duration, and subcellular localization of
cAMP (Houslay et al. 2007). In melanocytes the cAMP/
CREB signaling pathway is potently regulated by mela-
nocyte-stimulating hormone (MSH) via the G protein-
coupled receptor MC1R, which in turn transcriptionally
activates expression of MITF (Bertolotto et al. 1998; Price
et al. 1998), the master transcriptional regulator of me-
lanocyte development (Levy et al. 2006). In vitro and in
vivo, the up-regulation of cAMP in melanocytes leads to
increased pigment production and protection of the skin
against the deleterious effects of UV radiation (D’Orazio
et al. 2006). Skin is the most common organ to be affected
by cancer, and cutaneous malignancies most commonly
occur in fair-skinned people with a limited capacity to tan
(Fitzpatrick and Sober 1985; Fitzpatrick 1988; Sturm et al.
2003; Rijken et al. 2004). The role of pigmentation in
modulating human skin cancer risk continues to be in-
completely understood. However, darkly pigmented peo-
ple or those who tan easily (Fitzpatrick phototypes 3–6)
exhibit significantly diminished skin cancer risk relative
to those with light skin who tan poorly (Fitzpatrick
phototypes 1,2) (Lin and Fisher 2007).
Here we report that, in melanocytes, MITF directly
regulates the transcription of the PDE4D3 gene, creating
negative homeostatic control of the cAMP pathway and
lineage differentiation. These findings provide a rationale
for targeting PDE4D3 to modulate MITF expression and
control skin pigmentation.
Results and Discussion
PDE4D expression is MITF-dependent in melanocytes
We used annotated microarray data (http://biogps.gnf.
org/downloads) to search for PDEs that are expressed in
the melanocyte lineage. Among the PDEs known to
degrade cAMP (Conti and Beavo 2007; Omori and Kotera
2007), the following exhibited ‘‘present calls’’ in some or
most of the melanoma lines: PDE1C, PDE3A, PDE3B,
PDE4A, PDE4B, PDE4D, PDE6C, PDE7B, PDE8A, and
PDE10A. When the expression of these PDEs was com-
pared with that of MITF, we observed that two of them,
PDE4D and PDE4B, exhibited fluctuations in expression
that correlated with MITF expression, suggesting that
they could be transcriptionally linked to MITF (Supple-
mental Fig. S1). In this analysis, we used tyrosinase, a
known target of MITF, as a positive control (Bentley et al.
1994; Yasumoto et al. 1994). Interestingly, PDE4C did not
correlate with MITF.
To address this possibility, we first investigated whether
MITF was necessary for PDE4D and PDE4B expression
(Fig. 1A). siRNA-mediated knockdown of MITF in three
different primary human melanocyte cultures signifi-
cantly suppressed expression of PDE4D, but not PDE4B.
This experiment was carried out using primers able to
recognize all isoforms of the PDE4D gene. Expression of
LEF-1, an additional control, was also unaffected. The
knockdown efficiency of MITF was also verified by West-
ern blot (Supplemental Fig. S2).
PDE4D is a complex gene encoding multiple isoforms
(Supplemental Fig. S3) with distinct regulation and tissue
distribution (D’Sa et al. 2002; Richter et al. 2005). To
identify the specific isoforms expressed in primary mela-
nocytes, isoform-specific primers were used and revealed
expression of seven of the nine PDE4Ds known to be
expressed in humans (all except PDE4D4 and PDE4D8)
(Supplemental Fig. S4A). Since MITF is regulated by
signaling pathways using cAMP (Bertolotto et al. 1998;
Price et al. 1998), we examined the effect of forskolin on
the expression of the PDE4D isoforms in melanocytes.
Only PDE4D3 mRNAwas consistently up-regulated upon
forskolin stimulation (Supplemental Fig. S4B). We ob-
served induction of PDE4D3 mRNA levels with a slight
delay (first seen at 6 h), consistent with the possibility of
a mechanistic intermediate between forskolin/cAMP and
PDE4D3 transcription. Slight suppression of PDE4D3
levels at 2 h wasalso observed in fibroblasts (Supplemental
Fig. S5A), suggesting a mechanism unrelated to M-MITF.
Treatment of normal human fibroblasts with forskolin
failed to induce the expression of PDE4D (using common
region primers) or PDE4D3 (Supplemental Fig. S5A), al-
though it did induce phosphorylation of CREB, confirming
[Keywords: cAMP; MITF; PDE4D; skin; pigmentation]
E-MAIL firstname.lastname@example.org; FAX (617) 632-2085.
Articleis online at http://www.genesdev.org/cgi/doi/10.1101/gad.1937710.
2276 GENES & DEVELOPMENT 24:2276–2281 ? 2010 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/10; www.genesdev.org
the functionality of forskolin in fibroblasts (Supplemental
In rat cortical neurons, it has been observed previously
that cAMP signaling can directly target the expression of
Pde4b2 and Pde4d1/2 but not Pde4d3 (D’Sa et al. 2002).
Although we observed PDE4B2 to be expressed in melano-
cytes, its expression did not change in response to cAMP
signaling (datanot shown), suggestingthat tissue-restricted
mechanisms may regulate the precise PDE species in-
volved in feedback loops of different cell lineages.
MITF regulates the transcription of PDE4D3
To assess whether MITF is involved in the regulation of
PDE4D3 expression, melanocytes were stimulated with
forskolin in the presence of siRNA targeting MITF (or
control siRNA). Whereas PDE4D3 was up-regulated by
forskolin in control siRNA-treated cells, it was not up-
regulated in melanocytes transfected with either of two
siRNAs targeting MITF (Fig. 1B; Supplemental Fig. S6A),
indicating that up-regulation of PDE4D3 is MITF-depen-
dant in melanocytes. The MITF knockdown efficiency
was verified by Western blot (Supplemental Fig. S6B).
Although MITF-dependent PDE4D3 up-regulation by
forskolin was a consistent finding, we observed some
variability in the kinetics of its induction (between 6 and
12 h) that was linked to the donor source of the primary
human melanocytes. It is currently unknown if these
varied kinetics correlate with differing pigmen-
MITF, UACC62 melanoma cells were trans-
fected with a luciferase reporter driven by
1.8 kb upstream of the translation start site of
PDE4D3. We observed stimulation of luciferase
activity by forskolin, and this up-regulation was
lost when the consensus MITF-binding site
(Ebox) was mutated (CACATG to GAGATG)
(Fig. 1C). The ability of MITF to bind to the
Ebox of the 1.8-kb promoter region of PDE4D3
was verified by an electropheretic mobility shift
assay (Supplemental Fig. S7). The supershift
observed by adding an MITF antibody in the
binding reaction was successfully chased by
a nonbiotinylated Ebox wild-type probe and
failed when using the Ebox mutated probe.
To investigate if MITF directly occupies the
promoter region of the PDE4D3 gene, we per-
formed chromatin immunoprecipitation (ChIP)
using genomic DNA from primary human
melanocytes. When the relative occupancy of
MITF on the PDE4D3 promoter was quantified,
a significant enrichment was seen that was
comparable with MITF occupancy measured
at the tyrosinase promoter (Fig. 1D). As a nega-
tive control, detection of the exon-14/intron-14
boundary of PDE4D3 failed to show enrich-
ment of MITF binding. The transcriptional reg-
ulation of PDE4D3 by MITF is particularly
notable because MITF is a central regulator
of the pigmentation cascade via targeting of
numerous enzymes and other factors impor-
tant for melanocyte-specific synthesis, matura-
tion, and transport of melanin. Interestingly,
examination of a gene expression data set from
MITF-overexpressing melanoma cells showed
PDE4D up-regulation by MITF (Hoek et al. 2008). Also,
PDE4D was found to be regulated by the MC1R pathway
in mouse melanocytes (Le Pape et al. 2008).
Inhibition of PDE4 sustains CREB phosphorylation
and MITF expression in melanocytes
To explore the physiological role of PDE4D3 in melano-
cytes, we investigated the impact of its inhibition by
rolipram, a known PDE4 inhibitor that has been studied
previously as an anti-inflammatory and anti-depressant
agent (Bobon et al. 1988; Griswold et al. 1993). We tested
the toxicity of this inhibitor and its effect on the cells’
morphology. We observed that it does not significantly
affect human primary melanocyte growth or alter cellular
morphology (Supplemental Fig. S8A,B). We observed the
same results when the cells were exposed to forskolin
or the combination of forskolin and rolipram, although
the combination of forskolin and rolipram modestly in-
creased pigmentation by 48 h (data not shown). At the
cAMP signaling level, we observed that pretreatment
of melanocytes with rolipram prolonged CREB phos-
phorylation after forskolin stimulation (Fig. 2A). Several
PDE4D isoforms are activated via PKA phosphorylation,
creating a short-term feedback loop (Sette and Conti
1996; MacKenzie et al. 2002; Richter et al. 2005). This
phenomenon could explain why rolipram extends the
duration of forskolin-induced CREB phosphorylation.
Primary human melanocytes were transfected with a nontargeting control siRNA
(siCtrl) or a siRNA specific for MITF. Total mRNA was subjected to qPCR. The data
are normalized to b-actin. (B) Primary human melanocytes were transfected with
a nontargeting siRNA (siCtrl) or siRNAs targeting MITF and exposed to forskolin,
as indicated. Total mRNA was subjected to qPCR. Results are expressed as fold
stimulation and represent the mean 6 SD of three independent experiments. (C)
UACC62 cells were transiently transfected with reporter plasmids pPDE4D3 and
pPDE4D3 Ebox mutated. After 24 h of forskolin stimulation, firefly luciferase was
measured and normalized to the renilla luciferase control. Results are shown as fold
stimulation and represent the mean 6 SD of three experiments performed in
triplicate. (TSS) Transcription start site. (D) ChIP from primary human melanocytes
was performed using MITF or control antibody (see the Materials and Methods).
RNA levels were analyzed by qPCR. Data are plotted as fold change versus control
antibody, and each point is the mean 6 SD of three independent experiments.
P-values were obtained by a paired t-test. (*) P < 0.05; (**) P < 0.005.
cAMP-induced up-regulation of PDE4D3 is dependent on MITF. (A)
MITF regulates PDE4D3 in melanocytes
GENES & DEVELOPMENT2277
PDE4D3 has also been shown to be inhibited by phos-
phorylation by ERK2, although this inhibition is ablated
by subsequent PKA phosphorylation (Hoffmann et al.
1999), an event that is likely to occur in melanocytes
(Busca et al. 2000).
Since MITF expression is directly regulated by CREB,
we investigated the combined effect of rolipram and
forskolin on MITF mRNA expression. Primary melano-
cytes were pretreated with DMSO or rolipram for 30 min,
and then exposed to forskolin or DMSO for up to 4 h
before mRNAwas collected and analyzed by quantitative
PCR(qPCR)(Fig. 2B).Whereas thecombination of rolipram
plus forskolin only minimally affected peak MITF levels,
the duration of MITF induction was significantly pro-
longed. We also verified the effect of PDE4 inhibition of
MITF protein expression, and we observed that rolipram
did not significantly affect the expression level of MITF
after 8 h of stimulation, but prolonged its expression.
Indeed, at 24 h, MITF expression remained higher than
control only when the cells were exposed to the combi-
nation of rolipram and forskolin (Fig. 2C). These results
suggest that PDE4 activity plays a measurably important
role in modulating MITF, and suggest a testable strategy to
modulate skin pigmentation.
To further examine whether PDE4D3 regulation by
MITF is a physiologically relevant negative feedback loop,
we asked whether prior treatment with forskolin may
render melanocytes resistant to repeated forskolin treat-
ment due to negative feedback mediated by PDE4D3.
Primary melanocytes were treated on day 1 for 4 h with
forskolin (or DMSO control) and then returned to minimal
media until repeat forskolin challenge on day 2 in the
presence or absence of pretreatment with either siRNA
directed against PDE4D3 or control siRNA. Melanocytes
pretreated withDMSOonday1 weresensitivetoforskolin
on day 2, as measured by phosphorylation of CREB.
Melanocytes treated with forskolin on day 1 were insen-
phosphorylation. However, sensitivity to repeated forsko-
of PDE4D3 (Fig. 3A). The efficiency of PDE4D3 knock-
down was assessed by qPCR (Supplemental Fig. S9). The
same phenomenon was observed when looking at the
expression of a target of MITF: PMEL17. When primary
human melanocytes were exposed to DMSO on day 1, we
observed an up-regulation of PMEL17 mRNA expression
on day 2 after 8 h of forskolin treatment. However, if the
cells were exposed previously to forskolin, we did not
observe an up-regulation of PMEL17 after forskolin treat-
ment. The sensitivity to forskolin was restored when the
cells were transfected with an siRNA specific for PDE4D3
phosphorylation in primary human melanocytes. (A) Extracts from
primary human melanocytes placed in minimal media for 14 h
before being treated as indicated were immunoblotted using anti-
bodies against phospho-CREB and a-tubulin. (B) Primary human
melanocytes were exposed to the indicated conditions for 30 min
before being rinsed and placed in minimal media for the indicated
times. The results are normalized to b-actin and are shown as fold
induction of MITF. Each data point is the mean 6 SD of three
experiments. (*) P < 0.05. (C) Extracts from primary human mela-
nocytes were placed in minimal media for 14 h before being exposed
for 30 min to rolipram or DMSO prior to being treated with forskolin
for 30 min. Subsequently, the cells were rinsed and placed in
minimal media for the indicated times. The protein extracts were
submitted to immunoblotting using MITF and a-tubulin antibodies.
Inhibition of PDE4 potentiates forskolin-induced CREB
is reversible by PDE4D3 knockdown or rolipram treatment. (A)
Primary human melanocytes were transfected with a nontargeting
siRNA (siCtrl) or siRNAs targeting PDE4D3. Twenty-four hours
after transfection, the cells were placed in minimal media (MM) for
14 h, and were subsequently treated with forskolin for 4 h before
being returned to minimal media. The following day, the cells were
exposed to forskolin for the indicated times. Cell lysates were
subjected to Western blotting using antibodies specific for phos-
pho-CREB and a-tubulin. The intensity of the phospho-CREB
Western blot was measured and normalized to a-tubulin. (B) Primary
human melanocytes were incubated in minimal media for 14 h,
treated with forskolin or DMSO for 4 h, and returned to minimal
media. The following day, the cells were exposed to DMSO or
rolipram or Ro 20-1724 for 4 h and then rechallenged with forskolin
for the indicated times. Cell lysates were subjected to Western
blotting using antibodies specific for phospho-CREB and a-tubulin.
The intensity of the phospho-CREB Western blot was measured and
normalized to a-tubulin.
Forskolin pretreatment induces forskolin resistance that
Khaled et al.
2278 GENES & DEVELOPMENT
but not when they were transfected with a siRNA
directed against PDE4D5. This experiment suggests that
PDE4D3 isthe mainisoform ofPDE4D preventingchronic
activation of the cAMP pathway in human primary
melanocytes (Supplemental Fig. S10). The same experi-
ment was also performed by adding the PDE4 inhibitors
rolipram or Ro 20-1724 prior to forskolin stimulation on
day 2. Here, the drug treatments similarly restored CREB
phosphorylation by forskolin on day 2 (Fig. 3B). We
conclude that up-regulation of PDE4D3 by the MITF
autoregulatory loop normally limits repetitivestimulation
of the melanogenic pathway.
PDE4 inhibition synergizes with forskolin to induce
To study the relevance of the regulation of PDE4D3 by
MITF in vivo, we asked if topical PDE4 inhibitors may
enhance skin pigmentation in C57bl6e/eK14 SCF mice
(Kunisada et al. 1998; D’Orazio et al. 2006). This mouse
strain carries a frameshift mutation inMC1R, causing the
red hair phenotype, as well as a transgene (K14-SCF) that
produces ‘‘humanized’’ skin containing epidermal mela-
nocytes (in contrast to wild-type mice that lack melano-
cytes in the interfollicular epidermis) (Kunisada et al.
1998). After 5 d, a modest degree of melanization was
observed in the mice exposed to either forskolin or two
different PDE4 inhibitors: rolipram and Ro 20-1724.
However, the combination of forskolin with either of
the PDE4 inhibitors induced robust darkening (Fig. 4A).
C57bl6e/emice harboring no melanocytes in the epider-
mis served as a negative control. Fontana-Masson stain-
ing of skin showed that skin darkness correlated with
increased melanin pigment (Fig. 4B). Darkening of the
skin was quantified using reflective colorimetry (Supple-
mental Table 1; Park et al. 1999), and showed that the
combination of forskolin and rolipram or forskolin and
Ro 20-1724, applied once daily for five consecutive days,
induced pigmentation comparable with the effect ob-
served previously after 21 d of forskolin treatment alone
(D’Orazio et al. 2006).
Human epidemiological data and mouse models have
both demonstrated correlations between pigmentation
and skin cancer risk (D’Orazio et al. 2006; Lin and Fisher
2007). Whereas skin pigmentation may thus be protective
against UV carcinogenesis, the standard method of en-
hancing skin pigmentation (via UV exposure) carries
simultaneous carcinogenic risk. Parenteral peptide ana-
logs of MSH exhibit strong skin-darkening effects (Lerner
and McGuire 1961), although they require systemic de-
livery. Prior topical application of forskolin was seen to
darken the skin of redhead/light-skinned mice and confer
significant protection against subsequent UV carcino-
genic challenge (D’Orazio et al. 2006). However, forskolin
exhibits very poor human skin penetration and thus has
limited efficacy (data not shown). The development of
drugs that function as antagonists of inhibitors (such
as PDE4D inhibitors) may represent easier drug discov-
ery than identification of enzyme stimulators (such as
forskolin, which activates adenylate cyclase). Aside from
their potential roles in protection against skin cancer and
photo-aging, the use of topical agents affecting pigmen-
tation may provide opportunities to better understand
signaling and trafficking components of the pigmentation
response through the ability to pharmacologically and
synchronously activate melanogenesis. The homeostatic
pathway reported here (Fig. 5) uses a tissue-restricted
transcription factor (MITF) to target expression of a factor
that down-regulates a major stimulatory pathway. MITF
et al. 2005), and this PDE4D3 pathway may thus be sig-
nificantly altered, carrying therapeutic implications wor-
thy of study. PDEs modulate numerous signaling path-
ways in distinct lineage contexts. It is thus plausible that
comparable lineage-specific transcriptional mechanisms
may exist in other cell types that provide feedback con-
trol of cellular signaling
Materials and methods
Primary human melanocytes from neonatal foreskins were established in
Ham’s F10 (Invitrogen) medium containing 7% fetal bovine serum,
penicillin/streptomycin/L-glutamine (Invitrogen), 1 3 10?4M 3-isobutyl-
(Sigma), 1 mM Na3VO4, and 1 3 10?3M N6,29-O-dibutyryladenosine
3:5-cyclic monophosphate (Sigma). Minimal media was made with Ham’s
F10 medium supplemented with 3% fetal bovine serum and penicillin/
streptomycin/L-glutamine. UACC62 human melanoma cells were ob-
tained from NCI and grown in RPMI medium supplemented with 10%
fetal bovine serum and penicillin/streptomycin/L-glutamine.
ChIP was performed as described previously (Du et al. 2004). The
immunoprecipitations were performed with MITF polyclonal antibody
and Placental Protein 4 (control antibody) purchased from Assay Designs.
qPCR was carried out using primers specific for the promoter region
of human PDE4D3 (59-GAGAACAGCCAGCCTTATTATGGG-39 and
59-CTGCTCTGCAGGACAAGATTACCA-39) or spanning the junction
of exon 14 and intron 14 of human PDE4D3 (59-ACGTGGCATGGAGA
TAAGCCC-39 and 59- AACCAAATGCTAAAGCGGTAGCTC-39). The
human tyrosinase promoter primers used were 59-GTGGGATACGAGC
CAATTCGAAAG-39 and 59-TCCCACCTCCAGCATCAAACACTT-39.
sunless tanning in redhead/fair-skinned mice. (A) C57bl6e/e;K14-
SCF and C57bl6e/emice were treated for five consecutive days with
forskolin alone, rolipram alone, Ro 20-1724 alone, or the combina-
tions of forskolin/rolipram or forskolin/Ro 20-1724, once per day.
A representative group of mice out of three independent groups re-
ceiving the same treatment was photographed at day 5. (B) Fontana-
Masson-stained skin sections (for melanin) (black deposits, 633
PDE4 inhibition synergizes with forskolin to induce
MITF regulates PDE4D3 in melanocytes
GENES & DEVELOPMENT 2279
cDNA were obtained using total mRNA for primary human melanocytes
and sk-n-sh using SuperScript III Reverse Transcriptase (Invitrogen)
according to the manufacturer’s recommendations. Human heart quick-
clone cDNAwas purchased from Clontech. For primer sequences, refer to
the primer table in the Supplemental Material.
Primary human melanocytes were starved for 14 h before being treated as
specified in each experiment. RNA was isolated using RNeasy plus
minikits from Qiagen, and was subjected to one-step RT–PCR using
QuantiTect Probe RT–PCR kits from Qiagen and iQ Sybr-Green Supermix
(Bio-Rad). For each reaction, 20 ng of RNA was subjected to the following
steps: reverse transcription for 30 min at 48°C, hot-start Taq activation for
8.5 min at 95°C, and 40 cycles of PCR reaction for 15 sec at 95°C and for 30
sec at 62°C. The results are the average of three independent experiments.
For primer sequences, refer to the primer table in the Supplemental
Material (Supplemental Table 2).
Gel electrophoresis and immunoblotting
Primary human melanocytes were cultured in six-well dishes and exposed
to different agents for the times indicated in each figure legend. Sub-
sequently, the cells were lysed in buffer containing 50 mM Tris (pH 7.4),
150 mM NaCl, 1% Triton X-100, 10 mM leupeptin, 1 mM AEBSF, 100 U/
mL aprotinin, 10 mM NaF, and 1 mM Na3VO4. Samples (30 mg) were
resolved by 10% SDS-PAGE, transferred to nitrocellulose membranes, and
then exposed to the appropriate antibodies: anti-phospho-CREB (Cell
Signaling Technologies), anti-tubulin (Sigma Aldrich), and anti-MITF
(C5). Proteins were visualized with the ECL system from Perkin Elmer
Life Sciences using horseradish peroxidase-conjugated anti-rabbit or anti-
mouse secondary antibody. Western blot assays shown are representative
of at least three experiments.
Construction of pPDE4D3 reporter and mutagenesis
A 1.8-kb fragment upstream of the translation start site of PDE4D3 was
amplified from BAC clone RP11-A7, purchased from Children’s Hospital
Oakland Research Institute using primers 59-CCCAAGCTTGGTCATCTG
CAGCTAAATGGTTAC-39 and 59-CATGCCATGGTCGCAGATCTTCT
GTCATTAATA-39. The fragment was digested with HindIII and NcoI and
inserted into pGL4.11 (Promega). Site-directed mutagenesis was performed
using the QuickChange method from Stratagene according to the supplier’s
recommendations. The primers usedwere 59-TCCAAGGCAAAATATGAA
AAGCTCCGAGATGGTTTTGATAATAACAAAATAAAG-39 and 59- CTT
Transfection and dual luciferase reporter assay
UACC62 melanoma cells were seeded onto 24-well dishes and transfected
the following day in 1.5 mL of Lipofectamine 2000 (Invitrogen) with 0.5 mg
each of pGL4.11, pPDE4D3, and pPDE4D3mut. Test plasmids were
transfected with pGL2cmv renilla luciferase reporter to control for the
variability in transfection efficiency. After 24 h of transfection, the cells
were exposed to 20 mM fsk for 24 h. Cell lysates were prepared 48 h later,
and the activity of firefly and renilla luciferase was measured using
the Dual Luciferase kit (Promega) according to the manufacturer’s
Primary human melanocytes were seeded in six-well dishes transfected
three times with HiPerFect (Qiagen) according to the manufacturer’s
protocols and100 pmolof double-stranded siRNA per well(0.5 3106cells)
at 24-h intervals. Twelve hours after the last transfection, the cells were
placed in minimal media for 14 h and treated as indicated in the figures.
Nontargeting siRNA (Silencer Negative Control; Ambion) used were si#1,
MITF, 59-GGCUUUCUAGAAAGAAUAA-39 (Ambion); and si#2, MITF,
59-GGUGAAUCGGAUCAUCAAG-39 (Carreira et al. 2005) (Ambion).
si PDE4D3 pool 1 was a mix of 59-CACGAUAGCUGCUCAAACA-39
(Qiagen) and 59-UAACGUAGGAGACAAGAAA-39 (Qiagen), and si
PDE4D3 pool 2 was a mix of 59-UGAUGCACGUGAAUAAUUU-39
(Qiagen) and 59-GAGUUGGAAUUCAUCUGUA-39 (Qiagen) (Lynch
et al. 2005).
C57BL6e/eK14SCF and C57BL6e/emice were described previously
(D’Orazio et al. 2006) . Animals between 5 and 10 wk of age were depilated
in the dorsal area using bee’s wax. Crude extracts of ground-up Plec-
tranthus barbatus (forskolin) (ATZ natural) and rolipram and Ro20-1724
(Sigma-Aldrich) were dissolved in 70% ethanol and 30% propylene glycol
at a final concentration of 98 mM for forskolin and 18 mM for rolipram
and Ro20-1724. Eighty microliters of the different drugs or combinations
was applied to the skin once a day for 5 d. Skin reflective colorimetry
measurements were assessed with a CR-400 Colorimeter (Minolta Cor-
poration). The instrument was calibrated against the white standard
background provided by the manufacturer before use. Degree of melani-
zation (darkness) is described as the colorimetric measurement on the
*L axis (white–black axis) of the Center Internationale d’Eclairage (CIE)
L*a*b* color system (Park et al. 1999). The measurement is based on the
ability of colors to reflect light. The instrument emits flashes and
quantitatively measures reflected light. The results range from 0 to 100,
where 0 represents no reflected light (black) and 100 represents maximal
reflected light (white).
Animals were euthanized by CO2 inhalation prior to skin sampling.
Dorsal skin biopsies were kept in 10% buffered formalin until paraffin
embedding and sectioning. Fontana-Masson staining was performed
according to routine procedures.
P-values were calculated with a paired, two-sided t-test. P-values were
noted in each statistical analysis when P was <0.05.
We thank Andre Rosowsky, Kathleen C. Robinson, and Su-Jean Seo for
critical comments on the manuscript, and Scott Granter, Hans Widlund,
Akinori Kawakami, Rosa Veguilla, and Satoru Yokoyama for discussions
and help with technical aspects of the study. We also thank Riham
Carden and Abraham Cooper, who participated in early portions of the
project. We thank Takahiro Kunisada for providing the C57BL6 K14SCF
mouse. M.K. acknowledges the Philippe Foundation for their support.
D.E.F. gratefully acknowledges support for this research from the Na-
tional Institutes of Health (NIAMS AR043369-15 and 1AR058469), the
Melanoma Research Alliance, the US/Israel Binational Science Founda-
tion, The Adelson Medical Research Foundation, and the Doris Duke
Graphical representation of the cAMP–MITF–PDE4D3
Khaled et al.
2280GENES & DEVELOPMENT
Bentley NJ, Eisen T, Goding CR. 1994. Melanocyte-specific expression of
the human tyrosinase promoter: Activation by the microphthalmia
gene product and role of the initiator. Mol Cell Biol 14: 7996–8006.
Bertolotto C, Abbe P, Hemesath TJ, Bille K, Fisher DE, Ortonne JP,
Ballotti R. 1998. Microphthalmia gene product as a signal transducer
in cAMP-induced differentiation of melanocytes. J Cell Biol 142: 827–
Bobon D, Breulet M, Gerard-Vandenhove MA, Guiot-Goffioul F, Plomteux
G, Sastre-y-Hernandez M, Schratzer M, Troisfontaines B, von Frenckell
R, Wachtel H. 1988. Is phosphodiesterase inhibition a new mechanism
of antidepressant action? A double blind double-dummy study be-
tween rolipram and desipramine in hospitalized major and/or endog-
enous depressives. Eur Arch Psychiatry Neurol Sci 238: 2–6.
Busca R, Abbe P, Mantoux F, Aberdam E, Peyssonnaux C, Eychene A,
Ortonne JP, Ballotti R. 2000. Ras mediates the cAMP-dependent
activation of extracellular signal-regulated kinases (ERKs) in melano-
cytes. EMBO J 19: 2900–2910.
Carreira S, Goodall J, Aksan I, La Rocca SA, Galibert MD, Denat L, Larue
L, Goding CR. 2005. Mitf cooperates with Rb1 and activates p21Cip1
expression to regulate cell cycle progression. Nature 433: 764–769.
Conti M, Beavo J. 2007. Biochemistry and physiology of cyclic nucleotide
phosphodiesterases: Essential components in cyclic nucleotide sig-
naling. Annu Rev Biochem 76: 481–511.
D’Orazio JA, Nobuhisa T, Cui R, Arya M, Spry M, Wakamatsu K, Igras V,
Kunisada T, Granter SR, Nishimura EK, et al. 2006. Topical drug
rescue strategy and skin protection based on the role of Mc1r in
UV-induced tanning. Nature 443: 340–344.
D’Sa C, Tolbert LM, Conti M, Duman RS. 2002. Regulation of cAMP-
specific phosphodiesterases type 4B and 4D (PDE4) splice variants by
cAMP signaling in primary cortical neurons. J Neurochem 81: 745–757.
Du J, Widlund HR, Horstmann MA, Ramaswamy S, Ross K, Huber WE,
Nishimura EK, Golub TR, Fisher DE. 2004. Critical role of CDK2 for
melanoma growth linked to its melanocyte-specific transcriptional
regulation by MITF. Cancer Cell 6: 565–576.
Fitzpatrick TB. 1988. The validity and practicality of sun-reactive skin
types I through VI. Arch Dermatol 124: 869–871.
Fitzpatrick TB, Sober AJ. 1985. Sunlight and skin cancer. N Engl J Med
Garraway LA, Widlund HR, Rubin MA, Getz G, Berger AJ, Ramaswamy
S, Beroukhim R, Milner DA, Granter SR, Du J, et al. 2005. Integrative
genomic analyses identify MITF as a lineage survival oncogene
amplified in malignant melanoma. Nature 436: 117–122.
Griswold DE, Webb EF, Breton J, White JR, Marshall PJ, Torphy TJ. 1993.
Effect of selective phosphodiesterase type IV inhibitor, rolipram, on
fluid and cellular phases of inflammatory response. Inflammation 17:
Hoek KS, Schlegel NC, Eichhoff OM, Widmer DS, Praetorius C, Einarsson
SO, Valgeirsdottir S, Bergsteinsdottir K, Schepsky A, Dummer R, et al.
2008. Novel MITF targets identified using a two-step DNA microarray
strategy. Pigment Cell Melanoma Res 21: 665–676.
Hoffmann R, Baillie GS, MacKenzie SJ, Yarwood SJ, Houslay MD. 1999.
The MAP kinase ERK2 inhibits the cyclic AMP-specific phosphodi-
esterase HSPDE4D3 by phosphorylating it at Ser579. EMBO J 18:
Houslay MD, Baillie GS, Maurice DH. 2007. cAMP-specific phosphodi-
esterase-4 enzymes in the cardiovascular system: A molecular tool-
box for generating compartmentalized cAMP signaling. Circ Res 100:
Kunisada T, Lu SZ, Yoshida H, Nishikawa S, Mizoguchi M, Hayashi S,
Tyrrell L, Williams DA, Wang X, Longley BJ. 1998. Murine cutaneous
mastocytosis and epidermal melanocytosis induced by keratinocyte
expression of transgenic stem cell factor. J Exp Med 187: 1565–1573.
Le Pape E, Wakamatsu K, Ito S, Wolber R, Hearing VJ. 2008. Regulation of
eumelanin/pheomelanin synthesis and visible pigmentation in mela-
nocytes by ligands of the melanocortin 1 receptor. Pigment Cell
Melanoma Res 21: 477–486.
Lerner AB, McGuire JS. 1961. Effect of a- and b-melanocyte stimulating
hormones on the skin colour of man. Nature 189: 176–179.
Levy C, Khaled M, Fisher DE. 2006. MITF: Master regulator of melano-
cyte development and melanoma oncogene. Trends Mol Med 12: 406–
Lin JY, Fisher DE. 2007. Melanocyte biology and skin pigmentation.
Nature 445: 843–850.
Lynch MJ, Baillie GS, Mohamed A, Li X, Maisonneuve C, Klussmann E,
van Heeke G, Houslay MD. 2005. RNA silencing identifies PDE4D5
as the functionally relevant cAMP phosphodiesterase interacting
with barrestin to control the protein kinase A/AKAP79-mediated
switching of the b2-adrenergic receptor to activation of ERK in
HEK293B2 cells. J Biol Chem 280: 33178–33189.
MacKenzie SJ, Baillie GS, McPhee I, MacKenzie C, Seamons R, McSorley
T, Millen J, Beard MB, van Heeke G, Houslay MD. 2002. Long PDE4
cAMP specific phosphodiesterases are activated by protein kinase
A-mediated phosphorylation of a single serine residue in Upstream
Conserved Region 1 (UCR1). Br J Pharmacol 136: 421–433.
Omori K, Kotera J. 2007. Overview of PDEs and their regulation. Circ Res
Park SB, Suh DH, Youn JI. 1999. A long-term time course of colorimetric
evaluation of ultraviolet light-induced skin reactions. Clin Exp
Dermatol 24: 315–320.
Price ER, Horstmann MA, Wells AG, Weilbaecher KN, Takemoto CM,
Landis MW, Fisher DE. 1998. a-Melanocyte-stimulating hormone
signaling regulates expression of microphthalmia, a gene deficient in
Waardenburg syndrome. J Biol Chem 273: 33042–33047.
Richter W, Jin SL, Conti M. 2005. Splice variants of the cyclic nucleotide
phosphodiesterase PDE4D are differentially expressed and regulated
in rat tissue. Biochem J 388: 803–811.
Rijken F, Bruijnzeel PL, van Weelden H, Kiekens RC. 2004. Responses of
black and white skin to solar-simulating radiation: Differences in
DNA photodamage, infiltrating neutrophils, proteolytic enzymes
induced, keratinocyte activation, and IL-10 expression. J Invest
Dermatol 122: 1448–1455.
Sette C, Conti M. 1996. Phosphorylation and activation of a cAMP-
specific phosphodiesterase by the cAMP-dependent protein kinase.
Involvement of serine 54 in the enzyme activation. J Biol Chem 271:
Sturm RA, Duffy DL, Box NF, Chen W, Smit DJ, Brown DL, Stow JL,
Leonard JH, Martin NG. 2003. The role of melanocortin-1 receptor
polymorphism in skin cancer risk phenotypes. Pigment Cell Res 16:
Yasumoto K, Yokoyama K, Shibata K, Tomita Y, Shibahara S. 1994.
Microphthalmia-associated transcription factor as a regulator for
melanocyte-specific transcription of the human tyrosinase gene.
Mol Cell Biol 14: 8058–8070.
MITF regulates PDE4D3 in melanocytes
GENES & DEVELOPMENT2281