Download Immunomodulatory effects of phytocompounds characterized by in

J Biomed Sci (2008) 15:813–822
DOI 10.1007/s11373-008-9266-7
ORIGINAL PAPER
Immunomodulatory effects of phytocompounds characterized
by in vivo transgenic human GM-CSF promoter activity
in skin tissues
Pei-Fen Su Æ Vanisree Staniforth Æ Chin-Jin Li Æ
Chien-Yu Wang Æ Ming-Tsang Chiao Æ Sheng-Yang Wang Æ
Lie-Fen Shyur Æ Ning-Sun Yang
Received: 28 March 2008 / Accepted: 22 June 2008 / Published online: 13 July 2008
Ó National Science Council Taipei 2008
Abstract To investigate the immunomodulatory activities of phytocompounds for potential therapeutics, we
devised an in vivo, transgenic, human cytokine gene promoter assay using defined epidermal skin cells as test
tissue. Test compounds were topically applied to mouse
skin before or after gene gun transfection, using a cytokine
gene promoter-driven luciferase reporter. Croton oil, an
inflammation inducer, induced transgenic GM-CSF and
TNF-a promoter activities in skin epidermis 6-fold and 3.4fold, respectively; however, it produced a less than 1.5-fold
and 1.7-fold change in IL-1b and IL-18 promoter activity,
respectively. The phytocompound shikonin drastically
inhibited inducible GM-CSF promoter activity. However, a
fraction of Dioscorea batatas extract significantly
increased the GM-CSF promoter activity in normal and
inflamed skin. Shikonin suppressed the transcriptional
activity of GM-CSF promoter by inhibiting the binding of
TFIID protein complex (TBP) to TATA box. Our results
demonstrate that this in vivo transgenic promoter activity
assay system is cytokine gene-specific, and highly
responsive to pro-inflammatory or anti-inflammatory
stimuli. Currently it is difficult to profile the expression and
cross-talk of various types of cytokines in vivo. This
P.-F. Su V. Staniforth C.-J. Li C.-Y. Wang M.-T. Chiao L.-F. Shyur N.-S. Yang (&)
Agricultural Biotechnology Research Center, Academia Sinica,
Nankang, Taipei 11529, Taiwan, ROC
e-mail: [email protected]
P.-F. Su
Institute of Biomedical Sciences, Academia Sinica, Taipei,
Taiwan, ROC
S.-Y. Wang
Department of Forestry, National Chung-Hsing University,
Taichung, Taiwan, ROC
investigation has established a bona fide in vivo, in situ,
immune tissue system for research into cytokine response
to inflammation.
Keywords Dioscorea batatas Immune-modulating
activities In vivo assay Human granulocyte
macrophage-colony stimulating factor Phytocompounds Shikonin
Introduction
Cytokines, the key regulators of the immune system, have
been studied intensively as a class of powerful immunomodulators for clinical applications. However, the adverse
side effects, high cost, and labile features of cytokine
proteins often prohibit their routine use. Efforts are therefore underway to develop alternative immunomodulators
as therapies. One source of alternative immunomodulators
may be traditional herbs or their derived phytocompounds
reputed to confer medicinal efficacy [1]. The historically
successful development of medicinal compounds from
plants, such as aspirin and taxol, has led to research into the
potential of a large range of herbs and their derived phytocompounds as a source of immunomodulators [2, 3].
In drug discovery research, in vitro cell-based screening
systems are well established as methods for evaluation of
candidate lead compounds. For example, in vitro assays of
NF-jB [4, 5] and COX-2 [6], two examples of drug targets,
are employed to develop therapeutic strategies to counter
inflammation. However, it is also known that the regulation
of immune-modifiers and their gene expression is highly
dependent upon three dimensional microenvironments.
Therefore, an in vivo assay that can accurately evaluate the
effects of immune modulators/drugs on the expression
123
814
J Biomed Sci (2008) 15:813–822
level of key cytokines in targeted tissue is highly desirable
for applications in medical biotechnology. Keratinocytes
are known as the major source of cytokines in skin tissues
[7, 8], and these cells have also been shown to be the skin
cell type most susceptible to gene transfection with a gene
gun [9]. The purpose of this study was to establish the use
of promoter activities from a panel of cytokine genes as the
basis for a reporter system that can characterize the transcriptional control of cytokine gene(s) responsive to
topically applied pro- or/and anti-inflammatory
phytocompounds.
GM-CSF (granulocyte macrophage-colony stimulating
factor) is a cytokine known for its multi-functional effects
on various immune cell systems. Recombinant human GMCSF is used therapeutically for enhancing hematopoietic
recovery [10] and anti-infection activities [11]. Recently,
GM-CSF has also been actively evaluated as an immune
adjuvant to enhance maturation and function of dendritic
cells and to augment macrophage activities [12]. GM-CSF
is also applied as a vaccine adjuvant in various immunotherapy approaches, including cancer treatment [13–16].
However, dysregulated GM-CSF has been shown to be
associated with chronic inflammatory diseases such as
cutaneous dermatitis [17, 18] and skin cancer [19]; and has
thus been suggested as a therapeutic target for multiple
sclerosis [20]. Therefore, human GM-CSF was chosen as
the key cytokine for this study.
A number of candidate immunomodulatory agents,
including medicinal herb extracts of traditional Chinese
medicine and Western remedies, purified phytocompounds,
and anti-inflammatory prescription drugs were quantitatively analyzed for pro- or anti-inflammatory activities in
this study. Since different clusters of specific cytokine
genes are related to different types of skin inflammation
[21], to evaluate various cytokine gene responses to
inflammatory inducers, transgenic promoter activities of
GM-CSF, TNF-a, IL-1b and IL-18 gene were assayed in
this system. Correlations between inflammatory responses,
specific cytokine gene promoter activities, and inhibition or
Table 1 Primer sequences and
cloning sites of cytokine
promoters amplified by PCR
enhancement of these activities by specific phytocompounds or herbal extracts were elucidated. Information
from this study provides a quantifiable and quick in vivo
assay system for validating and characterizing potential
immunomodulators (e.g., phytocompounds) as candidate
therapeutics.
Materials and methods
Plasmids
Human genomic DNA was isolated from lymphocytes
using a Genomic DNA purification kit (Promega, Madison,
WI) and was then used as a template for amplifying the
promoter region of target cytokine genes by an expanded
long template PCR system (Roche, Mannhein, Germany).
The sequences of target promoter regions of cytokine genes
were obtained from either published results or the NCBI
database. For GM-CSF, a proximal promoter fragment of
620-bp (-620 to -1) and another fragment of 3.3-kb (3286 to -1) including the distal enhancer region were
PCR-amplified and individually cloned into a pGL-3 basic
(Promega), which is a luciferase reporter vector. The PCRamplified promoter sequences of IL-1b (881 bp, -853 to
+28), IL-18 (909 bp, -738 to +171), and TNF-a
(1097 bp, -1049 to +48) were also individually cloned
into pGL3-basic vector as depicted in Table 1. The proximal promoter motif of GM-CSF was also cloned into the
pb-Gal basic vector (Clontech, Palo Alto, CA) to generate
a pGM620-b-gal plasmid with a b-galactosidase reporter
gene driven by GM-CSF promoter.
Mice
Female BALB/c Byj mice aged 8–12 weeks, were purchased from the National Laboratory Animal Breeding and
Research Center, Taipei, Taiwan and maintained under
standard pathogen-free, animal room conditions.
Promoter of gene Primer sequences (50 ? 30 )
IL-1b
CGGGGTACCCTGGAGCAGGCACTTTGCTGGTGTC*a
Cloning site References
KpnI/BglII
[35]
SacI/NheI
[37]
GAAGATCTGTGCCTTGTGCCTCGAAGAGGTTTG*b
IL-18
AGTCTCGAGCTCTTTGCTATCATTCCAGGAA*a
[36]
GTACTGCTAGCTTTCCTCTTCCCGAAGCTG*b
TNF-a
CGGACGCGTGAAGCAAAGGAGAAGCTGAGAAGA*a MluI/BglII
CGGAGATCTCTCTTAGCTGGTCCTCTGCTGTC*b
*underline: restriction enzyme
recognition sequence
a
Sense primer sequence
b
Anti-sense primer sequence
123
L11698
[38]
GM-CSF
ACGCGTGAGAGTGGCTGCAGTCTCGCTGCTG*a
MluI/XhoI
AJ224149
(620 bp)
GM-CSF
CTCGAGGGGCTCACTGGCAAAAGAGCTCTTA*b
ACGCGTAGATCTCAGGTCCCCAGAGATAACA*a
MluI/XhoI
[39]
AJ224149
(3286 bp)
CTCGAGGGGCTCACTGGCAAAAGAGCTCTTA*b
[39]
J Biomed Sci (2008) 15:813–822
815
Herbal extracts preparation and reagents
Luciferase activity assay
All the medicinal herbs used in this study have been reported,
reputed or claimed to confer immunomodulatory activities in
traditional or folk medicine cultures. The tubers of the Dioscorea batatas, a traditional Chinese medicinal plant, were air
dried, ground into powder, and extracted by water followed by
50% ethanol. The resultant pellet was designated as D. batatas
fraction I (DsCE-I) and supernatant was again extracted by
75% ethanol to obtain DsCE-II fraction. Air dried aerial parts
of Bidens pilosa were ground into powder and extracted by
hot-water, the solvent was removed in a vacuum and the
resultant material was used as a crude extract. Total crude
extract of Echinacea purpurea whole plants was prepared
using 70% ethanol (v/v) extraction. DsCE-I and DsCE-II were
normalized by their sugar (mannose) contents (unpublished
results and communication from Dr. Shih-Hsiung Wu, Institute of Biological Chemistry, Academia Sinica, Taiwan).
Plant extracts from B. pilosa and E. purpurea were defined by
their specific metabolite profile or index compounds determined using HPLC and various spectral analyses [22, 23]. The
normalized and standardized extract preparations were
employed for routine promoter activity assays designed for
this study. Shikonin, an anti-inflammatory phytocompound of
Lithospermum erythrorhizon was obtained commercially
(TCI, Tokyo, Japan). The RubiDerm cream (Marco Polo
Technologies, MD) was used as a commercially available
paste containing plant extracts including shikonin. Croton oil,
an inflammation inducer, was purchased from Fluka, WI; and
Getason cream containing Gentamycin and b-methasone was
obtained from a local pharmacy.
Transfected skin tissues were collected by excision and
frozen in liquid nitrogen. Immediately prior to performing
the luciferase assay, frozen tissue samples were thawed and
scissor minced in 0.5 ml lysis buffer (1% Triton X-100 and
a protease inhibitor cocktail in PBS) [9]. Skin tissue lysates
were centrifuged at 15,000 rpm for 10 min at 4°C, and the
supernatant collected. Transgenic luciferase activity in cell
lysate was determined using a luminometer assay (Promega). The levels of luminescent intensities as relative light
units (RLU) were translated into picograms of luciferase
protein per transfected tissue site, using a standard curve for
pure luciferase enzyme assayed in parallel. Results shown
represent the mean of three independent transfections.
In vivo particle-mediated gene transfer
Electrophoretic mobility shift assay (EMSA)
The Helios gene gun system (Bio-Rad, Hercules, CA) was
employed to deliver plasmid DNA into mouse skin per
manufacturer’s instructions. In brief, plasmid DNA was
precipitated onto gold particles of 2 lm in size (Degussa
Corporation, Plainfield, NJ) at a loading rate of 2.5 lg
DNA/mg gold particles, and coated onto the inner surface
of Tefzel tubing. Each half-inch segment of tubing conferred the delivery of 0.5 mg of gold and 625 ng of plasmid
DNA/transfection. Clippers and a mild depilatory treatment
were used to remove animal hair from the abdominal area
of test mice one day prior to gene-gun bombardment.
Mouse skin (3.8 cm2/site) was transfected with test promoter/reporter gene constructs by using a discharge
pressure of 350 psi helium. The commercially available
drugs, herbal creams, or test herbal extracts were topically
applied onto mouse skin tissue as indicated. Normal skin
(negative control, without any treatment) and test gene
transfected skin treated with or without solvent were used
as controls.
Nuclear protein from mouse skin was isolated as described
previously [25]. EMSA was performed using a LightShift
Chemiluminescent EMSA kit (Pierce, Rockford, IL)
according to the manufacturer’s protocol. A biotin-labeled
30 bp double-stranded oligonucleotide corresponding to
the -32 to -3 nucleotide core promoter element of human
GM-CSF promoter, 50 - CCTCTGTGTATTTAAGAGCTC
TTTTGCCAG-30 was used as a probe to study the binding
of nuclear proteins from untreated and shikonin treated
skin samples.
X-Gal assay
Histochemical staining of b-galactosidase (b-gal) was
performed as previously described [9, 24] with some
modifications. X-gal substrate was prepared at a final
concentration of 40 mg/ml in DMF stock solution. Working solution was made by adding stock X-gal substrate to
assay buffer (44 mM HEPES, 3 mM K-ferricyanide,
3 mM K+ ferrocyanide, 15 mM NaCl and 1.3 mM MgCl2,
pH 7.4) to make a final 1 mg/ml solution. The excised
mouse skin specimen was stretched out on the surface of a
wax layer in a 60-mm dish using needles. Six milliliters of
X-gal buffer solution was added to each dish and incubated
for 2 h at 37°C. The skin tissues were observed under
dissection microscopy.
Results
Establishing the in vivo hGM-CSF gene
promoter-based assay system
In order to develop a system that can be used to evaluate
the skin immune responses to topically applied agents,
123
816
123
Luciferase protein level (pg/site)
A
1200
GM620 plasmid
1000
800
600
400
200
0
1st
2nd
None
Ace
Cro
GM-CSF GM-CSF GM-CSF
GM-CSF GM-CSF
Ace
Cro
Treatment
B
Relative luciferase activity (RLA)
pGM620 plasmid containing proximal human GM-CSF
promoter fragment (620-bp) was gene gun-transfected into
mouse abdominal skin. The basal transgenic promoter
activity of pGM620 was 20.0 ± 10.5 pg luciferase protein
per blasted skin tissue site (Fig. 1a); in comparison negligible luciferase activity was detected in skin tissue blasted
with the promoter-less pGL-3 negative vector (data not
shown). To compare the transcriptional activity of the
hGM-CSF promoter in pre-inflamed skin or recently
inflamed skin, croton oil (2%) was applied to test skin
tissue 4 h before or immediately after pGM620 DNA was
gene gun-transfected. Skin samples were harvested 8 h
post transfection and luciferase activity assayed. Both pretreatment and post-treatment with croton oil enhanced the
transcriptional activity of transgenic pGM620 promoter
(5.1-fold and 6.4-fold, respectively) compared to solvent
(acetone) treatment (Fig. 1a). Different concentrations of
croton oil were then tested to find the optimum conditions
for inducing skin inflammation. The pGM620 promoter
activities increased 3.3-, 5.1- and 5.0-fold over the solvent
control when 0.4%, 2% and 10% croton oil were applied
after transfection, respectively. Since similar transgenic
luciferase protein levels were induced by 2% and 10%
croton oil, skin inflammation was induced by 2% croton oil
for 8 h in the following experiments.
Previous studies on human GM-CSF promoter show that
a distal enhancer element may play an important role in
promoter activity regulation [26]. Therefore, we constructed pGM3286 plasmid containing GM-CSF promoter
that included the enhancer element. pGM620 and
pGM3286 plasmid constructs were analyzed in parallel to
evaluate the promoter activities under normal and inflamed
conditions. The two plasmids were individually transfected
into mouse skin and treated with croton oil or solvent
control. Transgenic luciferase proteins obtained from the
basal level promoter activities of pGM3286 and pGM620
vectors were 2.7 ± 1.0 RLA (relative luciferase activity in
arbitrary units) and 7.9 ± 2.1 RLA per blasted skin site,
respectively, indicating that the pGM620 promoter had a
2.9-fold higher basal level of promoter activity than the
pGM3286 (Fig. 1b, open bar). Induction of inflammation
by croton oil effectively enhanced pGM3286 and pGM620
promoter activities by 13.8-fold and 8.0-fold, respectively,
over the non-inflamed control (Fig. 1b, filled bar). Similar
to promoter activity levels observed for normal skin, a
higher level (1.7-fold) of pGM620 promoter activity than
pGM3286 activity was obtained in the inflamed skin tissue.
Although the 3.3-kb promoter fragment contained a distal
enhancer element, we show here that the promoter activity
of pGM620 was higher than pGM3286 in both normal and
inflamed skin conditions. We therefore suggest that a cissilencing element may exist in the region between -620
and -3286 nucleotide that can down regulate the
J Biomed Sci (2008) 15:813–822
80
Normal Skin
Normal
Skin
Inflamed Skin
Inflamed
Skin
60
40
20
0
pGL-3
GM3286
GM620
Fig. 1 Establishment of an in vivo, inflammation-responsive, transgenic hGM-CSF promoter activity assay. (a) Determination of basal level
and inducible transcriptional activity of human GM-CSF gene promoter
in mouse skin. pGM620 plasmid was transfected into BALB/c mouse
abdominal skin by bombardment with a gene gun, and inflammation was
induced by topical application of 2% croton oil 4 h before or immediately
after bombardment. The treatments were None/GM-CSF: no treatment of
blasted skin; Ace/GM-CSF and Cro/GM-CSF: skin treated with acetone
(Ace), a solvent base for croton oil (Cro) before gene bombardment; GMCSF/Ace and GM-CSF/Cro: skin bombarded before acetone or croton oil
treatment. Bombarded and treated skin tissues were harvested, extracted
and assayed for luciferase activity 8 h after bombardment. (b) Comparison of promoter activities of the extended length (3,286 bp) and the
proximal region (620 bp) of the hGM-CSF promoter. The pGL-3-basic,
pGM3286 and pGM620 expression plasmids were particle-delivered into
test mouse skin, transfected skin tissue sites were then immediately
treated with or without 2% croton oil, and skin tissues collected at 16 h
post-bombardment and assayed for luciferase activity. Relative luciferase activities (RLA, in arbitrary units) were obtained as comparable
luciferase equivalents using standardized, normalized Luminometer
assays. Transgenic luciferase activities obtained for 8 h versus 16 h postbombardment gave similar levels of activities. Data represents the
mean ± sd of triplicate skin tissue samples. Results are from one
representative experiment; two other independent experiments showed
similar results
transcriptional activity of pGM3286. Because of the
stronger basal level and inflammation-inducible promoter
activity of pGM620, we employed this construct in our
J Biomed Sci (2008) 15:813–822
assay system to evaluate the enhancing or inhibitory effects
of phytocompounds on GM-CSF promoter activity.
In situ analysis of hGM-CSF promoter activity
in croton oil-inflamed skin
Our results demonstrated that transgenic hGM-CSF promoter activity could be effectively induced in croton oilinflamed skin tissue. We then addressed the issue of
whether croton oil treatment had resulted in an increase in
transfection efficiency (e.g., an increase in the cell number of transgene-expressing cells), or whether it had
directly enhanced the level of transgene expression in
each transfected cell. To distinguish between these two
possibilities, X-gal histochemical staining of transgenic bgalactosidase activity were performed at the tissue and
cellular level of test skins. The pGM620-b-gal plasmid
and pb-Gal basic control vectors were gene gun-delivered
into mouse skin, and the skin was then inflamed using
croton oil. Treated skin tissues were collected at 8 h postgene gun bombardment and assayed by X-gal staining. No
Fig. 2 X-gal histochemical analysis of the effect of inflammation on
the transcriptional activity of transgenic hGM-CSF promoter. The pbgal basic and pGM620-b-gal plasmids were individually transfected
into BALB/c mouse skin using a gene gun. The blasted skin was then
treated with or without 2% croton oil, as inflamed or normal skin
tissues, respectively. Eight hours post-bombardment the b-galactosidase (b-gal) expressing cells in test skin tissues were examined by X-
817
blue-stained positive colonies were observed in skin
blasted by the promoter-less pb-gal basic vector (data not
shown). In contrast, b-gal expression driven by pGM620
in transfected skin was enhanced as a bluish ring structure
at the tissue level in the inflamed skin, as compared to the
normal skin (Fig. 2, 109 magnification, panel A vs. B).
At the cellular level (Fig. 2, 309 magnification, panel C
vs. D), a higher X-gal staining intensity was observed for
individual cells in the inflamed skin tissue than in the
normal tissue. By scoring numbers of individual blue
color cells per unit skin area in Fig. 2c and d,
389.7 ± 24.0 cells and 382.0 ± 32.9 cells were observed
as b-gal positive cells, respectively. Similar results were
observed in cell-scoring from other (n = 5) microscopic
fields. Therefore, very similar or equal numbers of skin
cells were transfected by gene gun bombardment in normal (non-croton oil treated) and inflamed skin tissues.
These results suggest that the action of skin inflammation
enhanced the transcriptional activity of the transgenic
GM-CSF promoter, rather than increased the efficiency of
transgene delivery into targeted mouse epidermal cells.
gal staining. Photomicrographs show b-gal expression for pGM620b-gal in normal skin tissues (a and c) and inflamed skin tissues (b and
d). No blue-stained cells were observed in skin tissues that were
blasted with pb-gal basic (vector only) plasmid (data not shown).
Results are representative of one experiment, other independent
experiments showed similar results
123
818
Evaluation of transcriptional activities of other
pro-inflammatory cytokine gene promoters in skin tissue
Evidence has shown that dermatitic inflammatory reactions
are associated with the activation of a group of specific
cytokine genes [21, 27]. In order to test the response of
other pro-inflammatory cytokine gene promoters in skin
tissues, three human cytokine gene (IL-1b, IL-18, and
TNF-a) promoters were PCR-amplified and cloned into
pGL-3 basic vector. These plasmid constructs were
123
Fold change in luciferase activity
In order to evaluate the use of the present in vivo transgenic
promoter activity assay for assessing immunomodulatory
activities, plant extracts from Bidens pilosa, E. purpurea and
D. batatas, were topically applied (at 10 lg/10 ll per tissue
site) to pGM620-blasted skin tissue. DsCE-I, a 50% ethanol
partitioned fraction of the D. batatas aqueous extract,
induced a substantial increase (1.8-fold higher) in pGM620
promoter activity (239.7 ± 19.2 RLA/blasted skin site) than
the water treated vehicle control (133 ± 27.3 RLA/site),
suggesting a modest pro-inflammatory effect. In comparison, no significant enhancement or suppression of promoter
activity was observed with other herbal extracts tested
(Fig. 3a). These results therefore show that the transgenic
hGM-CSF promoter assay in skin could selectively and
specifically respond to different medicinal plant extracts.
To investigate whether the pGM620 promoter assay
would respond to anti-inflammatory agents, the effect of
selected phytocompounds and medicinal plant extracts on
pGM620 promoter activity was examined in croton oilinflamed skin tissues. pGM620 was gene gun transfected
into mouse skin, and test epidermal tissues treated immediately with 2% croton oil for 1 h to induce inflammation.
One hour after inflammation induction, herbal extracts,
single phytocompounds, and the commercial drugs Getason and RubiDerm were applied individually onto inflamed
skin sites. Skin tissues inflamed with croton oil but not
treated with test agents were used as controls. As shown in
Fig. 3b, DsCE-I induced a significant 1.8-fold increase in
transcriptional activity of pGM620 over the control. A
marginal increase (&27%) in promoter activity was also
observed by treatment with B. pilosa plant extract. In
contrast, shikonin significantly suppressed pGMG620
promoter activity to 11.6% of the control value (100%).
RubiDerm and Getason inhibited pGM620 promoter
activity to 67% and 40%, respectively, of the control values. These results suggest that the current assay system
could effectively discriminate between phytocompounds
with potential pro- versus anti-inflammatory bioactivities.
A
2.5
Normal Skin
*
2.0
1.5
1.0
0.5
0.0
W
Di
Di
Ec
Bi
Ds
Ds
os
de
at
hi
CE osc
er
na
co CEns
or
-II
I
(c
ce
re
ea
pi
on
a
a
l
o
(
(w
7
tro
sa
0%
at
l)
er
Et
)
O
H)
B
Fold change in luciferase activity
Effect of medicinal plant extracts and derived
phytocompounds on transcriptional activity of
hGM-CSF promoter in normal or inflamed skin tissues
J Biomed Sci (2008) 15:813–822
2.5
Inflamed Skin
2.0
1.5
1.0
**
0.5
**
0.0
B
S
G
D
Wat
S
er (c aline idens p sCE-I hikonin RubiDe etason
ilosa
rm
ontr
ol)
Pre-treated with 2% croton oil
Fig. 3 Effects of phytocompounds or plant extracts on the transcriptional activity of the hGM-CSF promoter in mouse skin tissue. (a)
Normal skin. Fold-changes in luciferase activity were obtained by
comparing the values of phytocompound-treated test samples over the
water-treated (vehicle control) sample. pGM620 plasmid was transfected into BALB/c skin tissue, and 1 h later water or herbal extracts
prepared in water were applied to blasted skin at 10 lg/10 ll/tissue site.
Test skin tissues were harvested for assay of transgenic luciferase
activity/protein levels at 8 h post-treatment. Water: 10 ll/site; B. pilosa
hot water extract; E. purpurea 70% ethanolic extract; D. batatas 70%
ethanolic extract; DsCE-I and DsCE-II: D. batatas water extract
fraction as described in ‘‘Materials and Methods’’; and D. batatas hot
water extract. Data represent the results from triplicate skin samples.
Another experiment showed similar results. Significant increase is
indicated by * with a p value\0.05. (b) Inflamed skin. Fold change in
luciferase activity is the same as that defined above. pGM620 plasmid
was transfected into BALB/c mouse skin tissue and the skin was
inflamed with 2% croton oil as described in the legend of Fig. 2. An
hour later, test herbal extracts or commercial medicines were applied
onto inflamed or normal (untreated, control) skin. Test skin tissues were
harvested for transgenic luciferase activity assay 8 h post treatment
with test herbal extract or medicine. Test substances and applied doses
were as follows: water 10 ll/site; saline 10 ll/site; B. pilosa, hot water
extract (100 lg/site); DsCE-I: D. batatas water extract fraction I
(100 lg/site); Shikonin: 10 lg/site; RubiDerm (5.4 mg/site); and
Getason: b-methasone (54 lg/site). Data represents the mean ± sd of
triplicate tissue samples. Results are representative of one experiment.
Two other independent experiments showed similar results. Significant
inhibition is indicated by ** with a p value\0.01, Student’s t-test
J Biomed Sci (2008) 15:813–822
819
+Solvent
Inflamed Skin
Inflamed
Skin
(+2% croton
(+2%
crotonoil)oil)
Competition
Probe alone
Fold of Induction
(In luciferase activity)
6
Normal Skin
Normal
Skin
(respective
(respectivecontrols)
controls)
No treatment
pGM3286 transfected skin
8
1
2
3
4
5
Shikonin (µg)
10
100
6
7
4
TFIID
2
0
IL-1β
IL-18
GM-CSF
TNF-α
Fig. 4 Relative transcriptional activity of four different human
cytokine gene promoters in normal and inflamed mouse skin tissues.
Fold of induction of luciferase activity driven by each test promoter
was obtained by comparing the level obtained from the inflamed skin
over the correspondent normal control level. The plasmids containing
transgenic promoter sequences of IL-1b, IL-18, GM-CSF, and TNF-a
genes were individually transfected into normal and inflamed BALB/
c mouse skin. Inflammation was induced by topical treatment with 2%
croton oil immediately after particle bombardment, and skin tissues
collected for assay of transgenic luciferase assay 8 h post-bombardment. Data represents the mean ± sd of triplicate skin samples.
Results are representative of one experiment. Two other independent
experiments showed similar results
individually blasted into mouse skin and left untreated or
immediately treated with 2% croton oil to induce inflammation. Under normal skin conditions, the three tested
promoters revealed a wide range of basal level transcriptional activity, as detected at 8 h post-transfection. The
TNF-a promoter showed the highest level of transgenic
luciferase protein (326.3 ± 23.5 RLA/blasted skin site)
expression. In contrast, the promoters of IL-1b and IL-18
conferred 46.9 ± 10.7 RLA and 32.3 ± 8.0 RLA of
transgenic luciferase proteins per blasted skin site,
respectively. Upon inflammation of skin tissue, the promoter activities of TNF-a, IL-1band IL-18 increased 3.4-,
1.5- and 1.7-fold, respectively (Fig. 4). TNF-a promoter
was also sensitive to croton oil-induced inflammation in a
similar way to GM-CSF promoter (5 to 6-fold increase)
(Fig. 4). TNF-a promoter activity was routinely found to be
increased 3 to 4-fold (data not shown). Our data suggests
that the current transgenic assay could be effectively
employed to profile various cytokine promoter activities
in vivo and in situ in response to inflammation.
Molecular basis for shikonin inhibition of GM-CSF
promoter activity in skin tissue
Our results show that shikonin can significantly suppress
pGM620 promoter activity caused by croton-oil induced
inflammation. We have previously reported that shikonin
suppresses the transcription of TNF-a by inhibiting the
Free probe
Lane
1
Fig. 5 Inhibitory effect of shikonin on binding of TFIID complex
(TBP) to TATA box in the GM-CSF promoter. Mouse skin was gene
gun-transfected with pGM3286 DNA and topically untreated or
treated with solvent alone, or treated with a dose of 1, 10, or 100 lg
per site of shikonin. Skin samples were harvested after 1 h and
nuclear extracts were subjected to EMSA analysis as described in
‘‘Materials and Methods’’. Binding specificity of the complexes (lanes
2–7) was analyzed by competition against the GM-CSF probe using
1000-fold excess of unlabeled DNA probe (lane 3)
binding of TFIID protein complex (TBP) to TATA box [25].
We thus conducted an EMSA study to determine whether
shikonin suppresses GM-CSF promoter activity through
interfering with TBP binding activity. Mouse skin was
transfected with pGM3286 by bombardment, and left alone
(untreated), or treated with solvent alone or shikonin at
concentrations of 1, 10 and 100 lg per site per mouse. Skin
tissue samples were harvested after 1 h and nuclear extracts
were subjected to EMSA by using the 30-bp core promoter
region of human GM-CSF containing the TATA box element
as a DNA probe. In untreated and solvent-treated samples,
TFIID complex binding was clearly detectable (Fig. 5, lanes
2 and 4). Whereas, in shikonin treated samples the binding of
TFIID complex was clearly inhibited in a dose dependent
manner (lanes 5–7). The specificity of this binding was
indicated by the decrease of signal shift seen when a 1000fold excess of unlabeled DNA probe was used to compete out
the GM-CSF DNA probe (lane 3).
Discussion
It is well known that cytokine expression and function is a
complex process. Factors affecting the process include the
123
820
time of release of specific cytokines, the local environment
in which the cytokines act, the presence of antagonistic or
synergetic molecules, the density of cytokine receptors on
cells, and interactions with other cytokines. Consequently,
experimental studies on cytokine biology and related cellular immunology are challenging and sometimes
controversial. To measure the expression level of cytokine
genes in vivo, Northern blot, RT-PCR, ELISA, and
immunohistochemistry are currently employed; however,
these techniques are relatively time-consuming and labor
intensive. This study describes the establishment of an
in vivo, promoter activity-based assay system that can
effectively measure relative transgenic cytokine gene promoter activities in response to pro- or anti-inflammatory
stimuli and to specific phytocompounds. This in vivo gene
assay system may have broad medicinal and biotechnological applications.
As seen in Fig. 2, we have demonstrated here that our
present epidermal keratinocyte system can be effectively
visualized and employed at the cellular and tissue levels for
molecular biology and cell research studies of the immune
system of the skin under in vivo physiological or pathological conditions. Future experiments using confocal
microscopy are also expected to reveal sub-cellular or
organelle structures or features that are involved in various
skin inflammations under in vivo conditions.
Temporal and adjustable boosts to the immune system
are considered to be key therapeutic strategies. GM-CSF
has been considered as an adjuvant strategy to improve the
efficacy of candidate DNA cancer vaccines [16, 28]. The
results of our study showed that croton oil elevates GMCSF promoter activity in transfected skin cells (Fig. 1) and
in in situ b-gal activity-staining analyses (Fig. 2). This
result is similar to a previous report [29] that showed that
croton oil can induce the expression of GM-CSF in
keratinocytes and mouse skin. Our findings suggest that
phytocompounds exhibiting analogous proinflammatory
function to croton oil (which is quite toxic) may have the
potential to be developed as topically applied adjuvants for
immunotherapies or cancer vaccines. Taking this into
account, we showed that DsCE-I (a fraction of D. batatas
tuber extract) enhanced GM-CSF promoter activity in both
normal (Fig. 3a) and inflamed skin (Fig. 3b). Inflammation
is a crucial physiological reaction to infection and tissue
damage induced by physical and chemical factors. GMCSF has been considered to play a central role in maintaining chronic inflammation [20]. Our results (Fig. 3b)
demonstrated that b-methasone (e.g., as Getason)
decreased the transcriptional activity of GM-CSF promoter
in croton oil-inflamed skin. This result agrees with a previous study showing the suppression of GM-CSF
expression by glucocorticoids [30]. Shikonin and RubiDerm inhibited the activity of the transgenic GM-CSF
123
J Biomed Sci (2008) 15:813–822
promoter, whereas extract of B. pilosa showed no effect,
furthermore a fraction of D. batatas (DsCE-I) increased the
activity of GM-CSF promoter. Taken together, our results
demonstrate that this in vivo assay system can effectively
delineate both the activation and inhibition effects of test
materials on GM-CSF gene promoter.
Inflammation is involved in a spectrum of common skin
disorders, both intrinsic chronic disorders and extrinsic,
chemical or physical damage. Previous studies [31–33]
have shown that different types of skin inflammation may
be associated with the activation of different specific
groups of cytokines. For example, the chronic inflammatory skin disease atopic dermatitis, as compared with
psoriasis, was found to be associated with elevated Th2
cytokine (IL-4 and IL-13) expression and low levels of proinflammatory cytokines such as TNF-a, IFN-c, and IL-1b
[31] in skin. Irritant and allergic contact dermatitis are the
two most frequent manifestations of skin toxicity [32, 33].
Clinically, it is difficult to distinguish these two types of
dermatitis, but studies have suggested that both irritants
and allergens can activate TNF-a and GM-CSF [21]; while
only contact allergens up-regulate IL-a and IL-1b [21, 34].
We therefore evaluated whether an alternative strategy, the
current transgene assay system, could differentiate between
the cytokine expression profiles of these two types of
dermatitis. Croton oil is a known irritant that stimulates
keratinocytes to release inflammatory mediators such as
IL-1a, TNF-a, IL-8, and GM-CSF [26, 29]. Our assay
system (Fig. 4) shows that croton oil can significantly upregulate GM-CSF and TNF-a promoter activities while
only slightly enhancing IL-1b and IL-18 promoters in skin.
These results suggest that it may be possible to apply the
current transgene assay system to the systematic analysis of
cytokine expression profiles in skin or other epithelial
tissues.
In the present study we showed that shikonin, an antiinflammatory phytocompound, suppressed human GMCSF promoter activity by inhibiting TFIID (TBP) binding
to TATA box. This result is consistent with our previous
studies on TNF-a [25]. We suggest that shikonin may act
though a common molecular signaling pathway(s) to exert
its inhibitory role on the transcriptional regulation of a set
of functionally related cytokine genes. Future studies are
needed to determine how different cytokine genes are
differentially affected by shikonin and other anti-inflammatory compounds.
This study provides an in vivo three-dimensional tissue
assay system that has an advantage over the two-dimensional, (e.g., in vitro) cell culture systems. The findings of
this in vivo study provide useful information on the
application of phytocompounds or other agents as topical
immunomodulators on skin. This experimental approach
may also be employed to explore the underlying
J Biomed Sci (2008) 15:813–822
mechanisms of various cytokines or chemokines actively
involved in various immunological activities in skin.
Acknowledgements This work was supported by research grants of
No. 91S701911 and No. 94S-0402 from the National Science and
Technology Program for Agricultural Biotechnology, and an institutional grant from Academia Sinica, Taiwan, R.O.C.
References
1. Koop CE (2002) The future of medicine. Science 295:233
2. Borchers AT, Keen CL, Stern JS, Gershwin ME (2000) Inflammation and native American medicine: the role of botanicals. Am
J Clin Nutr 72:339–347
3. Hengge UR, Benninghoff B, Ruzicka T, Goos M (2001) Topical
immunomodulators–progress towards treating inflammation,
infection, and cancer. Lancet Infect Dis 1:189–198
4. Tak PP, Firestein GS (2001) NF-jB: a key role in inflammatory
diseases. J Clin Invest 107:7–11
5. Yamamoto Y, Gaynor RB (2001) Therapeutic potential of inhibition of the NF-jB pathway in the treatment of inflammation
and cancer. J Clin Invest 107:135–142
6. Subbaramaiah K, Bulic P, Lin Y, Dannenberg AJ, Pasco DS
(2001) Development and use of a gene promoter-based screen to
identify novel inhibitors of cyclooxygenase-2 transcription. J
Biomol Screen 6:101–110
7. Williams IR, Kupper TS (1996) Immunity at the surface:
homeostatic mechanisms of the skin immune system. Life Sci
58:1485–1507
8. Wittmann M, Werfel T (2006) Interaction of keratinocytes with
infiltrating lymphocytes in allergic eczematous skin diseases.
Curr Opin Allergy Clin Immunol 6:329–334
9. Yang NS, McCabe DE, Swain WF (1996) Methods for particlemediated gene transfer into skin. In: Robbins P (ed) Methods in
molecular medicine, gene therapy protocols. Totowa, Humana
Press Inc, pp 281–296
10. Armitage JO (1998) Emerging applications of recombinant
human granulocyte-macrophage colony-stimulating factor. Blood
92:4491–4508
11. Smith PD, Lamerson CL, Wong HL, Wahl LM, Wahl SM (1990)
Granulocyte-macrophage colony-stimulating factor augments
human monocyte fungicidal activity for Candida albicans. J
Infect Dis 161:999–1005
12. Shi Y, Liu CH, Roberts AI, Das J, Xu G, Ren G, Zhang Y, Zhang
L, Yuan ZR, Tan HSW, Das G, Devadas S (2006) Granulocytemacrophage colony-stimulating factor (GM-CSF) and T-cell
responses: what we do and don’t know. Cell Res 16:126–133
13. Liu XY, Gu JF (2006) Targeting gene-virotherapy of cancer. Cell
Res 16:25–30
14. Berinstein NL (2007) Enhancing cancer vaccines with immunomodulators. Vaccine 25:B72–B88
15. Dranoff G (2002) GM-CSF-based cancer vaccines. Immunol Rev
188:147–154
16. Rakhmilevich AL, Imboden M, Hao Z, Macklin MD, Roberts T,
Wright KM, Albertini MR, Yang NS, Sondel PM (2001) Effective particle-mediated vaccination against mouse melanoma by
coadministration of plasmid DNA encoding Gp100 and granulocyte-macrophage colony-stimulating factor. Clin Cancer Res
7:952–961
17. Pastore S, Belasio EF, Albanesi C, Chinni LM (1997) Granulocyte macrophage colony-stimulating factor is overproduced by
keratinocytes in atopic dermatitis. Implications for sustained
dendritic cell activation in the skin. J Clin Invest 99:3009–3017
821
18. Pastore S, Giustizieri ML, Mascia F, Giannetti A, Kaushansky K,
Girolomoni G (2000) Dysregulated activation of activator protein
1 in keratinocytes of atopic dermatitis patients with enhanced
expression of granulocyte/macrophage-colony stimulating factor.
J Invest Dermatol 115:1134–1143
19. Mann A, Breuhahn K, Schirmacher P, Wilhelmi A, Beyer C,
Rosenau A, Ozbek S, SR J, Blessing M (2001) Up- and downregulation of granulocyte/macrophage-colony stimulating factor
activity in murine skin increase susceptibility to skin carcinogenesis by independent mechanisms. Cancer Res 61:2311–2319
20. McQualter JL, Darwiche R, Ewing C, Onuki M, Kay TW,
Hamilton JA, Reid HH, BCC A (2001) Granulocyte macrophage
colony-stimulating factor: A new putative therapeutic target in
multiple sclerosis. J Exp Med 194:873–881
21. Corsini E, Galli CL (2000) Epidermal cytokines in experimental
contact dermatitis. Toxicology 142:203–211
22. Chiang YM, Chuang DY, Wang SY, Kuo YH, Tsai PW, Shyur LF
(2004) Metabolite profiling and chemopreventive bioactivity of
plant extracts from Bidens pilosa. J Ethnopharmacol 95:409–419
23. Perry NB, Burgess EJ, Glennie VL (2001) Echinacea standardization: analytical methods for phenolic compounds and typical
levels in medicinal species. J Agric Food Chem 49:1702–1706
24. Chen X, Oppenheim J, Zack Howard OM (2001) Shikonin, a
component of anti-inflammatory Chinese herbal medicine,
selectively blocks chemokine binding to CC chemokine receptor1. Int Immunopharmacol 1:229–236
25. Staniforth V, Shyur LF, Yang NS (2004) Shikonins, phytocompounds from Lithospermum erythrorhizon, inhibit the
transcriptional activation of human tumor necrosis factor promoter in vivo. J Biol Chem 279:5877–5885
26. Haas J, Lipkow T, Mohamadzadeh M, Kolde G, Knop J (1992)
Induction of inflammatory cytokines in murine keratinocytes
upon in vivo stimulation with contact sensitizers and tolerizing
analogues. Exp Dermatol 1:76–83
27. Leung DY (1997) Atopic dermatitis: immunobiology and treatment with immune modulators. Clin Exp Immunol 107 Suppl
1:25–30
28. Cassaday RD, Sondel PM, King DM, Macklin MD, Gan J,
Warner TF, Zuleger CL, Bridges AJ, Schalch HG, Kim KM,
Hank JA, Mahvi DM, Albertini MR (2007) A phase I study of
immunization using particle-mediated epidermal delivery of
genes for gp100 and GM-CSF into uninvolved skin of melanoma
patients. Clin Cancer Res 13:540–549
29. Wilmer JL, Burleson FG, Kayama F, Kanno J, Luster MI (1994)
Cytokine induction in human epidermal keratinocytes exposed to
contact irritants and its relation to chemical-induced inflammation in mouse skin. J Invest Dermatol 102:915–922
30. Smith PJ, Cousins DJ, Jee YK, Staynov DZ, Lee TH, Lavender P
(2001) Suppression of granulocyte-macrophage colony-stimulating factor expression by glucocorticoids involves inhibition of
enhancer function by the glucocorticoid receptor binding to
composite NF-AT/Activator Protein-1 elements. J Immunol
167:2502–2510
31. Nomura I, Goleva E, Howell MD, Hamid QA, Ong PY, Hall CF,
Darst MA, Gao B, Boguniewicz M, Travers JB, Leung DYM
(2003) Cytokine milieu of atopic dermatitis, as compared to
psoriasis, skin prevents induction of innate immune response
genes. J Immunol 171:3262–3269
32. Wakelin SH (2001) Contact urticaria. Clin Exp Dermatol 26:132–
136
33. Watanabe H, Unger M, Tuvel B, Wang B, Sauder DN (2002)
Contact hypersensitivity: the mechanism of immune responses
and T cell balance. J Interf Cytok Res 22:407–412
34. Muller G, Knop J, Enk AH (1996) Is cytokine expression
responsible for differences between allergens and irritants? Am J
Contact Dermat 7:177–184
123
822
35. Gray JG, Chandra G, Clay WC, Stinnett SW, Haneline SA,
Lorenz JJ, Patel IR, Wisely GB, Furdon PJ, Taylor JD (1993) A
CRE/ATF-like site in the upstream regulatory sequence of the
human interleukin 1 beta gene is necessary for induction in U937
and THP-1 monocytic cell lines. Mol Cell Biol 13:6678–6689
36. Clark BD, Collins KL, Gandy MS, Webb AC, Auron AE (1986)
Genomic sequence for human prointerleukin 1 beta: possible
evolution from a reverse transcribed prointerleukin 1 alpha gene.
Nucleic Acids Res 14:7897–7914
37. Giedraitis V, He B, Huang WX, Hillert J (2001) Cloning and
mutation analysis of the human IL-18 promoter: a possible role of
123
J Biomed Sci (2008) 15:813–822
polymorphisms in expression regulation. J Neuroimmunol
112:146–152
38. Takashiba S, Shapira L, Amar S, Van Dyke TE (1993) Cloning
and characterization of human TNF alpha promoter region. Gene
131:307–308
39. Jenkins F, Cockerill PN, Bohmann D, Shannon MF (1995)
Multiple signals are required for function of the human granulocyte-macrophage colony-stimulating factor gene promoter in T
cells. J Immunol 155:1240–1251