Accepted Manuscript Title: Transforming growth factor- (TGF-) signaling in healthy human fetal skin, a descriptive study Author: id="aut0005" > M. Walraven R.H.J. Beelen M.M.W. Ulrich PII: DOI: Reference:
S0923-1811(15)00054-7 http://dx.doi.org/doi:10.1016/j.jdermsci.2015.02.012 DESC 2795
To appear in:
Journal of Dermatological Science
Received date: Revised date: Accepted date:
18-9-2014 17-2-2015 18-2-2015
Please cite this article as: Walraven M, Beelen RHJ, Ulrich MMW, Transforming growth factor-rmbeta (TGF-rmbeta) signaling in healthy human fetal skin, a descriptive study, Journal of Dermatological Science (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.02.012 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Transforming growth factor-β (TGF-β) signaling in healthy human fetal skin, a descriptive study
a
ip t
Walraven M, MSc. a, Beelen RHJ, Prof. dr. a, Ulrich MMW, dr. a,b Dept. of Molecular Cell Biology & Immunology, VU University Medical Center, Amsterdam, The
Association of Dutch Burn Centres, Beverwijk, The Netherlands
us
b
cr
Netherlands
Corresponding author contact information
an
M. Walraven; Department of Molecular Cell Biology and Immunology (MCBI), VU University Medical Center (VUmc), Room MF J-289, Van der Boechorststraat 7, 1081 BT Amsterdam, The
d
Request for reprints contact information
M
Netherlands; T:+31204448079; F:0031-204448081; E: [email protected]
te
M.M.W. Ulrich; Department of Molecular Cell Biology and Immunology (MCBI), VU University Medical Center (VUmc), Room MF J-294, Van der Boechorststraat 7, 1081 BT Amsterdam, The
Ac ce p
Netherlands; T:+31204448058; F:0031-204448081; E: [email protected]
Funding
This work was supported by the Netherlands Institute for Regenerative Medicine (NIRM) [FES0908].
The authors have no conflict of interest to declare
Word count: 3687; Number of References: 31; Tables: 3; Figures: 3; Supplementary: 2;
1
Page 1 of 27
Abstract Background: TGF-β plays an important role in growth and development but is also involved in scarring and fibrosis. Differences for this growth factor are known between scarless fetal
ip t
wound healing and adult wound healing. Nonetheless, most of the data in this area are from animal studies or in vitro studies and, thus, information about the human situation is
cr
incomplete and scarce.
us
Objective: The aim of this study was to compare the canonical TGF-β signaling in unwounded human fetal and adult skin.
an
Methods: Q-PCR, immunohistochemistry, western blot and Luminex assays were used to determine gene expression, protein levels and protein localization of components of this
M
pathway in healthy skin.
Results: All components of the canonical TGF- β pathway were present in unwounded fetal
d
skin. Compared to adult skin, fetal skin had differential concentrations of the TGF-β isoforms,
te
had high levels of phosphorylated Receptor-Smads, especially in the epidermis, and had low
Ac ce p
expression of several fibrosis-associated target genes. Further, the results indicated that the processes of receptor endocytosis might also differ between fetal and adult skin. Conclusion: This descriptive study showed that there are differences in gene expression, protein concentrations and protein localization for most components of the canonical TGF-β pathway between fetal and adult skin. The findings of this study can be a starting point for further research into the role of TGF-β signaling in scarless healing.
2
Page 2 of 27
Introduction The intriguing phenomenon of scarless healing in early fetal skin is a quest that is studied intensively for decades, but still remains unsolved. A significant part of the research in this
ip t
area has focused on the role of transforming growth factor-β (TGF-β), as this growth factor
has a key function in scarring and fibrosis. Intriguingly, differences between fetal and adult
cr
skin have been found for the three TGF-β isoforms. These findings led to the assumption that
us
TGF-β3 has anti-fibrotic properties while TGF-β1 and –β2 are pro-fibrotic [1,2]. Although it is now known that the three isoforms have distinct functions during wound healing, all three
an
isoforms have the ability to activate the fibrosis-associated canonical TGF-β pathway [3,4]. Since all three isoforms activate the same pathway, the exact molecular mechanisms leading
M
to their differential function during wound healing are not clear [5]. Thus, the role of TGF-β in scarless healing has not been elucidated yet and the quest still remains.
d
The canonical TGF-β pathway consist of several extra- and intra-cellular components that,
te
together, form a cascade of events that results in upregulation of fibrosis-related genes. The
Ac ce p
cascade starts with the activation of inactive TGF-β from the latent TGF-β binding protein (LTBP) complex which is bound to the extracellular matrix (ECM). Upon activation, TGF-β binds to TGF-β Receptor II (TGF-βRII) which subsequently forms a complex with TGF-βRI, the so called TGF-βR complex. Formation of this complex leads to activation and phosphorylation of the R-Smads (Smad 2 and Smad 3) [6]. These phosphorylated Smads can then bind Smad 4 and, as a complex, migrate to the nucleus to activate transcription. After activation and phosphorylation of the R-Smads, the TGF-βR complex is taken up via Receptor endocytosis pathways; the receptors are either recycled back to the plasma membrane via clathrin-mediated endocytosis or degraded via caveolin-mediated endocytosis. Besides TGFβRI and –βRII, another receptor, TGF-βRIII, is identified, but its function in wound healing and fibrosis still remains unknown. 3
Page 3 of 27
We recently described the potential role that the different components of the TGF-β pathway might play in fetal scarless wound healing in a review article [7]. In contrast to the extensive studies on the TGF-β isoforms in the fetal skin, only little information is available about the
ip t
other components of the canonical TGF-β pathway such as the TGF-βRs or the intracellular Smads. In human skin, one group studied the expression of both TGF-β and the TGF-βRs [8].
cr
Several other groups studied some of the components of the pathway in rodents but little
attention has been devoted to the intracellular part of the pathway. The only data available on
us
this intracellular Smad-signaling are from in vitro studies in fibroblasts [9-12]. Therefore, only scarce information exists on the presence and function of most of the components of the
an
canonical pathway in human fetal skin.
M
The purpose of the present investigation was to evaluate all of the components of the canonical TGF-β pathway in unwounded human fetal and adult skin both on gene expression
d
as well as protein levels and protein localization. Besides the TGF-β isoforms, the LTBP, the
te
TGF-βRs, the receptor endocytosis markers, the R-Smads, and the inhibitory Smads (ISmads) were studied. Furthermore, expression levels of some direct target genes of the
Ac ce p
canonical TGF-β pathway were studied , i.e. TGF-β induced (TGF-βi) [13-15], connective tissue growth factor (CTGF) [16,17], and plasminogen activator inhibitor-1 (PAI-1) [18]. Finally, expression levels of some ECM genes that are also known to be involved in this pathway were studied, i.e. Collagen Type 1 (Col1), Collagen Type 3 (Col3) [19,20], Biglycan, Decorin and Fibromodulin [21-23].
4
Page 4 of 27
Materials and methods Tissue collection Healthy human adult skin was obtained from fully informed, consenting donors undergoing
ip t
abdominoplasty. Healthy human fetal skin was harvested from the limbs of aborted foetuses; patients undergoing elective terminations of pregnancy were fully informed and had given
us
overview of samples used for the different assays, see table 1.
cr
consent. Gestational age of the foetuses was estimated by last menstrual period. For a full
From all skin samples, subcutaneous tissue was carefully removed and two biopsies were
an
taken. One biopsy was fixed in kryofix (50% ethanol, 3% PEG300) for at least 24 hours (h) at 4°C, the other biopsy was snap-frozen and stored at -80°C. Samples that could not be snap-
M
frozen immediately were preserved in RNAlater (Life Technologies, Bleiswijk, the Netherlands) on ice for a maximum of 24 h before processing.
d
RNA isolation and cDNA synthesis
te
Frozen skin samples were manually cut into small slices and lysed using TRIzol reagent (Life
Ac ce p
Technologies). Adult skin was further lysed in a Tissue Lyser LT (Qiagen, Venlo, The Netherlands) for 5 minutes (min) at 65 Hertz. During all procedures samples were kept on ice. RNA was extracted according to the manufacturers’ instructions. RNA was dissolved in Diethylpyrocarbonate (DEPC) treated water and RNA concentration and purity were measured (Nanodrop, Nanodrop Technologies, Wilmington, DE, USA). Agarose gel electrophoresis was used to asses RNA quality, only RNA samples with clear 28S and 18S bands were used for cDNA synthesis. From these samples, 500 ng of RNA was reverse transcripted using the QuantiTect Reverse Transcription Kit (Qiagen) following the manufacturers’ instructions. The resulting cDNA was diluted 5 or 10 times in DEPC-treated water and stored at -20°C upon use.
5
Page 5 of 27
Quantitative Real-time polymerase chain reaction (Q-PCR) Primers for the different components of the TGF-β pathway were designed based on National Center for Biotechnology Information (NCBI) and the primers were synthesized by Invitrogen
ip t
(Life Technologies) (table 2). For the Q-PCR reaction, 2.5 µL of diluted cDNA was incubated with 7.5 µL of mastermix containing 300nM of the primer-pair dissolved in 5.0 µL
cr
SYBRGreen DNA polymerase (Applied Biosystems, CA, USA) and DEPC-treated water. All samples were analysed in triplicate for every primer. Q-PCR reaction was performed on a
us
StepOnePlus™ RT-PCR System (Applied Biosystems). First, SYBRGreen DNA polymerase was activated (1 min at 90°C) followed by 40 amplification cycles (15 seconds (sec) at 95°C,
an
1 min at 60°C) and a meltcurve analyses (stepwise reduction of temperature over time) was
M
performed. StepOnePlus™ software was used to determine Ct-values and relative gene expression was calculated with the ddCt method as normalized ratios between the target gene
te
Immunohistochemistry (IHC)
d
and the average of three reference genes (supp A).
Kryofix-treated samples were processed for paraffin embedding and 5 µm sections were
Ac ce p
sliced. Slides were stained with antibodies from Abcam (Cambridge, UK), R&D systems (Abingdon, UK), BD Biosciences (Breda, The Netherlands), LSBio (Seattle, WA, USA) or Santa Cruz Biotechnology Inc (Heidelberg, Germany) (supp B) and negative control slides were incubated in absence of the primary antibody. Appropriate secondary poly-horseradish peroxidase (poly-HRP) either goat-anti-mouse (1:400, P0260 clone) or goat-anti-rabbit (1:400, polyclonal) both from Dako (Glostrup, Denmark) were used followed by DAB substrate (Dako) incubation and Mayer’s Haematoxylin (Dako) as a counterstain. Localization and intensity of specific staining were evaluated and representative slides were photographed with a Zeiss BF microscope (Zeiss, Sliedrecht, The Netherlands). In case of p-Smad2/3, secondary donkey-anti-goat poly-biotin from Jackson ImmunoResearch (Suffolk, UK) was
6
Page 6 of 27
used followed by streptavidin Alexa-Fluor-555 (Life Technologies), DAPI (Invitrogen) was used to envision the nuclei. Slides were photographed with a Leica CTR6000 microscope (Leica-microsystems, Wetzlar, Germany).
ip t
Luminex Frozen skin samples were manually cut into small slices. Tissue was lysed in MILLIPLEX
cr
MAP Lysis Buffer (Millipore, Billerica, MA, USA), homogenized for 5 min at 65 Hertz in a
Tissue Lyser LT and centrifuged to remove debris and excess lipids. Two Milliplex map kits
us
were used, the TGF-β 1,2,3 Magnetic Bead Kit (TGF-β kit) and the Human TGF-β Signalling Magnetic Bead Panel 6-plex (6-plex kit) from Millipore. Assays were run following
an
manufacturer’s instructions on a Bioplex 200 system (Biorad, Veenendaal, The Netherlands).
M
For both kits, samples were analyzed in duplo. The TGF-β kit measured total protein levels (pg TGF-β/mg tissue) of TGF-β1,2,3. Inactive TGF-β was activated by lowering the pH under
d
3.0 (10 min, room temperature (RT)). Protein levels were corrected for the total protein levels
te
of the sample, which were quantified by means of a BCA assay (Pierce, Thermo Scientific, Rockford, IL, USA). The 6-plex kit measured protein levels (median fluorescence intensity
Ac ce p
(MFI)) of TGF-βRII and Smad 4 and phosphorylated levels of Smad 2 and Smad 3. MFI values were corrected for the amount of cells in these samples quantified by means of GAPDH with a Western Blot. Western Blot
Western Blot was performed to normalize for the amount of cells and to measure the receptor endocytosis markers in the Luminex samples. The lysis buffer samples (see Luminex) were loaded onto 10% sodium dodecyl sulphate polyacrylamide (SDS-PAGE) gels and the separated protein fractions were transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad). The membranes were blocked with 1:1 PBS-Odyssey Blocking Buffer (LI-COR Biosciences, Bad Homburg, Germany) and stained with primary antibodies (supp B) followed
7
Page 7 of 27
by incubation with a secondary antibody either fluorescently labelled goat-anti-mouse-800 or goat-anti-rabbit-680 (both from LI-COR Biosciences). An odyssey Infrared Imaging System (LI-COR Biosciences) was used to scan the blots and Odyssey V3.0 software was used to
ip t
measure the bands. Statistics
cr
Q-PCR, Luminex and Western Blot data were analysed by IBM SPSS software version 20 (IBM, Amsterdam, The Netherlands). All data were tested independently and non-
us
parametrically by means of a Mann-Whitney U statistical test, significance was set at p
S0923-1811(15)00054-7 http://dx.doi.org/doi:10.1016/j.jdermsci.2015.02.012 DESC 2795
To appear in:
Journal of Dermatological Science
Received date: Revised date: Accepted date:
18-9-2014 17-2-2015 18-2-2015
Please cite this article as: Walraven M, Beelen RHJ, Ulrich MMW, Transforming growth factor-rmbeta (TGF-rmbeta) signaling in healthy human fetal skin, a descriptive study, Journal of Dermatological Science (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.02.012 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Transforming growth factor-β (TGF-β) signaling in healthy human fetal skin, a descriptive study
a
ip t
Walraven M, MSc. a, Beelen RHJ, Prof. dr. a, Ulrich MMW, dr. a,b Dept. of Molecular Cell Biology & Immunology, VU University Medical Center, Amsterdam, The
Association of Dutch Burn Centres, Beverwijk, The Netherlands
us
b
cr
Netherlands
Corresponding author contact information
an
M. Walraven; Department of Molecular Cell Biology and Immunology (MCBI), VU University Medical Center (VUmc), Room MF J-289, Van der Boechorststraat 7, 1081 BT Amsterdam, The
d
Request for reprints contact information
M
Netherlands; T:+31204448079; F:0031-204448081; E: [email protected]
te
M.M.W. Ulrich; Department of Molecular Cell Biology and Immunology (MCBI), VU University Medical Center (VUmc), Room MF J-294, Van der Boechorststraat 7, 1081 BT Amsterdam, The
Ac ce p
Netherlands; T:+31204448058; F:0031-204448081; E: [email protected]
Funding
This work was supported by the Netherlands Institute for Regenerative Medicine (NIRM) [FES0908].
The authors have no conflict of interest to declare
Word count: 3687; Number of References: 31; Tables: 3; Figures: 3; Supplementary: 2;
1
Page 1 of 27
Abstract Background: TGF-β plays an important role in growth and development but is also involved in scarring and fibrosis. Differences for this growth factor are known between scarless fetal
ip t
wound healing and adult wound healing. Nonetheless, most of the data in this area are from animal studies or in vitro studies and, thus, information about the human situation is
cr
incomplete and scarce.
us
Objective: The aim of this study was to compare the canonical TGF-β signaling in unwounded human fetal and adult skin.
an
Methods: Q-PCR, immunohistochemistry, western blot and Luminex assays were used to determine gene expression, protein levels and protein localization of components of this
M
pathway in healthy skin.
Results: All components of the canonical TGF- β pathway were present in unwounded fetal
d
skin. Compared to adult skin, fetal skin had differential concentrations of the TGF-β isoforms,
te
had high levels of phosphorylated Receptor-Smads, especially in the epidermis, and had low
Ac ce p
expression of several fibrosis-associated target genes. Further, the results indicated that the processes of receptor endocytosis might also differ between fetal and adult skin. Conclusion: This descriptive study showed that there are differences in gene expression, protein concentrations and protein localization for most components of the canonical TGF-β pathway between fetal and adult skin. The findings of this study can be a starting point for further research into the role of TGF-β signaling in scarless healing.
2
Page 2 of 27
Introduction The intriguing phenomenon of scarless healing in early fetal skin is a quest that is studied intensively for decades, but still remains unsolved. A significant part of the research in this
ip t
area has focused on the role of transforming growth factor-β (TGF-β), as this growth factor
has a key function in scarring and fibrosis. Intriguingly, differences between fetal and adult
cr
skin have been found for the three TGF-β isoforms. These findings led to the assumption that
us
TGF-β3 has anti-fibrotic properties while TGF-β1 and –β2 are pro-fibrotic [1,2]. Although it is now known that the three isoforms have distinct functions during wound healing, all three
an
isoforms have the ability to activate the fibrosis-associated canonical TGF-β pathway [3,4]. Since all three isoforms activate the same pathway, the exact molecular mechanisms leading
M
to their differential function during wound healing are not clear [5]. Thus, the role of TGF-β in scarless healing has not been elucidated yet and the quest still remains.
d
The canonical TGF-β pathway consist of several extra- and intra-cellular components that,
te
together, form a cascade of events that results in upregulation of fibrosis-related genes. The
Ac ce p
cascade starts with the activation of inactive TGF-β from the latent TGF-β binding protein (LTBP) complex which is bound to the extracellular matrix (ECM). Upon activation, TGF-β binds to TGF-β Receptor II (TGF-βRII) which subsequently forms a complex with TGF-βRI, the so called TGF-βR complex. Formation of this complex leads to activation and phosphorylation of the R-Smads (Smad 2 and Smad 3) [6]. These phosphorylated Smads can then bind Smad 4 and, as a complex, migrate to the nucleus to activate transcription. After activation and phosphorylation of the R-Smads, the TGF-βR complex is taken up via Receptor endocytosis pathways; the receptors are either recycled back to the plasma membrane via clathrin-mediated endocytosis or degraded via caveolin-mediated endocytosis. Besides TGFβRI and –βRII, another receptor, TGF-βRIII, is identified, but its function in wound healing and fibrosis still remains unknown. 3
Page 3 of 27
We recently described the potential role that the different components of the TGF-β pathway might play in fetal scarless wound healing in a review article [7]. In contrast to the extensive studies on the TGF-β isoforms in the fetal skin, only little information is available about the
ip t
other components of the canonical TGF-β pathway such as the TGF-βRs or the intracellular Smads. In human skin, one group studied the expression of both TGF-β and the TGF-βRs [8].
cr
Several other groups studied some of the components of the pathway in rodents but little
attention has been devoted to the intracellular part of the pathway. The only data available on
us
this intracellular Smad-signaling are from in vitro studies in fibroblasts [9-12]. Therefore, only scarce information exists on the presence and function of most of the components of the
an
canonical pathway in human fetal skin.
M
The purpose of the present investigation was to evaluate all of the components of the canonical TGF-β pathway in unwounded human fetal and adult skin both on gene expression
d
as well as protein levels and protein localization. Besides the TGF-β isoforms, the LTBP, the
te
TGF-βRs, the receptor endocytosis markers, the R-Smads, and the inhibitory Smads (ISmads) were studied. Furthermore, expression levels of some direct target genes of the
Ac ce p
canonical TGF-β pathway were studied , i.e. TGF-β induced (TGF-βi) [13-15], connective tissue growth factor (CTGF) [16,17], and plasminogen activator inhibitor-1 (PAI-1) [18]. Finally, expression levels of some ECM genes that are also known to be involved in this pathway were studied, i.e. Collagen Type 1 (Col1), Collagen Type 3 (Col3) [19,20], Biglycan, Decorin and Fibromodulin [21-23].
4
Page 4 of 27
Materials and methods Tissue collection Healthy human adult skin was obtained from fully informed, consenting donors undergoing
ip t
abdominoplasty. Healthy human fetal skin was harvested from the limbs of aborted foetuses; patients undergoing elective terminations of pregnancy were fully informed and had given
us
overview of samples used for the different assays, see table 1.
cr
consent. Gestational age of the foetuses was estimated by last menstrual period. For a full
From all skin samples, subcutaneous tissue was carefully removed and two biopsies were
an
taken. One biopsy was fixed in kryofix (50% ethanol, 3% PEG300) for at least 24 hours (h) at 4°C, the other biopsy was snap-frozen and stored at -80°C. Samples that could not be snap-
M
frozen immediately were preserved in RNAlater (Life Technologies, Bleiswijk, the Netherlands) on ice for a maximum of 24 h before processing.
d
RNA isolation and cDNA synthesis
te
Frozen skin samples were manually cut into small slices and lysed using TRIzol reagent (Life
Ac ce p
Technologies). Adult skin was further lysed in a Tissue Lyser LT (Qiagen, Venlo, The Netherlands) for 5 minutes (min) at 65 Hertz. During all procedures samples were kept on ice. RNA was extracted according to the manufacturers’ instructions. RNA was dissolved in Diethylpyrocarbonate (DEPC) treated water and RNA concentration and purity were measured (Nanodrop, Nanodrop Technologies, Wilmington, DE, USA). Agarose gel electrophoresis was used to asses RNA quality, only RNA samples with clear 28S and 18S bands were used for cDNA synthesis. From these samples, 500 ng of RNA was reverse transcripted using the QuantiTect Reverse Transcription Kit (Qiagen) following the manufacturers’ instructions. The resulting cDNA was diluted 5 or 10 times in DEPC-treated water and stored at -20°C upon use.
5
Page 5 of 27
Quantitative Real-time polymerase chain reaction (Q-PCR) Primers for the different components of the TGF-β pathway were designed based on National Center for Biotechnology Information (NCBI) and the primers were synthesized by Invitrogen
ip t
(Life Technologies) (table 2). For the Q-PCR reaction, 2.5 µL of diluted cDNA was incubated with 7.5 µL of mastermix containing 300nM of the primer-pair dissolved in 5.0 µL
cr
SYBRGreen DNA polymerase (Applied Biosystems, CA, USA) and DEPC-treated water. All samples were analysed in triplicate for every primer. Q-PCR reaction was performed on a
us
StepOnePlus™ RT-PCR System (Applied Biosystems). First, SYBRGreen DNA polymerase was activated (1 min at 90°C) followed by 40 amplification cycles (15 seconds (sec) at 95°C,
an
1 min at 60°C) and a meltcurve analyses (stepwise reduction of temperature over time) was
M
performed. StepOnePlus™ software was used to determine Ct-values and relative gene expression was calculated with the ddCt method as normalized ratios between the target gene
te
Immunohistochemistry (IHC)
d
and the average of three reference genes (supp A).
Kryofix-treated samples were processed for paraffin embedding and 5 µm sections were
Ac ce p
sliced. Slides were stained with antibodies from Abcam (Cambridge, UK), R&D systems (Abingdon, UK), BD Biosciences (Breda, The Netherlands), LSBio (Seattle, WA, USA) or Santa Cruz Biotechnology Inc (Heidelberg, Germany) (supp B) and negative control slides were incubated in absence of the primary antibody. Appropriate secondary poly-horseradish peroxidase (poly-HRP) either goat-anti-mouse (1:400, P0260 clone) or goat-anti-rabbit (1:400, polyclonal) both from Dako (Glostrup, Denmark) were used followed by DAB substrate (Dako) incubation and Mayer’s Haematoxylin (Dako) as a counterstain. Localization and intensity of specific staining were evaluated and representative slides were photographed with a Zeiss BF microscope (Zeiss, Sliedrecht, The Netherlands). In case of p-Smad2/3, secondary donkey-anti-goat poly-biotin from Jackson ImmunoResearch (Suffolk, UK) was
6
Page 6 of 27
used followed by streptavidin Alexa-Fluor-555 (Life Technologies), DAPI (Invitrogen) was used to envision the nuclei. Slides were photographed with a Leica CTR6000 microscope (Leica-microsystems, Wetzlar, Germany).
ip t
Luminex Frozen skin samples were manually cut into small slices. Tissue was lysed in MILLIPLEX
cr
MAP Lysis Buffer (Millipore, Billerica, MA, USA), homogenized for 5 min at 65 Hertz in a
Tissue Lyser LT and centrifuged to remove debris and excess lipids. Two Milliplex map kits
us
were used, the TGF-β 1,2,3 Magnetic Bead Kit (TGF-β kit) and the Human TGF-β Signalling Magnetic Bead Panel 6-plex (6-plex kit) from Millipore. Assays were run following
an
manufacturer’s instructions on a Bioplex 200 system (Biorad, Veenendaal, The Netherlands).
M
For both kits, samples were analyzed in duplo. The TGF-β kit measured total protein levels (pg TGF-β/mg tissue) of TGF-β1,2,3. Inactive TGF-β was activated by lowering the pH under
d
3.0 (10 min, room temperature (RT)). Protein levels were corrected for the total protein levels
te
of the sample, which were quantified by means of a BCA assay (Pierce, Thermo Scientific, Rockford, IL, USA). The 6-plex kit measured protein levels (median fluorescence intensity
Ac ce p
(MFI)) of TGF-βRII and Smad 4 and phosphorylated levels of Smad 2 and Smad 3. MFI values were corrected for the amount of cells in these samples quantified by means of GAPDH with a Western Blot. Western Blot
Western Blot was performed to normalize for the amount of cells and to measure the receptor endocytosis markers in the Luminex samples. The lysis buffer samples (see Luminex) were loaded onto 10% sodium dodecyl sulphate polyacrylamide (SDS-PAGE) gels and the separated protein fractions were transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad). The membranes were blocked with 1:1 PBS-Odyssey Blocking Buffer (LI-COR Biosciences, Bad Homburg, Germany) and stained with primary antibodies (supp B) followed
7
Page 7 of 27
by incubation with a secondary antibody either fluorescently labelled goat-anti-mouse-800 or goat-anti-rabbit-680 (both from LI-COR Biosciences). An odyssey Infrared Imaging System (LI-COR Biosciences) was used to scan the blots and Odyssey V3.0 software was used to
ip t
measure the bands. Statistics
cr
Q-PCR, Luminex and Western Blot data were analysed by IBM SPSS software version 20 (IBM, Amsterdam, The Netherlands). All data were tested independently and non-
us
parametrically by means of a Mann-Whitney U statistical test, significance was set at p
