Joctrtrtrl 111Derrrfirtol~~~lctrlS&~ce. 3 ( I YY2) 13 1-I 36 c 1’492 Elsevier Science Publishers B.V. All rights reserved. 0923-1811~92SO5.00
DESC
131
OOl3h
Skin distribution and differential expression of transforming growth factor pl and p2 Vincent Falanga ‘, Carolyn 0. Gerhardt ‘, James R. Dasch’, and George A. Ksander’
Kazuhiko
Takehara3
’ l_triwr~rtv of Altrrttti Scl~ool c!f Medicitw. Deparrmett/ of DermtrlologJs. Miami, Florida; ‘Ce1tri.u Laboratories. Ptrlo .-ll~o. Cul~forttia. tYS.4: attd ‘L’tziversit~~of‘ Tok.~~t,Departmettt of Dertnatology. Tokyo. Japatt
(Received 6 September
Key \\ords:
Transforming
199 I: accepted 27 December
growth factor p: Skin; Scleroderma;
1991)
Expression
Abstract
l~ransforming grouth factor /I (TGF-/I) I and 2 have both become increasingly important in cutaneous biology, but their expression and distribution in human skin are not entirely clear. In this report, normal forearm skin from four volunteers was investigated tbr TGF-11’1and/Q immunostaining with antibodies that detect preferentially either cell- or matrix-associated forms ofthese peptrdes. Marked cell-associated TGF-/?I was found in the dermis, particularly around blood vessels and ducts; cellular TGF#l immunostaining was less prominent, and was predominantly around blood vessels. Neither TGF-/II nor -82 could be detected in the epidermis or cpithelial structures. and the dermal matrix contained minimally detectable amounts of the two Isoforms. In all cases. dermal matrix and cells contained greater amounts of TGF-fll than TGF-P2. Previous studies have shown that both TGF-/,‘I and -/(2 can induce dramatic increases in extracellular matrix, and both peptides have been implicated in the pathoyenesrs of fibrosis. We therefore investigated TGF-/II and $2 immunostaining in involved forearm skin of four patients with systemic sclerosis. Compared to normal skin. fibrotic specimens showed increased amounts of matrix and epidermal TGF+I, hut not TGF-/Q. Vv’ec~mcludc that TGF-/iI and -82 expression in human skin is differentially regulated, and that their distribution is varied and complex.
Introduction -rransforming growth factor j? (TGF-P) 1 and 2 are polypeptidcs that have been shown to regulate many fundamental cellular activities, including cell proliferation. immunologic function, and
C‘,~f.~t\/),~/‘rlcri~(~ rc>:V. Falanga. University of Miami School of Mcdicinc, Department of Dermatology. P.O. Box 016250, Miami. FL. 33101. I:SA.
extracellular matrix formation [ 1.21. Although both peptides have become increasingly important in cutaneous biology, their expression and distribution in human skin arc not entirely clear. In this report. we describe skin immunostaining with antibodies to TGF-PI and -/32 that detect preferentially either cell- or matrix-associated forms of these peptides. Our findings show that TGF-/31 and -1~2 expression in human skin is differentially regulated.
132
Materials and Methods Subjects Normal skin was obtained by excisional biopsies from the dorsal forearm of four healthy adults. Similarly, biopsies were obtained from involved (indurated) dorsal forearms of four patients with systemic sclerosis. Healthy subjects and patients were matched for age and sex. Patients with systemic sclerosis were diagnosed according to published criteria [3] had diffuse disease that was less than 1 year in duration, and were untreated. The biopsy specimens were immediately placed in 10% formalin, and were embedded in paraffin within 24 h.
Antibodies to TGF$l and $2 Three different TGF-/I antibodies were used for immunostaining. Antibody CC-Al/30 was raised against a synthetic peptide consisting of the first 30 amino-terminal amino acids of mature bovine TGF-Pl and has been previously described [4]. This reagent predominantly recognizes matrix-associated TGF-pl following digestion of tissue sections with hyaluronidase [ 51. The RAB4 antibody was also raised against a synthetic peptide consisting of the first 30 aminoterminal amino acids of mature bovine TGF-Pl and has not been previously described. RAB4 detects primarily cellular TGF-Pl following double digestion with protease and hyaluronidase. Western blots show that this reagent is highly specific for TGF$l with little or no TGF-p2 cross-reactivity (data not shown). This antibody behaves in a manner similar to that described for the reagent designated LC-Al/30 [6]. Antibody CL-B l/29 has been described previously [ 71. This reagent was raised against a synthetic peptide consisting of the first 29 amino-terminal amino acids of mature bovine TGF-P2 and is specific for TGF-JJ2. Following digestion of tissue sections with hyaluronidase, CL-B l/29 primarily detects matrix-associated TGF-/?2. After digestion of tissue sections with protease, CL-B l/29 primarily detects cell-associated TGF-fi2.
Immunostaining Immediately following surgical removal, tissue was fixed in 10% formalin for 24 h. Tissue sections, 3-5 pm in thickness, were mounted on slides coated with polylysine. The sections were deparafIinized in xylene and rehydrated in graded ethanols to deionized water. Sections were rinsed in 0.05 M Tris buffer containing 0.15 M NaCl, pH 7.6 (TBS). Tissues were digested for 30 min with 0.1 y0 (w/v) bovine testicular hyaluronidase (Sigma) to visualize matrix-associated antigen, or with 0.1 y0 protease (Streptomyces Pronase E, Sigma) to detect cell-associated antigen. Hyaluronidase digestions were carried out in 0.06 M sodium acetate-buffered saline, pH 5.5, at 37 “C, for 60 min. Pronase E digestions were carried out in 0.05 M Tris containing 0.15 M NaCl, pH 7.6 (TBS) for 5 min. Sections for use in RAB4 staining were digested with Pronase E in TBS for 5 min, rinsed three times in TBS, and digested with hyaluronidase in acetate-buffered saline for 60 min. Endogenous peroxidase was quenched with 1 y0 (v/v) HzO, in TB S for 30 min at room temperature with continuous gentle agitation, followed by three more rinses in TBS. Non-specific staining was blocked with 10 % normal goat serum in TBSBT (TBS containing 0.5% bovine serum albumin (B SA, Sigma)) and 0.05 % Tween 20 (Sigma) for 45 min and the slides were drained. Sections were then stained with primary antibody diluted to 0.1 mg/ml in TBSBT for 60 min at room temperature, followed by three 5-min rinses in TBS containing 0.05% Tween 20 (TBST). Non-specific staining was investigated by replacing the primary antibody with normal rabbit IgG (Zymed). Sections were then incubated for 10 min in biotinylated goat anti-rabbit IgG (Zymed No 62-9640) diluted to 0.015 mg/ml in TBSBT, rinsed three times in TBST, further incubated for 5 min, in streptavidin-horseradish peroxidase (Zymed, No. 43-4323) diluted to 0.01 mg/ml in TBSBT at room temperature, and finally rinsed three times in TBST. The sections were then incubated for l-2 min in diaminobenzidine (Zymed, No. 00-2014) followed by a
133
washing in running tap water for 2 min. Sections were counterstained in 0.1 T0 aqueous methyl green (Sigma M-50 15) for 30-60 s, quickly rinsed in running tap water, dehydrated in ethanols, cleared in xylene and mounted with Permount (Fisher Scientific). Results The results of the expression and distribution of TGF-/I1 and -,!I3 were the same in all four normal
Fig. 1. Cell-associated peroxidase body
RAB4
TGF-01
TGF-PI
stain of normal and
in vessels There
in normal
skin section
demonstrating (closed
marked
arrows)
is no epidermal
and ducts staining.
withm
primary
antibody
cndothelial
RAB4.
Marked
cells and pericytes. mis are also stained.
Immuno-
Fig. ?. Cell-associated
with anti-
were
incubated
cell-associated
ment
(open
association
arrows).
x 350.
TGF-/?2 Bl/29
Minimal
with
blood dermal
skin. Immunowere incubated
staining Other
with
of tissue.
x 100.
Fig. 2. Cell-associated TGF-/II in normal prroxidase stain of normal skin. Sections with
skin.
incubated
specimens. Marked cell-associated TGF-Pl was found in the dermis, particularly around blood vessels and ducts (Figs. 1 and 2); cellular TGF-/IQ immunostaining was less marked (Fig. 3) and was observed predominantly around blood vessels. Neither TGF-/I1 nor -82 could be detected in the epidermis or epithelial structures (Figs. 2 and 3). The dermal matrix contained minimally detectable amounts of TGF-/I1 (Fig. 4) or TGF-P2. Both TGF-fll and -82 have been shown to
is evident
cells in the der-
in normal
antibody cellular
vessels
after staining
(arrows).
staining.
x
skin.
Sections
Pronase
E treat-
is detectable There
in
is no epi-
100.
Fig. 4. Matrix-associated TGF$l in normal skin. Immunoperoxidase stain of normal skin section incubated with antibody
.4 1.30.
Minimal
homogeneous X 100.
staining
is present.
stain from a patient with systemic sclerosis. ConsecuFig. 6. Matrix-associated TGF-bl in systemic sclerosis. Immunoperoxidase tive sections were incubated either with control IgG without primary antibody (A) or with primary antibody A l/30 (B). In B one sees marked staining (arrows) which is more prevalent in papillary dermis. x 100.
Fig. 5. Matrix-associated TGF-/I1 in systemic sclerosis. Immunoperoxidase stain ofskin from a patient with systemic sclerosis. Sections were incubated with primary antibody A l/30. Marked staining is seen localized to matrix material in the papillary dermis. Areas of intense staining are indicated by arrows. x 100.
Fig. 7. Cell-associated TGF-/I1 in systemic sclerosis. Immunoperoxidase stain of skin from a patient with systemic sclerosis. Sections were incubated with primary antibody RAB4. This is a representative photomicrograph showing diffuse staining of epidermal cells (closed arrows). Staining is also prominent in association with vessels (open arrow). Cells in connective tissue, probably tibroblasts or macrophages, are also stained. x 250.
135
induce dramatic increases in extracellular matrix [ 8,9], and both have been implicated in the pathogenesis of fibrosis [l&-13]. We therefore investigated TGF-fil and $2 immunostaining in involvcd (indurated) forearm skin of four patients with early systemic sclerosis. Compared to normal skin. fibrotic specimens showed increased amounts of matrix TGF-PI, particularly in the papillary dermis (Fig. 5; observed in three of four patients). Figs. 6A,B shows representative examples of results of normal IgG as negative control in sections of fibrotic skin stained for matrix TGF-PI. Epidermal immunostaining with TGF-PI (Fig. 7). but not TGF-P2, was observed in all four fibrotic specimens. TGF-82 showed minimal or non-detectable expression in fibrotic skin. Discussion
Thus far, investigations in the presence and distribution of TGF-/?I and $2 in the skin have been largely limited to studies of TGF-P gene transcripts in cells by in situ hybridization [ l&-13]: the matrix distribution of the peptides has not been characterized, and generally only TGF+l or -[D has been examined [11-131. In this report, we describe the expression and distribution of TGF$l and $2 in human skin. Our report shows that dermal cells in normal skin contain marked amounts of TGF-Pl protein, in agreement with previous findings of TGF-Pl gene transcripts in normal dermis [lo]. As in previous limited reports, cellular TGF-01 and -82 occurred predominantly around blood vessels and ducts [ 10,13]. Our report shows that the epidermis and cpithelial structures do not contain detectable amounts (of either isoforms, and that minimal amounts of either peptide are diffusely associated with extracellular dermal matrix. These results clearly, indicate that the expression of TGF-PI and $2 in normal skin is differentially regulated, and that their distribution is varied and complex. Although our results are not quantitative, they do suggest that in normal skin TGF-fll is the predominant isoform.
TGF-PI and $2 are both capable of stimulating extracellular matrix protein formation [ 8,9], and both have been implicated in the pathogenesis of fibrosis [ 10-141. TGF-fil in particular has been reported to induce excessive accumulation of collagenous protein [ 151. Thus, it is of interest that in this report three of the four specimens from patients with systemic sclerosis contained greater than normal amounts of detectable TGF$,l in the dermal matrix. However, it is possible that normal matrix may contain similar amounts of TGF-j?l, and that its immunoreactivity may be unmasked under certain pathologic conditions, including fibrosis. Proteolytic enzymes and acidic conditions are some of the ways TGF-fil and $2 may become activated [ 1,2]. In fact. it should be noted that no antibody presently available can differentiate biologically active TGF-fll and $2 from their larger, inactive precursors. Another finding was that minimal or nondetectable amounts of TGF-P2 were observed in fibrotic PSS skin. This is to be contrasted with a recent report showing TGF-fl2 mRNA in PSS [ 131. Possibly, although gene transcription of TGF-fi2 is present, the amount of synthesized protein is minimal. In summary, we show evidence that TGF-pl and -82. expression in normal and fibrotic skin is differentially regulated. Acknowledgement
This work was supported in part by National Institutes of Health Grant AR 39658 to V.F. References Massague J: The transforming growth factor-b family. Plnnu Rev Cell Biol 6: 597-641, 1990. Barnard JA. Lyons RM, Moses HL: The cell biology of transforming growth factor-p. Biochim Biophys Acta 1032: 79-87, 1990. Medsger TA Jr: Systemic sclerosis, eosinophihc fasciitis and calcinosis, in: Arthritis and Allied Conditiom. Edited by DJ McCarty. Lea & Febiger. Philadelphia. PA. 1985: pp. 966-1036. Ellingsworth LR. Brennan JE, Fok K, Rosen DM, Bentz H, Piez K, Seyeden SM: Antibodies to the N-terminal
136
5
6
7
8
9
10
portion of cartilage-inducing factor A and transforming growth factor /I. J Biol Chem 261: 12362-12367, 1986. Heine UI, Munoz EF, FLanders KC, Ellingsworth LR, Peter Lam HY, Thompson NL, Sporn MB: Role of transforming growth factor-p in the development of the mouse embryo. J Cell Biol 105: 2861-2867, 1987. Flanders KC, Thompson, NL, Cissel DS, Van Obberghen-Shilling E, Baker CC, Kass ME, Ellingsworth LR, Roberts AB, Sporn MB: Transforming growth factor-/%: histochemical localization with antibodies to different epitopes. J Cell Biol 108: 653-660, 1989. Ksander GA, Gerhardt Co, Dasch JR, Ellingsworth LR: A novel polyclonal antibody (CL-B l/29) for immunolocalization of transforming growth factor-P2 (TGF-P2) in adult mouse. J Histochem Cytochem. 38: 1831-1840, 1990. Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS, Wakelield LM, Heine UI, Liotta LA, Falanga V, Kehrl JH, Fauci AS: Transforming growth factor-j: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc Nat1 Acad Sci USA 83: 4167-4171, 1986. Ksander GA, Ogawa Y, Chu GH, McMullin H, Rosenblatt JS, McPherson JM: Exogenous transforming growth factor-p2 enhances connective tissue formation and wound strength in guinea pig dermal wounds healing by secondary intent. Ann Surg 211: 288-294, 1989. Gruschwitz M. Muller PU, Sepp N, Hofer E, Fontana A,
11
12
13
14
15
Wick G: Transcription and expression of transforming growth factor type beta in the skin of progressive systemic sclerosis: a mediator of fibrosis? J Invest Dermatol 94: 197-203, 1990. Peltonen J, Kahari L, Jaakkola S, Kahari V-M, Varga J, Uitto J, Jimenez SA: Evaluation of transforming growth factor land type I procollagen gene expression in fibrotic skin disease by in situ hybridization. J Invest Dermatol 94: 365-371, 1990. Peltonen J, Varga J, Sollberg S, Uitto J. Jimenez SA: Elevated expression of the genes for transforming growth factor-/Q and type VI collagen in diffuse fasciitis associated with the eosinophilia-myalgia syndrome. J Invest Dermatol 96: 20-25, 1991. Kulozik M, Hogg A, Lankat-Buttgereit B, Krieg T: Colocalization of transforming growth factor 82 with r l(1) procollagen mRNA in tissue sections of patients with systemic sclerosis. J Clin Invest 86: 917-922, 1990. Falanga V, Tiegs SL, Alstadt SP, Roberts AB. Sporn MB: Transforming growth factor-/?: Selective increase in glycosaminoglycan synthesis by cultures of libroblasts from patients with progressive systemic sclerosis. J Invest Dermatol 89: 100-104, 1987. Krummel TM, Michna BA, Thomas BL, Sporn MB, Nelson JM, Salzberg AM, Cohen IK, Diegelmann RF: Transforming growth factor beta (TGF-/?) induces librosis in a fetal wound model. J Pediatr Surg 23: 647-652, 1988.
DESC
131
OOl3h
Skin distribution and differential expression of transforming growth factor pl and p2 Vincent Falanga ‘, Carolyn 0. Gerhardt ‘, James R. Dasch’, and George A. Ksander’
Kazuhiko
Takehara3
’ l_triwr~rtv of Altrrttti Scl~ool c!f Medicitw. Deparrmett/ of DermtrlologJs. Miami, Florida; ‘Ce1tri.u Laboratories. Ptrlo .-ll~o. Cul~forttia. tYS.4: attd ‘L’tziversit~~of‘ Tok.~~t,Departmettt of Dertnatology. Tokyo. Japatt
(Received 6 September
Key \\ords:
Transforming
199 I: accepted 27 December
growth factor p: Skin; Scleroderma;
1991)
Expression
Abstract
l~ransforming grouth factor /I (TGF-/I) I and 2 have both become increasingly important in cutaneous biology, but their expression and distribution in human skin are not entirely clear. In this report, normal forearm skin from four volunteers was investigated tbr TGF-11’1and/Q immunostaining with antibodies that detect preferentially either cell- or matrix-associated forms ofthese peptrdes. Marked cell-associated TGF-/?I was found in the dermis, particularly around blood vessels and ducts; cellular TGF#l immunostaining was less prominent, and was predominantly around blood vessels. Neither TGF-/II nor -82 could be detected in the epidermis or cpithelial structures. and the dermal matrix contained minimally detectable amounts of the two Isoforms. In all cases. dermal matrix and cells contained greater amounts of TGF-fll than TGF-P2. Previous studies have shown that both TGF-/,‘I and -/(2 can induce dramatic increases in extracellular matrix, and both peptides have been implicated in the pathoyenesrs of fibrosis. We therefore investigated TGF-/II and $2 immunostaining in involved forearm skin of four patients with systemic sclerosis. Compared to normal skin. fibrotic specimens showed increased amounts of matrix and epidermal TGF+I, hut not TGF-/Q. Vv’ec~mcludc that TGF-/iI and -82 expression in human skin is differentially regulated, and that their distribution is varied and complex.
Introduction -rransforming growth factor j? (TGF-P) 1 and 2 are polypeptidcs that have been shown to regulate many fundamental cellular activities, including cell proliferation. immunologic function, and
C‘,~f.~t\/),~/‘rlcri~(~ rc>:V. Falanga. University of Miami School of Mcdicinc, Department of Dermatology. P.O. Box 016250, Miami. FL. 33101. I:SA.
extracellular matrix formation [ 1.21. Although both peptides have become increasingly important in cutaneous biology, their expression and distribution in human skin arc not entirely clear. In this report. we describe skin immunostaining with antibodies to TGF-PI and -/32 that detect preferentially either cell- or matrix-associated forms of these peptides. Our findings show that TGF-/31 and -1~2 expression in human skin is differentially regulated.
132
Materials and Methods Subjects Normal skin was obtained by excisional biopsies from the dorsal forearm of four healthy adults. Similarly, biopsies were obtained from involved (indurated) dorsal forearms of four patients with systemic sclerosis. Healthy subjects and patients were matched for age and sex. Patients with systemic sclerosis were diagnosed according to published criteria [3] had diffuse disease that was less than 1 year in duration, and were untreated. The biopsy specimens were immediately placed in 10% formalin, and were embedded in paraffin within 24 h.
Antibodies to TGF$l and $2 Three different TGF-/I antibodies were used for immunostaining. Antibody CC-Al/30 was raised against a synthetic peptide consisting of the first 30 amino-terminal amino acids of mature bovine TGF-Pl and has been previously described [4]. This reagent predominantly recognizes matrix-associated TGF-pl following digestion of tissue sections with hyaluronidase [ 51. The RAB4 antibody was also raised against a synthetic peptide consisting of the first 30 aminoterminal amino acids of mature bovine TGF-Pl and has not been previously described. RAB4 detects primarily cellular TGF-Pl following double digestion with protease and hyaluronidase. Western blots show that this reagent is highly specific for TGF$l with little or no TGF-p2 cross-reactivity (data not shown). This antibody behaves in a manner similar to that described for the reagent designated LC-Al/30 [6]. Antibody CL-B l/29 has been described previously [ 71. This reagent was raised against a synthetic peptide consisting of the first 29 amino-terminal amino acids of mature bovine TGF-P2 and is specific for TGF-JJ2. Following digestion of tissue sections with hyaluronidase, CL-B l/29 primarily detects matrix-associated TGF-/?2. After digestion of tissue sections with protease, CL-B l/29 primarily detects cell-associated TGF-fi2.
Immunostaining Immediately following surgical removal, tissue was fixed in 10% formalin for 24 h. Tissue sections, 3-5 pm in thickness, were mounted on slides coated with polylysine. The sections were deparafIinized in xylene and rehydrated in graded ethanols to deionized water. Sections were rinsed in 0.05 M Tris buffer containing 0.15 M NaCl, pH 7.6 (TBS). Tissues were digested for 30 min with 0.1 y0 (w/v) bovine testicular hyaluronidase (Sigma) to visualize matrix-associated antigen, or with 0.1 y0 protease (Streptomyces Pronase E, Sigma) to detect cell-associated antigen. Hyaluronidase digestions were carried out in 0.06 M sodium acetate-buffered saline, pH 5.5, at 37 “C, for 60 min. Pronase E digestions were carried out in 0.05 M Tris containing 0.15 M NaCl, pH 7.6 (TBS) for 5 min. Sections for use in RAB4 staining were digested with Pronase E in TBS for 5 min, rinsed three times in TBS, and digested with hyaluronidase in acetate-buffered saline for 60 min. Endogenous peroxidase was quenched with 1 y0 (v/v) HzO, in TB S for 30 min at room temperature with continuous gentle agitation, followed by three more rinses in TBS. Non-specific staining was blocked with 10 % normal goat serum in TBSBT (TBS containing 0.5% bovine serum albumin (B SA, Sigma)) and 0.05 % Tween 20 (Sigma) for 45 min and the slides were drained. Sections were then stained with primary antibody diluted to 0.1 mg/ml in TBSBT for 60 min at room temperature, followed by three 5-min rinses in TBS containing 0.05% Tween 20 (TBST). Non-specific staining was investigated by replacing the primary antibody with normal rabbit IgG (Zymed). Sections were then incubated for 10 min in biotinylated goat anti-rabbit IgG (Zymed No 62-9640) diluted to 0.015 mg/ml in TBSBT, rinsed three times in TBST, further incubated for 5 min, in streptavidin-horseradish peroxidase (Zymed, No. 43-4323) diluted to 0.01 mg/ml in TBSBT at room temperature, and finally rinsed three times in TBST. The sections were then incubated for l-2 min in diaminobenzidine (Zymed, No. 00-2014) followed by a
133
washing in running tap water for 2 min. Sections were counterstained in 0.1 T0 aqueous methyl green (Sigma M-50 15) for 30-60 s, quickly rinsed in running tap water, dehydrated in ethanols, cleared in xylene and mounted with Permount (Fisher Scientific). Results The results of the expression and distribution of TGF-/I1 and -,!I3 were the same in all four normal
Fig. 1. Cell-associated peroxidase body
RAB4
TGF-01
TGF-PI
stain of normal and
in vessels There
in normal
skin section
demonstrating (closed
marked
arrows)
is no epidermal
and ducts staining.
withm
primary
antibody
cndothelial
RAB4.
Marked
cells and pericytes. mis are also stained.
Immuno-
Fig. ?. Cell-associated
with anti-
were
incubated
cell-associated
ment
(open
association
arrows).
x 350.
TGF-/?2 Bl/29
Minimal
with
blood dermal
skin. Immunowere incubated
staining Other
with
of tissue.
x 100.
Fig. 2. Cell-associated TGF-/II in normal prroxidase stain of normal skin. Sections with
skin.
incubated
specimens. Marked cell-associated TGF-Pl was found in the dermis, particularly around blood vessels and ducts (Figs. 1 and 2); cellular TGF-/IQ immunostaining was less marked (Fig. 3) and was observed predominantly around blood vessels. Neither TGF-/I1 nor -82 could be detected in the epidermis or epithelial structures (Figs. 2 and 3). The dermal matrix contained minimally detectable amounts of TGF-/I1 (Fig. 4) or TGF-P2. Both TGF-fll and -82 have been shown to
is evident
cells in the der-
in normal
antibody cellular
vessels
after staining
(arrows).
staining.
x
skin.
Sections
Pronase
E treat-
is detectable There
in
is no epi-
100.
Fig. 4. Matrix-associated TGF$l in normal skin. Immunoperoxidase stain of normal skin section incubated with antibody
.4 1.30.
Minimal
homogeneous X 100.
staining
is present.
stain from a patient with systemic sclerosis. ConsecuFig. 6. Matrix-associated TGF-bl in systemic sclerosis. Immunoperoxidase tive sections were incubated either with control IgG without primary antibody (A) or with primary antibody A l/30 (B). In B one sees marked staining (arrows) which is more prevalent in papillary dermis. x 100.
Fig. 5. Matrix-associated TGF-/I1 in systemic sclerosis. Immunoperoxidase stain ofskin from a patient with systemic sclerosis. Sections were incubated with primary antibody A l/30. Marked staining is seen localized to matrix material in the papillary dermis. Areas of intense staining are indicated by arrows. x 100.
Fig. 7. Cell-associated TGF-/I1 in systemic sclerosis. Immunoperoxidase stain of skin from a patient with systemic sclerosis. Sections were incubated with primary antibody RAB4. This is a representative photomicrograph showing diffuse staining of epidermal cells (closed arrows). Staining is also prominent in association with vessels (open arrow). Cells in connective tissue, probably tibroblasts or macrophages, are also stained. x 250.
135
induce dramatic increases in extracellular matrix [ 8,9], and both have been implicated in the pathogenesis of fibrosis [l&-13]. We therefore investigated TGF-fil and $2 immunostaining in involvcd (indurated) forearm skin of four patients with early systemic sclerosis. Compared to normal skin. fibrotic specimens showed increased amounts of matrix TGF-PI, particularly in the papillary dermis (Fig. 5; observed in three of four patients). Figs. 6A,B shows representative examples of results of normal IgG as negative control in sections of fibrotic skin stained for matrix TGF-PI. Epidermal immunostaining with TGF-PI (Fig. 7). but not TGF-P2, was observed in all four fibrotic specimens. TGF-82 showed minimal or non-detectable expression in fibrotic skin. Discussion
Thus far, investigations in the presence and distribution of TGF-/?I and $2 in the skin have been largely limited to studies of TGF-P gene transcripts in cells by in situ hybridization [ l&-13]: the matrix distribution of the peptides has not been characterized, and generally only TGF+l or -[D has been examined [11-131. In this report, we describe the expression and distribution of TGF$l and $2 in human skin. Our report shows that dermal cells in normal skin contain marked amounts of TGF-Pl protein, in agreement with previous findings of TGF-Pl gene transcripts in normal dermis [lo]. As in previous limited reports, cellular TGF-01 and -82 occurred predominantly around blood vessels and ducts [ 10,13]. Our report shows that the epidermis and cpithelial structures do not contain detectable amounts (of either isoforms, and that minimal amounts of either peptide are diffusely associated with extracellular dermal matrix. These results clearly, indicate that the expression of TGF-PI and $2 in normal skin is differentially regulated, and that their distribution is varied and complex. Although our results are not quantitative, they do suggest that in normal skin TGF-fll is the predominant isoform.
TGF-PI and $2 are both capable of stimulating extracellular matrix protein formation [ 8,9], and both have been implicated in the pathogenesis of fibrosis [ 10-141. TGF-fil in particular has been reported to induce excessive accumulation of collagenous protein [ 151. Thus, it is of interest that in this report three of the four specimens from patients with systemic sclerosis contained greater than normal amounts of detectable TGF$,l in the dermal matrix. However, it is possible that normal matrix may contain similar amounts of TGF-j?l, and that its immunoreactivity may be unmasked under certain pathologic conditions, including fibrosis. Proteolytic enzymes and acidic conditions are some of the ways TGF-fil and $2 may become activated [ 1,2]. In fact. it should be noted that no antibody presently available can differentiate biologically active TGF-fll and $2 from their larger, inactive precursors. Another finding was that minimal or nondetectable amounts of TGF-P2 were observed in fibrotic PSS skin. This is to be contrasted with a recent report showing TGF-fl2 mRNA in PSS [ 131. Possibly, although gene transcription of TGF-fi2 is present, the amount of synthesized protein is minimal. In summary, we show evidence that TGF-pl and -82. expression in normal and fibrotic skin is differentially regulated. Acknowledgement
This work was supported in part by National Institutes of Health Grant AR 39658 to V.F. References Massague J: The transforming growth factor-b family. Plnnu Rev Cell Biol 6: 597-641, 1990. Barnard JA. Lyons RM, Moses HL: The cell biology of transforming growth factor-p. Biochim Biophys Acta 1032: 79-87, 1990. Medsger TA Jr: Systemic sclerosis, eosinophihc fasciitis and calcinosis, in: Arthritis and Allied Conditiom. Edited by DJ McCarty. Lea & Febiger. Philadelphia. PA. 1985: pp. 966-1036. Ellingsworth LR. Brennan JE, Fok K, Rosen DM, Bentz H, Piez K, Seyeden SM: Antibodies to the N-terminal
136
5
6
7
8
9
10
portion of cartilage-inducing factor A and transforming growth factor /I. J Biol Chem 261: 12362-12367, 1986. Heine UI, Munoz EF, FLanders KC, Ellingsworth LR, Peter Lam HY, Thompson NL, Sporn MB: Role of transforming growth factor-p in the development of the mouse embryo. J Cell Biol 105: 2861-2867, 1987. Flanders KC, Thompson, NL, Cissel DS, Van Obberghen-Shilling E, Baker CC, Kass ME, Ellingsworth LR, Roberts AB, Sporn MB: Transforming growth factor-/%: histochemical localization with antibodies to different epitopes. J Cell Biol 108: 653-660, 1989. Ksander GA, Gerhardt Co, Dasch JR, Ellingsworth LR: A novel polyclonal antibody (CL-B l/29) for immunolocalization of transforming growth factor-P2 (TGF-P2) in adult mouse. J Histochem Cytochem. 38: 1831-1840, 1990. Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS, Wakelield LM, Heine UI, Liotta LA, Falanga V, Kehrl JH, Fauci AS: Transforming growth factor-j: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc Nat1 Acad Sci USA 83: 4167-4171, 1986. Ksander GA, Ogawa Y, Chu GH, McMullin H, Rosenblatt JS, McPherson JM: Exogenous transforming growth factor-p2 enhances connective tissue formation and wound strength in guinea pig dermal wounds healing by secondary intent. Ann Surg 211: 288-294, 1989. Gruschwitz M. Muller PU, Sepp N, Hofer E, Fontana A,
11
12
13
14
15
Wick G: Transcription and expression of transforming growth factor type beta in the skin of progressive systemic sclerosis: a mediator of fibrosis? J Invest Dermatol 94: 197-203, 1990. Peltonen J, Kahari L, Jaakkola S, Kahari V-M, Varga J, Uitto J, Jimenez SA: Evaluation of transforming growth factor land type I procollagen gene expression in fibrotic skin disease by in situ hybridization. J Invest Dermatol 94: 365-371, 1990. Peltonen J, Varga J, Sollberg S, Uitto J. Jimenez SA: Elevated expression of the genes for transforming growth factor-/Q and type VI collagen in diffuse fasciitis associated with the eosinophilia-myalgia syndrome. J Invest Dermatol 96: 20-25, 1991. Kulozik M, Hogg A, Lankat-Buttgereit B, Krieg T: Colocalization of transforming growth factor 82 with r l(1) procollagen mRNA in tissue sections of patients with systemic sclerosis. J Clin Invest 86: 917-922, 1990. Falanga V, Tiegs SL, Alstadt SP, Roberts AB. Sporn MB: Transforming growth factor-/?: Selective increase in glycosaminoglycan synthesis by cultures of libroblasts from patients with progressive systemic sclerosis. J Invest Dermatol 89: 100-104, 1987. Krummel TM, Michna BA, Thomas BL, Sporn MB, Nelson JM, Salzberg AM, Cohen IK, Diegelmann RF: Transforming growth factor beta (TGF-/?) induces librosis in a fetal wound model. J Pediatr Surg 23: 647-652, 1988.
