Aging is a deteriorative process marked by a functional decline, a decreased capacity for adaptive responsiveness, an increased susceptibility to cancer and other diseases, and an increased probability of death. Contributing to the global "functional decline" are multiple changes in gene expression patterns, which have been characterized in a variety of tissues and cell culture models (Cristofalo et al, 1998b).
The EPC-1 (early population doubling level cDNA-1) gene is one gene whose expression has been shown to decline dramatically during cellular aging in normal human lung and skin fibroblast lines (Doggett et al, 1992;Pignolo et al, 1993;Tresini et al, 1999;Pignolo et al, 2003). EPC-1 is expressed in several growth-arrested (G0) non-transformed cell lines, including human lung, skin, and endometrial stromal fibroblasts, vascular smooth muscle cells, retinal-pigmented epithelial cells, and T lymphocytes (Pignolo et al, 1995;Tombran-Tink et al, 1996). In addition, EPC-1 mRNA expression has been detected in vivo in a broad range of fetal and adult tissues including the ovary, testis, lung, liver, skeletal muscle, and small intestine (Tombran-Tink et al, 1996).
The EPC-1 gene product is a 50 kDa secreted protein that was shown to be identical to retinal pigmented epithelium-derived factor (PEDF) and the mouse homolog, caspin (collagen-associated serpin;Tombran-Tink et al, 1991;Doggett et al, 1992;Pignolo et al, 1993;Shirozu et al, 1996;Kozaki et al, 1998). Since EPC-1 is secreted by G0 fibroblasts, we have suggested that EPC-1 plays a role in the maintenance of or entry into G0 (Pignolo et al, 2003). This is supported by the fact that cells exhibiting a senescent or transformed phenotype, which are therefore incapable of entering a G0 growth-arrested state, produce little or no EPC-1 mRNA (our unpublished results;Pignolo et al, 2003). Other functional studies have shown that EPC-1/PEDF/caspin (EPC-1) can induce neurite-like outgrowth (Becerra et al, 1993;Steele et al, 1993;Seigel et al, 1994;Stratikos et al, 1996) and protects against glutamate- and hydrogen peroxide-induced toxicity in retinal neurons and cerebellar granule cells (Taniwaki et al, 1995;Taniwaki et al, 1997;Cao et al, 1999;DeCoster et al, 1999;Cao et al, 2001). The latter effect has been reported to be due to protection from apoptosis through the activation of NF-
B (Araki et al, 1998;Yabe et al, 2001).
As a means of examining the functional role of EPC-1/PEDF,Dawson et al (1999) have shown that EPC-1/PEDF is a powerful inhibitor of endothelial cell migration in vitro and neovascularization in the eye in vivo, suggesting its role as a major inhibitor of angiogenesis. Additional reports have suggested that loss of EPC-1 expression in the eye, as a result of either increasing age (Tombran-Tink et al, 1995) or disease, may play a role in the development of the wet form of macular degeneration or diabetic retinopathies (Ohno-Matsui et al, 2001;Rasmussen et al, 2001;Spranger et al, 2001;Holekamp et al, 2002;Ogata et al, 2002;Semkova et al, 2002;Yamagishi et al, 2002). These reports suggest that the balance between levels of EPC-1 and pro-angiogenic agents such as vascular endothelial cell growth factor (VEGF) in the eye is important for preventing certain debilitating retinal diseases that are caused by neovascularization.
As an inhibitor of angiogenesis, EPC-1 has been suggested to regulate the vasculature and mass of the pancreas and prostate (Doll et al, 2003). Lack of EPC-1/PEDF in knockout mice resulted in increased stromal blood vessels and epithelial hyperplasia. We believe that this will also be true in the skin, which contains a rich source of stromal fibroblasts that produce EPC-1 and microcapillaries that can respond to angiogenic activators. Thus, if EPC-1 expression declines with age in tissues as predicted by cell culture models, an environment would be created that is more permissive for cancer growth and metastasis.
Control of the angiogenic environment is more complex than accounting for the abundance of just one modulator, EPC-1. Although we predict an increase in vascularity due to a reduction in EPC-1 expression, angiogenesis in general has been shown to be delayed or impaired as a function of increasing age (Yamaura and Matsuzawa, 1980;Pili et al, 1994;Rivard et al, 1999;Swift et al, 1999;Sadoun and Reed, 2003). Some of these studies, however, are confounded by the use of pathological models such as creating ischemia in the hind limbs of rabbits (Rivard et al, 1999) and examining excisional wounds in mice (Swift et al, 1999). A recent report bySadoun and Reed (2003) characterized the delayed angiogenesis in healthy old mice compared with young mice by examining the neovascular response in implanted polyvinyl alcohol sponges. The only drawback to this model is that it might be considered an injury or wound response. This group showed a decline in the abundance of pro-angiogenic molecules VEGF and transforming growth factor (TGF)—
1 and an increase in the angiogenic inhibitor thrombospondin-2 (TSP-2). Furthermore, there was a decline in collagen abundance. All of this contributed to approximately a 40% reduction in capillary density without changing microvessel morphology and basement membrane composition.
Expression of the EPC-1 gene in vivo as a function of age in humans has not yet been evaluated. The purpose of this study was to determine if the decline of EPC-1 expression observed during replicative senescence of normal fibroblasts occurs in situ in the skin of donors of different ages. Here, we demonstrate that EPC-1 mRNA is detected in the dermal layer of the skin and that it exhibits an age-associated decline in expression. The largest significant decline occurs between young and middle-aged donors, suggesting that reduction in EPC-1 expression may be one of the earliest aging events that disturbs the delicate homeostatic balance of angiogenic modulators in the tissue microenvironment. We also show that the EPC-1 protein is localized to the sub-epidermal region of the skin, suggesting a possible role as a contributor to a protective barrier against abnormal (for example, sun-induced) skin angiogenesis (Bielenberg et al, 1998). Finally, we evaluated the expression of other known angiogenic modulators to obtain an overall picture of the angiogenic environment in the skin during human aging. Our results are consistent with the previously observed age-associated decline of EPC-1 expression in replicative senescence, and suggest that EPC-1 may play an important role in tissue homeostasis and contribute to the understanding of the angiogenic balance during aging.
Results
EPC-1 mRNA expression in the skin
We have shown previously that the abundance of EPC-1 mRNA and protein is dramatically lower in late passage cultured WI38 human lung fibroblasts than in early passage cultures (Doggett et al, 1992;Pignolo et al, 1993,2003). This was also true of skin fibroblast cultures derived from donors of different ages (Tresini et al, 1999). As an extension of that study, we explored the idea that expression of EPC-1 would be detected in the skin and that loss of EPC-1 expression during cellular aging would be reflected in vivo in the skin as a function of donor age. We examined normal skin sections for EPC-1 mRNA expression using in situ hybridization with anti-sense and sense strand RNA probes to determine specific and non-specific hybridization, respectively. Our results show that the majority of EPC-1 mRNA expression was detected in the dermal layer of the skin whereas, consistent with our in vitro data (Pignolo et al, 1995), keratinocytes in the epidermis expressed very little EPC-1 mRNA Figure 1a and b. This result was observed regardless of the donor age and is consistent with the fact that fibroblasts, one of the main producers of EPC-1, are the major cell type in the dermis (Sauer and Hall, 1996).
Figure 1.
EPC-1 mRNA expression is detected in the dermal layer of the skin. Representative dermal sections from an 18-y-old donor after in situ hybridization using an anti-sense (A) or sense (B) strand EPC-1 in vitro-transcribed probe, autoradiography, and staining with hematoxylin and eosin. The above are representative digital images taken on a Zeiss microscope. The majority of the hybridization is detected in the dermis (D) rather than the epidermis (E). Scale bar=50 
m.
EPC-1 mRNA expression in the dermis as a function of age
In order to determine if EPC-1 expression declines as a function of donor age, we examined multiple skin sections from donors ranging in age from 14 to 82 y old. Representative sections from two young and two old donors are shown in Figure 2. Although we do see variability in tissue sections, the results clearly show that there is much more hybridization (i.e., grains representing EPC-1 mRNA) in young donors than in old donors. To quantify the relative abundance of EPC-1 mRNA in the dermis of each donor, we used a grain counting program to determine the percentage of the total dermal area analyzed that reflected the area covered by grains. The relative abundance of EPC-1 mRNA detected in the dermis exhibited a significant downward trend with increasing donor age Figure 3. Furthermore, group analyses showed that EPC-1 mRNA levels were approximately 3–4-fold higher in young donors (
30 y old) than in middle-aged (31–60 y old; ANOVA, Tukey HSD unequal n, p=0.0018) or old donors (>60 years old; ANOVA, Tukey HSD unequal n, p=0.00038) (data not shown). In order to determine if this observation was due to a reduction in the expression per cell or to a decline in the number of cells per measured area that express EPC-1 mRNA, the cell number per dermal area (cell density) was also enumerated. Analysis of 3–6 images per slide showed that as a function of donor age, there is no significant difference in the cell density in the dermal area immediately adjacent to the epidermis Figure 4a. Previous reports have shown a decrease in the total number of cells in the dermis of abdominal skin (Andrew et al, 1964) and a decrease in average skin thickness as measured in the radial aspect of the forearm (Black, 1969). Our results may be different from these reports due to the small area in which we determined cell density and the close proximity to the epidermal border.
Figure 2.
EPC-1 mRNA expression in dermal sections of young and old donors. Representative dermal sections from young and old donors after in situ hybridization using an EPC-1 in vitro-transcribed probe as described. The epidermal (E) and dermal (D) layers of the skin are indicated. The above are representative areas from the samples examined. Scale bar=50 m.
Figure 3.
EPC-1 mRNA expression declines with increasing donor age.In situ hybridization was used to determine EPC-1 mRNA expression in skin sections from donors of different ages. The average grain density per dermal area was determined for each donor. Donors within a given experiment were normalized to the lowest donor. A power regression line is shown (n=31, r=0.63, and p=0.000216 as determined by the Pearson-Product Moment).
Full figure and legend (14K)Figure 4.
EPC-1 mRNA expression as a function of cell number. Following in situ hybridization, the cell number was determined in the dermal area of each analyzed image (A). Relative abundance of EPC-1 mRNA was then determined as a function of cell number (B). Young: 30 y old; middle-aged: between 30 and 60 y old; and old: >60 y old. Data are expressed as mean
SD of six independent experiments. Reduction of relative EPC-1 mRNA abundance compared with young donors, *p=0.013 and **p=0.0026.
In order to account for the slight differences in cell density per section, we calculated the average grain density per cell Figure 4b. These results show that in addition to reduced grain density per tissue area, middle-aged (M) and older (O) individuals show a significant reduction in the grain density per cell compared with young (Y) donors (M, p=0.013 and O, p=0.0026, ANOVA, LSD test). These results in combination with the previously published findings, which show an inverse relationship between EPC-1 mRNA expression and the in vitro age of the fibroblasts in culture (Pignolo et al, 1993), suggest that the fibroblasts in the skin of older individuals in situ exhibit characteristics of older (but not senescent) fibroblasts in culture.
EPC-1 protein abundance in the skin as a function of age
EPC-1 mRNA is expressed primarily in the dermis. At least in vitro, however, EPC-1 is a secreted protein (Pignolo et al, 2003), and its putative receptor has not been fully characterized. Only binding properties of EPC-1 to a potential binding protein in the neural retina and on specific neurons have been elucidated (Alberdi et al, 1999,2003;Aymerich et al, 2001;Bilak et al, 2002). Therefore, it was important to establish the tissue distribution of the protein itself. For this, we examined multiple skin sections by immunofluorescence for the location of the EPC-1 protein. Since EPC-1 is a secreted protein, we expected to detect the protein in a gradient radiating from dermal fibroblasts. Although we do see some staining in the dermal extracellular matrix, our results showed that the bulk of the protein accumulated in the sub-epidermal region as indicated by the bracket Figure 5. In co-staining experiments, we also examined the tissues for the presence of heparan sulfate proteoglycan (green fluorescence), which is a major component of basement membranes and identifies the dermal–epidermal border. It is clear from Figure 5b (identical to Figure 5a without the DAPI staining) that the bulk of the EPC-1 protein is just below the basement membrane. Although there is some variability, our results also suggested that there was less accumulated EPC-1 protein in older donors compared with young individuals (Figure 5a and Figure c are 22 and 24 y old, respectively, and E and F are 68 and 74 y old, respectively). This expression pattern suggests that EPC-1 may be playing a role as a barrier protein as described for thrombospondin (TSP)-1 and -2 (Wight et al, 1985;Detmar 2000). The potential barrier function of EPC-1 would help prevent endothelial cell migration out of the vascularized dermal layer of the skin.
Figure 5.
EPC-1 protein accumulates in the sub-epidermal layer of the epidermis and abundance is reduced in old donors. Skin sections from young (A–D) and old (E,F) donors were processed by immunofluorescence to determine EPC-1 protein localization (Cy3, red fluorescence), which accumulates in the sub-epidermal region (brackets). All sections were also co-stained for heparan sulfate proteoglycan (HSP), which is detected with an FITC-labeled secondary antibody (green fluorescence in B). HSPs are the major component of basement membranes and are detected at the dermal–epidermal border and in blood vessels (arrows). All images (except B) are overlaid with the DAPI stain (blue fluorescence) image to visualize the nuclei of all the cells in the tissue. A representative negative control is shown (D, same donor as C). Donors: 22 y old (A, B), 24 y old (C, D), 68 y old (E) and 74 y old (F). These are representative images from young and old donors. The epidermal (E) and dermal (D) layers of the skin are indicated. Scale bar=100 m.
The angiogenic environment in the skin
Angiogenesis is a complex balance of modulators that can stimulate or inhibit new blood vessel growth in a given tissue. We additionally evaluated some of these modulators and microvessel density (MVD) counts to determine how EPC-1 abundance may contribute to the larger picture of the angiogenic environment in the skin. We examined the tissue sections by immunofluorescence so that we can overlay images to show DAPI staining and multiple proteins. In Figure 6 we show the localization of TSP-1, a potent inhibitor of angiogenesis. There is generally a greater deposition of TSP-1 in the extracellular matrix (dermis) of older donors (68 and 74 y old) than in younger donors (22 and 24 y old). We also examined TSP-2 expression and see a similar increase in older donors (data not shown) as has been shown by others (Sadoun and Reed, 2003).
Figure 6.
Age-associated expression of thrombospondin-1 (TSP-1). Skin sections from young and old donors were processed by immunofluorescence to determine TSP-1 accumulation (FITC). DAPI images were overlaid to identify the nuclei of all cells. The epidermal (E) and dermal (D) layers of the skin are indicated. The donors' ages are indicated and these are representative images from young and old donors. Scale bar=100 m.
In contrast to the inhibitors, stimulators of new blood vessel growth trigger angiogenesis and VEGF is the most potent mitogen for endothelial cells both in vivo and in vitro. Under conditions that trigger a vascular response (ischemia or PVA sponges), others have shown that the expression of VEGF is decreased in tissue from old animals. We examined VEGF expression by immunostaining in multiple normal skin sections from young and old donors and show Figure 7 that expression is detected in the epidermis of the skin. This is consistent with previous observations that detected VEGF mRNA and protein expression in normal keratinocytes in the epidermis, but not the dermis (Weninger et al, 1996;Kishimoto et al, 2000). Although some donors show strong VEGF staining, however, others show little staining. This is also consistent with the transgenic mouse model using the VEGF promoter driving green fluorescent protein (GFP), which shows occasional GFP staining in the unstimulated epidermis (Kishimoto et al, 2000). Our analyses of young and old donors show no consistent age difference as shown with the two young and two old donors in this figure. We attribute this result to the fact that we are examining the angiogenic environment in normal tissue, rather than skin that has been triggered to undergo neovascularization.
Figure 7.
Age-associated expression of VEGF. Skin sections from young and old donors were processed by immunofluorescence to determine VEGF accumulation. DAPI images were overlaid to identify the nuclei of all cells. Donors: 24 y old (A), 27 y old (B), 68 y old (C) and 74 y old (D and E). A negative control is shown for Cy3 staining in E. The epidermal (E) and dermal (D) layers of the skin are indicated. These are representative images from young and old donors. Scale bar=100 m.
Finally, we examined MVD in skin sections as a function of age (young vs. old). Using an antibody to the endothelial cell marker CD31, we visualized microvessels in the dermis and counted the number per field in multiple sections from each donor examined Figure 8a. Skin sections from old donors showed a significantly decreased vessel density. We also show CD31 staining (green fluorescence with * indicating stained microvessels in the dermis) in random fields of three young and three old donors Figure 8b. Although we predicted that the decline in expression of the EPC-1 angiogenic inhibitor would lead to an increase in dermal microvessels, our results are consistent with the age-related decline in MVD as reported in the literature (Yamaura and Matsuzawa, 1980;Pili et al, 1994;Sadoun and Reed, 2003). These results point to the complexity of the angiogenic environment and that there is a complicated balance of multiple stimulators and inhibitors. Although there is a decline in blood vessel density with increasing age and in some cases an increase in microvessels that result under pathological conditions, angiogenesis is not completely absent in elderly people. We suggest that the decline in EPC-1 expression helps to balance or reverse the age-related decline in angiogenesis.
Figure 8.
Microvessel density. Skin sections from young and old donors were processed by immunofluorescence to determine CD31 to identify endothelial cells in the dermis and stained with DAPI to identify nuclei of all cells in the tissue sections. Microvessels were enumerated in five to eight sections per donor and the average number per section is shown in A (p<0.005, Student's t test, Excel). Representative images of the CD31 staining (green fluorescence) in three random young (B–D) and old (F–H) donors are indicated. Donors: 24 y old (B), 24 y old (C), 27 y old (D), 68 y old (E) and 74 y old (F), and 78 y old (G). * indicates microvessels stained with CD31. Scale bar=100 m.
Discussion
EPC-1/PEDF expression has been shown to decline during replicative aging in cell culture models (Doggett et al, 1992;Pignolo et al, 1993,2003;Tombran-Tink et al, 1995;Li et al, 1999;Palmieri et al, 1999;Tresini et al, 1999). There has not been a study, however, to determine if EPC-1 expression declines as a function of age in vivo. In a previous report,Tresini et al (1999) showed that, regardless of donor age, skin fibroblast lines derived from donors of different ages (young adults, 17–33 y old and old adults, 78–94 y old) exhibited an in vitro age-dependent decline in EPC-1 mRNA expression. No correlation, however, was observed between EPC-1 mRNA abundance and donor age when examined at approximately 50%–60% lifespan completed. One interpretation of this result is that down-regulation of EPC-1 mRNA abundance is not a biomarker of aging in vivo, but a senescence-specific change in gene expression linked to the replicative age of the cells in culture. An alternative explanation is that the established dermal fibroblast cultures are not a true representation of the age of the donor since the resultant fibroblast culture is highly selected for an outgrowth of the most robust (or youthful) cells from the skin biopsy. In fact, when the health status and biopsy conditions are controlled, the replicative lifespan of fibroblast lines derived from adult skin does not correlate with donor age (Cristofalo et al, 1998a), suggesting that they do not represent a realistic picture of gene expression in vivo. Therefore, it is not surprising that there is no difference in EPC-1 mRNA abundance between skin fibroblast lines derived from young and old donors at the same in vitro age.
In order to obtain an accurate representation of EPC-1 expression as a function of donor age, here we examined skin sections in situ and demonstrated for the first time that EPC-1 mRNA abundance declines in the dermal layer of the skin during aging in vivo. Our results show that the largest drop in abundance is between young and middle-aged donors and only a small but not statistically significant additional drop occurs between middle-aged and old donors. This trajectory of decline is reflected in the best-fit power regression line displayed in Figure 3. In contrast, EPC-1 mRNA abundance in senescent cultures in vitro exhibits a gradual decline during the lifespan of the cells (Tresini et al, 1999). The apparent difference in the trajectory of the age-associated decline in relative EPC-1 mRNA abundance between the cell culture model and the tissue may be the result of the selection for the "youngest" cells that occurs in the process of establishing cultures. Nonetheless, the general decline observed during the replicative lifespan of cells in culture is recapitulated in aging dermal tissue.
One alternative explanation for the age-related decline in EPC-1 mRNA abundance in the dermis is sun damage rather than aging per se. Most of the tissues that were examined were from the arm, face, back, or leg, all of which are from sun-exposed areas. Photodamage increases with age and at this point it is not clear if the decrease in expression is due solely to aging or to sun-exposure. Since we have examined, however, only a few tissues from non-sun-exposed areas, which also show a decline in EPC-1 mRNA abundance, on-going studies are focused on deciphering the effects of sun-exposure on EPC-1 expression.
Given that EPC-1 has been shown to be an inhibitor of angiogenesis, we suggest that the loss of EPC-1 mRNA expression in middle-aged individuals is an early event that begins to disturb the homeostatic balance between various angiogenic modulators. In turn, this leads to the establishment of an environment that is permissive for neovascularization and perhaps tumor growth via angiogenesis. Our results may contribute to a better understanding of the role of EPC-1 in this process.
Interestingly, our results also show that, although EPC-1 mRNA is detected in the dermal layer of the skin, the secreted EPC-1 protein accumulates in the sub-epidermal region of the skin. In light of the robust anti-angiogenic activity of EPC-1, and the fact that the dermis supports a rich vasculature consisting of capillaries, arterioles, venules, and lymph capillaries (Okun et al, 1988), EPC-1 may contribute to the formation of a barrier between the skin surface (epidermis) and the dermal layer of the skin. This barrier would prevent the migration of dermal microvascular cells into the epidermis. A similar hypothesis was proposed for the anti-angiogenic molecules thrombospondin (TSP)-1 and -2, which are deposited in the basement membrane area of the skin (Wight et al, 1985;Detmar, 2000). In both cases, these molecules probably contribute to the barrier function that blocks blood vessel ingrowth from the vascularized dermis into the epidermis (Detmar, 2000). If EPC-1 contributes similarly to barrier function at this interface but declines with increasing age, this suggests that there would be an increase in vascular dysplasias and an increase in diseases that result from abnormal blood vessel growth. In fact, it is well documented that vascular dysplasias, such as senile angiomata, cherry angiomatas, de Morgan spots, angiokeratomas, purpura, palor, venous lake formation, and telangiectasias, are observed in virtually all individuals over the age of 70 (Ryan, 1969;Montagna and Carlisle, 1979;Balin and Lin, 1989;Tsuchida, 1993). Despite these numerous cutaneous disorders in the elderly, however, it is generally accepted that angiogenesis is delayed or impaired during aging (Yamaura and Matsuzawa, 1980;Pili et al, 1994;Sadoun and Reed, 2003). This points to the fact that the triggering of angiogenesis is complex and relies on the balance of many inhibitory and stimulatory molecules. Whereas our studies have focused on the age-related decline in the expression of EPC-1, an inhibitor, there is also an increase in the abundance of other inhibitors, the thrombospondins and at least in some cases a decline in VEGF expression (Rivard et al, 1999;Sadoun and Reed, 2003). This gives the impression that many of these modulators become imbalanced during aging. We suggest that the loss of EPC-1 expression may reflect a compensatory change that helps to restore the balance that is disturbed by the changing concentrations and responses of other modulators in older human donors and animal models. In so doing, we suggest that this helps establish a pro-angiogenic environment with numerous cutaneous vascular diseases, and perhaps tumor growth and metastasis, developing in the elderly. Of interest is that the microvascular diseases are likely to result from capillary proliferation and dilation at focal points. This is consistent with our overall hypothesis that EPC-1 plays a role in blocking capillary proliferation in young individuals, whereas in the elderly there is a decline in the production of EPC-1 in a mosaic fashion, thus resulting in age-associated cutaneous vascular dysplasias.
Materials and Methods
Plasmids and preparation of RNA probes
The full-length PEDF cDNA gene (kindly provided J. Tombran-Tink) and the truncated fragment (797–1489 bp; PEDF
5') were cloned into the pBluescript plasmid (pBS; Stratagene; La Jolla, California). The full-length clone was linearized with XhoI to generate an anti-sense probe at the 3'-end of PEDF using T3 RNA polymerase. The PEDF5' clone was linearized with EcoRI to generate the sense strand probe with T7 RNA polymerase. Sense and anti-sense RNA probes were generated from the linearized templates by using an in vitro transcription kit (Promega; Madison, Wisconsin) and the probes were labeled with [35S]-UTP (Amersham Biosciences; Piscataway, New Jersey). The radioactively labeled RNA probes were purified by NucTrap probe purification columns according to the manufacturer's instructions (Stratagene).
Tissue sections
Burows triangles of normal skin were obtained during surgical repair of defects following Mohs micrographic surgery (sun-exposed) or during the harvesting of skin grafts (non sun-exposed) to repair defects following cancer removal. Individuals ranged in age from 14 to 82 years old and samples were categorized as young (30 years old, n=11), middle-aged (31–60 years old, n=10), and old (>60 years old, n=14). The use of human skin sections in this study was approved by the Main Line Hospitals (MLH) Institutional Review Board and performed in accordance with the Declaration of Helsinki Guidelines and with the MLH federal-wide assurance. Immediately following resection, the tissue was flash frozen and placed in OCT embedding compound. Samples from the frozen tissue blocks were cut into 5 m sections, mounted on clean Snowcoat glass slides, and stored at -80°C in the presence of Dri-Rite.
In situ Hybridization
The tissue sections were briefly air-dried (15–20 min) and fixed in 4% paraformaldehyde in PBS (phosphate-buffered saline), pH 7.4 (all solutions were made with 0.1% diethyl pyrocarbonate-treated water). The fixed tissue sections were acetylated in 0.25% acetic anhydride in 0.1 M triethanolamine (pH 8.0) for 10 min, rinsed in 2 
SSC (0.3 M sodium chloride, 0.03 M sodium citrate), de-hydrated through a graded series of ethanols, and then air-dried. The hybridization solution consisted of 200 L of the 35S-labeled RNA probe (8 107 cpm of the probe in nuclease-free water containing 50 mM DTT (dithiothreotol) and 2.5 mg salmon sperm DNA) and 800 L of the hybridization buffer (62.5% formamide, 12.5% dextran sulfate, 0.25 M NaCl, 1.25 Denhardt's solution, 12.5 mM Tris, pH 7.5, 1.25 mM EDTA, pH 8.0 in nuclease-free water). A 50 L aliquot of the hybridization solution was applied to each tissue section, covered with a coverslip, and incubated for 3.5 h at 55°C in a humidified chamber. After hybridization, the sections were washed at room temperature (RT) in 4 SSC with gentle agitation (40 min). Coverslips were gently removed and sections were washed 5 more times in 4X SSC (5 min each, RT). This was followed by treatment with RNase A to remove the unhybridized probe (20 g/mL RNase A in 0.5 M NaCl, 10 mM Tris, pH 7.5, 1 mM EDTA; 30 min at 37°C). The tissue sections were then washed (RT) in two changes of 2 SSC, one wash 1 SSC, and one wash of 0.5 SSC, all containing 1 mM DTT. This was followed by a high stringency wash in 0.1 SSC containing 1 mM DTT (55°C, 30 min). Following five dips in 0.1 SSC (RT), the sections were then dehydrated sequentially in 50%, 75%, and 95% ethanol (in 1 mM DTT and 0.1 SSC) for 3 min each, and finally three changes of 100% ethanol (3 min). Air-dried slides were exposed to Amersham Hyperfilm for 1–4 d at 4°C to determine optimal emulsion exposure time based on negative controls and a 14C standard slide.
Autoradiography was performed by dipping the slides in Kodak NTB2 nuclear track emulsion that was diluted 1:1 with sterile distilled water (Eastman Kodak Company, Rochester, NY). After exposure in the dark at 4°C (3 d in emulsion for every day on film), the slides were developed in Kodak D-19 developer for 5 min, fixed in Kodak fixer for 5 min, rinsed extensively in water, and stained with hematoxylin and eosin. To evaluate the specificity of the 35S-labeled anti-sense EPC-1 probes to the tissue sections, control experiments were performed using a 35S-labeled sense EPC-1 probe separately for each sample. The hybridized and stained skin sections were examined by light microscopy using a Zeiss microscope and images were recorded using a digital camera (AxioCam) with the KS300 v 3.0 software (Carl Zeiss Microimaging, Thornwood, New York). In addition to recorded images of sections hybridized with sense and anti-sense probes, we also recorded images away from the tissue to determine background probe binding to the slide. To analyze each image for grain abundance, a region in each image was circumscribed and a grain counting Macro written for the KS300 imaging software (Zeiss) was used to determine the total area, the percent area representing grains, and the grain density for that image. For each experiment, usually two slides per donor were hybridized with the anti-sense probe and 1 with the sense probe. Between four and six images were examined at 630, recorded, and analyzed per slide.
Statistical analysis
For each in situ hybridization experiment, multiple images from young, middle-aged, and old donors were analyzed for grain density and grains per cell of the circumscribed dermal area of the image. After correcting for background hybridization away from the tissue and from the sense strand probe, we normalized each experiment to the lowest value, which was given a value of 1. Data representing the relative grain density per area and per cell from 35 donors ranging in age from 14 to 82 years were compared using ANOVA. Post hoc comparisons were made using least significant difference (LSD) or the Tukey honest significant difference (HSD) for unequal N tests. Microsoft Excel (Excel 2000, Microsoft, Redmond, WA) and the Statistica (StatSoft, v.5.5A; Statsoft, Tulsa, OK) software packages were used in these analyses. All data in the bar graphs are reported as the meanstandard deviation.
Tissue immunofluorescence
Prepared as described above, slides containing 5 m thick sections were removed from -80°C, air-dried and then fixed in cold methanol or in cold acetone for 2 min at -20°C followed by 80% methanol at 4°C for 5 min (M. Detmar, personal communication). Tissue sections were blocked (10% goat serum or donkey serum in PBS) and then incubated with specified primary antibodies diluted in 10% serum in PBS overnight at 4°C. After washing sections, they were incubated with the appropriate secondary antibody conjugated to either FITC (fluorescein isothiocyanate) or Cy3 (Jackson ImmunoResearch Labs; West Grove, Pennsylvania) for 1 h at RT in the dark. Following incubation with the secondary antibody and washes, coverslips were mounted with VECTASHIELD mounting medium containing 4',6-diamidino-2-phenylindole (DAPI; Vector Labs, Burlingame, California) to counterstain DNA and visualize the nuclei of all cells. The following primary antibodies were used: a rabbit polyclonal antibody to PEDF (BioProducts Maryland, Middletown, Maryland), a mouse monoclonal anti-human CD31 to detect endothelial cells (R&D Systems; Minneapolis, Minnesota), a rat monoclonal antibody to heparan sulfate proteoglycans to detect basement membranes (Biomeda; Foster City, California), a rabbit polyclonal antibody to Ki67 to detect proliferating cells (Novocastra Laboratories, Newcastle-upon-Tyne, UK, Distributed by US Vector Labs, Burlingame, CA), a chicken anti-human TSP-1 (thrombospondin, kindly provided by Robert Swerlick, Emory University), a rabbit polyclonal antibody to TSP-2 (kindly provided by Michael Detmar, Harvard Medical School/Massachusetts General Hospital), and an affinity purified rabbit polyclonal antibody to VEGF (A-20, Santa Cruz Biotechnology; Santa Cruz, California). In all experiments, secondary antibody alone served as negative controls. Immunostained sections were examined at 400 by fluorescence microscopy using a Zeiss microscope. Images were captured as described above using a digital camera with the AxioVision software (Carl Zeiss Microimaging).
MVD counts
Cryostat sections (5 m) were prepared as described above and stained with an anti-human CD31 mouse monoclonal antibody as described above to identify endothelial cells that line blood vessels. Vessels were identified by fluorescence microscopy using an FITC or Cy3 filter depending on the fluorophore-coupled secondary antibody. Vessels were counted at 250 in five to eight randomly selected fields per section. Five young and four old donors were examined.
References
| 1. | Alberdi E, Aymerich MS & Becerra SP. Binding of pigment epithelium-derived factor (PEDF) to retinoblastoma cells and cerebellar granule neurons. Evidence for a PEDF receptor. J Biol Chem (1999) 274: 31605–31612 10.1074/jbc.274.44.31605. | Article | PubMed | ISI | ChemPort | |
| 2. | Alberdi EM, Weldon JE & Becerra SP. Glycosaminoglycans in human retinoblastoma cells: Heparan sulfate, a modulator of the pigment epithelium-derived factor-receptor interactions. BMC Biochem (2003) 4: 1 10.1186/1471-2091-4-1. | Article | PubMed | |
| 3. | Andrew W, Behnke R & Sato T. Changes with advancing age in the cell population of human dermis. Gerontologica (1964) 10: 1–19. | ChemPort | |
| 4. | Araki T, Taniwaki T, Becerra SP, Chader GJ & Schwartz JP. Pigment epithelium-derived factor (PEDF) differentially protects immature but not mature cerebellar granule cells against apoptotic cell death. J Neurosci Res (1998) 53: 7–15 10.1002/(SICI)1097-4547(19980701)53:1<7::AID-JNR2>3.3.CO;2-H. | Article | PubMed | ISI | ChemPort | |
| 5. | Aymerich MS, Alberdi EM, Martinez A & Becerra SP. Evidence for pigment epithelium-derived factor receptors in the neural retina. Invest Ophthalmol Vis Sci (2001) 42: 3287–3293. | PubMed | ISI | ChemPort | |
| 6. | Balin AK & Lin AN. Skin changes as a biological marker for measuring the rate of human aging. In: Balin AK, Kligman AM (eds)Aging and the Skin (1989) New York: Raven Press p 43–75. |
| 7. | Becerra SP, Palmer I, Kumar A, Steele F, Shiloach J, Notario V & Chader GJ. Overexpression of fetal human pigment epithelium-derived factor in Escherichia coli. A functionally active neurotrophic factor. J Biol Chem (1993) 268: 23148–23156. | PubMed | ISI | ChemPort | |
| 8. | Bielenberg DR, Bucana CD, Sanchez R, Donawho CK, Kripke ML & Fidler IJ. Molecular regulation of UVB-induced cutaneous angiogenesis. J Invest Dermatol (1998) 111: 864–872. | Article | PubMed | ISI | ChemPort | |
| 9. | Bilak MM, Becerra SP, Vincent AM, Moss BH, Aymerich MS & Kuncl RW. Identification of the neuroprotective molecular region of pigment epithelium-derived factor and its binding sites on motor neurons. J Neurosci (2002) 22: 9378–9386. | PubMed | ISI | ChemPort | |
| 10. | Black M. A modified radiographic method for measuring skin thickness. Br J Dermatol (1969) 81: 661–666. | PubMed | ISI | ChemPort | |
| 11. | Cao W, Tombran-Tink J, Chen W, Mrazek D, Elias R & McGinnis JF. Pigment epithelium-derived factor protects cultured retinal neurons against hydrogen peroxide-induced cell death. J Neurosci Res (1999) 57: 789–800. | Article | PubMed | ISI | ChemPort | |
| 12. | Cao W, Tombran-Tink J, Elias R, Sezate S, Mrazek D & McGinnis JF. In vivo protection of photoreceptors from light damage by pigment epithelium-derived factor. Invest Ophthalmol Vis Sci (2001) 42: 1646–3293. | PubMed | ISI | ChemPort | |
| 13. | Cristofalo VJ, Allen RG, Pignolo RJ, Martin BG & Beck JC. Relationship between donor age and the replicative lifespan of human cells in culture: A reevaluation. Proc Natl Acad Sci USA (1998a) 95: 10614–10619. | Article | PubMed | ChemPort | |
| 14. | Cristofalo VJ, Volker C, Francis MK & Tresini M. Age-dependent modifications of gene expression in human fibroblasts. Crit Rev Eukaryot Gene Expr (1998b) 8: 43–80. | PubMed | ISI | ChemPort | |
| 15. | Dawson DW, Volpert OV, Gillis P, Crawford SE, Xu H, Benedict W & Bouck NP. Pigment epithelium-derived factor: A potent inhibitor of angiogenesis. Science (1999) 285: 245–248 10.1126/science.285.5425.245. | Article | PubMed | ISI | ChemPort | |
| 16. | DeCoster MA, Schabelman E, Tombran-Tink J & Bazan NG. Neuroprotection by pigment epithelial-derived factor against glutamate toxicity in developing primary hippocampal neurons. J Neurosci Res (1999) 56: 604–610 10.1002/(SICI)1097-4547(19990615)56:6<604::AID-JNR6>3.3.CO;2-2. | Article | PubMed | ISI | ChemPort | |
| 17. | Detmar M. Tumor angiogenesis. J Investig Dermatol Symp Proc (2000) 5: 20–23 10.1046/j.1087-0024.2000.00003.x. | Article | PubMed | ISI | ChemPort | |
| 18. | Doggett DL, Rotenberg MO, Pignolo RJ, Phillips PD & Cristofalo VJ. Differential gene expression between young and senescent, quiescent WI-38 cells. Mech Ageing Dev (1992) 65: 239–255 10.1016/0047-6374(92)90039-G. | Article | PubMed | ISI | ChemPort | |
| 19. | Doll JA, Stellmach VM & Bouck NP. et al Pigment epithelium-derived factor regulates the vasculature and mass of the prostate and pancreas. Nat Med (2003) 12: 12. |
| 20. | Holekamp NM, Bouck N & Volpert O. Pigment epithelium-derived factor is deficient in the vitreous of patients with choroidal neovascularization due to age-related macular degeneration. Am J Ophthalmol (2002) 134: 220–227 10.1016/S0002-9394(02)01549-0. | Article | PubMed | ISI | ChemPort | |
| 21. | Kishimoto J, Ehama R, Ge Y, Kobayashi T, Nishiyama T, Detmar M & Burgeson RE. In vivo detection of human vascular endothelial growth factor promoter activity in transgenic mouse skin. Am J Pathol (2000) 157: 103–110. | PubMed | ISI | ChemPort | |
| 22. | Kozaki K, Miyaishi O & Koiwai O. et al Isolation, purification, and characterization of a collagen-associated serpin, caspin, produced by murine colon adenocarcinoma cells. J Biol Chem (1998) 273: 15125–15130 10.1074/jbc.273.24.15125. | Article | PubMed | ISI | ChemPort | |
| 23. | Li W, Yanoff M, Li Y & He Z. Artificial senescence of bovine retinal pigment epithelial cells induced by near-ultraviolet in vitro. Mech Ageing Dev (1999) 110: 137–155 10.1016/S0047-6374(99)00047-0. | Article | PubMed | ISI | ChemPort | |
| 24. | Montagna W & Carlisle K. Structural changes in aging human skin. J Invest Dermatol (1979) 73: 47–53 10.1111/1523-1747.ep12532761. | Article | PubMed | ISI | ChemPort | |
| 25. | Ogata N, Nishikawa M, Nishimura T, Mitsuma Y & Matsumura M. Unbalanced vitreous levels of pigment epithelium-derived factor and vascular endothelial growth factor in diabetic retinopathy. Am J Ophthalmol (2002) 134: 348–353 10.1016/S0002-9394(02)01568-4. | Article | PubMed | ISI | ChemPort | |
| 26. | Ohno-Matsui K, Morita IT & ombran-Tink J. et al Novel mechanism for age-related macular degeneration: An equilibrium shift between the angiogenesis factors VEGF and PEDF. J Cell Physiol (2001) 189: 323–333 10.1002/jcp.10026. | Article | PubMed | ISI | ChemPort | |
| 27. | Okun MR, Edelstein LM & Fisher BK. Gross and Microscopic Pathology of the Skin (1988) Canton, MA: Dermatopathology Foundation Press p 14–49. |
| 28. | Palmieri D, Watson JM & Rinehart CA. Age-related expression of PEDF/EPC-1 in human endometrial stromal fibroblasts: Implications for interactive senescence. Exp Cell Res (1999) 247: 142–147 10.1006/excr.1998.4341. | Article | PubMed | ISI | ChemPort | |
| 29. | Pignolo RJ, Cristofalo VJ & Rotenberg MO. Senescent WI-38 cells fail to express EPC-1, a gene induced in young cells upon entry into the G0 state. J Biol Chem (1993) 268: 8949–8957. | PubMed | ISI | ChemPort | |
| 30. | Pignolo RJ, Francis MK, Rotenberg MO & Cristofalo VJ. Putative role for EPC-1/PEDF in the G0 growth arrest of human diploid fibroblasts. J Cell Physiol (2003) 195: 12–20 10.1002/jcp.10212. | Article | PubMed | ISI | ChemPort | |
| 31. | Pignolo RJ, Rotenberg MO & Cristofalo VJ. Analysis of EPC-1 growth state-dependent expression, specificity, and conservation of related sequences. J Cell Physiol (1995) 162: 110–118. | Article | PubMed | ISI | ChemPort | |
| 32. | Pili R, Guo Y, Chang J, Nakanishi H, Martin GR & Passaniti A. Altered angiogenesis underlying age-dependent changes in tumor growth. J Natl Cancer Inst (1994) 86: 1303–1314. | PubMed | ChemPort | |
| 33. | Rasmussen H, Chu KW & Campochiaro P. Clinical protocol. An open-label, phase I, single administration, dose-escalation study of ADGVPEDF.11D (ADPEDF) in neovascular age-related macular degeneration (AMD). Hum Gene Ther (2001) 12: 2029–2032. | PubMed | ChemPort | |
| 34. | Rivard A, Fabre JE & Silver M. et al Age-dependent impairment of angiogenesis. Circulation (1999) 99: 111–120. | PubMed | ISI | ChemPort | |
| 35. | Ryan TJ. The epidermis and its blood supply in venous disorders of the leg. Trans St Johns Hosp Dermatol Soc (1969) 55: 51–63. | PubMed | ChemPort | |
| 36. | Sadoun E & Reed MJ. Impaired angiogenesis in aging is associated with alterations in vessel density, matrix composition, inflammatory response, and growth factor expression. J Histochem Cytochem (2003) 51: 1119–1130. | PubMed | ISI | ChemPort | |
| 37. | Sauer GC & Hall JC. Manual of Skin Diseases (1996) Philadelphia: Lippincott-Raven Publishers p 1–8. |
| 38. | Seigel GM, Tombran-Tink J & Becerra SP. et al Differentiation of Y79 retinoblastoma cells with pigment epithelial-derived factor and interphotoreceptor matrix wash: Effects on tumorigenicity. Growth Factors (1994) 10: 289–297. | PubMed | ISI | ChemPort | |
| 39. | Semkova I, Kreppel F & Welsandt G. et al Autologous transplantation of genetically modified iris pigment epithelial cells: A promising concept for the treatment of age-related macular degeneration and other disorders of the eye. Proc Natl Acad Sci USA (2002) 99: 13090–13095 10.1073/pnas.202486199. | Article | PubMed | ChemPort | |
| 40. | Shirozu M, Tada H & Tashiro K. et al Characterization of novel secreted and membrane proteins isolated by the signal sequence trap method. Genomics (1996) 37: 273–280 10.1006/geno.1996.0560. | Article | PubMed | ISI | ChemPort | |
| 41. | Spranger J, Osterhoff M & Reimann M. et al Loss of the antiangiogenic pigment epithelium-derived factor in patients with angiogenic eye disease. Diabetes (2001) 50: 2641–2645. | PubMed | ISI | ChemPort | |
| 42. | Steele FR, Chader GJ, Johnson LV & Tombran-Tink J. Pigment epithelium-derived factor: Neurotrophic activity and identification as a member of the serine protease inhibitor gene family. Proc Natl Acad Sci USA (1993) 90: 1526–1530. | PubMed | ChemPort | |
| 43. | Stratikos E, Alberdi E, Gettins PG & Becerra SP. Recombinant human pigment epithelium-derived factor (PEDF): Characterization of PEDF overexpressed and secreted by eukaryotic cells. Protein Sci (1996) 5: 2575–2582. | PubMed | ISI | ChemPort | |
| 44. | Swift ME, Kleinman HK & DiPietro LA. Impaired wound repair and delayed angiogenesis in aged mice. Lab Invest (1999) 79: 1479–1487. | PubMed | ISI | ChemPort | |
| 45. | Taniwaki T, Becerra SP, Chader GJ & Schwartz JP. Pigment epithelium-derived factor is a survival factor for cerebellar granule cells in culture. J Neurochem (1995) 64: 2509–2517. | PubMed | ISI | ChemPort | |
| 46. | Taniwaki T, Hirashima N, Becerra SP, Chader GJ, Etcheberrigaray R & Schwartz JP. Pigment epithelium-derived factor protects cultured cerebellar granule cells against glutamate-induced neurotoxicity. J Neurochem (1997) 68: 26–32. | PubMed | ISI | ChemPort | |
| 47. | Tombran-Tink J, Chader GG & Johnson LV. PEDF: A pigment epithelium-derived factor with potent neuronal differentiative activity. Exp Eye Res (1991) 53: 411–414 10.1016/0014-4835(91)90248-D. | Article | PubMed | ChemPort | |
| 48. | Tombran-Tink J, Mazuruk K, Rodriguez IR, Chung D, Linker T, Englander E & Chader GJ. Organization, evolutionary conservation, expression and unusual Alu density of the human gene for pigment epithelium-derived factor, a unique neurotrophic serpin. Mol Vis (1996) 2: 11. | PubMed | ChemPort | |
| 49. | Tombran-Tink J, Shivaram SM, Chader GJ, Johnson LV & Bok D. Expression, secretion, and age-related downregulation of pigment epithelium-derived factor, a serpin with neurotrophic activity. J Neurosci (1995) 15: 4992–5003. | PubMed | ChemPort | |
| 50. | Tresini M, Pignolo RJ, Allen RG & Cristofalo VJ. Effects of donor age on the expression of a marker of replicative senescence (EPC-1) in human dermal fibroblasts. J Cell Physiol (1999) 179: 11–17 10.1002/(SICI)1097-4652(199904)179:1<11::AID-JCP2>3.3.CO;2-Z. | Article | PubMed | ISI | ChemPort | |
| 51. | Tsuchida Y. The effect of aging and arteriosclerosis on human skin blood flow. J Dermatol Sci (1993) 5: 175–181. | Article | PubMed | ChemPort | |
| 52. | Weninger W, Uthman A & Pammer J. et al Vascular endothelial growth factor production in normal epidermis and in benign and malignant epithelial skin tumors. Lab Invest (1996) 75: 647–657. | PubMed | ISI | ChemPort | |
| 53. | Wight TN, Raugi GJ, Mumby SM & Bornstein P. Light microscopic immunolocation of thrombospondin in human tissues. J Histochem Cytochem (1985) 33: 295–302. | PubMed | ISI | ChemPort | |
| 54. | Yabe T, Wilson D & Schwartz JP. NFkappaB activation is required for the neuroprotective effects of pigment epithelium-derived factor (PEDF) on cerebellar granule neurons. J Biol Chem (2001) 276: 43313–43319 10.1074/jbc.M107831200. | Article | PubMed | ISI | ChemPort | |
| 55. | Yamagishi S, Inagaki Y, Amano S, Okamoto T, Takeuchi M & Makita Z. Pigment epithelium-derived factor protects cultured retinal pericytes from advanced glycation end product-induced injury through its antioxidative properties. Biochem Biophys Res Commun (2002) 296: 877–882 10.1016/S0006-291X(02)00940-3. | Article | PubMed | ISI | ChemPort | |
| 56. | Yamaura H & Matsuzawa T. Decrease in capillary growth during aging. Exp Gerontol (1980) 15: 145–150 10.1016/0531-5565(80)90086-8. | Article | PubMed | ISI | ChemPort | |
Acknowledgments
This work was supported by NIH grants AG19490 (M. K. F.) and AG00378 and AG20955 (V. J. C.) and by the Lankenau Hospital Foundation. We thank M. Liu and Dr R. G. Allen for help with data and statistical analyses. We thank Drs John Furth, Christian Sell, and Felipe Sierra for helpful discussions and critical reading of the manuscript.
