Nerve growth factor levels in cultured human skin cells: effect of gestation and viral transformation

ELSEVIER

Neuroscience Letters 184 (1995) 157-160

HtHTERIOHSCIHC[

Nerve growth factor levels in cultured human skin cells: effect of gestation and viral transformation P. A n a n d a,*, P. F o l e y a, H . A . N a v s a r i a b, D. S i n i c r o p i c, R . E . W i l l i a m s - C h e s t n u t c, I . M . L e i g h b aDepartment of Neurology, Royal London Hospital, Whitechapel, London E1 1BB, UK bDepartment of Experimental Dermatology, London Hospital Medical College, Royal London Hospital, Whitechapel, London El 1BB, UK eDepartment of Medicinal and Analytical Chemistry, Genentech Inc., San Francisco, CA, USA

Received 1 July 1994; revised version received 24 October 1994; accepted 25 November 1994

Abstract

Extracts of cultured human keratinocytes and fibroblasts were assayed for nerve growth factorqike immunoreactivity (NGF) by a specific enzyme-linked immunoabsorbant assay. NGF levels were higher in primary cultured keratinocytes than in freshly isolated keratinocytes or culture through multiple passages. Viral transformation of keratinocytes with the human papilloma 'virus (HPV16) significantly increased NGF levels, whilst transformation with the simian virus (SV40), which induces simple epithelial differentiation, reduced the concentration of NGE Passaged epidermal keratinocytes contained more than twice as much NGF as did passaged fibroblasts. Oral keratinocytes and fibroblasts, and psoriatic fibroblasts, all from high turnover tissues, did not contain significantly different levels of NGF in culture than dermal keratinocytes or fibroblasts. Foetal fibroblasts contained five times as much NGF as did adult fibroblasts. These results suggest that basal keratinocytes are a major but not sole source of NGF in human skin, and that NGF may play a role in human skin development. Keywords: Nerve growth factor; Human keratinocytes; Fibroblasts; Skin development

Nerve growth factor (NGF) is produced by skin, and is neurotrophic to sympathetic neurons and a subset of sensory neurons [4,10,21]. Changes of NGF expression in skin may be associated with pathological conditions: skin NGF levels were shown to be undetectable in a sensory and autonomic neuropathic syndrome [2], and raised levels of NGF associated with inflammatory or hyperalgesic conditions in animal models (see Ref. [11]). For these reasons, it is important to establish in vitro systems to test modulation of NGF synthesis and secretion in skinderived cells. Previous bioassay and NGF m R N A studies have shown that cultured skin cells, including cultured human keratinocytes, can synthesise and secrete NGF [5,22,24]. A study using an anti-mouse NGF enzyme-linked immunosorbant assay (ELISA) of human biopsies indicated that there may be regional differences in cutaneous NGF levels [2]. Some of these studies need re-evaluation in the * Corresponding author, ,Tel.: +44 71 377 7234/7000; Fax: +44 71 377 7008.

light of the discovery of a family of neurotrophins structurally related to NGF [7-9] which can interfere in NGF bioassays, and bind to antibodies raised against NGF [ 1]. In this study, we have measured levels of NGF in normal and transformed human skin cells in culture, using a characterised specific anti-human recombinant NGF assay. Epidermal keratinocytes were cultured by the methods described by Wu et al. (1982), as modified by Tatnall et al. (1990). Briefly, single cell suspensions, derived from normal adult human breast and abdominal skin by trypsinisation were cultured in the presence of a feeder layer of gamma-irradiated 3T3 cells, in a medium comprising a 3:1 mixture of Dulbecco's modified Eagle's medium (DMEM) and H a m ' s F 12 medium with 10% fetal calf serum (FCS), supplemented with the following growth factors: hydrocortisone 0.4 mg/ml, cholera toxin 10-1°M (ICN Biomedicals, UK), transferrin 5 mg/ml, triiodothyronine 2 × 10-~l M and epidermal growth factor 20 ng/ml (Flow Laboratories, UK). Primary keratinocytes were subcultured in the absence of feeder cells, at a

0304-3940/95/$09~50 © 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0304-3940(94)1 I 195-L

158

P. Anand et al. / Neuroscience Letters 184 (1995) 157-160

seeding density of 1 × 105 cells per plate for 5 days in the standard keratinocyte medium, washed with calcium-free phosphate buffered saline (PBS) and switched to low calcium medium (<0.05 mmol) [15] for 3 days to induce desquamation of suprabasal cells, leaving a keratinocyte monolayer. Keratinocytes from normal buccal mucosa, were treated in the same way. Studies were performed using the initial keratinocyte suspension (uncultured skin keratinocytes) and the same keratinocytes at multiple passages. A line of keratinocytes immortalised with SV 40 (SVK 14) [20] was used between the 10th and 20th passage and was cultured in RPMI 1640 plus 10% FCS. Oral keratinocytes immortalised with human papilloma virus 16 (T121CN) [17] were also studied and cultured in DMEM plus 10% FCS. Foetal (12-16weeks gestation) fibroblasts, psoriatic fibroblasts (involved skin), and buccal fibroblasts were cultured as previously described [18]. All fibroblasts were grown in DMEM plus 10% FCS. All cultured cells were treated as follows. They were washed in PBS, and then resuspended in 1 ml of PBS. A 100/.tl aliquot was taken for protein analysis, spun down and resuspended in 0.5 M NaOH. The remaining 900/~ 1 of cells in PBS were resuspended in 500/tl of NGF extraction buffer (100 mM Tris-HC1, 0.2 M NaCI, 2% BSA, 0.05% sodium azide, 1% Triton X-100, l mM phenylmethylsulfonyl fluoride, 4 m M ethylenediaminetetraacetic acid, 7/~g/ml bovine aprotinin, pH 7.0) and frozen at -20°C. The samples were kept at -70°C until assayed. Protein levels were measured by the method of Lowry et al. [ 13] adapted for microtiter plates (Costar EIA/RIA plate). In each well 50/~1 of sample or standard was added to 250~1 of freshly prepared buffer (185 mM NazCO 3, 98 mM NaOH, 400/zM CuSO4, 1.4 mM sodium potassium tartrate) followed by 10/zl of colour reagent (Folin reagent diluted 2-fold with distilled water). On each plate there was a standard curve of BSA from 10/~g/ml to 10 mg/ml. Standards and samples were assayed in duplicate. The plates were read on a microtiter plate reader (Anthos 200, Labtec Instruments) at 670 nm after 30 rain. The protein concentrations of the samples were assessed by comparing their readings with those of the standard curve. Nerve growth factor-like immunoreactivity (NGF-LI) was measured by a fluorometric ELISA using recombinant human NGF (rhNGF) to set standards. The NGF antibody was obtained from a mouse hybridoma and raised against rhNGF (Genentech, Inc.). Microtitre plate wells were coated with 100#1 of anti-rhNGF mAb at a dilution of 125 ng/ml for 18 h at 4°C. After washing the plates, non-specific binding sites were blocked with PBS containing 2% bovine serum albumin and 0.05% sodium aside for 2 h at 37°C. After washing the plates, standards and samples were added (50/~1, in duplicate), and incubated for 18 h at 4°C. After the wells were washed, 100 ~1 of biotinylated anti-rhNGF

mAb was added to each well. The plates were then incubated for 2 h at 37°C after which they were again washed, and 100/~l of streptavidin-beta-galactosidase added to each well. After a further wash, 50#1 of substrate solution (4-methylumbelliferyl-beta-D-galactopyranoside) was added, and the plates incubated at room temperature in the dark for 24 h. The reaction was stopped by the addition of 200ktl of 0.15 M glycine at pH 10.5, and the fluorescence read at an excitation wavelength of 355 nm and an emission wavelength of 480 nm. The detection limit of the assay was 5 pg/ml. There was no appreciable cross-reactivity in the ELISA when the following molecules were tested at concentrations up to 1000-fold greater than rhNGF: BDNF, NT3, NT4/5, CNTF, IGF-1, insulin and TGF-beta (Sinicropi and Williams-Chestnut, unpublished data). The data were analysed using Student's t-test. The results of the study are presented in Table 1. Uncultured freshly dissociated and passaged keratinocytes contained significantly less NGF than the same ceils in primary culture (P < 0.001), but no significant difference could be found between cultured oral and epidermal keratinocytes. The HPV16 immortalised line contained significantly increased NGF levels (P < 0.05), whilst the SVK14 cells showed reduced concentration of NGF with respect to passaged skin keratinocytes (P < 0.001). Passaged epidermal keratinocytes contained more than twice as much NGF as did passaged normal adult fibroblasts (P < 0.05). Foetal fibroblasts contained 5 times as much NGF as did adult fibroblasts, on a per gram protein basis (P<0.01). Neither psoriatic nor oral fibroblasts contained significantly different levels of NGF compared to normal dermal fibroblasts. NGF was not detected in any of the media used for cell cultures. The 3T3 feeder cells contained very high levels of NGF; gamma-irradiation led to a dramatic fall of NGF levels. These cells were removed by passaging and the presence of low calcium media. This study demonstrates the production of NGF by human keratinocytes and fibroblasts following culture, using a specific enzyme-linked immunoabsorbant assay. Primary cultures of rapidly growing keratinocytes contained much more NGF than did either uncultured keratinocytes or those which had been passaged and were confluent. Using a bioassay, Tron et al. (1990) reported NGF levels in freshly isolated murine keratinocytes (Bal B/c) of about 100 pg/ml; our NGF levels in freshly isolated human keratinocytes are consistent with these results, assuming a protein concentration per unit wet weight of about 100 mg/g. Di Marco et al. (1991) showed that the expression of NGF mRNA was maximal when cultured human keratinocytes were in the exponential phase of growth. Cultured keratinocytes appeared to contain equivalent amounts of NGF whether they originated from oral epithelium or from skin. As in vitro systems provide optimal proliferation conditions, they may fail to

P. Anand et al. / Neuroscience Letters 184 (1995) 157-160 Table 1 NGF levels in cultured human skin cells NGF concentration (pg/g protein) Mean _+SEM Keratinocytes Fresh skin keratinocytes (uncultured) Primary cultured skin keratinocytes Cultured skin keratinocytes P1-4 Oral keratinocytes T I 2 I C N cells SVKI4 cells Fibroblasts Normal fibroblasts P2-9 Foetal fibroblasts Psoriatic fibroblasts Oral fibroblasts 3T3 cells Irradiated 3T3 cells

24984 +_5984 122808 - 4610 18443 +_ 2493 15124 + 2937 45194 _+ 10767 1926 -!-_384

7480 38035 6871 2568 >150236 3215

+ 2597 _+8529 _+919 _+ 1717 --2-29979 _+ 331

n

6 3 4 6 5 4

5 5 5 3 4 3

The data are means _+ SEM which were calculated from the values which were obtained from samples (number in parentheses) which were assayed in duplicate.

discriminate between dermal keratinocytes and those from a tissue with high turnover in vivo, such as the oral epithelium. The keratinocyte line transformed by SV40 expressed simple epithelial differentiation [20], and these keratinocytes had decreased NGF content. The oral keratinocyte line transformed with the human papilloma virus (HPV16) showed an over-production of NGF. Since HPV transformation down-regulates expression of p53 and RB by binding to both HPV16 E6 and E7 proteins, the relation of NGF production Io p53 mutation/stabilisation in human keratinocytes requires further investigation. The levels of NGF that we have measured in fibroblasts are consistent with the report of Murase et al. (1992) who showed, using a recombinant human NGFELISA, that levels of NGF are modulated in a growthdependent manner in the human fibroblast cell line WS- 1. In our study, foetal hutnan fibroblasts in culture consistently contained five times more NGF than normal adult fibroblasts grown in the same culture conditions, which indicates a role for NGF in developing skin. We have not as yet measured NGF levels in foetal cultured keratinocytes or foetal skin. It has been shown that NGF synthesis in developing mouse skin begins with the arrival of its sensory innervation, and that the bulk of the NGF synthesis, measured as NGF mRNA, occurs in the epithelium [4]. Apoptosis of neurons in the trigeminal ganglion occurs when the cutaneous projection has started to transport NGF back to the perikarya. The normal development of sub-populations of primary afferent fibres is dependent on NGF (see Ref. [11]). In adult rats, however, although sympathetic adrenergic fibres continue to depend on NGF

159

for survival, unmyelinated sensory neurons require NGF only for phenotypic properties, including the induction of neuropeptides substance P and calcitonin gene-related peptide [ 12]. Normal adult cultured epidermal keratinocytes contained more than four times as much NGF as did fibroblasts from the same tissue, and differentiation of keratinocytes by transformation decreased NGF levels. In accord, in situ hybridisation studies in rat skin [14], and immunostaining studies in human skin (our unpublished observations), demonstrate NGF production in the basal keratinocyte layer. The same region shows immunoreactivity for the low-affinity NGF receptor [ 16], which is coexpressed with a high-affinity NGF receptor trkA by unmyelinated sensory neurons. These studies support the view that NGF is trophic to the unmyelinated subpopulation of sensory fibres, which terminate in or adjacent to the region of the basal keratinocyte layer. Recent studies have suggested that NGF may stimulate keratinocyte growth in an autocrine manner [6],, and nerve sprouting in classical neurotrophic fashion in denervated skin [14]. A number of non-neuronal cells other than keratinocytes in skin may also respond to NGF. Melanocytes in culture express NGF receptors, and NGF may regulate their differentiation and migration [24]. Merkel cells have been shown to possess NGF-receptor immunoreactivity [16]. In vitro systems provide a model to study the molecular basis of such paracrine interactions, as well as the role of cytokines and other agents in the modulation of NGF activity in skin. Such studies may be helpful in understanding the role of NGF in diverse pathological processes, including sensory and autonomic neuropathies, skin inflamtnation and pigmentation, and wound healing. PA thanks the Special Trustees of the London Hospital Medical College and the Medical Research Council UK for financial support. [1] Acheson, A., Barker, P.A., Alderson, R.F., Miller', F.D. and Murphy, R.A., Detection of brain-derived neurotrophic factor-like activity in fibroblasts and Schwann ceils: inhibition by antibodies to NGF, Neuron, 7 (1991) 265-275. [2] Anand, P., Rudge, P., Mathias, CJ., Springall, D.R., Ghatei, M.A., Nfiher-NoE, M., Sharief, M., Misra, V.P., Polak, J.M., Bloom, S.R. and Thomas, P.K, New autonomic and sensory neuropathy with loss of adrenergie sympathetic function and sensory neuropeptides, Lancet, 337 ( [99 [) 1253-1254. [3] Brennan, J.K., Mansky, J., Roberts, G. and Lichtman, M.A., Improved methods for reducing calcium and magnesium concentralions in tissue culture medium: application to studies of lymphoblast proliferation in vitro, In Vitro, 11 (1975) 354-359. [4] Davies, A.M., Brandtlow, C., Heumann, R., Korsching, S., Rohrer, H. and Thoenen, H., Timing and site of nerve growth factor synthesis in developing skin in relation to innervation and expression of the receptor, Nature, 326 (1987) 353-:358. [5] Di Marco, E., Marcbisio, P.C., Bondanza, S., Franzi, A.T., Cancedda, R. and De Luca, M., Growth regulated synthesis and secretion of biologically active nerve growth ['actor by human keratinocytes, J. Biol. Chem., 266 (1991) 21718-21722.

160

P. Anand et al. / Neuroscience Letters 184 (1995) 157-160

[6] Di Marco, E., Mathor, E., Bondanza, S., Cutuli, N., Marchisio, P.C., Cancedda, R, and De Luea, M., Nerve growth factor binds to normal human keratinocytes through high and low affinity receptors and stimulates their growth by a novel autocrine loop, J. Biol. Chem., 268 (1993) 22838-22846. [7] Hofer, M., Pagliusi, S.R., Hohn, A., Leibrock, J. and Barde, Y.A., Regional distribution of brain derived neurotrophic factor mRNA in the adult mouse brain, EMBO J., 9 (1990) 2459-2464. [8] Hohn, A., Leibrock, J., Bailey, K. and Barde, Y.-A., Identification of a novel member of the nerve growth factor-brain derived neurotrophie factor family, Nature, 344 (1990) 339-341. [9] Ib~nez, C.F., Hallbrook, F., Godeau, F. and Persson, H., Expression of neurotrophin-4 mRNA during oogenesis in Xenopus laevis, Int. J. Dev. Biol., 36 (1992) 239-245. [10] Korsching, S. and Thoenen, H., Nerve growth factor in sympathetic ganglia and corresponding target organs of the rat: correlation with density of sympathetic innervation, Proc. Natl. Acad. Sci. USA, 80 (1983) 3513-3516. [11] Lewin, G. and Mendell, L.M., Nerve growth factor and nociception, Trends Neurosci., 16 (1993) 353-359. [12] Lindsay, R.M. and Harmar, A.J., Nerve growth factor regulates expression of neuropeptide genes in adult senso~ neurones, Nature, 337 (1989) 362-364. [13] Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275. [14] Mearow, K.M., Kril, Y. and Diamond, J., Increased NGF mRNA expression in denervated rat skin, NeuroReport, 4 (1993) 351354. [15] Murase, K., Murakami, Y., Furukawa, Y. and Hayashi, K., Human fibroblast cells synthesise and secrete nerve growth factor in culture, Biochem. Biophys. Res. Commun., 184 (1992) 373-379.

[16] Narisawa, Y.. Hashimoto, K., Nihei, Y. and Pietruk, T., Biological significance of dermal Merkel cells in development of cutaneous nerves in fetal human skin, J. Histochem. Cytochem., 40 (1992) 65-71. [17] Sexton, C.J., Proby, C.M., Banks, L, Stables, J.N., Powell, K., Navsaria, H. and Leigh, I.M., Characterisation of factors involved in human papillomavirus type 16-mediated immortalisation of oral keratinocytes, J. Gen. Virol., 74 (1993)755-761. [18] Sly, W.S. and Grubb, J., Isolation of fibroblasts from patients, Methods Enzymol., 58,444--450. [19] Tatnall, F.M., Leigh, I.M. and Gibson, J.R., Comparative study of antiseptic toxicity on basal keratinocytes, transformed human keratinocytes and fibroblasts, Skin Pharmacol., 3 (1990) 157-163. [20] Taylor-Papadimitriou, J., Purkis, P., Lane, E.B. et al., Effects of SV40 transformation on the cytoskeleton and behavioural properties of human keratinocytes, Cell Diff., 11 (1972) 169-179. [21] Thoenen, H., Brandtlow, C. and Heumann, R., The physiological function of nerve growth factor in central nervous system: comparison with the periphery, Rev. Physiol. Biochem. Pharmacol., 109 (1987) 145-159. [22] Tron, V.A., Coughlin, M.D., Jang, D.E., Stanisz, J. and Sauder, D.N., Expression and modulation of nerve growth factor in murine keratinocytes (PAM 212), J. Clin. Invest., 85 (1990) 1085-1089. [23] Wu, J.-Y., Parker, L.M., Binder, N.E., et al., The mesothelial keratins: a new family of cytoskeletal proteins identified in cultured mesothelial cells and nonkeratinizing epithelia, Cell, 31 (1982) 693-703. [24] Yaar, M., Grossman, K., Eller, M. and Gilchrest, B.A., Evidence for nerve growth factor mediated paracrine effects in human epidermis, J. Cell. Biol., 115(1991) 821-828.