Rapid desensitization of PC12 cells stimulated with high concentrations of extracellular S100

Pergamon

PII:

Neuroscience Vol. 89, No. 3, pp. 991–997, 1999 Copyright  1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/99 $19.00+0.00 S0306-4522(98)00386-8

RAPID DESENSITIZATION OF PC12 CELLS STIMULATED WITH HIGH CONCENTRATIONS OF EXTRACELLULAR S100 S. FULLE,*0 M. A. MARIGGIO v ,†0 S. BELIA,*0 C. PETRELLI,‡ P. BALLARINI,‡ S. GUARNIERI† and G. FANO v †§ *Dipartimento di Biologia Cellulare e Molecolare Sezione di Fisiologia e Biofisica, Universita` di Perugia, Perugia, Italy †Dipartimento di Scienze Biomediche, Lab. Fisiologia Cellulare, Universita` ‘‘G. D’Annunzio’’, Via dei Vestini 29, 66013 Chieti, Italy ‡Dipartimento di Biologia Cellulare Animale e Molecolare, Universita` di Camerino, Camerino, Italy Abstract––Undifferentiated PC12 cells undergo apoptosis, via a calcium-induced calcium release mechanism, when the calcium-binding protein purified from bovine brain (native S100) is present in micromolar concentration in the medium. This process begins when S100 binds to specific membrane binding sites and involves up to 50% of the cell population. In the experiments reported here, we demonstrate that, by utilizing [3H]S100, the S100 protein can be displaced from its binding sites only during the first 10 min of incubation. This fact is due to an internalization mechanism, having a time-course with a plateau after 10–20 min of incubation. The native form of S100 is a mixture of two different S100 isoforms: S100A1 (20%) and S100B (80%). Using confocal microscopy and monoclonal antibodies, we demonstrated that only one of these isoforms, S100A1, was autoexpressed in more than 50% of the PC12 cells analysed. After cell incubation with 2 µM native S100, S100B also appears in PC12 cells, with a maximum presence after 10 min of incubation. This fact seems to indicate that this isoform, at least, is effectively translocated when stimulated with external native S100. From the data reported, it is possible to hypothesize that, in PC12 cells, a possible homeostatic mechanism is present that can counteract the effect of a continuously applied lethal stimulus (stimuli) on cell viability.  1999 IBRO. Published by Elsevier Science Ltd. Key words: S100, [3H]S100, apoptosis, PC12, desensitization.

S100 protein, mainly the B isoform, is a member of a Ca2+-binding protein family which affects the survival and/or death of brain cells. It is present in the mammalian nervous system as a mixture of two dimers (S100A1 and S100B),9 in both cellular (glial and neuronal cytoplasm) and extracellular compartments. Increased levels of S100 extracellular pool have been observed in several regions of the brain not only during physiological development,12 but also in the cerebral fluids from patients with Alzheimer’s disease,8,13 Down’s syndrome1 or acquired immunodeficiency syndrome-associated dementia.11 Under these conditions, the extracellular increase of S100 is correlated with an increase in neuronal death.2 The S100 family consists of 17 members,10 most of which appear to function as intracellular calciummodulated protein, because they contain two EFhand calcium-binding sites. Furthermore, different §To whom correspondence should be addressed. 0These authors contributed equally to this work. Abbreviations: EGTA, ethyleneglycolbis(aminoethyl ether)tetra-acetate; [3H]S100, tritium-labelled form of S100; NGF, nerve growth factor; PBS, phosphate-buffered saline; PC12, cell line from rat pheochromocytoma; S100, calcium-binding protein. 991

possible intracellular targets for S100 have been determined by using cell-free systems as substrates, but conclusive tests which demonstrate that this also occurs in intact cells are lacking.5 However, S100B, the isoform present in the nervous system, and some other S100 species, may also cause modifications in cell activity when added to the extracellular space of intact cells.3 In the past, several in vitro systems have been utilized to study the cellular aspects of the action mechanism by which S100 can induce significant events related to cell survival. In particular, it has been demonstrated that, in the presence of nanomolar S100 concentrations, neurite extension is induced, as well as other differentiative features, in primary neuronal cell cultures.14 In contrast, when higher levels of the protein (>0.5 µM) are present in the medium, the effects induced in PC12 cells and in some neuroblastoma cell lines may be directly related to the triggering of programmed cell death.7 This event is precluded if nerve growth factor (NGF) is also present in the medium.4 The effects attributed to extracellular S100B levels, which are derived from a possible internalization of this protein, can be hypothesized. It is possible, as with other growth factors, that the extracellular pool

992

S. Fulle et al.

of S100 may regulate the cell activity by first inducing activation of a second messenger pathway as an initial response, followed by a triggering of an agonist/receptor internalization process. To study these sequences, it is necessary to have a cellular model that satisfies the following prerequisites: (i) specific sensibility to the S100B extracellular pool; (ii) absence of S100B autoexpression. As already described, brain-derived S100 (mainly the S100B isoform) can induce apoptosis in a neuron-like PC12 cell line. This effect starts when S100 binds to specific sites on cells,3 which then induce a rapid calcium-induced calcium release-dependent increase of the intracellular calcium concentration.7 Zimmer and Landar15 recently observed that, in PC12 cells, only S100A1 is genetically present, while S100B is completely lacking. These facts indicate that PC12 cells are a suitable substrate for studying the S100B internalization process. EXPERIMENTAL PROCEDURES

Materials The experiments were carried out on PC12 cells (clone A1), a cell line derived from a rat pheochromocytoma which can differentiate into a sympathetic phenotype when incubated in the presence of NGF.6 The cells were cultured as described previously in RPMI 1640 medium supplemented with 10% horse serum and 5% fetal calf serum (Gibco BRL-Life Technologies, Paisley, U.K.). S100, a mixture of the two isoforms S100A1 and S100B, was purified from bovine brain as described previously;5 3 [ H]S100 (specific activity 4.31 Ci/mmol and where indicated 9.38 Ci/mmol) was prepared as already reported.3 EGTA, poly--lysine, antibodies to the S100 á- and â-subunits, trypsin, Tween 20 and antibody to mouse immunoglobulin G were obtained from Sigma (St Louis, MO, U.S.A.). Displacement and internalization methods PC12 cells (2.5106) were incubated for 5 or 30 min in an oscillating bath at 37C in medium-plus-sera containing 50 nM [3H]S100, in a final volume of 0.5 ml. At the end of the incubation time, different concentrations (25–250 nM) of unlabelled S100 were added to the medium in order to measure the [3H]S100 displacement from the S100 specific binding sites. The cells were rapidly filtered through a Bio-Rad apparatus (Richmond, CA, U.S.A.) utilizing Whatman GF/C filters (Springfield Mill, Maidstone, Kent, U.K.). The filters were washed twice with cold phosphatebuffered saline (PBS) solution, placed in scintillation liquid and the radioactivity bound to the cells determined using a liquid scintillation counter (Beckman LS-1800). Another experimental set was prepared to measure the time dependence of [3H]S100 displacement. For this, after 5 or 30 min of incubation in the presence of 100 nM [3H]S100, an excess of 10 µM native S100 was added to each sample, which were then filtered after an additional incubation period (0– 45 min). The filters were then processed and their radioactivity read as described above. A third series of experiments was performed to calculate the effect of the addition of 1 µM unlabelled S100 on the 100 nM [3H]S100 internalization. First, to control the effect of trypsin and EGTA on the binding of [3H]S100 on the cell membrane, we processed samples of 2.5106 cells in a final volume of 0.5 ml preincubated for 5 min at 37C in medium containing 0.05% trypsin in the presence or absence of 1 mM EGTA. Then,

Fig. 1. Displacement of [3H]S100 by different concentrations of native S100. PC12 cells (2.5106) were incubated with 50 nM [3H]S100 for 30 min (a) or 5 min (b) at 37C, and then with different concentrations (0–250 nM) of unlabelled S100. The points (meanS.D., n=5) represent the cell-associated [3H]S100 (pmol/107 cells). 100 nM [3H]S100 was added to the cells. After 5 min incubation, the cells were filtered, washed with PBS and the radioactivity of the filters was read. Other samples of 2.5106 cells in a final volume of 0.5 ml were incubated for 5 min in the presence of 100 nM [3H]S100. After this time, 1 µM unlabelled S100 was added to the medium at selected times (0–60 min). The cells were then filtered and washed as described above. After this, a solution containing 1 mM EGTA and 0.05% trypsin was added to the filters for 5 min to avoid any non-specific bound [3H]S100. After a second filter washing with PBS plus 1 mM EGTA, the radioactivity was read using the above-described procedure. Immunofluorescence preparations PC12 cells (usually 50,000), processed essentially as reported by Zimmer and Landar,15 were attached on glass coverslips (previously treated with 10 µg/ml poly--lysine) and incubated with 2 µM native S100 at 37C for different incubation times (0–60 min). After a rapid (by means of a pipette) double washing with Ca2+-free PBS containing 10 µM EGTA, the cells were fixed with 100 µl cold absolute methanol for 8 min at
20C. This solution was then substituted with PBS containing 0.1% Triton X-100 and, after 5 min, the glasses were washed (three times) with PBS. To introduce the antibodies, the cells were first incubated for 1 h with 10% horse serum at room temperature and then with the first monoclonal antibody (anti-S100A1 or antiS100B), diluted 1:100 with PBS. After three washings with PBS (10 min each), the cells were incubated with the second antibody (anti-mouse immunoglobulin G, fluorescein isothiocyanate conjugate) diluted 1:100 with PBS for 45 min at

S100 desensitization in PC12 cells

993

Fig. 2. Displacement of [3H]S100 after different times of incubation in the presence of unlabelled S100. PC12 cells (2.5106) were incubated with 100 nM [3H]S100 for 5 min at 37C and then with 10 µM unlabelled S100 for different times (0–45 min). The points (meanS.D., n=5) indicate the residual cell-associated [3H]S100. The data reported are representative of one of three different experiments.

37C. Finally, after another three washings (5 min in the presence of 0.1% Tween 20) with PBS at room temperature, the glasses were placed on objective slides, and were ready to be processed by confocal microscopy. Confocal microscopy parameters The samples were observed with a confocal microscope (Bio-Rad MCR 600) connected to an inverted optical microscope (Nikon Diaphot) with an immersion objective (60; NA 1.4). The image acquisitions were performed with a krypton–argon laser source using a wavelength selection filter (BHS) corresponding to an excitation wavelength of 488 nm. To attenuate the laser intensity, the neutral density filter was set at 10% transmission. To minimize the artefacts during the acquisition, all images were collected activating photon counting at the same mode, setting the accumulate filter at 35 scans. The software used was COMOS (Bio-Rad, Richmond, CA, U.S.A.). The peculiarity of this apparatus is that it effects a stratified observation of the sample, in that the laser illuminates a single 1-µm-thick section. The general image was obtained by superimposing the single sections. RESULTS

Fig. 3. S100 internalization. (a) Effect of 0.05% trypsin and/or 1 mM EGTA on [3H]S100 binding in PC12 cells. The cells (2.5106) were incubated in medium in the presence of 0.05% trypsin with or without 1 mM EGTA. After this time, 100 nM [3H]S100 was added for 5 min. Each bar represents the meanS.D. (n=5; P<0.05 or P<0.01 indicate statistical significance with respect to the presence of 1 mM EGTA; **P<0.01 with respect to the presence of 0.05% trypsin). (b) Effect of 1 µM unlabelled S100 on the membrane association of 100 nM [3H]S100. PC12 cells (2.5106) were incubated with 100 nM [3H]S100 (specific activity 9.38 Ci/mmol) for 5 min at 37C and then with 1 µM unlabelled S100 for different times (0–60 min). After incubation, the cells were filtered and then incubated for 5 min with 0.05% trypsin plus 1 mM EGTA. The points (meanS.D., n=3) indicate the residual radioactivity present in the cells (expressed as c.p.m./107 cells).

Binding experiments As demonstrated previously, the effect of S100 in undifferentiated PC12 cells begins with the binding of this protein with specific binding sites, which became saturated at 200 nM.3 The available kinetic data seem to indicate the presence, in both undifferentiated and differentiated PC12 cells, of at least one class of high-affinity binding sites, having Kd values of 9.4 and 18.910
8 M, respectively.3,4 Figure 1a reports data derived from experiments in which 2.5106 cells were incubated for 30 min in the presence of 50 nM [3H]S100, a 3H-labelled form of S100 which has the same chemical and physical properties and induces the same biological effects as the native protein.3 Subsequently, different concentrations of unlabelled S100 (0–250 nM) were added and, after another 30 min of incubation, cells were rapidly filtered. As can be seen from the graph, increased levels of unlabelled S100 in the cell suspension did

not significantly modify the radioactivity bound to the cells. Results from similar experiments were significantly different, when the preincubation time in the presence of 50 nM [3H]S100 alone was limited to 5 min. Under these conditions, the radioactivity measured after addition of 50–250 nM S100 was drastically reduced, to a level of approximately 50% of the initial value (Fig. 1b). Considering the characteristics of the S100 binding, this indicates that [3H]S100 was displaced from its specific sites. The results of this first series of experiments seem to demonstrate that the cell-bound [3H]S100 was available to interact with the external solution for a very short time. This consideration is also supported by the data reported in Fig. 2, in which it is possible to note that if the cells were incubated for only 5 min with 100 nM [3H]S100, a subsequent addition of 10 µM unlabelled S100 induced an apparent displace-

994

S. Fulle et al.

Fig. 4. Confocal analysis with different S100 monoclonal antibodies. The images (1 µm thick), derived from an intermediate section (4 µm from the external surface), are of three different cellular pools which were incubated without antibodies (a), with the monoclonal antibody for S100B (b), or with anti-S100A1 alone (c). Scale bar=10 µm.

ment essentially completed in 20 min. Under these experimental conditions, a long preincubation time with [3H]S100 (up to 20 min) prevented any modification in radioactivity (data not shown). Even if, in PC12 cells, the S100 binding sites are saturated at protein concentrations starting from 200 nM, we also verified the possibility that S100 concentrations of 1 µM, known to induce apoptosis in PC12 cells,7 could modify the displacement just described. To verify this, we preincubated the cells for 5 min with 100 nM [3H]S100 as a tracer. After this, 1 µM unlabelled S100 was added to the cells for selected periods (0–60 min). At the end of each incubation period (in order to prevent further S100 binding to the external cell membrane), we washed the cells with 1 mM EGTA and 0.05% trypsin which, as shown in Fig. 3a, are able to block the binding of S100 to PC12 cells (see Experimental Procedures for more details). Under these conditions, all of the radioactivity measured must be derived from internalized (or included into the membrane) [3H]S100. Figure 3b reports the time-related level of radioactivity in the above-mentioned conditions and, as can be seen from the graph, only during the first 10 min does the radioactivity accumulation increase in a time-dependent fashion, after which the values remain statistically unchanged.

These data indicate that at least one of the S100 isoforms (S100B or S100A1) present in the mixture utilized in this experiment was involved in an internalization process. However, two important questions remain. (1) Which S100 isoform is really internalized? (2) Where is it localized in the cell? We attempted to answer these questions by carrying out a series of experiments using confocal microscopy with a double antibody immunofluorescent technique. Confocal microscopy As a preliminary step, we checked for the presence of S100A1 and S100B in the PC12 cells using a double antibody technique (see Experimental Procedures for more details). Figure 4 shows the confocal analysis of the intermediate section (4 µm from the external surface) of three different cells (containing the fluorescent second antibody) which were incubated with no antibody, the monoclonal antibody for S100B, or anti-S100A1 alone. The figure shows that, under our conditions, and as reported previously by Zimmer and Landar,15 only S100A1 was autoexpressed in more than 50% of the PC12 cells checked. Since, in all fields examined, S100B was not present, the prerequisite for studying

S100 desensitization in PC12 cells

995

Fig. 5. Temporal analysis of S100B. Temporal confocal analysis of single, thin intermediate sections (1 µm) of different cellular pools incubated with 2 µM native S100 for selected times (0, 1, 3, 5, 10 and 30 min) and then with S100B monoclonal antibody. Scale bar=10 µm.

a possible mechanism of internalization involving S100B added in the extracellular compartments was met. A second experimental set was prepared to study the expression of the two S100 species during a brief (0–30 min) exposure of PC12 cells to 2 µM native S100 added to the medium. At selected times, the cells were washed twice in the presence of 10 µM EGTA and processed, using monoclonal antibodies (anti-S100A1 or anti-S100B), to reveal the presence of S100B or S100A1 in different cellular pools. The series of images reported in Fig. 5 shows the time analysis of different cellular pools treated with antiS100B. The figure indicates that, for the first 10 min of incubation, the intermediate confocal section shows a direct correlation between the presence of intracellular S100B and the incubation time of the cells in the presence of native S100. After this, the correlation became inverted and the fluorescence, indicating the translocation of S100B, was less evident. We also tried to measure the time-course of the presence of S100A1 after incubating different PC12 pools with 2 µM native S100 but, probably due to the high fluorescent signal caused by the autoexpression of this S100 isoform, no significant

modifications were observed (data not shown). In Fig. 6, we show the first four sections (1 µm in thickness) derived from the confocal application on S100B-responsive cells. Under these conditions, it was possible to more clearly define the distribution of the protein after 10 min of incubation with extracellular native S100 (2 µM). Thus, it seems evident that S100B is distributed in the equatorial layer of the cells examined, while the nuclear zone is never involved in the localization process of this protein. DISCUSSION

When PC12 cells are incubated in the presence of 0.5 µM or higher concentrations of the native S100 mixture (80% S100B and 20% S100A1), the apoptosis of a relatively large part of the cell population (20–50%) is a direct consequence of the binding of this protein to specific sites present in the responsive cells.3 In other agonist/receptor models, the binding of the first messenger to specific membrane sites usually induces a cascade of events, which are directly or indirectly related to a transmembrane mechanism of cell activation. However, the cell must

996

S. Fulle et al.

Fig. 6. Intracellular distribution of S100B. A sequence of four sections (1 µm thick), starting from the cellular surface obtained from confocal application on a cellular pool incubated for 10 min in the presence of 2 µM native S100 and with monoclonal antibody S100B. Scale bar=10 µm.

be rapidly and simultaneously inactivated to permit the correct development of the newly arrived message. One of the more frequently described mechanisms by which this can be obtained is the internalization of the agonist/receptor complex. This rapid event is usually completed in a few minutes. This same sequence seems to be followed in our experimental model, because both the internalization and fluorescent data confirm that, after more than 10 min of incubation, the S100 bound to the specific sites of cell membranes cannot be displaced from these sites. It is also important to note that, at the same time, the level of internalized S100B reaches its maximum. From our data, it is not possible to determine if the S100A1 isoform can also inform the responsive cells through some internalization process. The confocal analysis of the cells preincubated with S100A1 monoclonal antibody showed a high fluorescence emission in the controls (without the extracellular presence of native S100), as well as in the time-series of cells incubated in the presence of 2 µM S100. An attempt to define an intracellular localization of the two isoforms was unsuccessful. Information about the eventual participation of S100A1 in controlling some mechanisms of cell activity (including cell death) is of critical importance considering that an autocrine mechanism implicating S100A1 release from responsive cells could exist. In fact, Zimmer and Landar15 have clearly

demonstrated that the expression of S100A1 in PC12 cells increased during the NGF-induced differentiative process. This increase was attributed to a post-transcriptional mechanism, because the specific S100A1 mRNA level remained unchanged. These authors also hypothesized that S100A1 may be released by PC12 cells, thereby determining one of the first steps of apoptosis of this cell population.16 This possibility, although in accordance with previously reported data on the ability of an external S100A1 pool3 to induce apoptosis in PC12 cells, has not been demonstrated directly and, for this reason, it can only be considered an attractive hypothesis. The results presented here clearly show that, at least for the added S100B, there is direct evidence that this isoform binds to PC12 cells, causing the desensitization of activity by an internalization mechanism. Thus, a series of steps can be proposed, which may elucidate the early events by which S100 induces apoptosis in PC12 cells. The activation of responsive cells, induced directly by external S100B or indirectly by an autocrine S100A1-mediated mechanism, starts with the binding of the protein to specific binding sites present on the cell membrane. After this, a rapid and large increase of intracellular Ca2+ via a calciuminduced calcium release mechanism appears, and this stimulus triggers the pathway that activates cell death.7 Whereas the overall apoptotic processes

S100 desensitization in PC12 cells

induced by S100 in PC12 cells require at least 24–48 h to reach completion, after only a few minutes the S100 reactive system present in the cell membranes was already completely inactivated and the cells were no longer responsive. In addition, only 20–50% of the cells present in the suspension died when high S100 concentrations were present in the medium.3 This fact suggests that a homeostatic mechanism exists in a large percentage of the cell population. In this case, the presence of a rapid desensitization system would not only be helpful but quite necessary to counteract the effect of a continuously applied lethal stimulus on cell viability.

997 CONCLUSIONS

If one considers that high S100 levels are related in vivo to several physiopathological conditions of the nervous system,2,10,13 it is evident that a homeostatic system in the neurons, similar to that just described, assumes a critical role. Acknowledgements—We wish to thank Peter A. Mattei for a critical reading of the manuscript. This work was supported by research grants from M.U.R.S.T. (Ministero dell’Universita` e della Ricerca Scientifica e Tecnologica, 40%) and from the University of Chieti, Italy (60% research grant) to G.F.

REFERENCES

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

Allore R., O’Hanlon D., Price R., Neilson K., Willard H. F., Cox D. R., Marks A. and Dunn R. J. (1988) Gene encoding the subunit of S-100 protein is on chromosome 21: implications for Down Syndrome. Science 239, 1311–1313. Fano` G., Biocca S., Fulle S., Mariggio` M. A., Belia S. and Calissano P. (1995) S-100 protein: a family in search of a function. Prog. Neurobiol. 46, 71–82. Fano` G., Mariggio` M. A., Angelella P., Nicoletti I., Antonica A., Fulle S. and Calissano P. (1993) The S-100 protein causes an increase of intracellular calcium and death of PC12 cells. Neuroscience 53, 919–925. Fulle S., Mariggio` M. A., Belia S., Nicoletti I. and Fano` G. (1997) Nerve growth factor inhibits apoptosis induced by S-100 binding in neuronal PC12 cells. Neuroscience 76, 159–166. Fulle S., Mariggio` M. A., Fano` G., Salvatore A. M., Mercanti D., Petrelli C. and Calissano P. (1992) Activation of brain adenylate cyclase by S-100 protein via a possible interaction with G-protein(s). Neurosci. Res. Commun. 10, 37–43. Greene L. A. and Tischler A. S. (1976) Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc. natn. Acad. Sci. U.S.A. 73, 2424–2428. Mariggio` M. A., Fulle S., Calissano P., Nicoletti I. and Fano` G. (1994) The brain protein S-100ab induces apoptosis in PC12 cells. Neuroscience 60, 29–35. Marshak D. R. and Pena L. A. (1992) Potential role of S-100 in Alzheimer’s disease: a hypothesis involving mitotic protein kinases. Prog. clin. biol. Res. 379, 289–307. Scha¨fer B. W., Wicki R., Engelkamp D., Mattei M. G. and Heizmann C. W. (1995) Isolation of a YAC clone covering a cluster of nine S100 genes on human chromosome 1q21: rationale for a new nomenclature of the S100 calcium-binding protein family. Genomics 25, 638–643. Scha¨fer B. W. and Heizmann C. W. (1996) The S100 family of EF-hand calcium-binding proteins: functions and pathology. Trends biochem. Sci. 21, 134–140. Stanley L. C., Mrak R. E., Woody R. C., Perrot L. J., Zhang S., Marshak D. R., Nelson S. J. and Griffin W. S. (1994) Glial cytokines as neuropathogenic factors in HIV infection: pathogenic similarities to Alzheimer’s disease. J. Neuropath. exp. Neurol. 53, 231–238. Stewart J. A. and Urban M. I. (1972) The postnatal accumulation of S-100 protein in the mouse central nervous system. Modulation of protein synthesis and degradation. Devl Biol. 29, 372–384. Van Eldik L. J. and Griffin W. S. (1994) S100 beta expression in Alzheimer’s disease: relation to neuropathology in brain regions. Biochim. biophys. Acta 1223, 398–403. Winningham-Major F., Staecker J. L., Barger S. W., Coats S. and Van Eldik L. J. (1989) Neurite extension and neuronal survival activities of recombinant S-100b proteins that differ in the content and position of cysteine residues. J. Cell Biol. 109, 3063–3071. Zimmer D. B. and Landar A. (1995) Analysis of S100A1 expression during skeletal muscle and neuronal cell differentiation. J. Neurochem. 64, 2727–2736. Zimmer D. B., Cornwall F. H., Landar A. and Song W. (1995) The S100 protein family: history, function and expression. Brain Res. Bull. 37, 417–429. (Accepted 8 July 1998)