HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ridinger, K. Articles by Heizmann, C. W. Search for Related Content PubMed PubMed Citation Articles by Ridinger, K. Articles by Heizmann, C. W. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 12, 8686-8694, March 24, 2000

S100A13
BIOCHEMICAL CHARACTERIZATION AND SUBCELLULAR LOCALIZATION IN DIFFERENT CELL LINES^*

Katrin Ridinger, Beat W. Schäfer, Isabelle Durussel^§, Jos A. Cox^§¶, and Claus W. Heizmann

From the  Department of Pediatrics, Division of Clinical Chemistry and Biochemistry, University of Zurich, 8032 Zurich, Switzerland and the ^§ Department of Biochemistry, University of Geneva, 1211 Geneva, Switzerland

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

S100 proteins became of major interest because of their divergent cell- and tissue-specific expression, their close association with a number of human diseases, and their importance for clinical diagnostics. Here, we report for the first time the purification and characterization of human recombinant S100A13. Flow dialysis revealed that the homodimeric S100A13 binds four Ca²⁺ in two sets of binding sites, both displaying positive cooperativity but of very different affinity. Fluorescence and difference spectrophotometry indicate that the Trp/Tyr signal changes are almost complete upon binding of Ca²⁺ to the two high affinity sites, which probably correspond to the C-terminal EF-hands in each subunit. The far-UV circular dichroic signal also changes upon binding of the first two Ca²⁺. So far, the tissue distribution of S100A13 has not been well characterized. Here, we show that S100A13 is widely expressed in various types of tissues with a high expression level in thyroid gland. Using specific antisera against S100A13, high protein expression was detected in follicle cells of thyroid, Leydig cells of testis, and specific cells of brain. In human smooth muscle cells, which co-express S100A2 in the nucleus and S100A1 in stress fibers, S100A13 shows a unique subcellular localization in the perinuclear area. These data suggest diverse functions for this protein in signal transduction.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The S100 protein family is characterized by two sets of different EF-hands in the dimer and currently numbers at least 18 different members. In contrast to the ubiquitously expressed calmodulin, the expression of S100 proteins is cell- and tissue-specific. Some are expressed mainly in a few specific tissues, such as S100A3 in human hair cuticle cells (1, 2) or S100A8/A9/A12 in myeloid cells (e.g. monocytes and granulocytes) (3-5). Other members are detected in a broader range of tissues, such as S100A2 in lung, kidney, liver, cardiac, and skeletal muscle (6-8), or S100A6, which is overexpressed in many tumor cells (9, 10).

S100 proteins show different binding affinities for Ca²⁺, Cu²⁺, and Zn²⁺, yielding new clues to their functional significance in cellular events (11). In general, they can bind four Ca²⁺ per dimer and display rather low affinities, with [Ca²⁺]_0.5 values of 100-300 µM. Some parameters such as high ionic strength can decrease the Ca²⁺ affinity, e.g. in S100A1 (12) and S100A6 (13, 14). In contrast, Zn²⁺ increases the Ca²⁺ affinity of S100A12 (15) and S100B (12) and enhances the calcium-dependent association of S100B to target peptides (16). Recently, the Ndr kinase was shown to be a target protein of S100B whereby the interaction is regulated by changes in the intracellular calcium concentration (17). Hence, calcium seems to be an important regulator of S100 interactions with target proteins.

The cDNA of human and murine S100A13 was first identified by screening expressed sequence tag data bases (18). 15 genes coding for S100 proteins, including two epidermal differentiation genes with a S100 domain, are arranged in a cluster on human chromosome 1q21 (18, 19). The human S100A13 was shown to neighbor S100A1 on chromosome 1q21, at least 35 kilobase pairs apart from the subgroup of the closely linked S100A2-S100A3-S100A4-S100A5-S100A6 genes (18). Recently, we demonstrated that the S100 gene cluster also exists in mouse and is structurally conserved during evolution (20). Interestingly, in contrast to the linkage relationship between S100A8 and S100A9 and S100A3-S100A4-S100A5-S100A6, the mouse S100 genes S100A1 and S100A13 are separated in comparison with the human linkage group. Generally, the interspecies homology between the known mouse and human S100 cDNAs ranges from 85 to 95%, whereas mouse S100A13 cDNA shows only 79.6% sequence identity to the human cDNA. According to the hypothesis that S100 genes originated from a primordial gene through duplication, the lower homology of mouse and human S100A13 might suggest that the divergence between these two genes occurred earlier in evolution.

Expression of S100A13 mRNA has so far been detected in skeletal muscle, heart, kidney, pancreas, ovary, spleen, and small intestine (18). Comparing the protein sequence of human S100A13 with other S100 proteins, we observed interesting differences. The second EF hand contains an unusual lysine in the x position which might influence the Ca²⁺ binding properties. Furthermore, the last 11 C-terminal amino acids contain six lysine and two arginine residues. This positively charged C-terminal region could potentially be involved in specific protein interactions as was reported for the positive C-terminal tail of syntaxin, which interacts with synaptotagmin (21).

S100A13 seems to function in exocytosis, since it is one of the targets of two antiallergic drugs, amlexanox and cromolyn (22), which inhibit degranulation of mast cells (23). Recently, association of S100A13 with the fibroblast growth factor 1 (FGF-1)¹/p40 synaptotagmin-1 (p40Syn-1) aggregate was shown, and amlexanox is able to repress this release (24). FGF-1 is an extracellular mitogenic factor for several cell types, which is secreted e.g. in response to heat shock. These findings suggest that S100A13 might be involved in the regulation of FGF-1 and p40Syn-1 release in response to heat shock. Another possibility might be that S100A13 is secreted together with the FGF-1/p40Syn-1 aggregate. Extracellular activities have also been reported for S100A2, S100A4, S100A7, S100A8, S100A9, S100B, and S100A12.

Less is known about the biochemical features and the expression of the S100A13 protein. In order to determine the biochemical characteristics of S100A13, we purified the human recombinant protein for the first time. We then characterized the Ca²⁺ binding properties by flow dialysis and probed the conformational changes upon binding of cations. We produced and characterized different S100A13 antibodies, which allowed us to demonstrate the expression of S100A13 in specific cells of the thyroid, brain, and testis. Furthermore, we report a particular subcellular localization of S100A13 in human smooth muscle and endothelial cells by confocal laser-scanning microscopy.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

S100A13 Constructs-- The expressed sequence tag clone (IMAGE T78482) containing the human S100A13 cDNA was supplied by the HGMP Resource Center. The coding region of S100A13 was amplified by polymerase chain reaction using specific oligonucleotides containing a 5' NdeI site (5'-GTCCATATGGCAGAACCACTGA- 3') and a BamHI site (5'-CTAGGATCCGAAGAGTGCGGTTCTGCT-3') under the following conditions: 95 °C for 1 min, 59 °C for 1 min, 72 °C for 2 min, 30 cycles, followed by 10 min of 72 °C extension. The polymerase chain reaction product was cut with the corresponding restriction enzymes and cloned into the pGEMEX vector. The plasmids were sequenced by an ALF DNA sequencer (Amersham Pharmacia Biotech) to verify the open reading frame.

Northern Blot Hybridization-- A commercially available multitissue Northern blot was used (CLONTECH, Palo Alto, CA) and hybridized with [³²P]dATP-labeled (Prime-a-gene, Promega, Madison, WI) S100A13 cDNA as described by Sambrook et al. (25).

Expression and Purification of Recombinant S100A13-- The S100A13 pGEMEX construct was transformed into the Bl-21 Lys S strain for expression of the recombinant protein. After the addition of 0.5 mM isopropyl-1-thio--D-galactopyranoside, the recombinant S100 protein was purified as described by Pedrocchi et al. (14) with modifications. The flow-through of the phenyl-Sepharose column containing the S100A13 protein was dialyzed against 20 mM Tris-HCl, 2 mM Ca²⁺, pH 7.5, and applied to a MonoQ-1 column. The flow-through was concentrated and loaded onto a 1 × 100-cm column of Sephacryl S200 equilibrated in buffer A containing 50 µM Ca²⁺. The sharp peak corresponding to an apparent molecular mass of 25-30 kDa was dialyzed against 20 mM Tris-HCl buffer containing 10 µM Ca²⁺, passed over a 1.5 × 10-cm column of Q-Sepharose Fast Flow equilibrated in the same buffer, and eluted with a linear gradient of 0-0.3 M KCl. Homogeneous S100A13 protein eluted in a sharp peak at a gradient conductivity of 10 millimho, corresponding to 80 mM KCl.

Characterization of S100A13 by Mass Spectrometry, Amino Acid Analysis, and Denaturating Gel Electrophoresis-- Electrospray ionization mass spectra were obtained with a Perkin-Elmer Sciex API 365 LC/MS/MS system mass spectrometer equipped with an atmospheric pressure electrospray ion source. Amino acid analysis was performed by gas phase HCl hydrolysis and analyzed by ion exchange chromatography on a Beckman system 6300. Tricine SDS-PAGE (15%) under reducing conditions was performed as described (26). The proteins were visualized by silver staining as described (27).

Metal Ion Removal and Ca²⁺ Binding-- S100A13 was dialyzed against 50 mM Tris-HCl buffer, 150 mM KCl (buffer A) and passed over a Sephadex G25 column (0.8 × 40 cm) equilibrated in buffer A. The residual Ca²⁺ contamination in the protein fractions, checked by atomic absorption, was always less than 0.05 mol/mol of protein. All the experiments were performed in buffer A starting with this "metal-free" solution. Total Ca²⁺, Mg²⁺, and Zn²⁺ concentrations were determined with a Perkin-Elmer 2380 atomic absorption spectrophotometer. The protein concentration was determined from the ultraviolet absorption spectrum using a molar extinction coefficient, E_M, 278 nm, of 16,000 M¹ cm¹ for the dimer.² This value was determined by quantitative amino acid analysis on a sample of known optical density and is quite close to the theoretical value of 14,000 based on the content of aromatic residues.

Ca²⁺ binding was measured at 25 °C by the flow dialysis method (28) in buffer A. Protein concentrations were 10-15 µM. Treatment of the raw data and evaluation of the intrinsic metal-binding constants was as described (29). Since the Ca²⁺ binding isotherms display two sets of different affinities with positive cooperativity in each set, the data were analyzed with the Adair equation for two binding sites,

(Eq. 1)

where K₁, K₂, K₃, and K₄ are the stoichiometric association constants for the binding of the successive Ca²⁺ ions to the protein dimer.

Secondary Structure Monitored by Circular Dichroism-- CD spectra were acquired with a Jasco J-715 spectropolarimeter at 25 °C on solutions of 0.25 mg/ml protein in buffer A in a cell of 1-mm optical path. Ellipticities were normalized for the residue concentration using the relationship _MRW = _oM_r/lc, where _o is the observed ellipticity in millidegrees, M_r is the average molecular weight of an amino acid in the polypeptide (116.8 for S100A5), l is the path length in mm, and c is the protein concentration in mg/ml.

Conformational Changes Monitored by Trp Fluorescence and Near-UV Absorption-- Emission fluorescence spectra were taken with a Perkin-Elmer LS-5B spectrofluorimeter on 2 µM metal-free protein in buffer A at 25 °C with excitation at 278 nm and slits of 5 and 10 nm. 20 µM EGTA or 2 mM Mg²⁺, 2 mM Ca²⁺, 0.15 mM Zn²⁺, or 4 M guanidine-HCl was added to monitor the effect of the respective ions or to obtain the spectrum of the denatured protein.

UV spectra and difference spectra were measured with a Perkin-Elmer 5 spectrophotometer at room temperature. To the metal-free protein solution (55 µM) in buffer A, 2 mM Ca²⁺, 0.15 mM Zn²⁺, or 4 M guanidine-HCl was added. Difference spectra were normalized to an optical density of 1.0 at 278 nm.

Interaction of S100A13 with the Hydrophobic Probe TNS-- Changes in the fluorescence properties of 2-p-toluidinylnaphthalene-6-sulfonate (TNS) were monitored by fluorimetry as described (30). After incubation of 2 µM protein in buffer A containing either 20 µM EGTA, 2 mM Ca²⁺, 2 mM Mg²⁺, 0.15 mM Zn²⁺, or 0.1 mM Cu²⁺ with 40 µM TNS for 5 min, the solutions were excited at 328 nm, and the emission spectra were recorded at 25 °C with 5-nm slits. In control experiments, it was established that the metal ions do not directly affect the TNS fluorescence.

Production of Polyclonal Antibodies and Western Blot Analysis-- Antisera against human recombinant S100A13 were produced in two rabbits (2505 and 2506). Animals were immunized five times at 3-week intervals by injection of initially 500 µg of protein, followed by 250 µg of protein for further boosts in an emulsion with Hunter's Titer Max (Cyt RX, Norcross, GA). The titers of the antisera were determined using 1 µg of recombinant protein on Western blots as described by Ilg et al. (8). A dilution of maximally 1:50,000 (antisera 2505/2506) is possible for the detection of 1 µg of recombinant protein.

Western Blot Analysis-- Total cell extracts were prepared as described previously (8). After separation with Tricine SDS-PAGE, the proteins were blotted at 50 V for 1 h onto 0.2-µm nitrocellulose (Schleicher and Schuell, Germany) and cross-linked to the membrane by UV irradiation (480 mJ). Protein transfer was detected by Ponceau S staining. The membrane was blocked overnight in TBST (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.02% Tween) containing 3% milk powder and incubated for 1 h with S100A13 antiserum (antiserum 2505; 1:500). After washing the membrane three times in TBST, the blot was incubated with goat anti-rabbit IgG Peroxidase (Sigma; 1:50,000). The bands were detected using ECL solution (Amersham Pharmacia Biotech) and exposed to Kodak films for 5-30 min.

Cell Culture-- The two human smooth muscle cell lines HISMC and HA-VSMC were obtained from ATCC (Manassas, VA), and the endothelial cell line ECV was a gift from Dr. T. Maciag (South Portland, Maine). HISMC cells were cultivated in Dulbecco's modified Eagle's medium, 10% fetal calf serum, and HVSMC in Ham's F-12K medium with 2 mM L-glutamine adjusted to 1.5 g/liter sodium bicarbonate and supplemented with 10 mM Hepes, 10 mM TES, 50 µg/ml ascorbic acid, 10 µg/ml insulin, 10 µg/ml transferrin, 10 ng/ml sodium selenite, 10 µg/ml endothelial cell growth supplement, and 10% fetal calf serum. The ECV cell line was cultivated in M199 medium containing 10% fetal calf serum.

Immunofluorescence Labeling and Confocal Laser-scanning Microscopy-- For immunofluorescence staining, the different cell lines were transferred into 24-well plates. The stainings were performed as described (8). The polyclonal serum (rabbit 2505) and the preimmune serum were diluted at 1:100. S100A2 was stained with the polyclonal rabbit antiserum F4023 (1:100) and S100A1 with the polyclonal rabbit serum F3792 (1:100). The secondary Cy3-conjugated affinity-purified goat anti-rabbit IgG (H + L) antibody was used (Jackson ImmunoResearch Laboratories, Baltimore, MD) at a dilution of 1:200. The coverslips were mounted on glass slides and embedded in Mowiol (Hoechst, Frankfurt).

A Zeiss Axioplan fluorescence microscope (Zeiss, Germany) equipped with a confocal scanning unit MRC-600 (Bio-Rad, UK) and an argon-krypton laser was used to examine the cells. Subsequently, images were processed using Photoshop (Adobe Systems, San Jose, CA).

Immunohistochemistry-- Multitissue slides containing different human tissue sections (Biomeda, Foster City, CA) were stained as described (8). The antiserum against S100A13 was diluted 1:200.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Purification of Human Recombinant S100A13-- In order to study the biochemical properties and to produce antibodies, we purified recombinant S100A13 protein. To this end, the S100A13 cDNA was cloned into the prokaryotic expression vector pGEMEX and transformed into Escherichia coli cells. The protein was expressed by the addition of isopropyl-1-thio--D-galactopyranoside, and extracts were prepared before and after induction. On a SDS gel, a 10-kDa band was evident with and without the addition of isopropyl-1-thio--D-galactopyranoside (Fig. 1A, lanes 1 and 2), indicating that some S100A13 is also expressed in the noninduced state. The overexpressed S100A13 corresponds to about 5% of the total bacterial protein content, and about 5 mg of protein could be expressed in a 250-ml bacteria culture. Typically, S100 proteins can be purified through hydrophobic interaction with a phenyl-Sepharose matrix. However, S100A13 behaved differently from all other S100 proteins, since it was not retained on the column. The flow-through was applied on a MonoQ-1 column, which retained nearly all of the contaminating proteins, whereas S100A13 appeared again in the flow-through. The protein was over 95% pure by SDS-PAGE (Fig. 1A, lane 3) although with an atypical UV spectrum (maximum at 260 nm). Therefore, it was further purified by gel filtration on Sephacryl S200 followed by ion exchange on QS fast-flow, essentially to remove nonprotein UV-absorbing material. After this additional chromatographic step, the protein was homogeneous in SDS-PAGE and displayed, in the metal-free as well as in the denatured form, a near-UV spectrum, which is consistent with the presence of one Trp and one Tyr (Fig. 4, left inset).

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Purification of recombinant S100A13 (A) and characterization of the polyclonal antibodies (B). A, protein extracts of 20 µl of S100A13 transformed Bl 21 Lys S cells before (lane 1) and after 3-h induction with isopropyl-1-thio--D-galactopyranoside (lane 2), separated on 15% SDS Tricine-PAGE under reducing conditions, followed by silver staining. 150 ng of purified S100A13 protein are shown in lane 3. B, Western blot analysis of 2 µg of human recombinant proteins S100B (lane 1), S100A6 (lane 2), S100A5 (lane 3), S100A4 (lane 4), S100A2 (lane 5), S100A1 (lane 6), and S100A13 (lane 7). The proteins were blotted onto nitrocellulose membrane and stained with Ponceau S (loading control); they were then stained with the two different antibodies (2505 and 2506) against S100A13 in a dilution of 1:1000. The bands were visualized using a horseradish peroxidase- coupled antibody diluted 1:2000.

Biophysical Characterization-- The molecular mass and purity of S100A13 were checked by mass spectrometry. The molecular mass of 11343 Da (data not shown) is in good agreement with the theoretically calculated molecular mass without the first methionine (11341.1 Da). We therefore conclude that the first methionine is cleaved off during expression. The native molecular mass of S100A13, determined by Sephacryl S200 gel filtration in buffer A containing either 2 mM Ca²⁺ or 1 mM EGTA, amounts to 27 kDa, strongly suggesting that the protein is a homodimer as was reported for most other members of this family.

Direct Cation Binding Studies-- Ca²⁺ binding by flow dialysis under physiological ionic strength conditions (Fig. 2) revealed that each dimer binds four Ca²⁺ with two different affinities: two with a [Ca²⁺]_0.5 of 8 µM and two with a [Ca²⁺]_0.5 of 400 µM. However, within each set there is positive cooperativity with a Hill coefficient, n_H, of 1.33, apparently as a result of allostery between the subunits. The analysis with the Adair equation (Equation 1) yielded the stoichiometric binding constants 1.25 × 10⁵ M¹ for K₁ and K₂, and 2.5 × 10³ M¹ for K₃ and K₄.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Ca2+ binding to S100A13 and effect of Mg2+. Ca2+ binding was measured in duplicate (closed and open symbols) by flow dialysis at 25 °C in buffer A. Circles, no additional ions; rectangles, 2 mM Mg2+. The lines connecting the symbols were generated using the Adair equation with the following stoichiometric constants: 1.25 × 105, 1.25 × 105, 2.5 × 103, and 2.5 × 103 106 M1. The Hill coefficient amounts to 1.33 for both groups of sites.

Secondary Structure Probed by Circular Dichroism-- The far-UV CD spectra shown in Fig. 3 are atypical, since the minimum at 222 nm found in most other -helix-rich proteins, including the other S100 proteins (31), is shifted to 228 nm, perhaps as a result of particular contributions of the tandem Trp-Tyr at the end of the second Ca²⁺-binding loop. A quantitative analysis of the metal-free form according to Johnson (32) yielded the following estimates of secondary structure: 49% -helix, 9% 3₁₀-helix, 2% -pleated sheet, 9% -turn, 7% Pro-type helix, and 24% "others." The addition of 100 µM Ca²⁺, which saturates only the high affinity sites, leads to an increase of the ellipticity, especially at 238 nm (see difference spectrum of Fig. 3) and an increase of -helix content of 5%. The addition of 0.15 mM Zn²⁺ or 2 mM Mg²⁺ to the metal-free sample had no effect (not shown). The addition of 2 mM Ca²⁺ in order to saturate the low affinity sites leads to an additional very small change. It can thus be concluded that, contrary to S100A2 (31), S100P (33), or S100A5,³ the -helix content increases upon binding of Ca²⁺.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Far-UV CD spectra of recombinant S100A13. Spectra were recorded at room temperature on 0.25 mg/ml protein in 5 mM Tris-HCl buffer, pH 7.5, in the metal-free state (dotted lines), in the presence of 100 µM Ca2+ (dashed lines) and 2 mM Ca2+ (solid line). The dashed and dotted line represents the difference spectrum between fully saturated and metal-free protein.

Trp Fluorescence Properties (Data Not Shown)-- S100A13 contains only one Tyr⁷⁶ and one Trp⁷⁷ at the end of the Ca²⁺-binding loop in the second EF-hand. The emission fluorescence spectrum of the metal-free protein has a maximum at 332 nm, and its intensity is 1.3-fold higher than that of the denatured protein. Contrary to the case for nearly all of the other S100 proteins, Ca²⁺ binding to S100A13 leads to a 1.3-fold decrease together with a slight red shift. Thus, the Ca²⁺ form differs from the denatured one only by its 12-nm blue shift. The addition of 2 mM Mg²⁺ or 0.15 mM Zn²⁺ to the metal-free protein has nearly no effect, pointing to the Ca²⁺-specific character of the sites.

Near-UV Difference Spectrophotometry-- The near-UV spectrum (Fig. 4, left inset) shows a maximum at 280 nm with a shoulder at 290 nm, typical for a protein containing Trp. The near-UV difference spectrum induced by Ca²⁺ binding shows positive and negative peaks in the 270-310-nm region, quite similar to those of calretinin and calretinin-22k (34) and indicative of rearrangements in the environment of the single Trp. Qualitatively, 0.15 mM Zn²⁺ has the same effects as Ca²⁺, except that the intensities of the peaks are 2.5-fold less (dashed line in Fig. 4). Cu²⁺ (0.1 mM) has a similar effect as Zn²⁺, except that the base line continuously rises at lower wavelengths (data not shown). Interestingly, Ca²⁺ binding and denaturation by guanidine-HCl produce quite similar spectral changes, supporting the fluorescence data. The Ca²⁺ titration of 55 µM S100A13 (Fig. 4, right inset) shows a linear dependence of the optical signal change on the added Ca²⁺ concentration with a clear inflection point at a ratio of 2.15 Ca²⁺/mol of dimeric protein. Thus, conformational changes are completed after binding of one Ca²⁺ to the high affinity site.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Difference spectra of 55 µM S100A13 dimer in buffer A at room temperature after the addition of 2 mM Ca2+ (solid line), 0.15 mM Zn2+ (dashed line), or 4 M guanidine HCl (dashed and dotted line) to the metal-free protein. The difference optical density (DOD) was expressed for a protein solution with an optical density of 1.0 at 280 nm. Left inset, UV spectrum of metal-free (solid line) and denatured (dotted and dashed line) S100A13. Right inset, Ca2+ dependence of the difference spectral changes of S100A13. Rectangles, S(290-285 nm); circles, S(290-296 nm).

TNS Fluorescence Enhancement-- TNS becomes fluorescent when associated noncovalently to hydrophobic regions in proteins. The emission fluorescence of TNS alone is very weak; metal-free recombinant S100A13 induces a 17-fold enhancement, whereas only a slight enhancement occurs with the Ca²⁺-loaded form (Fig. 5). These data suggest that, contrary to nearly all of the documented cases of the other S100 proteins, Ca²⁺ binding does not lead to the exposure of a hydrophobic patch at the surface of S100A13, and they explain why S100A13 does not bind to phenyl-Sepharose. Zn²⁺ and Cu²⁺ have no pronounced effect on the apo form and no effect at all on the Ca²⁺-saturated protein.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   TNS fluorescence spectra of S100A13. Hydrophobic exposure as monitored by fluorescence enhancement of TNS is shown. Spectra of 40 µM TNS in the presence of 2 µM recombinant S100A13 in the absence of divalent cations (dotted lines) or in the presence of 2 mM Ca2+ (solid line), 2 mM Mg2+ (dashed line), or 0.15 mM Zn2+ (thin dashed and dotted line) are shown. The thick dashed and dotted lines represent the fluorescence of 40 µM TNS alone.

Generation of Specific Antibodies against Recombinant S100A13-- To study the expression pattern and localization of S100A13 in different human tissues and cell lines, we raised polyclonal antisera against the recombinant protein. Antisera were first tested for their cross-reactivity against other S100 proteins on Western blots (Fig. 1B). The polyclonal antibody 2505 recognized specifically the corresponding antigen (lane 7). A faint cross-reactivity against S100A5 was only observed after a longer exposure time (lane 3). Antiserum 2506 recognized predominantly the S100A13 antigen with a faint cross-reactivity against S100A1 (lane 6), S100A2 (lane 5), and S100B (lane 1). For further studies, we chose therefore the antiserum 2505 because it has the highest specificity for S100A13. In addition to the monomer, both antisera recognized a faint band of about 27 kDa, which might indicate a low amount of dimer.

Gene Expression of S100A13 in Various Tissues-- Little is known about the tissue distribution of S100A13. Preliminary studies demonstrated high levels of S100A13 mRNA in skeletal muscle, heart, kidney, pancreas, ovary, spleen, and small intestine (18). Here, we examined the expression of S100A13 on RNA blots isolated from 50 different human tissues (Fig. 6). We confirmed the high S100A13 mRNA levels in the above mentioned tissues with the exception of skeletal muscle and pancreas. Our data also demonstrate that S100A13 is expressed in various regions of the brain (Fig. 6, A and B, 1-8). Furthermore, S100A13 is expressed in bladder, uterus, thyroid gland, mammary gland, appendix, lung, trachea, and placenta. During development, fetal expression could be detected in heart, kidney, spleen, and lung. We therefore conclude that S100A13 is widely expressed, similar to S100A2 but different from most other S100 members (11).

View larger version (68K): [in this window] [in a new window]   Fig. 6.   Expression of S100A13 in different human tissues. Commercially available Northern blots were hybridized with the labeled S100A13 cDNA. High mRNA levels were found in several regions of the brain, heart, colon, bladder, uterus, ovary, thyroid, mammary gland, kidney, small intestine, spleen, appendix, lung, trachea, and placenta. Ubiquitin was rehybridized with the same filter as a loading control.

Subcellular Localization of S100A13 in Different Tissues and Cell Lines-- At the protein level, the expression of S100A13 in different tissues was analyzed by staining a panel of normal tissue slides with antiserum 2505. S100A13 was highly expressed in follicle cells of thyroid (Fig. 7), in agreement with the high mRNA level detected on Northern blot (Fig. 6). In the brain, expression was detected in only a few cells. Furthermore, S100A13 was also detected in the Leydig cells of testis (Fig. 7). S100A13 is therefore expressed in a cell type-specific manner in these tissues.

View larger version (163K): [in this window] [in a new window]   Fig. 7.   Localization of S100A13 in human tissues. Paraffin-embedded human tissue sections were immunohistochemically stained with polyclonal antibody against S100A13. Leydig cells in testis and follicle cells in thyroid are depicted by black arrows. As a control, tissue sections were incubated with preimmune serum. Calibration bar, 0.1 mm.

Since S100 proteins show specific subcellular expression patterns, distinct for each member of the family, we were interested in the subcellular localization of S100A13 in different cell lines. Therefore, smooth muscle cells (HA-VSMC) were stained with polyclonal antibodies. S100A13 was specifically localized in the perinuclear compartment (Fig. 8, A-C) compared with the control with the preimmune serum (Fig. 8D). This localization is different from that of S100A2, specifically expressed in the nucleus, and S100A1, which is associated with stress fiber-like structures (Fig. 8, E and F) in smooth muscle cells. Similar results were obtained with HISMC cells and endothelial cells ECV (data not shown). S100A13 has therefore a unique subcellular distribution differing from that of all other S100 proteins in HA-VSMC cells.

View larger version (151K): [in this window] [in a new window]   Fig. 8.   Subcellular immunolocalization of S100A13 in human vascular smooth muscle cells (HA-VSMC). A-C, the localization of S100A13 was visualized by confocal laser-scanning microscopy after indirect immunofluorescent staining with a polyclonal antibody against S100A13 (antibody 2505; 1:100); D, the preimmune serum (1:100) was applied; E, HA-VSMC stained with a polyclonal antiserum (F4023; 1:100 dilution) against S100A2; F, HA-VSMC stained with a polyclonal serum (F3792; 1:100 dilution) against S100A1. The secondary Cy3-conjugated affinity-purified goat anti-rabbit IgG (H + L) antibody was used in a concentration of 1:200. A-C show three optical sections from the top down to the cell body in a distance of 0.56 µm; in DF, average pictures were selected.

To confirm these data, whole-cell extracts were prepared and analyzed by immunoblotting (Fig. 9). In each cell line (HISMC (lane 1), HA-VSMC (lane 2), and ECV (lane 3)), a similar amount of S100A13 (5 ng/100 µg of cell extract) was detected.

View larger version (50K): [in this window] [in a new window]   Fig. 9.   Western blot analysis of S100A13 in cell extracts. 100 µg of whole cell protein extract of the different cell lines were separated on SDS-PAGE and blotted onto a nitrocellulose membrane and stained with the polyclonal antibody 2505 (lanes 1-3) or preimmune serum (lanes 4-6). Both sera were diluted 1:1000. lanes 1 and 4, extract of HISMC; lanes 2 and 5, extract of HA-VSMC; lanes 3 and 6, extract of ECV.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

A hallmark in the molecular mechanism of the activation of most S100 proteins is the exposure of two hydrophobic patches in the presence of calcium ions, which allows their interaction with target proteins (11). The comparison of the three-dimensional structures of Ca²⁺-free and -saturated S100B revealed that in the dimer the helices III and III' state swing out upon binding of calcium, thereby uncovering hydrophobic residues. This feature also allows most S100 proteins to bind to a phenyl-Sepharose matrix. However, in this respect, S100A13 is distinct. We and others (23) observed that only a negligible amount of recombinant S100A13 can actually bind to phenyl-Sepharose. Moreover, Ca²⁺-saturated S100A13 does not interact with the hydrophobic probe TNS. The strong enhancement of the TNS fluorescence by the metal-free form is intriguing and may be due to the molten-globule character of this form of S100A13. Indeed, extrinsic hydrophobic probes show a large increase in fluorescence quantum yield upon binding to cavities within noncompact proteins (35). Similar experiments with the hydrophobic probe bis-ANS showed a very strong enhancement with the metal-free protein and also with the Ca²⁺ form. Probably, the very hydrophobic bis-ANS actively penetrates the protein core and induces a molten globule state as was reported for -lactalbumin (36) and Ca²⁺-binding proteins such as grancalcin⁴ and sarcoplasmic Ca²⁺-binding protein.⁵ The singularity of S100A13 is also underlined by the unusual CD spectra, the decrease of Trp fluorescence upon Ca²⁺ binding, and changes in the near-UV spectra that resemble those induced by a denaturing agent. Future three-dimensional studies on S100A13 may reveal how much the metal-bound and the metal-free forms of S100A13 differ from the well studied three-dimensional structures of other S100 proteins.

S100A13 is also particular in that it possesses two sets of Ca²⁺-binding sites with very different affinities. S100A2, S100A3, S100A4, S100A5, S100A6, and S100A11, studied under identical conditions by our group (for a review, see Ref. 11), all displayed positive cooperativity for the four sites in the dimer. Of interest, within each set of sites in S100A13, there is also moderate positive cooperativity. Direct binding data do not allow us to distinguish between the following two binding models. 1) The first allosteric transition encompasses two sites within the same subunit, and Ca²⁺ occupation to that subunit weakens Ca²⁺ binding to the second subunit. 2) The first allosteric transition is an intersubunit event (e.g. between the C-terminal EF-hands in each subunit), and Ca²⁺ binding to this site hinders Ca²⁺ binding to the noncanonical N-terminal EF-hands in each subunit. Conformational data favor the second working hypothesis. Indeed, Trp⁷⁶ and Tyr⁷⁷ are adjacent to the Ca²⁺-binding loop of the second EF-hand and probably "sensors" of the cation occupancy of that particular site. Since changes in the microenvironment of these two residues are completed after binding of the first two Ca²⁺ to the dimeric protein, we hypothesize that the two high affinity sites correspond to the C-terminal EF-hands. The sharp intersection in Fig. 4 (right inset) points to a high affinity and positive cooperativity between the C-terminal EF-hands in the dimer. Interestingly, far-UV CD changes are also complete upon binding of Ca²⁺ to the high affinity sites. Based on the observed conformational changes, it can be inferred that Zn²⁺ interacts with S100A13, although our data do not give more insight on the stoichiometry and affinity.

S100 proteins show a characteristic cell- and tissue-specific expression. They are either expressed in a subset of cells, i.e. S100A3 in human hair cuticle cells (1, 2) and S100A5 in the olfactory bulb,³ or they are found in a broader range of tissues like S100A2. The expression of S100A13 is most similar to S100A2, since high RNA expression is observed in a number of different tissues. Moreover, searching in an expressed sequence tag DNA data base revealed S100A13 clones in heart, lung, kidney, and brain, thus confirming the high expression levels in these tissues. It remains to be investigated if the S100A13 expression is altered in tumor tissues as described for a number of other S100 proteins (7, 8, 10, 37).

Furthermore, we investigated the immunohistochemical localization of the S100A13 protein in different tissues and cell lines. To this end, we produced a specific polyclonal antibody that detected S100A13 expression in Leydig cells of testis, in follicle cells of thyroid, in distinct cells of brain, and in certain cell lines such as smooth muscle, endothelial, and cervix epithelial carcinoma (HeLa) (data not shown), confirming S100A13 expression in various tissues. These observations suggest that S100A13 might regulate Ca²⁺ signaling in various tissues and cell types.

Recently, S100A13 was shown to be present in a multiaggregate complex of FGF-1 and p40 synaptotagmin (24). FGF-1 is an extracellular mitogenic factor for a diverse population of cells and uses a secretion pathway that seems to be independent of the conventional endoplasmic reticulum/Golgi pathway (38, 39). Hence, it was proposed that S100A13 could be involved in the regulation of such secretory processes. Indeed, we observed that S100A13 is predominantly localized in the perinuclear area of smooth muscle and endothelial cells, most likely in association with the endoplasmic reticulum/Golgi compartment. Specific relocations of some S100 proteins have been demonstrated in response to an increase of the intracellular calcium level (40, 41). Future studies should elucidate if this complex is translocated in response to a rise in the intracellular calcium level or other stimuli. Interestingly, S100A4 interacts with FGF-2, resulting in a modulation of the metalloproteinase induction (42). Perhaps the interaction of S100A13 with FGF-1 could similarly regulate enzyme activities required for the extracellular function of FGF-1. FGF-1 is highly expressed at sites of inflammation, including inflamed cartilage (43). It has been reported to be involved in vascular physiologic stress (44, 45) and is secreted in response to heat shock. Recently, S100A12 and S100A13 were shown to be targets of the antiallergic drugs amlexanox, tranilast, and cromolyn, which inhibit the degranulation of mast cells (23). We therefore speculate that S100A13 might play an important role in immunological reactions.

Very recently, S100A12 was demonstrated to trigger a signal transduction cascade by binding the cell surface receptor RAGE in endothelium, lymphocytes, and mononuclear phagocytes, resulting in the activation of the transcription factor NF-B (46). There, for the first time, a molecular model for the involvement of a extracellular S100 protein in inflammatory processes was proposed. S100A13 also may be involved in inflammatory and immunological responses via RAGE or a related receptor, but the protein most likely fulfills other functions as suggested by its expression in a wide range of tissues and cell lines. In contrast to S100A8/S100A9, which are specific for myeloid cells, and to S100A12, which is mostly expressed in neutrophils, S100A13 was demonstrated to be expressed in other cells such as smooth muscle or HeLa.

In summary, we conclude that S100A13 is a particular member of the S100 protein family. It differs from the others in its very broad Ca²⁺ sensitivity, its unusual metal-free state and the absence of a surface-exposed hydrophobic patch in the Ca²⁺-saturated state. The expression in a number of different tissues might support a general physiological role.

	ACKNOWLEDGEMENTS

We thank Dr. H. Troxler for the mass spectrometry analyses, M. Killen for critical reading of the manuscript, U. Redweik for the amino acid analysis, and Dr. M. Höchli and Prof. Dr. T. Bächi for advice with confocal microscopy.

	FOOTNOTES

^* This work was supported by BIOMED 2, European Union Grant BMH4CT950319/BBW Grant 950215-1, and Swiss National Science Foundation Project 31-50510.97.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: University of Geneva, 30 Quai Ernest Ansermet, 1211 Geneva 12 CH, Switzerland. Tel.: 41 22 7026491; Fax: 41 22 7026483; E-mail: Jos.Cox@biochem.unige.ch.

² All of the molar concentrations of S100A13 are calculated for the dimer form.

³ B. W. Schäfer, unpublished results.

⁴ K. Lollike, A. H. Johnsen, I. Durussel, N. Borregaard, J. A. Cox, submitted for publication.

⁵ J. A. Cox, unpublished results.

	ABBREVIATIONS

The abbreviations used are: FGF-1, fibroblast growth factor-1; p40Syn-1, p40 synaptotagmin-1; [Ca²⁺]_0.5, cation concentrations at half-maximal change; PAGE, polyacrylamide gel electrophoresis; TNS, 2-p-toluidinylnaphthalene-6-sulfonate; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; TES, 2-{[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]amino}ethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. C. Brisset, H. Hao, E. Camenzind, M. Bacchetta, A. Geinoz, J.-C. Sanchez, C. Chaponnier, G. Gabbiani, and M.-L. Bochaton-Piallat Intimal Smooth Muscle Cells of Porcine and Human Coronary Artery Express S100A4, a Marker of the Rhomboid Phenotype In Vitro Circ. Res., April 13, 2007; 100(7): 1055 - 1062. [Abstract] [Full Text] [PDF]

				E. Sturchler, J. A. Cox, I. Durussel, M. Weibel, and C. W. Heizmann S100A16, a Novel Calcium-binding Protein of the EF-hand Superfamily J. Biol. Chem., December 15, 2006; 281(50): 38905 - 38917. [Abstract] [Full Text] [PDF]

				V. Sivaraja, T. K. S. Kumar, D. Rajalingam, I. Graziani, I. Prudovsky, and C. Yu Copper Binding Affinity of S100A13, a Key Component of the FGF-1 Nonclassical Copper-Dependent Release Complex Biophys. J., September 1, 2006; 91(5): 1832 - 1843. [Abstract] [Full Text] [PDF]

				H.-L. Hsieh, B. W. Schafer, J. A. Cox, and C. W. Heizmann S100A13 and S100A6 exhibit distinct translocation pathways in endothelial cells J. Cell Sci., January 8, 2002; 115(15): 3149 - 3158. [Abstract] [Full Text] [PDF]