| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Departament de Genètica, Universitat de Barcelona, 08028-Barcelona, Spain; and Departament de Química, Universitat Autònoma de Barcelona, 08193-Bellaterra, Barcelona, Spain
1To whom requests for reprints should be addressed at Departament de Genètica, Universitat de Barcelona, 08028-Barcelona, Spain. E-mail: satrian{at}ub.edu
Abstract
Recombinant (E. coli ) synthesis of mammalian MT1 and MT4 domains as separate peptides in Zn(II) and Cd(II) enriched growth media has rendered metal complexes containing sulfide anions as additional ligands. The Cd preparations show higher sulfide content than the Zn preparations. Also, the ßMT1 and ßMT4 fragments exhibit higher sulfide/peptide ratios than the respective 
fragments. Titration of Zn3-ßMT1 with Cd(II) followed by addition of several sodium sulfide equivalents shows that the Cd(II)-ßMT1 species can incorporate sulfide ligands in vitro, with a concomitant evolution of their UV-vis and CD fingerprints to those characteristic of the Cd-S2– chromophores. Current results have also provided full understanding of previous data collected by this group in the characterization of the Cd-ßMT1 preparations obtained from large-scale fermentor synthesis by allowing identification of at least 2S2– ligands per Cd-ßMT1 species. Furthermore, the results here presented have revealed that synthesis of ßMT4 in Cd-supplemented cultures yielded Cd,S2––containing clusters instead of the proposed heterometallic Zn,Cd-ßMT4 complexes. Finally, a global evaluation of our results suggests that the higher the Cu-thionein character of a MT peptide, the higher is its tendency to harbor nonproteic ligands (i.e., sulfide anions) when building divalent metal clusters, especially Cd-MT complexes.
Key Words: domain • ß domain • metallothionein • MT1 • MT4 • sulfide ligands
Introduction
Metallothioneins (MTs) are a superfamily of atypical small proteins, ubiquitous but probably polyphyletic, which coordinate heavy metal ions through metal-thiolate bonds established by the highly abundant cysteine residues of their sequence (1). Currently, and after half a century of multidisciplinary research, the biological structure of MTs and their contribution to a variety of physiological processes in the most diverse organisms still remain undetermined (2, 3). Two main reasons should be connected with this fact. First, most of the existing data refer to mammalian MTs, while the extreme sequence heterogeneity of this superfamily of proteins precludes any homology-driven structural, functional, or evolutionary inference (cf. Web page: http://www.expasy.ch/cgi-bin/lists?metallo.txt). Second, the difficulties found when trying to obtain homogeneous native metal-MT complexes, together with the impossibility of purifying them of some organisms or in non-metal-induced forms, forced the use of indirect methods for their preparation, always on the assumption that genuine and functional native MT structures were reproduced. The most common methodology applied relied on the in vitro reconstitution of metal-MT complexes from native apo-forms obtained after heavy acidification, or even from synthetic peptides, which does not necessarily guarantee the physiological significance of the recovered species. A further advance was attained through the recombinant synthesis of MTs in heterologous hosts, which enabled both the study of metal-MT species conformed in vivo (i.e., in a physiological environment) and the application of nonaggressive purification strategies.
Nearly ten years ago we developed an Escherichia coli expression system that allows the biosynthesis of ZnII-, CdII- and CuI-MT intact MT complexes, isolated domains, and mutant variants, with sufficient quantity and purity to allow for analytical, spectrometric, and spectroscopic characterization. Since studies on the mouse Zn-MT1 system (4, 5) fully validated the correspondence between native and recombinant complexes, studies were expanded from mammalian MT1 (6–8) to mammalian MT4 (9), the crustacean MTH (10), the Drosophila MTN (11) and MTO (12), and the plant Quercus suber QsMT (13). Recently, after the recombinant synthesis of several MTs of different organisms in E. coli and the thorough analysis of the chemical/structural features of their in vivo conformed metal-MT complexes, we identified a third component in the metal-MT clusters: inorganic sulfur anions behaving as sulfide ligands. Hence, using analytical, spectroscopic, and spectrometric techniques, we provided qualitative and quantitative evidence that S2– ligands were present in nearly all the ZnII- and CdII-MT complexes of the studied MTs, always in a larger quantity in the latter than in the former (14). The general features of these clusters correlated well with those of the plant and yeast Zn- or Cd-
-glutamyl peptides, thus bridging the gap between both kinds of metal binding biomolecules (15, 16). Also, this finding enlarges the physiological potential of these poorly understood proteins, in view of the emerging evidence of S2– involvement in relevant cell events, such as electron transfer (17), redox equilibrium (18), and neurotransmission and neuro-modulation (19).
While the crucial question of whether sulfide ligands are also present in native MT forms is addressed, we have also focused our efforts in analyzing the sulfide ligand presence in the separate ß and domains of the mammalian MT1 and MT4 isoforms. This is of great interest, first, due to the intrinsic importance of the mammalian proteins in biomedical research, and second, because characterization of the recombinant mammalian domains had been carried out before the discovery of the S2– presence in the MT recombinant preparations. Therefore, we present here the results of the analysis of the Zn(II)- and Cd(II)-species of the separate ßMT1, ßMT4, MT1, and MT4 domains obtained by the same rationale used to detect and quantify sulfide in the full length MTs (14). Furthermore, this work includes the comparison of their features with those of the entire MT1 and MT4, and the reconsideration of some data reported in the literature for the Cd(II)-complexes of the MT1 and MT4 separate domains (6, 9).
Material and Methods
Recombinant Syntheses of the Metal-MT Complexes.
The construction of the cDNA encoding for the separate MT domains and their cloning in the pGEX expression vector have been previously reported (MT1 [6], MT4 [11]). The synthesis and purification of the corresponding Zn(II)- and Cd(II) complexes in E. coli LB (Luria Bertani) culture media supplemented with 300 µM ZnII or CdII final concentration, respectively, has also been carried out as explained there.
Inductively Coupled Plasma-Atomic Emission Spectroscopy (ICP-AES) and Amino Acid Analysis.
S, Zn, and Cd sample contents were determined by ICP-AES, using a Polyscan 61E (Thermo Jarrell Ash; Thermo Electron Corporation, Waltham, MA) spectropolarimeter and measuring S at 182.040 nm, Zn at 213.856 nm, and Cd at 228.802 nm, and sample preparation was performed according to two alternatives. Conventional ICP implied no previous treatment of the sample, while acid ICP included an acidification of the sample (incubation in 1 M HCl at 65°C for 5 mins) prior to ICP measures (14). In both cases, protein concentration was calculated from the S content of the sample, assuming that S was only contributed by the Cys and Met residues of the MT. In addition, protein concentration was assessed by standard amino acid analysis (hydrolysis in 6 M HCl at 110°C for 22 hrs) on an Alpha Plus Amino Acid Autoanalyzer (Pharmacia LKB, Cambridge, England). Ser, Lys and Gly contents were used to extrapolate MT concentrations.
In Vitro Cadmium and Sulfide Binding Studies.
Cd(II) binding studies of Zn3-ßMT1 were undertaken following the procedures already described in the literature for Cd(II) titrations (4). After four Cd(II) equivalents added to Zn3-ßMT1, aliquots of a standard 3.14 mM Na2S solution, prepared as described in (14), were added until a ratio of 5 S2– per ßMT1 was achieved.
Mass Spectrometry and CD-UV Spectroscopy.
The molecular mass of the metal-MT species was determined by electrospray ionization mass spectrometry (ESI-MS) on a Fisons Platform II Instrument (Fisons Instruments Inc, Beverly, MA), equipped with MassLynx software and calibrated with horse-heart myoglobin (0.1 mg/ml). The assay conditions were as follows: 20 µl of protein solution injected at 40 µl/min; the use of an HPLC Uptisphere (Interchim, Montluçon, France) C4 33 mm x 2 mm x 5 µm column to separate analytes; capillary counter-electrode voltage, 4.5 kV; lens counterelectrode voltage, 1.0 kV; cone potential, 60 V; source temperature, 120°C; m/z range, 850–1950; scanning rate, 3 secs/scan; interscan delay, 0.3 secs. In all cases, the running buffer was an appropriate mixture of acetonitrile and 5 mM ammonium acetate/ammonia, pH 7.5. Electronic absorption measurements were performed on an HP-8453 diode array UV-visible spectrophotometer. A Jasco spectropolarimeter (J-715; Jasco, Easton, MD) interfaced to a computer (GRAMS/32 software) was used for CD determinations. The temperature for all measurements was kept at 25°C by means of a Peltier PTC-351S apparatus (TE Technology Inc., Traverse City, MI).
Gas Chromatography with Flame Photometric Detection (GC-FPD).
GC-FPD (14, 20) was used to measure sulfide at low concentrations without the need of a derivatization step. H2S was generated by strong sample acidification (H2SO4, pH 0.0) in order to ensure the metal-MT complex disruption as well as to avoid the precipitation of the insoluble ZnS and CdS which might have been generated. From a nominal 1000 ppm S2– solution (14), dilute standards of 0, 0.25, 0.5, 1, 1.5, 2.5, and 3 ppm sulfide concentration were used to draw the corresponding calibration curve. Sample aliquots, as well as the standard solutions, were transferred to airtight 2-ml vials, acidified to a final volume of 0.5 ml, and immediately sealed. Vials were then incubated at 40°C for 2 hours with agitation (250 rpm) in order to accelerate the evolution of hydrogen sulfide from the aqueous phase and equilibration of gas phase in the headspace. Five hundred microliters of the headspace gas were subjected to gas chromatography (HP 5890 Series II coupled to a FPD80 CE Instruments [Thermo Finnigan detector; Thermo Electron Corporation]). The gaseous mixture was carried by a 6.6 ml/min flux of He through the GC glass column (SPB 608 30 m x 0.25 mm I.D. with 0.5 µm of particle size). Both the injection and the detection port were kept at 110°C, while the column was operated at a constant temperature of 35°C. The H2S peak generated from MT samples was readily identified by its retention time. All determinations were done in duplicate to ensure reproducibility.
Results and Discussion
Zn- and Cd-Complexes of Mammalian ßMT1 and MT1 Domains.
Analytical data in Table 1
show that recombinant synthesis of the ßMT1 and MT1 domains as separate peptides in metal supplemented media renders S2–-containing Zn- and Cd-species, in correspondence with the results obtained for the entire MT1 (14). In the case of the ß clusters, the presence of the sulfide ligands is quantitatively more important for the cadmium than for the zinc preparations, whereas both cluster preparations exhibit approximately equivalent sulfide content.
|
, respectively) fully reproduced those reported, although for lower metal-to-protein ratios. Interpretation of these results are rather straightforward, considering that the Zn3-ßMT1 solution used in this work contains a non-negligible amount of S2–, as revealed by the analytical results (cf. Table 1
) and also confirmed by the ESI-MS spectrum (Fig. 2A). Therefore, quantification of the initial Zn3-ßMT1 sample by conventional ICP-AES led to an overestimation of the actual MT concentration, and consequently the number of equivalents added during the titration were also overestimated (for a full explanation of conventional vs. acid ICP-AES MT quantification methodology, see 14). After the final stage of the Cd(II)/Zn3-ßMT1 titration, increasing amounts of S2–, from one to five equivalents per Cd3-ßMT1, were added to the sample. Spectroscopic data (Fig. 1D, E, and F) and ESI-MS results (Fig. 2B) unequivocally evidenced the incorporation of S2– ligands to the metal clusters. This implies the conversion of the 250-nm centred exciton coupling, attributable to the Cd(SCys)4 chromophores, to a Gaussian band at the same wavelength, together with a redshift of the initially 275-nm centered band to ca. 300 nm. In parallel, UV-vis and difference UV-vis spectra also show increase in the absorptions between 260 and 320 nm, with a clear redshift of the maxima, in accordance with literature data correlating higher wavelength absorptions with bigger Cd-S crystallites (21).
|
|
). In spite of the lack of conclusive results of the ESI-MS analysis on that occasion, our current results strongly suggest the presence of S2– ligands in the Cd-ßMT1 species recovered after large-scale synthesis, at an approximate ratio of 2 S2–/ßMT1. This significant presence of S2–ligands per ßMT1 peptide would also easily explain the low Cd/ßMT1 stoichiometry then reported, in accordance with our verification that conventional (i.e., non-acid) ICP-AES measurements lead to an overestimation of the protein concentration in S2–-containing MT samples, and thus to false low metal-to-protein ratios (14). A plausible explanation of the reasons determining large-scale synthesis yielding metal-MT species with increased S2– content remains to be put forward. These productions are carried out in 50-liter fermentors, with good aeration systems (6). This implies that the E. coli culture grows on aerobiosis, and some of the cell responses to adapt to oxidative status may end up favoring metabolic pathways that increase intra-cellular sulfide availability.
|
MT4 Domains. shows that not only the entire protein, but also its two constituting ß and domains, contain sulfide ligands in their Zn- and Cd-species recovered after recombinant synthesis. As expected, the Cd-preparations show higher S2– content than the Zn samples, and in particular, Cd-ßMT4 shows the highest value found in any mammalian MT fragment. Interestingly, this allowed us to revisit some of the results we recently reported for this MT isoform. Hence, when the in vivo Cd-ßMT4 preparations were characterized (9), we interpreted the ESI-MS data as corresponding to a mixture of Cd3, Cd2Zn1, and Cd1Zn2 species. However, the ICP-AES figures somewhat questioned this possibility, since Cd(II) was the unique metal ion detected in the samples. By considering the data obtained for in vivo Cd-ßMT4 in the current work, we can now suggest that all the assumed heterometallic Zn,Cd-ßMT4 species were in fact Cd-ßMT4 complexes that included sulfide ligands, on the basis of the mass equivalence between one Zn and two S. Consequently, the ESI-MS spectra of the in vivo Cd-ßMT4 preparation should be reinterpreted as shown in Figure 4, which shows the coexistence of the Cd3- and Cd2S2-ßMT4 species. This reinterpretation of the ESI-MS results corresponding to the in vivo synthesis of Cd-ßMT4 cannot be directly extended to the full-length MT4 protein, which also gave rise to a mixture of Zn,Cd-species, as in that case the ICP-AES data also confirmed their heterometallic nature. However, in view of the high S2– content measured in the Cd-MT4 preparations (Table 1), it is sensible to postulate now that probably part of the alleged Zn,Cd-MT4 clusters are in fact Cd,S2–-MT4 species.
|
This work, properly completing the results reported in (14), shows that recombinant synthesis of mammalian MT1 and MT4 domains in the presence of Zn(II) and Cd(II) gives rise to sulfide-containing clusters. The Cd-complexes have a higher number of sulfide ligands than the Zn-species, and among the former, those of the ß domain higher than those of the domain. Interestingly, ßMT1 and ßMT4 have been described by this group as Cu-thioneins (9, 10), with ßMT4, which harbors the highest sulfide content among all the peptides analyzed (Table 1), exhibiting a higher Cu-thionein character than ßMT1. Consequently, it seems sensible to hypothesize that the higher the Cu-thionein character of a MT peptide, the higher the sulfide content in its divalent-metal, especially cadmium, complexes. This is also patently clear if considering the Drosophila MTs, MTN and MTO, both classified as Cu-thioneins (11, 12, 14). Maybe the higher predisposition of a polypeptide to accommodate the small Cu(I) ions in linear or trigonal coordination environments is correlated with a relatively higher difficulty in providing Zn and Cd divalent ions, and especially the bulkier Cd(II), with a strict tetrahedral coordination environment, and thus the participation of extra (nonproteic) ligands to stabilize their metal complexes would be significantly favored.
Acknowledgments
We thank R. Bofill and A. Pagani for the preparation of the figures and helpful comments when preparing this manuscript, and the Servei d’Anàlisi Química de la Universitat Autònoma de Barcelona (AAS, CD, UV-vis) and the Serveis Científico-Tècnics de la Universitat de Barcelona (ICP-AES, ESI-MS, GC-FPD) for allocating instrument time. We especially thank Dra. Pilar Teixidor and Dra. Lourdes Berdié for their technical assistance in the GC-FPD measurements. Finally, we thank Prof. J. Kägi for fruitful and stimulating discussion on the subject of this work.
Footnotes
This work was supported by Spanish Ministerio de Ciencia y Tecnología grants CTQ2005–01946/BQU to M.C. and BIO2003–03892 to S.A. M.C. also wants to thank the Vicerrectorat d’Investigació of the Universitat Autònoma de Barcelona for additional financial support.
Authors Tío and Villarreal made equal contributions to this work.
References
metallothionein 1 domains and characterization of their cadmium(II) binding capacity. Cel Mol Life Sci 53:681–688, 1997.-glutamyl peptide complex of Schizosaccharomyces pombe. J Biol Chem 263: 12832–12835, 1998.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
