Enzymatic immobilization of organometallic species: biosilification of NCN- and PCP-pincer metal species using demosponge axial filaments

Abstract

Silicatein protein filaments isolated from marine demosponges have been used to influence the condensation of siloxanes bearing organometallic pincer complexes. The siliceous material is formed under remarkably mild conditions and the organometallic pincer becomes an intrinsic part of the silica. The immobilisation of a metal pincer, which acts as a sensor and initial results on the immobilisation of a pre-catalytic pincer species are reported.

---

The formation of solid silica materials with controlled structures occurs in vivo in diatoms and demosponges.^1–3 The silica is formed under the direction of proteins and polysaccharides, under the mild conditions experienced in vivo. In contrast to this, laboratory methods for the production of silica-type materials can be harsh using extremes of pH and temperature that are not compatible with the production of silicas bearing labile organic or organometallic moieties.^4–6 For example, many methods for the production of sol-gels involve the use of a pH of less than 3.^6b–6c

The silica producing demosponge Tethya aurantia contains silica spicules from which protein filaments can be isolated that are key to the in vivo formation of silica deposits. The protein filaments, consisting of multiple subunits of an enzyme named silicatein, have been used as a research model for investigating the mechanism and control of the biosilification process.^7,8Tetraethoxysilane (TEOS) and trialkoxysilanes, which bear a small organic moiety on the silicon, have been hydrolysed and condensed to form silica structures with controlled shapes at room temperature.^8,9

The potential enzymatic silification of trialkoxysilanes bearing an organometallic species would represent a very mild method for the immobilisation of organometallic moieties. We have tested this technique in the immobilisation of so-called pincer complexes. These systems are used here as models for future applications in the immobilisation of, e.g.catalytic species, which are incompatible with more standard immobilisation techniques.^10–12 In essence, this study represents a proof of concept for the feasibility of using enzymatic silification to immobilise organometallic compounds, which may be active catalysts or pre-catalysts.

Previous work in this area has concentrated on strategies for the immobilization of organometallic pincer metal complexes on, e.g.dendrimers¹³ and pre-formed solids,¹⁴ as well as on the catalytic activity of these materials. The nature of the pincer species involves a carbon–metal bond, stabilized by two additional bis-ortho chelating donor interactions, which makes it quite stable in an aqueous environment, thus maximizing the chance the complex will survive the immobilization process intact. The precursors used in the present study consist of the required pincer metal moiety attached via the para-position of the phenyl ring to a carbamate linker, which terminates in a triethoxysilyl moiety (Fig. 1). Siloxy functionalized pincer Pt and Pd complexes 1–3 were prepared using reported methods.¹⁴

Fig. 1 Pincer metal complexes.

The silicatein filaments were isolated and prepared as previously reported and outlined in Fig. 2.³ A series of standard reactions with tetraethoxysilane (TEOS) was used to assess the activity of each batch of protein. Visual inspection and/or a molybdate assay³ of the product of the test reactions gave a good indication of the most active concentration of the protein preparation, which was used in the subsequent study.

Fig. 2 Isolation of silicatein filaments from silica spicules.

Functionalized silicas were prepared by condensation of alkoxy silanes 1–3 onto the silicatein filaments; however, the amount of condensation achieved was small. The initial method was thus altered to provide an additional source of silicon. The altered method used a mixture of TEOS and the functionalized siloxanes. A suspension of the filaments in water was introduced to the reaction vial and to this was added a solution of the organometallic alkoxysilane (2 or 4 mg) in toluene (60 µL) and TEOS (50 µL). The mixture was mixed to form an effective emulsion and then lightly agitated for 16 h. Work-up consisted of one high-speed centrifugation step, to remove the supernatant, and washing with ethanol.† The pellets were examined visually and by SEM. When the protein filaments were excluded from the reaction medium no visible reaction was found to take place even over a period of days, which indicated a key role in the process for the protein filaments.

The enzymatically catalyzed silification process incorporated both the TEOS and the organometallic component into the silica material. The majority of the material reported in this study was prepared in this manner. NCN-Pd pincer compound 1 was also treated with the protein filament without the addition of the TEOS. Though, as outlined above, yields in using this technique had been generally poor, we were gratified to obtain 27 mg of a silica product representing a yield of approx. 66% in this case.

Once isolated, the silica type materials were studied in a number of ways. Initially, a sample of each material was studied viascanning electron microscopy (SEM) to establish whether any visible order was present in the materials and to determine if the coating on the protein filament could be observed. Fig. 3 shows three SEM images. Image (A) shows protein filaments with a very small amount of silica type material visible on the surface. This product is the result of the initial condensation of the NCN-Pd pincer alkoxysilane 1 in the presence of the protein with no TEOS added, showing a very low level of condensation onto the protein filament. An FTIR spectrum of this material showed a peak at 1714 cm⁻¹ consistent with the carbamate linker. We thought that the steric bulk of the organometallic group might have been inhibiting further condensation of the alkoxysilane to the silicatein and thus TEOS was used subsequently in the reaction mixture to allow co-condensation of the organometallic siloxane to occur.

Fig. 3 Scanning electron microscopy (SEM) images of products of enzymatically catalysed silification reactions. (A) Limited condensation of NCN-Pd pincer alkoxysilane 1 in the presence of protein filament with no TEOS added. (B) Condensation of the PCP-Pd pincer alkoxysilane 3 and TEOS in the presence of the silicatein filaments. (C) Extensive condensation of the NCN-Pt pincer alkoxysilane 2 and TEOS in the presence of the silicatein filaments.

The second SEM image (B) shows the product of the co-condensation of the PCP-Pd pincer alkoxysilane 3 and TEOS in the presence of the protein filaments. In this case more extensive condensation has occurred on the filaments. Indeed in this case parallel reactions produced sufficient product (77 mg) to be able to test the solid obtained to see if the physico-chemical character of the pincer complex was maintained during the enzymatically-catalysed silification process. The third SEM image (C) shows the product obtained from the co-condensation of NCN-Pt pincer alkoxysilane 2 and TEOS in the presence of the silicatein filaments. The protein filaments are no longer visible in the material isolated. Once begun, the reaction appears to continue even at sites now remote from the protein.

ICP-MS was used with two samples of the NCN-Pt biosilica. The first sample (2 mg of precursor 2 with 50 µL of TEOS) was found to have a Pt content of 0.4%. The second sample (4 mg of 2 with 50 µL of TEOS) was found to have a Pt content of 0.7% (corresponding to 77% incorporation of organometallic moiety into the isolated solid).

The material with a Pt content of 0.7% was subjected to a ¹³C NMR study using magic angle spinning (MAS). The product of 11 parallel reactions was used but had to be diluted with an inert material to fill the MAS tube. The resulting spectrum was compared to the ¹³C NMR MAS spectrum of a corresponding NCN-Pd carbamate, which had been grafted onto commercial silicavia chemical methods.¹⁴ The similarity of the spectra obtained (Fig. 4) from the two solid samples indicated that the metal pincer had been incorporated into the silica material during the enzymatically-catalyzed silification process.

Fig. 4 ¹³C MAS NMR spectra of chemically and enzymatically immobilised NCN-metal pincers.

Upon exposure to SO₂ gas, NCN-Pt halide complexes are known to reversibly bind one molecule of SO₂via Pt–S bond formation, which is accompanied by a reversible colourless to yellow/orange colour change (Fig. 5).¹⁵ This sensing behaviour is metal and donor atom specific and will only give a positive result if the pincer system is as described above, including the metal halide bond. The silica derived from the NCN-Pt complex 2 was tested in this manner. Reversible binding of SO₂ to the silica material, indicated by reversible coloration and distinctive changes in the IR spectrum of the material, allowed us to determine qualitatively that the pincer metal moiety survived the enzymatically-catalyzed silification process intact. Washing the silica with toluene and testing the solution with SO₂ showed no leaching of free pincer.

Fig. 5 Reversible binding of SO₂ to an NCN-Pt pincer (X = halide).

PCP-Pd pincer species have been shown to be pre-catalysts of Heck reactions, such as that shown in Fig. 6, a synthetically useful method of forming carbon–carbon bonds.¹⁶ This reaction, which is relatively non-demanding, was used to provide a proof of principle of the catalytic activity of the enzymatically engineered silica. The catalytic activity of the solid produced from the PCP-palladium alkoxysilane 3via the enzymatic silification process with TEOS was tested in two subsequent runs that were allowed to go to 50% conversion. In both runs the catalytic activity was observed and, clearly, the activity and nature of the catalytic material changed in accordance with the pre-catalyst description of the pincer metal species. The reaction profiles are similar to those of the immobilized catalyst obtained by chemical means.¹⁷ This observation indicates the potential of this new immobilization method in the field of immobilized organometallic catalysts. It is envisaged that our method of immobilization may be expanded to more versatile and more demanding catalytic species.

Fig. 6 Plot of the progress of the Heck reactionvs. time using enzymatically formed catalyst.

In summary, we have successfully employed an organometallic moiety as a starting material for an enzymatically catalysed silification process; the first example of such a process. This study has successfully combined the areas of molecular biology, catalysis studies and both organic and inorganic chemistry, creating new materials, which are of interest in all these areas and more importantly demonstrating the application of biosilification to a wider area of research.

Notes and references

1. S. Mann, Biomineralisation: Principles and Concepts in Bioinorganic Materials Chemistry, Oxford University Press, 2001.
2. E. G. Vrieling, T. P. M. Beelen, R. A. van Santen and W. W. C. Gieskes, J. Biotechnol., 1999, 70, 39.
3. D. E. Morse, Trends Biotechnol., 1999, 17, 230.
4. G. Oye, J. Sjoblom and M. Stocker, Adv. Colloid Int. Sci., 2001, 89, 439.
5. J. L. Blin, A. Leonard and B. L. Su, Chem. Mater., 2001, 13, 3542.
6. (a) P. J. Bruinsma, A. Y. Kim, J. Liu and S. Baskaran, Chem. Mater., 1997, 9, 2507; (b) G.-Z. Fan, T. Li and G.-X. Li, Appl. Organomet. Chem., 2006, 20, 656; (c) J. Zhao, Y. Zhang, J. Han and Y. Jiao, J. Mol. Catal. A: Chem., 2005, 241, 238.
7. For silica immobilization of an enzyme, see: N. Poulsen, C. Berne, J. Spain and N. Kröger, Angew. Chem., Int. Ed., 2007, 46, 1843.
8. Y. Zhou, K. Shimizu, J. N. Cha, G. D. Stucky and D. E. Morse, Angew. Chem., Int. Ed., 1999, 38, 780.
9. K. Shimizu, J. Cha, G. D. Stucky and D. E. Morse, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 6234.
10. K. J. C. van Bommel, A. Friggeri and S. Shinkai, Angew. Chem., Int. Ed., 2003, 42, 980.
11. J. A. Gladysz, Chem. Rev., 2002, 102, 3215 , and following articles.
12. S. C. Schurer, S. Gessler, N. Buschmann and S. Blechert, Angew. Chem., Int. Ed., 2000, 39, 3898.
13. (a) M. Albrecht and G. van Koten, Angew. Chem., Int. Ed., 2001, 40, 3750; (b) A. W. Kleij, R. A. Gossage, R. J. M. Klein Gebbink, N. Brinkmann, E. J. Reijerse, U. Kragl, M. Lutz, A. L. Spek and G. van Koten, J. Am. Chem. Soc., 2000, 122, 12112; (c) R. van de Coevering, A. P. Alfers, J. D. Meeldijk, E. Martínez-Viviente, P. S. Pregosin, R. J. M. Klein Gebbink and G. van Koten, J. Am. Chem. Soc., 2006, 128, 12700; (d) N. J. M. Pijnenburg, T. J. Korstanje, G. van Koten, R. J. M. Klein Gebbink, in The Chemistry of Pincer Compounds, ed. J. Dupont and M. Pfeffer, Wiley VCH Verlag, Weinheim, 2008, pp. 361–398.
14. (a) N. C. Mehendale, C. Bezemer, C. A. van Walree, R. J. M. Klein Gebbink and G. van Koten, J. Mol. Catal. A, 2006, 257, 167; (b) N. C. Mehendale, J. R. A. Sietsma, K. P. de Jong, C. A. van Walree, R. J. M. Klein Gebbink and G. van Koten, Adv. Synth. Catal., 2007, 349, 2619; (c) N. C. Mehendale, M. Lutz, A. L. Spek, R. J. M. Klein Gebbink and G. van Koten, J. Organomet. Chem., 2008, 693, 2971.
15. (a) M. Albrecht, M. Lutz, A. L. Spek and G. van Koten, Nature, 2000, 406, 970; (b) M. Albrecht, R. A. Gossage, M. Lutz, A. L. Spek and G. van Koten, Chem.–Eur. J., 2000, 6, 1431.
16. (a) W. J. Sommer, K. Q. Yu, J. S. Sears, Y. Y. Ji, X. L. Zheng, R. J. Davis, C. D. Sherrill, C. W. Jones and M. Weck, Organometallics, 2005, 24, 4351; (b) M. Ohff, A. Ohff, M. E. Van der Boom and D. Milstein, J. Am. Chem. Soc., 1998, 120, 3273.
17. N. C. Mehendale, R. J. M. Klein Gebbink, G. van Koten, unpublished results.

---

Footnote

† NCN-PtI carbamate enzymatic silification (0.4% Pt by weight product). A solution of NCNPtI carbamate (1) (2 mg) in toluene (60 µL) was added to a centrifuge tube containing water (160 µL), protein preparation (40 µL) and TEOS (50 µL). An emulsion was created either by repeatedly drawing the liquid into a pipette and expelling it rapidly or by using a vortexer. The reaction mixture was then placed on a rotor and gently agitated for 16 h. The work-up consisted of careful removal of the top layer followed by dilution with ethanol (1 mL), brief agitation, re-centrifugation, removal of as much liquid as possible, dilution, agitation, centrifugation and removal of liquid followed by air drying. A sample of the product was tested using SO₂ and gave a positive indication for pincer incorporation. In total, 19 parallel reactions were conducted yielding 145 mg of silica-type product.

---