Elsevier

Analytica Chimica Acta

Volume 481, Issue 1, 28 March 2003, Pages 41-53
Analytica Chimica Acta

Molecular recognition properties of peptide mixtures obtained by polymerisation of amino acids in the presence of estradiol

https://doi.org/10.1016/S0003-2670(03)00062-XGet rights and content

Abstract

In this paper we show that the carbodiimide-induced polymerisation of amino acid mixtures in aqueous medium and in presence of estradiol produces the mixtures of peptides with an average molecular weight of 2–6 kDa that are characterised by possessing molecular recognition properties towards estradiol. After the removal of the templating molecule, the binding properties of the peptide mixtures were studied using spectrophotometric and immunochemical methods. The experimental results show the presence of molecular recognition behaviour for all the peptide mixtures obtained by polymerisation in presence of estradiol, with affinity constant values between 0.44×109 and 6.6×109 M−1, while the same mixtures obtained without estradiol show lower affinity constant values between 2.2×106 and 1.3×109 M−1. The molecular recognition behaviour was found to be highly selective, as the binding constants of peptides towards the structural homologues testosterone and progesterone are lower than three orders of magnitude. Peptide fractions separated by ion-exchange chromatography show the same molecular recognition properties, with affinity constant values between 3.2×106 and 7.1×109 M−1. Similarities and differences between this polymerisation technique and the molecular imprinting technique are briefly discussed.

Introduction

In nature, molecular recognition plays a decisive role in the chemistry of biological phenomena. In the attempts to mimic these systems, the scientific community has developed increasingly synthetic recognition systems based on tailor-made supramolecular organic structures [1], [2]. In these systems selectivity often is enhanced if, as happen in nature, a cavity exists that has been shaped to match that of the substance to be recognised. In the molecular imprinted technique these recognition systems can be obtained in highly cross-linked organic polymers: in fact complementary shaped cross-linked cavities can be formed around a molecule that acts as a template during the polymerisation step, and when the template is removed, an imprinted cavity with reversible binding capacity remains behind in the polymer [3], [4]. The molecular imprinting of organic polymers is today a promising area that includes many research fields like chromatography [5], [6], catalysis [7], and sensor technology [8]. Anyway, all artificial systems based on this kind of technique are rigid organic polymers that work successfully in organic solvents, but are far from the natural molecular recognition systems based on protein structures. Nevertheless, some theoretical studies suggest the possibility of preparing “protein-like” polymers containing active sites capable of recognising a given target molecule and characterised by a well-defined memory effect towards the template [9], [10], [11].

We adopted a different approach to “classical” molecular imprinting to mimic the binding function of protein-based natural recognition systems. We thought that in a mixture of natural amino acids, non-covalent interactions between a template and the amino acids should be able to shape the structure of growing peptides, forcing the polymerisation process to produce a template-complementary sequence. Moreover, the large structural heterogeneity in the amino acid side chains introduces a great variety of chemical properties, which can be useful in this kind of template polymerisation in order to exploit many different interactions with the template (i.e. ionic, hydrogen-bond, charge-transfer, hydrophobic), increasing the possibility of obtaining an efficient recognition system also in a polar medium such as water; thus mimicking the binding environment of natural receptors like proteins.

The aims of our work were to obtain tailor-made synthetic polypeptides with molecular recognition properties towards low molecular mass molecules. As briefly reported previously [12], we are attempting to demonstrate the feasibility of this approach by preparing mixtures of water-soluble synthetic peptides with molecular recognition properties towards the steroidal hormone estradiol.

Section snippets

Materials and methods

l-Amino acids, methanol, N,N-dimethylformamide (DMF), N,N-dicyclohexylcarbodiimide (DCCD), N,N′-diisopropylcarbodiimide, all chemicals for buffers were from Merck (Darmstadt, Germany). Estradiol and peptides used to measure molecular weights were from Sigma (Milwaukee, USA). Labelled steroids were from Amersham (Uppsala, Sweden) and were in a toluene:ethanol (9:1) solution at the concentration of 25 μCi/ml. The specific activities of the markers were 85 Ci/mmol for testosterone and progesterone,

Synthesis and purification of the peptides

The amino acid composition of the polymerisation mixtures was chosen to be similar to the bovine serum albumin, as it appears to be relatively similar to the average composition of animal proteins. To study the binding properties of both cross-linked and linear structures different amino acid mixtures were polymerised, as reported in Table 1. The first ones contained Lys, Asp and Glu amino acids, able to give cross-linked structures “branched mixtures”; the second ones were different only for

Conclusions

The experimental results reported in this paper show that by polymerising a mixture of amino acids in presence of a molecule able to act as a template, it is possible to obtain a mixture of peptides that show selective molecular recognition properties towards the template itself.

The presence of cross-linking amino acids seems to influence the resulting peptide binding properties. The same occurs by increasing the polymerisation time. As the affinity of peptide mixtures obtained by

Acknowledgements

This work was supported by the Italian Ministry of University and Scientific & Technological Research (MURST), project COFIN99-9903032732-005 “New immunometric methodologies for the detection of endocrine modulators in waters and biological matrices”.

References (14)

  • K. Haupt et al.

    Trends Biotechnol.

    (1998)
  • S.A. Piletsky et al.

    Trends Biotechnol.

    (2001)
  • L.I. Andersson

    J. Chromatogr. B

    (2000)
  • G. Giraudi et al.

    Anal. Chim. Acta

    (1999)
  • D.J. Cram

    Angew. Chem. Int. Ed. Engl.

    (1988)
  • J.M. Lehn

    Angew. Chem. Int. Ed. Engl.

    (1990)
  • F. Lanza et al.

    Adv. Chromatogr.

    (2001)
There are more references available in the full text version of this article.

Cited by (12)

  • Emerging bio-sensing methods for mycotoxin analysis

    2011, Determining Mycotoxins and Mycotoxigenic Fungi in Food and Feed
  • Biosensors for cancer markers diagnosis

    2009, Seminars in Cell and Developmental Biology
  • Affinity chromatography techniques based on the immobilisation of peptides exhibiting specific binding activity

    2003, Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences
View all citing articles on Scopus
View full text