Quantification of cell surface proteins with bispecific antibodies.

Flow cytometry is an established method for fast and accurate quantitation of cellular protein levels and requires fluorescently labeled antibodies as well as calibration standards. A critical step for quantitation remains the production of suitable detection antibodies with a precisely defined ratio of antigen-binding sites to fluorophores. Problems often arise as a consequence of inefficient and unspecific labeling which can influence antibody properties. In addition, the number of incorporated fluorophores necessitates a special normalization step for quantitation. To address these problems, we constructed different mono- and bivalent bispecific antibodies with binding site(s) for the cell surface antigens, cMET, EGFR1/HER1, ErbB2/HER2 or ErbB3/HER3 and with an additional digoxigenin-binding single-chain Fv fusion. The fluorophore Cy5 was covalently coupled to digoxigenin and quantitatively bound by the bispecific antibody. A panel of tumor cell lines was assessed under different culture conditions for absolute receptor expression levels of the indicated antigens and the data were set in relation to mRNA, gene count and immunoblot data. We could reproducibly quantify these receptors, omit the otherwise required normalization step and demonstrate the superiority of a 1 + 1 bispecific antibody. The same antibodies were also used to quantify the number of proteins in intracellular vesicles in confocal microscopy. The antibodies can be stored like regular antibodies and can be coupled with different digoxigenin-labeled fluorophores which makes them excellent tools for FACS and imaging-based experiments.

Stigmata: an algorithm to determine structural commonalities in diverse datasets.

An algorithm, Stigmata, is described, which extracts structural commonalities from chemical datasets. It is discussed using several illustrative examples and a pharmaceutically interesting set of dopamine D2 agonists. The commonalities are determined using two-dimensional topological chemical descriptions and are incorporated into the key feature of the algorithm, the modal fingerprint. Flexibility is built into the algorithm by means of a user-defined threshold value, which affects the information content of the modal fingerprint. The use of the modal fingerprint as a diversity assessment tool, as a database similarity query, and as a basis for color mapping the determined commonalities back onto the chemical structures is demonstrated.

CHORTLES: a method for representing oligomeric and template-based mixtures.

Screening mixtures of synthetic oligomers or fixed templates (e.g., rings) with varying substituents is increasingly the focus of drug discovery programs. CHORTLES is designed and implemented to facilitate representation, storage, and searching of oligomeric and template-based mixtures of any size. Building upon the CHUCKLES method of representing oligomers as both monomer-based sequences and all-atom structures, CHORTLES compactly represents a mixture without explicitly enumerating individual molecules. This method lends itself to a hierarchy relating mixtures to submixtures and individual compounds, as one finds when deconvoluting mixtures in drug lead discovery programs. In addition, we describe two methods of searching mixtures at the monomer level. We also present a simple pictorial representation for describing all components in a mixture, which becomes essential as the list of monomer names is expanded beyond common names (e.g., amino acids).

CHUCKLES: a method for representing and searching peptide and peptoid sequences on both monomer and atomic levels.

Dual representation of peptide and non-peptide structures in a chemical database as atomic-level molecular graphs and sequence strings permits chemical substructure and similarity searches as well as sequence-based substring and regular expression searches. CHUCKLES interconverts monomer-based sequences with SMILES, which represent atomic-level molecular graphs. Forward-translation maps peptide or other sequences into SMILES. Back-translation extracts monomer sequences from SMILES. This approach permits a generalized representation of monomers allowing user specification of any monomer. CHUCKLES allows mixing of atoms with user-defined monomer names; that is, monomer representation is consistent with SMILES notation. In addition, oligomer branching and cyclization are handled.

ALADDIN: an integrated tool for computer-assisted molecular design and pharmacophore recognition from geometric, steric, and substructure searching of three-dimensional molecular structures.

ALADDIN is a computer program for the design or recognition of compounds that meet geometric, steric, and substructural criteria. ALADDIN searches a database of three-dimensional structures, marks atoms that meet substructural criteria, evaluates geometric criteria, and prepares a number of files that are input for molecular modification and coordinate generation as well as for molecular graphics. Properties calculated from the three-dimensional structure are described by either properties calculated from the molecule itself or from the molecule as compared to a reference molecule and associated surfaces. ALADDIN was used to design analogues to probe a bioactive conformation of a small molecule and a peptide, to test alternative superposition rules for receptor mapping of the D2 dopamine receptor, to recognize unexpected D2 dopamine agonist activity of existing compounds, and to design compounds to fit a binding site on a protein of known structure. We have found that series designed by ALADDIN show much more subtle variation in shape than do those designed by traditional methods and that compounds can be designed to be very close matches to the objective.

MENTHOR, a database system for the storage and retrieval of three-dimensional molecular structures and associated data searchable by substructural, biologic, physical, or geometric properties.

MENTHOR is a database system for the storage and retrieval of three-dimensional coordinate and charge information on molecules as well as of traditional biological and physical properties. Our molecular graphics system retrieves from MENTHOR structural information in individual molecules and receptor map/macromolecular binding site hypotheses. Substructural searches of MENTHOR are used to find starting coordinates for molecular modeling and traditional database searches of MENTHOR identify compounds for which modeling is needed. It also forms the data to be searched with ALLADDIN, our substructure/geometric search program. MENTHOR expedites molecular modeling by organizing previous work and facilitating transmission of information between individuals. Examples from modeling of D-2 receptor agonists are shown.