Structure Elucidation of Human Lens Proteins by Mass Spectrometry

Abstract

Development of fast atom bombardment and electrospray ionization mass spectrometry have greatly facilitated investigations of proteins and peptides. The extraordinary accuracy of the molecular weights determined by these techniques permit proteins and their modifications to be identified with much less ambiguity than was possible with previously available techniques. Combining fast atom bombardment mass spectrometry with directly-coupled HPLC and with collision-induced dissociation mass spectrometry has further extended the information that can be obtained about the primary structures of proteins. We describe the use of these techniques in investigations into the structure of the proteins present in the human eye lens. Modifications of the lens proteins associated with age are of particular interest because lens proteins do not turnover; i.e., the lens of an old person contains even the proteins that were present in the fetus. It is also an important area of investigation because elucidation of the structure of old lens proteins may lead to an understanding of the mechanisms of cataract formation, the most common cause of blindness, worldwide.

Delivery of the first commercial mass spectrometer, manufactured by Consolidated Electronics Corporation, in 1942 is a benchmark for analytical mass spectrometry. During the subsequent forty years, mass spectrometry became a principal tool for identification and quantification of volatile organic materials. Peptides, even those with molecular weights less than 500, are not volatile because they are highly polar, and required chemical derivatization before mass spectrometric analysis¹. In 1980, Barber et al. described the first use of fast atom bombardment mass spectrometry (FABMS) for analysis of involatile substances². During the ensuing 15 years, FABMS has proved extremely useful for determining the molecular weights of peptides with as many as 40–50 residues^3–5. The utility of FABMS has been enhanced by joining this ionization method with directly-coupled HPLC^6,7 and with collision-induced dissociation (CID) mass spectrometry.^8,9

The notion that molecules with an excess of positive or negative charge in solution might be analyzed directly from solution by mass spectrometry had been proposed, but not clearly demonstrated, prior to 1988 when Fenn et al. demonstrated the reality of this wishful-thinking¹⁰. Specifically, they showed that analytically useful ions could be produced when a solution is sprayed from a small needle into a strong electric field. During the past several years, this approach has become known as electrospray ionization mass spectrometry (ESIMS), and is now one of the most important analytical tools available for determining the primary structures of peptides and proteins^11–13.The molecular weights of peptides and proteins to M_R 75,000 are routinely determined with error less than 0.01–0.05% using commercial instrumentation. With this accuracy, it is now possible to determine the molecular weights of proteins with M_R 20,000 (typical of proteins isolated from the eye lens) with an uncertainty of approximately ±2 Da^14,15.

In protein structure studies, ESIMS is used most often to determine the molecular weights of peptides and proteins. The use of this information is limited only by the investigator’s imagination. For example, when the molecular weight of a protein matches, within a few Daltons, the molecular weight calculated from the amino acid sequence of a proposed protein, it may be concluded that the protein has been tentatively identified. If the found molecular weight differs slightly from the expected molecular weight, posttranslational modifications may be considered. For example, acetylation of the N-terminus of a protein, a common in vivo posttranslational modification, increases the molecular weight by 42 Da. This increase is easily detected by ESIMS analysis of most proteins. Furthermore, analysis of a mixture of mono-acetylated and non-acetylated forms of a protein with M_R 50,000 gives separated peaks in the ESI mass spectrum. When compared with traditional methods of determining the molecular weights of proteins, such as sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), ESIMS has vastly higher accuracy, substantially better resolution, and detection limits somewhat lower than those typically achieved by silver staining techniques.

The high performance of ESIMS makes it particularly useful for investigations of the primary structures of complex mixtures of proteins, such as structural proteins extracted from human eye lenses. This tissue, whose function is to focus images on the retina of the eye, is approximately 37% structural protein. The eye lens is unusual because it continues to grow throughout the life of an individual by adding more layers of cells, and because the protein never turns-over. As a result, copies of the lens proteins present in the inner portion (the nucleus), of the lens, were present at birth. Identification of which gene products are actually expressed, as well as posttranslational modifications that occur during the life of an individual, are important for advancing our understanding of the natural in vivo aging of lens proteins and of the possible causes of cataract. The physiology and biochemistry of the lens have been described in detail elsewhere^16,17.

Approximately 15 different genes code for the most abundant structural proteins found in the lens. Fractionation of lens extracts by gel filtration chromatography (Sephadex G-200) may be used to isolate the three general classes, α-, β-, and γ-crystallins, of lens proteins^18,19. The α-crystallins consist of two gene products (αA and αB), which aggregate to form large amorphous complexes with an average molecular weight of 500,000 Da. The β-crystallins include at least 8 different gene products, which aggregate to form complexes with average molecular weights of 80,000 to 150,000 Da. The γ-crystallins consist of 6 different gene products and do not aggregate under normal conditions. All of the crystallin monomers have molecular weights in the 20–30 kDa range. The N-termini of all the α- and β-crystallins and γs-crystallin are acetylated.

Because individual copies of each protein are as old as the cell in which they were formed, and those in the center of the lens are as old as the person, one can imagine that a lens from an elderly person has crystallins that have undergone a variety of modifications. For example, modifications that have been found in the human eye lens include phosphorylation,^20–22 C-terminal truncation,^14,23 deamidation,^22,24 and disulfide bonding.^22,25,26 Oxidation of Met, Trp, and Cys have also been proposed as likely modifications. Understanding the aging of proteins, as well as chemical reactions causing or accompanying cataract, poses a significant analytical challenge. One would like to know which modifications occur, and specifically on which residue in a particular gene product. Furthermore, one would like to know the extent to which these modifications are present, as well as the point in lens development at which they occurred.

Our general approach to identifying crystallins present in human eye lenses, including their modified forms, starts with their separation into α-, β-, and γ- fractions by gel filtration chromatography. Constituents of these fractions are further purified by a combination of ion exchange and reversed phase high performance liquid chromatography (HPLC). Although the final products of this isolation procedure may contain only one gene product, different modifications (phosphorylation, deamidation, truncation) may be included, and these modifications may be located at different sites along the peptide backbone. These semi-pure samples are analyzed by ESIMS to determine the molecular weights of the constituent proteins, and to assess the purity of the sample. The molecular weights may be used to identify the protein as belonging to a particular gene, and to tentatively identify the existence of posttranslational modifications. The sample can also be fragmented into peptides whose molecular weights are determined either by ESIMS or FABMS. This molecular weight information may be used to relate large but incomplete protein fragments to specific genes, and to identify specific posttranslational modifications. Peptides that cannot be identified by systematically searching proposed sequences of proteins believed to be present in the sample²⁷ may be sequenced by CID mass spectrometry. This technique is also useful for determining which residues are modified. The purpose of this communication is to describe how we have used mass spectrometry to investigate the primary structure of human lens crystallins.

Represents person presenting Paper