The analysis of the whole tissue proteome requires powerful separation methods to allow sensitive and specific detection and quantification of the different proteins in each sample. This can be achieved by gel-based (two-dimensional polyacrylamide gel electrophoresis, 2D-PAGE, coupled to MS) or gel-free approaches (LC-MS-based methods). The first method is very powerful in providing immediate visualization of the abnormal species [8, 10, 11]. The digital images of the 2D-PAGE gels from patients and controls samples are overlaid, to detect and identify the abnormal protein spots, which are then cut from the gel, digested, and identified by MS. However, this approach is more labor intensive and less automatable than gel-free approaches and, for the clinical purposes, has generally been abandoned in favor of the latter.

The gel-free (so-called “shotgun”) methodologies employ enzymatic digestion of the protein sample (usually with trypsin) prior to analysis, and separation of the peptides by one-dimensional (reverse phase on C18 resin columns) or two-dimensional (reverse phase coupled to strong cation exchange) chromatography. The second approach is employed in the MudPIT technology (Multidimensional Protein Identification Technology), in use by the Pavia/ITB-CNR center (Fig. 24.2). 2D chromatography separates the peptides, which can directly enter the mass spectrometer, where their mass, and the mass of their fragments, is measured. A list of identified proteins is provided after searching the MS data against protein sequence databases (such as NCBInr). The diseased vs. control proteome comparison is performed through dedicated softwares (such as MaProMa [7, 9], developed in-house at ITB-CNR, or commercial ones).

An important step in sample preparation, especially when working with unfixed specimens, is blood removal through washing. In fact, the systemic amyloidosis precursors are blood-borne, and traces of their normal counterparts may be detected in case of blood carryover. Since minor amounts of blood contaminants are often detected in samples, we have developed an algorithm that provides a quantitative parameter (designated α-value [7]) to evaluate the overrepresentation of a specific amyloidogenic protein compared to the other ones. This simplifies the interpretation of proteomic results: amyloid type attribution is achieved not only from the presence of an amyloidogenic protein in the list of identified species but also from its greater abundance (Fig. 24.3).

Abdominal subcutaneous fat, acquired by fine-needle aspiration or by surgical biopsy, can be effectively analyzed by diagnostic proteomics. The tissue preparation workflow is shown in Fig. 24.4 [7–11]. Adipose tissue must be collected in clean microcentrifuge tubes and stored frozen (−80 °C) without fixatives. Care must be paid in freezing the sample as soon as possible (not later than 30 min, during which time it must be kept on ice), in order to prevent protein degradation. Protein extraction is achieved by manually crushing the sample (10–20 mg) in a lysis buffer. Our preferred lysis buffer contains chaotropes (8 M urea, or 7 M urea plus 2 M thiourea), detergents (4 % CHAPS), and a reducing agent (0.1 M DTT). This buffer has been demonstrated to effectively solubilize the proteins species present in amyloid deposits, including those forming amyloid fibrils. Short pulses of sonication help disaggregate the tissue and extract proteins. Delipidation must be performed after protein extraction and prior to further processing, and is obtained through centrifugation (12,000–16,000 rpm). The clear solution below the floating lipids must be carefully recovered using a gel-loader tip. This protein extract would now be ready for protein quantification and 2D-PAGE analysis [10, 11]. In order to proceed to gel-free proteomic analysis, instead, detergents and other interfering buffer components must be removed [7] and proteins must be subjected to protease digestion.

Amyloid typing can also be performed on formalin-fixed paraffin-embedded (FFPE) samples. Notably, aldehyde fixation cross-links proteins and makes these samples unsuited for 2D-PAGE separation. The tissue (1–2 slices, 10-μm thick) is subjected to paraffin removal by incubation in 100 % xylene (3 times, 5 min each), followed by rehydration through consecutive incubations in absolute ethanol, 70 % ethanol, and water. The deparaffinized samples are incubated in a buffer containing 100 mM NH₄HCO₃ at 100 °C for 20 min and then at 100 °C for 2 h. Prior to trypsin digestion, the MS-compatible surfactant Rapigest (Waters) is added to the sample, to a final concentration of 0.5 % in 10 % acetonitrile. Proteins are digested with sequencing grade trypsin in a ratio 1:50 enzyme–substrate and incubated at 37 °C overnight. Peptides are desalted and loaded for LC-MS analysis (using mono- or two-dimensional chromatography). The mass spectrometry and bioinformatic workflow, as well as data interpretation through α-value, have been described above, and are analogous to what performed for proteins extracted from fresh adipose tissue. For each tissue type, a reference proteome map derived from unaffected individuals is used as comparison. Our center currently performs amyloid typing on heart, liver, kidney, lymph nodes, intestine, and salivary glands, using LC-MS.