Abstract

Background: Amyloidosis caused by immunoglobulin light chain (IGLC) deposition, so-called AL-type or primary amyloidosis, is the most common type of amyloidosis. It has been long believed that IGLC variable regions form the core of the AL-type amyloid deposits and peptides derived from IGLC constant region peptides are only occasionally integrated into this core. For this reason, the scientific effort to identify thge risk factors for development of AL amyloidosis and the biochemical characteristics amyloid deposits has focused on IGLC variable region derived proteins. To understand the peptide constituents of AL amyloidosis better, we undertook a comprehensive study of AL amyloidosis using a novel mass spectrometry based proteomic analysis approach.

Methods: Paraffin embedded tissue from 100 cases of AL amyloidosis was studied. In each case amyloid type was previously established by clinical and pathological examination. Congo red stained paraffin sections were prepared and amyloid deposits were microdissected by laser microdissection microscopy. The microdissected tissue fragments were processed and trypsin digested into peptides. The peptides were analyzed by nano-flow liquid chromatography electrospray tandem mass spectrometry (LC-MS/MS). The resulting LC-MS/MS data were correlated to theoretical fragmentation patterns of tryptic peptide sequences from the Swissprot database using Scaffold (Mascot, Sequest, and X!Tandem search algorithms). Peptide identifications were accepted if they could be established at greater than 90.0% probability and protein identifications were accepted if they could be established at greater than 90.0% probability and contain at least 2 identified spectra. The identified proteins were subsequently examined for the presence or absence of amyloid related peptides.

Results and Discussion: LC-MS/MS gave peptide profiles consistent with AL amyloidosis in each case. The analysis showed IGLC-lambda deposition in 66 cases and IGLC-kappa deposition in 34 of cases. In each case, LC MS/MS confirmed the previous clinicopathological diagnosis. Interestingly peptides representing IGLC constant region were present in each case. Using this LC-MS/MS methodology, theoretically it is possible to cover 78% of the IGLC-lambda and 87% IGLC-kappa constant regions. In our samples, the average coverage of the IGLC-lambda and IGLC-kappa constant regions were 40% (range 14–78%)and 55% (range 16–87%) respectively. Additionally, the distribution of the peptides suggested that in the majority of the cases whole of the IGLC constant region was deposited. LC MS/MS also identified IGLC-lambda variable region peptides in 37 of 66 cases and IGLC-kappa variable region peptides in 29 of 34 cases studied. The variable region coverage was more restricted and the peptides identified were frequently within the framework segments. It is likely that the peptides derived from CDR segments were present but not detected by the methodology as somatic hypermutation randomly alters the amino acid sequence in the CDR segments and such new sequences are not available in public databases used by algorithms for peptide identification. In the cases with the IGLC variable region hits, it was also possible to assign variable region family usage. IGLC-lambda cases frequently used IGLC-lambda variable region I, II and III families whereas, in IGLC-kappa cases, IGLC-kappa variable region I and III families dominated.

Conclusions:

  1. AL amyloidosis can be accurately diagnosed using laser microdissection and LC-MS/MS based proteomic analysis in routine clinical specimens.

  2. AL amyloidosis invariably contains IGLC constant region peptides and, frequently, the whole of the constant region is deposited. This finding suggests that studies on molecular pathogenesis of amyloidosis should not only consider the IGLC-variable region but also the constant region.

  3. It is possible to identify IGLC variable region family usage in AL amyloidosis using LC MS/MS based proteomic analysis. In the clinical setting, this information may be helpful in predicting organ distribution and clinical outcome.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author