Protein biomarkers and patterns of protein biomarkers may provide potential for diagnostics, intelligent drug design and intervention monitoring.
One approach is to identify disease biomarkers upon quantification of a global pattern of proteins and peptides in serum or CSF of AD and controls. The most used techniques for protein profiling and identification of potential markers are 2-D gel electrophoresis and the protein chip arrays on which bound proteins are detected by a mass spectrometer and identified (surface enhanced laser desorption ionization (SELDI)- time of flight (TOF) mass spectrometry (MS) system). SELDI-TOF MS allows rapid protein profiling as well as identification and characterization of novel protein biomarkers .
Although these techniques are powerful protein profiling tools, they do not allow profiling of the whole proteome for potential biomarkers, and thus a combination of (complementary) techniques is necessary. However, when a specific question is if a specific isoform pattern is involved, there is no need to cover the whole proteome, and "focussed or targetted proteomics" may be used.
Difficulties encountered with the identification of protein biomarkers for a disease in biological fluids are, that newly translated proteins are extensively modified (glycosylation, phosphorylation, proteolytic processing) which determines their functional activity, and that detection of many low abundant proteins in CSF and serum is hampered by the presence of highly anbundant proteins (albumin, immnoglobulins, beta-trace protein etc). Especially searching for plasma biomarkers for AD is hampered, because levels of proteins of interest are much lower in plasma than in CSF (blood brain barier), whereas the levels of high abundant proteins are much higher in plasma than CSF.
To improve the detection of low-abundant proteins, methods (organic solvent precipitation, affinity dye based depletion, ultrafiltration, size-exclusion chromatography) are used to remove these proteins or only the high molecular weight proteins from the sample. Another approach is to enrich for the low-abundance proteins / peptides of interest in CSF or serum by immunoprecipitation (IP) or affinity chromatography. The selective capturing of a specific protein/peptide with antibodies in combination with mass spectrometry (IP-MS) is most used, however also a class of proteins, such as phospho-proteins can be captured. An advantage of this technique over depletion of high abundant proteins is, that with depletion, the peptides/proteins of interest may be lost, when these are associated with the high abundant carrier proteins. Even, in a recent study the binding of various factors to albumin was used to analyze the albumin enriched low molecular weight proteome in serum, which resulted in the identification of three peptides/proteins as potential biomarkers .
Most proteomic studies to identify biomarkers for AD have used an unbiased approach using 2D- gels and MALDI TOF (matrix-assisted laser desorption/ionisation- time of flight), or searched for specific markers with SELDI-TOF MS. Stable isotope labeling is now also applied to detect and identify proteins that are expressed differently in AD and controls [2,313].
By now (see the elegant and comprehensive overview on biomarker discovery in neurodegenerative diseases by Zetterberg et al ) in a large number of studies, a number of factors that are either up- or downregulated or of which certain isoforms or splice products are detectable, have been reported. These include different N-truncated Ap forms, as well as different isoforms of apoE and apoA and complement factors (see Tabel 1).
Of interest is that with use of a combination of techniques (separation of serum proteins with LC and 2D-gels and identification with MALDI TOF and ion-trap MS) several inflammation related proteins, including protein factors C4 and C3 and factor H, a cofactor for inactivation of activated C3, are found upregulated in serum of AD compared to controls [121,314]. Polymorphism (Y402H) in Factor H has been associated with age-related macular degeneration, and recently also with AD , the pathologies of which are both associated with Ap deposition  and APOE polymorphism .
Using the SELDI TOF approach 15 biomarkers could be selected from CSF that could distinguish between patients with stable MCI and patients with MCI who progressed to AD. Regression analysis was used to determine the best combination of markers for distinguishing AD from CTL. The resulting multi marker model consisted of Cystatin C, N-terminally truncated cystatin C, Ap1-40, C3a anaphylatoxin des-Arg) and an unidentified 4,0 kD peak, combined with ELISA data for Ap1-42 and Tau . Interestingly, this set contains some factors (cystatin C and complement activation products) already known from immunohistochemical and animal model studies to be of interest.
The approach chosen, at least in part, seems to determine the kind of biomarkers evolving from the search. For example, the combination of 2-D gels and MS in one study yielded five proteins differentially-expressed in AD and controls. Apo A-1, cathepsin D and transthyretin (TTR), were significantly reduced in AD, whereas hemopexin (HPX) and two pigment epithelium-derived factor (PEDF) isoforms were increased in AD CSF .
In another study, 2-D difference gel electrophoresis (2-D DIGE) was used to identify candidate markers differentially-expressed in individual CSF samples from subjects with very mild dementia (believed to be clinically due to AD) and controls after depletion of high-abundant proteins, yielded 11 spots. Upon identification by MS/MS Cystatin C, Prostaglandin D2 synthase, p2-microglobulin, GP-39 cartilage protein and thioredoxin were found to be increased, and 1 P-glycoprotein decreased in the very mild demented group . Differences between the latter studies may be due to the different stage of the disease process (mild demented versus advanced AD stage), or to the technical approach. Removal of abundant proteins may also lead to loss of factors of interest that may co-precipitate or co-purify with these proteins.
Limitations of 2-D gels are, that only a limited number of spots can be matched across a large number of gels / images, due to inconsistency of the 2-D gel methods resulting in image artifacts, such as inadequate resolution, vertical and horizontal streaking, and particularly, local geometric distortions. Intrinsic to the 2-DE-based methodology is the separation of multiple isoforms for each protein, which implies that changes in abundance for a protein spot do not necessarily correlate with the change in total abundance of the corresponding protein . The application of orthogonal methodologies, such as ICAT and the newly developed ITRAQ (i.e. an amine-reactive isobaric tagging reagent-based protein quantification method) [2,313] may prove to be the most powerful discovery approach for AD biomarkers .
Was this article helpful?