Antibody-based Protein Profiling of Cerebrospinal Fluid in Multiple Sclerosis

In recent years, multiplexed and high-throughput protein profiling of unfractionated plasma and serum samples using a direct labelling approach has facilitated affinity-based biomarker discovery. The Human Protein Atlas, version 11.0, for example,  hosts  more than 18,000 antibodies to approximately 75% of the human protein encoding genes (>15,000 genes), enabling the study of protein expression in normal and cancerous tissues and cell lines (www.proteinatlas.org).¹

Anna Häggmark, PhD, and colleagues from the Royal Institute of Technology (Stockholm, Sweden) now describe the development and application of an assay for  protein profiling of cerebrospinal fluid (CSF) in multiple sclerosis (MS), adapting an established protocol for multiplexed analysis of serum and plasma.^2,3  

MS, a neurological disease characterized by the demyelination of axons, occurs in one  of four subtypes—relapsing remitting MS (RRMS), primary progressive MS (PPMS),  secondary progressive MS (SPMS) and progressive relapsing MS (PRMS), based on the extent of remission between clinical episodes and recovery, if any, between relapses. In the absence of a quick diagnostic tool to detect variation in the disease spectrum, researchers have felt the need for reliable biomarkers and an understanding of the underlying disease mechanisms.

Using an antibody suspension bead array technology, the Swedish researchers have created protein profiles of CSF that suggest disease-related proteins in MS. The technique utilizes biotin for direct labelling of proteins via their primary amines. The diluted CSF  samples [1:2 in PBS or PBS supplemented with 0.5% (w/v) bovine serum albumin (BSA, Sigma) and 0.1% (w/v) rabbit IgG (Bethyl), denoted BIG buffer] were labelled at 4°C with a 10-fold molar excess of biotin (NHS-PEG4-Biotin, Thermo Scientific) over total protein content (approximated to 0.5 mg/ml for all crude CSF samples; average protein size estimated at 70 kDa). The labelling was terminated a couple of hours later by adding 1M Tris-HCl in a 250-fold molar excess over biotin, and samples were further processed. The dilution factors were modified to account for the lower total protein concentration in CSF compared to plasma, and also tested for the direct labelling procedure as well as subjected to two heat treatment temperatures (at 56°C and 72°C) for epitope retrieval. The assay was used to profile 339 CSF samples from MS patients and controls, with 101 antibodies targeting 43 different proteins. Of the 339 CSF and plasma samples analyzed, 209 were either from patients diagnosed with PPMS, RRMS and SPMS subtypes or from patients with one episode of a clinically isolated syndrome of neurological symptoms. The 130 control samples included patients with other neurological diseases, either with or without signs of inflammation.

In the current study, the investigators added an excess of BSA and IgG to all samples before sample biotinylation in order to adjust for the relative differences in total protein amounts among samples and to formulate a more similar ratio of protein content and biotin in all samples. Thereby, the researchers successfully reduced the differences, in terms of biological content and composition of  samples, without affecting the ability to detect the desired protein targets using the assay.

Albumin and IgG are the two most abundant proteins in CSF. The composition of high abundance proteins can vary in CSF depending on disease status, potentially skewing the results in group comparison studies. The researchers also assessed the effects of heat treatment on the analysis of CSF proteins.

The results showed that the growth-associated presynaptic protein GAP43 was detected at significantly lower levels in late-stage secondary progressive MS, compared to early stages of MS and the control group of other neurological diseases. GAP43 plays a key role in axonal growth,⁴ with decreased GAP43 levels linked to demyelinated brain lesions. The finding of decreased levels of GAP43 in SPMS patients in the current study suggests increased axonal damage and worse prognosis. The importance of CSF analysis can be appreciated by the differences in GAP43 content between RRMS and SPMS in CSF samples alone, but not in plasma or blood-derived samples obtained from these individuals.

The investigators noticed elevated levels of serpin peptidase inhibitor, clade A, member 3 (SERPINA3), an acute phase protein that is released early in response to inflammation, in all MS patients (compared to controls); it was also seen in the control patients with inflammation, at both temperatures. The antibodies targeting SERPINA3 revealed higher relative intensities in all MS subtypes and  in the control patients diagnosed with other neurological diseases with signs of inflammation. The investigators attribute the increased levels of SERPINA3 to an overall pattern of inflammation in the body and not MS alone.

The researchers state that their technology opens up possibilities to perform broad-scale, antibody-based protein profiling of CSF. “Using the single binder assay setup and a direct labeling approach, protein supplementation of CSF prior to labeling reduces the effect of differential total protein concentration,” the researchers conclude. The study paves the way for discovery of new candidate molecules for diagnostics based on the assessment of the CSF protein content, regardless of albumin and IgG levels.

References

1. Uhlen, M., et al. (2010) “Towards  a  knowledge-based Human Protein Atlas,” Nature Biotechnology, 28 (pp. 1248–1250), doi:10.1038/nbt1210-1248.

2. Häggmark, A., et al. (2013, May 21) “Antibody-based profiling of cerebrospinal fluid within multiple sclerosis,” Proteomics, doi: 10.1002/pmic.201200580.

3.  Schwenk, J.M., et al. (2008) “Antibody suspension bead arrays within serum proteomics,” Journal of Proteome Research, 7 (pp. 3168–3179), doi: 10.1021/pr700890b.

4. Oestreicher, A.B., et al. (1997) “B-50, the growth associated protein-43: modulation of cell morphology and communication in the nervous system,” Progress in Neurobiology, 53 (pp. 627–686), http://dx.doi.org/10.1016/S0301-0082(97)00043-9.