Evaluation of Apolipoprotein E Fragmentation as a Biomarker for Alzheimer’s Disease

Recent studies have supported a role for the proteolytic cleavage of apolipoprotein E4 (APOE4) as a potential mechanism for the enhanced dementia risk associated with Alzheimer’s disease. To determine whether APOE4 fragmentation is correlated with AD, ELISA assays were performed with cerebral spinal fluid (CSF) and plasma samples utilizing an antibody that specifically detects a 17 kDa amino-terminal fragment (p17) of APOE (nApoECF antibody). In CSF samples, levels of APOE fragmentation were minimal in both neuropathological normals (NPNs) and AD cases and there were no significant differences between the two cohorts across APOE genotypes. Similar results were found in plasma samples where the p17 APOE fragment comprised only 8.4% of the total level of identified APOE. As with CSF, there were no significant differences found between NPNs and AD cases in terms of the amount of nApoECF quantified. Taken together, these results suggest that the p17 amino-terminal fragment of APOE is not correlated with AD or APOE genotype in the plasma or CSF.

Keywords: Apolipoprotein E; Cerebral spinal fluid; Alzheimer’s disease; Plasma; Biomarker; Proteolysis

The apolipoprotein E (APOE) allele is polymorphic leading to the expression of three proteins (APOE2, E3, and E4) that differ by single amino acid substitutions of cysteine to arginine at positions 112 and 158, respectively, of the full-length form of the protein [1]. Harboring the APOE3 allele is believed to neither increase nor decrease one’s risk of AD, while having the E2 form may decrease one’s risk. In contrast, inheritance of one copy of the APOE4 allele increases disease risk fourfold, while two copies raises the risk tenfold [2]. It has been well established that harboring the APOE4 allele enhances dementia associated with Alzheimer’s disease (AD), and several studies have supported a role of APOE4 proteolysis as an important factor that may contribute to this risk [2,3-10]. Thus, the high susceptibility of APOE4 to proteolytic cleavage may be the driving force behind the enhanced risk for AD in individuals who are either heterozygous or homozygous for the APOE4 allele. Recently, we reported the presence of a 17 kDa fragment (p17) of APOE4 in the AD brain (corresponding to amino acids 1-151) that localized intracellularly, predominantly within the nucleus of microglia and neurons and the levels of this fragment was significantly higher as compared to neuropathological normals (NPNs) [11]. These findings were accomplished using a site-directed cleavage antibody to APOE that is highly specific for a 17 kDa amino-terminal fragment of APOE3 or E4 but does not react with the mature, full-length form of APOE [11]. In the present study, we sought to examine the utility of this antibody (termed nApoECFp17 antibody) to detect the p17 fragment in either cerebrospinal fluid or plasma samples. The results indicated the levels of the p17 fragment were very low in both fluids and no significant differences between the NPN and AD cohorts were observed. Therefore, the 17 kDa amino-terminal fragment of APOE does not appear to be a viable biomarker for predicting AD.

The in house, rabbit polyclonal nApoECFp17 antibody was synthesized and extensively characterized as previously described and detects the amino-terminal fragment of APOE corresponding to amino acids 1-151 [11]. The nApoECFp17 antibody was synthesized using the 7-mer peptide C-RKRLLRD, which represents the N-terminal upstream neoepitope fragment of APOE4 and E3 that would be generated following cleavage after the terminal aspartic acid residue at position D151.The anti-APOE4 N-terminal rabbit polyclonal antibody was purchased from Aviva Systems Biology Corp. (San Diego, CA). Peroxidase-conjugated AffiniPure goat anti-rabbit (IgG) was purchased from Jackson ImmunoResearch Labs (West Grove, PA). This antibody is known to recognize the extreme N-terminal region of human APOE. The TMB microwell peroxidase substrate system was purchased from KPL, Inc. (Gaithersburg, MD). 96-well, clear, flat bottom, high binding, polystyrene plates were purchased from Corning (Kennebunk, ME).

CSF and plasma samples from NPNs as well as AD subjects were provided by the Institute for Brain Aging and Dementia Tissue Repositories at the University of California, Irvine. Approval from the Boise State University Institutional Review Board was not obtained due to the exemption granted that all samples were received from University of California, Irvine. Fluid samples obtained from University of California, Irvine were anonymized and never identified except by case number. Donors or their next of kin provided informed signed consents to the Institute for Memory Impairments and Neurological Disorders for the use of their biological samples in research (IRB 2014-1526). Case demographics are presented in Table 1.

Protein content for CSF and plasma samples was determined using the BCA method (Pierce) to ensure equal protein loading between samples. In 96-well plates, 50 μl of CSF and/or 25 μl of plasma diluted with 100 mM carbonate coating buffer (pH, 9.7) to a final volume of 100 μl and were incubated overnight at 4 °C on a slow rocker. Wells were washed 3X in tween, tris-buffered saline (TBS, 10 mM TRIS, 100 mM NaCl, and .05% Tween 20). Following this wash step, 300 μl of blocking solution (3% non-fat dried milk in TBS) was added and the plate was incubated for 1 hour at 37 °C. Following a wash step (3X), wells were then incubated for 2 hours at 37 °C with the nApoECFp17 antibody at a 1:500 dilution in TBS containing 0.3% non-fat dried milk. The two hour incubation time and temperature of incubation at 37 °C were determined to be optimal parameters based on our previous studies [11]. Following the wash step (3X), wells were then incubated for 1 hour at 37 °C with anti-rabbit secondary HRP antibody at a 1:5,000 dilution in TBS containing 0.3% non-fat dried milk. The plate was then washed (3X) in TBS followed by the development of color utilizing the TMB microwell peroxidase substrate system. Color was allowed to develop for 4 minutes before the addition of a stop solution. The relative absorbance at 450 nm was recorded using a BioTek Synergy Mx microplate reader. All samples were evaluated in duplicate and data are representative of three independent experiments ±S.E.M. To calculate the percent of total nApoECFp17 fragment present in either CSF or plasma, we used the overall OD value obtained using the full-length APOE antibody as representing the total amount of both full-length and any associated amino-terminal fragments of APOE. The value for nApoECF was then subtracted by this value and then divided by the value obtained by full-length APOE. For example, to calculate the percent of nApoECF found in CSF samples, 1.57 (averaged OD for full-length APOE)-.090 (averaged OD for nApoECF/1.57 (averaged OD for full-length APOE), which was equal to 94.2%. This number was then subtracted from 100% to give a value of 5.8% nApoECF fragment present in the CSF samples. A similar calculation was undertaken to determine the percent of nApoECF found in plasma samples.

Statistical differences in this study were determined using Student’s two-tailed T-test employing Microsoft Office Excel to determine significant differences between samples. P-values greater than 0.05 were considered non-significant.

In a previous study, we designed and characterized a site-directed cleavage antibody directed towards position D151 of the mature form of APOE. This antibody (termed the nApoECFp17 antibody) does not react with full-length APOE and specifically detects a 17 kDa fragment in AD cases that localizes predominantly within the nucleus of microglia [11]. Immunohistochemical analysis of AD cases indicated staining of nApoECFp17 was exclusively intracellular with no evidence of immunoreactivity in extracellular, plaque-rich regions. However, we did observe staining along blood vessels and therefore hypothesized that the p17 fragment of APOE may serve as a biomarker in AD. To test this hypothesis, we first examined whether this amino-terminal fragment of APOE could be detected in CSF in both AD and neuropathological normal (NPN) cases and compared these results to those obtained using an antibody that detects full-length APOE. It is important to note that the nApoECFp17 antibody does not distinguish between the various isoforms of APOE and will readily detect amino-terminal fragments of APOE3 and E4 [11].

As shown in Figure 1, ELISA analysis of CSF using the nApoECFp17 antibody showed no significant differences between NPNs and AD subjects (Figure 1A). Indeed, there was a trend for lower levels of the p17 fragment in AD cases as compared to NPNs. It is noteworthy that all data were expressed in relative units due to lack of any control antigen that could possibly provide for a standard curve.

Previous studies have supported the hypothesis that APOE4 is more susceptible to proteolytic cleavage than APOE2 and E3 (for a recent review see [5]). Therefore, we sought to determine whether there were any differences in the levels of the p17 fragment based on APOE genotype alone (both AD and NPN subjects). Due to the rarity of the APOE2 allele, we analyzed all heterozygous APOE2/X together for this experiment. There were no significant differences across APOE genotypes following this analysis (Figure 1B). In contrast to these results, the use of a positive control consisting of the immunogen used to generate the nApoECFp17 antibody consisting of the 7-mer peptide C-RKRLLRD, resulted in significantly higher OD values (green bar labeled positive control, Figure 1A).

The results from Figure 1 indicated not only little difference in the levels of the p17 fragment between cohorts, but also very little fragment was identified as compared to background controls (typically in the range of .035-.045 Relative OD units). To confirm this finding, we repeated ELISAs using a commercially available antibody to the extreme amino-terminus of APOE that detects full-length APOE (299 amino acids) as well as any putative amino-terminal fragments. As Figure 2A indicates, robust levels of full-length APOE were detected under these conditions, however, like the p17 fragment, no significant differences were found between the NPN and AD cohorts. Comparison of the two antibodies indicated that approximately 94.2% of the APOE detected in the AD cohorts was full-length, while only 5.8% was equated to the p17 amino-terminal fragment of APOE (Figure 2B).

We also assessed plasma samples for the p17 fragment of APOE and obtained similar results as those reported for CSF (Figure 3). In this regard, no significant difference was observed between the NPN and AD cohorts (Figure 3A), and overall the p17 fragment made up only 8.4% of the total APOE detected in plasma AD samples (Figure 3B). There was a trend for lower levels of full-length APOE in the AD cohort but the data did not reach statistical significance (Figure 3B).

Human APOE is polymorphic with three major isoforms, APOE2, APOE3 and APOE4, which differ by single amino acid substitutions involving cysteine-arginine replacements at positions 112 and 158 [1]. Harboring the APOE3 allele is believed to neither increase nor decrease one’s risk of AD, while having the E2 form may decrease one’s risk. In contrast, inheritance of one copy of the APOE4 allele increases disease risk fourfold, while two copies raises the risk tenfold [2]. It is noteworthy that 65-80% of all AD patients have at least one APOE4 allele [12,13].

Numerous studies have supported a role for apoE4 proteolysis as a putative mechanism linking APOE4 to the dementia risk associated with AD [3-10]. Thus, proteolysis of APOE4 may promote AD pathogenesis through a loss of function (i.e., inability to shuttle cholesterol in the CNS) or due to a toxic-gain of function whereby the cleaved amino-terminal domain becomes neurotoxic [5]. We recently characterized a novel antibody that specifically detects a 17 kDa fragment (p17) of APOE2, E3 and E4. The antibody shows no reactivity to full-length APOE but readily identifies the p17 fragment by Western blot in the AD brain [11]. In addition, by immunohistochemistry it was demonstrated that the fragment preferentially localizes to the nucleus of microglia cells in the AD brain [11].

The purpose of the current study was to examine whether the p17 fragment identified by nApoECFp17 antibody could serve as a useful biomarker in AD. Identifying protein biomarkers of AD could help uncover novel treatment approaches by allowing for early diagnosis and treatment initiation. As an initial approach, we screened CSF samples for the presence of the p17 fragment utilizing the nApoECFp17 antibody and ELISA. Very little of the p17 fragment was identified, and there were no significant differences between controls and AD cohorts or across APOE genotypes. A comparison of the amount of full-length APOE versus the p17 fragment revealed 89% and 11%, respectively, in AD CSF samples. Similar findings were obtained following analysis of plasma samples: little of the p17 fragment was identified (5% as compared to full-length APOE), and there were no significant differences between NPN and AD cohorts. Interestingly, for full-length APOE in the plasma, there appeared to be a trend for lower levels in the AD cohort as compared to NPNs, but this did not reach statistical significance (p=0.27). These data do suggest that proteolysis of APOE within the plasma compartment may be occurring to a greater degree in AD perhaps generating a unique fragment distinct from the p17 fragment.

In conclusion, we attempted to verify the presence of an amino-terminal fragment of APOE in fluid samples in the AD brain and to determine whether it could serve as a potential biomarker for this disease. Our results support the conclusion that there are very low levels of the p17 fragment in extracellular fluids of the AD brain including the CSF and plasma suggesting that this proteolytic product is not a viable biomarker for AD. These data support our previous findings that the labeling of nApoECFp17 antibody in the AD brain is exclusively intracellular [11]. Future studies will be necessary to determine if other, unique proteolytic fragments of APOE4 are present in AD plasma samples and could serve as potential biomarkers for this neurodegenerative disorder.

This work was funded by National Institutes of Health Grant 1R15AG042781-01A1 and AG0165773 to WWP. This study was also supported by the KO Alzheimer’s Dementia Foundation (Boise, ID). The project described was supported by Institutional Development Awards (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health under Grants #P20GM103408 and P20GM109095. We also acknowledge support from The Biomolecular Research Center at Boise State with funding from the National Science Foundation, Grants #0619793 and #0923535; the MJ Murdock Charitable Trust; and the Idaho State Board of Education. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.