Interrelationship between the Levels of C9orf72 and Amyloid-β Protein Precursor and Amyloid-β in Human Cells and Brain Samples

Abstract

A subset of C9orf72 repeat expansion-carrying frontotemporal dementia patients display an Alzheimer-like decrease in cerebrospinal fluid amyloid-β (Aβ) biomarker levels. We report that downregulation of C9orf72 in non-neuronal human cells overexpressing amyloid-β protein precursor (AβPP) resulted in increased levels of secreted AβPP fragments and Aβ, while levels of AβPP or its C-terminal fragments (CTFs) remained unchanged. In neuronal cells, AβPP and C83 CTF levels were decreased upon C9orf72 knockdown, but those of secreted AβPP fragments or Aβ remained unchanged. C9orf72 protein levels significantly increased in human brain with advancing neurofibrillary pathology and positively correlated with brain Aβ₄₂ levels. Our data suggest that altered C9orf72 levels may lead to cell-type specific alterations in AβPP processing, but warrant further studies to clarify the underlying mechanisms.

Keywords

Alzheimer’s disease amyloid-β amyloid-β protein precursor frontotemporal dementia

INTRODUCTION

The GGGGCC hexanucleotide repeat expansion in C9orf72 gene is a major genetic cause of frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS) [1, 2]. The frequency of the C9orf72 repeat expansion varies in different countries. For example, 2–19% of sporadic and 14–48% familial FTD cases, with Finland showing the highest frequency, are expansion carriers [3]. The suggested pathogenic mechanisms in the carriers include haploinsufficiency and formation of RNA foci and dipeptide-repeat (DPR)-containing protein inclusions in cells [1 , 4–8]. In humans, C9orf72 encodes three mRNA variants, which yield two protein isoforms, the long isoform a (∼50 Da) and the short isoform b (∼25 kDa). Isoform a is implicated in vesicular trafficking in the endosomal and autophagosomal/lysosomal pathways [9 –11] and isoform b in nucleocytoplasmic trafficking [12].

In addition to FTD or ALS, C9orf72 repeat expansion has been identified in some patients with other neurodegenerative disorders, including Alzheimer’s disease (AD). However, the significance of the repeat expansion beyond disorders of the FTD-ALS spectrum is not clear [13 –15]. AD is the most common cause of dementia and its central neuropathological hallmarks include intraneuronal neurofibrillary tangles, composed of hyperphosphorylated tau protein, and amyloid-β (Aβ) plaques in the brain parenchyma [16]. Increased levels of total and phosphorylated tau and decreased levels of Aβ₄₂ peptides in the cerebrospinal fluid (CSF) are used as diagnostic biomarkers for AD [17, 18]. We have previously reported that 25% of Finnish C9orf72 repeat expansion-carrying FTD patients show decreased CSF Aβ_1 - 42 levels [19]. The patients with such altered biomarker status fulfilled the clinical criteria of the behavioral variant FTD with no clinical signs of AD. In this study, neuropathological confirmation was available only for one patient, showing TAR DNA-binding protein 43 (TDP-43) pathology, but no Aβ or tau pathology [19]. Albeit the reason for the decrease in the CSF Aβ_1 - 42 levels remains thus far unknown, these results suggest that decreased CSF Aβ1-42 levels may not exclude C9orf72 repeat expansion-carrying FTD patients in clinical diagnostics [19]. Given this AD-like CSF Aβ finding in a subset of C9orf72 expansion-carrying FTD patients, we asked here if modifying the levels of C9orf72 affects AβPP processing and generation of Aβ peptides.

MATERIALS AND METHODS

siRNAs and cDNA constructs

Dharmacon (ON-TARGETplus, L-013341-01-005; siRNA 1) or Santa Cruz (sc-92761; siRNA 2) C9orf72 siRNA pools were used for C9orf72 knockdown. Silencer Negative control siRNA (Ambion, AM4611) was used as a control. Expression cDNA constructs encoding C9orf72 isoform a containing a C-terminal green fluorescent protein (GFP) tag (isoform a-GFP) [11] and C9orf72 isoform b containing a C-terminal myc-DDK tag (isoform b-myc-DDK) were purchased from Origene. Empty plasmid (pcDNA3.1, Invitrogen) was used as a control in the cDNA transfections.

Cell culture, transfection, and treatments

Human embryonic kidney HEK293-AP-AβPP cells overexpressing AβPP751 with a thermostable N-terminal alkaline phosphatase (AP) tag were cultured as previously [20, 21]. Human H4 neuroglioma cells overexpressing AβPP751 (H4-AβPP751) were cultured as in [22]. Twenty-five nM of each siRNA or 4μg of plasmid DNA and 10μl Lipofectamine 2000 reagent (Invitrogen) were used per transfection according to manufacturer’s instructions to knockdown C9orf72 or overexpress C9orf72 protein isoforms. Fresh media were added 6 h post transfection. A subset of HEK293-AP-AβPP or H4-AβPP751 cells were treated with 100 ng/ml phorbol myristate acetate (PMA) and 250 nM γ-secretase inhibitor N-[N-(3,5-Difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester (DAPT) for 3 h to enhance the generation of AβPP C-terminal fragments (CTFs) C83 and C99. These samples were used as positive controls for validation of the correct C83 and C99 bands in the western blot analyses. Cells and conditioned media were collected 48 (overexpression) or 96 h (knockdown) post transfection for analyses.

Protein extraction from cells and western blotting

Protein extraction and western blotting were performed as in [21]. Following primary antibodies were used: anti-C9orf72 (1:1000; Santa Cruz, sc-13876); anti-AβPP C-terminus (1:2000; Sigma, A8717); anti-AβPP N-terminus [1:1000; Millipore, MAB348; used to detect total amount of sAβPP fragments (sAβPPtot = sAβPPα + sAβPPβ)]; anti-β-amyloid (1:1000; BioSite, 6E10; used to specifically detect sAβPPα forms); anti-sAβPPβ (1:500, Covance; to specifically detect AβPPβ forms); and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 1:5000, Abcam, ab8245). Levels of the investigated proteins were quantified using Quantity One or Image Lab (BioRad) and normalized to those of GAPDH in the same samples. Protein levels in conditioned media were normalized to total protein lysate concentrations in each sample. Data are shown as % of the level compared to that with control siRNA or plasmid-transfected cells (set to 100%).

Soluble AβPP measurement using alkaline phosphatase assay

Soluble AβPP fragments containing the N-terminal AP tag (AP-sAβPP) and released into the conditioned media were measured as previously by alkaline phosphatase assay [20].

Aβ ELISA

Aβ_x - 40 and Aβ_x - 42 levels in conditioned media were measured using Human/Rat β Amyloid 40 (Cat no 294-64701) and Human/Rat β Amyloid 42 (Cat no 292-64501) ELISA Kits (Wako) according to kit instructions as previously [20]. Aβ concentrations were normalized to total protein lysate concentrations in each sample.

Human brain sample cohort

The human brain samples from the inferior temporal cortex used in the analyses have been previously described [23]. The samples were classified into groups according to Braak staging (0–VI) depending on the extent of neurofibrillary pathology [24]. The Ethics Committee of the Kuopio University Hospital has approved the study.

Extraction and analysis of C9orf72 RNA and protein from the frozen brain tissue samples

RNA and protein from 71 human brain tissue samples were extracted as described previously [25, 26]. Global expression of C9orf72 from 60 RNA samples from the inferior temporal cortex was investigated by using exon array probe FILL_23_P405873 (NM_145005). Assessment was implemented using Agilent One-Color Microarray-Based Exon Analysis as described in [23]. Global expression of C9orf72 was investigated from 36 protein samples from the inferior temporal cortex using SysQuant™ global proteomics (Proteome Sciences) as described in [26]. In total, 25 human brain tissue samples underwent both exon array and proteomic analysis.

Statistical analyses

Cell-based data are shown as mean % ±standard error of mean (SEM) of control plasmid- or siRNA-transfected cells. Statistical analyses were performed using GraphPad Prism5 or R. One-way analysis of variance (ANOVA), followed by Newman-Keuls or Fisher’s least significant difference post-hoc test or Mann-Whitney U test, was used to test statistical significance between sample groups. Correlation between variables was assessed using Spearman’s rank-order correlation. The level of statistical significance was defined as p < 0.05.

RESULTS

To find out if decreased levels of C9orf72 result in altered AβPP processing and Aβ generation, we first used siRNA-mediated knockdown of C9orf72 in non-neuronal HEK293-AP-AβPP cells that have been widely used to screen for novel substrates for the AD-associated secretases and factors influencing AβPP processing [20, 27]. Knockdown of C9orf72 with two different siRNA pools (C9orf72 siRNA 1 and 2) led to significantly decreased levels of the C9orf72 protein isoform a by 78.4% and 73.4%, respectively, as assessed by western blotting (Fig. 1A, C). Endogenous expression of isoform b was not detectable. We observed a moderate, but statistically not significant, increase in the levels of full-length AP-AβPP and C83 AβPP C-terminal fragment (CTF) in C9orf72 siRNA-transfected cells as compared to control cells (Fig. 1A, B). The β-secretase-cleaved C99 CTFs were not detectable. As a positive control to verify the correct sizes of the observed AβPP CTFs, a subset of cells was treated with PMA and the γ-secretase inhibitor DAPT. Under these conditions, increase in both C83 and C99 levels could be observed (Fig. 1A). Western blotting of conditioned media from the same cells showed a significant increase in the levels of soluble AP-AβPPα (AP-sAβPPα) and total soluble AP-AβPP (AP-sAβPPtot = AP-sAβPPα + AP-sAβPPβ) fragments upon C9orf72 knockdown. The levels of sAβPPβ, as assessed by an antibody specifically recognizing this soluble fragment in the media, were not detectable (Fig. 1D, E). An alkaline phosphatase-based assay [20], which detects the levels of AP-sAβPPtot, also indicated a significant increase in the conditioned media of C9orf72 siRNA-transfected cells as compared to control cells (Fig. 1F), confirming the result from Western blot analysis. There were no statistically significant alterations in the levels of endogenous sAβPPα or sAβPPtot, even though there was a small trend toward increased sAβPPtot levels (Fig. 1D, E). A significant increase in soluble Aβ₄₀ and Aβ₄₂ levels in the conditioned media of C9orf72 siRNA-transfected cells was evident compared to control cells. There were no differences in the Aβ42/Aβ40 ratio (Fig. 1G).

Fig.1

Knockdown of C9orf72 significantly increases the levels of soluble AβPP fragments and Aβ₄₀ and Aβ₄₂ and mildly the levels of full-length AβPP or C83 C-terminal fragment in non-neuronal HEK293-AP-AβPP cells. A) Western blot of total protein lysates of HEK293-AP-AβPP cells transfected with C9orf72 siRNA 1 or 2 or control siRNA showing the levels of full-length AP-AβPP (AP-AβPP), C83 AβPP C-terminal fragment, endogenous C9orf72 isoform a, or GAPDH (loading control). Treatment with 100 ng/ml PMA and 250 nM DAPT for 3 h (positive control for the CTFs on the blot) leads to increased levels of both C83 and C99 AβPP CTFs. Endogenous C9orf72 isoform b levels were not detectable. All samples were run on the same gel. Molecular weight markers are indicated on the left as kDa. An unspecific band, detected by the C9orf72 antibody and not affected by the siRNA treatment, is indicated by an asterisk (*). B) Quantification of full-length AP-AβPP and C83 levels normalized to those of GAPDH from A. C83 levels were also normalized to those of AP-AβPP. Data are shown as mean % ±SEM of the protein levels compared to those in the control cells. AP-AβPP n = 9; C83/GAPDH n = 6; C83/AP-AβPP n = 6; One-way ANOVA, Newman-Keuls, not significant. C) Quantification of C9orf72 levels normalized to those of GAPDH from A. Data are shown as mean % ±SEM of the protein levels compared to those in the control cells. n = 9 for all transfections, One-way ANOVA, Newman-Keuls, ^***p≤0.001. D) Western blot of conditioned media of HEK293-AP-AβPP cells transfected with C9orf72 siRNA 1 or 2 or control siRNA showing the levels of the endogenous soluble AβPPα (sAβPPα) and soluble AβPPtot (sAβPPtot = sAβPPα + sAβPPβ) and soluble AP-AβPPα (AP-sAβPPα) and AP-AβPPtot (AP-sAβPPtot) derived from overexpressed AP-AβPP. Soluble sAβPPβ levels remained undetectable. Molecular weight markers are indicated on the left as kDa. E) Quantification of sAβPPα, sAβPPtot, AP-sAβPPα, and AP-sAβPPtot from D. Data are shown as mean % ±SEM of the protein levels compared to those in the control cells. n = 6 for all transfections, One-way ANOVA, Newman-Keuls, ^*p≤0.05, ^**p≤0.01, ^***p≤0.001. F) Alkaline phosphatase assay-based determination of the soluble AP-sAβPPtot levels in the conditioned media of HEK293-AP-AβPP cells transfected with C9orf72 siRNA 1 or 2 or control siRNA. The assay detects both AP-sAβPPα and AP-sAβPPβ forms (=AP-sAβPPtot). Data are shown as mean % ±SEM of the protein levels compared to those in the control cells. n = 6 for all transfections, One-way ANOVA, Newman-Keuls, ^***p≤0.001. G) The levels of Aβ₄₀ or Aβ₄₂, and Aβ₄₀/Aβ₄₂ ratio in the conditioned media from HEK293-AP-AβPP cells transfected with C9orf72 siRNA 1 or 2 or control siRNA. Data are shown as mean % ±SEM of the Aβ levels compared to those in control cells. n = 6 for all transfections, One-way ANOVA, Newman-Keuls, ^**p≤0.01, ^***p≤0.001.

To study the effects of C9orf72 knockdown in neuronal cells, H4-AβPP751 cells were transfected with C9orf72 siRNA1. Similar to HEK293-AP-AβPP cells, a significant decrease in C9orf72 levels by 73.2% was observed (Fig. 2C). Opposite to HEK293-AP-AβPP cells, the levels of mature and immature forms of AβPP and C83 CTF were observed to decrease upon C9orf72 knockdown in H4-AβPP751 cells. The decrease in the levels of immature AβPP was statistically significant. C99 levels were not detectable in the H4-AβPP751 cells even after PMA and DAPT treatment (Fig. 2A, B). Also in contrast to the HEK293-AP-AβPP cells, the levels of the soluble sAβPP (sAβPPtot or sAβPPα) fragments remained unchanged in C9orf72 siRNA-transfected H4-AβPP751 cells as compared to control cells. sAβPPβ levels were undetectable in H4-AβPP751 cells (Fig. 2D, E). There were no differences in the Aβ₄₀ and Aβ₄₂ levels or Aβ₄₂/Aβ₄₀ ratio in the conditioned media between C9orf72 siRNA-transfected and control cells (Fig. 2F). Similar results related to AβPP processing upon C9orf72 knockdown were also obtained in H4 cells without AβPP overexpression (data not shown).

Fig.2

Knockdown of C9orf72 decreases the levels of mature and immature forms of full-length AβPP and C83 C-terminal fragment, but does not lead to changes in the levels of soluble sAβPP fragments or Aβ in neuronal H4-AβPP751 cells. A) Western blot of total protein lysates of H4-AβPP751 cells transfected with C9orf72 siRNA 1 or control siRNA showing the levels of mature and immature forms of AβPP, C83 AβPP C-terminal fragment, endogenous C9orf72 isoform a, or GAPDH (loading control). Treatment with 100 ng/ml PMA and 250 nM DAPT for 3 h (positive control for the CTFs on the blot) leads to increased levels of C83 AβPP CTFs. Endogenous C9orf72 isoform b levels were not detectable. All samples were run on the same gel. Molecular weight markers are indicated on the left as kDa. An unspecific band, detected by the C9orf72 antibody and not affected by the siRNA treatment, is indicated by an asterisk (*). B) Quantification of mature and immature AβPP and C83 levels normalized to those of GAPDH from A. C83 levels were also normalized to those of total AβPP (mature + immature AβPP). Data are shown as mean % ±SEM of the protein levels compared to those in the control cells. AP-AβPP n = 6; C83/GAPDH n = 6; C83/AP-AβPP n = 6; One-way ANOVA, Newman-Keuls, ^**p≤0.01. C) Quantification of C9orf72 levels normalized to those of GAPDH from A. Data are shown as mean % ±SEM of the protein levels compared to those in the control cells. n = 6 for all transfections, Mann-Whitney U, ^**p≤0.01. D) Western blot of conditioned media of H4-AβPP751 cells transfected with C9orf72 siRNA 1 or control siRNA showing the levels of the soluble AβPPα (sAβPPα) and soluble AβPPtot (sAβPPtot = sAβPPα + sAβPPβ). Soluble sAβPPβ levels remained undetectable. Molecular weight markers are indicated on the left as kDa. E) Quantification of sAβPP and sAβPPtot from D. Data are shown as mean % ±SEM of the protein levels compared to those in the control cells. n = 6 for all transfections, One-way ANOVA, Newman-Keuls, not significant. F) The levels of Aβ₄₀ or Aβ₄₂, and Aβ₄₀/Aβ₄₂ ratio in the conditioned media from H4-AβPP751 cells transfected with C9orf72 siRNA 1 or control siRNA. Data are shown as mean % ±SEM of the Aβ levels compared to those in control cells. n = 6 for all transfections, One-way ANOVA, Newman-Keuls, not significant.

To test whether overexpression of the individual C9orf72 protein isoforms affected AβPP processing, transfection of HEK293-AP-AβPP cells with C9orf72 isoform a-GFP [11] or isoform b-myc-DDK was performed. Western blot analysis of cell lysates from isoform a-GFP- or isoform b-myc-DDK-transfected cells did not indicate differences in full-length AP-AβPP or C83AβPP CTF levels between cells overexpressing the two isoforms or the control cells (Supplementary Figure 1A, B). Western blot analysis of conditioned media from the same cells did not reveal statistically significant differences in AP-sAβPPtot or AP-sAβPPα levels (Supplementary Figure 1C, D). Levels of the endogenously expressed sAβPPα or sAβPPtot did not show alterations by western blotting (Supplementary Figure 1C, D) or AP-based assay (Supplementary Figure 1E). The levels of secreted Aβ₄₀ or Aβ₄₂ or Aβ₄₀/Aβ₄₂ ratio remained unchanged in the conditioned media of cells transfected with either of the C9orf72 isoforms compared to control cells (Supplementary Figure 1F).

Finally, the levels of C9orf72 mRNA and protein in an existing human brain sample cohort from temporal cortex subcategorized according to advancing neurofibrillary pathology (Braak staging 0–VI) were examined. We have previously used this cohort to assess AD-related changes in gene expression [23, 27]. We observed a small decrease in C9orf72 mRNA levels at Braak stage IV as compared to Braak stage 0, but overall the mRNA levels of C9orf72 did not show major alterations at different Braak stages (Fig. 3A). However, the C9orf72 protein levels showed a clear increase along with increasing neurofibrillary pathology and these increases were statistically significant at Braak stages II–VI as compared to Braak stage 0 (Fig. 3B). Correlation analyses indicated a highly significant positive correlation between C9orf72 and AβPP mRNA levels (Fig. 3C). A similar mild, but not statistically significant, trend in the correlation between C9orf72 and AβPP protein levels could also be observed (Fig. 3D). The C9orf72 protein, but not mRNA, levels showed a statistically significant positive correlation with brain Aβ₄₂ levels (Fig. 3E, F).

Fig.3

The C9orf72 mRNA levels slightly decrease and protein levels show an increase in relation to increased neurofibrillary pathology in human temporal cortex. A) C9orf72 mRNA expression in the human temporal cortex samples was assessed using a FILL_23_P405873 probe, detecting global expression of all C9orf72 mRNAs. The samples were subcategorized according to the extent of neurofibrillary pathology as assessed by Braak staging. Braak 0 stage indicates no detectable neurofibrillary pathology and Braak I–VI stages indicate increasing neurofibrillary pathology (I = mildest, VI = most severe). C9orf72 mRNA levels show a slight decrease according to increasing neurofibrillary pathology and the decrease at Braak stage IV is statistically significant (^*p≤0.05) as compared to stage 0. Box plots show the median, 25th and 75th percentiles, error bars show 1.5 interquartile ranges. ANOVA, LSD, ^*p < 0.05, n = 60. B) C9orf72 protein levels were assessed by LC-MS/MS from a subset of patients, identifying a C9orf72-specific peptide (DVLMTF). C9orf72 protein levels are significantly increased at Braak stages II–VI according to increasing severity of the neurofibrillary pathology. Box plots show the median, 25th and 75th percentiles, error bars show 1.5 interquartile ranges. ANOVA, LSD, ^*p < 0.05, n = 36. C) C9orf72 mRNA levels show a significant positive correlation with AβPP mRNA levels in human brain. D) C9orf72 and AβPP protein levels show a trend towards positive correlation in human brain. E) C9orf72 mRNA levels and Aβ₄₂ levels do not correlate with each other in human brain. F) C9orf72 protein and brain Aβ₄₂ levels indicate a significant positive correlation in human brain. Spearman’s rank-order correlation coefficient, C-D; n = 55, E-F; n = 36.

DISCUSSION

The C9orf72 hexanucleotide repeat expansion is suggested to lead to FTD or ALS pathogenesis through haploinsufficiency, resulting in decreased levels of C9orf72 mRNA and protein [1, 28], and formation of nuclear RNA foci and cytoplasmic inclusions containing dipeptide repeat (DPR)-containing proteins [1 , 4–8]. Also, some patients with other neurodegenerative disorders, including AD, have been reported to carry C9orf72 repeat expansion, suggesting its potential involvement in neurodegeneration beyond the FTD/ALS spectrum [13 –15]. On the other hand, in our previous study, we found that 25% of Finnish C9orf72 repeat expansion-carrying FTD patients display decreased CSF Aβ_1 - 42 biomarker levels similarly to AD patients [19].

To assess the potential link between C9orf72 and altered AβPP processing and Aβ production, we modulated expression of C9orf72 by RNA interference-mediated knockdown or overexpression of the individual C9orf72 protein isoforms a and b in non-neuronal HEK293-AP-AβPP or neuronal H4-AβPP751 cells. The knockdown approach was used to model the suggested haploinsufficiency and the overexpression to allow investigation of the effects of individual C9orf72 protein isoforms. Overexpression of either C9orf72 protein isoform did not significantly affect AβPP processing in HEK293-AP-AβPP cells. However, a moderate increase in the levels of full-length AβPP or its C83 CTF upon C9orf72 knockdown was observed. In addition, the levels of secreted sAβPP fragments and Aβ₄₀ and Aβ₄₂ were significantly increased in HEK293-AP-AβPP cells with decreased C9orf72 expression. In contrast to these non-neuronal cells, knockdown of C9orf72 in the neuronal H4-AβPP751 cells overexpressing AβPP, or H4 cells expressing endogenous AβPP, led to decreased levels of full-length AβPP and C83 CTF, but did not influence the levels of the secreted sAβPP fragments or Aβ peptides. These results altogether indicate that alterations in C9orf72 expression may have an impact on AβPP levels and processing, but such effects may be cell type-specific in non-neuronal and neuronal cells.

C9orf72 proteins, especially the isoform a, are implicated in protein trafficking in the endosomal-lysosomal and autophagosomal pathways and suggested to function as GDP-GTP exchange factors for Rab proteins, key regulators of vesicular trafficking [9, 11]. It is well established that AβPP trafficking and subcellular localization determine whether AβPP goes through amyloidogenic or non-amyloidogenic processing. AβPP undergoes maturation by glycosylation when trafficking on the secretory pathway from ER to Golgi [29]. The mature, N- and O-glycosylated AβPP then traffics to plasma membrane and is cleaved by α-secretases. This cleavage prevents Aβ generation and yields sAβPPα potentially having neuroprotective properties [30, 31]. Alternatively, AβPP is internalized from the plasma membrane and either recycled via the early endosomal compartment or targeted to later endosomal compartments and lysosomes. Endosomes are the major sites of AβPP amyloidogenic processing by β- and γ-secretase and Aβ production [32, 33]. Our results suggest that decreased or increased levels of C9orf72 might not directly affect AβPP cleavage by the secretases, because we did not observe major differences in the levels of α-secretase- or β-secretase-cleaved AβPP CTFs. Moreover, the levels of C83 AβPP CTF normalized to those of full-length AβPP were not altered, also further inferring that the α-secretase-mediated cleavage of AβPP was not affected. The fact that C9orf72 knockdown did not result in increased levels of sAβPPβ or C99 AβPP CTF argues against an enhanced β-secretase-mediated AβPP cleavage. On the other hand, investigations in both non-neuronal and neuronal cells suggested that C9orf72 knockdown may lead to altered levels of full-length AβPP, but whether the AβPP levels increase or decrease, appears to depend on the cell type. Increased levels of both secreted sAβPP and Aβ in HEK293-AP-AβPP cells upon C9orf72 knockdown imply that decreased levels of C9orf72 may lead to alterations in the secretion of these proteins at least in these non-neuronal cells. Alternatively, these increases might result from the initially slightly increased full-length AβPP levels.

Our findings in human brain samples suggest that the expression levels of C9orf72 and AβPP positively correlate with each other. This correlation was statistically significant at the mRNA level and there was a trend toward a positive correlation also at the protein level. These data are in line with the observed decrease of AβPP levels in the neuronal H4 cells either overexpressing or endogenously expressing AβPP upon C9orf72 downregulation. Also, the levels of C9orf72 protein with the brain Aβ₄₂ levels displayed a statistically significant positive correlation. These results together point out that the hexanucleotide repeat expansion-associated haploinsufficiency leading to decreased levels of C9orf72 may have an impact on AβPP expression or the levels of Aβ in human brain.

There are several possible mechanisms, which might underlie the previously observed decrease in CSF Aβ_1 - 42 levels in repeat expansion-carrying FTD patients, including changes in Aβ secretion, clearance, synaptic pools, transport through the blood-brain barrier, or degradation by Aβ-degrading enzymes. Our cell-based findings suggesting potential alterations in the expression of AβPP or secretion of sAβPP and Aβ agree well with these patient data, even though uncovering the specific molecular mechanisms in different cell types still requires further investigations. Therefore, it would be important to investigate the levels of AβPP, its cleavage products, or Aβ in the cells or brain samples derived from the hexanucleotide repeat carriers. To our knowledge, we provide here for the first time information on the relationship between C9orf72 and AβPP levels and processing. However, further investigations are warranted in different model systems and patient-derived cells or tissue samples with decreased C9orf72 expression to reveal the underlying molecular mechanisms leading to potentially altered AβPP levels and/or processing.

Footnotes

ACKNOWLEDGMENTS

The authors would like to express their gratitude to Dr. Stefan Lichtenthaler (DZNE, Munich, Germany) for generously providing the HEK293-AP-AβPP cells. This study was supported by VTR grants 5772795 and 5772816 of Kuopio University Hospital, Sigrid Jusélius Foundation, Academy of Finland (no 307866), the Strategic Neuroscience Funding of the University of Eastern Finland, FP7; Grant Agreement no 601055, VPH Dementia Research Enabled by IT VPH-DARE@IT, EADB project in the JPND-CO-FUND program (no 301220) and Doctoral Program of Molecular Medicine (DPMM), University of Eastern Finland.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

DeJesus-Hernandez

, Mackenzie

, Boeve

, Boxer

, Baker

, Rutherford

, Nicholson

, Finch

, Flynn

, Adamson

, Kouri

, Wojtas

, Sengdy

, Hsiung

, Karydas

, Seeley

, Josephs

, Coppola

, Geschwind

, Wszolek

, Feldman

, Knopman

, Petersen

, Miller

, Dickson

, Boylan

, Graff-Radford

, Rademakers

(2011) Expanded GGGGCC hexanucleotide repeat in noncoding region of C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron 72, 245–256.

Renton

, Majounie

, Waite

, Simon-Sanchez

, Rollinson

, Gibbs

, Schymick

, Laaksovirta

, van Swieten

, Myllykangas

, Kalimo

, Paetau

, Abramzon

, Remes

, Kaganovich

, Scholz

, Duckworth

, Ding

, Harmer

, Hernandez

, Johnson

, Mok

, Ryten

, Trabzuni

, Guerreiro

, Orrell

, Neal

, Murray

, Pearson

, Jansen

, Sondervan

, Seelaar

, Blake

, Young

, Halliwell

, Callister

, Toulson

, Richardson

, Gerhard

, Snowden

, Mann

, Neary

, Nalls

, Peuralinna

, Jansson

, Isoviita

, Kaivorinne

, Holtta-Vuori

, Ikonen

, Sulkava

, Benatar

, Wuu

, Chio

, Restagno

, Borghero

, Sabatelli

, Consortium ITALSGEN, Heckerman

, Rogaeva

, Zinman

, Rothstein

, Sendtner

, Drepper

, Eichler

, Alkan

, Abdullaev

, Pack

, Dutra

, Pak

, Hardy

, Singleton

, Williams

, Heutink

, Pickering-Brown

, Morris

, Tienari

, Traynor

(2011) A hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21-linked ALS-FTD. Neuron 72, 257–268.

Majounie

, Renton

, Mok

, Dopper

, Waite

, Rollinson

, Chio

, Restagno

, Nicolaou

, Simon-Sanchez

, van Swieten

, Abramzon

, Johnson

, Sendtner

, Pamphlett

, Orrell

, Mead

, Sidle

, Houlden

, Rohrer

, Morrison

, Pall

, Talbot

, Ansorge

, Chromosome 9-ALS/FTD Consortium, French research network on FTLD/FTLD/ALS ITALSGEN Consortium, Hernandez

, Arepalli

, Sabatelli

, Mora

, Corbo

, Giannini

, Calvo

, Englund

, Borghero

, Floris

, Remes

, Laaksovirta

, McCluskey

, Trojanowski

, Van Deerlin

, Schellenberg

, Nalls

, Drory

, Lu

, Yeh

, Ishiura

, Takahashi

, Tsuji

, Le Ber

, Brice

, Drepper

, Williams

, Kirby

, Shaw

, Hardy

, Tienari

, Heutink

, Morris

, Pickering-Brown

, Traynor

(2012) Frequency of the C9orf72 hexanucleotide repeat expansion in patients with amyotrophic lateral sclerosis and frontotemporal dementia: A cross-sectional study. Lancet Neurol 11, 323–330.

Gendron

, Bieniek

, Zhang

, Jansen-West

, Ash

, Caulfield

, Daughrity

, Dunmore

, Castanedes-Casey

, Chew

, Cosio

, van Blitterswijk

, Lee

, Rademakers

, Boylan

, Dickson

, Petrucelli

(2013) Antisense transcripts of the expanded C9ORF72 hexanucleotide repeat form nuclear RNA foci and undergo repeat-associated non-ATG translation in c9FTD/ALS. Acta Neuropathol 126, 829–844.

Mori

, Weng

, Arzberger

, May

, Rentzsch

, Kremmer

, Schmid

, Kretzschmar

, Cruts

, Van Broeckhoven

, Haass

, Edbauer

(2013) The C9orf72 GGGGCC repeat is translated into aggregating dipeptide-repeat proteins in FTLD/ALS. Science 339, 1335–1338.

, Poidevin

, Li

, Shu

, Nelson

, Li

, Hales

, Gearing

, Wingo

, Jin

(2013) Expanded GGGGCC repeat RNA associated with amyotrophic lateral sclerosis and frontotemporal dementia causes neurodegeneration. Proc Natl Acad Sci U S A 110, 7778–7783.

Donnelly

, Zhang

, Pham

, Haeusler

, Mistry

, Vidensky

, Daley

, Poth

, Hoover

, Fines

, Maragakis

, Tienari

, Petrucelli

, Traynor

, Wang

, Rigo

, Bennett

, Blackshaw

, Sattler

, Rothstein

(2013) RNA toxicity from the ALS/FTD C9ORF72 expansion is mitigated by antisense intervention. Neuron 80, 415–428.

Chew

, Gendron

, Prudencio

, Sasaguri

, Zhang

, Castanedes-Casey

, Lee

, Jansen-West

, Kurti

, Murray

, Bieniek

, Bauer

, Whitelaw

, Rousseau

, Stankowski

, Stetler

, Daughrity

, Perkerson

, Desaro

, Johnston

, Overstreet

, Edbauer

, Rademakers

, Boylan

, Dickson

, Fryer

, Petrucelli

(2015) Neurodegeneration. C9ORF72 repeat expansions in mice cause TDP-43 pathology, neuronal loss, and behavioral deficits. Science 348, 1151–1154.

Levine

, Daniels

, Gatta

, Wong

, Hayes

(2013) The product of C9orf72, a gene strongly implicated in neurodegeneration, is structurally related to DENN Rab-GEFs. Bioinformatics 29, 499–503.

10.

Zhang

, Iyer

, He

, Aravind

(2012) Discovery of novel DENN proteins: Implications for the evolution of eukaryotic intracellular membrane structures and human disease. Front Genet 3, 283.

11.

Farg

, Sundaramoorthy

, Sultana

, Yang

, Atkinson

, Levina

, Halloran

, Gleeson

, Blair

, Soo

, King

, Atkin

(2014) C9ORF72, implicated in amytrophic lateral sclerosis and frontotemporal dementia, regulates endosomal trafficking. Hum Mol Genet 23, 3579–3595.

12.

Xiao

, MacNair

, McGoldrick

, McKeever

, McLean

, Zhang

, Keith

, Zinman

, Rogaeva

, Robertson

(2015) Isoform-specific antibodies reveal distinct subcellular localizations of C9orf72 in amyotrophic lateral sclerosis. Ann Neurol 78, 568–583.

13.

Beck

, Poulter

, Hensman

, Rohrer

, Mahoney

, Adamson

, Campbell

, Uphill

, Borg

, Fratta

, Orrell

, Malaspina

, Rowe

, Brown

, Hodges

, Sidle

, Polke

, Houlden

, Schott

, Fox

, Rossor

, Tabrizi

, Isaacs

, Hardy

, Warren

, Collinge

, Mead

(2013) Large C9orf72 hexanucleotide repeat expansions are seen in multiple neurodegenerative syndromes and are more frequent than expected in the UK population. Am J Hum Genet 92, 345–353.

14.

Harms

, Benitez

, Cairns

, Cooper

, Mayo

, Carrell

, Faber

, Williamson

, Bird

, Diaz-Arrastia

, Foroud

, Boeve

, Graff-Radford

, Mayeux

, Chakraverty

, Goate

, Cruchaga

, NIA-LOAD/NCRAD Family Study Consortium (2013) C9orf72 hexanucleotide repeat expansions in clinical Alzheimer disease. JAMA Neurol 70, 736–741.

15.

Davidson

, Robinson

, Snowden

, Mann

(2013) Pathological assessments for the presence of hexanucleotide repeat expansions in C9ORF72 in Alzheimer’s disease. Acta Neuropathol Commun 1, 5059601–50.

16.

Ballard

, Gauthier

, Corbett

, Brayne

, Aarsland

, Jones

(2011) Alzheimer’s disease. Lancet 377, 1019–1031.

17.

Herukka

, Hallikainen

, Soininen

, Pirttila

(2005) CSF Abeta42 and tau or phosphorylated tau and prediction of progressive mild cognitive impairment. Neurology 64, 1294–1297.

18.

Tapiola

, Alafuzoff

, Herukka

, Parkkinen

, Hartikainen

, Soininen

, Pirttila

(2009) Cerebrospinal fluid beta-amyloid 42 and tau proteins as biomarkers of Alzheimer-type pathologic changes in the brain. Arch Neurol 66, 382–389.

19.

Kamalainen

, Herukka

, Hartikainen

, Helisalmi

, Moilanen

, Knuuttila

, Jansson

, Tienari

, Remes

(2015) Cerebrospinal fluid biomarkers for Alzheimer’s disease in patients with frontotemporal lobar degeneration and amyotrophic lateral sclerosis with the C9ORF72 repeat expansion. Dement Geriatr Cogn Disord 39, 287–293.

20.

Lichtenthaler

, Dominguez

, Westmeyer

, Reiss

, Haass

, Saftig

, De Strooper

, Seed

(2003) The cell adhesion protein P-selectin glycoprotein ligand-1 is a substrate for the aspartyl protease BACE1. J Biol Chem 278, 48713–48719.

21.

Viswanathan

, Haapasalo

, Bottcher

, Miettinen

, Kurkinen

, Lu

, Thomas

, Maynard

, Romano

, Hyman

, Berezovska

, Bertram

, Soininen

, Dantuma

, Tanzi

, Hiltunen

(2011) Alzheimer’s disease-associated ubiquilin-1 regulates presenilin-1 accumulation and aggresome formation. Traffic 12, 330–348.

22.

Natunen

, Takalo

, Kemppainen

, Leskela

, Marttinen

, Kurkinen

, Pursiheimo

, Sarajarvi

, Viswanathan

, Gabbouj

, Solje

, Tahvanainen

, Pirttimaki

, Kurki

, Paananen

, Rauramaa

, Miettinen

, Makinen

, Leinonen

, Soininen

, Airenne

, Tanzi

, Tanila

, Haapasalo

, Hiltunen

(2016) Relationship between ubiquilin-1 and BACE1 in human Alzheimer’s disease and APdE9 transgenic mouse brain and cell-based models. Neurobiol Dis 85, 187–205.

23.

Martiskainen

, Viswanathan

, Nykanen

, Kurki

, Helisalmi

, Natunen

, Sarajarvi

, Kurkinen

, Pursiheimo

, Rauramaa

, Alafuzoff

, Jaaskelainen

, Leinonen

, Soininen

, Haapasalo

, Huttunen

, Hiltunen

(2015) Transcriptomics and mechanistic elucidation of Alzheimer’s disease risk genes in the brain and in vitro models. Neurobiol Aging 36, 1221.e15–1221.e28.

24.

Braak

, Alafuzoff

, Arzberger

, Kretzschmar

, Del Tredici

(2006) Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry. Acta Neuropathol 112, 389–404.

25.

Natunen

, Parrado

, Helisalmi

, Pursiheimo

, Sarajarvi

, Makinen

, Kurkinen

, Mullin

, Alafuzoff

, Haapasalo

, Bertram

, Soininen

, Tanzi

, Hiltunen

(2013) Elucidation of the BACE1 regulating factor GGA3 in Alzheimer’s disease. J Alzheimers Dis 37, 217–232.

26.

Britton

, Zen

, Quaglia

, Selzer

, Mitra

, Lobetaner

, Jung

, Bohm

, Schmid

, Prefot

, Hoehle

, Koncarevic

, Gee

, Nicholson

, Ward

, Castellano

, Stebbing

, Zucht

, Sarker

, Heaton

, Pike

(2014) Quantification of pancreatic cancer proteome and phosphorylome: Indicates molecular events likely contributing to cancer and activity of drug targets. PLoS One 9, e90948.

27.

Kurkinen

, Marttinen

, Turner

, Natunen

, Makinen

, Haapalinna

, Sarajarvi

, Gabbouj

, Kurki

, Paananen

, Koivisto

, Rauramaa

, Leinonen

, Tanila

, Soininen

, Lucas

, Haapasalo

, Hiltunen

(2016) SEPT8 modulates beta-amyloidogenic processing of APP by affecting the sorting and accumulation of BACE1. J Cell Sci 129, 2224–2238.

28.

Waite

, Baumer

, East

, Neal

, Morris

, Ansorge

, Blake

(2014) Reduced C9orf72 protein levels in frontal cortex of amyotrophic lateral sclerosis and frontotemporal degeneration brain with the C9ORF72 hexanucleotide repeat expansion. Neurobiol Aging 35, 1779.e5–1779.e13.

29.

Thinakaran

, Koo

(2008) Amyloid precursor protein trafficking, processing, and function. J Biol Chem 283, 29615–29619.

30.

Hiltunen

, van Groen

, Jolkkonen

(2009) Functional roles of amyloid-beta protein precursor and amyloid-beta peptides: Evidence from experimental studies. J Alzheimers Dis 18, 401–412.

31.

Fol

, Braudeau

, Ludewig

, Abel

, Weyer

, Roederer

, Brod

, Audrain

, Bemelmans

, Buchholz

, Korte

, Cartier

, Muller

(2016) Viral gene transfer of APPsalpha rescues synaptic failure in an Alzheimer’s disease mouse model. Acta Neuropathol 131, 247–266.

32.

Jiang

, Li

, Zhang

, Bu

, Xu

, Zhang

(2014) Trafficking regulation of proteins in Alzheimer’s disease. Mol Neurodegener 9, 613269–6.

33.

Vassar

, Bennett

, Babu-Khan

, Kahn

, Mendiaz

, Denis

, Teplow

, Ross

, Amarante

, Loeloff

, Luo

, Fisher

, Fuller

, Edenson

, Lile

, Jarosinski

, Biere

, Curran

, Burgess

, Louis

, Collins

, Treanor

, Rogers

, Citron

(1999) Beta-secretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE. Science 286, 735–741.