Differential Effects of Yeast NADH Dehydrogenase (Ndi1) Expression on Mitochondrial Function and Inclusion Formation in a Cell Culture Model of Sporadic Parkinson's Disease

Abstract

Parkinson's disease (PD) is a neurodegenerative disorder that exhibits aberrant protein aggregation and mitochondrial dysfunction. Ndi1, the yeast mitochondrial NADH dehydrogenase (complex I) enzyme, is a single subunit, internal matrix-facing protein. Previous studies have shown that Ndi1 expression leads to improved mitochondrial function in models of complex I-mediated mitochondrial dysfunction. The trans-mitochondrial cybrid cell model of PD was created by fusing mitochondrial DNA-depleted SH-SY5Y cells with platelets from a sporadic PD patient. PD cybrid cells reproduce the mitochondrial dysfunction observed in a patient's brain and periphery and form intracellular, cybrid Lewy bodies comparable to Lewy bodies in PD brain. To improve mitochondrial function and alter the formation of protein aggregates, Ndi1 was expressed in PD cybrid cells and parent SH-SY5Y cells. We observed a dramatic increase in mitochondrial respiration, increased mitochondrial gene expression, and increased PGC-1α gene expression in PD cybrid cells expressing Ndi1. Total cellular aggregated protein content was decreased but Ndi1 expression was insufficient to prevent cybrid Lewy body formation. Ndi1 expression leads to improved mitochondrial function and biogenesis signaling, both processes that could improve neuron survival during disease. However, other aspects of PD pathology such as cybrid Lewy body formation were not reduced. Consequently, resolution of mitochondrial dysfunction alone may not be sufficient to overcome other aspects of PD-related cellular pathology.

Keywords: Lewy body 2; Ndi1 4; alpha-synuclein 6; mitochondrial dysfunction 3; oxygen utilization 5; trans-mitochondrial cybrid 1.

Figure 1

Expression of Ndi1 was greater in PD61^Ndi1 than in SH-SY5Y^Ndi1 based on immunofluorescence, Western blot analysis and qRT-PCR. (A,B) Mitochondria in PD61^Ndi1 and SH-SY5Y^Ndi1 cells stained with Ndi1 (green), complex Vα (red) and nuclei labeled with DAPI (blue). Box outline designates an enlarged area of merged image (high mag). Scale bars = 10 μm; high mag scale bar = 2 μm. (C) Western blot visualization of Ndi1 protein levels in SH-SY5Y^Ndi1 and PD61^Ndi1. Values shown are integrated intensities normalized to actin. (D) Quantitative real-time (qRT-PCR) analysis of NDI1 gene expression levels in PD61^Ndi1 and SH-SY5Y^Ndi1 and shown as fold change from SH-SY5Y^Ndi1 levels. There was an approximately three-fold increase in PD61^Ndi1. Students t-test, n = 3, # p < 0.005.

Figure 2

PD61^Ndi1 exhibited improved mitochondrial respiration. (A) Representative traces of oxygen consumption rates (OCR) from PD61 and PD61^Ndi1. Indicated inhibitors were added sequentially. (B) ECAR (glycolysis) rates in PD61 and PD61^Ndi1. (C) Normalized basal, ATP-linked respiration, max capacity, spare capacity, and leak rates for PD61 and PD61^Ndi1. (D,E) Normalized basal respiration rates from rotenone (−) and rotenone (+)/Pre-RTN experiments using PD61 (D) and PD61^Ndi1 (E). Rotenone (−) experiments were performed using normal running media. Rotenone (+)/Pre-RTN experiments (D,E) were performed with a 30 nM rotenone pretreatment to inhibit endogenous complex I. Two-way ANOVA with Bonferroni post-hoc analysis, n = 10, * p < 0.01, # p < 0.001.

Figure 3

Ndi1 expression in SH-SY5Y^Ndi1 improved mitochondrial respiration. (A) Representative traces of oxygen consumption rates (OCR) from SH-SY5Y and Sh-SY5Y^Ndi1. Indicated inhibitors were added sequentially. (B) ECAR (glycolysis) rates in SH-SY5Y and SH-SY5Y^Ndi1. (C) Normalized basal, ATP-linked respiration, max capacity, spare capacity and leak rates for PD61 and PD61^Ndi1. (D,E) Normalized basal respiration rates from rotenone (−) and rotenone (+)/Pre-RTN experiments using SH-SY5Y (D) and SH-SY5Y^Ndi1 (E). Rotenone (−) experiments were performed using normal running media. Rotenone (+)/Pre-RTN experiments (D,E) were performed with a 30 nM rotenone pretreatment to inhibit endogenous complex I. Two-way ANOVA with Bonferroni post-hoc analysis, n = 10, # p < 0.001.

Figure 4

Mitochondrial movement by axonal transport is altered in PD^Ndi1. (A) Histogram of average mitochondrial velocity (PD61- 8–50 mitochondria/process in 39 processes from four different cultures; PD61^Ndi1 9–44 mitochondria/process in 36 processes from six different cultures). (B) Histogram of average mitochondrial velocity (SH-SY5Y 8–46 mitochondria/process in 59 different processes in four different cultures; SH-SY5Y^Ndi1 5–34 mitochondria/processes in 59 different processes from five different cultures).

Figure 5

Mitochondrial gene expression and copy number after Ndi1 expression. qRT-PCR analysis of mtDNA-encoded gene expression is shown as fold change from SH-SY5Y levels (dashed line at 1.0). (A) Gene expression was significantly increased for ND2, ND4, and COX3 in PD61^Ndi1 compared to PD61. (B) mtDNA-encoded gene expression was not significantly changed in SH-SY5Y^Ndi1 compared to SH-SY5Y. qRT-PCR analysis of mtDNA-encoded gene copy number shown as fold change from SH- SY5Y levels (dashed line at 1.0). (C) Gene copy number was elevated in all four genes measured (ND2, ND4, COX3, and 12s) in PD61^Ndi1 compared to PD61. (D) There was no change in mtDNA-encoded copy number in SH-SY5Y^Ndi1 relative to SH-SY5Y. Student’s t-test, n = 3, * p < 0.05, ** p < 0.01, # p < 0.005.

Figure 6

Mitochondrial biogenesis and electron transport chain assembly after Ndi1 expression. (A) qRT-PCR analysis shows that PGC-1α is increased in PD61^Ndi1 relative to PD61 and in SH-SY5Y^Ndi1 versus SH-SY5Y. Dashed line represents SH-SY5Y mean. Student’s t-test, n = 3, * p < 0.05. (B) qRT-PCR analysis shows that porin is not significantly different in PD61, PD61^Ndi1, SH-SY5Y or SH-SY5Y^Ndi1. (C,D) Graphs showing normalized integrated intensities of ETC complex assembly for all five ETC proteins, normalized to porin (B). (C) ETC complex assembly was increased in complexes III and V in PD61^Ndi1 compared to PD61. (D) ETC assembly was increased in complexes I-IV for SH-SY5Y^Ndi1 compared to SH-SY5Y. Student’s t-test, n = 3 for SY5Yand SY5Y^Ndi1, n = 4 for PD61 and PD61^Ndi1, * p < 0.05.

Figure 7

Aggregated protein and αSYN expression gene expression. (A) Congo red pixels were counted and normalized to cell number as a measure of cellular aggregated protein content. Both SH-SY5Y^Ndi1 and PD61^Ndi1 showed a decrease in total aggregated protein content after Ndi1 expression. Student’s t-test, n = 3. * p < 0.05, ** p < 0.01 (B) qRT-PCR for αSYN showed no significant change in αSYN expression after Ndi1 expression in SH-SY5Y^Ndi1. PD61^Ndi1 shows a trend towards increased αSYN gene expression (p = 0.055). Student’s t-test, n = 3.

Figure 8

Cybrid Lewy body expression after NDI1 gene expression. (A,B) Representative CLB from PD61 and PD61^Ndi1 showing colocalization of αSYN with Lamp2A (lysosome-associated membrane protein). Scale bars = 10 μm (C,D) Representative images from PD61^Ndi1 showing CLB present in cells expressing Ndi1, Lamp2A and porin. Scale bars = 5 μm.