Abstract

Recent studies provide evidence that wild-type Cu/Zn-superoxide dismutase (SOD1(hWT)) might be an important factor in mutant SOD1-mediated amyotrophic lateral sclerosis (ALS). In order to investigate its functional role in the pathogenesis of ALS, we designed fusion proteins of two SOD1 monomers linked by a polypeptide. We demonstrated that wild-type-like mutants, but not SOD1(G85R) homodimers, as well as mutant heterodimers including SOD1(G85R)-SOD1(hWT) display dismutase activity. Mutant homodimers showed an increased aggregation compared with the corresponding heterodimers in cell cultures and transgenic Caenorhabditis elegans, although SOD1(G85R) heterodimers are more toxic in functional assays. Our data show that (i) toxicity of mutant SOD1 is not correlated to its aggregation potential; (ii) dismutase-inactive mutants form dismutase-active heterodimers with SOD1(hWT); (iii) SOD1(hWT) can be converted to contribute to disease by forming active heterodimers. Therefore, we conclude that toxicity of mutant SOD1 is at least partially mediated through heterodimer formation with SOD1(hWT) in vivo and does not correlate with the aggregation potential of individual mutants.

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is the most prominent adult-onset motor neuron disease. Sporadic and familial forms which are partly caused by mutations in the Cu/Zn superoxide dismutase (sod1) gene are clinically and pathologically indistinguishable. Therefore, cellular and animal systems expressing mutant SOD1 are well-acknowledged models of the human disease, even though toxic properties and the site of action are still not understood (reviewed in 1). Recent studies documented that non-neuronal cells such as microglia next to motor neurons greatly affect their degeneration (2,3).

SOD1 is a primarily cytosolic enzyme of the cellular oxidative defense and acts as a protein homodimer with each monomer containing one complexed copper and zinc ion (4). About 120 point mutations in the sod1 gene are known to cause disease (http://alsod.iop.kcl.ac.uk/Als/index.aspx). Mutant proteins can be divided into at least two classes according to their biophysical characteristics representing wild-type-like proteins, which bind metal ions and show dismutase activity (e.g. SOD1(G37R)), and impaired metal-binding mutants, which have a largely reduced activity (e.g. SOD1(G85R)) (reviewed in 5). On the basis of the observations that wild-type-like and dismutase-inactive mutants cause the disease alike and that the deletion of endogenous SOD1 does not induce ALS in mice (6), it is hypothesized that mutant SOD1 causes ALS through a gain of one or more so far unidentified toxic properties.

One of the proposed toxic functions that have been described for mutant SOD1 is an aberrant enzymatic activity, such as peroxidase or superoxide-reducing activities (7–9) and peroxinitrite catalysis (10). Although discussed recently (11,12), the presence of oxidative stress markers in patient tissues and animal models (reviewed in 13) supports the oxidative hypothesis.

A second hypothesis, which is not mutually exclusive, is based on the high tendency of mutant SOD1 to form aggregates in vitro (14). They are a pathological hallmark of both ALS forms and have been found in the cytosol as well as in mitochondria in transgenic SOD1 mice (15,16). The aggregation process was proposed to depend on the presence of monomeric SOD1 (17,18) which might be a consequence of a structural instability of functional dimers (19,20).

The role of SOD1(hWT) for the pathogenesis of ALS is still elusive. At least in humans, the vast majority of SOD1 mutations are dominantly inherited, resulting in the presence of wild-type and mutant SOD1 subunits. In fact, there is compelling evidence from three independent groups that the co-expression of human mutant and SOD1(hWT) in mice accelerated disease (21–23). In addition, unaffected SOD1(A4V) mutant mice developed ALS-like disease only when mated with SOD1(hWT) overexpressing mice (21). These results could be confirmed in a novel mutant SOD1-zebrafish model in which the co-expression of SOD1(hWT) aggravates the motor neuron-specific phenotype (24). In contrast, expression of the mutant SOD1(G85R) in a mouse SOD1-deleted background did not ameliorate disease (25).

In order to address the question of how SOD1(hWT) contributes to the disease, we designed fusion proteins consisting of two SOD1 molecules (wild-type and/or mutant forms) that are connected through a protein linker, thereby promoting the dimerization between the two monomers. We studied the aggregation potential of mutant homo- and heterodimers after transient transfections and in neuronal cell lines stably expressing SOD1(G37R)- and SOD1(G85R)-bearing dimers. Furthermore, we analyzed the toxicity of these fusion proteins in survival assays and generated a novel transgenic Caenorhabditis elegans model. On the basis of our results, we conclude that protein aggregation cannot account for the primary toxicity in SOD1-mediated ALS. Inactive mutant SOD1 might acquire toxicity by heterodimerization with SOD1(hWT), thus converting SOD1(hWT) to contribute to pathogenesis.

RESULTS

Co-expression of SOD1(hWT) and mutant SOD1 results in the formation of dismutase-active heterodimers

To investigate the potential of mutant SOD1 to heterodimerize with SOD1(hWT), we transiently transfected human HEK293 cells with monomeric EGFP-tagged human SOD1(A4V), SOD1(G37R), SOD1(G85R), SOD1(G93C) and SOD1(hWT). The presence of EGFP enabled us to monitor the formation of homo- and heterodimers between endogenously and ectopically expressed SOD1 as three distinct activity spots (Fig. 1A). The formation of dismutase-active heterodimers between wild-type mouse or SOD1(hWT) with some mutant SOD1 variants has been described earlier (19,26). We could confirm that the wild-type-like mutants SOD1(G37R) and SOD1(G93C) form dismutase-active homodimers and heterodimers with SOD1(hWT). We could also detect active heterodimers between SOD1(hWT) and the SOD1(A4V) mutant, whereas SOD1(A4V) homodimers were likely not to be formed owing to mutations at the dimer interface. Cu2+-free SOD1(G85R) showed no activity where we expected mutant homodimers, but displayed some dismutase activity when forming heterodimers with SOD1(hWT).

Figure 1.

Heterodimer formation of mutant and SOD1(hWT) after transient transfection of EGFP-tagged SOD1 in HEK293 cells. (A) Human HEK293 cells were transiently transfected with monomeric human SOD1-EGFP fusion constructs as indicated. Extracts were separated by native gel electrophoresis on a 15% acryamide gel and assayed for superoxide dismutase activity. All constructs formed dismutase-active heterodimers with endogenous, untagged SOD1(hWT). In contrast to SOD1(G85R) and SOD1(A4V), SOD1(hWT), SOD1(G37R) and SOD1(G93C) formed homodimers with detectable dismutase activity. (B) Total cell extracts were assayed for SOD1-EGFP fusion protein levels by standard SDS–PAGE on 12% gels and an antibody detecting primarily human SOD1 (SOD100). Note that SOD1(G85R) EGFP ran faster than the other fusion proteins as previously observed. (C) Equal loading of the gel was documented by detecting endogenous SOD1(hWT) on the same membrane.

Figure 1.

Heterodimer formation of mutant and SOD1(hWT) after transient transfection of EGFP-tagged SOD1 in HEK293 cells. (A) Human HEK293 cells were transiently transfected with monomeric human SOD1-EGFP fusion constructs as indicated. Extracts were separated by native gel electrophoresis on a 15% acryamide gel and assayed for superoxide dismutase activity. All constructs formed dismutase-active heterodimers with endogenous, untagged SOD1(hWT). In contrast to SOD1(G85R) and SOD1(A4V), SOD1(hWT), SOD1(G37R) and SOD1(G93C) formed homodimers with detectable dismutase activity. (B) Total cell extracts were assayed for SOD1-EGFP fusion protein levels by standard SDS–PAGE on 12% gels and an antibody detecting primarily human SOD1 (SOD100). Note that SOD1(G85R) EGFP ran faster than the other fusion proteins as previously observed. (C) Equal loading of the gel was documented by detecting endogenous SOD1(hWT) on the same membrane.

Mutant homo- and heterodimeric fusion proteins are expressed and form functional enzymes

In order to analyze mutant SOD1 homo- and heterodimers separately, we generated C-terminally EGFP-tagged fusion proteins consisting of two SOD1 monomers connected by a protein linker with SOD1(A4V), SOD1(G37R), SOD1(G85R), SOD1(G93C) and SOD1(hWT) (Fig. 2A and B). The protein linker that consists mainly of uncharged glycine and serine residues (Supplementary Material, Table S1) was previously used to study transcription factor activities (27). The fusion proteins were expressed after transient transfection in HEK293 cells and migrated according to their predicted molecular weights at ∼70 kDa on denaturing SDS–PAGE gels (Fig. 2C, top). It is consistent with previous results that dimers containing SOD1(G85R) run slightly faster compared with the other dimer proteins. The detection of endogenous monomeric SOD1 serves as a loading control (Fig. 2C, middle). The same samples were analyzed for their dismutase activity in native gels. All dimer proteins were functionally active except the SOD1(G85R) homodimer (Fig. 2C, bottom) which had no activity as previously described (28). Consistent with our earlier results (Fig. 1), SOD1(G85R)-SOD1(hWT) heterodimers had a substantial dismutase activity. These data indicate that the linker as well as the EGFP-tag did not interfere with the proper folding and metal loading of enzymatically active SOD1 proteins.

Figure 2.

Homo- and heterodimeric SOD1 fusion proteins resemble functional properties as if SOD1 is expressed as monomer. (A) Schematic representation of the dimer constructs. Two SOD1 monomer cDNAs were connected by a linker sequence. In addition, the dimers were tagged with EGFP. (B) The figure represents the three dimensional structure of the fusion dimer without EGFP. The crystal structure of the wild-type dimer is shown with the linker peptide as modeled using Modeller software. The linker peptide (blue) is long enough to connect the C-terminus of the first SOD1 monomer (red) with the N-terminus of the second monomer (purple) without interfering with the formation of a functional enzyme. Sites of point mutations are labeled. (C) Dismutase-active dimers were expressed in HEK293 cells after transient transfection. Total extracts of HEK293 cells transfected with the dimer constructs indicated were separated on 12% standard SDS–PAGE gels and analyzed by Western blot using the SOD100 antibody (top). A high-molecular-weight band is visible most prominently in extracts from cells transfected with mutant homodimers. The endogenous SOD1(hWT) serves as a loading control (middle). All dimer constructs, but not the SOD1(G85R) homodimer, show dismutase activity when separated on native 15% acrylamide gels (bottom). The lower activity represents homodimers formed by endogenous SOD1.

Figure 2.

Homo- and heterodimeric SOD1 fusion proteins resemble functional properties as if SOD1 is expressed as monomer. (A) Schematic representation of the dimer constructs. Two SOD1 monomer cDNAs were connected by a linker sequence. In addition, the dimers were tagged with EGFP. (B) The figure represents the three dimensional structure of the fusion dimer without EGFP. The crystal structure of the wild-type dimer is shown with the linker peptide as modeled using Modeller software. The linker peptide (blue) is long enough to connect the C-terminus of the first SOD1 monomer (red) with the N-terminus of the second monomer (purple) without interfering with the formation of a functional enzyme. Sites of point mutations are labeled. (C) Dismutase-active dimers were expressed in HEK293 cells after transient transfection. Total extracts of HEK293 cells transfected with the dimer constructs indicated were separated on 12% standard SDS–PAGE gels and analyzed by Western blot using the SOD100 antibody (top). A high-molecular-weight band is visible most prominently in extracts from cells transfected with mutant homodimers. The endogenous SOD1(hWT) serves as a loading control (middle). All dimer constructs, but not the SOD1(G85R) homodimer, show dismutase activity when separated on native 15% acrylamide gels (bottom). The lower activity represents homodimers formed by endogenous SOD1.

Mutant heterodimers have a reduced aggregation potential

Previous reports showed that the transient expression of mutant SOD1 proteins in different cell types resulted in the formation of protein aggregates (29,30). In line with these experiments, the transient expression of mutant homodimers consisting of SOD1(G37R), SOD1(G93C), SOD1(G85R) and SOD1(A4V) induced protein aggregation in HEK293 cells (Fig. 3C, E, G and I, respectively). In contrast, transfection with vector-only (MockEGFP) and SOD1(hWT) homodimers showed a homogenous distribution of EGFP throughout the cells and the cytosol, respectively (Fig. 3A and B). We did not observe visible protein aggregates in cells expressing hetereodimers of wild-type-like mutants SOD1(G37R) and SOD1(G93C) (Fig. 3D and F) and only some in cells expressing heterodimers containing SOD1(G85R) and SOD1(A4V) (Fig. 3H and J), indicating that dimer formation with SOD1(hWT) reduced the formation of aggregates. In order to quantify the formation of aggregates, total PBS extracts of transiently transfected cells were subjected to a centrifugation assay (Fig. 3K, top). The quality of the supernatant and pellet fractions was monitored by endogenous SOD1 and histon H3, respectively (Fig. 3K, middle and bottom). All mutant SOD1-containing dimers, but not wild-type like containing SOD1(G37R) and SOD1(G93C) heterodimers, significantly shifted towards the pellet fraction compared with SOD1(hWT) homodimers (Fig. 3L). Furthermore, all mutant SOD1 homodimers had a statistically significant higher aggregation potential than their heterodimeric counterparts, which was also represented by the higher molecular weight bands present in the pellet fractions (Figs. 2C and 3K). These results clearly document that mutant SOD1 homodimers and those consisting of mutant and SOD1(hWT) display different aggregation properties.

Figure 3.

Mutant homodimers have a higher aggregation potential than mutant heterodimers. (A)–(J) SOD dimer constructs were transiently transfected into HEK293 cells. Forty-eight hours after transfection, cells were fixed and EGFP fluorescence pictures were taken. Aggregates, characterized by a granular appearance, were prominently seen with all mutant homodimers (arrows). When aggregates appear, the remaining EGFP-fluorescence in the cytosol is low. SOD1(hWT) homodimers (A), SOD1(G37R) (D) and SOD1(G93C) (F) heterodimers were evenly distributed throughout the cytosol, whereas SOD1(G85R) (H) and SOD1(A4V) (J) heterodimers additionally developed some visible aggregates, but less than their homodimeric counterparts (G, I, respectively). (K) Total PBS extracts of transiently transfected cells were separated into supernatant and pellet fractions by centrifugation. Fifteen micrograms of protein were separated on 12% standard SDS–PAGE gels and analyzed by Western blot using SOD100 antibodies to monitor the distribution of fusion proteins (top), endogenous SOD1(hWT) (cytosolic fraction; middle) and histon H3 (pellet fraction; bottom). (L) Results of the densitometric analysis of the fusion protein bands are expressed as the ratio between pellet and soluble fractions. Bars represent mean values ± SD of at least three independent experiments. All mutant-containing dimers have a statistically significant higher aggregation potential than SOD1(hWT) homodimers (‘*’ on bars), and all mutant homodimers are more prone to aggregate than their heterodimeric counterparts (‘*’ on brackets). One-way ANOVA followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001; n ≥ 3.

Figure 3.

Mutant homodimers have a higher aggregation potential than mutant heterodimers. (A)–(J) SOD dimer constructs were transiently transfected into HEK293 cells. Forty-eight hours after transfection, cells were fixed and EGFP fluorescence pictures were taken. Aggregates, characterized by a granular appearance, were prominently seen with all mutant homodimers (arrows). When aggregates appear, the remaining EGFP-fluorescence in the cytosol is low. SOD1(hWT) homodimers (A), SOD1(G37R) (D) and SOD1(G93C) (F) heterodimers were evenly distributed throughout the cytosol, whereas SOD1(G85R) (H) and SOD1(A4V) (J) heterodimers additionally developed some visible aggregates, but less than their homodimeric counterparts (G, I, respectively). (K) Total PBS extracts of transiently transfected cells were separated into supernatant and pellet fractions by centrifugation. Fifteen micrograms of protein were separated on 12% standard SDS–PAGE gels and analyzed by Western blot using SOD100 antibodies to monitor the distribution of fusion proteins (top), endogenous SOD1(hWT) (cytosolic fraction; middle) and histon H3 (pellet fraction; bottom). (L) Results of the densitometric analysis of the fusion protein bands are expressed as the ratio between pellet and soluble fractions. Bars represent mean values ± SD of at least three independent experiments. All mutant-containing dimers have a statistically significant higher aggregation potential than SOD1(hWT) homodimers (‘*’ on bars), and all mutant homodimers are more prone to aggregate than their heterodimeric counterparts (‘*’ on brackets). One-way ANOVA followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001; n ≥ 3.

N2A cells can tolerate low levels of homo- and heterodimeric mutant SOD1 proteins

To study the functional properties of dimeric SOD1 proteins, we established neuroblastoma N2A cell lines stably expressing SOD1(G85R) and SOD1(G37R) homo- and heterodimers. Two independent lines were generated to avoid clonal effects. The expression levels of the dimeric proteins were determined by Western blots using anti-GFP (Fig. 4A) and the SOD100 antibodies (Supplementary Material, Fig. S1A) and represented by EGFP-fluorescence (Supplementary Material, Fig. S2). In addition, we analyzed the dismutase activity levels, whereby only the SOD1(hWT) homodimer and both SOD1(G37R)-bearing dimers showed detectable activities according to the protein levels. Although the SOD1(G85R) heterodimer had significant dismutase activity after transient transfection, the expression level was not high enough to detect activity in stable lines (Supplementary Material, Fig. S1B). It is noteworthy that the two lines expressing the SOD1(G85R) heterodimer were the highest expressing lines after intensive screening, indicating that SOD1(G85R) heterodimers were not well tolerated. In addition, we analyzed the aggregation behavior of the dimer proteins stably expressed by the N2A cells. Consistent with earlier experiments, SOD1(G85R) and SOD1(G37R) homodimers were more enriched in the pellet fractions compared with their heterodimeric counterparts (Fig. 4B and C), although visible aggregates were not detected (Supplementary Material, Fig. S2). The aggregation potential of mutant homodimers was increased compared with their heterodimeric counterparts. It is noteworthy and consistent with earlier results that the overall aggregate formation for any given fusion protein was reduced in stable lines compared with transient transfections because of reduced levels of mutant proteins.

Figure 4.

Aggregate formation in neuroblastoma N2A cells stably expressing SOD1(G37R)- and SOD1(G85R)-bearing homo- and heterodimers. (A) The levels of dimer expression in stably transfected N2A cells have been determined after separation of total cell extracts on standard 8% SDS–PAGE by Western blot analysis using an anti-GFP antibody (top). Two individual cell lines per mutant-bearing construct have been generated. Levels are normalized to tubulin that was detected on the same membrane (middle). The expression levels of mutant-bearing dimers are shown as the fraction of SOD1(hWT) homodimer expression (n = 3; bottom). (B) PBS extracts of N2A cells stably expressing dimer constructs were divided into supernatant and pellet fractions by centrifugation and then separated on 8% SDS–PAGE. The Western blots were developed with the SOD100 antibody. Three independent experiments are shown. (C) Quantification of Western blots is shown as the ratio between pellet and supernatant fractions. Asterisks on bars represent significance to the aggregation of SOD1(hWT) homodimer; asterisks on brackets, between the dimers indicated. Values are expressed as the mean ± SD (n = 3). One-way ANOVA followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001; n = 3.

Figure 4.

Aggregate formation in neuroblastoma N2A cells stably expressing SOD1(G37R)- and SOD1(G85R)-bearing homo- and heterodimers. (A) The levels of dimer expression in stably transfected N2A cells have been determined after separation of total cell extracts on standard 8% SDS–PAGE by Western blot analysis using an anti-GFP antibody (top). Two individual cell lines per mutant-bearing construct have been generated. Levels are normalized to tubulin that was detected on the same membrane (middle). The expression levels of mutant-bearing dimers are shown as the fraction of SOD1(hWT) homodimer expression (n = 3; bottom). (B) PBS extracts of N2A cells stably expressing dimer constructs were divided into supernatant and pellet fractions by centrifugation and then separated on 8% SDS–PAGE. The Western blots were developed with the SOD100 antibody. Three independent experiments are shown. (C) Quantification of Western blots is shown as the ratio between pellet and supernatant fractions. Asterisks on bars represent significance to the aggregation of SOD1(hWT) homodimer; asterisks on brackets, between the dimers indicated. Values are expressed as the mean ± SD (n = 3). One-way ANOVA followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001; n = 3.

SOD1(hWT)-SOD1(G85R) heterodimer-expressing cells are more vulnerable to secondary stress than SOD1(G85R) homodimer cells

N2A cells expressing monomeric mutant SOD1 are vulnerable in a variety of stress paradigms (31,32). We tested whether differentiated N2A cells stably expressing SOD1 dimer proteins respond differently to oxidative stress as induced by H2O2 or xanthine/xanthine oxidase treatment. We observed that cells expressing SOD1(G37R)-containing dimers, both homo- and heterodimers, developed a similar toxicity, but were more vulnerable in both stress paradigms compared with cells expressing SOD1(hWT) homodimers (Fig. 5B and D). In contrast, cells expressing SOD1(G85R) homo- and heterodimers responded differently in toxicity assays, but cells expressing a given dimer protein behaved similarly. SOD1(G85R) homodimer cells displayed a reduced survival rate after xanthine/xanthine oxidase exposure (Fig. 5A), whereas the survival was not statistically different to SOD(hWT) homodimer-expressing cells after H2O2 treatment (Fig. 5C). SOD1(G85R) heterodimer-expressing cells suffered a pronounced cell death after exposure to both stressors. This is even more striking, as the expression level of SOD1(G85R) heterodimer is less than half of the SOD1(G85R) homodimer, resulting in an almost four-fold reduction of mutant monomer equivalents. Taken together, we can state that all active mutants, the SOD1(G37R) homo- and heterodimers as well as the SOD1(G85R) heterodimers strongly reduced viability of N2A cells, whereas the inactive SOD1(G85R) homodimers only showed a minor toxicity when challenged with H2O2. In summary, these results indicated that the cellular toxicity of SOD1 dimer constructs is not associated with their aggregation potential, but is rather correlated with the enzymatic activity of the fusion proteins.

Figure 5.

N2A cells expressing dismutase-active dimers are more susceptible to secondary stress than SOD1(G85R) homodimers. (A)–(D) Differentiated N2A cells expressing SOD1(G85R) [(A) and (C)] and SOD1(G37R) [(B) and (D)] bearing homo- and heterodimers were challenged for 24 h with either xanthine in combination with 10 mU/ml xanthine oxidase [(A) and (B)] or H2O2 [(C) and (D)] with the concentrations indicated. Viability of the cells after treatment was assessed with MTT and expressed as percent survival normalized to the survival of untreated cells. Data represent the mean values ± SD of four (xanthine/xanthine oxidase) and six (H2O2) independent experiments done in six wells per line and condition. Results were compared with the survival of SOD1(hWT) homodimer-expressing cells (‘*’ on bars) and between mutant homo- and heterodimer cells (‘*’ on brackets). (E)–(H) Steady-state carbonylation level of dimer expressing N2A lines as a marker for oxidative stress. Total SDS extracts of N2A cells expressing dimer constructs and differentiated for 6 days were treated with DNPH to modulate carbonyl groups on proteins. The resulting DNP groups were detected after separation of the proteins of SOD1(G37R) and SOD1(G85R) dimer cells on 12% SDS–PAGE gels with a monoclonal anti-DNP antibody on Western blot membranes [(E) and (G), respectively]. One representative blot is shown. The ratio of the DNP signal and the actin signal which was detected on a separate membrane done in parallel (E, G lower panels) was calculated. Values for SOD1(G37R) (n = 4; F) and SOD1(G85R) (n = 3; H) bearing dimer cells are expressed as percentage of SOD1(hWT) homodimer-expressing cells ± SD. Statistical analysis for (A)–(D), (F) and (H) were performed with one-way ANOVA where possible or Kruskal–Wallis one-way ANOVA, followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 5.

N2A cells expressing dismutase-active dimers are more susceptible to secondary stress than SOD1(G85R) homodimers. (A)–(D) Differentiated N2A cells expressing SOD1(G85R) [(A) and (C)] and SOD1(G37R) [(B) and (D)] bearing homo- and heterodimers were challenged for 24 h with either xanthine in combination with 10 mU/ml xanthine oxidase [(A) and (B)] or H2O2 [(C) and (D)] with the concentrations indicated. Viability of the cells after treatment was assessed with MTT and expressed as percent survival normalized to the survival of untreated cells. Data represent the mean values ± SD of four (xanthine/xanthine oxidase) and six (H2O2) independent experiments done in six wells per line and condition. Results were compared with the survival of SOD1(hWT) homodimer-expressing cells (‘*’ on bars) and between mutant homo- and heterodimer cells (‘*’ on brackets). (E)–(H) Steady-state carbonylation level of dimer expressing N2A lines as a marker for oxidative stress. Total SDS extracts of N2A cells expressing dimer constructs and differentiated for 6 days were treated with DNPH to modulate carbonyl groups on proteins. The resulting DNP groups were detected after separation of the proteins of SOD1(G37R) and SOD1(G85R) dimer cells on 12% SDS–PAGE gels with a monoclonal anti-DNP antibody on Western blot membranes [(E) and (G), respectively]. One representative blot is shown. The ratio of the DNP signal and the actin signal which was detected on a separate membrane done in parallel (E, G lower panels) was calculated. Values for SOD1(G37R) (n = 4; F) and SOD1(G85R) (n = 3; H) bearing dimer cells are expressed as percentage of SOD1(hWT) homodimer-expressing cells ± SD. Statistical analysis for (A)–(D), (F) and (H) were performed with one-way ANOVA where possible or Kruskal–Wallis one-way ANOVA, followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001.

Steady-state carbonylation levels of SOD1(G85R) and SOD1(G37R) heterodimer-expressing cells are elevated

As the toxicity of SOD1(G37R) and SOD1(G85R)-containing dimers is not reflected in their aggregation behavior, we analyzed the steady-state carbonylation levels as a measure for oxidative modifications. Protein carbonylation can be introduced directly by reactive oxygen species or as a secondary reaction with reactive carbonyl derivatives of carbohydrates and lipids (reviewed in 33) and is increased in a mouse model of ALS (34) as well as in ALS patients (35). We assayed the levels of protein carbonylation with the dinitrophenylhydrazine (DNPH) method. Western blots of DNPH-derivatized proteins were quantified and normalized to actin which was assayed on a membrane done in parallel (Fig. 5E–H). The presence of carbonylated proteins was significantly higher in cells expressing SOD1(G85R) heterodimers compared with their mutant homodimer counterparts (Fig. 5H). These data were in line with the hypothesis that some aberrant oxidative activity might be necessary to mediate mutant SOD1 toxicity. The carbonylation levels of cells expressing SOD1(G37R) heterodimers were also higher compared with SOD1(G37R) homodimer-expressing cells, although both dimers developed similar toxicities in oxidative stress paradigms (Fig. 5B and D). A possible explanation might be that SOD1(G37R) heterodimers form more stable dimers compared with mutant homodimers, thus generating more oxidative damage under steady-state conditions.

Neuronal expression of SOD1(G85R) heterodimers in C. elegans causes reduced survival after exposure to paraquat and induces motor deficits

C. elegans has been used as a valuable in vivo model system to study proteins involved in neurodegenerative diseases such as Alzheimer’s and Huntington’s disease and ALS (36–38). In the latter, it was established that mutant SOD1 aggregation can be induced upon paraquat exposure, an herbicide inducing the production of superoxide radicals. We generated transgenic worms expressing SOD1(hWT)- and SOD1(G85R)-bearing dimers under the control of a neuron-specific synaptogyrin-1 promoter (sng-1), resulting in a neuronal expression pattern (Fig. 6A). To avoid clonal effects, two independent mutant SOD1-bearing lines were generated, resulting in comparable results in all assays performed. The SOD1(hWT) homodimer and the SOD1(G85R) heterodimer, but not the SOD1(G85R) homodimer, had detectable dismutase activity (Fig. 6B). In order to obtain comparable results in functional assays, we selected integrated SOD1(G85R) homo- and heterodimeric lines with similar expression levels (Fig. 6B and C). All lines shared a comparable lifespan (data not shown). Large aggregates were detected inside the motor neurons and small neurons along the ventral cord expressing SOD1(G85R) homodimers, whereas SOD1(hWT) homodimers were evenly distributed throughout the cells (Fig. 7B and C). In contrast to SOD1(G85R) homodimers, SOD1(G85R) heterodimers formed aggregates only in some cells, but were uniformly distributed in others (Fig. 7D). We could not observe a punctate staining along the ventral cord, indicating a lack of aggregates in the small neurons. This phenotype was also represented in the biochemical analysis, in which SOD1(G85R) homodimers had the highest aggregation potential (Fig. 7E). Aggregation of SOD1(G85R) heterodimers was statistically higher compared with SOD1(hWT) homodimers, but profoundly reduced compared with SOD1(G85R) homodimers. To evaluate, if the expression of SOD1(hWT)- and SOD1(G85R)-bearing dimers has also behavioral consequences, we first monitored motor performance. The locomotion of worms expressing mutant SOD1(G85R) heterodimers was significantly impaired compared with those expressing SOD1(hWT) homodimers (Fig. 7F). Remarkably, SOD1(G85R) heterodimeric C. elegans showed a highly significantly reduced number of deflections compared with SOD1(G85R) homodimer-expressing nematodes. In accordance with the survival assays of N2A cells expressing SOD1(G85R) dimers, the SOD1(G85R) heterodimeric nematodes had a reduced survival rate after exposure to paraquat (2.5 mm) compared with the SOD1(G85R) homodimer expressing worms (Figs. 7G). Paraquat is a herbicide inducing superoxide generation. All tested nematode lines reacted in a dose-dependent manner in response to higher paraquat concentrations (5 mm; Fig. 7G). At higher concentrations, the paraquat effect overrode the toxic effect of the SOD1(G85R) heterodimer expression. Under control conditions, no change of survival has been observed (Supplementary Material, Fig. S3). As the dimer proteins are expressed exclusively in neurons, we conclude that the expression of SOD1(G85R) heterodimers is more harmful to neuronal function, although their aggregation potential is reduced compared with SOD1(G85R) homodimers.

Figure 6.

SOD1(G85R) heterodimers display dismutase activity in C. elegans. (A) Expression of EGFP and EGFP-tagged SOD1(hWT) and SOD1(G85R) homodimer and SOD1(G85R) heterodimer constructs are driven by the neuron specific C. elegans sng-1 promotor. Constructs were co-injected with the pRF4 plasmid inducing the roller phenotype. The large motor neurons in the ventral cord are easily detectable (arrow). A fluorescent picture of a nematode expressing SOD1(hWT) homodimers is shown. (B) Expression levels were determined by Western blot with anti-GFP antibodies after separating an extract of 10 worms per line on 12% SDS–PAGE (top panel). Tubulin served as a loading control on the same blot (middle panel). Extracts of 10 worms were loaded on native 6% PAGE gels to monitor dismutase activity (bottom panel). SOD1(hWT) homodimers and SOD1(G85R) heterodimers were active, whereas SOD1(G85R) homodimers had no dismutase activity. (C) The expression levels of SOD1(hWT) and SOD1(G85R) homo- and heterodimer-expressing transgenic worms were quantified and the ratio between the GFP and tubulin signals was calculated. Error bars represent standard deviation (n = 3).

Figure 6.

SOD1(G85R) heterodimers display dismutase activity in C. elegans. (A) Expression of EGFP and EGFP-tagged SOD1(hWT) and SOD1(G85R) homodimer and SOD1(G85R) heterodimer constructs are driven by the neuron specific C. elegans sng-1 promotor. Constructs were co-injected with the pRF4 plasmid inducing the roller phenotype. The large motor neurons in the ventral cord are easily detectable (arrow). A fluorescent picture of a nematode expressing SOD1(hWT) homodimers is shown. (B) Expression levels were determined by Western blot with anti-GFP antibodies after separating an extract of 10 worms per line on 12% SDS–PAGE (top panel). Tubulin served as a loading control on the same blot (middle panel). Extracts of 10 worms were loaded on native 6% PAGE gels to monitor dismutase activity (bottom panel). SOD1(hWT) homodimers and SOD1(G85R) heterodimers were active, whereas SOD1(G85R) homodimers had no dismutase activity. (C) The expression levels of SOD1(hWT) and SOD1(G85R) homo- and heterodimer-expressing transgenic worms were quantified and the ratio between the GFP and tubulin signals was calculated. Error bars represent standard deviation (n = 3).

Figure 7.

Neuronal expression of SOD1(G85R) heterodimers in C. elegans results in impaired motility and reduced stress resistance. (A)–(D) High magnification view of worms expressing EGFP (A), SOD1(hWT) (B) and SOD1(G85R) homodimers (C) and the SOD1(G85R) heterodimer (D). SOD1(G85R) homodimers form large aggregates (arrow). These are located inside the large neurons and visible even in the small neurons along the ventral cord (open arrows). SOD1(G85R) heterodimers are distributed throughout the cells, but form also aggregates in other cells. (E) Aggregation behavior of SOD1 dimer proteins was determined by Western blot. PBS extracts of mixed cultures of nematodes were separated by centrifugation. Equal protein amounts of supernatant and pellet fractions were loaded on a 12% SDS–PAGE gel. EGFP-tagged proteins were detected using a GFP antibody (top panel). Tubulin as a marker for the cytosolic fraction was detected on the same membrane (middle panel). Photoluminescent signals were quantified and the ratio of pellet and supernatant fractions were calculated (bottom panel). Data are represented as mean ± SD (n = 3). Asterisk on bars represents significance to SOD1(G85R) homodimers. (F) Locomotion of transgenic L4 larvae. Animals were placed in a drop of M9 buffer. After recovery (1 min), deflections within 30 s were counted. Fifty animals per line were monitored and data presented in a scatter plot. The line represents the mean. Asterisks on bar represent the statistical significance to the SOD1(hWT) homodimer-expressing line; asterisks on the bracket, between the two lines indicated. (G) Twenty L4 larvae of controls (Mock), SOD1(hWT) homodimers (WTWT) and SOD1(G85R) homo-(8585) and heterodimers (85WT) were placed on FUDR plates containing 2.5 (black patterned lines) or 5 mm paraquat (colored solid lines) and cultured at 20°C. Every 24 h, survival of nematodes was assayed. Viability was expressed as percent survival ± SD normalized to the survival at day 1 (n = 5); ‘#’ represents statistical significance to SOD1(hWT) homodimers, ‘*’ to SOD1(G85R) homodimers. No statistical significance between the different worm lines was observed with 5 mm paraquat. (E) Statistical analysis was performed with the Kruskal–Wallis one-way ANOVA; (F) and (G) One-way ANOVA. Both were followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 7.

Neuronal expression of SOD1(G85R) heterodimers in C. elegans results in impaired motility and reduced stress resistance. (A)–(D) High magnification view of worms expressing EGFP (A), SOD1(hWT) (B) and SOD1(G85R) homodimers (C) and the SOD1(G85R) heterodimer (D). SOD1(G85R) homodimers form large aggregates (arrow). These are located inside the large neurons and visible even in the small neurons along the ventral cord (open arrows). SOD1(G85R) heterodimers are distributed throughout the cells, but form also aggregates in other cells. (E) Aggregation behavior of SOD1 dimer proteins was determined by Western blot. PBS extracts of mixed cultures of nematodes were separated by centrifugation. Equal protein amounts of supernatant and pellet fractions were loaded on a 12% SDS–PAGE gel. EGFP-tagged proteins were detected using a GFP antibody (top panel). Tubulin as a marker for the cytosolic fraction was detected on the same membrane (middle panel). Photoluminescent signals were quantified and the ratio of pellet and supernatant fractions were calculated (bottom panel). Data are represented as mean ± SD (n = 3). Asterisk on bars represents significance to SOD1(G85R) homodimers. (F) Locomotion of transgenic L4 larvae. Animals were placed in a drop of M9 buffer. After recovery (1 min), deflections within 30 s were counted. Fifty animals per line were monitored and data presented in a scatter plot. The line represents the mean. Asterisks on bar represent the statistical significance to the SOD1(hWT) homodimer-expressing line; asterisks on the bracket, between the two lines indicated. (G) Twenty L4 larvae of controls (Mock), SOD1(hWT) homodimers (WTWT) and SOD1(G85R) homo-(8585) and heterodimers (85WT) were placed on FUDR plates containing 2.5 (black patterned lines) or 5 mm paraquat (colored solid lines) and cultured at 20°C. Every 24 h, survival of nematodes was assayed. Viability was expressed as percent survival ± SD normalized to the survival at day 1 (n = 5); ‘#’ represents statistical significance to SOD1(hWT) homodimers, ‘*’ to SOD1(G85R) homodimers. No statistical significance between the different worm lines was observed with 5 mm paraquat. (E) Statistical analysis was performed with the Kruskal–Wallis one-way ANOVA; (F) and (G) One-way ANOVA. Both were followed by Student–Newman–Keul’s post hoc analysis; *P < 0.05; **P < 0.01; ***P < 0.001.

DISCUSSION

There is an ongoing discussion up to now whether and in which way SOD1(hWT) participates in mutant SOD1-mediated ALS. It has been established in a zebrafish and at least three mouse models of mutant SOD1 toxicity that the co-expression of SOD1(hWT) hastens pathology, although the mode of action is yet unresolved (21–25). It has recently been shown that SOD1(hWT) acquires toxic, disease-promoting properties when it is oxidatively damaged (17,39). In addition, oxidized SOD1(hWT) has been identified in aggregates containing non-native disulfide-linked SOD1 located in mitochondria in vivo in double transgenic mice (21,40). Furthermore, the heterodimerization of mutant and SOD1(hWT) might actually stabilize mutant proteins as proposed previously (22). Here, we provide in vitro and in vivo evidence that SOD1(hWT) even as a native protein might actually directly be involved in the pathogenesis of ALS through the heterodimerization with mutant SOD1.

We followed the hypothesis that mutant and SOD1(hWT) might form heterodimers and thus change the functional properties of SOD1(hWT). It has been shown earlier that mutant-wild-type heterodimers are formed in vivo in extracts of lymphocytes derived from SOD1(G37R) patients (28). Although there are conflicting results on whether different, particularly dismutase-inactive mutant SOD1, proteins are able to dimerize with SOD1(hWT) and whether this interaction affects SOD1(hWT) function (19,26), we could detect dismutase-active heterodimers of SOD1(hWT) with SOD1(A4V), SOD1(G37R), SOD1(G93C), and even with the inactive SOD1(G85R) mutant (Fig. 1). In order to characterize the functional properties of SOD1 homo- and heterodimers, we designed fusion proteins of two SOD1 monomers linked by a flexible polypeptide. The expressed proteins shared many properties of SOD1 dimers that were formed between two single monomers. These features included migration in native as well as in denaturing gels, aggregation behavior and dismutase activity. This indicates that the fusion proteins were folded properly and were sufficiently loaded with Cu2+- and Zn2+-ions. The interaction of the two SOD1 monomers within the fusion protein forming a functional enzyme was preferred as we barely, if at all, observe multimers of fusion proteins or fusion proteins and endogenous untagged SOD1 monomers through domain swapping. Nevertheless, the interaction of the two monomers is not static and does not prevent aggregate formation as it has been observed as a result of a direct covalent linkage of two monomers (41). We now for the first time have tools in hand to study the functional properties of mutant homo- and heterodimeric SOD1.

All mutant-containing dimers exert a higher aggregation potential than SOD1(hWT) homodimers. In fact, the increased aggregate formation of mutant homodimers compared with their heterodimeric counterparts might reflect a change in dimer stability (22), as dimer instability and the presence of monomeric SOD1 are postulated to be required for aggregate formation (17,20,42,43). Consistent with this is a recent evidence that co-incubation of Zn-deficient SOD1 mutants with SOD1(hWT) reduces aggregation in vitro by the formation of stable heterodimers (44). The differences in aggregate formation of our SOD1 dimers were consistently seen in three different experimental settings, including a novel C. elegans model. Intriguingly, the aggregation potential of mutant homo- and heterodimers did not correlate with their toxicity. This is a particularly striking result because we detected massive aggregates in SOD1(G85R) homodimer expressing nematodes (Fig. 7). In line with these results is the observation that upon over-expression of the copper chaperon for SOD1 mutant, SOD1 mice develop an accelerated disease without the formation of any detectable aggregates or SOD1 multimers (45).

Our data indicate that toxicity of mutant containing SOD1 dimers is rather correlated with the presence of enzymatic activity in two SOD1 dimer-expressing systems (Table 1). The presence of elevated protein carbonylation even at steady-state levels is indeed consistent with the view that mutant heterodimers cause increased oxidative stress. We could clearly show that all SOD1 mutants tested were able to form dismutase-active heterodimers. Even heterodimers of dismutase-inactive SOD1(G85R) with SOD1(hWT) exert Cu-dependent redox chemistry. This activity is indeed proposed to be necessary for an aberrant enzymatic activity, as it has been shown for wild-type-like, Cu-loaded SOD1 mutants (reviewed in 5).

Table 1.

Aggregation behavior and toxicity of SOD1 dimer constructs

Construct Activity Aggregates in N2A cells/C. elegans Toxicity in N2A cells/C. elegans 
SOD1(hWT)-SOD1(hWT) −/− −/− 
SOD1(G37R)-SOD1(hWT) −/ND +++/ND 
SOD1(G37R)-SOD1(G37R) ++/ND +++/ND 
SOD1(G85R)-SOD1(hWT) −/+ +++/+++ 
SOD1(G85R)-SOD1(G85R) − ++/++ +a/− 
Construct Activity Aggregates in N2A cells/C. elegans Toxicity in N2A cells/C. elegans 
SOD1(hWT)-SOD1(hWT) −/− −/− 
SOD1(G37R)-SOD1(hWT) −/ND +++/ND 
SOD1(G37R)-SOD1(G37R) ++/ND +++/ND 
SOD1(G85R)-SOD1(hWT) −/+ +++/+++ 
SOD1(G85R)-SOD1(G85R) − ++/++ +a/− 

ND, not determined.

aNo toxicity in survival assay with H2O2, but with xanthine/XO.

We hypothesize that an aberrant enzymatic activity of mutant heterodimers might be induced by structural alterations in SOD1(hWT) upon heterodimerization, particularly with mutant SOD1s that display large conformational changes similar to SOD1(G85R). Studies on other homodimeric enzymes such as glutathione S-transferase A1-1 and galacto-1-P uridylyltransferase indicated that heterodimers of mutant and wild-type enzymes form, but have an altered activity (46,47). This functional change is mediated through the structural interaction of mutant with wild-type monomers, as the active center of the wild-type protein is altered in the crystal structure of glutathione S-transferase A1-1 heterodimers (47). Although the currently available crystal structures of homodimeric apo- and holo wild-type enzymes (48) and different mutant SOD1s (42,49) do not have the strength to pinpoint a significant structural crosstalk between two SOD1 monomers, the analysis of wild-type-like mutants SOD1(A4V) and SOD1(I113T) showed that rather subtle changes in the loop regions which determine substrate specificity and the relative subunit orientation can mediate toxicity (42). It is noteworthy that the metal-binding cavity and the active site were not altered compared with SOD1(hWT) homodimers. In light of these data, we hypothesize that rather small structural alterations of SOD1(hWT) induced by heterodimerization with mutant SOD1 might change substrate specificity and therefore increase an aberrant oxidative activity (7–10) and/or lead to an accelerated S-nitrosothiol breakdown (50,51). Although the formation of SOD1(G85R) heterodimers is likely not preferred as it has not been observed earlier (26) and as the dismutase activity is low but clearly detectable (this study), it is accepted that only minute quantities of mutant protein contribute to disease (52,53).

It is noteworthy that Goldsteins et al. (54) very recently observed a deleterious effect of SOD1 in the mitochondrial intermembrane space. Their data clearly showed that elevated levels of even wild-type SOD1 activity lead to increased generation of reactive oxygen species in the mitochondria because of an imbalance of hydroperoxid. On the basis of these data, the formation of dismutase-active heterodimers and their localization in mitochondria might be sufficient to induce cellular toxicity.

On the basis of the analysis of homo- and heterodimeric SOD1 fusion proteins (Table 1), we can now provide evidence that the aggregation potential per se does not correlate with the increased toxic properties in a cell culture and C. elegans model. We could show that even Cu-deficient SOD1 mutants form enzymatically active SOD1 dimers when heterodimerized with SOD1(hWT). A gain of aberrant enzymatic activity might be the initial step in line of events contributing to pathogenesis and possibly lead to the formation of oxidized SOD1 which, in turn, causes protein oligomerization, aggregation and aberrant subcellular localization. The formation of heterodimers of mutant and SOD1(hWT) might actually provide the basis for the dominant inheritance of mutant SOD1-mediated ALS. Further studies are under way to pinpoint the toxic properties of heterodimeric SOD1.

MATERIALS AND METHODS

A detailed version of the Materials and Methods can be found in the Supplementary Data.

Generation of monomer and dimer constructs

Human SOD1 cDNAs (SOD1(G37R), SOD1(G85R), SOD1(G93C), SOD1(A4V) and SOD1(hWT)) were inserted in the pEGFP-N1 Vector (Clontech) to allow the expression of EGFP-tagged monomeric SOD1. To generate EGFP-tagged dimeric fusion proteins, two SOD1 cDNAs and a linker oligonucleotide (Supplementary Material, Table S1; 27) were sequentially cloned into the pEGFP-N1 vector. The expression of SOD1(G85R)-SOD1(G85R), SOD1(hWT)-SOD1(hWT), SOD1(G85R)-SOD1(hWT) dimers in C. elegans was driven by a neuron-specific sng-1 promotor.

Expression of SOD1 constructs in cell lines

Human HEK293 and mouse N2A cells were maintained in DMEM supplemented with 10% (v/v) fetal bovine serum, 1% (v/v) pyruvate, 100 U/ml penicillin and 100 µg/ml streptomycin at 37°C in a 5% CO2-humidified atmosphere. Transient transfections of HEK293 cells were carried out by the calcium phosphate precipitation method (55) and analyzed 48 h after transfection. Microscopic analysis was carried out with an inverted Zeiss Axiovert 200 microscope equipped with a Spot-RT-SE camera (Diagnostic Instruments, Visitron). Cell lines stably expressing SOD1 dimer constructs were generated by transfecting N2A cells with Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. Clonal lines were maintained with 600 µg/ml Geneticin (G418; Invitrogen).

Survival assay

N2A cells were plated in 96-well-plates and differentiated for 48 h in standard medium supplemented with 2% (v/v) fetal calf serum and 1 µm retinoic acid. Cells were treated for 24 h with 200 and 300 µm H2O2 or 100 and 200 µm xanthine with 10 mU/ml xanthine oxidase. Viability was assessed with the MTT method (see Supplementary Data).

Assessment of SOD1 activity

Cells or nematodes were extracted in PBS containing protease inhibitor cocktail tablets (Roche) and sonicated four times at 50 Hz for 10 s. Thirty microgram cell extract or 60 µg nematode extract was loaded on a non-denaturing polyacrylamide gel [15% (v/v) for cell extracts; 6% (v/v) for nematode extracts]. The staining procedure is detailed in Supplementary Material.

Protein extraction and Western blot

Cells or nematodes were extracted in PBS containing protease inhibitor cocktail tablets (Roche) and sonicated four times at 50 Hz for 10 s [referred to as total lysate (L)]. Insoluble proteins were pelleted at 15 500g and 4°C for 15 min. The supernatant (S) was collected and the pellet was washed, resuspended in 300 µl SDS-containing buffer and sonicated four times at 50 Hz for 10 s (P). Expression levels of stable C. elegans lines were determined by extraction of 10 adult nematodes in sample loading buffer containing SDS and β-mercaptoethanol and sonification at 50 Hz for two times. Equal amounts of protein or nematode extract were separated on denaturing SDS–PAGE gels as specified and subsequently transferred on nitrocellulose membranes. Membranes were blocked and incubated with polyclonal SOD-100 antibody (1:2500; Stressgene), monoclonal GFP-antibody (1:1000, Covance), monoclonal histone H3 antibody (Abcam), monoclonal tubulin antibody (1:1000, Sigma) or polyclonal actin antibody (1:1000, Sigma). Blots were developed with peroxidase-conjugated secondary antibody (1:10 000, Jackson), and chemoluminescent signals were detected with the LAS300 system (Fuji, Raytest) and quantified using the Aida software (Raytest).

Detection of carbonylated protein residues

N2A cells stably expressing SOD1 dimer fusion constructs were extracted and sonicated four times at 50 Hz for 10 s in SDS-containing lysis buffer after 6 days of differentiation with differentiation medium. Carbonylated proteins were derivatized with 2,4-DNPH and subsequently detected on Western blot membranes using a polyclonal DNP antibody (1:1000, Molecular Probes) as specified in Supplementary Material.

C. elegans methods

For the generation of transgenic animals, plasmids encoding EGFP, SOD1(hWT)-SOD1(hWT)-EGFP, SOD1(G85R)-SOD1(G85R)-EGFP and SOD1(G85R)-SOD1(hWT)-EGFP under the control of the sng-1 promoter were co-injected with the selection marker pRF4 [rol6 (su1006dm)] in N2 and integrated by UV irradiation as detailed in Supplementary Material. For each mutant construct, at least two independent stable lines were created and analyzed. The motility assay was carried out at 22°C with L4 larvae. Thrashing rate was determined by placing individual nematodes in one drop of M9 buffer. After 1 min of recovery, deflections of the nematodes within 30 s were counted. Paraquat sensitivity assays were carried out on NGM plates containing 2.5 and 5 mm paraquat, respectively. Twenty L4 larvae were placed on paraquat or control plates containing 0.1 mg/ml 5-fluorodeoxyuridine (FUDR) to avoid progeny (56) and cultured at 20°C. Every 24 h, survival was scored as the nematodes response to gentle touching with a platinum wire. Missing nematodes were eliminated from the statistics.

Statistical analysis

To determine whether data are statistically significant, we performed Kruskal–Wallis one-way Anova on ranks tests for multiple comparisons. One-way Anova was used if data passed the test for equal variance. Both tests were combined with the Student–Newman–Keul’s method for post hoc analysis.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG Online

FUNDING

H.W. is a member of the neuroscience graduate school at the University of Mainz (DFG GRK1044); A.K. was in part supported by a scholarship of the Boehringer Ingelheim Fonds. Financial support was provided by the Fritz und Hildegard Berg-Stiftung (C.B.), the MAIFOR program (A.M.C.), the IFZN of the University of Mainz (A.M.C.) and Stiftung fuer Innovation Rheinland-Pfalz (A.M.C., C.B.).

ACKNOWLEDGEMENTS

We thank K. Hüsken and R. Leube (University of Mainz) for the generous gift of the pVH10.10 vector modified with the sng-1 promoter (1727) and helpful discussions, Rueyling Lin (UTSW Medical Center, Dallas) for introducing A.K. into C. elegans methods, Harald Neumann (University of Bonn) for stimulating discussions, Brigitte Welsch for expert technical assistance and Parvana Hajieva (University of Mainz) for help with the carbonylation assay.

Conflict of Interest statement. None declared.

REFERENCES

1
Boillee
S.
Velde Vande
C.
Cleveland
D.W.
ALS: a disease of motor neurons and their nonneuronal neighbors
Neuron
 , 
2006
, vol. 
52
 (pg. 
39
-
59
)
2
Clement
A.M.
Nguyen
M.D.
Roberts
E.A.
Garcia
M.L.
Boillee
S.
Rule
M.
McMahon
A.P.
Doucette
W.
Siwek
D.
Ferrante
R.J.
, et al.  . 
Wild-type nonneuronal cells extend survival of SOD1 mutant motor neurons in ALS mice
Science
 , 
2003
, vol. 
302
 (pg. 
113
-
117
)
3
Boillee
S.
Yamanaka
K.
Lobsiger
C.S.
Copeland
N.G.
Jenkins
N.A.
Kassiotis
G.
Kollias
G.
Cleveland
D.W.
Onset and progression in inherited ALS determined by motor neurons and microglia
Science
 , 
2006
, vol. 
312
 (pg. 
1389
-
1392
)
4
Fridovich
I.
Superoxide dismutases
Annu. Rev. Biochem.
 , 
1975
, vol. 
44
 (pg. 
147
-
159
)
5
Valentine
S.J.
Doucette
P.A.
Zittin Potter
S.
Copper–zinc superoxide dismutase and amyotrophic lateral sclerosis
Annu. Rev. Biochem.
 , 
2005
, vol. 
74
 (pg. 
563
-
593
)
6
Reaume
A.G.
Elliott
J.L.
Hoffman
E.K.
Kowall
N.W.
Ferrante
R.J.
Siwek
D.F.
Wilcox
H.M.
Flood
D.G.
Beal
M.F.
Brown
R.H.
Jr.
, et al.  . 
Motor neurons in Cu/Zn superoxide dismutase-deficient mice develop normally but exhibit enhanced cell death after axonal injury
Nat. Genet.
 , 
1996
, vol. 
13
 (pg. 
43
-
47
)
7
Beckman
J.S.
Carson
M.
Smith
C.D.
Koppenol
W.H.
ALS, SOD and peroxynitrite
Nature
 , 
1993
, vol. 
364
 pg. 
584
 
8
Yim
M.B.
Kang
J.H.
Yim
H.S.
Kwak
H.S.
Chock
P.B.
Stadtman
E.R.
A gain-of-function of an amyotrophic lateral sclerosis-associated Cu,Zn-superoxide dismutase mutant: an enhancement of free radical formation due to a decrease in Km for hydrogen peroxide
Proc. Natl Acad. Sci. USA
 , 
1996
, vol. 
93
 (pg. 
5709
-
5714
)
9
Wiedau-Pazos
M.
Goto
J.J.
Rabizadeh
S.
Gralla
E.B.
Roe
J.A.
Lee
M.K.
Valentine
J.S.
Bredesen
D.E.
Altered reactivity of superoxide dismutase in familial amyotrophic lateral sclerosis
Science
 , 
1996
, vol. 
271
 (pg. 
515
-
518
)
10
Crow
J.P.
Sampson
J.B.
Zhuang
Y.
Thompson
J.A.
Beckman
J.S.
Decreased zinc affinity of amyotrophic lateral sclerosis-associated superoxide dismutase mutants leads to enhanced catalysis of tyrosine nitration by peroxynitrite
J. Neurochem.
 , 
1997
, vol. 
69
 (pg. 
1936
-
1944
)
11
Subramaniam
J.R.
Lyons
W.E.
Liu
J.
Bartnikas
T.B.
Rothstein
J.
Price
D.L.
Cleveland
D.W.
Gitlin
J.D.
Wong
P.C.
Mutant SOD1 causes motor neuron disease independent of copper chaperone-mediated copper loading
Nat. Neurosci.
 , 
2002
, vol. 
5
 (pg. 
301
-
307
)
12
Wang
J.
Slunt
H.
Gonzales
V.
Fromholt
D.
Coonfield
M.
Copeland
N.G.
Jenkins
N.A.
Borchelt
D.R.
Copper-binding-site-null SOD1 causes ALS in transgenic mice: aggregates of non-native SOD1 delineate a common feature
Hum. Mol. Genet.
 , 
2003
, vol. 
12
 (pg. 
2753
-
2764
)
13
Barber
S.C.
Mead
R.J.
Shaw
P.J.
Oxidative stress in ALS: a mechanism of neurodegeneration and a therapeutic target
Biochim. Biophys. Acta
 , 
2006
, vol. 
1762
 (pg. 
1051
-
1067
)
14
Stathopulos
P.B.
Rumfeldt
J.A.
Scholz
G.A.
Irani
R.A.
Frey
H.E.
Hallewell
R.A.
Lepock
J.R.
Meiering
E.M.
Cu/Zn superoxide dismutase mutants associated with amyotrophic lateral sclerosis show enhanced formation of aggregates in vitro
Proc. Natl Acad. Sci. USA
 , 
2003
, vol. 
100
 (pg. 
7021
-
7026
)
15
Liu
J.
Lillo
C.
Jonsson
P.A.
Vande Velde
C.
Ward
C.M.
Miller
T.M.
Subramaniam
J.R.
Rothstein
J.D.
Marklund
S.
Andersen
P.M.
, et al.  . 
Toxicity of familial ALS-linked SOD1 mutants from selective recruitment to spinal mitochondria
Neuron
 , 
2004
, vol. 
43
 (pg. 
5
-
17
)
16
Pasinelli
P.
Belford
M.E.
Lennon
N.
Bacskai
B.J.
Hyman
B.T.
Trotti
D.
Brown
R.H.
Jr.
Amyotrophic lateral sclerosis-associated SOD1 mutant proteins bind and aggregate with Bcl-2 in spinal cord mitochondria
Neuron
 , 
2004
, vol. 
43
 (pg. 
19
-
30
)
17
Rakhit
R.
Crow
J.P.
Lepock
J.R.
Kondejewski
L.H.
Cashman
N.R.
Chakrabartty
A.
Monomeric Cu,Zn-superoxide dismutase is a common misfolding intermediate in the oxidation models of sporadic and familial amyotrophic lateral sclerosis
J. Biol. Chem.
 , 
2004
, vol. 
279
 (pg. 
15499
-
15504
)
18
Rakhit
R.
Robertson
J.
Velde
C.V.
Horne
P.
Ruth
D.M.
Griffin
J.
Cleveland
D.W.
Cashman
N.R.
Chakrabartty
A.
An immunological epitope selective for pathological monomer-misfolded SOD1 in ALS
Nat. Med.
 , 
2007
, vol. 
13
 (pg. 
754
-
759
)
19
Wang
J.
Caruano-Yzermans
A.
Rodriguez
A.
Scheurmann
J.P.
Slunt
H.H.
Cao
X.
Gitlin
J.
Hart
P.J.
Borchelt
D.R.
Disease-associated mutations at copper ligand histidine residues of superoxide dismutase 1 diminish the binding of copper and compromise dimer stability
J. Biol. Chem.
 , 
2007
, vol. 
282
 (pg. 
345
-
352
)
20
Lindberg
M.J.
Bystrom
R.
Boknas
N.
Andersen
P.M.
Oliveberg
M.
Systematically perturbed folding patterns of amyotrophic lateral sclerosis (ALS)-associated SOD1 mutants
Proc. Natl Acad. Sci. USA
 , 
2005
, vol. 
102
 (pg. 
9754
-
9759
)
21
Deng
H.X.
Shi
Y.
Furukawa
Y.
Zhai
H.
Fu
R.
Liu
E.
Gorrie
G.H.
Khan
M.S.
Hung
W.Y.
Bigio
E.H.
, et al.  . 
Conversion to the amyotrophic lateral sclerosis phenotype is associated with intermolecular linked insoluble aggregates of SOD1 in mitochondria
Proc. Natl Acad. Sci. USA
 , 
2006
, vol. 
103
 (pg. 
7142
-
7147
)
22
Fukada
K.
Nagano
S.
Satoh
M.
Tohyama
C.
Nakanishi
T.
Shimizu
A.
Yanagihara
T.
Sakoda
S.
Stabilization of mutant Cu/Zn superoxide dismutase (SOD1) protein by coexpressed wild SOD1 protein accelerates the disease progression in familial amyotrophic lateral sclerosis mice
Eur. J. Neurosci.
 , 
2001
, vol. 
14
 (pg. 
2032
-
2036
)
23
Jaarsma
D.
Haasdijk
E.D.
Grashorn
J.A.
Hawkins
R.
van Duijn
W.
Verspaget
H.W.
London
J.
Holstege
J.C.
Human Cu/Zn superoxide dismutase (SOD1) overexpression in mice causes mitochondrial vacuolization, axonal degeneration, and premature motoneuron death and accelerates motoneuron disease in mice expressing a familial amyotrophic lateral sclerosis mutant SOD1
Neurobiol. Dis.
 , 
2000
, vol. 
7
 (pg. 
623
-
643
)
24
Lemmens
R.
Van Hoecke
A.
Hersmus
N.
Geelen
V.
D’Hollander
I.
Thijs
V.
Van Den Bosch
L.
Carmeliet
P.
Robberecht
W.
Overexpression of mutant superoxide dismutase 1 causes a motor axonopathy in the zebrafish
Hum. Mol. Genet.
 , 
2007
, vol. 
16
 (pg. 
2359
-
2365
)
25
Bruijn
L.I.
Houseweart
M.K.
Kato
S.
Anderson
K.L.
Anderson
S.D.
Ohama
E.
Reaume
A.G.
Scott
R.W.
Cleveland
D.W.
Aggregation and motor neuron toxicity of an ALS-linked SOD1 mutant independent from wild-type SOD1
Science
 , 
1998
, vol. 
281
 (pg. 
1851
-
1854
)
26
Borchelt
D.R.
Guarnieri
M.
Wong
P.C.
Lee
M.K.
Slunt
H.S.
Xu
Z.S.
Sisodia
S.S.
Price
D.L.
Cleveland
D.W.
Superoxide dismutase 1 subunits with mutations linked to familial amyotrophic lateral sclerosis do not affect wild-type subunit function
J. Biol. Chem.
 , 
1995
, vol. 
270
 (pg. 
3234
-
3238
)
27
Robinson
C.R.
Sauer
R.T.
Optimizing the stability of single-chain proteins by linker length and composition mutagenesis
Proc. Natl Acad. Sci. USA
 , 
1998
, vol. 
95
 (pg. 
5929
-
5934
)
28
Borchelt
D.R.
Lee
M.K.
Slunt
H.S.
Guarnieri
M.
Xu
Z.S.
Wong
P.C.
Brown
R.H.
Jr
Price
D.L.
Sisodia
S.S.
Cleveland
D.W.
Superoxide dismutase 1 with mutations linked to familial amyotrophic lateral sclerosis possesses significant activity
Proc. Natl Acad. Sci. USA
 , 
1994
, vol. 
91
 (pg. 
8292
-
8296
)
29
Shinder
G.A.
Lacourse
M.C.
Minotti
S.
Durham
H.D.
Mutant Cu/Zn-superoxide dismutase proteins have altered solubility and interact with heat shock/stress proteins in models of amyotrophic lateral sclerosis
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
12791
-
12796
)
30
Matsumoto
G.
Stojanovic
A.
Holmberg
C.I.
Kim
S.
Morimoto
R.I.
Structural properties and neuronal toxicity of amyotrophic lateral sclerosis-associated Cu/Zn superoxide dismutase 1 aggregates
J. Cell. Biol.
 , 
2005
, vol. 
171
 (pg. 
75
-
85
)
31
Pasinelli
P.
Borchelt
D.R.
Houseweart
M.K.
Cleveland
D.W.
Brown
R.H.
Jr.
Caspase-1 is activated in neural cells and tissue with amyotrophic lateral sclerosis-associated mutations in copper–zinc superoxide dismutase
Proc. Natl Acad. Sci. USA
 , 
1998
, vol. 
95
 (pg. 
15763
-
15768
)
32
Krishnan
J.
Lemmens
R.
Robberecht
W.
Van Den Bosch
L.
Role of heat shock response and Hsp27 in mutant SOD1-dependent cell death
Exp. Neurol.
 , 
2006
, vol. 
200
 (pg. 
301
-
310
)
33
Dalle-Donne
I.
Aldini
G.
Carini
M.
Colombo
R.
Rossi
R.
Milzani
A.
Protein carbonylation, cellular dysfunction, and disease progression
J. Cell. Mol. Med.
 , 
2006
, vol. 
10
 (pg. 
389
-
406
)
34
Andrus
P.K.
Fleck
T.J.
Gurney
M.E.
Hall
E.D.
Protein oxidative damage in a transgenic mouse model of familial amyotrophic lateral sclerosis
J. Neurochem.
 , 
1998
, vol. 
71
 (pg. 
2041
-
2048
)
35
Bowling
A.C.
Schulz
J.B.
Brown
R.H.
Jr.
Beal
M.F.
Superoxide dismutase activity, oxidative damage, and mitochondrial energy metabolism in familial and sporadic amyotrophic lateral sclerosis
J. Neurochem.
 , 
1993
, vol. 
61
 (pg. 
2322
-
2325
)
36
Link
C.D.
Expression of human beta-amyloid peptide in transgenic Caenorhabditis elegans
Proc. Natl Acad. Sci. USA
 , 
1995
, vol. 
92
 (pg. 
9368
-
9372
)
37
Satyal
S.H.
Schmidt
E.
Kitagawa
K.
Sondheimer
N.
Lindquist
S.
Kramer
J.M.
Morimoto
R.I.
Polyglutamine aggregates alter protein folding homeostasis in Caenorhabditis elegans
Proc. Natl Acad. Sci. USA
 , 
2000
, vol. 
97
 (pg. 
5750
-
5755
)
38
Oeda
T.
Shimohama
S.
Kitagawa
N.
Kohno
R.
Imura
T.
Shibasaki
H.
Ishii
N.
Oxidative stress causes abnormal accumulation of familial amyotrophic lateral sclerosis-related mutant SOD1 in transgenic Caenorhabditis elegans
Hum. Mol. Genet.
 , 
2001
, vol. 
10
 (pg. 
2013
-
2023
)
39
Ezzi
S.A.
Urushitani
M.
Julien
J.P.
Wild-type superoxide dismutase acquires binding and toxic properties of ALS-linked mutant forms through oxidation
J. Neurochem.
 , 
2007
, vol. 
102
 (pg. 
170
-
178
)
40
Furukawa
Y.
Fu
R.
Deng
H.X.
Siddique
T.
O’Halloran
T.V.
Disulfide cross-linked protein represents a significant fraction of ALS-associated Cu, Zn-superoxide dismutase aggregates in spinal cords of model mice
Proc. Natl Acad. Sci. USA
 , 
2006
, vol. 
103
 (pg. 
7148
-
7153
)
41
Ray
S.S.
Nowak
R.J.
Strokovich
K.
Brown
R.H.
Jr.
Walz
T.
Lansbury
P.T.
Jr.
An intersubunit disulfide bond prevents in vitro aggregation of a superoxide dismutase-1 mutant linked to familial amytrophic lateral sclerosis
Biochemistry
 , 
2004
, vol. 
43
 (pg. 
4899
-
4905
)
42
Hough
M.A.
Grossmann
J.G.
Antonyuk
S.V.
Strange
R.W.
Doucette
P.A.
Rodriguez
J.A.
Whitson
L.J.
Hart
P.J.
Hayward
L.J.
Valentine
J.S.
, et al.  . 
Dimer destabilization in superoxide dismutase may result in disease-causing properties: structures of motor neuron disease mutants
Proc. Natl Acad. Sci. USA
 , 
2004
, vol. 
101
 (pg. 
5976
-
5981
)
43
Doucette
P.A.
Whitson
L.J.
Cao
X.
Schirf
V.
Demeler
B.
Valentine
J.S.
Hansen
J.C.
Hart
P.J.
Dissociation of human copper–zinc superoxide dismutase dimers using chaotrope and reductant. Insights into the molecular basis for dimer stability
J. Biol. Chem.
 , 
2004
, vol. 
279
 (pg. 
54558
-
54566
)
44
Roberts
B.R.
Tainer
J.A.
Getzoff
E.D.
Malencik
D.A.
Anderson
S.R.
Bomben
V.C.
Meyers
K.R.
Karplus
P.A.
Beckman
J.S.
Structural characterization of zinc-deficient human superoxide dismutase and implications for ALS
J. Mol. Biol.
 , 
2007
, vol. 
373
 (pg. 
877
-
890
)
45
Son
M.
Puttaparthi
K.
Kawamata
H.
Rajendran
B.
Boyer
P.J.
Manfredi
G.
Elliott
J.L.
Overexpression of CCS in G93A-SOD1 mice leads to accelerated neurological deficits with severe mitochondrial pathology
Proc. Natl Acad. Sci. USA
 , 
2007
, vol. 
104
 (pg. 
6072
-
6077
)
46
Christacos
N.C.
Fridovich-Keil
J.L.
Impact of patient mutations on heterodimer formation and function in human galactose-1-P uridylyltransferase
Mol. Genet. Metab.
 , 
2002
, vol. 
76
 (pg. 
319
-
326
)
47
Misquitta
S.A.
Colman
R.F.
Communication between the two active sites of glutathione S-transferase A1-1, probed using wild-type-mutant heterodimers
Biochemistry
 , 
2005
, vol. 
44
 (pg. 
8608
-
8619
)
48
Strange
R.W.
Antonyuk
S.
Hough
M.A.
Doucette
P.A.
Rodriguez
J.A.
Hart
P.J.
Hayward
L.J.
Valentine
J.S.
Hasnain
S.S.
The structure of holo and metal-deficient wild-type human Cu,Zn superoxide dismutase and its relevance to familial amyotrophic lateral sclerosis
J. Mol. Biol.
 , 
2003
, vol. 
328
 (pg. 
877
-
891
)
49
Ferraroni
M.
Rypniewski
W.
Wilson
K.S.
Viezzoli
M.S.
Banci
L.
Bertini
I.
Mangani
S.
The crystal structure of the monomeric human SOD mutant F50E/G51E/E133Q at atomic resolution. The enzyme mechanism revisited
J. Mol. Biol.
 , 
1999
, vol. 
288
 (pg. 
413
-
426
)
50
Schonhoff
C.M.
Matsuoka
M.
Tummala
H.
Johnson
M.A.
Estevez
A.G.
Wu
R.
Kamaid
A.
Ricart
K.C.
Hashimoto
Y.
Gaston
B.
, et al.  . 
S-nitrosothiol depletion in amyotrophic lateral sclerosis
Proc. Natl Acad. Sci. USA
 , 
2006
, vol. 
103
 (pg. 
2404
-
2409
)
51
Johnson
M.A.
Macdonald
T.L.
Mannick
J.B.
Conaway
M.R.
Gaston
B.
Accelerated s-nitrosothiol breakdown by amyotrophic lateral sclerosis mutant copper,zinc-superoxide dismutase
J. Biol. Chem.
 , 
2001
, vol. 
276
 (pg. 
39872
-
39878
)
52
Jonsson
P.A.
Ernhill
K.
Andersen
P.M.
Bergemalm
D.
Brannstrom
T.
Gredal
O.
Nilsson
P.
Marklund
S.L.
Minute quantities of misfolded mutant superoxide dismutase-1 cause amyotrophic lateral sclerosis
Brain
 , 
2004
, vol. 
127
 (pg. 
73
-
88
)
53
Gruzman
A.
Wood
W.L.
Alpert
E.
Prasad
M.D.
Miller
R.G.
Rothstein
J.D.
Bowser
R.
Hamilton
R.
Wood
T.D.
Cleveland
D.W.
, et al.  . 
Common molecular signature in SOD1 for both sporadic and familial amyotrophic lateral sclerosis
Proc. Natl Acad. Sci. USA
 , 
2007
, vol. 
104
 (pg. 
12524
-
12529
)
54
Goldsteins
G.
Keksa-Goldsteine
V.
Ahtoniemi
T.
Jaronen
M.
Arens
E.
Akerman
K.
Chan
P.H.
Koistnaho
J.
Deleterious role of superoxide dismutase in the mitochondrial intermembrane space
J. Biol. Chem.
 , 
2008
 
(Jan 2, epub ahead of print)
55
Graham
F.L.
van der Eb
A.J.
Transformation of rat cells by DNA of human adenovirus 5. Virology
1973
, vol. 
54
 (pg. 
536
-
539
)
56
Hosono
R.
Sterilization and growth inhibition of Caenorhabditis elegans by 5-fluorodeoxyuridine
Exp. Gerontol.
 , 
1978
, vol. 
13
 (pg. 
369
-
374
)