Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Variation in the vulnerability of mice expressing human superoxide dismutase 1 to prion-like seeding: a study of the influence of primary amino acid sequence

Variation in the vulnerability of mice expressing human superoxide dismutase 1 to prion-like... Misfolded forms of superoxide dismutase 1 (SOD1) with mutations associated with familial amyotrophic lateral sclerosis (fALS) exhibit prion characteristics, including the ability to act as seeds to accelerate motor neuron disease in mouse models. A key feature of infectious prion seeding is that the efficiency of transmission is governed by the primary sequence of prion protein (PrP). Isologous seeding, where the sequence of the PrP in the seed matches that of the host, is generally much more efficient than when there is a sequence mis‑match. Here, we used paradigms in which mutant SOD1 seeding homogenates were injected intraspinally in newborn mice or into the sciatic nerve of adult mice, to assess the influence of SOD1 primary sequence on seeding efficiency. We observed a spectrum of seeding efficiencies depending upon both the SOD1 expressed by mice injected with seeds and the origin of the seed preparations. Mice expressing WT human SOD1 or the disease variant G37R were resistant to isologous seeding. Mice expressing G93A SOD1 were also largely resistant to isologous seeding, with limited success in one line of mice that express at low levels. By contrast, mice expressing human G85R‑SOD1 were highly susceptible to isologous seed‑ ing but resistant to heterologous seeding by homogenates from paralyzed mice over‑ expressing mouse SOD1‑ G86R. In other seeding experiments with G85R SOD1:YFP mice, we observed that homogenates from paralyzed animals expressing the H46R or G37R variants of human SOD1 were less effective than seeds prepared from mice expressing the human G93A variant. These sequence mis‑match effects were less pronounced when we used purified recombi‑ nant SOD1 that had been fibrilized in vitro as the seeding preparation. Collectively, our findings demonstrate diversity in the abilities of ALS variants of SOD1 to initiate or sustain prion‑like propagation of misfolded conformations that produce motor neuron disease. anatomically connected pathways, leading to general- Introduction ized paralysis [1]. In a subset of cases, weakness first Amyotrophic lateral sclerosis (ALS) can present clini- appears in muscles of the head and neck before spread- cally as focal weakness in a limb, hand, or foot that ing to include the diaphragm and intercostal muscles. progressively worsens before weakness spreads along The symptoms of weakness originate from dysfunction of both upper and lower motor neurons, with dysfunc- *Correspondence: drb1@ufl.edu 1 tion of the upper neurons causing symptoms of spasticity Department of Neuroscience, Center for Translational Research in Neurodegenerative Disease (CTRND), University of Florida, Box 100159, before paralysis makes voluntary movements impossible. Gainesville, FL 32610, USA The vast majority of ALS cases have no clear etiology and Full list of author information is available at the end of the article © The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Ayers et al. acta neuropathol commun (2021) 9:92 Page 2 of 24 no obvious family history; however, up to 25% of patients most unstable variants of mutant SOD1 we have exam- have a family history of disease with a subset of these clas- ined [20, 21]. Notably, however, in pilot studies with small sified as familial based on the inheritance of rare genetic numbers of animals that express WT human SOD1, the variants [2]. Approximately 15% of familial ALS (fALS) G37R variant, or the G93A variant of SOD1, we did not and 2% of sporadic cases are associated with mutations observe the same robust accelerated onset of paralysis or in the gene encoding the antioxidant enzyme known as induction of more severe inclusion pathology [15]. superoxide dismutase 1 (SOD1) [2]. To date, more than The present study describes a broader analysis of seed - 170 missense mutations at more than 80 different amino ing across a larger panel of SOD1 variants to character- acids of this 153 amino acid protein have been associated ize the influence of primary sequence on the efficacy of with ALS (https:// alsod. uk). Though the effects of most seeding to induce motor neuron disease in SOD1 trans- of these mutations on enzymatic function have not been genic mice. We examine both isologous and heterologous characterized, studies of a random subset of mutations seeding efficiencies across multiple lines of SOD1 trans - have shown that many disease-associated mutants of genic mice. Our findings confirm that mice expressing SOD1 retain high enzymatic activity [3]. These mutants WT, G93A, or G37R human SOD1 are relatively resist- have been classified as wild-type-like (WT-like) mutants. ant to isologous prion-like seeding. Mice expressing Other mutations, however, can be highly destabilizing to the G85R and L126Z variants of fALS were susceptible the normal protein conformation and render the enzyme to both isologous and heterologous seeding by tissue inactive [3]. Work from multiple laboratories including homogenates from paralyzed mice expressing mutant ours has demonstrated that mutations associated with human SOD1. We noted that seeds prepared from mice ALS cause conformational changes in SOD1 that induce or rats that express the G37R or H46R variants of slowly the protein to misfold and self-associate into insoluble progressing ALS were among the least efficient in induc - aggregates and pathological inclusions (reviewed in [4]). ing early paralysis in G85R-SOD1:YFP mice. Notably, we Indeed, SOD1 inclusion pathology is a common feature observed that mice expressing human G85R SOD1, or of SOD1-linked ALS [5]. Whether misfolded SOD1 is G85R-SOD1:YFP, were resistant to seeding by prepara- also a common feature of sporadic ALS is less certain as tions from spinal cords of paralyzed mice over-expressing there have been contradictory reports in the literature mouse SOD1 with the same mutation. Other examples of [6–13]. Importantly, studies in cell culture models have sequence specificity in seeding implicated an amyloido - established the potential for misfolded WT SOD1 to genic element bordered by amino acids 31–38 as poten- propagate between cells, leading to the hypothesis that tially important in the propagation of misfolded SOD1 prion-like propagation of misfolded WT SOD1 could be conformations. Collectively, our studies demonstrate that involved in the progressive spread of weakness in spo- the primary sequence of SOD1 exerts significant influ - radic ALS patients [6, 14]. ence over the efficacy of the seed preparation and the We have previously demonstrated that injecting spi- susceptibility of the host-recipient in the prion-like prop- nal cord homogenate prepared from paralyzed mutant agation of misfolded SOD1 conformations. SOD1 transgenic mice can accelerate motor neuron dis- ease (MND) in transgenic mice expressing the G85R vari- Methods ant of SOD1 [15–17]. To date, most of the work in our Mice. All of the transgenic mice expressing human laboratory has used a line of mice that express the G85R SOD1 variants that were used in this study as recipients variant of human SOD1 fused to YFP (G85R-SOD1:YFP) of seeding injections have been previously described: that is largely free of disease until 20–24  months of age. G85R-SOD1:YFP and WT-SOD1:YFP mice [22], Injection of tissue homogenates, or purified protein, con - L126Z Line 45 mice [20], G37R Line 29 mice [23], PrP. taining misfolded SOD1 into the spinal cords of newborn G37R Line 110 mice [24], QV103Z hSOD1 Line D14 mice, or injecting misfolded SOD1 into the sciatic nerve mice [21], G85R Line 148 mice [25], GurWT mice of young adult mice, induces paralysis 2–15 months post- (B6SJL-Tg(SOD1)2Gur/J; stock no. 002297, Jackson injection [15–17]. Using mice that express untagged Laboratories) [26], Thy1-G93A Line T3 mice (FVB(Cg)- human G85R SOD1, Bidhendi and colleagues demon- Tg(y1- Th SOD1*G93A)T3Hgrd/J; stock no. 008230, strated accelerated onset of paralysis in mice when young Jackson Laboratories)[27], and VLE G93A mice (B6SJL- adult animals were similarly injected with preparations Tg(SOD1*G93A)1Gur/ThpaJ; stock no. 032166, Jackson derived from spinal cords of paralyzed mutant SOD1 ani- Laboratories [28]. The G85R-SOD1:YFP, WT-SOD1:YFP, mals or from human SOD1 ALS cases [18, 19]. We have and Thy1-G93A mice were maintained on the FVB/ also observed accelerated onset of paralysis in mice that NJ background. The GurWT mice were backcrossed to express untagged versions of two other mutant human B6/C3F1 hybrid mice for more than 10 generations and SOD1 truncation variants [17], which are among the maintained on this background. The VLE G93A mice A yers et al. acta neuropathol commun (2021) 9:92 Page 3 of 24 were maintained on the B6SJL hybrid background. All detect emission using the Gen5 software (v1.10.8). Incu- other lines of mutant SOD1 mice were maintained on bations were halted when maximum fluorescence was a hybrid background of C57Bl/6J and C3H/HeJ, which achieved. The presence of aggregated SOD1 was con - were the strains used in the initial generation of the mice. firmed by filter trap assay and electron microscopy as For identification of genotype, DNA was extracted from previously described [17]. mouse tail biopsies and analyzed by PCR as previously Seeding Experiments. Newborn mice were injected described [20, 24]. intraspinally (ISP) as previously described [15]. Briefly, Spinal tissues were harvested from transgenic animals P0 neonatal pups were placed in aluminum foil and sur- in our in-house colonies that became paralyzed. In addi- rounded with ice until movement ceased and skin tone tion to the mice listed above, the source for tissues from became cyanotic (5–10  min). Using a 10  μl syringe MoG86R Line M1 was FVB-Tg(SOD1*G86R)M1Jwg/J equipped with a 1-inch, 30-gauge needle with 30° bevel, (stock no. 005110, Jackson Laboratories, Bar Harbor ME) the needle was inserted through the skin at midline, [29]. The original source for GurG93A mice was B6SJL- about 5  mm from the base of the tail, and then into the Tg(SOD1*G93A)1Gur/J (stock no. 002726 Jackson Labo- vertebral column before 1 μl of the inoculum was slowly ratories) [26]. The GurG93A mice were backcrossed to injected. For cerebral ventricular injections, 2 μl of inoc- B6/C3F1 hybrid mice for more than 10 generations and ulum was slowly injected into each cerebral ventricle by maintained on this background in our colony. Spinal tis- inserting the needle approximately 2  mm into the skull, sues from paralyzed H46R rats [30] were a kind gift of Dr. penetrating the skull near bregma. For bilateral muscle Christine Vande Velde (University of Montreal, CHUM injections, 2  μl of inoculum was slowly injected into the Research Center, Montreal Canada). Spinal tissues from lower hindlimb targeting the gastrocnemius muscles. paralyzed G85R mice (Line 148) [25] were a kind gift of The accuracy of these techniques of injection was veri - Dr. Don Cleveland (University of California San Diego, fied in separate cohorts of newborn mice injected with San Diego CA). PBS containing 1% Evans blue dye [32]. After injections, All animals were housed one to five to a cage and main - pups were allowed to completely recover on a warming tained on ad libitum food and water with a 14 h light and blanket and then returned to the home cage where they 10 h dark cycle. were monitored to ensure full mobility and no signs of Preparation of inoculum. Spinal cord tissues were impairment. homogenized in PBS creating a 10% homogenate (w/v), Sciatic nerve injections were performed as previously containing 1:100 v/v protease inhibitor cocktail (Sigma, described [16]. Prior to the injection, mice were injected St. Louis, MO) as previously described [15]. Tissues were subcutaneously with 2  mg/kg Meloxicam (Norbrook, disrupted by sonication 4 times for 20 s each, with cool- Overland Park, KS, USA) to relieve pain and the injection ing on ice between bouts of sonication. Homogenates site was shaved and sterilized. Mice were anesthetized were then clarified by a low-speed spin at ~ 800 × g for with isoflurane and a small incision was then made in the 10  min and the supernatants were aliquoted and placed skin of the hindlimb before the sciatic nerve was exposed at -80 °C. In studies involving human tissues, we concen- at the popliteal fossa. A 30-gauge needle containing the trated 200 µl of the clarified homogenates by centrifuga - inoculum was inserted in the sciatic nerve and recipro- tion in an AirFuge at maximum speed for 20  min. The cated 10 times, which has been shown to greatly enhance resulting pellet was resuspended in 40 µl of PBS with the the efficiency of prion transport to the spinal cord [33]. protease inhibitor cocktail and solubilized by pulse soni- Two microliters of homogenate were injected under the cation as described above. perineurium of the sciatic nerve and the incision was Recombinant SOD1 purification and fibrillization. then closed with stainless steel clips and cleaned. Follow- Recombinant hSOD1 proteins were expressed and puri- ing surgery, 2  mg/kg Meloxicam was administered at 24 fied as previously described [31]. Fibrillar aggregates of and 48 h post-surgery. purified SOD1 were generated in 200  µl solutions con - Tissue collection. Mice were anesthetized with isoflu - taining 50  μM of protein in 20  mM potassium phos- rane and perfused transcardially with 20  ml of PBS. The phate, pH 7.2 with the addition of 10  mM TCEP. The spinal cord and brain were immediately removed and protein solutions were incubated in a 96-well plate with the brains were bisected sagittally with one hemisphere the addition of a Teflon ball (1/8-in diameter) at 37  °C drop fixed in 4% paraformaldehyde in PBS for 24–48  h with constant agitation in a Synergy HT plate reader at 4  °C and the other flash frozen on dry ice. The spinal (BIO-TEK, Winooski, VT). Parallel wells containing the columns were removed and cut into 4 sections. Cervi- same components with 4  μM Thioflavin T were moni - cal and lumbar segments were de-roofed and drop fixed tored by fluorescence measurements every 15 min using in 4% paraformaldehyde in PBS for 24–48 h at 4 °C. The a λex = 440/30 filter to excite and a λem = 485/20 filter to other segments were dissected and flash frozen on dry Ayers et al. acta neuropathol commun (2021) 9:92 Page 4 of 24 ice. Frozen tissues were stored at −80 °C. Fixed tissues appeared as dark structures of various sizes, including were either embedded in paraffin for sectioning at 7  µm small neuropil puncta, fiber-like structures in the neuro - or immersed in 30% sucrose in PBS, mounted in OCT pil, and accumulations of puncta within cell bodies. media (Sakura, The Netherlands) and sectioned to 30 μm using a cryostat. The sections were placed in a dish con - Results taining anti-freeze solution (100  mM sodium acetate, We have demonstrated previously that intraspinal injec- 250 mM polyvinylpyrrolidone, 40% ethylene glycol) at pH tion of spinal homogenates from paralyzed mutant SOD1 6.5, and stored at 4 °C. Sections were then mounted onto mice into newborn G85R-SOD1:YFP mice accelerates slides, air-dried overnight, and cover-slipped in mount- the onset of paralysis and intraspinal SOD1 inclusion ing media containing DAPI (Vector, Burlingame, CA). pathology [15, 17]. Importantly, we have previously dem- Neuropathology. For imaging YFP fluorescence in par - onstrated that paralysis in G85R-SOD1:YFP mice is not affin embedded tissues or from cryostat sections, images induced by injection of PBS or spinal tissue homogen- were captured by epifluorescence on an Olympus BX60 ates from various types of controls (Additional file  2: microscope. Table  S1) [17]. The controls tested include homogenates For immunohistochemistry and histochemical stain- from non-transgenic mice, and from transgenic mice that ing, we examined paraffin embedded tissues. For immu - express mutant human tau or mutant human α-Synuclein nostaining with C4F6 antibody (Medimabs, Montreal, (αSyn), and which develop motor phenotypes. Other Quebec, Canada), deparaffinized sections were incubated controls include spinal homogenates from young asymp- in 95% formic acid for 10  min and then washed in PBS. tomatic G85R-SOD1:YFP mice. Additionally, nontrans- Sections were incubated in 0.3% H O in PBS for 20 min genic (NTg) littermates injected with homogenates 2 2 and then incubated in PBS-T with 3% normal goat serum from paralyzed mutant SOD1 do not develop paralysis for 1  h before the primary antibody C4F6 was added at or exhibit evidence of inclusion pathology (Additional a dilution of 1:500 and incubated at 4  °C overnight. The file  1: Data File 1 – NTg data tab). Collectively, these pre- sections were then incubated with a biotinylated sec- vious studies have established the specificity with which ondary anti-mouse antibody (Vector Laboratories, Burl- spinal homogenates from paralyzed mutant SOD1 mice ingame, CA) diluted 1:500 in PBS-T with 3% normal goat induce an accelerated paralysis, and inclusion pathology, serum followed by incubation with the ABC-horserad- when injected into the spinal cords of newborn G85R- ish peroxidase staining kit (Vector Laboratories). Sec- SOD1:YFP mice. tions were developed using the DAB staining kit (KPL, Gaithersburg, MD) and counterstained with hema- Comparative analysis of seeding efficiency by spinal toxylin. Images were captured using an Olympus BX60 homogenates from paralyzed G37R and G93A SOD1 mice microscope. In the present study, we have focused on examining the SOD1 inclusions were also visualized by Campbell- role of the primary sequence of the SOD1 seed and the Switzer silver staining using a protocol provided by Dr. SOD1 gene expressed by the recipient in the efficiency Robert Switzer (NeuroScience Associates, Knoxville, of prion-like seeding. We have previously reported that TN; Campbell S, Switzer R, Martin T (1987) Alzhei- spinal homogenates from paralyzed mice expressing the mer’s plaques and tangles: a controlled and enhanced sil- G37R variant, which is associated with disease durations ver staining method. Soc Neurosci Abst 13:678) [34]. A in humans that average 18.7  years [35], can seed early subset of the silver-stained sections was counterstained paralysis and pathology in G85R-SOD1:YFP mice [15, with hematoxylin and all slides were cover slipped using 17]. Here, we have reanalyzed previously published data, ™ ™ Thermo Scientific Cytoseal 60 mounting medium. along with additional new data, to assess the variability Images were obtained using an Olympus BX60 micro- and efficiency of G37R-SOD1 seeds in G85R-SOD1:YFP scope or an Aperio Scanscope XT image scanner (Aperio, mice. Overall, homogenates prepared from paralyzed Vista, CA, USA). Images captured by the scanner were G37R mice induced early paralysis in only 4 of the 15 digitally cropped to generate final images. injected G85R-SOD1:YFP mice (Fig. 1a, black circles). All The severity of inclusion pathology revealed by direct the mice that developed paralysis also developed inclu- visualization of fluorescence or after silver staining was sion pathology, which appeared as fluorescent puncta in scored by a blinded observer (Additional file  1: data file the neuropil with variable levels of accumulation in cell 1). In the G85R-SOD1:YFP mice inclusions appeared as bodies (Fig. 1b, c). Additionally, there were accumulations fluorescent puncta in the neuropil, punctate or fibrillar within processes that appeared as fiber-like structures accumulations in the cell body, or as fiber-like structures in the neuropil (Fig.  1c–g). The initial description of the in the neuropil due to accumulations within processes G85R-SOD1:YFP mice demonstrated that this transgene (axons or dendrites). In silver stained sections, inclusions is highly expressed in neurons, with minimal evidence A yers et al. acta neuropathol commun (2021) 9:92 Page 5 of 24 of astrocytic expression [22]. Therefore, it seems likely As compared to seeding with spinal homogenates from that most of the inclusion pathology occurs in neurons paralyzed G37R mice, the efficiency of seeding paralysis or neuronal processes. Notably, there were two asymp- by homogenates from paralyzed G93A mice was more tomatic animals that were euthanized at 19  months that consistent. In experiments using 4 independently gen- exhibited extensive inclusion pathology (Fig.  1a gray erated homogenates, we observed paralysis in at least circles; Fig.  1d). Curiously, all 4 of the G85R-SOD1:YFP half of the injected G85R-SOD1:YFP recipients for each mice that developed paralysis after seeding were from a cohort (Fig. 2a, black circles). Mice that developed paral- single litter. In the subset of mice that developed paraly- ysis tended to do so before the age of 12 months (Fig. 2a). sis, the average age to paralysis was 12.2 months, but the In three of the 4 experiments, the average age to paraly- range was from 6.5 to 18 months (Additional file  1: Data sis was 5.7, 4.3, and 4.4 mo post-injection (Additional File 1). file  1: Data File 1). In cohort #4, three of the 7 mice that In prior studies, we have demonstrated that the effi - were injected developed paralysis before 16  months of ciency of seeding improved when we passaged G37R age when we terminated the experiment (average age to seeds from these initial G85R-SOD1:YFP mice to naïve paralysis was 10 mo post injection) (Fig.  2a, G93A Sp recipients [17]. To assess the variability of second pas- Cord #4). Of the 25 total mice that were injected with sage G37R seeds, we compared second passage data G93A seeds, 11 lived to 12  months post-injection with- for 3 different isolates. Second passage of seeds from out developing paralysis (Fig.  2a; Additional file  1: Data the 19-month-old asymptomatic animal that exhibited File 1). All mice that developed paralysis exhibited inclu- inclusion pathology (see Fig.  1d) induced paralysis in sion pathology within the spinal cord (Fig.  2b; Addi- 3 of 5 injected G85R-SOD1:YFP mice at an average age tional file  1: Data File 1). Interestingly, as was the case of 12  months post-injection (Fig.  1a). Second passage in G37R first-passage mice, we observed several cases in of seeds from an 18-month-old symptomatic animal which asymptomatic mice were euthanized at advanced that exhibited inclusion pathology (Fig.  1c) also induced ages (> 12  months) and found to exhibit extensive inclu- paralysis in 4 of 7 injected recipients at an average age sion pathology (Fig.  2c; Additional file  1: Data File 1). of 17  months post-injection (Fig.  1a). Second passage of As previously reported [15, 17], second passage of seeds G85R−SOD1:YFP seeds from a 6.5 month old symptomatic animal induced from P1-G93A recipients that became para- paralysis in 6 of 7 injected recipients at an average age of lyzed produced more efficient induction of paralysis at 6.2  months post-injection (Fig.  1a) [17]. Passage of this earlier ages (Fig.  2a) with inclusion pathology (Fig.  2d). isolate for a third time produced paralysis in 7 of 8 recipi- The average age to paralysis for second passage G93A ents at an average of 6.2 months post-injection (Fig.  1a). seeds was ~ 3  months post-injection (Additional file  1: All the paralyzed mice in second or third passage experi- Data File 1). These findings are consistent with the idea ments exhibited inclusion pathology to varying degrees that the G93A variant of SOD1 misfolds into a confor- (Fig.  1e–h). The level of inclusion pathology seen in the mation that produces relatively efficient seeds, imprint - 19-month-old P1 animals that were used to prepare ing conformations to G85R-SOD1:YFP that can produce seeds for second passage (Fig.  1d) was not obviously dif- relatively short incubation periods as seeds in second ferent from that of similarly aged paralyzed mice (com- passage. pare Fig.  1b–d). Overall, the data are consistent with the idea that the efficiency of seeding by homogenates con - Low seeding efficiency by spinal homogenates taining the G37R variant of misfolded SOD1 is relatively from paralyzed H46R‑SOD1 transgenic rats low and that seeds imprinted by the G37R conformation The average duration of disease in individuals with the can exhibit relatively long incubation periods in second H46R mutation is 17.0  years [35]). Spinal cords from passage. paralyzed rats that were heterozygous or homozygous (See figure on next page.) Fig. 1 Spinal homogenates from paralyzed G37R SOD1 mice seed G85R‑SOD1:YFP mice inefficiently with variable incubation periods. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates. Scatter plots in this figure G85R− and figures that follow were generated in GraphPad Prism v9. TgG37R denotes spinal homogenate from paralyzed G37R‑Line 29 mice. P1‑ G37R SOD1:YFP G85R−SOD1:YFP denotes inoculum from a paralyzed first ‑passage G85R‑SOD1:YFP animal. P2‑ G37R denotes inoculum prepared from a paralyzed second‑passage G85R‑SOD1:YFP animal. The dashed lines with arrows mark animals from the initial passage that were used to prepare inoculum for second passage. b–h Representative images of inclusion pathology in cryostat sections that were induced in the spinal cord of G85R‑SOD1:YFP mice that had been injected intraspinally (ISP) with spinal homogenates from G37R mice with subsequent passages (images representative of 2–3 sections per mouse). The arrows identify pathological specimens that are related due to passage. b–d Pathology specimens from first passage rd animals. e–g Representative pathology specimens from 2nd passage animals. h A representative image of pathology specimens from 3 passage animals. Raw data provided in Additional file 1: Data File 1. Scale bars = 50 µm Ayers et al. acta neuropathol commun (2021) 9:92 Page 6 of 24 for the transgene (described in [30]) were homog- developed paralysis and the experiment was terminated enized and injected into the spinal cords of newborn at 15–16  months post-injection to examine pathol- G85R-SOD1:YFP mice. None of the injected animals ogy (Fig.  3a), finding that most of these asymptomatic A yers et al. acta neuropathol commun (2021) 9:92 Page 7 of 24 Fig. 2 Variability in seeding efficiency of spinal homogenates from paralyzed G93A SOD1 mice in G85R‑SOD1:YFP mice. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates. TgG93A denotes spinal homogenate G85R−SOD1:YFP from paralyzed GurG93A mice. P1‑ G93A denotes inoculum from a paralyzed first ‑passage G85R‑SOD1:YFP animal. b–d Representative images of inclusion pathology in cryostat sections induced in the spinal cord of G85R‑SOD1:YFP mice that had been injected with spinal homogenates from G93A mice with subsequent passages (images representative of 2–3 sections per mouse). Raw data provided in Additional file 1: Data File 1. Scale bars = 50 µm mice injected with homogenate from the homozygous paralysis in G85R-SOD1:YFP mice rather inefficiently, H46R rats had developed inclusion pathology (Fig. 3b). despite a capability to induce inclusion pathology. To assess whether a higher dose of H46R seeds may be more effective, we concentrated spinal homogenates Seeding G85R‑SOD1:YFP mice with recombinant SOD1 from paralyzed homozygous H46R rats by fivefold and fibrils injected them into newborn G85R-SOD1:YFP mice. In prior studies, we have demonstrated the induction of One animal of the 5 that were injected developed fore- paralysis and inclusion pathology in G85R-SOD1:YFP limb weakness at 7.9  months of age, and this animal mice by the injection of fibrils produced by recombi - exhibited inclusion pathology (Fig.  3c). The remaining nant human SOD1 (WT or G93A variants) [17, 36]. To 4 animals from this cohort were asymptomatic when expand the number of different sequence variants we we terminated the experiment at 12  months of age, could assess in seeding, we produced recombinant SOD1 and lacked evidence of inclusion pathology (Fig.  3a; encoding eight different fALS mutations and fibrilized Additional file  1: Data File 1). These data suggest these proteins in  vitro to produce inoculum for injec- that homogenates from paralyzed H46R rats induce tion (Table  1). To confirm the formation of fibrils, we analyzed the samples both by electron microscopy (EM) Ayers et al. acta neuropathol commun (2021) 9:92 Page 8 of 24 Fig. 3 Spinal homogenates from paralyzed H46R rats seed G85R‑SOD1:YFP mice inefficiently. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates from paralyzed H46R rats. b, c Representative images of inclusion pathology seen in paraffin sections from the spinal cord of G85R‑SOD1:YFP mice that had been injected with spinal homogenates from heterozygous and homozygous paralyzed H46R rats, with a fivefold concentrated inoculum prepared from homogenate from a paralyzed homozygous rat (images representative of 2–3 sections per mouse). Raw data provided in Additional file 1: Data File 1. Scale bar = 100 µm (b) or 50 µm (c) (Fig.  4a–i; Additional file  2: Fig. S1) and by using a cel- are associated with rapidly progressing disease, we only lulose filter trap assay (Table  1). Somewhat surprisingly, observed debris that was similar to what we observed over multiple assays, we were unable to detect the gen- in the control blank (Fig.  4a, f, and i; Table  1). Based on eration of fibrillar aggregates by 2 of the 7 mutant SOD1 prior studies of the H46R variant in vitro [37, 38], we had proteins examined (Table 1; G85R mutant was not exam- expected poor fibrilization of this variant in our experi - ined by EM). For the G41S and D101N mutants, which ments. However, we were able to produce fibrilized H46R A yers et al. acta neuropathol commun (2021) 9:92 Page 9 of 24 Table 1 Summary of data for recombinant SOD1 protein H46R, and other seeds, may further reveal unique attrib- seeding experiments utes of this variant. Mice that displayed paralysis from injection of these SOD1 protein Avg survival Fibrils visible Induced time in years by EM pathology in seeds were found to produce inclusion pathology (Addi- G85R‑ tional file  2: Fig. S2; Additional file  1: Data File 1). The SOD1:YFP inclusion pathology induced by recombinant WT-SOD1 slice culture fibrils was distributed in cell bodies and neuropil (Addi - WT N/A Yes Yes tional file  2: Fig.  2a arrows; also see [17]); whereas, for A4V 1.2 ± 0.9 Yes Yes all of the mutants, the dominant pathology was located G37R 18.7 ± 11.4 Yes Yes in the neuropil, and appeared as fiber-like and punctate G41S 0.9 ± 0.2 No Yes structures (Additional file  2: Fig.  2b-i). Interestingly, all H46R 17.0 ± 7 Yes Yes but one of the 17-month-old asymptomatic mice injected G85R 6.0 ± 4.5 N.D Yes with recombinant H46R fibrils were found to have high G93C 13.0 ± 0.4 Yes Yes levels of inclusion pathology (Fig.  4j; Additional file  2: E100K 12.0 ± 4.1 Yes Yes Fig. S2e). Collectively, these findings demonstrate that D101N 2.4 ± 0.9 No Yes purified recombinant SOD1 with mutations associated Data originally reported in[35] with fALS can produce prion-like seeds that can acceler- Inclusive of data originally reported in[17] ate paralytic disease in the G85R-SOD1:YFP model. The efficiency of seeding did not obviously relate to whether the injected recombinant protein formed large visible human SOD (Fig. 4e). Prior studies have noted that SOD1 fibrils, or whether the mutant was associated with rap - fibrilization can be somewhat stochastic and in competi - idly or slowly progressing disease. While the behavior of tion with the formation of amorphous aggregates [39]. the recombinant H46R-SOD1 seeds was similar to that Due to the uncertainty surrounding the SOD1 species of spinal homogenates from H46R-SOD1 rats, the effi - responsible for prion-like seeding in G85R-SOD1:YFP ciency of seeding with recombinant G37R-SOD1 seeds mice, we decided to test all 8 of the SOD1 mutants, using was relatively high with incubation periods that were protein that had been treated in a manner to produce similar to variants associated with rapidly progressing fibrilization. Initially, we used a spinal cord slice culture disease (Fig.  4j). Apart from the H46R variant, we did model from the G85R-SOD1:YFP mice to assess patho- not observe an obvious distinction in the performance of logical seeding activity [17]. We treated the slice cultures recombinant SOD1 seeds from slowly progressing vari- with 1  μl of each of the proteins (50  μM) and incubated ants (G37R, G93C, E100K) as compared to seeds from the sections for one month while monitoring for the rapidly progressing variants (A4V, G41S, D101N). induction of SOD1-YFP inclusion pathology. All eight of the mutant SOD1 protein preparations were observed to Low seeding efficiency with spinal homogenates induce the misfolding of G85R-SOD1:YFP in the slice cul- from older GurWT mice ture model (Table  1; primary data not shown). We then Previously, we reported that injection of spinal injected the mutant recombinant SOD1 preparations homogenates from older GurWT mice into P0 G85R- into the spinal cords of newborn G85R-SOD1:YFP mice SOD1:YFP mice produced paralysis with inclusion to assess whether we could induce accelerated paralysis. pathology in one of the 3 injected animals [15]. One of All preparations injected were capable of inducing early these 3 injected animals was euthanized at 20  months disease with the age to paralysis ranging between 8- and of age with no signs of paralysis or pathology (Fig.  5a, 16-months post-injection (Fig.  4j). The efficacy of induc - b) and the other was found dead at 19  months of age ing paralysis across the mutants was similar to WT SOD1 with no prior signs of paresis. To further examine fibrils with the notable exception in that most of the the seeding activity of WT-SOD1 in spinal cords of mice injected with fibrilized recombinant H46R-SOD1 aged GurWT mice, we injected 9 G85R-SOD1:YFP reached a pre-determined aging endpoint of 17  months mice. One of these animals developed abnormali- without developing symptoms (Fig.  4j). Further study in ties at 12  months of age that were described as asym- second and third passage experiments with spinal cords metrical weakness with swelling in the weak limb and of G85R-SOD1:YFP mice seeded with recombinant partial paralysis. Despite the suggestion that ALS-like Ayers et al. acta neuropathol commun (2021) 9:92 Page 10 of 24 Fig. 4 Analysis of the seeding efficiency of recombinant SOD1 fibrils in G85R‑SOD1:YFP mice. a–i Representative images of fibrils formed by recombinant SOD1 aggregated in vitro. j Scatter plot of the ages at which the G85R‑SOD1:YFP mice developed paralysis or were euthanized after intraspinal injection of recombinant SOD1 fibrils (P0 injection). Raw data provided in Additional file 1: Data File 1. All scale bars (below each panel) represent 500 nm disease was induced, this animal exhibited only sparse puncta (Additional file  1: data file 1). To follow up on fluorescent puncta (Fig.  5c). Another animal developed the one G85R-SOD1:YFP animal that developed paral- partial paralysis at 13 months of age with signs of with ysis by seeding with GurWT spinal homogenates, we injuries to the hindlimbs from fighting or self-mutila - attempted to passage seeds from this animal into naïve G85R−SOD1:YFP tion. This animal also showed only sparse fluorescent G85R-SOD1:YFP (P1-WT ). In a small A yers et al. acta neuropathol commun (2021) 9:92 Page 11 of 24 Fig. 5 Spinal homogenates from aged GurWT SOD1 mice seed G85R‑SOD1:YFP mice inefficiently. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates. TgGurWT denotes spinal homogenate from G85R−SOD1:YFP aged GurWT mice. P1‑ GurWT denotes inoculum from a paralyzed first ‑passage G85R‑SOD1:YFP animal that has been previously described [15]. The data graphed here include 3 first ‑passage animals described in [15]. b–d Representative images of inclusion pathology seen in cryostat sections from the spinal cord of G85R‑SOD1:YFP mice that had been injected with spinal homogenates from aged GurWT mice or with homogenate from the one first ‑passage animal that developed paralysis (images representative of 2–3 sections per mouse). Raw data provided in Additional file 1: Data File 1. Scale bars = 20 µm cohort of three mice, none developed paralysis, and exposure of the G85R-SOD1:YFP protein to a misfolded when these animals were euthanized at 16.4  months untagged mutant SOD1. We have previously observed of age we once again observed only sparse fluorescent that co-expression of human G93A-SOD1 with G85R- puncta (Fig.  5d). These data indicate that the efficiency SOD1:YFP produced abundant YFP inclusion pathology of seeding paralysis, or inclusion pathology, in G85R- [15]. This outcome presaged our observation that injec - SOD1:YFP mice by injecting spinal cords of older tion of spinal homogenates from paralyzed G93A mice GurWT mice is relatively low. induced an early paralysis with inclusion pathology in G85R-SOD1:YFP mice [15]. We were therefore interested Comparison of prion‑like propagation to transgenic to examine other mutants in co-expression models. Con- co‑expression of WT and mutant SOD1 sistent with L126Z-SOD1 seeding of G85R-SOD1:YFP with G85R‑SOD1:YFP described previously [17], we found that co-expression The co-expression of mutant SOD1 with G85R- of the L126Z-SOD1 variant (Line 45) at levels sufficient SOD1:YFP presents a scenario of sustained high-level to cause disease was able to induce the aggregation of Ayers et al. acta neuropathol commun (2021) 9:92 Page 12 of 24 Fig. 6 Induction of G85R‑SOD1:YFP inclusion pathology by co ‑ expression of untagged mutant SOD1. a–c Kaplan–Meier survival curves and representative pathologic images resulting from G85R‑SOD1:YFP mice crossed with mice expressing untagged L126Z (Line 45) (L126Z mice n = 3, G85R‑ YFP mice n = 5, L45xG85R‑ YFP mice n = 6); untagged G37R (Line 29) (G37R mice n = 6, G85R‑ YFP mice n = 10, L29xG85R‑ YFP mice n = 6), or untagged WT human SOD1 (GurWT n = 14, G85R‑ YFP n = 5, GurWTxG85R‑ YFP n = 8), respectively. Survival plots in this figure and figures that follow were generated in GraphPad Prism v9. Representative images of inclusion pathology seen in cryostat sections from the spinal cord of paralyzed bigenic mice created in each crossing experiment (images representative of 2–3 sections per mouse). Scale bars = 50 µm G85R-SOD1:YFP (Fig.  6). Notably, the age to paralysis induced pathology was a mixture of fibrillar and punctate was modestly earlier in L126Z/G85R-SOD1:YFP mice structures. Our findings from these crossing experiments relative to littermates that expressed the L126Z-SOD1 would suggest that seeds prepared from older GurWT alone (Fig.  6a). In the bigenic mice, we observed abun- mice should have been more effective than observed. dant fibrillar YFP containing inclusion pathology that resembled pathology induced in G85R-SOD1:YFP mice Comparison of injection routes by seeding with spinal homogenates from paralyzed In a previous study, we have shown that injection of tis- L126Z-SOD1 mice [17]. Bigenic mice co-expressing sue homogenates from paralyzed mutant SOD1 mice into G37R-SOD1 (Line 29) and G85R-SOD1:YFP developed sciatic nerve directly can seed earlier onset of paralysis disease much earlier than mice expressing G37R-SOD1 with intraspinal inclusion pathology [16]. To determine alone (Fig. 6b). Paralyzed bigenic G37R/G85R-SOD1:YFP whether we would achieve the same efficiency of seed - mice exhibited robust fluorescent inclusion pathology ing by intramuscular injection, we injected the hindlimb (Fig. 6b). Bigenic mice co-expressing WT-SOD1 (GurWT muscles of newborn G85R-SOD1:YFP mice with tissue line) with G85R-SOD1:YFP also developed paralysis at homogenates from paralyzed G93A mice. In this cohort, early ages with robust inclusion pathology (Fig.  6c). The we observed that all five mice lived to approximately A yers et al. acta neuropathol commun (2021) 9:92 Page 13 of 24 Fig. 7 Inefficient seeding of human G85R‑SOD1 mice by misfolded murine G86R‑SOD1. a Kaplan–Meier survival curves resulting from untagged G85R SOD1 mice injected with spinal homogenates from paralyzed G85R SOD1 mice, paralyzed murine G86R SOD1 mice, NTg mice, or uninjected mice. b Representative images of inclusion pathology seen by silver staining of paraffin sections from paralyzed recipient G85R mice (images representative of 2–3 sections per mouse). Scale bars = 20 µm 20  months of age without developing paralysis (Addi- newborn G85R-SOD1:YFP mice. Using spinal homoge- tional file  2: Fig. S3). Notably, however, one of these mice nates from G85R:YFP mice that had been injected with G85R−YFP developed inclusion pathology and hence there may have G93A homogenate (P1-G93A mice) as seeds, been a low level of seed uptake at nerve terminals, result- we examined the efficacy of the ICV route in inducing ing in inefficient seeding and reduced penetrance (Addi - paralysis. In a cohort of 6 animals, we noted that the tional file 2: Fig. S3). age of paralytic onset was similar to what was observed We next examined the relative efficacy of intracer - in ISP injected mice (range of 4–10  months) (Addi- ebral ventricular (ICV) injection of SOD1 seeds into tional file  2: Fig. S3). All paralyzed animals exhibited Ayers et al. acta neuropathol commun (2021) 9:92 Page 14 of 24 abundant inclusion pathology. Interestingly, in the ICV mutants (L126Z and QV103Z) show accelerated paralysis seeding paradigm, all the mice first presented with fore - after isologous seeding [17]. Both of these lines of mice limb weakness, suggesting a rostral to caudal spread of will develop paralysis on their own, with the L126Z mice the disease-causing conformation. developing disease at 7–9 months of age and the QV103Z mice developing paralysis at > 14  months of age [20, 21]. Isologous and heterologous seeding in untagged G85R In subsequent studies of seeding in these lines of mice, SOD1 mice we have now found an unexpected distinction in seeding As previously reported [19], mice expressing untagged activity. L126Z-SOD1 mice responded to seeding with G85R-SOD1 were susceptible to seeding by spinal homogenates from paralyzed L126Z or QV103Z mice homogenates prepared from paralyzed G85R-SOD1 (Fig.  8a), with the paralyzed mice showing varied levels mice. The line of G85R-SOD1 mice used here develops of silver positive inclusion structures (Fig.  8b–d; Addi- paralysis between 11 and 14 months of age [25] (Fig. 7a). tional file  2: Fig. S6). By contrast, QV103Z mice were not Although there was some variability in the age to paral- responsive to seeding by homogenates from L126Z mice ysis in the seeded G85R mice (range 2.1–6  months), all (Fig.  8e). In silver stains of the asymptomatic QV103Z 11 of the injected G85R mice developed paralysis early mice we observed occasional red blood cells (Fig.  8f, (Fig. 7a). By contrast, spinal homogenates from mice that g), whereas the paralyzed mice seeded with QV103Z express a version of murine Sod1 with the G85R muta- homogenates exhibited obvious inclusion pathology tion (originally noted as G86R by codon numbering (Fig.  8h; Additional file  2: Fig. S7). These studies suggest [29]), did not induce earlier paralysis in human G85R- that QV103Z seeds can propagate misfolded conforma- SOD1 mice (Fig.  7a). G85R-SOD1 mice injected with tions to L126Z host SOD1, but when the QV103Z-SOD1 spinal cord homogenates from paralyzed MoG86R-Sod1 is the host, the conformation carried by L126Z seeds can- mice developed paralysis in the same time frame as mice not propagate. injected with nontransgenic (NTg) homogenates or PBS (Fig. 7a). The homogenate from paralyzed MoG86R-Sod1 Limited prion‑like seeding in mice expressing WT‑like mice was also ineffective when these seed preparations mutants of SOD1 were injected into G85R-SOD1:YFP mice (Additional We have previously attempted to seed earlier onset dis- file 1: Data File 1; Additional file 2: Fig. S4). ease in Gur1-G93A SOD1 mice G37R-SOD1 mice with- To visualize inclusion pathology in these seeded out success (Additional file  2: Table  S2; data from [15]). untagged G85R mice, we used two methods. The inclu - The GurG93A-SOD1 mice have been reported to have sion pathology in the G85R mice seeded with spinal inherently high levels of misfolded SOD1 from very homogenates from paralyzed G85R mice was immu- young ages [41], and thus it could be argued that these noreactive with C4F6 SOD1 antibody as was pathology mice are not responsive to seeding because the patho- observed in older G85R mice seeded with NTg homoge- genic misfolding of SOD1 is already at saturation. Thus, nates or vehicle control, PBS (Additional file  2: Fig. S5). we turned to lines of mice that express G93A-SOD1 Due to variation in the performance of this antibody, we at much lower levels; Thy1-G93A mice (line T3) [27] switched to a new method of detecting inclusion pathol- and a substrain of the G93A mice termed the very low ogy in untagged SOD1 mice termed Campbell-Switzer expressing (VLE) G93A mice [28]. Mice from the T3 line silver stain [34], which we have recently found to be spe- that are heterozygous for the transgene do not develop cific for mutant SOD1 inclusions [40]. As was seen in paralysis by 20  months of age; whereas, homozygous sections stained by C4F6, silver staining revealed a mix- mice become paralyzed between 15 and 24 months of age ture of neuropil and cell-body inclusions in all paralyzed [27]. Heterozygous Thy-1 G93A mice injected with spinal G85R mice, regardless of age at which paralysis devel- homogenates from paralyzed GurG93A or G37R-SOD1 oped or injection inoculum (Fig.  7b). Importantly, mice mice did not develop paralysis or pathology (Table  2; that became paralyzed early as a result of seeding showed Additional file  2: Fig. S8). The VLE-G93A mice do not abundant inclusion pathology. These findings indicate develop paralysis until > 22 months of age [28]. For these that sequence differences between mouse and human experiments, the source of the seed was spinal homoge- SOD1 inhibit the ability of misfolded G85R mouse Sod1 nates from paralyzed GurG93A-SOD1 (FVB/NJ strain). to seed aggregation of human G85R-SOD1. We used both the newborn intraspinal route of inocu- lation and the sciatic nerve injection route. For the first Isologous and heterologous seeding in the L126Z‑SOD1 time in mice expressing the G93A variant, we observed and QV103Z‑SOD1 mice seeding to produce an earlier onset to paralysis in 6 of In prior studies, we have reported that mice express- the mice injected by the two routes (Table 2). Silver stain- ing untagged mutant SOD1 variants that are truncation ing confirmed that only the paralyzed mice had inclusion A yers et al. acta neuropathol commun (2021) 9:92 Page 15 of 24 Fig. 8 Comparison of the seeding efficiency of L126Z and QV103Z mice in isologous and heterologous seeding. a–d Kaplan–Meier survival curves and representative pathology resulting from untagged L126Z SOD1 mice injected with spinal homogenates from paralyzed L126Z or QV103Z mice, and NTg mice. e–h Kaplan–Meier survival curves and representative pathology resulting from untagged QV103Z SOD1 mice injected with spinal homogenates from paralyzed L126Z or QV103Z mice, and NTg mice. For each animal 2–3 sections were examined to identify representative images. Paraffin sections of spinal cord from each animal was stained by Campbell‑Switzer method as described in Methods . Scale bars = 50 µm pathology (Additional file  2: Fig. S9). Thus, we demon - G37R-SOD1 – Line 110) [24]. Mice from Line 110 that strate that VLE-G93A mice are modestly susceptible to are heterozygous for the transgene develop paralysis at isologous seeding with G93A spinal cord homogenates. 20 + months of age, whereas homozygous mice develop We also extended our investigations in seeding dis- paralysis between 8 and 11 months of age [24]. In a prior ease in mice that express the G37R variant under the study, we reported that we were not able to accelerate transcriptional control of the mouse PrP vector (PrP. disease in this model by isologous seeding using either an Ayers et al. acta neuropathol commun (2021) 9:92 Page 16 of 24 Table 2 Summary of data from seeding Thy1‑ G93A Line T3 mice and VLE‑ G93A Host Inoculum (Route)# inoculated # Age to paralysis or euthanasia paralyzed Thy1‑ G93A PBS 2 0 14.8, 16.6 mo Thy1‑ G93A NTg Sp Cord (ISP) 3 0 17.7, 18.2, 19.1 mo Thy1‑ G93A Gur G93A Sp Cord (ISP) 4 0 Terminated at 19 mo Thy1‑ G93A G37R‑L29 Sp Cord (ISP) 4 0 Terminated at 18 mo VLE‑ G93A G93A(FVB) Sp Cord (ISP) 8 3 5.3, 8.3, 13.4 VLE‑ G93A G93A(FVB) Sp Cord (ScN) 11 3 5.8, 8, 10 VLE‑ G93A PBS (ScN) 5 0 Terminated at 12 mo Mice that were euthanized for non-MND related conditions, such as fight wounds or tumors, before 12 months of age were excluded, unless otherwise noted was hampered by unexpectedly high losses of mice in Table 3 Summary of data from seeding heterozygous PrP.G37R‑ SOD1 Line 110 mice the cohort due to fighting injuries. We also attempted to induce paralysis by seeding mice that over-express Inoculum (ISP) # inoculated Age of euthanasia due untagged WT-SOD1 (GurWT-SOD1 mice), again to age or non‑ ALS health issues observing no induction of paralysis in a small cohort of mice out to 15  months of age (Additional file  2: PBS 5 Terminated at 16 mo Table  S3; Data from [15]). To more rigorously address PrP.G37R‑SOD1 Line 110 6 Terminated at 16 mo whether it is possible to induce paralytic disease in Sp Cord mice that over-express WT-SOD1 we first considered G37R‑L29 Sp Cord 4 Terminated at 16 mo G85R−SOD1:YFP what type of misfolded SOD1 seed would potentially be P1 G37R Sp 7 Terminated at 9 mo Cord efficacious in mice expressing WT-SOD1. GurG93A Sp Cord 3 Terminated at 17 mo To determine what type of seed might be efficacious GurWT Sp Cord 4 Terminated at 16.8 mo in WT-SOD1 mice, we examined pathology in mice that co-express WT-SOD1:YFP with G93A- and WT- SOD1 (Additional file  2: Fig. S11). We have previously reported that WT-SOD1:YFP is not induced to form ICV or a direct sciatic nerve injection route [15]. Here, inclusion pathology by co-expression of L126Z-SOD1 we used our paradigm of newborn intraspinal injec- [21]. Bigenic mice produced from crosses of GurG93A tion to seed with spinal homogenates from paralyzed mice and WT-SOD1:YFP mice developed paralysis at homozygous PrP.G37R-SOD1, paralyzed G37R-SOD1 about the same time as mice that express only G93A- G85R−YFP Line 29, paralyzed P1-G37R , or paralyzed SOD1 (Additional file  2: Fig. S11a). As compared to GurG93A mice. None of these inocula induced para- mice that express only WT-SOD1:YFP, in paralyzed lytic disease (Table  3). In these older heterozygous PrP- GurG93A/WT-SOD1:YFP mice we observed redis- G37R mice, we observed some instances of silver-positive tribution of the YFP fluorescence into fibrillary or inclusion pathology, but the levels of pathology were punctate inclusion-like pathology (Additional file  2: similar between mice injected with PBS or transgenic spi- Fig. S11b & c). Remarkably, a similar type of punc- nal homogenates (Additional file  2: Fig. S10). To date, we tate inclusion-like pathology was observed in very old have not observed accelerated disease in any attempt to bigenic mice generated by crosses of GurWT mice and seed mice expressing the G37R variant of human SOD1. WT-SOD1:YFP mice (Additional file  2: Fig. S11d). We also noted a remarkable level of fluorescent inclusion Mice expressing WT‑SOD1 are largely resistant to prion‑like pathology in the cerebellar dentate (Additional file  2: seeding Fig. S12). Although these mice showed some obvious In addition to our attempts to seed mutant SOD1 mice, gait abnormalities, which occur in older GurWT mice we have also examined whether it is possible to induce [42], these bigenic GurWT/WT-SOD1:YFP mice did paralysis in mice that express WT-SOD1 [15]. In a not develop the paralytic phenotypes of mice express- pilot study with a relatively small cohort of mice that ing mutant SOD1. express WT-SOD1:YFP, we did not observe induction The foregoing data suggested that spinal homogen - of paralysis by 9–13 months of age by seeding with spi- ates from aged GurWT mice or paralyzed G93A mice nal homogenates from paralyzed G93A or G37R mice could potentially seed aggregation and disease in (Additional file  2: Table S3; Data from [15]). This study A yers et al. acta neuropathol commun (2021) 9:92 Page 17 of 24 Table 4 Summary of data from seeding WT‑SOD1:YFP and GurWT ‑SOD1 mice Host Inoculum (ISP) # inoculated # paralysis Age of euthanasia due to age WT‑SOD1:YFP PBS 3 0 17—20 mo (a) WT‑SOD1:YFP NTg Sp Cord 4 0 20 mo (b) WT‑SOD1:YFP GurWT Sp Cord 15 0 19–20 mo (b) WT‑SOD1:YFP recWT fibrils 10 0 16–20 mo (d) GurWT PBS 3 0 20 mo (e) GurWT NTg Sp Cord 3 0 20 mo (f ) GurWT GurWT Sp Cord 6 0 19 mo GurWT recWT SOD1 fibrils 7 0 16–20 mo GurWT GurG93A Sp Cord 4 0 17–20 mo GurWT L126Z Sp Cord 3 0 18–19 mo (g) GurWT QV103Z Sp Cord 3 1 19–20 mo (h) (a) Data originally reported in [15] (b) 1 found dead @ 7 mo. Data originally reported in [15] (c) 1 found dead @ 4.3 mo, 2 euthanized for non ALS-related issues at 8.7 and 15 mo. Includes data for 5 animals originally reported in [15] (d) 1 found dead @ 5.4 mo, 2 euthanized for non-ALS-related issues at 7 and 18 mo (e) 3 found dead @ 4.5 -5.4 mo, 1 euthanized at 3 mo (f ) Data originally reported in [15] (g) 1 found dead at 18.8 mo – no prior clinical signs (h) 1 animal showed bilateral weakness at 20 months when euthanized but lacked inclusion pathology WT-SOD1:YFP mice. Aged GurWT mice accumulate testing whether we could induce paralysis by the injec- detergent insoluble WT-SOD1 (albeit at lower levels tion of fibrilized WT-SOD1, or spinal homogenates of than mutant mice) [15]. We have now examined 15 aged GurWT mice, or spinal homogenates of paralyzed WT-SOD1:YFP mice injected with spinal homogen- mice expressing the QV103Z or L126Z mutants. Cohorts ates from aged GurWT mice, finding no animals with of mice were aged to 17–20  months of age with none of ALS-like symptoms by 16–20  months post-injection the animals developing any type of paralytic phenotype (Table 3). Additionally, there was not obvious patholog- (Table  4). Additionally, we could not pathologically dis- ical distinction between WT-SOD1:YFP mice injected tinguish mice injected with GurWT spinal homogenates with GurWT spinal homogenates and mice with PBS or fibrilized WT SOD1 from mice injected with PBS or or spinal homogenates of NTg mice (Additional file  2: NTg spinal homogenates (Table  4; Additional file  2: Fig. Fig. S13). To further attempt to induce paralytic dis- S14). Overall, the data show that WT-SOD1 is relatively ease in the WT-SOD1:YFP mice, we injected animals resistant to prion-like seeding towards induction of with recombinant WT human SOD1 that had been motor neuron disease or the type of inclusion pathology fibrilized in  vitro. No animals developed paralysis by present in mice expressing fALS mutant SOD1. 16–20  months post-injection, and none showed an obvious induction of WT-SOD1:YFP inclusion pathol- Discussion ogy at the time of euthanasia (Additional file  2: Fig. The present study builds on prior work to examine the S13). Altogether, these findings indicate that WT- sources of variability in the efficiency of prion-like seed - SOD1:YFP mice are relatively resistant to developing ing in mutant SOD1 mice. From the effort we describe paralytic disease, or inclusion pathology, by injection here, prior work in our laboratories [15, 17], and data with preparations of misfolded WT or mutant SOD1. described by Bedhendi [18], it is clear that first pas - In parallel to our effort in the WT-SOD1:YFP mice, we sage G85R seeds efficiently seed G85R mice, and that have attempted to induce paralytic disease in the GurWT- first passage L126Z seeds efficiently seed L126Z mice SOD1 mice by newborn ISP injection. In a prior study, (Fig.  9a). By contrast, mice expressing WT-, G37R-, we reported that by 15  months of age, small cohorts or G93A-SOD1 variants were partially or fully resist- (n = 4) of GurWT mice injected with spinal homogen- ant to isologous seeding (Fig.  9a). Notably, mice that ates from GurG93A mice had not developed paralysis express human G85R-SOD1 fused to YFP were found to [15]. We have now extended this effort considerably, be susceptible to seeding with a wide range of seeding Ayers et al. acta neuropathol commun (2021) 9:92 Page 18 of 24 Fig. 9 Summary diagrams illustrating the relative efficiency of seeding in mouse models. a Summary of findings in seeding experiments involving various lines of SOD1 mice express untagged WT or mutant SOD1. b Summary of findings in seeding experiments with G85R‑SOD1:YFP mice preparations, including tissue homogenates from para- considerable variation in the susceptibility of differ - lyzed G37R and G93A mice and preparations of recom- ent lines of mutant SOD1 mice to seeding, and that the binant SOD1 fibrilized in  vitro (Fig.  9b). In general, sequence of the misfolded SOD1 in the seeds can influ - heterologous seeding of G85R-SOD1:YFP mice was less ence prion-like transmissibility to SOD1 transgenic efficient than isologous seeding, with seed preparations hosts. that contained the G37R or H46R variant being among the least efficient. We further observed that the seeds Role of structural stability in the susceptibility of SOD1 that arose in G85R-SOD1:YFP mice after seeding by variants to seeding preparations from paralyzed G37R mice appeared to A key distinguishing feature of transgenic SOD1 mice acquire attributes that led to longer incubation times that were vulnerable to seeding versus those that were upon second passage. Collectively, these findings show more resistant could be related to the structural sta- that sequence variation and seed source have profound bility of the SOD1 variant expressed by the host. Mice effects on the prion-like activity of misfolded mutant that express WT, G37R, or G93A human SOD1 show SOD1. high steady-state levels of soluble protein prior to the In assessing the significance of animals that failed to onset of paralysis, with the accumulation of misfolded develop paralysis or pathology as a result of intraspinal SOD1 aggregates occurring as paralysis develops [41, or sciatic nerve seeding, we must consider the possi- 43, 44]. By contrast, the steady-state levels of soluble bility of experimentation error. It is important to note mutant SOD1 in G85R mice are relatively low in pro- that all of the injections performed in this study were portion to transgene mRNA levels [43]. Both the L126Z conducted by only two operators that had extensively and QV103Z truncation variants are unstable and show trained to perform the injections. Although we cannot very low steady state levels until they begin to aggregate rule out the possibility of occasional injection error, we as paralysis develops [20, 21]. Thus, for the G85R, L126Z, have no indication that operator error would explain and QV103Z mutants, their susceptibility to isologous, or the data in cases where few or no animals in a cohort heterologous, seeding may be linked to poor structural developed paralysis. Thus, we conclude that there is stability. A yers et al. acta neuropathol commun (2021) 9:92 Page 19 of 24 In WT SOD1 a structurally important intramolecular these variants resist seeding because bound Zn limits disulfide bond links Cys 57 to Cys 146 [45]. In studies of prion-like propagation. SOD1 aggregation and fibrillization in vitro, the presence Overall, our data indicate that WT SOD1 and WT- of the normal intramolecular disulfide bond appears to like variants are less likely to exist in conformations that be a key factor in modulating aggregation [46]. In mutant are susceptible to prion-like conformational templat- SOD1 mice that develop paralysis due to over-expression, ing as compared to mutants that are more destabilized. the misfolded mutant protein that accumulates as patho- Whether any specific structure-stabilizing feature is criti - logical inclusions develop lacks the normal intramolecu- cal in mediating susceptibility to seeding is unclear. What lar disulfide bond [43, 44]. Previous analysis of the same is clear is that mice expressing the more destabilized vari- G85R mice used here for seeding studies found that all of ants, G85R, L126Z, and QV103Z were far more suscepti- the mutant protein present in spinal cord of pre-symp- ble to prion-like seeding. tomatic mice lacks this critical structural stabilizer [43]. Similarly, the L126Z and QV103Z variants would not be Do ALS‑associated mutations in SOD1 imprint strain‑like able to produce a normal intramolecular disulfide bond. attributes that influence prion‑like activity? It is important to note, however, that in GurWT and Based on our observation that we can induce and moni- Gur1G93A SOD1 mice, 8–14% of the over-expressed tor the spread of G85R-SOD1:YFP aggregation by seed- protein present in spinal cords of pre-symptomatic ing into the sciatic nerve of this model, we had begun mice also lacks the normal disulfide bond [43, 44]. Thus, to link the progressive spread of weakness in SOD1- whether the G85R, L126Z, and QV103Z mice are suscep- ALS to the prion-like spread of misfolded conforma- tible to seeding solely because the expressed mutant lacks tions [16]. The question that arises is whether there may an intramolecular disulfide bond is unclear. be conformational attributes that are enciphered by the Another important post-translation modification in the primary sequence of mutant SOD1 that, by some man- structural stability of SOD1 is the binding of Cu and Zn ner, modulate the ability of misfolded SOD1 to spread [47]. Similar to WT SOD1, the G93A or G37R variants, within the CNS. It is important to note that in the trans- are capable of binding Cu and Zn with high affinity [3, genic mouse models that over-express mutant SOD1, 23, 26]. The G85R variant has been shown to have a weak the mutation-specific differences in disease duration affinity for Cu and Zn [48, 49] and the enzyme appears that is seen in humans are not recapitulated. For exam- to be inactive in vivo [50]. Crystal structures of the G85R ple, mice that express the G93A and G37R variants at SOD1 variant demonstrate disorder in the Zn binding similar levels exhibit disease onset and duration that is and electrostatic loop elements of SOD1 (amino acids comparable; 4–6  months to onset and 3–4  weeks dura- 50–83 and 121–142, respectively) [48]. The L126Z and tion [23, 26]. Presumably the high-level expression of experimental QV103Z truncation mutants have not been mutant SOD1 throughout the CNS that is required to crystallized, but neither of these variants possess intact produce disease within the lifespan of the animal masks electrostatic loop elements. Whether the L126Z trun- the mutation-specific attribute that causes very slow pro - cation mutant can bind Cu or Zn is unknown, but the gression in persons with the G37R mutation. Here, we experimental QV103Z variant that we studied was engi- have asked whether the G37R, or H46R, mutations could neered to remove the critical Cu-binding sites in SOD1 impart some specific conformational information to the [21]. It is notable, however, that the majority of over- misfolded SOD1 that accumulates in these animals that expressed WT or G93A SOD1 present in the spinal cords would lead to distinct behavior in prion-like transmission of Gur1 G93A or Gur WT mice is inactive due to insuf- studies. ficient acquisition of Cu [43]. Thus, the ability of WT and Our newborn injection paradigm is one way to assess G93A SOD1 to bind Cu does not seem to fully explain the seeding competency of different SOD1 variants. In the low susceptibility of mice over-expressing these pro- heterologous seeding in G85R-SOD1:YFP mice by spi- teins to prion-like seeding if poor Cu binding were the nal homogenates from paralyzed G93A or L126Z mice only structural feature involved. (avg. durations of 2.4 and 3.8  years in humans, respec- The binding of Zn by nascent SOD1 appears to be criti - tively [35]), the efficiency of seeding at first passage was cal to achieving a native conformation [51, 52]. Analy- variable. Second passage of G93A seeds through G85R- sis of partially purified homodimeric SOD1 from spinal SOD1:YFP mice, however, produced a consistently earlier cords of the GurWT, G37R-Line 29, and Gur1G93A mice onset of paralysis (avg. 2.8  months post-injection). Our found the over-expressed protein appeared have no defi - study of L126Z seeds in G85R-SOD1:YFP mice is more ciency in the level of bound Zn; measured to be between limited, but we did similarly observe consistent accel- 2.5 and 3 atoms per dimer [53]. Thus, it is possible that eration of paralysis by second-passage L126Z seeds in a small cohort of animals (Additional file  2: Fig. S15; data Ayers et al. acta neuropathol commun (2021) 9:92 Page 20 of 24 from [17]). Thus, for two mutants associated with rapidly that seeding could accelerate the appearance of paralysis progressing disease we observe relative ease of first pas - albeit at a lower frequency than we would have expected sage with accelerated and highly efficient second passage for isologous seeding. As noted above, in the Gur1-G93A seeding. mice, a substantial portion of the over-expressed protein For the H46R and G37R variants, associated with lacks Cu and a normal disulfide bond [43]. The absence of slowly progressing ALS in humans, we observed poor these modifications appears to be related to over-expres - seeding activity in first passage, and with G37R-derived sion; increasing Cu loading in Gur1G93A mice dramati- seeds we osbserved variable and protracted incubation cally mitigates its misfolding and toxicity [57]. The Cu periods. These data are consistent with the idea that loading and disulfide status of G93A SOD1 in spinal mutations associated with slowly progressing disease cords of VLE-G93A mice has not been examined, but the may be less efficient in prion-like seeding. lower expression of mutant protein would be expected An important consideration in interpreting these to result in lower levels of incompletely modified mutant results is the potential impact of the host G85R SOD1 SOD1. variant on any conformational change propagated by The absence of seeding efficacy in the y Th -1 G93A mice exposure to seeds from mice expressing any other vari- could potentially be related to the pattern of transgene ant. In PrP prions, sequence differences between the pri - expression for this model. It is also possible that the level mary sequence of PrP in seeds and the sequence of PrP of G93A SOD1 expression in Thy-1 G93A mice is just in the host can modify strain attributes [54]. Moreover, below threshold to sustain the propagation of misfolded single amino acid differences of the PrP sequence in the conformations. Additional studies of seeding efficacy in seed and host can create a species barrier that lowers these two lines of G93A mice could provide insight into transmission efficiency [55]. To date, G85R-SOD1 mice the basis of vulnerability. At the time of writing, we are have shown efficient seeding with spinal cord prepara - not able to explain why the Gur1-G93A or Thy-1 G93A tions containing misfolded A4V, G85R, D90A, G93A, mice did not respond to seeding or why the VLE-G93A and L126 or G127 truncation mutants [15, 17–19]. We showed a modest response. If vulnerability to seeding is observed that fibrilization treatments of 7 different by some means related to the folding state of the SOD1 recombinant SOD1 proteins could produce seeds that protein in the host, we would have expected the Gur1- induce early paralysis in G85R-SOD1:YFP mice. These G93A mice to be more susceptible to seeding due to a data indicate that G85R-SOD1 is broadly susceptible to higher level of incompletely modified mutant protein. At heterologous conformational templating. Indeed, seeds face value, the poor seeding of G93A mice could foster from G37R mice were more effective in G85R SOD1 skepticism over the role of prion-like spread in the patho- mice than G37R SOD1 mice. u Th s, the poor seeding of genesis of SOD1-linked fALS; however, there are mitigat- G85R-SOD1:YFP mice by spinal tissues from H46R mice ing factors that merit discussion. and the long incubation periods produced by G37R seeds A key variable in our experiments is the level of bioac- may indicate that these variants produce strains of mis- tive misfolded SOD1 seeds in the spinal homogenates we folded SOD1 that are less effective as seeds to propagate have used for seeding. Previous studies of mutant SOD1 misfolded conformations. mice that develop paralysis have demonstrated that the levels of misfolded mutant SOD1 in the spinal cords rise Is prion‑like spreading a common feature in SOD1 ALS? steadily and reach maximum when mice are paralyzed Our observation that mice expressing the G93A variant [20, 41, 44]. We originally had assumed that at endstage, are not particularly receptive to seeding would appear to regardless of age, the levels of misfolded SOD1 in any run counter to the notion that prion-like mechanisms of given animal for a given mutant would be at maximal lev- spreading apply to all fALS variants. As indicated above, els. For all of our mouse to mouse propagation studies, we the G93A mutation is associated with relatively rap- used 10% homogenates of spinal cord that had been clari- idly progressing disease [35, 56], and thus the expecta- fied by low speed centrifugation. In our isologous seeding tion would be that mice expressing G93A-SOD1 should studies of G85R-SOD1:YFP, G85R, and L126Z mice, the be highly permissive to seeding, particularly isologous levels of the seeds in homogenates from paralyzed trans- seeding. We have previously reported that the com- genic mice were clearly sufficient to induce accelerated monly used Gur1-G93A mice did not exhibit acceler- paralysis with high efficiency. For the WT-SOD1:YFP, ated paralysis after intraspinal seeding [15]. To assess WT-, G93A-, and G37R-SOD1 expressing mice, which whether the level of G93A expression may have been a did not respond to isologous seeding, it is possible that factor, we tested two additional lines of G93A mice. The injection of a higher amount of seeds would have induced y Th -1 G93A mice were unresponsive to seeding; how - disease. Notably, our G93A and G37R seed prepara- ever, with the VLE-G93A mice we began to see evidence tions were effective, to varying degrees, in heterologous A yers et al. acta neuropathol commun (2021) 9:92 Page 21 of 24 seeding of G85R-SOD1:YFP mice. Moreover, recombi- Intriguingly, we observed multiple examples in which nant fibrils of WT human SOD1 were highly effective in seeded mice reached relatively old ages without devel- seeding G85R-SOD1:YFP mice, with no activity in WT oping symptoms despite significant inclusion pathology or WT-SOD1:YFP mice. Thus, it is clear that the lines of burden. The most striking example was in the cohort of mice expressing WT and WT-like SOD1 mutants (G37R G85R-SOD1:YFP mice seeded with spinal homogenates and G93A) are less susceptible to prion-like seeding with from paralyzed H46R rats. We also observed high levels preparations that effectively induce early disease in mice of inclusion pathology in aged bigenic mice created by expressing the unstable G85R mutant. Whether the lower crossing GurWT mice with WT-SOD1:YFP mice. These susceptibility of WT and WT-like variants to seeding intriguing findings raise the possibility that some con - could be overcome by injection of higher amounts of mis- formations of misfolded SOD1 may be non-toxic. Addi- folded SOD1 seeds will require additional investigation. tional studies are required to understand whether strain An additional consideration in assessing the suscepti- variants of misfolded mutant SOD1 may exist that prop- bility of WT and WT-like SOD1 mutants to prion-like agate between cells more slowly, or produce less toxic propagation is the timing of exposure to the seed. We aggregates, and whether such strain variations explain have relied heavily on a paradigm in which the seeds aspects of human SOD1 linked ALS. were injected into newborn mice, primarily as a means to facilitate wide-spread dissemination of the seeds. The Role of WT‑SOD1 in sporadic ALS earliest age to paralysis in mice expressing high levels of The role of prion-like propagation in sporadic ALS, and mutant SOD1 that has been reported is 3–4  months of the identity of the misfolded propagon, remains to be age [23, 26, 29], suggesting that early in life the toxicity of determined. Multiple studies have examined whether the mutant SOD1 is suppressed by yet to be defined pro - a misfolded form of WT-SOD1 could be propagat- tective factors. It is possible that mice expressing WT-, ing throughout the neuraxis in sporadic ALS patients G93A-, or G37R-SOD1 do not respond to seeding in (reviewed in [58]). Recombinant WT-SOD1 that has the newborn injection paradigm because a combination been fibrilized in  vitro clearly possesses high seeding of protective factors and the natural propensity of these activity when injected into G85R-SOD1:YFP mice [17], mutants to fold into a more native-like enzyme, sup- but these same preparations showed no activity in the presses the establishment of a self-sustaining propaga- WT-SOD1:YFP or GurWT mice. From our experiments tive process. It is also possible that the capacity of WT or in which mutant SOD1 was co-expressed with WT- WT-like SOD1 variants to fold and mature could change SOD1:YFP, it was clear that with sustained exposure it is with aging as proteostatic protective factors decline. possible to induce WT-SOD1:YFP to produce inclusions. In humans, where disease usually occurs late in life, the Collectively, our data indicate that although it is pos- WT-like variants of SOD1 could become more prone to sible to create a seeding-competent conformer of WT- adopt conformations that are susceptible to prion-like SOD1, it does not appear to transmit to WT substrates seeding. efficiently in short-term seeding paradigms. It is possible Although our studies, and others [18, 19], demonstrate that propagation of misfolded conformations by WT- that mutant SOD1 has the potential to acquire prion-like SOD1 is inhibited by the high propensity of WT-SOD1 capabilities, there are aspects of the mouse models that to acquire a stable native conformation. Whether other provided sources of misfolded SOD1 seeds that should types of conformational changes in WT-SOD1 that could be considered. In all the mouse or rat models used to contribute to ALS, and possibly propagate in a prion-like produce inoculum, mutant SOD1 is highly expressed fashion as suggested by other studies [10, 58, 59], has throughout the CNS. In this scenario, the development not been ruled out by our study. However, we have not of disease would not necessarily depend on the prion- observed WT SOD1 to be able to support propagation of like spread and therefore there would be little selective a disease-causing misfolded SOD1 conformation thus far. pressure towards the accumulation of misfolded SOD1 proteins that can efficiently propagate between cells. Potential role of amyloidogenic segments in prion‑like It is possible that the absence of such selective pressure propagation of misfolded SOD1 leads to the accumulation of mixtures of misfolded SOD1 Although the G85R-SOD1:YFP, or untagged G85R, mice strains in the transgenic mice that were used to produce appear to be highly susceptible to seeding, we observed the seeds. Any given individual animal could potentially that seeds prepared from mice that over-express murine harbor a unique mixture of strains and as such could SOD1 with the G85R mutation were not effective in account for some of the variability we observe in first- either host. Incompatibility between the misfolded seed passage seeding experiments. and the host was also noted when QV103Z mice were seeded with spinal homogenates from L126Z mice, but Ayers et al. acta neuropathol commun (2021) 9:92 Page 22 of 24 not vice versa. These seemingly disparate outcomes may efficiently seed G93A mice weakens the idea that prion-like provide clues to initial sites of interaction between the propagation mediates the perceived spread of weakness propagating seed and the naïve host protein. Ivanova and in all cases of SOD1-linked ALS. Additional studies are colleagues identified four amino acid segments in SOD1 required to understand the poor response of mice express- that were highly prone to aggregation [60]. The four seg - ing the G93A variant to seeding; perhaps it is a question of ments were distributed across the protein at residues timing or dose of the seed. Our study suggests that further 14–21, 30–38, 101–107, and 147–153. The ability of mis - investigations of the prion-like propagation of misfolded folded QV103Z to propagate to itself or L126Z-SOD1 conformations in mutant SOD1 could reveal mechanisms indicates that if these segments are involved in prion-like that underlie disparate durations of illness in individuals propagation, then the segments 14–21 and 30–38 must with different mutations in the same gene. be involved. For the L126Z mutant, segment 101–107 could also be important because although L126Z could Supplementary Information self-seed, it could not cross seed to QV103Z. It is also The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40478‑ 021‑ 01191‑w. noteworthy that mice expressing the G37R variant are largely resistant to seeding, and similarly, a synthetic pep- Additional file 1. Excel spreadsheet that lists all of the transgenic animals, tide of the amyloidogenic segment 30–38 with the G37R and a subset of the nontransgenic animals, used in this study, with details mutation was much slower to aggregate in  vitro [60]. on the inoculum injected and incubation period postinjection. See Sup‑ Segment 30–38 is also of interest because of significant plementary Materials for additional information. sequence divergence between mouse and human SOD1 Additional file 2. This file contains all Supplementary Material, including Supplementary Tables S1‑S3 and Supplementary Figures S1‑S15. within this segment; the human sequence is KVWG- SIKGL and the mouse sequence is VLSGQITGL. Our studies clearly show that misfolded mouse G85R-SOD1 Acknowledgements We would like to thank Dr. Paramita Chakrabarty for helpful suggestions in the seeds are not transmissible to mice expressing human preparation of this report, and Susan Fromholt for assistance in the manage‑ G85R-SOD1. In segments 14–21 and 101–107 there ment of transgenic animal colonies. is only one amino acid difference between mouse and Authors’ contributions human SOD1, and in segment 147–153 there are no JIA and DRB conceived the described experiments, directed the efforts differences. One of the differences between human and of support staff, interpreted the data, and composed the manuscript. GX mouse sequence in segment 30–38 is a W to S change performed histological examinations and contributed to the production of figures. KD administered inocula, monitored animals, and harvested tissues. at position 32, which has been implicated in prion-like QL and ZC examined histological data. JB, AKMR, DLZ, & AG produced and propagation of misfolded SOD1 [14, 36, 61, 62]. Col- purified recombinant SOD1 proteins. All authors read and approved the final lectively, these data suggest the amyloidogenic segment manuscript. between residues 30–38 could be an important modula- Funding tor of prion-like propagation of misfolded SOD1. A.G. was supported by St. Mary’s University Research Grant and the Biaggini Research Program. This work was supported by an NIH Shared Instrumenta‑ tion Grant (S10OD020026), grants from the National Institutes of Neurological Conclusions Disorders and Stroke (1R21NS088839; 1R01NS092788), and the Packard Center The collective data of the present study and prior work for ALS Research at Johns Hopkins University. [15–17, 19] clearly show that spinal cords of paralyzed Availability of data and materials mutant SOD1 mice contain seed-competent forms of mis- Any data generated or analyzed during this study that are not included in folded SOD1 that can induce a self-propagating process, this published article and its supplementary files, are available from the cor ‑ responding author on reasonable request. leading to accelerated motor neuron disease and inclu- sion pathology. Similar seeds are found in human spinal Declarations tissues from SOD1-linked ALS patients [19]. Overall, our findings generally align with the hypothesis that prion-like Ethics approval and consent to participate propagation of misfolded conformations could mediate the All studies involving mice were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Florida in accordance with all progressive spread of weakness from limb to limb that is a state and federal guidelines. defining feature of the disease. Tissue homogenates from mice that express mutations associated with more rapidly Consent for publication Not applicable. progressing ALS (G93A, G85R, L126Z) were generally more effective in seeding than homogenates prepared from Competing Interests mice expressing mutants associated with slowly progress- The authors declare that they have no competing interests. ing disease (G37R, H46R). Mice expressing the G85R and Author details L126Z mutants were more permissive to seeding than mice 1 Department of Neuroscience, Center for Translational Research in Neurode‑ expressing the G37R variant. At face value, the inability to generative Disease (CTRND), University of Florida, Box 100159, Gainesville, FL A yers et al. acta neuropathol commun (2021) 9:92 Page 23 of 24 32610, USA. Institute for Neurodegenerative Disease, Weill Institute for Neuro‑ 18. Bidhendi EE, Bergh J, Zetterstrom P, Andersen PM, Marklund SL, sciences, University of California, San Francisco, CA 94143, USA. Depar tment Brannstrom T (2016) Two superoxide dismutase prion strains transmit of Neurology, Weill Institute for Neurosciences, University of California, San amyotrophic lateral sclerosis‑like disease. J Clin Invest 126:2249–2253 Francisco, CA 94143, USA. Department of Biological Sciences, St. Mary’s Uni‑ 19. Ekhtiari Bidhendi E, Bergh J, Zetterström P, Forsberg K, Pakkenberg B, versity, San Antonio, TX 78228, USA. SantaFe HealthCare Alzheimer’s Disease Andersen PM et al (2018) Mutant superoxide dismutase aggregates from Research Center, McKnight Brain Institute, University of Florida, Gainesville, FL human spinal cord transmit amyotrophic lateral sclerosis. Acta Neuro‑ 32610, USA. pathol 136:939–953 20. Wang J, Xu G, Li H, Gonzales V, Fromholt D, Karch C et al (2005) Soma‑ Received: 24 March 2021 Accepted: 3 May 2021 todendritic accumulation of misfolded SOD1‑L126Z in motor neurons mediates degeneration: {alpha}B‑ crystallin modulates aggregation. Hum Mol Genet 14:2335–2347 21. Xu G, Ayers JII, Roberts BLL, Brown H, Fromholt S, Green C et al (2015) Direct and indirect mechanisms for wild‑type SOD1 to enhance the toxicity of mutant SOD1 in bigenic transgenic mice. Hum Mol Genet References 24:1019–1035 1. Ravits JM, La SAR (2009) ALS motor phenotype heterogeneity, focality, 22. Wang J, Farr GW, Zeiss CJ, Rodriguez‑ Gil DJ, Wilson JH, Furtak K et al and spread: deconstructing motor neuron degeneration. Neurology (2009) Progressive aggregation despite chaperone associations of a 73:805–811 mutant SOD1‑ YFP in transgenic mice that develop ALS. Proc Natl Acad 2. Volk AE, Weishaupt JH, Andersen PM, Ludolph AC, Kubisch C (2018) Sci USA 106:1392–1397 Current knowledge and recent insights into the genetic basis of amyo‑ 23. Wong PCC, Pardo CAA, Borchelt DRR, Lee MKK, Copeland NGG, Jenkins trophic lateral sclerosis. Med Genet Mitteilungsblatt des Berufsverbandes NAA et al (1995) An adverse property of a familial ALS‑linked SOD1 muta‑ Medizinische Genet eV 30:252–258 tion causes motor neuron disease characterized by vacuolar degenera‑ 3. Potter SZ, Valentine JS (2003) The perplexing role of copper‑zinc superox ‑ tion of mitochondria. Neuron 14:1105–1116 ide dismutase in amyotrophic lateral sclerosis (Lou Gehrig’s disease). J Biol 24. Wang J, Xu G, Slunt HHH, Gonzales V, Coonfield M, Fromholt D et al (2005) Inorg Chem 8:373–380 Coincident thresholds of mutant protein for paralytic disease and protein 4. Sheng Y, Chattopadhyay M, Whitelegge J, Valentine JS. SOD1 aggrega‑ aggregation caused by restrictively expressed superoxide dismutase tion and ALS: role of metallation states and disulfide status. Curr Top Med cDNA. Neurobiol Dis 20:943–952 Chem. United Arab Emirates; 2012;12:2560–72. 25. Bruijn LI, Becher MW, Lee MK, Anderson KL, Jenkins NA, Copeland NG 5. Sangwan S, Eisenberg DS (2016) Perspective on SOD1 mediated toxicity et al (1997) ALS‑linked SOD1 mutant G85R mediates damage to astro ‑ in amyotrophic lateral sclerosis. Postepy Biochem Poland 62:362–369 cytes and promotes rapidly progressive disease with SOD1‑ containing 6. Grad LI, Yerbury JJ, Turner BJ, Guest WC, Pokrishevsky E, O’Neill MA et al inclusions. Neuron 18:327–338 (2014) Intercellular propagated misfolding of wild‑type Cu/Zn superoxide 26. Gurney ME, Pu H, Chiu AY, Canto MCD, Polchow CY, Alexander DD et al dismutase occurs via exosome‑ dependent and ‑independent mecha‑ (1994) Motor neuron degeneration in mice that express a human Cu, Zn nisms. Proc Natl Acad Sci USA 111:3620–3625 superoxide dismutase mutation. Science 264:1772–1775 7. Bosco DA, Morfini G, Karabacak NM, Song Y, Gros‑Louis F, Pasinelli P et al 27. Jaarsma D, Teuling E, Haasdijk ED, De ZCI, Hoogenraad CC (2008) (2010) Wild‑type and mutant SOD1 share an aberrant conformation and Neuron‑specific expression of mutant superoxide dismutase is sufficient a common pathogenic pathway in ALS. Nat Neurosci 13:1396–1403 to induce amyotrophic lateral sclerosis in transgenic mice. J Neurosci 8. Forsberg K, Jonsson PA, Andersen PM, Bergemalm D, Graffmo KS, Hultdin 28:2075–2088 M, et al. Novel antibodies reveal inclusions containing non‑native SOD1 28. Deitch JS, Alexander GM, Bensinger A, Yang S, Jiang JT, Heiman‑Patterson in sporadic ALS patients. PLoSONE. 2010;5:e11552. TD. Phenotype of transgenic mice carrying a very low copy number of 9. Tokuda E, Takei Y‑I, Ohara S, Fujiwara N, Hozumi I, Furukawa Y (2019) Wild‑ the mutant human G93A superoxide dismutase‑1 gene associated with type Cu/Zn‑superoxide dismutase is misfolded in cerebrospinal fluid of amyotrophic lateral sclerosis. PLoS One. 2014;9:e99879. sporadic amyotrophic lateral sclerosis. Mol Neurodegener 14:42 29. Ripps ME, Huntley GW, Hof PR, Morrison JH, Gordon JW. Transgenic mice 10. Paré B, Lehmann M, Beaudin M, Nordström U, Saikali S, Julien J‑P et al expressing an altered murine superoxide dismutase gene provide an (2018) Misfolded SOD1 pathology in sporadic amyotrophic lateral sclero‑ animal model of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A. sis. Sci Rep 8:14223 United States; 1995;92:689–93. 11. Brotherton TE, Li Y, Cooper D, Gearing M, Julien JP, Rothstein JD et al 30. Aoki M, Kato S, Nagai M, Itoyama Y (2005) Development of a rat model (2012) Localization of a toxic form of superoxide dismutase 1 protein to of amyotrophic lateral sclerosis expressing a human SOD1 transgene. pathologically affected tissues in familial ALS. Proc Natl Acad Sci USA Neuropathology Australia 25:365–370 109:5505–5510 31. Seetharaman SV, Taylor AB, Holloway S, Hart PJ (2010) Structures of 12. Ayers JI, Xu G, Pletnikova O, Troncoso JCC, Hart PJJ, Borchelt DRR (2014) mouse SOD1 and human/mouse SOD1 chimeras. Arch Biochem Biophys Conformatonal specificity of the C4F6 SOD1 antibody; low frequency of United States 503:183–190 reactivity in sporadic ALS cases. Acta Neuropathol Commun 2:55 32. Ayers JII, Fromholt S, Sinyavskaya O, Siemienski Z, Rosario AMM, Li A et al 13. Da Cruz S, Bui A, Saberi S, Lee SK, Stauffer J, McAlonis‑Downes M et al (2015) Widespread and efficient transduction of spinal cord and brain (2017) Misfolded SOD1 is not a primary component of sporadic ALS. Acta following neonatal AAV injection and potential disease modifying effect Neuropathol Germany 134:97–111 in ALS mice. Mol Ther 23:53–62 14. Grad LI, Guest WC, Yanai A, Pokrishevsky E, O’Neill MA, Gibbs E et al (2011) 33. Bartz JC, Kincaid AE, Bessen RA (2002) Retrograde transport of transmis‑ Intermolecular transmission of superoxide dismutase 1 misfolding in sible mink encephalopathy within descending motor tracts. J Virol living cells. Proc Natl Acad Sci USA 108:16398–16403 76:5759–5768 15. Ayers JI, Fromholt S, Koch M, DeBosier A, McMahon B, Xu G et al (2014) 34. Vallet PG, Guntern R, Hof PR, Golaz J, Delacourte A, Robakis NK et al (1992) Experimental transmissibility of mutant SOD1 motor neuron disease. Acta A comparative study of histological and immunohistochemical methods Neuropathol 128:791–803 for neurofibrillary tangles and senile plaques in Alzheimer’s disease. Acta 16. Ayers JI, Fromholt SE, O’Neal VM, Diamond JH, O’Neal VM, Diamond JH Neuropathol 83:170–178 et al (2016) Prion‑like propagation of mutant SOD1 misfolding and motor 35. Prudencio M, Hart PJJ, Borchelt DRR, Andersen PMM (2009) Variation in neuron disease spread along neuroanatomical pathways. Acta Neuro‑ aggregation propensities among ALS‑associated variants of SOD1: cor ‑ pathol 131:103–114 relation to human disease. Hum Mol Genet 18:3217–3226 17. Ayers JI, Diamond J, Sari A, Fromholt S, Galaleldeen A, Ostrow LW et al 36. Crown A, McAlary L, Fagerli E, Brown H, Yerbury JJ, Galaleldeen A, et al. (2016) Distinct conformers of transmissible misfolded SOD1 distinguish Tryptophan residue 32 in human Cu‑Zn superoxide dismutase modulates human SOD1‑FALS from other forms of familial and sporadic ALS. Acta prion‑like propagation and strain selection. PLoS One. 2020;15:e0227655. Neuropathol 132:827–840 37. Abdolvahabi A, Shi Y, Rasouli S, Croom CM, Aliyan A, Martí AA et al (2017) Kaplan‑Meier Meets Chemical Kinetics: Intrinsic Rate of SOD1 Ayers et al. acta neuropathol commun (2021) 9:92 Page 24 of 24 Amyloidogenesis Decreased by Subset of ALS Mutations and Cannot 50. Bruijn LI, Houseweart MK, Kato S, Anderson KL, Anderson SD, Ohama Fully Explain Age of Disease Onset. ACS Chem Neurosci United States E et al (1998) Aggregation and motor neuron toxicity of an ALS‑linked 8:1378–1389 SOD1 mutant independent from wild‑type SOD1. Science 281:1851–1854 38. McAlary L, Aquilina JA, Yerbury JJ. Susceptibility of Mutant SOD1 to 51. Kayatekin C, Zitzewitz JA, Matthews CR (2008) Zinc binding modulates Form a Destabilized Monomer Predicts Cellular Aggregation and Toxicity the entire folding free energy surface of human Cu. Zn superoxide but Not In vitro Aggregation Propensity. Front Neurosci. Switzerland; dismutase J Mol Biol 384:540–555 2016;10:499. 52. Potter SZ, Zhu H, Shaw BF, Rodriguez JA, Doucette PA, Sohn SH et al 39. Abdolvahabi A, Shi Y, Chuprin A, Rasouli S, Shaw BF (2016) Stochastic (2007) Binding of a single zinc ion to one subunit of copper‑zinc super ‑ formation of fibrillar and amorphous superoxide dismutase oligomers oxide dismutase apoprotein substantially influences the structure and linked to amyotrophic lateral sclerosis. ACS Chem Neurosci 7:799–810 stability of the entire homodimeric protein. J Am Chem Soc United States 40. Xia Y, Chen Z, Xu G, Borchelt DR, Ayers JI, Giasson BI. Novel SOD1 Mono‑ 129:4575–4583 clonal Antibodies against the Electrostatic Loop Preferentially Detect 53. Lelie HLL, Liba A, Bourassa MWW, Chattopadhyay M, Chan PKK, Gralla EBB Misfolded SOD1 Aggregates. Neurosci Lett. Ireland; 2020;135553. et al (2011) Copper and zinc metallation status of copper‑zinc superoxide 41. Lang L, Zetterstrom P, Brannstrom T, Marklund SL, Danielsson J, Oliveberg dismutase from amyotrophic lateral sclerosis transgenic mice. J Biol M (2015) SOD1 aggregation in ALS mice shows simplistic test tube Chem 286:2795–2806 behavior. Proc Natl Acad Sci U S A 112:9878–9883 54. Cortez LM, Sim VL (2013) Implications of prion polymorphisms. Prion 42. Graffmo KS, Forsberg K, Bergh J, Birve A, Zetterstrom P, Andersen PM et al 7:276–279 (2013) Expression of wild‑type human superoxide dismutase ‑1 in mice 55. Hsiao KK, Groth D, Scott M, Yang SL, Serban H, Rapp D et al (1994) Serial causes amyotrophic lateral sclerosis. Hum Mol Genet 22:51–60 transmission in rodents of neurodegeneration from transgenic mice 43. Jonsson PA, Graffmo KS, Andersen PM, Brannstrom T, Lindberg M, expressing mutant prion protein. Proc Natl Acad Sci U S A 91:9126–9130 Oliveberg M et al (2006) Disulphide‑reduced superoxide dismutase ‑1 in 56. Cleveland DW, Laing N, Hurse PV, Brown RHJ (1995) Toxic mutants in CNS of transgenic amyotrophic lateral sclerosis models. Brain; a J Neurol Charcot’s sclerosis. Nature England 378:342–343 129:451–464 57. Williams JR, Trias E, Beilby PR, Lopez NI, Labut EM, Bradford CS et al (2016) 44. Karch CMM, Prudencio M, Winkler DDD, Hart PJJ, Borchelt DRR (2009) Role Copper delivery to the CNS by CuATSM effectively treats motor neuron of mutant SOD1 disulfide oxidation and aggregation in the pathogenesis disease in SOD(G93A) mice co‑ expressing the Copper‑ Chaperone‑for ‑ of familial ALS. Proc Natl Acad Sci U S A 106:7774–7779 SOD. Neurobiol Dis 89:1–9 45. Hart PJ, Liu H, Pellegrini M, Nersissian AM, Gralla EB, Valentine JS et al 58. Grad LI, Cashman NR (2014) Prion‑like activity of Cu/Zn superoxide (1998) Subunit asymmetry in the three‑ dimensional structure of a human dismutase: implications for amyotrophic lateral sclerosis. Prion 8:33–41 CuZnSOD mutant found in familial amyotrophic lateral sclerosis. Protein 59. Furukawa Y, Tokuda E (2020) Does wild‑type Cu/Zn‑superoxide dismutase Sci 7:545–555 have pathogenic roles in amyotrophic lateral sclerosis? Transl Neurode‑ 46. Chattopadhyay M, Nwadibia E, Strong CD, Gralla EB, Valentine JS, gener 9:33 Whitelegge JP (2015) The Disulfide Bond, but Not Zinc or Dimerization, 60. Ivanova MI, Sievers SA, Guenther EL, Johnson LM, Winkler DD, Galalel‑ Controls Initiation and Seeded Growth in Amyotrophic Lateral Sclerosis‑ deen A et al (2014) Aggregation‑triggering segments of SOD1 fibril linked Cu, Zn Superoxide Dismutase (SOD1) Fibrillation. J Biol Chem formation support a common pathway for familial and sporadic ALS. Proc 290:30624–30636 Natl Acad Sci U S A 111:197–201 47. Oztug Durer ZA, Cohlberg JA, Dinh P, Padua S, Ehrenclou K, Downes S, 61. Pokrishevsky E, McAlary L, Farrawell NE, Zhao B, Sher M, Yerbury JJ, et al. et al. Loss of metal ions, disulfide reduction and mutations related to Tryptophan 32‑mediated SOD1 aggregation is attenuated by pyrimidine ‑ familial ALS promote formation of amyloid‑like aggregates from superox ‑ like compounds in living cells. Sci Rep. England; 2018;8:15590. ide dismutase. PLoS One. United States; 2009;4:e5004. 62. DuVal MG, Hinge VK, Snyder N, Kanyo R, Bratvold J, Pokrishevsky E et al 48. Cao X, Antonyuk SV, Seetharaman SV, Whitson LJ, Taylor AB, Holloway SP (2019) Tryptophan 32 mediates SOD1 toxicity in a in vivo motor neuron et al (2008) Structures of the G85R variant of SOD1 in familial amyo‑ model of ALS and is a promising target for small molecule therapeutics. trophic lateral sclerosis. J Biol Chem 283:16169–16177 Neurobiol Dis United States 124:297–310 49. Ratovitski T, Corson LB, Strain J, Wong P, Cleveland DW, Culotta VC et al (1999) Variation in the biochemical/biophysical properties of mutant Publisher’s Note superoxide dismutase 1 enzymes and the rate of disease progres‑ Springer Nature remains neutral with regard to jurisdictional claims in pub‑ sion in familial amyotrophic lateral sclerosis kindreds. Hum Mol Genet lished maps and institutional affiliations. 8:1451–1460 Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals

Variation in the vulnerability of mice expressing human superoxide dismutase 1 to prion-like seeding: a study of the influence of primary amino acid sequence

Loading next page...
 
/lp/springer-journals/variation-in-the-vulnerability-of-mice-expressing-human-superoxide-XOo2ucie9q

References (76)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2051-5960
DOI
10.1186/s40478-021-01191-w
Publisher site
See Article on Publisher Site

Abstract

Misfolded forms of superoxide dismutase 1 (SOD1) with mutations associated with familial amyotrophic lateral sclerosis (fALS) exhibit prion characteristics, including the ability to act as seeds to accelerate motor neuron disease in mouse models. A key feature of infectious prion seeding is that the efficiency of transmission is governed by the primary sequence of prion protein (PrP). Isologous seeding, where the sequence of the PrP in the seed matches that of the host, is generally much more efficient than when there is a sequence mis‑match. Here, we used paradigms in which mutant SOD1 seeding homogenates were injected intraspinally in newborn mice or into the sciatic nerve of adult mice, to assess the influence of SOD1 primary sequence on seeding efficiency. We observed a spectrum of seeding efficiencies depending upon both the SOD1 expressed by mice injected with seeds and the origin of the seed preparations. Mice expressing WT human SOD1 or the disease variant G37R were resistant to isologous seeding. Mice expressing G93A SOD1 were also largely resistant to isologous seeding, with limited success in one line of mice that express at low levels. By contrast, mice expressing human G85R‑SOD1 were highly susceptible to isologous seed‑ ing but resistant to heterologous seeding by homogenates from paralyzed mice over‑ expressing mouse SOD1‑ G86R. In other seeding experiments with G85R SOD1:YFP mice, we observed that homogenates from paralyzed animals expressing the H46R or G37R variants of human SOD1 were less effective than seeds prepared from mice expressing the human G93A variant. These sequence mis‑match effects were less pronounced when we used purified recombi‑ nant SOD1 that had been fibrilized in vitro as the seeding preparation. Collectively, our findings demonstrate diversity in the abilities of ALS variants of SOD1 to initiate or sustain prion‑like propagation of misfolded conformations that produce motor neuron disease. anatomically connected pathways, leading to general- Introduction ized paralysis [1]. In a subset of cases, weakness first Amyotrophic lateral sclerosis (ALS) can present clini- appears in muscles of the head and neck before spread- cally as focal weakness in a limb, hand, or foot that ing to include the diaphragm and intercostal muscles. progressively worsens before weakness spreads along The symptoms of weakness originate from dysfunction of both upper and lower motor neurons, with dysfunc- *Correspondence: drb1@ufl.edu 1 tion of the upper neurons causing symptoms of spasticity Department of Neuroscience, Center for Translational Research in Neurodegenerative Disease (CTRND), University of Florida, Box 100159, before paralysis makes voluntary movements impossible. Gainesville, FL 32610, USA The vast majority of ALS cases have no clear etiology and Full list of author information is available at the end of the article © The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Ayers et al. acta neuropathol commun (2021) 9:92 Page 2 of 24 no obvious family history; however, up to 25% of patients most unstable variants of mutant SOD1 we have exam- have a family history of disease with a subset of these clas- ined [20, 21]. Notably, however, in pilot studies with small sified as familial based on the inheritance of rare genetic numbers of animals that express WT human SOD1, the variants [2]. Approximately 15% of familial ALS (fALS) G37R variant, or the G93A variant of SOD1, we did not and 2% of sporadic cases are associated with mutations observe the same robust accelerated onset of paralysis or in the gene encoding the antioxidant enzyme known as induction of more severe inclusion pathology [15]. superoxide dismutase 1 (SOD1) [2]. To date, more than The present study describes a broader analysis of seed - 170 missense mutations at more than 80 different amino ing across a larger panel of SOD1 variants to character- acids of this 153 amino acid protein have been associated ize the influence of primary sequence on the efficacy of with ALS (https:// alsod. uk). Though the effects of most seeding to induce motor neuron disease in SOD1 trans- of these mutations on enzymatic function have not been genic mice. We examine both isologous and heterologous characterized, studies of a random subset of mutations seeding efficiencies across multiple lines of SOD1 trans - have shown that many disease-associated mutants of genic mice. Our findings confirm that mice expressing SOD1 retain high enzymatic activity [3]. These mutants WT, G93A, or G37R human SOD1 are relatively resist- have been classified as wild-type-like (WT-like) mutants. ant to isologous prion-like seeding. Mice expressing Other mutations, however, can be highly destabilizing to the G85R and L126Z variants of fALS were susceptible the normal protein conformation and render the enzyme to both isologous and heterologous seeding by tissue inactive [3]. Work from multiple laboratories including homogenates from paralyzed mice expressing mutant ours has demonstrated that mutations associated with human SOD1. We noted that seeds prepared from mice ALS cause conformational changes in SOD1 that induce or rats that express the G37R or H46R variants of slowly the protein to misfold and self-associate into insoluble progressing ALS were among the least efficient in induc - aggregates and pathological inclusions (reviewed in [4]). ing early paralysis in G85R-SOD1:YFP mice. Notably, we Indeed, SOD1 inclusion pathology is a common feature observed that mice expressing human G85R SOD1, or of SOD1-linked ALS [5]. Whether misfolded SOD1 is G85R-SOD1:YFP, were resistant to seeding by prepara- also a common feature of sporadic ALS is less certain as tions from spinal cords of paralyzed mice over-expressing there have been contradictory reports in the literature mouse SOD1 with the same mutation. Other examples of [6–13]. Importantly, studies in cell culture models have sequence specificity in seeding implicated an amyloido - established the potential for misfolded WT SOD1 to genic element bordered by amino acids 31–38 as poten- propagate between cells, leading to the hypothesis that tially important in the propagation of misfolded SOD1 prion-like propagation of misfolded WT SOD1 could be conformations. Collectively, our studies demonstrate that involved in the progressive spread of weakness in spo- the primary sequence of SOD1 exerts significant influ - radic ALS patients [6, 14]. ence over the efficacy of the seed preparation and the We have previously demonstrated that injecting spi- susceptibility of the host-recipient in the prion-like prop- nal cord homogenate prepared from paralyzed mutant agation of misfolded SOD1 conformations. SOD1 transgenic mice can accelerate motor neuron dis- ease (MND) in transgenic mice expressing the G85R vari- Methods ant of SOD1 [15–17]. To date, most of the work in our Mice. All of the transgenic mice expressing human laboratory has used a line of mice that express the G85R SOD1 variants that were used in this study as recipients variant of human SOD1 fused to YFP (G85R-SOD1:YFP) of seeding injections have been previously described: that is largely free of disease until 20–24  months of age. G85R-SOD1:YFP and WT-SOD1:YFP mice [22], Injection of tissue homogenates, or purified protein, con - L126Z Line 45 mice [20], G37R Line 29 mice [23], PrP. taining misfolded SOD1 into the spinal cords of newborn G37R Line 110 mice [24], QV103Z hSOD1 Line D14 mice, or injecting misfolded SOD1 into the sciatic nerve mice [21], G85R Line 148 mice [25], GurWT mice of young adult mice, induces paralysis 2–15 months post- (B6SJL-Tg(SOD1)2Gur/J; stock no. 002297, Jackson injection [15–17]. Using mice that express untagged Laboratories) [26], Thy1-G93A Line T3 mice (FVB(Cg)- human G85R SOD1, Bidhendi and colleagues demon- Tg(y1- Th SOD1*G93A)T3Hgrd/J; stock no. 008230, strated accelerated onset of paralysis in mice when young Jackson Laboratories)[27], and VLE G93A mice (B6SJL- adult animals were similarly injected with preparations Tg(SOD1*G93A)1Gur/ThpaJ; stock no. 032166, Jackson derived from spinal cords of paralyzed mutant SOD1 ani- Laboratories [28]. The G85R-SOD1:YFP, WT-SOD1:YFP, mals or from human SOD1 ALS cases [18, 19]. We have and Thy1-G93A mice were maintained on the FVB/ also observed accelerated onset of paralysis in mice that NJ background. The GurWT mice were backcrossed to express untagged versions of two other mutant human B6/C3F1 hybrid mice for more than 10 generations and SOD1 truncation variants [17], which are among the maintained on this background. The VLE G93A mice A yers et al. acta neuropathol commun (2021) 9:92 Page 3 of 24 were maintained on the B6SJL hybrid background. All detect emission using the Gen5 software (v1.10.8). Incu- other lines of mutant SOD1 mice were maintained on bations were halted when maximum fluorescence was a hybrid background of C57Bl/6J and C3H/HeJ, which achieved. The presence of aggregated SOD1 was con - were the strains used in the initial generation of the mice. firmed by filter trap assay and electron microscopy as For identification of genotype, DNA was extracted from previously described [17]. mouse tail biopsies and analyzed by PCR as previously Seeding Experiments. Newborn mice were injected described [20, 24]. intraspinally (ISP) as previously described [15]. Briefly, Spinal tissues were harvested from transgenic animals P0 neonatal pups were placed in aluminum foil and sur- in our in-house colonies that became paralyzed. In addi- rounded with ice until movement ceased and skin tone tion to the mice listed above, the source for tissues from became cyanotic (5–10  min). Using a 10  μl syringe MoG86R Line M1 was FVB-Tg(SOD1*G86R)M1Jwg/J equipped with a 1-inch, 30-gauge needle with 30° bevel, (stock no. 005110, Jackson Laboratories, Bar Harbor ME) the needle was inserted through the skin at midline, [29]. The original source for GurG93A mice was B6SJL- about 5  mm from the base of the tail, and then into the Tg(SOD1*G93A)1Gur/J (stock no. 002726 Jackson Labo- vertebral column before 1 μl of the inoculum was slowly ratories) [26]. The GurG93A mice were backcrossed to injected. For cerebral ventricular injections, 2 μl of inoc- B6/C3F1 hybrid mice for more than 10 generations and ulum was slowly injected into each cerebral ventricle by maintained on this background in our colony. Spinal tis- inserting the needle approximately 2  mm into the skull, sues from paralyzed H46R rats [30] were a kind gift of Dr. penetrating the skull near bregma. For bilateral muscle Christine Vande Velde (University of Montreal, CHUM injections, 2  μl of inoculum was slowly injected into the Research Center, Montreal Canada). Spinal tissues from lower hindlimb targeting the gastrocnemius muscles. paralyzed G85R mice (Line 148) [25] were a kind gift of The accuracy of these techniques of injection was veri - Dr. Don Cleveland (University of California San Diego, fied in separate cohorts of newborn mice injected with San Diego CA). PBS containing 1% Evans blue dye [32]. After injections, All animals were housed one to five to a cage and main - pups were allowed to completely recover on a warming tained on ad libitum food and water with a 14 h light and blanket and then returned to the home cage where they 10 h dark cycle. were monitored to ensure full mobility and no signs of Preparation of inoculum. Spinal cord tissues were impairment. homogenized in PBS creating a 10% homogenate (w/v), Sciatic nerve injections were performed as previously containing 1:100 v/v protease inhibitor cocktail (Sigma, described [16]. Prior to the injection, mice were injected St. Louis, MO) as previously described [15]. Tissues were subcutaneously with 2  mg/kg Meloxicam (Norbrook, disrupted by sonication 4 times for 20 s each, with cool- Overland Park, KS, USA) to relieve pain and the injection ing on ice between bouts of sonication. Homogenates site was shaved and sterilized. Mice were anesthetized were then clarified by a low-speed spin at ~ 800 × g for with isoflurane and a small incision was then made in the 10  min and the supernatants were aliquoted and placed skin of the hindlimb before the sciatic nerve was exposed at -80 °C. In studies involving human tissues, we concen- at the popliteal fossa. A 30-gauge needle containing the trated 200 µl of the clarified homogenates by centrifuga - inoculum was inserted in the sciatic nerve and recipro- tion in an AirFuge at maximum speed for 20  min. The cated 10 times, which has been shown to greatly enhance resulting pellet was resuspended in 40 µl of PBS with the the efficiency of prion transport to the spinal cord [33]. protease inhibitor cocktail and solubilized by pulse soni- Two microliters of homogenate were injected under the cation as described above. perineurium of the sciatic nerve and the incision was Recombinant SOD1 purification and fibrillization. then closed with stainless steel clips and cleaned. Follow- Recombinant hSOD1 proteins were expressed and puri- ing surgery, 2  mg/kg Meloxicam was administered at 24 fied as previously described [31]. Fibrillar aggregates of and 48 h post-surgery. purified SOD1 were generated in 200  µl solutions con - Tissue collection. Mice were anesthetized with isoflu - taining 50  μM of protein in 20  mM potassium phos- rane and perfused transcardially with 20  ml of PBS. The phate, pH 7.2 with the addition of 10  mM TCEP. The spinal cord and brain were immediately removed and protein solutions were incubated in a 96-well plate with the brains were bisected sagittally with one hemisphere the addition of a Teflon ball (1/8-in diameter) at 37  °C drop fixed in 4% paraformaldehyde in PBS for 24–48  h with constant agitation in a Synergy HT plate reader at 4  °C and the other flash frozen on dry ice. The spinal (BIO-TEK, Winooski, VT). Parallel wells containing the columns were removed and cut into 4 sections. Cervi- same components with 4  μM Thioflavin T were moni - cal and lumbar segments were de-roofed and drop fixed tored by fluorescence measurements every 15 min using in 4% paraformaldehyde in PBS for 24–48 h at 4 °C. The a λex = 440/30 filter to excite and a λem = 485/20 filter to other segments were dissected and flash frozen on dry Ayers et al. acta neuropathol commun (2021) 9:92 Page 4 of 24 ice. Frozen tissues were stored at −80 °C. Fixed tissues appeared as dark structures of various sizes, including were either embedded in paraffin for sectioning at 7  µm small neuropil puncta, fiber-like structures in the neuro - or immersed in 30% sucrose in PBS, mounted in OCT pil, and accumulations of puncta within cell bodies. media (Sakura, The Netherlands) and sectioned to 30 μm using a cryostat. The sections were placed in a dish con - Results taining anti-freeze solution (100  mM sodium acetate, We have demonstrated previously that intraspinal injec- 250 mM polyvinylpyrrolidone, 40% ethylene glycol) at pH tion of spinal homogenates from paralyzed mutant SOD1 6.5, and stored at 4 °C. Sections were then mounted onto mice into newborn G85R-SOD1:YFP mice accelerates slides, air-dried overnight, and cover-slipped in mount- the onset of paralysis and intraspinal SOD1 inclusion ing media containing DAPI (Vector, Burlingame, CA). pathology [15, 17]. Importantly, we have previously dem- Neuropathology. For imaging YFP fluorescence in par - onstrated that paralysis in G85R-SOD1:YFP mice is not affin embedded tissues or from cryostat sections, images induced by injection of PBS or spinal tissue homogen- were captured by epifluorescence on an Olympus BX60 ates from various types of controls (Additional file  2: microscope. Table  S1) [17]. The controls tested include homogenates For immunohistochemistry and histochemical stain- from non-transgenic mice, and from transgenic mice that ing, we examined paraffin embedded tissues. For immu - express mutant human tau or mutant human α-Synuclein nostaining with C4F6 antibody (Medimabs, Montreal, (αSyn), and which develop motor phenotypes. Other Quebec, Canada), deparaffinized sections were incubated controls include spinal homogenates from young asymp- in 95% formic acid for 10  min and then washed in PBS. tomatic G85R-SOD1:YFP mice. Additionally, nontrans- Sections were incubated in 0.3% H O in PBS for 20 min genic (NTg) littermates injected with homogenates 2 2 and then incubated in PBS-T with 3% normal goat serum from paralyzed mutant SOD1 do not develop paralysis for 1  h before the primary antibody C4F6 was added at or exhibit evidence of inclusion pathology (Additional a dilution of 1:500 and incubated at 4  °C overnight. The file  1: Data File 1 – NTg data tab). Collectively, these pre- sections were then incubated with a biotinylated sec- vious studies have established the specificity with which ondary anti-mouse antibody (Vector Laboratories, Burl- spinal homogenates from paralyzed mutant SOD1 mice ingame, CA) diluted 1:500 in PBS-T with 3% normal goat induce an accelerated paralysis, and inclusion pathology, serum followed by incubation with the ABC-horserad- when injected into the spinal cords of newborn G85R- ish peroxidase staining kit (Vector Laboratories). Sec- SOD1:YFP mice. tions were developed using the DAB staining kit (KPL, Gaithersburg, MD) and counterstained with hema- Comparative analysis of seeding efficiency by spinal toxylin. Images were captured using an Olympus BX60 homogenates from paralyzed G37R and G93A SOD1 mice microscope. In the present study, we have focused on examining the SOD1 inclusions were also visualized by Campbell- role of the primary sequence of the SOD1 seed and the Switzer silver staining using a protocol provided by Dr. SOD1 gene expressed by the recipient in the efficiency Robert Switzer (NeuroScience Associates, Knoxville, of prion-like seeding. We have previously reported that TN; Campbell S, Switzer R, Martin T (1987) Alzhei- spinal homogenates from paralyzed mice expressing the mer’s plaques and tangles: a controlled and enhanced sil- G37R variant, which is associated with disease durations ver staining method. Soc Neurosci Abst 13:678) [34]. A in humans that average 18.7  years [35], can seed early subset of the silver-stained sections was counterstained paralysis and pathology in G85R-SOD1:YFP mice [15, with hematoxylin and all slides were cover slipped using 17]. Here, we have reanalyzed previously published data, ™ ™ Thermo Scientific Cytoseal 60 mounting medium. along with additional new data, to assess the variability Images were obtained using an Olympus BX60 micro- and efficiency of G37R-SOD1 seeds in G85R-SOD1:YFP scope or an Aperio Scanscope XT image scanner (Aperio, mice. Overall, homogenates prepared from paralyzed Vista, CA, USA). Images captured by the scanner were G37R mice induced early paralysis in only 4 of the 15 digitally cropped to generate final images. injected G85R-SOD1:YFP mice (Fig. 1a, black circles). All The severity of inclusion pathology revealed by direct the mice that developed paralysis also developed inclu- visualization of fluorescence or after silver staining was sion pathology, which appeared as fluorescent puncta in scored by a blinded observer (Additional file  1: data file the neuropil with variable levels of accumulation in cell 1). In the G85R-SOD1:YFP mice inclusions appeared as bodies (Fig. 1b, c). Additionally, there were accumulations fluorescent puncta in the neuropil, punctate or fibrillar within processes that appeared as fiber-like structures accumulations in the cell body, or as fiber-like structures in the neuropil (Fig.  1c–g). The initial description of the in the neuropil due to accumulations within processes G85R-SOD1:YFP mice demonstrated that this transgene (axons or dendrites). In silver stained sections, inclusions is highly expressed in neurons, with minimal evidence A yers et al. acta neuropathol commun (2021) 9:92 Page 5 of 24 of astrocytic expression [22]. Therefore, it seems likely As compared to seeding with spinal homogenates from that most of the inclusion pathology occurs in neurons paralyzed G37R mice, the efficiency of seeding paralysis or neuronal processes. Notably, there were two asymp- by homogenates from paralyzed G93A mice was more tomatic animals that were euthanized at 19  months that consistent. In experiments using 4 independently gen- exhibited extensive inclusion pathology (Fig.  1a gray erated homogenates, we observed paralysis in at least circles; Fig.  1d). Curiously, all 4 of the G85R-SOD1:YFP half of the injected G85R-SOD1:YFP recipients for each mice that developed paralysis after seeding were from a cohort (Fig. 2a, black circles). Mice that developed paral- single litter. In the subset of mice that developed paraly- ysis tended to do so before the age of 12 months (Fig. 2a). sis, the average age to paralysis was 12.2 months, but the In three of the 4 experiments, the average age to paraly- range was from 6.5 to 18 months (Additional file  1: Data sis was 5.7, 4.3, and 4.4 mo post-injection (Additional File 1). file  1: Data File 1). In cohort #4, three of the 7 mice that In prior studies, we have demonstrated that the effi - were injected developed paralysis before 16  months of ciency of seeding improved when we passaged G37R age when we terminated the experiment (average age to seeds from these initial G85R-SOD1:YFP mice to naïve paralysis was 10 mo post injection) (Fig.  2a, G93A Sp recipients [17]. To assess the variability of second pas- Cord #4). Of the 25 total mice that were injected with sage G37R seeds, we compared second passage data G93A seeds, 11 lived to 12  months post-injection with- for 3 different isolates. Second passage of seeds from out developing paralysis (Fig.  2a; Additional file  1: Data the 19-month-old asymptomatic animal that exhibited File 1). All mice that developed paralysis exhibited inclu- inclusion pathology (see Fig.  1d) induced paralysis in sion pathology within the spinal cord (Fig.  2b; Addi- 3 of 5 injected G85R-SOD1:YFP mice at an average age tional file  1: Data File 1). Interestingly, as was the case of 12  months post-injection (Fig.  1a). Second passage in G37R first-passage mice, we observed several cases in of seeds from an 18-month-old symptomatic animal which asymptomatic mice were euthanized at advanced that exhibited inclusion pathology (Fig.  1c) also induced ages (> 12  months) and found to exhibit extensive inclu- paralysis in 4 of 7 injected recipients at an average age sion pathology (Fig.  2c; Additional file  1: Data File 1). of 17  months post-injection (Fig.  1a). Second passage of As previously reported [15, 17], second passage of seeds G85R−SOD1:YFP seeds from a 6.5 month old symptomatic animal induced from P1-G93A recipients that became para- paralysis in 6 of 7 injected recipients at an average age of lyzed produced more efficient induction of paralysis at 6.2  months post-injection (Fig.  1a) [17]. Passage of this earlier ages (Fig.  2a) with inclusion pathology (Fig.  2d). isolate for a third time produced paralysis in 7 of 8 recipi- The average age to paralysis for second passage G93A ents at an average of 6.2 months post-injection (Fig.  1a). seeds was ~ 3  months post-injection (Additional file  1: All the paralyzed mice in second or third passage experi- Data File 1). These findings are consistent with the idea ments exhibited inclusion pathology to varying degrees that the G93A variant of SOD1 misfolds into a confor- (Fig.  1e–h). The level of inclusion pathology seen in the mation that produces relatively efficient seeds, imprint - 19-month-old P1 animals that were used to prepare ing conformations to G85R-SOD1:YFP that can produce seeds for second passage (Fig.  1d) was not obviously dif- relatively short incubation periods as seeds in second ferent from that of similarly aged paralyzed mice (com- passage. pare Fig.  1b–d). Overall, the data are consistent with the idea that the efficiency of seeding by homogenates con - Low seeding efficiency by spinal homogenates taining the G37R variant of misfolded SOD1 is relatively from paralyzed H46R‑SOD1 transgenic rats low and that seeds imprinted by the G37R conformation The average duration of disease in individuals with the can exhibit relatively long incubation periods in second H46R mutation is 17.0  years [35]). Spinal cords from passage. paralyzed rats that were heterozygous or homozygous (See figure on next page.) Fig. 1 Spinal homogenates from paralyzed G37R SOD1 mice seed G85R‑SOD1:YFP mice inefficiently with variable incubation periods. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates. Scatter plots in this figure G85R− and figures that follow were generated in GraphPad Prism v9. TgG37R denotes spinal homogenate from paralyzed G37R‑Line 29 mice. P1‑ G37R SOD1:YFP G85R−SOD1:YFP denotes inoculum from a paralyzed first ‑passage G85R‑SOD1:YFP animal. P2‑ G37R denotes inoculum prepared from a paralyzed second‑passage G85R‑SOD1:YFP animal. The dashed lines with arrows mark animals from the initial passage that were used to prepare inoculum for second passage. b–h Representative images of inclusion pathology in cryostat sections that were induced in the spinal cord of G85R‑SOD1:YFP mice that had been injected intraspinally (ISP) with spinal homogenates from G37R mice with subsequent passages (images representative of 2–3 sections per mouse). The arrows identify pathological specimens that are related due to passage. b–d Pathology specimens from first passage rd animals. e–g Representative pathology specimens from 2nd passage animals. h A representative image of pathology specimens from 3 passage animals. Raw data provided in Additional file 1: Data File 1. Scale bars = 50 µm Ayers et al. acta neuropathol commun (2021) 9:92 Page 6 of 24 for the transgene (described in [30]) were homog- developed paralysis and the experiment was terminated enized and injected into the spinal cords of newborn at 15–16  months post-injection to examine pathol- G85R-SOD1:YFP mice. None of the injected animals ogy (Fig.  3a), finding that most of these asymptomatic A yers et al. acta neuropathol commun (2021) 9:92 Page 7 of 24 Fig. 2 Variability in seeding efficiency of spinal homogenates from paralyzed G93A SOD1 mice in G85R‑SOD1:YFP mice. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates. TgG93A denotes spinal homogenate G85R−SOD1:YFP from paralyzed GurG93A mice. P1‑ G93A denotes inoculum from a paralyzed first ‑passage G85R‑SOD1:YFP animal. b–d Representative images of inclusion pathology in cryostat sections induced in the spinal cord of G85R‑SOD1:YFP mice that had been injected with spinal homogenates from G93A mice with subsequent passages (images representative of 2–3 sections per mouse). Raw data provided in Additional file 1: Data File 1. Scale bars = 50 µm mice injected with homogenate from the homozygous paralysis in G85R-SOD1:YFP mice rather inefficiently, H46R rats had developed inclusion pathology (Fig. 3b). despite a capability to induce inclusion pathology. To assess whether a higher dose of H46R seeds may be more effective, we concentrated spinal homogenates Seeding G85R‑SOD1:YFP mice with recombinant SOD1 from paralyzed homozygous H46R rats by fivefold and fibrils injected them into newborn G85R-SOD1:YFP mice. In prior studies, we have demonstrated the induction of One animal of the 5 that were injected developed fore- paralysis and inclusion pathology in G85R-SOD1:YFP limb weakness at 7.9  months of age, and this animal mice by the injection of fibrils produced by recombi - exhibited inclusion pathology (Fig.  3c). The remaining nant human SOD1 (WT or G93A variants) [17, 36]. To 4 animals from this cohort were asymptomatic when expand the number of different sequence variants we we terminated the experiment at 12  months of age, could assess in seeding, we produced recombinant SOD1 and lacked evidence of inclusion pathology (Fig.  3a; encoding eight different fALS mutations and fibrilized Additional file  1: Data File 1). These data suggest these proteins in  vitro to produce inoculum for injec- that homogenates from paralyzed H46R rats induce tion (Table  1). To confirm the formation of fibrils, we analyzed the samples both by electron microscopy (EM) Ayers et al. acta neuropathol commun (2021) 9:92 Page 8 of 24 Fig. 3 Spinal homogenates from paralyzed H46R rats seed G85R‑SOD1:YFP mice inefficiently. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates from paralyzed H46R rats. b, c Representative images of inclusion pathology seen in paraffin sections from the spinal cord of G85R‑SOD1:YFP mice that had been injected with spinal homogenates from heterozygous and homozygous paralyzed H46R rats, with a fivefold concentrated inoculum prepared from homogenate from a paralyzed homozygous rat (images representative of 2–3 sections per mouse). Raw data provided in Additional file 1: Data File 1. Scale bar = 100 µm (b) or 50 µm (c) (Fig.  4a–i; Additional file  2: Fig. S1) and by using a cel- are associated with rapidly progressing disease, we only lulose filter trap assay (Table  1). Somewhat surprisingly, observed debris that was similar to what we observed over multiple assays, we were unable to detect the gen- in the control blank (Fig.  4a, f, and i; Table  1). Based on eration of fibrillar aggregates by 2 of the 7 mutant SOD1 prior studies of the H46R variant in vitro [37, 38], we had proteins examined (Table 1; G85R mutant was not exam- expected poor fibrilization of this variant in our experi - ined by EM). For the G41S and D101N mutants, which ments. However, we were able to produce fibrilized H46R A yers et al. acta neuropathol commun (2021) 9:92 Page 9 of 24 Table 1 Summary of data for recombinant SOD1 protein H46R, and other seeds, may further reveal unique attrib- seeding experiments utes of this variant. Mice that displayed paralysis from injection of these SOD1 protein Avg survival Fibrils visible Induced time in years by EM pathology in seeds were found to produce inclusion pathology (Addi- G85R‑ tional file  2: Fig. S2; Additional file  1: Data File 1). The SOD1:YFP inclusion pathology induced by recombinant WT-SOD1 slice culture fibrils was distributed in cell bodies and neuropil (Addi - WT N/A Yes Yes tional file  2: Fig.  2a arrows; also see [17]); whereas, for A4V 1.2 ± 0.9 Yes Yes all of the mutants, the dominant pathology was located G37R 18.7 ± 11.4 Yes Yes in the neuropil, and appeared as fiber-like and punctate G41S 0.9 ± 0.2 No Yes structures (Additional file  2: Fig.  2b-i). Interestingly, all H46R 17.0 ± 7 Yes Yes but one of the 17-month-old asymptomatic mice injected G85R 6.0 ± 4.5 N.D Yes with recombinant H46R fibrils were found to have high G93C 13.0 ± 0.4 Yes Yes levels of inclusion pathology (Fig.  4j; Additional file  2: E100K 12.0 ± 4.1 Yes Yes Fig. S2e). Collectively, these findings demonstrate that D101N 2.4 ± 0.9 No Yes purified recombinant SOD1 with mutations associated Data originally reported in[35] with fALS can produce prion-like seeds that can acceler- Inclusive of data originally reported in[17] ate paralytic disease in the G85R-SOD1:YFP model. The efficiency of seeding did not obviously relate to whether the injected recombinant protein formed large visible human SOD (Fig. 4e). Prior studies have noted that SOD1 fibrils, or whether the mutant was associated with rap - fibrilization can be somewhat stochastic and in competi - idly or slowly progressing disease. While the behavior of tion with the formation of amorphous aggregates [39]. the recombinant H46R-SOD1 seeds was similar to that Due to the uncertainty surrounding the SOD1 species of spinal homogenates from H46R-SOD1 rats, the effi - responsible for prion-like seeding in G85R-SOD1:YFP ciency of seeding with recombinant G37R-SOD1 seeds mice, we decided to test all 8 of the SOD1 mutants, using was relatively high with incubation periods that were protein that had been treated in a manner to produce similar to variants associated with rapidly progressing fibrilization. Initially, we used a spinal cord slice culture disease (Fig.  4j). Apart from the H46R variant, we did model from the G85R-SOD1:YFP mice to assess patho- not observe an obvious distinction in the performance of logical seeding activity [17]. We treated the slice cultures recombinant SOD1 seeds from slowly progressing vari- with 1  μl of each of the proteins (50  μM) and incubated ants (G37R, G93C, E100K) as compared to seeds from the sections for one month while monitoring for the rapidly progressing variants (A4V, G41S, D101N). induction of SOD1-YFP inclusion pathology. All eight of the mutant SOD1 protein preparations were observed to Low seeding efficiency with spinal homogenates induce the misfolding of G85R-SOD1:YFP in the slice cul- from older GurWT mice ture model (Table  1; primary data not shown). We then Previously, we reported that injection of spinal injected the mutant recombinant SOD1 preparations homogenates from older GurWT mice into P0 G85R- into the spinal cords of newborn G85R-SOD1:YFP mice SOD1:YFP mice produced paralysis with inclusion to assess whether we could induce accelerated paralysis. pathology in one of the 3 injected animals [15]. One of All preparations injected were capable of inducing early these 3 injected animals was euthanized at 20  months disease with the age to paralysis ranging between 8- and of age with no signs of paralysis or pathology (Fig.  5a, 16-months post-injection (Fig.  4j). The efficacy of induc - b) and the other was found dead at 19  months of age ing paralysis across the mutants was similar to WT SOD1 with no prior signs of paresis. To further examine fibrils with the notable exception in that most of the the seeding activity of WT-SOD1 in spinal cords of mice injected with fibrilized recombinant H46R-SOD1 aged GurWT mice, we injected 9 G85R-SOD1:YFP reached a pre-determined aging endpoint of 17  months mice. One of these animals developed abnormali- without developing symptoms (Fig.  4j). Further study in ties at 12  months of age that were described as asym- second and third passage experiments with spinal cords metrical weakness with swelling in the weak limb and of G85R-SOD1:YFP mice seeded with recombinant partial paralysis. Despite the suggestion that ALS-like Ayers et al. acta neuropathol commun (2021) 9:92 Page 10 of 24 Fig. 4 Analysis of the seeding efficiency of recombinant SOD1 fibrils in G85R‑SOD1:YFP mice. a–i Representative images of fibrils formed by recombinant SOD1 aggregated in vitro. j Scatter plot of the ages at which the G85R‑SOD1:YFP mice developed paralysis or were euthanized after intraspinal injection of recombinant SOD1 fibrils (P0 injection). Raw data provided in Additional file 1: Data File 1. All scale bars (below each panel) represent 500 nm disease was induced, this animal exhibited only sparse puncta (Additional file  1: data file 1). To follow up on fluorescent puncta (Fig.  5c). Another animal developed the one G85R-SOD1:YFP animal that developed paral- partial paralysis at 13 months of age with signs of with ysis by seeding with GurWT spinal homogenates, we injuries to the hindlimbs from fighting or self-mutila - attempted to passage seeds from this animal into naïve G85R−SOD1:YFP tion. This animal also showed only sparse fluorescent G85R-SOD1:YFP (P1-WT ). In a small A yers et al. acta neuropathol commun (2021) 9:92 Page 11 of 24 Fig. 5 Spinal homogenates from aged GurWT SOD1 mice seed G85R‑SOD1:YFP mice inefficiently. a Scatter plot of the ages at which G85R‑SOD1:YFP mice developed paralysis or were euthanized after injection of spinal homogenates. TgGurWT denotes spinal homogenate from G85R−SOD1:YFP aged GurWT mice. P1‑ GurWT denotes inoculum from a paralyzed first ‑passage G85R‑SOD1:YFP animal that has been previously described [15]. The data graphed here include 3 first ‑passage animals described in [15]. b–d Representative images of inclusion pathology seen in cryostat sections from the spinal cord of G85R‑SOD1:YFP mice that had been injected with spinal homogenates from aged GurWT mice or with homogenate from the one first ‑passage animal that developed paralysis (images representative of 2–3 sections per mouse). Raw data provided in Additional file 1: Data File 1. Scale bars = 20 µm cohort of three mice, none developed paralysis, and exposure of the G85R-SOD1:YFP protein to a misfolded when these animals were euthanized at 16.4  months untagged mutant SOD1. We have previously observed of age we once again observed only sparse fluorescent that co-expression of human G93A-SOD1 with G85R- puncta (Fig.  5d). These data indicate that the efficiency SOD1:YFP produced abundant YFP inclusion pathology of seeding paralysis, or inclusion pathology, in G85R- [15]. This outcome presaged our observation that injec - SOD1:YFP mice by injecting spinal cords of older tion of spinal homogenates from paralyzed G93A mice GurWT mice is relatively low. induced an early paralysis with inclusion pathology in G85R-SOD1:YFP mice [15]. We were therefore interested Comparison of prion‑like propagation to transgenic to examine other mutants in co-expression models. Con- co‑expression of WT and mutant SOD1 sistent with L126Z-SOD1 seeding of G85R-SOD1:YFP with G85R‑SOD1:YFP described previously [17], we found that co-expression The co-expression of mutant SOD1 with G85R- of the L126Z-SOD1 variant (Line 45) at levels sufficient SOD1:YFP presents a scenario of sustained high-level to cause disease was able to induce the aggregation of Ayers et al. acta neuropathol commun (2021) 9:92 Page 12 of 24 Fig. 6 Induction of G85R‑SOD1:YFP inclusion pathology by co ‑ expression of untagged mutant SOD1. a–c Kaplan–Meier survival curves and representative pathologic images resulting from G85R‑SOD1:YFP mice crossed with mice expressing untagged L126Z (Line 45) (L126Z mice n = 3, G85R‑ YFP mice n = 5, L45xG85R‑ YFP mice n = 6); untagged G37R (Line 29) (G37R mice n = 6, G85R‑ YFP mice n = 10, L29xG85R‑ YFP mice n = 6), or untagged WT human SOD1 (GurWT n = 14, G85R‑ YFP n = 5, GurWTxG85R‑ YFP n = 8), respectively. Survival plots in this figure and figures that follow were generated in GraphPad Prism v9. Representative images of inclusion pathology seen in cryostat sections from the spinal cord of paralyzed bigenic mice created in each crossing experiment (images representative of 2–3 sections per mouse). Scale bars = 50 µm G85R-SOD1:YFP (Fig.  6). Notably, the age to paralysis induced pathology was a mixture of fibrillar and punctate was modestly earlier in L126Z/G85R-SOD1:YFP mice structures. Our findings from these crossing experiments relative to littermates that expressed the L126Z-SOD1 would suggest that seeds prepared from older GurWT alone (Fig.  6a). In the bigenic mice, we observed abun- mice should have been more effective than observed. dant fibrillar YFP containing inclusion pathology that resembled pathology induced in G85R-SOD1:YFP mice Comparison of injection routes by seeding with spinal homogenates from paralyzed In a previous study, we have shown that injection of tis- L126Z-SOD1 mice [17]. Bigenic mice co-expressing sue homogenates from paralyzed mutant SOD1 mice into G37R-SOD1 (Line 29) and G85R-SOD1:YFP developed sciatic nerve directly can seed earlier onset of paralysis disease much earlier than mice expressing G37R-SOD1 with intraspinal inclusion pathology [16]. To determine alone (Fig. 6b). Paralyzed bigenic G37R/G85R-SOD1:YFP whether we would achieve the same efficiency of seed - mice exhibited robust fluorescent inclusion pathology ing by intramuscular injection, we injected the hindlimb (Fig. 6b). Bigenic mice co-expressing WT-SOD1 (GurWT muscles of newborn G85R-SOD1:YFP mice with tissue line) with G85R-SOD1:YFP also developed paralysis at homogenates from paralyzed G93A mice. In this cohort, early ages with robust inclusion pathology (Fig.  6c). The we observed that all five mice lived to approximately A yers et al. acta neuropathol commun (2021) 9:92 Page 13 of 24 Fig. 7 Inefficient seeding of human G85R‑SOD1 mice by misfolded murine G86R‑SOD1. a Kaplan–Meier survival curves resulting from untagged G85R SOD1 mice injected with spinal homogenates from paralyzed G85R SOD1 mice, paralyzed murine G86R SOD1 mice, NTg mice, or uninjected mice. b Representative images of inclusion pathology seen by silver staining of paraffin sections from paralyzed recipient G85R mice (images representative of 2–3 sections per mouse). Scale bars = 20 µm 20  months of age without developing paralysis (Addi- newborn G85R-SOD1:YFP mice. Using spinal homoge- tional file  2: Fig. S3). Notably, however, one of these mice nates from G85R:YFP mice that had been injected with G85R−YFP developed inclusion pathology and hence there may have G93A homogenate (P1-G93A mice) as seeds, been a low level of seed uptake at nerve terminals, result- we examined the efficacy of the ICV route in inducing ing in inefficient seeding and reduced penetrance (Addi - paralysis. In a cohort of 6 animals, we noted that the tional file 2: Fig. S3). age of paralytic onset was similar to what was observed We next examined the relative efficacy of intracer - in ISP injected mice (range of 4–10  months) (Addi- ebral ventricular (ICV) injection of SOD1 seeds into tional file  2: Fig. S3). All paralyzed animals exhibited Ayers et al. acta neuropathol commun (2021) 9:92 Page 14 of 24 abundant inclusion pathology. Interestingly, in the ICV mutants (L126Z and QV103Z) show accelerated paralysis seeding paradigm, all the mice first presented with fore - after isologous seeding [17]. Both of these lines of mice limb weakness, suggesting a rostral to caudal spread of will develop paralysis on their own, with the L126Z mice the disease-causing conformation. developing disease at 7–9 months of age and the QV103Z mice developing paralysis at > 14  months of age [20, 21]. Isologous and heterologous seeding in untagged G85R In subsequent studies of seeding in these lines of mice, SOD1 mice we have now found an unexpected distinction in seeding As previously reported [19], mice expressing untagged activity. L126Z-SOD1 mice responded to seeding with G85R-SOD1 were susceptible to seeding by spinal homogenates from paralyzed L126Z or QV103Z mice homogenates prepared from paralyzed G85R-SOD1 (Fig.  8a), with the paralyzed mice showing varied levels mice. The line of G85R-SOD1 mice used here develops of silver positive inclusion structures (Fig.  8b–d; Addi- paralysis between 11 and 14 months of age [25] (Fig. 7a). tional file  2: Fig. S6). By contrast, QV103Z mice were not Although there was some variability in the age to paral- responsive to seeding by homogenates from L126Z mice ysis in the seeded G85R mice (range 2.1–6  months), all (Fig.  8e). In silver stains of the asymptomatic QV103Z 11 of the injected G85R mice developed paralysis early mice we observed occasional red blood cells (Fig.  8f, (Fig. 7a). By contrast, spinal homogenates from mice that g), whereas the paralyzed mice seeded with QV103Z express a version of murine Sod1 with the G85R muta- homogenates exhibited obvious inclusion pathology tion (originally noted as G86R by codon numbering (Fig.  8h; Additional file  2: Fig. S7). These studies suggest [29]), did not induce earlier paralysis in human G85R- that QV103Z seeds can propagate misfolded conforma- SOD1 mice (Fig.  7a). G85R-SOD1 mice injected with tions to L126Z host SOD1, but when the QV103Z-SOD1 spinal cord homogenates from paralyzed MoG86R-Sod1 is the host, the conformation carried by L126Z seeds can- mice developed paralysis in the same time frame as mice not propagate. injected with nontransgenic (NTg) homogenates or PBS (Fig. 7a). The homogenate from paralyzed MoG86R-Sod1 Limited prion‑like seeding in mice expressing WT‑like mice was also ineffective when these seed preparations mutants of SOD1 were injected into G85R-SOD1:YFP mice (Additional We have previously attempted to seed earlier onset dis- file 1: Data File 1; Additional file 2: Fig. S4). ease in Gur1-G93A SOD1 mice G37R-SOD1 mice with- To visualize inclusion pathology in these seeded out success (Additional file  2: Table  S2; data from [15]). untagged G85R mice, we used two methods. The inclu - The GurG93A-SOD1 mice have been reported to have sion pathology in the G85R mice seeded with spinal inherently high levels of misfolded SOD1 from very homogenates from paralyzed G85R mice was immu- young ages [41], and thus it could be argued that these noreactive with C4F6 SOD1 antibody as was pathology mice are not responsive to seeding because the patho- observed in older G85R mice seeded with NTg homoge- genic misfolding of SOD1 is already at saturation. Thus, nates or vehicle control, PBS (Additional file  2: Fig. S5). we turned to lines of mice that express G93A-SOD1 Due to variation in the performance of this antibody, we at much lower levels; Thy1-G93A mice (line T3) [27] switched to a new method of detecting inclusion pathol- and a substrain of the G93A mice termed the very low ogy in untagged SOD1 mice termed Campbell-Switzer expressing (VLE) G93A mice [28]. Mice from the T3 line silver stain [34], which we have recently found to be spe- that are heterozygous for the transgene do not develop cific for mutant SOD1 inclusions [40]. As was seen in paralysis by 20  months of age; whereas, homozygous sections stained by C4F6, silver staining revealed a mix- mice become paralyzed between 15 and 24 months of age ture of neuropil and cell-body inclusions in all paralyzed [27]. Heterozygous Thy-1 G93A mice injected with spinal G85R mice, regardless of age at which paralysis devel- homogenates from paralyzed GurG93A or G37R-SOD1 oped or injection inoculum (Fig.  7b). Importantly, mice mice did not develop paralysis or pathology (Table  2; that became paralyzed early as a result of seeding showed Additional file  2: Fig. S8). The VLE-G93A mice do not abundant inclusion pathology. These findings indicate develop paralysis until > 22 months of age [28]. For these that sequence differences between mouse and human experiments, the source of the seed was spinal homoge- SOD1 inhibit the ability of misfolded G85R mouse Sod1 nates from paralyzed GurG93A-SOD1 (FVB/NJ strain). to seed aggregation of human G85R-SOD1. We used both the newborn intraspinal route of inocu- lation and the sciatic nerve injection route. For the first Isologous and heterologous seeding in the L126Z‑SOD1 time in mice expressing the G93A variant, we observed and QV103Z‑SOD1 mice seeding to produce an earlier onset to paralysis in 6 of In prior studies, we have reported that mice express- the mice injected by the two routes (Table 2). Silver stain- ing untagged mutant SOD1 variants that are truncation ing confirmed that only the paralyzed mice had inclusion A yers et al. acta neuropathol commun (2021) 9:92 Page 15 of 24 Fig. 8 Comparison of the seeding efficiency of L126Z and QV103Z mice in isologous and heterologous seeding. a–d Kaplan–Meier survival curves and representative pathology resulting from untagged L126Z SOD1 mice injected with spinal homogenates from paralyzed L126Z or QV103Z mice, and NTg mice. e–h Kaplan–Meier survival curves and representative pathology resulting from untagged QV103Z SOD1 mice injected with spinal homogenates from paralyzed L126Z or QV103Z mice, and NTg mice. For each animal 2–3 sections were examined to identify representative images. Paraffin sections of spinal cord from each animal was stained by Campbell‑Switzer method as described in Methods . Scale bars = 50 µm pathology (Additional file  2: Fig. S9). Thus, we demon - G37R-SOD1 – Line 110) [24]. Mice from Line 110 that strate that VLE-G93A mice are modestly susceptible to are heterozygous for the transgene develop paralysis at isologous seeding with G93A spinal cord homogenates. 20 + months of age, whereas homozygous mice develop We also extended our investigations in seeding dis- paralysis between 8 and 11 months of age [24]. In a prior ease in mice that express the G37R variant under the study, we reported that we were not able to accelerate transcriptional control of the mouse PrP vector (PrP. disease in this model by isologous seeding using either an Ayers et al. acta neuropathol commun (2021) 9:92 Page 16 of 24 Table 2 Summary of data from seeding Thy1‑ G93A Line T3 mice and VLE‑ G93A Host Inoculum (Route)# inoculated # Age to paralysis or euthanasia paralyzed Thy1‑ G93A PBS 2 0 14.8, 16.6 mo Thy1‑ G93A NTg Sp Cord (ISP) 3 0 17.7, 18.2, 19.1 mo Thy1‑ G93A Gur G93A Sp Cord (ISP) 4 0 Terminated at 19 mo Thy1‑ G93A G37R‑L29 Sp Cord (ISP) 4 0 Terminated at 18 mo VLE‑ G93A G93A(FVB) Sp Cord (ISP) 8 3 5.3, 8.3, 13.4 VLE‑ G93A G93A(FVB) Sp Cord (ScN) 11 3 5.8, 8, 10 VLE‑ G93A PBS (ScN) 5 0 Terminated at 12 mo Mice that were euthanized for non-MND related conditions, such as fight wounds or tumors, before 12 months of age were excluded, unless otherwise noted was hampered by unexpectedly high losses of mice in Table 3 Summary of data from seeding heterozygous PrP.G37R‑ SOD1 Line 110 mice the cohort due to fighting injuries. We also attempted to induce paralysis by seeding mice that over-express Inoculum (ISP) # inoculated Age of euthanasia due untagged WT-SOD1 (GurWT-SOD1 mice), again to age or non‑ ALS health issues observing no induction of paralysis in a small cohort of mice out to 15  months of age (Additional file  2: PBS 5 Terminated at 16 mo Table  S3; Data from [15]). To more rigorously address PrP.G37R‑SOD1 Line 110 6 Terminated at 16 mo whether it is possible to induce paralytic disease in Sp Cord mice that over-express WT-SOD1 we first considered G37R‑L29 Sp Cord 4 Terminated at 16 mo G85R−SOD1:YFP what type of misfolded SOD1 seed would potentially be P1 G37R Sp 7 Terminated at 9 mo Cord efficacious in mice expressing WT-SOD1. GurG93A Sp Cord 3 Terminated at 17 mo To determine what type of seed might be efficacious GurWT Sp Cord 4 Terminated at 16.8 mo in WT-SOD1 mice, we examined pathology in mice that co-express WT-SOD1:YFP with G93A- and WT- SOD1 (Additional file  2: Fig. S11). We have previously reported that WT-SOD1:YFP is not induced to form ICV or a direct sciatic nerve injection route [15]. Here, inclusion pathology by co-expression of L126Z-SOD1 we used our paradigm of newborn intraspinal injec- [21]. Bigenic mice produced from crosses of GurG93A tion to seed with spinal homogenates from paralyzed mice and WT-SOD1:YFP mice developed paralysis at homozygous PrP.G37R-SOD1, paralyzed G37R-SOD1 about the same time as mice that express only G93A- G85R−YFP Line 29, paralyzed P1-G37R , or paralyzed SOD1 (Additional file  2: Fig. S11a). As compared to GurG93A mice. None of these inocula induced para- mice that express only WT-SOD1:YFP, in paralyzed lytic disease (Table  3). In these older heterozygous PrP- GurG93A/WT-SOD1:YFP mice we observed redis- G37R mice, we observed some instances of silver-positive tribution of the YFP fluorescence into fibrillary or inclusion pathology, but the levels of pathology were punctate inclusion-like pathology (Additional file  2: similar between mice injected with PBS or transgenic spi- Fig. S11b & c). Remarkably, a similar type of punc- nal homogenates (Additional file  2: Fig. S10). To date, we tate inclusion-like pathology was observed in very old have not observed accelerated disease in any attempt to bigenic mice generated by crosses of GurWT mice and seed mice expressing the G37R variant of human SOD1. WT-SOD1:YFP mice (Additional file  2: Fig. S11d). We also noted a remarkable level of fluorescent inclusion Mice expressing WT‑SOD1 are largely resistant to prion‑like pathology in the cerebellar dentate (Additional file  2: seeding Fig. S12). Although these mice showed some obvious In addition to our attempts to seed mutant SOD1 mice, gait abnormalities, which occur in older GurWT mice we have also examined whether it is possible to induce [42], these bigenic GurWT/WT-SOD1:YFP mice did paralysis in mice that express WT-SOD1 [15]. In a not develop the paralytic phenotypes of mice express- pilot study with a relatively small cohort of mice that ing mutant SOD1. express WT-SOD1:YFP, we did not observe induction The foregoing data suggested that spinal homogen - of paralysis by 9–13 months of age by seeding with spi- ates from aged GurWT mice or paralyzed G93A mice nal homogenates from paralyzed G93A or G37R mice could potentially seed aggregation and disease in (Additional file  2: Table S3; Data from [15]). This study A yers et al. acta neuropathol commun (2021) 9:92 Page 17 of 24 Table 4 Summary of data from seeding WT‑SOD1:YFP and GurWT ‑SOD1 mice Host Inoculum (ISP) # inoculated # paralysis Age of euthanasia due to age WT‑SOD1:YFP PBS 3 0 17—20 mo (a) WT‑SOD1:YFP NTg Sp Cord 4 0 20 mo (b) WT‑SOD1:YFP GurWT Sp Cord 15 0 19–20 mo (b) WT‑SOD1:YFP recWT fibrils 10 0 16–20 mo (d) GurWT PBS 3 0 20 mo (e) GurWT NTg Sp Cord 3 0 20 mo (f ) GurWT GurWT Sp Cord 6 0 19 mo GurWT recWT SOD1 fibrils 7 0 16–20 mo GurWT GurG93A Sp Cord 4 0 17–20 mo GurWT L126Z Sp Cord 3 0 18–19 mo (g) GurWT QV103Z Sp Cord 3 1 19–20 mo (h) (a) Data originally reported in [15] (b) 1 found dead @ 7 mo. Data originally reported in [15] (c) 1 found dead @ 4.3 mo, 2 euthanized for non ALS-related issues at 8.7 and 15 mo. Includes data for 5 animals originally reported in [15] (d) 1 found dead @ 5.4 mo, 2 euthanized for non-ALS-related issues at 7 and 18 mo (e) 3 found dead @ 4.5 -5.4 mo, 1 euthanized at 3 mo (f ) Data originally reported in [15] (g) 1 found dead at 18.8 mo – no prior clinical signs (h) 1 animal showed bilateral weakness at 20 months when euthanized but lacked inclusion pathology WT-SOD1:YFP mice. Aged GurWT mice accumulate testing whether we could induce paralysis by the injec- detergent insoluble WT-SOD1 (albeit at lower levels tion of fibrilized WT-SOD1, or spinal homogenates of than mutant mice) [15]. We have now examined 15 aged GurWT mice, or spinal homogenates of paralyzed WT-SOD1:YFP mice injected with spinal homogen- mice expressing the QV103Z or L126Z mutants. Cohorts ates from aged GurWT mice, finding no animals with of mice were aged to 17–20  months of age with none of ALS-like symptoms by 16–20  months post-injection the animals developing any type of paralytic phenotype (Table 3). Additionally, there was not obvious patholog- (Table  4). Additionally, we could not pathologically dis- ical distinction between WT-SOD1:YFP mice injected tinguish mice injected with GurWT spinal homogenates with GurWT spinal homogenates and mice with PBS or fibrilized WT SOD1 from mice injected with PBS or or spinal homogenates of NTg mice (Additional file  2: NTg spinal homogenates (Table  4; Additional file  2: Fig. Fig. S13). To further attempt to induce paralytic dis- S14). Overall, the data show that WT-SOD1 is relatively ease in the WT-SOD1:YFP mice, we injected animals resistant to prion-like seeding towards induction of with recombinant WT human SOD1 that had been motor neuron disease or the type of inclusion pathology fibrilized in  vitro. No animals developed paralysis by present in mice expressing fALS mutant SOD1. 16–20  months post-injection, and none showed an obvious induction of WT-SOD1:YFP inclusion pathol- Discussion ogy at the time of euthanasia (Additional file  2: Fig. The present study builds on prior work to examine the S13). Altogether, these findings indicate that WT- sources of variability in the efficiency of prion-like seed - SOD1:YFP mice are relatively resistant to developing ing in mutant SOD1 mice. From the effort we describe paralytic disease, or inclusion pathology, by injection here, prior work in our laboratories [15, 17], and data with preparations of misfolded WT or mutant SOD1. described by Bedhendi [18], it is clear that first pas - In parallel to our effort in the WT-SOD1:YFP mice, we sage G85R seeds efficiently seed G85R mice, and that have attempted to induce paralytic disease in the GurWT- first passage L126Z seeds efficiently seed L126Z mice SOD1 mice by newborn ISP injection. In a prior study, (Fig.  9a). By contrast, mice expressing WT-, G37R-, we reported that by 15  months of age, small cohorts or G93A-SOD1 variants were partially or fully resist- (n = 4) of GurWT mice injected with spinal homogen- ant to isologous seeding (Fig.  9a). Notably, mice that ates from GurG93A mice had not developed paralysis express human G85R-SOD1 fused to YFP were found to [15]. We have now extended this effort considerably, be susceptible to seeding with a wide range of seeding Ayers et al. acta neuropathol commun (2021) 9:92 Page 18 of 24 Fig. 9 Summary diagrams illustrating the relative efficiency of seeding in mouse models. a Summary of findings in seeding experiments involving various lines of SOD1 mice express untagged WT or mutant SOD1. b Summary of findings in seeding experiments with G85R‑SOD1:YFP mice preparations, including tissue homogenates from para- considerable variation in the susceptibility of differ - lyzed G37R and G93A mice and preparations of recom- ent lines of mutant SOD1 mice to seeding, and that the binant SOD1 fibrilized in  vitro (Fig.  9b). In general, sequence of the misfolded SOD1 in the seeds can influ - heterologous seeding of G85R-SOD1:YFP mice was less ence prion-like transmissibility to SOD1 transgenic efficient than isologous seeding, with seed preparations hosts. that contained the G37R or H46R variant being among the least efficient. We further observed that the seeds Role of structural stability in the susceptibility of SOD1 that arose in G85R-SOD1:YFP mice after seeding by variants to seeding preparations from paralyzed G37R mice appeared to A key distinguishing feature of transgenic SOD1 mice acquire attributes that led to longer incubation times that were vulnerable to seeding versus those that were upon second passage. Collectively, these findings show more resistant could be related to the structural sta- that sequence variation and seed source have profound bility of the SOD1 variant expressed by the host. Mice effects on the prion-like activity of misfolded mutant that express WT, G37R, or G93A human SOD1 show SOD1. high steady-state levels of soluble protein prior to the In assessing the significance of animals that failed to onset of paralysis, with the accumulation of misfolded develop paralysis or pathology as a result of intraspinal SOD1 aggregates occurring as paralysis develops [41, or sciatic nerve seeding, we must consider the possi- 43, 44]. By contrast, the steady-state levels of soluble bility of experimentation error. It is important to note mutant SOD1 in G85R mice are relatively low in pro- that all of the injections performed in this study were portion to transgene mRNA levels [43]. Both the L126Z conducted by only two operators that had extensively and QV103Z truncation variants are unstable and show trained to perform the injections. Although we cannot very low steady state levels until they begin to aggregate rule out the possibility of occasional injection error, we as paralysis develops [20, 21]. Thus, for the G85R, L126Z, have no indication that operator error would explain and QV103Z mutants, their susceptibility to isologous, or the data in cases where few or no animals in a cohort heterologous, seeding may be linked to poor structural developed paralysis. Thus, we conclude that there is stability. A yers et al. acta neuropathol commun (2021) 9:92 Page 19 of 24 In WT SOD1 a structurally important intramolecular these variants resist seeding because bound Zn limits disulfide bond links Cys 57 to Cys 146 [45]. In studies of prion-like propagation. SOD1 aggregation and fibrillization in vitro, the presence Overall, our data indicate that WT SOD1 and WT- of the normal intramolecular disulfide bond appears to like variants are less likely to exist in conformations that be a key factor in modulating aggregation [46]. In mutant are susceptible to prion-like conformational templat- SOD1 mice that develop paralysis due to over-expression, ing as compared to mutants that are more destabilized. the misfolded mutant protein that accumulates as patho- Whether any specific structure-stabilizing feature is criti - logical inclusions develop lacks the normal intramolecu- cal in mediating susceptibility to seeding is unclear. What lar disulfide bond [43, 44]. Previous analysis of the same is clear is that mice expressing the more destabilized vari- G85R mice used here for seeding studies found that all of ants, G85R, L126Z, and QV103Z were far more suscepti- the mutant protein present in spinal cord of pre-symp- ble to prion-like seeding. tomatic mice lacks this critical structural stabilizer [43]. Similarly, the L126Z and QV103Z variants would not be Do ALS‑associated mutations in SOD1 imprint strain‑like able to produce a normal intramolecular disulfide bond. attributes that influence prion‑like activity? It is important to note, however, that in GurWT and Based on our observation that we can induce and moni- Gur1G93A SOD1 mice, 8–14% of the over-expressed tor the spread of G85R-SOD1:YFP aggregation by seed- protein present in spinal cords of pre-symptomatic ing into the sciatic nerve of this model, we had begun mice also lacks the normal disulfide bond [43, 44]. Thus, to link the progressive spread of weakness in SOD1- whether the G85R, L126Z, and QV103Z mice are suscep- ALS to the prion-like spread of misfolded conforma- tible to seeding solely because the expressed mutant lacks tions [16]. The question that arises is whether there may an intramolecular disulfide bond is unclear. be conformational attributes that are enciphered by the Another important post-translation modification in the primary sequence of mutant SOD1 that, by some man- structural stability of SOD1 is the binding of Cu and Zn ner, modulate the ability of misfolded SOD1 to spread [47]. Similar to WT SOD1, the G93A or G37R variants, within the CNS. It is important to note that in the trans- are capable of binding Cu and Zn with high affinity [3, genic mouse models that over-express mutant SOD1, 23, 26]. The G85R variant has been shown to have a weak the mutation-specific differences in disease duration affinity for Cu and Zn [48, 49] and the enzyme appears that is seen in humans are not recapitulated. For exam- to be inactive in vivo [50]. Crystal structures of the G85R ple, mice that express the G93A and G37R variants at SOD1 variant demonstrate disorder in the Zn binding similar levels exhibit disease onset and duration that is and electrostatic loop elements of SOD1 (amino acids comparable; 4–6  months to onset and 3–4  weeks dura- 50–83 and 121–142, respectively) [48]. The L126Z and tion [23, 26]. Presumably the high-level expression of experimental QV103Z truncation mutants have not been mutant SOD1 throughout the CNS that is required to crystallized, but neither of these variants possess intact produce disease within the lifespan of the animal masks electrostatic loop elements. Whether the L126Z trun- the mutation-specific attribute that causes very slow pro - cation mutant can bind Cu or Zn is unknown, but the gression in persons with the G37R mutation. Here, we experimental QV103Z variant that we studied was engi- have asked whether the G37R, or H46R, mutations could neered to remove the critical Cu-binding sites in SOD1 impart some specific conformational information to the [21]. It is notable, however, that the majority of over- misfolded SOD1 that accumulates in these animals that expressed WT or G93A SOD1 present in the spinal cords would lead to distinct behavior in prion-like transmission of Gur1 G93A or Gur WT mice is inactive due to insuf- studies. ficient acquisition of Cu [43]. Thus, the ability of WT and Our newborn injection paradigm is one way to assess G93A SOD1 to bind Cu does not seem to fully explain the seeding competency of different SOD1 variants. In the low susceptibility of mice over-expressing these pro- heterologous seeding in G85R-SOD1:YFP mice by spi- teins to prion-like seeding if poor Cu binding were the nal homogenates from paralyzed G93A or L126Z mice only structural feature involved. (avg. durations of 2.4 and 3.8  years in humans, respec- The binding of Zn by nascent SOD1 appears to be criti - tively [35]), the efficiency of seeding at first passage was cal to achieving a native conformation [51, 52]. Analy- variable. Second passage of G93A seeds through G85R- sis of partially purified homodimeric SOD1 from spinal SOD1:YFP mice, however, produced a consistently earlier cords of the GurWT, G37R-Line 29, and Gur1G93A mice onset of paralysis (avg. 2.8  months post-injection). Our found the over-expressed protein appeared have no defi - study of L126Z seeds in G85R-SOD1:YFP mice is more ciency in the level of bound Zn; measured to be between limited, but we did similarly observe consistent accel- 2.5 and 3 atoms per dimer [53]. Thus, it is possible that eration of paralysis by second-passage L126Z seeds in a small cohort of animals (Additional file  2: Fig. S15; data Ayers et al. acta neuropathol commun (2021) 9:92 Page 20 of 24 from [17]). Thus, for two mutants associated with rapidly that seeding could accelerate the appearance of paralysis progressing disease we observe relative ease of first pas - albeit at a lower frequency than we would have expected sage with accelerated and highly efficient second passage for isologous seeding. As noted above, in the Gur1-G93A seeding. mice, a substantial portion of the over-expressed protein For the H46R and G37R variants, associated with lacks Cu and a normal disulfide bond [43]. The absence of slowly progressing ALS in humans, we observed poor these modifications appears to be related to over-expres - seeding activity in first passage, and with G37R-derived sion; increasing Cu loading in Gur1G93A mice dramati- seeds we osbserved variable and protracted incubation cally mitigates its misfolding and toxicity [57]. The Cu periods. These data are consistent with the idea that loading and disulfide status of G93A SOD1 in spinal mutations associated with slowly progressing disease cords of VLE-G93A mice has not been examined, but the may be less efficient in prion-like seeding. lower expression of mutant protein would be expected An important consideration in interpreting these to result in lower levels of incompletely modified mutant results is the potential impact of the host G85R SOD1 SOD1. variant on any conformational change propagated by The absence of seeding efficacy in the y Th -1 G93A mice exposure to seeds from mice expressing any other vari- could potentially be related to the pattern of transgene ant. In PrP prions, sequence differences between the pri - expression for this model. It is also possible that the level mary sequence of PrP in seeds and the sequence of PrP of G93A SOD1 expression in Thy-1 G93A mice is just in the host can modify strain attributes [54]. Moreover, below threshold to sustain the propagation of misfolded single amino acid differences of the PrP sequence in the conformations. Additional studies of seeding efficacy in seed and host can create a species barrier that lowers these two lines of G93A mice could provide insight into transmission efficiency [55]. To date, G85R-SOD1 mice the basis of vulnerability. At the time of writing, we are have shown efficient seeding with spinal cord prepara - not able to explain why the Gur1-G93A or Thy-1 G93A tions containing misfolded A4V, G85R, D90A, G93A, mice did not respond to seeding or why the VLE-G93A and L126 or G127 truncation mutants [15, 17–19]. We showed a modest response. If vulnerability to seeding is observed that fibrilization treatments of 7 different by some means related to the folding state of the SOD1 recombinant SOD1 proteins could produce seeds that protein in the host, we would have expected the Gur1- induce early paralysis in G85R-SOD1:YFP mice. These G93A mice to be more susceptible to seeding due to a data indicate that G85R-SOD1 is broadly susceptible to higher level of incompletely modified mutant protein. At heterologous conformational templating. Indeed, seeds face value, the poor seeding of G93A mice could foster from G37R mice were more effective in G85R SOD1 skepticism over the role of prion-like spread in the patho- mice than G37R SOD1 mice. u Th s, the poor seeding of genesis of SOD1-linked fALS; however, there are mitigat- G85R-SOD1:YFP mice by spinal tissues from H46R mice ing factors that merit discussion. and the long incubation periods produced by G37R seeds A key variable in our experiments is the level of bioac- may indicate that these variants produce strains of mis- tive misfolded SOD1 seeds in the spinal homogenates we folded SOD1 that are less effective as seeds to propagate have used for seeding. Previous studies of mutant SOD1 misfolded conformations. mice that develop paralysis have demonstrated that the levels of misfolded mutant SOD1 in the spinal cords rise Is prion‑like spreading a common feature in SOD1 ALS? steadily and reach maximum when mice are paralyzed Our observation that mice expressing the G93A variant [20, 41, 44]. We originally had assumed that at endstage, are not particularly receptive to seeding would appear to regardless of age, the levels of misfolded SOD1 in any run counter to the notion that prion-like mechanisms of given animal for a given mutant would be at maximal lev- spreading apply to all fALS variants. As indicated above, els. For all of our mouse to mouse propagation studies, we the G93A mutation is associated with relatively rap- used 10% homogenates of spinal cord that had been clari- idly progressing disease [35, 56], and thus the expecta- fied by low speed centrifugation. In our isologous seeding tion would be that mice expressing G93A-SOD1 should studies of G85R-SOD1:YFP, G85R, and L126Z mice, the be highly permissive to seeding, particularly isologous levels of the seeds in homogenates from paralyzed trans- seeding. We have previously reported that the com- genic mice were clearly sufficient to induce accelerated monly used Gur1-G93A mice did not exhibit acceler- paralysis with high efficiency. For the WT-SOD1:YFP, ated paralysis after intraspinal seeding [15]. To assess WT-, G93A-, and G37R-SOD1 expressing mice, which whether the level of G93A expression may have been a did not respond to isologous seeding, it is possible that factor, we tested two additional lines of G93A mice. The injection of a higher amount of seeds would have induced y Th -1 G93A mice were unresponsive to seeding; how - disease. Notably, our G93A and G37R seed prepara- ever, with the VLE-G93A mice we began to see evidence tions were effective, to varying degrees, in heterologous A yers et al. acta neuropathol commun (2021) 9:92 Page 21 of 24 seeding of G85R-SOD1:YFP mice. Moreover, recombi- Intriguingly, we observed multiple examples in which nant fibrils of WT human SOD1 were highly effective in seeded mice reached relatively old ages without devel- seeding G85R-SOD1:YFP mice, with no activity in WT oping symptoms despite significant inclusion pathology or WT-SOD1:YFP mice. Thus, it is clear that the lines of burden. The most striking example was in the cohort of mice expressing WT and WT-like SOD1 mutants (G37R G85R-SOD1:YFP mice seeded with spinal homogenates and G93A) are less susceptible to prion-like seeding with from paralyzed H46R rats. We also observed high levels preparations that effectively induce early disease in mice of inclusion pathology in aged bigenic mice created by expressing the unstable G85R mutant. Whether the lower crossing GurWT mice with WT-SOD1:YFP mice. These susceptibility of WT and WT-like variants to seeding intriguing findings raise the possibility that some con - could be overcome by injection of higher amounts of mis- formations of misfolded SOD1 may be non-toxic. Addi- folded SOD1 seeds will require additional investigation. tional studies are required to understand whether strain An additional consideration in assessing the suscepti- variants of misfolded mutant SOD1 may exist that prop- bility of WT and WT-like SOD1 mutants to prion-like agate between cells more slowly, or produce less toxic propagation is the timing of exposure to the seed. We aggregates, and whether such strain variations explain have relied heavily on a paradigm in which the seeds aspects of human SOD1 linked ALS. were injected into newborn mice, primarily as a means to facilitate wide-spread dissemination of the seeds. The Role of WT‑SOD1 in sporadic ALS earliest age to paralysis in mice expressing high levels of The role of prion-like propagation in sporadic ALS, and mutant SOD1 that has been reported is 3–4  months of the identity of the misfolded propagon, remains to be age [23, 26, 29], suggesting that early in life the toxicity of determined. Multiple studies have examined whether the mutant SOD1 is suppressed by yet to be defined pro - a misfolded form of WT-SOD1 could be propagat- tective factors. It is possible that mice expressing WT-, ing throughout the neuraxis in sporadic ALS patients G93A-, or G37R-SOD1 do not respond to seeding in (reviewed in [58]). Recombinant WT-SOD1 that has the newborn injection paradigm because a combination been fibrilized in  vitro clearly possesses high seeding of protective factors and the natural propensity of these activity when injected into G85R-SOD1:YFP mice [17], mutants to fold into a more native-like enzyme, sup- but these same preparations showed no activity in the presses the establishment of a self-sustaining propaga- WT-SOD1:YFP or GurWT mice. From our experiments tive process. It is also possible that the capacity of WT or in which mutant SOD1 was co-expressed with WT- WT-like SOD1 variants to fold and mature could change SOD1:YFP, it was clear that with sustained exposure it is with aging as proteostatic protective factors decline. possible to induce WT-SOD1:YFP to produce inclusions. In humans, where disease usually occurs late in life, the Collectively, our data indicate that although it is pos- WT-like variants of SOD1 could become more prone to sible to create a seeding-competent conformer of WT- adopt conformations that are susceptible to prion-like SOD1, it does not appear to transmit to WT substrates seeding. efficiently in short-term seeding paradigms. It is possible Although our studies, and others [18, 19], demonstrate that propagation of misfolded conformations by WT- that mutant SOD1 has the potential to acquire prion-like SOD1 is inhibited by the high propensity of WT-SOD1 capabilities, there are aspects of the mouse models that to acquire a stable native conformation. Whether other provided sources of misfolded SOD1 seeds that should types of conformational changes in WT-SOD1 that could be considered. In all the mouse or rat models used to contribute to ALS, and possibly propagate in a prion-like produce inoculum, mutant SOD1 is highly expressed fashion as suggested by other studies [10, 58, 59], has throughout the CNS. In this scenario, the development not been ruled out by our study. However, we have not of disease would not necessarily depend on the prion- observed WT SOD1 to be able to support propagation of like spread and therefore there would be little selective a disease-causing misfolded SOD1 conformation thus far. pressure towards the accumulation of misfolded SOD1 proteins that can efficiently propagate between cells. Potential role of amyloidogenic segments in prion‑like It is possible that the absence of such selective pressure propagation of misfolded SOD1 leads to the accumulation of mixtures of misfolded SOD1 Although the G85R-SOD1:YFP, or untagged G85R, mice strains in the transgenic mice that were used to produce appear to be highly susceptible to seeding, we observed the seeds. Any given individual animal could potentially that seeds prepared from mice that over-express murine harbor a unique mixture of strains and as such could SOD1 with the G85R mutation were not effective in account for some of the variability we observe in first- either host. Incompatibility between the misfolded seed passage seeding experiments. and the host was also noted when QV103Z mice were seeded with spinal homogenates from L126Z mice, but Ayers et al. acta neuropathol commun (2021) 9:92 Page 22 of 24 not vice versa. These seemingly disparate outcomes may efficiently seed G93A mice weakens the idea that prion-like provide clues to initial sites of interaction between the propagation mediates the perceived spread of weakness propagating seed and the naïve host protein. Ivanova and in all cases of SOD1-linked ALS. Additional studies are colleagues identified four amino acid segments in SOD1 required to understand the poor response of mice express- that were highly prone to aggregation [60]. The four seg - ing the G93A variant to seeding; perhaps it is a question of ments were distributed across the protein at residues timing or dose of the seed. Our study suggests that further 14–21, 30–38, 101–107, and 147–153. The ability of mis - investigations of the prion-like propagation of misfolded folded QV103Z to propagate to itself or L126Z-SOD1 conformations in mutant SOD1 could reveal mechanisms indicates that if these segments are involved in prion-like that underlie disparate durations of illness in individuals propagation, then the segments 14–21 and 30–38 must with different mutations in the same gene. be involved. For the L126Z mutant, segment 101–107 could also be important because although L126Z could Supplementary Information self-seed, it could not cross seed to QV103Z. It is also The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s40478‑ 021‑ 01191‑w. noteworthy that mice expressing the G37R variant are largely resistant to seeding, and similarly, a synthetic pep- Additional file 1. Excel spreadsheet that lists all of the transgenic animals, tide of the amyloidogenic segment 30–38 with the G37R and a subset of the nontransgenic animals, used in this study, with details mutation was much slower to aggregate in  vitro [60]. on the inoculum injected and incubation period postinjection. See Sup‑ Segment 30–38 is also of interest because of significant plementary Materials for additional information. sequence divergence between mouse and human SOD1 Additional file 2. This file contains all Supplementary Material, including Supplementary Tables S1‑S3 and Supplementary Figures S1‑S15. within this segment; the human sequence is KVWG- SIKGL and the mouse sequence is VLSGQITGL. Our studies clearly show that misfolded mouse G85R-SOD1 Acknowledgements We would like to thank Dr. Paramita Chakrabarty for helpful suggestions in the seeds are not transmissible to mice expressing human preparation of this report, and Susan Fromholt for assistance in the manage‑ G85R-SOD1. In segments 14–21 and 101–107 there ment of transgenic animal colonies. is only one amino acid difference between mouse and Authors’ contributions human SOD1, and in segment 147–153 there are no JIA and DRB conceived the described experiments, directed the efforts differences. One of the differences between human and of support staff, interpreted the data, and composed the manuscript. GX mouse sequence in segment 30–38 is a W to S change performed histological examinations and contributed to the production of figures. KD administered inocula, monitored animals, and harvested tissues. at position 32, which has been implicated in prion-like QL and ZC examined histological data. JB, AKMR, DLZ, & AG produced and propagation of misfolded SOD1 [14, 36, 61, 62]. Col- purified recombinant SOD1 proteins. All authors read and approved the final lectively, these data suggest the amyloidogenic segment manuscript. between residues 30–38 could be an important modula- Funding tor of prion-like propagation of misfolded SOD1. A.G. was supported by St. Mary’s University Research Grant and the Biaggini Research Program. This work was supported by an NIH Shared Instrumenta‑ tion Grant (S10OD020026), grants from the National Institutes of Neurological Conclusions Disorders and Stroke (1R21NS088839; 1R01NS092788), and the Packard Center The collective data of the present study and prior work for ALS Research at Johns Hopkins University. [15–17, 19] clearly show that spinal cords of paralyzed Availability of data and materials mutant SOD1 mice contain seed-competent forms of mis- Any data generated or analyzed during this study that are not included in folded SOD1 that can induce a self-propagating process, this published article and its supplementary files, are available from the cor ‑ responding author on reasonable request. leading to accelerated motor neuron disease and inclu- sion pathology. Similar seeds are found in human spinal Declarations tissues from SOD1-linked ALS patients [19]. Overall, our findings generally align with the hypothesis that prion-like Ethics approval and consent to participate propagation of misfolded conformations could mediate the All studies involving mice were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Florida in accordance with all progressive spread of weakness from limb to limb that is a state and federal guidelines. defining feature of the disease. Tissue homogenates from mice that express mutations associated with more rapidly Consent for publication Not applicable. progressing ALS (G93A, G85R, L126Z) were generally more effective in seeding than homogenates prepared from Competing Interests mice expressing mutants associated with slowly progress- The authors declare that they have no competing interests. ing disease (G37R, H46R). Mice expressing the G85R and Author details L126Z mutants were more permissive to seeding than mice 1 Department of Neuroscience, Center for Translational Research in Neurode‑ expressing the G37R variant. At face value, the inability to generative Disease (CTRND), University of Florida, Box 100159, Gainesville, FL A yers et al. acta neuropathol commun (2021) 9:92 Page 23 of 24 32610, USA. Institute for Neurodegenerative Disease, Weill Institute for Neuro‑ 18. Bidhendi EE, Bergh J, Zetterstrom P, Andersen PM, Marklund SL, sciences, University of California, San Francisco, CA 94143, USA. Depar tment Brannstrom T (2016) Two superoxide dismutase prion strains transmit of Neurology, Weill Institute for Neurosciences, University of California, San amyotrophic lateral sclerosis‑like disease. J Clin Invest 126:2249–2253 Francisco, CA 94143, USA. Department of Biological Sciences, St. Mary’s Uni‑ 19. Ekhtiari Bidhendi E, Bergh J, Zetterström P, Forsberg K, Pakkenberg B, versity, San Antonio, TX 78228, USA. SantaFe HealthCare Alzheimer’s Disease Andersen PM et al (2018) Mutant superoxide dismutase aggregates from Research Center, McKnight Brain Institute, University of Florida, Gainesville, FL human spinal cord transmit amyotrophic lateral sclerosis. Acta Neuro‑ 32610, USA. pathol 136:939–953 20. Wang J, Xu G, Li H, Gonzales V, Fromholt D, Karch C et al (2005) Soma‑ Received: 24 March 2021 Accepted: 3 May 2021 todendritic accumulation of misfolded SOD1‑L126Z in motor neurons mediates degeneration: {alpha}B‑ crystallin modulates aggregation. Hum Mol Genet 14:2335–2347 21. Xu G, Ayers JII, Roberts BLL, Brown H, Fromholt S, Green C et al (2015) Direct and indirect mechanisms for wild‑type SOD1 to enhance the toxicity of mutant SOD1 in bigenic transgenic mice. Hum Mol Genet References 24:1019–1035 1. Ravits JM, La SAR (2009) ALS motor phenotype heterogeneity, focality, 22. Wang J, Farr GW, Zeiss CJ, Rodriguez‑ Gil DJ, Wilson JH, Furtak K et al and spread: deconstructing motor neuron degeneration. Neurology (2009) Progressive aggregation despite chaperone associations of a 73:805–811 mutant SOD1‑ YFP in transgenic mice that develop ALS. Proc Natl Acad 2. Volk AE, Weishaupt JH, Andersen PM, Ludolph AC, Kubisch C (2018) Sci USA 106:1392–1397 Current knowledge and recent insights into the genetic basis of amyo‑ 23. Wong PCC, Pardo CAA, Borchelt DRR, Lee MKK, Copeland NGG, Jenkins trophic lateral sclerosis. Med Genet Mitteilungsblatt des Berufsverbandes NAA et al (1995) An adverse property of a familial ALS‑linked SOD1 muta‑ Medizinische Genet eV 30:252–258 tion causes motor neuron disease characterized by vacuolar degenera‑ 3. Potter SZ, Valentine JS (2003) The perplexing role of copper‑zinc superox ‑ tion of mitochondria. Neuron 14:1105–1116 ide dismutase in amyotrophic lateral sclerosis (Lou Gehrig’s disease). J Biol 24. Wang J, Xu G, Slunt HHH, Gonzales V, Coonfield M, Fromholt D et al (2005) Inorg Chem 8:373–380 Coincident thresholds of mutant protein for paralytic disease and protein 4. Sheng Y, Chattopadhyay M, Whitelegge J, Valentine JS. SOD1 aggrega‑ aggregation caused by restrictively expressed superoxide dismutase tion and ALS: role of metallation states and disulfide status. Curr Top Med cDNA. Neurobiol Dis 20:943–952 Chem. United Arab Emirates; 2012;12:2560–72. 25. Bruijn LI, Becher MW, Lee MK, Anderson KL, Jenkins NA, Copeland NG 5. Sangwan S, Eisenberg DS (2016) Perspective on SOD1 mediated toxicity et al (1997) ALS‑linked SOD1 mutant G85R mediates damage to astro ‑ in amyotrophic lateral sclerosis. Postepy Biochem Poland 62:362–369 cytes and promotes rapidly progressive disease with SOD1‑ containing 6. Grad LI, Yerbury JJ, Turner BJ, Guest WC, Pokrishevsky E, O’Neill MA et al inclusions. Neuron 18:327–338 (2014) Intercellular propagated misfolding of wild‑type Cu/Zn superoxide 26. Gurney ME, Pu H, Chiu AY, Canto MCD, Polchow CY, Alexander DD et al dismutase occurs via exosome‑ dependent and ‑independent mecha‑ (1994) Motor neuron degeneration in mice that express a human Cu, Zn nisms. Proc Natl Acad Sci USA 111:3620–3625 superoxide dismutase mutation. Science 264:1772–1775 7. Bosco DA, Morfini G, Karabacak NM, Song Y, Gros‑Louis F, Pasinelli P et al 27. Jaarsma D, Teuling E, Haasdijk ED, De ZCI, Hoogenraad CC (2008) (2010) Wild‑type and mutant SOD1 share an aberrant conformation and Neuron‑specific expression of mutant superoxide dismutase is sufficient a common pathogenic pathway in ALS. Nat Neurosci 13:1396–1403 to induce amyotrophic lateral sclerosis in transgenic mice. J Neurosci 8. Forsberg K, Jonsson PA, Andersen PM, Bergemalm D, Graffmo KS, Hultdin 28:2075–2088 M, et al. Novel antibodies reveal inclusions containing non‑native SOD1 28. Deitch JS, Alexander GM, Bensinger A, Yang S, Jiang JT, Heiman‑Patterson in sporadic ALS patients. PLoSONE. 2010;5:e11552. TD. Phenotype of transgenic mice carrying a very low copy number of 9. Tokuda E, Takei Y‑I, Ohara S, Fujiwara N, Hozumi I, Furukawa Y (2019) Wild‑ the mutant human G93A superoxide dismutase‑1 gene associated with type Cu/Zn‑superoxide dismutase is misfolded in cerebrospinal fluid of amyotrophic lateral sclerosis. PLoS One. 2014;9:e99879. sporadic amyotrophic lateral sclerosis. Mol Neurodegener 14:42 29. Ripps ME, Huntley GW, Hof PR, Morrison JH, Gordon JW. Transgenic mice 10. Paré B, Lehmann M, Beaudin M, Nordström U, Saikali S, Julien J‑P et al expressing an altered murine superoxide dismutase gene provide an (2018) Misfolded SOD1 pathology in sporadic amyotrophic lateral sclero‑ animal model of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A. sis. Sci Rep 8:14223 United States; 1995;92:689–93. 11. Brotherton TE, Li Y, Cooper D, Gearing M, Julien JP, Rothstein JD et al 30. Aoki M, Kato S, Nagai M, Itoyama Y (2005) Development of a rat model (2012) Localization of a toxic form of superoxide dismutase 1 protein to of amyotrophic lateral sclerosis expressing a human SOD1 transgene. pathologically affected tissues in familial ALS. Proc Natl Acad Sci USA Neuropathology Australia 25:365–370 109:5505–5510 31. Seetharaman SV, Taylor AB, Holloway S, Hart PJ (2010) Structures of 12. Ayers JI, Xu G, Pletnikova O, Troncoso JCC, Hart PJJ, Borchelt DRR (2014) mouse SOD1 and human/mouse SOD1 chimeras. Arch Biochem Biophys Conformatonal specificity of the C4F6 SOD1 antibody; low frequency of United States 503:183–190 reactivity in sporadic ALS cases. Acta Neuropathol Commun 2:55 32. Ayers JII, Fromholt S, Sinyavskaya O, Siemienski Z, Rosario AMM, Li A et al 13. Da Cruz S, Bui A, Saberi S, Lee SK, Stauffer J, McAlonis‑Downes M et al (2015) Widespread and efficient transduction of spinal cord and brain (2017) Misfolded SOD1 is not a primary component of sporadic ALS. Acta following neonatal AAV injection and potential disease modifying effect Neuropathol Germany 134:97–111 in ALS mice. Mol Ther 23:53–62 14. Grad LI, Guest WC, Yanai A, Pokrishevsky E, O’Neill MA, Gibbs E et al (2011) 33. Bartz JC, Kincaid AE, Bessen RA (2002) Retrograde transport of transmis‑ Intermolecular transmission of superoxide dismutase 1 misfolding in sible mink encephalopathy within descending motor tracts. J Virol living cells. Proc Natl Acad Sci USA 108:16398–16403 76:5759–5768 15. Ayers JI, Fromholt S, Koch M, DeBosier A, McMahon B, Xu G et al (2014) 34. Vallet PG, Guntern R, Hof PR, Golaz J, Delacourte A, Robakis NK et al (1992) Experimental transmissibility of mutant SOD1 motor neuron disease. Acta A comparative study of histological and immunohistochemical methods Neuropathol 128:791–803 for neurofibrillary tangles and senile plaques in Alzheimer’s disease. Acta 16. Ayers JI, Fromholt SE, O’Neal VM, Diamond JH, O’Neal VM, Diamond JH Neuropathol 83:170–178 et al (2016) Prion‑like propagation of mutant SOD1 misfolding and motor 35. Prudencio M, Hart PJJ, Borchelt DRR, Andersen PMM (2009) Variation in neuron disease spread along neuroanatomical pathways. Acta Neuro‑ aggregation propensities among ALS‑associated variants of SOD1: cor ‑ pathol 131:103–114 relation to human disease. Hum Mol Genet 18:3217–3226 17. Ayers JI, Diamond J, Sari A, Fromholt S, Galaleldeen A, Ostrow LW et al 36. Crown A, McAlary L, Fagerli E, Brown H, Yerbury JJ, Galaleldeen A, et al. (2016) Distinct conformers of transmissible misfolded SOD1 distinguish Tryptophan residue 32 in human Cu‑Zn superoxide dismutase modulates human SOD1‑FALS from other forms of familial and sporadic ALS. Acta prion‑like propagation and strain selection. PLoS One. 2020;15:e0227655. Neuropathol 132:827–840 37. Abdolvahabi A, Shi Y, Rasouli S, Croom CM, Aliyan A, Martí AA et al (2017) Kaplan‑Meier Meets Chemical Kinetics: Intrinsic Rate of SOD1 Ayers et al. acta neuropathol commun (2021) 9:92 Page 24 of 24 Amyloidogenesis Decreased by Subset of ALS Mutations and Cannot 50. Bruijn LI, Houseweart MK, Kato S, Anderson KL, Anderson SD, Ohama Fully Explain Age of Disease Onset. ACS Chem Neurosci United States E et al (1998) Aggregation and motor neuron toxicity of an ALS‑linked 8:1378–1389 SOD1 mutant independent from wild‑type SOD1. Science 281:1851–1854 38. McAlary L, Aquilina JA, Yerbury JJ. Susceptibility of Mutant SOD1 to 51. Kayatekin C, Zitzewitz JA, Matthews CR (2008) Zinc binding modulates Form a Destabilized Monomer Predicts Cellular Aggregation and Toxicity the entire folding free energy surface of human Cu. Zn superoxide but Not In vitro Aggregation Propensity. Front Neurosci. Switzerland; dismutase J Mol Biol 384:540–555 2016;10:499. 52. Potter SZ, Zhu H, Shaw BF, Rodriguez JA, Doucette PA, Sohn SH et al 39. Abdolvahabi A, Shi Y, Chuprin A, Rasouli S, Shaw BF (2016) Stochastic (2007) Binding of a single zinc ion to one subunit of copper‑zinc super ‑ formation of fibrillar and amorphous superoxide dismutase oligomers oxide dismutase apoprotein substantially influences the structure and linked to amyotrophic lateral sclerosis. ACS Chem Neurosci 7:799–810 stability of the entire homodimeric protein. J Am Chem Soc United States 40. Xia Y, Chen Z, Xu G, Borchelt DR, Ayers JI, Giasson BI. Novel SOD1 Mono‑ 129:4575–4583 clonal Antibodies against the Electrostatic Loop Preferentially Detect 53. Lelie HLL, Liba A, Bourassa MWW, Chattopadhyay M, Chan PKK, Gralla EBB Misfolded SOD1 Aggregates. Neurosci Lett. Ireland; 2020;135553. et al (2011) Copper and zinc metallation status of copper‑zinc superoxide 41. Lang L, Zetterstrom P, Brannstrom T, Marklund SL, Danielsson J, Oliveberg dismutase from amyotrophic lateral sclerosis transgenic mice. J Biol M (2015) SOD1 aggregation in ALS mice shows simplistic test tube Chem 286:2795–2806 behavior. Proc Natl Acad Sci U S A 112:9878–9883 54. Cortez LM, Sim VL (2013) Implications of prion polymorphisms. Prion 42. Graffmo KS, Forsberg K, Bergh J, Birve A, Zetterstrom P, Andersen PM et al 7:276–279 (2013) Expression of wild‑type human superoxide dismutase ‑1 in mice 55. Hsiao KK, Groth D, Scott M, Yang SL, Serban H, Rapp D et al (1994) Serial causes amyotrophic lateral sclerosis. Hum Mol Genet 22:51–60 transmission in rodents of neurodegeneration from transgenic mice 43. Jonsson PA, Graffmo KS, Andersen PM, Brannstrom T, Lindberg M, expressing mutant prion protein. Proc Natl Acad Sci U S A 91:9126–9130 Oliveberg M et al (2006) Disulphide‑reduced superoxide dismutase ‑1 in 56. Cleveland DW, Laing N, Hurse PV, Brown RHJ (1995) Toxic mutants in CNS of transgenic amyotrophic lateral sclerosis models. Brain; a J Neurol Charcot’s sclerosis. Nature England 378:342–343 129:451–464 57. Williams JR, Trias E, Beilby PR, Lopez NI, Labut EM, Bradford CS et al (2016) 44. Karch CMM, Prudencio M, Winkler DDD, Hart PJJ, Borchelt DRR (2009) Role Copper delivery to the CNS by CuATSM effectively treats motor neuron of mutant SOD1 disulfide oxidation and aggregation in the pathogenesis disease in SOD(G93A) mice co‑ expressing the Copper‑ Chaperone‑for ‑ of familial ALS. Proc Natl Acad Sci U S A 106:7774–7779 SOD. Neurobiol Dis 89:1–9 45. Hart PJ, Liu H, Pellegrini M, Nersissian AM, Gralla EB, Valentine JS et al 58. Grad LI, Cashman NR (2014) Prion‑like activity of Cu/Zn superoxide (1998) Subunit asymmetry in the three‑ dimensional structure of a human dismutase: implications for amyotrophic lateral sclerosis. Prion 8:33–41 CuZnSOD mutant found in familial amyotrophic lateral sclerosis. Protein 59. Furukawa Y, Tokuda E (2020) Does wild‑type Cu/Zn‑superoxide dismutase Sci 7:545–555 have pathogenic roles in amyotrophic lateral sclerosis? Transl Neurode‑ 46. Chattopadhyay M, Nwadibia E, Strong CD, Gralla EB, Valentine JS, gener 9:33 Whitelegge JP (2015) The Disulfide Bond, but Not Zinc or Dimerization, 60. Ivanova MI, Sievers SA, Guenther EL, Johnson LM, Winkler DD, Galalel‑ Controls Initiation and Seeded Growth in Amyotrophic Lateral Sclerosis‑ deen A et al (2014) Aggregation‑triggering segments of SOD1 fibril linked Cu, Zn Superoxide Dismutase (SOD1) Fibrillation. J Biol Chem formation support a common pathway for familial and sporadic ALS. Proc 290:30624–30636 Natl Acad Sci U S A 111:197–201 47. Oztug Durer ZA, Cohlberg JA, Dinh P, Padua S, Ehrenclou K, Downes S, 61. Pokrishevsky E, McAlary L, Farrawell NE, Zhao B, Sher M, Yerbury JJ, et al. et al. Loss of metal ions, disulfide reduction and mutations related to Tryptophan 32‑mediated SOD1 aggregation is attenuated by pyrimidine ‑ familial ALS promote formation of amyloid‑like aggregates from superox ‑ like compounds in living cells. Sci Rep. England; 2018;8:15590. ide dismutase. PLoS One. United States; 2009;4:e5004. 62. DuVal MG, Hinge VK, Snyder N, Kanyo R, Bratvold J, Pokrishevsky E et al 48. Cao X, Antonyuk SV, Seetharaman SV, Whitson LJ, Taylor AB, Holloway SP (2019) Tryptophan 32 mediates SOD1 toxicity in a in vivo motor neuron et al (2008) Structures of the G85R variant of SOD1 in familial amyo‑ model of ALS and is a promising target for small molecule therapeutics. trophic lateral sclerosis. J Biol Chem 283:16169–16177 Neurobiol Dis United States 124:297–310 49. Ratovitski T, Corson LB, Strain J, Wong P, Cleveland DW, Culotta VC et al (1999) Variation in the biochemical/biophysical properties of mutant Publisher’s Note superoxide dismutase 1 enzymes and the rate of disease progres‑ Springer Nature remains neutral with regard to jurisdictional claims in pub‑ sion in familial amyotrophic lateral sclerosis kindreds. Hum Mol Genet lished maps and institutional affiliations. 8:1451–1460 Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Acta Neuropathologica CommunicationsSpringer Journals

Published: May 20, 2021

There are no references for this article.