Complement components are upregulated and correlate with disease progression in the TDP-43Q331K mouse model of amyotrophic lateral sclerosis

Complement components are upregulated and correlate with disease progression in the TDP-43Q331K... Background: Components of the innate immune complement system have been implicated in the pathogenesis of amyotrophic lateral sclerosis (ALS) specifically using hSOD1 transgenic animals; however, a comprehensive examination of complement expression in other transgenic ALS models has not been performed. This study therefore aimed to determine the expression of several key complement components and regulators in the Q331K lumbar spinal cord and tibialis anterior muscle of TDP-43 mice during different disease ages. WT Q331K Methods: Non-transgenic, TDP-43 and TDP-43 mice were examined at three different ages of disease progression. Expression of complement components and their regulators were examined using real-time quantitative PCR and enzyme-linked immunosorbent assay. Localisation of terminal complement component receptor C5aR1 within the lumbar spinal cord was also investigated using immunohistochemistry. Results: Altered levels of several major complement factors, including C5a, in the spinal cord and tibialis anterior Q331K muscle of TDP-43 mice were observed as disease progressed, suggesting overall increased complement Q331K activation in TDP-43 mice. C5aR1 increased during disease progression, with immuno-localisation demonstrating expression on motor neurons and expression on microglia surrounding the regions of motor neuron death. There was a strong negative linear relationship between spinal cord C1qB, C3 and C5aR1 mRNA levels with hind-limb grip strength. Conclusions: These results indicate that similar to SOD1 transgenic animals, local complement activation and increased Q331K expression of C5aR1 may contribute to motor neuron death and neuromuscular junction denervation in the TDP-43 mouse ALS model. This further validates C5aR1 as a potential therapeutic target for ALS. Background (~ 10%), the disease has a familial component, and it is Amyotrophic lateral sclerosis (ALS), also known as due to specific genetic mutations. Some of the genes that motor neuron disease, is a devastatingly fatal neurode- have been implicated in ALS include C9orf72, VCP, FUS, generative disorder for which there are few effective SOD1 and TARDBP (TDP-43) [1–5]. Despite these treatments. ALS is characterised by loss of cortical spinal differing aetiologies, sporadic and familial ALS patients neurons of the motor cortex and alpha motor neurons are clinically and pathologically indistinguishable, sug- within the brainstem and spinal cord, which results in gesting that regardless of whether an ALS patient carries skeletal muscle atrophy and progressive paralysis, even- a known ALS mutation, or is sporadic, the underlying tually leading to death within 2 to 5 years of diagnosis. mechanism of motor neuron dysfunction is similar [6]. In the vast majority of ALS patients (~ 90%), the disease Numerous mechanisms have been proposed to contribute develops sporadically; however, in a minority of cases to ALS pathophysiology, including neuroinflammation. A key mediator of neuroinflammation is the chronic activa- * Correspondence: j.lee9@uq.edu.au; p.noakes@uq.edu.au tion of the complement system, proposed to drive ALS School of Biomedical Sciences, the University of Queensland, St Lucia, disease progression [7]. Brisbane, QLD 4072, Australia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 2 of 13 Multiple clinical and experimental studies have shown 16 months based on their motor deficits were used in this WT compelling evidence that complement activation is in- study [14]. TDP-43 mice have moderate overexpression volved in the pathogenesis of ALS, whereby components levels in total TDP-43, with a 1.5-fold increase in total of all the complement pathways are upregulated in the TDP-43 expression and similar levels of human wild-type serum, cerebrospinal fluid, skeletal muscles and neuro- TDP-43 levels compared with endogenous TDP-43 in NTg Q331K logical tissue (spinal cord and motor cortex) of ALS pa- mice. TDP-43 mice also have moderate overexpres- tients, as well as in transgenic SOD1 animal models of sion levels in total TDP-43, with a 2.5-fold increase in total ALS [8]. Chronic complement activation is proposed to TDP-43 expression and 1.5-fold greater expression of hu- drive ALS disease progression through the actions of the man Q331K TDP-43 levels compared with endogenous pro-inflammatory complement peptide, C5a, signalling TDP-43 in NTg mice [13]. through its main receptor C5aR1 [9, 10]. This patho- genic role of C5a-C5aR1 is proposed to drive disease Hind-limb grip strength test WT progression through inducing glial chemotaxis, activa- The hind-limb grip strength of NTg, TDP-43 and Q331K tion of local immune cells and infiltration of macro- TDP-43 female mice (n = 15 per genotype) was phages into skeletal muscles, thereby inducing an overall measured by using a grip strength meter (IMADA, To- increase in inflammation/neuroinflammation and thus yohashi, Japan) at 3, 10 and 16 months of age at the neurodegeneration [9–12]. However, the pathogenic role same time (14:00 h) of the day as previously described of C5a-C5aR1 signalling in ALS has primarily been by us [9, 15, 16]. In brief, mice were held by their tails shown in transgenic SOD1 rodent models of ALS and lowered until they grasped the T-bar connected to the [9, 12]. Hence, it is unknown whether C5a-C5aR1 patho- digital force transducer with their hind-limbs. The tail was genic signalling is specific to ALS cases not characterised lowered until the body was horizontal, and the mouse was by SOD1 pathology. Thus, as majority of ALS patients pulled away from the T-bar with a smooth steady pull show TDP-43 pathology (~ 95%), the current study until both hind-limbs released the T-bar. The strength of aimed to investigate complement in a recently developed the grip was measured in newtons. Each mouse was given Q331K TDP-43 mouse model of ALS [13]. 10 attempts, and the maximum strength was recorded [9]. We examined the expression of major complement factors and of C5a and its receptor C5aR1, within the Tissue preparation for microglia/astrocyte quantification lumbar spinal cord and tibialis anterior (TA) leg muscle and immunohistochemistry WT Q331K at three different ages during disease progression in Female NTg, TDP-43 and TDP-43 mice (n = 4 per Q331K TDP-43 mice, in order to provide a comprehensive genotype) were euthanized by intraperitoneal injection of overview of the potential involvement of complement in zolazapam (50 mg/kg; Zoletil, Lyppard) and xylazine an alternative mouse model of ALS. Our findings dem- (10 mg/kg; Xylazil, Lyppard). Mice were then fixed by onstrate that a global dysregulation of complement transcardiac perfusion with 2% sodium nitrite in 0.1 M system is involved in this TDP-43 familial mouse model phosphate buffer (pH 7.4; Sigma-Aldrich, St Louis, MO, of ALS, suggesting that complement/C5aR1 could be a USA) followed by 4% paraformaldehyde in 0.1 M phos- potential therapeutic target in most forms of ALS. phate buffer (4% PFA-PB; pH 7.4; Sigma-Aldrich, St Louis, MO, USA) at the previously mentioned ages. Lumbar Methods spinal cords were collected and placed into 4% PFA-PB Animals for 2 h at 4 °C. Following this incubation, the spinal cords WT Q331K Transgenic TDP-43 (Line 96) and TDP-43 (Line were washed 3 × 5 min in phosphate-buffered saline (PBS; 103) mice were obtained from the Jackson Laboratory pH 7.4), followed by submersion in sucrose solution at (Bar Harbor, ME, USA) and were bred on a C57BL/6J back- 15% then 30% in PBS (pH 7.4). Lumbar spinal cords were WT Q331K ground to produce TDP-43 ,TDP-43 and respective then embedded in optimal cutting temperature compound WT non-transgenic (NTg) control mice. TDP-43 transgenic (Sakura, Finetek, Torrance, CA, USA) then snap frozen in mice express a myc-tagged human non-mutated version of liquid nitrogen. Lumbar spinal cords were sectioned into Q331K the TDP-43 cDNA sequence and TDP-43 mice ex- 16-μm-thick transverse and coronal sections and dry press a myc-tagged, human TDP-43 cDNA modified to mounted onto Superfrost Plus slides (Menzel-Glaser, have an ALS-linked glutamine to lysine residue mutation at Braunschweig, Germany) for estimation of astrocytes, position 331, under the direction of the mouse prion pro- microglia and immunohistochemistry as detailed below. tein promoter. The prion protein promoter ensures that the transgene expression is directed primarily to the central Estimation of astrocytes and microglia nervous system—the brain and spinal cord—and is very For estimation of astrocytes and microglia within the lum- low in other peripheral tissues [13]. Female NTg, bar spinal cord, sections were rehydrated in PBS (pH 7.4) WT Q331K TDP-43 and TDP-43 mice at 3, 10 months and then blocked in PBS containing 3% bovine serum albumin Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 3 of 13 (BSA) for 1 h at room temperature. Sections were incu- 555 goat anti-rat, Alexa 594 donkey anti-rat, Alexa 488 goat bated overnight at 4 °C with the astrocyte (mouse anti-mouse, Alexa 488 goat anti-rabbit and Alexa 488 don- anti-GFAP; 1:1000, BD Biosciences, San Diego, CA, USA) key anti-goat (Invitrogen, Eugene, OR, USA). All secondary and microglia (rat anti-CD11b; 1:500, Abcam, Cambridge, antibodies were diluted in PBS (pH 7.4) containing 1% BSA MA, USA) markers. Sections were washed with PBS for or 1% DS (1:1,1000 for Alexa 555, 1:500 for Alexa 594 and 3 × 10 min prior to incubation overnight at 4 °C with the 1:600 for Alexa 488). Following 3 × 5 min washes in PBS, Alexa secondary cocktail: Alexa Fluor 555 dye-conjugated all sections were mounted with Prolong Gold Anti-Fade goat anti-rat (1:1000, Invitrogen, Eugene, OR, USA) and medium containing DAPI (Invitrogen, Eugene, OR, USA). Alexa Fluor 488 dye-conjugated goat anti-mouse (1:600, Sections with no primary antibodies were used as negative Invitrogen, Eugene, OR, USA) antibody. All primary and controls for all immunohistochemistry experiments to give secondary antibodies were diluted in PBS (pH 7.4) con- a measure of non-specific background staining. taining 1% BSA. Sections were then washed for 3 × 5 min Quantification of immunofluorescence for C1q was in PBS, then mounted with Prolong Gold Anti-Fade performed on ~ 25 to 35 lumbar spinal cord sections medium containing 4, 6-diamidino-2-phenylindole (DAPI; (per animal; n = 4) spaced 160 μm apart and expressed Invitrogen, Eugene, OR, USA). Quantification of GFAP as the percentage immunoreactive area per section. and CD11b was performed on ~ 11 to 14 lumbar spinal Staining procedures and image exposures were all stan- cord sections spaced 320 μm apart and expressed as the dardised between genotype and between sections [15]. percentage immunoreactive area per section [12]. Quanti- fication was within the second lumbar dorsal root ganglia (L2) to the fifth lumbar dorsal root ganglia (L5), selected Real-time quantitative PCR with the aid of the mouse spinal cord atlas [17]. Staining Total RNA was isolated from the spinal cord and TA WT Q331K procedures and image exposures were all standardised be- muscle of NTg, TDP-43 and TDP-43 mice using tween genotypes and between sections. The mouse geno- RNeasy Lipid Tissue extraction kit according to manu- type was not made available to the researchers until the facturer’s instructions (QIAGEN, CA, USA). Total RNA completion of the study. was purified from genomic DNA contamination using Turbo DNAse treatment (Ambion, NY, USA), then Quantification of activated microglia numbers converted to cDNA using AffinityScript cDNA synthesis The cell body of microglia was labelled with the nuclear kit according to manufacturer’s instructions (Agilent marker, DAPI. As microglia are known to display morpho- Technologies, CA, USA). Commercially available logical changes when they become activated, such as an gene-specific Taqman probes for complement increase in cell body size, thickening of proximal processes component 1, q subcomponent, beta polypeptide and a decrease in the ramification of distal branches [18], (C1qB; Mm01179619_m1), complement component 4 activated microglia were defined by the presence of one (C4; Mm00437893_g1), complement factor B (Cfb; DAPI stain, an amoeboid cell body and proximal pro- Mm00433909_m1), complement component 3 (C3; cesses length ≤ 1–2 μm[19]. The total number of acti- Mm01232779_m1), CD55 antigen (Cd55; Mm00438 vated microglia was determined by the average of 11–14 377_m1), CD59a antigen (Cd59a; Mm00483149_m1) sections, with the overall average multiplied by the num- and complement component 5a receptor 1 (C5ar1; ber of sections within L2–L5. The mouse genotype was Mm00500292_s1) were used to amplify target gene of not made available to the researchers until quantification interest (Applied Biosystems, MA, USA). Relative target was completed. gene expression to geometric mean of reference genes glyceraldehyde-3-phosphate dehydrogenase (Gapdh; Immunohistochemistry Mm99999915_g1), beta actin (Actb; Mm02619580_g1) Fluorescence double-labelling immunohistochemistry was and hypoxanthine guanine phosphoribosyl transferase performed to localise C1q and C5aR1 expression with spe- (Hprt; Mm03024075_m1) was determined using this -ΔCT cific cell-type markers for astrocytes, microglia and motor formula: 2 where ΔCT = (Ct − Ct (Target gene) (Gapdh, neurons as previously described [15]. Briefly, sections were ), as per our previous studies [15, 16]. Final Actb and Hprt) rehydrated in PBS (pH 7.4) for 10 min, then blocked in PBS measures are presented as relative levels of gene ex- WT Q331K containing 3% BSA or 3% donkey serum (DS) for 1 h at pression in TDP-43 and TDP-43 mice com- room temperature. Sections were incubated overnight at pared with expression in NTg controls. Probe sets were 4 °C with combination of primary antibodies outlined in tested over a serial cDNA concentration for amplifica- Table 1. All primary antibodies were diluted in PBS tion efficiency. No reverse transcription and water as (pH 7.4) containing 1% BSA or 1% DS. Sections were no template control was used as negative controls. All washed 3 × 10 min with PBS prior to incubation with an samples were run in triplicate and were tested in three appropriate Alexa-conjugated secondary cocktail: Alexa separate experiments. Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 4 of 13 Table 1 Summary of antibodies used for immunohistochemistry Antibody Manufacturer Dilution In combination with one of the following antibodies Rat anti-mouse C5aR1 Bio-Rad 1:250 GFAP, Iba-1 or ChAT Rat anti-mouse C1q Hycult Biotechnology 1:1000 GFAP, Iba-1 or ChAT Goat anti-mouse ChAT Merck 1:100 C1q or C5aR1 Rabbit anti-mouse Iba-1 Wako 1:400 C1q or C5aR1 Mouse anti-rat GFAP BD Biosciences 1:1000 C1q or C5aR1 Enzyme-linked immunosorbent assay mice (Fig. 1a). However, by 10 months, we observed a Ninety-six-well plates (Greiner Bio-One, Frickenhausen, significant reduction in hind-limb grip strength in Q331K Germany) were pre-coated with rat anti-mouse C5a cap- TDP-43 mice when compared with NTg and WT ture antibody (4 μg/mL; R&D Systems, Minneapolis, TDP-43 mice (~ 30% reduction, n = 15, ****P < 0.0001, MN, USA) diluted in PBS (pH 7.4) overnight at room ++++P < 0.0001; Fig. 1a). Furthermore, at 16 months, temperature in a sealed humidified container. Following there was a progressive decline in hind-limb grip Q331K the plate being blocked for 1 h at room temperature strength in TDP-43 mice (~ 45% reduction, n = 15, with reagent diluent (1% BSA in PBS), C5a standard, ****P < 0.0001, ++++P < 0.0001; Fig. 1a) when compared WT spinal cord and TA muscle homogenates were incubated with NTg and TDP-43 mice. Importantly, we found for 2 h at room temperature. The plates were subse- that the decline in hind-limb grip strength in Q331K quently incubated with biotinylated goat anti-mouse C5a TDP-43 mice closely correlated with the increase detection antibody (0.2 μg/mL, R&D Systems, Minneap- in immunoreactive area of astrocytes using immuno- olis, MN, USA) for 1 h at room temperature, and then fluorescence staining in the lumbar spinal cords at incubated with Streptavidin-HRP conjugate for 20 min 10 months (~ 200% increase, n =4, **P < 0.01; Fig. 1b, c) at room temperature. Tetramethylbenzidine (Sigma-Al- and 16 months (~ 380% increase, n = 4, ****P < 0.0001; drich, Saint Louis, MO, USA) substrate was used as the Fig. 1b, c). Similarly, we also found that the decline Q331K chromogen, and the plate was read at 450 nm. Levels of in hind-limb grip strength in TDP-43 mice C5a in spinal cord and TA muscle samples were ad- closely correlated with the increase in immunoreactive justed to micrograms per protein and expressed as nano- area of microglia and number of activated microglia grams of C5a per microgram of protein. in the lumbar spinal cord at 10 months (140~230% increase, n =4, **P < 0.01, ***P < 0.001, ****P < 0.0001; Statistical analysis Fig. 1d–f) and 16 months (130~280% increase, n =4, All statistical analyses were performed using GraphPad *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; Fig. 1d–f). Prism 7.0 (GraphPad Software Inc., San Diego, CA, Taken together, these data reveal an age-related decline in Q331K USA). For the results from GFAP and CD11b quantifica- hind-limb grip strength of TDP-43 mice associated tion, quantitative real-time PCR and enzyme-linked im- with increased glial activation in the lumbar spinal cord. munosorbent assay, statistical differences between NTg, WT Q331K TDP-43 and TDP-43 mice were determined Components of the classical/lectin pathways of using one-way ANOVA with Tukey’s post hoc test for complement are upregulated along with decreased each age group. To assess the linear association between expression levels of complement regulator CD55 in TDP- Q331K complement mRNA transcript levels and hind-limb grip 43 mice WT Q331K strength of NTg, TDP-43 and TDP-43 mice, The complement system is part of the innate immune Pearson’s correlation was used. All data was presented as system that can contribute to neuroinflammation in mean ± standard error of mean and differences consid- many neurodegenerative diseases, including ALS [8]. ered significant when P < 0.05. Previous studies, including our own, have identified major complement components are upregulated in the G93A Results lumbar spinal cord of hSOD1 mice [15]. However, Q331K Motor deficits in TDP-43 mice correlate with there is no comprehensive overview of complement sys- increases in astrocytes and microglia during disease tem in different animal models of ALS other than progression hSOD1 transgenic mice [8]. Therefore, we measured the To monitor the decline in motor performance during mRNA levels of key components of the classical/lectin Q331K disease onset and progression in TDP-43 mice, we pathway (C1qB and C4), alternative pathway (fB), the performed hind-limb grip strength tests in animals. At central component to all pathways (C3) and the comple- 3 months, there was no difference in hind-limb grip ment regulators (CD55 and CD59a) in the lumbar spinal WT Q331K WT Q331K strength between NTg, TDP-43 and TDP-43 cord of NTg, TDP-43 and TDP-43 mice using Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 5 of 13 Fig. 1 Decline in hind-limb grip strength during ALS progression correlates with increase in astrocytes and microglia in the lumbar spinal cord of Q331K WT Q331K TDP-43 mice. a Hind-limb grip strength (N) from NTg, TDP-43 and TDP-43 mice at 3, 10 and 16 months of age. At 3 months, no Q331K significant difference in force is present between genotypes. By 10 months, TDP-43 mice display a significant decrease in hind-limb grip WT Q331K WT Q331K strength compared with both NTg and TDP-43 mice (~ 30% reduction; NTg vs TDP-43 ****P < 0.0001 and TDP-43 vs TDP-43 +++ Q331K +P < 0.0001; n = 15). A progressive decline in hind-limb grip strength is present at 16 months, with TDP43 mice showing significantly lower WT Q331K WT hind-limb grip strength compared with NTg and TDP-43 mice (~ 45% reduction; NTg vs TDP-43 ****P < 0.0001 and TDP-43 vs TDP- Q331K WT 43 ++++P < 0.0001; n = 15). b Representative images of the GFAP-positive astrocytes in the lumbar spinal cord of NTg, TDP-43 and TDP- Q331K 43 animals at 16 months. Dashed line shows the outline of the ventral horn with higher magnification of the white square. Scale bars = Q331K WT 100 μm. c Increased astrocyte expression in TDP-43 mice (blue bars) compared with NTg (orange bars) and TDP-43 mice (green bars) at ages 10 and 16 months (**P < 0.01, ****P < 0.0001; n = 4). d Representative images of CD11b-positive microglia in the lumbar spinal cord of NTg, WT Q331K TDP-43 and TDP-43 animals at 16 months. Dashed line shows the outline of the ventral horn with higher magnification of the white Q331K WT square. Scale bar = 100 μm. e Increased microglia expression in TDP-43 mice (blue bars) compared with NTg (orange bars) and TDP-43 Q331K mice (green bars) at ages 10 and 16 months (*P < 0.05, **P < 0.01; n = 4). f Increased number of activated microglia (amoeboid) in TDP-43 WT mice (blue bars) compared with NTg (orange bars) and TDP-43 (green bars) mice at 10 and 16 months (***P < 0.001, ****P < 0.0001; n = 4). Data are presented as mean ± SEM; one-way ANOVA with Tukey’s post hoc test for each age quantitative real-time PCR during disease progression of bar compared to orange and green bars respectively in ALS (3, 10 and 16 months). Fig. 2b; n =5, ***P <0.001, ****P < 0.0001). By contrast, fB Quantitative real-time PCR analyses showed significant did not show any significant changes at 10 and 16 months Q331K Q331K increases of the C1qB transcript in TDP-43 mice by of age in TDP-43 mice when compared to NTg and WT 1.3-fold at 10 months of age when compared to NTg and TDP-43 mice (blue bar compared to orange and green WT TDP-43 mice; at 16 months of age, the increase was bars respectively in Fig. 2c; n =5, P > 0.05). The central 1.5-fold and 1.4-fold when compared to NTg and component of complement, C3, was also increased in the WT Q331K TDP-43 mice, respectively (blue bar compared to lumbar spinal cord of TDP-43 mice, with a 1.6-fold orange and green bars respectively in Fig. 2a; n =5, and 1.4-fold increase at 10 months of age when compared WT ***P < 0.001, ****P < 0.0001). C4 transcript was also in- to NTg and TDP-43 mice, respectively; at 16 months creased by 2.4-fold and 3.3-fold at 10 months of age of age, the increase was 1.7-fold when compared to both and 2.0-fold and 2.7-fold at 16 months of age when genotypes (blue bar compared to orange and green bars WT compared to NTg and TDP-43 mice, respectively (blue respectively in Fig. 2d; n =5, ***P < 0.001, ****P < 0.0001). Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 6 of 13 Q331K Fig. 2 Dysregulation of complement components in the lumbar spinal cord of TDP-43 mice at three different ages of disease progression. a–f The mRNA expression profiles of the following complement components: C1qB (a, classical pathway), C4 (b, classical/lectin pathway), fB Q331K (c, alternative pathway), C3 (d, central component), CD55 (e, regulator) and CD59a (f, regulator) in the lumbar spinal cord of TDP-43 mice WT (blue bars) relative to non-transgenic (NTg, orange bars) and TDP-43 (green bars) mice during 3, 10 and 16 months of age. Data are expressed as means ± SEM (n = 5 mice/group, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, one-way ANOVA with Tukey’s post hoc test) WT The negative regulators of the complement system, NTg and TDP-43 mice (blue bar compared to orange CD55 and CD59a, were also investigated due to their im- and green bars in Fig. 3a; n =4, *P < 0.05). We also ob- Q331K portance in maintaining homeostasis and keeping the served that the marked increase of C1q in TDP-43 complement system in its proper physiological state in re- mice was localised to motor neurons and microglia sponse to altered physiological conditions (i.e. infection (white arrows in Fig. 3d, g), compared with NTg and WT and/or neurodegeneration). CD55 mRNA expression in TDP-43 mice where little to no C1q was observed Q331K TDP-43 mice was decreased at 10 months of age by (Fig. 3b, c, e, f). We did not observe C1q on astrocytes WT Q331K 0.3-fold and 0.4-fold, and by 0.4-fold and 0.5-fold at in either NTg, TDP-43 or TDP-43 mice 16 months of age, when compared with NTg and (Fig. 3h–j). WT TDP-43 mice (blue bar compared to orange and green bars respectively in Fig. 2e; n =5, *P <0.05, **P <0.01, The terminal complement pathway C5a receptor, C5aR1, ****P < 0.0001). By contrast, CD59a mRNA expression did is upregulated and expressed on motor neurons and Q331K not significantly alter at 10 and 16 months of age in microglia in the lumbar spinal cord of TDP-43 mice Q331K TDP-43 mice when compared to NTg and Previous studies have shown increases in C5aR1 expres- WT TDP-43 mice (blue bar compared to orange and green sion in the central nervous system of multiple rodent bars respectively in Fig. 2f; n =5, P > 0.05). These results models of ALS, with many studies suggesting a pathogenic suggest widespread complement perturbation occurs in role for C5aR1 in the disease progression of ALS in Q331K G93A the lumbar spinal cord of TDP-43 mice, which may hSOD1 mice [9, 10, 12, 20]. C5a, the ligand for contribute to glial activation and neuroinflammation, and C5aR1, is an activation fragment of the terminal comple- ultimately disease progression in this model. ment cascade that is rapidly generated following comple- Upregulation of C1q at 16 months of age was also ment cascade initiation [21]. We therefore examined the Q331K confirmed using immunofluorescence, where there was protein levels of C5a in the spinal cord of TDP-43 Q331K marked increase in TDP-43 mice compared with mice using enzyme-linked immunosorbent assay as a Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 7 of 13 WT Q331K Fig. 3 Expression and localisation of C1q in the spinal cord of non-transgenic, TDP-43 and TDP-43 mice at 16 months of age. a The WT Q331K immunoreactive area of C1q in the spinal cord of non-transgenic (NTg, orange bars), TDP-43 (green bars) and TDP-43 (blue bars) at 16 months. Data are expressed as means ± SEM (n = 4 mice/group, *P < 0.05, one-way ANOVA with Tukey’s post hoc test). b–j Double immunolabelling of C1q (red) with cellular markers (green) for motor neurons (ChAT; b–d), microglia (Iba-1; e–g) and astrocyte (GFAP; h–j) in the WT Q331K ventral lumbar spinal cord of NTg, TDP-43 and TDP-43 mice at 16 months of age. C1q was co-localised with ChAT-positive motor neurons Q331K (d, white arrow) and microglia (g, white arrow) in TDP-43 mice. Note that white arrows indicate red and green fluorescent signal merge to Q331K orange. In TDP-43 mice, immunolabelling of C1q was also evident on other cell types, indicated by lack of co-localisation with anti-ChAT, anti-Iba-1 and anti-GFAP (d, g, j). Scale bars for all panels = 20 μm biomarker for terminal complement activation. Interest- whereas it was not observed in GFAP-positive astrocytes ingly, the results showed no change in C5a protein levels (Fig. 4i–k). Together, the results above indicate that Q331K in TDP-43 mice at 16 months of age when compared C5a-C5aR1 signalling could play a role in facilitating WT to NTg and TDP-43 mice (blue bar compared to or- microglia activation and phagocytosis ultimately leading ange and green bars in Fig. 4a; n =5, P > 0.05). In contrast to motor neuron death in these animals. to protein levels of C5a, C5aR1 mRNA expression was sig- nificantly increased by 1.4-fold and 1.2-fold at 10 months of age and by 1.8-fold and 1.6-fold at 16 months of age, Dysregulation of complement in the tibialis anterior WT Q331K when compared to NTg and TDP-43 mice (blue bar muscle of TDP-43 mice compared to orange and green bars in Fig. 4b; n =5, In addition to showing altered levels of complement Q331K *P < 0.05, ***P < 0.001, ****P < 0.0001). components in the spinal cord of TDP-43 mice, we We next aimed to determine the cellular localisation also investigated the level of major complement compo- Q331K of C5aR1 that could explain the increased expression nents in the TA muscle of TDP-43 mice, as it has Q331K G93A seen in TDP-43 mice. To achieve this, we per- been shown previously in hSOD1 mice that comple- formed immunolabelling for C5aR1 on the lumbar spinal ment is upregulated in this muscle. To investigate this, WT Q331K cord from NTg, TDP-43 and TDP-43 mice. we measured the mRNA levels of key components of the These sections were immunostained for C5aR1 with spe- complement pathways, which include the classical/lectin cific cellular markers to identify motor neurons (anti-C- (C1qB and C4), alternative (fB) and terminal pathways hAT), astrocytes (anti-GFAP) and microglia (anti-Iba-1). (C5a and C5aR1), as well as the major complement reg- C5aR1 localised to ChAT-positive motor neurons and ulators (CD55 and CD59a) using quantitative real-time WT Iba-1-positive microglia in NTg, TDP-43 and PCR and enzyme-linked immunosorbent assay for C5a Q331K Q331K TDP-43 mice at 16 months of age (white arrows in in TA muscle of TDP-43 mice during disease pro- Fig. 4c–e for motor neurons and Fig. 4f–h for microglia), gression of ALS (3, 10 and 16 months). Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 8 of 13 Q331K Fig. 4 Expression of C5a and C5aR1 in the spinal cord of TDP-43 mice at three different ages of disease progression. a The protein WT Q331K expression of C5a in the spinal cord of non-transgenic (NTg, orange bars), TDP-43 (green bars) and TDP-43 (blue bars) at 16 months. b Q331K WT mRNA expression of C5aR1 in the spinal cord of TDP-43 mice relative to age-matched NTg and TDP-43 mice at 3, 10 and 16 months of age. Data are expressed as means ± SEM (n = 5 mice/group, *P < 0.05, ***P < 0.001, ****P < 0.0001, one-way ANOVA with Tukey’s post hoc test for each age). c–k Double immunolabelling of C5aR1 (red) with cellular markers (green) for motor neurons (ChAT; c–e), microglia (Iba-1; f–h) and WT Q331K astrocyte (GFAP; i–k) in the ventral lumbar spinal cord of NTg, TDP-43 and TDP-43 mice at 16 months of age. C5aR1 was co-localised with WT Q331K ChAT-positive motor neurons (c–e, white arrows) and microglia (f–h, white arrows) in NTg, TDP-43 and TDP-43 mice. Note that white arrows indicate red and green fluorescent signal merge to orange. Scale bars for all panels = 20 μm WT C1qB and C4 transcripts were significantly increased 10 months of age when compared to NTg and TDP-43 by 1.6-fold and 1.9-fold when compared to NTg mice mice, whereas 1.5-fold increase at 16 months of age when WT WT and by 1.5-fold when compared to TDP-43 mice at compared to TDP-43 mice (blue bar compared to or- 10 months of age, respectively (blue bar compared to or- ange and green bars in Fig. 5c; n =5, *P <0.05, **P <0.01). ange and green bars in Fig. 5a, b; n =5, *P < 0.05, By contrast, C3 was surprisingly decreased by 0.3-fold in Q331K **P < 0.01, ***P < 0.001). C1qB and C4 transcripts were TA muscle of TDP-43 mice when compared to NTg WT also increased by 1.7-fold and 1.5-fold when compared and TDP-43 mice at 10 months of age (blue bar com- to NTg mice and by 1.7-fold and 1.9-fold when com- pared to orange and green bars in Fig. 5d; n =5, *P <0.05, WT pared to TDP-43 mice at 16 months of age, respect- **P < 0.01). The regulator, CD55, was increased by 1.3-fold Q331K ively (blue bar compared to orange and green bars in at 10 months of age in TDP-43 mice when compared Fig. 5a, b; n =5, *P < 0.05, **P < 0.01, ***P < 0.001). In to NTg mice (blue bar compared to orange bar in Fig. 5e; addition to C1qB and C4, fB also displayed a marked in- n =5, *P < 0.05), while there was no significant change in Q331K crease in mRNA levels by 1.4-fold and 1.3-fold at CD59a in TDP-43 mice when compared to NTg and Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 9 of 13 Q331K Fig. 5 Dysregulation of complement components in tibialis anterior muscle of TDP-43 mice at three different ages of disease progression. a– f The mRNA expression profiles of the following complement components: C1qB (a, classical pathway), C4 (b, classical/lectin pathway), fB (c, Q331K alternative pathway), C3 (d, central component), CD55 (e, regulator) and CD59a (f, regulator) in tibialis anterior muscle of TDP-43 mice (blue WT bars) relative to non-transgenic (NTg, orange bars) and TDP-43 (green bars) mice during 3, 10 and 16 months of age. g The protein expression WT Q331K of C5a in the tibialis anterior muscle of NTg (orange bars), TDP-43 (green bars) and TDP-43 (blue bars) mice at 16 months. h mRNA Q331K WT expression of C5aR1 in the tibialis anterior muscle of TDP-43 mice relative to age-matched NTg and TDP-43 mice at three different ages. Data are expressed as means ± SEM (n = 5 mice/group, *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with Tukey’s post hoc test for each age) WT TDP-43 controls (Fig. 5f; n =5, P > 0.05). Lastly, the ter- with their hind-limb grip strength. We found a strong minal complement pathway component, C5a, and its re- negative correlation between lumbar spinal cord mRNA ceptor C5aR1 were investigated. The results showed expression of C1qB (r = − 0.6824), C3 (r = − 0.8282) and significant increases in C5a protein at 16 months of age by C5aR1 (r = − 0.7428) to hind-limb grip strength of these 1.8-fold and 2.0-fold when compared to NTg and animals, with C3 presenting the strongest correlation WT TDP-43 mice respectively (blue bar compared to or- (n = 30, ****P < 0.0001; Fig. 6a–c). Furthermore, a moder- ange and green bars in Fig. 5g; n =5, ***P <0.001). C5aR1 ate to strong negative correlation between TA mRNA mRNA expression was also significantly increased by levels of C1qB (r = − 0.6624) and C5aR1 (r = − 0.4116) to 1.7-fold and 1.6-fold at 10 months of age and by 1.7-fold hind-limb grip strength was observed, whereas a moderate and 2.0-fold at 16 months of age when compared to NTg positive correlation was identified between C3 mRNA WT and TDP-43 controls, respectively (blue bar compared levels (r = 0.4149) to hind-limb grip strength (n =25, to orange and green bars in Fig. 5h; n =5, *P <0.05, *P < 0.05, ***P < 0.001; Fig. 6d–f). These results indicate **P <0.01, ***P < 0.001). Taken together, these results sug- that changes in complement transcript levels in the spinal gest that dysregulation of the complement system also oc- cord and TA muscle directly correlates with the decrease Q331K curs in the TA muscle of TDP-43 mice, which could in hind-limb grip strength (i.e. increased in ALS symp- contribute to the disease pathology in these animals. toms) in this ALS model. Q331K Motor dysfunction in TDP-43 mice strongly Discussion correlates with lumbar spinal cord levels of C1qb, C3 and The major findings of the current study are that compo- C5aR1 nents of the complement system are dysregulated in a We next examined if there was any correlation between transgenic mutant TDP43-based mouse model of ALS. major complement transcript levels to the hind-limb It has now been well documented that the complement Q331K grip strength of TDP-43 mice. To investigate this cascade is synthesised by neurons, astrocytes and micro- aspect, we performed a Pearson correlation to measure glia within the central nervous system and is involved in the strength of the linear relationship between C1qB, C3 the disease progression of ALS, with evidence from both and C5aR1 mRNA levels in the lumbar spinal cord and human patients and rodent models [8]. The present WT Q331K TA muscle of NTg, TDP-43 and TDP-43 mice study further adds to this knowledge, demonstrating Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 10 of 13 Q331K Fig. 6 Motor dysfunction in TDP-43 mice strongly correlates with lumbar spinal cord/tibialis anterior muscle levels of C1qB, C3 and C5aR1. A strong negative correlation between lumbar spinal cord mRNA expression of C1qB (a, r = − 0.6824), C3 (b, r = − 0.8282) and C5aR1 (c, r = − 0.7428) WT Q331K with hind-limb grip strength of non-transgenic (NTg), TDP-43 and TDP-43 mice, where C3 demonstrated the strongest correlation (n = 30, ****P < 0.0001, Pearson correlation). A moderate to strong negative correlation between tibialis anterior muscle mRNA levels of C1qB (d, r = − WT Q331K 0.6624) and C5aR1 (f, r = − 0.4116) with hind-limb grip strength of NTg, TDP-43 and TDP-43 mice, whereas a moderate positive correlation between C3 mRNA levels (e, r = 0.4149) was observed (n = 25, *P < 0.05, ***P < 0.001, Pearson correlation). Pearson correlation coefficient was used to measure the strength of the linear relationship between complement mRNA expression and hind-limb grip strength Q331K alteration of mRNA expression in major complement fB in the TA muscles of TDP-43 mice, however, factors including C1qB, C4, fB, C3, C5a, C5aR1 and founddownregulation of C3inthe TA musclesof Q331K regulator CD55 in the spinal cord and TA muscles of TDP-43 mice, which is contradictory of what is Q331K G93A TDP-43 mice, which is suggestive of a progressive showninhSOD1 mice. The upregulation of C1qB dysregulation of complement in this model. These results in the TA muscles could assist in the removal of degen- are similar to our previous demonstration of complement erating neuromuscular synapses via phagocytosis dur- Q331K dysregulation in the spinal cord and TA muscle of ing disease progression in TDP-43 mice, similar to G93A G93A hSOD1 mice [11, 15], indicating that complement ac- what has been shown in hSOD1 mice [22]. The tivation occurs in response to motor neuron death and reason for the differential expression of C3 in the TA Q331K muscle denervation regardless of which ALS-related gene muscle of TDP-43 mice compared with G93A mutation is present. Furthermore, complement activation, hSOD1 mice is unclear, but could be due to the differ- Q331K and subsequent C5aR1 activation, could be a common ence in the severity of the disease with TDP-43 mice mechanism of pathology in most forms of ALS. showing milder ALS-like symptoms compared to G93A Thepresent studyprovidedevidencefor thedysregula- hSOD1 mice [13, 14]. tion of classical/lectin, alternative and central component We additionally showed decreased mRNA expression to all pathways of the complement system in the spinal levels of CD55 in the spinal cord and increased expression Q331K cord and TA muscle of TDP-43 mice during ALS levels in TA muscle, which suggests that homeostatic bal- diseaseprogression.Thisisinlinewithnumerousstud- ance of the complement system may be perturbed in Q331K ies including our own, where increased C1qB, C4 and TDP-43 mice, leading to the over activation of com- G93A C3 was found in the spinal cord of hSOD1 mice plement system. These findings support other studies, [11, 15, 22, 23]. This upregulation of C1qB and C3 in which have also shown decreased CD55 mRNA levels in the spinal cord could assist in the removal of dying the spinal cord, with deficiency in CD55 exacerbating neu- motor neurons via opsonisation through microglia acti- rodegeneration [15, 22, 24]. Our findings also support pre- Q331K vation during disease progression in TDP-43 mice, vious studies identifying increased CD55 expression in G93A similar to what is evident in hSOD1 mice [15]. intercostal muscles of ALS patients and TA muscles of G93A Similarly, we identified upregulation of C1qB, C4 and hSOD1 mice [11, 25]. Upregulation of CD55 at the Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 11 of 13 G93A Q331K motor end-plates of ALS patients and hSOD1 mice expression of C5aR1 in TDP-43 spinal cord and could be a mechanism to protect against high levels of TA muscles is elevated, confirming dysregulation of complement activation at the neuromuscular junction. downstream terminal complement pathway in these ani- Interestingly, the current study did not demonstrate any mals. This suggests that enhanced C5a-C5aR1 signalling changes in the other regulator CD59a in the spinal cord may affect the disease progression of ALS in both spinal Q331K Q331K or TA muscle of TDP-43 mice. This is in contrast to cord and TA muscles of TDP-43 mice, similar to G93A previous studies, where CD59a was decreased in the what is evident in hSOD1 mice [9, 11, 15]. In G93A Q331K spinal cord of hSOD1 mice, while it increased in the addition to an increase in C5aR1 in TDP-43 mice, intercostal muscle of ALS patients and TA muscle of the present study also revealed C5aR1 on motor neurons G93A WT Q331K hSOD1 mice [11, 15, 25]. The difference in the ex- and microglia in NTg, TDP-43 and TDP-43 pression changes between different models and patients mice. This is in concordance with previous studies could be attributed to the severity of the disease, as showing C5aR1 on motor neurons and microglia in Q331K TDP-43 mice show milder ALS-like symptoms com- other mouse models of ALS [15, 20], indicating that G93A pared to hSOD1 mice [9, 13–15]. However, further C5a-C5aR1 signalling could play a role in facilitating investigation is required into the expression and localisa- microglia activation and motor neuron death regardless Q331K tion of membrane attack complex in TDP-43 mice, of which ALS-related gene mutation is present. and its putative correlation with motor neuron loss. Re- In line with increased complement components in Q331K gardless, collectively our data adds further support to the TDP-43 mice, the present study also revealed a notion that dysregulated complement activity may play an strong correlation between C1qB, C3 and C5aR1 important role in accelerating motor neuron loss and mRNA levels in the spinal cord and TA muscles to the neuromuscular junction denervation, ultimately driving hind-limb grip strength of these animals, suggesting the progression of ALS. that the mRNA levels of complement system could be a Among the complement activation effector molecules, direct correlate of motor neuron loss and neuromuscu- C5a is considered the most potent peptide, with its signal- lar junction denervation throughout disease progres- ling through its main receptor C5aR1 having detrimental sion, that we and others have previously demonstrated effects in multiple neurodegenerative diseases, including in this mouse model [13, 14]. These finding are con- ALS [9, 10, 12, 26, 27]. To obtain a better understanding of sistent with increased astrocyte and microglia numbers/ its role in disease progression of ALS, we analysed C5a and activation in these animals, indicating that neuroin- its receptor C5aR1 in the spinal cord and TA muscle of flammation could be a good indicator of disease sever- Q331K TDP-43 mice. C5a protein levels in the spinal cord ity and pathology. WT Q331K did not change between NTg, TDP-43 and TDP-43 mice, whereas it increased in the TA muscles of Q331K Conclusions TDP-43 mice. This is in line with C5a protein levels at In summary, the current study has demonstrated upreg- mid-symptomatic stage in the spinal cord and TA muscle G93A ulation of major complement factors, together with de- of hSOD1 mice [11], suggesting that enhanced creased levels of the negative complement regulator C5a-C5aR1 signalling may affect disease progression of Q331K G93A CD55, in TDP-43 mice. This suggests that comple- ALS, similar to what has been found in hSOD1 mice. ment activation and/or its dysregulation could play an This finding is interesting given that C3 mRNA transcript important role in motor neuron loss and neuromuscular levels were higher in the spinal cord compared to TA junction denervation in this TDP43-based mouse model muscle in these animals. The reason for the increase in of ALS. Expression of the C5a receptor, C5aR1, was also muscle C5a, despite reduced C3 transcript expression is un- Q331K upregulated in TDP-43 mice, predominantly due to clear; however, it should be noted that mRNA expression increased microglial/macrophage C5aR1 expression, and does not directly equate with the degree of complement was strongly correlated with resulting motor decline. activation. Furthermore, there is a well-described Taken together, these results indicate that heightened C3-bypass (extrinsic) complement activation pathway that complement activation and enhanced C5aR1 signalling enables C3-independent cleavage of C5 [28]. Thus, C5a may play a crucial role in pathophysiology of the protein levels do not necessarily correlate with C3 mRNA Q331K TDP-43 ALS model, further validating C5aR1 as a expression within a tissue. potential therapeutic target for all forms of ALS. Numerous studies have demonstrated upregulation of C5aR1 within the spinal cord and TA muscles of G93A hSOD1 rats and mice, as well as human ALS pa- Abbreviations tients, suggesting that heightened C5a-C5aR1 signalling ALS: Amyotrophic lateral sclerosis; BSA: Bovine serum albumin; plays a role in ALS pathology [9–12, 15, 29]. In the DAPI: Diamidino-2-phenylindole; DS: Donkey serum; NTg: Non-transgenic; present study, we demonstrated that the mRNA TA: Tibialis anterior Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 12 of 13 Acknowledgements 6. Taylor JP, Brown RH Jr, Cleveland DW. Decoding ALS: from genes to The authors would like to sincerely thank Associate Professor Massimo A. mechanism. Nature. 2016;539:197–206. Hilliard for carefully going through the manuscript and giving us feedback to 7. Woodruff TM, Costantini KJ, Taylor SM, Noakes PG. Role of complement in improve our manuscript. We also thank Maryam Shayegh for her technical motor neuron disease: animal models and therapeutic potential of support with genotyping mice, Mary White for general laboratory support complement inhibitors. Adv Exp Med Biol. 2008;632:143–58. and Dr. Shaun Walters for his support with microscopy. 8. Brennan FH, Lee JD, Ruitenberg MJ, Woodruff TM. Therapeutic targeting of complement to modify disease course and improve outcomes in Funding neurological conditions. Semin Immunol. 2016;28:292–308. JDL holds a Motor Neuron Disease Research Institute of Australia (MNDRIA) 9. Lee JD, Kumar V, Fung JN, Ruitenberg MJ, Noakes PG, Woodruff TM. Postdoctoral Fellowship. TMW is supported by a NHMRC Career Pharmacological inhibition of complement C5a-C5a1 receptor signalling Development Fellowship. The research was funded by grants from the ameliorates disease pathology in the hSOD1(G93A) mouse model of MNDRIA to PGN (GIA1429 and GIA1547), TMW (GIA1865) and JDL (GIA1830) amyotrophic lateral sclerosis. Br J Pharmacol. 2017;174:689–99. and a National Health and Medical Research Council (NHMRC) Project grant 10. Woodruff TM, Lee JD, Noakes PG. Role for terminal complement activation (APP1082271) to TMW. in amyotrophic lateral sclerosis disease progression. Proc Natl Acad Sci U S A. 2014;111:E3–4. 11. Wang HA, Lee JD, Lee KM, Woodruff TM, Noakes PG. Complement C5a-C5aR1 Authors’ contributions signalling drives skeletal muscle macrophage recruitment in the hSOD1(G93A) JDL, TMW and PGN conceived the project. JDL, TMW and PGN designed the mouse model of amyotrophic lateral sclerosis. Skelet Muscle. 2017;7:10. study. JDL, SCL, EFW and RL performed the experiments in the laboratories 12. Woodruff TM, Costantini KJ, Crane JW, Atkin JD, Monk PN, Taylor SM, of PGN and TMW. All authors contributed to the analyses and/or interpreted Noakes PG. The complement factor C5a contributes to pathology in a rat the data. JDL wrote the paper with contributions from TMW and PGN. All model of amyotrophic lateral sclerosis. J Immunol. 2008;181:8727–34. authors read and approved the final manuscript. 13. Arnold ES, Ling SC, Huelga SC, Lagier-Tourenne C, Polymenidou M, Ditsworth D, Kordasiewicz HB, McAlonis-Downes M, Platoshyn O, Parone PA, Ethics approval et al. ALS-linked TDP-43 mutations produce aberrant RNA splicing and All experimental procedures were approved by the University of Queensland adult-onset motor neuron disease without aggregation or loss of nuclear Animal Ethics Committee and complied with the policies and regulations TDP-43. Proc Natl Acad Sci U S A. 2013;110:E736–45. regarding animal experimentation. They were conducted in accordance with 14. Chand KK, Lee KM, Lee JD, Qiu H, Willis EF, Lavidis NA, Hilliard MA, Noakes the Queensland Government Animal Research Act 2001, associated Animal PG. Defects in synaptic transmission at the neuromuscular junction Care and Protection Regulations (2002 and 2008) and the Australian Code of precedes motor deficits in a TDP-43(Q331K) transgenic mouse model of Practice for the Care and Use of Animals for Scientific Purposes, 8th Edition amyotrophic lateral sclerosis. FASEB J. 2018;32:2676–89. (National Health and Medical Research Council, 2013). ARRIVE guidelines 15. Lee JD, Kamaruzaman NA, Fung JN, Taylor SM, Turner BJ, Atkin JD, Woodruff have been followed in the preparation of the manuscript. TM, Noakes PG. Dysregulation of the complement cascade in the hSOD1G93A transgenic mouse model of amyotrophic lateral sclerosis. J Competing interests Neuroinflammation. 2013;10:119. The authors declare that they have no competing interests. 16. Lee JY, Lee JD, Phipps S, Noakes PG, Woodruff TM. Absence of toll-like receptor 4 (TLR4) extends survival in the hSOD1 G93A mouse model of amyotrophic lateral sclerosis. J Neuroinflammation. 2015;12:90. Publisher’sNote 17. Watson C, Paxinos G, Kayalioglu G. The spinal cord. 1st ed. London: Springer Nature remains neutral with regard to jurisdictional claims in published Academic Press; 2009. maps and institutional affiliations. 18. Li Q, Barres BA. Microglia and macrophages in brain homeostasis and disease. Nat Rev Immunol. 2017;18:225–42. Author details School of Biomedical Sciences, the University of Queensland, St Lucia, 19. Philips T, Robberecht W. Neuroinflammation in amyotrophic lateral Brisbane, QLD 4072, Australia. University of Queensland Centre for Clinical sclerosis: role of glial activation in motor neuron disease. Lancet Neurol. Research, the University of Queensland, Herston, Brisbane, QLD 4029, 2011;10:253–63. Australia. Queensland Brain Institute, the University of Queensland, St Lucia, 20. Humayun S, Gohar M, Volkening K, Moisse K, Leystra-Lantz C, Mepham Brisbane, QLD 4072, Australia. J, McLean J, Strong MJ. The complement factor C5a receptor is upregulated in NFL−/− mouse motor neurons. J Neuroimmunol. Received: 5 March 2018 Accepted: 24 May 2018 2009;210:52–62. 21. Woodruff TM, Ager RR, Tenner AJ, Noakes PG, Taylor SM. The role of the complement system and the activation fragment C5a in the central nervous system. NeuroMolecular Med. 2010;12:179–92. References 22. Heurich B, El Idrissi NB, Donev RM, Petri S, Claus P, Neal J, Morgan BP, 1. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, Ramaglia V. Complement upregulation and activation on motor neurons Rutherford NJ, Nicholson AM, Finch NA,Flynn H, Adamson J,etal. and neuromuscular junction in the SOD1 G93A mouse model of familial Expanded GGGGCC hexanucleotide repeat in noncoding region of amyotrophic lateral sclerosis. J Neuroimmunol. 2011;235:104–9. C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron. 23. Lobsiger CS, Boillee S, Pozniak C, Khan AM, McAlonis-Downes M, Lewcock 2011;72:245–56. JW, Cleveland DW. C1q induction and global complement pathway 2. Kwiatkowski TJ Jr, Bosco DA, Leclerc AL, Tamrazian E, Vanderburg CR, Russ activation do not contribute to ALS toxicity in mutant SOD1 mice. Proc Natl C, Davis A, Gilchrist J, Kasarskis EJ, Munsat T, et al. Mutations in the FUS/TLS Acad Sci U S A. 2013;110:E4385–92. gene on chromosome 16 cause familial amyotrophic lateral sclerosis. 24. Wang Y, Li Y, Dalle Lucca SL, Simovic M, Tsokos GC, Dalle Lucca JJ. Decay Science. 2009;323:1205–8. accelerating factor (CD55) protects neuronal cells from chemical hypoxia- 3. Rosen DR, Siddique T, Patterson D, Figlewicz DA, Sapp P, Hentati A, induced injury. J Neuroinflammation. 2010;7:24. Donaldson D, Goto J, O'Regan JP, Deng H-X. Mutations in Cu/Zn superoxide 25. Bahia El Idrissi N, Bosch S, Ramaglia V, Aronica E, Baas F, Troost D. dismutase gene are associated with familial amyotrophic lateral sclerosis. Complement activation at the motor end-plates in amyotrophic lateral Nature. 1993;362:59–62. sclerosis. J Neuroinflammation. 2016;13:72. 4. Sreedharan J, Blair IP, Tripathi VB, Hu X, Vance C, Rogelj B, Ackerley S, Durnall JC, Williams KL, Buratti E, et al. TDP-43 mutations in familial and 26. Fonseca MI, Ager RR, Chu SH, Yazan O, Sanderson SD, LaFerla FM, Taylor sporadic amyotrophic lateral sclerosis. Science. 2008;319:1668–72. SM, Woodruff TM, Tenner AJ. Treatment with a C5aR antagonist decreases 5. Williams KL, Warraich ST, Yang S, Solski JA, Fernando R, Rouleau GA, pathology and enhances behavioral performance in murine models of Nicholson GA, Blair IP. UBQLN2/ubiquilin 2 mutation and pathology in Alzheimer's disease. J Immunol. 2009;183:1375–83. familial amyotrophic lateral sclerosis. Neurobiol Aging. 2012;33:2527 27. Woodruff TM,Crane JW, Proctor LM,Buller KM, Shek AB,deVos K, e2523–10. Pollitt S, Williams HM, Shiels IA, Monk PN, Taylor SM. Therapeutic Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 13 of 13 activity of C5a receptor antagonists in a rat model of neurodegeneration. FASEB J. 2006;20:1407–17. 28. Huber-Lang M, Younkin EM, Sarma JV, Riedemann N, McGuire SR, Lu KT, Kunkel R, Younger JG, Zetoune FS, Ward PA. Generation of C5a by phagocytic cells. Am J Pathol. 2002;161:1849–59. 29. Mantovani S, Gordon R, Macmaw JK, Pfluger CM, Henderson RD, Noakes PG, McCombe PA, Woodruff TM. Elevation of the terminal complement activation products C5a and C5b-9 in ALS patient blood. J Neuroimmunol. 2014;276:213–8. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Neuroinflammation Springer Journals

Complement components are upregulated and correlate with disease progression in the TDP-43Q331K mouse model of amyotrophic lateral sclerosis

Free
13 pages
Loading next page...
 
/lp/springer_journal/complement-components-are-upregulated-and-correlate-with-disease-DrwO5vDuxO
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Neurology; Neurobiology; Immunology
eISSN
1742-2094
D.O.I.
10.1186/s12974-018-1217-2
Publisher site
See Article on Publisher Site

Abstract

Background: Components of the innate immune complement system have been implicated in the pathogenesis of amyotrophic lateral sclerosis (ALS) specifically using hSOD1 transgenic animals; however, a comprehensive examination of complement expression in other transgenic ALS models has not been performed. This study therefore aimed to determine the expression of several key complement components and regulators in the Q331K lumbar spinal cord and tibialis anterior muscle of TDP-43 mice during different disease ages. WT Q331K Methods: Non-transgenic, TDP-43 and TDP-43 mice were examined at three different ages of disease progression. Expression of complement components and their regulators were examined using real-time quantitative PCR and enzyme-linked immunosorbent assay. Localisation of terminal complement component receptor C5aR1 within the lumbar spinal cord was also investigated using immunohistochemistry. Results: Altered levels of several major complement factors, including C5a, in the spinal cord and tibialis anterior Q331K muscle of TDP-43 mice were observed as disease progressed, suggesting overall increased complement Q331K activation in TDP-43 mice. C5aR1 increased during disease progression, with immuno-localisation demonstrating expression on motor neurons and expression on microglia surrounding the regions of motor neuron death. There was a strong negative linear relationship between spinal cord C1qB, C3 and C5aR1 mRNA levels with hind-limb grip strength. Conclusions: These results indicate that similar to SOD1 transgenic animals, local complement activation and increased Q331K expression of C5aR1 may contribute to motor neuron death and neuromuscular junction denervation in the TDP-43 mouse ALS model. This further validates C5aR1 as a potential therapeutic target for ALS. Background (~ 10%), the disease has a familial component, and it is Amyotrophic lateral sclerosis (ALS), also known as due to specific genetic mutations. Some of the genes that motor neuron disease, is a devastatingly fatal neurode- have been implicated in ALS include C9orf72, VCP, FUS, generative disorder for which there are few effective SOD1 and TARDBP (TDP-43) [1–5]. Despite these treatments. ALS is characterised by loss of cortical spinal differing aetiologies, sporadic and familial ALS patients neurons of the motor cortex and alpha motor neurons are clinically and pathologically indistinguishable, sug- within the brainstem and spinal cord, which results in gesting that regardless of whether an ALS patient carries skeletal muscle atrophy and progressive paralysis, even- a known ALS mutation, or is sporadic, the underlying tually leading to death within 2 to 5 years of diagnosis. mechanism of motor neuron dysfunction is similar [6]. In the vast majority of ALS patients (~ 90%), the disease Numerous mechanisms have been proposed to contribute develops sporadically; however, in a minority of cases to ALS pathophysiology, including neuroinflammation. A key mediator of neuroinflammation is the chronic activa- * Correspondence: j.lee9@uq.edu.au; p.noakes@uq.edu.au tion of the complement system, proposed to drive ALS School of Biomedical Sciences, the University of Queensland, St Lucia, disease progression [7]. Brisbane, QLD 4072, Australia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 2 of 13 Multiple clinical and experimental studies have shown 16 months based on their motor deficits were used in this WT compelling evidence that complement activation is in- study [14]. TDP-43 mice have moderate overexpression volved in the pathogenesis of ALS, whereby components levels in total TDP-43, with a 1.5-fold increase in total of all the complement pathways are upregulated in the TDP-43 expression and similar levels of human wild-type serum, cerebrospinal fluid, skeletal muscles and neuro- TDP-43 levels compared with endogenous TDP-43 in NTg Q331K logical tissue (spinal cord and motor cortex) of ALS pa- mice. TDP-43 mice also have moderate overexpres- tients, as well as in transgenic SOD1 animal models of sion levels in total TDP-43, with a 2.5-fold increase in total ALS [8]. Chronic complement activation is proposed to TDP-43 expression and 1.5-fold greater expression of hu- drive ALS disease progression through the actions of the man Q331K TDP-43 levels compared with endogenous pro-inflammatory complement peptide, C5a, signalling TDP-43 in NTg mice [13]. through its main receptor C5aR1 [9, 10]. This patho- genic role of C5a-C5aR1 is proposed to drive disease Hind-limb grip strength test WT progression through inducing glial chemotaxis, activa- The hind-limb grip strength of NTg, TDP-43 and Q331K tion of local immune cells and infiltration of macro- TDP-43 female mice (n = 15 per genotype) was phages into skeletal muscles, thereby inducing an overall measured by using a grip strength meter (IMADA, To- increase in inflammation/neuroinflammation and thus yohashi, Japan) at 3, 10 and 16 months of age at the neurodegeneration [9–12]. However, the pathogenic role same time (14:00 h) of the day as previously described of C5a-C5aR1 signalling in ALS has primarily been by us [9, 15, 16]. In brief, mice were held by their tails shown in transgenic SOD1 rodent models of ALS and lowered until they grasped the T-bar connected to the [9, 12]. Hence, it is unknown whether C5a-C5aR1 patho- digital force transducer with their hind-limbs. The tail was genic signalling is specific to ALS cases not characterised lowered until the body was horizontal, and the mouse was by SOD1 pathology. Thus, as majority of ALS patients pulled away from the T-bar with a smooth steady pull show TDP-43 pathology (~ 95%), the current study until both hind-limbs released the T-bar. The strength of aimed to investigate complement in a recently developed the grip was measured in newtons. Each mouse was given Q331K TDP-43 mouse model of ALS [13]. 10 attempts, and the maximum strength was recorded [9]. We examined the expression of major complement factors and of C5a and its receptor C5aR1, within the Tissue preparation for microglia/astrocyte quantification lumbar spinal cord and tibialis anterior (TA) leg muscle and immunohistochemistry WT Q331K at three different ages during disease progression in Female NTg, TDP-43 and TDP-43 mice (n = 4 per Q331K TDP-43 mice, in order to provide a comprehensive genotype) were euthanized by intraperitoneal injection of overview of the potential involvement of complement in zolazapam (50 mg/kg; Zoletil, Lyppard) and xylazine an alternative mouse model of ALS. Our findings dem- (10 mg/kg; Xylazil, Lyppard). Mice were then fixed by onstrate that a global dysregulation of complement transcardiac perfusion with 2% sodium nitrite in 0.1 M system is involved in this TDP-43 familial mouse model phosphate buffer (pH 7.4; Sigma-Aldrich, St Louis, MO, of ALS, suggesting that complement/C5aR1 could be a USA) followed by 4% paraformaldehyde in 0.1 M phos- potential therapeutic target in most forms of ALS. phate buffer (4% PFA-PB; pH 7.4; Sigma-Aldrich, St Louis, MO, USA) at the previously mentioned ages. Lumbar Methods spinal cords were collected and placed into 4% PFA-PB Animals for 2 h at 4 °C. Following this incubation, the spinal cords WT Q331K Transgenic TDP-43 (Line 96) and TDP-43 (Line were washed 3 × 5 min in phosphate-buffered saline (PBS; 103) mice were obtained from the Jackson Laboratory pH 7.4), followed by submersion in sucrose solution at (Bar Harbor, ME, USA) and were bred on a C57BL/6J back- 15% then 30% in PBS (pH 7.4). Lumbar spinal cords were WT Q331K ground to produce TDP-43 ,TDP-43 and respective then embedded in optimal cutting temperature compound WT non-transgenic (NTg) control mice. TDP-43 transgenic (Sakura, Finetek, Torrance, CA, USA) then snap frozen in mice express a myc-tagged human non-mutated version of liquid nitrogen. Lumbar spinal cords were sectioned into Q331K the TDP-43 cDNA sequence and TDP-43 mice ex- 16-μm-thick transverse and coronal sections and dry press a myc-tagged, human TDP-43 cDNA modified to mounted onto Superfrost Plus slides (Menzel-Glaser, have an ALS-linked glutamine to lysine residue mutation at Braunschweig, Germany) for estimation of astrocytes, position 331, under the direction of the mouse prion pro- microglia and immunohistochemistry as detailed below. tein promoter. The prion protein promoter ensures that the transgene expression is directed primarily to the central Estimation of astrocytes and microglia nervous system—the brain and spinal cord—and is very For estimation of astrocytes and microglia within the lum- low in other peripheral tissues [13]. Female NTg, bar spinal cord, sections were rehydrated in PBS (pH 7.4) WT Q331K TDP-43 and TDP-43 mice at 3, 10 months and then blocked in PBS containing 3% bovine serum albumin Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 3 of 13 (BSA) for 1 h at room temperature. Sections were incu- 555 goat anti-rat, Alexa 594 donkey anti-rat, Alexa 488 goat bated overnight at 4 °C with the astrocyte (mouse anti-mouse, Alexa 488 goat anti-rabbit and Alexa 488 don- anti-GFAP; 1:1000, BD Biosciences, San Diego, CA, USA) key anti-goat (Invitrogen, Eugene, OR, USA). All secondary and microglia (rat anti-CD11b; 1:500, Abcam, Cambridge, antibodies were diluted in PBS (pH 7.4) containing 1% BSA MA, USA) markers. Sections were washed with PBS for or 1% DS (1:1,1000 for Alexa 555, 1:500 for Alexa 594 and 3 × 10 min prior to incubation overnight at 4 °C with the 1:600 for Alexa 488). Following 3 × 5 min washes in PBS, Alexa secondary cocktail: Alexa Fluor 555 dye-conjugated all sections were mounted with Prolong Gold Anti-Fade goat anti-rat (1:1000, Invitrogen, Eugene, OR, USA) and medium containing DAPI (Invitrogen, Eugene, OR, USA). Alexa Fluor 488 dye-conjugated goat anti-mouse (1:600, Sections with no primary antibodies were used as negative Invitrogen, Eugene, OR, USA) antibody. All primary and controls for all immunohistochemistry experiments to give secondary antibodies were diluted in PBS (pH 7.4) con- a measure of non-specific background staining. taining 1% BSA. Sections were then washed for 3 × 5 min Quantification of immunofluorescence for C1q was in PBS, then mounted with Prolong Gold Anti-Fade performed on ~ 25 to 35 lumbar spinal cord sections medium containing 4, 6-diamidino-2-phenylindole (DAPI; (per animal; n = 4) spaced 160 μm apart and expressed Invitrogen, Eugene, OR, USA). Quantification of GFAP as the percentage immunoreactive area per section. and CD11b was performed on ~ 11 to 14 lumbar spinal Staining procedures and image exposures were all stan- cord sections spaced 320 μm apart and expressed as the dardised between genotype and between sections [15]. percentage immunoreactive area per section [12]. Quanti- fication was within the second lumbar dorsal root ganglia (L2) to the fifth lumbar dorsal root ganglia (L5), selected Real-time quantitative PCR with the aid of the mouse spinal cord atlas [17]. Staining Total RNA was isolated from the spinal cord and TA WT Q331K procedures and image exposures were all standardised be- muscle of NTg, TDP-43 and TDP-43 mice using tween genotypes and between sections. The mouse geno- RNeasy Lipid Tissue extraction kit according to manu- type was not made available to the researchers until the facturer’s instructions (QIAGEN, CA, USA). Total RNA completion of the study. was purified from genomic DNA contamination using Turbo DNAse treatment (Ambion, NY, USA), then Quantification of activated microglia numbers converted to cDNA using AffinityScript cDNA synthesis The cell body of microglia was labelled with the nuclear kit according to manufacturer’s instructions (Agilent marker, DAPI. As microglia are known to display morpho- Technologies, CA, USA). Commercially available logical changes when they become activated, such as an gene-specific Taqman probes for complement increase in cell body size, thickening of proximal processes component 1, q subcomponent, beta polypeptide and a decrease in the ramification of distal branches [18], (C1qB; Mm01179619_m1), complement component 4 activated microglia were defined by the presence of one (C4; Mm00437893_g1), complement factor B (Cfb; DAPI stain, an amoeboid cell body and proximal pro- Mm00433909_m1), complement component 3 (C3; cesses length ≤ 1–2 μm[19]. The total number of acti- Mm01232779_m1), CD55 antigen (Cd55; Mm00438 vated microglia was determined by the average of 11–14 377_m1), CD59a antigen (Cd59a; Mm00483149_m1) sections, with the overall average multiplied by the num- and complement component 5a receptor 1 (C5ar1; ber of sections within L2–L5. The mouse genotype was Mm00500292_s1) were used to amplify target gene of not made available to the researchers until quantification interest (Applied Biosystems, MA, USA). Relative target was completed. gene expression to geometric mean of reference genes glyceraldehyde-3-phosphate dehydrogenase (Gapdh; Immunohistochemistry Mm99999915_g1), beta actin (Actb; Mm02619580_g1) Fluorescence double-labelling immunohistochemistry was and hypoxanthine guanine phosphoribosyl transferase performed to localise C1q and C5aR1 expression with spe- (Hprt; Mm03024075_m1) was determined using this -ΔCT cific cell-type markers for astrocytes, microglia and motor formula: 2 where ΔCT = (Ct − Ct (Target gene) (Gapdh, neurons as previously described [15]. Briefly, sections were ), as per our previous studies [15, 16]. Final Actb and Hprt) rehydrated in PBS (pH 7.4) for 10 min, then blocked in PBS measures are presented as relative levels of gene ex- WT Q331K containing 3% BSA or 3% donkey serum (DS) for 1 h at pression in TDP-43 and TDP-43 mice com- room temperature. Sections were incubated overnight at pared with expression in NTg controls. Probe sets were 4 °C with combination of primary antibodies outlined in tested over a serial cDNA concentration for amplifica- Table 1. All primary antibodies were diluted in PBS tion efficiency. No reverse transcription and water as (pH 7.4) containing 1% BSA or 1% DS. Sections were no template control was used as negative controls. All washed 3 × 10 min with PBS prior to incubation with an samples were run in triplicate and were tested in three appropriate Alexa-conjugated secondary cocktail: Alexa separate experiments. Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 4 of 13 Table 1 Summary of antibodies used for immunohistochemistry Antibody Manufacturer Dilution In combination with one of the following antibodies Rat anti-mouse C5aR1 Bio-Rad 1:250 GFAP, Iba-1 or ChAT Rat anti-mouse C1q Hycult Biotechnology 1:1000 GFAP, Iba-1 or ChAT Goat anti-mouse ChAT Merck 1:100 C1q or C5aR1 Rabbit anti-mouse Iba-1 Wako 1:400 C1q or C5aR1 Mouse anti-rat GFAP BD Biosciences 1:1000 C1q or C5aR1 Enzyme-linked immunosorbent assay mice (Fig. 1a). However, by 10 months, we observed a Ninety-six-well plates (Greiner Bio-One, Frickenhausen, significant reduction in hind-limb grip strength in Q331K Germany) were pre-coated with rat anti-mouse C5a cap- TDP-43 mice when compared with NTg and WT ture antibody (4 μg/mL; R&D Systems, Minneapolis, TDP-43 mice (~ 30% reduction, n = 15, ****P < 0.0001, MN, USA) diluted in PBS (pH 7.4) overnight at room ++++P < 0.0001; Fig. 1a). Furthermore, at 16 months, temperature in a sealed humidified container. Following there was a progressive decline in hind-limb grip Q331K the plate being blocked for 1 h at room temperature strength in TDP-43 mice (~ 45% reduction, n = 15, with reagent diluent (1% BSA in PBS), C5a standard, ****P < 0.0001, ++++P < 0.0001; Fig. 1a) when compared WT spinal cord and TA muscle homogenates were incubated with NTg and TDP-43 mice. Importantly, we found for 2 h at room temperature. The plates were subse- that the decline in hind-limb grip strength in Q331K quently incubated with biotinylated goat anti-mouse C5a TDP-43 mice closely correlated with the increase detection antibody (0.2 μg/mL, R&D Systems, Minneap- in immunoreactive area of astrocytes using immuno- olis, MN, USA) for 1 h at room temperature, and then fluorescence staining in the lumbar spinal cords at incubated with Streptavidin-HRP conjugate for 20 min 10 months (~ 200% increase, n =4, **P < 0.01; Fig. 1b, c) at room temperature. Tetramethylbenzidine (Sigma-Al- and 16 months (~ 380% increase, n = 4, ****P < 0.0001; drich, Saint Louis, MO, USA) substrate was used as the Fig. 1b, c). Similarly, we also found that the decline Q331K chromogen, and the plate was read at 450 nm. Levels of in hind-limb grip strength in TDP-43 mice C5a in spinal cord and TA muscle samples were ad- closely correlated with the increase in immunoreactive justed to micrograms per protein and expressed as nano- area of microglia and number of activated microglia grams of C5a per microgram of protein. in the lumbar spinal cord at 10 months (140~230% increase, n =4, **P < 0.01, ***P < 0.001, ****P < 0.0001; Statistical analysis Fig. 1d–f) and 16 months (130~280% increase, n =4, All statistical analyses were performed using GraphPad *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; Fig. 1d–f). Prism 7.0 (GraphPad Software Inc., San Diego, CA, Taken together, these data reveal an age-related decline in Q331K USA). For the results from GFAP and CD11b quantifica- hind-limb grip strength of TDP-43 mice associated tion, quantitative real-time PCR and enzyme-linked im- with increased glial activation in the lumbar spinal cord. munosorbent assay, statistical differences between NTg, WT Q331K TDP-43 and TDP-43 mice were determined Components of the classical/lectin pathways of using one-way ANOVA with Tukey’s post hoc test for complement are upregulated along with decreased each age group. To assess the linear association between expression levels of complement regulator CD55 in TDP- Q331K complement mRNA transcript levels and hind-limb grip 43 mice WT Q331K strength of NTg, TDP-43 and TDP-43 mice, The complement system is part of the innate immune Pearson’s correlation was used. All data was presented as system that can contribute to neuroinflammation in mean ± standard error of mean and differences consid- many neurodegenerative diseases, including ALS [8]. ered significant when P < 0.05. Previous studies, including our own, have identified major complement components are upregulated in the G93A Results lumbar spinal cord of hSOD1 mice [15]. However, Q331K Motor deficits in TDP-43 mice correlate with there is no comprehensive overview of complement sys- increases in astrocytes and microglia during disease tem in different animal models of ALS other than progression hSOD1 transgenic mice [8]. Therefore, we measured the To monitor the decline in motor performance during mRNA levels of key components of the classical/lectin Q331K disease onset and progression in TDP-43 mice, we pathway (C1qB and C4), alternative pathway (fB), the performed hind-limb grip strength tests in animals. At central component to all pathways (C3) and the comple- 3 months, there was no difference in hind-limb grip ment regulators (CD55 and CD59a) in the lumbar spinal WT Q331K WT Q331K strength between NTg, TDP-43 and TDP-43 cord of NTg, TDP-43 and TDP-43 mice using Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 5 of 13 Fig. 1 Decline in hind-limb grip strength during ALS progression correlates with increase in astrocytes and microglia in the lumbar spinal cord of Q331K WT Q331K TDP-43 mice. a Hind-limb grip strength (N) from NTg, TDP-43 and TDP-43 mice at 3, 10 and 16 months of age. At 3 months, no Q331K significant difference in force is present between genotypes. By 10 months, TDP-43 mice display a significant decrease in hind-limb grip WT Q331K WT Q331K strength compared with both NTg and TDP-43 mice (~ 30% reduction; NTg vs TDP-43 ****P < 0.0001 and TDP-43 vs TDP-43 +++ Q331K +P < 0.0001; n = 15). A progressive decline in hind-limb grip strength is present at 16 months, with TDP43 mice showing significantly lower WT Q331K WT hind-limb grip strength compared with NTg and TDP-43 mice (~ 45% reduction; NTg vs TDP-43 ****P < 0.0001 and TDP-43 vs TDP- Q331K WT 43 ++++P < 0.0001; n = 15). b Representative images of the GFAP-positive astrocytes in the lumbar spinal cord of NTg, TDP-43 and TDP- Q331K 43 animals at 16 months. Dashed line shows the outline of the ventral horn with higher magnification of the white square. Scale bars = Q331K WT 100 μm. c Increased astrocyte expression in TDP-43 mice (blue bars) compared with NTg (orange bars) and TDP-43 mice (green bars) at ages 10 and 16 months (**P < 0.01, ****P < 0.0001; n = 4). d Representative images of CD11b-positive microglia in the lumbar spinal cord of NTg, WT Q331K TDP-43 and TDP-43 animals at 16 months. Dashed line shows the outline of the ventral horn with higher magnification of the white Q331K WT square. Scale bar = 100 μm. e Increased microglia expression in TDP-43 mice (blue bars) compared with NTg (orange bars) and TDP-43 Q331K mice (green bars) at ages 10 and 16 months (*P < 0.05, **P < 0.01; n = 4). f Increased number of activated microglia (amoeboid) in TDP-43 WT mice (blue bars) compared with NTg (orange bars) and TDP-43 (green bars) mice at 10 and 16 months (***P < 0.001, ****P < 0.0001; n = 4). Data are presented as mean ± SEM; one-way ANOVA with Tukey’s post hoc test for each age quantitative real-time PCR during disease progression of bar compared to orange and green bars respectively in ALS (3, 10 and 16 months). Fig. 2b; n =5, ***P <0.001, ****P < 0.0001). By contrast, fB Quantitative real-time PCR analyses showed significant did not show any significant changes at 10 and 16 months Q331K Q331K increases of the C1qB transcript in TDP-43 mice by of age in TDP-43 mice when compared to NTg and WT 1.3-fold at 10 months of age when compared to NTg and TDP-43 mice (blue bar compared to orange and green WT TDP-43 mice; at 16 months of age, the increase was bars respectively in Fig. 2c; n =5, P > 0.05). The central 1.5-fold and 1.4-fold when compared to NTg and component of complement, C3, was also increased in the WT Q331K TDP-43 mice, respectively (blue bar compared to lumbar spinal cord of TDP-43 mice, with a 1.6-fold orange and green bars respectively in Fig. 2a; n =5, and 1.4-fold increase at 10 months of age when compared WT ***P < 0.001, ****P < 0.0001). C4 transcript was also in- to NTg and TDP-43 mice, respectively; at 16 months creased by 2.4-fold and 3.3-fold at 10 months of age of age, the increase was 1.7-fold when compared to both and 2.0-fold and 2.7-fold at 16 months of age when genotypes (blue bar compared to orange and green bars WT compared to NTg and TDP-43 mice, respectively (blue respectively in Fig. 2d; n =5, ***P < 0.001, ****P < 0.0001). Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 6 of 13 Q331K Fig. 2 Dysregulation of complement components in the lumbar spinal cord of TDP-43 mice at three different ages of disease progression. a–f The mRNA expression profiles of the following complement components: C1qB (a, classical pathway), C4 (b, classical/lectin pathway), fB Q331K (c, alternative pathway), C3 (d, central component), CD55 (e, regulator) and CD59a (f, regulator) in the lumbar spinal cord of TDP-43 mice WT (blue bars) relative to non-transgenic (NTg, orange bars) and TDP-43 (green bars) mice during 3, 10 and 16 months of age. Data are expressed as means ± SEM (n = 5 mice/group, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, one-way ANOVA with Tukey’s post hoc test) WT The negative regulators of the complement system, NTg and TDP-43 mice (blue bar compared to orange CD55 and CD59a, were also investigated due to their im- and green bars in Fig. 3a; n =4, *P < 0.05). We also ob- Q331K portance in maintaining homeostasis and keeping the served that the marked increase of C1q in TDP-43 complement system in its proper physiological state in re- mice was localised to motor neurons and microglia sponse to altered physiological conditions (i.e. infection (white arrows in Fig. 3d, g), compared with NTg and WT and/or neurodegeneration). CD55 mRNA expression in TDP-43 mice where little to no C1q was observed Q331K TDP-43 mice was decreased at 10 months of age by (Fig. 3b, c, e, f). We did not observe C1q on astrocytes WT Q331K 0.3-fold and 0.4-fold, and by 0.4-fold and 0.5-fold at in either NTg, TDP-43 or TDP-43 mice 16 months of age, when compared with NTg and (Fig. 3h–j). WT TDP-43 mice (blue bar compared to orange and green bars respectively in Fig. 2e; n =5, *P <0.05, **P <0.01, The terminal complement pathway C5a receptor, C5aR1, ****P < 0.0001). By contrast, CD59a mRNA expression did is upregulated and expressed on motor neurons and Q331K not significantly alter at 10 and 16 months of age in microglia in the lumbar spinal cord of TDP-43 mice Q331K TDP-43 mice when compared to NTg and Previous studies have shown increases in C5aR1 expres- WT TDP-43 mice (blue bar compared to orange and green sion in the central nervous system of multiple rodent bars respectively in Fig. 2f; n =5, P > 0.05). These results models of ALS, with many studies suggesting a pathogenic suggest widespread complement perturbation occurs in role for C5aR1 in the disease progression of ALS in Q331K G93A the lumbar spinal cord of TDP-43 mice, which may hSOD1 mice [9, 10, 12, 20]. C5a, the ligand for contribute to glial activation and neuroinflammation, and C5aR1, is an activation fragment of the terminal comple- ultimately disease progression in this model. ment cascade that is rapidly generated following comple- Upregulation of C1q at 16 months of age was also ment cascade initiation [21]. We therefore examined the Q331K confirmed using immunofluorescence, where there was protein levels of C5a in the spinal cord of TDP-43 Q331K marked increase in TDP-43 mice compared with mice using enzyme-linked immunosorbent assay as a Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 7 of 13 WT Q331K Fig. 3 Expression and localisation of C1q in the spinal cord of non-transgenic, TDP-43 and TDP-43 mice at 16 months of age. a The WT Q331K immunoreactive area of C1q in the spinal cord of non-transgenic (NTg, orange bars), TDP-43 (green bars) and TDP-43 (blue bars) at 16 months. Data are expressed as means ± SEM (n = 4 mice/group, *P < 0.05, one-way ANOVA with Tukey’s post hoc test). b–j Double immunolabelling of C1q (red) with cellular markers (green) for motor neurons (ChAT; b–d), microglia (Iba-1; e–g) and astrocyte (GFAP; h–j) in the WT Q331K ventral lumbar spinal cord of NTg, TDP-43 and TDP-43 mice at 16 months of age. C1q was co-localised with ChAT-positive motor neurons Q331K (d, white arrow) and microglia (g, white arrow) in TDP-43 mice. Note that white arrows indicate red and green fluorescent signal merge to Q331K orange. In TDP-43 mice, immunolabelling of C1q was also evident on other cell types, indicated by lack of co-localisation with anti-ChAT, anti-Iba-1 and anti-GFAP (d, g, j). Scale bars for all panels = 20 μm biomarker for terminal complement activation. Interest- whereas it was not observed in GFAP-positive astrocytes ingly, the results showed no change in C5a protein levels (Fig. 4i–k). Together, the results above indicate that Q331K in TDP-43 mice at 16 months of age when compared C5a-C5aR1 signalling could play a role in facilitating WT to NTg and TDP-43 mice (blue bar compared to or- microglia activation and phagocytosis ultimately leading ange and green bars in Fig. 4a; n =5, P > 0.05). In contrast to motor neuron death in these animals. to protein levels of C5a, C5aR1 mRNA expression was sig- nificantly increased by 1.4-fold and 1.2-fold at 10 months of age and by 1.8-fold and 1.6-fold at 16 months of age, Dysregulation of complement in the tibialis anterior WT Q331K when compared to NTg and TDP-43 mice (blue bar muscle of TDP-43 mice compared to orange and green bars in Fig. 4b; n =5, In addition to showing altered levels of complement Q331K *P < 0.05, ***P < 0.001, ****P < 0.0001). components in the spinal cord of TDP-43 mice, we We next aimed to determine the cellular localisation also investigated the level of major complement compo- Q331K of C5aR1 that could explain the increased expression nents in the TA muscle of TDP-43 mice, as it has Q331K G93A seen in TDP-43 mice. To achieve this, we per- been shown previously in hSOD1 mice that comple- formed immunolabelling for C5aR1 on the lumbar spinal ment is upregulated in this muscle. To investigate this, WT Q331K cord from NTg, TDP-43 and TDP-43 mice. we measured the mRNA levels of key components of the These sections were immunostained for C5aR1 with spe- complement pathways, which include the classical/lectin cific cellular markers to identify motor neurons (anti-C- (C1qB and C4), alternative (fB) and terminal pathways hAT), astrocytes (anti-GFAP) and microglia (anti-Iba-1). (C5a and C5aR1), as well as the major complement reg- C5aR1 localised to ChAT-positive motor neurons and ulators (CD55 and CD59a) using quantitative real-time WT Iba-1-positive microglia in NTg, TDP-43 and PCR and enzyme-linked immunosorbent assay for C5a Q331K Q331K TDP-43 mice at 16 months of age (white arrows in in TA muscle of TDP-43 mice during disease pro- Fig. 4c–e for motor neurons and Fig. 4f–h for microglia), gression of ALS (3, 10 and 16 months). Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 8 of 13 Q331K Fig. 4 Expression of C5a and C5aR1 in the spinal cord of TDP-43 mice at three different ages of disease progression. a The protein WT Q331K expression of C5a in the spinal cord of non-transgenic (NTg, orange bars), TDP-43 (green bars) and TDP-43 (blue bars) at 16 months. b Q331K WT mRNA expression of C5aR1 in the spinal cord of TDP-43 mice relative to age-matched NTg and TDP-43 mice at 3, 10 and 16 months of age. Data are expressed as means ± SEM (n = 5 mice/group, *P < 0.05, ***P < 0.001, ****P < 0.0001, one-way ANOVA with Tukey’s post hoc test for each age). c–k Double immunolabelling of C5aR1 (red) with cellular markers (green) for motor neurons (ChAT; c–e), microglia (Iba-1; f–h) and WT Q331K astrocyte (GFAP; i–k) in the ventral lumbar spinal cord of NTg, TDP-43 and TDP-43 mice at 16 months of age. C5aR1 was co-localised with WT Q331K ChAT-positive motor neurons (c–e, white arrows) and microglia (f–h, white arrows) in NTg, TDP-43 and TDP-43 mice. Note that white arrows indicate red and green fluorescent signal merge to orange. Scale bars for all panels = 20 μm WT C1qB and C4 transcripts were significantly increased 10 months of age when compared to NTg and TDP-43 by 1.6-fold and 1.9-fold when compared to NTg mice mice, whereas 1.5-fold increase at 16 months of age when WT WT and by 1.5-fold when compared to TDP-43 mice at compared to TDP-43 mice (blue bar compared to or- 10 months of age, respectively (blue bar compared to or- ange and green bars in Fig. 5c; n =5, *P <0.05, **P <0.01). ange and green bars in Fig. 5a, b; n =5, *P < 0.05, By contrast, C3 was surprisingly decreased by 0.3-fold in Q331K **P < 0.01, ***P < 0.001). C1qB and C4 transcripts were TA muscle of TDP-43 mice when compared to NTg WT also increased by 1.7-fold and 1.5-fold when compared and TDP-43 mice at 10 months of age (blue bar com- to NTg mice and by 1.7-fold and 1.9-fold when com- pared to orange and green bars in Fig. 5d; n =5, *P <0.05, WT pared to TDP-43 mice at 16 months of age, respect- **P < 0.01). The regulator, CD55, was increased by 1.3-fold Q331K ively (blue bar compared to orange and green bars in at 10 months of age in TDP-43 mice when compared Fig. 5a, b; n =5, *P < 0.05, **P < 0.01, ***P < 0.001). In to NTg mice (blue bar compared to orange bar in Fig. 5e; addition to C1qB and C4, fB also displayed a marked in- n =5, *P < 0.05), while there was no significant change in Q331K crease in mRNA levels by 1.4-fold and 1.3-fold at CD59a in TDP-43 mice when compared to NTg and Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 9 of 13 Q331K Fig. 5 Dysregulation of complement components in tibialis anterior muscle of TDP-43 mice at three different ages of disease progression. a– f The mRNA expression profiles of the following complement components: C1qB (a, classical pathway), C4 (b, classical/lectin pathway), fB (c, Q331K alternative pathway), C3 (d, central component), CD55 (e, regulator) and CD59a (f, regulator) in tibialis anterior muscle of TDP-43 mice (blue WT bars) relative to non-transgenic (NTg, orange bars) and TDP-43 (green bars) mice during 3, 10 and 16 months of age. g The protein expression WT Q331K of C5a in the tibialis anterior muscle of NTg (orange bars), TDP-43 (green bars) and TDP-43 (blue bars) mice at 16 months. h mRNA Q331K WT expression of C5aR1 in the tibialis anterior muscle of TDP-43 mice relative to age-matched NTg and TDP-43 mice at three different ages. Data are expressed as means ± SEM (n = 5 mice/group, *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with Tukey’s post hoc test for each age) WT TDP-43 controls (Fig. 5f; n =5, P > 0.05). Lastly, the ter- with their hind-limb grip strength. We found a strong minal complement pathway component, C5a, and its re- negative correlation between lumbar spinal cord mRNA ceptor C5aR1 were investigated. The results showed expression of C1qB (r = − 0.6824), C3 (r = − 0.8282) and significant increases in C5a protein at 16 months of age by C5aR1 (r = − 0.7428) to hind-limb grip strength of these 1.8-fold and 2.0-fold when compared to NTg and animals, with C3 presenting the strongest correlation WT TDP-43 mice respectively (blue bar compared to or- (n = 30, ****P < 0.0001; Fig. 6a–c). Furthermore, a moder- ange and green bars in Fig. 5g; n =5, ***P <0.001). C5aR1 ate to strong negative correlation between TA mRNA mRNA expression was also significantly increased by levels of C1qB (r = − 0.6624) and C5aR1 (r = − 0.4116) to 1.7-fold and 1.6-fold at 10 months of age and by 1.7-fold hind-limb grip strength was observed, whereas a moderate and 2.0-fold at 16 months of age when compared to NTg positive correlation was identified between C3 mRNA WT and TDP-43 controls, respectively (blue bar compared levels (r = 0.4149) to hind-limb grip strength (n =25, to orange and green bars in Fig. 5h; n =5, *P <0.05, *P < 0.05, ***P < 0.001; Fig. 6d–f). These results indicate **P <0.01, ***P < 0.001). Taken together, these results sug- that changes in complement transcript levels in the spinal gest that dysregulation of the complement system also oc- cord and TA muscle directly correlates with the decrease Q331K curs in the TA muscle of TDP-43 mice, which could in hind-limb grip strength (i.e. increased in ALS symp- contribute to the disease pathology in these animals. toms) in this ALS model. Q331K Motor dysfunction in TDP-43 mice strongly Discussion correlates with lumbar spinal cord levels of C1qb, C3 and The major findings of the current study are that compo- C5aR1 nents of the complement system are dysregulated in a We next examined if there was any correlation between transgenic mutant TDP43-based mouse model of ALS. major complement transcript levels to the hind-limb It has now been well documented that the complement Q331K grip strength of TDP-43 mice. To investigate this cascade is synthesised by neurons, astrocytes and micro- aspect, we performed a Pearson correlation to measure glia within the central nervous system and is involved in the strength of the linear relationship between C1qB, C3 the disease progression of ALS, with evidence from both and C5aR1 mRNA levels in the lumbar spinal cord and human patients and rodent models [8]. The present WT Q331K TA muscle of NTg, TDP-43 and TDP-43 mice study further adds to this knowledge, demonstrating Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 10 of 13 Q331K Fig. 6 Motor dysfunction in TDP-43 mice strongly correlates with lumbar spinal cord/tibialis anterior muscle levels of C1qB, C3 and C5aR1. A strong negative correlation between lumbar spinal cord mRNA expression of C1qB (a, r = − 0.6824), C3 (b, r = − 0.8282) and C5aR1 (c, r = − 0.7428) WT Q331K with hind-limb grip strength of non-transgenic (NTg), TDP-43 and TDP-43 mice, where C3 demonstrated the strongest correlation (n = 30, ****P < 0.0001, Pearson correlation). A moderate to strong negative correlation between tibialis anterior muscle mRNA levels of C1qB (d, r = − WT Q331K 0.6624) and C5aR1 (f, r = − 0.4116) with hind-limb grip strength of NTg, TDP-43 and TDP-43 mice, whereas a moderate positive correlation between C3 mRNA levels (e, r = 0.4149) was observed (n = 25, *P < 0.05, ***P < 0.001, Pearson correlation). Pearson correlation coefficient was used to measure the strength of the linear relationship between complement mRNA expression and hind-limb grip strength Q331K alteration of mRNA expression in major complement fB in the TA muscles of TDP-43 mice, however, factors including C1qB, C4, fB, C3, C5a, C5aR1 and founddownregulation of C3inthe TA musclesof Q331K regulator CD55 in the spinal cord and TA muscles of TDP-43 mice, which is contradictory of what is Q331K G93A TDP-43 mice, which is suggestive of a progressive showninhSOD1 mice. The upregulation of C1qB dysregulation of complement in this model. These results in the TA muscles could assist in the removal of degen- are similar to our previous demonstration of complement erating neuromuscular synapses via phagocytosis dur- Q331K dysregulation in the spinal cord and TA muscle of ing disease progression in TDP-43 mice, similar to G93A G93A hSOD1 mice [11, 15], indicating that complement ac- what has been shown in hSOD1 mice [22]. The tivation occurs in response to motor neuron death and reason for the differential expression of C3 in the TA Q331K muscle denervation regardless of which ALS-related gene muscle of TDP-43 mice compared with G93A mutation is present. Furthermore, complement activation, hSOD1 mice is unclear, but could be due to the differ- Q331K and subsequent C5aR1 activation, could be a common ence in the severity of the disease with TDP-43 mice mechanism of pathology in most forms of ALS. showing milder ALS-like symptoms compared to G93A Thepresent studyprovidedevidencefor thedysregula- hSOD1 mice [13, 14]. tion of classical/lectin, alternative and central component We additionally showed decreased mRNA expression to all pathways of the complement system in the spinal levels of CD55 in the spinal cord and increased expression Q331K cord and TA muscle of TDP-43 mice during ALS levels in TA muscle, which suggests that homeostatic bal- diseaseprogression.Thisisinlinewithnumerousstud- ance of the complement system may be perturbed in Q331K ies including our own, where increased C1qB, C4 and TDP-43 mice, leading to the over activation of com- G93A C3 was found in the spinal cord of hSOD1 mice plement system. These findings support other studies, [11, 15, 22, 23]. This upregulation of C1qB and C3 in which have also shown decreased CD55 mRNA levels in the spinal cord could assist in the removal of dying the spinal cord, with deficiency in CD55 exacerbating neu- motor neurons via opsonisation through microglia acti- rodegeneration [15, 22, 24]. Our findings also support pre- Q331K vation during disease progression in TDP-43 mice, vious studies identifying increased CD55 expression in G93A similar to what is evident in hSOD1 mice [15]. intercostal muscles of ALS patients and TA muscles of G93A Similarly, we identified upregulation of C1qB, C4 and hSOD1 mice [11, 25]. Upregulation of CD55 at the Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 11 of 13 G93A Q331K motor end-plates of ALS patients and hSOD1 mice expression of C5aR1 in TDP-43 spinal cord and could be a mechanism to protect against high levels of TA muscles is elevated, confirming dysregulation of complement activation at the neuromuscular junction. downstream terminal complement pathway in these ani- Interestingly, the current study did not demonstrate any mals. This suggests that enhanced C5a-C5aR1 signalling changes in the other regulator CD59a in the spinal cord may affect the disease progression of ALS in both spinal Q331K Q331K or TA muscle of TDP-43 mice. This is in contrast to cord and TA muscles of TDP-43 mice, similar to G93A previous studies, where CD59a was decreased in the what is evident in hSOD1 mice [9, 11, 15]. In G93A Q331K spinal cord of hSOD1 mice, while it increased in the addition to an increase in C5aR1 in TDP-43 mice, intercostal muscle of ALS patients and TA muscle of the present study also revealed C5aR1 on motor neurons G93A WT Q331K hSOD1 mice [11, 15, 25]. The difference in the ex- and microglia in NTg, TDP-43 and TDP-43 pression changes between different models and patients mice. This is in concordance with previous studies could be attributed to the severity of the disease, as showing C5aR1 on motor neurons and microglia in Q331K TDP-43 mice show milder ALS-like symptoms com- other mouse models of ALS [15, 20], indicating that G93A pared to hSOD1 mice [9, 13–15]. However, further C5a-C5aR1 signalling could play a role in facilitating investigation is required into the expression and localisa- microglia activation and motor neuron death regardless Q331K tion of membrane attack complex in TDP-43 mice, of which ALS-related gene mutation is present. and its putative correlation with motor neuron loss. Re- In line with increased complement components in Q331K gardless, collectively our data adds further support to the TDP-43 mice, the present study also revealed a notion that dysregulated complement activity may play an strong correlation between C1qB, C3 and C5aR1 important role in accelerating motor neuron loss and mRNA levels in the spinal cord and TA muscles to the neuromuscular junction denervation, ultimately driving hind-limb grip strength of these animals, suggesting the progression of ALS. that the mRNA levels of complement system could be a Among the complement activation effector molecules, direct correlate of motor neuron loss and neuromuscu- C5a is considered the most potent peptide, with its signal- lar junction denervation throughout disease progres- ling through its main receptor C5aR1 having detrimental sion, that we and others have previously demonstrated effects in multiple neurodegenerative diseases, including in this mouse model [13, 14]. These finding are con- ALS [9, 10, 12, 26, 27]. To obtain a better understanding of sistent with increased astrocyte and microglia numbers/ its role in disease progression of ALS, we analysed C5a and activation in these animals, indicating that neuroin- its receptor C5aR1 in the spinal cord and TA muscle of flammation could be a good indicator of disease sever- Q331K TDP-43 mice. C5a protein levels in the spinal cord ity and pathology. WT Q331K did not change between NTg, TDP-43 and TDP-43 mice, whereas it increased in the TA muscles of Q331K Conclusions TDP-43 mice. This is in line with C5a protein levels at In summary, the current study has demonstrated upreg- mid-symptomatic stage in the spinal cord and TA muscle G93A ulation of major complement factors, together with de- of hSOD1 mice [11], suggesting that enhanced creased levels of the negative complement regulator C5a-C5aR1 signalling may affect disease progression of Q331K G93A CD55, in TDP-43 mice. This suggests that comple- ALS, similar to what has been found in hSOD1 mice. ment activation and/or its dysregulation could play an This finding is interesting given that C3 mRNA transcript important role in motor neuron loss and neuromuscular levels were higher in the spinal cord compared to TA junction denervation in this TDP43-based mouse model muscle in these animals. The reason for the increase in of ALS. Expression of the C5a receptor, C5aR1, was also muscle C5a, despite reduced C3 transcript expression is un- Q331K upregulated in TDP-43 mice, predominantly due to clear; however, it should be noted that mRNA expression increased microglial/macrophage C5aR1 expression, and does not directly equate with the degree of complement was strongly correlated with resulting motor decline. activation. Furthermore, there is a well-described Taken together, these results indicate that heightened C3-bypass (extrinsic) complement activation pathway that complement activation and enhanced C5aR1 signalling enables C3-independent cleavage of C5 [28]. Thus, C5a may play a crucial role in pathophysiology of the protein levels do not necessarily correlate with C3 mRNA Q331K TDP-43 ALS model, further validating C5aR1 as a expression within a tissue. potential therapeutic target for all forms of ALS. Numerous studies have demonstrated upregulation of C5aR1 within the spinal cord and TA muscles of G93A hSOD1 rats and mice, as well as human ALS pa- Abbreviations tients, suggesting that heightened C5a-C5aR1 signalling ALS: Amyotrophic lateral sclerosis; BSA: Bovine serum albumin; plays a role in ALS pathology [9–12, 15, 29]. In the DAPI: Diamidino-2-phenylindole; DS: Donkey serum; NTg: Non-transgenic; present study, we demonstrated that the mRNA TA: Tibialis anterior Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 12 of 13 Acknowledgements 6. Taylor JP, Brown RH Jr, Cleveland DW. Decoding ALS: from genes to The authors would like to sincerely thank Associate Professor Massimo A. mechanism. Nature. 2016;539:197–206. Hilliard for carefully going through the manuscript and giving us feedback to 7. Woodruff TM, Costantini KJ, Taylor SM, Noakes PG. Role of complement in improve our manuscript. We also thank Maryam Shayegh for her technical motor neuron disease: animal models and therapeutic potential of support with genotyping mice, Mary White for general laboratory support complement inhibitors. Adv Exp Med Biol. 2008;632:143–58. and Dr. Shaun Walters for his support with microscopy. 8. Brennan FH, Lee JD, Ruitenberg MJ, Woodruff TM. Therapeutic targeting of complement to modify disease course and improve outcomes in Funding neurological conditions. Semin Immunol. 2016;28:292–308. JDL holds a Motor Neuron Disease Research Institute of Australia (MNDRIA) 9. Lee JD, Kumar V, Fung JN, Ruitenberg MJ, Noakes PG, Woodruff TM. Postdoctoral Fellowship. TMW is supported by a NHMRC Career Pharmacological inhibition of complement C5a-C5a1 receptor signalling Development Fellowship. The research was funded by grants from the ameliorates disease pathology in the hSOD1(G93A) mouse model of MNDRIA to PGN (GIA1429 and GIA1547), TMW (GIA1865) and JDL (GIA1830) amyotrophic lateral sclerosis. Br J Pharmacol. 2017;174:689–99. and a National Health and Medical Research Council (NHMRC) Project grant 10. Woodruff TM, Lee JD, Noakes PG. Role for terminal complement activation (APP1082271) to TMW. in amyotrophic lateral sclerosis disease progression. Proc Natl Acad Sci U S A. 2014;111:E3–4. 11. Wang HA, Lee JD, Lee KM, Woodruff TM, Noakes PG. Complement C5a-C5aR1 Authors’ contributions signalling drives skeletal muscle macrophage recruitment in the hSOD1(G93A) JDL, TMW and PGN conceived the project. JDL, TMW and PGN designed the mouse model of amyotrophic lateral sclerosis. Skelet Muscle. 2017;7:10. study. JDL, SCL, EFW and RL performed the experiments in the laboratories 12. Woodruff TM, Costantini KJ, Crane JW, Atkin JD, Monk PN, Taylor SM, of PGN and TMW. All authors contributed to the analyses and/or interpreted Noakes PG. The complement factor C5a contributes to pathology in a rat the data. JDL wrote the paper with contributions from TMW and PGN. All model of amyotrophic lateral sclerosis. J Immunol. 2008;181:8727–34. authors read and approved the final manuscript. 13. Arnold ES, Ling SC, Huelga SC, Lagier-Tourenne C, Polymenidou M, Ditsworth D, Kordasiewicz HB, McAlonis-Downes M, Platoshyn O, Parone PA, Ethics approval et al. ALS-linked TDP-43 mutations produce aberrant RNA splicing and All experimental procedures were approved by the University of Queensland adult-onset motor neuron disease without aggregation or loss of nuclear Animal Ethics Committee and complied with the policies and regulations TDP-43. Proc Natl Acad Sci U S A. 2013;110:E736–45. regarding animal experimentation. They were conducted in accordance with 14. Chand KK, Lee KM, Lee JD, Qiu H, Willis EF, Lavidis NA, Hilliard MA, Noakes the Queensland Government Animal Research Act 2001, associated Animal PG. Defects in synaptic transmission at the neuromuscular junction Care and Protection Regulations (2002 and 2008) and the Australian Code of precedes motor deficits in a TDP-43(Q331K) transgenic mouse model of Practice for the Care and Use of Animals for Scientific Purposes, 8th Edition amyotrophic lateral sclerosis. FASEB J. 2018;32:2676–89. (National Health and Medical Research Council, 2013). ARRIVE guidelines 15. Lee JD, Kamaruzaman NA, Fung JN, Taylor SM, Turner BJ, Atkin JD, Woodruff have been followed in the preparation of the manuscript. TM, Noakes PG. Dysregulation of the complement cascade in the hSOD1G93A transgenic mouse model of amyotrophic lateral sclerosis. J Competing interests Neuroinflammation. 2013;10:119. The authors declare that they have no competing interests. 16. Lee JY, Lee JD, Phipps S, Noakes PG, Woodruff TM. Absence of toll-like receptor 4 (TLR4) extends survival in the hSOD1 G93A mouse model of amyotrophic lateral sclerosis. J Neuroinflammation. 2015;12:90. Publisher’sNote 17. Watson C, Paxinos G, Kayalioglu G. The spinal cord. 1st ed. London: Springer Nature remains neutral with regard to jurisdictional claims in published Academic Press; 2009. maps and institutional affiliations. 18. Li Q, Barres BA. Microglia and macrophages in brain homeostasis and disease. Nat Rev Immunol. 2017;18:225–42. Author details School of Biomedical Sciences, the University of Queensland, St Lucia, 19. Philips T, Robberecht W. Neuroinflammation in amyotrophic lateral Brisbane, QLD 4072, Australia. University of Queensland Centre for Clinical sclerosis: role of glial activation in motor neuron disease. Lancet Neurol. Research, the University of Queensland, Herston, Brisbane, QLD 4029, 2011;10:253–63. Australia. Queensland Brain Institute, the University of Queensland, St Lucia, 20. Humayun S, Gohar M, Volkening K, Moisse K, Leystra-Lantz C, Mepham Brisbane, QLD 4072, Australia. J, McLean J, Strong MJ. The complement factor C5a receptor is upregulated in NFL−/− mouse motor neurons. J Neuroimmunol. Received: 5 March 2018 Accepted: 24 May 2018 2009;210:52–62. 21. Woodruff TM, Ager RR, Tenner AJ, Noakes PG, Taylor SM. The role of the complement system and the activation fragment C5a in the central nervous system. NeuroMolecular Med. 2010;12:179–92. References 22. Heurich B, El Idrissi NB, Donev RM, Petri S, Claus P, Neal J, Morgan BP, 1. DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, Ramaglia V. Complement upregulation and activation on motor neurons Rutherford NJ, Nicholson AM, Finch NA,Flynn H, Adamson J,etal. and neuromuscular junction in the SOD1 G93A mouse model of familial Expanded GGGGCC hexanucleotide repeat in noncoding region of amyotrophic lateral sclerosis. J Neuroimmunol. 2011;235:104–9. C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron. 23. Lobsiger CS, Boillee S, Pozniak C, Khan AM, McAlonis-Downes M, Lewcock 2011;72:245–56. JW, Cleveland DW. C1q induction and global complement pathway 2. Kwiatkowski TJ Jr, Bosco DA, Leclerc AL, Tamrazian E, Vanderburg CR, Russ activation do not contribute to ALS toxicity in mutant SOD1 mice. Proc Natl C, Davis A, Gilchrist J, Kasarskis EJ, Munsat T, et al. Mutations in the FUS/TLS Acad Sci U S A. 2013;110:E4385–92. gene on chromosome 16 cause familial amyotrophic lateral sclerosis. 24. Wang Y, Li Y, Dalle Lucca SL, Simovic M, Tsokos GC, Dalle Lucca JJ. Decay Science. 2009;323:1205–8. accelerating factor (CD55) protects neuronal cells from chemical hypoxia- 3. Rosen DR, Siddique T, Patterson D, Figlewicz DA, Sapp P, Hentati A, induced injury. J Neuroinflammation. 2010;7:24. Donaldson D, Goto J, O'Regan JP, Deng H-X. Mutations in Cu/Zn superoxide 25. Bahia El Idrissi N, Bosch S, Ramaglia V, Aronica E, Baas F, Troost D. dismutase gene are associated with familial amyotrophic lateral sclerosis. Complement activation at the motor end-plates in amyotrophic lateral Nature. 1993;362:59–62. sclerosis. J Neuroinflammation. 2016;13:72. 4. Sreedharan J, Blair IP, Tripathi VB, Hu X, Vance C, Rogelj B, Ackerley S, Durnall JC, Williams KL, Buratti E, et al. TDP-43 mutations in familial and 26. Fonseca MI, Ager RR, Chu SH, Yazan O, Sanderson SD, LaFerla FM, Taylor sporadic amyotrophic lateral sclerosis. Science. 2008;319:1668–72. SM, Woodruff TM, Tenner AJ. Treatment with a C5aR antagonist decreases 5. Williams KL, Warraich ST, Yang S, Solski JA, Fernando R, Rouleau GA, pathology and enhances behavioral performance in murine models of Nicholson GA, Blair IP. UBQLN2/ubiquilin 2 mutation and pathology in Alzheimer's disease. J Immunol. 2009;183:1375–83. familial amyotrophic lateral sclerosis. Neurobiol Aging. 2012;33:2527 27. Woodruff TM,Crane JW, Proctor LM,Buller KM, Shek AB,deVos K, e2523–10. Pollitt S, Williams HM, Shiels IA, Monk PN, Taylor SM. Therapeutic Lee et al. Journal of Neuroinflammation (2018) 15:171 Page 13 of 13 activity of C5a receptor antagonists in a rat model of neurodegeneration. FASEB J. 2006;20:1407–17. 28. Huber-Lang M, Younkin EM, Sarma JV, Riedemann N, McGuire SR, Lu KT, Kunkel R, Younger JG, Zetoune FS, Ward PA. Generation of C5a by phagocytic cells. Am J Pathol. 2002;161:1849–59. 29. Mantovani S, Gordon R, Macmaw JK, Pfluger CM, Henderson RD, Noakes PG, McCombe PA, Woodruff TM. Elevation of the terminal complement activation products C5a and C5b-9 in ALS patient blood. J Neuroimmunol. 2014;276:213–8.

Journal

Journal of NeuroinflammationSpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off