RESEARCH ARTICLE OPEN ACCESS Neuroinflammation-Related Proteins NOD2 and Spp1 Are Abnormally Upregulated in Amyotrophic Lateral Sclerosis Noemı́ de Luna, PhD, Álvaro Carbayo, MD, Oriol Dols-Icardo, PhD, Janina Turon-Sans, MD, Correspondence Dr. Rojas-Garcia David Reyes-Leiva, MD, Ignacio Illan-Gala, MD, PhD, Ivonne Jericó, MD, PhD, Inma Pagola-Lorz, MD, PhD, rrojas@santpau.cat Cinta Lleixà, MD, PhD, Luis Querol, MD, PhD, Sara Rubio-Guerra, MD, Daniel Alcolea, MD, PhD, Juan Fortea, MD, PhD, Alberto Lleó, MD, PhD, Elena Cortés-Vicente, MD, PhD,* and Ricardo Rojas-Garcia, MD, PhD* Neurol Neuroimmunol Neuroinflamm 2023;10:e200072. doi:10.1212/NXI.0000000000200072 Abstract Background and Objectives Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease of unknown etiology and poorly understood pathophysiology. There is no specific biomarker either for diagnosis or prognosis. The aim of our study was to investigate differentially expressed proteins in the CSF and serum from patients with ALS to determine their role in the disease process and evaluate their utility as diagnostic or prognostic biomarkers. Methods We performed mass spectrometry in the CSF from 3 patients with ALS and 3 healthy controls (HCs). The results were compared with motor cortex dysregulated transcripts obtained from 11patients with sporadic ALS and 8 HCs. Candidate proteins were tested using ELISA in the serum of 123 patients with ALS, 30 patients with Alzheimer disease (AD), 28 patients with frontotemporal dementia (FTD), and 102 HCs. Patients with ALS, AD, and FTD were prospectively recruited from January 2003 to December 2020. A group of age-matched HCs was randomly selected from the Sant Pau Initiative on Neurodegeneration cohort of the Sant Pau Memory Unit. Results Nucleotide-binding oligomerization domain–containing protein 2 (NOD2) and osteopontin (Spp1) were differentially expressed in the CSF and the motor cortex transcriptome of patients with ALS compared with that in HCs (p < 0.05). NOD2 and Spp1 levels were significantly higher in sera from patients with ALS than in HCs (p < 0.001). Receiver operating characteristic analysis showed an area under the curve of 0.63 for NOD2 and 0.81 for Spp1. NOD2 levels were significantly lower in patients with AD and FTD than in patients with ALS (p < 0.0001), but we found no significant differences in Spp1 levels between patients with ALS, AD (p = 0.51), and FTD (p = 0.42). We found a negative correlation between Spp1 levels and ALS functional rating scale (r = −0.24, p = 0.009). Discussion Our discovery-based approach identified NOD2 as a novel biomarker in ALS and adds evidence to the contribution of Spp1 in the disease process. Both proteins are involved in innate immunity and autophagy and are increased in the serum from patients with ALS. Our data support a relevant *These authors contributed equally to this work. From the Neuromuscular Diseases Laboratory (N.d.C., A.C., D.R.-L., C.L., L.Q., E.C.-V., R.R.-G.), Institut de Recerca Hospital de la Santa Creu i Sant Pau (IIB Sant-Pau), Barcelona; Centro de Investigación Biomédica en Red de Enfermedades Raras (CIBERER) (N.d.C., A.C., J.T.-S., D.R.-L., L.Q., E.C.-V., R.R.-G.), Madrid; Motor Neuron Diseases Clinic (A.C., J.T.-S., D.R.-L., L.Q., E.C.-V., R.R.-G.), Neuromuscular Diseases Unit, Department of Neurology, Hospital de la Santa Creu i Sant Pau, Departament de Medicina, Universitat Autònoma de Barcelona; Sant Pau Memory Unit (O.D.-I., I.I.-G., S.R.-G., D.A., J.F., A.L.), Department of Neurology, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona; Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED) (O.D.-I., I.I.-G., S.R.-G., D.A., J.F., A.L.), Madrid, Spain; and Neuromuscular and Motor Neuron Diseases Research Group-Health Research Institute of Navarra (IdisNA) (I.J., I.P.-L.). Go to Neurology.org/NN for full disclosures. Funding information is provided at the end of the article. The Article Processing Charge was funded by Fondo de Investigación Sanitaria (FIS), ISCIII (PI15/01618 and PI 19/01543). This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND), which permits downloading and sharing the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal. Copyright © 2022 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. 1 Glossary AD = Alzheimer disease; ALS = amyotrophic lateral sclerosis; ALSFRS-R = ALS functional rating scale; AUC = area under the curve; bvFTD = behavioral variant of frontotemporal dementia; DAMPs = danger-associated molecular patterns; ELISA = Enzyme-Linked Immunosorbent Assays; FTD = frontotemporal dementia; FUS = fused in sarcoma; HCs = healthy controls; IFN = interferon; MAVS = mitochondrial antiviral signaling; MNDs = motor neuron diseases; NLRs = nucleotide-binding domain leucine-rich repeat-containing receptors; NOD2 = nucleotide-binding oligomerization domain–containing protein 2; PAMPs = pathogen-associated molecular patterns; RIG-I = retinoic acid–inducible gene I; ROC = receiver operating characteristic; SPIN = Sant Pau Initiative on Neurodegeneration; TAR = transactivating response; TDP-43 = TAR DNA binding protein–43. role of neuroinflammation in the pathophysiology of the disease and may identify targets for disease-modifying treatments in ALS. Further longitudinal studies should investigate the diagnostic and prognostic value of NOD2 and Spp1 in clinical practice. Amyotrophic lateral sclerosis (ALS) is a rapidly progressive pathways could help to understand the pathophysiologic neurodegenerative disease characterized by selective loss of up- mechanisms involved in the initiation and progression of the per and lower motor neurons.1 Clinical presentation is highly disease. The identification of diagnostic or prognostic bio- variable, depending on the site of onset and the degree of upper markers could also be useful in clinical practice and may provide and lower motor neuron involvement, with some patients pre- important insights into the clinical and pathologic heterogeneity senting only upper or only lower motor neuron signs at onset. by tracking various aspects of pathophysiology. This variability and the absence of a diagnostic biomarker can make diagnosis difficult. Survival is also highly heterogeneous, and no reliable prognostic biomarker for ALS has been reported Methods to date.2,3 In this study, we differentiated 2 phases. In the discovery Although the pathophysiology of ALS remains unknown, it has phase, we performed a proteomic analysis of the CSF from been suggested that an altered protein homeostasis may play a patients with ALS and controls to identify dysregulated pro- role in the pathogenesis of the disease. The neuropathologic teins in patients with ALS in comparison with those in con- hallmark of ALS is the aberrant cytoplasmic aggregation of trols. In the second phase, the candidate biomarkers selected mislocalized proteins, mainly transactive response DNA bind- were tested in a validation cohort including patients with ing protein 43 (TDP-43) and fused in sarcoma (FUS), in ALS, controls, and individuals with other neurodegenerative degenerating neurons.4 The presence of stress granules and disorders. neuroinflammation are also consistent features in the CNS of Discovery Phase patients with ALS.5,6 Study Population and Sample Preparation for One explanation for the formation of cytoplasmic protein Proteomic Analysis aggregates is a defective autophagy process.7 Aberrant accumu- The CSF from 3 patients with ALS (mean age 67.7 ± 2.3 lation of misfolded proteins could work as danger-associated years; 2 females) and 3 healthy controls (HCs), age matched molecular patterns (DAMPs) and may trigger a combined and sex matched, were used to analyze differentially expressed innate/adaptive immune response with direct activation of in- proteins in the CSF in both groups. Patients with ALS were flammation pathways.8 Neurodegeneration involves neuro- selected from our Motor Neuron Diseases (MND) Clinic and inflammatory components that may accelerate ALS progression fulfilled El Escorial revised criteria for definite ALS, and and accentuate disease severity.9 Inflammasomes are multiprotein controls were randomly selected from the Sant Pau Initiative complexes that are responsible for detecting and eliminating on Neurodegeneration (SPIN) cohort (SPIN cohort: sant- pathogen-associated molecular patterns (PAMPs) and DAMPs, paumemoryunit.com/our-research/SPIN-cohort/).14 Sam- and they play critical roles in neurodegenerative diseases. ples of CSF were obtained through lumbar puncture using Inflammasomes can also be activated by abnormal protein ag- previously published methods.15 CSF samples were collected gregation,10 and the resulting overexpression of proinflammatory following international consensus recommendations. CSF cytokines can aggravate the chronic inflammatory state.11-13 samples were stored immediately at −80°C until analysis. In this study, we performed a proteomic analysis of the CSF Proteins present in 1 mL of CSF were precipitated using 2-D from patients with ALS and controls to identify potentially Clean up kit (GE Healthcare) following the manufacturer’s dysregulated proteins in ALS. The investigation of protein bio- instructions. Protein pellets were resuspended in radio- markers involved in neuroinflammatory and neurodegenerative immunoprecipitation assay buffer supplemented with 2 Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 Neurology.org/NN protease inhibitors. The protein concentration was measured showed symptoms of cognitive or behavior impairment nor using the reducing agent and detergent compatible Protein had a family history of ALS or dementia, and C9orf72 hex- Assay (BioRad). Fifty micrograms of protein was loaded into a anucleotide repeat expansion was not present in any of the 12% acrylamide gel. cases. Demographic and clinical data (sex, age at onset, disease duration, and site of onset) have been previously described.19 In-Gel Digestion and Liquid Chromatography–Mass Spectrometry Analysis Validation Phase In-gel digestion protocol and liquid chromatography–mass spectrometry (LC-MS) analysis protocol were both per- Study Population and Classification formed as previously reported.16 Patients with ALS were prospectively recruited from the MND Clinic at Hospital de la Santa Creu i Sant Pau between Protein Identification and Quantitative Differential January 2003 and December 2020. We included patients of Analysis both sexes, older than 18 years, and categorized as probable Protein identification and quantitative differential analysis laboratory-supported or definite ALS according to El Escorial protocols were adapted from previous reports.17,18 Progenesis revised criteria.2 QI for proteomics software v3.0 (Nonlinear dynamics, UK) was used for MS data analysis using default settings. The results Patients with Alzheimer disease (AD) with pathophysiologic from each of the 3 gel fractions were analyzed independently. evidence in the CSF and patients with frontotemporal de- LC-MS runs were automatically aligned to an automatically mentia (FTD) were selected from the SPIN cohort.14 All selected reference sample—with manual supervision—and patients underwent comprehensive cognitive and behavioral automatically normalized to all features. Features were con- assessments. According to this assessment, patients with ALS sidered for identification and quantification only if they were were classified to one of 2 groups: ALS with no cognitive or within the 400–1,500 m/z range, had a 15- to 80-minute re- behavioral impairment (ALS no-CBi) and ALS-FTD fulfilling tention time, and showed positive charges between 2 and 5. criteria for FTD. Patients with ALS-FTD were included in the Peak lists were generated by Progenesis and analyzed using the ALS group for the main group comparisons. In accordance Mascot search engine (v2.2, Matrix Science, UK). Protein with current diagnostic criteria, patients with FTD were identification was performed using the SwissProt-human da- classified into 3 clinical groups: those presenting a behavioral tabase (2017_01: 20,171 entries). Precursor mass tolerance variant of frontotemporal dementia (bvFTD), those with a was set to 10 ppm and fragment mass tolerance to 0.8 Da. semantic variant of primary progressive aphasia,20 and those Oxidized methionine was considered as variable amino acid with a nonfluent agrammatic primary progressive aphasia.11,21 modification and carbamidomethylation of cysteines as fixed modification. Trypsin was selected as the enzyme allowing up We performed an active screening for signs suggestive of to 2 missed cleavages. The significant threshold for peptide motor neuron involvement in patients with FTD. If present, identification was set to p < 0.05 for the probability-based they were referred to the MND clinic for further clinical and Mascot score and a minimum ions score of 20. Finally, the 3 neurophysiologic evaluation. If motor neuron involvement fractions were combined into a single Progenesis dataset. Label- was confirmed during follow-up, they were switched to the free protein abundance quantification was based on the sum of ALS group and analyzed within this group.11 the peak areas within the isotope boundaries of peptide ion peaks. Only unique nonconflicting peptides were used for A group of age-matched cognitively normal controls was ran- quantification. To allow comparison of sample runs, the domly selected from the SPIN cohort.14 We registered the abundance of each protein was normalized to all proteins. Only revised ALS functional rating scale (ALSFRS-R) during serum proteins that were quantified and identified with at least 2 sampling and calculated the ALS progression rate as originally unique peptides, a p value <0.05, and a fold change >|2| were described.11,22 considered meaningful for subsequent analyses.17,18 All patients provided written informed consent before par- RNA sequencing ticipation. The study was approved by the local ethics com- The differentially expressed proteins detected by mass spec- mittee and was conducted in compliance with the Declaration trometry were compared with the results obtained using total of Helsinki. RNA sequencing data derived from the motor cortex of 11 patients with sporadic ALS and 8 HCs. Proteins overlapping Enzyme-Linked Immunosorbent Assays with significantly differentially expressed genes were priori- Nucleotide-binding oligomerization domain–containing protein tized for validation.19 Human motor cortex (Brodmann area 2 (NOD2) levels in the CSF and serum of patients and controls 4) samples were provided by the Neurologic Tissue Bank of were detected using human NOD2 ELISA (My Biosource, San the Biobanc-Hospital Clinic-IDIBAPS. Samples from patients Diego, CA), following the manufacturer’s instructions. with ALS fulfilled El Escorial revised criteria for definite ALS during life and presented TDP43 inclusions in the motor Scavenger receptor cysteine-rich type 1 protein M130 (CD163), cortex at postmortem examination. None of the patients C-X-C motif chemokine 16 (CXCL16), and Human Neurology.org/NN Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 3 Osteopontin (Spp1) ELISAs were purchased from Biotechne Four of the 16 proteins, the nucleotide-binding oligomeriza- R&D Systems (Minneapolis, MN). All samples were analyzed in tion domain-containing protein 2 (NOD2), the scavenger duplicate. receptor cysteine-rich type 1 protein M130 (CD163), the C-X-C motif chemokine 16 (CXCL16), and the osteopontin Statistical Analysis (Spp1) showed significant differential gene expression We selected the proteins that had an analysis of variance test with changes in RNA sequencing data from the motor cortex of a p < 0.05 and a >2-fold change. Multiple comparisons were patients with sporadic ALS when compared with HCs (p < analyzed using a 1-way analysis of variance test, followed by the 0.05 and log2–fold change >0.45)19 (Figure 1). Kruskal-Wallis post hoc test. A Shapiro-Wilk test showed that the distribution of NOD2 and Spp1 values departed significantly We performed commercially available ELISAs to detect the from normality (W = 0.8, p < 0.0001 and W = 0.94, p < 0.0001, expression of NOD2, CD163, CXCL16, and Spp1 in the CSF respectively). Based on this outcome, a nonparametric test (the and serum from patients with ALS and HCs. None of these Mann-Whitney U test) was used to compare the protein con- candidate proteins were detected by ELISA in the CSF in centration between patient groups. We evaluated the discrimi- either group. For this reason, we performed a preliminary nation power between patients with ALS and HCs by means of study analyzing protein levels in the serum from 20 randomly the corresponding receiver operating characteristic (ROC) curve selected patients and 20 age-matched and sex-matched HCs. for both proteins. The Spearman rank correlation was computed No significant differences were observed in CD163 or in to assess the relationship between NOD2 and Spp1 and total CXCL16 expression in either group. However, we observed ALSFRS-R, time from blood sampling and disease onset, ALS significant differences between the 2 groups in both NOD2 progression rate, and survival. Data analysis and graphical rep- and Ssp1 expression. Thus, we increased the number of resentation was performed using the STATA 13 software for samples analyzed to 123 serum samples from patients with Windows (StataCorp LP, College Station, TX). ALS and 102 control serum samples. We also analyzed the expression of NOD2 and Spp1 in the serum of 28 patients Standard Protocol Approvals, Registrations, with FTD and 30 patients with AD. and Patient Consents This study was approved by the Hospital de la Santa Creu I NOD2 Analysis Sant Pau Ethics Committee. The CSF and serum samples NOD2 levels in serum differed significantly between patients were obtained and registered according to standard protocols with ALS (273.6 pg/mL, SD 257.8) and healthy individuals and previous written informed consent by the patients. Brain (164.6 pg/mL, SD 181.9) (p = 0.0007). ROC analysis showed tissue samples were provided by the Neurologic Tissue Bank an area under the curve (AUC) of 0.63 (Figure 2A). Statistically of the Biobanc-Hospital Clinic-IDIBAPS. All participants or significant differences were observed between patients with ALS their relatives had given their written informed consent to use and those with FTD (57.8 pg/mL, SD 78.2) (p < 0.0001) or AD the sample for this study. (75.0 pg/mL, SD 126.2) (p < 0.0001) (Figure 3A). We found no correlation between NOD2 levels and ALSFRS-R (p = 0.66), Data Availability time from blood sampling to disease onset (p = 0.82), survival (p Anonymized data not published within this article will be = 0.57), and ALS progression (p = 0.71). made available by request from any qualified investigator. Spp1 Analysis Spp1 expression showed statistically significant differences Results between patients with ALS (38.7 ng/mL, SD 21.0) and Sample Composition of the Validation Cohort healthy individuals (mean 17.8 ng/mL, SD 14.0) (p < We included 123 patients with ALS, 28 patients with FTD, 30 0.0001). ROC analysis showed an AUC of 0.81 (Figure 2B). patients with AD dementia, and 102 HCs. Twenty-three of 123 No significant differences were observed in Spp1 levels be- patients with ALS had associated FTD. Eleven patients had tween patients with ALS and patients with AD (34.0, SD 12.7, genetically confirmed ALS (6 C9orf72, 2 SOD1, 1 FUS, 1TBK1, p = 0.51) and those with FTD (39.9, SD 14.7, p = 0.42) and 1 TARDBP). Among patients with FTD, we included 16 (Figure 3A). We found a negative correlation between Spp1 cases of bvFTD, and 14 were primary progressive aphasia. Two levels and ALSFRS-R (r = −0.24, p = 0.009) (Figure 4). No patients with FTD later developed motor neuron involvement correlation was found between Spp1 levels and time from during follow-up and were moved to the ALS group for analysis. blood sampling to disease onset (p = 0.30), survival (p = 0.25) and ALS progression (p = 0.90). Proteomic Analysis CSF analysis using mass spectrometry detected 801 proteins, 450 of which were identified with more than 1 peptide ho- Discussion mology and 25 were found to be differentially expressed proteins (p < 0.05). If we restrict the number of recognized Our CSF discovery–based screening revealed 16 proteins peptides to more than 2, 16 proteins were found upregulated were differentially expressed in patients with ALS compared in patients with ALS compared with HCs. with that in HCs. Four of these 16 proteins were also 4 Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 Neurology.org/NN Figure 1 Box Plot Showing the Normalized RNA Expression Values of the 4 Genes Encoding the Prioritized Proteins in the Motor Cortex of Patients With Sporadic ALS and Healthy Controls Obtained Through RNA Sequencing *p < 0.05; **p < 0.01 ***p < 0.001. sALS = sporadic amyotrophic lateral sclerosis. significantly upregulated in the transcriptome of the motor patients with ALS than in those with other neurodegenerative cortex from patients with ALS. We therefore prioritized these diseases. NOD2 gene expression was also upregulated in the 4 proteins for further analyses. Our results were validated by motor cortex from patients with ALS when compared with ELISA in sera samples and showed differentially increased that in controls. NOD2 belongs to the nucleotide-binding levels in 2 of these proteins, NOD2 and Spp1. No significant domain leucine-rich repeat-containing receptors (NLRs) differences were found in CXCL16 and CD163 expression family. It can interact with multiple proteins and modulates in the serum from patients with ALS compared with those immune responses in a stimuli-dependent manner. The NLR from HCs. NOD2 serum levels were significantly higher in family, together with the toll-like receptor family and retinoic Figure 2 Analysis of Diagnostic Utility of Serum Oligomerization Domain–Containing Protein 2 (NOD2) and Osteopontin (Spp1) Through Receiver Operating Characteristic (ROC) Curves Sensitivity and specificity of serum (A) oligomerization domain–containing protein 2 (NOD2) and (B) osteopontin (Spp1) levels in patients with ALS and healthy controls. ALS = amyotrophic lateral sclerosis; HC = healthy control. Neurology.org/NN Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 5 Figure 3 Levels of Nucleotide-Binding Oligomerization Domain–Containing Protein 2 (NOD2) and Osteopontin (Spp1) Across Diagnostic Groups (Controls, n = 102; ALS, n = 123; AD, n = 30; and FTD, n = 28) Group differences between serum biomarker raw levels of (A) NOD2 and (B) Spp1. Error bars represent the 10th to 90th percentile range of data, and circles represent outliers. *<0.001. AD = Alzheimer disease; ALS = amyotrophic lateral sclerosis; FTD = frontotemporal dementia; HC = healthy control. acid-inducible gene I (RIG-I)–like helicase receptor family 12 and is essential in type I immunity.26 Expression of Spp1 (including RIG-I), belongs to pathogen recognition receptors can be upregulated by proinflammatory cytokines, under that are activated either by evolutionary conserved PAMPs or hypoxic conditions or in response to stress or cellular by DAMPs. RIG-I is an RNA helicase that recognizes viral injury.27-29 double-stranded RNA and initiates an innate immune response through mitochondrial antiviral signaling (MAVS). NOD2 Spp1 could induce autocrine and paracrine signaling by senses single-stranded RNA from viruses and interacts with binding to the cell surface receptors integrin αvβ2 and/or MAVS, activating interferon (IFN) regulatory factor–3 and CD44 and transduce cell matrix signaling into cell regulating expression of type I IFN. Both NOD2 and RIG-I exhibit broad- proliferation, survival, migration, and angiogenesis.27 CD44 spectrum anti-PAMPs and DAMPS. NOD2 activation mediates acts as a ligand receptor by interacting with Spp1. It has been numerous interactions and responses that include regulation implicated in the pathogenesis of inflammation in the CNS, of inflammasome, production of proinflammatory cytokines, contributing to the recruitment of inflammatory cells. It is production of type I IFNs, and triggering autophagy.23 expressed in most astrocytes and in a portion of microglia.30,31 Cytoplasmic protein aggregates of TDP-43 and stress gran- ules, formed by ribonucleoproteins are pathologic markers of ALS.5,24 RIG-I has been described as a direct target of TDP-43 Figure 4 Correlation Between the Raw Osteopontin (Spp1) Levels and the ALSFRS-R Score at the Moment of because its mRNA contains (uracil guanine)3-5 repeats, which Sample Extraction in the ALS Group are known binding motifs of TDP-43.25 Similarly, in silico analysis of NOD2 mRNA showed that it is rich in conserved uracil guanine/guanine uracil rich repeats interspaced by other nucleotides, a major factor influencing the interaction of NOD2 mRNA and TDP-43. These RNA molecules from cytoplasmic aggregates could activate NOD2, triggering the neuroinflammatory response observed in patients with ALS. Neurodegeneration involves a neuroinflammatory compo- nent that accelerates ALS progression and accentuates disease severity,9 and NOD2 could have a role in the inflammatory process observed in patients with ALS. These results suggest NOD2 has a specific role of in ALS pathophysiology because it is not elevated in other neurodegenerative diseases. We also found high levels of Spp1 in the serum of patients ALS = amyotrophic lateral sclerosis; ALSFRS-R = Revised ALS functional rating with ALS. Spp1 is a multifunctional, proinflammatory cyto- scale. kine that upregulates the expression of IFN-γ and interleukin 6 Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 Neurology.org/NN During the pathogenesis of ALS, astrocytes and microglia are NOD2 and SPP1 genes in brain tissue of patients with ALS by activated,32 contributing to the progression of disease.30 RNA sequencing, unfortunately, we were unable to detect our candidate biomarkers using ELISA in the CSF. This would have Although the role of Spp1 in ALS is largely unknown, CSF been desirable for a more detailed understanding of their pos- levels of this protein have been proposed as a potential bio- sible role in the underlying physiopathogenic mechanism and marker in ALS.33,34 A recent study showed that Spp1 is over- disease process. However, we were able to detect these proteins expressed in perivascular fibroblasts from the spinal cord of in sera, which is an easier option to use in clinical practice. patients with ALS.35 The authors also found that high plasma levels of Spp1 correlated with shorter survival.35 By contrast, we Our discovery-based approach allowed us to identify a novel found that Spp1 serum levels in our ALS group correlated with biomarker that points to a relevant role of neuroinflammatory more severe disease (lower ALSFRS-R scores) during sample activity in the pathophysiology of ALS and adds evidence to collection. However, this correlation was weak. In addition, we the involvement of Spp1 in the disease process. Nevertheless, did not find a correlation either with the disease progression the precise role of inflammation in neurodegenerative disor- rate or survival. Longitudinal studies with follow-up samples ders can be difficult to clarify. Our data do not prove a would be necessary to evaluate Spp1 level use in clinical causative role of inflammation in neurodegeneration. In fact, it practice. Although Spp1 levels were found to be increased in could be a response to an active degenerative process. How- patients with ALS in comparison with HCs, we did not find ever, our results provide evidence of dysregulated pathways differences in Spp1 levels in patients with ALS when compared that can be a potential target for therapeutic interventions. with levels in patients with other neurodegenerative diseases such as AD and FTD. It therefore seems that increased levels of Our data suggest that an impaired innate immunity, impaired Spp1 are not a distinct finding of ALS alone, but rather a autophagy, and protein homeostasis can facilitate protein aggre- neurodegeneration biomarker. Further prospective studies are gation, thereby influencing disease progression. These findings needed to elucidate the role of upregulation of Spp1 and its shed light on the cell signaling mechanisms involved in ALS. utility in the diagnosis and prognosis of the disease. These mechanisms may serve as potential targets for therapeutic strategies and may promote the development of disease-modifying Although the differences in serum levels of NOD2 between our treatments for ALS, aimed at modulating neuroinflammation. patients and controls were significant, ROC analysis suggests the value of this protein as a diagnostic marker is limited. Further longitudinal studies, including other neurodegenerative NOD2 levels were also significantly higher in patients with ALS disorders, such as Parkinson disease, are needed to elucidate the than in the groups with other neurodegenerative diseases, but clinical utility of NOD2 and Spp1 levels as diagnostic and we cannot conclude from this that NOD2 provides a definitive prognostic biomarkers in clinical practice and to understand their practical utility to differentiate ALS from other diseases. role in the disease process and progression of ALS. However, these results suggest a different pathophysiologic pathway is involved in neurodegeneration. Similarly, we found Acknowledgment Spp1 levels in patients with ALS were higher than those in The authors thank the patients and their relatives for their HCs. In this case, ROC analysis showed an AUC of 0.82, support for this study. The authors dedicate this work to our indicating it can discriminate between the 2 groups with good master and colleague Isabel Illa. sensitivity and specificity and suggesting that Spp1 may be a good biomarker in the diagnostic process of ALS. Study Funding This work was supported by grants from the Fondo de The main strengths of this study are the prospective design and Investigación Sanitaria (FIS), Instituto de Salud Carlos III the detailed phenotype description that includes a standardized (PI15/01618 and PI 19/01543 to R. Rojas-Garcia and PI21/ cognitive and behavioral assessment in all patients. The ho- 00791 to I. Illan-Gala), and cofunded by European Union mogenous assessment applied within a single center and the (ERDF/ESF, “A way to make Europe”/“Investing in your inclusion of other neurodegenerative diseases allowed us to in- future”); CERCA programme/Generalitat de Catalunya; vestigate the role of biomarkers in detail in different settings. The FUNDELA (Por un mundo sin ELA, 2019 to O. Dols- main limitation of the study is that although we found significant Icardo); A. Carbayo is supported by the Rio Hortega contract differences between patients with ALS and controls in the serum (cm21/00057) from Instituto de Salud Carlos III. O. Dols- levels of NOD2 and Spp1, we did not find serum levels of Spp1 Icardo is a recipient of a grant by The Association for Fron- differed significantly between patients with ALS and those with totemporal Degeneration (Clinical Research Postdoctoral other neurodegenerative conditions, and we observed a mod- Fellowship, AFTD 2019–2021); E. Cortés-Vicente and I. erate overlap between groups. Although this overlap might limit Illan-Gala are supported by the Juan Rodés Contract (JR19/ the value of Spp1 and NOD2 as diagnostic markers, our results 00037 and JR20/0018 respectively) from Instituto de Salud may help understand the role of neuroinflammation in the Carlos III. I. Illan-Gala is also an Atlantic Fellow for Equity in pathophysiology of neurodegenerative diseases. In addition, al- Brain Health at the Global Brain Health Institute (GBHI) and though our results showed differentially expressed NOD2 and is supported with funding from the Alzheimer’s Association Spp1 in CSF tested by mass spectrometry and upregulation of (GBHI ALZ UK-21-720973 and AACSF-21-850193). N. de Neurology.org/NN Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 7 Luna, E. Cortés-Vicente, J. Turon-Sans, L. Querol, and R. Rojas-Garcia are members of the ERN Euro-NMD. Appendix (continued) Name Location Contribution Disclosure N. de Luna, A. Carbayo, O. Dols-Icardo, J. Turon-Sans, D. Janina Centro de Investigación Drafting/revision of the Turon-Sans, Biomédica en Red de article for content, including Reyes-Leiva, D. Reyes-Leiva, I. Illán-Gala, I. Jericó, I. Pagola- MD Enfermedades Raras medical writing for content; Lorz, C. Lleixà, L. Querol, and S. Rubio-Guerra report no (CIBERER), Madrid, Spain analysis or interpretation of Motor Neuron Diseases Clinic, data disclosures. D. Alcolea participated in advisory boards for Neuromuscular Diseases Unit, Fujirebio-Europe and Roche Diagnostics and received Department of Neurology, Hospital de la Santa Creu i Sant speaker honoraria from Fujirebio-Europe, Roche Diagnostics, Pau, Departament de Nutricia, Krka Farmacéutica S.L., Zambon S.A.U. and Esteve Medicina, Universitat Autònoma de Barcelona, Spain Pharmaceuticals. S.A. D. Alcolea declares a filed patent ap- plication (WO2019175379 A1 Markers of synaptopathy in David Neuromuscular Diseases Drafting/revision of the Reyes- Laboratory, Institut de article for content, including neurodegenerative disease). J. Fortea, A. Lleó, E. Cortés- Leiva, MD Recerca Hospital de la Santa medical writing for content; Vicente, and R. Rojas-Garcı́a report no disclosures. Go to Creu i Sant Pau (IIB Sant-Pau), analysis or interpretation of Barcelona, Spain data Neurology.org/NN for full disclosures. Centro de Investigación Biomédica en Red de Publication History Enfermedades Raras (CIBERER), Madrid, Spain Received by Neurology: Neuroimmunology & Neuroinflammation Motor Neuron Diseases July 12, 2022. Accepted in final form October 13, 2022. Submitted Clinic, Neuromuscular Diseases Unit, Department of and externally peer reviewed. The handling editor was Josep O. Dalmau, Neurology, Hospital de la MD, PhD, FAAN. Santa Creu i Sant Pau, Departament de Medicina, Universitat Autònoma de Barcelona, Spain Appendix Authors Ignacio Sant Pau Memory Unit, Drafting/revision of the Name Location Contribution Illan-Gala, Department of Neurology, article for content, including MD, PhD Hospital de la Santa Creu i Sant medical writing for content Noemı́ de Neuromuscular Diseases Drafting/revision of the Pau, Universitat Autònoma de Luna, PhD Laboratory, Institut de article for content, including Barcelona, Spain Recerca Hospital de la Santa medical writing for content; Centro de Investigación Creu i Sant Pau (IIB Sant-Pau), major role in the acquisition Biomédica en Red de Barcelona, Spain of data; study concept or Enfermedades Centro de Investigación design; and analysis or Neurodegenerativas Biomédica en Red de interpretation of data (CIBERNED), Madrid, Spain Enfermedades Raras (CIBERER), Madrid, Spain Ivonne Neuromuscular and motor Drafting/revision of the Jericó, MD, neuron diseases research article for content, including Álvaro Neuromuscular Diseases Drafting/revision of the PhD Group-Health Research medical writing for content; Carbayo, Laboratory, Institut de article for content, including Institute of Navarra (IdisNA) analysis or interpretation of MD Recerca Hospital de la Santa medical writing for content; data Creu i Sant Pau (IIB Sant-Pau), analysis or interpretation of Barcelona, Spain data Inma Neuromuscular and motor Drafting/revision of the Centro de Investigación Pagola- neuron diseases research article for content, including Biomédica en Red de Lorz, MD, Group-Health Research medical writing for content; Enfermedades Raras PhD Institute of Navarra (IdisNA) analysis or interpretation of (CIBERER), Madrid, Spain data Motor Neuron Diseases Clinic, Neuromuscular Diseases Unit, Cinta Neuromuscular Diseases Drafting/revision of the Department of Neurology, Lleixà, MD, Laboratory, Institut de article for content, including Hospital de la Santa Creu i Sant PhD Recerca Hospital de la Santa medical writing for content; Pau, Departament de Creu i Sant Pau (IIB Sant-Pau), major role in the acquisition Medicina, Universitat Barcelona, Spain of data Autònoma de Barcelona, Spain Luis Neuromuscular Diseases Drafting/revision of the Oriol Dols- Sant Pau Memory Unit, Drafting/revision of the Querol, MD, Laboratory, Institut de article for content, including Icardo, PhD Department of Neurology, article for content, including PhD Recerca Hospital de la Santa medical writing for content Hospital de la Santa Creu i medical writing for content; Creu i Sant Pau (IIB Sant-Pau), Sant Pau, Universitat major role in the acquisition Barcelona, Spain Autònoma de Barcelona, of data; study concept or Centro de Investigación Spain design; and analysis or Biomédica en Red de Centro de Investigación interpretation of data Enfermedades Raras Biomédica en Red de (CIBERER), Madrid, Spain Enfermedades Motor Neuron Diseases Neurodegenerativas Clinic, Neuromuscular (CIBERNED), Madrid, Spain Diseases Unit, Department of Centro de Investigación Neurology, Hospital de la Biomédica en Red de Santa Creu i Sant Pau, Enfermedades Departament de Medicina, Neurodegenerativas Universitat Autònoma de (CIBERNED), Madrid, Spain Barcelona, Spain 8 Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 Neurology.org/NN Appendix (continued) Appendix (continued) Name Location Contribution Name Location Contribution Sara Rubio- Sant Pau Memory Unit, Drafting/revision of the Investigación Biomédica en design; and analysis or Guerra, MD Department of Neurology, article for content, including Red de Enfermedades Raras interpretation of data Hospital de la Santa Creu i medical writing for content (CIBERER), Madrid, Spain; Sant Pau, Universitat Motor Neuron Diseases Autònoma de Barcelona, Clinic, Neuromuscular Spain; Centro de Diseases Unit, Department of Investigación Biomédica en Neurology, Hospital de la Red de Enfermedades Santa Creu i Sant Pau, Neurodegenerativas Departament de Medicina, (CIBERNED), Madrid, Spain; Universitat Autònoma de Centro de Investigación Barcelona, Spain Biomédica en Red de Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain References Daniel Sant Pau Memory Unit, Drafting/revision of the 1. Brown RH, Al-Chalabi A. Amyotrophic lateral sclerosis. N Engl J Med. 2017;377(2): Alcolea, Department of Neurology, article for content, including 162-172. doi: 10.1056/nejmra1603471. MD, PhD Hospital de la Santa Creu i medical writing for content 2. Brooks BR, Miller RG, Swash M, Munsat TL. El Escorial revisited: revised criteria for Sant Pau, Universitat the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Mot Autònoma de Barcelona, Neuron Disord. 2000;1(5):293-299. doi: 10.1080/146608200300079536. Spain; Centro de 3. Turner MR, Hardiman O, Benatar M, et al. Controversies and priorities in amyotrophic Investigación Biomédica en lateral sclerosis. Lancet Neurol. 2013;12(3):310-322. doi: 10.1016/s1474-4422(13)70036-x. Red de Enfermedades 4. Brettschneider J, Del Tredici K, Toledo JB, et al. Stages of pTDP-43 pathology in Neurodegenerativas amyotrophic lateral sclerosis. Ann Neurol. 2013;74(1):20-38. doi: 10.1002/ana.23937. (CIBERNED), Madrid, Spain 5. Aksoy YA, Deng W, Stoddart J, et al. Stressed out: the role of FUS and TDP-43 in amyotrophic lateral sclerosis. Int J Biochem Cell Biol. 2020;126:105821. doi: 10.1016/ Juan Sant Pau Memory Unit, Drafting/revision of the j.biocel.2020.105821. Fortea, MD, Department of Neurology, article for content, including 6. Liu J, Wang F. Role of neuroinflammation in amyotrophic lateral sclerosis: cellular PhD Hospital de la Santa Creu i medical writing for content mechanisms and therapeutic implications. Front Immunol. 2017;8:1005. doi: 10.3389/ Sant Pau, Universitat fimmu.2017.01005. Autònoma de Barcelona, 7. Amin A, Perera ND, Beart PM, Turner BJ, Shabanpoor F. Amyotrophic lateral Spain; Centro de sclerosis and autophagy: dysfunction and therapeutic targeting. Cells. 2020;9(11): Investigación Biomédica en 2413. doi: 10.3390/cells9112413. Red de Enfermedades 8. Michaelson N, Facciponte D, Bradley W, Stommel E. Cytokine expression levels in Neurodegenerativas ALS: a potential link between inflammation and BMAA-triggered protein misfolding. (CIBERNED), Madrid, Spain Cytokine Growth Factor Rev. 2017;37:81-88. doi: 10.1016/j.cytogfr.2017.05.001. 9. Beers DR, Zhao W, Neal DW, et al. Elevated acute phase proteins reflect peripheral Alberto Sant Pau Memory Unit, Drafting/revision of the inflammation and disease severity in patients with amyotrophic lateral sclerosis. Sci Lleó, MD, Department of Neurology, article for content, including Rep. 2020;10(1):15295. doi: 10.1038/s41598-020-72247-5. PhD Hospital de la Santa Creu i medical writing for content 10. Guan Y, Han F. Key mechanisms and potential targets of the NLRP3 inflammasome Sant Pau, Universitat in neurodegenerative diseases. Front Integr Neurosci. 2020;14:37. doi: 10.3389/ Autònoma de Barcelona, fnint.2020.00037. Spain; Centro de 11. Illan-Gala I, Alcolea D, Montal V, et al. CSF sAPPβ, YKL-40, and NfL along the ALS-FTD Investigación Biomédica en spectrum. Neurology. 2018;91(17):e1619-e1628. doi: 10.1212/wnl.0000000000006383. Red de Enfermedades 12. Steinacker P, Verde F, Fang L, et al. Chitotriosidase (CHIT1) is increased in microglia Neurodegenerativas and macrophages in spinal cord of amyotrophic lateral sclerosis and cerebrospinal (CIBERNED), Madrid, Spain fluid levels correlate with disease severity and progression. J Neurol Neurosurg Psy- chiatry. 2018;89(3):239-247. doi: 10.1136/jnnp-2017-317138. Elena Neuromuscular Diseases Drafting/revision of the 13. Thompson AG, Gray E, Thezenas ML, et al. Cerebrospinal fluid macrophage bio- Cortés- Laboratory, Institut de article for content, including markers in amyotrophic lateral sclerosis. Ann Neurol. 2018;83(2):258-268. doi: Vicente, Recerca Hospital de la Santa medical writing for content; 10.1002/ana.25143. MD, PhD Creu i Sant Pau (IIB Sant-Pau), major role in the acquisition 14. Alcolea D, Clarimón J, Carmona-Iragui M, et al. The Sant Pau Initiative on Neuro- Barcelona, Spain; Centro de of data; study concept or degeneration (SPIN) cohort: a dataset for biomarker discovery and validation in Investigación Biomédica en design; and analysis or neurodegenerative disorders. Alzheimers Dement (N Y). 2019;5:597-609. Red de Enfermedades Raras interpretation of data 15. Alcolea D, Martinez-Lage P, Sanchez-Juan P, et al. Amyloid precursor protein me- (CIBERER), Madrid, Spain; tabolism and inflammation markers in preclinical Alzheimer disease. Neurology. 2015; Motor Neuron Diseases 85(7):626-633. doi: 10.1212/wnl.0000000000001859. Clinic, Neuromuscular 16. Barbero F, Michelini S, Moriones OH, et al. Role of common cell culture media Diseases Unit, Department of supplements on citrate-stabilized gold nanoparticle protein corona formation, ag- Neurology, Hospital de la gregation state, and the consequent impact on cellular uptake. Bioconjug Chem. 2022; Santa Creu i Sant Pau, 33(8):1505-1514. doi: 10.1021/acs.bioconjchem.2c00232. Departament de Medicina, 17. Garcia-Berrocoso T, Llombart V, Colas-Campas L, et al. Single cell immuno-laser Universitat Autònoma de microdissection coupled to label-free proteomics to reveal the proteotypes of human Barcelona, Spain brain cells after ischemia. Mol Cell Proteomics. 2018;17(1):175-189. doi: 10.1074/ mcp.ra117.000419. Ricardo Neuromuscular Diseases Drafting/revision of the 18. Simats Oriol A. Potential Terapèutic de Nous Biomarcadors d’isquèmia Cerebral. Uni- Rojas- Laboratory, Institut de article for content, including versitat Autònoma de Barcelona Departament de Bioquı́mica i Biologia Molecular; Garcia, MD, Recerca Hospital de la Santa medical writing for content; 2018. hdl.handle.net/10803/662614. PhD Creu i Sant Pau (IIB Sant-Pau), major role in the acquisition 19. Dols-Icardo O, Montal V, Sirisi S, et al. Motor cortex transcriptome reveals microglial Barcelona, Spain; Centro de of data; study concept or key events in amyotrophic lateral sclerosis. Neurol Neuroimmunol Neuroinflamm. 2020;7(5):e829. doi: 10.1212/nxi.0000000000000829. Neurology.org/NN Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 9 20. Gorno-Tempini ML, Hillis AE, Weintraub S, et al. Classification of primary pro- 28. Sodhi CP, Phadke SA, Batlle D, Sahai A. Hypoxia and high glucose cause exaggerated gressive aphasia and its variants. Neurology. 2011;76(11):1006-1014. doi: 10.1212/ mesangial cell growth and collagen synthesis: role of osteopontin. Am J Physiol Ren wnl.0b013e31821103e6. Physiol. 2001;280(4):F667-F674. doi: 10.1152/ajprenal.2001.280.4.f667. 21. Rascovsky K, Hodges JR, Knopman D, et al. Sensitivity of revised diagnostic criteria 29. Uchibori T, Matsuda K, Shimodaira T, Sugano M, Uehara T, Honda T. IL-6 trans- for the behavioural variant of frontotemporal dementia. Brain. 2011;134(pt 9): signaling is another pathway to upregulate Osteopontin. Cytokine. 2017;90:88-95. doi: 2456-2477. doi: 10.1093/brain/awr179. 10.1016/j.cyto.2016.11.006. 22. Labra J, Menon P, Byth K, Morrison S, Vucic S. Rate of disease progression: a 30. Matsumoto T, Imagama S, Hirano K, et al. CD44 expression in astrocytes and prognostic biomarker in ALS. J Neurol Neurosurg Psychiatry. 2016;87(6):628-632. doi: microglia is associated with ALS progression in a mouse model. Neurosci Lett. 2012; 10.1136/jnnp-2015-310998. 520(1):115-120. doi: 10.1016/j.neulet.2012.05.048. 23. Dominguez-Martinez DA, Nunez-Avellaneda D, Castanon-Sanchez CA, Salazar MI. 31. Nikodemova M, Small AL, Smith SM, Mitchell GS, Watters JJ. Spinal but not cortical NOD2: activation during bacterial and viral infections, polymorphisms and potential microglia acquire an atypical phenotype with high VEGF, galectin-3 and osteopontin, as therapeutic target. Rev Invest Clin. 2018;70(1):18-28. doi: 10.24875/ric.17002327. and blunted inflammatory responses in ALS rats. Neurobiol Dis. 2014;69:43-53. doi: 24. Mackenzie IR, Rademakers R, Neumann M. TDP-43 and FUS in amyotrophic lateral 10.1016/j.nbd.2013.11.009. sclerosis and frontotemporal dementia. Lancet Neurol. 2010;9(10):995-1007. doi: 32. Beers DR, Zhao W, Liao B, et al. Neuroinflammation modulates distinct regional and 10.1016/s1474-4422(10)70195-2. temporal clinical responses in ALS mice. Brain Behav Immun. 2011;25(5):1025-1035. 25. MacNair L, Xiao S, Miletic D, et al. MTHFSD and DDX58 are novel RNA-binding doi: 10.1016/j.bbi.2010.12.008. proteins abnormally regulated in amyotrophic lateral sclerosis. Brain. 2016;139(1): 33. Varghese AM, Sharma A, Mishra P, et al. Chitotriosidase–a putative biomarker for 86-100. doi: 10.1093/brain/awv308. sporadic amyotrophic lateral sclerosis. Clin Proteomics. 2013;10(1):19. doi: 10.1186/ 26. Zhao K, Zhang M, Zhang L, et al. Intracellular osteopontin stabilizes TRAF3 to 1559-0275-10-19. positively regulate innate antiviral response. Sci Rep. 2016;6(1):23771. doi: 10.1038/ 34. von Neuhoff N, Oumeraci T, Wolf T, et al. Monitoring CSF proteome alterations in srep23771. amyotrophic lateral sclerosis: obstacles and perspectives in translating a novel marker 27. Mahmud FJ, Du Y, Greif E, et al. Osteopontin/secreted phosphoprotein-1 behaves as panel to the clinic. PLoS One. 2012;7(9):e44401. doi: 10.1371/journal.pone.0044401. a molecular brake regulating the neuroinflammatory response to chronic viral in- 35. Manberg A, Skene N, Sanders F, et al. Altered perivascular fibroblast activity precedes fection. J Neuroinflammation. 2020;17(1):273. doi: 10.1186/s12974-020-01949-4. ALS disease onset. Nat Med. 2021;27(4):640-646. doi: 10.1038/s41591-021-01295-9. 10 Neurology: Neuroimmunology & Neuroinflammation | Volume 10, Number 2 | March 2023 Neurology.org/NN