Amyotrophic Lateral Sclerosis (ALS) is a fatal and relentlessly progressive neurodegenerative disorder. Although, clinically characterized by the loss of both upper and lower motor neurons, it is a multi-systemic condition driven by several cell non-autonomous processes. Glial dysregulation in particular can exacerbate disease progression and is necessary for motor neuronal death to occur (1–3). Multiple lines of evidence have shown that this dysregulation extends to the peripheral innate immune system. Patient monocytes have a pro-inflammatory transcriptomic profile (4), secrete increased levels of pro-inflammatory cytokines (5), and can infiltrate the central nervous system (CNS) (6); furthermore, these alterations can influence disease progression. Crucially, monocytes can be readily sampled and differentiated to macrophages ex vivo and although ontogenetically different, monocyte-derived macrophages (MoMas) and microglia functionally complement each other (7). Studying macrophages may therefore help understand disease-associated inflammatory sequelae in the CNS.

In lieu of this, reports of elevated cerebrospinal fluid (CSF) levels of chitinases in multiple neurodegenerative conditions are particularly intriguing, as they are considered markers of chronic gliosis (8–10). The chitinases, including CHIT1, CHI3L1, and CHI3L2, belong to the family 18 glycosyl hydrolases and bind to chitin, a natural polysaccharide found in the coating of various pathogens with high affinity. Dysregulated chitinase levels have been noted in a range of non-infectious diseases, including asthma, chronic obstructive pulmonary disease, multiple sclerosis, and even Alzheimer's disease. Multiple lines of evidence suggest that chitinases aren't merely markers of disease status, but are active components of the immunological response in pathological conditions characterized by chronic inflammation.

In ALS, these moieties exacerbate neuroinflammation and directly affect neuronal viability (11–13). While studies using post-mortem motor cortex and spinal cord tissue from ALS patients have reported micro-and astroglial expression of CHIT1 and CHI3L1, respectively, the cellular origins of these targets remain to be fully understood. In vitro studies using healthy controls (HCs) have shown that the chitinases are produced by mature macrophages, wherein they display distinct temporal expression patterns (14–16). Therefore, the objective of this pilot study was to examine whether these cells also contribute to chitinase dysregulation in ALS. To do so, we examined the “baseline” expression of CHIT1, CHI3L1 and, CHI3L2 in non-polarized monocyte-derived macrophages (MoMas) in patients with ALS relative to HCs.

The final cohort from which MoMas were generated included 8 patients with ALS and 8 HCs. ALS patients were significantly older than HCs (ALS = 60.5 ± 7.7 years vs. HC = 51 ± 7.9 years, t (14) = 2.403, p = 0.03) and both groups had a greater proportion of males (ALS = 6, HC = 7) than females. Four patients were receiving riluzole for ≥2 months at the time of sampling. None of the ALS patients had active cancer or manifest diabetes. The additional NDC cohort recruited to allow CSF chitinase analyses was representative of the HC cohort both in terms of age and sex distribution. Further demographic details are outlined in Table 1. Additional diagnostic information for the NDC cohort is provided in Supplementary Table 1.

As seen in Figure 1, CHIT1, CHI3L1, and CHI3L2 were detectable in MoMas from both ALS patients and HCs. However, they displayed distinct regulatory profiles. Temporal expression patterns were similar for CHIT1 (Figures 1A,D) and CHI3L1 (Figures 1B,E) in both ALS and HCs: relative gene and protein expression for both targets were minimal at earlier time-points, increased over time, and peaked on Day 9. However, no such temporal regulation was observed for CHI3L2 at either the transcriptomic (Figure 1C) or the protein level (Figure 1F) in either group. To illustrate, in HCs, mean secreted CHIT1 and CHI3L1 levels increased by 42.6 and 625%, respectively, from Day 1 to 9; conversely, for CHI3L2 only a 9% increase was observed. Indeed, secreted CHIT1 and CHI3L1 levels on D9 correlated significantly with each other but not with CHI3L2 (r_s = 0.69, p = 0.003).

Profound between-group differences were observed for CHIT1 and CHI3L1 at later time-points. Relative CHIT1 expression was significantly higher in ALS MoMas on day 6 (F(2, 12) = 17.93, p = 0.001, partial η² = 0.6) and day 9 (F(2, 12) = 15.42, p = 0.002, partial η² = 0.56). This effect was recapitulated at the protein level, wherein a statistically significant time × group interaction was observed despite the inclusion of age as a covariate, thus underscoring the effect of group on CHIT1 levels over time (F(3, 39) = 4.97, p = 0.005, partial η² = 0.27). ALS MoMas secreted significantly higher CHIT1 levels than HC MoMas on Day 9 (ALS = 39.4 ng/ml, [15.3, 53.5] vs. HC = 6.5 ng/ml, [−17.5, 30.6]) (F(1, 13) = 15.7, p = 0.002, partial η² = 0.55), despite having lower levels on Day 1 (ALS = 3.6 ng/ml [3.1, 4.2] vs. HC = 4.6 ng/ml [4.1,5.2] (F(1, 13) = 5.77, p = 0.032, partial η² = 0.31). An analogous trend was noted for CHI3LI: as seen in Figure 1B, relative CHI3L1 expression was higher in the ALS group at all time-points and particularly so at Day 6. However, this effect did not reach statistical significance (F(1.6, 19.2) = 1.59, p = 0.23, partial η² = 0.12). At the protein level however (Figure 1E), a statistically significant time × group interaction was noted (F(1.96, 27.43) = 7.09, p = 0.003, partial η² =.34, Greenhouse-Geisser correction χ²(5) = 13.04, p = 0.02). Further, univariate analyses showed that ALS MoMas secreted significantly higher CHI3L1 levels than HC MoMas on Day 6 (ALS = 153.18 ng/ml, [91.6, 214.7] vs. HC = 51.2 ng/ml [−10.2, 112.7]) (F(1, 14) = 8.92, p = 0.01, partial η² = 0.39), and Day 9 (ALS = 199.7 ng/ml [91.2, 308.2] vs. HC = 57.3 ng/ml [-51.2, 165.8]) (F(1, 14) = 5.83, p = 0.03, partial η² = 0.29). Crucially, this group effect did not retain significance after the inclusion of age as a covariate (F(2.08, 26.99) = 2.60, p = 0.09, partial η² = 0.17).

As seen in Figure 2, between-group biofluid analyses revealed that a disease-associated chitinase upregulation was only evident in CSF rather than plasma. Both CHIT1 and CHI3L1 plasma levels were largely similar between ALS patients and NDCs. Conversely, CSF CHI3L1 levels were significantly upregulated within the ALS group relative to the NDC group (ALS = 398.4 ng/ml, [256, 540.8] vs. NDC = 218.9 ng/ml [65.25, 372.5], U = 6, p = 0.017). While a considerable upregulation was also noted for CSF CHIT1 levels, the effect did not reach significance (ALS = 14.56 ng/ml, [−6.02, 35.14] vs. NDC = 2.65 ng/ml [−0.31, 5.62], U = 6, p = 0.017).

Finally, within the ALS group, no significant correlations were observed between secreted chitinase levels on Day 9 (this time-point was selected as this is when transcriptomic and protein expression peaked) and the total ALSFRS-R score, calculated PR, D50 and rD50 (data not shown).

To our knowledge, this pilot study is the first to report a dysregulated chitinase profile in peripheral innate immune cells from ALS patients. By studying the transcriptomic and protein expression of key chitinases in non-polarized MoMas, we show here that macrophages in ALS have an intrinsically augmented capacity to secrete chitinases. To begin with, the temporal regulation patterns observed here are in keeping with previous studies; CHIT1 and CHI3L1 are minimally expressed in monocytes and highly upregulated during later stages of macrophage differentiation (14, 15). Conversely, CHI3L2 expression remains minimal across the differentiation process and is only upregulated as a result of stimulation (16). Here, the static and minimal CHI3L2 expression in both groups also serves to reinforce that the cells were at “baseline” and not stimulated as a result of the differentiation process itself. This, coupled with the absence of a group-associated effect, suggests that CNS rather than systemic immune cells likely contribute to the CHI3L2 elevations reported in the CSF of ALS patients. It is therefore unsurprising that the profound elevations we observed in CHIT1 and CHI3L1 expression in the ALS group were only evident at later time-points i.e. when cells were fully differentiated. Given the evidence that the chitinases are a feature of “M1-like” pro-inflammatory macrophages (18, 19), the upregulations observed here underscore how in ALS, peripheral myeloid cells are skewed toward a pro-inflammatory phenotype (4, 6). Indeed, monocytes from ALS patients are more readily differentiated toward an M1-like phenotype, wherein they produce higher levels of pro-inflammatory cytokines, including TNF-α and IL-6, than macrophages differentiated from HC monocytes (5). The data reported here are also interesting given that the chitinases themselves are active immune-modulators; for instance, stimulating monocytes with either CHI3L1 or CHIT1 resulted in the release of IL-8, MCP-1, and RANTES (20). Indeed, a “feed-forward” loop wherein the chitinases sustain neuroinflammation in ALS via their auto- and paracrine effects has already been postulated (12). For instance, Varghese et al. demonstrated that microglia appear to be the primary cellular source for CHIT1 in the CNS using murine cultures and that microglia themselves were susceptible to the effects of accumulated CHIT1, as they were chronically activated as a result of exposure (11). Another study also showed that conditioned medium from MoMas induced CHI3L1 transcription and morphological changes in cultured human astrocytes (19). Crucially, chitinase exposure was shown to increase leukocytic migratory capacity across an in vitro blood-brain barrier (BBB) model (20). Therefore, one might hypothesize that neuronal death and aggregate deposition could trigger chitinase expression by glial cells, thus creating a chemotactic axis recruiting circulating monocytes. Finally, the monocytes, by virtue of their intrinsically augmented chitinase synthesis capacity, exacerbate the neuroinflammatory milieu upon differentiation. In keeping with this hypothesis, Steinacker et al. (8) reported that in post-mortem spinal cord tissue from ALS patients, CHIT1 immuno-staining was primarily observed in CD68+ve macrophages: no expression was noted in tissue from HCs.

The upregulations in CSF CHIT1 and CHI3L1 levels in ALS patients relative to NDCs are concordant with previous studies (8, 21). Indeed, CSF CHIT1 in particular is now considered a surrogate marker of microglial activity and recommended for the differential diagnosis of ALS (22). As also previously reported in the literature, we noted no significant between-group differences in plasma levels, which suggests that the chitinase dysregulation observed in ALS MoMas is more reflective of the inflammatory microenvironment in the CNS than the periphery. This is reinforced by our observation that monocytic expression of chitinases in both ALS and HCs was almost negligible.

Undoubtedly, the effect of age must be considered; the results observed here are to be expected, given that chitinase levels, particularly those of CHI3L1, increase with age and are potentially indicative of the wider “inflammaging” process (23). While further studies with age-matched cohorts are warranted, we posit that the effects of age and disease on chitinase expression are not mutually exclusive and should not be studied as such, as the contribution of “immunosenescence” to neurodegenerative conditions has been extensively reported (24).

The present study is not without its limitations, with the restricted sample size being foremost. While it sufficed for demonstrating proof-of-principle, these results warrant validation within a more sizeable cohort. We believe this also explains why no correlations were observed with clinical indices. Detailed information on existing chronic comorbidities like diabetes was only available for some individuals. However, these have also been reported to influence chitinase levels (25).

Next, the present study did not assess enzymatic CHIT1 activity as genetic information for CHIT1 polymorphisms was not available for the cohort. The 24 bp duplication in exon 10 of the gene directly affects activity; heterozygous carriers display reduced activity and homozygous carriers display none at all (26). Therefore, the interpretation of these results would have been constrained, especially given that the prevalence of this polymorphism is almost 50% in European populations (27). However, given the observation that CHIT1 activity and protein levels are highly correlated, i.e., “elevated CHIT1 levels do not constitute inactive enzyme” (28), we posit that the results reported here are indeed evidence of a disease-associated CHIT1 upregulation in MoMas. Nevertheless, we recommend that future studies should include an assessment of CHIT1 activity.

Further studies with larger, age-matched and more representative cohorts can (1) help dissect the cumulative effect of age and disease on chitinase expression, (2) examine the implications for overall disease aggressiveness and acute activity, and (3) account for the dynamicity of the immune response by tracking chitinase expression across different disease phases.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethics committee of the Jena University Hospital (Jena, Germany, No. 3633-11/12). The patients/participants provided their written informed consent to participate in this study.

NG and JGr: study conceptualization and design. NG, EH, JGu, and AS: data acquisition, experimental execution, and troubleshooting. NG and TP: data analysis and visualization. NG: manuscript draft preparation. JGr, OW, RS and TP: revising the work for intellectual content. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.