Cerebrospinal fluid level of Nogo receptor 1 antagonist lateral olfactory tract usher substance (LOTUS) correlates inversely with the extent of neuroinflammation

Inflammation in the central nervous system (CNS) is associated with a wide range of neurological diseases. In particular, infection and autoimmunity cause acute or chronic inflammation in the CNS [1]. However, it is difficult to evaluate neuroinflammation in these diseases, in large part because of the lack of biomarkers suitable for judging their activities [2]. Although total protein concentrations or cell counts in the cerebrospinal fluid (CSF) are the most commonly used biomarkers for neuroinflammation in CNS diseases, they do not always vary in parallel with disease activities. For example, multiple sclerosis (MS), one of the most common neuroinflammatory diseases caused by an autoimmune mechanism, exhibits poor variation in conventional CSF biomarkers [3], making precise evaluation of disease activity difficult [4]. To complement CSF biomarkers, imaging biomarkers have recently been developed, including magnetic resonance imaging with ultra-small superparamagnetic iron oxide, a novel contrast agent, and positron-emission tomography using [¹¹C] (R)-PK11195 ligand. These new markers now play an important role in evaluation of the activities of neuroinflammatory diseases, including MS [3, 5]. However, the availability of cutting-edge imaging techniques is generally limited, even in the advanced treatment hospitals and especially in emergency situations. Therefore, the utility of currently available biomarkers for evaluating neuroinflammation is unsatisfactory, creating a demand for new and practical biomarkers.

Recently, we showed that lateral olfactory tract usher substance (LOTUS, also called Crtac1B) is critical for axonal growth. LOTUS acts as a Nogo receptor 1 (NgR1) antagonist, preventing Nogo from binding to NgR1 [6‐8]. Furthermore, we showed that LOTUS in the human CSF closely associated with the disease activity of MS, a representative neuroinflammatory disease accompanied by neurodegeneration [9‐11]. Because various molecules associated with axonal growth during brain development mediate not only axonal guidance but also immune responses [12‐17], our previous findings suggest that LOTUS is intimately involved in neuroinflammation, as well as axonal growth and/or degeneration in MS [18]. Therefore, we hypothesized that variations of LOTUS concentrations in the CSF may also reflect disease activity in other neuroinflammatory diseases.

In this study, to investigate the relationship between LOTUS and neuroinflammation, we analyzed the variation of LOTUS concentrations in CSF of patients with bacterial and viral meningitis. These patients’ clinical courses were objectively verified using established CSF markers such as total protein concentration, cell counts, and glucose, in addition to clinical findings, and we also performed comparisons between the acute and recovery phases in the same patients. Furthermore, we investigated the mechanisms underlying the observations in human patients using lipopolysaccharide-injected mice as a disease model of neuroinflammation.

The characteristics of the patients enrolled in this study are summarized in Table 1. As shown in Table 1, the timing of sample collection did not differ significantly between viral and bacterial meningitis in either the acute or recovery phases (p = 0.31 and 0.53, respectively).

First, we examined conventional CSF biomarkers of meningitis, such as total protein and cell counts, in the acute and recovery phases in the same patients. Objective improvements in these markers and clinical recovery were confirmed in all patients with meningitis (Figs. 1 and 2).

We next assessed the CSF LOTUS concentration in the acute and recovery phases by immunoblot analysis (Fig. 3a). The standard value for CSF LOTUS concentration was defined as ≥ 192.1 ng/ml, based on the results in healthy controls (Table 1). Significant differences in LOTUS concentrations were observed between the acute and recovery phases regardless of whether the meningitis was viral or bacterial. The CSF levels of LOTUS significantly decreased in the acute phase of viral and bacterial meningitis, but returned to the normal level in the recovery phase (Fig. 3b). Individual analysis also revealed an increase in LOTUS concentration in all patients with both viral and bacterial meningitis (Fig. 3c).

Finally, we asked whether neuroinflammation downregulates Lotus mRNA expression in the brain, using LPS-injected mice as a model. Lotus mRNA expression was downregulated in LPS-inflamed brains, followed by a decrease in LOTUS protein levels (Fig. 4).

Myelin components in the CNS are important environmental factors that exert an inhibitory effect on axonal growth. Nogo, myelin-associated glycoprotein (MAG), and oligodendrocyte myelin glycoprotein (OMgp) are representative myelin components [21]. These molecules bind to a common receptor, NgR1, and prevent axonal growth, thereby leading to neuronal degeneration [22]. Because LOTUS endogenously functions as an NgR1 antagonist, preventing Nogo from binding to NgR1, it can promote axonal growth and is thus a promising therapeutic target for neural regeneration [6]. These molecules have attracted attention due to their relationship with MS pathogenesis and also as therapeutic targets. In particular, recent studies have demonstrated that the activation of NgR1-mediated signaling plays a substantial role in axonal degeneration in MS. Furthermore, a previous study by our group showed that LOTUS levels in the CSF correlated inversely with disease activity of MS, suggesting that its reduction during the active phase may be involved in the pathogenesis of MS. Interestingly, a recent study suggested that Nogo and NgR1, as well as other molecules, are related not only to axonal guidance but also to the immune system. This observation demonstrated that lymphocytes express NgR1 and that Nogo–NgR1 signaling can alter lymphocyte phenotypes [17]. Accordingly, we hypothesized that LOTUS is also involved in the immune system and thus represents a candidate novel biomarker for disease activity associated with neuroinflammation. In this study, we elucidated the correlation between the CSF LOTUS level and neuroinflammation in meningitis, one of the most common neuroinflammatory diseases. Our results indicated that, similar to Nogo, LOTUS is involved in the immune system as well as playing a role in axonal guidance.

Currently, total protein concentration and cell counts in the CSF are widely used as biomarkers to evaluate disease activity, although they are not necessarily appropriate for monitoring neuroinflammation in most CNS diseases. For evaluation of meningitis, these conventional markers are well established; therefore, we first confirmed improvement of meningitis in reference to these CSF biomarkers, in addition to the clinical neurological findings. Then, we chronologically compared the CSF LOTUS concentration between acute and recovery phases. We observed that the LOTUS level was reduced in acute phase, followed by recovery to the healthy control level, accompanied by clinical improvement. These data demonstrated that the CSF LOTUS concentration is inversely correlated with the disease activity of meningitis, as we previously reported in MS.

We then considered two possible mechanisms for a decrease in LOTUS: (1) downregulation of Lotus mRNA expression by neuroinflammation, and (2) enhanced degradation of LOTUS protein. Accordingly, we investigated whether Lotus mRNA levels decreased with the severity of neuroinflammation in LPS-injected mice, which is widely used as an animal model for neuroinflammation [23‐25]. As expected, LPS-treated mice exhibited a significant downregulation of Lotus mRNA, along with a reduced level of LOTUS protein, in the brain. These results indicated that the mechanism underlying reduced CSF LOTUS concentration is downregulation of Lotus mRNA expression in response to neuroinflammation.

Previous proteomic studies reported LOTUS in CSF from normal subjects [26], and another group reported that LOTUS was not detectable in the CSF of patients with viral meningitis and CNS sarcoidosis [27]. Indeed, we confirmed a reduction in CSF concentration of LOTUS in acute phase of three patients with CNS sarcoidosis (61.6 ± 21.1 ng/ml). Thus, these findings support our results that reduced LOTUS concentration is associated with neuroinflammation in the acute phase of the diseases.

Interleukin-6 (IL-6) concentration in the CSF is often used as a biomarker of neuroinflammation in bacterial meningitis and progressive neuro-Behçet’s disease [28‐31]. However, the sensitivity of IL-6 is too low to estimate disease activity in non-progressive neuro-Behçet’s disease or MS [32‐34], indicating that IL-6 is of limited utility as a biomarker of neuroinflammation.

In contrast to IL-6, LOTUS may reflect disease activity of wider range of CNS diseases such as viral and bacterial meningitis, CNS sarcoidosis, and MS. Thus, the CSF LOTUS level is potentially of great diagnostic value for estimating neuroinflammation in a variety of CNS diseases, and we hypothesize that LOTUS might be critical to the pathogenesis of these diseases.

CSF levels of LOTUS correlated inversely with disease activity both in bacterial and viral meningitis, as in multiple sclerosis. Therefore, variation of CSF LOTUS is associated with neuroinflammation and is useful as a biomarker, leading accurate diagnosis, appropriate therapy, and comprehension of pathogenesis for a broader range of neuroinflammatory diseases.