The Promise of Niacin in Neurology

Abstract

Niacin (vitamin B3) is an essential nutrient that treats pellagra, and prior to the advent of statins, niacin was commonly used to counter dyslipidemia. Recent evidence has posited niacin as a promising therapeutic for several neurological disorders. In this review, we discuss the biochemistry of niacin, including its homeostatic roles in NAD⁺ supplementation and metabolism. Niacin also has roles outside of metabolism, largely through engaging hydroxycarboxylic acid receptor 2 (Hcar2). These receptor-mediated activities of niacin include regulation of immune responses, phagocytosis of myelin debris after demyelination or of amyloid beta in models of Alzheimer’s disease, and cholesterol efflux from cells. We describe the neurological disorders in which niacin has been investigated or has been proposed as a candidate medication. These are multiple sclerosis, Alzheimer’s disease, Parkinson’s disease, glioblastoma and amyotrophic lateral sclerosis. Finally, we explore the proposed mechanisms through which niacin may ameliorate neuropathology. While several questions remain, the prospect of niacin as a therapeutic to alleviate neurological impairment is promising.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13311-023-01376-2.

Introduction

Niacin, also known as vitamin B₃, is an essential nutrient obtained through dietary intake, where rich food sources include meat, fish, grains, and vegetables [1]. The importance of maintaining proper niacin levels is clearly established, as niacin-deficient individuals develop pellagra, a disease characterized by dementia, dermatitis, diarrhea, and, ultimately, death [2]. Niacin is thus a medication that is used to treat pellagra [3], and it was commonly indicated for dyslipidemia [4] prior to the advent of statins. In humans, the recommended minimum intake of niacin is between 15 and 20 mg/day [5], while pharmacological doses up to 3000 mg/day have been administered for dyslipidemia, demonstrating its tolerability across a broad range of doses [6].

Recent evidence has posited niacin as an exciting therapeutic option for a range of neurological disorders. Ranked as the third most promising repurposed drug candidate for progressive multiple sclerosis (MS) [7], niacin promotes phagocytosis of inhibitory myelin debris following demyelination in an animal model of MS, leading to remyelination [8]. In animal models of Parkinson’s [9] and Alzheimer’s disease [10], niacin ameliorates neuropathology through mechanisms such as immunomodulation and dopamine supplementation. Furthermore, niacin reduces tumour size and mortality in an animal model of glioblastoma [11] and is currently being investigated in a clinical trial of patients with glioblastoma [12]. Niacin also alleviates motor symptoms in patients with Parkinson’s disease [13], and dietary intake of niacin is associated with reduced incidence of Alzheimer’s disease and cognitive decline [14].

Despite the promise of niacin in neurological diseases, several unanswered questions remain regarding the actions of niacin in the central nervous system (CNS); these include the mechanisms of neuroprotection by niacin and whether it is a pro- or anti-inflammatory agent. In this review, we discuss the biochemistry and activity of niacin, and its receptor-dependent and -independent activities. We then consider the mechanisms of action of niacin within the CNS and explore its potential role as a therapeutic for neurological diseases.

Biology and Chemistry of Niacin

Niacin exists as the vitamers nicotinic acid and nicotinamide, which can give rise to the niacin derivatives nicotinamide riboside and nicotinamide mononucleotide. Niacin generates nicotinamide adenine dinucleotide (NAD⁺) through a series of metabolic pathways summarized in Fig. 1. In the Preiss-Handler pathway, nicotinic acid is converted into NAD⁺ in three steps, utilizing the intermediates nicotinic acid mononucleotide and nicotinic acid adenine dinucleotide [15]. The salvage pathway recycles nicotinamide, the by-product of enzymatic activities of NAD⁺, and dietary nicotinamide riboside to generate NAD⁺ [16]. Lastly, de novo biosynthesis of NAD⁺ is accomplished via the kynurenine pathway (Fig. 1), where dietary tryptophan serves as a precursor and is converted to NAD⁺ through a series of eight enzymatic steps; notably, this is the only NAD⁺ synthesis pathway that operates independently of niacin [17]. The de novo pathway is the longest and most energy-intensive; as a result, tryptophan is less efficient at increasing NAD⁺ levels compared to other precursors, and this pathway plays a modest role in NAD⁺ production [18, 19]. In contrast, the salvage pathway is the chief producer of NAD⁺ in mammalian cells and is largely responsible for maintaining homeostatic levels of this metabolite [18].

Fig. 1.

NAD⁺ biosynthesis pathways. In the kynurenine pathway, dietary tryptophan is first converted to N-formylkynurenine via tryptophan 2,3-dioxygenase (TDO) or indoleamine 2,3-dioxygenase (IDO). Through a series of four enzymatic steps, N-formylkynurenine generates quinolinic acid, which gives rise to nicotinic acid mononucleotide in a reaction catalyzed by quinolinic acid phosphoribosyl transferase (QPRT). In the final steps, nicotinic acid mononucleotide is converted to nicotinic acid adenine dinucleotide, which generates NAD⁺. In the Preiss-Handler pathway, dietary nicotinic acid (niacin) is converted to nicotinic acid mononucleotide via nicotinate phosphoribosyltransferase (NAPRT). Nicotinic acid mononucleotide is then converted to nicotinic acid adenine dinucleotide in a nicotinamide mononucleotide adenylyl transferase (NMNAT)-catalyzed reaction, and this gives rise to NAD⁺ via NAD⁺ synthase (NADS). In the salvage pathway, nicotinamide that has been recycled from the enzymatic activities of NAD⁺ is used to generate nicotinamide mononucleotide via nicotinamide phosphoribosyltransferase (NAMPT). Dietary nicotinamide riboside can produce either nicotinamide mononucleotide, or nicotinamide. In the final step of this pathway, nicotinamide mononucleotide gives rise to NAD⁺. Once generated, NAD⁺ is consumed by several enzymes, including sirtuins (SIRTs), poly(ADP-ribose) polymerases (PARPs), and sterile alpha and TIR motif-containing 1 (SARM1), as well as the cyclic ADP-ribose (cADPR) synthases CD38 and CD157. These enzymes generate nicotinamide as a by-product. Figure created using BioRender

Once generated, NAD⁺ serves as a precursor for its phosphorylated form, nicotinamide adenine dinucleotide phosphate (NADP), which is generated via NAD⁺ kinases [20, 21]. NAD(P) homeostasis involves a balance between biosynthesis and use by NAD⁺-consuming enzymes. NAD⁺ and NADP are critical coenzymes for oxidoreductases, and NAD⁺ also serves as a substrate for redox-independent enzymatic processes in the cell [22]. Thus, NAD(P) homeostasis is essential to the proper metabolic functioning of a cell. In the context of oxidation–reduction reactions such as glycolysis and oxidative phosphorylation, NAD⁺ serves as a proton acceptor, generating its reduced form NADH. NADH is then oxidized in the electron transport chain, contributing to the mitochondrial proton gradient and allowing for the generation of ATP via ATP synthase [23].

NAD⁺ also has redox-independent functions, serving as a cosubstrate for several important enzymes, which catabolize NAD⁺ and generate nicotinamide as a by-product (Fig. 1) [24]. These enzymes include sirtuins (SIRTs), poly(ADP-ribose) polymerases (PARPs), sterile alpha, and TIR motif-containing 1 (SARM1) and cyclic ADP-ribose (cADPR) synthases such as CD38 and CD157 [25]. Sirtuins are a family of protein deacetylases involved in processes such as cell metabolism, inflammation, and oxidative stress [26]. During a sirtuin-catalyzed deacetylation, NAD⁺ is cleaved into nicotinamide and ADP-ribose (ADPR). ADPR serves as an acyl acceptor, allowing for the removal of an acyl group from the substrate [27]. PARPs catalyze the transfer of multiple ADP-ribose groups from NAD⁺ onto target macromolecules in a process termed poly-ADP-ribosylation [28]. PARP activity is upregulated following DNA damage, leading to substantial decreases in total NAD⁺ levels and initiation of DNA repair signalling [29]. In response to neuronal damage or injury, SARM1 is activated, catalyzing the cleavage of NAD⁺ to nicotinamide, ADPR, and cADPR via its Tol/interleukin-1 receptor (TIR) motif; this in turn promotes axon degeneration [30–32]. Finally, CD38 and CD157 hydrolyze NAD⁺ to nicotinamide, ADPR, and cADPR, the latter of which serves as a second messenger in Ca²⁺ signalling [33, 34]. The activity of these enzymes is dependent on NAD⁺ levels. Thus, a disruption in NAD⁺ biosynthesis or a lack of NAD⁺ precursors such as niacin can lead to dysregulated cellular activities.

NADP plays an essential role in the pentose phosphate pathway (Fig. 2). Here, the reduction of NADP to NADPH is coupled to the synthesis of ribose-5-phosphate which serves as a precursor to many biological molecules including DNA and RNA. NADPH is then used as a reducing agent in the generation of molecules such as fatty acids, sterols, and nucleotides [35]. NADPH is also involved in the balance of oxidative stress, serving as a cofactor for glutathione reductase in the production of the antioxidant glutathione [36], and acting as a substrate for NADPH oxidases, donating an electron in the generation of reactive oxygen species [37, 38]. Thus, through supplementation of NAD(P), niacin restores proper metabolic functions (Fig. 2).

Fig. 2.

Homeostatic roles of niacin as a precursor to NAD⁺. Through the activity of NAD⁺-consuming enzymes, niacin is involved in the maintenance of cellular processes such as the DNA damage response and Ca²⁺ signalling. NAD⁺ is reduced to form NADH, which serves as a proton donor in the electron transport chain, generating the mitochondrial proton gradient and leading to the production of ATP. NAD⁺ is also phosphorylated to generate NADP. NADP serves as a precursor for ribose-5-phosphate, which gives rise to nucleic acids such as DNA and RNA. Finally, NADP is reduced to generate NADPH. NADPH is then used as a reducing agent in the generation of biological molecules such as fatty acids, sterols, and nucleotides. Figure created using BioRender

At endogenous levels, niacin works through supplementation of NAD⁺ and NADP in a receptor-independent manner, restoring proper metabolic function. In a rodent model of abdominal aortic aneurysm, nicotinic acid and nicotinamide reduce aneurysm incidence and pathology by increasing NAD⁺ levels and Sirtuin1 activity, rather than by activating a principal receptor for niacin, the hydroxycarboxylic acid receptor (Hcar2) [39]. In addition, patients with mitochondrial myopathy experience muscle weakness and fatigue, as well as reduced NAD⁺ levels likely caused by altered mitochondrial activity. Niacin supplementation in these patients increases their systemic and muscle levels of NAD⁺ and leads to increased muscle strength and improved mitochondrial biogenesis [40]. Further, decreased retinal levels of NAD⁺ have been observed with age and are associated with retinal pathologies such as glaucoma [41]. In a rodent model of glaucoma, prevention of NAD⁺ decline by supplementation with nicotinamide prevents retinal ganglion cell neurodegeneration [42] while also protecting against glaucoma-induced metabolic changes and preserving integrity of mitochondrial morphology [43].

Of note, it is unlikely that niacin acts through NAD⁺ supplementation in the brain, as levels of nicotinamide phosphoribosyltransferase (NAMPT) and nicotinate phosphoribosyltransferase (NAPRT), enzymes required for the synthesis of NAD⁺ from nicotinic acid, are much lower in the brain compared to other tissues [44, 45]. As a result, NAD⁺ in the CNS is likely derived from other sources, such as the kynurenine metabolism pathway [44]. Thus, while NAD⁺ supplementation may be the canonical role of niacin, there are other important pathways that mediate its effects in the body, particularly in the CNS.

Another notable biochemical feature of niacin is its rapid elimination from the body; maximum plasma concentrations are reported 30–60 min after oral intake, and it has an estimated half-life between 28 and 40 min in both humans and mice [46–48].

Niacin Receptors

Hydroxycarboxylic Acid Receptor (Hcar)2

The primary receptor for niacin is the inhibitory G_i/o protein coupled receptor (GPCR) Hcar2, also known as GPR109A [49, 50]. As a GPCR, Hcar2 is characterized by a seven-transmembrane domain structure with an extracellular N terminus that interacts with ligands and an intracellular C terminus that engages in signal transduction via interaction with heterotrimeric G proteins [51]. Once bound by an extracellular ligand such as niacin, Hcar2 undergoes a conformational change and binds to the G_i/o protein alpha subunit. In adipocytes, this stimulates a downstream signalling cascade which inhibits adenylyl cyclase activity, as well as reduces intracellular cyclic AMP levels [52, 53]. In immune cells, ligand binding to Hcar2 leads to a transient increase in intracellular Ca²⁺ (Fig. 3) [54].

Fig. 3.

Activity of niacin at Hcar2 in adipocytes and immune cells. In adipocytes, niacin acts through hydroxycarboxylic acid receptor (Hcar2) to inhibit adenylyl cyclase activity. Under normal conditions, adenylyl cyclase generates cAMP, which activates protein kinase A. Protein kinase A phosphorylates and activates hormone-sensitive lipase, which increases lipolysis. By inhibiting adenylyl cyclase, niacin reduces activity of this pathway and leads to suppression of lipolysis. In immune cells, niacin binding to Hcar2 leads to an increase in intracellular Ca^2+. Although the precise mechanism has yet to be elucidated, one model suggests that Hcar2 agonism activates phospholipase C (PLC), promoting the release of Ca²⁺ from intracellular stores within the endoplasmic reticulum. Ca²⁺ then acts as a second messenger, inhibiting the phosphorylation of p65 which is downstream from inflammatory NF-κB activation. Conversely, it is thought that the transient increase in Ca²⁺ could originate from extracellular sources. This would stabilize the intracellular Ca²⁺ stores found within the endoplasmic reticulum, making the cell more resistant to stress. Cellular stress leads to the activation of the NLRP3 inflammasome, which promotes cholesterol accumulation within macrophages and leads to a proinflammatory, detrimental immune cell phenotype. In the figure, the step of inhibition of inflammatory activity in immune cells subsequent to niacin/Hcar2 interaction is depicted by the red T sign. Figure created using BioRender

The primary endogenous ligands of Hcar2 are the short-chain fatty acid butyrate and the ketone body β-hydroxybutyrate [55–57]. At pharmacological doses, nicotinic acid, but not nicotinamide, serves as a ligand for Hcar2 [58, 59]. Of note, monomethyl fumarate, a metabolite of the MS medication dimethyl fumarate, is also able to stimulate the Hcar2 receptor, and this mediates at least part of its therapeutic response in a model of MS [60–62].

Hcar2 is widely expressed throughout the body, which likely explains the broad range of physiological and pharmacological effects that Hcar2 ligands such as niacin can have. On adipocytes of white and brown adipose tissue, Hcar2 acts as a metabolite sensor, inhibiting lipolysis in times of fasting and contributing to the lipid-modifying effects of niacin [58, 63]. Hcar2 is also expressed on immune cells such as macrophages, dendritic cells, and neutrophils, underlying the immunomodulatory properties of niacin (Fig. 3) [49, 64]. In addition, Hcar2 is expressed on epithelial cells of the retina, colon, and apical intestine, as well as keratinocytes of the skin [55, 65]. Activation of Hcar2 on the latter cell type leads to release of prostaglandins, which partially mediates the flushing response associated with niacin supplementation [66].

While endogenous levels of nicotinic acid (< 0.3 nM) [67] are too low to significantly impact the activity of Hcar2 and likely act through NAD⁺ supplementation, pharmacological doses are sufficient to activate Hcar2 and mediate a separate range of biological activities [68]. Indeed, humans taking 1 g of niacin for dyslipidemia have an average serum concentration of 120–230 μM of nicotinic acid, 1–2 h after intake [69]. Experimental studies have shown that the EC₅₀ for nicotinic acid on Hcar2 in humans is between 0.13 and 1 μM, with a maximal response achieved between 10 and 100 μM [50, 70, 71]. This demonstrates that the plasma levels of niacin obtained following pharmacological niacin supplementation are within the optimal window for Hcar2 activation.

Hcar2 Agonism

As aforementioned, Hcar2 is widely expressed on adipocytes. During periods of fasting or exercise, Hcar2 on adipocytes is activated by ketone bodies such as β-hydroxybutyrate, one of its primary endogenous ligands [72]. This activation leads to the inhibition of adenylyl cyclase, a concomitant reduction in cyclic AMP levels and protein kinase A activity, and ultimately decreased activity of hormone sensitive lipase, an enzyme responsible for lipolysis [73, 74]. In adipocytes cultured from wild-type mice, administration of β-hydroxybutyrate or nicotinic acid reduces the amount of free fatty acids present in the cell culture medium, indicating reduced lipolysis. Of note, this result is abolished in adipocytes from PUMA-G knockout mice, the rodent ortholog of Hcar2 [75], a finding that has been reported elsewhere [70]. In transgenic rats overexpressing human Hcar2, plasma free fatty acid levels are reduced during fasting, compared to wild-type controls [76]. Furthermore, nicotinic acid administration more strongly decreases free fatty acid levels in these transgenic mice, when compared to wild-type, indicating that niacin works through Hcar2 to decrease lipolysis (Fig. 3) [76]. As such, Hcar2 suppresses lipolysis during periods of fasting, reducing the levels of free fatty acids mobilized in the blood stream. This activity underlies the lipid-modifying aspects of niacin that have led to its use in the treatment of dyslipidemia.

The other primary cell type known to express Hcar2 is immune cells, including macrophages, microglia, dendritic cells, and neutrophils. While precise mechanisms remain unclear, there are two main hypotheses regarding the effects of Hcar2 agonism in immune cells. In a model of atherosclerosis, activation of Hcar2 on macrophages by nicotinic acid leads to an increase in intracellular Ca²⁺ from extracellular sources, inhibiting the phospholipase C (PLC)/inositol trisphosphate (IP₃) pathway that leads to the release of Ca²⁺ from the endoplasmic reticulum [77]. Notably, inhibition of PLC by Hcar2 is Ca²⁺ dependent, as PLC inhibition is lost following the application of an intracellular Ca²⁺ chelator [77]. Thus, it is hypothesized that the stabilization of intracellular Ca²⁺ stores prevents cellular stress and inflammasome activation (Fig. 3).

In contrast to the above hypothesis, work in the EAE model suggests that agonism of Hcar2 leads to a G_i/o signalling cascade in which the PLC pathway is activated. This leads to an elevation in intracellular Ca²⁺ from intracellular sources, preventing the hyperphosphorylation of p65 and downstream activation of NF-κb, a transcription factor that regulates the expression of many pro-inflammatory cytokines (Fig. 3) [60, 61]. Thus, while it is clear that agonism of Hcar2 by niacin or other compounds leads to an increase in intracellular Ca²⁺ and a reduction in inflammation, the specific pathways remain to be fully elucidated.

Expression of Hcar2 on immune cells contributes to the anti-atherogenic and immune modulatory properties of niacin. Indeed, while it was previously thought that the antiatherogenic properties of niacin were derived from its antilipolytic effects, where reduced plasma levels of free fatty acids would lead to decreased synthesis of VLDL and LDL, with an increase in HDL [78–80], current research shows that the antiatherogenic properties of niacin instead involve Hcar2 activation on immune cells. Indeed, in the apolipoprotein E knockout model of atherogenesis, activation of Hcar2 on macrophages promotes cholesterol efflux and reduces atherosclerotic plaque burden [77].

Agonism of Hcar2 expressed on immune cells also serves to modulate inflammation. In LPS-activated human monocytes and mouse macrophages, niacin treatment reduces the expression of pro-inflammatory cytokines such as TNFα and IL-6, an effect that is lost following knockdown of Hcar2 [81, 82]. This reduction in cytokine release likely occurs due to inhibited NF-κB signalling, as intermediates of the NF-κB pathway such as phosphorylated p-65 and phosphorylated NF-κB are decreased upon Hcar2 agonism [81, 82]. Although activation of Hcar2 by niacin has previously been described as primarily anti-inflammatory, the recent work demonstrates that niacin acts to promote a beneficial phenotype typified by altered cytokine production and enhanced phagocytosis, rather than simply suppressing immune activity. In support of this view, studies in several models of neurological disease have shown that niacin can enhance the activity of microglia/macrophages, promoting a beneficial phenotype characterized by increased phagocytosis of pathological particles and reduced pathology [8, 10, 11]. Furthermore, in a model of stroke, niacin treatment activates a population of Hcar2-expressing neuroprotective macrophages, which infiltrate into the ischemic brain and ameliorate neuropathology [83].

Hcar3

Another receptor that niacin has been suggested to act through is Hcar3, also known as GPR109B or HM74. Hcar3 is also primarily expressed on adipocytes and immune cells, where it regulates fatty acid metabolism and inflammation through its primary endogenous ligands 3-hydroxyoctanoic acid and kynurenic acid [84]. Similar to Hcar2, Hcar3 is a G_i/o-coupled GPCR that exerts its effects through downstream pathways involving modulation of adenylyl cyclase, cyclic AMP and intracellular Ca²⁺ levels [73, 85]. Although earlier studies reported Hcar3 as a second niacin receptor [86, 87], more recent work suggests that despite high homology with Hcar2, Hcar3 displays low affinity binding to niacin [88]. Indeed, Hcar3 binds niacin with approximately 1000-fold less affinity as compared to Hcar2 [50, 71]. As a result, the actions of niacin exerted at this receptor are minimal compared to that of Hcar2.

Transient Receptor Potential Cation Channel Subfamily V Member 1 (Trpv1)

Niacin has also been shown to bind to and activate the capsaicin receptor Trpv1, a nonselective cation channel [89]. Trpv1 is widely expressed throughout the body, and located in neurons, sensory cells of the gastrointestinal tract, immune cells, epithelial cells, and others [90]. As a heat-sensing receptor, the activation of Trpv1 leads to thermoregulatory processes including vasodilation. In mice, treatment with nicotinic acid leads to cutaneous vasodilation, as detected by Doppler perfusion imaging. This vasodilation was ameliorated in Trpv1 knockout mice [91] and following antagonism of the Trpv1 receptor [92]. Together, these results suggest that activation of Trpv1 by niacin is partly responsible for the flushing side effect, commonly reported in individuals taking niacin supplements.

Mechanisms and Applications of Niacin in Neurological Disease

Niacin and the Blood Brain Barrier

Before a medication can exert a direct pharmacological effect in the CNS, it must first cross the blood brain barrier, a quality of niacin that has been demonstrated. Indeed, cerebral uptake of nicotinic acid and nicotinamide is detected by positron emission tomography imaging following intravenous administration in both healthy volunteers and patients with neurodegenerative disease [93]. Furthermore, a significant increase in brain nicotinic acid levels is observed in mice following oral administration [10].

Multiple Sclerosis

MS is a chronic inflammatory disease of the CNS, characterized by demyelination, inflammation, and progressive neurodegeneration. In MS patients, remyelination is associated with reduced score of clinical disability [94], while in a model of MS, remyelination facilitates functional recovery [95]. Current treatments for MS focus on modulating inflammation to prevent relapses, but none of the approved therapeutics target remyelination [96]. Demyelination in MS leads to the generation of myelin debris, which inhibits remyelination by preventing maturation of oligodendrocyte progenitor cells [97]. As a result, increased phagocytic capacity of microglia and infiltrating macrophages, allowing for enhanced clearance of inhibitory myelin debris, would be a promising therapeutic strategy in MS.

Cholesterol is a major component of myelin and cannot be broken down by most mammalian cells [98]. Following the phagocytosis of lipid-rich myelin debris by microglia or macrophages, cholesterol undergoes reverse cholesterol transport or is esterified which is then stored in lipid droplets or released as lipoproteins [99]; these mechanisms protect the cell from the toxic effects of free cholesterol [100, 101]. In the CNS, cholesterol homeostasis is regulated by liver X receptors (LXRs), which form obligate heterodimers with retinoid X receptors [102]. Cholesterol derivatives such as oxysterols and desmosterol serve as ligands for LXRs, providing an indicator for the level of intracellular cholesterol [103]. Upon oxysterol binding, these receptors act as transcription factors, promoting the expression of apolipoprotein E (ApoE), as well as ATP binding cassette subfamily members ABCA1 and ABCG1 [103, 104]. ABCA1, and to a lesser extent, ABCG1, mediate the transport of cholesterol to the extracellular acceptor ApoE, leading to the generation of HDL-like particles that disperse lipids throughout the CNS [104, 105]. Once effluxed from macrophages, cholesterol may be taken up by oligodendrocytes and recycled for use in the creation of new myelin, thus facilitating remyelination (Fig. 4) [106].

Fig. 4.

Cholesterol recycling in the CNS and impact of niacin. Demyelination often occurs as a result of CNS insult or injury, producing myelin debris which is phagocytosed by microglia/macrophages in the CNS. This uptake of debris is promoted by niacin [8]. Following phagocytosis, myelin debris is partially degraded in the lysosome. Cholesterol, which cannot be broken down, is either esterified for storage in lipid droplets, or effluxed out of the cell. Impaired cholesterol processing leads to sustained cholesterol accumulation and the formation of cholesterol crystals, which promote an inflammatory macrophage phenotype. Cholesterol efflux is mediated by the ABCA1 and ABCG1 transporters, which transfer free cholesterol onto lipid-poor ApoE particles. There is evidence that niacin promotes the mRNA level of ABCA1 and ABCG1 [107, 108] and cholesterol efflux [106] although whether the latter is due to passive diffusion or through an ABCA1-dependent mechanism is unresolved. Together, cholesterol and ApoE generate an HDL-like particle, which distributes cholesterol throughout the CNS. Oligodendrocytes are one cell type that receives this free cholesterol, using it in the generation of new myelin. In the nucleus, cholesterol derivatives (e.g., oxysterol) bind to LXR. LXR forms a heterodimer with RXR and serves as a transcription factor, promoting transcription of ApoE, ABCA1 and ABCG1. Figure created using BioRender

Niacin modulates cholesterol recycling and efflux from macrophages. In vitro, HDL particles collected from healthy, niacin-treated humans promote cholesterol efflux from foamy macrophages in an ABCA1- and ABCG1-dependent manner [107]. Further, niacin treatment in ApoE knockout mice increases mRNA expression of Abca1, Abcg1, and LXR-α [108]. Incubation of human monocytes with niacin increases mRNA expression of ABCA1 and promotes cholesterol efflux to lipid-poor HDL particles [109]. In humans with a history of cardiovascular disease, a combination therapy of niacin and atorvastatin increases macrophage cholesterol efflux capacity beyond the levels observed in the statin-only group, although there was no observed increase in ABCA1-specific cholesterol efflux in this study [110]. Lastly, an in vitro experiment examining the effect of extended-release niacin found that while it increases passive diffusion-mediated cholesterol efflux capacity, there was no significant effect on ABCA1-mediated efflux [111]. Together, these studies demonstrate that while niacin promotes cholesterol efflux from macrophages, it remains to be determined whether this is done in an ABCA1-dependent or ABCA1-independent manner, or both.

Impaired cholesterol recycling and efflux can have negative consequences for the function and phenotype of macrophages. For instance, impaired cholesterol recycling and the concomitant production of intracellular cholesterol crystals lead to NLRP3 inflammasome activation [112, 113]. In the lysolecithin model of MS, impaired intracellular cholesterol esterification results in decreased lipid droplet formation and defective remyelination [114]. In addition, impaired cholesterol efflux and sustained myelin fragment accumulation within phagocytes produces an inflammatory foamy macrophage phenotype [115] which may lose their ability to effectively phagocytose debris [116]. Of note, dysregulated lipid metabolism has been implicated in the pathology of MS [117]. For example, LXR knockout mice have reduced CNS myelination [118], and serum levels of low-density lipoprotein cholesterol are positively associated with disease activity in MS patients, as indicated by MRI [119].

Cholesterol homeostasis in macrophages is related to their phagocytic capacity, as lipid-laden macrophages are unable to continue effectively phagocytosing debris and may eventually undergo apoptosis [116, 120]. Thus, it is thought that niacin works to promote remyelination in the MS context through increased phagocytosis of inhibitory myelin debris, perhaps linked to niacin’s ability to promote cholesterol efflux [107]. By enhancing cholesterol efflux that may in turn be consumed by cholesterol-rich membrane-forming regenerating oligodendrocytes, niacin would prevent prolonged lipid accumulation within macrophages and free up their phagolysosomes for further digestion of engulfed material, while also avoiding a detrimental inflammatory macrophage phenotype (Fig. 4).

Promising data has been obtained from preclinical studies, demonstrating that niacin ameliorates disease in models of MS [121]. In the lysolecithin demyelination model, niacin treatment promotes phagocytosis of inhibitory myelin debris by microglia/macrophages, leading to increased numbers of oligodendrocyte progenitor cells in the lesion site and improved remyelination [8]. Of note, this enhanced phagocytosis is thought to be due in part to increased expression of the scavenger receptor CD36 on microglia/macrophages [8]. EAE mice treated with niacin experience ameliorated neurological deficits, reduced number of inflammatory infiltrates, and enhanced neuroprotection [122]. Furthermore, young rats fed a nicotinic acid-rich diet have enhanced developmental myelination, when compared to controls [123]. As a result of this promising evidence, a recent study aiming to identify licensed drugs with potential to be repurposed for clinical trials involving progressive MS patients put forth niacin as one of three highly recommended candidate drugs [7]. Future studies will allow us to determine whether niacin has therapeutic effects in people with MS, given its potential remyelinating and immunomodulatory responses (Fig. 5).

Fig. 5.

Mechanisms of niacin in neurological disease. Niacin may act through a variety of mechanisms to alleviate pathology in neurological and neurodegenerative diseases. These putative mechanisms based on preclinical studies include enhanced phagocytosis and lipid recycling, immunomodulation, and ameliorated oxidative stress. Figure created using BioRender

Parkinson’s Disease

Parkinson’s disease (PD) is a neurodegenerative disorder primarily caused by a loss of dopaminergic neurons in the substantia nigra pars compactica [124]. This neuropathology leads to dopamine deficiency in the basal ganglia and manifests in patients as bradykinesia, tremor, and/or rigidity, as well as other nonmotor symptoms such as cognitive decline and depression [125]. The first line of treatment for patients with PD is levodopa, a precursor of dopamine. Although levodopa is effective at treating motor symptoms, it does not prevent neurodegeneration and does not impact underlying inflammation; therefore, there is a need for improved therapeutic options [126].

There are several mechanisms through which niacin may act to ameliorate PD pathology, the first of which is through immune modulation. Aberrant neuroinflammation is increasingly being recognized as a key driver of PD pathogenesis. Activated microglia are enriched in the post-mortem brains of people with PD, particularly in affected areas such as the substantia nigra [127, 128]. Furthermore, there is increased signalling of proinflammatory cytokines such as IL-6, IFN-γ, and IL-1β in the CSF and peripheral blood of PD patients compared to healthy controls [129]. As such, it may be beneficial to modulate immune activity in the context of PD, which niacin has been shown to do. Macrophages collected from niacin-treated PD patients show polarization towards a beneficial, rather than detrimental inflammatory phenotype [130]. In the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) model of PD, treatment with the niacin metabolite NADPH leads to reduced gliosis and inflammatory NF-κB activation [131].

Due to the loss of dopaminergic neurons, PD is characterized by decreased dopamine receptor signalling. Another mechanism through which niacin may modulate PD pathology is by supplementing dopamine levels. As mentioned, niacin gives rise to NAD⁺ and NADH, which are involved in the production of dopamine [132]. Indeed, incubation of rodent cells with NADH leads to increased dopamine biosynthesis via elevated recycling of quinonoid dihydrobiopterin to tetrahydrobiopterin [133]. Tetrahydrobiopterin is a cofactor necessary for the synthesis of tyrosine hydroxylase, a rate-limiting enzyme in dopamine production [134, 135]. Preclinical data has shown support for this mechanism of action, as niacin treatment in the rotenone model of PD increases brain dopamine levels, compared to vehicle controls [9]. In striatal brain slices, application of NADH leads to increased KCl-evoked dopamine release [136].

Oxidative stress is increasingly being recognized as a key driver of neurodegeneration in PD, possibly underlying neuronal loss [137]. Markers of oxidative stress are observed in the blood of PD patients [138], and genetic disruption of mitochondrial complex I in the dopaminergic neurons of rodents leads to progressive, human-like parkinsonism [139]. In models of PD, niacin has been shown to ameliorate PD-related oxidative stress. Indeed, niacin treatment in the rotenone model reduces malondialdehyde level that informs on oxidative stress and increases the levels of antioxidants glutathione and superoxide dismutase [9]. In the MPTP model, intraperitoneal administration of NADPH restores glutathione levels and reduces reactive oxygen species production [131].

In sum, it is likely that niacin acts through several neuroprotective mechanisms to modify the pathology of PD, including altered inflammatory state, dopamine supplementation, and reduced oxidative stress (Fig. 5). Preliminary clinical data shows that niacin is a promising treatment in people with PD, and several clinical trials are ongoing (Table 1). Intake of niacin-rich foods is negatively associated with the risk of developing PD [140, 141], and in people with PD, low-dose niacin supplementation over the course of 12 months improves the Unified Parkinson’s Disease Rating Scale, a measure of motor disability [13]. Furthermore, a case study demonstrated reduced rigidity and bradykinesia in a PD patient administered niacin for dyslipidemia [142].

Table 1.

Ongoing and completed trials of niacin in neurological disorders

Disorder	Trial details and status	Niacin formulation and oral dose	Primary outcome	Some secondary end points	Data
Glioblastoma	Phase I-II (NCT04677049); recruiting; planned study end date January 2026	Niacin CRT™. Phase I: Dose escalation study, in steps of 500 mg/d every 4 weeks up to 3 g/d Phase II: Maximally tolerated dose, as recommended from Phase I	Phase I: Maximally tolerated dose Phase II: Progression free survival at 6 months	Change in peripheral monocytes, QOL	Trial in progress, no available data
Parkinson’s disease	Phase II (NCT03462680); completed April 2020	Nicotinic acid; 250 mg/d	MDS-UPDRS, REM sleep pattern, sleep percentages, MMSE, Stroop test, fatigue	CSF and plasma cytokine changes, niacin levels in plasma and urine, Hcar2 expression, plasma serotonin levels	No significant difference in motor function [143]
Ischemic stroke	Phase II (NCT00796887); completed August 2012	Niaspan®; 500 or 1000 mg/d	Adverse events	Functional recovery	No significant differences in functional recovery. HDL cholesterol was significantly increased in niacin-treated group [144]
Parkinson’s disease	Phase II (NCT03808961); active, not recruiting; planned study end date April 2024	Nicotinic acid or nicotinamide; 100 mg tablets, twice daily	MDS-UPDRS, MMSE, Hcar2 expression, plasma cytokine levels, niacin levels in plasma and urine	VAFS, TMT, arm strength and fatigue	Trial in progress, no available data
Alzheimer’s disease or Mild cognitive impairment	Phase II (NCT03061474); active, not recruiting; planned study end date August 2022	Nicotinamide, sustained release tablet; 750 mg tablets, twice daily	Change in CSF p-tau231	Change in CSF p-tau 181 and total tau	Trial in progress, no available data
Alzheimer’s disease	Phase II (NCT00580931); completed July 2014	Endur-amide® (nicotinamide); 1500 mg tablets, twice daily	ADAS-Cog	CIBIC-Plus, ADCS-ADL, CDR	No significant difference in primary or secondary end points [145]
Parkinson’s disease	Phase I (NCT05589766); not yet recruiting; planned study end date December 2024	Niagen™ (nicotinamide riboside); dose escalation to a maximum of 3000 mg/d	Cerebral and CSF NAD levels, NRRP expression	Adverse events, QOL, NAD metabolite levels, MDS-NMS, MoCA, GIDS-PD, MDS-UPDRS	Trial in progress, no available data
Alzheimer’s disease	Phase I (NCT05617508); not yet recruiting; planned study end date December 2024	Niagen™ (nicotinamide riboside); dose escalation to a maximum of 3000 mg/d	Cerebral and CSF NAD levels, FDG-PET, 31P-MRS	Adverse events, ADAS-Cog, CDR-SB, MoCA, TMT, IADL, PSMS, NPI-Q, MADRS	Trial in progress, no available data
Parkinson’s disease	Phase II (NCT03568968); recruiting; planned study end date March 2024	Niagen™ (nicotinamide riboside); 1000 mg/d	MDS-UPDRS	NAD metabolite levels	Trial in progress, no available data
Parkinson’s disease	Phase I (NCT05344404); completed July 2022	Niagen™ (nicotinamide riboside); 1500 mg tablets, twice daily	Severe adverse events	Mild adverse events, NAD metabolome, MDS-UPDRS	Results not yet available
Alzheimer’s disease or Mild cognitive impairment	Early Phase I (NCT04430517); recruiting; planned study end date April 2025	Niagen™ (nicotinamide riboside); 250 mg tablets, four times per day	Brain NAD, Brain redox state	Mitochondrial function, GSH levels	Trial in progress, no available data
Mild cognitive impairment	Phase I (NCT02942888); completed August 2021	Niagen™ (nicotinamide riboside); dose escalation to a maximum of 1000 mg/d	MoCA	Cerebral blood flow, plasma NAD level, SPPB, IADL, arterial pressure, GDS, GAS, CLOX, EXIT, TAPS, Grip strength	Results not yet available
Mild cognitive impairment	Phase II (NCT04078178); planned study end date September 2022	Niagen™ (nicotinamide riboside); 1200 mg/d	RBANS	N/A	Trial in progress, no available data
Parkinson’s disease	Phase II (NCT03816020); completed February 2020	Nicotinamide riboside; 500 mg tablets, twice daily	PDRP changes	MDS-UPDRS	Significant increase in cerebral NAD levels and altered cerebral metabolism [146]

31P-MRS ³¹Phosphorous-magnetic resonance spectroscopy, ADAS-Cog Alzheimer’s Disease Assessment Scale-Cognitive Subscale, ADCS-ADL Alzheimer's Disease Cooperative Study-Activities of Daily Living Scale, CDR Clinical Dementia Rating Scale, CDR-SB Clinical Dementia Rating Scale Sum of Boxes, CIBIC-Plus Clinician's Interview-Based Impression of Change Plus Caregiver Input, CLOX Clock Drawing Task Protocol, CSF cerebrospinal fluid, EXIT Executive Interview, FDG-PET fluorodeoxyglucose-positron emission tomography, GAS Geriatric Anxiety Scale, GDS Geriatric Depression Scale, GSH glutathione, GIDS-PD Gastrointestinal Dysfunction Scale for Parkinson’s Disease, IADL Instrumental Activities of Daily Living, MADRS Montgomery-Asberg Depression Rating Scale, MDS-NMS Movement Disorder Society Non-Motor Rating Scale, MDS-UPDRS Movement Disorder Society Unified Parkinson’s Disease Rating Scale, MMSE Mini-Mental State Examination, MoCA Montreal Cognitive Assessment, NAD Nicotinamide adenine dinucleotide, NPI-Q Neuropsychiatric Inventory brief questionnaire form, NRRP Nicotinamide Riboside Related Pattern, PDRP Parkinson's disease related pattern, PSMS Physical Self-Maintenance Scale, RBANS Repeatable Battery for the Assessment of Neuropsychological Status, SPPB Short Physical Performance Battery, TAPS Test of Auditory Processing Skills, TMT Trail making test, QOL Quality of life, VAFS Visual analogue fatigue scale

Alzheimer’s Disease

Alzheimer’s disease (AD) is the most common form of dementia, and its prevalence and burden on the healthcare system will only increase as our population ages. The pathological hallmarks of AD include amyloid-β plaques and neurofibrillary tangles [147], and people with AD experience reduced cognition including memory, language, and executive functions [148]. Current treatment options for individuals with AD are limited and primarily focus on symptom management without affecting neuropathology [149].

The mechanisms through which niacin may act in AD have similarities to those employed in MS, particularly its regulation of liver X receptor expression and lipid recycling (Fig. 5). Altered lipid dynamics have been implicated in the pathogenesis of AD [150]. For instance, the ApoE gene is strongly implicated in AD pathology; while the ApoE ε4 allele is the strongest genetic risk factor for AD development, individuals carrying the ε2 allele experience protection from the disease [151]. Furthermore, neuronal cholesterol levels control amyloid-β production by regulating the cleavage of amyloid precursor protein [152, 153]. In addition, ABCA1- and ABCG1-mediated cholesterol efflux is impaired in the CSF of AD patients, when compared to healthy and non-AD dementia controls [154]. Thus, the lipid-modifying actions of niacin, where niacin promotes cholesterol efflux and regulates CNS cholesterol and lipid homeostasis (Fig. 4), may allow it to modulate AD pathogenesis.

In addition to its lipid-modifying properties, niacin alters AD pathology by promoting a rejuvenated microglia/macrophage phenotype, enhancing the phagocytosis of pathological amyloid-B plaques. In the 5xFAD transgenic mouse model of AD, treatment with slow-release niacin (niaspan^R) increases microglia engagement with plaques, reduces plaque number and area, and promotes expression of microglial genes related to phagocytosis [10]. This, in turn, has a positive effect on clinical aspects, reducing cognitive deficits [10].

Gene expression analysis has also been performed to identify novel mechanisms of action for niacin in the context of AD. In the APP/PS1 transgenic model of AD, mice that received niacin supplementation had enhanced cognition. Niacin-supplemented AD mice also had elevated expression of genes including Ctnnb1, Mdm2, and Pten, which are involved in processes such as Wnt signalling, posttranslational modifications, and regulation of mTOR signalling [155].

Longitudinal studies suggest that niacin may have therapeutic potential in the context of people with AD. For example, previous work has established that increased intake of dietary niacin is associated with improved cognition [156] and reduced risk of AD later in life [14]. Whether or not niacin is an effective treatment in patients with AD remains to be investigated but is a promising avenue for future research.

Glioblastoma

Glioblastomas (GBMs) are the most common primary tumours in the CNS in adults, and they affect approximately 2.3 people per 100,000 [157]. The current treatment involves surgical resection followed by radiation and chemotherapy with temozolomide, yet GBMs are one of the deadliest forms of cancer, with a median survival time of less than 15 months following diagnosis [158]. Treatment advances are in part halted by the self-renewing capacity of brain tumour initiating cells (BTICs), an immunosuppressive tumour microenvironment, and limited CNS access due to the BBB [159, 160]. Indeed, treatment of GBM has not improved since 2005 with the introduction of temozolomide into the therapeutic regimen, despite research efforts [161].

BTICs are a subclass of cancer cells that initiate glioblastoma growth and development due to their capacity for self-renewal and proliferation [162]; engraftment of as few as 100 human BTICs is sufficient to generate an identical tumour in recipient mice [163]. These cells have highly efficient DNA repair machinery, making them resistant to traditional radiation treatment that induces cell death by causing double-stranded DNA breaks [164]. BTICs have also demonstrated resistance to chemotherapy, although the mechanisms involved are not as clear [165]. Effective GBM treatment is further hindered by the presence of an immunosuppressive tumour microenvironment. Tumour-associated macrophages are the most abundant nontransformed cells in the tumour microenvironment, and they demonstrate a clear protumour phenotype, releasing anti-inflammatory cytokines and failing to initiate T cell responses [166]. Furthermore, tumour-infiltrating immune cells display an exhausted phenotype with reduced activity and fail to mount a proper immune response towards tumour cells [167]. Thus, a potential treatment for GBM would ably cross the BBB and stimulate immune activity, counteracting immune suppression and promoting recognition of BTICs by immune cells.

In a rodent model of GBM, treatment with niacin promotes beneficial immune modulation, rejuvenating immunosuppressive myeloid cells and increasing their tumour-fighting abilities (Fig. 5). Indeed, niacin activates myeloid cells derived from GBM patients in vitro, leading to their enhanced release of cytokines such as TNFα and IL-6, and niacin-treated monocytes attenuate the growth of GBM patient-derived BTICs [11]. Furthermore, niacin treatment in BTIC-implanted mice reduces brain tumour growth and prolongs survival. Following these promising preclinical results, the addition of slow-release niacin (niacin^CRT) to standard of care entered a Phase I/IIa clinical trial in individuals with GBM (clinicaltrials.gov NCT04677049) (Table 1). Results from this ongoing trial will increase our knowledge on the potential of niacin to promote antitumour immune functions in a clinical setting.

Amyotrophic Lateral Sclerosis

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disorder, characterized by the loss of both upper and lower motor neurons. Current treatments target symptoms such as muscle spasticity, sialorrhea, and pain, but disease-modifying therapies are nonexistent [168]. Recently, the gut microbiome has been implicated in the pathogenesis of many disorders, including ALS [169], with a particular interest being paid to the microbiota-gut-brain axis [170]. In animal models of ALS, impaired tight junction integrity and enhanced gut permeability are observed [171], and microbiome dysbiosis precedes motor deficits [172]. Furthermore, ALS patients have significantly altered gut microbiota composition compared to controls [173], and repeated antibiotic use is associated with increased risk of developing ALS [174]. Notably, the recent work described below suggests that niacin may exert its therapeutic effects in part through modulation of the gut microbiome.

In the Sod1 transgenic mouse model of ALS, supplementation of Akkermansia muciniphila increases the serum levels of nicotinamide and leads to improvements in motor and neurological function, suggesting that nicotinamide released from gut microbiota has a beneficial effect on ALS pathogenesis [175]. Further, ALS patients have altered serum levels of molecules involved in nicotinamide synthesis, and increased serum nicotinamide correlates with better functional status [175]. Beyond the ALS literature, in obese humans, low levels of dietary niacin intake are correlated with low alpha-diversity and reduced Bacteroidetes abundance [176], a microbe that is largely considered to be beneficial [177]. Administration with delayed-release nicotinic acid leads to a significant increase in Bacteroidetes abundance, indicating that niacin administration can modulate composition of the microbiome [176]. In ACE2 knockout mice, which are prone to colitis and experience altered microbiome ecology, nicotinamide administration restores the microbiome composition to control levels and ameliorates gastrointestinal symptoms of colitis [178]. Thus, by modulating the gut microbiome, niacin may serve as a promising therapeutic option for ALS and other CNS disorders (Fig. 5).

Conclusion and Remaining Questions

In conclusion, niacin is an essential vitamin that has long served as a well-tolerated treatment for a variety of disorders. While its canonical role is as a precursor for NAD⁺/NADP, niacin has additional mechanisms of action, including agonistic activity at the GPCR Hcar2, and modulation of the microbiome. Due to the expression of Hcar2 on immune cells, niacin is emerging as a potent modulator of the immune system and has been shown to promote a beneficial immune cell phenotype, enhancing phagocytosis of harmful debris and reducing neuropathology in several neurological disorders. Niacin also plays a role in regulating cholesterol recycling, which is critical following the uptake of lipid-rich debris such as myelin by CNS macrophages. The recent work is establishing niacin as a promising therapeutic option in a range of neurological diseases such as MS, Alzheimer’s disease, and glioblastoma.

Several questions remain about the mechanisms and utility of niacin. At what doses are its effects due solely to metabolism such as conversion to NAD⁺? When high pharmacological doses are used where Hcar2 stimulation is thought to be engaged, to what extent is the benefit conferred by metabolic mechanisms? Are there as yet unidentified receptors for niacin? Is niacin anti-inflammatory or pro-inflammatory, and is there a concentration range that separates these potentially divergent activities? Would long-term use in chronic conditions such as MS run into risks of pro-inflammatory responses? Can niacin be combined with other therapeutic agents, such as direct agonists at LXRs, for a more efficacious outcome? Are there additional neurological conditions that may benefit from niacin therapy? Future studies will elucidate the role of niacin as a neuroprotective agent, allowing for its widespread adoption into clinical practice.

Supplementary Information

Below is the link to the electronic supplementary material.