Introduction

Russian chemist V.S. Gulewitsch discovered the endogenous peptide carnosine (b-alanyl-l-histidine) more than a century ago as a main element of Liebig’s meat extract (Gulewitch and Amiradzibi 1900), and in the last 30 years much attention has been attracted to this substance, due to its promising beneficial health effects. In the first half of the twentieth century it was found that carnosine is synthesized and hydrolyzed by specific enzymes, namely carnosine synthase and carnosinase, and is accumulated in vertebrate brain and muscles in amounts proportionally to their functional activity. Its biological effects first were detected in muscle tissue, and the so-called Severin’s effect was outlined—prolongation of the muscle contractile capacity after addition of carnosine to the muscle exhausted by preceding exercise in vitro (Severin et al. 1953). The proposed mechanism of this effect was based on carnosine’s pH-buffering activity. Henceforth, the number of major researches in different tissues revealed many properties of carnosine, such as its direct antioxidant action (Boldyrev et al. 1988; Kohen et al. 1988), ability to modulate the endogenous antioxidant system (Boldyrev et al. 2007), to scavenge NO (Nicoletti et al. 2007), metal-chelating (Corona et al. 2011), telomerase-like activity (Holliday and McFarland 2000), chaperone activity (Villari et al. 2014), immunomodulation (Nagai and Suda 1986b), and protein carbonyl formation inhibition (Hipkiss et al. 1997). After a hundred years’ study of carnosine’s biological activity, its importance in the excitable tissues leaves few doubts, and the research of its biological role in skeletal muscles has already reached the practical stage being widely implemented in sport and nutritional science (Derave 2011; Hobson et al. 2012). However, the same cannot be said for the cerebral and neurological diseases. Despite the growing number of works, studying the effects of carnosine in neural tissue, there are still several difficult issues that delay the advancement of carnosine in the relevant areas: (1) carnosine distribution and bioavailability in the brain, (2) cellular cascades of its action in neuronal/glial cells, (3) its physiological actions in normal conditions and pathology in vivo, (4) its efficiency in human studies. The purpose of this review is to address these key questions summarizing the recent literature data and the original data obtained in our laboratory.

Carnosine distribution and bioavailability

Carnosine is a nonproteinogenic dipeptide, consisting of β-alanine an l-histidine. Though in mammalian tissues carnosine is prevalent in the skeletal muscle (10 mM and more), a number of analytical studies reveal that carnosine concentrations are comparable to those of the muscle in the olfactory bulb (1–2 mM), whereas lower concentrations (< 0.1 mM) of carnosine and homocarnosine occur in the brain and in the spinal cord (Margolis 1974; O’Dowd et al. 1990; Osborne et al. 1974; Pisano et al. 1961). In humans, brain homocarnosine concentration is rather high (0.5–1 mM) in contrast to carnosine and can be assessed by proton magnetic resonance spectroscopy (MRS) in vivo (Petroff et al. 1998).

The data on carnosine distribution in different brain regions is rather scarce, and the majority of the studies show the prevalence of brain carnosine in the olfactory bulb and epithelium (De Marchis et al. 2000; Margolis 1974). Nevertheless, in the study by Flancbaum (Flancbaum et al. 1990) in different murine species, the concentrations of carnosine in several brain regions (pituitary gland, olfactory bulb) are shown to be comparable to those in muscles and even higher in some cases (in CDF1 mice 25.97 ± 3.8 mkg/g in hypothalamus versus 8.03 ± 4.2 in muscle). According to more recent studies, the basal level of carnosine in mouse brain is around 10 mkg/g (Sariev et al. 2015). In the study by Ivanisevic (Ivanisevic et al. 2015), the evaluation of metabolic patterns across the brain regions uncovers high levels of carnosine along with uric acid in the frontal cortex of rats. Another recent study reveals an increased level of homocarnosine in the frontal cortex and hippocampus of human postmortem material (Lieblein-Boff et al. 2015). This data presumes the presence of rostro-caudal gradient of carnosine in the mature brain, possibly related to its prevalent synthesis in the olfactory bulb. However, within the first three postnatal weeks, the glial immunostaining of carnosine-related dipeptides shows a caudo-rostral gradient from the spinal cord to the forebrain which may be connected with the wave of glial maturation (De Marchis et al. 1997).

The first data, concerning the tight link between carnosine-related dipeptides and glial cells, has been acquired with an extensive mapping of the carnosine-like immunoreactivity in the mouse brain (Biffo et al. 1990). Only olfactory bulb neurons have revealed the presence of carnosine, whereas in other brain areas carnosine has been found mainly in glial cells. Other cells found to be highly positive for carnosine are ependymal cell precursors of rodents, generated by the neuroblasts of the sub-ependymal layer (SEL) (Bonfanti et al. 1999). Further studies by means of anti-carnosine serum in combination with different glial markers in the brain of adult rats show the presence of carnosine both in mature astrocytes and oligodendrocytes in the brain (De Marchis et al. 1997) and spinal cord (De Marchis et al. 2000), with no immunoreactivity in the peripheral nervous system. Significant changes in carnosine distribution are shown after brain ischemia (Rajanikant et al. 2007a). Whereas no carnosine immunoreactivity is observed in the core, a dramatic increase in the cellular immunostaining is observed at the periphery of the infarct area. In normal conditions, carnosine concentration in the brain is possibly affected mainly by its excretion from muscles (Nagai et al. 2003).

As per the infiltration of carnosine in the brain, the majority of data support the idea of its synthesis de novo rather than intact carnosine penetration through the BBB. In vitro models showed that glial cells might synthesize histidine-containing dipeptides. Carnosine and homocarnosine were first demonstrated to be synthesized by the rat glioma cell line and primary cultures of newborn mouse brain, rich with astroglia (Bauer et al. 1979, 1982). However, it is important to remember that in the living brain, only oligodendrocytes express carnosine-synthetase activity (Hoffmann et al. 1996). Similar to the muscle tissue, the limiting amino acid for the carnosine synthesis in the brain is β-alanine, which is mainly obtained as a product of the uracil and thymine catabolic pathways in the liver (Bauer et al. 1982). With regard to homocarnosine, which is more common to the brain, it is the other amino acid, aside from l-histidine, GABA, which is a widespread inhibitory neurotransmitter in the mammalian brain. Both β-alanine and l-histidine could be readily taken up from the blood into the brain through the amino acid transporters in the BBB (Hawkins et al. 2006). The interesting findings have been obtained for the nasal administration, which makes it rather prominent for the therapeutic use of carnosine. The irrigation of the mucosa with radiolabeled β-alanine and histidine has led to the rapid uptake of these precursors and their conversion into carnosine, which subsequently has been transported to the olfactory bulb with axonal flow (Margolis and Grillo 1977). Recently the data on carnosine pharmacokinetics has been obtained in a detailed study on mice with the use of HPLC–mass spectrometry (Sariev et al. 2015). After a single intraperitoneal injection (1 g/kg) of carnosine, its peak concentration in the brain (20.3 µg/g) was reached in 6 h, the mean residence time was 24.36 h and the calculated tissue availability in the brain was 59%. The carnosine pharmacokinetic curve in the brain is quite different from that in the blood, where the peak concentration is much higher (1081.75 µg/g) and can be reached in 0.25 h which is consistent with the earlier studies in rats (Guliaeva et al. 1989). This discrepancy assumes that cerebral carnosine is mainly synthesized de novo in glial cells after its breakdown in the blood flow, but this synthesis is highly affected by the altering level of the precursor amino acids.

Cellular mechanisms of carnosine action

The great importance of redox potential for the neural tissue is substantiated by its distinctive biochemical properties (Boldyrev et al. 2001). Both the neuronal cell-signaling cascade proteins and their electrical properties rely on the integrity of the neuronal membrane, which is rich in unsaturated fatty acids—the main substrate for the lipid peroxidation (Floyd 1999). The main sources of reactive oxygen species (ROS) in the brain are: elevated mitochondria metabolism (Chance et al. 1979; Fridovich 2004), catecholamine turnover (Lotharius and O’Malley 2000), excitotoxicity (Mark et al. 2001) and inflammation (Huang et al. 2006; Liu et al. 2003). These mechanisms contribute to the pathogenesis of ischemic injury (Guo et al. 2013; Suslina et al. 2007) and neurodegenerative diseases including Alzheimer’s, Parkinson’s dementia, Huntington chorea and lateral amyotrophic sclerosis (LAS) (Cantuti-Castelvetri et al. 2000; Floyd 1999; Shukla et al. 2011). More recently, the internal mechanisms of such conditions—depression, schizophrenia and attention deficit hyperactivity disorder (ADHD)—are also associated with oxidative damage (Chung et al. 2013; Popa-Wagner et al. 2013). The data on oxidative stress in normal physiological conditions is rather scarce (Aksu et al. 2009; Forman et al. 2010; Nilova and Polezhaeva 1994).

In his reviews, Boldyrev summarizes the main biochemical properties of carnosine in detail (Boldyrev 2012; Boldyrev et al. 2007, 2013). They include buffering activity, metal ion-chelating activity, direct and indirect antioxidant action, inhibition of protein carbonylation and glycoxidation, elimination of aberrant proteins, regulation of nitric oxide and possibly the NO- and ROS-signaling cascades. Despite the focus on direct antioxidant properties of carnosine in the majority of brain researches, it turns out that its action is always multifactorial. Even its oxidative stress-ameliorating properties could be mediated both by its direct antioxidant ability (Boldyrev et al. 1997) and its ability to inactivate products of lipid peroxidation, for example, 4-hydroxy-trans-2-nonenal (Marchette et al. 2012).

Carnosine’s direct effect on cell viability under various stressful conditions has been demonstrated both on neuronal (Lopachev et al. 2016; Ouyang et al. 2016) and glial cells (Shi et al. 2017). The latest researches in this field are listed in Table 1. We have shown the neuroprotective effect of carnosine (2 mM) in primary cultures of rat cerebellar cells under oxidative stress induced by 1 mM 2.2ʹ-azobis(2-amidinopropane)dihydrochloride (AAPH), which directly generates free radicals in the medium and in the cells, and 20 nM rotenone, which increases the amount of intracellular ROS. In both models, administration of 2 mM carnosine to the incubation medium decreased cell death. The antioxidant effect of carnosine inside cultured cells has been demonstrated using dichlorofluorescein fluorescent probe (Lopachev et al. 2016). The neuroprotective effect of carnosine is studied extensively in several toxin-induced parkinsonism models: salsolinol (Zhao et al. 2017) and 6-OHDA (Kulikova et al. 2016; Oh et al. 2009). Carnosine treatment (10 mM) also reduces ROS production and increases cell viability in the chemical hypoxia model induced by 10 uM antimycin A with glucose-free media (Park et al. 2014). Carnosine decreases cell death, increases extracellular GABA and decreases levels of extracellular glutamate, and reverses mitochondrial energy metabolism disorder in the oxygen–glucose deprivation model (90 min/2 or 24 h) in neuron/astrocyte co-cultures (Ouyang et al. 2016). Both this and other recent studies point out different actions of carnosine that determine its neuroprotective properties: metal-chelating activity in the model of heavy metal toxicity (Fedorova et al. 2016a, b), pH-buffering capacity in the model of NMDA excitotoxicity (Lopachev et al. 2017), the protection of membrane-bound Na,K–ATPase (Kurella et al. 1999), amyloid fibril inhibitory property in the model of Ab1-42 toxicity (Aloisi et al. 2013) and the ability to interact with intracellular signaling cascades (Oh et al. 2009). The last effect is especially important, because it is not limited to the pathological conditions and is closely related to the question on the role of carnosine in the neuronal tissue functioning.

Table 1 Summary of main studies reviewed reporting carnosine effects in vitro
Full size table

Yet, the data on this issue remain contradictory. The ability of carnosine to change the expression of bcl-2, bax, NF-kB and MN-SOD genes and alter the NMDA-induced activation profile of ERK ½, JNK and MAPK kinase cascades has been shown in the studies on PC-12 and cerebellar cell cultures (Kulebyakin et al. 2012). However, the latter study has shown no carnosine-induced changes in the level of apoptosis-regulating proteins of the Bcl-2 family and in the phosphorylation of MAP kinases. This suggests that carnosine possibly has minimal or no effects on proliferation and apoptosis control systems in normal cells (Lopachev et al. 2016). The recent studies also consider possible indirect mechanisms of carnosine action on the cell-signaling cascades via regulation of glutamate transporters (Ouyang et al. 2016; Shen et al. 2010) or the growth factor-mediated neuron-to-glia interactions (Yamashita et al. 2017). Despite the growing support to the notion of ROS regulatory role in neuronal functioning (Forman et al. 2010; Son et al. 2011) and, hence, the substantiation of the possible mechanism of carnosine action via oxidative stress-responsive genes, there are still not enough studies in this area (Calabrese et al. 2005; Kulebyakin et al. 2012; Spina-Purrello et al. 2010). Therefore, the action of carnosine on the neuronal and glial cells turns out to be multipotent and, despite more than a century of research, there are still “new concepts for the function of the old molecule” (Boldyrev 2012).

Carnosine actions in vivo

The vast majority of studies on carnosine physiological function and therapeutic potential in vivo are conducted in murine models of oxidative stress-related disease. It is driven from the fact that the major effects of carnosine are attributed to its free radical quenching and carbonyl species inactivating ability. The main research topics in this field are covered by Table 2. Since the 1950s, more than 500 works on the physiological role of carnosine in the brain have been published, according to Bellia’s review (Bellia et al. 2011). The most pronounced and well-documented physiological effect of carnosine is the reduction of ischemia/reperfusion (I/R) damage in different animal models in the brain and other organs. The anti-ischemic activity of carnosine has been extensively studied in vivo (Davis et al. 2016; Gallant et al. 2000; Stvolinsky and Dobrota 2000). In earlier studies 7-month-long carnosine supplementation (100 mg/kg with drinking water) had a complex effect on the permanent 120 h cerebral ischemia, induced by both side occlusion of the common carotid arteries: reduction of mortality from 55 to 17%, normalization of monoamine oxidase B (MAO B) activity, reduction of glutamate excitotoxicity and improvement of post-ischemic neurological symptomatic and learning ability in the Open-field and T-maze tests (Gallant et al. 2000). The extensive analysis of this study shows the dual effect of carnosine on the glutamate binding to NMDA receptors: it significantly increases in normal conditions (p < 0,01 carnosine vs. intact), but decreases in the ischemic conditions after carnosine treatment (p < 0,05 carnosine vs. intact) (Gallant et al. 2000). This study substantiates the positive effect of carnosine supplementation not only after ischemia, but also in intact animals that is also supported by improved learning results both before and after ischemia. In the following years, carnosine neuroprotective effects are shown both in permanent (Min et al. 2008; Rajanikant et al. 2007b) and transient ischemia (Park et al. 2014; Pekcetin et al. 2009) and in the model of intracerebral hemorrhage in rats (Xie et al. 2017). Interestingly, the anti-ischemic effect is dose dependent, limited to carnosine and is not possessed by its analogs or constituting amino acids (Bae and Majid 2013; Min et al. 2008). In the outstanding meta-analysis by Davis (Davis et al. 2016), carnosine treatment, both before and after ischemia, is shown to decrease the infarct volume from 24 to 34.9% on average and the therapeutic window is shown to extend up to 6 h after the ischemic episode. The neuroprotective effect of carnosine in doses less than 1000 mg/kg is shown only if administered i.p. 30 min before the ischemia onset and in the majority of studies with therapeutic carnosine use. The significant effects is demonstrated only in high doses (1000–2000 mg/kg) either i.p. (Rajanikant et al. 2007b) or i.v. (Bae et al. 2013). Further studies reveal the significant neuroprotective effect of carnosine in lower doses—the prevention of oxidative damage and apoptosis in the ischemic penumbra in both preventive (150 mg/kg) (Deviatov et al. 2017; Stvolinsky et al. 2017), (200, 500 mg/kg) (Ma et al. 2016) and post-ischemic administration (50 and 500 mg/kg) (Fedorova et al. 2017). We have found that acute administration of carnosine in the permanent 72 h focal ischemia (MCA coagulation) model in a daily dose of 50 and 500 mg/kg i.p. (15 min, 24, 48 h post-operation) promotes the decrease of the infarct area by 27% and 39%, respectively (Fedorova et al. 2017). The preventive treatment with carnosine in the 24 h transient ischemia (MCA occlusion) model in a dose of 150 mg/kg per os (with food) for 7 days prior to the operation promotes the decrease of the infarct area by 20% with respect to the control group (Stvolinsky et al. 2017). The main carnosine effects in ischemia studies were shown to be mediated by matrix metalloproteinases activity regulation (Rajanikant et al. 2007b), protection from MDA-induced neuronal injury (Cheng et al. 2011), lipid peroxidation prevention (Stvolinsky et al. 2017), antiexcitotoxic, antioxidant, and mitochondria protecting activity (Bae et al. 2013). In the study by Shen (Shen et al. 2010), it is clearly shown that carnosine effect on permanent cerebral ischemia is still observed in histidine decarboxylase knockout mice, and this effect is mediated by the attenuation of glutamate excitotoxicity by carnosine itself, not the constituting amino acids. Carnosine is also attributed with significant anti-hypoxic effect in vivo (Fedorova et al. 2016a, b; Berezhnoy et al. 2015; Fedorova et al. 2006), which may directly contribute to its anti-ischemic effects.

Table 2 Summary of main research topics on carnosine in vivo
Full size table

Carnosine effects in other neurological diseases gain less attention than ischemia–reperfusion injury, probably because of the greater uncertainty in the pathophysiological cascades involved. However, there is a growing amount of works, where the effects of carnosine in neurodegeneration have been studied. Models of Parkinson’s disease induced by ROS-generating toxins are widely used in the studies of carnosine therapeutic potential. In the model of Parkinson’s disease (PD) induced by systematic intraperitoneal 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) injection, carnosine administration (100 mg/kg simultaneously with MPTP) to senescence-accelerated mice (SAMP1) (8 days, 30 mg/kg daily), preserves motor activity, prevents the development of rigidity and suppresses oxidative stress development in the brain tissue by preventing the decrease in superoxide dismutase (SOD) activity and inhibiting the formation of protein carbonyls and lipid hydroperoxides (Boldyrev et al. 2004; Sorokina et al. 2003). The long-term preventive intake (4 weeks, 0.5, 1, and 2 g/L of drinking water) of carnosine also decreases the glutathione loss in the striatum, retains the activity of glutathione peroxidase (GPX) and SOD, diminishes oxidative and nitrosative stress and lowers inflammatory response in MPTP-treated mice (Tsai et al. 2010). More recent studies reveal the beneficial effect of acute carnosine treatment (250 mg/kg i.p. twice with a 24 h interval pre-surgery) in the unilateral intrastriatal 6-OHDA-lesioned rats. Carnosine preventive treatment significantly reduces rotational behavior, restores catalase activity, attenuates apoptosis, and restores malondialdehyde (MDA) and nitrite content (Afshin-Majd et al. 2015). Even lower concentration (50 μg/ml) of carnosine as shown in the study of salsolinol-induced neurotoxicity is capable of decreasing the level of apoptosis in the rat brain significantly and renormalize the level of MDA, glutathione (GSH), SOD and catalase activity (Zhao et al. 2017).

Only few studies of carnosine in animal models of Alzheimer’s disease (AD) are reported, and they are mainly conducted on transgenic animals (Corona et al. 2011; Herculano et al. 2013). It is reported that carnosine supplementation (10 mM L-Carnosine in tap water for 11–13 months) has a strong effect on restoring mitochondrial functioning, preventing Zn-ion toxicity and counteracting amyloid-b aggregation, but not the tau-pathology, in the triple transgenic 3xTg-AD mouse model (Corona et al. 2011). The positive effects on mitochondria are possibly related to Zn-chelating and antioxidant activity, while the anti-amyloid action is possibly associated with antiglycating and chaperone-like activity (Hobart et al. 2004).

The positive effects of carnosine are shown in a wide range of different pathologies, with pathophysiological cascades associated with oxidative stress: ethanol-induced brain damage (Turkcu et al. 2010), methamphetamine-induced gliosis (Pubill et al. 2002), acute spinal cord injury (Di Paola et al. 2011), etc. It is suggested that carnosine may work as an internal modulator, preserving the intracellular homeostasis, increasing resistance to stressful conditions (Nagai and Suda 1986a).

However, carnosine effects in pathology are not limited to oxidative stress-driven diseases and new studies broaden our perception of carnosine effects in the brain. One of such examples is a prominent action of carnosine in the models of epilepsy. To date, only a few studies have been published on the anticonvulsant role of carnosine in the experimental epilepsy models: amygdala kindled (Jin et al. 2005), PTZ induced (Wu et al. 2006; Zhu et al. 2007) and penicillin induced (Kozan et al. 2008). Jin et al. (2005) reports that intraperitoneal injection of carnosine (500, 1000, 1500 mg/kg) significantly decreases seizure stage and duration and also prolongs generalized seizure latency of amygdaloid-kindled seizures. The protection of carnosine is antagonized by histamine H1-antagonists pyrilamine and diphenhydramine, but not by histamine H2-antagonist zolantidine that assumes that carnosine acts via carnosine–histidine–histamine pathway (Flancbaum et al. 1990). This study also contains an interesting finding, concerning the neuromodulatory effect of carnosine: concentrations of glutamate in the hippocampus and amygdala significantly decrease 1–2 h following the increase of histamine levels after carnosine injection. In different models, the dose of 500 mg/kg is the most effective in ameliorating induced seizures (Kozan et al. 2008; Wu et al. 2006). These studies suggest that carnosine by itself, or via histidine, may play a role of an endogenous anticonvulsant factor in the brain. Though, some studies reveal the opposite effect: carnosine (25–100 mmol/l) application to CA1 region of intact rat hippocampus transforms population spikes with single spike into epileptiform multiple spikes and significantly decreases paired-pulse stimulating depression, similar to GABAA antagonist picrotoxin (Feng et al. 2009). Moreover, it should be noted, that the doses of carnosine used in this studies far exceed the normal physiological concentrations of this dipeptide in the brain (Boldyrev and Severin 1990). In the study with lower doses of carnosine (0.1, 1, 10, 100 mkg/mkl), its application into the dentate gyrus of rat hippocampus reveals the dose-dependent effect on long-term potentiation post-tetanic and induction phase: the population spike amplitude is decreased by the small doses (0.1, 1 mkg/mkl) and is increased by the dose of 100 mkg/mkl (Suer et al. 2009). These processes, relied on the Ca-concentration, suppose the potential role of carnosine in neurotransmission.

There is a wide range issues to discuss on this notion. Bearing in mind its synthesis and high concentrations in the olfactory bulb neurons, carnosine is hypothesized to be a neurotransmitter or a neuromodulator in the olfactory system. The experimental data on this issue remains contradictory- either supporting (Kanaki et al. 1997; Panzanelli et al. 1997; Sassoe-Pognetto et al. 1993) or disproving (Frosch and Dichter 1984; MacLeod and Straughan 1979; Nicoll et al. 1980). In the in vivo study on rabbit olfactory bulb, carnosine application (0,44 mol/l in 2-5 mkl) increases the frequency and decreases the phase of evoked potentials and produces the sustained oscillations in EEG within 2–7 min, supporting its excitatory role (Gonzalez-Estrada and Freeman 1980). Nevertheless, the following works with lower concentrations are incapable of finding the inward currents in identified neurons in response to carnosine (Frosch and Dichter 1984; MacLeod and Straughan 1979; Nicoll et al. 1980). It is possible that carnosine effects on neuronal activity are mediated by glutamate receptor modulation via its ability to chelate Zn ons or other indirect mechanisms (Gallant et al. 2000; Panzanelli et al. 1997), but this issue requires further research. The other potential mechanism of modulation involves glial cells, exhibiting glutamate-receptor mediated release of carnosine (Bakardjiev 1998) and carnosine-related release of brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF) (Yamashita et al. 2017). These mechanisms broaden the issue of the potential role of carnosine in neuron-to-glia along with neuron-to-neuron communication.

The other interesting study of carnosine functions in the brain lies around its effects on systemic stress-related response, sympathic system, hypothalamus, and various types of behavior. Carnosine is shown to ameliorate stress response in animals: blood and cerebral cortisone and noradrenaline are quickly normalized in carnosine-treated animals after their exposure to electric shock (Nagai et al. 1990) that is in line with carnosine action on the brain oxidative status in the same conditions (Guliaeva et al. 1989). The dietary supplementation with carnosine (150,300 mg/kg in drinking water for 7 days) is also shown to lower the level of corticosterone in stressed mice (Tsoi et al. 2011), though the data is controversial (Tomonaga et al. 2004). Carnosine administration (0.005 to 5 nmol per 300 g body weight, i.p.) suppresses the activity of sympathetic nerves innervating the adrenal glands, liver, and pancreas (Niijima et al. 2002; Yamano et al. 2001), lowering the blood pressure and glucose concentration. This effect is limited to L-carnosine, mediated by histaminergic neurons of hypothalamic suprachiasmatic nucleus (SCN) and blocked by the administration of a histamine H3 receptor antagonist (thioperamide), that supports the concept of carnosine–histidine–histamine pathway (Flancbaum et al. 1990). It is supposed, that sympathetic stimulation is mediated by H1 receptor, whereas the parasympathetic stimulation or sympathetic suppression is mediated by H3 receptor (Nagai et al. 2012). The dose-dependent biphasic effect of carnosine on renal sympathetic nerve activity, mean arterial pressure and body temperature, in line with its biphasic effects on hippocampal LTP (Suer et al. 2009), is also shown: after administration of small doses (1 mg/300 g i.v. or 0.01 mg/300 g i.c.v.) the reduced nerve activity and corresponding physiological parameters are observed, and large amounts administered (100 mg/300 g i.v. or 10 mg/300 g i.c.v.) show an opposite effect (Tanida et al. 2007, 2005). These effects are also relied on histamine mediated action of carnosine in SCN neurons.

Other potential mechanisms of carnosine action, not mediated by SCN, are shown in the studies of animal behavior in different models—the least studied effects of carnosine. Among the revealed behavioral effects there are: hyperactivity (Tomonaga et al. 2004, 2005), antidepressant-like activity (Tomonaga et al. 2008), the potentiating effect on learning in T-maze (Gallant et al. 2000), Morris water maze (Stvolinsky et al. 2012) and passive and active avoidance paradigms (Berezhnoy et al. 2016; Stvolinsky et al. 2014). In the study of different NOS inhibitors, it is found that hyperactivity in chicken induced after i.c.v. infusion of carnosine (3.2 μmol) is mediated by cNOS activation in astrocytes (Tomonaga et al. 2005). The authors also propose the mechanisms of action via GABAA receptor modulation. It is important to point out that these works deal with systemic actions of carnosine in normal animals, not burdened by any kind of pathological process. Although carnosine supplementation (100 mg/kg i.p. daily) has the beneficial effect on learning, it is rather expressed in aversive conditioning models, accompanied by emotional stress. The potential effects of carnosine in these conditions may be tied with alterations in the content of monoamine neurotransmitters (Berezhnoy et al. 2016) and their metabolism (Tomonaga et al. 2008) which are in line with modulation of the stress-activated hypothalamic–pituitary–adrenal axis (Tsoi et al. 2011). The specific behavioral effects are limited to carnosine, and infusion of constituting amino acids or “reverse-carnosine” leads to the opposing effects (Tomonaga et al. 2004; Tsuneyoshi et al. 2008).

The aforementioned effects—especially the structure–effect dependence (Tsuneyoshi et al. 2008), biphasic concentration dependence (Tanida et al. 2005) and pronounced systemic effects in small doses—assume that carnosine should be treated as a neuroactive peptide and its physiological effects may be pronounced in lesser concentrations than those, usually used in studies in vivo.

Efficiency of carnosine in human studies

Unlike the studies in animal models, the number of carnosine studies in humans demonstrating its clinical efficiency in the nervous system, is rather limited. It may be due to the problems with the sufficient dose maintenance in humans and due to the fact that in most countries carnosine is registered as dietary supplement, rather than a drug. Despite that, the amount of studies on the beneficial role of carnosine in cognitive performance and well-being in different conditions since recently has being growing. In the double-blind placebo-controlled study of carnosine/anserine supplementation (500 mg daily, per os for 3 month) in the elderly healthy people (40–78 years) beneficial cognitive and neurophysiological effects are shown (Rokicki et al. 2015). Carnosine increases the verbal episodic memory and restores the resting state network connectivity, based on the fMRI study. Another study reports the decrease in the level of proinflammatory cytokines in the blood and increase of the cerebral blood flow, mediating these beneficial changes (Hisatsune et al. 2016). In this study, histidine, used as control, did not reproduce the effects of carnosine. The clinical study in the Persian Gulf War veterans was conducted to estimate the effect of carnosine supplementation (3 month 500, 1000, and 1500 mg daily, increasing in 4 week intervals) on the well-being and cognitive function in this condition. The use of carnosine provides the significant increase on the WAIS-R (Wechsler Adult Intelligence Scale) and decrease in diarrhea associated with irritable bowel syndrome (Baraniuk et al. 2013).

The double-blind placebo-controlled study, conducted in autistic children, reveals that carnosine (800 mg daily, per os for 3 month) as a supplement to the basic therapeutic scheme leads to the increase in CARS (Childhood Autism Rating Scale) and E/ROWPVT (the Expressive and Receptive One-Word Picture Vocabulary tests). Though the mechanism of carnosine in these studies is not clear at all, it may be mediated by the alterations in the entorhinal or temporal cortex functioning (Chez et al. 2002). The authors report improved receptive speech and social attention, lessened apraxia, and overall global improvements. In the other similar clinical study, the long-term treatment with carnosine (500 mg daily, 2 month) reduces sleep disorders in autistic children (Mehrazad-Saber et al. 2018). In a recent metabolomic study, a reduction in the levels of urinary carnosine, β-alanine, and l-histidine is shown in children with autistic spectrum disorder that confirms the prospects of using carnosine for treatment of autistic patients (Ming et al. 2012).

Carnosine as an adjunctive treatment (2 g daily, 3 month) demonstrates beneficial effects on the symptomatically stable adults with chronic schizophrenia in the double-blind clinical trial (Chengappa et al. 2012). Carnosine-treated patients show significant improvement in cognitive tests [Strategic target detection test (STDT) and Set Shifting Test (SST)] without changes in most psychopathological tests. Although, the carnosine group reported more adverse events (30%) than the placebo group (14%), the acquired effect remains contradictory. The other ongoing psychopathological clinical study is dealing with the effects of carnosine on cognitive functions in bipolar disorder (ClinicalTrials.gov Identifier: NCT00177463).

The double-blind placebo-controlled study is performed in Russia, where « Sevitin » (commercially available source of carnosine) is added to the standard therapy in patients with chronic cerebral ischemia of different origin. Besides the evaluation of neurological symptomatic, the reaction of the cortical hearing center to paired pulses (potentials P300) and the endogenous antioxidant activity of blood plasma lipoproteins are analyzed. The 20 days treatment (2 g daily) resulted in the improvement of neurological symptomatic, discrimination of the paired pulses, and the recovery of endogenous antioxidant system (Fedorova et al. 2009). Administration of « Sevitin » in a dose of 1.5 g/day during 30 days as per the basic protocol of Parkinson’s disease treatment has resulted in significant improvement of neurological symptoms, along with an increase in red blood cell Cu/Zn-SOD and the decrease of protein carbonyls and lipid hydroperoxides in blood plasma. Thus, addition of carnosine to the basic therapy not only improves clinical indices and reduces toxic effects of the basic therapy, but also elevates antioxidant status of the organism in patients with Parkinson’s disease (Boldyrev et al. 2008). Therefore, these pilot studies, though being small-scale and limited in number, show the therapeutic potential of carnosine as an adjunctive treatment of neurological disorders.

Concluding remarks

We have reviewed the key issues for the research of carnosine functions in the brain and still it is impossible to rule out a single role it plays in the brain. Taking into account both the physiological and pathological oxidative-related cascades in the nervous system, carnosine as an endogenous antioxidant molecule might play a key role in the prevention and amelioration of cerebral damages. But it is unlikely that histidine-related peptides appear in the brain merely as endogenous anti-ischemic or anticonvulsant compounds. We know that carnosine effects in the brain are multipotent, tied both with neurons and glial cells, different in vivo effects are performed by different concentrations of carnosine and its concentration in different brain structures is controlled by a complex biochemical system. It beggars belief such a complexity is associated with this simple molecule, but it is possible that the simplicity and thus universality of this molecule is the key to the diversity of its roles. Starting as an endogenous protector against reactive oxygen species in the neural tissue, this molecule might have integrated into many different functional cascades in the nervous system, playing different roles in different brain structures. Some of its roles are well studied, such as its homeostatic effect, the others, for example, the neuroregulatory ones, are far from being clear, and some are possibly yet to be discovered. It is only clear by now that carnosine is effective in supporting nervous tissue in different conditions of extensive physical and psychological load, and the broadening of our knowledge about the functions of this molecule only opens up new horizons for its use.