Introduction

Aspalathus linearis represents a South African endemic species. Its oxidized leaves serve to make an infusion commonly known as “fermented” rooibos herbal tea (FRHT). Phytochemical analysis of herbal tea yielded big amount of polyphenols that play fundamental role in pharmacological activity of FRHT. The most important compounds comprise flavonoids, phenolic acids, flavanones, flavonols, flavones as well as flavonoid biosynthesis precursor PPAG (phenylpyruvic acid glucoside) [1, 2]. Among above components, aspalathin is found solely in Aspalathus spp., and nothofagin only in few species. A. linearis infusions are inter alia able to scavenge free radicals and influence antioxidant enzymes activity, therefore reveal antioxidant activity both in central nervous system and peripheral tissues. Rooibos tea regulates lipid and glucose metabolism, exerts protective effects against glucolipotoxicity in in vitro research and in pre-clinical animal models, as well as affects inflammatory processes, by e.g., influence on interleukins; thus, theoretically may decrease the risk of neurodegeneration. The recent extensive review of pharmacological activity of Aspalathus linearis extracts elaborated by Pyrzanowska [3] based on pre-clinical research suggests that FRHT affects the activity of central nervous system and may support prospective neuroprotection. In our previous experiments on FRHT, we have shown that long-term rooibos tea consumption increased cognitive performance of rats in the water maze as well as general motor activity and exploration in hole-board test. It changed monoamine and amino acid content in several brain structures—striatum, hippocampus, and prefrontal cortex [1, 4]. FRHT primarily increased striatal dopamine concentration that could correspond with enhanced voluntary locomotion, motivation, and cognitive abilities. Furthermore, rooibos tea revealed feasible protective impact by diminished striatal glutamate accumulation and greater content of taurine both in striatum and hippocampus. Decreased GABA level in prefrontal cortex suggested supporting impact to locomotor and cognitive behavior. Albeit hitherto little is known about the impact of rooibos tea on hypothalamus that regulates the neuroendocrine, autonomic and behavioral expression of emotions, and body homeostasis.

Therefore, present experiment is focused on hypothalamic acting, trying to answer whether FRHT is able to affect neurotransmission in terms of altered concentration of monoamines and amino acids, as well as neurotrophic system activity. Hypothalamus regulates body homeostasis by influencing endocrine system performance (e.g., hypothalamic–pituitary–adrenal/gonadal axis) and by its specific functions. The latter include control of body energy balance (i.e., regulation of food intake and energy expenditure by impact on hunger/satiety feeling and thermoregulation), coordination of circadian rhythms [5], and motivation-related behaviors covering arousal, aggression/defense, reproductive demeanor, and social affiliation [6,7,8,9]. Possible alterations in this brain structure may influence the endocrine and autonomic nervous system or its own function, thus modify, e.g., affective behavior evaluated here by social interaction test. Finally, the effect of rooibos tea on hypothalamic profile of monoamines, amino acids, BDNF, TrkB as well as social behavior is discussed in detail, on the background of our previous experiments and current state of knowledge.

Materials and methods

Plant material

Rooibos Ltd (RSA) supplied “fermented” Aspalathus linearis (Burm.f.) R. Dahlgren raw material (07I1D/115/QQ). The phytochemical analysis methods as well as detailed flavonoid composition of the extracts used in the experiment were already presented in our previous publication [1]. The chromatographic investigation of rooibos infusions showed significant amount of polyphenols, primarily flavonoids, including the specific for rooibos aspalathin and nothofagin. The teas were made with 1, 2, and 4 g of dry leaves used for 100 ml of boiling water, then brewed for half an hour at room temperature and finally given to R1, R2, and R4 groups of rats, respectively.

Animals

In the experiment, the impact of 3-month oral FRHT administration on monoamine, amino acids as well as BDNF and TrkB concentration in the hypothalamus was investigated in adult 9-month-old male Sprague–Dawley rats (n = 35). The randomly divided animals formed four groups: (1) control (Con, n = 9) and pre-treated with (2) 1:100 A. linearis infusion (AL1, n = 8), (3) 2:100 A. linearis infusion (AL2, n = 8) and (4) 4:100 A. linearis infusion (AL4, n = 10). They were caged in pairs and maintained in the same conditions as in Pyrzanowska et al. [1]. Body mass of animals was estimated each 2 weeks. The behavioral observation was conducted within the experimental groups in bodyweight matched pairs of animals (n = 34). All animal testing was performed according to the Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010 on the protection of animals used for scientific purposes, after approval of the Ethical Committee for Animal Experiments at the Medical University of Warsaw (LKE 430/2017). The ARRIVE guidelines 2.0 (Animal Research: Reporting of In Vivo Experiments) were implemented to ensure the highest quality of research standards.

High-performance liquid chromatography with electrochemical detection (HPLC-ECD) method for analysis of monoamines and amino acid content

The analysis was conducted as described precisely in previous articles [1, 10]. The rats’ brains were removed, and hypothalamus tissues excised similarly on an iced plate by standard techniques and criteria to select the sample, then swiftly weighed, frozen, and stored in a deep freezer at − 80 °C for future analysis. The concentrations of monoaminergic neurotransmitters (dopamine—DA, noradrenaline—NA, serotonin—5-HT) and their metabolites (3,4-dihydroxyphenylacetic acid—DOPAC, 3-MT—3-methoxytyramine, homovanillic acid—HVA, 3-methoxy-4-hydroxyphenylglycol—MHPG, and 5-hydroxyindoloacetic acid—5-HIAA) as well as amino acids (aspartic acid—AST, glutamic acid—GLU, gamma-amino butyric acid—GABA, taurine—TAU, alanine—ALA, histidine—HIS, and serine—SER) were evaluated in 35 rats. Also, the turnover of monoamines (as indicated by the values of DOPAC/DA, HVA/DA, MHPG/NA and 5-HIAA/5-HT ratios) was calculated. After the homogenization (Vir Sonic 60 sonicator; VIR TIS) in a 1000 μl of ice-cold 0.1 N perchloric acid, the tissues were centrifuged at 13,000 × g for 15 min at 4 °C (Labofuge 400R Haereus Instruments). Afterward, the supernatants were filtered (0.2 μm syringe filter, Whatman) and 20 μl aliquots were injected onto the HPLC apparatus. Levels of monoamines and metabolites were calculated as ng/g and amino acids as ng/mg of fresh tissue.

Enzyme-linked immunosorbent assay (ELISA) method for analysis of BDNF and TrkB level

The concentrations of brain-derived neurotrophic factor (BDNF) and tropomycin receptor kinase B (TrkB) in rat hypothalami were yielded using commercial ELISA kits (BDNF—MyBioSource MBS2088301; TrkB—Elabscience E-EL-R0656). Tissue samples were homogenized (Virsonic 60; Virtis Inc.) with PBS in a 1:50 ratio and centrifuged at 5000 × g for 5 min. The quantitative estimation of the BDNF and TrkB in tissue homogenates was conducted according to the manufacturer procedure description. The standards and samples were run in duplicate. Plates were read on a ELx800 Universal Microplate Reader (Bio-Tek, USA) and the results were presented as pg/mg of fresh tissue.

Social interaction test

Social behavior [11, 12] was assessed in a plexiglass arena (1 × 1 area, 0.3 m height). A habituation session was conducted 24 h before testing in the empty test apparatus. After 2.5 h of isolation, animals were paired with an unfamiliar weight-matched partner from the same experimental group but from another cage. At the beginning of a trial, the rat was placed in the dimly lit arena opposite the partner and observed for 300 s. Social interaction was assessed by the latency of first contact, total time spent on physical contact, frequency of active interaction episodes (including sniffing, allo-grooming, poking, crawling, and following), and total time spent on them. The aggressive behavior (submitting, biting, wrestling) was also evaluated. Stress level was assessed by the number of defecations during the session. The locomotor activity was scored for pairs of rats as combined time spent on movements. The arena was cleaned with 10% solution of ethanol after each session. The experiment was recorded by a camera situated above the arena as video files and further analyzed by an observer blind to treatment using EthoVision XT10 software (Noldus Information Technology, Netherland).

Statistical analysis

Statistical analysis was performed using Statistica v.13.1 software (Statsoft, PL). The results were presented as mean values and standard error. The normality of data distribution was assessed using Shapiro–Wilk test. In the behavioral and biochemical analysis, data that had not normal distribution were assessed using Kruskal–Wallis ANOVA as well as Dunn’s multiple comparison test and Mann–Whitney test (MW) for estimation of the differences between groups within particular parameters. All the hypotheses evaluated used a significance level of 0.05.

Results

Monoamine level in hypothalamus

The levels of monoamines and their metabolites as well as the MHPG/NA, DOPAC/DA, HVA/DA, and 5-HIAA/5-HT ratio values in the hypothalamus of adult Sprague–Dawley male rats after long-term FRHT administering are collected in the Table 1 and 2.

Table 1 Effects of long-term oral administration of FRHT on monoamine and metabolite levels in hypothalamus of adult Sprague–Dawley male rats
Full size table
Table 2 Effects of long-term oral administration of FRHT on metabolite turnover in the hypothalamus of adult Sprague–Dawley male rats
Full size table

The statistically significant differences among groups were noticed in the level of serotonin (5-HT: H(3,35) = 13.83, p = 0.003). The AL4 rats had lower concentration of 5-HT than all other groups of rats (Table 1) (Fig. 1a). There were no differences in other monoamines and their metabolites content in the prefrontal cortex (Kruskal–Wallis ANOVA: DA: H(3,35) = 1.26, p = 0.75; DOPAC: H(3,35) = 5.74, p = 0.16; HVA: H(3,35) = 0.53, p = 0.91; NA: H(3,35) = 3.32, p = 0.34; MHPG: H(3,35) = 3.25, p = 0.36 and 5-HIAA: H(3,35) = 4.06, p = 0.26). The 5-HIAA/5-HT ratio was different among groups (H(3,35) = 13.52, p = 0.004). The metabolite turnover was significantly increased in AL4 rats versus Con and AL2 rats (Table 2) (Fig. 1b). The values of MHPG/NA (H(3,35) = 4.2, p = 0.24), DOPAC/DA: H(3,35) = 6.36, p = 0.1, and HVA/DA (H(3,35) = 1.55, p = 0.67) ratios were the same among all groups of rats.

Fig. 1
figure 1

Long-term rooibos tea consumption affects the hypothalamic neurotransmission in adult SD male rats, primarily: a serotonin content, b serotonin turnover; c glutamic acid content. **p < 0.01 (vs. Con, Mann–Whitney test); °°p < 0.01 (vs. AL1, MW test); ••p < 0.01 (vs. AL2, MW test); ##p < 0.01 (vs. AL4, MW test)

Full size image

Amino acid level in hypothalamus

The levels of amino acids (alanine, aspartic acid, gamma-amino butyric acid—GABA, glutamic acid, histidine, serine, and taurine) in the hypothalami of adult Sprague–Dawley male rats after long-term FRHT administration are collected in Table 3.

Table 3 Effects of long-term oral administration of FRHT on amino acid levels in the hypothalamus of adult Sprague–Dawley male rats
Full size table

The concentration of glutamic acid varied within the groups of animals (Kruskal–Wallis ANOVA: GLU: H(3,35) = 12.99, p = 0.005). The rats from AL2 group had significantly increased levels of glutamic acid in comparison to controls as well as to both other treated groups (Table 3, Fig. 1c). There were no other differences in the tested amino acids (ALA: H(3,35) = 1.71, p = 0.63; ASP: H(3,35) = 4.34, p = 0.23; GABA: H(3,35) = 5.41, p = 0.14; HIS: H(3,35) = 2.15, p = 0.54; SER: H(3,35) = 3.02, p = 0.39; and TAU: H(3,35) = 6.89, p = 0.08).

BDNF and TrkB level in hypothalamus

The levels of BDNF (brain-derived neurotrophic factor) in the hypothalami were the same in all the groups of animals (Kruskal–Wallis ANOVA: GLU: H(3,35) = 0.93, p = 0.82), whereas the BDNF receptor—TrkB (tropomycin receptor kinase B) level alteration was observed (H(3,35) = 13.98, p = 0.003) (Table 4). All the treated animals had decreased expression of the hypothalamic TrkB receptors (Fig. 2a, b).

Table 4 Effects of long-term oral administration of FRHT on BDNF and TrkB content in the hypothalamus of adult Sprague–Dawley male rats
Full size table
Fig. 2
figure 2

Hypothalamus—the content of a BDNF and b TRkb. **p < 0.01 (vs. Con, Mann–Whitney test)

Full size image

Body weight

The mean body mass of the rats (Fig. 3) at the beginning of the experiment did not differ between groups (M1: Con: 428 ± 10.33 g, AL1: 433 ± 7.2 g, AL2: 432 ± 6.5 g, AL4: 422 ± 5.1 g; Kruskal–Wallis ANOVA H(3,35) = 2.25; p = 0.52). Three months of rooibos tea administration did not cause the differences in body mass of animals between groups (M2: Con: 473 ± 12.8 g, AL1: 484 ± 6.4 g, AL2: 471 ± 9.2 g, AL4:467 ± 5.5 g; H(3,35) = 3.05; p = 0.38).

Fig. 3
figure 3

Body weight of adult SD rats at the beginning (M1) and at the end (M2) of the experiment 3 months later

Full size image

Social interaction test

The results of the test are presented for the pairs of animals. The latency to first contact between animals was the same between all the groups of rats (H(3,17) = 2.93, p = 0.4), but rooibos tea consumption significantly decreased total time spent on social contact (both passive and active) during the trial (H(3,17) = 8.14, p = 0.04). The control animals (Con = 199.0 ± 9.28 s) shared together more time than other animals (AL1: 133.0 ± 14.75 s, p < 0.05 vs. Con, MW test; AL2: 141.5 ± 18.77 s, p < 0.06 vs. Con NS, MW test; AL4: 133.4 ± 10.7 s, p < 0.05 vs. Con, MW test) (Fig. 4a). There were also differences in total time of active social interactions (sniffing, mounting, allo-grooming and following) between the groups of rats (H(3,17) = 8.28, p = 0.04). The controls (Con = 153.0 ± 7.36 s) interacted longer than other animals (AL1: 108.5 ± 10.94 s, p < 0.05 vs. Con, MW test; AL2: 104.5 ± 7.19 s, p < 0.05 vs. Con, MW test; AL4: 119.2 ± 10.14 s, p < 0.06 vs. Con NS, MW test) (Fig. 4b). The frequency of interactions did not indicate significant group-to-group differences (H(3,17) = 5.16, p = 0.16), but the tendency to increase the number of social behavior episodes was seen in controls (Con = 18 ± 2.74) when compared to rooibos-treated rats (AL1: 13.25 ± 2.29; AL2: 11.75 ± 1.2; AL4: 12.8 ± 0.73). Time spent on movements during trial was the same in all groups of animals (H(3,17) = 1.05, p = 0.79; Con: 73.25 ± 13.0 s; AL1: 98.75 ± 13.37 s; AL2: 85.0 ± 15.97 s; AL4: 87.2 ± 17.17 s). Within the trial period, any aggressive behavior was observed between the rats. Also, any episodes of defecation happened in the arena during the testing time.

Fig. 4
figure 4

Social interaction test. Rooibos-administered rats spent less time than controls on: a total contact, b active social interactions. *p < 0.05 (vs. Con, Mann–Whitney test)

Full size image

Discussion

The present experiment focuses on hypothalamus as a crucial brain structure that regulates body homeostasis by affecting endocrine system performance (e.g., hypothalamic–pituitary–adrenal/gonadal axis) or by its own acting, recognized here as regulation of motivation-related behavior [5, 6].

Rooibos tea affects the behavior of rats in social interaction test

In the present research, long-term oral administration of rooibos infusions to adult male rats affected social behavior of animals. Hypothalamus is subjected to the impact of brain structures as prefrontal cortex, hippocampus, amygdala, and brainstem that deliver different sensory inputs as well as to the influence of gonadal or posterior pituitary hormones. This brain structure, which contains anatomically diverse areas including periventricular region involved principally in autonomic and endocrine adjustment, as well as medial and lateral zones responsible for motor and behavioral control, plays a role in coordination of social interactions, reproductive behavior, and aggression control [12,13,14]. Hypothalamus monitors internal body state, and its signaling is projected to such brain areas as periaqueductal gray (controlling of motivated and defensive behavior) or dopaminergic neurons of the ventral tegmental area and amygdala (regulating the goal-directed behavior) with subsequent harmonization of motivational processes by prefrontal cortex [13].

In our experiment, in social interaction test, the tea-administered rats showed decreased time spent both on contact and on active interactions, yet the initiation of social contact (measured as latency to first contact) as well as frequency of contact episodes did not differ significantly in all groups of animals. This suggests some social behavior deficits (decreased social interest), that does not have to be connected necessarily with social fear but could be related either to pro-conflict attitude or increased exploratory drive [9]. In our observation, the rooibos-drinking rats did not reveal any aggressive behavior either in their cages or in the course of behavioral testing. File and Seth [12] describe a decrease in social interaction as indicative of an anxiogenic effect, and vice versa an increase in active social contact without enhanced motor activity is linked with anxiolytic effect. The stress level of all animals in the social test, if correlated with the defecation rate, seemed to be the same between the groups. They presented also comparable locomotor activity. It should be also noticed that the low-lighted and familiar arena of the social test apparatus in our experiment, as the least imminent, support active interactions being the most sensitive to reveal any anxiety behavior of rats and anxiogenic effects of tested compounds, whereas bright light and unfamiliar arena are more sensitive to find their anxiolytic effects [12, 15]. Thus, the results of social test conducted in aversive conditions could add more to the characteristic of rooibos tea.

Moreover, when taking into account the results of previous research in the hole-board showing an increased exploration and significantly declined thigmotaxis in rats consuming rooibos tea [4], it cannot be excluded that shorter contact and lower active social interaction was rather due to their greater inclination to survey the apparatus and non-social targets than fear. The rats did not delay the first contact with the partner, and revealed no typical for anxiety escapes, freezing or alarm cries suggesting that they did not develop social fear. Nevertheless, the potential anxiogenic effects of chronic rooibos tea consumption need future evaluation in additional behavioral testing as, e.g., elevated plus-maze, and social avoidance in social preference-avoidance test.

Rooibos tea changes serotonin and glutamate content in hypothalamus

In the current experiment, the alterations of some hypothalamic neurotransmitters estimated in whole the structure were seen—the rats receiving the greatest concentration of the infusion had significantly lower concentration of serotonin and increased its turnover expressed as 5-HIAA/5-HT ratio. The concentration of glutamic acid was significantly increased in rats drinking the 2:100 infusion.

A reduction in brain serotonin content is described to correlate with decreased response to social clues [16]. Rats treated with para-chloroamphetamine, a potent serotonin neurotoxin, revealed smaller 5-HT concentration in hypothalamus, and developed decreased social behavior regarded as diminished number of social acts, i.e., sniffing, following, and contacting [17]. These observations are consistent with our behavioral testing showing some social interest decline in FRHT-administered rats.

Serotonergic hypothalamic signaling is linked also with aggressive and sexual behavior in male rats [18, 19]. Greater serotonin activity due to 5-HT precursors, reuptake inhibitors or 5-HT 1A/1B-receptor agonists can reduce aggressive behavior in rodents [20]. Glutamic acid is regarded as main excitatory neurotransmitter in the neuroendocrine hypothalamus, and “hypothalamic attack area” presents dense glutamatergic activity [21, 22]. A decrease in serotonin as well as increase in glutamate content seem to facilitate the expression of impulsive rage [23, 24], but such behavior was not observed in our experiment.

Rooibos tea does not implicate body weight of rats

Hypothalamus is also engaged in eating performance, which is substantially regulated both by the hypothalamic and brainstem homeostatic energy maintenance as well as by striatal reward systems affecting motivational features of food consumption [25]. The homeostatic and hedonic circles are anatomically and functionally coupled pointing to the significant regulatory role of neurotransmitters serotonin and dopamine, respectively.

Serotonergic neurotransmission in the central nervous system and modulation of numerous subtypes of 5-HT receptors showed their role in the regulation of eating behavior and long-term body weight [26]. Hypothalamus and brainstem are regarded as the essential brain structures of the homeostatic regulation of food intake. Meal induces the activity of serotonergic neurons in the dorsal raphe nucleus, and hypothalamic serotonergic signaling is linked with anorexigenic stimuli [26, 27]. Median raphe nucleus projects to some structures in the brain reward system, thus, may be involved in the motivational control of eating behavior [28]. Serotonin is able to affect food intake via activation of the anorexigenic melanocortin system involving 5HT2c receptors and by inhibition of the orexigenic neuropeptide Y-dependent system [26]. In animal models of obesity, baseline serotonin release in the hypothalamus is decreased [29]. Meal-induced decrease of hypothalamic serotonin release occurs early due to high-fat feeding, and it worsens over time until a complete absence of food-stimulated release [30]. It was also described that reduced content of serotonin in the hypothalamus results in hypothermic effects, whereas increased serotonin concentration activates post-synaptic 5-HT2 receptors and produces hyperthermic effects (due to an imbalance in heat production and loss) [31].

Furthermore, feeding is associated with rapid release of glutamate in the mediobasal hypothalamus [32] and administration of glutamate and GABA receptors agonists into hypothalamic nuclei stimulates food consumption [33].

Also, dopaminergic transmission could act as incentive signaling that orients attention of animals toward food seeking, increases the significance of food-related impulses, and potentiates efforts aimed at obtaining food [34]. It can be mentioned here that in our previous research [1], long-term rooibos tea administration led to a significant increase in striatal dopamine.

Nevertheless, in the current experiment, the FRHT-dependent alteration of hypothalamic neurotransmitters concentration, i.e., reduced 5-HT in AL4, and increased glutamate in AL2 rats, was not accompanied by increased body mass of animals indicating to other coexisting mechanisms that are able to counteract weight gain.

Rooibos tea may affect HPA axis via hypothalamic 5-HT

In the hypothalamus, serotonin activates the hypothalamic–pituitary–adrenal (HPA) axis by 5-HT2c receptor agonism [35], thus may implicate hormonal and behavioral outcomes (e.g., anxiety, affective dysregulation) [36]. Corticotropin-releasing hormone (CRH) is synthesized in paraventricular hypothalamus (PVH) and then released into the hypophyseal portal circulation stimulating the release of adrenocorticotropin (ACTH) from the anterior lobe of the pituitary, and subsequently corticosterone/cortisol (CORT) from the adrenal cortex. Reduction of the serotonin precursor or transporter reduces cortisol level, while potentiated serotonergic signaling increases plasma concentrations of CORT [37].

Long-term rooibos tea consumption may decrease the serotonin content in the hypothalamus suggesting that it can affect the activity of HPA axis. This finding is consistent with the results of in vivo studies showing decreased corticosterone and deoxycorticosterone levels as well as the CORT/testosterone ratio following rooibos administration [38]. In vitro testing presented possible A. linearis mechanisms of decreased glucocorticoid biosynthesis by inhibition of 11β-hydroxysteroid dehydrogenase type 1 (11βHSD1) [38, 39]. Hypothalamus is a part of brain network of stress-related neurocircuits. Stressors actuate serotonin neurons activity and augment extracellular 5-HT levels in the dorsal raphe nucleus, subsequently enabling the impact on HPA axis. The lower activation of HPA may support the decreased risk of glucocorticoid excess-related effects, e.g., increased feeding behavior, obesity, or mood lability [40] suggesting benefits of rooibos infusions drinking.

Hypothalamus covers not only the HPA axis that integrate the neuro-endocrine-immune responses to stress, but also other hypothalamic nuclei and systems crucial for the development of the depression symptoms, including alterations of circadian rhythm, reward feelings, disturbed sexual and cognitive ability; however, there is no facile straight interrelation between the monoaminergic systems and the depression incidents [41].

Glutamic acid and GABA play a major role in central integration of HPA stress responses. Glutamate activates the HPA axis, presumably by way of hypothalamic and brainstem projections to the PVN.

Rooibos tea alters hypothalamic BDNF–TrkB pathway

A. linearis extracts affected in our experiment the BDNF–TrkB pathway signaling in the hypothalamus. However, the concentration of brain-derived neurotrophic factor (BDNF) was not changed but BDNF receptor (tropomyosin receptor kinase B—TrkB) content was significantly decreased in all the rats drinking rooibos tea.

Brain-derived neurotrophic factor is a member of growth factors and binds to tropomyosin (tyrosine) receptor kinase TrkB. BDNF and its receptors were found across the central nervous system as well as in many structures of hypothalamus, both in neurons and glial cells [6]. Activation of TrkB triggers multiple intracellular signaling pathways including phospholipase PLCγ (production of diacylglycerol and an increase in intracellular calcium), phosphoinositide PI-3-kinase (anti-apoptotic effects), and MAP/ERK cascade (regulation of protein translation influencing cell survival). During development, BDNF supports dendritic growth, induces axon elongation and branching, increases cell survival as well as neuronal plasticity. It also exerts many functions in adult brain affecting neurogenesis, excitatory and inhibitory neurotransmission, modulating pre- and post-synaptic activity being released in either constitutive or activity-dependent way [6]. Deletion of either the TrkB or Bdnf gene leads to cell atrophy, dendritic degeneration, and neuronal loss, as shown in the excitatory neurons of the dorsal forebrain [42].

BDNF is able to affect the behavior of experimental animals [6]. Decrease of BDNF in the ventromedial hypothalamus (VMH) may reduce locomotion in mice [43] or not alter it [44]. Simultaneously an increase of BDNF in the PVH increases locomotor activity [45]. BDNF signaling is also important in social behavior. BDNF lost males show generally increased aggression [46, 47]. Some differences in observed level of aggressive behavior were linked to different Bdnf transcripts contributing to total BDNF content [48], area of hypothalamus affected [44] or by cell-autonomous regulation through TrkB signaling [49]; however, the detailed effect of receptor loss requires further examination to better understanding of how BDNF/TrkB pathway performs in aggressive behavior. A downregulation of TrkB receptors in the hypothalamus and unchanged BDNF content were found in mice with diminished social empathy, selected as resistant to emotional contagion in social modulation of pain response testing [50]. This remains in line with the results of our experiment, in which a TrkB deficiency, and preserved level of BDNF in whole hypothalamic samples from tea-drinking animals were accompanied by deficits of social contact and active interactions. Also lack of BDNF increase corresponded with no signs of aggression within our animals.

Moreover BDNF/TrkB pathway in hypothalamus performs an uppermost importance in the central regulation of energy balance [51]. The PVH and VMH are regarded as important brain areas that express BDNF to decrease food intake and support energy expenditure [43, 52]. BDNF/TrkB favors satiety and holds body weight and energy balance in the PVH via melanocortin 4 receptor (MC4R) signaling, in the VMH via leptin and glucose signaling, and in the dorsomedial hypothalamus (DMH) via unclear mechanisms. Known sources point to hypothalamic BDNF deficiency or TrkB deletion leading to increased body weight and hyperphagia [44, 51]. Liao et al. [53] showed that DMH neurons expressing TrkB are a population of neurons that are necessary and sufficient to suppress appetite. Mutations in the BDNF or the TrkB-encoding NTRK2 gene have been found to cause severe obesity in humans and mice. An et al. [52] demonstrated that PVH is a main site where TrkB signaling decreases food intake, as well as that deletion of Ntrk2 gene for TrkB within this structure leads to severe hyperphagic obesity. Acute stimulation of BDNF neurons in PVH promotes negative energy balance and weight loss [45].

In our experiment, the differences in mean body mass in the groups of rats both at the beginning and the end of the experiment were not observed suggesting that, in this case, the decrease in TrkB signaling was not sufficient to disturb central metabolic control in the hypothalamus. It should be also noticed that TrkB content in rooibos-tea-drinking rats was estimated in the whole hypothalamic structure; thus, possible differences in expression of TrkB in PVH could remain unnoticed.

Conclusion

Summarizing, it was shown that long-term “fermented” rooibos tea consumption was able to affect social behavior of Sprague–Dawley adult male rats in terms of decreased social interest but increased their exploration of non-social clues. Social deficits, expressed as diminished total time of contact as well as active social interaction, need more detailed evaluation whether they may be supported by increased anxiety. In behavioral evaluation, no aggressive activity was seen.

Neurochemical investigation exerted that in the hypothalamus, FRHT alters primarily the serotonergic, glutamatergic, and BDNF/TrkB pathways. Alteration in social interactions could be linked with hypothalamic serotonin decline as well as lower TrkB signaling. Decreased 5-HT and TrkB content suggested positive effect on central energy balance, yet the main body mass of animals in the experiment remained unaffected. On the other hand, the reduced hypothalamic serotonin signaling anticipated the influence on HPA axis and possible diminution of plasma corticosterone level with subsequent behavioral aftermath and lower obesity risk. The more detailed explanation of the influence of A. linearis infusions on behavior and central energy maintenance requires further research.