Background
para-Tyramine (
p-TA) is one of the trace amines derived from tyrosine and present in fermented foods such as cheese [
1,
2]. In the brain,
p-TA acts as a neuromodulator, which supports neuronal actions by several neurotransmitters such as
l-glutamate and norepinephrine [
3,
4]. In addition,
p-TA binds to trace amine-associated receptor 1, which is a G protein-coupled receptor and is a therapeutic target for schizophrenia [
5]. It has been reported that, after activating the receptor, serotonergic and dopaminergic neuronal actions via serotonin receptor 1A and presynaptic dopamine D2 receptor, respectively, are reduced [
6‐
10]. Since
p-TA has a variety of roles in neuronal signal transduction, it is important to understand the homeostatic systems governing the
p-TA concentration in the brain.
It is considered that the neural concentration of
p-TA is maintained by a balance between production and clearance and there are reports that the activity of in vivo blood-to-brain transport of
p-TA is low [
11,
12]. In addition, Faraj et al. [
13] have shown that the
p-TA concentration in the cerebrospinal fluid (CSF) of dogs is approximately 2.6-fold lower than that in plasma although
p-TA is produced in the brain. Therefore, it is conceivable that the
p-TA clearance from the brain is relatively greater compared with that of
p-TA production. Regarding the metabolic pathway of
p-TA, it is known to be metabolized to octopamine and
p-hydroxyphenylacetate via dopamine β-hydroxylase and monoamine oxidase, respectively [
14]. Brain barriers are involved in the clearance of several neural compounds: the blood-brain barrier (BBB), which is formed by brain capillary endothelial cells, directly separates the brain from the circulating blood [
15]. In addition, the cerebrospinal fluid (CSF) is separated from the circulating blood by the blood-CSF barrier (BCSFB) composed of choroid plexus epithelial cells [
15]. Recent reports have shown that these barriers play a role in brain-to-blood transport of endogenous anionic and cationic compounds such as homovanillic acid and histamine [
16,
17]. Therefore, there is a possibility that the BBB and/or BCSFB transport
p-TA from the brain to the circulating blood and play a role in the
p-TA clearance system in the brain.
These cells express a variety of plasma membrane transporters which contribute to compound transport across the BBB and BCSFB. In rat brain capillary endothelial cells, mRNAs of serotonin transporter [SERT/solute carrier (Slc) 6a4], organic cation transporter 1–3 (OCT1–3/Slc22a1–3), multidrug and toxin extrusion 1–2 (MATE1–2/Slc47a1–2), organic cation/carnitine transporter 1–2 (OCTN1–2/Slc22a4–5), and plasma membrane monoamine transporter (PMAT/Slc29a4) have been reported to be expressed [
18,
19]. Among these transporters, it has been demonstrated that PMAT is involved in 1-methyl-4-phenylpyridinium (MPP
+) efflux transport at the BBB, at least in part [
18]. Regarding the BCSFB, the mRNA expression of these transporters has also been reported [
18]. In addition, our previous studies have shown that PMAT and OCT3 are involved in the elimination of histamine and creatinine, respectively, from rat CSF [
16,
20]. Taking these reports into consideration, it is possible that several cationic transporters at these brain barriers take part in the clearance of cationic compounds from the brain. It is known that
p-TA is a substrate of plasma membrane transporters, such as SERT, OCT3, PMAT, and dopamine transporter (DAT/Slc6a3) [
21‐
24] and, thus, it is hypothesized that
p-TA in the brain is eliminated via these plasma membrane transporters at the brain barriers.
To increase our knowledge of the homeostatic mechanism(s) governing the cerebral p-TA concentration, the purpose of this study was to examine the role of the brain barriers in p-TA clearance from the brain. To examine p-TA elimination across the BBB, we used an intracerebral microinjection technique. To evaluate BCSFB-mediated p-TA efflux transport, in vivo intracerebroventricular administration and ex vivo transport studies using isolated choroid plexus were performed. In addition, the properties of p-TA uptake were examined using an in vitro model of rat BCSFB, conditionally immortalized rat choroid plexus epithelial cells (TR-CSFB3 cells).
Discussion
In this study,
p-TA elimination from the brain and CSF was investigated in rats (Fig.
1). To clarify the involvement of the BCSFB in
p-TA elimination from the CSF, a
p-TA transport study using isolated rat choroid plexus, which forms the BCSFB, was carried out (Fig.
2 and Table
1). Since carrier-mediated transport properties were observed in this study using rat choroid plexus,
p-TA transport characteristics were examined in in vitro rat BCSFB model cells (Figs.
3,
4, and Table
2).
The percentage of [
3H]
p-TA remaining in the ipsilateral cerebrum tended to be reduced in a time-dependent manner (Fig.
1a), although there was no significant difference between the values at examined time points. It is known that
p-TA is converted via enzymes in neural cells to several metabolites, such as octopamine [
14]. Thus, it is suggested that
p-TA and its metabolites are eliminated to a degree from the brain across the BBB. However, this
t
1/2, as a reference record (178 min), was 62-fold longer than that of the residual concentration of [
3H]
p-TA in the CSF (Fig.
1b). These results show that
p-TA elimination from the CSF makes a major contribution to cerebral
p-TA clearance relative to that from the brain parenchyma or interstitial fluid. In addition, the
CL
CSF for [
3H]
p-TA was 3.5-fold higher than that of [
14C]
d-mannitol, a marker compound for bulk flow of CSF [
26,
32]. Taking the inhibitory effect of
p-TA elimination from the CSF by excess unlabeled
p-TA (Fig.
1c) into consideration, this suggests that carrier-mediated transport system(s), which are distinct from CSF bulk flow, are involved in
p-TA elimination from the CSF.
After intracerebroventricular administration, it is possible that [
3H]
p-TA is distributed to the brain parenchyma and/or taken up into ependymal cells and choroid plexus epithelial cells [
33]. It is known that transporters at the BCSFB participate in the elimination of several compounds from the CSF [
15]. We have shown that [
3H]
p-TA was time-dependently taken up into ex vivo isolated rat choroid plexus (Fig.
2) and in vitro choroid plexus epithelial cells, which form the BCSFB (Fig.
3). In addition, this [
3H]
p-TA uptake was significantly inhibited by unlabeled 10 mM
p-TA (Tables
1,
2). Since the initial distribution volume of
p-TA was found to be 159 μL (Fig.
1b), the concentration in rat CSF of unlabeled
p-TA after a 10 μL intracerebroventricular microinjection of 75 mM unlabeled
p-TA was calculated to be 4.7 mM, which is similar to that in the in vitro self-inhibition studies. Therefore, it is suggested that some transport systems at the BCSFB take part in carrier-mediated elimination of
p-TA from the CSF.
Since the apical membrane of the choroid plexus faces the [
3H]
p-TA-containing buffer in the uptake study using isolated choroid plexus, it appears that the characteristics of [
3H]
p-TA uptake by rat choroid plexus reflects the CSF-to-cell transport direction at the BCSFB. In this study, the choroid plexus from the lateral ventricle was used in the
p-TA transport study. Although there are known regional differences in physiological and transport functions between the choroid plexuses from the lateral, third, and fourth ventricles [
34,
35], Ogawa et al. [
26] have reported that the in vivo uptake rate which is extrapolated from in vitro benzylpenicillin transport in rat choroid plexus from the lateral ventricle, is in good agreement with that of saturable in vivo elimination of benzylpenicillin from the CSF.
p-TA transport across the apical membrane of the isolated rat choroid plexus was found to be 7.14 μL/(min rat) since the total volume of rat choroid plexus has been reported to be 6 μL (7.14 μL/(min rat) = 1.19 μL/μL ChP × 6 μL/rat) [
26]. Regarding the polarity of TR-CSFB3 cells after seeding onto a culture plate and glass slide, it has been reported that membrane protein localization onto the plasma membrane of choroid plexus epithelial cells is mostly retained [
29]. As the surface area of TR-CSFB3 cells and rat lateral ventricle choroid plexus epithelium has been reported to be 20 cm
2/mg protein [
18] and 75 cm
2/rat [
36], respectively, the initial
p-TA uptake clearance estimated from the uptake study using TR-CSFB3 cells was calculated to be 11.0 μL/(min rat) (= 2.93 μL/(min mg protein) ÷ 20 cm
2/mg protein × 75 cm
2/rat). This estimated initial
p-TA uptake clearance from the transport study using TR-CSFB3 cells is consistent with that using isolated rat choroid plexus. Taking these findings into consideration, there is a high possibility that the characteristics of
p-TA uptake by TR-CSFB3 cells reflect the
p-TA transport properties of the apical membrane of the BCSFB.
These BCSFB-mediated [
3H]
p-TA uptake clearance values obtained from the uptake studies of isolated rat choroid plexus (7.14 μL/(min rat) and TR-CSFB3 cells (11.0 μL/(min rat)) were 18.5 and 28.5% of the in vivo total [
3H]
p-TA elimination clearance from rat CSF (38.6 μL/(min rat), Fig.
1b). As the elimination clearance of [
14C]
d-mannitol from rat CSF was found to be 10.9 μL (Fig.
1b), 28.2% of the total [
3H]
p-TA clearance would reflect the CSF bulk flow and diffusion to the brain parenchyma. As the other pathways for
p-TA elimination from the CSF, the remainder of which corresponds to (43.3–53.5%), the incorporation of
p-TA into the neural cells facing the cerebroventricles, including the ependymal cells, is a possibility. Further analysis to check the radioactivities in choroid plexus and neural cells surrounding the cerebroventricle could help these cells contribute to
p-TA elimination from the CSF. Nevertheless, it is considered that the BCSFB is involved in the carrier-mediated
p-TA elimination from the CSF, at least in part, since an inhibitory effect of unlabeled excess
p-TA on [
3H]
p-TA uptake by isolated rat choroid plexus (Table
1) and TR-CSFB3 cells (Table
2) was observed.
p-TA uptake by TR-CSFB3 cells exhibited saturable and non-saturable kinetics (Fig.
3b). The clearance of the saturable process of
p-TA transport (
V
max/
K
m) was found to be 2.09 μL/(min mg protein), which is 2.1-fold higher than that of the non-saturable process [
K
d, 0.978 μL/(min mg protein)]. This result indicates that the transporter-mediated process makes a major contribution to apical
p-TA transport at the BCSFB. It has been reported that human OCT3 and PMAT accept
p-TA as a substrate with a
K
m of 0.281 and 283 μM, respectively [
21,
22]. However, the
K
m value of
p-TA uptake by TR-CSFB3 cells (3.48 mM) was inconsistent with these values. mRNA expression of OCT1–2, OCTN1–2, and MATE in addition to OCT3 and PMAT as typical organic cation transporters has been reported [
18] but no attenuation of [
3H]
p-TA transport into TR-CSFB3 cells and/or isolated rat choroid plexus was observed in the presence of these transporter inhibitors, such as MPP
+, TEA,
l-carnitine, pyrimethamine, and cimetidine (Tables
1,
2). In summary, these lines of evidence suggest that typical organic cation transporters which are expressed in the BCSFB are not involved in
p-TA transport across the BCSFB. Other plasma membrane transporters for
p-TA are known: i.e. several Na
+-, Cl
−-dependent Slc molecules, such as DAT and SERT. However, in the uptake study using isolated rat choroid plexus and TR-CSFB3 cells, [
3H]
p-TA uptake was not attenuated in the absence of extracellular Na
+ and Cl
− (Tables
1,
2). The
K
m value of human DAT- and SERT-mediated
p-TA transport has been reported to be 1.7 and 52.7 μM, respectively [
23,
24], which is different from the
K
m value obtained for
p-TA uptake by TR-CSFB3 cells. In addition, [
3H]
p-TA uptake by both isolated rat choroid plexus and TR-CSFB3 cells was not significantly changed in the presence of serotonin, a substrate of SERT (Tables
1,
2). Consequently, we conclude that typical organic cation transporters and the Na
+- and Cl
−-dependent Slc family do not play a role in apical
p-TA transport at the BCSFB.
We showed that
p-TA transport into TR-CSFB3 cells exhibited an oppositely-directed H
+-gradient (Fig.
4). Cationic drug transport system(s) in several tissues and blood-central nervous system barriers have been proposed [
37‐
43]. It has been reported that human intestinal epithelial cells transport several cationic drugs, such as bisoprolol and metoprolol, in an extracellular pH-dependent manner [
37,
38]. Our previous reports have shown that nicotine is taken up into rat hepatocytes and lung via unidentified transport system(s) which recognize several cationic drugs and exhibit a H
+/substrate antiport behavior [
39,
40]. Moreover, it has been shown that H
+/substrate antiport system(s) involve blood-to-brain transport across the BBB of cationic compounds and drugs, such as pyrilamine, oxycodone, and nicotine, although the molecular identification of the system(s) has not been reported [
41‐
43].
p-TA uptake by isolated rat choroid plexus (Table
1) and TR-CSFB3 cells (Table
2) was significantly inhibited by cationic drugs which also inhibited the uptake of pyrilamine, oxycodone, and nicotine by an in vitro BBB model cell line [
41,
42]. Thus, it is suggested that cationic drug-sensitive transport system(s), similar to the systems in the BBB, are present on the apical membrane of the BCSFB and are involved in
p-TA elimination across the BCSFB from the CSF.
In patients with Parkinson’s disease and depression, monoamine oxidase inhibitors, such as selegiline, are regularly prescribed. It is widely known that patients who take
p-TA-enriched foods exhibit neural side-effects such as migraine when receiving pharmacotherapy with monoamine oxidase inhibitors [
44,
45]. As one reason for the above side-effects, the activation of noradrenergic signal transduction has been proposed by monoamine oxidase inhibition and, thus, an increase in the neural
p-TA concentration. Many monoamine oxidase inhibitors, including selegiline, are cationic and lipophilic drugs. This study has demonstrated that
p-TA elimination across the BCSFB from the CSF is inhibited by cationic and lipophilic compounds (Tables
1,
2). Thus, it is possible that the administration of cationic drugs causes an increase in
p-TA in the brain via the inhibition of BCSFB-mediated
p-TA elimination and exacerbates neuronal actions induced by excess
p-TA. Further studies on the identification of
p-TA transport molecule(s) which also recognize cationic drugs are needed, since this could help improve the pharmacotherapy of these patients in addition to increasing our understanding of the role of the BCSFB in homeostasis of
p-TA levels in the brain. Moreover, it is considered that our findings about the in vitro characteristics of tyramine transport at the BCSFB (Tables
1,
2), such as inhibition by cationic drugs and promotion by the presence of several compounds (i.e., tyrosine, choline, MPP
+, and TEA), are helpful for identifying the responsible molecule(s) for tyramine transport at the BCSFB.