Brain homeostasis for riboflavin

The concentration of total riboflavin in brain, unlike
in liver and kidney, is maintained relatively constant
even in the face of severe vitamin B2 defi ciency or
after massive doses of intravenous ribofl avin. As in
other tissues, ribofl avin that enters the brain is enzymatically
phosphorylated to FMN, which can then
be converted to FAD. In rat brain, more than 90% of
the total ribofl avin is present as FAD and FMN. About
10% of the total ribofl avin in the brain of normal rats
turns over per hour.

The concentration of total riboflavin in cerebrospinal
fl uid (CSF) is about 50% of that in plasma.
However, because roughly 50% of the total ribofl avin
in plasma is bound to serum proteins, the concentrations
of unbound vitamin in plasma and CSF are approximately equal. Because CSF is constantly leaving
the central nervous system, ribofl avin must be
continually supplied to the newly formed fl uid.
Ribofl avin, but not FMN or FAD, enters the central
nervous system principally through the blood–brain
barrier (Spector, 1980a). Uptake of [14C]ribofl avin by
rabbit brain slices in vitro took place by a saturable
system that depended on the conversion of accumulated
ribofl avin to FMN and FAD, i.e. intracellular
trapping of ribofl avin (Spector, 1980b). These enzymatic
conversions require ATP. Energy dependency
was shown by the inhibition of uptake by dinitrophenol
and low-temperature (1°C) incubation. The
system was one-half saturated at the normal plasma
concentration of riboflavin (~0.03 μM). This means
that the entry of excessive amounts of ribofl avin from
blood to brain is prohibited.

Spector & Boose (1979) studied the capacity of
the isolated choroid plexus, the anatomical locus of
the blood–CSF barrier, to transport [14C]ribofl avin.
With concentrations of [14C]ribofl avin of 0.7 μM or
greater in the incubation medium, the choroid plexus
accumulated [14C]ribofl avin against a large concentration
gradient, thus demonstrating active transport.
Uptake did not depend on intracellular binding or
phosphorylation of the vitamin. The half-saturation
concentration (Km) was 78 μM (cf. normal plasma
concentration of ribofl avin of ~0.03 μM), which
means that the system has the potential to transport
high concentrations of ribofl avin before it becomes
saturated. Studies using the isolated choroid plexus do
not show direction of transport.
Spector (1980a) injected [14C]ribofl avin directly
into the ventricular CSF of anaesthetized rabbits.
Some of the ribofl avin entered the brain by a saturable
accumulation system that depended in part on
phosphorylation of the ribofl avin, but the majority
of the labelled vitamin left the CSF extremely rapidly
and was not found in the brain. The disappearance of
the injected ribofl avin means that the choroid plexus
must have actively transported the vitamin from the
CSF into the bloodstream. Direct pictorial evidence of
this property of the choroid plexus was provided by
fl uorescence microscopy (Spector, 1980c).

In conclusion, the controlled entry and exit of ribofl
avin from the central nervous system provide the
means for maintaining total riboflavin homeostasis
in brain cells. The main entry is via the blood–brain
barrier followed by high-affinity saturable uptake by
brain cells and metabolic trapping. The choroid plexus
actively transports ribofl avin from blood into the
CSF, but perhaps more importantly it has the capacity
to transport excess ribofl avin in the reverse direction