The association of changes in exogenous or endogenous GH with altered CNS GHR gene expression does imply a central action of GH. Although the most economical hypothesis is that GH directly regulates itself via its own receptors, it is always difficult to exclude indirect effects of GH, either on whole body growth or on metabolism causing secondary changes via IGF-1. This can be partially overcome by demonstrating that direct administration of GH at doses below those causing peripheral changes still exerts central effects, but this then raises the question of how circulating GH can gain access to brain GHRs.
There is some evidence that GH can enter cerebral spinal fluid (CSF), at least at high secretory rates (55), and that exogenous GH can affect CSF endorphin levels (56). As mentioned earlier, GH might enter the CNS by receptor-mediated or GHBP-mediated transport in the choroid plexus as is suggested for prolactin (57). The choroid plexus has been shown to contain GHR binding sites in humans (58) and in the rat, GHR immunore-activity (5) and GHR gene transcripts (59) have been identified in this region. Alternatively, GH could reach some areas of the hypothalamus by retrograde transport from the pituitary up the pituitary stalk (60). Studies carried out on the ability of iodinated GH to cross the blood-brain barrier have mainly demonstrated low levels of GH uptake (61-63) apart from one earlier study in rat brain (64), although this could change if the blood-brain barrier is compromised, either from disturbances in the pituitary stalk vasculature secondary to tumor growth or to other effects on brain capillary permeability (65).
The possibility of an extrapituitary central source of GH cannot be completely excluded because GH transcripts as well as GH immunoreactivity have been detected in the rat brain (62,66). Although the existence and regulation of such central GH is still controversial, placental expression of GH variants is well established (67), lymphoid derived cells have been shown to produce GH (68), and recent studies also implicate mammary tissue as an alternative source in several species, including human (69,70). Again the presence of GH-like material in the CNS is not widely accepted and may not correspond to authentic pituitary GH, which could explain discrepant results with antibodies of different specificities. Irrespective of its source, the important question in this context is whether GH or GH-like molecules are present at the right sites and in sufficient amounts with the ability to activate or inhibit GHRs; this still remains to be established. On the other hand, the amounts of GH that would need to reach the CNS need not be large. Animal models bearing an hGH transgene with expression in the CNS show suppressed endogenous GHRH and GH expression and dominant dwarfism (71-73) with a net decrease in peripheral circulating GH, despite very small amounts of GH locally produced from such transgenes in the CNS.
Was this article helpful?