Abstract
Anti-Müllerian hormone (AMH) is produced by male embryos and triggers the regression of the Müllerian ducts, the precursor of the uterus. After birth, AMH is expressed by the gonads of both males and females, where it is a local regulator of gonadal function. However, AMH is also found in the blood, suggesting that AMH may have undetected hormonal actions. My thesis relates to this by examining AMH signalling, as localisation of the AMH receptors is the first step towards understanding the potential functions of AMH.
AMH is a member of the transforming growth factor ″ (TGF-″) superfamily, a large family of cytokines. The members of the TGF-″ superfamily classically signal via their C-terminal domain obtained after cleavage of the pro-molecule. Signalling involves two types of receptors, the type I and type II receptors. Both receptors are shared by several members of the family. However, the AMH receptor type II (AMHRII) appears to be specific to AMH, and its localisation may therefore be a clue to the functions of AMH. AMH may also signal in the absence of AMHRII as the amyloid precursor like protein 2 (APLP2), a member of the amyloid precursor protein family involved in Alzheimer's disease was recently described as a potential new AMH receptor. Part of the N-terminal region of AMH (npAMH) binds to APLP2, which recruits the extracellular regulated kinase (ERK) 1/2. Consistent with this, we detected proAMH and the cleaved n-cAMH (fAMH) in the blood. The presence of APLP2 in the brain suggests that APLP2 may be a receptor for AMH in the central nervous system. I therefore investigated whether AMH can signal via APLP2 in cultured embryonic hippocampal neurons by studying the phosphorylation of ERK1/2. The neurons were exposed chronically to fAMH and acutely to fAMH, npAMH and cAMH. The recruitment of the ERK1/2 intracellular signalling pathway by endogenous AMH was also examined using Amh-/- mice. Unlike the positive control, none of the forms of AMH increased the amount of phospho-ERK1/2 suggesting that the regulation of the brain by AMH does not involve APLP2. The initial report of APLP2 as a receptor for AMH used sperm, with nAMH increasing sperm survival. I therefore incubated mouse sperm with cAMH and npAMH but did not observe altered survival of sperm.
Two natural AMHRII splice-variants were recently characterised. One variant lacks part of the ligand binding domain and the other lacks the intracellular serine/threonine kinase domain involved in the recruitment of the signalling cascade. This raises the possibility that they might inhibit AMH signalling. Using a luciferase-based assay, I examined the influence of the AMHRII splice-variants on AMH signalling. Both splice-variants were natural antagonists of the AMHRII. The extend of the inhibition depended on the AMHRII/receptor variant ratio and on the concentration of AMH. The expression of the AMHRII splice-variants may therefore regulate AMH signalling.
Before puberty, when most of the brain development occurs, only the testes produce AMH. This results in AMH-dependant sex-biases in the brain. Both AMH and testosterone are produced by the testes, where they act as local regulators. The effect of AMH on the brain may therefore be due to a dual regulation by AMH and testosterone. This was investigated by studying an androgen-dependant sexually dimorphic nucleus of the spinal cord in AMH-deficient mice. The absence of AMH did not influence the number or size of the androgen-dependant neurons, indicating that AMH and testosterone regulate different populations of neurons.
In summary, AMH is postulated to have multiple functions, which occur at different concentrations and in different contexts. This raised the possibility that AMH signals through multiple receptors. The thesis implicates splicing variants of AMHRII as one of the factors that creates diversity of AMH signalling and provides arguments against the putative role of APLP2 as a AMH receptor.