We have identified a mouse PDZ protein that interacts with the activin type IIA receptor (ActRIIA), which we named activin receptor-interacting protein 1 (ARIP1). By using yeast two-hybrid screening, we isolated a cDNA clone of ARIP1 from a mouse brain cDNA library. We detected two forms of ARIP1, ARIP1-long and ARIP1-short, which may be produced by alternative splicing. ARIP1-long had one guanylate kinase domain in the NH 2 -terminal region, followed by two WW domains and five PDZ domains (PDZ1-5). ARIP1-short had a deletion in the NH 2 -terminal region and lacked the guanylate kinase domain. Both forms interacted with ActRIIA through PDZ5. The COOH-terminal residues of ActRIIA (ESSL) agree with a PDZ-binding consensus motif, and ARIP1 recognized the consensus sequence. ARIP1 interacts specifically with ActRIIA among the receptors for the transforming growth factor ␤ family. Interestingly, ARIP1 also interacted with Smad3, which is an activin/ transforming growth factor ␤ intracellular signaling molecule. The mRNA of ARIP1 was more abundant in the brain than in other tissues. Overexpression of ARIP1 controls activin-induced and Smad3-induced transcription in activin-responsive cell lines. These findings suggest that ARIP1 has a significant role in assembling activin signaling molecules at specific subcellular sites and in regulating signal transduction in neuronal cells.

Using yeast two-hybrid screening, we have identified a mouse Postsynaptic density 95/Discs large/Zona occludens-1 (PDZ) protein that interacts with activin type II receptors (ActRIIs). We named the protein activin receptor-interacting protein 2 (ARIP2). ARIP2 was found to have one PDZ domain in the NH 2 -terminal region and interact specifically with ActRIIs among the receptors for the transforming growth factor ␤ family by the PDZ domain. Interestingly, overexpression of ARIP2 enhances endocytosis of ActRIIs and reduces activin-induced transcription in Chinese hamster ovary K1 cells. In addition, immunofluorescence co-localization studies indicated the direct involvement of ARIP2 in the intracellular translocation of ActRIIs by PDZ domain-mediated interaction. Moreover, we have identified that the COOH-terminal region of ARIP2 interacts with Ralbinding protein 1 (RalBP1). RalBP1 is a potential effector protein of small GTP-binding protein Ral and regulates endocytosis of epidermal growth factor and insulin receptors. The studies using deletion mutants of RalBP1 and constitutively GTP and GDP binding forms of Ral indicate that ARIP2 regulates endocytosis of ActRIIs through the Ral/RalBP1-dependent pathway, and the GDP-GTP exchange of Ral is critical for this regulation.Activin, a member of the TGF-␤ 1 superfamily, has a broad range of physiological activities including hematopoiesis, bone morphogenesis, neurogenesis, and hormone action (1-5). These various actions on cell proliferation, differentiation, and apoptosis are dependent upon target cells. Activin transduces its signal via heteromeric complexes composed of two different serine/threonine kinase receptors, termed type I and type II (6 -8). Upon ligand binding, the type II receptor transphosphorylates and activates the type I receptor kinase at the membrane region. Then, the type I receptor cytoplasmic domain interacts with intracellular signaling molecules, Smads, which regulate transcription of selected genes in a cell-specific manner (9, 10).In the current model, the functions of the ActRIIs are limited to ligand binding, type I receptor recruitment, and transphosphorylation. However, it is speculated that there are specific roles of the ActRIIs in activin signaling. For example, the serine/threonine kinase domains of ActRIIs are constitutively activated, and ActRIIs exist as homooligomers (probably homodimers) even in the absence of ligands (11,12). These data suggest the existence of a strict monitoring mechanism for type II receptor kinase activity. Furthermore, multiple forms of ActRIIs exist; two subtypes of activin type II receptors, ActRIIA and ActRIIB, each of which is encoded by individual genes, are known (13,14). Several alternative splicing variants have also been found in each subtype (14 -16). For example, activin type IIA-N receptor, a splicing product of ActRIIA, is specifically expressed in neural cells and is thought to mediate neuronalspecific activin action. In a previous report, we noted the identification of a PDZ protein called a...

The present study was conducted to study the effect of immunoneutralization against endogenous inhibin on FSH, LH, oestradiol-17 beta and progesterone secretion and to investigate the effect of removal of endogenous inhibin on subsequent follicular development in the hamster. After treatment with anti-inhibin serum (inhibin-AS) at 1100 h on day 2 of the oestrous cycle (day 1 = day of ovulation), a marked increase in plasma levels of FSH and a slight increase in plasma levels of LH were noted and pituitary contents of FSH, but not LH, were also increased. In the group treated with inhibin-AS, superovulation occurred on day 1 of the following cycle. Plasma levels of oestradiol-17 beta markedly increased with the increase in the number of ovulations induced by human chorionic gonadotrophin (hCG) as compared with those in control animals. In the second cycle, plasma concentrations and pituitary contents of FSH in the animals given 200 microliters inhibin-AS still showed high values as compared with those in the animals treated with control serum, although superovulation did not occur on day 1 of the third cycle. Plasma concentrations and pituitary contents of LH in the hamster given 200 microliters inhibin-AS tended to decrease as compared with those in control animals during the second cycle. Plasma concentrations of oestradiol-17 beta in the animals treated with 200 microliters inhibin-AS changed in a similar way to controls. A marked increase in plasma concentrations of progesterone was noted on days 1 and 2 of the second cycle in the group receiving inhibin-AS. The twice daily injection of 1 IU hCG during the second cycle to the animals given 200 microliters inhibin-AS induced superovulation on day 1 of the third cycle. These results indicate that circulating inhibin may be an important indicator of the number of developing follicles and may maintain the species-specific number of developing follicles through suppression of FSH secretion in the cyclic hamster. They also suggest that high levels of inhibin slightly suppress plasma levels of LH, indicating that plasma LH may also regulate follicular development in the cyclic hamster.

No abstract

Plasma and ovarian levels of inhibin were determined by a radioimmunoassay (RIA) at 3-h intervals throughout the 4-day oestrous cycle of hamsters. Plasma concentrations of FSH, LH, progesterone, testosterone and oestradiol-17 beta were also determined by RIAs. In addition, hamsters were injected at various times with human chorionic gonadotrophin (hCG) to determine the follicular development. The changes in plasma concentrations of FSH after injection of antisera to oestradiol-17 beta (oestradiol-AS) and inhibin (inhibin-AS) on the morning of day 2 (day 1 = day of ovulation) were also determined. Plasma concentrations of inhibin showed a marked increase on the afternoon of day 1, remained at plateau levels until the morning of day 4, then increased abruptly on the afternoon of day 4 when preovulatory LH and FSH surges were initiated. A marked decrease in plasma concentrations of inhibin occurred during the process of ovulation after the preovulatory gonadotrophin surges. An inverse relationship between plasma levels of FSH and inhibin was observed when the secondary surge of FSH was in progress during the periovulatory period. Plasma concentrations of oestradiol-17 beta showed three increase phases and these changes differed from those of inhibin. Changes in plasma concentrations of oestradiol-17 beta correlated well with the maturation and regression of large antral follicles. Follicles capable of ovulating following hCG administration were first noted at 2300 h on day 1. The number of follicles capable of ovulating reached a maximum on the morning of day 3 (24.8 +/- 0.6), and decreased by 0500 h on day 4 (15.0 +/- 1.1), corresponding to the number of normal spontaneous ovulations.(ABSTRACT TRUNCATED AT 250 WORDS)