The arginine vasopressin (AVP) precursor gene of mammals contains three exons encoding the principal domains of the polyprotein precursor, including vasopressin (exon A), neurophysin (exon B), and glycopeptide (exon C). The AVP precursor (preprohormone) is processed and transported through the endoplasmic reticulum (ER), Golgi apparatus, and secretory vesicles, and finally, mature AVP is secreted from the posterior pituitary into the circulation. The exact steps of these processes during AVP translation and posttranslation events are not yet well elucidated. Defects in peptide processing are associated with several genetic disorders, including central diabetes insipidus (CDI). In the Brattleboro rat with CDI, the mRNA and protein of AVP are present in the hypothalamus, but no circulating AVP is detectable, thus suggesting a processing defect, transport defect, or both. The mutated AVP gene precursor of Brattleboro rat has a deletion of a single base, guanine, in the neurophysin coding region that leads to a frameshift resulting in the loss of the normal stop codon. It has been reported that the mutated precursor is trapped in the ER and does not reach the Golgi apparatus. Recent studies examined AVP secretion in cultured COS cells transfected with various constructs from wild-type and mutated Brattleboro AVP gene precursors. The wild-type in vitro studies demonstrated that intact neurophysin, but not the glycoprotein coding region, is necessary for normal AVP processing and secretion. Next, the results demonstrated that the guanine defect in the neurophysin coding region and the prolonged C-terminus accounted for the processing defect in the Brattleboro rat with CDI. These defects no doubt impair the folding and configuration necessary for normal processing of the AVP gene precursor in the ER. In hereditary CDI in humans, the majority of the mutations have also been shown to occur in the neurophysin coding region. However, in contrast to the recessive defect in the Brattleboro rat, in human CDI, neurotoxicity and denigration of the magnocellular neurons have been observed, and dominant inheritance occurs. Moreover, all mutations are missense, nonsense, or deletions in human CDI rather than the shift in reading frame and preserved neurons that is observed with the Brattleboro rat. Thus, the results from studies in the Brattleboro rat may only be partially applicable to hereditary CDI in humans.