Introduction

Diabetes insipidus (DI) is characterized by the kidney inability to concentrate urine leading to consistent hypotonic polyuria greater than 3 liters in 24 hours in adults and persisting even during water deprivation. Three main types of DI can be defined as follows: A. Central DI (CDI) due to a defect in arginin-vasopressin (AVP) synthesis; B. nephrogenic DI (NDI) characterized by kidney resistance to the AVP action; C. gestational DI caused by accelerated AVP degradation.Besides these forms excessive fluid intake can cause DI [1] [2].

Central DI is the most frequent form and can be due to damage at the level of the hypothalamus which might affect the supraoptic or paraventricular nuclei or the osmoreceptors committed to ‘sense’ the alterations in blood osmolality. In rare cases (1-2%) CDI can be due to mutations of the gene coding for the AVP receptor.

On the other hand, AVP resistance due to renal defect characterizes NDI. In NDI the kidney cannot concentrate urine leading to a risk of severe volume depletion, hypernatremia, and hyperchloremia. NDI can be congenital due to mutations in the vasopressin receptor V2R gene causing X-linked NDI accounting for about 90% of congenital NDI [3] or to mutations in the gene coding for the water channel Aquaporin 2 causing autosomal-recessive or autosomal dominant DI responsible for about 10% of congenital NDI characterized by altered expression or trafficking of the AQP2 water channels [4] (full text).

Much more frequent are the forms of acquired NDI associated to a wide range of conditions such as lithium ingestion, urinary obstructions, hypercalcemia and hypercalciuria [5]. 

Finally about 5% of all cases of NDI are genetically unresolved or have unidentified causative mutations [6].

The key advances in the understanding DI and NDI are shown in Figure 1.

History of Diabetes Insipidus: a disease identified about 200 years ago

DI has been described several thousand years later with respect to the similarly named Diabetes Mellitus (DM), a diseases already known in ancient Egypt, Greece and Asia. In fact, its description is date back to 1794, when Johann Peter Frank of the University of Pavia described patients characterized by “long continued abnormally increased secretion of nonsaccharine urine which is not caused by a diseased condition of the kidneys” and introduced the term “diabetes insipidus” derived from the french word “insipide” [7]. According to the historical documents, in 1841 Lancombe described a family with 8 members displaying the symptoms of DI first focusing the attention to the familial features of DI [8]. 

The familiarity of DI was subsequently described in a publication by McIlraith in 1892 entitled "Notes on some cases of diabetes insipidus with marked family and hereditary tendencies" [9]. In the following years it became clear that a defect in the hypothalamus was somehow responsible for the DI. In 1901 Magnus and Shaffer demonstrated that the posterior pituitary extract had a pressor and antidiuretic activities [10]. Few years later in 1913 Fariniand van den Velden successfully used posterior pituitary extracts to treat diabetes insipidus [11] [12].Subsequently Bailey and Ranson, two researchers working in Illinois, described a supraoptico-hypophyseal tract in animals that connects the hypothalamic supraoptic nuclei to the posterior pituitary and showed that an injury to this tract produced DI [13] [14]. In parallel in Europe, Camus and Roussy of the Faculté de Médecine of Paris discovered in dogs that, puncturing the hypothalamus but leaving the pituitary intact, produced polyuria [15]. In summary, the available evidence in the 1920s was conclusive enough to define DI as a disorder of the pituitary gland and actually named this disorder as ‘hypopituitary syndrome’ [16] [17].

Central Diabetes Insipidus is distinct from Nephrogenic Diabetes Insipidus

Regarding the two principal forms of diabetes, mellitus and nephrogenic, in 1670s Thomas Willis, professor of Natural Philosophy at Oxford already noted the difference in taste of urine from polyuric subjects compared with healthy individuals [18] [19] [20]. He used the term diabetes in the generic sense, meaning polyuria, however his observations led to the differentiation of diabetes mellitus from the more rare entity of diabetes insipidus a century later.

In the early 1928, a german scientist DeLange, first observed that some patients with DI did not respond to posterior pituitary extracts. Moreover those families appeared not to have a male to male transmission of the disease [21]. These observations were followed by more accurate analysis by Forssman [22] [23], who established that the kidney had a critical role in those forms of DI resistant to treatment with the posterior pituitary extracts. In an original document Waring described patients with “an unusual syndrome” that presented shortly after birth, characterized by polyuria, polydipsia, fever, and constipation vomiting, high serum Na and Cl, rapid dehydration, and inability to excrete hypertonic urine. He concluded that the condition was caused by “a specific defect in tubular reabsorption of water” and appeared more frequently in boys. This description is consistent with what we know today to be the congenital form of the X-linked Nephrogenic Diabetes Insipidus. In 1947 by Williams and Henry introduced the term “nephrogenic diabetes insipidus” for the congenital syndrome characterized by polyuria and renal concentrating defect but unaffected by vasopressin [24]. They noticed the inheritance pattern realizing that the defect was transmitted by asyntomatic female to male offspring. The authors concluded that the disease was due to a congenital defect in the loop of Henle and the distal convoluted tubule.

Purification and synthesis of vasopressin

In 1955, several years later the successful treatment of diabetes insipidus using posterior pituitary extracts in 1913, du Vigneaud of Cornell Medical College, received the 1955 Nobel Prize in chemistry for the first synthesis of a polypeptide hormone vasopressin [25]. With a seminal work, du Vigneaud purified both oxytocin and vasopressin by countercurrent distribution and then chemically synthesized oxytocin in 1953 and vasopressinin 1954 [25]. Further distribution studies on the oxytocic hormone of the posterior lobe of the pituitary gland and the preparation of an active crystalline flavianate [26] (full text) [27] (full text).

Within this scenario of this series of truly landmark achievements, the 2013 marked the 100^th anniversary of vasopressin treatment for DI [28].

Cloning of the V2R and identification of patients with X-linked NDI

The 1971 Nobel Prize in Physiology or Medicine was awarded to Sutherland “for his discoveries concerning the mechanisms of the action of hormones” (Sutherland EW. Studies on the mechanism of hormone action (Nobel Lecture) [29]. Sutherland promoted the idea that hormones activate the adenylyl cyclase molecule. Subsequently Pastan et al [30] introduced the concept that most hormones act at the cell surface and by binding to a finite number of molecules (receptor) on the cell surface.

In 1992 the AVPR2 gene that encodes the V2 vasopressin recep­tor was cloned and mutations in this gene were identi­fied in patients with X linked NDI. Specifically Lolait et al from the National Institute of Mental Health, Bethesda, cloned the rat kidney V2 receptor displaying a transmembrane topography characteristic of G protein-coupled receptors. The human V2 receptor gene was localized to the long arm of the X chromosome close to the locus for nephrogenic diabetes insipidus [31].

In parallel Rosenthal and co-workers from the Baylor College of Medicine, Houston, Texas (USA) also cloned the V2 receptor and reported the case of an affected male of a family with CDI having a deletion in the open reading frame of the V2 receptor gene, causing a frame shift and premature termination of translation in the third intracellular loop of the receptor protein [32]. In parallel van den Ouweland from the University Hospital Nijmegen (The Netherlands) described 3 patients with identified point mutations AVPR2 gene affected by NDI [33]. It became subsequently clear that mutations in AVPR2 are responsible for about 90% of the inherited forms [2] [34]. Since this gene is located on the X‑chromosome, the major­ity of patients with NDI are male.

To date more than 250 different AVPR2 mutations have been described and most of these mutations are missense and based on the in vitro data those mutations apparently encode for functional but misfolded receptors, retained and degraded in the endoplasmic reticulum [35]. These observations opened a research field aimed at correcting those forms of NDI trying to rescue the otherwise functional receptors from the endoplasmic reticulum to the plasma membrane.

Cloning of the AQP2 water channels and identification of patients with autosomal NDI

In 1993 the AQP2 gene was cloned by Sasaki group from the University of Tokyo, Japan, and found to be expressed in the renal collecting duct [36] [37]. In 1994 Deen and co-workers showed that mutations in AQP2 gene were found to be responsible for autosomal recessive NDI demonstrating that AQP2 water channel is required for vasopressin-dependent concentration of urine [38].

The cloning of the vasopressin sensitive water channel AQP2 has been obtained by homology cloning after the identification of the first water channel Aquaporin 1 by Peter Agre [39] (full text). The 2003 Nobel Prize in Chemistry was awarded jointly to Peter Agre, “for the discovery of water channels”, and to Roderick MacKinnon “for structural and mechanistic studies of ion channels”.

The discovery of Aquaporins allowed understanding the etiology of congenital NDI caused by mutation of the AQP2 gene. Using immunogold electron microscopy, Nielsen and co-workers localized AQP2 in intracellular vesicles in collecting duct principal cells. AQP2 vesicles fuse to the apical membrane under vasopressin action leading to an increase in the osmotic water permeability (see schematic model in Figure 2). Removal of vasopressin caused internalization and recycling of the APQ2 protein and markedly diminished water permeability [40] (full text). This mechanism allows water reabsorption from the lumen of collecting duct and concentration of urine. In 1994 Deen and co-workers demonstrated that aquaporin‑2 water channel is required for vasopressin-dependent concentration of urine [38]. In the same year van Lieburg identified mutations of the AQP2 gene that caused the autosomal recessive form of NDI, based on the observations that patients showed increased factor VIII responses to dDAVP, an extrarenal effect suggesting normal vasopressin receptors [41]. Subsequently, in 1998, the autosomal dominant form of NDI was identified [42]. So far 40 known mutations that cause autosomal recessive NDI have been described and 8 known mutations responsible for the autosomal dominant form of NDI [2].

Current treatment of congenital NDI focuses on dietary modification, thiazides and inhibitors of prostaglandin synthesis (see [2]). Treatment with statins have also been suggested to potentially be beneficial for ameliorate NDI [43] (full text) [44].

Novel therapies, such as mutation-specific treatment using molecular chaperones [45], have been investigated in animal models, but few data from clinical studies are currently available.

In the future, gene therapy aimed at delivering kidney-specific wild-type AVPR2 or AQP2 could potentially cure congenital NDI.

Summary and Conclusions

After 3000 years of the description of Diabetes Mellitus, a subset of polyuric patients was considered to be affected by Diabetes Insipidus. About one hundred years ago in 1913, posterior pituitary extracts (containing vasopressin and oxytocin) were used to treat Central Diabetes Insipidus and this approach represented one of the first successful therapies for a peptide hormone deficiency.

About 70 years ago, physicians recognized patients with DI who failed to respond to vasopressin and 60 years ago the Nobel Prize du Vigneaud was successful in isolation, sequencing, and chemical synthesis of oxytocin and vasopressin.

Starting 50 years ago, a multitude of cellular events at the target cell have been elucidated. An error at any step can result in defective water balance. Emerging concepts of receptors and recent genetic analysis led to the recognition of patients with mutations in the genes coding for the vasopressin receptor and for the AQP2 water channel. The etiology of the autosomal form of Neprogenic Diabetes Insipidus was delineated by the discovery of the first Aquaporin by Peter Agre who was awarded with the 2003 Nobel Prize for the discovery of the water channels. Our understanding of the molecular physiology of Diabetes Insipidus has greatly advanced in the past 25 years. It is known today that acquired forms of Neprogenic Diabetes Insipidus are very common and most of those forms are associated to lithium treatment.

Today, Diabetes Insipidus includes a large series of disorders. Many of the known disorders are now susceptible to symptomatic treatments or specific interventions such as on dietary modification, thiazides and inhibitors of prostaglandin synthesis. However these treatment approaches can only ameliorate the clinical phenotype of Neprogenic Diabetes Insipidus. In the future gene therapy to correct the deficiency of AVPR2 or AQP2 genes might potentially represent a successful approach.