Functional characterization of a loss-of-function mutant I324M of arginine vasopressin receptor 2 in X-linked nephrogenic diabetes insipidus

X-linked nephrogenic diabetes insipidus (X-linked NDI) is a rare inherited disease mainly caused by lost-of-function mutations in human AVPR2 gene encoding arginine vasopressin receptor 2 (V2R). Our focus of the current study is on exploration of the functional and biochemical properties of Ile324Met (I324M) mutation identified in a pedigree showing as typical recessive X-linked NDI. We demonstrated that I324M mutation interfered with the conformation of complex glycosylation of V2R. Moreover, almost all of the I324M-V2R failed to express on the cell surface due to being captured by the endoplasmic reticulum control system. We further examined the signaling activity of DDAVP-medicated cAMP and ERK1/2 pathways and the results revealed that the mutant receptor lost the ability in response to DDAVP stimulation contributed to the failure of accumulation of cAMP and phosphorylated ERK1/2. Based on the characteristics of molecular defects of I324M mutant, we selected two reagents (SR49059 and alvespimycin) to determine whether the functions of I324M-V2R can be restored and we found that both compounds can significantly “rescue” I324M mutation. Our findings may provide further insights for understanding the pathogenic mechanism of AVPR2 gene mutations and may offer some implications on development of promising treatments for patients with X-linked NDI.


Results
I324M mutant impaired the V2R maturation and expression on the plasma surface. V2R is a glycoprotein receptor with a glycosylation site at position 22. Glycosylation is one of the most important posttranslational modification, which plays an essential role in maintaining the structure and function of proteins. Therefore, identifying the degree of glycosylation of glycoproteins is important for understanding the pathogenic basis of V2R mutants. Endoglycosidase H (EndoH), peptide-N-glycosidase F (PNGase.F) and O-glycosidase & Sialidase can cleave N-linked high mannose groups, all N-linked sugar moieties and O-linked disaccharides respectively 23 . The maturity of this receptor can be assessed according to the different sensitivity of V2R to the deglycosylases as mentioned above 24 .
In order to detect the receptor maturation state of WT-V2R and I324M-V2R, HEK293T cells expressing WT-V2R and I324M-V2R were treated with or without EndoH, PNGase.F or O-glycosidase & Sialidase before analysis with western blotting. As shown in Fig. 1A, the group of WT-V2R was detected two main bands compared with control (Ctr) (Fig. 1A, compare lanes 1 and 2), one at ~ 48 kDa that represented the core glycosylated receptors (Core-Gly-V2R) and the other at ~ 63 kDa, which was thought to be the mature form of V2R with fully glycosylated modification. When the WT-V2R group was pretreated with Endo H, the position of the ~ 63 kDa band was not altered whereas ~ 48 kDa band was removed to ~ 42 kDa, which was described as de-glycosylated V2R (De-Gly-V2R) (Fig. 1A, compare lanes 2 and 3). EndoH resistance of the ~ 63 kDa form suggested that a considerable amount of V2R had been transported from the ER to Golgi apparatus or cell surface, so after processing with PNGase, F (Fig. 1B, compare lanes 1 and 2), we found that the band was removed to ~ 52 kDa and could be further dropped to ~ 42 kDa using O-glycosidase & Sialidase in the WT-V2R group (Fig. 1C, compare lanes 1, 2 and 3). In contrast to WT-V2R, I324M-V2R only formed a band at 48 kDa without the upper mature glycosylated band (Fig. 1A, compare lanes 2 and 4), but the sensitivity to EndoH was as same as in the WT-V2R group, revealing the process of immature glycosylation in ER (Fig. 1A, compare lanes 3 and 5). Furthermore, I324M-V2R also failed to form PNGase.F-sensitive band (Fig. 1B, compare lanes 3 and 4), confirming that the mutation did substantially impair receptor maturation, which was further illustrated by the disappearance of O-glycosylation moieties (Fig. 1C, compare lanes 4, 5 and 6). These findings confirmed that most, if not all, I324M-V2R were exist in an immature form.
I324M mutant hindered the transport of V2R, making almost all receptors trapped in the intracellular compartments. As mentioned before, I324M-V2R mutation blocked the protein proper glycosylation resulting in the dysmaturity of V2R. Mature glycosylated modification is essential for the transposition of glycoproteins from ER to Golgi apparatus and targeting to the cell surface. Therefore, whether the immaturity of V2R-I324M has a potential influence on its intracellular distribution was the focus of our further research. Immunofluorescence technology was applied to compare the trafficking difference of the WT-V2R  www.nature.com/scientificreports/ and I324M-V2R. COS7 cells were used to present the receptor's subcellular localization because of its prominent membrane structure that makes it easy to distinguish the specific location of V2R proteins. WT-V2R and I324M-V2R were co-stained with protein disulfide isomerase (PDI) and Golgi marker (GM130), which are used as the markers of endoplasmic reticulum (ER) and Golgi apparatus, respectively 25 . As shown in Fig. 2A, WT-V2R was predominantly localized on the plasma membrane as pointed by the white arrows ( Fig. 2A, the upper row). Furthermore, we can also note that a small part of receptors was co-located with PDI may be the result of the characteristics of COS7 cell line and the application of transient transfection technology. As expected, COS7 cells expressing I324M-V2R were barely detected fluorescence in cell surface, but almost completely co-stained with PDI ( Fig. 2A, the lower row). The performance of co-staining with GM130 is roughly similar to that of Fig. 2A, the difference in the fluorescence density of the receptor on plasma membrane between WT-V2R and I324M-V2R was still obvious (Fig. 2B, the upper row and lower row). That is to say, I324M-V2R mostly retained in intracellular regions compared with V2R-WT. These results indicated that I324M mutant was unable to target on the cell surface to perform a function, because of the receptor transport pathway disordered.
I324M mutant did not affect the production and stability of V2R proteins. Most V2R mutations result in protein retention in the endoplasmic reticulum (ER) and may trigger the endoplasmic reticulum-associated degradation (ERAD) pathway, thereby affecting the stability of the receptor. Therefore, the cycloheximide (CHX) chase assay was performed to test the protein stability in the groups of WT-V2R and I324M-V2R. As shown in Fig. 3A and B, we could see the degradation rate of I324M-V2R group was almost identical with WT-V2R group after treated with CHX for 4, 8 h. Although it looks like there was a difference in treatment with CHX for 12 h, unfortunately, the results had no statistical difference. Moreover, we can also observe that the production of V2R proteins in I324M-V2R was similar to WT-V2R (Fig. 3A). Consequently, these results illuminated that I324M mutation had no influence on the stability and total protein amount of V2R.
I324M mutant inactivated G protein-dependent cAMP signaling activity of V2R under the stimulation of DDAVP. The normality of cAMP production in response to AVP stimulation is the key for kidney to maintain body's water balance. Therefore, we monitor basal or DDAVP-promoted cAMP accumulation in transiently transfected HEK293T cells using cAMP kits to assess the function properties of V2R I324M www.nature.com/scientificreports/ mutant. As shown in Fig. 4A, the cAMP level of WT-V2R and I324M-V2R had no significant difference in absence of DDAVP. While in the presence of DDAVP, the concentration of cAMP in WT-V2R increased nearly 15-fold than basal state in all set-up concentration groups, and 100 nM DDAVP was chose as the optimum concentration for the cAMP generation. Completely opposite to WT-V2R, there was no obviously response to DDAVP-medicated cAMP generation in the group of I324M-V2R. Even in the case of high dose of DDAVP, the cAMP accumulation was still same as the control (Ctr). The activity of cAMP pathway was not associated with AVP concentration in I324M-V2R group indicate that the mutant perturbed the G protein-dependent cAMP signaling pathway and there was no production of functional receptors. At the same time, we examined the receptor expression from the same samples. Unexpectedly, most mature proteins were synthesized in WT-V2R transfected cells other than in V2R-I324M cells (Fig. 4B). The mature form of V2R can respond to treatment of DDAVP, then activated the cAMP increasement. In short, the loss of mature form in V2R-I324M was an important reason for the inactivation of V2R in cAMP signaling.   5A and B, compare lanes 1, 3 and 5). In the stimulation of V2R agonist DDAVP, we found that WT receptor was resulted in a marked ERK1/2 phosphorylation response whereas V2R-I324M was failed to response to DDAVP stimulation ( Fig. 5A and B, compare lanes 2, 4 and 6). These results indicated that I324M mutant also terminated the capacity of V2R to elicit the activity of ERK1/2 signaling pathway.   19 . Alvespimycin is an inhibitor of HSP90 that can increase the synthesis of mature receptors and thereby increasing the ability of V2R signal transduction 21 . Therefore, we selected these two compounds to determine whether they have the similar ability on rescuing the functions of I324M mutant. As shown in Values were mean ± SEM using One-way ANOVA. ***P < 0.001 (n = 3).  6A, there was almost the same amount of cAMP in the groups of I324M-V2R and WT-V2R in the absence of DDAVP (basal) or in the presence of SR49059 and alvespimycin. In the case of DDAVP (100 nM) pretreatment, the production of cAMP in WT-V2R group was increased by about 12-fold compare to basal level, while the cAMP level in I324M-V2R group has no any increasement, which was similar to the results as described above (Fig. 4A). Meanwhile cells were treated with DDAVP (100 nM) following incubation with SR49059 or alvespimycin, the cAMP accumulation in I324M-V2R group was increased approximately sevenfold and twofold respectively above non-SR49059 or non-alvespimycin treatment. Furthermore, SR49059 treatment has no significant effect on the group of WT-V2R, while alvespimycin treatment increased cAMP accumulation nearly 1.6-fold compare to the group of AVP (100 nM), which was due to the fact that alvespimycin can also increase the expression of V2R under normal conditions 21 . At the same time, we detected the mature expression of V2R in HEK293T cells upon incubation with SR49059 or alvespimycin. As presented in Fig. 6B, the expression of fully glycosylated receptor (mature form) was obviously increased after treatment with SR49059 or alvespimycin in the group of I324M-V2R compared to basal level (Fig. 6B, compare lanes 5, 6, 7), which may explain the www.nature.com/scientificreports/ recovery of DDAVP-medicated cAMP response. These results indicated that SR49059 and alvespimycin can play a role in rescuing the receptor maturation and cAMP signaling activity of I324M mutant.

Discussion
Over recent years, more and more mutations in AVPR2 gene have been identified in patients with X-linked NDI. Clarifying the molecular defects of these mutations is critical to find potential therapeutic targets for X-linked NDI. In this research, we have functional characterized a clinically relevant AVPR2 gene mutation (c.972C > G) previously reported in 2016 by our own team, a substitution of isoleucine at position 324 with methionine 22 . Through our studies, we firstly demonstrated the existence of abnormal glycosylation modification and intracellular transport disorder of the I324M mutant receptor. Endoplasmic reticulum (ER) is a strict quality control system, which allowed to be transported out are only properly folded and modified proteins, while misfolded proteins are naturally restricted in it and may target to ER-associated degradation pathway (ERAD) 2, 26, 27 . Subsequently, we performed CHX chase assay to compare protein stability between WT-V2R and I324M-V2R. We showed that the protein degradation rate of I324M-V2R was not affected. The key to determining the transduction of GPCR signals depends on the maturity and appropriate cell localization surface of the membrane receptors 28 . Therefore, we detected the V2R-dependent cAMP and p-ERK1/2 accumulation upon DDAVP stimulation. We found that the functional signaling pathways were inactivated due to maturation and localization errors of I324M-V2R, thus ultimately leading to renal water imbalance. V2R is an alpha helix receptor, which contains seven transmembrane domains (TMD), four extracellular regions and four intracellular regions 29 . The first and second extracellular regions are often considered to be the ligand binding sites. So, mutations reported in these regions, such as R104C, F105V, R181C, are sure to impair the ligand-receptor binding efficiency. Recently, M311V, M317S and M322S mutants present a reduction of AVPbinding efficiency, indicating that TMD VII is also a potential effect target of ligand 1 . While most of discovered V2R mutations were occurred in membrane spanning and usually exhibited strong retention in the ER 29 . I324M locates on the edge of the TMD VII of V2R, where it forms the special NPXXY motif with N321, P322, W323 and Y325. NPXXY motif is a highly conserved region, where the changes will not only interfere with receptor endocytosis and folding, but also undermine G protein signaling 30,31 . From previous studies, we know that the binding efficiency of N321Y-V2R, P322S-V2R and Y325F-V2R mutations to ligands was reduced in the case of a similar quantitative cell surface expression compare to WT -V2R, while W323-I324insR-V2R mutation was similar to I324M mutation researched in this paper, both of which lead to receptor misfolding and trapped in ER 31 . For one thing, the conversion of isoleucine (I) to methionine (M) has changes the NPXXY sequence and the additional S-CH3 structure may affect the proper formation of the receptor's spatial structure, which may be the interpretation for the misfolding and fault localization of I324M-V2R. For another, both of isoleucine and methionine are aliphatic hydrophobic amino acids, so that the replacement between amino acids with identical natures probably has no influence on protein stability and the total protein synthesis. V2R-promoted phosphorylation of ERK1/2 requires the support of β-arrestin. β-arrestin terminates the G protein coupling to participate in receptor endocytosis by competitively binding to V2R with AVP, and at the same time as an adaptor protein to activate ERK1/2 32,33 . Therefore, in addition to the importance of the NPXXY structure mentioned above, I324M-V2R-mediated silencing of the ERK1/2 pathway may imply that the stability of NPXXY sequence is also essential for V2R-mediate MAPK signaling.
At present, the clinical therapies of X-linked NDI are mainly symptomatic treatment for the clinical manifestation of urinary concentration dysfunction, including fluid administration, low-salt diet, and the application of drugs such as thiazide diuretics and potassium sparing diuretics 19,34 , the treatment effect however is obviously different among individuals that only alleviates the clinical symptoms in a certain extent. But a complete cure still cannot be achieved which resulted in the life quality of the vast majority of patients is still greatly affected. X-linked NDI is a rare inherited disorder, mostly with V2R single-amino acid abnormalities 35 , which is very difficult to treat due to its resistance to AVP. In recent years, V1a receptor antagonist (SR49059) has been reported as a promising treatment for X-linked NDI. On one hand, SR49059 can decrease 24-h urine volume and water intake and increase urine osmolality in vivo studies. On the other hand, it can increase the cell surface level and the cAMP signaling activity of mutant V2R (R137H, W164S, 185_193del) in vitro studies 20,36 . In our present studies, we have also shown that the functions of I324M mutant could be rescued to some extent by SR49059, showing as the increasement of mature V2R receptor and cAMP accumulation. However, this compound has no obvious effect on mutants with reduced AVP binding efficiency 20 , which may indirectly indicate that the I324M mutation does not affect the V2R-AVP binding which seems unlike to other mutations in NPXXY region at TMD VII that showed reduced binding affinity. SR49059 has been proved to have a side effect of hepatic toxicity during the second phase of clinical research 20 , so more promising treatments for X-linked NDI need to be explored. At present, alvespimycin is mainly used for the treatment of patients with various kinds of tumors with cells relying on HSP90 activity. Previous studies have demonstrated that alvespimycin has an effect of hyponatremia on patients with advanced solid tumors during the phase I/II study 37 . In response to the results of this literature, Qiong Y et al. believed that the main mechanism explanation of alvespimycin leading to hyponatremia was vasopressin or vasopressin-dependent signaling and shown that alvespimycin can enhance the cell surface level and receptor signaling of WT-V2R in HEK293T cells 21 . In this study, we found that it could also partially restore the maturation and cAMP signaling activity of I324M mutant. The effect of alvespimycin however was found not specific to V2R but common among many GPCRs in recent studies. It was found that alvespimycin can also increase the cell surface expression and fully glycosylated form of A2A adenosine receptor and α2C adrenergic receptor 38,39 , which may be the main limitation of clinical application of alvespimycin in the future. Therefore, in addition to further exploring more effective treatments for patients carrying I324M mutation, finding novel treatments for various types of V2R mutations is also the top priority of our next research. www.nature.com/scientificreports/ To sum up, we have elucidated the molecular basis for the loss of V2R function caused by the I324M mutation and the potential treatments for the functional recovery of I324M-V2R. These findings may provide new ideas for others to understand the molecular mechanism of V2R mutations and develop more promising treatment strategies for patients with X-linked NDI.
Penicillin-streptomycin mixture (100X) was from Solarbio Life Sciences (Beijing, China). Fetal bovine serum (FBS) was from Vistech (Austrilian). Phosphate buffer saline (PBS) was from Servicebio (Wuhan, China). Lipofectamine 2000 and 4-12% Bis-Tris Gel were from Thermo Fisher Scientific (Waltham, MA, US). Cycloheximide was from MilliporeSigma (Burlington, MA, USA). Vasopressin analogue desmopressin (DDAVP) was from Absin (Shanghai, China). SR49059, phosphatase inhibitor and proteinase inhibitor were from APExBIO Technology (Houston, TX, USA). Anti-C-Myc rabbit polyclonal antibody was from Immunology Consultants Laboratory (Portland, OR, USA). Anti-protein disulfide isomerase (PDI) mouse monoclonal antibody and cAMP Direct Immunoassay Kits were from Abcam (Cambridge, MA, USA). Anti-protein GM130 mouse monoclonal antibody was from BD Biosciences (Franklin Lakes, NJ, USA). Anti-protein HSP90 mouse monoclonal antibody and anti-protein β-tubulin mouse monoclonal antibody were from Santa Cruz Biotechnology (Dallas, TX, USA). Anti-ERK1/2 rabbit polyclonal antibody and antiphospho-ERK1/2 rabbit polyclonal antibody were from Cell Signaling Technology ( Western blotting. Prepared protein samples were separated using 4-12% Bis-Tris gradient gels and transferred on the nitrocellulose membranes. Before incubating with primary antibody overnight at 4 °C, the nitrocellulose membranes were blocked in 5% skimmed milk at room temperature for 1 h and then appropriate secondary antibodies conjugated with HRP were used at room temperature for 1 h. Imaging was captured after incubation with the Western ECL Substrate according to the manufacturer's instructions, using ChemiScope (CliNX science Instruments).
Immunofluorescence. Transiently transfected COS7 cells were seeded onto glass coverslips and cultured in incubator overnight. Cells were washed 5 min for three times with PBS and fixed with 4% paraformaldehyde for 20 min. Then fixed cells were permeabilized with 0.1% Triton X-100 at room temperature for 10 min and blocked with 5% BSA at room temperature for 1 h. After blocking, the samples were incubating with anti-Myc rabbit polyclonal antibody (1:2000), anti-protein disulfide isomerase (PDI) mouse monoclonal antibody (1:200) or anti-GM130 mouse monoclonal antibody (1:200) respectively at 4℃ overnight. Next day, cells were washed with PBS three times and then incubated with secondary antibody conjugated with different Alexafluor dyes at room temperature for 1 h (away from light). Finally, fluorescent images with a × 40 objective were collected under fluorescent microscope using Axio Imager M2 (Carl Zeiss, Oberkochen, Germany).
Cycloheximide chase assay. HEK293T cells expressing either WT-V2R or I324M-V2R were cultivated in 12-wells plates. After 24 h of transfection, the cells in each 12-well plate were digested into 4 parts and transferred to the 24-well plate, cultured overnight. Then cells were preincubated with 20 μg/ml cycloheximide (CHX) for 0, 4, 8 or 12 h before protein samples were collected.
Measurement of cAMP accumulation. For Basal [DDAVP (-)] or DDAVP-promoted cAMP production by using cAMP kits was carried out according to the manufacturer's introduction of cAMP BIOTRAK EIA W/O ACET REG ELISA Kits. Briefly, HEK293T cells were transiently transfected with Myc-tagged-WT-V2R or Myc-tagged-I324M-V2R. 24 h after transfection, cells were transferred to 96-well plates (160 µl/well), with cell concentrations between 10 4 -10 6 cells/well and incubated overnight at 37 °C. After additional 24 h, the cells were pretreated with or without 20 μl DDAVP solution (1 nM, 100 nM, 1 μM or 10 μM, respectively) for 15 min. The cells were lysed with 20ul lysis reagent and subjected to cAMP ELISA assay following the manufacturer's introduction. For functional rescue studies of I324M-V2R, the accumulation of DDAVP-promoted cAMP production in HEK293T cells after incubated with or without SR49059 and alvespimycin was determined by cAMP Statistical analysis. Data were analyzed using GraphPad Prism 7.0 (https:// www. graph pad-prism. cn/) software and values were presented as mean ± SEM. The significance among groups was evaluated using Oneway ANOVA or Two-way ANOVA. P < 0.01 was considered statistically significant.

Data availability
The data generated during and analysed during the current study are available from the corresponding author on reasonable request.