Mitchell-Riley Syndrome: Improving Clinical Outcomes and Searching for Functional Impact of RFX-6 Mutations

Abstract

Aims/hypothesis: Caused by biallelic mutations of the gene encoding the transcription factor RFX6, the rare Mitchell-Riley syndrome (MRS) comprises neonatal diabetes, pancreatic hypoplasia, gallbladder agenesis or hypoplasia, duodenal atresia, and severe chronic diarrhea. So far, sixteen cases have been reported, all with a poor prognosis. This study discusses the multidisciplinary intensive clinical management of 4 new cases of MRS that survived over the first 2 years of life. Moreover, it demonstrates how the mutations impair the RFX6 function.

Methods: Clinical records were analyzed and described in detail. The functional impact of two RFX6^R181W and RFX6^V506G variants was assessed by measuring their ability to transactivate insulin transcription and genes that encode the L-type calcium channels required for normal pancreatic beta-cell function.

Results: All four patients were small for gestational age (SGA) and prenatally diagnosed with duodenal atresia. They presented with neonatal diabetes early in life and were treated with intravenous insulin therapy before switching to subcutaneous insulin pump therapy. All patients faced recurrent hypoglycemic episodes, exacerbated when parenteral nutrition (PN) was disconnected. A sensor-augmented insulin pump therapy with a predictive low-glucose suspension system was installed with good results. One patient had a homozygous c.1517T>G (p.Val506Gly) mutation, two patients had a homozygous p.Arg181Trp mutation, and one patient presented with new compound heterozygosity. The RFX6^V506G and RFX6^R181W mutations failed to transactivate the expression of insulin and genes that encode L-type calcium channel subunits required for normal pancreatic beta-cell function.

Conclusions/interpretation: Multidisciplinary and intensive disease management improved the clinical outcomes in four patients with MRS, including adjustment of parenteral/oral nutrition progression and advanced diabetes technologies. A better understanding of RFX6 function, in both intestine and pancreas cells, may break ground in new therapies, particularly regarding the use of drugs that modulate the enteroendocrine system.

Keywords: Mitchell–Riley syndrome; RFX6; beta-cell function; diabetes technology; neonatal diabetes mellitus; parenteral nutrition.

Figure 1

Comparison of stature (A) and (B) Body Mass Index evolution of four patient using standard deviation scores according to WHO growth charts. BMI, Body M ass Index; P, patient, SD, Score Deviation.

Figure 2

Diabetes Treatment and Nutrition follow-up of four patients with RFX-6 mutation. PN, parenteral nutrition; CSII, Continuous Subcutaneous Infusion System; Hb1Ac, glycosylated hemoglobin; CGM, continuous monitoring system; GTT, gastrostomy; HypoMinimyzer, The Predictive Hypoglycemia Minimizer System.

Figure 3

Conti nous Glucose Monitoring da ta from Mini Med 640G System from patients 3- M utR181W (A) and 4- (c.505-2A>G/ c.2782A>G ) (B). The traces represents glucose variability during the day. Red colour cshowed hypoglycemic periods (less than 70mg/dl) and yellow periods represents hyperglycemias ( more than 140-180 mg/dl) In figure (A), patient 3 presents recurrent hypoglycemic episodes during daily period when she was disconnect from parenteral nutrition (4 a.m.) and frequent hyperglycemia during night period. Basal rate was 0.2UI of insulin during parenteral nutrition (around 10 times more than during the day) Patient 4 had lower glycemic variability, but some hypoglycemia du ring the day. He received parenteral nutrition and few oral nutrition.

Figure 4

Patient 2 (R181W) 3D-MRJ T2 (A) and Tl-weighted (B) with volume rendering reconstructions (C, D). The MRI shows a small pancreas (white arrowheads), especially in the bod y and tail, with a volume estimated between 13 and 15.5 cc. For comparison, an aged-matched control patient has a pancreatic volume of 45 cc.

Figure 5

Transactivation of Insulin promoter by RFX6^WT, RFX6^R181W and RFX6^V506G in EndoC-βH2 cells. (A) Schematic presentation of the functional domains in RFX6 protein and the position of mutations. (B) Insulin promoter activity in human beta cells model EndoC-βH2 cells determined by firefly luciferase (pGL4.12huINS-378to+42), which was cotransfected with either RFX6^WT or RFX6^R181W or RFX6^V506G and with Renilla luciferase (pGL4.72-TK[hRlucCP]) to correct for variation in transfection efficiency. Results are presented as fold increase over empty control vector. Da ta are mean of ± SEM of three to four experiments. ***p < 0.001; ns, non significant.

Figure 6

Differential activation of R FX6 targets gene expression by wtRFX6 or Mu tR181W-R FX 6 or MutV506G-RFX6 in EndoC-βH2 cells. (A) Schematic presentation of the constructs with trans-activation domain VP16 and I ERS-EGFP used to overexpress wt or Mut R FX6 proteins (B) Fort y-eight hours posttransfection, GFP+ cells were FACS sorted and analyzed for the expression of R FX6 by RTq PCR in EndoC-βH2. (C–F) RT qPCR analysis of CACNAlA, CACNA1C, CACNA1D and G PR 68 in EndoC-βH2 post 48h transfected with wt or Mut RFX6. Da ta are mean of ± SEM of three to four experiments. *p < 0.05; **p <0.01; ns, non significant.

Figure 7

Expression and localization of of wtRFX6 or MutR181W-RFX6 or MutV506G-RFX6 in HEK293. HEK293 cells were transiently transfected with pRIG-empty or pRIG-wtRFX6 or pRIG-MutR181W or pRIG­MutV506G -R FX6 constructs and immunostained for RFX6 post 48 h of transfection. Scale bar 25 μm. Legend: (A) Vector control, (B) wtRFX6, (C) Mut R181W-RFX6, (D) MutV506G-RFX6.