Noninvasive Immunohistochemical Diagnosis and Novel MUC1 Mutations Causing Autosomal Dominant Tubulointerstitial Kidney Disease

Abstract

Background: Autosomal dominant tubulointerstitial kidney disease caused by mucin-1 gene (MUC1) mutations (ADTKD-MUC1) is characterized by progressive kidney failure. Genetic evaluation for ADTKD-MUC1 specifically tests for a cytosine duplication that creates a unique frameshift protein (MUC1fs). Our goal was to develop immunohistochemical methods to detect the MUC1fs created by the cytosine duplication and, possibly, by other similar frameshift mutations and to identify novel MUC1 mutations in individuals with positive immunohistochemical staining for the MUC1fs protein.

Methods: We performed MUC1fs immunostaining on urinary cell smears and various tissues from ADTKD-MUC1-positive and -negative controls as well as in individuals from 37 ADTKD families that were negative for mutations in known ADTKD genes. We used novel analytic methods to identify MUC1 frameshift mutations.

Results: After technique refinement, the sensitivity and specificity for MUC1fs immunostaining of urinary cell smears were 94.2% and 88.6%, respectively. Further genetic testing on 17 families with positive MUC1fs immunostaining revealed six families with five novel MUC1 frameshift mutations that all predict production of the identical MUC1fs protein.

Conclusions: We developed a noninvasive immunohistochemical method to detect MUC1fs that, after further validation, may be useful in the future for diagnostic testing. Production of the MUC1fs protein may be central to the pathogenesis of ADTKD-MUC1.

Keywords: Autosomal Dominant Tubulo-Interstitial Kidney Disease; Inherited; MUC1; Mucin-1 Kidney Disease; diagnosis; immunostaining; kidney disease.

Copyright © 2018 by the American Society of Nephrology.

Figures

Graphical abstract

Figure 1.

Work-flow diagram shows the distribution of samples used for validation of the methodology and for identification of families with ADTKD-MUC1.

Figure 2.

Immunohistochemical staining of skin, sebaceous glands, and breast ducts shows positive MUC1fs staining in genetically affected individuals with ADTKD-MUC1 and negative MUC1fs staining in controls. (A) Survey of a kidney section showing positive staining for MUC1 in cortical distal tubules and collecting ducts in a patient with ADTKD-MUC1 and (B) a detailed view showing distal tubules positively stained for MUC1 with maximal staining intensity on the apical membranes of tubular cells. (C) Survey of a parallel kidney section from the patient with ADTKD-MUC1 stained with an antibody against MUC1fs with positivity in corresponding structures and (D) a detailed view showing finely granular intracellular MUC1fs staining pattern in distal tubules. (E) Survey of a control kidney section stained with an antibody against MUC1 and (F) a detailed view showing intracellular positivity in the distal tubule with an accent on the apical pole of the plasma membrane of tubular cells. (G) Survey of a parallel control kidney section stained with an antibody against MUC1fs and (H) a detailed view, both demonstrating negative staining in corresponding tubules. (I) Strong positivity for MUC1 in sebaceous glands in a skin biopsy specimen from a patient with ADTKD-MUC1 and (J) a detailed view. (K) Less intensive but distinct positivity of MUC1fs in sebaceous glands in a patient with ADTKD-MUC1 and (L) a detailed view of the patient’s MUC1fs-positive sebaceous glands. (M) Strong positivity of MUC1 in sebaceous glands in a control skin biopsy sample and (N) a detailed view. (O and P) Negative MUC1fs staining in sebaceous glands in a control skin biopsy sample demonstrated in a low power view (O) and (P) in detail. Epithelial cells in sweat glands (marked by arrows) displaying MUC1 positivity at the apical poles [demonstrated in a control skin biopsy sample in (M)] were MUC1fs-negative in a patient with ADTKD-MUC1 (K). (Q) Epithelial cells in breast ducts in a male patient with ADTKD-MUC1 display strong positivity of MUC1 at the apical poles and (R) distinct granular cytoplasmic positivity of MUC1fs. (S) Breast ducts in a male control display a similar staining pattern of MUC1 but (T) no immunostaining with the antibody detecting MUC1fs.

Figure 3.

Immunohistochemical staining of urinary smears from individuals with ADTKD-MUC1 reveals positive MUC1fs staining in all stages of kidney disease. Urinary pellets were smeared on glass slides, then fixed and stained with anti-MUC1fs and anti-MUC1 antibodies. Confocal images show DAPI (blue) and either MUC1fs or MUC1 (green). (A and B) Strong diffuse to granular intracellular staining of MUC1fs in superficial and intermediate urothelial cells in affected individuals. (C and D) Negative MUC1fs staining in urothelial cells from a healthy control. (E) Finely granular cytoplasmic staining in superficial urothelial cells and (F) distinct plasma membrane staining of MUC1 in intermediate urothelial cells in affected individuals that is similar to that in controls (G and H). Positive MUC1fs staining in urothelial cells was detected in (A and B) a genetically affected individual with normal kidney function, (I and J) in two individuals with advanced CKD on dialysis, (K) in a nondialyzed individual 1 month before transplantation, and (L) in another patient with advanced CKD on dialysis. Note that in individuals with advanced CKD, cells that stain strongly positive for both MUC1fs and MUC1 are much smaller in size (diameter ±15 µm) than controls (diameter 30–40 µm).

Figure 4.

Immunostaining for MUC1fs is positive in patients with different MUC1 mutations and is negative in a patient with ADTKD-UMOD. Urinary cell pellets were smeared on glass slides, then fixed and stained with anti-MUC1fs and anti-MUC1 antibodies. Merged confocal images show DAPI (blue), MUC1 (red), and MUC1fs (green). Panels illustrate the presence of MUC1fs in the cells of genotyped individuals with (A) the MUC1 27dupC mutation, (B) the MUC1 28dupA mutation, and (C) an unknown mutation. U/M/R negative denotes that the tested participant was negative for mutations in UMOD, REN, and the MUC1 mutations described here. Absence of MUC1fs is shown in an individual with (D) a UMOD mutation (ADTKD-UMOD) and (E) control. MUC1-positive cells outlined with a white rectangle are shown in detail in the corresponding image on the right.

Figure 5.

MUC1 VNTR sequencing identifies novel mutations causing ADTKD-MUC1. (A) Sequence logo showing the most conserved regions of the VNTR repeats. Corresponding amino acid sequences of wild-type MUC1 (wt_AA) and MUC1fs (mut_AA) are shown below. To find novel frameshift mutations that change the open reading frame, different conserved 10-mers of the wild-type repeat were used as sequence anchors (underlined DNA sequence as an example). For each anchor pair, all sequences delimited by these two anchors that are changing an open reading frame (i.e., adding or deleting nucleotides) were selected from the FASTQ file. (B) Sequences of the canonical 60 nucleotide long wild-type VNTR repeat (wt) and candidate frameshift mutations identified in this study. (C) Random mutations are generated in DNA molecules during PCR amplification step. To find true germline mutations, the percentage of reads with a given sequence (putative frameshift mutation) from all reads was calculated for each of the analyzed samples (y-axis), and this needed to be higher than the average+2 SD of the nine wild-type control samples. Indicated are numbers of controls (wt), patients with individual MUC1 mutations (27dupC, 28dupA, 26_27insG, 1–16dup, 23delinsAT, 51dupC), and individuals with still unknown MUC1 mutation(s) who have urinary cell smears positive for MUC1fs and who tested negative for 27dupC by conventional genotyping assay (unknown). (D) 27dupC, confirmed by a mass spectrometry-based primer extension assay. The 27dupC extension product is observed at 5904 D (red asterisk). (E) 28dupA, confirmed by a mass spectrometry-based assay. The 28dupA extension product is observed at 6571 D (red asterisk). (F) 26_27insG, confirmed by a mass spectrometry-based assay. The 26_27insG extension product is observed at 5944.85 D (red arrow). (G) 1–16dup confirmed by restriction analysis. The mutation creates new restriction site for EciI enzyme. The electrophoretogram shows amplified VNTR regions of the affected patient (P1), two healthy relatives (H1, H2), and one unrelated control (NC) after (EciI) and before restriction by EciI (PCR). The patient’s (P1) mutated allele (5000 bp) was cut into two fragments of 3000 and 2000 bp. (H) 23delinsAT, confirmed by restriction analysis. The mutation creates new restriction site for FokI enzyme. The electrophoretogram is showing amplified VNTR regions of two affected patients (P1, P2) and one unrelated control (NC) after (FokI) and before restriction by FokI (PCR). The patients’ (P1, P2) mutated alleles (3000 bp) were cut into two fragments of 2000 and 1000 bp.

Figure 6.

Pedigrees of families with novel MUC1 mutations show characteristic autosomal dominant transmission. Clinically affected family members (defined as stage III-V CKD or requiring dialysis/kidney transplant) are shown with a black symbol, clinically unaffected family members are shown with a white symbol. Gray symbols indicate that clinical status is unknown. The plus sign (+) indicates that genotyping was performed and a mutation resulting in MUC1fs was identified in the individual. A dash (−) means the patient was genotyped and found not to have an MUC1 mutation.