Reticular Dysgenesis and Mitochondriopathy Induced by Adenylate Kinase 2 Deficiency with Atypical Presentation

Abstract

Reticular dysgenesis is an autosomal recessive form of severe combined immunodeficiency (SCID) that usually manifests in newborns. It is a unique example of an immune deficiency that is linked to dysfunctional mitochondrial energy metabolism and caused by adenylate kinase 2 (AK2) deficiency. It is characterized by an early differentiation arrest in the myeloid lineage, impaired lymphoid maturation, and sensorineural hearing loss. In this study, a novel AK2 homozygous mutation, c.622 T > C [p.Ser208Pro], was identified in an Old Order Amish patient through whole exome sequencing. Functional studies showed that the patient's cells have no detectable AK2 protein, as well as low oxygen consumption rate (OCR), extracellular acidification rate (ECAR) and proton production rate (PPR). An increased production of reactive oxygen species, mitochondrial membrane permeability, and mitochondrial mass, and decreased ATP production, were also observed. The results confirm the pathogenicity of the AK2 mutation and demonstrate that reticular dysgenesis should be considered in Amish individuals presenting with immune deficiency. We also describe other pathophysiological aspects of AK2 deficiency not previously reported.

Figure 1

H&E staining of the bone marrow, Pedigree, AK2 gene and protein structure. (A) H&E staining of the bone marrow. Control; bone marrow from an age-matched individual showing adequate cellularity with all normal hemopoietic cell lines represented and without predominance of any particular lineage. Pre-Tx; pre-transplant bone marrow biopsy from the patient at 13 months of age before bone marrow transplant showing myeloid maturation only through the promyelocyte/myelocyte stage. Only occasional neutrophils were seen. Post-Tx; post-transplant bone marrow biopsy from affected patient after bone marrow transplant showing normocellular bone marrow with trilineage hematopoiesis (all images at 100X). (B) Pedigree of the family identified with mutations in the AK2 gene. Both parents and siblings are unaffected and heterozygous for the mutation while the patient is homozygous. Ages are representative of the individual ages at the time of manuscript submission. Asterisk denotes the age of the patient when he died following bone marrow transplant complication. (C) Structure of the AK2 gene (GenBank: NM_001625), location of the Ser208Pro mutation relative to the polypeptide stretch, and homology alignment of the AK2 β-strands IV (L125-I129) and VII (G206-A212) regions. The secondary structure assignments are according to the human AK2 crystal structure atomic coordinates, PDB 2C9Y, highlighted in pink boxes. Residues in bold capital letters are invariants in all species examined, only 13 are shown. Residues in capital letters (not bold) are highly conserved, and residues in small letters seem dispensable. Highlighted in green is the Ser208 position. (D) Ribbon representation of the AK2 protein 3D structure, PDB 2C9Y, with the position of Ser208 depicted (replaced with a Pro in the patient of this study), plus the position of other previously reported mutations in the AK2 protein. BATP (letters in yellow) is the ATP binding domain.

Figure 2

Protein and mRNA expression analysis of AK2 in cell lines derived control (CT) and patient (PT). (A) Fibroblast cells from CT and PT were analyzed for the presence of AK2 antigen by SDS-PAGE followed by western blotting with anti-AK2 and anti-β-actin antibodies. Twenty-five µg protein from fibroblasts’ lysates were loaded for both CT and PT (see Supplementary Figs S1 and S2 for full blot). (B,C) Immunofluorescence staining of fibroblasts and bone marrow from CT and PT. AK2 antigen was visualized with green fluorescently tagged antibodies and mitochondrial cytochrome c oxidase subunit 1 (MTCO1) was visualized with red fluorescently tagged antibodies. Nuclei were visualized with DAPI staining. The merged image shows co-localization of AK2 and MTCO1 in mitochondria as yellow/orange in the CT fibroblasts while there is essentially no green staining that would show co-localization with MTCO1 in the patient fibroblasts. Scale bar, 50 μm. (D) RT-PCR for AK2 and GAPDH on CT, PT and negative control (H₂O) samples. GAPDH was used as a standard reference gene. Both CT and PT PCR reactions resulted in similar cDNA band instensity (see Supplementary Fig. S3 for full image of agarose gel electrophoresis).

Figure 3

Mitochondrial studies in control (CT) and patient (PT). (A) Oxygen consumption rate (OCR) measured using both parameters with Seahorse XF^e96 Extracellular Flux Analyzer: Basal respiration and reserve capacity. Both were decreased in the PT compared with CT. Data reported in pmol/min/cells. (B) Extracellular acidification (ECAR) and the proton production (PPR) rate using Seahorse XF^e96 Extracellular Flux Analyzer. Both rates can be used as surrogates for lactate production attributed to anaerobic glycolysis. Both are decreased in PT compared to CT. (C) Reactive oxygen species (ROS) production by measuring superoxide production using MitoSOX Red with and without using glucose. ROS is higher in PT’s fibroblasts comparing with the CT, and is more pronounced when the media used is without glucose. (D) Measurement of the mitochondrial mass and membrane potential with MitoTracker Green and Red, respectively, revealed a significant increase in mitochondrial mass concurrent with an increase in mitochondrial membrane potential in patient fibroblasts. (E) ATP production assay showing significantly lower ATP in the patient cells (PT) compared to the CT.