Absence of an osteopetrosis phenotype in IKBKG (NEMO) mutation-positive women: A case-control study

Abstract

Background: NF-κB essential modulator (NEMO), encoded by IKBKG, is necessary for activation of the ubiquitous transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB). Animal studies suggest NEMO is required for NF-κB mediated bone homeostasis, but this has not been thoroughly studied in humans. IKBKG loss-of-function mutation causes incontinentia pigmenti (IP), a rare X-linked disease featuring linear hypopigmentation, alopecia, hypodontia, and immunodeficiency. Single case reports describe osteopetrosis (OPT) in boys carrying hypomorphic IKBKG mutations.

Method: We studied the bone phenotype in women with IP with evaluation of radiographs of the spine and non-dominant arm and leg; lumbar spine and femoral neck aBMD using DXA; μ-CT and histomorphometry of trans-iliac crest biopsy specimens; bone turnover markers; and cellular phenotype in bone marrow skeletal (stromal) stem cells (BM-MSCs) in a cross-sectional, age-, sex-, and BMI-matched case-control study. X-chromosome inactivation was measured in blood leucocytes and BM-MSCs using a PCR method with methylation of HpaII sites. NF-κB activity was quantitated in BM-MSCs using a luciferase NF-κB reporter assay.

Results: Seven Caucasian women with IP (age: 24-67 years and BMI: 20.0-35.2 kg/m²) and IKBKG mutation (del exon 4-10 (n = 4); c.460C>T (n = 3)) were compared to matched controls. The IKBKG mutation carriers had extremely skewed X-inactivation (>90:10%) in blood, but not in BM-MSCs. NF-κB activity was lower in BM-MSCs from IKBKG mutation carriers (n = 5) compared to controls (3094 ± 679 vs. 5422 ± 1038/μg protein, p < 0.01). However, no differences were identified on skeletal radiographics, aBMD, μ-architecture of the iliac crest, or bone turnover markers. The IKBKG mutation carriers had a 1.7-fold greater extent of eroded surfaces relative to osteoid surfaces (p < 0.01), and a 2.0-fold greater proportion of arrested reversal surface relative to active reversal surface (p < 0.01).

Conclusion: Unlike mutation-positive males, the IKBKG mutation-positive women did not manifest OPT.

Keywords: IKBKG; Incontinentia pigmenti; NEMO; NF-κB; Osteopetrosis; X-chromosome inactivation.

Figure 1.

Radiographic comparison of a IKBKG mutation-positive woman with IP versus control (A) Anteroposterior pelvis from IKBKG mutation-positive carrier # 1 (B) Anteroposterior pelvis from matched control. There is no suggestion of an osteopetrosis phenotype in the IKBKG-mutation positive carrier. Micro-CT based bone structure in mutation-positive women versus controls. 3D reconstruction of consecutive micro-CT images of trabecular bone in iliac crest biopsies from IKBKG mutation-positive carrier (C) and control (D). There is no suggestion of altered structure of ilial cancellous bone in the IKBKG mutation carrier.

Figure 2.

Histology and histomorphometry of iliac crest bone in IKBKG mutation-positive carriers versus controls. Micrographs of Goldner Trichrome stained sections printed as maps show eroded surface (ES), osteoid surface (OS), osteoclast surface (Oc.S), reversal surface (Rv.S), and quiescent surface (QS) marked on the bone surfaces. The prevalence’s were calculated by intercepts (indicated by a black circle) between the particular surfaces and a cycloid grid (A). Histological appearance of an active eroded surface with OC and osteoid surfaces (B) and arrested Rv.Ss with no neighboring osteoid or Oc surfaces showing superficial erosion (C) and deep erosion (D). IKBKG mutation carriers show 1.7–fold higher extent of eroded surfaces relative to osteoid surfaces (E) and 2.0–fold higher proportion of arrested vs. active Rv.S (F). The arrested vs. active Rv.S have a 5.2–fold higher proportion of superficial resorption depths relative to deep erosion depths (G). MS/BS was significantly higher in controls vs. IKBKG mutation carriers (0.022 [0.019–0.029] vs. 0.013 [0.010–0.016], p=0.03) (H), whereas no significant difference are found in mineral apposition rate (MAR) comparing IKBKG mutation carriers with controls (I). Fluorescent images of tetracycline double labeled iliac crest biopsy of an IKBKG mutation carrier and a control. The white arrows mark the tetracycline label (J).

Figure 3

(A) Number of TRAP-positive Ocs in IKBKG mutation carriers (n=7) vs. controls (n=7). Data are presented as mean ± SEM. No differences are found between IKBKG mut⁺ carriers vs. controls (121±57 vs. 150±61, p=0.75). (B) Representative pictures of TRAP (purple) staining of Ocs in controls and IKBKG mut⁺ carriers, (scale bar 200μm). (C) mRNA levels (A.U.) of p65 (RelA) in Ocs before and after LPS stimulation. IKBKG mut⁺ (n=7) vs. controls (n=7) (0.003 ±0.001 vs. 0.002±0.001, p=0.59) (D) mRNA levels of TNFα (A.U.) in Ocs before and after LPS. IKBKG mut⁺ vs. controls (0.061 ± 0.014 vs. 0.062 ± 0.028, p=0.98) Data are presented as mean ± SEM.

Figure 4

(A) NF-κB activity expressed as luciferase activity normalized to protein concentration (A.U.) in BM-MSCs before and after LPS stimulation (LPS+). Controls (n=5) vs. IKBKG mutation carrier (n=5). Data are presented as mean ± SEM. IKBKG mutation carriers have lower NF-κB activity in BM-MSCs compared to controls. (3094 ± 679 vs. 5425 ± 899 μg protein, p<0.01). (B) NF-κB activity in BM-MSCs expressed as luciferase activity normalized to protein concentration (A.U.) before and after LPS stimulation (LPS+). Controls (n=5) vs. IKBKGexon4_10del (n=2) and IKBKG, c.460C>T (n=3). Data are presented as mean ± SEM IKBKGexon4_10del showing lower NF-κB activity in BM-MSCs compared to IKBKG, c.460C>T and controls (2922±209 vs. 3209±922/μg protein, p<0.01). (C) Levels of mRNA, p65 (RelA) (A.U) in BM-MSCs before and after LPS stimulation. Controls vs. IKBKG mutation carriers. Data are presented as mean ± SEM. IKBKG mutation carriers showing lower gene expression of p65 (RELA) compared to controls. (0.0031 ± 0.0008, vs. 0.0020 ± 0.0008 p<0.01). (D) Levels of mRNA, TNFα (A.U) in BM-MSCs before and after LPS. Controls (n=5) vs. IKBKG mutation carriers (n=5). Data are presented as mean ± SEM. IKBKG mutation carriers have lower gene expression of TNFα compared to controls. (0.0039± 0.0020 vs. 0.0010 ± 0.0001, p<0.01). (E-F) Osteoblast differentiation potential of the BM-MSC evaluated by (E) quantification of ALP activity (day 10) (1.68 ± 0.76 vs. 1.96 ± 0.34, p=0.76) and (F) gene expression of osteoblastic marker genes in IKBKG mutation carriers (n=5) vs. controls (n=5): (RUNX2: 1.33±0.26 vs. 1.06 ±0.09, p=0.34; ALPL: 11.58±3.81 vs. 7.04±2.99, p= 0.38 ; Osteocalcin: 150.36 ±12.08 vs. 59.03 ± 25.19, p=0.01) represented as fold change (F.C.) over non-induced cells. Data are presented as mean ± SEM.