Ighmbp2 mutations and disease pathology: Defining differences that differentiate SMARD1 and CMT2S

Abstract

Mutations in the Immunoglobulin mu DNA binding protein 2 (IGHMBP2) gene result in two distinct diseases, SMA with Respiratory Distress Type I (SMARD1) and Charcot Marie Tooth Type 2S (CMT2S). To understand the phenotypic and molecular differences between SMARD1 and CMT2S, and the role of IGHMBP2 in disease development, we generated mouse models based on six IGHMBP2 patient mutations. Previously, we reported the development and characterization of Ighmbp2^D564N/D564N mice and in this manuscript, we examine two mutations: D565N (D564N in mice) and H924Y (H922Y in mice) in the Ighmbp2^H922Y/H922Y and Ighmbp2^D564N/H922Y contexts. We found significant differences between these mouse models, providing critical insight into the role of IGHMBP2 in the pathogenesis of SMARD1 and CMT2S. Importantly, these studies also demonstrate how disease pathogenesis is significantly altered in the context of Ighmbp2 D564N and H922Y homozygous recessive and compound heterozygous mutations. Notably, there were short-lived and long-lived lifespan cohorts within Ighmbp2^D564N/H922Y mice with early (P12/P16) respiratory pathology serving as a key predictor of lifespan. Despite differences in lifespan, motor function deficits initiated early and progressively worsened in all Ighmbp2^D564N/H922Y mice. There was decreased limb skeletal muscle fiber area and increased neuromuscular junction (NMJ) denervation in Ighmbp2^D564N/H922Y mice. Consistent with CMT2S, Ighmbp2^H922Y/H922Y mice did not have altered lifespans nor respiratory pathology. Interestingly, Ighmbp2^H922Y/H922Y limb muscle fibers demonstrated an increase in muscle fiber area followed by a reduction while changes in NMJ innervation were minimal even at P180. This is the first study that demonstrates differences associated with IGHMBP2 function within respiration with those within limb motor function. Significant to our understanding of IGHMBP2 function, we demonstrate that there is a direct correlation between disease pathogenesis associated with these IGHMBP2 patient mutations and IGHMBP2 biochemical activity. Importantly, these studies reveal the dynamic differences that are presented when either a single mutant protein is present (IGHMBP2-D564N or IGHMBP2-H922Y) or two mutant proteins are present (IGHMBP2-D564N and IGHMBP2-H922Y) within cells.

Keywords: IGHMBP2; Mouse models; Neurodegeneration; Respiration; SMARD1/CMT2S.

Fig. 1.

Survival and fitness were improved in Ighmbp2^D564N/H922Y and Ighmbp2^H922Y/H922Y mice. (A) Graphic representation of the mouse IGHMBP2 protein indicating the location of the D564N and H922Y mutations and the primary IGHMBP2 domains (helicase, R3H and zinc finger (ZnF)). (B) Kaplan-Meier plot of survival proportions. Ighmbp2^+/+ (black, n=16), Ighmbp2^D564N/D564N (orange, n=11), Ighmbp2^D564N/H922Y (blue, n=37), Ighmbp2^H922Y/H922Y (red, n=27). Log-rank (Mantel-Cox) test determined significance. (C) Representative photographs of Ighmbp2^+/+ and Ighmbp2^D564N/H922Y animals at P10 and P50. Representative photographs of Ighmbp2^+/+ and Ighmbp2^H922Y/H922Y animals at P50 and P100. (D) Weight graphed as mean ±SEM for Ighmbp2^+/+ (black, n=16), Ighmbp2^D564N/D564N (orange, n=14), Ighmbp2^D564N/H922Y (blue, n=9), Ighmbp2^H922Y/H922Y (red, n=21). Ordinary one-way ANOVA with Tukey’s multiple comparisons and an unpaired, two-tailed t test were used to determine statistical significance. n=number within sample.

Fig. 2.

Muscle weakness varied between Ighmbp2 mutant mice. (A) Latency to fall measured in seconds (mean ±SEM). Ighmbp2^+/+ (gold, n=11), Ighmbp2^H922Y/H922Y (red, n=9), Ighmbp2^D564N/H922Y (blue, n=11), Ighmbp2^D564N/D564N (orange, n=11). Ighmbp2^D564N/D564N mice were statistically different from Ighmbp2^+/+ mice on days 5–11 (P<0.0001). Ighmbp2^D564N/H922Y were statistically significant from Ighmbp2^+/+ mice from day 7. There was no statistical difference between Ighmbp2^+/+ and Ighmbp2^H922Y/H922Y mice at any timepoint. Representative latency to fall images of Ighmbp2^+/+ and Ighmbp2^D564N/H922Y mice at P8. (B) Time-to-right from P8–P100 for Ighmbp2^+/+ (gold, n=16), Ighmbp2^D564N/H922Y (blue, n=21), Ighmbp2^D564N/D564N (orange, n=8) and Ighmbp2^H922Y/H922Y (red, n=27) was measured in seconds (mean ±SEM). (C) Hindlimb splay assay from P8–P100 for Ighmbp2^+/+ (gold, mean=3.0, n=16), Ighmbp2^D564N/D564N (orange, mean=0.7, n=11), Ighmbp2^D564N/H922Y (blue, mean=0.5, n=21), Ighmbp2^H922Y/H922Y (red, mean=2.6, n=27) (mean ±SEM). There was statistical significance between Ighmbp2^+/+ and all mutants P<0.0001. Hindlimb weakness score 3=fully separated hindlimbs/paws past body, 2=hindlimbs separated to a hip-width distance, 1=hindlimbs placed at midline, 0=retracted hindlimbs/no hindlimb separation. (D) Speed measured at P22 using a CatWalk analysis system (mean cm/second ±SEM). Ighmbp2^+/+ mean=22.43 cm/second, Ighmbp2^D564N/H922Y mean=2.07 cm/second, Ighmbp2^H922Y/H922Y=20.36 cm/second (**** P<0.0001). There was not a significant difference between Ighmbp2^+/+ and Ighmbp2^H922Y/H922Y mice. (E) P100 Ighmbp2^H922Y/H922Y left and right paw print positions measured using a CatWalk analysis system. Left paw analyses (WT=0.111cm, n=11; H922Y/H922Y=0.176cm, n=20) and right paw analyses (WT=0.111cm, n=11; H922Y/H922Y=0.188cm, n=20). There was no statistical significance. Two-tailed, unpaired t test was used to determine significance. Two-way ANOVA multiple comparisons test with mixed-effects analysis was used for statistical analyses in A and B. Two-way ANOVA multiple comparisons test was used in C. One-way ANOVA multiple comparisons test was used to determine significance in D and an unpaired t test was used for statistical analyses in E. Data points/dots represent mice. N=number within sample, WT=wild type, cm=centimeter.

Fig. 3.

Ighmbp2^D564N/H922Y mice demonstrated a progressive reduction in muscle fiber area. Quantification of gastrocnemius and tibialis anterior (TA) muscle fiber area in Ighmbp2^+/+ (gold) and Ighmbp2^D564N/H922Y (blue) mice. (A) P12 gastrocnemius area WT=143.70μm², 1331 fibers, 6 animals; D564N/H922Y=123.60μm², 2046 fibers, 6 animals. (B) P26 gastrocnemius area WT=701.20μm², 725 fibers, 5 animals; D564N/H922Y=129.60μm², 249 fibers, 6 animals. (C) P50 gastrocnemius area WT=616.80μm², 636 fibers, 5 animals; D564N/H922Y=75.01μm², 651 fibers, 5 animals. (D) P12 TA area WT=344.30μm², 351 fibers, 4 animals; D564N/H922Y=153.70μm², 697 fibers, 4 animals. (E) P26 TA area WT=544.60μm², 290 fibers, 5 animals; D564N/H922Y=180.80μm², 620 fibers, 5 animals. (F) P50 TA area WT=638.90μm², 492 fibers, 6 animals; D564N/H922Y=148.90μm², 619 fibers, 6 animals. (G) Representative images of Ighmbp2^+/+ and Ighmbp2^D564N/H922Y gastrocnemius tissue cross section at P12 with anti-laminin immunohistochemistry. (H) Representative images of Ighmbp2^+/+ and Ighmbp2^D564N/H922Y gastrocnemius tissue cross section at P26 with anti-laminin immunohistochemistry. Statistical significance was determined using an unpaired, two-tailed t test with data expressed as mean ±SEM, ****P<0.0001. Data points/dots represent muscle fibers. WT=wild type.

Fig. 4.

Ighmbp2^H922Y/H922Y mice demonstrated muscle fiber size variation with disease progression. Quantification of gastrocnemius and TA muscle fiber area in Ighmbp2^+/+ (gold) and Ighmbp2^H922Y/H922Y (red) mice. (A) P12 gastrocnemius area (WT=210.10μm², 588 fibers, 3 animals; H922Y/H922Y=177.00μm², 803 fibers, 4 animals), P<0.0001. (B) P26 gastrocnemius area (WT=529.80μm², 285 fibers, 3 animals; H922Y/H922Y=557.50μm², 197 fibers, 3 animals. (C) P100 gastrocnemius area (WT=917.40μm², 168 fibers, 3 animals; H922Y/H922Y=709.00μm², 220 fibers, 3 animals), P<0.0001. (D) P180 gastrocnemius area (WT=1142.00μm², 149 fibers, 5 animals; H922Y/H922Y=921.20μm², 230 fibers, 5 animals), P<0.0001. (E) P12 TA area (WT=236.00μm², 398 fibers, 3 animals; H922Y/H922Y=213.70μm², 460 fibers, 4 animals), P=0.0011. (F) P26 TA area (WT=467.50μm², 318 fibers, 3 animals; H922Y/H922Y=603.40μm², 165 fibers, 3 animals), P<0.0001. (G) P100 TA area (WT=817.40μm², 197 fibers, 3 animals; H922Y/H922Y=743.30μm², 203 fibers, 3 animals), P=0.0440. (H) P180 TA area (WT=907.70μm², 121 fibers, 5 animals; H922Y/H922Y=489.10μm², 156 fibers, 5 animals), P<0.0001. Statistical significance was determined using an unpaired, two-tailed t test with data expressed as mean ±SEM. Data points/dots represent muscle fibers. WT=wild type.

Fig. 5.

Gastrocnemius and TA muscle NMJ denervation was prevalent early in Ighmbp2^D564N/H922Y mice. A-C Quantification of gastrocnemius percentage of endplates fully innervated (gold), partially innervated (blue) or fully denervated (red). (A) P12 WT FI=98.0%, PI=0.8%; D=1.2%; D564N/H922Y FI=66.3%, PI=10.3%, D=23.4%, (****P<0.0001), WT=16 animals, D564N/H922Y=21 animals. (B) P26 WT FI=99.3%, PI=0.7%; D564N/H922Y FI=21.8%, PI=16.2%, D=62.0%, (****P<0.0001), WT=18 animals, D564N/H922Y=16 animals. (C) P50 WT FI=95.7%, PI=3.5%, D=0.8%; D564N/H922Y FI=9.1%, PI=14.6%, D=76.3%, (****P<0.0001), WT=21 animals, D564N/H922Y=9 animals. D-F Quantification of tibialis anterior (TA) percentage of endplates fully innervated (gold), partially innervated (blue) or fully denervated (red). (D) P12 WT FI=90.9%, PI=3.7%, D=5.4%; D564N/H922Y FI=40.3%, PI=18.3%, D=41.4%, (****P<0.0001, ***P=0.0003), WT=20 animals, D564N/H922Y=25 animals. (E) P26 WT FI=100.0%; D564N/H922Y FI=33.2%, PI=17.4%, D=49.4%, (****P<0.0001, ***P=0.0009), WT=23 animals, D564N/H922Y=18 animals. (F) P50 WT FI=93.7%, PI=2.2%, D=4.1%; D564N/H922Y FI=31.8%, PI=26.2%, D=42.0%, (****P<0.0001), WT=28 animals, D564N/H922Y=19 animals. (G) Representative images of Ighmbp2^+/+ and Ighmbp2^D564N/H922Y P12 gastrocnemius. (H) Representative images of Ighmbp2^+/+ and Ighmbp2^D564N/H922Y P50 gastrocnemius. Anti-neurofilament heavy chain (NF-H) and anti-synaptic vesicle 2 (SV2) were used to label the axon and synaptic terminal. Acetylcholine receptors were labeled with Alexa Fluor 594-conjugated α-bungarotoxin. Images were analyzed based on the end plate overlap with the synaptic terminal. Statistical significance was determined using two-way ANOVA with Tukey’s multiple comparisons test with data expressed as mean ±SEM. Data points/dots represent mice. WT=Ighmbp2^+/+, FI=fully innervated, PI= partially innervated, D=fully denervated.

Fig. 6.

Respiratory deficits are present at P26 and P50 in Ighmbp2^D564N/H922Y mice. Quantitative whole-body plethysmography analyses of frequency, tidal volume, minute ventilation and mean inspiratory flow in Ighmbp2^+/+ (gold) and Ighmbp2^D564N/H922Y (blue) mice. These measurements were taken under three conditions normoxia, hypercapnia and maximum challenge. (A) P12, Ighmbp2^+/+ n=14, Ighmbp2^D564N/H922Y n=16, *P=0.0294. (B) P26, Ighmbp2^+/+ n=8, Ighmbp2^D564N/H922Y n=8, **P=0.0014, ****P<0.0001. (C) P50, Ighmbp2^+/+ n=7, Ighmbp2^D564N/H922Y n=7, **P=0.0014 frequency, **P=0.0021 mean inspiratory flow, ***P=0.0002, ****P<0.0001. (D) P100, Ighmbp2^+/+ n=9, Ighmbp2^D564N/H922Y n=6. Statistical significance was determined using two-way ANOVA with Tukey’s multiple comparisons test. Data is expressed as mean ±SEM. Data points/dots represent mice. mL=milliliter, n=number of animals.

Fig. 7.

P12/P16 plethysmography demonstrated short-lived Ighmbp2^D564N/H922Y mice were affected by significant respiratory deficiencies while long-lived mice were not. Quantitative whole-body plethysmography measured at P12 or P16 for frequency, tidal volume, minute ventilation and mean inspiratory flow in Ighmbp2^+/+ (black), long-lived Ighmbp2^D564N/H922Y (orange, P100+) and short-lived Ighmbp2^D564N/H922Y (blue, P25–30) mice. These measurements were taken under three conditions normoxia, hypercapnia and maximum challenge. (A) Frequency (WT N=282.60 breaths/minute, H=347.30 breaths/minute, MC=340.5 breaths/minute; long-lived D564N/H922Y N=303.80 breaths/minute, H=351.40 breaths/minute, MC=332.60 breaths/minute; short-lived D564N/H922Y N=261.30 breaths/minute, H=230.10 breaths/minute, MC=250.60 breaths/minute), *P=0.0215, **P=0.0025, ***P=0.0008. (B) Tidal volume (WT N=0.018 mL/gram, H=0.028 mL/gram, MC=0.027 mL/gram; long-lived D564N/H922Y N=0.020 mL/gram, H=0.033 mL/gram, MC=0.033 mL/gram; short-lived D564N/H922Y N=0.013 mL/gram, H=0.019 mL/gram, MC=0.017 mL/gram), *P=0.0367(normoxia), *P=0.0461(MC), ***P=0.0003, ****P<0.0001. (C) Minute ventilation (WT N=4.87 mL/minute/gram, H=9.70, MC=9.22; long-lived D564N/H922Y N=5.63 mL/minute/gram, H=11.55, MC=10.97; short-lived D564N/H922Y N=3.38 mL/minute/gram, H=4.41, MC=4.30), *P=0.0367(normoxia), ****P<0.0001. (D) Mean inspiratory flow (WT N=0.203 mL/second/gram, H=0.372, MC=0.334; long-lived D564N/H922Y N=0.238 mL/second/gram, H=0.422, MC=0.393; short-lived D564N/H922Y N=0.149 mL/second/gram, H=0.176, MC=0.171), ***P=0.0001, ****P<0.0001. Statistical significance was determined using two-way ANOVA with Tukey’s multiple comparisons test. Data is expressed as mean ±SEM. Data points/dots represent mice. mL=milliliter, n=number of animals, WT=wild type, N=normoxia, H=hypercapnia, MC=maximum challenge.

Fig. 8.

Diaphragm NMJ denervation was not progressive in Ighmbp2^D564N/H922Y nor Ighmbp2^H922Y/H922Y mice while diaphragm muscle fibers were reduced in size. A-D Quantification of diaphragm percentage of endplates fully innervated (gold), partially innervated (blue) or fully denervated (red). (A) P12 mice Ighmbp2^+/+ (9 animals) FI=99.4%, PI=0.6%; Ighmbp2^D564N/H922Y (10 animals) FI=89.6%, PI=7.4%, D=3.0% (P=0.0022). (B) P26 mice Ighmbp2^+/+ (15 animals) FI=98.1%, PI=1.4%, D=0.5%; Ighmbp2^D564N/H922Y (16 animals) FI=92.5%, PI=6.5%, D=1.0% (**P=0.0022, *P=0.0195). (C) P50 mice Ighmbp2^+/+ (17 animals) FI=96.9%, PI=1.7%, D=1.4%; Ighmbp2^D564N/H922Y (16 animals) FI=96.0%, PI=1.9%, D=2.1%. (D) P180 mice Ighmbp2^+/+ (12 animals) FI=98.5%, PI=0.5%, D=1.0%; Ighmbp2^H922Y/H922Y (29 animals) FI=98.4%, PI=1.4%, D=0.2%. Statistical significance was determined using two-way ANOVA with Tukey’s multiple comparisons test with data expressed as mean ±SEM. E-G Quantification of diaphragm muscle fiber area in Ighmbp2^+/+ (gold) and Ighmbp2^D564N/H922Y animals (blue). (E) P12 (Ighmbp2^+/+=243.10μm², fibers=324, 3 animals; Ighmbp2^D564N/H922Y=188.20μm², fibers=358, 3 animals), P<0.0001. (F) P26 (Ighmbp2^+/+=303.60μm², fibers=538, 3 animals; Ighmbp2^D564N/H922Y=207.50μm², fibers=512, 3 animals), P<0.0001. (G) P50 (Ighmbp2^+/+=357.70μm², fibers=525, 3 animals; Ighmbp2^D564N/H922Y=303.10μm², fibers=795, 3 animals), P<0.0001. (H) Quantification of diaphragm muscle fiber area in Ighmbp2^+/+ (gold) and Ighmbp2^H922Y/H922Y animals (red). P180 (Ighmbp2^+/+=266.80μm², fibers=387, 3 animals; Ighmbp2^H922Y/H922Y=236.80μm², fibers=363, 3 animals), P<0.0001. Statistical significance was determined using an unpaired, two-tailed t test with data expressed as mean ±SEM. Data points/dots represent mice in A-D. Data points/dots represent muscle fibers in E-H.

Fig. 9.

Quantification of diaphragm muscle fiber area in Ighmbp2^+/+ (gold) and Ighmbp2^D564N/H922Y (blue) mice. P12, P26 and P50 diaphragm cross sections of individual muscle fibers were binned by area into 100μm² intervals and expressed as percentage of muscle fibers. (A) P12 (WT 1=0.3%, 2=29%, 3=48%, 4=17%; D564N/H922Y 1=0.6%, 2=60%, 3=33%, 4=6%), *P=0.0109. (B) P26 (WT 1=2%, 2=12%, 3=39%, 4=28%; D564N/H922Y 1=15%, 2=20%, 3=25%, 4=24%). (C) P50 (WT 1=0.5%, 2=3%, 3=25%, 4=44%; D564N/H922Y 1=0.1%, 2=15%, 3=34%, 4=29%). WT=wild type, 1=0–99μm², 2=100–199μm², 3=200–299μm², 4=300–399μm². Statistical analyses using two-way ANOVA with Tukey’s multiple comparison test and row statistics were performed. Data points/dots represent mice. WT= wild type.