Metabolic phenotype of methylmalonic acidemia in mice and humans: the role of skeletal muscle

Abstract

Background: Mutations in methylmalonyl-CoA mutase cause methylmalonic acidemia, a common organic aciduria. Current treatment regimens rely on dietary management and, in severely affected patients, liver or combined liver-kidney transplantation. For undetermined reasons, transplantation does not correct the biochemical phenotype.

Methods: To study the metabolic disturbances seen in this disorder, we have created a murine model with a null allele at the methylmalonyl-CoA mutase locus and correlated the results observed in the knock-out mice to patient data. To gain insight into the origin and magnitude of methylmalonic acid (MMA) production in humans with methylmalonyl-CoA mutase deficiency, we evaluated two methylmalonic acidemia patients who had received different variants of combined liver-kidney transplants, one with a complete liver replacement-kidney transplant and the other with an auxiliary liver graft-kidney transplant, and compared their metabolite production to four untransplanted patients with intact renal function.

Results: Enzymatic, Western and Northern analyses demonstrated that the targeted allele was null and correctable by lentiviral complementation. Metabolite studies defined the magnitude and tempo of plasma MMA concentrations in the mice. Before a fatal metabolic crisis developed in the first 24-48 hours, the methylmalonic acid content per gram wet-weight was massively elevated in the skeletal muscle as well as the kidneys, liver and brain. Near the end of life, extreme elevations in tissue MMA were present primarily in the liver. The transplant patients studied when well and on dietary therapy, displayed massive elevations of MMA in the plasma and urine, comparable to the levels seen in the untransplanted patients with similar enzymatic phenotypes and dietary regimens.

Conclusion: The combined observations from the murine metabolite studies and patient investigations indicate that during homeostasis, a large portion of circulating MMA has an extra-heptorenal origin and likely derives from the skeletal muscle. Our studies suggest that modulating skeletal muscle metabolism may represent a strategy to increase metabolic capacity in methylmalonic acidemia as well as other organic acidurias. This mouse model will be useful for further investigations exploring disease mechanisms and therapeutic interventions in methylmalonic acidemia, a devastating disorder of intermediary metabolism.

Figure 1

Targeting construct and null allele. (A) An overview of the targeting construct and disrupted allele. Selected restriction sites are indicated. The initiator codon lies in the second coding exon and the coenzyme A binding pocket in the third coding exon. Recombination sites (loxP), the diptheria toxin A cassette (DTA) and neomycin resistance gene (NEO) are shown. The location of genotyping primers that flank the 5' loxP site are indicated by A and B.

Figure 2

Northern, Western, enzymatic and correction studies. (A) Northern analysis of total RNA extracted from wild-type and Mut knockout livers. The mutase message is apparent in the wild-type sample and absent in the Mut mutant liver. Actin hybridization after stripping and reprobing shows the RNA to be intact and equally reactive. (B) Western blotting using anti-mutase antisera reveals a band of ~80 kd in the wild-type liver extracts that is completely absent from the mutant liver extracts. (C) [1-¹⁴C] propionic acid incorporation, with and without vitamin B12, in various cell lines, expressed as percent of wild-type activity. The wild type activity varied between days and ranged from 5.0–15.0 nmol C14 propionate/mg protein/18 hrs. Error bars surround the standard deviation from triplicate measurements. The corrected Mut null murine cell line (labeled MUTO + MCM LENTI) shows restored propionate flux when compared with the Mut null and GFP-transduced cell lines.

Figure 3

Pre- and postnatal plasma and urine methylmalonic acid levels. (A) MMA concentrations (μM) measured in the plasma over time. Error bars surround the standard deviation. Wild-type (N = 4–8 age matched littermates at each time point, not displayed), Mut null prenatal [E19] (N = 3), Mut null postnatal 4–6 hours (N = 4), Mut null postnatal 8–12 hours (N = 4), and Mut null postnatal 20–24 hours (N = 4). Low-level wild-type values not displayed. An asterisk * designates a significant p-value, calculated using a one-way ANOVA with Tukey-Kramer adjustment, for comparison between advancing time points. Prenatal (value = 176 μM) vs. 4–6 hours [F (1,6) = 107; p < 0.0001], 4–6 hr vs 8–12 hours [F (1,6) = 54; p < 0.0001] and 8–12 hours vs 20–24 hours [F (1,4) = 152; p < 0.0001]. At all time points, the mutants were different from the unaffected littermates with p-values less than 0.001. (B) MMA concentrations (mM) in urine or amniotic fluid on day E19 (value = 117 μM) from the samples above. An asterisk * designates a significant p-value, calculated using a one-way ANOVA with Tukey-Kramer adjustment, for comparison between advancing time points. Prenatal vs. 4–6 hours [F (1,6) = 30; p = 0.0002] and 4–6 hr vs 8–12 hours [F (1,6) = 49; p < 0.0001]. However, at 8–12 hours vs 20–24 hours the differences were not significant [F (1,4) = 0.14; p = 0.7143]. In all cases, the mutants were different from the unaffected littermates (not displayed) with p-values less than 0.001.

Figure 4

Methylmalonic acid concentration by tissue type over time. (A) MMA concentrations expressed as nmol per gram of tissue. The values are averages from prenatal [embryonic day 19] (N = 3), neonatal [8–12 hour] (N = 3), and metabolic crisis [20–24 hour] (N = 2) mutant animals. Values for age-matched control littermates (N = 2) for each time point are not depicted on this graph but were used to plot the fold change, depicted in additional file 4. The error bars surround the standard deviation for the prenatal time point and the range observed at the stage of metabolic crisis. p-values for the mutant versus age matched controls were <0.01 for all tissues. An asterisk * designates a significant p-value. In the mutants, the MMA content of the liver at 24 hours was significantly greater than skeletal muscle (p = 0.002 at 8 hours, p = 0.05 at 24 hours), brain (p = 0.002 at 8 hours, p = 0.0016 at 24 hours), or kidney (p = 0.0023 at 8 hours, p = 0.0072 at 24 hours). However, only the difference between the liver samples at the 24-hour time point and other tissues at the 8 and 24-hour time points were significant with a p-value less than 0.05.

Figure 5

Biochemical parameters in MMA patients. Plasma MMA levels (μM) and 24-hour urine MMA output (mmol/kg/d) in non-transplanted patients (mut^o-patients 1–4) with preserved renal function (A) compared to patients who have received solid organ transplants (L(A)KT, LKT) (B). * p-values for the average plasma MMA concentrations and MMA output from the untransplanted patients versus the transplanted patients. Neither the plasma metabolites [F(1,4) = 0.40, p = 0.56] or the MMA output [F (1,4) = 0.70, p = 0.44] was significantly different between the groups.