A new hepatic biomarker in MMA: Mice help patients
Irini Manoli MD, PhD and Charles P. Venditti MD, PhD
Organic Acid Research Section
National Human Genome Research Institute
National Institutes of Health
Bethesda, MD 20892
Emails: firstname.lastname@example.org and email@example.com
This article provides a basic summary of a paper we recently published entitled “FGF21 underlies a hormetic response to metabolic stress in methylmalonic acidemia.” Our manuscript describes the construction of a new mouse model of MUT MMA, and its use, in combination with patient samples, to identify a previously unrecognized hormonal axis that is massively dysregulated in both the patients and mice. Of special importance, we further describe the use of this hormone, called Fibroblast Growth Factor 21 (FGF21), as a biomarker to measure the efficacy of liver-directed gene therapy in MMA.
Before we discuss the details of our scientific discovery, we would like to express our deepest appreciation and gratitude to all the patients and families who participated in our research at NIH. In our paper, we report the results of FGF21 levels in the blood and studies using liver and kidney samples that were donated by patients at the time of surgery. All these samples were of critical importance to prove that our observations in the mice were occurring in patients with MMA. We have published the information and our paper, including all the graphs, figures, tables, gene expression data, etc are freely available to anyone with an internet connection; the citation and links are at the end of this article.
Why do we need more mouse models and what can we learn from them? By the end of this article, we will hope to help you understand the importance of animal models, how they can be used to provide insight into disease processes, and what the ramifications can be for patients. With so many different mouse models of MMA, why do we need another? The first MMA mice we made, so called knock-out mice, were very sick animals, and usually die within the first few days of life. Having such severe and affected mice was indeed useful, and we published a number of very important papers testing gene therapy in these mice. In fact, they remain as the “gold standard” to test the efficacy of new treatments for MMA because the experimental read-out can be answered simply, and usually within a week, by answering the question: do the MMA mice survive after they receive the therapy? Using this approach, we have developed and tested both adenoviral and adeno-associated viral gene therapies for MMA with great success, and have used the neonatal lethal MMA knock out mice to find the lowest dose needed for treatment. Only the most potent treatments can rescue the full knock out mouse. However, the main limitation of the full knock out MMA (and PA) mice is the severity, and need to treat so early in life, which makes extension of the research observations more difficult. As examples, consider that a newborn mouse has the equivalent gestational age of a very premature human baby, perhaps in the 20-25 week of age range, and 9 mouse days = 1 human year.
Our first studies used these very sick MMA mice to answer the question: how effective is correction of the liver to the phenotype (appearance and behavior) seen in MMA mice? To do this, we used transgenesis to create mice that expressed the missing enzyme (MUT) only in the liver. These mice, called ALB MMA, are a rough equivalent of what it might be like if a baby with MMA was born with a liver transplant. As predicted, the ALB MMA mice were very healthy and active, and rescued from the lethal effects of mouse MMA. In parallel, we performed liver directed gene therapy experiments in the MMA knock out mice, and also could show that when a virus brought the MUT gene into the liver immediately after birth, the mice benefited like they did with a liver transgene, and were rescued from death.
We were next faced with a theoretical question: what would be the effect of restoring the missing enzyme in a given cell type, or tissue? Using an approach akin to what we did to probe liver effects, we made mice that expressed the MUT enzyme in the muscle (skeletal and cardiac) of the MMA knock out mice and explored the effects. We call these mice MCK MMA mice after the promoter used to drive the expression of MUT (MCK).
The first important observation made with these mice relates to survival: the expression of the MUT enzyme in the skeletal muscle rescued the MMA mice from lethality. This was very exciting because it further supported a previous observation that the MUT enzyme was expressed in all cells, and suggests that any cell that is missing the enzyme might be helped if it could make MUT. The other basic and critical observation was that the MCK MMA mice, while they survived, were fragile, growth retarded and had very high levels of methymalonic acid in the blood. If the mice were stressed, for example, by getting wet in their cages from the water bottles, they died. And despite being fed a mixture of high fat mouse food, a maple syrup like jelly that mice LOVE, and fruit, the MCK MMA mice do not grow like normal mice, in fact are severely growth retarded and typically only achieve about 50% the weight of the littermates that do not have MMA.
To understand why this was happening, we conducted a number of studies that are described in great detail in the manuscript. In brief, we suspected, and proved, that the MCK MMA mice had a severe mitochondrial dysfunction syndrome of the hepatocyte, the main cell of the liver, and the proximal tubule, similarly, an important cell in the kidney. We were able to characterize the cellular changes in the liver using electron microscopy, which enables the visualization of mitochondria (normally too small to see with a regular microscope), and make a comparison to the pattern seen in liver samples that were donated from patients. The combination of mouse and human studies enabled us to visually reconstruct the pathway that shows how mitochondria “self-destruct” due to MMA. This is depicted below in the patient samples where the progression begins with an enlarged and pale mitochondrial (far left), that then folds in upon itself, and we believe, eventually gets engulfed into a cellular structure called an autophagolysosome (far right).
Because the MCK MMA mice replicated the classic clinical findings seen in MMA patients, such as severe growth failure, liver mitochondrial dysfunction, chronic renal disease and a very low tolerance to different stressors (fasting, cold, high protein), we used the mice to model a metabolic crisis by exposing them to a fasting challenge. A major source of morbidity, hospital admissions and, in the most severe cases, mortality, are the episodes of metabolic instability and rapid deterioration when patients are stressed and have an accompanying reduced food intake. Control and mutant MCK MMA mice were therefore fasted and then sacrificed, and their livers removed. The livers were then used to profile the expression of all the genes in the genome ( ~ 20,000 genes) to identify markers that would be candidates to measure in MMA patients to assess the severity of the dysfunction in their hepatic mitochondria. We found many new candidate biomarkers with this genomic approach, and by analyzing the pattern of changes in the genes with our bioinformatics colleagues, discovered that MMA, whether in a mouse or person, causes stress pathways to be chronically activated to adapt (ei hormesis) to the metabolic injury that accompanies the enzyme deficiency.
One hormone, Fgf21 (fibroblast growth factor 21), was extremely elevated in the MCK MMA mice and further studied in patients. FGF21 is an important but only recently recognized hormone that is made by the liver. In patients with MMA, we found that the levels were highest in the sicker and more severely affected children, but normalize after liver transplant, with changes that were greater than conventional metabolites, such as serum methylmalonic acid levels. It should be noted that serum [MMA] does NOT normalize after liver, kidney or combined liver-kidney transplantation in patients with MMA. To directly test this prediction, we treated the MMA MCK mice with a liver-directed gene therapy vector. This virus, an adeno-associated virus (AAV), was designed to only express the MUT gene in the liver, nowhere else, and was given to the MCK MMA mice at a dose that has been given to other patients (not with MMA) in different studies. We found that after treatment with a single and relatively low dose of our new liver directed AAV gene therapy vector, the MMA MCK mice had marked metabolic improvement, increased weight gain and activity, and showed a drastic reduction in the levels of FGF21 in the blood. Thus, FGF21 represents an ideal biomarker to not only help predict which patient might have more severe liver mitochondrial involvement, but also to monitor the efficacy of new hepatic treatments, such as gene and mRNA therapy, for MMA and related disorders.
Hopefully, we have described how this new mouse model offers a unique platform to study disease mechanisms and test advanced therapies for MMA. As organ transplantation for methylmalonic acidemia becomes more available, and promising genomic therapies move closer to clinical trials, selecting patients who may derive the most benefit from available treatments will become challenging. Additional clinical and laboratory parameters that can indicate the severity of specific organ involvement can help families and their physicians decide between an isolated kidney or liver, as opposed to combined liver-kidney transplant, and perhaps between new experimental treatments. Moreover, monitoring each organ’s positive or adverse reaction to the different therapies will be critical for the success of any new treatment for MMA.
This massive study was conducted over many years at NIH, had over 30 co-authors with aggregate expertise in biochemical genetics, laboratory methods, bioinformatics, gene therapy, statistics and mouse genetics, and is the culmination of our deep collaboration with the patient community, who, with the NIH, has offered continued encouragement of our research efforts. Many of our students who helped with these mice were supported, in part, by the Angels for Alyssa (MMA Research Fund), Cure for Clark, and the Organic Acidemia Association. We would again like to express our appreciation for the amazing dedication of the patients and their families for participation in our clinical research protocol and invite comments and questions by email.
Manoli I, Sysol JR, Epping MW, Li L, Wang C, Sloan JL, Pass A, Gagné J, Ktena
YP, Li L, Trivedi NS, Ouattara B, Zerfas PM, Hoffmann V, Abu-Asab M, Tsokos MG,
Kleiner DE, Garone C, Cusmano-Ozog K, Enns GM, Vernon HJ, Andersson HC, Grunewald
S, Elkahloun AG, Girard CL, Schnermann J, DiMauro S, Andres-Mateos E,
Vandenberghe LH, Chandler RJ, Venditti CP. FGF21 underlies a hormetic response to
metabolic stress in methylmalonic acidemia. JCI Insight. 2018 3(23). PubMed PMID:
30518688; PubMed Central PMCID: PMC6328030.