*3.1. Bioenergetics Testing in Mitochondrial Disorders*

Impaired oxidative phosphorylation is a hallmark of mitochondrial disorders with progressive clinical manifestations that can vary from single organ to multisystem disorders, often affecting organs with high energetic demands [26]. Mitochondrial diseases have an estimated prevalence of one in 4300 adults, making them the most common inherited metabolic disorder [27–30]. Defects in CI are the most commonly observed respiratory chain disorder [31–33], making the complex a good candidate for bioenergetic testing.

In patients suspected to have mitochondrial disorders, genetic testing and metabolic screening are recommended before biochemical assays to evaluate mitochondrial function due to the precision in categorizing and less-invasive nature of these measurements. Analyte testing of blood, urine, or even cerebrospinal fluid (CSF) in specific cases, for pyruvate, lactate, amino acids, creatine kinase, and carnitine, can provide information about alterations in metabolism in these patients and provide a biochemical metabolic signature of mitochondrial disease [34,35]. Additionally, pathogenic variants in over 400 genes, both mitochondrial and nuclear, have been identified as contributors to mitochondrial disorders [26]. Pathogenic mutations in mtDNA and nuclear DNA (nDNA) can be used to confirm primary mitochondrial disease due to abnormal OXPHOS function. A genetics-first approach has been facilitated by the availability of exome and whole genome sequencing that has improved the identification of mitochondrial disease genes diagnosis [36–38]. Identification of nuclear mitochondrial disease genes are also used to diagnose mitochondrial dysfunctions [39–42]. Secondary mitochondrial dysfunction, which can be influenced by environmental factors in addition to genetic mutations, adds a further level of complexity and can lead to metabolic and neurodegenerative diseases [43]. While advances in sequencing have made rapid diagnosis possible in a large number of patients, mtDNA or nDNA mutations cannot always be identified even in the presence of a clinical mitochondrial phenotype. Therefore, tissue testing for functional validation may still be necessary [42,44].

Functional, biochemical assays to detect alterations in enzymatic activity, respiration capacity, or ATP synthesis in primary tissue can be used to confirm mitochondrial defects as well as histopathology and mtDNA copy number measurement. However, it is impor-

tant to note that these assays cannot distinguish between primary defects in OXPHOS enzymes and secondary mitochondrial dysfunctions. Measurement of respirometry and ATP synthesis requires fresh tissue since freezing will disrupt the inner mitochondrial membrane and uncouple oxygen consumption from ATP synthesis. Given the need for same day processing and assaying, the expertise required to run these samples, and limitations in the availability of tissue, bioenergetic alternatives to respirometry and ATP synthesis measurements can be run in frozen samples, such as enzymatic assays measuring ETC complex activities.

Although the electron transport chain and ATP synthesis uncouple when samples are frozen, the mitochondrial electron transport complexes remain structurally intact and functional [45,46]. For that reason, a common method used to assess mitochondrial function in frozen samples has been to measure spectrophotometric enzymatic assays that provide information on the activity of individual ETC complexes or the combination of CI + CIII or CII + CIII. The protocols to measure enzymatic activities often use supraphysiological concentrations of some reagents and non-physiological electron donors and acceptors. Table 3 provides an overview of assays to measure enzymatic activities in mitochondria. Although these measurements were successfully used in a relatively high-throughput manner to diagnose primary mitochondrial diseases, namely diseases caused by a primary defect in electron transport chain function [47–49], they cannot provide a single measurement of the coordinated function of the electron transport chain function working at more physiological rates. Taken together, these observations suggest that spectrophotometric assays might be less sensitive to detect defects in mitochondrial architecture and complex ultrastructure, such as the ones associated with secondary mitochondrial dysfunction. However, since the integrated measurement of mitochondrial respiration or ATP synthesis is often unfeasible in these samples these assays provide a useful alternative.

**Table 3.** Spectrophotometric approaches to measure mitochondrial enzymatic activities for assessment of mitochondrial bioenergetic function. The table presents the metabolic enzyme of interest, the metabolic pathway the measurement would evaluate, and methodology references. Abbreviations: ATP: Adenosine triphosphate; CI: Complex I; CII: Complex II; CIII: Complex III; CIV: Complex IV; DH: Dehydrogenase; ETC: Electron transport chain; NADH: Reduced nicotinamide adenine dinucleotide; OXPHOS: Oxidative phosphorylation; TCA: Tricarboxylic acid cycle.


#### *3.2. Skeletal Muscle Biopsies*

Muscle biopsies have long been the gold standard for investigating changes in mitochondrial function via respirometry in human samples [9,18]. Measurements of mitochondrial function can be performed in intact muscle fibers, permeabilized muscle fibers, or isolated mitochondria from muscle biopsies [13,18]. Skeletal muscle biopsies provide a terminally differentiated, post-mitotic sample that requires less mtDNA replication and may maintain stable levels of mtDNA heteroplasmy compared with mitotic cells [72–74]. Skeletal muscles have both high energy demand and mitochondrial content making skeletal muscle a valuable primary tissue sample to measure mitochondrial respiration and ATP synthesis. Measuring respiration in muscle fibers offers the ability to probe specific complexes of the ETC, providing information about underlying mechanisms of altered mitochondrial function. As mentioned previously, mitochondrial bioenergetics can also be assessed in these samples through spectrophotometric assays to measure different metabolic pathways (Table 3) and, in the case of citrate synthase, provide a measure of mitochondrial content. Histology can provide information about mitochondrial structure and bioenergetics through commonly used stains such as modified Gomori trichrome (muscle structure, mitochondrial accumulation, and ragged red fiber), cytochrome oxidase (CIV), succinate dehydrogenase (CII, nDNA encoded), and NADH dehydrogenase (CI), as well as stains to assess intramuscular lipid accumulation [75].

While a wealth of mitochondrial information can be gained from skeletal muscle, this technique can be invasive and not available outside clinical settings or the amount of material obtained can be limiting depending on the technique used [75,76]. For example, open biopsies that require a clinical setting are invasive but provided plenty of material while needle biopsies are less invasive, but the amount of sample obtained may be limiting. While a needle biopsy does not require an operative setting, it does require local anesthesia, sterile conditions, and some specialization to obtain the sample [75,77]. Therefore, there has been a push towards the use of more accessible samples with the development of new techniques. To that point, in vivo, non-invasive applications to measure mitochondrial respiratory capacity are available, such as near-infrared spectroscopy (NIRS), which corresponds with the high-resolution respirometry in muscle biopsies [78]. While the use of this and other in vivo techniques is suitable to study human mitochondrial function, there is still a need for more specific, higher throughput techniques that require less expertise to run for use in translational research studies and clinical applications.

While testing of muscle biopsies was considered the gold standard, before the genomics era, to measure mitochondrial bioenergetic function and diagnosis of primary mitochondrial disease [79], alternative methods would be preferentially useful for diagnosis and research when genetic screening is not enough for categorization of the deficiency [80]. Identifying suitable alternatives to muscle biopsies and systemic biomarkers of mitochondrial function has become a major research focus. This is due to the fact the evaluation of mitochondrial bioenergetic function remains an important parameter for basic research and, in some cases, diagnosis, as a way to determine the extent of mitochondrial dysfunction. Functional readouts of mitochondrial bioenergetics also have applications for translation research and for clinical trials where genetic monitoring may not provide a suitable alternative.

#### *3.3. Human Fibroblasts*

Fibroblasts from human patients can be generated from minimally invasive samples (1 mm punch skin biopsies) and maintain DNA mutations, although heteroplasmy is not necessarily maintained with passages, and cumulative cellular damage [81]. Respirometry measurements are straightforward in fibroblasts across multiple platforms [82–84]. Table 4 provides a general overview of the advantages and limitations of measuring mitochondrial bioenergetics in intact cells, permeabilized cells, and isolated mitochondria, all of which can be utilized in fibroblasts.

**Table 4.** Advantages and disadvantages of respirometry measurements in intact and permeabilized cells and isolated mitochondria. Adapted from [12]. Abbreviations: ETC: Electron transport chain; OCR: Oxygen consumption rate.


Fibroblast cell respirometry has been reported to be a faster and more sensitive measure of electron transport chain defects than traditional spectrophotometry enzymatic assays [85]. In a study of Leigh and Leigh-Like Syndrome patients, respirometry was able to detect impairment in the mitochondrial respiratory chain in 50% of patients that could not be identified with measuring individual ETC complex activities alone [84]. This study demonstrates that detection of causative mutations, biochemical, and respiratory defects vary between individuals and that using combinations of these approaches has the best diagnostic rates. Additionally, enzymatic impairment is not always consistent between muscle biopsies and fibroblasts, suggesting that testing multiple samples, when possible, is the best practice.

Patient fibroblasts provide an important research tool to measure mitochondrial bioenergetics as well as a full characterization of mitochondria including, but not limited to, mitochondrial morphology, turnover, and in-depth analysis of mitochondrial metabolism [86,87]. For research purposes, commercially available fibroblasts can be obtained for different mitochondrial disorders and diseases of aging. This makes it possible to better understand the full range of mitochondrial impairment caused by certain mutations or in particular diseases. It also opens up the ability for pre-clinical assessment of mitochondrial function in these cells in response to interventions, such as compound treatments. Fibroblasts can present the metabolic signatures observed in mitochondrial disease patients, such as a metabolic shift from oxidative phosphorylation to increased glycolysis for ATP synthesis [88,89]. Therefore, these cells can be a particularly important tool to test compounds that are not expected to improve mitochondrial function in healthy, control cells but only under mitochondrial dysfunction. Compound testing in fibroblasts can facilitate determining compound effects on metabolically stressed cells without needing to use exogenous stressors such as calcium, reactive oxygen species (ROS), or mitochondrial inhibitors that can be difficult to titrate and often have narrow exposure windows between impaired function and apoptosis.

#### Considerations for Utilizing Fibroblasts for Respirometry Studies

Given that fibroblasts proliferate in culture, the amount of sample is not limited to the same extent as with tissue biopsies. However, some considerations arise from measuring metabolism in cultured fibroblasts. In vitro culture conditions, such as the amount of glucose in the culture medium, have been shown to change cellular metabolism across multiple cell lines including fibroblasts [90–94]. Costa et al. demonstrate that culturing fibroblasts in low glucose conditions, a more physiological condition than standard high glucose conditions, results in remodeling of the mitochondrial network and a push towards a more oxidative phenotype without causing an extensive metabolic reconfiguration of substrate preference compared with high glucose cultured cells [95]. On the other hand, growing cells in galactose cause both mitochondrial and metabolic remodeling resulting in increased oxygen consumption rates, ATP levels, and mitochondrial biogenesis. Galactose culturing is a useful approach for screening for drug-induced mitochondrial toxicity as it increases sensitivity to mitochondrial respiratory chain inhibitors [96]. Galactose culturing can also improve the testing of interventions that increase mitochondrial bioenergetic function by preventing compensatory ATP synthesis through glycolysis [97]. There are additional considerations besides media composition, for example, modifying the culturing oxygen concentration has been shown to increase the in vitro lifespan of fibroblasts [98]. Therefore, it is important to consider the conditions under which fibroblasts are grown and how these conditions could affect mitochondrial function and metabolism when investigating mitochondrial phenotypes in vitro.

Furthermore, when utilizing primary fibroblasts it is important to consider population doubling time as these cells have limited proliferative capacity and will become senescent over time [99]. Indeed, human skin fibroblasts display hallmarks of aging senescence and are used as a model of aging in vitro [100–102]. However, changes in fibroblast metabolism with increasing passage numbers have technical implications for the experimental use of

these cells for the investigation of mitochondrial function. Glucose uptake and lactate production increase with fibroblast population doublings, which is consistent with a switch towards glycolysis as cells become senescent [103,104]. Senescent fibroblasts also display changes in mitochondrial bioenergetics with increased mitochondrial membrane potential heterogeneity, a decreased respiratory control ratio, and uncoupling of mitochondrial respiration from ATP synthesis without changes in total respiratory capacity [105]. There is also evidence that mitochondrial content increases in senescent fibroblasts [100]. However, this may represent compensation for increased cell volume that occurs during senescence and not necessarily a change in mitochondrial density [105]. Subculturing of fibroblasts can also induce shifts in mitochondrial heteroplasmy in cells with mtDNA mutations [106]. Taken together, these data highlight the importance of considering passage doubling when using fibroblasts as a model for studying mitochondrial function and comparing across similar passage numbers.

Primary fibroblasts represent the biological and chronological age of the subjects from which they are derived [81,107]. Fibroblasts derived from old subjects display slower proliferation rates and alterations in mitochondrial function and metabolism compared with fibroblasts from young subjects [107–109]. Therefore, a final consideration for using fibroblasts for investigating mitochondrial function is what control should be used for patient cells including considerations of age, sex, and genetics. Age and sex-matched fibroblasts should be used as controls and, whenever possible, those cells should be derived from a relative, such as an unaffected sibling, to better control for genetic differences between the patient and control fibroblasts. Another option is to generate isogenic controls using the corrected disease-causing mutation, for example by clustered regularly interspaced short palindromic repeats (CRISPR)—associated protein 9 (CRISPR-CAS9) correction or overexpressing the wild-type complementary DNA (cDNA), making the comparison more precise by keeping the same genetic background [110].
