Mitochondrial organization and structure are compromised in fibroblasts from patients with Huntington’s disease

ABSTRACT

Huntington´s disease (HD) is a progressive neurodegenerative disease with onset in adulthood that leads to a complete disability and death in approximately 20 years after onset of symptoms. HD is caused by an expansion of a CAG triplet in the gene for huntingtin. Although the disease causes most damage to striatal neurons, other parts of the nervous system and many peripheral tissues are also markedly affected. Besides huntingtin malfunction, mitochondrial impairment has been previously described as an important player in HD. This study focuses on mitochondrial structure and function in cultivated skin fibroblasts from 10 HD patients to demonstrate mitochondrial impairment in extra-neuronal tissue. Mitochondrial structure, mitochondrial fission, and cristae organization were significantly disrupted and signs of elevated apoptosis were found. In accordance with structural changes, we also found indicators of functional alteration of mitochondria. Mitochondrial disturbances presented in fibroblasts from HD patients confirm that the energy metabolism damage in HD is not localized only to the central nervous system, but also may play role in the pathogenesis of HD in peripheral tissues. Skin fibroblasts can thus serve as a suitable cellular model to make insight into HD pathobiochemical processes and for the identification of possible targets for new therapies.

KEYWORDS:

• Huntington’s disease
• fibroblasts
• oxidative phosphorylation system
• mitochondrial dysfunction
• mitochondrial network
• ultrastructure

Introduction

Mitochondria are essential eukaryotic organelles composed of an outer membrane, inner membrane arranged into cristae, and an internal matrix. More than 90% of cellular ATP is synthesized¹ in the oxidative phosphorylation system (OXPHOS) localized in the inner mitochondrial membrane.² ATP synthesis along with many other functions like purine synthesis, amino acid metabolism, fatty acid oxidation, or urea cycle running in mitochondria make them central metabolic organelles in eukaryotic cells.

Mitochondria create dynamic networks and their arrangement reflects the metabolic state of the cell and any disturbances can indicate pathological changes in mitochondrial functions.³ Similarly, ultrastructural changes (low cristae number, mitochondrial swelling) also indicate mitochondrial disruption.⁴ Besides primary mitochondrial disorders, caused by mutations in mitochondrial genes coded by mitochondrial DNA (mtDNA)⁵ or nuclear DNA (nDNA),⁶ there are disorders caused by mutations in non-mitochondrial genes, secondarily affecting mitochondrial metabolism. Such mitochondrial dysfunctions may act as important factors in the development of a disease pathology⁷ like Huntington’s chorea or Alzheimer’s disease.⁸

The autosomal dominant Huntington’s disease (HD) is caused by an expansion of a CAG triplet (>36 repeats) in the first exon of the gene for the huntingtin protein (Htt).⁹ The basic pathology mechanism is the misfolding of Htt and a consequent loss of its function caused by an extended N-terminal polyglutamine sequence.¹⁰

Htt is expressed in all tissues; nevertheless, the highest levels were found in the brain and testes.¹¹ Although the function of Htt is still not fully understood, it has been shown to participate in synaptic transduction and neuronal stability, vesicular and organelle trafficking,¹² regulation of gene expression,¹³ mitochondrial calcium handling¹⁴ or anti-apoptotic defense.¹⁵ Moreover, in the mouse model, a confirmed Htt knockout leads to early embryonic lethality.¹⁶

Symptoms of HD include progressive motor dysfunctions (chorea, dyskinesia, and dystonia) and psychiatric disturbances (cognitive decline, changes in personality, mood, and aggression).¹⁷

Despite the intensive research of HD, there is not yet a known cure. Recently, the new gene therapy for the disease is tested in experimental animals¹⁸ and at present under investigation in humans.¹⁹ Several independent clinical trials for HD gene therapy are ongoing in Phase III.²⁰ Unfortunately, some of them had to be stopped due to the negative impact of the normal htt protein.²¹ All therapeutic approaches will require time-consuming testing, where all possible tools will be needed, such as human cultured cells of various origins, including skin fibroblasts.

Htt malfunction affects a broad spectrum of cellular pathways with the most pathologic impact on the central nervous system, especially in the striatum.²² HD pathology is detectable also in peripheral tissues, for example, in the testes, blood, skeletal muscle, or fibroblasts.^23,24

The connection between HD and mitochondrial impairment was demonstrated on a decrease of mitochondrial energy metabolism,^25,26 higher level of reactive oxygen species,²⁷ calcium imbalance,²⁸ and a malfunction of the pyruvate dehydrogenase complex (PDHc).²⁹ Moreover, these mitochondrial changes in HD are detectable not only in the brain, but also in many other non-neural tissues like muscle,³⁰ heart,³¹ testes³² in TgHD models, or in muscle,³³ heart,³⁴ and peripheral blood³⁵ in HD patients, respectively.

Publications focusing on the mitochondrial status in HD human peripheral tissue, such as muscle,³⁶ blood (lymphocytes),³⁵ or fibroblasts^37,38 are scarce since these tissues from HD patients are not readily available. We studied mitochondrial impairment in cultivated skin fibroblasts from a group of 10 symptomatic heterozygous HD patients with the classical form of the disease, comparing them to healthy age-matched controls.

We observed decreased cristae number, lower mitochondrial density and a decreased amount of selected mitochondrial proteins (Optic atrophy protein 1 (Opa1), PDH 1-α subunit and Complex I NDUFA9 subunit) in patients’ fibroblasts. In concordance with the ultrastructural changes, we also detected functional impairment of the mitochondria.

Material and methods

Patient group

Ten lines of cultivated skin fibroblasts from symptomatic patients (P1 – P10) with confirmed HD were included in the experimental group (8 men, 2 women – P4, P6). All patients were heterozygotes with the classical form of the disease. CAG repeat number at the mutated allele varied between 42 and 62. Clinical symptoms, age of onset and sex are summarized in Table 1.

Table 1. Main characteristics of HD patients analyzed in experimental group.

Patient	Sex	CAG number	Age at disease onset	Age at skin biopsy	Clinical manifestation
P1	male	19/48	35	40	Generalized chorea, irritability, apathy
P2	male	17/45	30	35	Cognitive dysfunction, subdepression, irritability
P3	male	23/44	27	31	Mild chorea, mild apathy
P4	female	20/53	27	33	Generalized chorea
P5	male	24/42	28	42	Generalized chorea, dysphagia
P6	female	29/44	50	62	Preterminal stage, death at 3/2013
P7	male	17/48	22	33	Generalized chorea with mild facial dyskinesia
P8	male	16/42	49	50	Acral and perioral chorea
P9	male	17/50	25	38	Generalized chorea, dystonia, cognitive deficit
P10	male	28/49	30	43	Dystonia, very hard bradykinesia dysphagia, frequent infectious complications

The control group consists of 15 fibroblast lines obtained from healthy volunteers (9 men, 6 women aged between 21 and 50 years).

All work involving human samples was carried out in accordance with the Declaration of Helsinki of 1975 and was approved by the Ethical Committee of the General University Hospital in Prague, approval No. 115/15 Grant COST MSMT-COST 1. LF UK 29.9.2015 and No. 14/19 Grant AZV VES 2020 VFN 23.5.2019. The written informed consent was obtained from patients as well as control subjects prior to inclusion in the study.

Cultured fibroblasts

Cultivated skin fibroblasts were established from skin biopsy. Fibroblasts of all patients and controls were cultivated in Dulbecco’s modified Eagle’s medium (DMEM, Sigma Aldrich – Merck, Darmstadt, Germany) with 10% fetal bovine serum (FBS, Thermo Scientific, Waltham, MA) at 37°C under 5% CO₂ atmosphere. Fibroblasts for enzymatic measurements were grown to approximately 80% confluence, harvested by incubation (10 min, 37°C) with TE (Trypsin, 0.05% (w/V); EDTA (0.02%, w/V)), washed and suspended in PBS (1.5 mM KH₂PO_4; 8 mM Na₂HPO₄ · 12H₂O; 5.4 mM KCl; pH 7,2) and stored at −80°C until use.

Structural analyses

Transmission electron microscopy (TEM)

Fibroblasts were cultivated as mentioned previously to 90% confluence and fixed in phosphate-buffered saline (PBS) containing 2% (w/V) potassium permanganate for 15 min at room temperature (RT). After fixation, cells were dehydrated in an ethanol series (for 10 min in 50%, 70%, and 90% ethanol, for 30 min in 100% ethanol) at RT. Dehydrated cells were incubated in propylenoxide (2x15 min) and embedded in Durcupan Epon (Electron Microscopy Sciences, Hatfield, PA) (Durcupan Epon:Propyleonxide 1:1 for 2 hours, Durcupan Epon:Propylenoxide 1:3 overnight at 60°C). Polymerized blocks were sectioned by Ultracut III ultramicrotome (Reichert, Depew, NY) and 600–900 Å thick slices were stained with lead citrate and uranyl acetate.⁴ Pictures were taken on a transmission electron microscope JEOL JEM-1200 EX.

The electron microscopic images were analyzed semi-quantitatively and results were organized into five different categories of pathologies observed. The data for analysis were collected from approximately 100 cells on one grid for each sample (the cell represents a compact unit with a nucleus). Studied categories were: Cristae disruption,³⁹ Low cristae number,⁴⁰ Mitochondrial swelling,⁴¹ Degradation bodies,²³ Golgi complex structural abnormalities.⁴²

Fluorescence microscopy

Mitochondrial network staining: Fibroblasts from seven patients (P1-P7) and four control lines were cultivated on cover glasses in the DMEM medium and 10% FBS as described above. Mitochondria were stained using vital probe 10 nM MitoTracker Red CMX Ros (Molecular Probes, Eugene, OR) for 15 min, at 37°C. Cells were washed with PBS, mitochondrial network was detected by Nikon Diaphot 200 inverted microscope (Nikon, Tokyo, Japan) and pictures were taken with an Olympus DP50 CCD camera and Viewfinder Lite 1.0 software (Pixera, Santa Clara, CA). Mathematical analysis was carried out in the Fiji program (http://fiji.sc/Fiji). Two factors – mitochondrial density (MD) and the branching of mitochondria (branching factor, BF) were tested. MD characterizes the density of mitochondria per cells, BF compares the level of branching between mitochondrial networks in patients’ and control cells. The analysis was performed as published previously.⁴³ Obtained data were statistically assessed using a T-test.

Protein analysis (SDS-PAGE, Western blot, immunodetection)

Tricine SDS-PAGE was carried out under standard conditions with 12% polyacrylamide, 0.1% (w/v) SDS gels. Details are described in supplementary materials.

Functional analyses

Measurement of oxygen consumption in fibroblasts

The measurement was performed following a modified protocol of Kuznetsov et al.,⁴⁴ for details, see supplementary material. Results were normalized on protein concentration measured by the Lowry protocol.⁴⁵

Measurement of mitochondrial energy generating system capacity (MEGS)

The analysis was performed according to the Janssen et al. protocol⁴⁶ with [1⁻¹⁴C]pyruvate, [U-¹⁴C]malate and [1⁻¹⁴C]succinate as substrates. The composition of individual incubations is summarized in Table S1 (Supplementary material). Freshly harvested cells (cca 50 mg of wet pellet) were resuspended in 200 ul PBS and permeabilized by digitonin (0.3 µg/1 mg wet pellet, 20 min, 4°C). For one MEGS analysis, 30–50 ug of permeabilized cell suspension was used. The produced ¹⁴CO₂ was collected on a filter paper soaked with NaOH and measured in a liquid scintillation counter.

Measurement of enzymatic activity of selected mitochondrial enzymes

Basal and activated enzymatic activity of the PDHc with 5 mM DCA was assayed as the release of ¹⁴CO₂ from [1⁻¹⁴C]pyruvate.⁴⁷ The activities of NADH:coenzyme Q10 reductase (NQR, complex I), succinate:coenzyme Q10 reductase (SQR, complex II) and cytochrome c oxidase (COX, complex IV) and citrate synthase (CS) were measured spectrophotometrically according to Rustin et al.⁴⁸ About 80–100 µg of protein was used per one enzymatic assay. Total protein amount was determined by the method of Lowry⁴⁵.

Statistics

MEGS and enzyme activity data were statistically analyzed by a linear model (Software R, version 3.2.3, R Core Team (2015). Three parameters were analyzed: HD, gender and passage of fibroblasts cultivation. For the quantification of mitochondrial network parameters student T-test was used.

Results

Structural analyses

Mitochondrial ultrastructure

Samples from all 10 patients and 5 control lines were analyzed by TEM to study the ultrastructure of cells. There were five cellular and/or mitochondrial abnormalities evaluated in the patient's cell lines compared to the controls (arranged in the representative Figure 1(b-f), controls in Figure 1a). The three main pathological phenotypes across all patient's samples were: Cristae alteration (10/10), Low cristae number (10/10), Mitochondrial swelling (10/10) (Table 2). Furthermore, a higher number of degradation bodies (autophagic vacuoles) and lysosomes with accumulated material were detected in P2, P3, P4, and P7 cell lines, and abnormalities of Golgi complex architecture were observed in P2, P3, P5, P7, and P8. In the latter samples, the typical Golgi stacks of cisternae were disorganized into a higher number of vesicles (Golgi bodies) (Figure 1f). This semi-quantitative analysis was performed by comparing approximately 100 cells from each sample (1 grid); the presence and frequency of pathological markers in patient's fibroblasts are summarized in Table 2.

Figure 1. Ultrastructural alterations in fibroblasts of HD patients in comparison to controls. Pathological changes in ultrastructure of mitochondria and Golgi complex in fibroblasts of patients with HD visualized by transmission electron microscopy (TEM). Pictures were taken on a transmission electron microscope JEOL JEM-1200 EX. a) Normal mitochondria in adult control fibroblasts, b) Cristae disruption (P7), c) Low cristae number, marked by # (P3), d) Mitochondrial swelling, marked by * (P8), e) Degradation bodies, marked by < (P2), f) Structural abnormalities of Golgi complex (P2). Original magnification: 15 000 x (A-E), 25 000 x (f).

Table 2. Semi-quantitative analysis of ultrastructural alterations in fibroblasts of 10 HD patients. Columns present most common disturbances detected in patient’s fibroblasts. The number of crosses (+) defines the level of impairment in category, (-) the phenotype was not detected in the sample. Data were obtained in approximately 100 cells on one grid, where a compact unit with nucleus was considered as a cell.

Patients	Ultrastructural changes
	Mitochondria				Other organelles
	Cristae disruption	Low cristae number	Mitochondrial swelling	Increased degradation bodies	Structural abnormalities of Golgi complex
P1	+	+ +	+	-	-
P2	+	+	+	+	+
P3	+ +	+	+	+	+
P4	+ +	+	+	+ +	-
P5	+ +	+ +	+ +	-	±
P6	+	+	+	-	-
P7	+ +	+	+	+ + +	+ +
P8	+ + +	+ + +	+ +	-	+
P9	+ +	+	+	-	-
P10	+	+	+	-	-

Mitochondrial network

Native dying of the mitochondrial network in fibroblasts from P1-P7 and four controls showed unequally distributed mitochondria in patient's samples. A mild network dissipation with isolated mitochondrial spots (in Figure 2a, P2, P6, P7) was present. Mitochondrial density per cell (MD) analyzed in the Fiji program in patient's and control samples showed a significant decrease of MD in patient's cells (p = 0.0012). The branching factor (BF) was mildly decreased in patient's samples (Figure 2b). a)

Figure 2. Mitochondrial network visualization and quantitative analysis of mitochondrial density (MD) and branching factor (BF) in fibroblasts of 7 HD patients (P1 – P7) and 4 controls. a) Native dying of mitochondrial network in fibroblasts of HD patients P1-P7 and 4 controls: Cultivated skin fibroblasts were visualized in 10 nM MTR (MitoTracker RED, Molecular probes, Eugene, OR) 15 min at 37°C. Unequal distribution of mitochondrial network and mild fragmentation were detected in all 7 patients in comparison to controls, although the level of disruption was different in different lines. Original magnification: 600x. b) Mitochondrial density (MD) and Branching factor (BF) analysis in mitochondrial network in fibroblasts of 7 HD patients (P1 – P7) and 4 controls. quantitative analysis was provided in Fiji program. T-test was used for statistical analysis. MD and BF were significantly decreased in patient´s cells compared to controls (p = 0.0051) (p = 0.014),respectively.b)

Figure 2. (Continued).

Protein analysis of selected mitochondrial proteins

A spectrum of selected subunits from OXPHOS complexes (Complex I: NDUFA9, NDUFB6, Complex II: SDH70, SDH30, Complex III: Core2, Complex IV: Cox1, Cox2, CV: F1- α), PDH complex (E2, 1-α) were immunodetected in whole-cell lysates after SDS PAGE/WB (supplementary Figure S1). Evaluation of the results is summarized in Table 3. Patient's samples showed remarkably decreased levels of the NDUFA9, PDH 1- α subunit and the Opa 1 protein in comparison to controls. The complex I subunit NDUFA9 is reduced in P2-P8 and P10 (26–78% of control values). P1 and P9 showed normal levels of NDUFA9 subunit. The PDH complex protein analysis showed the decrease of PDH E1-α subunit in P1-P8 fibroblasts (34–72% of control values). P9 level was normal and P10 was not analyzed for this protein.

Table 3. Quantification of selected mitochondrial proteins expression in fibroblasts of HD patients and controls. SDS PAGE/WB signal (see Supplementary figure 1) in whole-cell lysates was quantified in G:Box (SynGene imaging system, SynGene UK). Values are in percentage, the average of signal of two adult controls is expressed as 100% of control value. Patient’s values ≤ 80% of controls are red, 120% ≥ in green. Results were normalized to total protein loading and α-tubulin protein. nd – not analyzed.

Three proteins important for the modulation of mitochondrial structure and the fusion/fission process were also analyzed: mitofilin, opa1 and drp1. Mitofilin showed normal levels except for P2, P9, and P10, where the levels decreased to as low as to 23.5% of control values (Table 3). Opa1 levels varied between 66.5% and 79.5% of the control value in 8 cases (P1-P3, P5-P7, P9, and P10). Drp1 protein levels were increased in P3, P4, P5, P7, and P8. Although the expression of Htt in fibroblasts was demonstrated earlier,⁴⁹ we verified this using an anti-Htt antibody (Mab-2166, Milipore), confirming the presence of the Htt protein in two of our control lines (data not shown).

Functional analyses

Measurement of oxygen consumption

Mitochondrial respiration of digitonin-permeabilized fibroblasts was measured in six patients (P1, P3, P4, P5, P6, and P10) and three adult controls. Results were normalized to protein concentration. A mild decrease of respiration after the addition of substrates for complex I (glutamate + malate) and complex II (succinate) and complex IV were detected in fibroblasts from P3, P4, and P5 compared to controls. The oxygen consumption parameters in various respiratory conditions are summarized in Figure 3a.

Figure 3. Alteration of mitochondrial functional parameters in fibroblasts from HD patients. a) Measurement of oxygen consumption. The ratios of average values of oxygen consumption in fibroblasts showed lower values in most of tested patient´s lines in comparison to controls Mitochondrial energy generating system capacity, oxidation of substrate in reaction containing [U-¹⁴C]malate+acetylcarnitine+malonate+ADP (b) and [1–4¹⁴C]succinate+acetylcarnitine+ADP (c) were decreased in HD cells; d) The activity of succinate – coenzyme Q reductase (SQR, complex II of respiratory chain) measured by spectrophotometry, showed a significant interaction of sex, HD and age (p = 0.0078), comparison related to HD and sex are shown. F- female, M- male. Ee) The activity of succinate – coenzyme Q reductase (SQR, complex II of respiratory chain) normalized to activity of citrate synthase (SQR/CS) was associated with sex (p = 0.0213) having a significant interaction of sex, HD and age (p = 0.0188). Comparison related to HD and sex are shown. Details of the effect of age are given in supplement, see Figure S4 and S5.

Mitochondrial energy generating system capacity (MEGS)

MEGS analysis in cultivated skin fibroblasts from 7 HD patients (P1, P2, P4, P6, P7, P9, and P10) revealed a significantly decreased oxidation rate in an incubation containing [U-14C] malate + acetylcarnitine + malonate + ADP (Figure 3b). Moreover, we have verified that this parameter in the controls and patient lines does not differ between different passages (Supplementary Figure S2).

Measurement of OXPHOS and PDH complexes activity

Values of SQR (Complex II) activity and its normalized activity expressed as the ratio of SQR/CS measurement were lower in HD fibroblast samples in comparison to controls. Moreover, SQR showed a significant interaction of sex, HD, and age (p = 0.0078), (Figure 3d). Normalized activity SQR/CS was associated with sex (p = 0.0213), and had a significant interaction of sex, HD, and age (p =0.0188 (Figure 3b)). No correlation between enzyme activity and HD for complex NQR, COX and CS, and ratios COX/CS and NQR/CS was found. Correlation with age and sex was found for COX (supplementary Figure S3).

In contrast to lowered PDH E1-α subunit, measurement of basal and DCA activated PDHc activity did not show significant differences in patients compared to controls (data not shown).

Discussion

Our study presents a comprehensive analysis of cultured skin fibroblasts from 10 HD patients and 15 healthy controls to map how a mutated Htt influences mitochondrial structure and function in a non-neural tissue with a low energy demand.

Skin biopsy of our HD patients was done at different times after disease onset, between 1 year (P8) and 14 years (P5) after the first recorded symptom. Specific mitochondrial dysfunctions or structural changes were well detectable in all patient lines.

Changes in mitochondrial ultrastructure or network organization reflecting functional disturbances of these organelles are usually observed in primary mitochondrial diseases.³ Dissipation of mitochondrial network in neurons was already described in HD,⁵⁰ Alzheimer’s disease⁵¹ and mentioned in other neurodegenerative diseases, like Charcot-Marie-Tooth type 2A,⁵² a peripheral neuropathy or dominant optic atrophy.⁵³ Squitieri’s work on mitochondrial ultrastructural changes in myoblasts and fibroblasts from 4 HD patients (2 homozygous, 2 heterozygous) described similar results and concluded that structural abnormalities are dose-dependent, because changes in the homozygous samples were more profound than in the heterozygous ones.²³

Consistent with these facts, we detected structural discrepancies, like mitochondrial swelling, decreased cristae number or cristae alteration by TEM in all HD patients’ fibroblasts but not in control lines. Patients’ mitochondria were distributed unequally through cells and their density (the area of mitochondria per cell) was lower, which was in accordance with previously published data from striatal cells from HD knock-in mice HdhQ111.⁵⁴ Some patients (P2, P5 in Fig.2a) showed a slightly dissipated mitochondrial network, usually associated with pro-apoptotic signals in cells.⁷ Additionally, TEM analysis showed that lines from P2-4 and P7 contained an increased level of degradation bodies, where cellular waste material is stored before its degradation. This fact, together with a mildly elevated drp1 (dynamin related protein 1; Table 3), responsible for mitochondrial fission, and a decreased level of opa1 (Table 3), responsible for mitochondrial fusion, suggests that HD patients’ mitochondria in fibroblasts increase pro-apoptotic stimuli. Both drp1 and opa1 are proteins maintaining the integrity of the mitochondrial membrane.⁵⁵ Increased drp1 causes mitochondrial fragmentation, which produces ultrastructural changes in mitochondrial cristae promoting, cytochrome c release, and increasing sensitivity to apoptosis.^39,56 Mitochondrial cristae are dynamic bioenergetic compartments whose shape determines the assembly and stability of respiratory chain complexes and therefore mitochondrial respiratory efficiency.⁵⁷ Thus, perturbations of the cristae integrity as a consequence of an imbalanced drp1, found in our cohort, could be responsible for the observed alterations in oxidative phosphorylation function. A similar situation was already described in TgHD minipig muscle, where the drp1 protein increased with development of the disease with parallely decreased respiratory chain complexes II and IV and a decreased oxidation rate of different substrates.³⁰ Aside from the pro-apoptotic signs, TEM revealed mitochondrial swelling and a decreased number of mitochondrial cristae. Our findings are in accordance with a previous report of mitochondrial swelling and cristae disruptions in primary striatal YAC128 mouse neurons and human HD neuronal cells.³⁹

Clearly, mitochondria do not function in isolation, but interact with other cell compartments, such as Golgi or endoplasmic reticulum through specialized membrane domains, signaling molecules (Ca²⁺) or proteins (drp1).⁵⁸ Mitochondrial disruption may thus affect other cellular compartments and the Golgi complex is no exception. A recent study, for example, described a reduced ATP production (induced by 2-deoxy-D-glucose treating) leading to reversible remodeling of Golgi architecture from compact cisternae to vesicular Golgi bodies in HepG2 cells.⁴²

Similarly to the mitochondrial network, structural changes of the Golgi apparatus during physiological/pathological processes like cell cycle,⁵⁹ neurodegeneration⁶⁰ or cellular ATP production⁴² were described. Research in the last decade has shown that the Golgi complex is responsible not only for the protein secretory pathway, but is also a highly dynamic organelle influenced by various cellular stimuli and also playing a role in HD pathogenesis. Dysfunctional mHtt disrupts vesicular transport from Golgi along microtubules and prevents the fusion of vesicles with the plasma membrane.⁶¹ Lower ATP production or increased cellular stress lead to modulation of steroidogenesis or membrane lipid homeostasis via the Golgi protein ACBD3.⁶² In HD, increased levels of the ACBD3 protein raise the levels of the ACBD3/mHtt/Rhes complex (Rhes, Ras homolog enriched in striatum, is a small GTPase), causing a higher cytotoxicity in HD via Rhes-induced sumoylation of mHtt.⁶³ We suggest that the observed changes in Golgi architecture in fibroblasts complement the ultrastructural and functional alteration of mitochondria caused by HD pathogenesis.

We analyzed a broad spectrum of mitochondrial proteins in fibroblasts by SDS PAGE/WB. The complex I subunit NDUFA9 (8/10) and the PDH complex subunit PDH 1-α (8/10) were decreased in most patients. Decreased NDUFA9, a subunit of complex I, together with decreased oxidation rate in the reaction containing malate as substrate (Figure 3b) would suggest a possible impairment in the first step of the oxidative phosphorylation system in HD fibroblasts.

Similarly, a deficiency of complex I was described in the brain of 3-nitropropionic acid-induced rat model of Huntington’s disease,⁶⁴ HD patients,⁶⁵ as well as in the TgHD minipig and the skeletal muscle of R6/2 mice.^30,66

Surprisingly, the decreased PDH 1-α subunit levels were not associated with a reduction in enzymatic activity of the PDH complex (Data not shown). We can thus assume that the threshold for PDH protein expression has not been exceeded and that PDH in the fibroblasts has a sufficient reserve in enzymatic capacity, able to cover the cell’s demands even with a reduced content of some subunits.

Decreased expression of the SDH30 complex II subunit was observed in 5/10 HD patients, together with a decreased oxidation rate in the incubation containing succinate (Figure 3c) compared to controls. Complex II of the respiratory chain has been shown affected several times in various tissues of HD patients – in the striatum,^67–69 peripheral lymphocytes,³⁵ and skeletal muscle,³³ as well as in HD models – in the mouse brain^70,71 and skeletal muscle.³⁰ However, our study has shown that the enzymatic activity of Complex II in fibroblasts significantly depends on sex and age (Figure 3d), so similar studies would need to be performed on a larger group of patients to support the significance of our findings. Functional insufficiency of mitochondria also causes lower respiration after the addition of substrates for complexes I, II, and IV of OXPHOS in most patients compared to control lines (Figure 3a).

The number of patients and some technique used represent the limitations of the study. 3D images from confocal microscopy could provide considerably more precise information about the mitochondrial network than only 2D section analysis. The classical techniques we used could perhaps benefit from a contribution from other multi-omics data or additional clinical patient data. A model study in human fibroblasts has been already used in patients with Parkinson’s disease to identify complex disease signatures.⁷² The diverse intensity of mitochondrial ultrastructural and functional defects in HD fibroblasts analyzed in this work can also be partially attributed to different genetic backgrounds of individual lines. Using a larger set of lines could reveal individual variation suggesting the presence of morphological features inherent to specific individuals. Such insights may uncover individual or patient-specific cellular phenotypes, with valuable implications for personalized medicine.

Conclusion

In conclusion, we have shown ultrastructural impairment and functional insufficiency of mitochondria in human cultured HD fibroblasts. Results of our study are valuable for the investigation of the particular pathobiochemical processes in Huntington’s disease in which mitochondria play a significant role, such as affected autophagy/mitophagy.^73,74 Our results may also serve as the pilot for an evaluation of possible biomarkers that may either be helpful to monitor the course of the disease or serve as a target for the new therapies.

Authors contributions

Acquisition, analysis, and interpretation of data MV, HS, MK, JK, TD; drafting the manuscript MV, HH; drafting figures MV, HS, others – recruitment of patients, collection of biological material JK, IR, JR; revising the manuscript JZ, HH

Acknowledgments

The authors are grateful to Dr. Vaclav Capek for the statistical analysis of the data.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/01913123.2022.2100951

Additional information

Funding

This study was supported by research grants from the Ministry of education, youth and sports (COST LD15099), the Ministry of Health of the Czech Republic (MZ CR AZV-NU20-04-00136 and RVO VFN 64165/2012), and institutional support of Charles University (SVV260516)