Protein expression profiles in patients carrying NFU1 mutations. Contribution to the pathophysiology of the disease
- First Online:
- Cite this article as:
- Ferrer-Cortès, X., Font, A., Bujan, N. et al. J Inherit Metab Dis (2013) 36: 841. doi:10.1007/s10545-012-9565-z
- 497 Views
Cofactor disorders of mitochondrial energy metabolism are a heterogeneous group of diseases with a wide variety of clinical symptoms, particular metabolic profiles and variable enzymatic defects. Mutations in NFU1 were recently identified in patients with fatal encephalopathy displaying a biochemical phenotype consistent with defects in lipoic acid-dependent enzymatic activities and respiratory chain complexes. This discovery highlighted the molecular function of NFU1 as an iron-sulfur(Fe-S) cluster protein necessary for lipoic acid biosynthesis and respiratory chain complexes activities. To understand the pathophysiological mechanisms underlying this disease we have characterized the protein expression profiles of patients carrying NFU1 mutations. Fibroblasts from patients with the p.Gly208Cys mutation showed complete absence of protein-bound lipoic acid and decreased SDHA and SDHB subunits of complex II. In contrast, subunits of other respiratory chain complexes were normal. Protein lipoylation was also decreased in muscle and liver but not in other tissues available (brain, kidney, lung) from NFU1 patients. Although levels of the respiratory chain subunits were unaltered in tissues, BN-PAGE showed an assembly defect for complex II in muscle, consistent with the low enzymatic activity of this complex. This study provides new insights into the molecular bases of NFU1 disease as well as into the regulation of NFU1 protein in human tissues. We demonstrate a ubiquitous expression of NFU1 protein and further suggest that defects in lipoic acid biosynthesis and complex II are the main molecular signature of this disease, particularly in patients carrying the p.Gly208Cys mutation.
Cofactor disorders of mitochondrial energy metabolism constitute a heterogeneous group of diseases with a wide variety of clinical symptoms associated with particular metabolic profiles and variable enzymatic defects of the energy metabolism. Although most of the genes involved in mitochondrial cofactor biosynthesis still remain to be elucidated, an increasing number of genes have recently been identified (Ylikallio and Suomalainen 2012; Navarro-Sastre et al 2011; Cameron et al 2011; Haack et al 2012; Mayr et al 2011a; b; Quinzii and Hirano 2011). The understanding of the pathophysiological and molecular bases of these diseases will help to develop potential therapeutic strategies.
In 2011 we identified mutations in NFU1 (MIM 608100) in ten patients with fatal infantile encephalopathy and/or pulmonary hypertension displaying a biochemical phenotype consistent with a defect in lipoic acid biosynthesis (Navarro-Sastre et al 2011). Lipoic acid is an essential cofactor which is covalently bound to the E2 subunits of the pyruvate dehydrogenase (PDH), α-ketoglutarate dehydrogenase (α -KGDH), branched-chain α -keto acid dehydrogenase (BCKD) complexes and to the H subunit of the glycine cleavage system (GCS) (Hiltunen JK et al 2010). Clinically, patients presented with acute neurological deterioration secondary to vacuolating leukoencephalophaty, pulmonary hypertension and metabolic acidosis. The biochemical phenotype included hyperglycinemia, lactic acidemia, and high urinary excretion of 2-ketoglutaric, 2-ketoadipic and 2-hydroxyadipic acids, among others. These patients also had low PDH activity and low or undetectable GCS activity, low rates of substrate oxidation and deficiency of the mitochondrial respiratory chain complex II (Navarro-Sastre et al 2011).
NFU1 is a conserved protein involved in the pathway of iron sulfur (Fe-S) cluster biogenesis. Fe-S clusters are essential protein cofactors that participate in a wide variety of enzymatic reactions, including electron transfer in the mitochondrial respiratory complexes I to III and sulfur donors in the synthesis of lipoic acid catalyzed by the mitochondrial enzyme lipoic acid synthase (LAS) (Johnson et al 2005; Beinert 2000; Booker SJ et al 2007; Lill et al 2012).
The discovery of human mutations in NFU1 allowed a better understanding of the molecular function of this protein and its relationship to lipoic acid-dependent pathways (Navarro-Sastre et al 2011; Cameron et al 2011). Indeed, patients carrying NFU1 mutations showed a strong reduction of protein-bound lipoic acid, suggesting that NFU1 might be involved in lipoic acid biosynthesis probably delivering Fe-S to LAS in the mitochondria. In addition, NFU1 patients showed heterogeneous patterns of mitochondrial respiratory chain alterations including isolated complex II deficiency (Navarro-Sastre et al 2011) or combined deficiencies of complexes I, II and III (Cameron et al 2011). However, the physiopathology of the disease caused by NFU1 mutations still remains to be completely understood.
The aim of this study was to characterize the expression pattern of a subset of proteins dependent on Fe-S clusters in a series of NFU1 patients to further delineate the molecular bases of the disease and to provide efficient tools for the diagnosis.
Material and methods
The seven patients involved in this study have previously been described and characterized clinically and biochemically (Navarro-Sastre et al 2011). These patients presented with neurological deterioration and pulmonary hypertension or failure to thrive with pulmonary hypertension and metabolic acidotic episodes. Symptoms started at 1–9 months of age and all of them died before the age of 15 months. Their biochemical phenotype was characterized by metabolic acidosis with variable lactic acidemia, hyperglycinemia and high urinary excretion of 2-ketoglutaric, 2-ketoadipic and 2-hydroxyadipic acids, among others. They all had low PDH and GCS activities and low rates of 14C substrate (pyruvate, glutamate and leucine) oxidation. The mitochondrial respiratory chain complexes analysis in frozen muscle showed slightly reduced complex II activity.
Biological samples used in this study
Effect on protein
c.622C > T/c.545 + 5 G > A
c.622C > T/c.622C > T
c.622C > T/c.622C > T
Fibroblast, kidney, brain
c.622C > T/c.622C > T
c.622C > T/c.622C > T
c.622C > T/c.622C > T
c.622C > T/c.622C > T
The controls used in this study were obtained from the tissue bank of our institution and corresponded to biopsies and necropsies from patients with suspicion of metabolic diseases. All of them were of pediatric age and the biochemical analysis showed no defects in the mitochondrial respiratory chain enzymes, PDHc and mtDNA.
Protein expression analysis and blue native gel electrophoresis
Fibroblasts and tissues obtained from patients and control individuals were homogenized in SETH buffer (10 mM Tris–HCl pH 7.4, 0.25 M sucrose, 2 mM EDTA, 5 × 104 U/l heparin). Cleared lysates were subjected to SDS-PAGE, electroblotted and proteins were visualized by immunostaining with specific antibodies followed by colorimetric detection (Opti-4CNTM Substrate Kit,Bio-Rad, U.S). IMAGEJ software was used for densitometry analysis of protein expression levels (Abramoff et al 2004). Anti-NFU1 antibody was obtained by immunizing rabbits with purified NFU1 protein (Navarro-Sastre et al 2011). Antibodies against SDHA, SDHB, NDUFA9, ATP5A and UQCRC2 were from Invitrogen (Paisley, UK). Protein conjugated-lipoic acid antibody (Calbiochem, Darmstadt, Germany), GAPDH (Santa Cruz Biotechnology, Heidelberg, Germany), PDHc (kindly donated by Dr. W. Ruitenbeek, the Netherlands) and UQCRFS1 (Abcam, Cambridge, UK) were also used in this study. The mitochondrial respiratory chain complexes assembly was analyzed by blue-native polyacrylamide gel electrophoresis (BN-PAGE) as described (Wittig I et al 2006).
Protein expression profiles in fibroblasts from NFU1 patients
Lipoylated proteins and NFU1 expression profiles in human control tissues
Lipoylated proteins and respiratory chain expression profiles in tissues from NFU1 patients
Age-related NFU1 protein expression
To study the age-related regulation of NFU1 we have analyzed control tissues from the first week of life until adulthood. Western blot analysis of muscle homogenates showed high levels of NFU1 protein until 3 months after birth and decreasing amounts along the first years of life until adult age (Fig. 3c).
The aim of this study was to provide a better understanding of the molecular and pathophysiological mechanisms involved in the fatal infantile encephalopathy caused by NFU1 mutations. As previously reported, these patients presented with early onset neurological deterioration and a biochemical phenotype compatible with defects in lipoic acid dependent enzymatic activities. NFU1 gene encodes for a conserved protein suggested to participate in Fe-S cluster biogenesis (Navarro-Sastre et al 2011; Cameron et al 2011). However, the exact role of NFU1 had not been completely understood until NFU1 disease-causing mutations were discovered to be associated with that particular phenotype (Navarro-Sastre et al 2011; Cameron et al 2011). These findings and the subsequent functional analyses pointed to the role of NFU1 as a late acting factor involved in Fe-S cluster delivery and maturation of a subset of mitochondrial proteins, including several subunits of the respiratory chain complexes and LAS (Navarro-Sastre et al 2011; Cameron et al 2011; Lill et al 2012; Rouault 2012).
Until now only two studies identified patients with mutations in this gene (Navarro-Sastre et al 2011; Cameron et al 2011). One of these studies was performed by our group and reported on nine patients carrying a homozygous NFU1 missense mutation (c.622 G > T, p.Gly208Cys) and one additional patient who was compound heterozygous for this and an additional splice site mutation (c.545 + 5 G > A) that generates an unstable transcript with a premature termination codon (Navarro-Sastre et al 2011). A parallel study by another group (Cameron et al 2011) identified a homozygous NFU1 null mutation (c.545 G > A) generating an aberrant splicing in three siblings. Phenotypically the disease was similar to that of our patients, but with a more severe presentation in the neonatal period, that led to death before 1 month of life (Seyda et al 2001; Cameron et al 2011).
To better understand the molecular bases of this disease we have analyzed the levels of lipoylated E2 subunits of PDH and KGHD complexes in fibroblasts and tissues from homozygous and compound heterozygous patients with NFU1 mutations. Because respiratory chain complexes I, II and III have components that contain Fe-S, we have also studied the levels of several subunits from these multienzymatic complexes including the iron-sulfur proteins SDHB and UQCRFS1 from complex II and III, respectively.
In agreement with previous observations in fibroblasts carrying NFU1 null mutations (Cameron et al 2011), our results showed almost complete absence of lipoic acid bound to E2 subunits of PDH and KGDH complexes (Fig. 1). However, when analyzing the respiratory chain subunits, both results differ because our patients with missense mutations (showing normal levels of NFU1) have decreased amounts of both subunits (SDHA and SDHB) of complex II (Fig. 1), while patients with null mutations (showing absence of NFU1) had decreased levels not only of the subunits of complex II but also of NDUFA9 (complex I) and UQCRFS1 (complex III) (Cameron et al 2011). This suggests a possible role of NFU1 both in the delivery of Fe-S clusters and also in the stability of some complexes of the respiratory chain. These observations fit well with the BN-PAGE results in muscle of one of our patients carrying NFU1 missense mutations (Fig. 4c) that showed defective assembly of complex II, but not of complexes I and III.
To our knowledge NFU1 protein expression has not been accurately determined in humans and its regulation remains to be elucidated. Only two studies documented the expression profile of NFU1 mRNA by northern blot in human and mouse tissues (Ganesh et al 2003; Lorain et al 2001). To further understand the regulation of NFU1 in human tissues as well as the requirement of NFU1 for protein lipoylation in physiological conditions we analyzed the expression levels of NFU1 and protein bound lipoic acid in a series of human control tissues (Fig. 2). Our results showed that the expression of NFU1 and of lipoylated proteins correlated with the levels of the UQCRC2 and citrate synthase activity (Fig. 2). These results complement previous observations made in human and mouse tissues by providing further evidence of the ubiquitous expression of NFU1 and they showed that the differences in several tissues are mainly due to the particular mitochondrial content of each tissue rather than a tissue-specific regulation of NFU1 expression.
With the aim of further extending the knowledge of the mitochondrial dysfunction caused by NFU1 disruption we performed similar studies in a series of available tissues from patients. Results showed that lipoylated E2 subunits of PDH and KGDH complexes were clearly reduced in patients’ liver and muscle (Fig. 3a) but surprisingly were unchanged in brain, kidney and lung (Supplementary Fig. 1). This suggests that proteins other than NFU1 may be involved in Fe-S cluster delivery to LAS and to other iron sulfur proteins in these tissues where they compensate for the NFU1 deficiency in a tissue-specific manner (Fig. 3a and Supplementary Fig. 1). To this effect, it has recently been reported that other proteins involved in Fe-S cluster biogenesis such as NDUFAB1, IBA57, ISCA1 and ISCA2 are necessary for lipoic acid biosynthesis since cell lines depleted of these proteins showed a strong reduction of protein-bound lipoic acid (Mühlenhoff et al 2011; Gelling et al 2008; Song et al 2009; Sheftel et al 2012; Feng et al 2009).
In contrast to fibroblasts none of the subunits of the mitochondrial respiratory chain, including SDHA and SDHB subunits, showed altered expression in the pathological tissues analyzed (Fig. 3a and Supplementary Fig. 1). However, the respiratory chain activities in frozen muscle from two of these patients had consistently reduced complex II but normal complex I, III and IV activities (Navarro-Sastre et al 2011). Accordingly, BN-PAGE performed in muscle from one of them demonstrated decreased amounts of fully assembled complex II, but normal complex I and III (Fig. 3b). These observations, together with the fact that the NFU1 p.Gly208Cys mutation impairs the transfer of Fe-S cluster to target apoproteins, support the proposed role of NFU1 in the delivery of Fe-S cluster to SDHB subunit of complex II and provide an explanation why complex II assembly is affected in muscle tissue even when normal levels of SDHB were detected (Navarro-Sastre et al 2011).
However, it is still intriguing that clinically affected tissues, such as brain or lung, showed no alterations in protein lipoylation and should be carefully studied in the future. The neurological regression affecting white matter of these patients (Navarro-Sastre et al 2011) could not be explained with our data, probably due to the fact that the samples used in this study could not be carefully dissected in order to discriminate the different parts of the brain. Therefore, further experimental work is needed to completely understand this phenomenon.
Another issue not yet clarified is why patients carrying missense mutations in NFU1 are able to bypass the requirement for full NFU1 function during the first months of life, as they do not show any symptoms until at least 1 month of age (Navarro-Sastre et al 2011), while patients carrying NFU1 null mutations presented symptoms during the first days of live and died before 1 month after birth (Seyda et al 2001; Cameron et al 2011). To address this question we studied age-related NFU1 expression in muscle tissue from control individuals. Interestingly, NFU1 protein levels seem to be higher in the first months of life suggesting that NFU1 protein functions might be specially required during the early neonatal period (Fig. 5). These observations fit well with a previous study that showed higher NFU1 mRNA levels in fetal tissues in comparison to their corresponding adult counterparts (Lorain et al 2001). Our results provide a possible explanation why NFU1 null mutant patients developed symptoms immediately after birth while patients with less severe NFU1 missense mutations presented later in infancy (Navarro-Sastre et al 2011). As we discussed above it is also possible that NFU1 plays other roles in addition to Fe-S cluster delivery to target apoproteins, since patients carrying NFU1 null mutations showed defects in subunits of the respiratory chain complexes I, II and III whereas missense mutations affect specifically complex II.
In summary, this study provides new insights into the molecular bases of NFU1 disease as well as into the regulation of NFU1 protein in human tissues. Our results demonstrated a ubiquitous expression of human NFU1 protein and further suggest that defects in lipoic acid biosynthesis and complex II are the main molecular signatures of this disease, particularly in patients carrying the p.Gly208Cys mutation. Finally, our observations suggest the analysis of protein lipoylation in fibroblasts as an approach to direct the diagnosis when a combined deficiency of PDHc and complex II or complex I-III is found.
We are grateful to the families involved in this study.
Conflict of interest
This research was supported in part by Centro de Investigación Biomédica en Red de Enfermedades Raras (CIBERER), an initiative of the Instituto de Salud Carlos III (Ministerio de Ciencia e Innovación, Spain) and the grants FIS PI12/01138, FIS PI08/90348 and FIS PI08/0307. The authors confirm independence from the sponsors. The content of this article has not been influenced by the sponsors.