We describe a 9 year-old female child from Palestinian related parents (first cousins) of Muslim origin. Pregnancy and delivery were unremarkable; the baby weighted 3 kg at birth. She was referred to have had normal development until 18th month of age, when parents started noticing unsteady gait with frequent falling. These symptoms were not cared of until 7 years of age, when learning difficulties became evident: she showed attention and memory deficit with difficulties manipulating objects and dysgraphia. Behavioral alterations were also noted, with abnormal sleep pattern characterized by night terrors and purposeless movements.

At first evaluation (at 7 years of age) she weighted 18 kg (5th centile), her height was 116 cm (25th centile), and her head circumference was normal (51.5 cm). Physical examination revealed neither facial dysmorphisms nor organomegaly. Neurological examination showed normal eye movements, cock-wheel rigidity of upper limbs with normal tone of legs, unsteady wide-based gait. Deep tendon reflexes were normal. Her blood and urine work including lactate, ammonia, CPK, hepatic assessment, urine organic acids, and plasma amino acid chromatography did not show any alteration.

Brain MRI showed mild cerebellar vermis atrophy and symmetrical bilateral hyperintensity of lentiform nucleus in T2 (Figure 1A) and diffusion weighted images, with cavitated aspects in FLAIR sequence, as seen in necrotic brain lesions (Figure 1B).

Her symptoms deteriorated progressively leading to tetraparesis and severe impairment of cognitive and language performances. She is now, at 9 years of age, bed-ridden. The parents signed an informed consent, approved by the ethics committee at Makassed Hospital, in agreement with the Declaration of Helsinki.

Measurement of the mitochondrial respiratory chain enzymes activity was performed by standard spectrophotometric techniques in muscle homogenate and in digitonin-treated cultures skin fibroblasts grown either in glucose-rich or in 5 mM galactose, glucose-free DMEM media for 48 h. (Bugiani et al., 2004). Oxygen consumption rate was measured using a SeaHorse FX-96 apparatus (Bioscience) in fibroblasts grown in glucose-rich DMEM medium and maximal respiration rate (MRR) was calculated as previously described (Invernizzi et al., 2012).

Total genomic DNA was extracted by standard methods from peripheral blood lymphocytes. Whole-exome sequencing (WES), using a commercial capture kit (Agilent SureSelect Human All Exon 50 Mb Kit), was performed as previously reported (Edvardson et al., 2013). Sanger sequencing of exon two of TTC19 was performed in Pt and her parents, using the following oligonucleotides: forward (Fw) TCCTGAGCTGGAGCCTGG and reverse (Rc) AAAGCCTGGATGGATTCCAAT.

Total RNA was extracted from culture fibroblasts by RNeasy Mini Kit (QIAGEN), then 200 ng of RNA were retro-transcribed using GoScript Reverse Transcriptase (Promega), following manufacturer’s recommendations. The expression levels of TTC19 and the housekeeping gene GAPDH, used for normalization, were analyzed by GoTaq qPCR Master Mix (Promega).

Primer sequences: for TTC19, Fw: CAAGCTGACCCCGTATAAATGC and Rc: AACAAGAAGGCCATCACTTACACTT; for GAPDH, Fw: CTCTGCTCCTCCTGTTCGAC and Rc: ACGACCAAATCCGTTGA.

To obtain a mitochondrial enriched fraction we digitonized culture fibroblasts. Briefly, approximately 2 × 10⁶ cells were pelleted, washed twice with PBS and incubated on ice with buffer A (MOPS 20 mM, sucrose 0,25 M, pH 7.4) and digitonin 100 μg/ml for 5 min. Then sample was centrifuged at 5,000 × g for 3 min and pellet was incubated on ice with buffer B (MOPS tampon 20 mM, sucrose 0.25 M, EDTA Na₄ 1 mM, pH 7.4) for 5 min, centrifuged at 10,000 × g for 3 min and the pellet obtained was used for western blot analysis. 50 μg of digitonin-treated culture fibroblasts from patient and controls were separated by 12% SDS-polyacrylamide gels and transferred to nitrocellulose membrane and incubated with antibodies against TTC19 (Sigma) and SDHA (Invitrogen). Immunoblot analysis was performed with the ECL-chemiluminescence kit (Amersham).

About 10⁶ culture fibroblasts from patient and controls were pelleted and incubated on ice with digitonin 0.4% for 10 min. Then samples were washed twice with PBS and incubated on ice with dodecylmaltoside 1% and NativePAGE Sample Buffer 1X (Invitrogen) for 15 min. After the incubation, samples were centrifuged at 20,000 × g for 30 min and the supernatants were measured out. Then 10 μg of supernatants plus the NativePAGE 0,25% G-250 sample additive (Invitrogen) were separated by 3–12% gradient NativePAGE Novex Bis-Tris Gels (Invitrogen) and transferred to nitrocellulose membranes and incubated with antibody against UQCRC1 (Invitrogen), NDUFS9 (Invitrogen) and SDHA (Invitrogen).

Spectrophotometric measurement of the activities of mitochondrial respiratory chain (MRC) complexes showed isolated, marked reduction of III in Pt muscle (28% of control mean, after normalization for the activity of citrate synthase, CS; Table 1). A decrease in cIII/CS was also observed in Pt fibroblasts grown in either glucose or galactose medium (55 and 52% of the control mean, respectively; Figure 2A). Accordingly, MRR of Pt fibroblasts was reduced compared to control cells, indicating reduced electron flow through the MRC (Figure 2B).

We performed whole exome sequencing on genomic DNA from Pt. After filtering steps to exclude common SNPs (frequency > 0.5%), we searched for homozygous variants, according to a predicted recessive mode of inheritance and because of the parents’ consanguinity. This strategy revealed the presence of a homozygous missense variant in TTC19. The TTC19 open reading frame contains two putative start codons, corresponding to methionines M₁ (NM_017775.2) and M₁₂₂ (NM_017775.3), but the 380-aminoacids protein, starting from the M₁₂₂, has been proposed to be the most likely human TTC19 protein (Ghezzi et al., 2011) and now the only one reported in most of the protein databases. This is corroborated from the fact that the region upstream the M₁₂₂ is not conserved among species. The variant found in Pt, predicted to cause the c.220G > C, p.Arg74His change in the longer isoform, was in fact localized in an untranslated region (c.–239G > C) upstream the AUG initiation codon of the standard TTC19 isoform. To exclude a deleterious effect of this variant on the stability of TTC19 mRNA, we quantified the levels of TTC19 transcript in Pt fibroblasts that were comparable to control subjects. Next, we checked the coverage of all TTC19 exons obtained by the exome analysis to identify any missing region: we noticed that exon two was poorly captured by the kit used for WES. We performed Sanger’s sequencing of exon two and identified a homozygous 17 bp duplication (c.213_229dup) in Pt (Figure 2C), that was present in heterozygous state in her parents (Figure 2D). This rearrangement is predicted to cause a frame-shift with the creation of a premature termination codon (p.Gln77Argfs^∗30). The analysis of the TTC19 mRNA obtained from Pt fibroblasts confirmed the presence of the 17 extra bases in the full-length transcript, without evidence of any further aberrant transcript (not shown) suggesting no influence of the duplication in the splicing processes.

As expected from the predicted consequence of the mutation on the protein, immunoblot analysis on digitonin-treated fibroblasts from Pt showed the absence of the mature TTC19 wild-type species (Figure 3A). To evaluate if the absence of TTC19 protein alters the assembly and stability of cIII we performed Blue-Native Electrophoresis analysis. No muscle was available for this assay, hence we used Pt fibroblasts. We did not observe a clear reduction in the amount of cIII holocomplex, but we detect the presence of cIII-specific assembly intermediates containing UQCRC1 subunit (Figure 3B), similar to those previously reported in muscle from mutant TTC19 subjects (Ghezzi et al., 2011).