Lost characters and lost taxonomists: Coelocarteriidae fam. nov., Poecilosclerida with and without chelae, to accommodate supposed phloeodictyids (Haplosclerida)

The taxonomic study of Great Amazon Reef System sponges yielded three Oceanapia-like (Phloeodictyidae, Haplosclerida) species of similar gross morphology, two preliminarily identified as O. bartschi and another as Coelocarteria (Poecilosclerida), due to the presence of common palmate isochelae. The remarkable overall similarity of all three species in gross morphology necessitated an integrative assessment of the phylogenetic affinities. A selection of haplosclerid and poecilosclerid sequences (18S and 28S rRNA) was gathered from Genbank and compared to sequences mapped to reference from metagenome of two Oceanapia-like species from the Amazon River mouth, one of which matched O. bartschi. Both Brazilian species clustered with Coelocarteria singaporensis (Singapore). These species nest in the Poecilosclerida, far from Oceanapia (sp. and isodictyiformis) and other haplosclerids (Amphimedon, Petrosia, Siphonodictyon and Xestospongia) but also far from the poecilosclerid Isodictya that is currently classified in the same family as Coelocarteria, the Isodictyidae. Specimens with chelae are named Coelocarteria alcoladoi sp. nov. herein, while those without chelae represent the other two species. One matched Inflatella bartschi (O. bartchi’s holotype, here re-examined), thereby supporting the transfer of this species to Coelocarteria. The other is proposed as C. amadoi sp. nov. and is the second known lipochelous species in this genus. The 28S phylogenies recovered Coelocarteria bartschi comb. nov. (formerly Oceanapia bartschi) in the Poecilosclerida clade that clustered with Coelocarteria spp., including the type species of this genus, C. singaporensis, with 100% support. Coelocarteria alcoladoi sp. nov., also without chelae, grouped in the same clade, thereby corroborating the classification in this genus. The clade composed of Coelocarteria spp. grouped with Cornulella sp., suggesting an affinity between these genera. Coelocarteria is currently situated within Isodictyidae and Cornulella in Acarnidae. Isodictya (Isodictyidae) grouped with Amphilectus (Esperiopsidae) with high support and is only distantly related to Coelocarteria. Acarnus (Acarnidae) grouped with Clathria, also with high support, far from Coelocarteria and Cornulella. These results suggest the polyphyletic nature of Isodictyidae and Acarnidae, for which reason we preferred to propose a new, currently monotypic family for Coelocarteria spp., Coelocarteriidae fam. nov. ZooBank: urn:lsid:zoobank.org:pub:71FDB6FD-4A5F-4180-8DA7-79EA4CB615D1

Alien hotspot: Benthic marine species introduced in the Brazilian semiarid coast

In this study, we provide a baseline assessment of introduced marine species along the extensive (~600 km) Brazilian semiarid coast. We reported 27 introduced and 26 cryptogenic species. Moreover, the main vectors of introduction were ballast water, shipping lines, oil and gas activities, biofouling, and rafting on plastic debris. The taxa were Ascidiacea (17 species) and Bryozoa (17), followed by Crustacea (6), Mollusca (6), Cnidaria (3), Echinodermata (3), and Porifera (1). Among these invertebrates, some species are recognized as drivers of impacts such as the invasive corals (Tubastraea tagusensis and Tubastraea coccinea), the bivalves Isognomom bicolor and Perna viridis, the crab Charybdis hellerii, the brittle star Ophiothela mirabilis, and, finally, the bryozoan Membraniporopsis tubigera. These species threaten the biodiversity of unique ecosystems such as intertidal sandstone reefs, shallow-water coral reefs, and mesophotic ecosystems. Moreover, the up-to-date results highlight that this region is a hotspot of bioinvasion in the tropical South Atlantic.

Taxonomy of Petrosiidae Van Soest, 1980 (Haplosclerida, Porifera) from Brazil

Petrosiidae (Haplosclerida, Demospongiae) is characterized by the hard and friable consistency, and an isotropic reticulated skeleton of oxeas/ strongyles. The Brazilian coast and oceanic islands harbor three out of four known genera of Petrosiidae: Neopetrosia de Laubenfels, 1949; Petrosia (Petrosia) Vosmaer, 1885 and Xestospongia de Laubenfels, 1932. A taxonomic survey of samples from several expeditions housed at the Porifera Collection of Museu Nacional, Universidade Federal do Rio de Janeiro, with additional comparative materials obtained from the Museu de Cincias Naturais, Fundao Zoobotnica do Rio Grande do Sul, and Universidade Federal de Pernambuco enabled the description of two new species: P. (P.) revizee sp. nov. and X. dorigo sp. nov. We also recorded a range extention for Neopetrosia proxima and Xestospongia muta, and Xestospongia grayi was synonymized with X. muta. Other species reported from the country are diagnosed, and an identification key for Petrosiidae from Brazil is provided.

Shipwrecks help invasive coral to expand range in the Atlantic Ocean

The invasive coral Tubastraea tagusensis (sun coral) is a habitat-forming species currently increasing its geographical range into the Atlantic Ocean, thereby causing negative ecological and socioeconomic impacts. Scuba divers observed this coral in the western equatorial Atlantic in January 2020, growing at high densities on a shipwreck from World War II (sunk in 1943) at a depth of approximately 32 m. Available footage from the beginning of the decade (2012-2018) shows no obvious signs of sun coral on this shipwreck, suggesting recent colonization and range expansion. The recent evidence of expansion was found 200 km east of the last record, which was also found on a WWII shipwreck (sunk in 1942) in 2016. We have identified hundreds of overlooked WWII shipwrecks, as well as new wrecks in shallow and mesophotic waters, that may provide stepping-stone habitats for this coral to expand its distribution in the Atlantic. We discuss the role of shipwrecks as a network of stepping stones for the sun coral spread, creating complementary paths for the invasiveness by overcoming physiological traits and the short lifespan of the coral larvae. Previous research underestimates the importance of these artificial stepping-stone patches in sustaining crucial dispersal events and range expansion of invasive species. These results are a call to action to manage the invasive Tubastraea corals at a national and international scale in the Atlantic basin.

Integrative taxonomy of Hemimycale (Hymedesmiidae: Poecilosclerida: Demospongiae) from Southeastern Brazil, with the description of two new species

Two new species of Hemimycale from Southeastern Brazil are described in this study with the use of morphological and molecular data. Hemimycale oxeata sp. nov. is orange salmon colored in life and has unique oxeote like tornote spicules in addition to the typical (subtylo-)strongyles of the genus. Hemimycale ceadensis sp. nov. is reddish orange in life and has raised, tiny pore sieves, and subtylostrongyles with asymmetrical ends and raphides as microscleres. The diagnosis of the genus Hemimycale was slightly modified for the inclusion of the new species. Maximum-Likelihood analyses of 18S rRNA and COI sequences resulted in the clustering of both new species with the type species of the genus, thus confirming our morphological identification. The value of anatomical characters of pore sieves for diagnosing species and the genus is discussed. A key for species identification is also furnished.

Cytotoxic Plakortides from the Brazilian Marine Sponge Plakortis angulospiculatus

Three new plakortides, 7,8-dihydroplakortide E (1), 2, and 10, along with known natural products 3, 4, spongosoritin A (5), 6-8, and plakortide P (9), were isolated from Brazilian specimens of Plakortis angulospiculatus. Compounds 2, 3, 5, and 7-9 displayed cytotoxic activities with IC50 values ranging from 0.2 to 10 μM. Compounds that contained a dihydrofuran ring were generally less active and displayed time dependence in their activity. The activities of compounds 2 and 7-9, carboxylic acids bearing a common six-membered endoperoxide, were higher overall than for compounds 3 and 5. The modes underlying the cytotoxic actions of plakortides 2, 3, 5, 7, and 9 were further investigated using HCT-116 cells. While dihydrofurans 3 and 5 induce a G0/G1 arrest, six-membered peroxides 2, 7, and 9 delivered a G2/M arrest and an accumulation of mitotic figures, indicating a distinctly different antimitotic response. Confocal analysis indicated that microtubules were not altered after treatment with 2, 7, or 9, therein suggesting that the mitotic arrest may be unrelated to cytoskeletal targets. Overall, we find that two related classes of natural products obtained from the same extract offer cytostatic activity, yet they do so through discrete pathways.

Neuropathological study of skeletal muscle, heart, liver, and brain in a neonatal form of glycogen storage disease type IV associated with a new mutation in GBE1 gene

Glycogen storage disease type IV (GSD IV, or Andersen disease) is an autosomal recessive disorder due to the deficiency of 1,4-alpha-glucan branching enzyme (or glycogen branching enzyme, GBE1), resulting in an accumulation of amylopectin-like polysaccharide in muscle, liver, heart and central and peripheral nervous system. Typically, the presentation is in childhood with liver involvement up to cirrhosis. The neuromuscular form varies in onset (congenital, perinatal, juvenile and adult) and in severity. Congenital cases are rare, and fewer than 20 cases have been described and genetically determined so far. This form is characterized by polyhydramnios, neonatal hypotonia, and neuronal involvement; hepatopathy is uncommon, and the babies usually die between 4 weeks and 4 months of age. We report the case of an infant who presented severe hypotonia, dilatative cardiomyopathy, mild hepatopathy, and brain lateral ventricle haemorrhage, features consistent with the congenital form of GSD IV. He died at one month of life of cardiorespiratory failure. Muscle biopsy and heart and liver autoptic specimens showed many vacuoles filled with PAS-positive diastase-resistant materials. Electron-microscopic analysis showed mainly polyglucosan accumulations in all the tissues examined. Postmortem examination showed the presence of vacuolated neurons containing this abnormal polysaccharide. GBE1 biochemical activity was virtually absent in muscle and fibroblasts, and totally lacking in liver and heart as well as glycogen synthase activity. GBE1 gene sequence analysis revealed a novel homozygous nonsense mutation, p.E152X, in exon 4, correlating with the lack of enzyme activity and with the severe neonatal involvement. Our findings contribute to increasing the spectrum of mutation associated with congenital GSD IV.

Hepatic and neuromuscular forms of glycogenosis type III: nine mutations in AGL

Glycogenosis type III (Cori disease) is an autosomal recessive disorder caused by the deficiency of the glycogen debranching enzyme, encoded by the AGL gene, and existing in six isoforms alternately spliced in a tissue-specific way. Generally, disease onset occurs early on starting from the first year of life, with hepatomegaly, hypoglycemia, hyperlipidemia, increased CK levels, and, in some cases, short stature and slight mental retardation. Frequently, hepatomegaly tends to resolve spontaneously and inexplicably during childhood, when myopathy, often associated with cardiomyopathy, arises. This disease is known to lack almost invariably clear links between the genotype and clinical phenotype. We describe nine new mutations in Italian patients: four nonsense (p.Arg285X, p.Lys422X, p.Arg910X, p.Arg977X), three frameshift (c.442delA, c.753_756delGACA, c.3963delG), and two missense (p.Ala1120Pro, p.Arg524His). Particularly, the nonsense p.Arg285X is linked to an exonic splicing enhancer and it was found to produce two species of transcripts at the same time. Moreover, we discuss a subgroup of subjects carrying c.2681+1G>A, which has proven to be the most frequent mutation among our patients. The previously described c.664+3A>G was also detected in two patients, both homozygous. The present work is yet another confirmation that the individual genetic background plays a pivotal role in influencing the phenotypes, as occurs in other metabolic diseases.

Neuronal differentiation of murine bone marrow Thy-1- and Sca-1-positive cells

Recent evidence suggests that cells from bone marrow can acquire neuroectodermal phenotypes in cell culture or after transplantation in animal models and in the human brain. However, isolation of the bone marrow cell subpopulation with neuronal differentiation potential remains a challenge. To isolate and expand neural progenitors from whole murine bone marrow, bone marrow was obtained from hind limb bone of C57BL6 mice and plated in culture with neuronal medium with basic fibroblast growth factor and epidermal growth factor. After 5-7 days in culture, cellular spheres similar to brain neurospheres appeared either floating or attached to culture dishes. These spheres were collected, dissociated, and expanded. The bone marrow-derived spheres were positive for nestin as assessed by immunocytochemistry and by reverse transcriptase polymerase chain reaction. Thy-1- and Sca-1-positive bone marrow cells selected by magnetic cell sorting resulted in a higher yield of nestin-positive spheres. After exposure to neuronal differentiative medium retinoic acid with and without Sonic hedgehog, cells positive for neuronal markers tubulin III (TuJ-1) and neurofilament (NF) were detected. The mRNA profile of these cells included the expression of TuJ-1, neuronal-specific enolase (NSE), and NF-light chain. To evaluate the in vivo behavior of these cells, spheres derived from bone marrow-derived cells of transgenic green fluorescent protein (GFP) mice were transplanted into newborn mouse brain. Two months later, the mouse neural cortex contained a minor proportion of GFP(+) cells co-expressing neuronal markers (TuJ-1, NF, MAP-2, NeuN). Although cell fusion phenomena with the host cells could not be ruled out, bone marrow-derived neurosphere transplantation could be a strategy for cellular mediated gene therapy.

Neuroectodermal and microglial differentiation of bone marrow cells in the mouse spinal cord and sensory ganglia

There is now evidence that bone marrow (BM) can generate cells expressing neuronal antigens in adult mouse brain. In the present study, we examined the spinal cord and dorsal root ganglia (DRG) of adult mice 3 months after BM cell transplantation from transgenic donor mice expressing the enhanced green fluorescent protein (GFP). To determine whether GFP(+) cells acquire neuroectodermal phenotypes, we tested, by immunocytochemistry followed by confocal analysis, the coexpression of the astrocytic marker glial fibrillary acidic protein (GFAP) and the neuronal markers NeuN, neurofilament (NF), and class III beta-tubulin (TuJ1). Rare GFP(+) cells coexpressing TuJ1, NF, and NeuN were found both in spinal cord and in sensory ganglia. These cells have small dimensions and short cytoplasmic processes, probably reflecting an immature phenotype. Double GFP and GFAP positivity was found only in spinal cord. To determine whether cell fusion with endogenous cells occurred, we investigated the nuclear content of cells coexpressing GFP and neuronal or astrocytic markers, demonstrating that these cells have only one nucleus and a DNA ploidy that it is not different from that of surrounding neurons and astrocytes. Large numbers of GFP(+) cells are also positively stained for F4/80, a microglial-recognizing antibody, and present a characteristic microglial-like morphology both in spinal cord and, with a higher frequency, in sensory ganglia. These data support a potential role for BM-derived stem cells in spinal cord neuroneogenesis. They also confirm that the microglial compartment within the CNS and in DRG undergoes a relatively fast turnover, with the contribution of hematopoietic stem cells. Both these findings might prove useful for the development of treatments for spinal cord neurodegenerative and acquired disorders.

Modulated generation of neuronal cells from bone marrow by expansion and mobilization of circulating stem cells with in vivo cytokine treatment

The aim of the present study is to determine whether the expansion and mobilization of circulating bone marrow (BM) stem cells by in vivo treatment with granulocyte-colony stimulating factor (G-CSF) and stem cell factor (SCF) increase the amount of BM-derived neuronal cells in mouse brain. The presence of BM-derived cells in the brain was traced by transplanting into lethally irradiated adults and newborns adult BM from transgenic mice that ubiquitously expressed enhanced green fluorescent protein (GFP). GFP+ and Y-chromosome+ donor-derived cells were present in several brain areas of all treated mice (cortical and subcortical areas, cerebellum, olfactory bulb). The presence of GFP+ cells expressing nuclear neural specific antigen (NeuN), neurofilament, and beta-III tubulin in cortical forebrain and olfactory bulb (OB) was higher in G-CSF-SCF treated groups (P < 0.05, analysis of variance, Fisher post hoc). We observed that overall the amount of double positive cells was higher in animals treated at birth than in adults and in OB than in forebrain areas (P < 0.05). Temporal cortical areas of cytokine-treated adult animals revealed a mean threefold increase in the number of GFP+ cells expressing the nuclear neural specific antigen (211 +/- 86 GFP+NeuN+/mm(3) in G-CSF + SCF treated mice and 66 +/- 33 GFP+NeuN+/mm(3) in control animals). GFP+ cells coexpressing neuronal markers contain only one nucleus and have a DNA index (a measure of DNA ploidy) identical to that of surrounding neurons, thus excluding donor cell fusion with endogenous cells as a relevant phenomenon under these experimental conditions. Our results indicate that G-CSF and SCF administration modulates the availability of GFP+ cells in the brain and enhances their capacity to acquire neuronal characteristics. Cytokine stimulation of autologous stem cells might be seen as a new strategy for neuronal repair in neurodegenerative diseases.

A subpopulation of murine bone marrow cells fully differentiates along the myogenic pathway and participates in muscle repair in the mdx dystrophic mouse

Bone marrow (BM) transplantation in mice suggests the existence of pluripotent cells able to differentiate into skeletal muscle tissue, although sustained myofiber reconstitution has not yet been achieved. We investigated the myogenic potential of mouse BM cells and evaluated whether a BM fraction enriched for cells expressing skeletal muscle markers would ameliorate muscle repair, when compared to whole BM, into the dystrophic mdx mouse. We demonstrate that cells expressing striated-muscle-specific proteins are already present in the BM independently from experimentally forced myogenic conversion. We observed the presence of both markers of early myogenic program such as Pax3, Myf5, MyoD, desmin, and late myogenesis such as myosin heavy chain and alpha-sarcomeric actin. These myogenic cells are more represented in the early nonadherent BM fraction, which generates clones able to fully differentiate into myotubes. Transplantation in mdx mice by intravenous injection of whole BM and a tenfold BM myogenic enriched fraction resulted in BM reconstitution and limited dystrophin restoration. Taken together, these data show that a fraction of BM cells have a definite potential for differentiation along the skeletal muscle pathway and can be recruited by muscle repair mechanisms. They also indicate that factors limiting the degree of muscle recruitment and the host stem cell competition should be assessed in order to evaluate the usefulness of BM-derived myogenic cells into the context of cell-mediated gene therapy of inherited muscle diseases.

Biodistribution studies of 99mTc-labeled myoblasts in a murine model of muscular dystrophy

The purpose of this study was twofold: first, to evaluate the myoblast labeling of various 99mTc complexes and to select the complex that best accomplishes this labeling, and second to evaluate the biodistribution of myoblasts labeled with this complex using mice with MDX muscular dystrophy (the murine homologue of Duchenne's muscular dystrophy). The following ligands were used to prepare the corresponding 99mTc complexes: hexakis-methoxy-isobutyl-isonitrile (MIBI), bis(2-ethoxyethyl)diphosphinoethane (Tf), (RR,SS)-4,8-diaza-3,6,6,9-tetramethyl-undecane-2,10-dione-bisoxime (HM-PAO), bis(N-ethyl)dithiocarbamate (NEt), and bis(N-ethoxy, N-ethyl)dithiocarbamate (NOEt). One million murine myoblasts were incubated for 30-60 minutes with 5 mCi of each of the 99mTc complexes prepared from the above ligands. Viability was assessed by microscopic counting after trypan blue staining, and the radioactivity absorbed in the cells was measured after centrifugation. The compound with the highest uptake in cellular pellets was [99mTc]N-NOEt. The biodistribution of myoblasts labeled with this complex was evaluated after intraaortic injection in dystrophic mice. Such an approach has the potential of effecting widespread gene transfer through the bloodstream to muscles lacking dystrophin.

Chemotactic factors enhance myogenic cell migration across an endothelial monolayer

Recent reports revealed that myogenic progenitors, derived from either bone marrow or muscle can migrate into muscle tissue and participate in myofiber regeneration, when injected in the peripheral circulation. This observation might open a new strategy for the treatment of muscular dystrophies. The signals involved in myoblast recruitment from circulation are at present poorly understood. To investigate myoblast migration we used a transwell assay in which murine myoblasts and myogenic cell lines were seeded on microporous membrane covered by an endothelial monolayer and chemotactic factors were added in the lower chamber. We demonstrated that myoblasts are able to cross the endothelium and that this process can be modulated. In particular among tested factors, we observed a gradient of chemotactic activity as follows: HGF >> RANTES > PDGF-A > PDGF-B > FGF >> TNF-alpha > IFN-gamma > EGF. Endothelial and myoblast expression of Pax3 (a transcription factor expressed by embryonic migrating myogenic cells) and cytokine transcripts (TNF-alpha, IFN-gamma) was also monitored either at the basal level and after transmigration. We observed increased Pax3 expression after interaction of C2C12 myoblasts with endothelial cells. We consider that any new report elucidating the molecular signals involved in myoblast migration may be useful toward the development of systemic cellular-mediated gene therapy of muscle diseases.

In vitro and in vivo tetracycline-controlled myogenic conversion of NIH-3T3 cells: evidence of programmed cell death after muscle cell transplantation

Ex vivo gene therapy of Duchenne muscular dystrophy based on autologous transplantation of genetically modified myoblasts is limited by their premature senescence. MyoD-converted fibroblasts represent an alternative source of myogenic cells. In this study the forced MyoD-dependent conversion of murine NIH-3T3 fibroblasts into myoblasts under the control of an inducible promoter silent in the presence of tetracycline was evaluated. After tetracycline withdrawal this promoter drives the transcription of MyoD in the engineered fibroblasts, inducing their myogenesis and giving rise to beta-galactosidase-positive cells. MyoD-expressing fibroblasts withdrew from the cell cycle, but were unable to fuse in vitro into multinucleated myotubes. Five days following implantation of engineered fibroblasts in muscles of C57BL/10J mice we observed a sevenfold increase of beta-galactosidase-positive regenerating myofibers in animals not treated with antibiotic compared with treated animals. After 1 week the number of positive fibers decreased and several apoptotic myonuclei were detected. Three weeks following implantation of MyoD-converted fibroblasts in recipient mice, no positive "blue" fiber was observed. Our results suggest that transactivation by tetracycline of MyoD may drive an in vivo myogenic conversion of NIH-3T3 fibroblasts and that, in this experimental setting, apoptosis plays a relevant role in limiting the efficacy of engineered fibroblast transplantation. This work opens the question whether apoptotic phenomena also play a general role as limiting factors of cell-mediated gene therapy of inherited muscle disorders.

T-antigen regulated expression reduces apoptosis of tag-transformed human myoblasts

The generation of human myogenic cell lines could potentially provide a valuable source for cell transplantation in myopathies. The dysregulation of proliferative-differentiative signals by viral oncogenes can result in the induction of apoptosis. Whether apoptosis occurred in myogenic cells expressing large T antigen (Tag) from SV40 upon differentiation was unknown. Human muscle satellite cells were transfected with two different constructs, containing either an origin-defective SV40 genome or Tag under vimentin promoter control. When differentiation was triggered, Tag expression reduced the formation of myotubes and dead cells showing apoptotic features were present. However, the cells expressing SV40 Tag under vimentin promoter control retained their capacity to form myotubes and expressed the myofibrillar proteins as myosin heavy chain and dystrophin when Tag expression was silent. Their apoptotic rate was similar to that of untransfected cells. The observation that apoptosis can be prevented by the down-regulation of Tag suggests that the programmed cell death induced in transformed cells can be reversed, and confirms the regulatory efficiency of the human vimentin promoter.

Partial depletion and multiple deletions of muscle mtDNA in familial MNGIE syndrome

OBJECTIVE

To describe the unique combination of partial depletion and multiple deletions of mitochondrial DNA (mtDNA) on muscle DNA analysis of three siblings with mitochondrial neurogastrointestinal encephalomyopathy (MNGIE).

BACKGROUND

MNGIE is a relatively homogeneous autosomal recessive disorder characterized by gastrointestinal dysmobility, ophthalmoparesis, peripheral neuropathy, mitochondrial myopathy, and altered white matter signal at brain imaging. Muscle multiple mtDNA deletions have been found in about half of the described cases.

METHODS

We studied three affected siblings (two were monozygotic twins) born to nonconsanguineous parents. Muscle mtDNA was investigated by quantitative Southern and Slot blot techniques and by PCR analysis. Morphologic confirmation in the muscle tissue was achieved by using in situ hybridization with a mtDNA probe complementary to an undeleted region and by DNA immunohistochemistry.

RESULTS

All three patients showed ragged red (RRF) and cytochrome c oxidase-negative fibers, as well as partial deficiency of complexes I and IV. Southern and Slot blot analyses showed mtDNA depletion in all patients. Multiple mtDNA deletions were also detected by PCR analysis. In situ hybridization demonstrated an overall signal weaker than controls, with a relatively higher signal in RRF. Antibodies against DNA showed a decreased cytoplasmic network.

CONCLUSIONS

The muscle histopathology and respiratory chain enzyme defects may be accounted for by the decreased mtDNA amount and by the presence of mtDNA deleted molecules; however, relative levels of mtDNA seem to correlate with life span in these patients. The combination of partial depletion and multiple deletions of mtDNA might indicate the derangement of a common genetic mechanism controlling mtDNA copy number and integrity.

Cytochrome c oxidase subunit I microdeletion in a patient with motor neuron disease

An out-of-frame mutation of the mitochondrial DNA-encoded subunit I of cytochrome c oxidase (COX) was discovered during investigation of a severe isolated muscle COX deficiency in a patient with motor neuron-like degeneration. The mutation is a heteroplasmic 5-bp microdeletion located in the 5' end of the COI gene, leading to premature termination of the corresponding translation product. Western blot analysis, immunohistochemistry, and single-fiber polymerase chain reaction demonstrated a tight correlation between COX defect, COX I expression, and percentage of mutation. COX subunits II, III, and IV were decreased as well, suggesting a defective assembly of COX holoenzyme. The mutation was associated with a clinical phenotype unusual for a mitochondrial disorder, that is, an isolated motor neuron disease (MND) with some atypical findings, including early onset, preferential involvement of the upper motor neuron, and increased cerebrospinal fluid protein content. MND may arise from impaired scavenging and overproduction of free oxygen radicals, a by-product of oxidative phosphorylation (OXPHOS). Our observation suggests that OXPHOS impairment could play a role in the pathogenesis of some MND cases.