Transcriptional changes in OXPHOS complex I deficiency are related to anti-oxidant pathways and could explain the disturbed calcium homeostasis.

Defective complex I (CI) is the most common type of oxidative phosphorylation disease, with an incidence of 1 in 5000 live births. Here, whole genome expression profiling of fibroblasts from CI deficient patients was performed to gain insight into the cell pathological mechanism. Our results suggest that patient fibroblasts responded to oxidative stress by Nrf2-mediated induction of the glutathione antioxidant system and Gadd45-mediated activation of the DNA damage response pathway. Furthermore, the observed reduced expression of selenoproteins, might explain the disturbed calcium homeostasis previously described for the patient fibroblasts and might be linked to endoplasmic reticulum stress. These results suggest that both glutathione and selenium metabolism are potentially therapeutic targets in CI deficiency.

Patient-derived fibroblasts indicate oxidative stress status and may justify antioxidant therapy in OXPHOS disorders.

Oxidative phosphorylation disorders are often associated with increased oxidative stress and antioxidant therapy is frequently given as treatment. However, the role of oxidative stress in oxidative phosphorylation disorders or patients is far from clear and consequently the preventive or therapeutic effect of antioxidants is highly anecdotic. Therefore, we performed a systematic study of a panel of oxidative stress parameters (reactive oxygen species levels, damage and defense) in fibroblasts of twelve well-characterized oxidative phosphorylation patients with a defect in the POLG1 gene, in the mitochondrial DNA-encoded tRNA-Leu gene (m.3243A>G or m.3302A>G) and in one of the mitochondrial DNA-encoded NADH dehydrogenase complex I (CI) subunits. All except two cell lines (one POLG1 and one tRNA-Leu) showed increased reactive oxygen species levels compared with controls, but only four (two CI and two tRNA-Leu) cell lines provided evidence for increased oxidative protein damage. The absence of a correlation between reactive oxygen species levels and oxidative protein damage implies differences in damage prevention or correction. This was investigated by gene expression studies, which showed adaptive and compensating changes involving antioxidants and the unfolded protein response, especially in the POLG1 group. This study indicated that patients display individual responses and that detailed analysis of fibroblasts enables the identification of patients that potentially benefit from antioxidant therapy. Furthermore, the fibroblast model can also be used to search for and test novel, more specific antioxidants or explore ways to stimulate compensatory mechanisms.

The role of mitochondrial OXPHOS dysfunction in the development of neurologic diseases.

The development of neurologic disease is a complex and multi-faceted process. Several factors, such as physiology, environment and genetics may play key roles in the manifestation of the associated illnesses. During the past decades, it has become clear that, at the cellular level, mitochondria function as more than "just" an energy source for our cells and plays a significant role in such aspects as neuronal development, maintenance and degeneration. Malfunctions in mitochondrial respiration and ATP production may prove disastrous for our cells and neurons, ultimately resulting in apoptosis, neurodegeneration and consequently, neurodegenerative diseases.

Tumoricidal activity of a novel anti-human DR5 monoclonal antibody without hepatocyte cytotoxicity.

A novel anti-human DR5 monoclonal antibody, TRA-8, induces apoptosis of most tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-sensitive tumor cells both in vitro and in vivo. In contrast to both the membrane-bound form of human TRAIL, which induced severe hepatitis in mice, and the soluble form of human TRAIL, which induced apoptosis of normal human hepatocytes in vitro, TRA-8 did not induce significant cell death of normal human hepatocytes. However, both primary hepatocellular carcinoma cells and an established liver cancer cell line were highly susceptible to the killing mediated by TRA-8. We show here that elevated levels of cell-surface expression of DR5 and increased susceptibility to DR5-mediated apoptosis are characteristics of malignant tumor cells. In contrast, DR5 alone is not sufficient to trigger apoptosis of normal hepatocytes. Therefore, selective, specific targeting of DR5 with an agonistic antibody might be a safe and effective strategy for cancer therapy.

In vitro regulation of IgA subclass synthesis. I. Discordance between plasma cell production and antibody secretion.

To investigate the in vitro regulation of IgA subclass synthesis, peripheral blood lymphocytes from healthy adults were cultured with the polyclonal B cell activator, pokeweed mitogen. Although 50% of the IgA plasma cells from a 7-d culture were positive for cytoplasmic IgA1 and 50% were positive for IgA2, less than 10% of the IgA released into the culture supernatant was IgA2. This discrepancy could not be explained by failure of the assay to detect in vitro synthesized IgA2, selective loss or destruction of IgA2 in culture media, delayed release of IgA2, or failure of IgA2 plasma cells to produce J chain. The results suggest that additional signals may be required for the differentiation of plasma cells into immunoglobulin-secreting cells.

VH gene expression is restricted in anti-IgG antibodies from MRL autoimmune mice.

Antibodies directed against IgG and DNA are found in the sera of autoimmune MRL/Mp lpr/lpr mice. Little is known of the molecular mechanisms underlying expression of such autoantibodies. We have investigated the binding diversity and pattern of VH gene expression in a panel of murine anti-IgG antibodies. We constructed eight hybridoma clones secreting IgM antibodies that bound to mouse IgG by using spleen cells from MRL/Mp lpr/lpr mice varying in age from 4 to 15 wk; one clone was derived from a 32-wk-old MRL +/+ mouse. The monoclonal IgM products exhibited varying binding specificities for intact mouse IgG, fragments of mouse IgG [Fc, Fab, (Fab')2], and heterologous IgG. Two of these antibodies crossreacted with B and/or Z DNA. Probes from seven of eight identified mouse VH gene families (7183, S107, Q52, J558, J606, 36-60, and 3609) were hybridized under high-stringency conditions with cytoplasmic RNA blots from each clone. Six clones hybridized only with the probe from the five-member 36-60 family. The remaining three clones crosshybridized with the 36-60 probe and the probe from the 60 member J558 family, perhaps reflecting somatic mutation from the original germline VH gene resulting in recognition by a probe from another family, in addition to the probe from the original germline family. Our results indicate that spontaneous MRL lpr/lpr anti-IgG antibodies are encoded predominantly by the 36-60 VH gene family and imply a nonrandom selection of this VH gene family in the production of these antibodies.

Human rheumatoid factors with restrictive specificity for rabbit immunoglobulin G: auto- and multi-reactivity, diverse VH gene segment usage and preferential usage of V lambda IIIb.

To determine the molecular and functional properties of human rheumatoid factors (RF), we established stable hybridomas and Epstein-Barr virus-transformed B cell lines from the synovial fluid or peripheral blood of three patients with rheumatoid arthritis and one patient with systemic lupus erythematosus. 17 cell lines were obtained that produced high-titer immunoglobulin M (IgM) RF that reacted exclusively with rabbit but not human IgG or IgG of other mammalian species. Certain anti-rabbit IgG RF also had specificity for other mammalian antigens (Ag), including cytoskeletal proteins and intracellular proteins found in HeLa cells, as well as for Ag present in an extract prepared from the cell wall of group A streptococci. 13 of the 17 RF contained lambda-type light (L) chains, of which 12 were classified serologically as members of the lambda-L chain variable region (V lambda) subgroup, designated V lambda III. The heavy chain V region (VH) and V lambda sequences of nine of these IgM lambda RF were determined at the cDNA level. Five VH genes in three VH families were used by these antibodies (Ab), including VH1 (dp21/1-4b and dp10 [51p1]/hv1051), VH3 (dp38/3-15 and dp77/13-21), and VH4 (dp70/4-4b). The deduced V gene-encoded amino acid sequences of the lambda chains of these IgM lambda RF confirmed their serological classification as lambda III, and they were further classified as members of the relatively uncommon V lambda III subgroup, designated V lambda IIIb. Based on cDNA analyses, nine were the product of three different V lambda III b germline genes. Two such genes, designated hsiggll150 and hsiggll295, were cloned and sequenced from genomic DNA. Unique combinations of these VH and V lambda III b genes could be related to distinctive patterns of reactivity among the IgM lambda RF. Although the VH and V lambda regions of these Abs were expressed primarily as germline-encoded sequences, four of nine multireactive Abs had extensive V region mutation, indicative of an Ag-driven process. The finding that lambda IIIb L chains are preferentially found among anti-rabbit IgG RF, and that some of these Ab have specificity for other protein, cellular, and bacterial Ag, provides new insight into the pathogenesis of RA and related diseases.

Accessory cells in murine Peyer's patch. I. Identification and enrichment of a functional dendritic cell.

Previous studies have suggested that in vitro and in vivo immune responses are defective in Peyer's patch (PP) as a result of a deficiency in accessory cell number or function. However, we report here that enzymatic dissociation of PP does release a cell population with accessory activity in oxidative mitogenesis, i.e., the proliferation of periodate-modified T cells. The accessory activity present in PP is quantitatively similar to that of spleen. Accessory function is mediated by a cell type(s) that has the following characteristics: low buoyant density, lack of adherence to plastic or glass surfaces, lack of Fc receptors, and presence of surface Ia and the 33D1 dendritic cell (DC)-specific determinants. This PP accessory cell was markedly enriched by a novel technique. PP cells formed large aggregates when cultured for 16 h with irradiated, periodate-treated spleen cells. From the clusters we obtained a low density cell population that was 60% Ia positive, 33D1 positive, non-T and non-B, Fc receptor-negative, and dendritic in morphology. The DC-enriched populations were 60-80-fold enriched in accessory function relative to unfractionated PP. We can now compare PP accessory cells with accessory cells from other organs, and try to determine how PP dendritic cells contribute to the unique functions of this lymphoid organ.

Production of predominantly polymeric IgA by human peripheral blood lymphocytes stimulated in vitro with mitogens.

Human peripheral blood lymphocytes (PBL) were cultured for various time periods (up to 8 d) in the presence of pokeweed mitogen (PWM), lipopolysaccharide, or Epstein-Barr virus. Cell-free supernates were fractionated on a standardized ultrogel AcA 22 column and the proportion of polymeric and monomeric IgA was determined by radioimmunoassay. The results demonstrate that PBL stimulated with these mitogens produce IgM and IgG with molecular characteristics identical to those found in serum, but that the IgA produced is predominantly of the polymeric type. To prove that this IgA represented disulfide bond-linked polymers rather than aggregated monomers, we have demonstrated that the high molecular weight IgA (a) maintains its polymeric form upon treatment with acidic buffers, (b) contains J chain, a glycoprotein associated only with polymeric immunoglobulins, and (c) dissociates to the monomeric form upon reduction of disulfide bonds. After 1 wk in culture, approximately 60% of the PWM-stimulated cells that contained IgA were positive for IgA2, whereas 40% were IgA1 positive as determined by immunofluorescence. Therefore, peripheral blood contains a population of lymphocytes with the potential to display, after appropriate stimulation and differentiation, characteristics similar to IgA cells found in external secretory tissues. The demonstration of the presence of such cells in the peripheral circulation suggests that these cells are precursors of IgA-producing plasma cells with the potential to populate mucosal tissues.

Preferential induction of polyclonal IgA secretion by murine Peyer's patch dendritic cell-T cell mixtures.

Polyclonal IgA secretion is inducible in murine B cells when DC-T from Peyer's patches (PP) provide the inducing stimulus. PP DC-T, which are composed predominantly of dendritic cells and Lyt-1+ T cells, are capable of dramatic augmentation of IgA secretion by PP or spleen B cells with minimal induction of IgM secretion. DC-T from spleen, however, are incapable of augmenting IgA secretion by either PP or spleen B cells. The level of IgA secretion is dependent upon the dose of DC-T providing the inducing stimulus and reaches a plateau with DC-T:B ratios of less than 1:1. This system for preferential induction of IgA responses should permit elucidation of cellular mechanisms involved in regulation of IgA secretion.

Murine Peyer's patch T cell clones. Characterization of antigen-specific helper T cells for immunoglobulin A responses.

We successfully cloned antigen-specific T cells from murine gut-associated lymphoreticular tissue, i.e., Peyer's patches, which are dependent upon T cell growth factor and independent of antigen for continuous growth. These clones exhibit helper activity for IgA responses to sheep erythrocytes (SRBC) and have been designated T helper (Th) A. Two broad categories of Th A clones have been maintained in continuous culture. The first group supports IgM and largely IgA anti-SRBC plaque-forming cell (PFC) responses in both normal and SRBC-primed splenic B cell cultures, whereas the second group supports low IgM, IgG1, and IgG2 and high IgA PFC responses. Subclones derived from single cells maintain the parent helper properties when propagated in culture for long periods (greater than 7 mo). Cloned Th A cells are antigen specific and do not support polyclonal or immune responses to other thymus dependent antigens in normal B cell cultures. Th A cells require full histocompatibility for helper functions because addition of cloned Th A cells to B cell cultures from other H-2 types does not result in IgA responses. Cloned Th A cells are Thy-1.2+ and Lyt-1+ and Lyt-2-, Ig-, and I-A-. Th A cells bear Fc receptors for IgA and do not possess receptors for IgM or IgG isotypes. Thus, T cells that primarily promote IgA isotype responses have been isolated in high frequency from murine PP, an anatomical site of major importance for induction and regulation of the IgA response.

Isotype specificity of helper T cell clones. Peyer's patch Th cells preferentially collaborate with mature IgA B cells for IgA responses.

The nature of the IgA B cell precursors that receive preferential help from selected clones of T helper cells from mouse Peyer's patches (PP Th A) were studied. Activation of the PP Th A clones required the presence of antigen, sheep erythrocytes (SRBC), in a culture system supporting development of antibody-secreting plasma cells. Two types of PP Th A cells were used. Both gave vigorous IgA responses; the first also supported low IgM, and the second low IgM and IgG subclass antibody responses. Removal of sIgA+ B cells from either splenic or PP B cell cultures selectively depleted precursors of IgA antibody producers. Cultures of purified sIgA+ B cells, cloned PP Th A cells and SRBC, selectively yielded IgA antibody producers. Finally, PP Th A cells did not support IgA responses in B cell cultures derived from spleens of young mice (days 1-25), and full IgA responses were not seen until the donor mice were 6-7 wk of age. These results suggest that cloned T helper cells can recognize and collaborate with mature, IgA committed B cells.

Interleukins and IgA synthesis. Human and murine interleukin 6 induce high rate IgA secretion in IgA-committed B cells.

Freshly isolated murine PP B cells were cultured with 10 different cytokines, including IL-1 alpha, IL-2, IL-4, IL-5, IL-6, IL-7, IFN-gamma, TNF-alpha, and TGF-beta, to investigate a possible role for these cytokines in induction of Ig synthesis. Of interest was the finding that only IL-5 and both mouse recombinant (mr) and human recombinant (hr) IL-6 enhanced IgA synthesis. The effect was greater with either mrIL-6 or hrIL-6 than with mrIL-5. IL-6 induced cycling mIgA+ PP B cells to secrete high levels of IgA (approximately 7-fold increase over control). Of importance was the finding that mrIL-6 had little effect on secretion of IgM or IgG by PP B cell cultures. hrIL-6 also increased IgA secretion by PP B cells and this enhancement was abolished by a goat anti-hrIL-6 antiserum. mrIL-6 did not cause B cell proliferation but induced a sharp increase in numbers of B cells secreting IgA. Isotype-switching was not a mechanism for this marked increase in IgA synthesis since mIgA- PP B cells were not induced to secrete IgA by mrIL-6. From these studies we conclude that IL-6 plays an important role in promoting the terminal differentiation of PP B cells to IgA-secreting plasma cells.

Suppression of in vitro monoclonal human rheumatoid factor synthesis by antiidiotypic antibody. Target cells and molecular requirements.

Previous studies have indicated that antiidiotypic antibody can modulate expression of idiotype both in vivo and in vitro. Although the precise mechanisms underlying modulation of idiotype expression by antiidiotype remains unclear, a requirement for intact IgG antiidiotypic antibody has been suggested and T cells appear to play a role in some systems. We have studied peripheral blood mononuclear leukocytes (MNL) from a patient with a B cell lymphoma and a circulating IgMK rheumatoid factor (RF) paraprotein in an effort to delineate mechanisms involved in regulation of idiotype expression by antiidiotypic antibody. 1-10% of MNL from this patient could be cytoplasmically stained with specific F(ab')2 antiidiotypic antibody. MNL from the patient spontaneously synthesized IgM RF in culture that possessed the same idiotype as the circulating IgM RF paraprotein. Production of RF by MNL was suppressed by pretreatment with either intact IgG or the F(ab')2 fragments of antiidiotypic antibody (50% inhibitory concentration was 0.2 and 1.1 micrograms/culture, respectively). In contrast, the Fab' fragment of antiidiotypic antibody was not inhibitory (up to 57 micrograms/culture) despite retaining demonstrable antiidiotype activity. Suppression of RF production was not observed over the same concentration range with the IgG or F(ab')2 fractions of a non-cross-reactive antiidiotypic antibody prepared against another monoclonal IgMK RF paraprotein or with IgG or F(ab')2 fractions prepared from normal rabbit serum. Inhibition of RF production by antiidiotypic antibody did not require T cells. Antiidiotypic antibody decreased intracellular and extracellular levels of idiotype indicating diminished synthesis of idiotype by the patient's B cells. Synthesis of IgM RF by MNL obtained from unrelated donors was not suppressed by the antiidiotypic antibody specific for the patient's paraprotein. The results indicate that (a) antiidiotypic antibody is capable of directly suppressing human B cell release of idiotype, (b) the bivalent antigen-binding fragment (F[ab']2) of antiidiotypic antibody is sufficient for mediating such suppression, (c) an intact Fc portion of antiidiotypic antibody enhances suppression of idiotype, and (d) antiidiotypic antibody inhibits idiotype expression by suppressing synthesis of idiotype.

Evidence of antigen receptor-influenced oligoclonal B lymphocyte expansion in the synovium of a patient with longstanding rheumatoid arthritis.

Plasma cell infiltration of synovium is common in longstanding rheumatoid arthritis (RA). The mechanism(s) underlying synovial B cell proliferation remains unclear. One theory invokes nonspecific polyclonal stimuli; another implicates antigen as the driving force. Antigen-driven repertoires are characteristically enriched for related sets of V gene segments containing similar sequence in the antigen binding site (complementarity-determining regions; CDRs). To study the forces shaping B cell proliferation, we analyzed V kappa transcripts expressed in the synovium of an RA patient. We found Humkv325, a developmentally regulated V kappa III gene segment associated with autoantibody reactivity, in > 10% of randomly-chosen synovial C kappa cDNAs. Two sets of sequences contained identical charged amino acid residues at the V kappa-J kappa join, apparently due to N region addition. We generated "signature" oligonucleotides from these CDR3s and probed PCR amplified V kappa products from the synovium and PBLs of the same patient, and from PBLs and spleen of individuals without rheumatic disease. Significant expression of transcripts containing these unique CDR3 sequences occurred only in the patient's synovium. Thus, in this synovium there is expansion of a limited set of B cell clones expressing antigen receptors that bear evidence of antigen selection.

Circulating immune complexes and immunoglobulin A rheumatoid factor in patients with mesangial immunoglobulin A nephropathies.

Circulating immune complexes (CIC) containing IgA and C3 were elevated in 48% of IgA nephropathy patients; IgA1 was the predominant subclass. IgA1-IgG CIC were detected in 44%, IgA2-IgG CIC in 7%, and IgM-IgA1 CIC in 16% of the patients. No IgM-IgA2 CIC were detectable. Sucrose gradient ultracentrifugation indicated that IgG-IgA1 CIC were predominantly of intermediate (13-19S) size whereas IgA1-C3 CIC sedimented from 11S to 19S. At acid pH, isolated CIC revealed the presence of substantial amounts of 7S IgA. One third of the patients had elevated serum IgA rheumatoid factor (RF) of both polymeric and monomeric forms despite normal levels of IgM-RF; 87% of patients with elevated IgA-RF had IgA1-IgG CIC. These results indicate that the IgA1 component of CIC in patients with IgA nephropathy is not necessarily of mucosal origin and suggest that a portion of these CIC consists of IgA RF immunologically complexed with autologous IgG.

Somatic mutation and CDR3 lengths of immunoglobulin kappa light chains expressed in patients with rheumatoid arthritis and in normal individuals.

Immunoglobulin secretion by plasma cells infiltrating synovial membranes is a prominent feature of RA. Previous analyses of a cDNA library generated from synovium of RA patient BC revealed immunoglobulin kappa light chain transcripts with extensive somatic mutation, frequent N region addition, and unexpected variation in the lengths of CDR3 regions which form the center of the antigen binding site. To determine if these characteristics are present in other individuals, we performed reverse transcription-polymerase chain reaction amplification and sequenced > or = 10 V kappa-containing amplicons from nine tissue samples: synovia of three individuals with long-standing RA (including patient BC), PBLs of two of these individuals, and PBLs or splenocytes of four normal individuals. Increased levels of somatic mutation in PBLs appeared to correlate with increased age, which may reflect accumulation of circulating memory cells and/or decreased bone marrow production of naive B lymphocytes. Two of three RA synovial samples and both RA PBL samples exhibited increased proportions of clones with unusual CDR3 lengths. Enrichment for these antibody binding sites could be due to abnormal regulation of the emerging repertoire or to selection for B lymphocytes bearing antibodies of unusual specificity, and may play a role in the pathogenesis of RA.

Impaired primary mouse myotube formation on crosslinked type I collagen films is enhanced by laminin and entactin.

In skeletal muscle, the stem cell niche is important for controlling the quiescent, proliferation and differentiation states of satellite cells, which are key for skeletal muscle regeneration after wounding. It has been shown that type I collagen, often used as 3D-scaffolds for regenerative medicine purposes, impairs myoblast differentiation. This is most likely due to the absence of specific extracellular matrix proteins providing attachment sites for myoblasts and/or myotubes. In this study we investigated the differentiation capacity of primary murine myoblasts on type I collagen films either untreated or modified with elastin, laminin, type IV collagen, laminin/entactin complex, combinations thereof, and Matrigel as a positive control. Additionally, increased reactive oxygen species (ROS) and ROCK signaling might also be involved. To measure ROS levels with live-cell microscopy, fibronectin-coated glass coverslips were additionally coated with type I collagen and Matrigel onto which myoblasts were differentiated. On type I collagen-coated coverslips, myotube formation was impaired while ROS levels were increased. However, anti-oxidant treatment did not enhance myotube formation. ROCK inhibition, which generally improve cellular attachment to uncoated surfaces or type I collagen, enhanced myoblast attachment to type I collagen-coated coverslips and -films, but slightly enhanced myotube formation. Only modification of type I collagen films by Matrigel and a combination of laminin/entactin significantly improved myotube formation. Our results indicate that type I collagen scaffolds can be modified by satellite cell niche factors of which specifically laminin and entactin enhanced myotube formation. This offers a promising approach for regenerative medicine purposes to heal skeletal muscle wounds. STATEMENT OF SIGNIFICANCE: In this manuscript we show for the first time that impaired myotube formation on type I collagen scaffolds can be completely restored by modification with laminin and entactin, two extracellular proteins from the satellite cell niche. This offers a promising approach for regenerative medicine approaches to heal skeletal muscle wounds.

Enteric neurons from Parkinson's disease patients display ex vivo aberrations in mitochondrial structure.

Based on autopsy material mitochondrial dysfunction has been proposed being part of the pathophysiological cascade of Parkinson's disease (PD). However, in living patients, evidence for such dysfunction is scarce. As the disease presumably starts at the enteric level, we studied ganglionic and mitochondrial morphometrics of enteric neurons. We compared 65 ganglia from 11 PD patients without intestinal symptoms and 41 ganglia from 4 age-matched control subjects. We found that colon ganglia from PD patients had smaller volume, contained significantly more mitochondria per ganglion volume, and displayed a higher total mitochondrial mass relative to controls. This suggests involvement of mitochondrial dysfunction in PD at the enteric level. Moreover, in PD patients the mean mitochondrial volume declined in parallel with motor performance. Ganglionic shrinking was evident in the right but not in the left colon. In contrast, mitochondrial changes prevailed in the left colon suggesting that a compensatory increase in mitochondrial mass might counterbalance mitochondrial dysfunction in the left colon but not in the right colon. Reduction in ganglia volume and combined mitochondrial morphometrics had both predictive power to discriminate between PD patients and control subjects, suggesting that both parameters could be used for early discrimination between PD patients and healthy individuals.