Mta, the maternally transmitted antigen, is determined jointly by the chromosomal Hmt and the extrachromosomal Mtf genes

Mitochondrial Proteins as Source of Cancer Neoantigens

In the past decade, anti-tumour immune responses have been successfully exploited to improve the outcome of patients with different cancers. Significant progress has been made in taking advantage of different types of T cell functions for therapeutic purposes. Despite these achievements, only a subset of patients respond favorably to immunotherapy. Therefore, there is a need of novel approaches to improve the effector functions of immune cells and to recognize the major targets of anti-tumour immunity. A major hallmark of cancer is metabolic rewiring associated with switch of mitochondrial functions. These changes are a consequence of high energy demand and increased macromolecular synthesis in cancer cells. Such adaptations in tumour cells might generate novel targets of tumour therapy, including the generation of neoantigens. Here, we review the most recent advances in research on the immune response to mitochondrial proteins in different cellular conditions.

Murine MHC class Ib gene, H2-M2, encodes a conserved surface-expressed glycoprotein

We have determined the genomic sequence of H2-M2 in seven haplotypes from nine inbred strains of mice and in five wild-derived haplotypes. Except for the spretus haplotype sp1 with a premature stop codon, we found only limited polymorphism. Four of the five amino acid substitutions in the alpha-helices are at positions that would point out from the antigen-binding groove, indicating that the polymorphism might influence receptor recognition rather than antigen binding. The rat homologue, RT1.M2(lv1), has 89% identity to H2-M2 at the nucleotide level and 91% at the amino acid level, and it also encodes an intact MHC class I glycoprotein. Chimeric proteins with alpha(1)alpha(2) or alpha(3)-transmembrane domains encoded by H2-Q9 were detectable on the surface of transfectants with monoclonal antibodies against Qa2, and the full-length M2 protein, labeled by fusion with green fluorescent protein, was detectable with S19.8 monoclonal antibodies. The H2-M2 protein was thus expressed on the cell surface, even in TAP-deficient RMA-S cells at 37 degrees C, suggesting that it is TAP-independent. We conclude that H2-M2 is a conserved mouse class Ib gene that is translated to a surface-expressed MHC class I molecule with a function still to be elucidated.

The Alloreactive T Cell Response Against the Class Ib Molecule H2-M3 Is Specific for High Affinity Peptides

MHC class Ib molecule H2-M3 presents N-formylated peptides to CD8+ CTLs. Endogenous formylated peptides can come from the N-terminus of each of the 13 proteins encoded by the mitochondrial genome. In peptide competition assays, two of these peptides bind with high affinity, six bind with intermediate affinity, three bind with low affinity, and two do not bind measurably. Alloreactive CTLs from M3-specific, mixed lymphocyte cultures responded strongly against the two peptides with high affinity for M3, occasionally to peptides with intermediate affinity, and not at all to the rest. Long term lines and CTL clones reacted with only the high affinity peptides, demonstrating that alloreactive CTLs depend on specific peptides and that peptide affinity for class I correlates with alloantigenicity.

Structural organization and transcription regulation of nuclear genes encoding the mammalian cytochrome c oxidase complex

Cytochrome c Oxidase (COX) is the terminal component of the bacterial as well as the mitochondrial respiratory chain complex that catalyzes the conversion of redox energy to ATP. In eukaryotes, the oligomeric enzyme is bound to mitochondrial innermembrane with subunits ranging from 7 to 13. Thus, its biosynthesis involves a coordinate interplay between nuclear and mitochondrial genomes. The largest subunits, I, II, and III, which represent the catalytic core of the enzyme, are encoded by the mitochondrial DNA and are synthesized within the mitochondria. The rest of the smaller subunits implicated in the regulatory function are encoded on the nuclear DNA and imported into mitochondria following their synthesis in the cytosol. Some of the nuclear coded subunits are expressed in tissue and developmental specific isologs. The ubiquitous subunits IV, Va, Vb, VIb, VIc, VIIb, VIIc, and VIII (L) are detected in all the tissues, although the mRNA levels for the individual subunits vary in different tissues. The tissue specific isologs VIa (H), VIIa (H), and VIII (H) are exclusive to heart and skeletal muscle. cDNA sequence analysis of nuclear coded subunits reveals 60 to 90% conservation among species both at the amino acid and nucleotide level, with the exception of subunit VIII, which exhibits 40 to 80% interspecies homology. Functional genes for COX subunits IV, Vb, VIa 'L' & 'H', VIIa 'L' & 'H', VIIc and VIII (H) from different mammalian species and their 5' flanking putative promoter regions have been sequenced and extensively characterized. The size of the genes range from 2 to 10 kb in length. Although the number of introns and exons are identical between different species for a given gene, the size varies across the species. A majority of COX genes investigated, with the exception of muscle-specific COXVIII(H) gene, lack the canonical 'TATAA' sequence and contain GC-rich sequences at the immediate upstream region of transcription start site(s). In this respect, the promoter structure of COX genes resemble those of many house-keeping genes. The ubiquitous COX genes show extensive 5' heterogeneity with multiple transcription initiation sites that bind to both general as well as specialized transcription factors such as YY1 and GABP (NRF2/ets). The transcription activity of the promoter in most of the ubiquitous genes is regulated by factors binding to the 5' upstream Sp1, NRF1, GABP (NRF2), and YY1 sites. Additionally, the murine COXVb promoter contains a negative regulatory region that encompasses the binding motifs with partial or full consensus to YY1, GTG, CArG, and ets. Interestingly, the muscle-specific COX genes contain a number of striated muscle-specific regulatory motifs such as E box, CArG, and MEF2 at the proximal promoter regions. While the regulation of COXVIa (H) gene involves factors binding to both MEF2 and E box in a skeletal muscle-specific fashion, the COXVIII (H) gene is regulated by factors binding to two tandomly duplicated E boxes in both skeletal and cardiac myocytes. The cardiac-specific factor has been suggested to be a novel bHLH protein. Mammalian COX genes provide a valuable system to study mechanisms of coordinated regulation of nuclear and mitochondrial genes. The presence of conserved sequence motifs common to several of the nuclear genes, which encode mitochondrial proteins, suggest a possible regulatory function by common physiological factors like heme/O2/carbon source. Thus, a well-orchestrated regulatory control and cross talks between the nuclear and mitochondrial genomes in response to changes in the mitochondrial metabolic conditions are key factors in the overall regulation of mitochondrial biogenesis.

H2-M3, a full-service class Ib histocompatibility antigen

H2-M3 is an MHC class Ib molecule of the mouse with a unique preference for N-formylated peptides, which may come from the N-termini of endogenous, mitochondrial proteins or foreign, bacterial proteins. The crystal structure of M3 revealed a hydrophobic peptide-binding groove with an occluded A pocket and the peptide shifted one residue relative to class Ia structures. The formyl group is held by a novel hydrogen bonding network, involving His9 on the bottom of the groove, and the side chain of the P1 methionine is lodged in the B pocket. M3 is a full-service histocompatibility (H) antigen, i.e. self-M3 can present endogenous peptides as minor H antigens and foreign, bacterial antigens in a defensive immune response to infection; and foreign M3 complexed with endogenous self-peptides.

Rotenone, a mitochondrial NADH dehydrogenase inhibitor, induces cell surface expression of CD13 and CD38 and apoptosis in HL-60 cells

We previously demonstrated that the mitochondrial NADH dehydrogenase subunit 2 (ND2) gene was overexpressed in human acute myelogenous leukemia (AML) cells. Since this finding suggested that ND2 gene expression was related to myeloid differentiation, we here investigated the effects of rotenone, a specific NADH dehydrogenase inhibitor, on HL-60 cell growth, differentiation and death. Fifty nM rotenone inhibited the growth of HL-60 cells and caused an increase in the cell population in the G(2) +M phase. In the quantitative comparison of myeloid antigen, the expression of CD13 and CD38 were relatively increased in the rotenone-treated cells. These findings suggest that the inhibition of NADH dehydrogenase changes the cell cycle and induces some specific surface antigens of HL-60 cells. On the other hand, the expression of ND2 gene remained unchanged after the rotenone treatment, suggesting the rotenone-mediated mitochondrial inhibition did not affect the mitochondrial gene expression. Five mu M rotenone strongly inhibited the cellular proliferation. Electron microscopy and an electrophoretic analysis of DNA showed that the majority of the HL-60 cells were induced into typical apoptosis within 24-48 hours. On the basis of this and other studies, we believe that mitochondrial function is directly involved in both cellular differentiation and apoptotic cell death.

Mitochondrial antigens, molecular mimicry and autoimmune disease

The immune system is normally tolerant to mitochondrial self-antigens, but responsive against bacteria. Low-titre anti-mitochondrial antibodies (AMA) might be involved in this discrimination. Tolerance is broken in diseases characterised by high titre AMA. Some of these AMA, against cardiolipin, cross-react with DNA. The best studied AMA are those characterising primary biliary cirrhosis (PBC). These are directed against E2 subunits of the oxo-acid dehydrogenase complexes, and also against subunits E1 alpha, E1 beta and X of the pyruvate dehydrogenase complex. AMA of PBC patients also react with bacterial E2s. Reactivities are primarily peptide-specific but with cross-reactivity between mitochondrial and microbial antigens and between E2s of respective complexes. Immunodominant epitopes, for anti E2 AMA, include the conserved sequence flanking the site of lipoyl attachment. It is proposed that the initial stimulus for antibody production is chronic urinary tract infection. AMA themselves are not pathogenic, but CD4+ T-cells would be primed, recognising the lipoyl domain epitope in association with class II HLA. Inappropriate expression of class II antigens on bile duct epithelia, (as found in PBC), might lead to presentation of a particular fragment of HLA-DR alpha, known to be a major MHC presented self-peptide in the mouse. That sequence strongly mimics the lipoyl domain and might be recognised by primed T-cells, initiating the autoimmune cascade. In the mouse, a peptide of ND1 of Complex I is presented in association with class I MHC. Cells exhibiting somatic mutation of such a peptide might thus be subject to attack by CD8+ T-cells. If such peptides were presented by class II HLA, autoimmune diseases might arise, related to mimicry between such peptides and microbial sequences and/or self-antigens. These considerations might apply in Leber's disease and in age-related pathology.

Elimination of paternal mitochondrial DNA in intraspecific crosses during early mouse embryogenesis

To examine whether mtDNA is uni- or biparentally transmitted in mice, we developed an assay that can detect sperm mtDNA in a single mouse embryo. In intraspecific hybrids of Mus musculus, paternal mtDNA was detected only through the early pronucleus stage, and its disappearance co-incided with loss of membrane potential in sperm-derived mitochondria. By contrast, in interspecific hybrids between M. musculus and Mus spretus, paternal mtDNA was detected throughout development from pronucleus stage to neonates. We propose that oocyte cytoplasm has a species-specific mechanism that recognizes and eliminates sperm mitochondria and mtDNA. This mechanism must recognize nuclearly encoded proteins in the sperm midpiece, and not the mtDNA or the proteins it encodes, because sperm mitochondria from the congenic strain B6.mtspr, which carries M. spretus mtDNA on background of M. musculus (B6) nuclear genes, were eliminated early by B6 oocytes as in intraspecific crosses. We conclude that cytoplasmic genomes are transmitted uniparentally in intraspecific crosses in mammals as in Chlamydomonas and that leakage of parental mtDNA is limited to interspecific crosses, which rarely occur in nature.

Functional expression of low density lipoprotein receptor-related protein is controlled by receptor-associated protein in vivo

The 39-kDa receptor-associated protein (RAP) associates with the multifunctional low density lipoprotein (LDL) receptor-related protein (LRP) and thereby prevents the binding of all known ligands, including alpha 2-macroglobulin and chylomicron remnants. RAP is predominantly localized in the endoplasmic reticulum, raising the possibility that it functions as a chaperone or escort protein in the biosynthesis or intracellular transport of LRP. Here we have used gene targeting to show that RAP promotes the expression of functional LRP in vivo. The amount of mature, processed LRP is reduced in liver and brain of RAP-deficient mice. As a result, hepatic clearance of alpha 2-macroglobulin is impaired and remnant lipoproteins accumulate in the plasma of RAP-deficient mice that also lack functional LDL receptors. These results are consistent with the hypothesis that RAP stabilizes LRP within the secretory pathway. They also suggest a further mechanism by which the activity of an endocytic receptor may be modulated in vivo.

HMT, encoded by H-2M3, is a neoclassical major histocompatibility class I antigen

H-2M3 encodes HMT, the major histocompatibility complex (MHC) class I heavy chain of the maternally transmitted antigen (Mta). Like classical MHC class I genes, the expression of M3 can be stimulated by gamma-interferon and its message can be detected from mid-gestational embryos (day 8) through adulthood. HMTb, a nonimmunogenic allelic form of HMT, differs from the common HMTa molecule by four amino acids, of which only two (residues 31 and 95) are located in the alpha 1 and alpha 2 domains that form the peptide-binding groove. Recognition of site-directed mutants by Mta-specific cytotoxic T lymphocytes was hardly affected by the substitution of Met for Val31 but was abolished by the substitution of Gln for Leu95, which is located in the beta-sheet floor of the peptide-binding groove. Thus a single amino acid difference is responsible for the immunological silence of HMTb.

Kinetics of killing by monoclonal cytotoxic T lymphocytes. I. Colorimetric detection of cryptic CTL determinants on adherent target cells and survivorship analysis

Some targets of cell-mediated cytolysis do not efficiently release 51Cr but manifestly are killed in direct viability assays. We characterize and validate an alternative and non-radioactive (colorimetric) method for measuring killing of adherent targets by monoclonal CTL. The method obviates concerns about the effects of trypsinization, is technically simple, quantitative and in some cases more sensitive than conventional 51Cr assays. Target loss obeyed first-order kinetics with respect both to [CTL] and time. These results are consistent with an exponential (Poisson) model of killing and support the use of a single kinetic parameter to describe the lytic activity of monoclonal CTL on adherent targets. When monoclonal CTL are used at appropriate effector:target ratios (less than or equal to 1:1), the residuals obtained after least squares linear regression are homoscedastic and normally distributed, justifying the use of commonly available statistical calculators or programs for the analysis of CTL data.

Ham-2 corrects the class I antigen-processing defect in RMA-S cells

The murine major histocompatibility complex (MHC) contains two genes (Ham-1 and Ham-2) that encode members of a super-family of ATP-dependent transport proteins. These genes are believed to mediate the transport of peptide antigen from the cytoplasm into the lumen of the endoplasmic reticulum for binding by MHC class I molecules. Evidence for such a function has come from the rescue of class I surface expression by a cloned copy of the human homologue of Ham-1, PSF-1, in a human cell line that is defective in antigen processing. A mutant murine cell line, RMA-S, has an identical antigen-processing-defective phenotype. Here we show that expression of a cloned copy of the Ham-2 gene in RMA-S cells results in recovery of the ability to process and present class I-restricted antigens to cytotoxic T lymphocytes, and in partial recovery of class I surface expression. Processing defects for classical (H-2 K and D) and non-classical (Qa1 and HMT) class I molecules are corrected by Ham-2. These data indicate that both MHC-linked transporter genes are probably required for class I antigen processing, and that the functional transporter in this pathway may consist of a Ham-1/Ham-2 heterodimer.

H-2M3 encodes the MHC class I molecule presenting the maternally transmitted antigen of the mouse

Mta, the maternally transmitted antigen of mice, is a hydrophobic, N-formylated mitochondrial peptide, MTF, presented on the cell surface to cytotoxic T lymphocytes by a novel major histocompatibility complex class I molecule, encoded by H-2M3. We have cloned and sequenced two alleles of M3, which differ in their ability to present MTF despite greater than 99% identity in the coding regions. M3 is as divergent from classical, antigen-presenting H-2 molecules as from other class I genes of the Hmt and the Qa/Tla regions. Amino acids critical for folding of class I molecules are conserved in M3. Noncharged amino acids lining the peptide-binding groove and phenylalanine 171 may explain the unique interaction with MTF, and leucine 95 appears critical for immunological activity.

Generation of T cells with lytic specificity for atypical antigens. III. Priming F1 animals with antigen-bearing cells also having reactivity for host alloantigens allows for potent lytic T cell responses

Here, we explore the conditions required for generating two different highly potent F1 antiparental killer cell populations to unusual antigens in rats. The first, L/DA anti-DA, has lytic specificity for two antigen systems: MTA, a mitochondrial antigen expressed on DA and DA Lewis (L) target cells restricted by RT1A class I molecules; and H, an antigen that maps to the class I-like RT1C region and is present only on parental target cells from donors homozygous at the major histocompatibility complex. The second killer population is generated in the reciprocal DA/L anti-DA combination and has lytic specificity only for the H antigen system. We show that the killer cells are T cells, and that generation of these F1 cytotoxic T lymphocytes (CTL) requires an in vivo priming step in which it is essential that the inoculated parental cells bear the relevant target antigens and possess alloreactivity for F1 host antigens. The requirement for alloreactivity and antigen on the same priming cell population suggests that these potent lytic responses depend on a situation akin to a hapten-carrier effect that bypasses otherwise ineffective helper responses by the host to these unusual antigens. Restimulation of F1 lymphocytes in culture is also necessary, requiring the presence of antigen on irradiated lymphoblast stimulator cells, but alloreactivity to responder cell antigens is not necessary; normal, nonactivated lymph node cells are completely ineffective as stimulators. For effective lysis, the target cells need not possess the potential for alloreactivity to responder F1 CTL. We also demonstrate in a preliminary way additional antigen systems defined by killer populations raised with other F1 antiparental strain combinations.

Generation of T cells with lytic specificity for atypical antigens. I. A mitochondrial antigen in the rat

F1 rats primed with normal parental strain lymphocyte populations and restimulated in culture with parental lymphoblasts generate potent cytotoxic T cell responses to unusual antigen systems. Here we describe in the Lewis (L)/DA anti-DA combination an antigen system most likely of mitochondrial origin with the following properties: it is transmitted maternally from DA strain females, inherited in an extra-chromosomal manner, restricted by class I RT1Aa major histocompatibility complex gene products, extinguished on target cells treated with chloramphenicol, and its pattern of expression in different rat strains correlates with restriction fragment-length polymorphisms of mitochondrial DNA. Sequence analysis of the rat ND1 gene indicates that the maternally transferred factor in the rat is not a homologue of the maternally transmitted factor responsible for the mitochondrial antigen in mice. In keeping with its inheritance from DA females, this antigen is present on target cells from (DA female x L male)F1 donors and all other F1 combinations derived from DA female parents, but absent from target cells from some F1 combinations (L/DA and Wistar-Furth [WF]/DA) derived from DA strain males. The presence of this antigen in other F1 combinations (Brown Norway [BN]/DA, August 2880 [AUG]/DA, and PVG/DA) indicates that this mitochondrial antigen system is shared by the DA, BN, and PVG strains, but not by the L and WF strains.

Variants of Mta, the mouse maternally transmitted antigen, determined by three alleles of the extrachromosomal gene Mtf

Two new forms of Mta, the maternally transmitted antigen of the mouse, were defined by cytotoxic T lymphocytes. One is found in laboratory mice carrying the beta allele of the maternally transmitted factor, Mtf, which were previously thought to be Mta-negative. The other is associated with a new allele, Mtf gamma, and is found in wild mice from Toulouse. Like the common form of Mta, determined by the extrachromosomal Mtf alpha, the new forms are also dependent on the a allele of Hmt, a chromosomal gene linked to H-2, and are associated with beta 2-microglobulin. Killers could be raised in all six combinations of alpha, beta, and gamma mice, showing that each form of Mtf determines a distinct (set of) epitope(s). We propose that the product of the Hmt gene, a class I MHC antigen, presents a peptide derived from a mitochondrial gene product on the cell surface and thereby creates the target antigen, Mta.

Molecular basis of mitochondrial autoreactivity in primary biliary cirrhosis

Primary biliary cirrhosis (PBC), styled as a model autoimmune disease, has a very close association with antibodies to mitochondrial antigens. The cloning of cDNAs for mitochondrial antigens led to their identification as the three enzymes of the 2-oxo-acid dehydrogenase family which includes pyruvate dehydrogenase (PDH). The autoantigen is associated with the E2 subunit of these enzymes. PDH is the 70-74 kDa antigen previously identified by immunoblotting. An immunodominant site (autoepitope) can be mapped to a sequence of PDH-E2 which is the site of attachment of the functionally important lipoic acid prosthetic group. Here, Ian Mackay and Eric Gershwin discuss how autoantigen identification and epitope mapping has now proceeded further in PBC than in other autoimmune diseases, and will allow access to critically important questions on pathogenesis of PBC in particular and autoimmunity in general.

Effect of mitochondrial protein synthesis inhibitors on erythroid differentiation of mouse erythroleukemia (Friend) cells

When mouse erythroleukemia (MEL) cells were incubated in the presence of chloramphenicol (a specific inhibitor for mitochondrial protein synthesis) during the early stage of in vitro erythroid differentiation, the number of induced erythroid cells was greatly reduced. By use of cell fusion between two genetically marked MEL cells, this finding was further investigated. We found that the drug, along with other agents which inhibit mitochondrial protein synthesis, blocked the induction and turnover of the DMSO-inducible intracellular-erythroid-inducing activity (differentiation-inducing factor II) in a manner similar to that of cycloheximide, an inhibitor for nuclear protein synthesis. The inhibitory effect was confirmed by directly assaying differentiation-inducing factor II in the cell extracts. These results strongly suggest that mitochondrial protein synthesis is closely associated with in vitro erythroid differentiation of MEL cells.

Effect of mitochondrial protein synthesis inhibitors on erythroid differentiation of mouse erythroleukemia (Friend) cells

When mouse erythroleukemia (MEL) cells were incubated in the presence of chloramphenicol (a specific inhibitor for mitochondrial protein synthesis) during the early stage of in vitro erythroid differentiation, the number of induced erythroid cells was greatly reduced. By use of cell fusion between two genetically marked MEL cells, this finding was further investigated. We found that the drug, along with other agents which inhibit mitochondrial protein synthesis, blocked the induction and turnover of the DMSO-inducible intracellular-erythroid-inducing activity (differentiation-inducing factor II) in a manner similar to that of cycloheximide, an inhibitor for nuclear protein synthesis. The inhibitory effect was confirmed by directly assaying differentiation-inducing factor II in the cell extracts. These results strongly suggest that mitochondrial protein synthesis is closely associated with in vitro erythroid differentiation of MEL cells.

The same MHC recombinational hot spots are active in crossing-over between wild/wild and wild/inbred mouse chromosomes

Four recombinational breakpoints were mapped in the K-A interval of the mouse major histocompatibility complex (MHC) by Southern blot analysis. The breakpoint in B10.SBR, containing a b/s recombinant MHC haplotype, is located about 45 kb upstream of the A beta 2 gene close to the breakpoint in B10.AQR. Crossover in two cas3/cas4 and one cas4/cas3 recombinant haplotypes has taken place in the previously identified K/A beta 3 and A beta 3/A beta 2 recombinational hot spots. The same hot spots are thus active in crossover between two Mus musculus castaneus MHC haplotypes and in crossover between a laboratory and a M. m. castaneus MHC haplotype.

Biogenesis of mitochondria and genetics of mitochondrial defects

Mitochondria are formed by the concerted action of two genetic systems: the nucleocytoplasmic system and the intrinsic mitochondrial system. The genetic contribution of the mitochondria is modest because the genetic potential of mtDNA of mammals is restricted to the equivalent of about 16,000 base pairs. For various animals and man the complete base sequence of mtDNA is known and all possible polypeptide genes have now been assigned to subunits of the respiratory enzymes. The mtDNA sequences are not present on the nuclear genome. From a genetic point of view it is important that the inheritance of mtDNA is strictly maternal. Mutations of mtDNA primarily lead to impairments of energy metabolism. In view of the indispensability of oxidative phosphorylation for obligatory aerobic organisms, such mutations should be lethal. However, there are various inborn errors of metabolism with tissue-specific manifestations, which are maternally inherited. The question discussed is whether these diseases can be explained on the basis of mutations of mitochondrial gene products. Tissue specificity poses a special problem, since it is not very attractive to assume that there is a heterogenous population of mtDNA molecules in the fertilized egg. Therefore, one should rather think in terms of a double mutational event, one tissue-specific cytoplasmic and the other general mitochondrial. These mutations only give rise to metabolic disturbances if they are expressed together in the same cell.