Abstract MOTIVATION: Polymerase chain reaction (PCR)-based RNA fingerprinting is a powerful tool for the isolation of differentially expressed genes in studies of neoplasia, differentiation or development. Arbitrarily primed RNA fingerprinting is capable of targeting coding regions of genes, as opposed to differential display techniques, which target 3' non-coding cDNA. In order to be of general use and to permit a systematic survey of differential gene expression, RNA fingerprinting has to be standardized and a number of highly efficient and selective arbitrary primers must be identified. RESULTS: We have applied a rational approach to generate a representative panel of high-efficiency oligonucleotides for RNA fingerprinting studies, which display marked affinity for coding portions of known genes and, as shown by preliminary results, of novel ones. The choice of oligonucleotides was driven by computer simulations of RNA fingerprinting reverse transcriptase (RT)-PCR experiments, performed on two custom-generated, non-redundant nucleotide databases, each containing the complete collection of deposited human or murine cDNAs. The simulation approach and experimental protocol proposed here permit the efficient isolation of coding cDNA fragments from differentially expressed genes. AVAILABILITY: Freely available on request from the authors. CONTACT: fesce.riccardo@hsr.it
T lymphocytes are defective in cystine uptake and thus require exogenous thiols for activation and function. Here we show that monocyte-derived human dendritic cells (DCs) release cysteine in the extracellular space. Cysteine generation is increased by lipopolysaccharide and tumor necrosis factor α, and by contact with T cells specifically recognizing soluble or alloantigens. These stimuli also induce thioredoxin (TRX) accumulation in DCs. However, only the contact with antigen-specific T cells triggers TRX secretion by the antigen-presenting cells. Fewer extracellular thiols are recovered after DC–T cell interactions when cystine uptake or TRX activity are inhibited. In addition, glutamate (Glu) and anti-TRX-inactivating antibodies inhibit antigen-dependent T lymphocyte proliferation. These findings indicate that, during antigen presentation, DCs uptake cystine and release cysteine and TRX, thus providing a reducing microenvironment that facilitates immune response.
Insufficient folding capacity of the endoplasmic reticulum (ER) activates the unfolded protein response (UPR) to restore homeostasis. Yet, how the UPR achieves ER homeostatic readjustment is poorly investigated, as in most studies the ER stress that is elicited cannot be overcome. Here we show that a proteostatic insult, provoked by persistent expression of the secretory heavy chain of immunoglobulin M (µs), is well-tolerated in HeLa cells. Upon µs expression, its levels temporarily eclipse those of the ER chaperone BiP, leading to acute, full-geared UPR activation. Once BiP is in excess again, the UPR transitions to chronic, submaximal activation, indicating that the UPR senses ER stress in a ratiometric fashion. In this process, the ER expands about three-fold and becomes dominated by BiP. As the UPR is essential for successful ER homeostatic readjustment in the HeLa-µs model, it provides an ideal system for dissecting the intricacies of how the UPR evaluates and alleviates ER stress.
Abstract Cells from the murine B lymphoma 1.29, expressing IgM or IgA of identical idiotype, were found inducible by lipopolysaccharide to differentiate into plasma cells. Within 3 days, differentiating cells lost membrane‐bound immunoglobulin (Ig) and accumu‐ lated large quantities of intracytoplasmic Ig. At day 6 of culture, IgA secretion increased 50‐100‐fold, as determined by enzyme‐linked immunoassay. Proliferation increased for the first days of culture but decreased thereafter; by day 10 very few viable cells were present in lipopolysaccharide‐stimulated cultures. Similar results were obtained by culturing 1.29 cells in the presence of supernatants of certain B cell lines (e.g. BF0.3). The finding of a strict correlation between the inductive activity and presence of contaminating Mycoplasma fermentans suggested that factor(s) released by mycoplasma were responsible for the mitogenic activities. This was further indicated by the findings that: (a) the supernatants of BF0.3 that were rendered free of mycoplasma were not inductive, and (b) a nonactive cell line could be made active by infection with supernatants of BF0.3 cells containing viable micro‐ organisms. Thus, supernatants of mycoplasma‐infected cell lines may act as potent polyclonal activators on both normal and malignant B lymphocytes. The ability to induce membrane Ig on 70Z/3 cells indicates that mycoplasma‐related mitogens are also active on pre‐B cells. The possibility of mycoplasma contamination should thus be carefully excluded when presumptive factors of cloned cell lines are being evalu‐ ated.
Abstract 131 I‐labeled IgD‐like molecules extracted from the membrane of mouse spleen, and lymph node cells and their constitutive chains have been characterized by sodium dodecyl sulfate polyacrylamide slab gel electrophoresis (SDS‐PAGE) followed by radioautography. Unreduced mouse IgD‐like molecules concentrated in two definite bands (IgD 1 and IgD 2 ). IgD, migrated slightly ahead of 131 I‐labeled mouse membrane IgM and displayed a mobility very similar to that of 131 I‐labeled IgD extracted from the membrane of human tonsil cells. IgD 2 migrated faster than both IgD 1 and human membrane IgD (mIgD), but slower than 14 C‐labeled mouse IgG2 b used as marker. The different mobility of IgD 1 and IgD 2 was due to difference in size of their respective H‐chains (δ 1 and δ 2 ), as shown by SDS‐PAGE analysis of the two molecules after reduction. Mol.wt. determination of IgD 1 , IgD 2 and of the corresponding constitutive chains are consistent with an H 2 L 2 structure. IgD 1 and IgD 2 were partially replaced following prolonged dialysis by molecules with a SDS‐PAGE mobility consistent with an HL structure, IgD 3 . In contrast to that found for spleen and lymph node, IgD 1 only was detected on the surface of bone marrow cells. Immunofluorescence experiments have shown that while the majority of mouse lymph node or spleen cells bear IgM and IgD, IgM was found to be the predominant Ig class on the surface of bone marrow cells, although a certain number of IgD‐IgM and a few IgD‐positive cells were found. Experiments where mouse lymphocytes have been treated with pronase have indicated that, in analogy with mlgD of primates, mouse IgD‐like molecules are very labile and can be removed from the cell surface by very mild proteolytic treatment.
IgM are glycoproteins secreted by plasma cells as (mu2L2)5+J or (mu2L2)6 polymers. In most species, mu- and J-chains bear five and one N -glycans, respectively. Here we compare the terminal glycosylation patterns of 4-hydroxy-3-nitrophenylacetyl (NP)-specific IgM secreted by transfectants of the J558L mouse myeloma deficient in the alpha2,6 sialyltransferase [alpha2,6ST(N)] or by a hybridoma expressing this enzyme (B1.8 cells). The absence of alpha2,6-sialylation results in an increased addition of alpha1, 3-galactosyl residues to mu- and J-chain N-glycans. Since alpha1, 3-galactosyltransferase (alpha1,3Gal-T) is similarly expressed in the two cell lines, these results indicate that a competition reaction occurs in vivo between alpha2,6ST(N) and alpha1,3Gal-T. In the alpha2,6ST(N) deficient transfectants, mu-chains lacking the C-subterminal Cys575 residue, which are secreted mainly in the form of mu2L2 monomers, are more efficiently capped by alpha1, 3-galactosyl residues, confirming that polymerization significantly reduces the accessibility of mu-chain glycans to the Golgi processing enzymes involved in the biogenesis of antennary sugars. Functional assays indicate that IgM sialylation affects antigen-binding and complement-dependent hemolysis of haptenated red blood cells.
