Genomic characterization of variable surface antigens reveals a telomere position effect as a prerequisite for RNA interference-mediated silencing in Paramecium tetraurelia.

UNLABELLED: Antigenic or phenotypic variation is a widespread phenomenon of expression of variable surface protein coats on eukaryotic microbes. To clarify the mechanism behind mutually exclusive gene expression, we characterized the genetic properties of the surface antigen multigene family in the ciliate Paramecium tetraurelia and the epigenetic factors controlling expression and silencing. Genome analysis indicated that the multigene family consists of intrachromosomal and subtelomeric genes; both classes apparently derive from different gene duplication events: whole-genome and intrachromosomal duplication. Expression analysis provides evidence for telomere position effects, because only subtelomeric genes follow mutually exclusive transcription. Microarray analysis of cultures deficient in Rdr3, an RNA-dependent RNA polymerase, in comparison to serotype-pure wild-type cultures, shows cotranscription of a subset of subtelomeric genes, indicating that the telomere position effect is due to a selective occurrence of Rdr3-mediated silencing in subtelomeric regions. We present a model of surface antigen evolution by intrachromosomal gene duplication involving the maintenance of positive selection of structurally relevant regions. Further analysis of chromosome heterogeneity shows that alternative telomere addition regions clearly affect transcription of closely related genes. Consequently, chromosome fragmentation appears to be of crucial importance for surface antigen expression and evolution. Our data suggest that RNAi-mediated control of this genetic network by trans-acting RNAs allows rapid epigenetic adaptation by phenotypic variation in combination with long-term genetic adaptation by Darwinian evolution of antigen genes. IMPORTANCE: Alternating surface protein structures have been described for almost all eukaryotic microbes, and a broad variety of functions have been described, such as virulence factors, adhesion molecules, and molecular camouflage. Mechanisms controlling gene expression of variable surface proteins therefore represent a powerful tool for rapid phenotypic variation across kingdoms in pathogenic as well as free-living eukaryotic microbes. However, the epigenetic mechanisms controlling synchronous expression and silencing of individual genes are hardly understood. Using the ciliate Paramecium tetraurelia as a (epi)genetic model, we showed that a subtelomeric gene position effect is associated with the selective occurrence of RNAi-mediated silencing of silent surface protein genes, suggesting small interfering RNA (siRNA)-mediated epigenetic cross talks between silent and active surface antigen genes. Our integrated genomic and molecular approach discloses the correlation between gene position effects and siRNA-mediated trans-silencing, thus providing two new parameters for regulation of mutually exclusive gene expression and the genomic organization of variant gene families.

Selective and programmed cleavage of GPI-anchored proteins from the surface membrane by phospholipase C.

Many surface proteins of eukaryotic cells are tethered to the membrane by a GPI-anchor which is enzymatically cleavable. Here, we investigate cleavage and release of different GPI-proteins by phospholipase C from the outer membrane of the ciliate Paramecium tetraurelia. Our data indicate that different GPI-proteins are not equally cleaved as proteins of the surface antigen family are preferentially released in vitro compared to several smaller GPI-proteins. Likewise, the analysis of culture medium indicates exclusive in vivo release of surface antigens by two phospholipase C isoforms (PLC2 and PLC6). This suggests that phospholipase C shows affinity for select groups of GPI-anchored proteins. Our data also reveal an up-regulation of PLC isoforms in GPI-anchored protein cleavage during antigenic switching. As a consequence, silencing of these PLCs leads to a drastic decrease of antigen concentration in the medium. These results suggest a higher order of GPI-regulation by phospholipase C as cleavage occurs programmed and specific for single GPI-proteins instead of an unspecific shedding of the entire surface membrane GPI-content.

Inefficient serotype knock down leads to stable coexistence of different surface antigens on the outer membrane in Paramecium tetraurelia.

Expression of surface antigens is usually mutually exclusive, meaning that only one protein is present on the cell surface. With the RNAi feeding technology we induce serotype shifts in Paramecium tetraurelia which are demonstrated to be incomplete, meaning that the cells remain in a shifting state. The coexpression of "old" and "new" protein on the surface can be detected to be stable for more than 15 divisions over a 5-day feeding procedure, a time period different from that reported for temperature-induced shifts. A characteristic heterogenic distribution of the different surface antigens is demonstrated by double indirect-immunofluorescent-staining and we show antigen transport mechanisms related to the tips of cilia. Therefore, we discuss release mechanisms, potential sorting mechanisms for glycosylphosphatidylinositol-anchored proteins and the localizations of surface antigens, which are important for the reported classical immobilization reaction.

Analysis of Paramecium tetraurelia A-51 surface antigen gene mutants reveals positive-feedback mechanisms for maintenance of expression and temperature-induced activation.

In Paramecium tetraurelia, variable surface antigen loci show mutually exclusive expression which is controlled primarily at the transcriptional level. Clonally stable expression of a single antigen has attracted models involving self-regulation by their gene products. However, direct demonstration of self-feedback at the molecular level has been complicated due to the inability to separate the functional gene from its product as well as copy number effects associated with injected extrachromosomal DNA in the polygenomic somatic nucleus. In this study, we exploited several germ line termination and frameshift mutations in the A-51 surface antigen gene to analyze variable surface antigen expression. These mutant alleles have the same copy number as the wild-type allele and therefore eliminate possible copy number effects. The mutant alleles were not transcribed at 27 degrees C, consistent with positive-feedback models for gene expression. However, further analysis showed that high temperatures (34 degrees C) induced transcription of the mutant A genes even in the presence of a different antigen on the cell surface. Thus, transcription was temperature dependent. Unlike wild-type cells, transcription of the mutant A genes at high temperatures was not maintained after temperature shift back to 27 degrees C in homozygous mutant cells. Importantly, transcription of the mutant allele was maintained at 27 degrees C in heterozygous cells with one copy of the wild-type allele. These results indicate that expression of the wild-type gene is required to stabilize its own transcriptional state at 27 degrees C.

A protective merozoite protein of Plasmodium falciparum shares an epitope with surface antigens of Paramecium.

A Plasmodium falciparum cDNA expression clone, lambdaPf9, had been identified earlier as a protective epitope, using anti-lambdaPf9 antibodies and combinatorial phagotopes. A segment of the Pf9 gene showed homology with Paramecium immobilization surface antigens such as 51B, 51A and 156G. A synthetic Pf9-peptide was designed from this region, and specific antibodies were raised. Each of these anti-Pf9 antibodies and combinatorial reagents, as well as anti-Paramecium 51B antibodies, recognized the Pf9-peptide on ELISA, and the same protein band in parasite immunoblots. The P. falciparum protein was released from the merozoite membrane fraction on treatment with PI-PLC, indicating the presence of a GPI anchor. Anti-Pf9-peptide antibodies specifically inhibited the growth of P. falciparum in culture. Immunofluorescence assays showed the reactivity of anti-Pf9-peptide sera with P. falciparum merozoites and gametocytes, as well as on the surface of Paramecium tetraurelia. The Pf9-peptide was able to induce proliferation of splenic lymphocytes obtained from mice infected with the rodent malarial parasites Plasmodium berghei and Plasmodium yoelii. These results point towards Plasmodium Pf9 as a conserved novel protective protein, sharing an epitope with Paramecium surface antigens.

Analysis of the conserved cysteine periodicity of Paramecium variable surface antigens.

The major surface antigens expressed by free-living and parasitic protozoa commonly contain repeating cysteine motifs. Despite the common occurrence of these repeats their functional significance remains largely unexplored. In this paper we investigate the conserved cysteine repeats within the variable surface antigens of Paramecium tetraurelia. We show that deletion of 2 entire repeating units or portions of repeats near the N-terminus does not prevent expression of the A51 variable surface antigen. Alteration of a single cysteine to serine residue also has no effect on A51 expression. In contrast, deletions near the C-terminus of the protein have identified a small segment within the repeats that is required for expression on the surface. The required region contains a number of conserved amino acid residues, yet site-directed mutagenesis of two residues (serine and threonine to alanine) did not prevent expression. These studies demonstrate the feasibility of using deletion analysis to identify regions critical for the expression of cysteine-rich surface antigens. The relationship of these results to the structure and expression of cysteine-rich surface proteins in other protozoa is discussed.

Evidence for transcriptional self-regulation of variable surface antigens in Paramecium tetraurelia.

Variable surface antigens are commonly found on free-living and parasitic protozoa, yet the regulation of antigen expression and switching is not fully understood in any system. A cell line of Paramecium tetraurelia stock 51 can express at least 11 different antigens yet only one type is found on the surface at any one time. Previous studies have shown that mutually exclusive expression of Paramecium surface antigens can be overcome if two antigen genes contain the same 5' coding region. In this article we utilize a gene chimera containing portions of A51 and B51 to analyze the effect of a frameshift mutation on transcription and steady-state mRNA levels. We show that a frameshift mutation near the 3' end prevents expression of the protein on the cell surface and reduces the rate of transcription of the corresponding gene. The difference in transcription is not the result of differences in plasmid copy number. We propose that expression of the antigen on the cell surface is part of a self-regulatory pathway that influences transcription of the corresponding gene. A model incorporating the previous and current data is presented.

Molecular characterization of the D surface protein gene subfamily in Paramecium tetraurelia.

When Paramecium tetraurelia expresses the D serotype, detectable by serum tests, high molecular mRNA could be isolated, which corresponds to the molecular mass of the D surface protein. Using this D specific mRNA as a probe for screenings in different genomic libraries a subfamily of five very similar genes was found, named alpha-51D, gamma 1-51D, gamma 2-51D, delta-51D, and epsilon-51D. Each of them is about 8-kb long, they show regions of identity to each other, and there is no evidence that any are defective genes or pseudogenes. Up to now serotype D is the only known serotype showing this phenomenon. Another novel feature is that two of the D isogenes are closely linked. The sequence for the entire coding region of the alpha-51D gene has been determined, as well as the upstream and downstream noncoding regions. Its deduced amino acid sequence shows the same characteristic cysteine periodicity displayed by all other immobilization antigen (i-ag) genes from Paramecium. However, in contrast to most other such genes, tandem repeats are missing from the 7599-bp long coding region of the alpha-51D gene. When the sequences of the type 51D genes are compared to each other, the similarity is very high and extends to coding as well as to noncoding regions. Similarity within noncoding regions is usually only observed for allelic i-ag genes. We conclude that the type D genes constitute a family of isogenes that are nonallelic. They contain slightly different consensus sequences with possible functions as regulatory regions.

The upstream region is required but not sufficient to control mutually exclusive expression of Paramecium surface antigen genes.

Paramecium tetraurelia stock 51 can express at least 11 different surface antigens, yet only one type is found on the surface of a cell at any given time. The mechanism that controls this mutually exclusive expression is unknown. A previous study has shown that the 51A surface antigen gene is regulated at the level of transcription (Gilley, D., Rudman, B. M., Preer, J. R., Jr., and Polisky, B. (1990) Mol. Cell. Biol. 10, 1538-1544). We show here that the 51B surface antigen gene is also transcriptionally regulated, and when the 51A and 51B genes are cotransformed into an A-, B- mutant, the 51A antigen is dominant at 27 degrees C just as in wild type cells. We have utilized this cotransformation system to experimentally determine that 273 base pairs of DNA upstream of the 51A gene is sufficient to allow the dominant expression of A, but 150 base pairs is not adequate. A hybrid gene that contains the upstream region of 51B attached to the 51A transcribed region was cotransformed with the 51B gene into the A-, B- mutant. Despite containing the same upstream sequences, only the hybrid 51B/A was transcribed at 27 degrees C. These results suggest the upstream region is required but not sufficient for mutually exclusive transcriptional control.

The dynein genes of Paramecium tetraurelia. Sequences adjacent to the catalytic P-loop identify cytoplasmic and axonemal heavy chain isoforms.

Paramecium tetraurelia is a unicellular organism that utilizes both axonemal and cytoplasmic dyneins. The highly conserved region containing the catalytic P-loop of the dynein heavy chain was amplified by RNA-directed polymerase chain reaction. Eight different P-loop-containing cDNA fragments were cloned. Southern hybridization analysis indicated that each fragment corresponds to a separate dynein gene and that there are at least 12 dynein heavy chain genes expressed in Paramecium. Seven of the eight cloned contain sequence motif A, which is found in axonemal dyneins, and one contains sequence motif B, which is found in the dyneins from cell types that do not have cilia or flagella. Two of the Paramecium dynein genes were further investigated: DHC-6 which contains motif A, and DHC-8 which contains motif B. Additional sequencing of the central portions of these genes showed that DHC-6 most closely matches sea urchin ciliary beta heavy chain and DHC-8 is similar to the cytoplasmic dynein from Dictyostelium. Deciliation of the cells resulted in a substantial increase in the steady state concentration of DHC-6 mRNA but only a small change in DHC-8 mRNA. Antisera were produced against synthetic peptides derived from sequence motifs A and B. Competitive solid-phase binding assays demonstrated that each antiserum was peptide-specific. In western blots, the antiserum to motif A reacted with both ciliary and cytoplasmic dyneins. In contrast, the antiserum to motif B reacted with the cytoplasmic dyneins of Paramecium and bovine brain but did not react with ciliary dynein.(ABSTRACT TRUNCATED AT 250 WORDS)

Ultrastructural and antigenic preservation of a delicate structure by cryopreparation: identification and immunogold localization during biogenesis of a secretory component (membrane-matrix connection) in Paramecium trichocysts.

Ultrastructure and antigenicity of the "mesh-like sheath" (MLS), a very delicate structure connecting the membrane and the paracrystalline matrix of Paramecium trichocysts, are well preserved after cryofixation (rapid freezing followed by freeze-substitution in methanol and embedding in Lowicryl K11M at 213K). The MLS is labeled by colloidal gold-bound antibodies (Ab-Au10nm) with primary antibody (Ab) against trichocyst components obtained by recloning hybridoma cells twice. We prepared Western blots from reduced gels obtained from subfractionated trichocysts. Trichocyst membranes displayed reactive bands of 68-70, 63-66, 43, 40 (strongest), and 57 and 54 KD, with a weak band of 38 KD. One of the most abundant protein bands of soluble secretory components (56-57 KD) was also strongly stained on blots. On ultra-thin sections pre-trichocysts display Ab-reactive material concentrated below the trichocyst membrane before the MLS can be recognized as a structural entity. Quantitative evaluation of Ab-Au10-labeled ultrathin sections also revealed passage of MLS materials through the very inconspicuous Golgi apparatus. This was substantiated by Ab-peroxidase labeling. We conclude that MLS components (whose ultrastructure is difficult to preserve) are largely membrane-associated, partly soluble proteins. They form a connection (released during exocytosis) between the abundant paracrystalline matrix components and the organelle membrane. MLS might thus maintain a peripheral aqueous space of functional importance.

Molecular and genetic analyses of the B type surface protein gene from Paramecium tetraurelia.

The gene encoding the B type variable surface protein from Paramecium tetraurelia stock 51 has been cloned and sequenced. The 7,182 nucleotide open reading frame contains no introns and encodes a cysteine-rich protein that has a periodic structure including three nearly perfect tandem repeats in the central region. Interestingly, the B gene is located near a macronuclear telomere as was shown previously for two other paramecium surface protein genes. In this paper, we characterize four independent mutants with complete macronuclear deletions of the B gene. Previous analysis of different macronuclear deletion mutants of the A surface protein gene demonstrated two types of inheritance: typical Mendelian segregation (as illustrated by d12) and cytoplasmic inheritance (shown by d48). F1 analysis of four B- mutants crossed with wild-type cells reveals heterozygous F1 cell lines derived from both parental cytoplasms contain approximately the same copy number of the B gene, as expected for a recessive Mendelian mutation. Analysis of F2 progeny from three of these four B- mutant crosses indicates that one of the three exhibits a Mendelian 1:1 segregation ratio of B+ and B- cell lines. The other two show a preponderance of B+ cells, but this is not correlated with the parental cytoplasmic type. In addition to having a large number of B+ individuals, the d12.144, A-, B- mutant produced some F2 progeny that stably maintain less than normal macronuclear amounts of the A gene and/or the B gene.