Sequence, biogenesis, and function of diverse small RNA classes bound to the Piwi family proteins of Tetrahymena thermophila

Abstract

PAZ/PIWI domain (PPD) proteins carrying small RNAs (sRNAs) function in gene and genome regulation. The ciliate Tetrahymena thermophila encodes numerous PPD proteins exclusively of the Piwi clade. We show that the three Tetrahymena Piwi family proteins (Twis) preferentially expressed in growing cells differ in their genetic essentiality and subcellular localization. Affinity purification of all eight distinct Twi proteins revealed unique properties of their bound sRNAs. Deep sequencing of Twi-bound and total sRNAs in strains disrupted for various silencing machinery uncovered an unanticipated diversity of 23- to 24-nt sRNA classes in growing cells, each with distinct genetic requirements for accumulation. Altogether, Twis distinguish sRNAs derived from loci of pseudogene families, three types of DNA repeats, structured RNAs, and EST-supported loci with convergent or paralogous transcripts. Most surprisingly, Twi7 binds complementary strands of unequal length, while Twi10 binds a specific permutation of the guanosine-rich telomeric repeat. These studies greatly expand the structural and functional repertoire of endogenous sRNAs and RNPs.

Figure 1.

Twi protein expression, localization, function, and bound sRNAs. (A) Northern blots to detect each TWI mRNA using total RNA isolated from cells in vegetative growth (Veg) or 9 h after initiating conjugation (Conj). (B) Imaging of live cells that expressed the indicated GFP-Twi fusion protein and were stained with the membrane-permeable dye Hoechst to visualize nuclei. (C) Southern blots to assess locus disruption by the neo2 selectable marker cassette. Restriction enzymes used for genomic DNA digestion are indicated along with the region used for probe (the thin gray line above the wild-type [WT] locus). (D) Silver-stained SDS-PAGE gel showing proteins obtained by affinity purification. Open arrowheads indicate the Twi protein after elution by TEV protease. The ∼30-kDa protein common to all lanes is recombinant TEV protease. Asterisk indicates purification from extract of conjugating cells harvested 6 h after conjugation initiation. (E) RNAs copurified with each Twi resolved by denaturing PAGE and stained with SYBR Gold. (M) Marker lane; (Total) gel-purified 23- to 24-nt sRNAs from strain CU522. Asterisk indicates purification from extract of conjugating cells harvested 4 or 10.5 h after conjugation initiation for Twi1 or Twi11, respectively. (F) End structure of sRNAs examined by β-elimination (Periodate) or 5′-monophosphate-dependent exonuclease treatment (Terminator exo). The control 24-nt RNA oligonucleotide has 2′- and 3′-hydroxyl groups but not a 5′ monophosphate.

Figure 2.

Deep sequencing library characteristics. (A) Proportions of sequence types in each unfiltered library. The number of total sequences from each library is indicated at right. (B) Proportions of 22- to 24-nt sequences in each library that mapped to the genome allowing a single 3′-nucleotide mismatch. Sequences indicated as rDNA were derived from the chromosome encoding the large ribosomal RNA precursor. Sequences indicated as mito-derived were from the mitochondrial genome. Further analysis was carried out on sequences represented by the yellow and red bars. (C) Fraction of distinct sequences (the inventory of unique sequences) calculated separately for genome-matching or definitively 3′-U-extended sRNAs. (D) Nucleotide frequency plots for genome-matching 22- to 24-nt sRNA sequences from each library. Nucleotide frequency plots for sRNAs with a definitively untemplated 3′ U are shown in Supplemental Figure S4.

Figure 3.

Pseudogene-derived sRNA clusters. (A) sRNA density plots for a representative pseudogene locus (IIIB). Density of sRNAs from each Twi-enriched library is shown superimposed on density from the total sRNA library of CU522. Peak height indicates the number of sRNA reads with the same 5′-end position at that genome location. Peaks above the zero axis represent sequences on the top strand and peaks below the zero axis represent sequences on the bottom strand. Annotations for predicted protein-coding genes (TTHERM numbers shown) are indicated at the bottom of the panel; yellow boxes indicate A-rich tracts. (B) Comparisons of sRNA numbers mapping to 13 genome locations of pseudogene loci. Number of sRNAs enriched by each Twi protein (top panel) or present in total sRNA from RDRC subunit knockout strains (bottom panel) was normalized to the corresponding wild-type library by the number of filtered sequences. Loci are indicated by the name of the ORF family (I, II, III, IV, or V) and cluster within that family (A, B, C, D). Note the difference in scale between the top and bottom panels of the figure. (C,D) Northern blot hybridization with an oligonucleotide probe complementary to a sRNA from pseudogene cluster IIIB or IB, as indicated. Below the blots, RNA loading is shown by SYBR Gold staining.

Figure 4.

Repeat-derived sRNA clusters. (A) sRNA density plots at a high-copy repeat locus. Gray-shaded bars represent the degenerate repeat unit. (B) Northern blot hybridization with an oligonucleotide probe. Below the blots, RNA loading is shown by SYBR Gold staining. (C) sRNA density plot at a low-copy repeat locus. Colored bars that are not visible are not hidden; they are not large enough to see on the given scale. (D) Northern blot hybridization with oligonucleotide probes indicated in the illustration at top. Below the blots, RNA loading is shown by SYBR Gold staining. The possible structure of sRNAs bound to Twi7 is also shown. (E) Illustration of a putative telo-sRNA precursor derived from telomere transcription. (F) Northern blot hybridization with an oligonucleotide probe complementary to the G-strand of telomeric repeats. Below the blots, RNA loading is shown by SYBR Gold staining.

Figure 5.

Phased sRNA clusters. (A) sRNA density plots at the phased cluster locus Ph3. A hypothetical sRNA complementary transcript is shown below the plot as well as a predicted secondary structure for the region shaded in gray. Blue shading in the secondary structure depiction indicates the complement of the sRNA in that region. (B) Northern blot hybridization with a sRNA complementary oligonucleotide probe. Below the blots, RNA loading is shown by SYBR Gold staining. (C) Northern blot hybridization with a hexamer-labeled probe covering much of the region shown in A. RPL21 expression is shown as a loading control.

Figure 6.

Twi8-associated sRNA clusters. (A) sRNA density plot at a structured RNA locus that is the most abundant cluster of Twi8 library sRNA. An EST that maps to the locus is indicated as well as predicted stem–loop structures for both sense strand and antisense strand transcripts. The green shading in stem–loop structures corresponds to the sRNAs numbered in the density plot. (B) Northern blot hybridization with an oligonucleotide probe complementary to the most abundant sRNA in A. Below the blots, RNA loading is shown by SYBR Gold staining. (C) sRNA density plots for a representative locus of Twi8-enriched sRNAs with convergently transcribed ORFs and/or ESTs that do not map uniquely. (D) sRNA density plots at TWI2-6. Only uniquely mapping sRNA sequences are included; see Supplemental Figure S6 for mapping of all sRNAs. Colored bars that are not visible are not large enough to see on the given scale. (E) Regulation of transgene-encoded Twi2 by the endogenous TWI2 locus. The presence of the transgene is indicated as mztTwi2. The top panels are mRNA Northern blots for expression of TWI2 and the RPL21 loading control; note that endogenous and transgene mRNAs are equally detected by the probe but differ slightly in size, as indicated. The bottom panel is an immunoblot detecting tagged Twi2; equal loading was confirmed by total protein staining (not shown).

Figure 7.

Summary of library representation and sRNA classes. (A) Representation of abundant and preferentially enriched sRNA classes. Only annotated classes with large enough representation to generate a visible pie slice in at least one library are included; these classes are listed in the legend at right. (B) Summary of sRNA classes, characteristics, and accumulation requirements.