Schistosome satellite DNA encodes active hammerhead ribozymes

Abstract

Using a computer program designed to search for RNA structural motifs in sequence databases, we have found a hammerhead ribozyme domain encoded in the Smalpha repetitive DNA of Schistosoma mansoni. Transcripts of these repeats are expressed as long multimeric precursor RNAs that cleave in vitro and in vivo into unit-length fragments. This RNA domain is able to engage in both cis and trans cleavage typical of the hammerhead ribozyme. Further computer analysis of S. mansoni DNA identified a potential trans cleavage site in the gene coding for a synaptobrevin-like protein, and RNA transcribed from this gene was efficiently cleaved by the Smalpha ribozyme in vitro. Similar families of repeats containing the hammerhead domain were found in the closely related Schistosoma haematobium and Schistosomatium douthitti species but were not present in Schistosoma japonicum or Heterobilharzia americana, suggesting that the hammerhead domain was not acquired from a common schistosome ancestor.

FIG. 1

Searching the GenBank database for hammerhead ribozyme RNA domains. (Top) The secondary structure of the hammerhead domain with the conserved nucleotides indicated. The part for which the descriptor was written is shown in boldface. (Bottom) The descriptor used in the program RNAMOT (23) was composed of the following features: s1 H1 s2 H1 s3. The s1 feature scans for a 12-nucleotide (nt) sequence of the form NNNNNCUGANGA, the first 5 nucleotides of which are represented by N (any nucleotide) and which includes the required sequence CUGANGA, which is part of the conserved catalytic core. The feature H1, s2, H1 corresponds to the helix II region closed by a loop of 4 to 10 nucleotides (s2). The numbers 4:4 refer to a helix of 4 bp having zero mismatches but allowing the wobble GU. In this helix, a GC pair is required at the base. The s3 feature requires an 8-nucleotide sequence of the form GAAASNNN, where S represents C or G. This feature contains the remaining part of the conserved catalytic core and the variable nucleotides of the 3′ recognition helix. In order to catalyze a cleavage reaction, RNA defined by this descriptor must recognize a substrate RNA by base pairing. The sequence of a possible substrate is represented, and the arrow indicates the cleavage site. The numbering system is that of Hertel et al. (19).

FIG. 2

A family of repetitive sequences coding for self-cleaving transcripts in S. mansoni. (A) General organization of Smα repeats into three regions: the 5′ region, which has similarity to tRNA and contains boxes A and B, required for transcription by RNA polymerase III; the middle region, encompassing the hammerhead domain (HH); and the 3′ region. Primers used for the experiments reported here are indicated as 1A, 2A, 1B, and 2B, and their positions on the map correspond to the parts of the sequence to which they are complementary. (B) Alignment of the tRNA and hammerhead (HH) ribozyme domains of 18 different Smα clones. The tRNA sequence is that of the serine 3 tRNA from the rat (accession no. K00371). In the tRNA domain, box A and B nucleotides are indicated in boldface. In the hammerhead domain, nucleotides essential for catalysis are indicated in boldface. (C) Summary of the hammerhead domain sequences found in 18 different Smα clones. The cleavage site is denoted by cs. Substitutions found in different isolates are denoted by arrows directed out of the structure, insertions are denoted by arrows directed toward the structure, and deletions are denoted by Δ.

FIG. 3

In vitro self-cleavage of the transcripts derived from Smα repeats. (A) Self-cleavage during in vitro transcription at 37°C of different Smα-derived templates. Lane 1, transcription products of the Sm1 template. Lane 2, product after shortening of the Sm1 template by the restriction endonuclease ClaI. Lane 3, product after shortening of the Sm1 template by restriction endonuclease NdeI. Lane 4, pattern obtained from the Sm3 template carrying a G5➛C base substitution. In vitro transcription was less efficient on the templates treated with restriction endonucleases. Numbers at right indicate sizes in base pairs. (B) Kinetics of self-cleavage at 30°C in 10 mM magnesium and at pH 8. Gel-purified full-length transcripts were incubated under the cleavage conditions described in Materials and Methods, and the reaction was stopped at the indicated time. The products were resolved on a 6% acrylamide–8 M urea gel. Numbers at right indicate size in base pairs. (C) The intensities of the bands in panel B were measured by densitometry, normalized to the background of degradation, and graphed on a semilog plot to calculate the rate of the reaction. S, concentration of substrate at time t; So, initial substrate concentration.

FIG. 4

Expression of the Smα family of repetitive DNA in vivo. (A) Northern blot analysis with total RNA from schistosomula (lane 1), adult males (lane 2), adult females (lane 3), and in vitro-transcribed Smα1 repeat (lane 4). The quantity of RNA used in each analysis was judged equivalent by ethidium bromide staining of the gel and by comparison of the intensity of the rRNA bands in each preparation. (B) Reverse transcription-PCR analysis using primers 2A and 2B from Fig. 2A. Total DNA-free RNAs from adult males (lane 1) or adult females (lane 2) were reverse transcribed with primer 2B. The cDNA was amplified by Vent DNA polymerase (see Materials and Methods). In lane 3, a mixture of both female and male RNA preparations was subjected to the same treatments as were the preparations in lanes 1 and 2, except that reverse transcriptase was omitted. Lane 4 is the 123-bp ladder from Gibco. (C) Primer extension analysis of in vivo transcripts from the Smα family of satellite DNA. In vitro-transcribed Sm1 repeat (lane 1) and total RNAs from adult males (lane 2) and adult females (lane 3) were annealed with primer 1B and reverse transcribed with Superscript Kit II from Gibco at 42°C. The products were resolved on a 6% acrylamide–8 M urea gel. Sequencing reactions performed with primer 1B on the Sm1 template were run in the same gel. Numbers to the right of panel A and the left of panels B and C indicate size in base pairs. The arrow on the right indicates the positions of the expected cleavage products.

FIG. 5

The Smα self-cleavage products as a trans-acting ribozyme. (A) A model of trans-acting ribozyme catalysis in the I/II format. The catalytic fragment is produced from self-cleavage of an active Smα repeat, which is shown in boldface. The substrate is a transcript of the Smα family with a disabled hammerhead ribozyme or the precursor mRNA of the synaptobrevin-like protein in the I/II format. The I/III format is shown in Fig. 1. (B) Cleavage kinetics of the trans hammerhead reaction with synaptobrevin-like protein precursor mRNA, obtained by in vitro transcription from a cloned template. Different concentrations of gel-purified 5′ and 3′ cleavage products from the self-cleavage reaction of Smα 1 transcripts (see Materials and Methods) were incubated in the cleavage conditions described in Materials and Methods with in vitro-transcribed synaptobrevin-like protein RNA for 4 h at 37°C. The products were resolved on a 6% acrylamide–8 M urea gel. Numbers at right indicate size in base pairs. (C) The intensity of the bands in panel B was measured by densitometry, normalized to the background of degradation, and graphed on a semilog plot to calculate the rate of the reaction. S/t, ratio of substrate concentration per time unit.

FIG. 6

Summary of sequences corresponding to the hammerhead domains found in S. haematobium (A) and Schistosomatium douthitti (B). Substitutions found in different isolates are denoted by arrows directed out of the structure; insertions are indicated by arrows directed toward the structure. CS, cleavage site.

FIG. 7

A model for the propagation and function of the α satellite DNA in schistosomes. Tandem repeats or monomeric α sequences are transcribed by RNA polymerase III. The long transcripts are processed by two mechanisms: (i) intrarepeat cleavage and (ii) interrepeat cleavage. The products of self-cleavage reactions then act in trans on other multimeric transcripts of the α family or in transcripts such as the one coding for the synaptobrevin-like protein gene or the OZ.A retroposon. Reintegration in the genome of reverse transcripts from cleaved repeats creates dead ends in the transposition process because these sequences possess the polymerase III promoter at their 3′ end. However, reintegration of nonprocessed multimeric transcripts creates new sites from which further transcription of Smα repetitive elements can occur.