RNA editing and mitochondrial genomic organization in the cryptobiid kinetoplastid protozoan Trypanoplasma borreli.

Abstract

The bodonids and cryptobiids represent an early diverged sister group to the trypanosomatids among the kinetoplastid protozoa. The trypanosome type of uridine insertion-deletion RNA editing was found to occur in the cryptobiid fish parasite Trypanoplasma borreli. A pan-edited ribosomal protein, S12, and a novel 3'-and 5'-edited cytochrome b, in addition to an unedited cytochrome oxidase III gene and an apparently unedited 12S rRNA gene, were found in a 6-kb fragment of the 80-to 90-kb mitochondrial genome. The gene order differs from that in trypanosomatids, as does the organization of putative guide RNA genes; guide RNA-like molecules are transcribed from tandemly repeated 1-kb sequences organized in 200-and 170-kb molecules instead of minicircles. The presence of pan-editing in this lineage is consistent with an ancient evolutionary origin of this process.The evolutionary origin of the uridine (U) insertion-deletion type of RNA editing which occurs in the mitochondrion of the trypanosomatid kinetoplastid protozoa is an intriguing problem. Within the kinetoplastid protozoa there are two suborders, the Trypanosomatina and the Bodonina, as determined by morphological characters and life cycles. The trypanosomatids consist of approximately 8 to 10 obligately parasitic genera in a single family, and the bodonids consist of two major families containing both free-living and parasitic organisms (the Bodonidae and the Cryptobiidae) (30). RNA editing in several trypanosomatids, including the digenetic genera Trypanosoma and Leishmania and the monogenetic genera Crithidia, Blastocrithidia, and Herpetomonas, has been investigated (7,26,28). It has been shown previously that extensive editing or pan-editing, which is mediated by multiple overlapping guide RNAs (gRNAs), represents a primitive character state within the trypanosomatid lineage and that 5' editing and even loss of editing represent derived traits which possibly arose on account of replacement of pan-edited original versions of the genes, with partially or fully edited versions derived from mRNAs (11)(12)(13)(14). Little is known about mitochondrial DNA in the bodonids and cryptobiids, which is also termed kinetoplast DNA since it is present as a large DNAcontaining structure situated within the single mitochondrion (Fig. 1). In the free-living bodonid Bodo caudatus, there is no network of catenated minicircles and maxicircles such as that which exists in the trypanosomatids; instead, a variety of large circular molecules with heterogeneous sizes is found (8). In this paper, we show the occurrence of the uridine insertiondeletion type of RNA editing in the cryptobiid Trypanoplasma borreli, with some novel features, and we show that small gRNA-like transcripts are encoded in tandemly repeated 1-kb

Is kinetoplastid pan-editing the result of an evolutionary balancing act?

“…Present day kinetoplastids, however, show an enormous diversity: maxicircles of up to 200 kb, and gRNA encoding minicircles of 1 -3 kb with only one to several gRNA genes to 200 kb circles possibly encoding hundreds (16). The minicircles clearly evolved later, possibly in response to coding capacity demands of the editing mechanism resulting in an increase of mt DNA size and network evolution to combat minicircle loss (27), although not in all species containing minicircles (28).…”

“…When freshly isolated from the host, L. tarentolae (LEM 125 strain) does show extensive editing of six transcripts (14,15). Extensive editing has further been observed in Trypanosoma cruzi, Trypanosoma species E1-CP and Herpetomonas muscarum and the bodonid T. borreli (16). A comparison of editing patterns and the kinetoplastid evolutionary tree leads to the surprising conclusion that pan-editing is an ancient trait (16 -18).…”

Proteomics and electron microscopic characterization of the unusual mitochondrial ribosome-related 45S complex in Leishmania tarentolae

“…In contrast to mammalian mitochondria, the first AUG or AUA codon at the 5' end of the mRNA is generally not used as the start codon in trypanosomatid mitochondria [4][5][6][7][8][9]. Non-canonical initiation codons may also be used in some cases [7,[10][11][12][13].…”

Mitochondrial minicircles in the free-living bodonid Bodo saltans contain two gRNA gene cassettes and are not found in large networks

“…Sedimentation analysis of B. saltans and C. fasciculata minicircles+ A: Sucrose gradient sedimentation of total B. saltans DNA was performed as described in Materials and Methods; DNA was isolated from each indicated fraction, spotted onto nitrocellulose filters, and hybridized with a B. saltans minicircle probe (pBME40)+ Radioactive signals were quantified with a phosphorimager and are indicated in Arbitrary Units (A+U+) on the vertical axis+ Fraction 1 indicates the bottom fraction of the gradient, top fractions are on the right+ B: Sucrose gradient sedimentation of total C. fasciculata DNA was performed as described in A for B. saltans DNA, with the difference that a C. fasciculata minicircle probe was used for hybridization (pDP312)+ not further analyze the contents of the middle and lower bands, but we infer that they contained the bulk of the nuclear DNA (and residual bacterial DNA) (Simpson, 1979;Maslov & Simpson, 1994)+ In line with the enrichment, EM pictures of the upper-band DNA showed the abundant presence of free 1+4-kb DNA circles, and a number of interlocked dimers and trimers in amounts of 13% and 3+0%, respectively, relative to the number of monomeric circles (see Fig+ 4C)+ Most importantly, even in the material enriched with mt DNA, minicircle networks were conspicuously absent+ This was confirmed by the migration patterns of the minicircles present in upper band DNA on agarose gels (Fig+ 6C, lane 1), which were virtually identical to those obtained with total DNA (Fig+ 6A), including the absence of hybridizing material in the slot+ …”

“…For further identification, each of the hybridizing bands of lane 1 of Figure 6A was isolated from gel and analyzed by EM+ In the lower-molecular-mass region of the gel, we found two bands predominantly consisting of 1+4-kb DNA circles and one band containing 1+4-kb linear molecules+ Digestion of the DNA with restriction enzymes prior to electrophoresis resulted in an increase in the intensity of the linear DNA band, at the expense of the two circular forms (results not shown)+ This clearly identified this band as the linear form of minicircles (indicated by l in Fig+ 6)+ In analogy to the pattern of monomeric C. fasciculata minicircles (Kitchin et al+, 1984; see Fig+ 6A, lane 2) and plasmid DNA (Sambrook et al+, 1989), we infer that the lower and the higher circular DNAs represented closed and nicked circular minicircles, respectively (indicated as c and n in Fig+ 6)+ Depending on the DNA preparation, the circular forms of the B. saltans minicircles together comprised 55-75% of the total amount of minicircles, whereas about 15-30% was found in the linear form+ We were unable to identify by EM the bands of hybridizing material migrating at higher-molecular-mass positions in the gel, most likely because of the low abundance of these molecules+ However, from the position to which they migrated in the gels, we infer that they correspond to the catenanes of two and three , were used in PCR reactions in combination with both BM5 (5-lanes) and BM6 (6-lanes) with total B. saltans DNA+ Products were analyzed on 0+8% agarose gels, of which the ethidium-bromide stains are shown+ The appearance of a product with BM5 (and not with BM6) demonstrates that the corresponding gRNA is encoded by cassette II+ A product generated with BM6 (and not with BM5) indicates that the gRNA in question is encoded by cassette I+ M: 100-nt ladder marker+ ϩ (Positive control): PCR primed with BM25, derived from the ND5 gRNA gene in cassette 1 (Fig+ 1B), and BM6+ Ϫ (Negative control): PCR primed with BM25 and BM5+ Note that the picture is a compilation of several separate experiments+ interlocked minicircles (indicated in Fig+ 6 as c2 and c3, respectively)+ The fact that these forms migrated as doublet bands is most probably related to the fact that different combinations of n and c forms can exist (e+g+, n homodimers have a different mobility than n+c heterodimers)+ Quantification of the hybridization signals of the (putative) dimers and trimers revealed that they represented 5-10% and 1-5%, respectively, of the total amount of circular minicircles+ In view of the fact that the CHEF gels were run in the absence of ethidium bromide, the identification of the different hybridizing bands on Southern blots of these gels was less clearcut+ The assignments given in Figure 6B were inferred from EM analysis and a comparison between the hybridization patterns obtained with blots of CHEF gels and those of standard gels (cf+ Figs+ 6A and 6C)+ Finally, we performed CsCl gradient buoyant density centrifugation of B. saltans DNA in the presence of the AϩT-binding dye Hoechst 33258, using a procedure designed to enrich for (AT-rich) mt DNA of other kinetoplastids (see, e+g+, Simpson, 1979;Maslov & Simpson, 1994)+ After centrifugation, three discrete bands were detected (upper, middle, and lower bands in Fig+ 6C) and DNA was isolated from each of them+ As can be deduced from Figure 6C, in which similar amounts of DNA from the three bands were el...…”