Protist. Vol. 150, 33-42, March 1999 © Urban & Fischer Verlag http://www.urbanfischer.de/journals/protist Protist ORIGINAL PAPER Phylogenetic Affinities of Dip/onema within the Euglenozoa as Inferred from the SSU rRNA Gene and Partial COl Protein Sequences Dmitri A. Maslova,1, Shinji Yasuhira b,2, and Larry Simpson b aDepartment of Biology, University of California, Riverside, CA 92521, USA bHoward Hughes Medical Institute, Department of Molecular, Cell and Developmental Biology and Department of Medical Microbiology, Immunology and Molecular Genetics, University of California, Los Angeles, CA 900951662, USA Submitted November 4, 1998; Accepted February 6, 1999 Monitoring Editor: Mitchell L. Sogin In order to shed light on the phylogenetic position of diplonemids within the phylum Euglenozoa, we have sequenced small subunit rRNA (SSU rRNA) genes from Dip/onema (syn. /sonema) papillatum and Dip/onema sp. We have also analyzed a partial sequence of the mitochondrial gene for cytochrome c oxidase subunit I from D. papillatum. With both markers, the maximum likelihood method favored a closer grouping of diplonemids with kinetoplastids, while the parsimony and distance suggested a closer relationship of diplonemids with euglenoids. In each case, the differences between the best tree and the alternative trees were small. The frequency of codon usage in the partial D. papillatum COl was different from both related groups; however, as is the case in kinetoplastids but not in Eug/ena, both the non-canonical UGA codon and the canonical UGG codon were used to encode tryptophan in Dip/onema. Introduction The phylum, Euglenozoa, includes four major groups: euglenoids, kinetoplastids (consisting of bodonids and trypanosomatids), diplonemids (including the genera Diplonema (syn. Isonema) and Rhynchopus) and Postgaardi (incertae sedis) (Cavalier-Smith 1993; Corliss 1994; Simpson 1997). This grouping is supported by a large amount of morphological and ultrastructural data; however, the relationship among these groups is not resolved. In par1 Corresponding author; fax 1-909-787-4286 e-mail maslov@ucrac1.ucr.edu 2 Present address: Research Reactor Institute, Kyoto University, Kumatori-cho, Noda, Sennan-gun, Osaka 590-04, Japan ticular, the phylogenetic affinity of Diplonema is uncertain, with these organisms sharing a variety of features with both euglenoids and kinetoplastids, but clearly differing from either group (Triemer 1992; Triemer and Farmer 1991; Triemer and Ott 1990). Initially it was proposed that Diplonema represents an evolutionary link between bodonids, which were considered ancestral to the entire group, and euglenoids (Kivic and Walne 1984). Subsequent scenarios considered an early divergence of all three lineages from the common ancestor, and in one case Diplonema was proposed to be a close relative of euglenoids (Willey et al. 1988), while in another it represented a separate branch (Triemer 1992; Triemer and Farmer 1991). Although morphological data 1434-4610/99/150/01-33 $ 12.00/0 34 D. A. Maslov, S. Yasuhira, and L. Simpson proved to be insufficient to corroborate or refute any of these hypotheses, no molecular data on diplonemids has been obtained until now. An usual base (J) was recently found in the nuclear DNA of Dip/onema and kinetoplastids, but no search was performed for this base in euglenoids (Van Leeuwen et al. 1998). In the present work, we have attempted to resolve the relationships among the major euglenozoan groups by a molecular phylogenetic analysis of the small subunit (SSU) ribosomal RNA and cytochrome c oxidase subunit I (COl) polypeptide sequences. Results Phylogenetic Position of Diplonema Based on the SSU Data The SSU rRNA gene sequences from D. papillatum and Dip/onema sp. were amplified from the total cell DNA with the 5' and 3' conserved sequence oligonucleotides, S762 and S763, respectively, and aligned with the sequences from kinetoplastids and euglenoids. The alignment contained 17 taxa and 1349 alignable characters including gaps. The kine- COl 56 100 D. papillatum toplastids included representatives of the major trypanosomatid clades (Leishmania tarento/ae, Herpetomonas muscarum, Phytomonas serpens, B/astocrithidia culicis and Trypanosoma bruce/) (Hollar et al. 1998) and four currently available bodonids (Dimastigella trypaniformis, Rhynchobodo sp., Trypanop/asma borreli and Bodo caudatus). Four euglenoid sequences (Eug/ena gracilis, Khawkinea quartana, Lepocinc/is ovata and Peta/omonas cantuscygm) , were retrieved from the GenBank™ database. The alignment contained two slowly evolving outgroup sequences - Physarum po/ycepha/um and Saccharomyces cerevisiae. Analysis of the data with the maximum likelihood method using the PAUP* program produced the tree shown in Figure 1. Three major groups of Euglenozoa are represented by the corresponding monophyletic clades, and the bootstrap support for their monophyly is high. The clade of Dip/onema is shown as a sister-group of kinetoplastids. However this affinity is not supported by the bootstrap analysis. Indeed, the maximum likelihood consensus majority tree (not shown) shows the clade of diplonemids as a sister-group of euglenoids with a 77% bootstrap level. The branching order which united kinetoplas- SSU 79 95 Leishmania tarentolae Figure 1. Maximum likelihood SSU rRNA and COl protein phyPhytomonas serpens 86 logenetic trees of Euglenozoa. 100 L. tarentolae Blastocrithidia culicis Ln likelihood of the SSU tree is 79 9936.69451. Bootstrap values Trypanosoma brucei T. brucei Dimastigella trypaniformis were obtained for 100 pseu79 doreplicates. The sequences E. gracilis Rhynchobodo sp. have the following Genbank™ 100 T. aestivum accession numbers: L. tarentoTrypanoplasma borreli /ae - M84225, H. muscarum <50 Bodo caudatus S. cerevisiae L18872, P. serpens - AF016323, Diplonema papillatum B. culicis - U05679, T. brucei 100 M12676, D. trypaniformis Diplonema sp. 100 X76494, Rhynchobodo sp. Euglena gracilis U67183, T. borreli - L14840, B. caudatus - X53910, D. papillaKhawkinea quarlana tum - AF119811 (this work), Lepocinclis ovata Dip/onema sp. - AF119812 (this ' - - - - - - Petalomonas cantuscygni work), E. gracilis - M12677, K. quartana - U84732, L. ovata ' - - - - - - Physarum polycephalum AF061338, P. cantuscygni Saccharomyces cerevisiae U84731, P. po/ycepha/um X13160, S. cerevisiae - J01353. Ln likelihood of the COl tree is -3572.64 +/- 97.57. Bootstrap values were obtained for 1000 pseudoreplicates. The corresponding Genbank™ accession numbers are: L. tarento/ae - M1 0126, T. brucei - M94286, T. borreli - U11683, D. papillatum - AF119813 (this work), E. gracilis - U49052, T. aestivum - Y00417, S. cerevisiae - M97514. T. borreli 91 Herpetomonas muscarum Phylogeny of Dip/onema tids and diplonemids as shown in Fig. 1 occurred in only 19% of the bootstrap replicates. This value may be an underestimate of the actual support level: in order to ease the computations, the bootstrap analysis, unlike the search for the best tree, was performed assuming that all sites evolve with an equal rate, and this simplified condition favors the affinity of diplonemids with euglenoids. In order to additionally address possible effects of unequal rates of sequence evolution in different lineages (Felsenstein 1978) on the support of the diplonemid-kinetoplastid clade, we also performed a bootstrap analysis after omitting the long branch of P. cantuscygni and the short trypanosomatid branches. Although the support level increased to 47% (not shown), it remained statistically irrelevant. With three monophyletic euglenozoan clades diplonemids (D), kinetoplastids (I<) and euglenoids (E) - and the clade of an outgroup, there can be only three alternative trees. Each of these alternatives was evaluated using the likelihood, parsimony and distance algorithms by comparing trees without topological constraints with trees constructed under the corresponding user-defined constraints. The results are presented in Table 1. The best unconstrained maximum likelihood tree shows diplonemids and kinetoplastids as sister-groups - ((K,D),E) (see also Fig. 1), while the shortest unconstrained parsimony and distance trees contain diplonemids as the closest relatives of euglenoids - (K,(D,E)). However, the trees with enforced alternative topologies - (K,(D,E)) for likelihood and ((K,D),E) for parsimony and distance - are not significantly different from the corresponding unconstrained trees. With each of these methods, the topology which shows 35 diplonemids as an earliest branch of Euglenozoa (D,(K,E)) - was the least favorable. It should also be noticed that using the SSU rRNA sequences of a diplomonad (Giardia /amblia) and a microsporidian (Vairimorpha necatrix) as outgroups substantially increases a bootstrap support for the diplonemid-kinetoplastid clade (up to 90% with some alignments, data not shown). However, such a high level of support may be an artefact associated with a biased nucleotide composition and fast substitution rates observed in the outgroup sequences. Cloning and Analysis of the COl mRNA Sequence In order to obtain additional information for resolving the phylogenetic position of Dip/onema, we analyzed the sequence of the mitochondrial COl protein. The usefulness of this marker is validated by previous work which showed that mitochondria are monophyletic and that nuclear and mitochondrial phylogenies are congruent (Inagaki et al. 1997; Sicheritz-PontEm et al. 1998; Tessier et al. 1997; Viale and Arakaki 1994; Yasuhira and Simpson 1997). Total cell RNA was isolated from D. papillatum. cDNA to polyadenylated mRNA was synthesized by reverse transcription using dT20 N primer, and a partial COl mRNA was amplified by PCR using the oligonucleotides C112 and dT2o N. The conserved sequence-specific oligonucleotide C112 anneals to a site within the central region of the kinetoplastid COl mRNA approximately 700 nt from the 5' end or 800-900 nt from the 3' end. The PCR yielded several Table 1. Parameters of the phylogenetic trees of Euglenozoa with or without specific topological constraints. Tree topology Marker Likelihood 1 Parsimony2 Distance3 ((K, D), E) SSU COl SSU COl SSU COl -9936.69451 -3572.64 -9936.72057 -3575.73 -9938.29060 -3574.26 1912 536 1900 534 1914 537 1.48893 (2 trees) (K, (0, E)) ((K, E), D) 1.48734 (2 trees) 1.52350 (2 trees) Ln likelihood value found by a heuristic search using empirical nucleotide frequencies, assuming two substitution types, and estimating a proportion of invariable sites and a transition/transversion ratio via maximum likelihood. 2 Performed by a branch-and-bound search. The score indicates a number of steps. 3 Minimal evolution score found by a heuristic search with Kimura 3-parameter distances and allowing for among site variation; starting tree obtained through neighbor-joining followed by global rearrangements. The two trees in each case differ only by arrangement of L. tarento/ae, H. muscarum and P. serpens branches. The similar differences in tree scores were also obtained for sums of unweighted least squares. For each optimality criterion, the underlined tree corresponds to the constrained tree which coincides with the best unconstrained tree. 1 3 " (") en ::;- ::!: 0 en C" :T III 0 III III a. III E 3 0 < <: ~ III < en:l ::J. Q~o.CD3:JOIl"'<tN_0enzc,» CDOCDIll ~neDC3 '"~.0-DC '"?~-)"(_T~:C.0:! ~3 ,...., - a.DCl:_C'" CD ~l:Z~DC~3C'"~l:"C-r:~0 __. 0 -"'C ..... :r " 0 UJCD n e ~ :T CD g ~ a. <0' ~ x' + 0 _ III :r _. . _. g:l (") O'» ~ :r - :r c: 0 a. co W C" :l :r c: E:~ C). CD ~ CD"'C:l <: CD OJ III co en Q.. = :r C: lio.~ 7': JO-CD:l~O: JOIl Il Oeno"'C~ - 3- 2. - co ~ CD :r _. c: (") ~ ::J. III en .en :I: 0 a. <: ~ III CD _ III 0 @ CD C" - CD 0 :E :so :r III o::J·:r ~ ~ !~il ~ ; ~I CD~ I~i ~ i~[ CD _..... ~:li g ~ :E i& [~ :Ea»CDo.~ic5 o.~ @ "'C:r ~ (") - ~. . Q.. p 3' 0 CD :E ~ ~ I : ~ i g ! 1 ~ [ . 0 : ; ~ l : 0 - ~ . 0 : ~ 8 t mo ~ P ±~"CD ~8l 1~ J: iP<g ,i~ 8;ne. Qi ~g i~ z ~.Q ~ gE:§~DC< ~nel ~-»w3l 1~_)"(l:_~g- r:3C'"DCl: CO~ - _. - :l <; ~ (") en (") c: :T C" (") (") ::;- III a. W ~ a. en W a. Qj 0 3:T 0 CD 0 "'C III 0 0 III -< _. 7' :E CD CD W CD .... 0 _ _. - ~ 0 en CD ~ ~ 0Q. III '<.... (") (") 0 (") ~ a. coC). ~ c: Q.. 3 3en '< = ... CD a. "'C _ .... III _ .... ::;- CIl @CD ::!: C" ~ - - a. CD 0 CD III '<!:!. '< ~ 0 (") en 0 (") T CD 3 mOo....... 0 """'C.:l:l ~ :r III c: 0 ~ 3:E C" zne3~ ~ CD a. - Z C" - III _CD»Ill'<_ ~ III :l <; ~ ~_ en ~ 0~ CD ~ III co _. 0 - ~ "'C III :T 3 :E 0 Z CD"'C»O~ CD,<o.l III 0en en CD ... SR DC~ 0'" Q. n, ~ CD CD ~ (") CIl co (") ::;- ~ ~ g- en _. - CD en _. :E m .!D 0" - CIl CD (") "'C ()) ~-~ CD» ..c c: C'"~ CD.... :l a. (") :r OCll co C' ~_ :.:=:1. . . C" ~ T III o' S. ~ ~ a. (") CD ~ ::J en - _ en III 0 c: 3 -0 en to en 0 :T III C" (") (") ::;- 5' 0 :r a. - :l CD 0 :r to CD 0 0 III co Z » :CJ ~E: C"c:en:E "'C l 13DC~ III 0 ::J 0 0 Q » - CD - CD en 0 G)::J _. a. CD 3 3 a7':l::J,. ~: JOo.CD-< _:l ~ CD ~ (")0 m ~ ::J@:T ~ »' 8 c5 ~ III . 3 ~ g co = ~ en _ en CD :r (") a. ::J 3 ~ :l - ~ :r . III :r - 0 III III 0.:E co CD CD 0 ~ ::!:!:.. ~ CD (J) ~ e!.:l:::: ::J _ :l WtOCD::D CD :T:E-o:r en~:r~ >< » _. 3 CD <@en-·;::+:"::JCD0..cCDQ.CDCIl CIl" III C:o (") co -' <..... .... 3 CD CD 3 c: CD 0 ~ ~ _. en :r <p CD CD .... c: Q. ~j ~ [~; CDI l ; _. 0 CD CD::J ~ g, 0 - C" Z 0 - III ..c - CD ~g; OJo.~ as. .... 0 CD ~ :E CD C" ~ "E. 5 CD il ~ lIC~ ; ~ : a. ~m 5~ ~O@T:~ e » CD s: e - a. a. ~-<':r(")Il CD 0 a. Z _ - CDO::;-CD ~ :E 0 - r:l " r:~zo _. 0 III III _. _._0 :E - 0 Q CD III CD :!1 CD III _. :r c' ~ _. - '< _. C::l :r CD CD 3 0- :r = co - CD :r <: CD :lI _a. co -< ::J (") en CD a. CD 0_ CIl , I III » III,CD, CD » :l CIl,:l ::J 0.,< :r :r CD ::J CD <; • • _:r _ 3 ::::.: ,a. ~ ~ -i::J ~. -....J g} GUAC~ F V ~ H H M F T V G L D V D T R V Y F S SAT L L _. CIl, en .... ~ I ALP T S I K V F S ~ AUGGUGGGACUGCGUAGGAUGGCACUAGCCAGCAGCGCUGCquGGpACGUCGUGGCCUUCCUCCUCAUGUUCCUCCUAGGUGGUGUCACAGGGCUGGUGCUAGCCAACAGUGAGGUGGAC 361 M V G L R R MAL A S S A A ~ V V A F L L M F L L G G V T G L V LAN S E V D CUCGUAAUGCAUGACAGCUAUUACGUCGUGGCCCACUUCCACUAUGUGCUCUCAUUAGGAGCGGUCUUCGGUCUGCUCAACGGUGUCCUCUCAUGCCACGAGCUGUGCUCUGGGUAUAGG 481 L V M H D S Y Y V V A H F H Y V L S L G A V F G L LNG V L S C H E L C S G Y R AGCU~ S A A ~ D G Y V ~ L R V Q V V L L T V V S T T C GIL T T C I F V L L A V ~G 601 M H L S G T L G L P LEA S L GAUC~ R R V P D A P T Q Q L R 721 L G V ALL L CAS ;x> s:: ~ 0" .< ~ ~ c: ;;; ~. III a. "'C:r - CD 0 241 Y !=' ::J ... GGUCUACGUGAUCAUACUCCCGGUGUUCGGGCUGGUAUCCCAUUCCCUACAUCGAGGAGGACUUUUUUCGCUCUACAACAUGCUGGGCAUGGUGUACGCCAUGAUAGCCAUAGCCGUGGUC 121 ... V Y V I I L P V F G L V S H S L H R G G L F S L Y N M L G M V YAM I A I A V V G en "'C :l ~ Ill(")(") ~g.Q:cl I~mO 3- DC 3~ ::J oe:E a. _. CD 0 III to CD Q. eDC)"(J:III co _. :l CD c: 0_ CIl CD :l 0 CD ~ 0 :l 0 3 Q. III a. (") a. "'C" _ a. CD CD, , 0__ en CD C Q. , :coDC)"(-~ ~ 0 W ~ GCCACCACUAGGAGCACCAACCCUGGGAUGCACAACCACAUGCUCAGUGGCAUGCACAGCCUUGACCACGCUACUCGGGGUCAGCUGAUGCUACAUACAUGUGUGUCUACGUCCCACAGA 841 A T T R S T N P G M H N H M L S G M H S L D HAT R G Q L M L H T C V S T S H R GCAUCCAUGCAUCUGGAGGAGUAGU-po1y(A) A S M H L E E Ter Figure 2. Partial mRNA and deduced protein sequence of the COl from D. papillatum. All tryptophan codons are boxed, UGA codons are additionally shown in bold face and the UGA codon conserved among kinetoplastids is underlined. r- C/l 3' "0 en o ::J Phylogeny of Diplonema 121 2 A 121 2 37 B Hoechst Figure 3. Fractionation of the total cell DNA from D. papillatum by equilibrium density centrifugation in CsCI gradients with ethidium bromide (A) and Hoechst 33258 (8). The gradient-purified DNA from the upper and lower band of each gradient, either undigested (lane 1) or digested with Mspl (lane 2) was analyzed in a 1% agarose gel shown to the left of the corresponding gradient. Results of the restriction digests shown in Figure 4A and 4C indicate that each digest contains a discrete set of nonstoichiometric fragments with the sizes not exceeding several kilobases. Some hexanucleotide recognizing enzymes (8glll, Hindlll, EcoRI and others) apparently do not digest this DNA, while most tetranucleotide cutters do. There is no evidence for the presence of uniform length minicircle-like molecules as those seen in digests of kinetoplast DNA from trypanosomatids. Therefore, this mtDNA is probably composed of several types of circles. The exact size, number of types and topological arrangement of these molecules remain to be investigated. Probing of the mtDNA digests with the cloned COl cDNA revealed hybridization with one or a few restriction fragments or hybridization with undigested molecules, depending on the enzyme used (Fig. 48). Specific restriction fragments producing hybridization signals could be identified in most cases. Cloning and sequence analysis of the mitochondrial genomic COl sequence are in progress. Figure 2 shows the cloned 3' end segment of the COl mRNA along with the deduced protein. The polypeptide sequence is 62% similar (41 % identical) to the corresponding portion of the E. gracilis COl sequence and 65% similar (37% identical) to the T. borreli sequence. There is no similarity at the C-ter- minus. The amplified COl sequence is likely to be derived from a polyadenylated mRNA and there is only a single nucleotide between the UAG termination codon and the poly(A) tail. This arrangement was identical in all three cDNA clones sequenced. However, there is also a possibility that the 3' UTR was indeed longer and contained an A-rich sequence which acted as an annealing site for the dT20 N primer. Future studies should clarify this question. An important feature of the COl sequence is the use of UGA codons as tryptophan instead of termination. The sequence contains seven deduced tryptophan residues, four of which are encoded by standard UGG codons, while three tryptophans, including one conserved residue, are encoded by noncanonical UGA codons. The codon usage is summarized in Table 2. The usage is biased towards codons with G or C in the 3rd position (65.3% of all codons). This reflects the 55.3% G+C content of this fragment and correlates with a relatively high G+C-richness of this mtDNA as mentioned above. Phylogenetic Analysis of COl Sequences The amplified 3' partial sequence of D. papillatum COl and additional 53 residues from the 5' region of 38 D. A. Maslov, S. Yasuhira, and L. Simpson the sequence were aligned with the corresponding sequences from E. gracilis, T. borreli, T. brucei and L. tarento/ae. The yeast and wheat cal sequences were used as outgroups. After elimination of the nonalignable C-terminus and a few short internal regions, the alignment contained 293 characters. As for the SSU genes, analysis of the data with maximum likelihood (PROTML) resulted in a tree which showed Dip/onema as a sister group to kinetoplastids (Fig. 1). However, two alternative topologies were not significantly different (Table 1). Protein parsimony analysis performed with PHYLIP resulted in two equally short trees, one of them containing Dip/onema as a sister lineage to kinetoplastids and another one showing it as a sister lineage to E. gracilis (data not shown). Protein parsimony analysis performed with PAUP slightly favored the second of these two options (Table 1). Protein distance analysis done with PHYLIP using the Kimura distance formula or the Dayhoff PAM matrix for distance computation combined with two different methods of tree construction (least squares and neighbor-joining), showed Dip/onema as the closest relative of kinetoplastids (not shown). 01 N 00 N ,... N ~ II) N • N M N N N u 00 Discussion II I I I C! C! II) M OeD C'i"; I C! ... I "'! o We have analyzed the phylogenetic relationships of Dip/onema with other Euglenozoa using SSU rRNA gene and mitochondrial cal protein sequences. Confirming results of the previous morphological studies (Triemer 1992; Triemer and Farmer 1991), our analysis showed that diplonemids rep- Phylogeny of Dip/onema 39 Table 2. Codon frequency in the partial COl mRNA of D. papillatum. Gly Gly Gly Gly Glu Glu Asp Asp Val Val Val Val Ala Ala Ala Ala GGG GGA GGU GGC GAG GAA GAU GAC GUG GUA GUU GUC GCG GCA GCU GCC 6.00 7.00 8.00 3.00 5.00 0.00 4.00 4.00 16.00 9.00 0.00 0.00 3.00 8.00 3.00 10.00 Arg Arg Ser Ser Lys Lys Asn Asn Met lie lie lie Thr Thr Thr Thr AGG AGA AGU AGC AAG AM AAU AAC AUG AUA AUU AUC ACG ACA ACU ACC 4.00 2.00 3.00 7.00 1.00 0.00 0.00 5.00 14.00 5.00 0.00 3.00 3.00 5.00 3.00 7.00 resent a third lineage within Euglenozoa in addition to kinetoplastids and euglenoids, and this lineage branches deeply within this group. However, molecular phylogeny, as well as the previous morphological studies, fail to provide an unequivocal answer to the question which of these two clades represents a closest relative of diplonemids. Maximum likelihood, a proven powerful and accurate method of phylogenetic reconstruction, favors a closer relationship of diplonemids with kinetoplastids rather than euglenoids with both markers, while parsimony and distance supported the alternative topology. However, differences between the best tree and suboptimal trees are small with each method. One interesting feature of the mitochondrial genetic code of D. papillatum , the use of UGA and UGG as tryptophan codons in cal mRNA is also shared by kinetoplastids (de la Cruz et al. 1984; Simpson et al. 1987). In cal from E. gracilis and the euglenoid, Eutreptiella gymnastica, only the UGG codon is used for this purpose (Inagaki et al. 1997; Tessier et al. 1997; Yasuhira and Simpson 1997). In addition, cal mRNA in E. gracilis does not contain a poly-A tail (Yasuhira and Simpson 1997), while the D. papillatum cal mRNA apparently is polyadenylated, as are mitochondrial messengers in kinetoplastids (Bhat et al. 1992). Although these observations support a closer affinity of diplonemids with kinetoplastids, they are based only on a single mitochondrial gene and need to be verified by analysis of additional genes. Our preliminary analysis of the mitochondrial DNA from D. papillatum indicates that it contains covalently closed molecules. No evidence for a kinetoplast-like network could be seen, although existence of some catenation needs to be investigated. Trp Trp Cys Cys End End Tyr Tyr Leu Leu Phe Phe Ser Ser Ser Ser UGG UGA UGU UGC UAG UAA UAU UAC UUG UUA UUU UUC UCG UCA UCU UCC 4.00 3.00 3.00 3.00 1.00 0.00 3.00 8.00 5.00 2.00 1.00 10.00 1.00 2.00 2.00 10.00 Arg Arg Arg Arg Gin Gin His His Leu Leu Leu Leu Pro Pro Pro Pro CGG CGA CGU CGC CAG CAA CAU CAC CUG CUA CUU CUC CCG CCA CCU CCC 1.00 1.00 4.00 0.00 4.00 0.00 5.00 11.00 15.00 10.00 2.00 13.00 1.00 2.00 3.00 0.00 It is unclear if topologically relaxed molecules also present in our preparations are due to the damage caused by handling or whether they represent replication intermediates. In any case, isolated mtDNA of D. papillatum is quite dissimilar from the heterogeneous linear molecules isolated from mitochondria of E. gracilis (yasuhira and Simpson 1997), and also from trypanosomatid networks (Simpson 1986; Simpson 1987). Although only limited information is available on the molecular organization of mtDNA of bodonids (Blom et al. 1998; Hajduk et al. 1986; Lukes et al. 1998; Yasuhira and Simpson 1996), it seems the non-catenated or weakly catenated structure is more similar to that seen in Dip/onema. Finally, although the data obtained in this work favor a closer association of diplonemids with kinetoplastids than with euglenoids, it is evident that more work is required in order to fully resolve phylogenetic relationships among these organisms. Methods Strains and cultivation conditions: Dip/onema (syn. /sonema) papillatum (ATCC50162) and Dip/onema sp. 2, strain IIIGPC, (ATCC50224) were obtained from the American Type Culture Collection. The strains were cultivated at 27°C in an enriched /sonema medium (ATCC Culture Medium 1728) containing 10% heat-inactivated fetal bovine serum in stationary flasks. For cultures of D. papillatum , stationary phase of growth was achieved at 107 cells/ml on the fourth day after inoculation with a starting cell density of 0.5-1.0 x 106 cells/ml. For Dip/onema sp., only a ten fold lower density of cells could be obtained. 40 D. A. Maslov, S. Yasuhira, and L. Simpson Isolation of total cell DNA and RNA: Cells from stationary cultures were washed with 10 mM TrisHCI (pH 8.0), 150 mM NaCI, 100 mM EDTA, and lysed with 2% Sarcosyl, 0.5 mg/ml pronase at 65 DC for 30 min. The lysate was extracted with buffered (pH 8.0) phenol, phenol-chloroform (1 :1) and chloroform. Nucleic acids were precipitated with an equal volume of isopropanol. The pellet was rinsed with ethanol, dissolved and reprecipitated with ethanol. Total cell RNA was isolated by guanidinium thiocyanate/phenol-chloroform extraction procedure using the RNA Isolation Kit (Stratagene). Isolation of mitochondrial DNA: Total cell DNA isolated from 0.5 x 1010 cells of D. papillatum was fractionated using CsCI-Hoechst 33258 or CsCIethidium bromide equilibrium density centrifugation in a VTi50 rotor at 45,000 rpm for 20 h. Other conditions were as described previously (Maslov and Simpson 1994; Simpson 1979). peR amplification and sequencing of the SSU genes: Small subunit rRNA genes were amplified using the oligonucleotides S762 (GACTITTGCTTCCTCTAWTG) and S763 (CATATGCTTGTTTCAAGGAC), cloned in the vector pT7Blue (Novagen), and sequenced using the Sequenase kit (version 2.0, Amersham) as described previously (Maslov et al. 1996). The first strand of the SSU gene of D. papillatum was sequenced with the following oligonucleotide primers (listed in the order of occurrence): S847: CATATGCTTGTTTCAAGGACTWAGCCATGCATGCC; S1404: CTGAGAACGGCTACCACATC; S713: CCGCGGTAATTCCAGCTCC; S1381: ACGGTCGACCACCGATGTTA; S757: TCAGGGGGGAGTACGTTCGC; S1402: TTGTAGGGGGTGTCTITTGG; S828: CAACAGCAGGTCTGTGATGC; The second strand was sequenced with the primers: S1401: AGCAACGACGGGCGGTGTGT; S829: GCATCACAGACCTGCTGTTG; S1378: CACACAATTCATCGAGAAAG; S714: CGTCAATTTCTTTAAGTTTC; S1382: GAAACTCAAAAGAGAACCGC; S1403: CACCATTACCACCGTTCATA; S1380: TCAGCAGTGTTGCTATTGGG; For the first strand of the Dip/onema sp. SSU gene we used: S847 (see above); S1645: CCCGCAAGAGTATCTGCCCTATC; S1736: GGAATTAGGGTTCGATTCCG; S713; S1644: ~GTCA GTCA S757 and S828. For the second strand of the same gene we used: S829; S714; S1643: GGTTTGGAGCCTTACCTTAAATTAT; S1739: CGGGTITTGATCTTCAACAG; S755: CTACGAACCCTTTAACAGCA; S1646: GATGTGGTAGCCGTTTCTCAGGCT. The SSU rRNA sequences are deposited in the GenBank™ database under the following accession numbers: AF119811 (D. papillatum) and AF119812 (Dip/onema sp.). Other nucleic acid manipulations: The conditions for RT-PCR and TA-cloning were as described previously (Simpson et al. 1996). Briefly, the cDNA from D. papillatum was synthesized using total cell RNA and an oligo-dT primer (dT20 N). The oligonucleotide C112 (TTYTGRTTYTTYGGNCAYCCNGA) (Lukes et al. 1994), which corresponds to a highly conserved region of eukaryotic cal sequences, was used together with the oligonucleotide dT20 N to PCR amplify a partial cal mRNA. The determined partial cal mRNA sequence is deposited in GenBank™ database under the accession number AF119813. Agarose gel electrophoresis, restriction digestion and Southern blotting were performed by standard protocols. Hybridization with a DNA probe labeled by random priming (Stratagene) was performed at 68 DC in 6 x SSC, 5 x Denhardt's solution, 100 mg/ ml salmon sperm DNA and 0.5% SDS. Phylogenetic analysis: Alignments were generated manually using the program Seq Edit, version 3.1 (Olsen 1990). The SSU and cal alignments are available on request or can be downloaded from the following URL: http://www.lifesci.ucla.edu/RNAItrypanosome/alignments.html. Maximum likelihood, parsimony and distance analyses were performed as described previously (Lukes et al. 1997) using PAUP* 4.0 beta version (Swofford 1998) (Sinauer Associates, Inc.) and PHYLIP 3.5 (Felsenstein 1995). In addition, maximum likelihood analysis of the cal polypeptide sequences was performed with PROTML, version 2.2, (Adachi et al. 1992) with the JTT amino acid substitution matrix (Jones et al. 1992). Acknowledgements We would like to thank Dr. J. Lukes for the oligonucleotide C112. This work was supported in part by the Burroughs Wellcome New Investigator Award in Molecular Parasitology to D.M. and the grant AI09102 from National Institutes of Health to L.S. References Adachi J, Hasegawa M (1992) MOLPHY: Programs for molecular phylogenetics I - PROTML: maximum Iikeli- Phylogeny of Dip/onema hood inference of protein phylogeny. Computer science monographs, No. 27. Institute of Statistical Mathematics, Tokyo Bhat GJ, Souza AE, Feagin JE, Stuart K (1992) Transcript-specific developmental regulation of polyadenylation in Trypanosoma brucei mitochondria. Mol Biochem Parasitol52: 231-240 Blom D, De Haan A, Van den Berg M, SI001 P, Jirku M, Lukes J, Benne R (1998) RNA editing in the free-living bodonid Bodo saltans. Nucl Acids Res 26: 1205-1213 Cavalier-Smith T (1993) Kingdom Protozoa and its 18 phyla. Microbiol Rev 57: 953-994 Corliss JO (1994) An interim utilitarian ("user-friendly") hierarchical classification and characterization of the protists. Acta Protozool33: 1-51 de la Cruz V, Neckelmann N, Simpson L (1984) Sequences of six structural genes and several open reading frames in the kinetoplast maxicircle DNA of Leishmania tarentolae. J Bioi Chem 259: 15136-15147 Felsenstein J (1978) Cases in which parsimony or compatibility methods will be positively misleading. SystZooI27:401-410 Felsenstein J (1995) PHYLIP - Phylogeny Inference Package, version 3.5. University of Washington, Seattle. Distributed by the Author Hajduk S, Siqueira A, Vickerman K (1986) Kinetoplast DNA of Bodo caudatus: a noncatenated structure. Mol Cell Bioi 6: 4372-4378 Hollar L, Lukes J, Maslov DA (1998) Monophyly of endosymbiont containing trypanosomatids: Phylogeny versus taxonomy. J Euk Microbiol45: 293-297 Inagaki Y, Hayashi-Ishimaru Y, Ehara R, Igarashi I, Ohama T (1997) Algae or protozoa: Phylogenetic position of euglenophytes and dinoflagellates as inferred from mitochondrial sequences. J Mol Evol 45: 295-300 Jones DT, Taylor WR, Thornton JM (1992) The rapid generation of mutation data matrices from protein sequences. Comput Appl Biosci 8: 275-282 Kivic PA, Walne PL (1984) An evaluation of a possible phylogenetic relationship between the euglenophyta and kinetoplastida. Origins of Life 13: 269-288 Lukes J, Jirku M, Avliyakulov N, Benada 0 (1998) Pankinetoplast DNA structure in a primitive bodonid flagellate, Cryptobia helicis. EMBO J 17: 838-846 Lukes J, Jirku M, Dolezel D, Kral'ov8 I, Hollar L, Maslov DA (1997) Analysis of ribosomal RNA genes suggests that trypanosomes are monophyletic. J Mol EvoI44:521-527 LukesJ, Arts GJ, Van den Burg J, De Haan A, Opperdoes F, SI001 P, Benne R (1994) Novel pattern of editing regions in mitochondrial transcripts of the cryptobiid Trypanoplasma borreli. EMBO J 13: 5086-5098 41 Maslov DA, Simpson L (1994) RNA editing and mitochondrial genomic organization in the cryptobiid kinetoplastid protozoan, Trypanoplasma borreli. Mol Cell Bioi 14:8174-8182 Maslov DA, Lukes J, Jirku M, Simpson L (1996) Phylogeny of trypanosomes as inferred from the small and large subunit rRNAs: implications for the evolution of parasitism in the trypanosomatid protozoa. Mol Biochem Parasitol 75: 197-205 Olsen GJ (1990) Sequence editor and analysis program. University of Illinois, Urbana, computer program distributed by the author Sicheritz-PontEm T, Kurland CG, Andersson SGE (1998) A phylogenetic analysis of the cytochrome band cytochrome c oxidase I genes supports an origin of mitochondria from within the Rickettsiaceae. Biochim Biophys Acta: Bio-Energetics 1365: 545-551 Simpson AGB (1997) The identity and composition of the Euglenozoa. Arch Protistenkd 148: 318-328 Simpson L (1979) Isolation of maxicircle component of kinetoplast DNA from hemoflagellate protozoa. Proc Natl Acad Sci USA 76: 1585-1588 Simpson L (1986) Kinetoplast DNA in trypanosomid flagellates. Int Rev Cytol 99: 119-179 Simpson L (1987) The mitochondrial genome of kinetoplastid protozoa: Genomic organization, transcription, replication and evolution. Annu Rev Microbiol 41: 363-382 Simpson L, Frech GC, Maslov DA (1996) RNA editing in trypanosomatid mitochondria. Methods Enzymol 264:99-121 Simpson L, Neckelmann N, de la Cruz V, Simpson A, Feagin J, Jasmer D, Stuart K (1987) Comparison of the maxicircle (mitochondrial) genomes of Leishmania tarentolae and Trypanosoma brucei at the level of nucleotide sequence. J Bioi Chem 262: 6182-6196 Swofford DL (1998) PAUP* 4.0: Phylogenetic Analysis Using Parsimony (and Other Methods), beta version. Sinauer Associates, Sunderland, Massachusetts Tessier LH, Van der Speck H, Gualberto JM, Grienenberger JM (1997) The cox1 gene from Euglena gracilis: A protist mitochondrial gene without introns and genetic code modifications. Current Genet 31: 208-213 Triemer RE (1992) Ultrastructure of mitosis in Diplonema ambulator Larsen and Patterson (Euglenozoa). Europ J Protistol 28: 398-404 Triemer RE, Farmer MA (1991) An ultrastructural comparison of the mitotic apparatus, feeding apparatus, flagellar apparatus and cytoskeleton in euglenoids and kinetoplastids. Protoplasma 164: 91-104 Triemer RE, Ott DW (1990) Ultrastructure of Diplonema ambulator Larsen & Petersen (Euglenozoa) and its relationship to Isonema. Europ J Protistol 25: 316-320 42 D. A. Maslov, S. Yasuhira, and L. Simpson Van Leeuwen F, Taylor Me, Mondragon A, Moreau H Gibson W, Kieft R, Borst P (1998) ­yh-lsoculg-D~ d;oxymethyluracil is a conserved DNA modification in kinetoplastid protozoans and is abundant in their telomeres. Proc Natl Acad Sci USA 95: 2366-2371 the origins of the euglenoid flagellates GRG Grit Rev Plant Sci 7: 303-340 Viale AM, Arakaki AK (1994) The chaperone connection to the origins of the eukaryotic organelles. FEBS Lett 341: 146-151 Yasuhira S, Simpson L (1997) Phylogenetic affinity of mitochondria of Eug/ena gracilis and kinetoplastids using cytochrome oxidase I and hsp60. J Mol Evol 44: 341-347 Willey RL, Walne PL, Kivic P (1988) Phagotrophy and Yasuhira S, Simpson L (1996) Guide RNAs and guide RNA genes in the cryptobiid kinetoplastid protozoan, Trypanop/asma borre/i. RNA 2: 1153-1160