CC BY
48
Protistology 2 (3), 185-188 (2002)
Protistology
Discrepancy between morphological and molecular biological characters in a strain of Hartmannella vermiformis Page 1967 (Lobosea, Gymnamoebia)
Julia Walochnik1, Rolf Michel2 and Horst Aspock1
1 Dept. of Med. Parasitology, Clin. Inst. of Hygiene, University of Vienna, Vienna, Austria
2 Central Institute of the Federal Armed Forces Medical Service Koblenz, Labgroup Microbiology (Parasitology), Koblenz, Germany
Summary
Hartmannella strain C3/8, isolated from the sediments of a water reservoir in Germany, has been affiliated to H. vermiformis as all light-microscopical characters of the trophozoites as well as their ultrastructure and locomotion behavior correspond completely to the original isolates of H. vermiformis. However, in contrast to the type strain of this species, the cysts of the strain C3/8 exhibit a regular spatial separation of the cyst wall into the endocyst and a wavy ectocyst. As cyst structure is considered to be a stable taxonomic character among Gymnamoebia, we sequenced the 18S rRNA gene of this particular Hartmannella strain and compared it to four strains of H. vermiformis. It was shown that the sequence dissimilarity of the Hartmannella strain C3/8 to several strains of H. vermiformis is smaller than the sequence dissimilarities between different strains of the species with identical cyst morphology. Moreover, it was shown that in the genus Hartmannella there seems to be no correlation between the geographical origin of the strains and their relatedness. Altogether the results of our study suggest the existence of polymorphism within the species H. vermiformis as it has already been demonstrated for various other Gymnamoebia species.
Key words: Hartmannella, identification, 18S rDNA, Gymnamoebia
Introduction
The Hartmannella strain C3/8 was isolated from the sediments of a water reservoir in Germany and was subsequently affiliated to H. vermiformis by morphological features obtained by light- and electron microscopy, as all light-microscopical characters of the trophozoites as well as their ultrastructure and locomotion behaviour correspond completely to the original isolates of H. vermiformis. However, in contrast to the type strain of this species the cysts of the strain C3/8 exhibit a regular spatial separation of the cyst wall into
the endocyst and a wavy ectocyst (Smirnov and Michel, 1999).
Page (1967, 1986) described the species Hartmannella vermiformis particularly also on the basis of typical double-walled cysts, the ecto- and the endocyst being either closely apposed or partially separated. Cyst structure is considered to be an important and stable taxonomic character among Gymnamoebia.
A reliable identification and classification of strains of Hartmannella is highly desirable, not only because evidence for its association with human diseases has
© 2002 by Russia, Protistology
been increasing in the past years. Free-living amoebae of the genus Hartmannella are of substantial medical relevance as potential vectors of nosocomial legionellosis (Fields et al., 1990) and have recently also been associated with keratitis in contact lens wearers (Aitken et al., 1996, Inoue et al., 1998). However, it is still unclear whether Hartmannella species are indeed potential causative agents of human keratitis or only harmless contaminants of the otherwise infected eye (De Jonckheere and Brown, 1998, 1999).
As many problems related to the identification and classification of protozoan organisms may be addressed by comparing DNA sequences, we sequenced the 18S rRNA gene of this particular Hartmannella strain and compared it to four strains of H. vermiformis. At present combining data obtained from electron microscopy and molecular biology is the most promising approach to an acceptable classification and a convincing system for the protozoa that might reflect the phylogeny of these organisms (Corliss, 2001).
Material and methods
The strain C3/8 of Hartmannella vermiformis was isolated from the sediments of a water reservoir of a potable water-treatment plant near Bonn (Germany). For details see Smirnov and Michel (1999).
ISOLATION OF DNA
For molecular biological investigations actively multiplying (dividing) amoebae (~106 cells) were harvested from plate cultures with a sterile cotton tipped applicator and washed 3x in sterile 0,9% NaCl by centrifugation 500 g/ 7 min. Whole-cell DNA was isolated by a modified UNSET-procedure (Hugo et al., 1992). Briefly, the pellet was re-suspended in 500 ці of UNSET-lysis buffer, overlaid with 500 ці phenol-chloroform-isoamylalcohol (PCI) and shaken gently for 5 h. DNA was extracted by multiple PCI-extraction, precipitated in alcohol, air dried and re-suspended in 30 ці of sterile double-distilled water.
AMPLIFICATION
The 18S rRNA-gene was amplified using the SSU1 and SSU2 primers (Gast et al., 1996). These are universal eukaryotic primers, complementary to the strongly conserved ends of the eukaryotic 18S rRNA genes. A standard amplification program with 30 cycles; 95°C 1 min, 51°C 2 min., 72°C 3 min. was used for PCR. The amplification of the 18S rRNA gene was visualized by ethidium-bromide in an agarose-gel electrophoresis and the amplified gene was sequenced stepwise by direct
sequencing from the PCR-product using the Thermo SequenaseTM II sequencing kit (Amersham Pharmacia Biotech GmbH, Wien, Austria) and subsequent construction of complementary internal primers. Sequences were obtained from both strands. Sequencing was carried out in a 310 ABI PRISM® automated sequencer (PE Applied Biosystems, Langen, Germany).
18S rDNA IDENTIFICATION
Sequence data were processed with the GeneDoc (Nicholas et al., 1997) sequence editor and sequences were compared to the ones of published strains using BLAST Search (Altschul et al., 1990). ClustalX (Thompson et al., 1997) was used for pairwise alignment and the percentages of sequence dissimilarities were calculated.
Sequence data reported in this paper are available at GenBank under the following accession number: AF426157.
Results
The 18S rRNA gene of Hartmannella sp. strain C3/ 8 is 1841 bp in length, with a G+C content of49,75%. It exhibits the highest sequence similarity to the CCAP 1534/7B strain of H. vermiformis (GenBank X75515). Sequence dissimilarities of only 0,60% to the CCAP 1534/7B strain, and 0,65% to the 0S-101 (GenBank X75514) and the CDC-19 (GenBank X75513) strains of H. vermiformis, respectively, support the affiliation of strain C3/8 to the species H. vermiformis despite its divergent cyst morphology.
The sequence dissimilarities to all Hartmannella strains available in GenBank and to other Gymn-amoebia are shown in Table 1. For alignment and calculation of percentages of sequence dissimilarities within the Gymnamoebia we included the sequences of the next closest related representatives of the Gymnamoebia, namely Echinamoeba sp. strain SH274 (GenBank AF293895) and Balamuthia mandrillaris strain CDC:V039 (GenBank AF019071), which are both the only sequenced strains of their genera and the type strain of Acanthamoeba castellanii, strain Castellani (GenBank U07413). Among the few amoeba genera sequenced so far Echinamoeba is obviously the most closely related genus to Hartmannella.
Discussion
It was shown that the sequence dissimilarities of the Hartmannella strain C3/8 to three of the H. vermiformis strains sequenced so far is smaller than the sequence dissimilarities within the strains presently
Protistology ■ 187
Table 1. Percentages of sequence dissimilarities within H. vermiformis and to the most closely related
Gymnamoebia (H. vermiformis).
Strain CCAP 1534/1B (H ve/mf/mi) C3/8 (H ve/mf/mi) 0S-1G1 (H vemf/m/j) CDC-19 Balamuth (H ve/mf/mi) SH214 E ex/z/a'an) CDC:V039 B жadУ/a/i> Castellani (A. cat/a/)
CCAP 1534/1B (H ve/mf/m/j) G 0.60 G.44 G.49 1.2G 11.11 2G.21 2G.6G
C3/8 (H ve/mf/m/j) 0.60 G G.65 G.65 1.2G 11.82 2G.26 2G.8G
0S-1G1 (H ve/mf/m/j) G.44 G.65 G G.38 1.G9 11.12 2G.GG 2G.16
CDC-19 G.49 G.65 G.38 G G.81 11.11 2G.11 2G.11
Balamuth 1.2G 1.2G 1.G9 G.81 G 11.66 2G.GG 2G.64
SH214 E ex/z/a'an) 11.11 11.82 11.12 11.11 11.66 G 2G.91 11.19
CDC:V039 B madlai) 2G.21 2G.26 2G.GG 2G.11 2G.GG 2G.91 G 24.85
Castellani (A afe//an) 20.60 2G.8G 2G.16 2G.11 2G.64 11.19 24.85 G
assigned to H. vermiformis. Thus, in spite of the untypical cyst structure, we consider the strain to be indeed a representative of H. vermiformis as had already been proposed by Smirnov and Michel (1999) on the basis of morphological features of the trophozoites. The high 18S rRNA gene sequence identity of the Hartmannella strain C3/8 with H. vermiformis suggests a more pronounced intraspecific morphological variation than initially considered by Page (1986).
As cysts with similar spatial separation of the ectocyst are also known from Paratetramitus jugosus, Vahlkampfia russelli, and even from V. ovis, the cysts of the present Hartmannella strain can be easily confused with the species mentioned. Again it is shown that cyst morphology alone is not sufficient to exclude misiden-tification of various species of Gymnamoebia. It must be emphasized once more that corresponding trophozoites have to be thoroughly studied in the hanging drop with respect to shape, locomotion behaviour, nuclear division, etc. Moreover, it has become apparent that the shape of cyst walls can be altered by varying growth conditions and polymorphisms even within one clone, as repeatedly shown for Acanthamoeba, represent a considerable obstacle in a morphological approach to classification. The genus Naegleria, another free-living amoeba, is currently more clearly defined by 18S rDNA sequence data than it is by phenotypic characters (Brown and De Jonckheere, 1999). Probably the most reliable method for identification and classification of protozoa to date is to accept the evidence from both structural and molecular data and use accumulated information from several sources as proposed by Cavalier-Smith (1998) and corroborated by Corliss
(2001). It is of the highest importance to compare organisms from more than one viewpoint. A major problem in the identification of free-living amoebae including Hartmannella is their lack of any sexual processes implying general difficulties in species determination well known from other non- sexual organisms.
The strain C3/8 shows a sequence identity of more than 99% to three of the known H. vermiformis strains. In H. vermiformis sequence variation is extremely low (Weekers et al., 1994), we found up to 1.2% sequence dissimilarities within this species, and all known Hartmannella sequences cluster together in phylogenetic analysis (data not shown). Thus, Hartmannella is one of the few amoebic genera which do not show any suspicion of polyphyly. Several genera of the free-living amoebae have been demonstrated to be poly-phyletic, as e.g. Vahlkampfia (Brown and De Jonckheere, 1999). Recently, polyphyly has also been detected within the leptomyxid amoebae (Amaral-Zettler et al., 2000).
In spite of very low sequence variation, Wfeekers et al. (1994) assume a separation between European and North American strains of H. vermiformis, because in their study European and North American strains clustered separately. Interestingly, also the C3/8 strain deriving from sediments of a water reservoir in Germany showed the highest sequence identity to a European isolate of H. vermiformis, namely to the CCAP 1534/7B strain, which had been isolated in the United Kingdom. However, the C3/8 strain showed equal sequence dissimilarity to the other European isolate, the 0S-101 strain, and to one of the isolates
from the United States, the CDC-19 strain, which argues against the proposed geographical correlation.
Among the few amoeba genera sequenced so far, Echinamoeba is the genus which is most closely related to Hartmannella. This has also been stated by Amaral-Zettler et al. (2000). In fact, Echinamoeba exundans was first described as a species of Hartmannella (Page, 1967). Amaral-Zettler et al. (2000) postulate that Echinamoeba evolved relatively recently from a Hartmannella-like ancestor. Several species of both genera are highly thermotolerant and both can harbour Legionella pneumophila. However, unlike Hartmannella, Echinamoeba species may also exhibit finely-pointed subpseudopodia (echino-podia) and thus resemble tiny acanthamoebids. This would correlate to the sequence dissimilarity, which is as high between Echinamoeba and Acanthamoeba as it is between Echinamoeba and Hartmannella.
Altogether, the results of our study suggest the existence of cyst polymorphism also within the species H. vermiformis as it has already been demonstrated for various other amoeboid species and this once again questions the value of morphological characters for identification of free-living amoebae at the species level. Moreover, these results once again touch upon the problem of the application of the classical binomial nomenclature to organisms without sexual reproduction.
References
Aitken D., Hay J., Kinnear F.B., Kirkness C.M., Lee W.R. and Seal D.V. 1996. Amebic keratitis in a wearer of disposable contact lenses due to a mixed Vahlkampfia and Hartmannella infection. Ophthalmology. 103, 485-494.
Altschul S.F., Gish W., Miller W, Myers E.W. and Lipman D. 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403-410.
Amaral-Zettler L.A., Nerad T.A., O’Kelly C.J., Peglar M.T., Gillevet P.M., Silberman J.D. and Sogin M.L. 2000. A molecular reassessment of the Leptomyxid amoebae. Protist. 151, 275-282. Brown S. and De Jonckheere J.F. 1999. A reevaluation of the amoeba genus Vahlkampfia based on SSUrDNA sequences. Europ. J. Protistol. 35, 49-54. Cavalier-Smith T. 1998. A revised six-kingdom system of life. Biol. Rev. 73, 203-266.
Corliss J.O. 2001. Protist systematics. Encyclopeida ofLife Sciences. Nature Publishing Group. www.els.net
De Jonckheere J.F. and Brown S. 1998. Is the free-living amoeba Hartmannella causing keratitis ? Clin. Infect. 36, 349-350.
De Jonckheere J.F. and Brown S. 1999. Non-Acanth-amoeba amoebic keratitis. Cornea. 18, 499-501.
Fields B.S., Nerad T.A., Sawyer T.K., King C.H., Barbaree J.M., Martin W.T., Morrill W.E. and Sanden G.N. 1990. Characterization of an axenic strain of Hartmannella vermiformis obtained from an investigation of nosocomial legionellosis. J. Protozool. 37, 581-583.
Gast R.J., Ledee D.R., Fuerst P.A. and Byers T. 1996. Subgenus systematics of Acanthamoeba: Four nuclear 18S rDNA sequence types. J. Euk. Microbiol. 43, 498-504.
Hugo E.R., Stewart V.J., Gast R.J. and Byers T.J. 1992. Purification of amoeba mtDNA using the UNSET procedure. In: Protocols in protozoology (Eds. Soldo A. T. and Lee J. J.) Allen, Lawrence, Kansas.
Inoue T., Asari S., Tahara K., Hayashi K., Kiritoshi A. and Shimomura Y 1998. Acanthamoeba keratitis with symbiosis of Hartmannella ameba. Am. J. 0phthalmol. 125, 721-723.
Nicholas K.B., Nicholas H.B.Jr. and Deerfield D.WII. 1997. GeneDoc: Analysis and Visualization of Genetic Variation. Embnew. News. 4, 14.
Page F.C. 1967. Taxonomic criteria for limax amoebae, with description of three new species of Hartmannella and three of Vahlkampfia. J. Protozool. 14, 499-521.
Page F.C. 1986. The limax amoebae: comparative fine structure of the Hartmannellidae (Lobosea) and further comparisons with the Vahlkampfiidae (Heterolobosea). Protistologica. 21, 361-383.
Smirnov A.V. and Michel R. 1999. New data on the cyst structure of Hartmannella vermiformis Page, 1967 (Lobosea, Gymnamoebia). Protistology. 1, 82-85.
Thompson J.D., Gibson T.J., Plewniak F., Jeanmougin F. and Higgins D.G. 1997. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 24, 4876-4882.
Weekers P.H., Gast R.J., Fuerst P.A. and Byers T.J. 1994. Sequence variations in small-subunit ribosomal RNAs of Hartmannella vermiformis and their phylogenetic implications. Mol. Biol. Evol. 11, 684-690.
Address for correspondence: Horst Aspock. Department for Medical Parasitology, Clinical Institute of Hygiene, University ofVienna, Kinderspitalgasse 15, 1095 Vienna, Austria. Phone: 043-1-4277-79430. Fax: 043-1-42779794. E-mail: Horst.Aspoeck@univie.ac.at.
The manuscript is presented by Norbert Hulsmann
