Naked Foraminiferans Revealed

Home , Allogromia, Amoeba, Reticulomyxa

scientific correspondence Naked foraminiferans revealed t is generally assumed that the first fossil We have obtained ribosomal DNA and analysis of foraminiferan actin-coding Iappearance of a group of organisms corre- actin gene sequences from Reticulomyxa genes. Two actin gene families were found sponds to its evolutionary origin. But we filosa, a well-known “giant freshwater in Reticulomyxa as well as in the two have molecular evidence that extant mem- amoeba”4 that lacks a shell but possesses foraminiferans examined (Allogromia sp. bers of the most abundant microfossil- pseudopodia that are strikingly similar to and Ammonia sp.). Both actin gene forming group, the Foraminifera, include those of foraminiferans5. Phylogenetic sequences branch next to the Euglenozoa in ‘naked’ amoeboid species, indicating that analyses of these molecular data show that the lower part of the phylogenetic tree ancestral foraminiferans could be unfos- R. filosa is actually a foraminiferan, indicat- (Fig. 1b) in a position remarkably similar to silized. This means that the origin of the ing that certain foraminiferans might lack a that suggested by analysis of the SSU rDNA group might be much earlier than has been shell and live in freshwater environments. data. Each actin family forms a clade sup- deduced from the fossil record. This might The complete small subunit (SSU) ported by a high bootstrap value (98% and help to explain the conflicting molecular rDNA sequences were obtained from 100%) in which Allogromia appears as a sis- and fossil data on the origin of the R. filosa and five foraminiferan species. The ter group to Reticulomyxa and Ammonia. Foraminifera. SSU rDNA of R. filosa shows several com- These relationships are supported by high Foraminifera are classically defined as mon features with other foraminiferans, bootstrap values (92% and 96%) and marine, shelled protists. Although the including a very low G+C content (32.6%) remain stable in all analyses based on foraminiferan fossil record extends to the and several foraminiferan-specific inser- amino acid and DNA sequences. Cambrian period1, molecular data2 indicate tions located in the conserved regions of the The placement of Reticulomyxa within a much earlier origin for these important gene that make it unusually long (3,347 the Foraminifera leads us to challenge the microfossils. This disparity suggests that nucleotides). Phylogenetic analyses of SSU micropalaeontological bias imparted to this microfossils do not provide adequate evi- rDNA sequences, using different computa- group: the Foraminifera can no longer be dence for the origin of this group. Such a tional methods, consistently show that defined strictly as shelled protists. Indeed, gap in the fossil record could be explained Reticulomyxa branches within the their only taxonomically consistent mor- by the existence of non-skeletonizing Foraminifera (Fig. 1a). Its position within phological character is the presence of gran- (‘naked’) foraminiferans that were not pre- the Foraminifera is not well established, but uloreticulopodia. Although R. filosa served in sediments. However, as a comparison of more than 100 probably lost its shell secondarily as an foraminiferans did not include a naked foraminiferan partial SSU rDNA sequences adaptation to the freshwater environment, species, we looked for the potential (not shown) suggests that the genus is most our findings nevertheless suggest that foraminiferan ancestor within the class closely related to some primitive thecate ancestral foraminiferans could have been Athalamea, a sister group to the Allogromiida. naked and existed well before the dawn of Foraminifera3. These results were confirmed by an skeletonization. This hypothesis helps to reconcile the conflicting molecular and fos- ab sil data on the origin of the Foraminifera. Our data also provide support for recent H. sapiens H. sapiens 6,7 100 64 ANIMALIA molecular studies that implicate a pro- X. laevis X. laevis 100 longed period of Precambrian diversifica- S. cerevisiae S. pombe tion among the major eukaryotic groups. 100 59 FUNGI We predict that naked foraminiferans 0.05 100 80 P. carinii P. carinii similar to their hypothetical ancestors 93 E. histolytica D. discoideum might still be living in the marine environ- 100 53 AMOEBAE ment. Finding these species is essential for D. discoideum P. polycephalum 98 SLIME MOULDS future molecular studies of the group. P. polycephalum E. histolytica Jan Pawlowski*†, Ignacio Bolivar*, Ammonia sp. Ammonia sp. José Fahrni*, Colomban de Vargas*, FORAMINIFERA 92 78 78 Samuel S. Bowser‡ Trochammina sp. R. filosa 98 72 *Department of Zoology and Animal Biology, 2 98 100 R. filosa Allogromia sp. University of Geneva,

100 Allogromia sp. Ammonia sp. 1224 Chêne-Bougeries, Geneva, Switzerland 96 e-mail: [email protected] S. orbiculus R. filosa 100 100 †Museum of Natural History, 1 P. pertusus Allogromia sp. 1211 Geneva 4, Switzerland ‡Wadsworth Center, T. cruzi T. cruzi 74 EUGLENOZOA 100 New York State Department of Health, E. gracilis L. major P.O. Box 509, Albany, New York 12201, USA

G. ardea ARCHAEZOA G. ardea 1. Culver, S. J. Science 254, 689–691 (1991). 2. Pawlowski, J. et al. Mol. Biol. Evol. 14, 498–505 (1997). 3. Lee J. J., Hutner, S. H. & Bovee, E. C. (eds) An Illustrated Guide to the Protozoa (Allen, Lawrence, Kansas, 1985). Figure 1 Phylogenetic position of Reticulomyxa filosa and foraminiferans. The position was inferred from 4. Koonce, M. P., Euteneuer, U. & Schliwa, M. J. Cell Sci. (suppl. 5) complete sequences of SSU rDNA (a) and actin amino acids (b), using the neighbour-joining method with 145–159 (1986). Kimura 2 and Dayhoff’s PAM (Percent Accepted Mutations) matrix distances, respectively. The two 5. Travis, J. L. & Bowser, S. S. in Biology of Foraminifera (eds foraminiferan actin gene families are numbered 1 and 2. The scale bar corresponds to the number of sub- Lee, J. J. & Anderson, O. R.) 91–156 (Academic, London, 1991). 6. Cooper, A. & Fortey, R. Trends Ecol. Evol. 13, 151–156 (1998). stitutions per site. The GenBank accession numbers of new sequences reported here are AJ 132367 to 7. Bromham, L., Rambaut, A., Fortey, R., Cooper, A. & Penny, D. AJ 132375. Proc. Natl Acad. Sci. USA 95, 12386–12389 (1998).